Your IP : 216.73.216.48
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Version Control with Subversion</title><link rel="stylesheet" type="text/css" href="styles.css"></link><meta name="generator" content="DocBook XSL Stylesheets V1.76.1"></meta></head><body><div class="book" title="Version Control with Subversion"><div class="titlepage"><div><div><h1 class="title"><a id="svn"></a>Version Control with Subversion</h1></div><div><h2 class="subtitle">For Subversion 1.7</h2></div><div><h2 class="subtitle">(Compiled from r5136)</h2></div><div><div class="authorgroup"><div class="author"><h3 class="author"><span class="firstname">Ben</span> <span class="surname">Collins-Sussman</span></h3></div><div class="author"><h3 class="author"><span class="firstname">Brian W.</span> <span class="surname">Fitzpatrick</span></h3></div><div class="author"><h3 class="author"><span class="firstname">C. Michael</span> <span class="surname">Pilato</span></h3></div></div></div><div><p class="copyright">Copyright © 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011 Ben Collins-Sussman, Brian W. Fitzpatrick, C. Michael Pilato</p></div><div><div class="legalnotice" title="Legal Notice"><a id="idp166320"></a><p>This work is licensed under the Creative
Commons Attribution License. To view a copy of this license,
visit <a class="ulink" href="http://creativecommons.org/licenses/by/2.0/" target="_top">http://creativecommons.org/licenses/by/2.0/</a>
or send a letter to Creative Commons, 559 Nathan Abbott Way,
Stanford, California 94305, USA.</p></div></div></div><hr></hr></div><div class="toc"><p><strong>Table of Contents</strong></p><dl><dt><span class="preface"><a href="#svn.foreword">Foreword</a></span></dt><dt><span class="preface"><a href="#svn.preface">Preface</a></span></dt><dd><dl><dt><span class="sect1"><a href="#svn.intro.whatis">What Is Subversion?</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.intro.righttool">Is Subversion the Right Tool?</a></span></dt><dt><span class="sect2"><a href="#svn.intro.history">Subversion's History</a></span></dt><dt><span class="sect2"><a href="#svn.intro.architecture">Subversion's Architecture</a></span></dt><dt><span class="sect2"><a href="#svn.intro.components">Subversion's Components</a></span></dt><dt><span class="sect2"><a href="#svn.intro.whatsnew">What's New in Subversion</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.preface.audience">Audience</a></span></dt><dt><span class="sect1"><a href="#svn.preface.howread">How to Read This Book</a></span></dt><dt><span class="sect1"><a href="#svn.preface.organization">Organization of This Book</a></span></dt><dt><span class="sect1"><a href="#svn.preface.free">This Book Is Free</a></span></dt><dt><span class="sect1"><a href="#svn.preface.acks">Acknowledgments</a></span></dt></dl></dd><dt><span class="chapter"><a href="#svn.basic">1. Fundamental Concepts</a></span></dt><dd><dl><dt><span class="sect1"><a href="#svn.basic.version-control-basics">Version Control Basics</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.basic.repository">The Repository</a></span></dt><dt><span class="sect2"><a href="#svn.basic.working-copy">The Working Copy</a></span></dt><dt><span class="sect2"><a href="#svn.basic.vsn-models">Versioning Models</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.basic.vsn-models.problem-sharing">The problem of file sharing</a></span></dt><dt><span class="sect3"><a href="#svn.basic.vsn-models.lock-unlock">The lock-modify-unlock solution</a></span></dt><dt><span class="sect3"><a href="#svn.basic.vsn-models.copy-merge">The copy-modify-merge solution</a></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.basic.in-action">Version Control the Subversion Way</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.basic.svn-repositories">Subversion Repositories</a></span></dt><dt><span class="sect2"><a href="#svn.basic.in-action.revs">Revisions</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.reposurls">Addressing the Repository</a></span></dt><dt><span class="sect2"><a href="#svn.basic.in-action.wc">Subversion Working Copies</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.basic.in-action.track-repos">How the working copy works</a></span></dt><dt><span class="sect3"><a href="#svn.basic.in-action.wc-funcdamentals">Fundamental working copy interactions</a></span></dt><dt><span class="sect3"><a href="#svn.basic.in-action.mixedrevs">Mixed-revision working copies</a></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.basic.summary">Summary</a></span></dt></dl></dd><dt><span class="chapter"><a href="#svn.tour">2. Basic Usage</a></span></dt><dd><dl><dt><span class="sect1"><a href="#svn.tour.help">Help!</a></span></dt><dt><span class="sect1"><a href="#svn.tour.importing">Getting Data into Your Repository</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.tour.importing.import">Importing Files and Directories</a></span></dt><dt><span class="sect2"><a href="#svn.tour.importing.layout">Recommended Repository Layout</a></span></dt><dt><span class="sect2"><a href="#svn.tour.importing.naming">What's In a Name?</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.tour.initial">Creating a Working Copy</a></span></dt><dt><span class="sect1"><a href="#svn.tour.cycle">Basic Work Cycle</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.tour.cycle.update">Update Your Working Copy</a></span></dt><dt><span class="sect2"><a href="#svn.tour.cycle.edit">Make Your Changes </a></span></dt><dt><span class="sect2"><a href="#svn.tour.cycle.examine">Review Your Changes</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.tour.cycle.examine.status">See an overview of your changes</a></span></dt><dt><span class="sect3"><a href="#svn.tour.cycle.examine.diff">Examine the details of your local modifications</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.tour.cycle.revert">Fix Your Mistakes</a></span></dt><dt><span class="sect2"><a href="#svn.tour.cycle.resolve">Resolve Any Conflicts</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.tour.cycle.resolve.diff">Viewing conflict differences interactively</a></span></dt><dt><span class="sect3"><a href="#svn.tour.cycle.resolve.resolve">Resolving conflict differences interactively</a></span></dt><dt><span class="sect3"><a href="#svn.tour.cycle.resolve.pending">Postponing conflict resolution</a></span></dt><dt><span class="sect3"><a href="#svn.tour.cycle.resolve.byhand">Merging conflicts by hand</a></span></dt><dt><span class="sect3"><a href="#svn.tour.cycle.resolve.theirsfull">Discarding your changes in favor of a newly fetched
revision</a></span></dt><dt><span class="sect3"><a href="#svn.tour.cycle.resolve.revert">Punting: using svn revert</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.tour.cycle.commit">Commit Your Changes</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.tour.history">Examining History</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.tour.history.diff">Examining the Details of Historical Changes</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.tour.history.diff.local">Examining local changes</a></span></dt><dt><span class="sect3"><a href="#svn.tour.history.diff.wcrepos">Comparing working copy to repository</a></span></dt><dt><span class="sect3"><a href="#svn.tour.history.diff.reposrepos">Comparing repository revisions</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.tour.history.log">Generating a List of Historical Changes</a></span></dt><dt><span class="sect2"><a href="#svn.tour.history.browsing">Browsing the Repository</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.tour.history.browsing.cat">Displaying file contents</a></span></dt><dt><span class="sect3"><a href="#svn.tour.history.browsing.annotate">Displaying line-by-line change attribution</a></span></dt><dt><span class="sect3"><a href="#svn.tour.history.browsing.list">Listing versioned directories</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.tour.history.snapshots">Fetching Older Repository Snapshots</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.tour.cleanup">Sometimes You Just Need to Clean Up</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.tour.cleanup.disposal">Disposing of a Working Copy</a></span></dt><dt><span class="sect2"><a href="#svn.tour.cleanup.interruption">Recovering from an Interruption</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.tour.treeconflicts">Dealing with Structural Conflicts</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.tour.treeconflicts.example">An Example Tree Conflict</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.tour.summary">Summary</a></span></dt></dl></dd><dt><span class="chapter"><a href="#svn.advanced">3. Advanced Topics</a></span></dt><dd><dl><dt><span class="sect1"><a href="#svn.tour.revs.specifiers">Revision Specifiers</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.tour.revs.keywords">Revision Keywords</a></span></dt><dt><span class="sect2"><a href="#svn.tour.revs.dates">Revision Dates</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.advanced.pegrevs">Peg and Operative Revisions</a></span></dt><dt><span class="sect1"><a href="#svn.advanced.props">Properties</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.advanced.props.why">Why Properties?</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.props.manip">Manipulating Properties</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.props.workflow">Properties and the Subversion Workflow</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.props.auto">Automatic Property Setting</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.advanced.props.file-portability">File Portability</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.advanced.props.special.mime-type">File Content Type</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.props.special.executable">File Executability</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.props.special.eol-style">End-of-Line Character Sequences</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.advanced.props.special.ignore">Ignoring Unversioned Items</a></span></dt><dt><span class="sect1"><a href="#svn.advanced.props.special.keywords">Keyword Substitution</a></span></dt><dt><span class="sect1"><a href="#svn.advanced.sparsedirs">Sparse Directories</a></span></dt><dt><span class="sect1"><a href="#svn.advanced.locking">Locking</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.advanced.locking.creation">Creating Locks</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.locking.discovery">Discovering Locks</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.locking.break-steal">Breaking and Stealing Locks</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.locking.lock-communication">Lock Communication</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.advanced.externals">Externals Definitions</a></span></dt><dt><span class="sect1"><a href="#svn.advanced.changelists">Changelists</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.advanced.changelists.creating">Creating and Modifying Changelists</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.changelists.asfilters">Changelists As Operation Filters</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.changelists.limitations">Changelist Limitations</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.serverconfig.netmodel">Network Model</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.serverconfig.netmodel.reqresp">Requests and Responses</a></span></dt><dt><span class="sect2"><a href="#svn.serverconfig.netmodel.creds">Client Credentials</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.serverconfig.netmodel.credcache">Caching credentials</a></span></dt><dt><span class="sect3"><a href="#svn.tour.initial.disabling-password-caching">Disabling password caching</a></span></dt><dt><span class="sect3"><a href="#svn.tour.initial.authn-cache-purge">Removing cached credentials</a></span></dt><dt><span class="sect3"><a href="#svn.tour.initial.different-user">Command-line authentication</a></span></dt><dt><span class="sect3"><a href="#svn.tour.initial.authn-wrapup">Authentication wrap-up</a></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.advanced.summary">Summary</a></span></dt></dl></dd><dt><span class="chapter"><a href="#svn.branchmerge">4. Branching and Merging</a></span></dt><dd><dl><dt><span class="sect1"><a href="#svn.branchmerge.whatis">What's a Branch?</a></span></dt><dt><span class="sect1"><a href="#svn.branchmerge.using">Using Branches</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.branchmerge.using.create">Creating a Branch</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.using.work">Working with Your Branch</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.using.concepts">The Key Concepts Behind Branching</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.branchmerge.basicmerging">Basic Merging</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.branchmerge.changesets">Changesets</a></span></dt><dt><span class="sect2"><a href="#svn.branchemerge.basicmerging.stayinsync">Keeping a Branch in Sync</a></span></dt><dt><span class="sect2"><a href="#svn.branchemerge.basicmerging.reintegrate">Reintegrating a Branch</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.basicmerging.mergeinfo">Mergeinfo and Previews</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.basicmerging.undo">Undoing Changes</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.basicmerging.resurrect">Resurrecting Deleted Items</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.branchmerge.advanced">Advanced Merging</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.branchmerge.cherrypicking">Cherrypicking</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.advanced.advancedsyntax">Merge Syntax: Full Disclosure</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.nomergedata">Merges Without Mergeinfo</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.advanced.mergeconflicts">More on Merge Conflicts</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.advanced.blockchanges">Blocking Changes</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.advanced.reintegratetwice">Keeping a Reintegrated Branch Alive</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.advanced.logblame">Merge-Sensitive Logs and Annotations</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.advanced.ancestry">Noticing or Ignoring Ancestry</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.advanced.moves">Merges and Moves</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.advanced.pre1.5clients">Blocking Merge-Unaware Clients</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.advanced.finalword">The Final Word on Merge Tracking</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.branchmerge.switchwc">Traversing Branches</a></span></dt><dt><span class="sect1"><a href="#svn.branchmerge.tags">Tags</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.branchmerge.tags.mksimple">Creating a Simple Tag</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.tags.mkcomplex">Creating a Complex Tag</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.branchmerge.maint">Branch Maintenance</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.branchmerge.maint.layout">Repository Layout</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.maint.lifetime">Data Lifetimes</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.branchmerge.commonpatterns">Common Branching Patterns</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.branchmerge.commonpatterns.release">Release Branches</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.commonpatterns.feature">Feature Branches</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.advanced.vendorbr">Vendor Branches</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.advanced.vendorbr.general">General Vendor Branch Management Procedure</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.vendorbr.svn_load_dirs">svn_load_dirs.pl</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.branchmerge.when">To Branch or Not to Branch?</a></span></dt><dt><span class="sect1"><a href="#svn.branchmerge.summary">Summary</a></span></dt></dl></dd><dt><span class="chapter"><a href="#svn.reposadmin">5. Repository Administration</a></span></dt><dd><dl><dt><span class="sect1"><a href="#svn.reposadmin.basics">The Subversion Repository, Defined</a></span></dt><dt><span class="sect1"><a href="#svn.reposadmin.planning">Strategies for Repository Deployment</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.reposadmin.projects.chooselayout">Planning Your Repository Organization</a></span></dt><dt><span class="sect2"><a href="#svn.reposadmin.basics.hosting">Deciding Where and How to Host Your Repository</a></span></dt><dt><span class="sect2"><a href="#svn.reposadmin.basics.backends">Choosing a Data Store</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.reposadmin.basics.backends.bdb">Berkeley DB</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.basics.backends.fsfs">FSFS</a></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.reposadmin.create">Creating and Configuring Your Repository</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.reposadmin.basics.creating">Creating the Repository</a></span></dt><dt><span class="sect2"><a href="#svn.reposadmin.create.hooks">Implementing Repository Hooks</a></span></dt><dt><span class="sect2"><a href="#svn.reposadmin.create.bdb">Berkeley DB Configuration</a></span></dt><dt><span class="sect2"><a href="#svn.reposadmin.create.fsfs">FSFS Configuration</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.reposadmin.maint">Repository Maintenance</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.reposadmin.maint.tk">An Administrator's Toolkit</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.reposadmin.maint.tk.svnadmin">svnadmin</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.tk.svnlook">svnlook</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.tk.svndumpfilter">svndumpfilter</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.tk.svnrdump">svnrdump</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.tk.svnsync">svnsync</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.tk.fsfsreshard">fsfs-reshard.py</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.tk.bdbutil">Berkeley DB utilities</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.reposadmin.maint.setlog">Commit Log Message Correction</a></span></dt><dt><span class="sect2"><a href="#svn.reposadmin.maint.diskspace">Managing Disk Space</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.reposadmin.maint.diskspace.deltas">How Subversion saves disk space</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.diskspace.deadtxns">Removing dead transactions</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.diskspace.bdblogs">Purging unused Berkeley DB logfiles</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.diskspace.fsfspacking">Packing FSFS filesystems</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.reposadmin.maint.recovery">Berkeley DB Recovery</a></span></dt><dt><span class="sect2"><a href="#svn.reposadmin.maint.migrate">Migrating Repository Data Elsewhere</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.reposadmin.maint.migrate.svnadmin">Repository data migration using svnadmin</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.migrate.svnrdump">Repository data migration using svnrdump</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.reposadmin.maint.filtering">Filtering Repository History</a></span></dt><dt><span class="sect2"><a href="#svn.reposadmin.maint.replication">Repository Replication</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.reposadmin.maint.replication.svnsync">Replication with svnsync</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.replication.svnsync-partial">Partial replication with svnsync</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.replication.svnsync-init-nonempty">A quick trick for mirror creation</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.replication.wrapup">Replication wrap-up</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.reposadmin.maint.backup">Repository Backup</a></span></dt><dt><span class="sect2"><a href="#svn.reposadmin.maint.uuids">Managing Repository UUIDs</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.reposadmin.maint.moving-and-removing">Moving and Removing Repositories</a></span></dt><dt><span class="sect1"><a href="#svn.reposadmin.summary">Summary</a></span></dt></dl></dd><dt><span class="chapter"><a href="#svn.serverconfig">6. Server Configuration</a></span></dt><dd><dl><dt><span class="sect1"><a href="#svn.serverconfig.overview">Overview</a></span></dt><dt><span class="sect1"><a href="#svn.serverconfig.choosing">Choosing a Server Configuration</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.serverconfig.choosing.svnserve">The svnserve Server</a></span></dt><dt><span class="sect2"><a href="#svn.serverconfig.choosing.svn-ssh">svnserve over SSH</a></span></dt><dt><span class="sect2"><a href="#svn.serverconfig.choosing.apache">The Apache HTTP Server</a></span></dt><dt><span class="sect2"><a href="#svn.serverconfig.choosing.recommendations">Recommendations</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.serverconfig.svnserve">svnserve, a Custom Server</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.serverconfig.svnserve.invoking">Invoking the Server</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.invoking.daemon">svnserve as daemon</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.invoking.inetd">svnserve via inetd</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.invoking.tunnel">svnserve over a tunnel</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.invoking.winservice">svnserve as a Windows service</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.invoking.launchd">svnserve as a launchd job</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.serverconfig.svnserve.auth">Built-in Authentication and Authorization</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.auth.users">Create a users file and realm</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.auth.general">Set access controls</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.serverconfig.svnserve.sasl">Using <span class="command"><strong>svnserve</strong></span> with SASL</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.sasl.authn">Authenticating with SASL</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.sasl.encryption">SASL encryption</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.serverconfig.svnserve.sshauth">Tunneling over SSH</a></span></dt><dt><span class="sect2"><a href="#svn.serverconfig.svnserve.sshtricks">SSH Configuration Tricks</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.sshtricks.setup">Initial setup</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.sshtricks.fixedcmd">Controlling the invoked command</a></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.serverconfig.httpd">httpd, the Apache HTTP Server</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.serverconfig.httpd.prereqs">Prerequisites</a></span></dt><dt><span class="sect2"><a href="#svn.serverconfig.httpd.basic">Basic Apache Configuration</a></span></dt><dt><span class="sect2"><a href="#svn.serverconfig.httpd.authn">Authentication Options</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.serverconfig.httpd.authn.basic">Basic authentication</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.httpd.authn.digest">Digest authentication</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.serverconfig.httpd.authz">Authorization Options</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.serverconfig.httpd.authz.blanket">Blanket access control</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.httpd.authz.perdir">Per-directory access control</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.httpd.authz.pathauthzoff">Disabling path-based checks</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.serverconfig.httpd.ssl">Protecting network traffic with SSL</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.serverconfig.httpd.ssl.server">Subversion server SSL certificate configuration</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.httpd.ssl.client">Subversion client SSL certificate management</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.serverconfig.httpd.extra">Extra Goodies</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.serverconfig.httpd.extra.browsing">Repository browsing</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.httpd.extra.logging">Apache logging</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.httpd.extra.writethruproxy">Write-through proxying</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.httpd.extra.other">Other Apache features</a></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.serverconfig.pathbasedauthz">Path-Based Authorization</a></span></dt><dt><span class="sect1"><a href="#svn.serverconfig.operational-logging">High-level Logging</a></span></dt><dt><span class="sect1"><a href="#svn.serverconfig.optimization">Server Optimization</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.serverconfig.optimization.caching">Data Caching</a></span></dt><dt><span class="sect2"><a href="#svn.serverconfig.optimization.compression">Network Compression of Data</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.serverconfig.multimethod">Supporting Multiple Repository Access Methods</a></span></dt></dl></dd><dt><span class="chapter"><a href="#svn.customization">7. Customizing Your Subversion Experience</a></span></dt><dd><dl><dt><span class="sect1"><a href="#svn.advanced.confarea">Runtime Configuration Area</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.advanced.confarea.layout">Configuration Area Layout</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.confarea.windows-registry">Configuration and the Windows Registry</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.confarea.opts">Configuration Options</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.advanced.confarea.opts.servers">Servers</a></span></dt><dt><span class="sect3"><a href="#svn.advanced.confarea.opts.config">Config</a></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.advanced.l10n">Localization</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.advanced.l10n.understanding">Understanding Locales</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.l10n.svnuse">Subversion's Use of Locales</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.advanced.externaleditors">Using External Editors</a></span></dt><dt><span class="sect1"><a href="#svn.advanced.externaldifftools">Using External Differencing and Merge Tools</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.advanced.externaldifftools.diff">External diff</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.externaldifftools.diff3">External diff3</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.externaldifftools.merge">External merge</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.customization.summary">Summary</a></span></dt></dl></dd><dt><span class="chapter"><a href="#svn.developer">8. Embedding Subversion</a></span></dt><dd><dl><dt><span class="sect1"><a href="#svn.developer.layerlib">Layered Library Design</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.developer.layerlib.repos">Repository Layer</a></span></dt><dt><span class="sect2"><a href="#svn.developer.layerlib.ra">Repository Access Layer</a></span></dt><dt><span class="sect2"><a href="#svn.developer.layerlib.client">Client Layer</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.developer.usingapi">Using the APIs</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.developer.usingapi.apr">The Apache Portable Runtime Library</a></span></dt><dt><span class="sect2"><a href="#svn.developer.usingapi.funcsbatons">Functions and Batons</a></span></dt><dt><span class="sect2"><a href="#svn.developer.usingapi.urlpath">URL and Path Requirements</a></span></dt><dt><span class="sect2"><a href="#svn.developer.usingapi.otherlangs">Using Languages Other Than C and C++</a></span></dt><dt><span class="sect2"><a href="#svn.developer.usingapi.codesamples">Code Samples</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.developer.summary">Summary</a></span></dt></dl></dd><dt><span class="chapter"><a href="#svn.ref">9. Subversion Complete Reference</a></span></dt><dd><dl><dt><span class="sect1"><a href="#svn.ref.svn">svn—Subversion Command-Line Client</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.ref.svn.sw">svn Options</a></span></dt><dt><span class="sect2"><a href="#svn.ref.svn.c">svn Subcommands</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.add">svn add</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.blame">svn blame (praise, annotate, ann)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.cat">svn cat</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.changelist">svn changelist (cl)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.checkout">svn checkout (co)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.cleanup">svn cleanup</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.commit">svn commit (ci)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.copy">svn copy (cp)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.delete">svn delete (del, remove, rm)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.diff">svn diff (di)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.export">svn export</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.help">svn help (h, ?)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.import">svn import</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.info">svn info</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.list">svn list (ls)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.lock">svn lock</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.log">svn log</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.merge">svn merge</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.mergeinfo">svn mergeinfo</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.mkdir">svn mkdir</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.move">svn move (mv)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.patch">svn patch</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.propdel">svn propdel (pdel, pd)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.propedit">svn propedit (pedit, pe)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.propget">svn propget (pget, pg)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.proplist">svn proplist (plist, pl)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.propset">svn propset (pset, ps)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.relocate">svn relocate</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.resolve">svn resolve</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.resolved">svn resolved</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.revert">svn revert</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.status">svn status (stat, st)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.switch">svn switch (sw)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.unlock">svn unlock</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.update">svn update (up)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.upgrade">svn upgrade</a></span><span class="refpurpose"></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.ref.svnadmin">svnadmin—Subversion Repository Administration</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.ref.svnadmin.sw">svnadmin Options</a></span></dt><dt><span class="sect2"><a href="#svn.ref.svnadmin.c">svnadmin Subcommands</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.crashtest">svnadmin crashtest</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.create">svnadmin create</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.deltify">svnadmin deltify</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.dump">svnadmin dump</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.help">svnadmin help (h, ?)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.hotcopy">svnadmin hotcopy</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.list-dblogs">svnadmin list-dblogs</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.list-unused-dblogs">svnadmin list-unused-dblogs</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.load">svnadmin load</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.lslocks">svnadmin lslocks</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.lstxns">svnadmin lstxns</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.pack">svnadmin pack</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.recover">svnadmin recover</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.rmlocks">svnadmin rmlocks</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.rmtxns">svnadmin rmtxns</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.setlog">svnadmin setlog</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.setrevprop">svnadmin setrevprop</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.setuuid">svnadmin setuuid</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.upgrade">svnadmin upgrade</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.verify">svnadmin verify</a></span><span class="refpurpose"></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.ref.svnlook">svnlook—Subversion Repository Examination</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.ref.svnlook.sw">svnlook Options</a></span></dt><dt><span class="sect2"><a href="#svn.ref.svnlook.c">svnlook Subcommands</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.author">svnlook author</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.cat">svnlook cat</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.changed">svnlook changed</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.date">svnlook date</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.diff">svnlook diff</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.dirs-changed">svnlook dirs-changed</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.filesize">svnlook filesize</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.help">svnlook help (h, ?)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.history">svnlook history</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.info">svnlook info</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.lock">svnlook lock</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.log">svnlook log</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.propget">svnlook propget (pget, pg)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.proplist">svnlook proplist (plist, pl)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.tree">svnlook tree</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.uuid">svnlook uuid</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.youngest">svnlook youngest</a></span><span class="refpurpose"></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.ref.svnsync">svnsync—Subversion Repository Mirroring</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.ref.svnsync.sw">svnsync Options</a></span></dt><dt><span class="sect2"><a href="#svn.ref.svnsync.c">svnsync Subcommands</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.svnsync.c.copy-revprops">svnsync copy-revprops</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnsync.c.help">svnsync help</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnsync.c.info">svnsync info</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnsync.c.init">svnsync initialize (init)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnsync.c.sync">svnsync synchronize (sync)</a></span><span class="refpurpose"></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.ref.svnrdump">svnrdump—Remote Subversion Repository Data Migration</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.ref.svnrdump.sw">svnrdump Options</a></span></dt><dt><span class="sect2"><a href="#svn.ref.svnrdump.c">svnrdump Subcommands</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.svnrdump.c.dump">svnrdump dump</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnrdump.c.help">svnrdump help</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnrdump.c.load">svnrdump load</a></span><span class="refpurpose"></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.ref.svnserve">svnserve—Custom Subversion Server</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.ref.svnserve.sw">svnserve Options</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.ref.svndumpfilter">svndumpfilter—Subversion History Filtering</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.ref.svndumpfilter.options">svndumpfilter Options</a></span></dt><dt><span class="sect2"><a href="#svn.ref.svndumpfilter.c">svndumpfilter Subcommands</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.svndumpfilter.commands.c.exclude">svndumpfilter exclude</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svndumpfilter.commands.c.include">svndumpfilter include</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svndumpfilter.commands.c.help">svndumpfilter help</a></span><span class="refpurpose"></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.ref.svnversion">svnversion—Subversion Working Copy Version Info</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.svnversion.re">svnversion</a></span><span class="refpurpose"></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.ref.mod_dav_svn">mod_dav_svn—Subversion Apache HTTP Server Module</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.mod_dav_svn.conf">mod_dav_svn Configuration Directives</a></span><span class="refpurpose"></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.ref.mod_authz_svn">mod_authz_svn—Subversion Apache HTTP Authorization Module</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.mod_authz_svn.conf">mod_authz_svn Configuration Directives</a></span><span class="refpurpose"></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.ref.properties">Subversion Properties</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.ref.properties.versioned-props">Versioned Properties</a></span></dt><dt><span class="sect2"><a href="#svn.ref.properties.unversioned-props">Unversioned Properties</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.ref.reposhooks">Repository Hooks</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.reposhooks.start-commit">start-commit</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.reposhooks.pre-commit">pre-commit</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.reposhooks.post-commit">post-commit</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.reposhooks.pre-revprop-change">pre-revprop-change</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.reposhooks.post-revprop-change">post-revprop-change</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.reposhooks.pre-lock">pre-lock</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.reposhooks.post-lock">post-lock</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.reposhooks.pre-unlock">pre-unlock</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.reposhooks.post-unlock">post-unlock</a></span><span class="refpurpose"></span></dt></dl></dd></dl></dd><dt><span class="appendix"><a href="#svn.intro">A. Subversion Quick-Start Guide</a></span></dt><dd><dl><dt><span class="sect1"><a href="#svn.intro.install">Installing Subversion</a></span></dt><dt><span class="sect1"><a href="#svn.intro.quickstart">High-Speed Tutorial</a></span></dt></dl></dd><dt><span class="appendix"><a href="#svn.forcvs">B. Subversion for CVS Users</a></span></dt><dd><dl><dt><span class="sect1"><a href="#svn.forcvs.revnums">Revision Numbers Are Different Now</a></span></dt><dt><span class="sect1"><a href="#svn.forcvs.directories">Directory Versions</a></span></dt><dt><span class="sect1"><a href="#svn.forcvs.disconnected">More Disconnected Operations</a></span></dt><dt><span class="sect1"><a href="#svn.forcvs.status-vs-update">Distinction Between Status and Update</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.forcvs.status-vs-update.status">Status</a></span></dt><dt><span class="sect2"><a href="#svn.forcvs.status-vs-update.update">Update</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.forcvs.branches-and-tags">Branches and Tags</a></span></dt><dt><span class="sect1"><a href="#svn.forcvs.properties">Metadata Properties</a></span></dt><dt><span class="sect1"><a href="#svn.forcvs.conflicts">Conflict Resolution</a></span></dt><dt><span class="sect1"><a href="#svn.forcvs.binary-and-trans">Binary Files and Translation</a></span></dt><dt><span class="sect1"><a href="#svn.forcvs.modules">Versioned Modules</a></span></dt><dt><span class="sect1"><a href="#svn.forcvs.auth">Authentication</a></span></dt><dt><span class="sect1"><a href="#svn.forcvs.convert">Converting a Repository from CVS to Subversion</a></span></dt></dl></dd><dt><span class="appendix"><a href="#svn.webdav">C. WebDAV and Autoversioning</a></span></dt><dd><dl><dt><span class="sect1"><a href="#svn.webdav.basic">What Is WebDAV?</a></span></dt><dt><span class="sect1"><a href="#svn.webdav.autoversioning">Autoversioning</a></span></dt><dt><span class="sect1"><a href="#svn.webdav.clients">Client Interoperability</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.webdav.clients.standalone">Standalone WebDAV Applications</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.webdav.clients.standalone.windows">Microsoft Office, Dreamweaver, Photoshop</a></span></dt><dt><span class="sect3"><a href="#svn.webdav.clients.standalone.free">cadaver, DAV Explorer</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.webdav.clients.file-explorer-extensions">File-Explorer WebDAV Extensions</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.webdav.clients.file-explorer-extensions.windows">Microsoft Web Folders</a></span></dt><dt><span class="sect3"><a href="#svn.webdav.clients.file-explorer-extensions.linux-de">Nautilus, Konqueror</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.webdav.clients.fs-impl">WebDAV Filesystem Implementation</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.webdav.clients.fs-impl.windows">WebDrive, NetDrive</a></span></dt><dt><span class="sect3"><a href="#svn.webdav.clients.fs-impl.macosx">Mac OS X</a></span></dt><dt><span class="sect3"><a href="#svn.webdav.clients.fs-impl.linux">Linux davfs2</a></span></dt></dl></dd></dl></dd></dl></dd><dt><span class="appendix"><a href="#svn.copyright">D. Copyright</a></span></dt><dt><span class="index"><a href="#svn.index">Index</a></span></dt></dl></div><div class="list-of-figures"><p><strong>List of Figures</strong></p><dl><dt>1. <a href="#svn.intro.architecture.dia-1">Subversion's architecture</a></dt><dt>1.1. <a href="#svn.basic.repository.dia-1">A typical client/server system</a></dt><dt>1.2. <a href="#svn.basic.vsn-models.problem-sharing.dia-1">The problem to avoid</a></dt><dt>1.3. <a href="#svn.basic.vsn-models.lock-unlock.dia-1">The lock-modify-unlock solution</a></dt><dt>1.4. <a href="#svn.basic.vsn-models.copy-merge.dia-1">The copy-modify-merge solution</a></dt><dt>1.5. <a href="#svn.basic.vsn-models.copy-merge.dia-2">The copy-modify-merge solution (continued)</a></dt><dt>1.6. <a href="#svn.basic.in-action.revs.dia-1">Tree changes over time</a></dt><dt>1.7. <a href="#svn.basic.in-action.wc.dia-1">The repository's filesystem</a></dt><dt>4.1. <a href="#svn.branchmerge.whatis.dia-1">Branches of development</a></dt><dt>4.2. <a href="#svn.branchmerge.using.dia-1">Starting repository layout</a></dt><dt>4.3. <a href="#svn.branchmerge.using.create.dia-1">Repository with new copy</a></dt><dt>4.4. <a href="#svn.branchmerge.using.work.dia-1">The branching of one file's history</a></dt><dt>8.1. <a href="#svn.developer.layerlib.repos.dia-1">Files and directories in two dimensions</a></dt><dt>8.2. <a href="#svn.developer.layerlib.repos.dia-2">Versioning time—the third dimension!</a></dt></dl></div><div class="list-of-tables"><p><strong>List of Tables</strong></p><dl><dt>1.1. <a href="#svn.basic.in-action.wc.tbl-1">Repository access URLs</a></dt><dt>2.1. <a href="#svn.tour.history.log.tbl-1">Common log requests</a></dt><dt>4.1. <a href="#svn.branchemerge.summary.tbl-1">Branching and merging commands</a></dt><dt>5.1. <a href="#svn.reposadmin.basics.backends.tbl-1">Repository data store comparison</a></dt><dt>6.1. <a href="#svn.serverconfig.overview.tbl-1">Comparison of subversion server options</a></dt><dt>C.1. <a href="#svn.webdav.clients.tbl-1">Common WebDAV clients</a></dt></dl></div><div class="list-of-examples"><p><strong>List of Examples</strong></p><dl><dt>4.1. <a href="#svn.branchmerge.advanced.hook-ex1">Merge-tracking gatekeeper start-commit hook script</a></dt><dt>5.1. <a href="#svn.reposadmin.maint.diskspace.deadtxns.ex-1">txn-info.sh (reporting outstanding transactions)</a></dt><dt>5.2. <a href="#svn.reposadmin.maint.replication.pre-revprop-change">Mirror repository's pre-revprop-change hook script</a></dt><dt>5.3. <a href="#svn.reposadmin.maint.replication.start-commit">Mirror repository's start-commit hook script</a></dt><dt>6.1. <a href="#svn.serverconfig.svnserve.invoking.launchd.ex-1">A sample svnserve launchd job definition</a></dt><dt>6.2. <a href="#svn.serverconfig.httpd.authz.perdir.ex-1">A sample configuration for anonymous access</a></dt><dt>6.3. <a href="#svn.serverconfig.httpd.authz.perdir.ex-2">A sample configuration for authenticated access</a></dt><dt>6.4. <a href="#svn.serverconfig.httpd.authz.perdir.ex-3">A sample configuration for mixed
authenticated/anonymous access</a></dt><dt>6.5. <a href="#svn.serverconfig.httpd.authz.pathauthzoff.ex-1">Disabling path checks altogether</a></dt><dt>7.1. <a href="#svn.advanced.confarea.windows-registry.ex-1">Sample registration entries (.reg) file</a></dt><dt>7.2. <a href="#svn.advanced.externaldifftools.diff.ex-1">diffwrap.py</a></dt><dt>7.3. <a href="#svn.advanced.externaldifftools.diff.ex-2">diffwrap.bat</a></dt><dt>7.4. <a href="#svn.advanced.externaldifftools.diff3.ex-1">diff3wrap.py</a></dt><dt>7.5. <a href="#svn.advanced.externaldifftools.diff3.ex-2">diff3wrap.bat</a></dt><dt>7.6. <a href="#svn.advanced.externaldifftools.merge.ex-1">mergewrap.py</a></dt><dt>7.7. <a href="#svn.advanced.externaldifftools.merge.ex-2">mergewrap.bat</a></dt><dt>8.1. <a href="#svn.developer.layerlib.repos.ex-1">Using the repository layer</a></dt><dt>8.2. <a href="#svn.developer.usingapi.otherlangs.ex-1">Using the repository layer with Python</a></dt><dt>8.3. <a href="#svn.developer.usingapi.otherlangs.ex-2">A Python status crawler</a></dt></dl></div><div class="preface" title="Foreword"><div class="titlepage"><div><div><h2 class="title"><a id="svn.foreword"></a>Foreword</h2></div><div><div class="author"><h3 class="author"><span class="firstname">Karl</span> <span class="surname">Fogel</span></h3></div></div><div><p class="pubdate">Chicago, March 14, 2004.</p></div></div></div><p>A bad Frequently Asked Questions (FAQ) sheet is one that is
composed not of the questions people actually ask, but of the
questions the FAQ's author <span class="emphasis"><em>wishes</em></span> people
would ask. Perhaps you've seen the type before:</p><div class="blockquote"><blockquote class="blockquote"><p>Q: How can I use Glorbosoft XYZ to maximize team
productivity?</p></blockquote></div><div class="blockquote"><blockquote class="blockquote"><p>A: Many of our customers want to know how they can
maximize productivity through our patented office groupware
innovations. The answer is simple. First, click on the
<code class="literal">File</code> menu, scroll down to
<code class="literal">Increase Productivity</code>,
then…</p></blockquote></div><p>The problem with such FAQs is that they are not, in a
literal sense, FAQs at all. No one ever called the tech support
line and asked, <span class="quote">“<span class="quote">How can we maximize
productivity?</span>”</span> Rather, people asked highly specific
questions, such as <span class="quote">“<span class="quote">How can we change the calendaring system
to send reminders two days in advance instead of one?</span>”</span>
and so on. But it's a lot easier to make up imaginary
Frequently Asked Questions than it is to discover the real ones.
Compiling a true FAQ sheet requires a sustained, organized
effort: over the lifetime of the software, incoming questions
must be tracked, responses monitored, and all gathered into a
coherent, searchable whole that reflects the collective
experience of users in the wild. It calls for the patient,
observant attitude of a field naturalist. No grand
hypothesizing, no visionary pronouncements here—open eyes
and accurate note-taking are what's needed most.</p><p>What I love about this book is that it grew out of just such
a process, and shows it on every page. It is the direct result
of the authors' encounters with users. It began with Ben
Collins-Sussman's observation that people were asking the same
basic questions over and over on the Subversion mailing lists:
what are the standard workflows to use with Subversion? Do
branches and tags work the same way as in other version control
systems? How can I find out who made a particular change?</p><p>Frustrated at seeing the same questions day after day, Ben
worked intensely over a month in the summer of 2002 to write
<em class="citetitle">The Subversion Handbook</em>, a 60-page
manual that covered all the basics of using Subversion. The
manual made no pretense of being complete, but it was
distributed with Subversion and got users over that initial hump
in the learning curve. When O'Reilly decided to
publish a full-length Subversion book, the path of least
resistance was obvious: just expand the Subversion
handbook.</p><p>The three coauthors of the new book were thus presented
with an unusual opportunity. Officially, their task was to
write a book top-down, starting from a table of contents and an
initial draft. But they also had access to a steady
stream—indeed, an uncontrollable geyser—of bottom-up
source material. Subversion was already in the hands of
thousands of early adopters, and those users were giving tons of
feedback, not only about Subversion, but also about its existing
documentation.</p><p>During the entire time they wrote this book, Ben, Mike, and
Brian haunted the Subversion mailing lists and chat rooms
incessantly, carefully noting the problems users were having in
real-life situations. Monitoring such feedback was part of their
job descriptions at CollabNet anyway, and it gave them a huge
advantage when they set out to document Subversion. The book
they produced is grounded firmly in the bedrock of experience,
not in the shifting sands of wishful thinking; it combines the
best aspects of user manual and FAQ sheet. This duality might
not be noticeable on a first reading. Taken in order, front to
back, the book is simply a straightforward description of a
piece of software. There's the overview, the obligatory guided
tour, the chapter on administrative configuration, some advanced
topics, and of course, a command reference and troubleshooting
guide. Only when you come back to it later, seeking the
solution to some specific problem, does its authenticity shine
out: the telling details that can only result from encounters
with the unexpected, the examples honed from genuine use cases,
and most of all the sensitivity to the user's needs and the
user's point of view.</p><p>Of course, no one can promise that this book will answer
every question you have about Subversion. Sometimes the
precision with which it anticipates your questions will seem
eerily telepathic; yet occasionally, you will stumble into a
hole in the community's knowledge and come away empty-handed.
When this happens, the best thing you can do is email
<code class="email"><<a class="email" href="mailto:users@subversion.apache.org">users@subversion.apache.org</a>></code> and present your
problem. The authors are still there and still watching, and the
authors include not just the three listed on the cover, but many others
who contributed corrections and original material. From the
community's point of view, solving your problem is merely a
pleasant side effect of a much larger project—namely,
slowly adjusting this book, and ultimately Subversion itself, to
more closely match the way people actually use it. They are
eager to hear from you, not only because they can help you, but
because you can help them. With Subversion, as with all active
free software projects, <span class="emphasis"><em>you are not
alone</em></span>.</p><p>Let this book be your first companion.</p></div><div class="preface" title="Preface"><div class="titlepage"><div><div><h2 class="title"><a id="svn.preface"></a>Preface</h2></div></div></div><div class="toc"><p><strong>Table of Contents</strong></p><dl><dt><span class="sect1"><a href="#svn.intro.whatis">What Is Subversion?</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.intro.righttool">Is Subversion the Right Tool?</a></span></dt><dt><span class="sect2"><a href="#svn.intro.history">Subversion's History</a></span></dt><dt><span class="sect2"><a href="#svn.intro.architecture">Subversion's Architecture</a></span></dt><dt><span class="sect2"><a href="#svn.intro.components">Subversion's Components</a></span></dt><dt><span class="sect2"><a href="#svn.intro.whatsnew">What's New in Subversion</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.preface.audience">Audience</a></span></dt><dt><span class="sect1"><a href="#svn.preface.howread">How to Read This Book</a></span></dt><dt><span class="sect1"><a href="#svn.preface.organization">Organization of This Book</a></span></dt><dt><span class="sect1"><a href="#svn.preface.free">This Book Is Free</a></span></dt><dt><span class="sect1"><a href="#svn.preface.acks">Acknowledgments</a></span></dt></dl></div><div class="blockquote"><table border="0" width="100%" cellspacing="0" cellpadding="0" class="blockquote" summary="Block quote"><tr><td width="10%" valign="top"> </td><td width="80%" valign="top"><p><span class="quote">“<span class="quote">It is important not to let the perfect become the
enemy of the good, even when you can agree on what perfect is.
Doubly so when you can't. As unpleasant as it is to be trapped by
past mistakes, you can't make any progress by being afraid of your
own shadow during design.</span>”</span></p></td><td width="10%" valign="top"> </td></tr><tr><td width="10%" valign="top"> </td><td colspan="2" align="right" valign="top">--<span class="attribution">Greg Hudson, Subversion developer</span></td></tr></table></div><a id="idp43536" class="indexterm"></a><a id="idp44592" class="indexterm"></a><p>In the world of open source software, the Concurrent Versions
System (CVS) was the tool of choice for version control for many
years. And rightly so. CVS was open source software itself, and
its nonrestrictive modus operandi and support for networked
operation allowed dozens of geographically dispersed programmers
to share their work. It fit the collaborative nature of the
open source world very well. CVS and its semi-chaotic development
model have since become cornerstones of open source
culture.</p><p>But CVS was not without its flaws, and simply fixing those
flaws promised to be an enormous effort. Enter Subversion.
Subversion was designed to be a successor to CVS, and its
originators set out to win the hearts of CVS users in two
ways—by creating an open source system with a design
(and <span class="quote">“<span class="quote">look and feel</span>”</span>) similar to CVS, and by
attempting to avoid most of CVS's noticeable flaws. While the
result wasn't—and isn't—the next great evolution in
version control design, Subversion <span class="emphasis"><em>is</em></span> very
powerful, very usable, and very flexible.</p><p>This book is written to document the 1.7 series of the Apache
Subversion™<sup>[<a id="idp5515232" href="#ftn.idp5515232" class="footnote">1</a>]</sup> version control system. We have made
every attempt to be thorough in our coverage. However, Subversion
has a thriving and energetic development community, so already a
number of features and improvements are planned for future
versions that may change some of the commands and specific notes
in this book.</p><div class="sect1" title="What Is Subversion?"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.intro.whatis"></a>What Is Subversion?</h2></div></div></div><a id="idp5517984" class="indexterm"></a><a id="idp5519664" class="indexterm"></a><a id="idp5520736" class="indexterm"></a><p>Subversion is a free/open source <em class="firstterm">version control
system</em> (VCS). That is, Subversion manages files and
directories, and the changes made to them, over time. This
allows you to recover older versions of your data or examine the
history of how your data changed. In this regard, many people
think of a version control system as a sort of <span class="quote">“<span class="quote">time
machine.</span>”</span></p><p>Subversion can operate across networks, which allows it to
be used by people on different computers. At some level, the
ability for various people to modify and manage the same set of
data from their respective locations fosters collaboration.
Progress can occur more quickly without a single conduit through
which all modifications must occur. And because the work is
versioned, you need not fear that quality is the trade-off for
losing that conduit—if some incorrect change is made to
the data, just undo that change.</p><a id="idp5524656" class="indexterm"></a><a id="idp5525872" class="indexterm"></a><p>Some version control systems are also <em class="firstterm">software
configuration management</em> (SCM) systems. These
systems are specifically tailored to manage trees of source code
and have many features that are specific to software
development—such as natively understanding programming
languages, or supplying tools for building software.
Subversion, however, is not one of these systems. It is a
general system that can be used to manage
<span class="emphasis"><em>any</em></span> collection of files. For you, those
files might be source code—for others, anything from
grocery shopping lists to digital video mixdowns and
beyond.</p><div class="sect2" title="Is Subversion the Right Tool?"><div class="titlepage"><div><div><h3 class="title"><a id="svn.intro.righttool"></a>Is Subversion the Right Tool?</h3></div></div></div><p>If you're a user or system administrator pondering the use
of Subversion, the first question you should ask yourself is:
"Is this the right tool for the job?" Subversion is a
fantastic hammer, but be careful not to view every problem as
a nail.</p><p>As a first step, you need to decide if version control in
general is required for your purposes. If you need to archive
old versions of files and directories, possibly resurrect
them, and examine logs of how they've changed over time, then
version control tools can do that. If you need to collaborate
with people on documents (usually over a network) and keep
track of who made which changes, a version control tool can do
that, too. In fact, this is why version control tools such as
Subversion are so often used in software development
environments—working on a development team is an
inherently social activity where changes to source code files
are constantly being discussed, made, evaluated, and even
sometimes unmade. Version control tools facilitate that sort
of collaboration.</p><p>There is cost associated with using version control, too.
Unless you can outsource the administration of your version
control system to a third-party, you'll have the obvious costs
of performing that administration yourself. When working with
the data on a daily basis, you won't be able to copy, move,
rename, or delete files the way you usually do. Instead,
you'll have to do all of those things through the version
control system.</p><p>Even assuming that you are okay with the cost/benefit
tradeoff afforded by a version control system, you shouldn't
choose to use one merely because it <span class="emphasis"><em>can</em></span>
do what you want. Consider whether your needs are better
addressed by other tools. For example, because Subversion
replicates data to all the collaborators involved, a common
misuse is to treat it as a generic distribution system.
People will sometimes use Subversion to distribute huge
collections of photos, digital music, or software packages.
The problem is that this sort of data usually isn't changing
at all. The collection itself grows over time, but the
individual files within the collection aren't being changed.
In this case, using Subversion is
<span class="quote">“<span class="quote">overkill.</span>”</span><sup>[<a id="idp5535168" href="#ftn.idp5535168" class="footnote">2</a>]</sup> There are simpler tools that
efficiently replicate data <span class="emphasis"><em>without</em></span> the
overhead of tracking changes, such as <span class="command"><strong>rsync</strong></span>
or <span class="command"><strong>unison</strong></span>.</p><p>Once you've decided that you need a version control
solution, you'll find no shortage of available options. When
Subversion was first designed and released, the predominant
methodology of version control was <em class="firstterm">centralized
version control</em>—a single remote master
storehouse of versioned data with individual users operating
locally against shallow copies of that data's version history.
Subversion quickly emerged after its initial introduction as
the clear leader in this field of version control, earning
widespread adoption and supplanting installations of many
older version control systems. It continues to hold that
prominent position today.</p><p>Much has changed since that time, though. In the years
since the Subversion project began its life, a newer
methodology of version control called <em class="firstterm">distributed
version control</em> has likewise garnered widespread
attention and adoption. Tools such as Git
(<a class="ulink" href="http://git-scm.com/" target="_top">http://git-scm.com/</a>) and Mercurial
(<a class="ulink" href="http://mercurial.selenic.com/" target="_top">http://mercurial.selenic.com/</a>) quickly rose
to the tops of the distributed version control system (DVCS)
ranks. Distributed version control harnesses the growing
ubiquity of high-speed network connections and low storage
costs to offer an approach which differs from the centralized
model in key ways. First and most obvious is the fact that
there is no remote, central storehouse of versioned data.
Rather, each user keeps and operates against very
deep—complete, in a sense—local version history
data stores. Collaboration still occurs, but is accomplished
by trading <em class="firstterm">changesets</em> (collections of
changes made to versioned items) directly between users' local
data stores, not via a centralized master data store. In
fact, any semblance of a canonical <span class="quote">“<span class="quote">master</span>”</span>
source of a project's versioned data is by convention only, a
status attributed by the various collaborators on that
project.</p><p>There are pros and cons to each version control approach.
Perhaps the two biggest benefits delivered by the DVCS tools
are incredible performance for day-to-day operations (because
the primary data store is locally held) and vastly better
support for merging between branches (because merge algorithms
serve as the very core of how DVCSes work at all). The
downside is that distributed version control is an inherently
more complicated model, which can present a non-negligible
challenge to comfortable collaboration. Also, DVCS tools do
what they do well in part because of a certain degree of
control withheld from the user which centalized systems freely
offer—the ability to implement path-based access
control, the flexibility to update or backdate individual
versioned data items, etc. Fortunately, many wise
organizations have discovered that this needn't be a religious
debate, and that Subversion and a DVCS tool such as Git can be
used together harmoniously within the organization, each
serving the purposes best suited to the tool.</p><p>Alas, this book is about Subversion, so we'll not attempt
a full comparison of Subversion and other tools. Readers who
have the option of choosing their version control system are
encouraged to research the available options and make the
determination that works best for themselves and their fellow
collaborators. And if, after doing so, Subversion is the
chosen tool, there's <span class="emphasis"><em>plenty</em></span> of detailed
information about how to use it successfully in the chapters
that follow!</p></div><div class="sect2" title="Subversion's History"><div class="titlepage"><div><div><h3 class="title"><a id="svn.intro.history"></a>Subversion's History</h3></div></div></div><a id="idp5547696" class="indexterm"></a><a id="idp5549088" class="indexterm"></a><p>In early 2000, CollabNet,
Inc. (<a class="ulink" href="http://www.collab.net" target="_top">http://www.collab.net</a>) began seeking
developers to write a replacement for CVS. CollabNet
offered<sup>[<a id="idp5550960" href="#ftn.idp5550960" class="footnote">3</a>]</sup> a collaboration software suite
called CollabNet Enterprise Edition (CEE), of which one
component was version control. Although CEE used CVS as its
initial version control system, CVS's limitations were obvious
from the beginning, and CollabNet knew it would eventually
have to find something better. Unfortunately, CVS had become
the de facto standard in the open source world largely because
there <span class="emphasis"><em>wasn't</em></span> anything better, at least
not under a free license. So CollabNet determined to write a
new version control system from scratch, retaining the basic
ideas of CVS, but without the bugs and misfeatures.</p><p>In February 2000, they contacted Karl Fogel, the author of
<em class="citetitle">Open Source Development with CVS</em>
(Coriolis, 1999), and asked if he'd like to work on this new
project. Coincidentally, at the time Karl was already
discussing a design for a new version control system with his
friend Jim Blandy. In 1995, the two had started Cyclic
Software, a company providing CVS support contracts, and
although they later sold the business, they still used CVS
every day at their jobs. Their frustration with CVS had led
Jim to think carefully about better ways to manage versioned
data, and he'd already come up with not only the Subversion
name, but also the basic design of the Subversion data store.
When CollabNet called, Karl immediately agreed to work on the
project, and Jim got his employer, Red Hat Software, to
essentially donate him to the project for an indefinite period
of time. CollabNet hired Karl and Ben Collins-Sussman, and
detailed design work began in May 2000. With the help of some
well-placed prods from Brian Behlendorf and Jason Robbins of
CollabNet, and from Greg Stein (at the time an independent
developer active in the WebDAV/DeltaV specification process),
Subversion quickly attracted a community of active developers.
It turned out that many people had encountered the same
frustrating experiences with CVS and welcomed the chance to
finally do something about it.</p><p>The original design team settled on some simple goals. They
didn't want to break new ground in version control methodology,
they just wanted to fix CVS. They decided that Subversion would
match CVS's features and preserve the same development model,
but not duplicate CVS's most obvious flaws. And although it did
not need to be a drop-in replacement for CVS, it should be
similar enough that any CVS user could make the switch with
little effort.</p><p>After 14 months of coding, Subversion became
<span class="quote">“<span class="quote">self-hosting</span>”</span> on August 31, 2001. That is,
Subversion developers stopped using CVS to manage Subversion's
own source code and started using Subversion instead.</p><p>While CollabNet started the project, and still funds a
large chunk of the work (it pays the salaries of a few
full-time Subversion developers), Subversion is run like most
open source projects, governed by a loose, transparent set of
rules that encourage meritocracy. In 2009, CollabNet worked
with the Subversion developers towards the goal of integrating
the Subversion project into the Apache Software Foundation
(ASF), one of the most well-known collectives of open source
projects in the world. Subversion's technical roots,
community priorities, and development practices were a perfect
fit for the ASF, many of whose members were already active
Subversion contributors. In early 2010, Subversion was fully
adopted into the ASF's family of top-level projects, moved its
project web presence to
<a class="ulink" href="http://subversion.apache.org" target="_top">http://subversion.apache.org</a>, and was
rechristened <span class="quote">“<span class="quote">Apache Subversion</span>”</span>.</p></div><div class="sect2" title="Subversion's Architecture"><div class="titlepage"><div><div><h3 class="title"><a id="svn.intro.architecture"></a>Subversion's Architecture</h3></div></div></div><a id="idp5560560" class="indexterm"></a><p><a class="xref" href="#svn.intro.architecture.dia-1" title="Figure 1. Subversion's architecture">Figure 1, “Subversion's architecture”</a> illustrates
a <span class="quote">“<span class="quote">mile-high</span>”</span> view of Subversion's
design.</p><div class="figure"><a id="svn.intro.architecture.dia-1"></a><p class="title"><strong>Figure 1. Subversion's architecture</strong></p><div class="figure-contents"><div><img src="images/ch01dia1.png" alt="Subversion's architecture"></img></div></div></div><br class="figure-break"></br><p>On one end is a Subversion repository that holds all of
your versioned data. On the other end is your Subversion
client program, which manages local reflections of portions of
that versioned data. Between these extremes are multiple
routes through a Repository Access (RA) layer, some of which
go across computer networks and through network servers which
then access the repository, others of which bypass the network
altogether and access the repository directly.</p></div><div class="sect2" title="Subversion's Components"><div class="titlepage"><div><div><h3 class="title"><a id="svn.intro.components"></a>Subversion's Components</h3></div></div></div><a id="idp5579408" class="indexterm"></a><p>Subversion, once installed, has a number of different
pieces. The following is a quick overview of what you get.
Don't be alarmed if the brief descriptions leave you
scratching your head—<span class="emphasis"><em>plenty</em></span> more
pages in this book are devoted to alleviating that
confusion.</p><div class="variablelist"><a id="idp5582048" class="indexterm"></a><a id="idp5583056" class="indexterm"></a><a id="idp5584064" class="indexterm"></a><a id="idp5585072" class="indexterm"></a><a id="idp5586080" class="indexterm"></a><a id="idp5587088" class="indexterm"></a><a id="idp5588096" class="indexterm"></a><a id="idp5589104" class="indexterm"></a><dl><dt><span class="term">svn</span></dt><dd><p>The command-line client program</p></dd><dt><span class="term">svnversion</span></dt><dd><p>A program for reporting the state (in terms of
revisions of the items present) of a working copy</p></dd><dt><span class="term">svnlook</span></dt><dd><p>A tool for directly inspecting a Subversion repository</p></dd><dt><span class="term">svnadmin</span></dt><dd><p>A tool for creating, tweaking, or repairing a Subversion
repository</p></dd><dt><span class="term">mod_dav_svn</span></dt><dd><p>A plug-in module for the Apache HTTP Server, used to
make your repository available to others over a
network</p></dd><dt><span class="term">svnserve</span></dt><dd><p>A custom standalone server program, runnable as a
daemon process or invokable by SSH; another way to make
your repository available to others over a network</p></dd><dt><span class="term">svndumpfilter</span></dt><dd><p>A program for filtering Subversion repository dump
streams</p></dd><dt><span class="term">svnsync</span></dt><dd><p>A program for incrementally mirroring one
repository to another over a network</p></dd><dt><span class="term">svnrdump</span></dt><dd><p>A program for performing repository history dumps
and loads over a network</p></dd></dl></div></div><div class="sect2" title="What's New in Subversion"><div class="titlepage"><div><div><h3 class="title"><a id="svn.intro.whatsnew"></a>What's New in Subversion</h3></div></div></div><a id="idp5606400" class="indexterm"></a><p>The first edition of this book was published by O'Reilly
Media in 2004, shortly after Subversion had reached 1.0.
Since that time, the Subversion project has continued to
release new major releases of the software. Here's a quick
summary of major new changes since Subversion 1.0. Note that
this is not a complete list; for full details, please visit
Subversion's web site at
<a class="ulink" href="http://subversion.apache.org" target="_top">http://subversion.apache.org</a>.</p><div class="variablelist"><dl><dt><span class="term">Subversion 1.1 (September 2004)</span></dt><dd><p>Release 1.1 introduced FSFS, a flat-file repository
storage option for the repository. While the Berkeley DB
backend is still widely used and supported, FSFS has
since become the default choice for
newly created repositories due to its low barrier to
entry and minimal maintenance requirements. Also in
this release came the ability to put symbolic links
under version control, auto-escaping of URLs, and a
localized user interface.</p></dd><dt><span class="term">Subversion 1.2 (May 2005)</span></dt><dd><p>Release 1.2 introduced the ability to create
server-side locks on files, thus serializing commit
access to certain resources. While Subversion is still
a fundamentally concurrent version control system,
certain types of binary files (e.g. art assets)
cannot be merged together. The locking feature fulfills
the need to version and protect such resources. With
locking also came a complete WebDAV auto-versioning
implementation, allowing Subversion repositories to be
mounted as network folders. Finally, Subversion 1.2
began using a new, faster binary-differencing algorithm
to compress and retrieve old versions of files.</p></dd><dt><span class="term">Subversion 1.3 (December 2005)</span></dt><dd><p>Release 1.3 brought path-based authorization
controls to the <span class="command"><strong>svnserve</strong></span> server,
matching a feature formerly found only in the Apache
server. The Apache server, however, gained some new
logging features of its own, and Subversion's API
bindings to other languages also made great leaps
forward.</p></dd><dt><span class="term">Subversion 1.4 (September 2006)</span></dt><dd><p>Release 1.4 introduced a whole new
tool—<span class="command"><strong>svnsync</strong></span>—for doing
one-way repository replication over a network. Major
parts of the working copy metadata were revamped to no
longer use XML (resulting in client-side speed gains),
while the Berkeley DB repository backend gained the
ability to automatically recover itself after a server
crash.</p></dd><dt><span class="term">Subversion 1.5 (June 2008)</span></dt><dd><p>Release 1.5 took much longer to finish than prior
releases, but the headliner feature was gigantic:
semi-automated tracking of branching and merging. This
was a huge boon for users, and pushed Subversion far
beyond the abilities of CVS and into the ranks of
commercial competitors such as Perforce and ClearCase.
Subversion 1.5 also introduced a bevy of other
user-focused features, such as interactive resolution of
file conflicts, sparse checkouts, client-side
management of changelists, powerful new syntax for
externals definitions, and SASL authentication support for
the <span class="command"><strong>svnserve</strong></span> server.</p></dd><dt><span class="term">Subversion 1.6 (March 2009)</span></dt><dd><p>Release 1.6 continued to make branching and merging
more robust by introducing tree conflicts, and offered
improvements to several other existing features: more
interactive conflict resolution options; de-telescoping
and outright exclusion support for sparse checkouts;
file-based externals definitions; and operational
logging support for <span class="command"><strong>svnserve</strong></span> similar
to what <span class="command"><strong>mod_dav_svn</strong></span> offered. Also,
the command-line client introduced a new shortcut syntax
for referring to Subversion repository URLs.</p></dd><dt><span class="term">Subversion 1.7 (October 2011)</span></dt><dd><p>Release 1.7 was primarily a delivery vehicle for two
big plumbing overhauls of existing Subversion
components. The largest and most impactful of these was
the so-called <span class="quote">“<span class="quote">WC-NG</span>”</span>—a complete
rewrite of the <span class="command"><strong>libsvn_wc</strong></span> working copy
management library. The second change was the
introduction of a sleeker HTTP protocol for Subversion
client/server interaction. Subversion 1.7 delivered a
handful of additional features, many bug fixes, and some
notable performance improvements, too.</p></dd></dl></div></div></div><div class="sect1" title="Audience"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.preface.audience"></a>Audience</h2></div></div></div><p>This book is written for computer-literate folk who want to
use Subversion to manage their data. While Subversion runs on a
number of different operating systems, its primary user
interface is command-line-based. That command-line tool
(<span class="command"><strong>svn</strong></span>), and some additional auxiliary
programs, are the focus of this book.</p><p>For consistency, the examples in this book assume that the reader
is using a Unix-like operating system and is relatively comfortable
with Unix and command-line interfaces. That said, the
<span class="command"><strong>svn</strong></span> program also runs on non-Unix platforms
such as Microsoft Windows. With a few minor exceptions, such as
the use of backward slashes (<code class="literal">\</code>) instead of
forward slashes (<code class="literal">/</code>) for path separators, the
input to and output from this tool when run on Windows are
identical to that of its Unix counterpart.</p><p>Most readers are probably programmers or system
administrators who need to track changes to source code. This
is the most common use for Subversion, and therefore it is the
scenario underlying all of the book's examples. But Subversion
can be used to manage changes to any sort of
information—images, music, databases, documentation, and
so on. To Subversion, all data is just data.</p><p>While this book is written with the assumption that the
reader has never used a version control system, we've also tried
to make it easy for users of CVS (and other systems) to make a
painless leap into Subversion. Special sidebars may mention
other version control systems from time to time, and
<a class="xref" href="#svn.forcvs" title="Appendix B. Subversion for CVS Users">Appendix B, <em>Subversion for CVS Users</em></a> summarizes many of the differences
between CVS and Subversion.</p><p>Note also that the source code examples used throughout the
book are only examples. While they will compile with the proper
compiler incantations, they are intended to illustrate a
particular scenario and not necessarily to serve as examples of good
programming style or practices.</p></div><div class="sect1" title="How to Read This Book"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.preface.howread"></a>How to Read This Book</h2></div></div></div><p>Technical books always face a certain dilemma: whether to
cater to <em class="firstterm">top-down</em>
or to <em class="firstterm">bottom-up</em> learners. A top-down
learner prefers to read or skim documentation, getting a large
overview of how the system works; only then does she actually
start using the software. A bottom-up learner is a <span class="quote">“<span class="quote">learn by
doing</span>”</span> person—someone who just wants to dive into the
software and figure it out as she goes, referring to book
sections when necessary. Most books tend to be written for one
type of person or the other, and this book is undoubtedly biased
toward top-down learners. (And if you're actually reading this
section, you're probably already a top-down learner yourself!)
However, if you're a bottom-up person, don't despair. While the
book may be laid out as a broad survey of Subversion topics, the
content of each section tends to be heavy with specific
examples that you can try-by-doing. For the impatient folks who
just want to get going, you can jump right to
<a class="xref" href="#svn.intro" title="Appendix A. Subversion Quick-Start Guide">Appendix A, <em>Subversion Quick-Start Guide</em></a>.</p><p>Regardless of your learning style, this book aims to be
useful to people of widely different backgrounds—from
those with no previous experience in version control to
experienced system administrators. Depending on your own
background, certain chapters may be more or less important to
you. The following can be considered a
<span class="quote">“<span class="quote">recommended reading list</span>”</span> for various types of
readers:</p><div class="variablelist"><dl><dt><span class="term">Experienced system administrators</span></dt><dd><p>The assumption here is that you've probably used
version control before and are dying to get a
Subversion server up and running ASAP.
<a class="xref" href="#svn.reposadmin" title="Chapter 5. Repository Administration">Chapter 5, <em>Repository Administration</em></a> and
<a class="xref" href="#svn.serverconfig" title="Chapter 6. Server Configuration">Chapter 6, <em>Server Configuration</em></a> will show you how to
create your first repository and make it available over
the network. After that's done,
<a class="xref" href="#svn.tour" title="Chapter 2. Basic Usage">Chapter 2, <em>Basic Usage</em></a> and
<a class="xref" href="#svn.forcvs" title="Appendix B. Subversion for CVS Users">Appendix B, <em>Subversion for CVS Users</em></a> are the fastest routes to
learning the Subversion client.</p></dd><dt><span class="term">New users</span></dt><dd><p>Your administrator has probably set up Subversion
already, and you need to learn how to use the client. If
you've never used a version control system, then
<a class="xref" href="#svn.basic" title="Chapter 1. Fundamental Concepts">Chapter 1, <em>Fundamental Concepts</em></a> is a vital introduction to the
ideas behind version control. <a class="xref" href="#svn.tour" title="Chapter 2. Basic Usage">Chapter 2, <em>Basic Usage</em></a>
is a guided tour of the Subversion client.</p></dd><dt><span class="term">Advanced users</span></dt><dd><p>Whether you're a user or administrator, eventually
your project will grow larger. You're going to want to
learn how to do more advanced things with Subversion, such
as how to use Subversion's property support
(<a class="xref" href="#svn.advanced" title="Chapter 3. Advanced Topics">Chapter 3, <em>Advanced Topics</em></a>), how to use branches and
perform merges (<a class="xref" href="#svn.branchmerge" title="Chapter 4. Branching and Merging">Chapter 4, <em>Branching and Merging</em></a>), how to
configure runtime options
(<a class="xref" href="#svn.customization" title="Chapter 7. Customizing Your Subversion Experience">Chapter 7, <em>Customizing Your Subversion Experience</em></a>), and other things.
These chapters aren't critical at first, but be sure to
read them once you're comfortable with the basics.</p></dd><dt><span class="term">Developers</span></dt><dd><p>Presumably, you're already familiar with Subversion,
and now want to either extend it or build new software on
top of its many APIs. <a class="xref" href="#svn.developer" title="Chapter 8. Embedding Subversion">Chapter 8, <em>Embedding Subversion</em></a> is
just for you.</p></dd></dl></div><p>The book ends with reference material—<a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a> is a reference guide for all Subversion
commands, and the appendixes cover a number of useful topics.
These are the chapters you're mostly likely to come back to
after you've finished the book.</p></div><div class="sect1" title="Organization of This Book"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.preface.organization"></a>Organization of This Book</h2></div></div></div><p>The chapters that follow and their contents are listed
here:</p><div class="variablelist"><dl><dt><span class="term"><a class="xref" href="#svn.basic" title="Chapter 1. Fundamental Concepts">Chapter 1, <em>Fundamental Concepts</em></a></span></dt><dd><p>Explains the basics of version control and different
versioning models, along with Subversion's repository,
working copies, and revisions.</p></dd><dt><span class="term"><a class="xref" href="#svn.tour" title="Chapter 2. Basic Usage">Chapter 2, <em>Basic Usage</em></a></span></dt><dd><p>Walks you through a day in the life of a Subversion
user. It demonstrates how to use a Subversion client to
obtain, modify, and commit data.</p></dd><dt><span class="term"><a class="xref" href="#svn.advanced" title="Chapter 3. Advanced Topics">Chapter 3, <em>Advanced Topics</em></a></span></dt><dd><p>Covers more complex features that regular users will
eventually come into contact with, such as versioned
metadata, file locking, and peg revisions.</p></dd><dt><span class="term"><a class="xref" href="#svn.branchmerge" title="Chapter 4. Branching and Merging">Chapter 4, <em>Branching and Merging</em></a></span></dt><dd><p>Discusses branches, merges, and tagging, including
best practices for branching and merging, common
use cases, how to undo changes, and how to easily swing
from one branch to the next.</p></dd><dt><span class="term"><a class="xref" href="#svn.reposadmin" title="Chapter 5. Repository Administration">Chapter 5, <em>Repository Administration</em></a></span></dt><dd><p>Describes the basics of the Subversion repository,
how to create, configure, and maintain a repository, and
the tools you can use to do all of this.</p></dd><dt><span class="term"><a class="xref" href="#svn.serverconfig" title="Chapter 6. Server Configuration">Chapter 6, <em>Server Configuration</em></a></span></dt><dd><p>Explains how to configure your Subversion server and
offers different ways to access your repository:
<code class="literal">HTTP</code>, the <code class="literal">svn</code>
protocol, and local disk access. It also covers the details
of authentication, authorization and anonymous
access.</p></dd><dt><span class="term"><a class="xref" href="#svn.customization" title="Chapter 7. Customizing Your Subversion Experience">Chapter 7, <em>Customizing Your Subversion Experience</em></a></span></dt><dd><p>Explores the Subversion client configuration files,
the handling of internationalized text, and how to make
external tools cooperate with Subversion.</p></dd><dt><span class="term"><a class="xref" href="#svn.developer" title="Chapter 8. Embedding Subversion">Chapter 8, <em>Embedding Subversion</em></a></span></dt><dd><p>Describes the internals of Subversion, the
Subversion filesystem, and the working copy
administrative areas from a programmer's point of view.
It also demonstrates how to use the public APIs to write a
program that uses Subversion.</p></dd><dt><span class="term"><a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a></span></dt><dd><p>Explains in great detail every subcommand of
<span class="command"><strong>svn</strong></span>, <span class="command"><strong>svnadmin</strong></span>, and
<span class="command"><strong>svnlook</strong></span> with plenty of examples for
the whole family!</p></dd><dt><span class="term"><a class="xref" href="#svn.intro" title="Appendix A. Subversion Quick-Start Guide">Appendix A, <em>Subversion Quick-Start Guide</em></a></span></dt><dd><p>For the impatient, a whirlwind explanation of how to
install Subversion and start using it immediately. You
have been warned.</p></dd><dt><span class="term"><a class="xref" href="#svn.forcvs" title="Appendix B. Subversion for CVS Users">Appendix B, <em>Subversion for CVS Users</em></a></span></dt><dd><p>Covers the similarities and differences between
Subversion and CVS, with numerous suggestions on how to
break all the bad habits you picked up from years of
using CVS. Included are descriptions of Subversion
revision numbers, versioned directories, offline
operations, <span class="command"><strong>update</strong></span>
versus <span class="command"><strong>status</strong></span>, branches, tags, metadata,
conflict resolution, and authentication.</p></dd><dt><span class="term"><a class="xref" href="#svn.webdav" title="Appendix C. WebDAV and Autoversioning">Appendix C, <em>WebDAV and Autoversioning</em></a></span></dt><dd><p>Describes the details of WebDAV and DeltaV and how
you can configure your Subversion repository to be
mounted read/write as a DAV share.</p></dd><dt><span class="term"><a class="xref" href="#svn.copyright" title="Appendix D. Copyright">Appendix D, <em>Copyright</em></a></span></dt><dd><p>A copy of the Creative Commons Attribution License,
under which this book is licensed.</p></dd></dl></div></div><div class="sect1" title="This Book Is Free"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.preface.free"></a>This Book Is Free</h2></div></div></div><p>This book started out as bits of documentation written by
Subversion project developers, which were then coalesced into a
single work and rewritten. As such, it has always been under a
free license (see <a class="xref" href="#svn.copyright" title="Appendix D. Copyright">Appendix D, <em>Copyright</em></a>). In fact,
the book was written in the public eye, originally as part of
the Subversion project itself. This means two things:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>You will always find the latest version of this book in
the book's own Subversion repository.</p></li><li class="listitem"><p>You can make changes to this book and redistribute it
however you wish—it's under a free license. Your only
obligation is to maintain proper attribution to the original
authors. Of course, we'd much rather you send feedback and
patches to the Subversion developer community, instead of
distributing your private version of this book.</p></li></ul></div><p>The online home of this book's development and most of the
volunteer-driven translation efforts regarding it is
<a class="ulink" href="http://svnbook.red-bean.com" target="_top">http://svnbook.red-bean.com</a>. There you can find
links to the latest releases and tagged versions of the book in
various formats, as well as instructions for accessing the
book's Subversion repository (where its DocBook XML source
code lives). Feedback is welcomed—encouraged, even. Please
submit all comments, complaints, and patches against the book
sources to <code class="email"><<a class="email" href="mailto:svnbook-dev@red-bean.com">svnbook-dev@red-bean.com</a>></code>.</p></div><div class="sect1" title="Acknowledgments"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.preface.acks"></a>Acknowledgments</h2></div></div></div><p>This book would not be possible (nor very useful) if
Subversion did not exist. For that, the authors would like to
thank Brian Behlendorf and CollabNet for the vision to fund such
a risky and ambitious new open source project; Jim Blandy for
the original Subversion name and design—we love you, Jim;
and Karl Fogel for being such a good friend and a great
community leader, in that order.<sup>[<a id="idp5699648" href="#ftn.idp5699648" class="footnote">4</a>]</sup></p><p>Thanks to O'Reilly and the team of professional editors who
have helped us polish this text at various stages of its
evolution: Chuck Toporek, Linda Mui, Tatiana Apandi, Mary Brady,
and Mary Treseler. Your patience and support has been
tremendous.</p><p>Finally, we thank the countless people who contributed to
this book with informal reviews, suggestions, and patches. An
exhaustive listing of those folks' names would be impractical to
print and maintain here, but may their names live on forever in
this book's version control history!</p></div><div class="footnotes"><br></br><hr width="100" align="left"></hr><div class="footnote"><p><sup>[<a id="ftn.idp5515232" href="#idp5515232" class="para">1</a>] </sup>We'll refer to it simply
as <span class="quote">“<span class="quote">Subversion</span>”</span> throughout this book. You'll thank
us when you realize just how much space that
saves!</p></div><div class="footnote"><p><sup>[<a id="ftn.idp5535168" href="#idp5535168" class="para">2</a>] </sup>Or as a friend puts
it, <span class="quote">“<span class="quote">swatting a fly with a
Buick.</span>”</span></p></div><div class="footnote"><p><sup>[<a id="ftn.idp5550960" href="#idp5550960" class="para">3</a>] </sup>CollabNet Enterprise Edition has since
been replaced by a new product line called CollabNet
TeamForge.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp5699648" href="#idp5699648" class="para">4</a>] </sup>Oh, and thanks,
Karl, for being too overworked to write this book
yourself.</p></div></div></div><div class="chapter" title="Chapter 1. Fundamental Concepts"><div class="titlepage"><div><div><h2 class="title"><a id="svn.basic"></a>Chapter 1. Fundamental Concepts</h2></div></div></div><div class="toc"><p><strong>Table of Contents</strong></p><dl><dt><span class="sect1"><a href="#svn.basic.version-control-basics">Version Control Basics</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.basic.repository">The Repository</a></span></dt><dt><span class="sect2"><a href="#svn.basic.working-copy">The Working Copy</a></span></dt><dt><span class="sect2"><a href="#svn.basic.vsn-models">Versioning Models</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.basic.vsn-models.problem-sharing">The problem of file sharing</a></span></dt><dt><span class="sect3"><a href="#svn.basic.vsn-models.lock-unlock">The lock-modify-unlock solution</a></span></dt><dt><span class="sect3"><a href="#svn.basic.vsn-models.copy-merge">The copy-modify-merge solution</a></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.basic.in-action">Version Control the Subversion Way</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.basic.svn-repositories">Subversion Repositories</a></span></dt><dt><span class="sect2"><a href="#svn.basic.in-action.revs">Revisions</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.reposurls">Addressing the Repository</a></span></dt><dt><span class="sect2"><a href="#svn.basic.in-action.wc">Subversion Working Copies</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.basic.in-action.track-repos">How the working copy works</a></span></dt><dt><span class="sect3"><a href="#svn.basic.in-action.wc-funcdamentals">Fundamental working copy interactions</a></span></dt><dt><span class="sect3"><a href="#svn.basic.in-action.mixedrevs">Mixed-revision working copies</a></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.basic.summary">Summary</a></span></dt></dl></div><p>This chapter is a short, casual introduction to Subversion and
its approach to version control. We begin with a discussion of
general version control concepts, work our way into the specific
ideas behind Subversion, and show some simple examples of
Subversion in use.</p><p>Even though the examples in this chapter show people sharing
collections of program source code, keep in mind that Subversion
can manage any sort of file collection—it's not limited to
helping computer programmers.</p><div class="sect1" title="Version Control Basics"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.basic.version-control-basics"></a>Version Control Basics</h2></div></div></div><a id="idp5855344" class="indexterm"></a><p>A version control system (or revision control system) is a
system that tracks incremental versions (or revisions) of files
and, in some cases, directories over time. Of course, merely
tracking the various versions of a user's (or group of users')
files and directories isn't very interesting in itself. What
makes a version control system useful is the fact that it allows
you to explore the changes which resulted in each of those
versions and facilitates the arbitrary recall of the
same.</p><p>In this section, we'll introduce some fairly high-level
version control system components and concepts. We'll limit our
discussion to modern version control systems—in today's
interconnected world, there is very little point in
acknowledging version control systems which cannot
operate across wide-area networks.</p><div class="sect2" title="The Repository"><div class="titlepage"><div><div><h3 class="title"><a id="svn.basic.repository"></a>The Repository</h3></div></div></div><a id="idp5902128" class="indexterm"></a><p>At the core of the version control system is a repository,
which is the central store of that system's data. The
repository usually stores information in the form of a
<em class="firstterm">filesystem tree</em>—a hierarchy of
files and directories. Any number of
<em class="firstterm">clients</em> connect to the repository, and
then read or write to these files. By writing data, a client
makes the information available to others; by reading data,
the client receives information from others.
<a class="xref" href="#svn.basic.repository.dia-1" title="Figure 1.1. A typical client/server system">Figure 1.1, “A typical client/server system”</a> illustrates
this.</p><div class="figure"><a id="svn.basic.repository.dia-1"></a><p class="title"><strong>Figure 1.1. A typical client/server system</strong></p><div class="figure-contents"><div><img src="images/ch02dia1.png" alt="A typical client/server system"></img></div></div></div><br class="figure-break"></br><p>Why is this interesting? So far, this sounds like the
definition of a typical file server. And indeed, the
repository <span class="emphasis"><em>is</em></span> a kind of file server, but
it's not your usual breed. What makes the repository special
is that as the files in the repository are changed, the
repository remembers each version of those files.</p><p>When a client reads data from the repository, it normally
sees only the latest version of the filesystem tree. But what
makes a version control client interesting is that it also has
the ability to request previous states of the filesystem from
the repository. A version control client can ask historical
questions such as <span class="quote">“<span class="quote">What did this directory contain last
Wednesday?</span>”</span> and <span class="quote">“<span class="quote">Who was the last person to
change this file, and what changes did he make?</span>”</span>
These are the sorts of questions that are at the heart of any
version control system.</p></div><div class="sect2" title="The Working Copy"><div class="titlepage"><div><div><h3 class="title"><a id="svn.basic.working-copy"></a>The Working Copy</h3></div></div></div><a id="idp5911712" class="indexterm"></a><p>A version control system's value comes from the fact that it
tracks versions of files and directories, but the rest of the
software universe doesn't operate on <span class="quote">“<span class="quote">versions of files
and directories</span>”</span>. Most software programs understand
how to operate only on a <span class="emphasis"><em>single</em></span> version
of a specific type of file. So how does a version control
user interact with an abstract—and, often,
remote—repository full of multiple versions of various
files in a concrete fashion? How does his or her word
processing software, presentation software, source code
editor, web design software, or some other program—all
of which trade in the currency of simple data files—get
access to such files? The answer is found in the version
control construct known as a <em class="firstterm">working
copy</em>.</p><p>A working copy is, quite literally, a local copy of a
particular version of a user's VCS-managed data upon which
that user is free to work. Working copies<sup>[<a id="idp5915856" href="#ftn.idp5915856" class="footnote">5</a>]</sup> appear to other software
just as any other local directory full of files, so those
programs don't have to be <span class="quote">“<span class="quote">version-control-aware</span>”</span>
in order to read from and write to that data. The task of
managing the working copy and communicating changes made to
its contents to and from the repository falls squarely to the
version control system's client software.</p></div><div class="sect2" title="Versioning Models"><div class="titlepage"><div><div><h3 class="title"><a id="svn.basic.vsn-models"></a>Versioning Models</h3></div></div></div><p>If the primary mission of a version control system is to
track the various versions of digital information over time, a
very close secondary mission in any modern version control
system is to enable collaborative editing and sharing of that
data. But different systems use different strategies to
achieve this. It's important to understand these different
strategies, for a couple of reasons. First, it will help you
compare and contrast existing version control systems, in case
you encounter other systems similar to Subversion. Beyond
that, it will also help you make more effective use of
Subversion, since Subversion itself supports a couple of
different ways of working.</p><div class="sect3" title="The problem of file sharing"><div class="titlepage"><div><div><h4 class="title"><a id="svn.basic.vsn-models.problem-sharing"></a>The problem of file sharing</h4></div></div></div><p>All version control systems have to solve the same
fundamental problem: how will the system allow users to
share information, but prevent them from accidentally
stepping on each other's feet? It's all too easy for users
to accidentally overwrite each other's changes in the
repository.</p><p>Consider the scenario shown in
<a class="xref" href="#svn.basic.vsn-models.problem-sharing.dia-1" title="Figure 1.2. The problem to avoid">Figure 1.2, “The problem to avoid”</a>.
Suppose we have two coworkers, Harry and Sally. They each
decide to edit the same repository file at the same time.
If Harry saves his changes to the repository first, it's
possible that (a few moments later) Sally could accidentally
overwrite them with her own new version of the file. While
Harry's version of the file won't be lost forever (because
the system remembers every change), any changes Harry made
<span class="emphasis"><em>won't</em></span> be present in Sally's newer version
of the file, because she never saw Harry's changes to begin
with. Harry's work is still effectively lost—or at
least missing from the latest version of the file—and
probably by accident. This is definitely a situation we want
to avoid!</p><div class="figure"><a id="svn.basic.vsn-models.problem-sharing.dia-1"></a><p class="title"><strong>Figure 1.2. The problem to avoid</strong></p><div class="figure-contents"><div><img src="images/ch02dia2.png" alt="The problem to avoid"></img></div></div></div><br class="figure-break"></br></div><div class="sect3" title="The lock-modify-unlock solution"><div class="titlepage"><div><div><h4 class="title"><a id="svn.basic.vsn-models.lock-unlock"></a>The lock-modify-unlock solution</h4></div></div></div><a id="idp5862624" class="indexterm"></a><p>Many version control systems use a
<em class="firstterm">lock-modify-unlock</em> model to address
the problem of many authors clobbering each other's work.
In this model, the repository allows only one person to
change a file at a time. This exclusivity policy is managed
using locks. Harry must <span class="quote">“<span class="quote">lock</span>”</span> a file before
he can begin making changes to it. If Harry has locked a
file, Sally cannot also lock it, and therefore cannot make
any changes to that file. All she can do is read the file
and wait for Harry to finish his changes and release his
lock. After Harry unlocks the file, Sally can take her turn
by locking and editing the file.
<a class="xref" href="#svn.basic.vsn-models.lock-unlock.dia-1" title="Figure 1.3. The lock-modify-unlock solution">Figure 1.3, “The lock-modify-unlock solution”</a>
demonstrates this simple solution.</p><div class="figure"><a id="svn.basic.vsn-models.lock-unlock.dia-1"></a><p class="title"><strong>Figure 1.3. The lock-modify-unlock solution</strong></p><div class="figure-contents"><div><img src="images/ch02dia3.png" alt="The lock-modify-unlock solution"></img></div></div></div><br class="figure-break"></br><p>The problem with the lock-modify-unlock model is that it's
a bit restrictive and often becomes a roadblock for
users:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p><span class="emphasis"><em>Locking may cause administrative
problems.</em></span>
Sometimes Harry will lock a file and then forget about it.
Meanwhile, because Sally is still waiting to edit the file,
her hands are tied. And then Harry goes on vacation. Now
Sally has to get an administrator to release Harry's lock.
The situation ends up causing a lot of unnecessary delay
and wasted time.</p></li><li class="listitem"><p><span class="emphasis"><em>Locking may cause unnecessary
serialization.</em></span>
What if Harry is editing the beginning of a text file,
and Sally simply wants to edit the end of the same file?
These changes don't overlap at all. They could easily
edit the file simultaneously, and no great harm would
come, assuming the changes were properly merged together.
There's no need for them to take turns in this
situation.</p></li><li class="listitem"><p><span class="emphasis"><em>Locking may create a false sense of
security.</em></span>
Suppose Harry locks and edits file A, while Sally
simultaneously locks and edits file B. But what if A and
B depend on one another, and the changes made to each are
semantically incompatible? Suddenly A and B don't work
together anymore. The locking system was powerless to
prevent the problem—yet it somehow provided a false
sense of security. It's easy for Harry and Sally to
imagine that by locking files, each is beginning a safe,
insulated task, and thus they need not bother discussing
their incompatible changes early on. Locking often
becomes a substitute for real communication.</p></li></ul></div></div><div class="sect3" title="The copy-modify-merge solution"><div class="titlepage"><div><div><h4 class="title"><a id="svn.basic.vsn-models.copy-merge"></a>The copy-modify-merge solution</h4></div></div></div><a id="idp5949552" class="indexterm"></a><p>Subversion, CVS, and many other version control systems
use a <em class="firstterm">copy-modify-merge</em> model as an
alternative to locking. In this model, each user's client
contacts the project repository and creates a personal
working copy. Users then work simultaneously and
independently, modifying their private copies. Finally, the
private copies are merged together into a new, final
version. The version control system often assists with the
merging, but ultimately, a human being is responsible for
making it happen correctly.</p><p>Here's an example. Say that Harry and Sally each create
working copies of the same project, copied from the
repository. They work concurrently and make changes to the
same file A within their copies. Sally saves her changes to
the repository first. When Harry attempts to save his changes
later, the repository informs him that his file A is
<em class="firstterm">out of date</em>. In other words, file A
in the repository has somehow changed since he last copied
it. So Harry asks his client
to <em class="firstterm">merge</em> any new changes from the
repository into his working copy of file A. Chances are
that Sally's changes don't overlap with his own; once he has
both sets of changes integrated, he saves his working copy
back to the repository.
<a class="xref" href="#svn.basic.vsn-models.copy-merge.dia-1" title="Figure 1.4. The copy-modify-merge solution">Figure 1.4, “The copy-modify-merge solution”</a> and
<a class="xref" href="#svn.basic.vsn-models.copy-merge.dia-2" title="Figure 1.5. The copy-modify-merge solution (continued)">Figure 1.5, “The copy-modify-merge solution (continued)”</a> show
this process.</p><div class="figure"><a id="svn.basic.vsn-models.copy-merge.dia-1"></a><p class="title"><strong>Figure 1.4. The copy-modify-merge solution</strong></p><div class="figure-contents"><div><img src="images/ch02dia4.png" alt="The copy-modify-merge solution"></img></div></div></div><br class="figure-break"></br><div class="figure"><a id="svn.basic.vsn-models.copy-merge.dia-2"></a><p class="title"><strong>Figure 1.5. The copy-modify-merge solution (continued)</strong></p><div class="figure-contents"><div><img src="images/ch02dia5.png" alt="The copy-modify-merge solution (continued)"></img></div></div></div><br class="figure-break"></br><a id="idp5959216" class="indexterm"></a><p>But what if Sally's changes <span class="emphasis"><em>do</em></span> overlap
with Harry's changes? What then? This situation is called a
<em class="firstterm">conflict</em>, and it's usually not much
of a problem. When Harry asks his client to merge the
latest repository changes into his working copy, his copy of
file A is somehow flagged as being in a state of conflict:
he'll be able to see both sets of conflicting changes and
manually choose between them. Note that software can't
automatically resolve conflicts; only humans are capable of
understanding and making the necessary intelligent choices.
Once Harry has manually resolved the overlapping
changes—perhaps after a discussion with Sally—he
can safely save the merged file back to the
repository.</p><p>The copy-modify-merge model may sound a bit chaotic, but
in practice, it runs extremely smoothly. Users can work in
parallel, never waiting for one another. When they work on
the same files, it turns out that most of their concurrent
changes don't overlap at all; conflicts are infrequent. And
the amount of time it takes to resolve conflicts is usually
far less than the time lost by a locking system.</p><p>In the end, it all comes down to one critical factor:
user communication. When users communicate poorly, both
syntactic and semantic conflicts increase. No system can
force users to communicate perfectly, and no system can
detect semantic conflicts. So there's no point in being
lulled into a false sense of security that a locking system
will somehow prevent conflicts; in practice, locking seems
to inhibit productivity more than anything else.</p><div class="sidebar" title="When Locking Is Necessary"><a id="svn.basic.vsn-models.copy-merge.sb-1"></a><div class="titlepage"><div><div><p class="title"><strong>When Locking Is Necessary</strong></p></div></div></div><p>While the lock-modify-unlock model is considered
generally harmful to collaboration, sometimes
locking is appropriate.</p><p>The copy-modify-merge model is based on the assumption
that files are contextually mergeable—that is, that the
majority of the files in the repository are line-based text
files (such as program source code). But for files with
binary formats, such as artwork or sound, it's often
impossible to merge conflicting changes. In these
situations, it really is necessary for users to take strict
turns when changing the file. Without serialized access,
somebody ends up wasting time on changes that are ultimately
discarded.</p><p>While Subversion is primarily a copy-modify-merge
system, it still recognizes the need to lock an occasional
file, and thus provides mechanisms for this. We discuss
this feature in <a class="xref" href="#svn.advanced.locking" title="Locking">the section called “Locking”</a>.</p></div></div></div></div><div class="sect1" title="Version Control the Subversion Way"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.basic.in-action"></a>Version Control the Subversion Way</h2></div></div></div><p>We've mentioned already that Subversion is a modern,
network-aware version control system. As we described in
<a class="xref" href="#svn.basic.version-control-basics" title="Version Control Basics">the section called “Version Control Basics”</a> (our
high-level version control overview), a repository serves as the
core storage mechanism for Subversion's versioned data, and it's
via working copies that users and their software programs
interact with that data. In this section, we'll begin to
introduce the specific ways in which Subversion implements
version control.</p><div class="sect2" title="Subversion Repositories"><div class="titlepage"><div><div><h3 class="title"><a id="svn.basic.svn-repositories"></a>Subversion Repositories</h3></div></div></div><p>Subversion implements the concept of a version control
repository much as any other modern version control system
would. Unlike a working copy, a Subversion repository is an
abstract entity, able to be operated upon almost exclusively
by Subversion's own libraries and tools. As most of a user's
Subversion interactions involve the use of the Subversion
client and occur in the context of a working copy, we spend
the majority of this book discussing the Subversion working
copy and how to manipulate it. For the finer details of the
repository, though, check out
<a class="xref" href="#svn.reposadmin" title="Chapter 5. Repository Administration">Chapter 5, <em>Repository Administration</em></a>.</p></div><div class="sect2" title="Revisions"><div class="titlepage"><div><div><h3 class="title"><a id="svn.basic.in-action.revs"></a>Revisions</h3></div></div></div><a id="idp5976192" class="indexterm"></a><p>A Subversion client commits (that is, communicates the
changes made to) any number of files and directories as a
single atomic transaction. By atomic transaction, we mean
simply this: either all of the changes are accepted into the
repository, or none of them is. Subversion tries to retain
this atomicity in the face of program crashes, system crashes,
network problems, and other users' actions.</p><p>Each time the repository accepts a commit, this creates a
new state of the filesystem tree, called a
<em class="firstterm">revision</em>. Each revision is assigned a
unique natural number, one greater than the number assigned to
the previous revision. The initial revision of a freshly
created repository is numbered 0 and consists of nothing but
an empty root directory.</p><p><a class="xref" href="#svn.basic.in-action.revs.dia-1" title="Figure 1.6. Tree changes over time">Figure 1.6, “Tree changes over time”</a>
illustrates a nice way to visualize the repository. Imagine
an array of revision numbers, starting at 0, stretching from
left to right. Each revision number has a filesystem tree
hanging below it, and each tree is a <span class="quote">“<span class="quote">snapshot</span>”</span>
of the way the repository looked after a commit.</p><div class="figure"><a id="svn.basic.in-action.revs.dia-1"></a><p class="title"><strong>Figure 1.6. Tree changes over time</strong></p><div class="figure-contents"><div><img src="images/ch02dia7.png" alt="Tree changes over time"></img></div></div></div><br class="figure-break"></br><div class="sidebar" title="Global Revision Numbers"><div class="titlepage"><div><div><p class="title"><strong>Global Revision Numbers</strong></p></div></div></div><a id="idp5983408" class="indexterm"></a><p>Unlike most version control systems, Subversion's
revision numbers apply to <span class="emphasis"><em>the entire repository
tree</em></span>, not individual files. Each revision number
selects an entire tree, a particular state of the repository
after some committed change. Another way to think about it
is that revision N represents the state of the repository
filesystem after the Nth commit. When Subversion users talk
about <span class="quote">“<span class="quote">revision 5 of
<code class="filename">foo.c</code>,</span>”</span> they really mean
<span class="quote">“<span class="quote"><code class="filename">foo.c</code> as it appears in revision
5.</span>”</span> Notice that in general, revisions N and M of a
file do <span class="emphasis"><em>not</em></span> necessarily differ! Many
other version control systems use per-file revision numbers,
so this concept may seem unusual at first. (Former CVS users
might want to see <a class="xref" href="#svn.forcvs" title="Appendix B. Subversion for CVS Users">Appendix B, <em>Subversion for CVS Users</em></a> for more
details.)</p></div></div><div class="sect2" title="Addressing the Repository"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.reposurls"></a>Addressing the Repository</h3></div></div></div><a id="idp5990624" class="indexterm"></a><a id="idp5992400" class="indexterm"></a><p>Subversion client programs use URLs to identify versioned
files and directories in Subversion repositories. For the
most part, these URLs use the standard syntax, allowing for
server names and port numbers to be specified as part of the
URL.</p><div class="informalexample"><div class="itemizedlist"><ul class="itemizedlist" type="disc" compact><li class="listitem">http://svn.example.com/svn/project</li><li class="listitem">http://svn.example.com:9834/repos</li></ul></div></div><p>Subversion repository URLs aren't limited to only
the <code class="literal">http://</code> variety. Because Subversion
offers several different ways for its clients to communicate
with its servers, the URLs used to address the repository
differ subtly depending on which repository access mechanism
is employed. <a class="xref" href="#svn.basic.in-action.wc.tbl-1" title="Table 1.1. Repository access URLs">Table 1.1, “Repository access URLs”</a>
describes how different URL schemes map to the available
repository access methods. For more details about
Subversion's server options, see
<a class="xref" href="#svn.serverconfig" title="Chapter 6. Server Configuration">Chapter 6, <em>Server Configuration</em></a>.</p><div class="table"><a id="svn.basic.in-action.wc.tbl-1"></a><p class="title"><strong>Table 1.1. Repository access URLs</strong></p><div class="table-contents"><table summary="Repository access URLs" border="1"><colgroup><col></col><col></col></colgroup><thead><tr><th>Schema</th><th>Access method</th></tr></thead><tbody><tr><td><code class="literal">file:///</code></td><td>Direct repository access (on local disk)</td></tr><tr><td><code class="literal">http://</code></td><td>Access via WebDAV protocol to Subversion-aware
Apache server</td></tr><tr><td><code class="literal">https://</code></td><td>Same as <code class="literal">http://</code>, but with
SSL encapsulation (encryption and authentication)</td></tr><tr><td><code class="literal">svn://</code></td><td>Access via custom protocol to an
<code class="literal">svnserve</code> server</td></tr><tr><td><code class="literal">svn+ssh://</code></td><td>Same as <code class="literal">svn://</code>, but through
an SSH tunnel</td></tr></tbody></table></div></div><br class="table-break"></br><p>Subversion's handling of URLs has some notable nuances.
For example, URLs containing the <code class="literal">file://</code>
access method (used for local repositories) must, in
accordance with convention, have either a server name
of <code class="literal">localhost</code> or no server name at
all:</p><div class="informalexample"><div class="itemizedlist"><ul class="itemizedlist" type="disc" compact><li class="listitem">file:///var/svn/repos</li><li class="listitem">file://localhost/var/svn/repos</li></ul></div></div><p>Also, users of the <code class="literal">file://</code> scheme on
Windows platforms will need to use an unofficially
<span class="quote">“<span class="quote">standard</span>”</span> syntax for accessing repositories
that are on the same machine, but on a different drive than
the client's current working drive. Either of the two
following URL path syntaxes will work, where
<code class="literal">X</code> is the drive on which the repository
resides:</p><div class="informalexample"><div class="itemizedlist"><ul class="itemizedlist" type="disc" compact><li class="listitem">file:///X:/var/svn/repos</li><li class="listitem">file:///X|/var/svn/repos</li></ul></div></div><p>Note that a URL uses forward slashes even though the
native (non-URL) form of a path on Windows uses backslashes.
Also note that when using
the <code class="literal">file:///<em class="replaceable"><code>X</code></em>|/</code>
form at the command line, you need to quote the URL (wrap it
in quotation marks) so that the vertical bar character is not
interpreted as a pipe.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>You cannot use Subversion's <code class="literal">file://</code> URLs
in a regular web browser the way typical
<code class="literal">file://</code> URLs can. When you attempt to view
a <code class="literal">file://</code> URL in a regular web browser, it
reads and displays the contents of the file at that location
by examining the filesystem directly. However, Subversion's
resources exist in a virtual filesystem (see <a class="xref" href="#svn.developer.layerlib.repos" title="Repository Layer">the section called “Repository Layer”</a>), and your browser
will not understand how to interact with that
filesystem.</p></td></tr></table></div><p>The Subversion client will automatically encode URLs as
necessary, just like a web browser does. For example, the URL
<code class="literal">http://host/path with space/project/españa</code>
— which contains both spaces and upper-ASCII characters
— will be automatically interpreted by Subversion as if
you'd provided
<code class="literal">http://host/path%20with%20space/project/espa%C3%B1a</code>.
If the URL contains spaces, be sure to place it within
quotation marks at the command line so that your shell treats
the whole thing as a single argument to the program.</p><p>There is one notable exception to Subversion's handling of
URLs which also applies to its handling of local paths in many
contexts, too. If the final path component of your URL or
local path contains an at sign (<code class="literal">@</code>), you need
to use a special syntax—described in
<a class="xref" href="#svn.advanced.pegrevs" title="Peg and Operative Revisions">the section called “Peg and Operative Revisions”</a>—in order to make
Subversion properly address that resource.</p><p>In Subversion 1.6, a new caret (<code class="literal">^</code>)
notation was introduced as a shorthand for <span class="quote">“<span class="quote">the URL of
the repository's root directory</span>”</span>. For example, you can
use the <code class="literal">^/tags/bigsandwich/</code> to refer to the
URL of the <code class="filename">/tags/bigsandwich</code> directory in
the root of the repository. Note that this URL syntax works
only when your current working directory is a working
copy—the command-line client knows the repository's root
URL by looking at the working copy's metadata. Also note that
when you wish to refer precisely to the root directory of the
repository, you must do so using <code class="literal">^/</code> (with
the trailing slash character), not merely
<code class="literal">^</code>.</p></div><div class="sect2" title="Subversion Working Copies"><div class="titlepage"><div><div><h3 class="title"><a id="svn.basic.in-action.wc"></a>Subversion Working Copies</h3></div></div></div><a id="idp6037792" class="indexterm"></a><p>A Subversion working copy is an ordinary directory tree on
your local system, containing a collection of files. You can
edit these files however you wish, and if they're source code
files, you can compile your program from them in the usual
way. Your working copy is your own private work area:
Subversion will never incorporate other people's changes, nor
make your own changes available to others, until you
explicitly tell it to do so. You can even have multiple
working copies of the same project.</p><p>After you've made some changes to the files in your
working copy and verified that they work properly, Subversion
provides you with commands to <span class="quote">“<span class="quote">publish</span>”</span> your
changes to the other people working with you on your project
(by writing to the repository). If other people publish their
own changes, Subversion provides you with commands to merge
those changes into your working copy (by reading from the
repository).</p><p>A working copy also contains some extra files, created and
maintained by Subversion, to help it carry out these commands.
In particular, each working copy contains a subdirectory
named <code class="filename">.svn</code>, also known as the working
copy's <em class="firstterm">administrative directory</em>. The
files in the administrative directory help Subversion
recognize which of your versioned files contain unpublished
changes, and which files are out of date with respect to
others' work.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>Prior to version 1.7, Subversion
maintained <code class="filename">.svn</code> administrative
subdirectories in <span class="emphasis"><em>every</em></span> versioned
directory of your working copy. Subversion 1.7 offers a
completely new approach to how working copy metadata is
stored and maintained, and chief among the visible changes
to this approach is that each working copy now has only
one <code class="filename">.svn</code> subdirectory which is an
immediate child of the root of that working copy.</p></td></tr></table></div><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>While <code class="filename">.svn</code> is the de facto name of
the Subversion administrative directory, Windows users may
run into problems with the ASP.NET Web application framework
disallowing access to directories whose names begin with a
dot (<code class="literal">.</code>). As a special consideration to
users in such situations, Subversion will instead use
<code class="literal">_svn</code> as the administrative directory name
if it finds a variable
named <code class="literal">SVN_ASP_DOT_NET_HACK</code> in its
operating environment. Throughout this book, any reference
you find to <code class="filename">.svn</code> applies also
to <code class="literal">_svn</code> when this <span class="quote">“<span class="quote">ASP.NET
hack</span>”</span> is in use.</p></td></tr></table></div><div class="sect3" title="How the working copy works"><div class="titlepage"><div><div><h4 class="title"><a id="svn.basic.in-action.track-repos"></a>How the working copy works</h4></div></div></div><p>For each file in a working directory, Subversion records
(among other things) two essential pieces of information:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>What revision your working file is based on (this is
called the file's <em class="firstterm">working
revision</em>)</p></li><li class="listitem"><p>A timestamp recording when the local copy was last
updated by the repository</p></li></ul></div><p>Given this information, by talking to the repository,
Subversion can tell which of the following four states a
working file is in:</p><div class="variablelist"><dl><dt><span class="term">Unchanged, and current</span></dt><dd><p>The file is unchanged in the working directory, and
no changes to that file have been committed to the
repository since its working revision. An <span class="command"><strong>svn
commit</strong></span> of the file will do nothing, and an
<span class="command"><strong>svn update</strong></span> of the file will do
nothing.</p></dd><dt><span class="term">Locally changed, and current</span></dt><dd><p>The file has been changed in the working directory,
and no changes to that file have been committed to the
repository since you last updated. There are local
changes that have not been committed to the repository;
thus an <span class="command"><strong>svn commit</strong></span> of the file will
succeed in publishing your changes, and an <span class="command"><strong>svn
update</strong></span> of the file will do nothing.</p></dd><dt><span class="term">Unchanged, and out of date</span></dt><dd><p>The file has not been changed in the working
directory, but it has been changed in the repository.
The file should eventually be updated in order to make
it current with the latest public revision.
An <span class="command"><strong>svn commit</strong></span> of the file will do
nothing, and an
<span class="command"><strong>svn update</strong></span> of the file will fold the
latest changes into your working copy.</p></dd><dt><span class="term">Locally changed, and out of date</span></dt><dd><p>The file has been changed both in the working
directory and in the repository. An <span class="command"><strong>svn
commit</strong></span> of the file will fail with an
<span class="quote">“<span class="quote">out-of-date</span>”</span> error. The file should be
updated first; an <span class="command"><strong>svn update</strong></span> command
will attempt to merge the public changes with the local
changes. If Subversion can't complete the merge in a
plausible way automatically, it leaves it to the user to
resolve the conflict.</p></dd></dl></div></div><div class="sect3" title="Fundamental working copy interactions"><div class="titlepage"><div><div><h4 class="title"><a id="svn.basic.in-action.wc-funcdamentals"></a>Fundamental working copy interactions</h4></div></div></div><p>A typical Subversion repository often holds the files (or
source code) for several projects; usually, each project is a
subdirectory in the repository's filesystem tree. In this
arrangement, a user's working copy will usually correspond to
a particular subtree of the repository.</p><p>For example, suppose you have a repository that contains
two software projects, <code class="literal">paint</code> and
<code class="literal">calc</code>. Each project lives in its own
top-level subdirectory, as shown in <a class="xref" href="#svn.basic.in-action.wc.dia-1" title="Figure 1.7. The repository's filesystem">Figure 1.7, “The repository's filesystem”</a>.</p><div class="figure"><a id="svn.basic.in-action.wc.dia-1"></a><p class="title"><strong>Figure 1.7. The repository's filesystem</strong></p><div class="figure-contents"><div><img src="images/ch02dia6.png" alt="The repository's filesystem"></img></div></div></div><br class="figure-break"></br><a id="idp6075872" class="indexterm"></a><a id="idp6077648" class="indexterm"></a><a id="idp6079040" class="indexterm"></a><p>To get a working copy, you must <em class="firstterm">check
out</em> some subtree of the repository. (The term
<span class="emphasis"><em>check out</em></span> may sound like it has something to do
with locking or reserving resources, but it doesn't; it simply
creates a working copy of the project for you.) For example,
if you check out <code class="filename">/calc</code>, you will get a
working copy like this:</p><div class="informalexample"><pre class="screen">
$ svn checkout http://svn.example.com/repos/calc
A calc/Makefile
A calc/integer.c
A calc/button.c
Checked out revision 56.
$ ls -A calc
Makefile button.c integer.c .svn/
$
</pre></div><p>The list of letter <code class="literal">A</code>s in the left
margin indicates that Subversion is adding a number of items
to your working copy. You now have a personal copy of the
repository's <code class="filename">/calc</code> directory, with one
additional entry—<code class="filename">.svn</code>—which
holds the extra information needed by Subversion, as mentioned
earlier.</p><p>Suppose you make changes to <code class="filename">button.c</code>.
Since the <code class="filename">.svn</code> directory remembers the
file's original modification date and contents, Subversion can
tell that you've changed the file. However, Subversion does
not make your changes public until you explicitly tell it to.
The act of publishing your changes is more commonly known as
<em class="firstterm">committing</em> (or <em class="firstterm">checking
in</em>) changes to the repository.</p><a id="idp6088928" class="indexterm"></a><a id="idp6090704" class="indexterm"></a><a id="idp6092096" class="indexterm"></a><p>To publish your changes to others, you can use
Subversion's <span class="command"><strong>svn commit</strong></span> command:</p><div class="informalexample"><pre class="screen">
$ svn commit button.c -m "Fixed a typo in button.c."
Sending button.c
Transmitting file data .
Committed revision 57.
$
</pre></div><p>Now your changes to <code class="filename">button.c</code> have
been committed to the repository, with a note describing your
change (namely, that you fixed a typo). If another user
checks out a working copy of <code class="filename">/calc</code>, she
will see your changes in the latest version of the
file.</p><p>Suppose you have a collaborator, Sally, who checked out a
working copy of <code class="filename">/calc</code> at the same time
you did. When you commit your change to
<code class="filename">button.c</code>, Sally's working copy is left
unchanged; Subversion modifies working copies only at the
user's request.</p><a id="idp6099632" class="indexterm"></a><a id="idp6101408" class="indexterm"></a><a id="idp6102800" class="indexterm"></a><p>To bring her project up to date, Sally can ask Subversion
to <em class="firstterm">update</em> her working copy, by using
the <span class="command"><strong>svn update</strong></span> command. This will incorporate
your changes into her working copy, as well as any others that
have been committed since she checked it out.</p><div class="informalexample"><pre class="screen">
$ pwd
/home/sally/calc
$ ls -A
Makefile button.c integer.c .svn/
$ svn update
Updating '.':
U button.c
Updated to revision 57.
$
</pre></div><p>The output from the <span class="command"><strong>svn update</strong></span> command
indicates that Subversion updated the contents of
<code class="filename">button.c</code>. Note that Sally didn't need to
specify which files to update; Subversion uses the information
in the <code class="filename">.svn</code> directory as well as further
information in the repository, to decide which files need to
be brought up to date.</p></div><div class="sect3" title="Mixed-revision working copies"><div class="titlepage"><div><div><h4 class="title"><a id="svn.basic.in-action.mixedrevs"></a>Mixed-revision working copies</h4></div></div></div><a id="idp6110976" class="indexterm"></a><p>As a general principle, Subversion tries to be as flexible
as possible. One special kind of flexibility is the ability
to have a working copy containing files and directories with a
mix of different working revision numbers. Subversion working
copies do not always correspond to any single revision in the
repository; they may contain files from several different
revisions. For example, suppose you check out a working copy
from a repository whose most recent revision is 4:</p><div class="informalexample"><div class="literallayout"><p><br></br>
calc/<br></br>
Makefile:4<br></br>
integer.c:4<br></br>
button.c:4<br></br>
</p></div></div><p>At the moment, this working directory corresponds exactly
to revision 4 in the repository. However, suppose you make a
change to <code class="filename">button.c</code>, and commit that
change. Assuming no other commits have taken place, your
commit will create revision 5 of the repository, and your
working copy will now look like this:</p><div class="informalexample"><div class="literallayout"><p><br></br>
calc/<br></br>
Makefile:4<br></br>
integer.c:4<br></br>
button.c:5<br></br>
</p></div></div><p>Suppose that, at this point, Sally commits a change to
<code class="filename">integer.c</code>, creating revision 6. If you
use <span class="command"><strong>svn update</strong></span> to bring your working copy
up to date, it will look like this:</p><div class="informalexample"><div class="literallayout"><p><br></br>
calc/<br></br>
Makefile:6<br></br>
integer.c:6<br></br>
button.c:6<br></br>
</p></div></div><p>Sally's change to <code class="filename">integer.c</code> will
appear in your working copy, and your change will still be
present in <code class="filename">button.c</code>. In this example,
the text of <code class="filename">Makefile</code> is identical in
revisions 4, 5, and 6, but Subversion will mark your working
copy of <code class="filename">Makefile</code> with revision 6 to
indicate that it is still current. So, after you do a clean
update at the top of your working copy, it will generally
correspond to exactly one revision in the repository.</p><div class="sect4" title="Updates and commits are separate"><div class="titlepage"><div><div><h5 class="title"><a id="svn.basic.in-action.mixedrevs.update-commit"></a>Updates and commits are separate</h5></div></div></div><p>One of the fundamental rules of Subversion is that
a <span class="quote">“<span class="quote">push</span>”</span> action does not cause
a <span class="quote">“<span class="quote">pull</span>”</span> nor vice versa. Just
because you're ready to submit new changes to the repository
doesn't mean you're ready to receive changes from other
people. And if you have new changes still in progress,
<span class="command"><strong>svn update</strong></span> should gracefully merge
repository changes into your own, rather than forcing you to
publish them.</p><p>The main side effect of this rule is that it means a
working copy has to do extra bookkeeping to track mixed
revisions as well as be tolerant of the mixture. It's made
more complicated by the fact that directories themselves are
versioned.</p><p>For example, suppose you have a working copy entirely at
revision 10. You edit the
file <code class="filename">foo.html</code> and then perform
an <span class="command"><strong>svn commit</strong></span>, which creates revision 15
in the repository. After the commit succeeds, many new
users would expect the working copy to be entirely at
revision 15, but that's not the case! Any number of changes
might have happened in the repository between revisions 10
and 15. The client knows nothing of those changes in the
repository, since you haven't yet run <span class="command"><strong>svn
update</strong></span>, and <span class="command"><strong>svn commit</strong></span> doesn't
pull down new changes. If, on the other hand,
<span class="command"><strong>svn commit</strong></span> were to automatically download
the newest changes, it would be possible to set the
entire working copy to revision 15—but then we'd be
breaking the fundamental rule of <span class="quote">“<span class="quote">push</span>”</span>
and <span class="quote">“<span class="quote">pull</span>”</span> remaining separate actions.
Therefore, the only safe thing the Subversion client can do
is mark the one
file—<code class="filename">foo.html</code>—as being at
revision 15. The rest of the working copy remains at
revision 10. Only by running <span class="command"><strong>svn update</strong></span>
can the latest changes be downloaded and the whole working
copy be marked as revision 15.</p></div><div class="sect4" title="Mixed revisions are normal"><div class="titlepage"><div><div><h5 class="title"><a id="svn.basic.in-action.mixedrevs.normal"></a>Mixed revisions are normal</h5></div></div></div><p>The fact is, <span class="emphasis"><em>every time</em></span> you run
<span class="command"><strong>svn commit</strong></span> your working copy ends up
with some mixture of revisions. The things you just
committed are marked as having larger working revisions than
everything else. After several commits (with no updates
in between), your working copy will contain a whole mixture
of revisions. Even if you're the only person using the
repository, you will still see this phenomenon. To examine
your mixture of working revisions, use the <span class="command"><strong>svn
status</strong></span> command with the <code class="option">--verbose</code>
(<code class="option">-v</code>) option (see
<a class="xref" href="#svn.tour.cycle.examine.status" title="See an overview of your changes">the section called “See an overview of your changes”</a> for more
information).</p><p>Often, new users are completely unaware that their
working copy contains mixed revisions. This can be
confusing, because many client commands are sensitive to the
working revision of the item they're examining. For
example, the <span class="command"><strong>svn log</strong></span> command is used to
display the history of changes to a file or directory (see
<a class="xref" href="#svn.tour.history.log" title="Generating a List of Historical Changes">the section called “Generating a List of Historical Changes”</a>). When the user
invokes this command on a working copy object, he expects
to see the entire history of the object. But if the
object's working revision is quite old (often because
<span class="command"><strong>svn update</strong></span> hasn't been run in a long
time), the history of the <span class="emphasis"><em>older</em></span>
version of the object is shown.</p></div><div class="sect4" title="Mixed revisions are useful"><div class="titlepage"><div><div><h5 class="title"><a id="svn.basic.in-action.mixedrevs.useful"></a>Mixed revisions are useful</h5></div></div></div><p>If your project is sufficiently complex, you'll discover
that it's sometimes nice to
forcibly <em class="firstterm">backdate</em> (or update to a
revision older than the one you already have) portions of
your working copy to an earlier revision; you'll learn how
to do that in <a class="xref" href="#svn.tour" title="Chapter 2. Basic Usage">Chapter 2, <em>Basic Usage</em></a>. Perhaps you'd
like to test an earlier version of a submodule contained in
a subdirectory, or perhaps you'd like to figure out when a
bug first came into existence in a specific file. This is
the <span class="quote">“<span class="quote">time machine</span>”</span> aspect of a version control
system—the feature that allows you to move any
portion of your working copy forward and backward in
history.</p></div><div class="sect4" title="Mixed revisions have limitations"><div class="titlepage"><div><div><h5 class="title"><a id="svn.basic.in-action.mixedrevs.limits"></a>Mixed revisions have limitations</h5></div></div></div><p>However you make use of mixed revisions in your working
copy, there are limitations to this flexibility.</p><p>First, you cannot commit the deletion of a file or
directory that isn't fully up to date. If a newer version
of the item exists in the repository, your attempt to delete
will be rejected to prevent you from accidentally
destroying changes you've not yet seen.</p><p>Second, you cannot commit a metadata change to a
directory unless it's fully up to date. You'll learn about
attaching <span class="quote">“<span class="quote">properties</span>”</span> to items in <a class="xref" href="#svn.advanced" title="Chapter 3. Advanced Topics">Chapter 3, <em>Advanced Topics</em></a>. A directory's working revision
defines a specific set of entries and properties, and thus
committing a property change to an out-of-date directory may
destroy properties you've not yet seen.</p><p>Finally, beginning in Subversion 1.7, you cannot by
default use a mixed-revision working copy as the target of
a merge operation. (This new requirement was introduced
to prevent common problems which stem from doing so.)</p></div></div></div></div><div class="sect1" title="Summary"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.basic.summary"></a>Summary</h2></div></div></div><p>We covered a number of fundamental Subversion concepts in
this chapter:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>We introduced the notions of the central repository,
the client working copy, and the array of repository
revision trees.</p></li><li class="listitem"><p>We saw some simple examples of how two collaborators
can use Subversion to publish and receive changes from one
another, using the <span class="quote">“<span class="quote">copy-modify-merge</span>”</span>
model.</p></li><li class="listitem"><p>We talked a bit about the way Subversion tracks and
manages information in a working copy.</p></li></ul></div><p>At this point, you should have a good idea of how Subversion
works in the most general sense. Armed with this knowledge, you
should now be ready to move into the next chapter, which is a
detailed tour of Subversion's commands and features.</p></div><div class="footnotes"><br></br><hr width="100" align="left"></hr><div class="footnote"><p><sup>[<a id="ftn.idp5915856" href="#idp5915856" class="para">5</a>] </sup>The
term <span class="quote">“<span class="quote">working copy</span>”</span> can be generally applied to
any one file version's local instance. When most folks use
the term, though, they are referring to a whole directory tree
containing files and subdirectories managed by the version
control system.</p></div></div></div><div class="chapter" title="Chapter 2. Basic Usage"><div class="titlepage"><div><div><h2 class="title"><a id="svn.tour"></a>Chapter 2. Basic Usage</h2></div></div></div><div class="toc"><p><strong>Table of Contents</strong></p><dl><dt><span class="sect1"><a href="#svn.tour.help">Help!</a></span></dt><dt><span class="sect1"><a href="#svn.tour.importing">Getting Data into Your Repository</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.tour.importing.import">Importing Files and Directories</a></span></dt><dt><span class="sect2"><a href="#svn.tour.importing.layout">Recommended Repository Layout</a></span></dt><dt><span class="sect2"><a href="#svn.tour.importing.naming">What's In a Name?</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.tour.initial">Creating a Working Copy</a></span></dt><dt><span class="sect1"><a href="#svn.tour.cycle">Basic Work Cycle</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.tour.cycle.update">Update Your Working Copy</a></span></dt><dt><span class="sect2"><a href="#svn.tour.cycle.edit">Make Your Changes </a></span></dt><dt><span class="sect2"><a href="#svn.tour.cycle.examine">Review Your Changes</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.tour.cycle.examine.status">See an overview of your changes</a></span></dt><dt><span class="sect3"><a href="#svn.tour.cycle.examine.diff">Examine the details of your local modifications</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.tour.cycle.revert">Fix Your Mistakes</a></span></dt><dt><span class="sect2"><a href="#svn.tour.cycle.resolve">Resolve Any Conflicts</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.tour.cycle.resolve.diff">Viewing conflict differences interactively</a></span></dt><dt><span class="sect3"><a href="#svn.tour.cycle.resolve.resolve">Resolving conflict differences interactively</a></span></dt><dt><span class="sect3"><a href="#svn.tour.cycle.resolve.pending">Postponing conflict resolution</a></span></dt><dt><span class="sect3"><a href="#svn.tour.cycle.resolve.byhand">Merging conflicts by hand</a></span></dt><dt><span class="sect3"><a href="#svn.tour.cycle.resolve.theirsfull">Discarding your changes in favor of a newly fetched
revision</a></span></dt><dt><span class="sect3"><a href="#svn.tour.cycle.resolve.revert">Punting: using svn revert</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.tour.cycle.commit">Commit Your Changes</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.tour.history">Examining History</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.tour.history.diff">Examining the Details of Historical Changes</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.tour.history.diff.local">Examining local changes</a></span></dt><dt><span class="sect3"><a href="#svn.tour.history.diff.wcrepos">Comparing working copy to repository</a></span></dt><dt><span class="sect3"><a href="#svn.tour.history.diff.reposrepos">Comparing repository revisions</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.tour.history.log">Generating a List of Historical Changes</a></span></dt><dt><span class="sect2"><a href="#svn.tour.history.browsing">Browsing the Repository</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.tour.history.browsing.cat">Displaying file contents</a></span></dt><dt><span class="sect3"><a href="#svn.tour.history.browsing.annotate">Displaying line-by-line change attribution</a></span></dt><dt><span class="sect3"><a href="#svn.tour.history.browsing.list">Listing versioned directories</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.tour.history.snapshots">Fetching Older Repository Snapshots</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.tour.cleanup">Sometimes You Just Need to Clean Up</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.tour.cleanup.disposal">Disposing of a Working Copy</a></span></dt><dt><span class="sect2"><a href="#svn.tour.cleanup.interruption">Recovering from an Interruption</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.tour.treeconflicts">Dealing with Structural Conflicts</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.tour.treeconflicts.example">An Example Tree Conflict</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.tour.summary">Summary</a></span></dt></dl></div><p>Theory is useful, but its application is just plain fun.
Let's move now into the details of using Subversion. By the time
you reach the end of this chapter, you will be able to perform all
the tasks you need to use Subversion in a normal day's work.
You'll start with getting your files into Subversion, followed by
an initial checkout of your code. We'll then walk you through
making changes and examining those changes. You'll also see how
to bring changes made by others into your working copy, examine
them, and work through any conflicts that might arise.</p><p>This chapter will not provide exhaustive coverage
of all of Subversion's commands—rather, it's a conversational
introduction to the most common Subversion tasks that you'll
encounter. This chapter assumes that you've read and understood
<a class="xref" href="#svn.basic" title="Chapter 1. Fundamental Concepts">Chapter 1, <em>Fundamental Concepts</em></a> and are familiar with the general
model of Subversion. For a complete reference of all commands,
see <a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a>.</p><p>Also, this chapter assumes that the reader is seeking
information about how to interact in a basic fashion with an
existing Subversion repository. No repository means no working
copy; no working copy means not much of interest in this chapter.
There are many Internet sites which offer free or inexpensive
Subversion repository hosting services. Or, if you'd prefer to
set up and administer your own repositories, check out
<a class="xref" href="#svn.reposadmin" title="Chapter 5. Repository Administration">Chapter 5, <em>Repository Administration</em></a>. But don't expect the
examples in this chapter to work without the user having access
to a Subversion repository.</p><p>Finally, any Subversion operation that contacts the repository
over a network may potentially require that the user authenticate.
For the sake of simplicity, our examples throughout this chapter
avoid demonstrating and discussing authentication. Be aware that
if you hope to apply the knowledge herein to an existing,
real-world Subversion instance, you'll probably be forced to
provide at least a username and password to the server. See
<a class="xref" href="#svn.serverconfig.netmodel.creds" title="Client Credentials">the section called “Client Credentials”</a> for a detailed
description of Subversion's handling of authentication and client
credentials.</p><div class="sect1" title="Help!"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.tour.help"></a>Help!</h2></div></div></div><a id="idp6418080" class="indexterm"></a><p>It goes without saying that this book exists to be a source
of information and assistance for Subversion users new and old.
Conveniently, though, the Subversion command-line is
self-documenting, alleviating the need to grab a book off the
shelf (wooden, virtual, or otherwise). The <span class="command"><strong>svn
help</strong></span> command is your gateway to that built-in
documentation:</p><div class="informalexample"><pre class="screen">
$ svn help
Subversion command-line client, version 1.7.0.
Type 'svn help <subcommand>' for help on a specific subcommand.
Type 'svn --version' to see the program version and RA modules
or 'svn --version --quiet' to see just the version number.
Most subcommands take file and/or directory arguments, recursing
on the directories. If no arguments are supplied to such a
command, it recurses on the current directory (inclusive) by default.
Available subcommands:
add
blame (praise, annotate, ann)
cat
…
</pre></div><p>As described in the previous output, you can ask for help on
a particular subcommand by running <strong class="userinput"><code>svn help
<em class="replaceable"><code>SUBCOMMAND</code></em></code></strong>. Subversion
will respond with the full usage message for that subcommand,
including its syntax, options, and behavior:</p><div class="informalexample"><pre class="screen">
$ svn help help
help (?, h): Describe the usage of this program or its subcommands.
usage: help [SUBCOMMAND...]
Global options:
--username ARG : specify a username ARG
--password ARG : specify a password ARG
…
</pre></div><div class="sidebar" title="Options and Switches and Flags, Oh My!"><div class="titlepage"><div><div><p class="title"><strong>Options and Switches and Flags, Oh My!</strong></p></div></div></div><a id="idp6426160" class="indexterm"></a><p>The Subversion command-line client has numerous command
modifiers. Some folks refer to such things
as <span class="quote">“<span class="quote">switches</span>”</span> or <span class="quote">“<span class="quote">flags</span>”</span>—in
this book, we'll call them <span class="quote">“<span class="quote">options</span>”</span>. You'll
find the options supported by a given <span class="command"><strong>svn</strong></span>
subcommand, plus a set of options which are globally supported
by all subcommands, listed near the bottom of the built-in
usage message for that subcommand.</p><p>Subversion's options have two distinct forms: short
options are a single hyphen followed by a single letter, and
long options consist of two hyphens followed by several
letters and hyphens (e.g., <code class="literal">-s</code>
and <code class="literal">--this-is-a-long-option</code>, respectively).
Every option has at least one long format. Some, such as
the <code class="option">--changelist</code> option, feature an
abbreviated long-format alias (<code class="option">--cl</code>, in this
case). Only certain options—generally the most-used
ones—have an additional short format. To maintain
clarity in this book, we usually use the long form in code
examples, but when describing options, if there's a short
form, we'll provide the long form (to improve clarity) and the
short form (to make it easier to remember). Use the form
you're more comfortable with when executing your own
Subversion commands.</p></div><p>Many Unix-based distributions of Subversion include manual
pages of the sort that can be invoked using
the <span class="command"><strong>man</strong></span> program, but those tend to carry only
pointers to other sources of real help, such as the project's
website and to the website which hosts this book. Also, several
companies offer Subversion help and support, too, usually via a
mixture of web-based discussion forums and fee-based consulting.
And of course, the Internet holds a decade's worth of
Subversion-related discussions just begging to be located by
your favorite search engine. Subversion help is never too far
away.</p></div><div class="sect1" title="Getting Data into Your Repository"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.tour.importing"></a>Getting Data into Your Repository</h2></div></div></div><p>You can get new files into your Subversion
repository in two ways: <span class="command"><strong>svn import</strong></span> and <span class="command"><strong>svn
add</strong></span>. We'll discuss <span class="command"><strong>svn import</strong></span> now
and will discuss <span class="command"><strong>svn add</strong></span> later in this
chapter when we review a typical day with Subversion.</p><div class="sect2" title="Importing Files and Directories"><div class="titlepage"><div><div><h3 class="title"><a id="svn.tour.importing.import"></a>Importing Files and Directories</h3></div></div></div><a id="idp6441568" class="indexterm"></a><p>The <span class="command"><strong>svn import</strong></span> command is a quick way to
copy an unversioned tree of files into a repository, creating
intermediate directories as necessary. <span class="command"><strong>svn
import</strong></span> doesn't require a working copy, and your files
are immediately committed to the repository. You typically
use this when you have an existing tree of files that you want to
begin tracking in your Subversion repository. For example:</p><div class="informalexample"><pre class="screen">
$ svn import /path/to/mytree \
http://svn.example.com/svn/repo/some/project \
-m "Initial import"
Adding mytree/foo.c
Adding mytree/bar.c
Adding mytree/subdir
Adding mytree/subdir/quux.h
Committed revision 1.
$
</pre></div><p>The previous example copied the contents of the local
directory <code class="filename">mytree</code> into the directory
<code class="filename">some/project</code> in the repository. Note
that you didn't have to create that new directory
first—<span class="command"><strong>svn import</strong></span> does that for you.
Immediately after the commit, you can see your data in the
repository:</p><div class="informalexample"><pre class="screen">
$ svn list http://svn.example.com/svn/repo/some/project
bar.c
foo.c
subdir/
$
</pre></div><p>Note that after the import is finished, the original local
directory is <span class="emphasis"><em>not</em></span> converted into a working
copy. To begin working on that data in a versioned fashion,
you still need to create a fresh working copy of that
tree.</p></div><div class="sect2" title="Recommended Repository Layout"><div class="titlepage"><div><div><h3 class="title"><a id="svn.tour.importing.layout"></a>Recommended Repository Layout</h3></div></div></div><a id="idp6452800" class="indexterm"></a><a id="idp6453808" class="indexterm"></a><a id="idp6454816" class="indexterm"></a><a id="idp6455824" class="indexterm"></a><p>Subversion provides the ultimate flexibility in terms of
how you arrange your data. Because it simply versions
directories and files, and because it ascribes no particular
meaning to any of those objects, you may arrange the data in
your repository in any way that you choose. Unfortunately,
this flexibility also means that it's easy to find
yourself <span class="quote">“<span class="quote">lost without a roadmap</span>”</span> as you attempt
to navigate different Subversion repositories which may carry
completely different and unpredictable arrangements of the
data within them.</p><p>To counteract this confusion, we recommend that you follow
a repository layout convention (established long ago, in the
nascency of the Subversion project itself) in which a handful
of strategically named Subversion repository directories
convey valuable meaning about the data they hold. Most
projects have a recognizable <span class="quote">“<span class="quote">main line</span>”</span>,
or <em class="firstterm">trunk</em>, of development;
some <em class="firstterm">branches</em>, which are divergent
copies of development lines; and
some <em class="firstterm">tags</em>, which are named, stable
snapshots of a particular line of development. So we first
recommend that each project have a
recognizable <em class="firstterm">project root</em> in the
repository, a directory under which all of the versioned
information for that project—and only that
project—lives. Secondly, we suggest that each project
root contain a <code class="filename">trunk</code> subdirectory for the
main development line, a
<code class="filename">branches</code> subdirectory in which specific
branches (or collections of branches) will be created, and
a <code class="filename">tags</code> subdirectory in which specific
tags (or collections of tags) will be created. Of course, if
a repository houses only a single project, the root of the
repository can serve as the project root, too.</p><p>Here are some examples:</p><div class="informalexample"><pre class="screen">
$ svn list file:///var/svn/single-project-repo
trunk/
branches/
tags/
$ svn list file:///var/svn/multi-project-repo
project-A/
project-B/
$ svn list file:///var/svn/multi-project-repo/project-A
trunk/
branches/
tags/
$
</pre></div><p>We talk much more about tags and branches in
<a class="xref" href="#svn.branchmerge" title="Chapter 4. Branching and Merging">Chapter 4, <em>Branching and Merging</em></a>. For details and some
advice on how to set up repositories when you have multiple
projects, see <a class="xref" href="#svn.branchmerge.maint.layout" title="Repository Layout">the section called “Repository Layout”</a>.
Finally, we discuss project roots more in
<a class="xref" href="#svn.reposadmin.projects.chooselayout" title="Planning Your Repository Organization">the section called “Planning Your Repository Organization”</a>.</p></div><div class="sect2" title="What's In a Name?"><div class="titlepage"><div><div><h3 class="title"><a id="svn.tour.importing.naming"></a>What's In a Name?</h3></div></div></div><p>Subversion tries hard not to limit the type of data you
can place under version control. The contents of files and
property values are stored and transmitted as binary data, and
<a class="xref" href="#svn.advanced.props.special.mime-type" title="File Content Type">the section called “File Content Type”</a>
tells you how to give Subversion a hint that
<span class="quote">“<span class="quote">textual</span>”</span> operations don't make sense for a
particular file. There are a few places, however, where
Subversion places restrictions on information it
stores.</p><p>Subversion internally handles certain bits of
data—for example, property names, pathnames, and log
messages—as UTF-8-encoded Unicode. This is not to say
that all your interactions with Subversion must involve UTF-8,
though. As a general rule, Subversion clients will gracefully
and transparently handle conversions between UTF-8 and the
encoding system in use on your computer, if such a conversion
can meaningfully be done (which is the case for most common
encodings in use today).</p><p>In WebDAV exchanges and older versions of some of
Subversion's administrative files, paths are used as XML
attribute values, and property names in XML tag names. This
means that pathnames can contain only legal XML (1.0)
characters, and properties are further limited to ASCII
characters. Subversion also prohibits <code class="literal">TAB</code>,
<code class="literal">CR</code>, and <code class="literal">LF</code> characters in
path names to prevent paths from being broken up in diffs or
in the output of commands such as <span class="command"><strong>svn log</strong></span>
or <span class="command"><strong>svn status</strong></span>.</p><p>While it may seem like a lot to remember, in practice
these limitations are rarely a problem. As long as your
locale settings are compatible with UTF-8 and you don't use
control characters in path names, you should have no trouble
communicating with Subversion. The command-line client adds
an extra bit of help—to create
<span class="quote">“<span class="quote">legally correct</span>”</span> versions for internal
use it will automatically escape illegal
path characters as needed in URLs that you type.</p></div></div><div class="sect1" title="Creating a Working Copy"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.tour.initial"></a>Creating a Working Copy</h2></div></div></div><a id="idp6478976" class="indexterm"></a><p>Most of the time, you will start using a Subversion
repository by performing a <em class="firstterm">checkout</em> of
your project. Checking out a directory from a repository
creates a working copy of that directory on your local machine.
Unless otherwise specified, this copy contains the youngest
(that is, most recently created or modified) versions of the
directory and its children found in the Subversion
repository:</p><div class="informalexample"><pre class="screen">
$ svn checkout http://svn.example.com/svn/repo/trunk
A trunk/README
A trunk/INSTALL
A trunk/src/main.c
A trunk/src/header.h
…
Checked out revision 8810.
$
</pre></div><p>Although the preceding example checks out the trunk
directory, you can just as easily check out a deeper
subdirectory of a repository by specifying that subdirectory's
URL as the checkout URL:</p><div class="informalexample"><pre class="screen">
$ svn checkout http://svn.example.com/svn/repo/trunk/src
A src/main.c
A src/header.h
A src/lib/helpers.c
…
Checked out revision 8810.
$
</pre></div><p>Since Subversion uses a copy-modify-merge model instead of
lock-modify-unlock (see <a class="xref" href="#svn.basic.vsn-models" title="Versioning Models">the section called “Versioning Models”</a>),
you can immediately make changes to the files and directories in
your working copy. Your working copy is just like any other
collection of files and directories on your system. You can
edit the files inside it, rename it, even delete the entire
working copy and forget about it.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>While your working copy is <span class="quote">“<span class="quote">just like any other
collection of files and directories on your system,</span>”</span>
you can edit files at will, but you must tell Subversion
about <span class="emphasis"><em>everything else</em></span> that you do. For
example, if you want to copy or move an item in a working
copy, you should use <span class="command"><strong>svn copy</strong></span> or
<span class="command"><strong>svn move</strong></span> instead of the copy and move
commands provided by your operating system. We'll talk more
about them later in this chapter.</p></td></tr></table></div><p>Unless you're ready to commit the addition of a new file or
directory or changes to existing ones, there's no need to
further notify the Subversion server that you've done
anything.</p><div class="sidebar" title="What Is This .svn Directory?"><div class="titlepage"><div><div><p class="title"><strong>What Is This .svn Directory?</strong></p></div></div></div><p>The topmost directory of a working copy—and prior to
version 1.7, every versioned subdirectory
thereof—contains a special administrative subdirectory
named <code class="filename">.svn</code>. Usually, your operating
system's directory listing commands won't show this
subdirectory, but it is nevertheless an important directory.
Whatever you do, don't delete or change anything in the
administrative area! Subversion uses that directory and its
contents to manage your working copy.</p></div><p>Notice that in the previous pair of examples, Subversion
chose to create a working copy in a directory named for the
final component of the checkout URL. This occurs only as a
convenience to the user when the checkout URL is the only bit of
information provided to the <span class="command"><strong>svn checkout</strong></span>
command. Subversion's command-line client gives you additional
flexibility, though, allowing you to optionally specify the
local directory name that Subversion should use for the working
copy it creates. For example:</p><div class="informalexample"><pre class="screen">
$ svn checkout http://svn.example.com/svn/repo/trunk my-working-copy
A my-working-copy/README
A my-working-copy/INSTALL
A my-working-copy/src/main.c
A my-working-copy/src/header.h
…
Checked out revision 8810.
$
</pre></div><p>If the local directory you specify doesn't yet
exist, that's okay—<span class="command"><strong>svn checkout</strong></span> will
create it for you.</p></div><div class="sect1" title="Basic Work Cycle"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.tour.cycle"></a>Basic Work Cycle</h2></div></div></div><p>Subversion has numerous features, options, bells, and
whistles, but on a day-to-day basis, odds are that you will use
only a few of them. In this section, we'll run through the most
common things that you might find yourself doing with Subversion
in the course of a day's work.</p><p>The typical work cycle looks like this:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p><span class="emphasis"><em>Update your working copy.</em></span> This
involves the use of the <span class="command"><strong>svn update</strong></span>
command.</p></li><li class="listitem"><p><span class="emphasis"><em>Make your changes.</em></span> The most common
changes that you'll make are edits to the contents of your
existing files. But sometimes you need to add, remove, copy
and move files and directories—the <span class="command"><strong>svn
add</strong></span>, <span class="command"><strong>svn delete</strong></span>, <span class="command"><strong>svn
copy</strong></span>, and <span class="command"><strong>svn move</strong></span> commands
handle those sorts of structural changes within the working
copy.</p></li><li class="listitem"><p><span class="emphasis"><em>Review your changes.</em></span>
The <span class="command"><strong>svn status</strong></span> and <span class="command"><strong>svn
diff</strong></span> commands are critical to reviewing the
changes you've made in your working copy.</p></li><li class="listitem"><p><span class="emphasis"><em>Fix your mistakes.</em></span> Nobody's
perfect, so as you review your changes, you may spot
something that's not quite right. Sometimes the easiest way
to fix a mistake is start all over again from scratch.
The <span class="command"><strong>svn revert</strong></span> command restores a file
or directory to its unmodified state.</p></li><li class="listitem"><p><span class="emphasis"><em>Resolve any conflicts (merge others'
changes).</em></span> In the time it takes you to make and
review your changes, others might have made and published
changes, too. You'll want to integrate their changes into
your working copy to avoid the potential out-of-dateness
scenarios when you attempt to publish your own. Again,
the <span class="command"><strong>svn update</strong></span> command is the way to do
this. If this results in local conflicts, you'll need to
resolve those using the <span class="command"><strong>svn resolve</strong></span>
command.</p></li><li class="listitem"><p><span class="emphasis"><em>Publish (commit) your changes.</em></span>
The <span class="command"><strong>svn commit</strong></span> command transmits your
changes to the repository where, if they are accepted, they
create the newest versions of all the things you modified.
Now others can see your work, too!</p></li></ol></div><div class="sect2" title="Update Your Working Copy"><div class="titlepage"><div><div><h3 class="title"><a id="svn.tour.cycle.update"></a>Update Your Working Copy</h3></div></div></div><a id="idp6516192" class="indexterm"></a><p>When working on a project that is being modified via
multiple working copies, you'll want to update your working
copy to receive any changes committed from other working
copies since your last update. These might be changes that
other members of your project team have made, or they might
simply be changes you've made yourself from a different
computer. To protect your data, Subversion won't
allow you commit new changes to out-of-date files and
directories, so it's best to have the latest versions
of all your project's files and directories before making new
changes of your own.</p><p>Use <span class="command"><strong>svn update</strong></span> to bring your working
copy into sync with the latest revision in the
repository:</p><div class="informalexample"><pre class="screen">
$ svn update
Updating '.':
U foo.c
U bar.c
Updated to revision 2.
$
</pre></div><p>In this case, it appears that someone checked in
modifications to both <code class="filename">foo.c</code>
and <code class="filename">bar.c</code> since the last time you
updated, and Subversion has updated your working copy to
include those changes.</p><p>When the server sends changes to your working copy via
<span class="command"><strong>svn update</strong></span>, a letter code is displayed next
to each item to let you know what actions Subversion performed
to bring your working copy up to date. To find out what these
letters mean, run <strong class="userinput"><code>svn help update</code></strong> or
see <a class="xref" href="#svn.ref.svn.c.update" title="svn update (up)">svn update (up)</a> in
<a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a>.</p></div><div class="sect2" title="Make Your Changes"><div class="titlepage"><div><div><h3 class="title"><a id="svn.tour.cycle.edit"></a>Make Your Changes </h3></div></div></div><p>Now you can get to work and make changes in your working
copy. You can make two kinds of changes to your working
copy: <em class="firstterm">file changes</em> and <em class="firstterm">tree
changes</em>. You don't need to tell Subversion that
you intend to change a file; just make your changes using your
text editor, word processor, graphics program, or whatever
tool you would normally use. Subversion automatically detects
which files have been changed, and in addition, it handles
binary files just as easily as it handles text files—and
just as efficiently, too. Tree changes are different, and
involve changes to a directory's structure. Such changes
include adding and removing files, renaming files or
directories, and copying files or directories to new
locations. For tree changes, you use Subversion operations
to <span class="quote">“<span class="quote">schedule</span>”</span> files and directories for removal,
addition, copying, or moving. These changes may take place
immediately in your working copy, but no additions or removals
will happen in the repository until you commit them.</p><div class="sidebar" title="Versioning Symbolic Links"><div class="titlepage"><div><div><p class="title"><strong>Versioning Symbolic Links</strong></p></div></div></div><p>On non-Windows platforms, Subversion is able to version
files of the special type <em class="firstterm">symbolic
link</em> (or <span class="quote">“<span class="quote">symlink</span>”</span>). A symlink is
a file that acts as a sort of transparent reference to some
other object in the filesystem, allowing programs to read
and write to those objects indirectly by performing
operations on the symlink itself.</p><p>When a symlink is committed into a Subversion
repository, Subversion remembers that the file was in fact a
symlink, as well as the object to which the symlink
<span class="quote">“<span class="quote">points.</span>”</span> When that symlink is checked out to
another working copy on a non-Windows system, Subversion
reconstructs a real filesystem-level symbolic link from the
versioned symlink. But that doesn't in any way limit the
usability of working copies on systems such as Windows that
do not support symlinks. On such systems, Subversion simply
creates a regular text file whose contents are the path to
which the original symlink pointed. While that file
can't be used as a symlink on a Windows system, it also
won't prevent Windows users from performing their other
Subversion-related activities.</p></div><p>Here is an overview of the five Subversion subcommands
that you'll use most often to make tree changes:</p><div class="variablelist"><dl><dt><span class="term"><strong class="userinput"><code>svn add FOO</code></strong></span></dt><dd><a id="idp6537472" class="indexterm"></a><p>Use this to schedule the file, directory, or
symbolic link <code class="filename">FOO</code> to be added to
the repository. When you next
commit, <code class="filename">FOO</code> will become a child of
its parent directory. Note that if
<code class="filename">FOO</code> is a directory, everything
underneath <code class="filename">FOO</code> will be scheduled
for addition. If you want only to add
<code class="filename">FOO</code> itself, pass the
<code class="option">--depth=empty</code> option.</p></dd><dt><span class="term"><strong class="userinput"><code>svn delete FOO</code></strong></span></dt><dd><a id="idp6545056" class="indexterm"></a><p>Use this to schedule the file, directory, or
symbolic link <code class="filename">FOO</code> to be deleted
from the repository. If <code class="filename">FOO</code> is a
file or link, it is immediately deleted from your
working copy. If <code class="filename">FOO</code> is a
directory, it is not deleted, but Subversion schedules
it for deletion. When you commit your
changes, <code class="filename">FOO</code> will be entirely
removed from your working copy and the
repository.<sup>[<a id="idp6550064" href="#ftn.idp6550064" class="footnote">6</a>]</sup></p></dd><dt><span class="term"><strong class="userinput"><code>svn copy FOO BAR</code></strong></span></dt><dd><a id="idp6554448" class="indexterm"></a><p>Create a new item <code class="filename">BAR</code> as a
duplicate of <code class="filename">FOO</code> and automatically
schedule <code class="filename">BAR</code> for addition. When
<code class="filename">BAR</code> is added to the repository on
the next commit, its copy history is recorded (as having
originally come from <code class="filename">FOO</code>).
<span class="command"><strong>svn copy</strong></span> does not create intermediate
directories unless you pass the
<code class="option">--parents</code> option.</p></dd><dt><span class="term"><strong class="userinput"><code>svn move FOO BAR</code></strong></span></dt><dd><a id="idp6562528" class="indexterm"></a><p>This command is exactly the same as running
<strong class="userinput"><code>svn copy FOO BAR; svn delete FOO</code></strong>.
That is, <code class="filename">BAR</code> is scheduled for
addition as a copy of <code class="filename">FOO</code>, and
<code class="filename">FOO</code> is scheduled for removal.
<span class="command"><strong>svn move</strong></span> does not create intermediate
directories unless you pass the
<code class="option">--parents</code> option.</p></dd><dt><span class="term"><strong class="userinput"><code>svn mkdir FOO</code></strong></span></dt><dd><a id="idp6569824" class="indexterm"></a><p>This command is exactly the same as running
<strong class="userinput"><code>mkdir FOO; svn add FOO</code></strong>. That is,
a new directory named <code class="filename">FOO</code> is
created and scheduled for addition.</p></dd></dl></div><div class="sidebar" title="Changing the Repository Without a Working Copy"><div class="titlepage"><div><div><p class="title"><strong>Changing the Repository Without a Working Copy</strong></p></div></div></div><p>Subversion <span class="emphasis"><em>does</em></span> offer ways to
immediately commit tree changes to the repository without an
explicit commit action. In particular, specific uses
of <span class="command"><strong>svn mkdir</strong></span>, <span class="command"><strong>svn
copy</strong></span>, <span class="command"><strong>svn move</strong></span>, and
<span class="command"><strong>svn delete</strong></span> can operate directly on
repository URLs as well as on working copy paths. Of
course, as previously mentioned, <span class="command"><strong>svn
import</strong></span> always makes direct changes to the
repository.</p><p>There are pros and cons to performing URL-based
operations. One obvious advantage to doing so is speed:
sometimes, checking out a working copy that you don't
already have solely to perform some seemingly simple action
is an overbearing cost. A disadvantage is that you are
generally limited to a single, or single type of, operation
at a time when operating directly on URLs. Finally, the
primary advantage of a working copy is in its utility as a
sort of <span class="quote">“<span class="quote">staging area</span>”</span> for changes. You can
make sure that the changes you are about to commit make
sense in the larger scope of your project before committing
them. And, of course, these staged changes can be as
complex or as a simple as they need to be, yet result in but
a single new revision when committed.</p></div></div><div class="sect2" title="Review Your Changes"><div class="titlepage"><div><div><h3 class="title"><a id="svn.tour.cycle.examine"></a>Review Your Changes</h3></div></div></div><a id="idp6581776" class="indexterm"></a><p>Once you've finished making changes, you need to commit
them to the repository, but before you do so, it's usually a
good idea to take a look at exactly what you've changed. By
examining your changes before you commit, you can compose a
more accurate <em class="firstterm">log message</em> (a
human-readable description of the committed changes stored
alongside those changes in the repository). You may also
discover that you've inadvertently changed a file, and that
you need to undo that change before committing. Additionally,
this is a good opportunity to review and scrutinize changes
before publishing them. You can see an overview of the
changes you've made by using the <span class="command"><strong>svn status</strong></span>
command, and you can dig into the details of those changes by
using the <span class="command"><strong>svn diff</strong></span> command.</p><div class="sidebar" title="Look Ma! No Network!"><div class="titlepage"><div><div><p class="title"><strong>Look Ma! No Network!</strong></p></div></div></div><p>You can use the commands <span class="command"><strong>svn status</strong></span>,
<span class="command"><strong>svn diff</strong></span>, and <span class="command"><strong>svn
revert</strong></span> without any network access even if your
repository <span class="emphasis"><em>is</em></span> across the network. This
makes it easy to manage and review your changes-in-progress
when you are working offline or are otherwise unable to
contact your repository over the network.</p><a id="idp6589184" class="indexterm"></a><a id="idp6590192" class="indexterm"></a><p>Subversion does this by keeping private caches of
pristine, unmodified versions of each versioned file inside
its working copy administrative area (or prior to version 1.7,
potentially multiple administrative areas). This allows
Subversion to report—and revert—local
modifications to those files <span class="emphasis"><em>without network
access</em></span>. This cache (called the
<em class="firstterm">text-base</em>) also allows Subversion to
send the user's local modifications during a commit to the
server as a compressed <em class="firstterm">delta</em> (or
<span class="quote">“<span class="quote">difference</span>”</span>) against the pristine version.
Having this cache is a tremendous benefit—even if you
have a fast Internet connection, it's generally much faster
to send only a file's changes rather than the whole file to
the server.</p></div><div class="sect3" title="See an overview of your changes"><div class="titlepage"><div><div><h4 class="title"><a id="svn.tour.cycle.examine.status"></a>See an overview of your changes</h4></div></div></div><a id="idp6595392" class="indexterm"></a><p>To get an overview of your changes, use the
<span class="command"><strong>svn status</strong></span> command. You'll probably use
<span class="command"><strong>svn status</strong></span> more than any other Subversion
command.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>Because the <span class="command"><strong>cvs status</strong></span> command's
output was so noisy, and because <span class="command"><strong>cvs
update</strong></span> not only performs an update, but also
reports the status of your local changes, most CVS users
have grown accustomed to using <span class="command"><strong>cvs
update</strong></span> to report their changes. In Subversion,
the update and status reporting facilities are completely
separate. See
<a class="xref" href="#svn.forcvs.status-vs-update" title="Distinction Between Status and Update">the section called “Distinction Between Status and Update”</a> for more
details.</p></td></tr></table></div><p>If you run <strong class="userinput"><code>svn status</code></strong> at the top
of your working copy with no additional arguments, it will
detect and report all file and tree changes you've
made.</p><div class="informalexample"><pre class="screen">
$ svn status
? scratch.c
A stuff/loot
A stuff/loot/new.c
D stuff/old.c
M bar.c
$
</pre></div><p>In its default output mode, <span class="command"><strong>svn
status</strong></span> prints seven columns of characters,
followed by several whitespace characters, followed by a
file or directory name. The first column tells the status
of a file or directory and/or its contents. Some of the
most common codes that <span class="command"><strong>svn status</strong></span>
displays are:</p><div class="variablelist"><dl><dt><span class="term"><code class="computeroutput">? item</code></span></dt><dd><p>The file, directory, or symbolic link
<code class="filename">item</code> is not under version
control.</p></dd><dt><span class="term"><code class="computeroutput">A item</code></span></dt><dd><p>The file, directory, or symbolic link
<code class="filename">item</code> has been scheduled for
addition into the repository.</p></dd><dt><span class="term"><code class="computeroutput">C item</code></span></dt><dd><p>The file <code class="filename">item</code> is in a state
of conflict. That is, changes received from the
server during an update overlap with local changes
that you have in your working copy (and weren't
resolved during the update). You must resolve this
conflict before committing your changes to the
repository.</p></dd><dt><span class="term"><code class="computeroutput">D item</code></span></dt><dd><p>The file, directory, or symbolic link
<code class="filename">item</code> has been scheduled for
deletion from the repository.</p></dd><dt><span class="term"><code class="computeroutput">M item</code></span></dt><dd><p>The contents of the file <code class="filename">item</code>
have been modified.</p></dd></dl></div><p>If you pass a specific path to <span class="command"><strong>svn
status</strong></span>, you get information about that item
alone:</p><div class="informalexample"><pre class="screen">
$ svn status stuff/fish.c
D stuff/fish.c
</pre></div><p><span class="command"><strong>svn status</strong></span> also has a
<code class="option">--verbose</code> (<code class="option">-v</code>) option,
which will show you the status of <span class="emphasis"><em>every</em></span>
item in your working copy, even if it has not been
changed:</p><div class="informalexample"><pre class="screen">
$ svn status -v
M 44 23 sally README
44 30 sally INSTALL
M 44 20 harry bar.c
44 18 ira stuff
44 35 harry stuff/trout.c
D 44 19 ira stuff/fish.c
44 21 sally stuff/things
A 0 ? ? stuff/things/bloo.h
44 36 harry stuff/things/gloo.c
</pre></div><p>This is the <span class="quote">“<span class="quote">long form</span>”</span> output of
<span class="command"><strong>svn status</strong></span>. The letters in the first
column mean the same as before, but the second column shows
the working revision of the item. The third and fourth
columns show the revision in which the item last changed,
and who changed it.</p><p>None of the prior invocations to <span class="command"><strong>svn
status</strong></span> contact the repository—they merely
report what is known about the working copy items based on
the records stored in the working copy administrative area
and on the timestamps and contents of modified files. But
sometimes it is useful to see which of the items in your
working copy have been modified in the repository since the
last time you updated your working copy. For
this, <span class="command"><strong>svn status</strong></span> offers the
<code class="option">--show-updates</code> (<code class="option">-u</code>)
option, which contacts the repository and adds information
about items that are out of date:</p><div class="informalexample"><pre class="screen">
$ svn status -u -v
M * 44 23 sally README
M 44 20 harry bar.c
* 44 35 harry stuff/trout.c
D 44 19 ira stuff/fish.c
A 0 ? ? stuff/things/bloo.h
Status against revision: 46
</pre></div><p>Notice in the previous example the two asterisks: if you
were to run <strong class="userinput"><code>svn update</code></strong> at this point,
you would receive changes to <code class="filename">README</code>
and <code class="filename">trout.c</code>. This tells you some very
useful information—because one of those items is also
one that you have locally modified (the
file <code class="filename">README</code>), you'll need to update and
get the servers changes for that file before you commit, or
the repository will reject your commit for being out of
date. We discuss this in more detail later.</p><p><span class="command"><strong>svn status</strong></span> can display much more
information about the files and directories in your working
copy than we've shown here—for an exhaustive
description of <span class="command"><strong>svn status</strong></span> and its output,
run <strong class="userinput"><code>svn help status</code></strong> or see
<a class="xref" href="#svn.ref.svn.c.status" title="svn status (stat, st)">svn status (stat, st)</a> in
<a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a>.</p></div><div class="sect3" title="Examine the details of your local modifications"><div class="titlepage"><div><div><h4 class="title"><a id="svn.tour.cycle.examine.diff"></a>Examine the details of your local modifications</h4></div></div></div><a id="idp6639792" class="indexterm"></a><a id="idp6641568" class="indexterm"></a><p>Another way to examine your changes is with the
<span class="command"><strong>svn diff</strong></span> command, which displays
differences in file content. When you run <strong class="userinput"><code>svn
diff</code></strong> at the top of your working copy with no
arguments, Subversion will print the changes you've made to
human-readable files in your working copy. It displays
those changes in <em class="firstterm">unified diff</em> format,
a format which describes changes as <span class="quote">“<span class="quote">hunks</span>”</span>
(or <span class="quote">“<span class="quote">snippets</span>”</span>) of a file's content where each
line of text is prefixed with a single-character code: a
space, which means the line was unchanged; a minus sign
(<code class="literal">-</code>), which means the line was removed
from the file; or a plus sign (<code class="literal">+</code>), which
means the line was added to the file. In the context
of <span class="command"><strong>svn diff</strong></span>, those minus-sign- and
plus-sign-prefixed lines show how the lines looked before
and after your modifications, respectively.</p><p>Here's an example:</p><div class="informalexample"><pre class="screen">
$ svn diff
Index: bar.c
===================================================================
--- bar.c (revision 3)
+++ bar.c (working copy)
@@ -1,7 +1,12 @@
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <unistd.h>
+
+#include <stdio.h>
int main(void) {
- printf("Sixty-four slices of American Cheese...\n");
+ printf("Sixty-five slices of American Cheese...\n");
return 0;
}
Index: README
===================================================================
--- README (revision 3)
+++ README (working copy)
@@ -193,3 +193,4 @@
+Note to self: pick up laundry.
Index: stuff/fish.c
===================================================================
--- stuff/fish.c (revision 1)
+++ stuff/fish.c (working copy)
-Welcome to the file known as 'fish'.
-Information on fish will be here soon.
Index: stuff/things/bloo.h
===================================================================
--- stuff/things/bloo.h (revision 8)
+++ stuff/things/bloo.h (working copy)
+Here is a new file to describe
+things about bloo.
</pre></div><a id="idp6372432" class="indexterm"></a><p>The <span class="command"><strong>svn diff</strong></span> command produces this
output by comparing your working files against its pristine
text-base. Files scheduled for addition are displayed as
files in which every line was added; files scheduled for
deletion are displayed as if every line was removed from
those files. The output from <span class="command"><strong>svn diff</strong></span> is
somehwat compatible with the <span class="command"><strong>patch</strong></span>
program—more so with the <span class="command"><strong>svn patch</strong></span>
subcommand introduced in Subversion 1.7. Patch processing
commands such as these read and apply <em class="firstterm">patch
files</em> (or <span class="quote">“<span class="quote">patches</span>”</span>), which are
files that describe differences made to one or more files.
Because of this, you can share the changes you've made in
your working copy with someone else without first committing
those changes by creating a patch file from the redirected
output of <span class="command"><strong>svn diff</strong></span>:</p><div class="informalexample"><pre class="screen">
$ svn diff > patchfile
$
</pre></div><p>Subversion uses its internal diff engine, which produces
unified diff format, by default. If you want diff output in
a different format, specify an external diff program using
<code class="option">--diff-cmd</code> and pass any additional flags
that it needs via the <code class="option">--extensions</code>
(<code class="option">-x</code>) option. For example, you might want
Subversion to defer its difference calculation and display
to the GNU <span class="command"><strong>diff</strong></span> program, asking that
program to print local modifications made to the
file <code class="filename">foo.c</code> in context diff format
(another flavor of difference format) while ignoring changes
made only to the case of the letters used in the file's
contents:</p><div class="informalexample"><pre class="screen">
$ svn diff --diff-cmd /usr/bin/diff -x "-i" foo.c
…
$
</pre></div></div></div><div class="sect2" title="Fix Your Mistakes"><div class="titlepage"><div><div><h3 class="title"><a id="svn.tour.cycle.revert"></a>Fix Your Mistakes</h3></div></div></div><p>Suppose while viewing the output of <span class="command"><strong>svn
diff</strong></span> you determine that all the changes you made to
a particular file are mistakes. Maybe you shouldn't have
changed the file at all, or perhaps it would be easier to make
different changes starting from scratch. You could edit the
file again and unmake all those changes. You could try to
find a copy of how the file looked before you changed it, and
then copy its contents atop your modified version. You
could attempt to apply those changes to the file again in
reverse using <strong class="userinput"><code>patch -R</code></strong>. And there are
probably other approaches you could take.</p><a id="idp6676144" class="indexterm"></a><p>Fortunately in Subversion, undoing your work and starting
over from scratch doesn't require such acrobatics. Just use
the <span class="command"><strong>svn revert</strong></span> command:</p><div class="informalexample"><pre class="screen">
$ svn status README
M README
$ svn revert README
Reverted 'README'
$ svn status README
$
</pre></div><p>In this example, Subversion has reverted the file to its
premodified state by overwriting it with the pristine version
of the file cached in the text-base area. But note that
<span class="command"><strong>svn revert</strong></span> can undo
<span class="emphasis"><em>any</em></span> scheduled operation—for
example, you might decide that you don't want to add a new
file after all:</p><div class="informalexample"><pre class="screen">
$ svn status new-file.txt
? new-file.txt
$ svn add new-file.txt
A new-file.txt
$ svn revert new-file.txt
Reverted 'new-file.txt'
$ svn status new-file.txt
? new-file.txt
$
</pre></div><p>Or perhaps you mistakenly removed a file from version
control:</p><div class="informalexample"><pre class="screen">
$ svn status README
$ svn delete README
D README
$ svn revert README
Reverted 'README'
$ svn status README
$
</pre></div><p>The <span class="command"><strong>svn revert</strong></span> command offers salvation
for imperfect people. It can save you huge amounts of time
and energy that would otherwise be spent manually unmaking
changes or, worse, disposing of your working copy and checking
out a fresh one just to have a clean slate to work with
again.</p></div><div class="sect2" title="Resolve Any Conflicts"><div class="titlepage"><div><div><h3 class="title"><a id="svn.tour.cycle.resolve"></a>Resolve Any Conflicts</h3></div></div></div><a id="idp6687520" class="indexterm"></a><p>We've already seen how <strong class="userinput"><code>svn status
-u</code></strong> can predict conflicts, but dealing with those
conflicts is still something that remains to be done.
Conflicts can occur any time you attempt to merge or integrate
(in a very general sense) changes from the repository into
your working copy. By now you know that <span class="command"><strong>svn
update</strong></span> creates exactly that sort of
scenario—that command's very purpose is to bring your
working copy up to date with the repository by merging all the
changes made since your last update into your working
copy. So how does Subversion report these conflicts to you,
and how do you deal with them?</p><p>Suppose you run <strong class="userinput"><code>svn update</code></strong> and you
see this sort of interesting output:</p><div class="informalexample"><pre class="screen">
$ svn update
Updating '.':
U INSTALL
G README
Conflict discovered in 'bar.c'.
Select: (p) postpone, (df) diff-full, (e) edit,
(mc) mine-conflict, (tc) theirs-conflict,
(s) show all options:
</pre></div><p>The <code class="computeroutput">U</code> (which stands for
<span class="quote">“<span class="quote">Updated</span>”</span>) and <code class="computeroutput">G</code>
(for <span class="quote">“<span class="quote">merGed</span>”</span>) codes are no cause for concern;
those files cleanly absorbed changes from the repository. A
file marked with <code class="computeroutput">U</code> contains
no local changes but was updated with changes from the
repository. One marked with
<code class="computeroutput">G</code> had local changes to begin
with, but the changes coming from the repository didn't
overlap with those local changes.</p><p>It's the next few lines which are interesting. First,
Subversion reports to you that in its attempt to merge
outstanding server changes into the
file <code class="filename">bar.c</code>, it has detected that some of
those changes clash with local modifications you've made to
that file in your working copy but have not yet committed.
Perhaps someone has changed the same line of text you also
changed. Whatever the reason, Subversion instantly flags this
file as being in a state of conflict. It then asks you what
you want to do about the problem, allowing you to
interactively choose an action to take toward resolving the
conflict. The most commonly used options are displayed, but
you can see all of the options by
typing <em class="replaceable"><code>s</code></em>:</p><div class="informalexample"><pre class="screen">
…
Select: (p) postpone, (df) diff-full, (e) edit,
(mc) mine-conflict, (tc) theirs-conflict,
(s) show all options: s
(e) edit - change merged file in an editor
(df) diff-full - show all changes made to merged file
(r) resolved - accept merged version of file
(dc) display-conflict - show all conflicts (ignoring merged version)
(mc) mine-conflict - accept my version for all conflicts (same)
(tc) theirs-conflict - accept their version for all conflicts (same)
(mf) mine-full - accept my version of entire file (even non-conflicts)
(tf) theirs-full - accept their version of entire file (same)
(p) postpone - mark the conflict to be resolved later
(l) launch - launch external tool to resolve conflict
(s) show all - show this list
Select: (p) postpone, (df) diff-full, (e) edit,
(mc) mine-conflict, (tc) theirs-conflict,
(s) show all options:
</pre></div><p>Let's briefly review each of these options before we go
into detail on what each option means.</p><div class="variablelist"><dl><dt><span class="term"><code class="computeroutput">(e) edit</code></span></dt><dd><p>Open the file in conflict with your favorite editor,
as set in the environment variable
<code class="literal">EDITOR</code>.</p></dd><dt><span class="term"><code class="computeroutput">(df) diff-full</code></span></dt><dd><p>Display the differences between the base revision
and the conflicted file itself in unified diff format.</p></dd><dt><span class="term"><code class="computeroutput">(r) resolved</code></span></dt><dd><p>After editing a file, tell
<span class="command"><strong>svn</strong></span> that you've resolved the
conflicts in the file and that it should accept the
current contents—basically that you've
<span class="quote">“<span class="quote">resolved</span>”</span> the conflict.</p></dd><dt><span class="term"><code class="computeroutput">(dc) display-conflict</code></span></dt><dd><p>Display all conflicting regions of the file,
ignoring changes which were successfully merged.</p></dd><dt><span class="term"><code class="computeroutput">(mc) mine-conflict</code></span></dt><dd><p>Discard any newly received changes from the server
which conflict with your local changes to the file under
review. However, accept and merge all non-conflicting
changes received from the server for that file.</p></dd><dt><span class="term"><code class="computeroutput">(tc) theirs-conflict</code></span></dt><dd><p>Discard any local changes which conflict with
incoming changes from the server for the file under
review. However, preserve all non-conflicting local
changes to that file.</p></dd><dt><span class="term"><code class="computeroutput">(mf) mine-full</code></span></dt><dd><p>Discard all newly received changes from the server
for the file under review, but preserve all your local
changes for that file.</p></dd><dt><span class="term"><code class="computeroutput">(tf) theirs-full</code></span></dt><dd><p>Discard all your local changes to the file under
review and use only the newly received changes from the
server for that file.</p></dd><dt><span class="term"><code class="computeroutput">(p) postpone</code></span></dt><dd><p>Leave the file in a conflicted state for you to
resolve after your update is complete.</p></dd><dt><span class="term"><code class="computeroutput">(l) launch</code></span></dt><dd><p>Launch an external program to perform the conflict
resolution. This requires a bit of preparation
beforehand.</p></dd><dt><span class="term"><code class="computeroutput">(s) show all</code></span></dt><dd><p>Show the list of all possible commands you can use
in interactive conflict resolution.</p></dd></dl></div><p>We'll cover these commands in more detail now, grouping
them together by related functionality.</p><div class="sect3" title="Viewing conflict differences interactively"><div class="titlepage"><div><div><h4 class="title"><a id="svn.tour.cycle.resolve.diff"></a>Viewing conflict differences interactively</h4></div></div></div><a id="idp6728640" class="indexterm"></a><p>Before deciding how to attack a conflict interactively,
odds are that you'd like to see exactly what is in conflict.
Two of the commands available at the interactive conflict
resolution prompt can assist you here. The first is
the <span class="quote">“<span class="quote">diff-full</span>”</span> command
(<strong class="userinput"><code>df</code></strong>), which displays all the local
modifications to the file in question plus any conflict
regions:</p><div class="informalexample"><pre class="screen">
…
Select: (p) postpone, (df) diff-full, (e) edit,
(mc) mine-conflict, (tc) theirs-conflict,
(s) show all options: df
--- .svn/text-base/sandwich.txt.svn-base Tue Dec 11 21:33:57 2007
+++ .svn/tmp/tempfile.32.tmp Tue Dec 11 21:34:33 2007
@@ -1 +1,5 @@
-Just buy a sandwich.
+<<<<<<< .mine
+Go pick up a cheesesteak.
+=======
+Bring me a taco!
+>>>>>>> .r32
…
</pre></div><p>The first line of the diff content shows the previous
contents of the working copy (the <code class="literal">BASE</code>
revision), the next content line is your change, and the
last content line is the change that was just received from
the server (<span class="emphasis"><em>usually</em></span> the
<code class="literal">HEAD</code> revision).</p><p>The second command is similar to the first, but
the <span class="quote">“<span class="quote">display-conflict</span>”</span>
(<strong class="userinput"><code>dc</code></strong>) command shows only the conflict
regions, not all the changes made to the file.
Additionally, this command uses a slightly different display
format for the conflict regions which allows you to more
easily compare the file's contents in those regions as they
would appear in each of three states: original and unedited;
with your local changes applied and the server's conflicting
changes ignored; and with only the server's incoming changes
applied and your local, conflicting changes reverted.</p><p>After reviewing the information provided by these
commands, you're ready to move on to the next action.</p></div><div class="sect3" title="Resolving conflict differences interactively"><div class="titlepage"><div><div><h4 class="title"><a id="svn.tour.cycle.resolve.resolve"></a>Resolving conflict differences interactively</h4></div></div></div><a id="idp6739472" class="indexterm"></a><p>There are several different ways to resolve conflicts
interactively—two of which allow you to selectively
merge and edit changes, the rest of which allow you to
simply pick a version of the file and move along.</p><p>If you wish to choose some combination of your local
changes, you can use the <span class="quote">“<span class="quote">edit</span>”</span> command
(<strong class="userinput"><code>e</code></strong>) to manually edit the file with
conflict markers in a text editor (configured per the
instructions in <a class="xref" href="#svn.advanced.externaleditors" title="Using External Editors">the section called “Using External Editors”</a>). After you've edited the file, if you're satisfied with
the changes you've made, you can tell Subversion that the
edited file is no longer in conflict by using
the <span class="quote">“<span class="quote">resolved</span>”</span> command
(<code class="literal">r</code>).</p><p>Regardless of what your local Unix snob will likely tell
you, editing the file by hand in your favorite text editor
is a somewhat low-tech way of remedying conflicts (see
<a class="xref" href="#svn.tour.cycle.resolve.byhand" title="Merging conflicts by hand">the section called “Merging conflicts by hand”</a> for a
walkthrough). For this reason, Subversion provides
the <span class="quote">“<span class="quote">launch</span>”</span> resolution command
(<strong class="userinput"><code>l</code></strong>) to fire up a fancy graphical
merge tool instead (see
<a class="xref" href="#svn.advanced.externaldifftools.merge" title="External merge">the section called “External merge”</a>).</p><p>If you decide that you don't need to merge any changes,
but just want to accept one version of the file or the
other, you can either choose your changes (a.k.a.
<span class="quote">“<span class="quote">mine</span>”</span>) by using the <span class="quote">“<span class="quote">mine-full</span>”</span>
command (<strong class="userinput"><code>mf</code></strong>) or choose theirs by using the
<span class="quote">“<span class="quote">theirs-full</span>”</span> command
(<strong class="userinput"><code>tf</code></strong>).</p><p>Finally, there is also a pair of compromise options
available. The <span class="quote">“<span class="quote">mine-conflict</span>”</span>
(<strong class="userinput"><code>mc</code></strong>)
and <span class="quote">“<span class="quote">theirs-conflict</span>”</span>
(<strong class="userinput"><code>tc</code></strong>) commands instruct Subversion to
select your local changes or the server's incoming changes,
respectively, as the <span class="quote">“<span class="quote">winner</span>”</span> for all conflicts
in the file. But, unlike the <span class="quote">“<span class="quote">mine-full</span>”</span>
and <span class="quote">“<span class="quote">theirs-full</span>”</span> commands, these commands
preserve both your local changes and changes received from
the server in regions of the file where no conflict was
detected.</p></div><div class="sect3" title="Postponing conflict resolution"><div class="titlepage"><div><div><h4 class="title"><a id="svn.tour.cycle.resolve.pending"></a>Postponing conflict resolution</h4></div></div></div><a id="idp6756448" class="indexterm"></a><p>This may sound like an appropriate section for avoiding
marital disagreements, but it's actually still about
Subversion, so read on. If you're doing an update and
encounter a conflict that you're not prepared to review or
resolve, you can type <strong class="userinput"><code>p</code></strong> to postpone
resolving a conflict on a file-by-file basis when you run
<strong class="userinput"><code>svn update</code></strong>. If you know in advance
that you don't want to resolve any conflicts interactively,
you can pass the <code class="option">--non-interactive</code> option
to <span class="command"><strong>svn update</strong></span>, and any file in conflict
will be marked with a <code class="computeroutput">C</code>
automatically.</p><p>The <code class="computeroutput">C</code>
(for <span class="quote">“<span class="quote">Conflicted</span>”</span>) means that the changes from the
server overlapped with your own, and now you have to
manually choose between them after the update has completed.
When you postpone a conflict resolution,
<span class="command"><strong>svn</strong></span> typically does three things to assist
you in noticing and resolving that conflict:</p><div class="itemizedlist"><a id="idp6764752" class="indexterm"></a><ul class="itemizedlist" type="disc"><li class="listitem"><p>Subversion prints a <code class="computeroutput">C</code>
during the update and remembers that the file is in a
state of conflict.</p></li><li class="listitem"><p>If Subversion considers the file to be mergeable, it
places <em class="firstterm">conflict
markers</em>—special strings of text that
delimit the <span class="quote">“<span class="quote">sides</span>”</span> of the
conflict—into the file to visibly demonstrate the
overlapping areas. (Subversion uses the
<code class="literal">svn:mime-type</code> property to decide whether a
file is capable of contextual, line-based merging. See
<a class="xref" href="#svn.advanced.props.special.mime-type" title="File Content Type">the section called “File Content Type”</a>
to learn more.)</p></li><li class="listitem"><p>For every conflicted file, Subversion places three
extra unversioned files in your working copy:</p><div class="variablelist"><dl><dt><span class="term"><code class="filename">filename.mine</code></span></dt><dd><p>This is the file as it existed in your working
copy before you began the update process. It
contains any local modifications you had made to
the file up to that point. (If Subversion
considers the file to be unmergeable,
the <code class="filename">.mine</code> file isn't created,
since it would be identical to the working
file.)</p></dd><dt><span class="term"><code class="filename">filename.r<em class="replaceable"><code>OLDREV</code></em>
</code></span></dt><dd><p>This is the file as it existed in the
<code class="literal">BASE</code> revision—that is,
the unmodified revision of the file in your
working copy <span class="emphasis"><em>before</em></span> you began
the update process—where
<em class="replaceable"><code>OLDREV</code></em> is that base
revision number.</p></dd><dt><span class="term"><code class="filename">filename.r<em class="replaceable"><code>NEWREV</code></em>
</code></span></dt><dd><p>This is the file that your Subversion client
just received from the server via the update of
your working copy, where
<em class="replaceable"><code>NEWREV</code></em> corresponds to
the revision number to which you were updating
(<code class="literal">HEAD</code>, unless otherwise
requested).</p></dd></dl></div></li></ul></div><p>For example, Sally makes changes to the file
<code class="filename">sandwich.txt</code>, but does not yet commit
those changes. Meanwhile, Harry commits changes to that
same file. Sally updates her working copy before committing
and she gets a conflict, which she postpones:</p><div class="informalexample"><pre class="screen">
$ svn update
Updating '.':
Conflict discovered in 'sandwich.txt'.
Select: (p) postpone, (df) diff-full, (e) edit,
(mc) mine-conflict, (tc) theirs-conflict,
(s) show all options: p
C sandwich.txt
Updated to revision 2.
Summary of conflicts:
Text conflicts: 1
$ ls -1
sandwich.txt
sandwich.txt.mine
sandwich.txt.r1
sandwich.txt.r2
</pre></div><p>At this point, Subversion will <span class="emphasis"><em>not</em></span>
allow Sally to commit the file
<code class="filename">sandwich.txt</code> until the three temporary
files are removed:</p><div class="informalexample"><pre class="screen">
$ svn commit -m "Add a few more things"
svn: E155015: Commit failed (details follow):
svn: E155015: Aborting commit: '/home/sally/svn-work/sandwich.txt' remains in conflict
</pre></div><p>If you've postponed a conflict, you need to resolve the
conflict before Subversion will allow you to commit your
changes. You'll do this with the <span class="command"><strong>svn
resolve</strong></span> command and one of several arguments to
the <code class="option">--accept</code> option.</p><p>If you want to choose the version of the file that you
last checked out before making your edits, choose
the <em class="replaceable"><code>base</code></em> argument.</p><p>If you want to choose the version that contains only
your edits, choose the <em class="replaceable"><code>mine-full</code></em>
argument.</p><p>If you want to choose the version that your most recent
update pulled from the server (and thus discarding your
edits entirely), choose
the <em class="replaceable"><code>theirs-full</code></em> argument.</p><p>However, if you want to pick and choose from your
changes and the changes that your update fetched from the
server, merge the conflicted text <span class="quote">“<span class="quote">by hand</span>”</span> (by
examining and editing the conflict markers within the file)
and then choose the <em class="replaceable"><code>working</code></em>
argument.</p><p><span class="command"><strong>svn resolve</strong></span> removes the three
temporary files and accepts the version of the file that you
specified with the <code class="option">--accept</code> option, and
Subversion no longer considers the file to be in a state of
conflict:</p><div class="informalexample"><pre class="screen">
$ svn resolve --accept working sandwich.txt
Resolved conflicted state of 'sandwich.txt'
</pre></div></div><div class="sect3" title="Merging conflicts by hand"><div class="titlepage"><div><div><h4 class="title"><a id="svn.tour.cycle.resolve.byhand"></a>Merging conflicts by hand</h4></div></div></div><a id="idp6797792" class="indexterm"></a><p>Merging conflicts by hand can be quite intimidating the
first time you attempt it, but with a little practice, it
can become as easy as falling off a bike.</p><p>Here's an example. Due to a miscommunication, you and
Sally, your collaborator, both edit the file
<code class="filename">sandwich.txt</code> at the same time. Sally
commits her changes, and when you go to update your working
copy, you get a conflict and you're going to have to edit
<code class="filename">sandwich.txt</code> to resolve the conflict.
First, let's take a look at the file:</p><div class="informalexample"><pre class="screen">
$ cat sandwich.txt
Top piece of bread
Mayonnaise
Lettuce
Tomato
Provolone
<<<<<<< .mine
Salami
Mortadella
Prosciutto
=======
Sauerkraut
Grilled Chicken
>>>>>>> .r2
Creole Mustard
Bottom piece of bread
</pre></div><p>The strings of less-than signs, equals signs, and
greater-than signs are conflict markers and are not part of
the actual data in conflict. You generally want to ensure
that those are removed from the file before your next
commit. The text between the first two sets of markers is
composed of the changes you made in the conflicting
area:</p><div class="informalexample"><pre class="screen">
<<<<<<< .mine
Salami
Mortadella
Prosciutto
=======
</pre></div><p>The text between the second and third sets of conflict
markers is the text from Sally's commit:</p><div class="informalexample"><pre class="screen">
=======
Sauerkraut
Grilled Chicken
>>>>>>> .r2
</pre></div><p>Usually you won't want to just delete the conflict
markers and Sally's changes—she's going to be awfully
surprised when the sandwich arrives and it's not what she
wanted. This is where you pick up the phone or walk across
the office and explain to Sally that you can't get
sauerkraut from an Italian deli.<sup>[<a id="idp6778080" href="#ftn.idp6778080" class="footnote">7</a>]</sup> Once you've agreed on the changes
you will commit, edit your file and remove the conflict
markers:</p><div class="informalexample"><pre class="screen">
Top piece of bread
Mayonnaise
Lettuce
Tomato
Provolone
Salami
Mortadella
Prosciutto
Creole Mustard
Bottom piece of bread
</pre></div><p>Now use <span class="command"><strong>svn resolve</strong></span>, and you're
ready to commit your changes:</p><div class="informalexample"><pre class="screen">
$ svn resolve --accept working sandwich.txt
Resolved conflicted state of 'sandwich.txt'
$ svn commit -m "Go ahead and use my sandwich, discarding Sally's edits."
</pre></div><p>Note that <span class="command"><strong>svn resolve</strong></span>, unlike most of
the other commands we deal with in this chapter, requires
that you explicitly list any filenames that you wish to
resolve. In any case, you want to be careful and use
<span class="command"><strong>svn resolve</strong></span> only when you're certain that
you've fixed the conflict in your file—once the
temporary files are removed, Subversion will let you commit
the file even if it still contains conflict markers.</p><p>If you ever get confused while editing the conflicted
file, you can always consult the three files that Subversion
creates for you in your working copy—including your
file as it was before you updated. You can even use a
third-party interactive merging tool to examine those three
files.</p></div><div class="sect3" title="Discarding your changes in favor of a newly fetched revision"><div class="titlepage"><div><div><h4 class="title"><a id="svn.tour.cycle.resolve.theirsfull"></a>Discarding your changes in favor of a newly fetched
revision</h4></div></div></div><a id="idp6816384" class="indexterm"></a><p>If you get a conflict and decide that you want to throw
out your changes, you can run <strong class="userinput"><code>svn resolve
--accept theirs-full
<em class="replaceable"><code>CONFLICTED-PATH</code></em></code></strong> and
Subversion will discard your edits and remove the temporary
files:</p><div class="informalexample"><pre class="screen">
$ svn update
Updating '.':
Conflict discovered in 'sandwich.txt'.
Select: (p) postpone, (df) diff-full, (e) edit,
(mc) mine-conflict, (tc) theirs-conflict,
(s) show all options: p
C sandwich.txt
Updated to revision 2.
Summary of conflicts:
Text conflicts: 1
$ ls sandwich.*
sandwich.txt sandwich.txt.mine sandwich.txt.r2 sandwich.txt.r1
$ svn resolve --accept theirs-full sandwich.txt
Resolved conflicted state of 'sandwich.txt'
$
</pre></div></div><div class="sect3" title="Punting: using svn revert"><div class="titlepage"><div><div><h4 class="title"><a id="svn.tour.cycle.resolve.revert"></a>Punting: using svn revert</h4></div></div></div><p>If you decide that you want to throw out your changes
and start your edits again (whether this occurs after a
conflict or anytime), just revert your changes:</p><div class="informalexample"><pre class="screen">
$ svn revert sandwich.txt
Reverted 'sandwich.txt'
$ ls sandwich.*
sandwich.txt
$
</pre></div><p>Note that when you revert a conflicted file, you don't
have to use <span class="command"><strong>svn resolve</strong></span>.</p></div></div><div class="sect2" title="Commit Your Changes"><div class="titlepage"><div><div><h3 class="title"><a id="svn.tour.cycle.commit"></a>Commit Your Changes</h3></div></div></div><p>Finally! Your edits are finished, you've merged all
changes from the server, and you're ready to commit your
changes to the repository.</p><p>The <span class="command"><strong>svn commit</strong></span> command sends all of
your changes to the repository. When you commit a change, you
need to supply a log message describing your change. Your log
message will be attached to the new revision you create. If
your log message is brief, you may wish to supply it on the
command line using the <code class="option">--message</code>
(<code class="option">-m</code>) option:</p><div class="informalexample"><pre class="screen">
$ svn commit -m "Corrected number of cheese slices."
Sending sandwich.txt
Transmitting file data .
Committed revision 3.
</pre></div><p>However, if you've been composing your log message in some
other text file as you work, you may want to tell Subversion
to get the message from that file by passing its filename as
the value of the <code class="option">--file</code> (<code class="option">-F</code>)
option:</p><div class="informalexample"><pre class="screen">
$ svn commit -F logmsg
Sending sandwich.txt
Transmitting file data .
Committed revision 4.
</pre></div><p>If you fail to specify either the
<code class="option">--message</code> (<code class="option">-m</code>)
or <code class="option">--file</code> (<code class="option">-F</code>) option,
Subversion will automatically launch your favorite editor (see
the information on <code class="literal">editor-cmd</code> in
<a class="xref" href="#svn.advanced.confarea.opts.config" title="Config">the section called “Config”</a>) for
composing a log message.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>If you're in your editor writing a commit message and
decide that you want to cancel your commit, you can just
quit your editor without saving changes. If you've already
saved your commit message, simply delete all the text, save
again, and then abort:</p><div class="informalexample"><pre class="screen">
$ svn commit
Waiting for Emacs...Done
Log message unchanged or not specified
(a)bort, (c)ontinue, (e)dit
a
$
</pre></div></td></tr></table></div><p>The repository doesn't know or care whether your changes make
any sense as a whole; it checks only to make sure nobody
else has changed any of the same files that you did when you
weren't looking. If somebody <span class="emphasis"><em>has</em></span> done
that, the entire commit will fail with a message informing you
that one or more of your files are out of date:</p><div class="informalexample"><pre class="screen">
$ svn commit -m "Add another rule"
Sending rules.txt
svn: E155011: Commit failed (details follow):
svn: E155011: File '/home/sally/svn-work/sandwich.txt' is out of date
…
</pre></div><p>(The exact wording of this error message depends on the
network protocol and server you're using, but the idea is the
same in all cases.)</p><p>At this point, you need to run <strong class="userinput"><code>svn
update</code></strong>, deal with any merges or conflicts that
result, and attempt your commit again.</p><p>That covers the basic work cycle for using Subversion.
Subversion offers many other features that you can use
to manage your repository and working copy, but most of your
day-to-day use of Subversion will involve only the commands
that we've discussed so far in this chapter. We will,
however, cover a few more commands that you'll use fairly
often.</p></div></div><div class="sect1" title="Examining History"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.tour.history"></a>Examining History</h2></div></div></div><p>Your Subversion repository is like a time machine. It keeps
a record of every change ever committed and allows you to
explore this history by examining previous versions of files and
directories as well as the metadata that accompanies them. With
a single Subversion command, you can check out the repository
(or restore an existing working copy) exactly as it was at any
date or revision number in the past. However, sometimes you
just want to <span class="emphasis"><em>peer into</em></span> the past instead of
<span class="emphasis"><em>going into</em></span> it.</p><p>Several commands can provide you with historical data from
the repository:</p><div class="variablelist"><dl><dt><span class="term"><span class="command"><strong>svn diff</strong></span></span></dt><dd><p>Shows line-level details of a particular change</p></dd><dt><span class="term"><span class="command"><strong>svn log</strong></span></span></dt><dd><p>Shows you broad information: log messages with date
and author information attached to revisions and which
paths changed in each revision</p></dd><dt><span class="term"><span class="command"><strong>svn cat</strong></span></span></dt><dd><p>Retrieves a file as it existed in a particular
revision number and displays it on your screen</p></dd><dt><span class="term"><span class="command"><strong>svn annotate</strong></span></span></dt><dd><p>Retrieves a human-readable file as it existed in a
particular revision number, displaying its contents in a
tabular form with last-changed information attributed to
each line of the file.</p></dd><dt><span class="term"><span class="command"><strong>svn list</strong></span></span></dt><dd><p>Displays the files in a directory for any given
revision</p></dd></dl></div><div class="sect2" title="Examining the Details of Historical Changes"><div class="titlepage"><div><div><h3 class="title"><a id="svn.tour.history.diff"></a>Examining the Details of Historical Changes</h3></div></div></div><p>We've already seen <span class="command"><strong>svn diff</strong></span>
before—it displays file differences in unified diff
format; we used it to show the local modifications made to
our working copy before committing to the repository.</p><p>In fact, it turns out that there are
<span class="emphasis"><em>three</em></span> distinct uses of <span class="command"><strong>svn
diff</strong></span>:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>Examining local changes</p></li><li class="listitem"><p>Comparing your working copy to the repository</p></li><li class="listitem"><p>Comparing repository revisions</p></li></ul></div><div class="sect3" title="Examining local changes"><div class="titlepage"><div><div><h4 class="title"><a id="svn.tour.history.diff.local"></a>Examining local changes</h4></div></div></div><p>As we've seen, invoking <strong class="userinput"><code>svn diff</code></strong> with
no options will compare your working files to the cached
<span class="quote">“<span class="quote">pristine</span>”</span> copies in
the <code class="filename">.svn</code> area:</p><div class="informalexample"><pre class="screen">
$ svn diff
Index: rules.txt
===================================================================
--- rules.txt (revision 3)
+++ rules.txt (working copy)
@@ -1,4 +1,5 @@
Be kind to others
Freedom = Responsibility
Everything in moderation
-Chew with your mouth open
+Chew with your mouth closed
+Listen when others are speaking
$
</pre></div></div><div class="sect3" title="Comparing working copy to repository"><div class="titlepage"><div><div><h4 class="title"><a id="svn.tour.history.diff.wcrepos"></a>Comparing working copy to repository</h4></div></div></div><p>If a single <code class="option">--revision</code>
(<code class="option">-r</code>) number is passed, your
working copy is compared to the specified revision in the
repository:</p><div class="informalexample"><pre class="screen">
$ svn diff -r 3 rules.txt
Index: rules.txt
===================================================================
--- rules.txt (revision 3)
+++ rules.txt (working copy)
@@ -1,4 +1,5 @@
Be kind to others
Freedom = Responsibility
Everything in moderation
-Chew with your mouth open
+Chew with your mouth closed
+Listen when others are speaking
$
</pre></div></div><div class="sect3" title="Comparing repository revisions"><div class="titlepage"><div><div><h4 class="title"><a id="svn.tour.history.diff.reposrepos"></a>Comparing repository revisions</h4></div></div></div><p>If two revision numbers, separated by a colon, are
passed via <code class="option">--revision</code>
(<code class="option">-r</code>), the two revisions are directly
compared:</p><div class="informalexample"><pre class="screen">
$ svn diff -r 2:3 rules.txt
Index: rules.txt
===================================================================
--- rules.txt (revision 2)
+++ rules.txt (revision 3)
@@ -1,4 +1,4 @@
Be kind to others
-Freedom = Chocolate Ice Cream
+Freedom = Responsibility
Everything in moderation
Chew with your mouth open
$
</pre></div><p>A more convenient way of comparing one revision to the
previous revision is to use the <code class="option">--change</code>
(<code class="option">-c</code>) option:</p><div class="informalexample"><pre class="screen">
$ svn diff -c 3 rules.txt
Index: rules.txt
===================================================================
--- rules.txt (revision 2)
+++ rules.txt (revision 3)
@@ -1,4 +1,4 @@
Be kind to others
-Freedom = Chocolate Ice Cream
+Freedom = Responsibility
Everything in moderation
Chew with your mouth open
$
</pre></div><p>Lastly, you can compare repository revisions even when
you don't have a working copy on your local machine, just by
including the appropriate URL on the command line:</p><div class="informalexample"><pre class="screen">
$ svn diff -c 5 http://svn.example.com/repos/example/trunk/text/rules.txt
…
$
</pre></div></div></div><div class="sect2" title="Generating a List of Historical Changes"><div class="titlepage"><div><div><h3 class="title"><a id="svn.tour.history.log"></a>Generating a List of Historical Changes</h3></div></div></div><p>To find information about the history of a file or
directory, use the <span class="command"><strong>svn log</strong></span>
command. <span class="command"><strong>svn log</strong></span> will provide you with a
record of who made changes to a file or directory, at what
revision it changed, the time and date of that revision,
and—if it was provided—the log message that accompanied
the commit:</p><div class="informalexample"><pre class="screen">
$ svn log
------------------------------------------------------------------------
r3 | sally | 2008-05-15 23:09:28 -0500 (Thu, 15 May 2008) | 1 line
Added include lines and corrected # of cheese slices.
------------------------------------------------------------------------
r2 | harry | 2008-05-14 18:43:15 -0500 (Wed, 14 May 2008) | 1 line
Added main() methods.
------------------------------------------------------------------------
r1 | sally | 2008-05-10 19:50:31 -0500 (Sat, 10 May 2008) | 1 line
Initial import
------------------------------------------------------------------------
</pre></div><p>Note that the log messages are printed in
<span class="emphasis"><em>reverse chronological order</em></span> by default.
If you wish to see a different range of revisions in a
particular order or just a single revision, pass the
<code class="option">--revision</code> (<code class="option">-r</code>)
option:</p><div class="table"><a id="svn.tour.history.log.tbl-1"></a><p class="title"><strong>Table 2.1. Common log requests</strong></p><div class="table-contents"><table summary="Common log requests" border="1"><colgroup><col></col><col></col></colgroup><thead><tr><th>Command</th><th>Description</th></tr></thead><tbody><tr><td><strong class="userinput"><code>svn log -r 5:19</code></strong></td><td>Display logs for revisions 5 through 19 in
chronological order</td></tr><tr><td><strong class="userinput"><code>svn log -r 19:5</code></strong></td><td>Display logs for revisions 5 through 19 in
reverse chronological order</td></tr><tr><td><strong class="userinput"><code>svn log -r 8</code></strong></td><td>Display logs for revision 8 only</td></tr></tbody></table></div></div><br class="table-break"></br><p>You can also examine the log history of a single file or
directory. For example:</p><div class="informalexample"><pre class="screen">
$ svn log foo.c
…
$ svn log http://foo.com/svn/trunk/code/foo.c
…
</pre></div><p>These will display log messages <span class="emphasis"><em>only</em></span>
for those revisions in which the working file (or URL)
changed.</p><div class="sidebar" title="Why Does svn log Not Show Me What I Just Committed?"><div class="titlepage"><div><div><p class="title"><strong>Why Does svn log Not Show Me What I
Just Committed?</strong></p></div></div></div><p>If you make a commit and immediately type <strong class="userinput"><code>svn
log</code></strong> with no arguments, you may notice that your
most recent commit doesn't show up in the list of log
messages. This is due to a combination of the behavior of
<span class="command"><strong>svn commit</strong></span> and the default behavior of
<span class="command"><strong>svn log</strong></span>. First, when you commit changes
to the repository, <span class="command"><strong>svn</strong></span> bumps only the
revision of files (and directories) that it commits, so
usually the parent directory remains at the older revision
(See
<a class="xref" href="#svn.basic.in-action.mixedrevs.update-commit" title="Updates and commits are separate">the section called “Updates and commits are separate”</a>
for an explanation of why). <span class="command"><strong>svn log</strong></span> then
defaults to fetching the history of the directory at its
current revision, and thus you don't see the newly committed
changes. The solution here is to either update your working
copy or explicitly provide a revision number to <span class="command"><strong>svn
log</strong></span> by using the <code class="option">--revision</code>
(<code class="option">-r</code>) option.</p></div><p>If you want even more information about a file or
directory, <span class="command"><strong>svn log</strong></span> also takes a
<code class="option">--verbose</code> (<code class="option">-v</code>) option.
Because Subversion allows you to move and copy files and
directories, it is important to be able to track path changes
in the filesystem. So, in verbose mode, <span class="command"><strong>svn
log</strong></span> will include a list of changed paths in a
revision in its output:</p><div class="informalexample"><pre class="screen">
$ svn log -r 8 -v
------------------------------------------------------------------------
r8 | sally | 2008-05-21 13:19:25 -0500 (Wed, 21 May 2008) | 1 line
Changed paths:
M /trunk/code/foo.c
M /trunk/code/bar.h
A /trunk/code/doc/README
Frozzled the sub-space winch.
------------------------------------------------------------------------
</pre></div><p><span class="command"><strong>svn log</strong></span> also takes
a <code class="option">--quiet</code> (<code class="option">-q</code>) option, which
suppresses the body of the log message. When combined
with <code class="option">--verbose</code> (<code class="option">-v</code>), it
gives just the names of the changed files.</p><div class="sidebar" title="Why Does svn log Give Me an Empty Response?"><div class="titlepage"><div><div><p class="title"><strong>Why Does svn log Give Me an Empty
Response?</strong></p></div></div></div><p>After working with Subversion for a bit, most users will
come across something like this:</p><div class="informalexample"><pre class="screen">
$ svn log -r 2
------------------------------------------------------------------------
$
</pre></div><p>At first glance, this seems like an error. But recall
that while revisions are repository-wide, <span class="command"><strong>svn
log</strong></span> operates on a path in the repository. If you
supply no path, Subversion uses the current working
directory as the default target. As a result, if you're
operating in a subdirectory of your working copy and attempt
to see the log of a revision in which neither that directory
nor any of its children was changed, Subversion will show you
an empty log. If you want to see what changed in that
revision, try pointing <span class="command"><strong>svn log</strong></span> directly at
the topmost URL of your repository, as in <strong class="userinput"><code>svn log
-r 2 ^/</code></strong>.</p></div><p>As of Subversion 1.7, users of the Subversion command-line
can also take advantage of a special output mode
for <span class="command"><strong>svn log</strong></span> which integrates a difference
report such as is generated by the <span class="command"><strong>svn diff</strong></span>
command we introduced earlier. When you invoke <span class="command"><strong>svn
log</strong></span> with the <code class="option">--diff</code> option,
Subversion will append to each revision log chunk in the log
report a <span class="command"><strong>diff</strong></span>-style difference report.
This is a very convenient way to see both the high-level, semantic
changes and the line-based modifications of a revision all at
the same time!</p></div><div class="sect2" title="Browsing the Repository"><div class="titlepage"><div><div><h3 class="title"><a id="svn.tour.history.browsing"></a>Browsing the Repository</h3></div></div></div><p>Using <span class="command"><strong>svn cat</strong></span> and <span class="command"><strong>svn
list</strong></span>, you can view various revisions of files and
directories without changing the working revision of your
working copy. In fact, you don't even need a working copy to
use either one.</p><div class="sect3" title="Displaying file contents"><div class="titlepage"><div><div><h4 class="title"><a id="svn.tour.history.browsing.cat"></a>Displaying file contents</h4></div></div></div><p>If you want to examine an earlier version of a file and
not necessarily the differences between two files, you can use
<span class="command"><strong>svn cat</strong></span>:</p><div class="informalexample"><pre class="screen">
$ svn cat -r 2 rules.txt
Be kind to others
Freedom = Chocolate Ice Cream
Everything in moderation
Chew with your mouth open
$
</pre></div><p>You can also redirect the output directly into a
file:</p><div class="informalexample"><pre class="screen">
$ svn cat -r 2 rules.txt > rules.txt.v2
$
</pre></div></div><div class="sect3" title="Displaying line-by-line change attribution"><div class="titlepage"><div><div><h4 class="title"><a id="svn.tour.history.browsing.annotate"></a>Displaying line-by-line change attribution</h4></div></div></div><p>Very similar to the <span class="command"><strong>svn cat</strong></span> command
we discussed in the previous section is the <span class="command"><strong>svn
annotate</strong></span> command. This command also displays the
contents of a versioned file, but it does so using a tabular
format. Each line of output shows not only a line of the
file's content but also the username, the revision number
and (optionally) the datestamp of the revision in which that
line was last modified.</p><p>When used with a working copy file target, <span class="command"><strong>svn
annotate</strong></span> will by default show line-by-line
attribution of the file as it currently appears in the
working copy.</p><div class="informalexample"><pre class="screen">
$ svn annotate rules.txt
1 harry Be kind to others
3 sally Freedom = Responsibility
1 harry Everything in moderation
- - Chew with your mouth closed
- - Listen when others are speaking
</pre></div><p>Notice that for some lines, there is no attribution
provided. In this case, that's because those lines have
been modified in the working copy's version of the file. In
this way, <span class="command"><strong>svn annotate</strong></span> becomes another
way for you to see which lines in the file you have
changed. You can use the <code class="literal">BASE</code> revision
keyword (see <a class="xref" href="#svn.tour.revs.keywords" title="Revision Keywords">the section called “Revision Keywords”</a>) to
instead see the unmodified form of the file as it resides
in your working copy.</p><div class="informalexample"><pre class="screen">
$ svn annotate rules.txt@BASE
1 harry Be kind to others
3 sally Freedom = Responsibility
1 harry Everything in moderation
1 harry Chew with your mouth open
</pre></div><p>The <code class="option">--verbose (-v)</code> option causes
<span class="command"><strong>svn annotate</strong></span> to also include on each line
the datestamp associated with that line's reported revision
number. (This adds a significant amount of width to each
line of ouput, so we'll skip the demonstration here.)</p><p>As with <span class="command"><strong>svn cat</strong></span>, you can also
ask <span class="command"><strong>svn annotate</strong></span> to display previous
versions of the file. This can be a handy trick when, after
finding out who most recently modified a particular line of
interest in the file, you then wish to see who modified the
same line prior to that.</p><div class="informalexample"><pre class="screen">
$ svn blame rules.txt -r 2
1 harry Be kind to others
1 harry Freedom = Chocolate Ice Cream
1 harry Everything in moderation
1 harry Chew with your mouth open
</pre></div><p>Unlike the <span class="command"><strong>svn cat</strong></span> command, the
functionality of <span class="command"><strong>svn annotate</strong></span> is tied
heavily to the idea of <span class="quote">“<span class="quote">lines</span>”</span> of text in a
human-readable file. As such, if you attempt to run the
command on a file that Subversion has determined is
<span class="emphasis"><em>not</em></span> human-readable (per the file's
<code class="literal">svn:mime-type</code> property—see <a class="xref" href="#svn.advanced.props.special.mime-type" title="File Content Type">the section called “File Content Type”</a> for
details), you'll get an error message.</p><div class="informalexample"><pre class="screen">
$ svn annotate images/logo.png
Skipping binary file: 'images/logo.png'
$
</pre></div><p>As revealed in the error message, you can use
the <code class="option">--force</code> option to disable this check
and proceed with the annotation as if the file's contents
are, in fact, human-readable and line-based. Naturally, if
you force Subversion to try to perform line-based annotation
on a nontextual file, you'll get what you asked for: a
screenful of nonsense.</p><div class="informalexample"><pre class="screen">
$ svn annotate images/logo.png --force
6 harry \211PNG
6 harry ^Z
6 harry
7 harry \274\361\MI\300\365\353^X\300…
</pre></div><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>Depending on your mood at the time you execute this
command and your reasons for doing so, you may find
yourself typing <strong class="userinput"><code>svn blame …</code></strong>
or <strong class="userinput"><code>svn praise …</code></strong> instead of
using the canonical <span class="command"><strong>svn annotate</strong></span>
command form. That's okay—the Subversion developers
anticipated as much, so those particular command aliases
work, too!</p></td></tr></table></div><p>Finally, as with many of Subversion's informational
commands, you can also reference files in your <span class="command"><strong>svn
annotate</strong></span> command invocations by their repository
URLs, allowing access to this information even when you
don't have ready access to a working copy.</p></div><div class="sect3" title="Listing versioned directories"><div class="titlepage"><div><div><h4 class="title"><a id="svn.tour.history.browsing.list"></a>Listing versioned directories</h4></div></div></div><p>The <span class="command"><strong>svn list</strong></span> command shows you what
files are in a repository directory without actually
downloading the files to your local machine:</p><div class="informalexample"><pre class="screen">
$ svn list http://svn.example.com/repo/project
README
branches/
tags/
trunk/
</pre></div><p>If you want a more detailed listing, pass the
<code class="option">--verbose</code> (<code class="option">-v</code>) flag to get
output like this:</p><div class="informalexample"><pre class="screen">
$ svn list -v http://svn.example.com/repo/project
23351 sally Feb 05 13:26 ./
20620 harry 1084 Jul 13 2006 README
23339 harry Feb 04 01:40 branches/
23198 harry Jan 23 17:17 tags/
23351 sally Feb 05 13:26 trunk/
</pre></div><p>The columns tell you the revision at which the file or
directory was last modified, the user who modified it, the size
if it is a file, the date it was last modified, and the item's
name.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>The <strong class="userinput"><code>svn list</code></strong> command with no
arguments defaults to the <span class="emphasis"><em>repository
URL</em></span> of the current working
directory, <span class="emphasis"><em>not</em></span> the local working copy
directory. After all, if you want a listing of your local
directory, you could use just plain
<span class="command"><strong>ls</strong></span> (or any reasonable non-Unixy
equivalent).</p></td></tr></table></div></div></div><div class="sect2" title="Fetching Older Repository Snapshots"><div class="titlepage"><div><div><h3 class="title"><a id="svn.tour.history.snapshots"></a>Fetching Older Repository Snapshots</h3></div></div></div><p>In addition to all of the previous commands, you can use
the <code class="option">--revision</code> (<code class="option">-r</code>) option
with <span class="command"><strong>svn update</strong></span> to take an entire working
copy <span class="quote">“<span class="quote">back in time</span>”</span>:<sup>[<a id="idp6969152" href="#ftn.idp6969152" class="footnote">8</a>]</sup></p><div class="informalexample"><pre class="screen">
# Make the current directory look like it did in r1729.
$ svn update -r 1729
Updating '.':
…
$
</pre></div><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>Many Subversion newcomers attempt to use the preceding
<span class="command"><strong>svn update</strong></span> example to <span class="quote">“<span class="quote">undo</span>”</span>
committed changes, but this won't work as you can't commit
changes that you obtain from backdating a working copy if
the changed files have newer revisions. See <a class="xref" href="#svn.branchmerge.basicmerging.resurrect" title="Resurrecting Deleted Items">the section called “Resurrecting Deleted Items”</a> for a
description of how to <span class="quote">“<span class="quote">undo</span>”</span> a commit.</p></td></tr></table></div><p>If you'd prefer to create a whole new working copy from an
older snapshot, you can do so by modifying the typical
<span class="command"><strong>svn checkout</strong></span> command. As with <span class="command"><strong>svn
update</strong></span>, you can provide
the <code class="option">--revision</code> (<code class="option">-r</code>) option.
But for reasons that we cover in
<a class="xref" href="#svn.advanced.pegrevs" title="Peg and Operative Revisions">the section called “Peg and Operative Revisions”</a>, you might instead want
to specify the target revision as part of Subversion's
expanded URL syntax.</p><div class="informalexample"><pre class="screen">
# Checkout the trunk from r1729.
$ svn checkout http://svn.example.com/svn/repo/trunk@1729 trunk-1729
…
# Checkout the current trunk as it looked in r1729.
$ svn checkout http://svn.example.com/svn/repo/trunk -r 1729 trunk-1729
…
$
</pre></div><p>Lastly, if you're building a release and wish to bundle up
your versioned files and directories, you can use <span class="command"><strong>svn
export</strong></span> to create a local copy of all or part of your
repository without any <code class="filename">.svn</code>
administrative directories included. The basic syntax of this
subcommand is identical to that of <span class="command"><strong>svn
checkout</strong></span>:</p><div class="informalexample"><pre class="screen">
# Export the trunk from the latest revision.
$ svn export http://svn.example.com/svn/repo/trunk trunk-export
…
# Export the trunk from r1729.
$ svn export http://svn.example.com/svn/repo/trunk@1729 trunk-1729
…
# Export the current trunk as it looked in r1729.
$ svn export http://svn.example.com/svn/repo/trunk -r 1729 trunk-1729
…
$
</pre></div></div></div><div class="sect1" title="Sometimes You Just Need to Clean Up"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.tour.cleanup"></a>Sometimes You Just Need to Clean Up</h2></div></div></div><p>Now that we've covered the day-to-day tasks that you'll
frequently use Subversion for, we'll review a few administrative
tasks relating to your working copy.</p><div class="sect2" title="Disposing of a Working Copy"><div class="titlepage"><div><div><h3 class="title"><a id="svn.tour.cleanup.disposal"></a>Disposing of a Working Copy</h3></div></div></div><p>Subversion doesn't track either the state or the existence of
working copies on the server, so there's no server overhead to
keeping working copies around. Likewise, there's no need to
let the server know that you're going to delete a working
copy.</p><p>If you're likely to use a working copy again, there's
nothing wrong with just leaving it on disk until you're ready
to use it again, at which point all it takes is an
<span class="command"><strong>svn update</strong></span> to bring it up to date and ready
for use.</p><p>However, if you're definitely not going to use a working
copy again, you can safely delete the entire thing using
whatever directory removal capabilities your operating system
offers. We recommend that before you do so you
run <strong class="userinput"><code>svn status</code></strong> and review any files
listed in its output that are prefixed with a
<code class="literal">?</code> to make certain that they're not of
importance.</p></div><div class="sect2" title="Recovering from an Interruption"><div class="titlepage"><div><div><h3 class="title"><a id="svn.tour.cleanup.interruption"></a>Recovering from an Interruption</h3></div></div></div><p>When Subversion modifies your working copy—either
your files or its own administrative state—it tries to do
so as safely as possible. Before changing the working copy,
Subversion logs its intentions in a private <span class="quote">“<span class="quote">to-do
list</span>”</span>, of sorts. Next, it performs those actions to
effect the desired change, holding a lock on the relevant part
of the working copy while it works. This prevents other
Subversion clients from accessing the working copy mid-change.
Finally, Subversion releases its lock and cleans up its
private to-do list. Architecturally, this is similar to a
journaled filesystem. If a Subversion operation is
interrupted (e.g, if the process is killed or if the machine
crashes), the private to-do list remains on disk. This allows
Subversion to return to that list later to complete any
unfinished operations and return your working copy to a
consistent state.</p><p>This is exactly what <span class="command"><strong>svn cleanup</strong></span> does:
it searches your working copy and runs any leftover to-do
items, removing working copy locks as it completes those
operations. If Subversion ever tells you that some part of
your working copy is <span class="quote">“<span class="quote">locked,</span>”</span> run <span class="command"><strong>svn
cleanup</strong></span> to remedy the problem. The <span class="command"><strong>svn
status</strong></span> command will inform you about administrative
locks in the working copy, too, by displaying
an <code class="literal">L</code> next to those locked paths:</p><div class="informalexample"><pre class="screen">
$ svn status
L somedir
M somedir/foo.c
$ svn cleanup
$ svn status
M somedir/foo.c
</pre></div><p>Don't confuse these working copy administrative locks with
the user-managed locks that Subversion users create when using
the lock-modify-unlock model of concurrent version control;
see the sidebar
<a class="xref" href="#svn.advanced.locking.meanings" title="The Three Meanings of “Lock”">The Three Meanings of <span class="quote">“<span class="quote">Lock</span>”</span></a> for
clarification.</p></div></div><div class="sect1" title="Dealing with Structural Conflicts"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.tour.treeconflicts"></a>Dealing with Structural Conflicts</h2></div></div></div><p>So far, we have only talked about conflicts at the level of
file content. When you and your collaborators make overlapping
changes within the same file, Subversion forces you to merge
those changes before you can commit.<sup>[<a id="idp7001952" href="#ftn.idp7001952" class="footnote">9</a>]</sup></p><p>But what happens if your collaborators move or delete a file
that you are still working on? Maybe there was a
miscommunication, and one person thinks the file should be
deleted, while another person still wants to commit changes to
the file. Or maybe your collaborators did some refactoring,
renaming files and moving around directories in the process. If
you were still working on these files, those modifications may
need to be applied to the files at their new location. Such
conflicts manifest themselves at the directory tree structure
level rather than at the file content level, and are known
as <em class="firstterm">tree conflicts</em>.</p><div class="sidebar" title="Tree conflicts prior to Subversion 1.6"><div class="titlepage"><div><div><p class="title"><strong>Tree conflicts prior to Subversion 1.6</strong></p></div></div></div><p>Prior to Subversion 1.6, tree conflicts could yield rather
unexpected results. For example, if a file was locally
modified, but had been renamed in the repository,
running <span class="command"><strong>svn update</strong></span> would make Subversion
carry out the following steps:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>Check the file to be renamed for local
modifications.</p></li><li class="listitem"><p>Delete the file at its old location, and if it
had local modifications, keep an on-disk copy of the file at
the old location. This on-disk copy now appears as an
unversioned file in the working copy.</p></li><li class="listitem"><p>Add the file, as it exists in the repository,
at its new location.</p></li></ul></div><p>When this situation arises, there is the possibility that
the user makes a commit without realizing that local
modifications have been left in a now-unversioned file in the
working copy, and have not reached the repository. This gets
more and more likely (and tedious) if the number of files
affected by this problem is large.</p><p>Since Subversion 1.6, this and other similar situations
are flagged as conflicts in the working copy.</p></div><p>As with textual conflicts, tree conflicts prevent a commit
from being made from the conflicted state, giving the user the
opportunity to examine the state of the working copy for
potential problems arising from the tree conflict, and resolving
any such problems before committing.</p><div class="sect2" title="An Example Tree Conflict"><div class="titlepage"><div><div><h3 class="title"><a id="svn.tour.treeconflicts.example"></a>An Example Tree Conflict</h3></div></div></div><p>Suppose a software project you were working on currently
looked like this:</p><div class="informalexample"><pre class="screen">
$ svn list -Rv svn://svn.example.com/trunk/
13 harry Sep 06 10:34 ./
13 harry 27 Sep 06 10:34 COPYING
13 harry 41 Sep 06 10:32 Makefile
13 harry 53 Sep 06 10:34 README
13 harry Sep 06 10:32 code/
13 harry 54 Sep 06 10:32 code/bar.c
13 harry 130 Sep 06 10:32 code/foo.c
$
</pre></div><p>Later, in revision 14, your collaborator Harry renames the file
<code class="filename">bar.c</code> to <code class="filename">baz.c</code>.
Unfortunately, you don't realize this yet. As it turns out,
you are busy in your working copy composing a different set of
changes, some of which also involve modifications
to <code class="filename">bar.c</code>:</p><div class="informalexample"><pre class="screen">
$ svn diff
Index: code/foo.c
===================================================================
--- code/foo.c (revision 13)
+++ code/foo.c (working copy)
@@ -3,5 +3,5 @@
int main(int argc, char *argv[])
{
printf("I don't like being moved around!\n%s", bar());
- return 0;
+ return 1;
}
Index: code/bar.c
===================================================================
--- code/bar.c (revision 13)
+++ code/bar.c (working copy)
@@ -1,4 +1,4 @@
const char *bar(void)
{
- return "Me neither!\n";
+ return "Well, I do like being moved around!\n";
}
$
</pre></div><p>You first realize that someone else has
changed <code class="filename">bar.c</code> when your own commit
attempt fails:</p><div class="informalexample"><pre class="screen">
$ svn commit -m "Small fixes"
Sending code/bar.c
svn: E155011: Commit failed (details follow):
svn: E155011: File '/home/svn/project/code/bar.c' is out of date
svn: E160013: File not found: transaction '14-e', path '/code/bar.c'
$
</pre></div><p>At this point, you need to run <span class="command"><strong>svn update</strong></span>.
Besides bringing our working copy up to date so that you can
see Harry's changes, this also flags a tree conflict so you
have the opportunity to evaluate and properly resolve it.</p><div class="informalexample"><pre class="screen">
$ svn update
Updating '.':
C code/bar.c
A code/baz.c
U Makefile
Updated to revision 14.
Summary of conflicts:
Tree conflicts: 1
$
</pre></div><p>In its output, <span class="command"><strong>svn update</strong></span> signifies tree
conflicts using a capital C in the fourth output column.
<span class="command"><strong>svn status</strong></span> reveals additional details of the
conflict:</p><div class="informalexample"><pre class="screen">
$ svn status
M code/foo.c
A + C code/bar.c
> local edit, incoming delete upon update
Summary of conflicts:
Tree conflicts: 1
$
</pre></div><p>Note how <code class="filename">bar.c</code> is automatically
scheduled for re-addition in your working copy, which
simplifies things in case you want to keep the file.</p><p>Because a move in Subversion is implemented as a copy
operation followed by a delete operation, and these two
operations cannot be easily related to one another during an
update, all Subversion can warn you about is an incoming
delete operation on a locally modified file. This delete
operation <span class="emphasis"><em>may</em></span> be part of a move, or it
could be a genuine delete operation. Determining exactly what
semantic change was made to the repository is
important—you want to know just how your own edits fit
into the overall trajectory of the project. So read log
messages, talk to your collaborators, study the line-based
differences—do whatever you must do—to determine
your best course of action.</p><p>In this case, Harry's commit log message tells you what
you need to know.</p><div class="informalexample"><pre class="screen">
$ svn log -r14 ^/trunk
------------------------------------------------------------------------
r14 | harry | 2011-09-06 10:38:17 -0400 (Tue, 06 Sep 2011) | 1 line
Changed paths:
M /Makefile
D /code/bar.c
A /code/baz.c (from /code/bar.c:13)
Rename bar.c to baz.c, and adjust Makefile accordingly.
------------------------------------------------------------------------
$
</pre></div><p><span class="command"><strong>svn info</strong></span> shows the URLs of the items
involved in the conflict. The <span class="emphasis"><em>left</em></span> URL
shows the source of the local side of the conflict, while
the <span class="emphasis"><em>right</em></span> URL shows the source of the
incoming side of the conflict. These URLs indicate where you
should start searching the repository's history for the change
which conflicts with your local change.</p><div class="informalexample"><pre class="screen">
$ svn info code/bar.c
Path: code/bar.c
Name: bar.c
URL: http://svn.example.com/svn/repo/trunk/code/bar.c
…
Tree conflict: local edit, incoming delete upon update
Source left: (file) ^/trunk/code/bar.c@4
Source right: (none) ^/trunk/code/bar.c@5
$
</pre></div><p><code class="filename">bar.c</code> is now said to be the
<em class="firstterm">victim</em> of a tree conflict.
It cannot be committed until the conflict is resolved:</p><div class="informalexample"><pre class="screen">
$ svn commit -m "Small fixes"
svn: E155015: Commit failed (details follow):
svn: E155015: Aborting commit: '/home/svn/project/code/bar.c' remains in confl
ict
$
</pre></div><p>To resolve this conflict, you must either agree or
disagree with the move that Harry made.</p><p>If you agree with the move, your <code class="filename">bar.c</code>
is superfluous. You'll want to delete it and mark the tree
conflict as resolved. But wait: you made changes to that
file! Before deleting <code class="filename">bar.c</code>, you need to
decide if the changes you made to it need to be applied
elsewhere, for example to the new <code class="filename">baz.c</code>
file where all of <code class="filename">bar.c</code>'s code now lives.
Let's assume that your changes do need to <span class="quote">“<span class="quote">follow the
move</span>”</span>. Subversion isn't smart enough to do this work
for you<sup>[<a id="idp7041344" href="#ftn.idp7041344" class="footnote">10</a>]</sup>, so you need to migrate your
changes manually.</p><p>In our example, you could manually re-make your change
to <code class="filename">bar.c</code> pretty easily—it was,
after all, a single-line change. That's not always the case,
though, so we'll show a more scalable approach. We'll first
use <span class="command"><strong>svn diff</strong></span> to create a patch file. Then
we'll edit the headers of that patch file to point to the new
name of our renamed file. Finally, we re-apply the modified
patch to our working copy.</p><div class="informalexample"><pre class="screen">
$ svn diff code/bar.c > PATCHFILE
$ cat PATCHFILE
Index: code/bar.c
===================================================================
--- code/bar.c (working copy)
+++ code/bar.c (working copy)
@@ -1,4 +1,4 @@
const char *bar(void)
{
- return "Me neither!\n";
+ return "Well, I do like being moved around!\n";
}
$ ### Edit PATCHFILE to refer to code/baz.c instead of code/bar.c
$ cat PATCHFILE
Index: code/baz.c
===================================================================
--- code/baz.c (working copy)
+++ code/baz.c (working copy)
@@ -1,4 +1,4 @@
const char *bar(void)
{
- return "Me neither!\n";
+ return "Well, I do like being moved around!\n";
}
$ svn patch PATCHFILE
U code/baz.c
$
</pre></div><p>Now that the changes you originally made
to <code class="filename">bar.c</code> have been successfully
reproduced in <code class="filename">baz.c</code>, you can
delete <code class="filename">bar.c</code> and resolve the conflict,
instructing the resolution logic to accept what is currently
in the working copy as the desired result.</p><div class="informalexample"><pre class="screen">
$ svn delete --force code/bar.c
D code/bar.c
$ svn resolve --accept=working code/bar.c
Resolved conflicted state of 'code/bar.c'
$ svn status
M code/foo.c
M code/baz.c
$ svn diff
Index: code/foo.c
===================================================================
--- code/foo.c (revision 14)
+++ code/foo.c (working copy)
@@ -3,5 +3,5 @@
int main(int argc, char *argv[])
{
printf("I don't like being moved around!\n%s", bar());
- return 0;
+ return 1;
}
Index: code/baz.c
===================================================================
--- code/baz.c (revision 14)
+++ code/baz.c (working copy)
@@ -1,4 +1,4 @@
const char *bar(void)
{
- return "Me neither!\n";
+ return "Well, I do like being moved around!\n";
}
$
</pre></div><p>But what if you do not agree with the move? Well, in that
case, you can delete <code class="filename">baz.c</code> instead, after
making sure any changes made to it after it was renamed are
either preserved or not worth keeping. (Do not forget to also
revert the changes Harry made to <code class="filename">Makefile</code>.)
Since <code class="filename">bar.c</code> is already scheduled for
re-addition, there is nothing else left to do, and the
conflict can be marked resolved:</p><div class="informalexample"><pre class="screen">
$ svn delete --force code/baz.c
D code/baz.c
$ svn resolve --accept=working code/bar.c
Resolved conflicted state of 'code/bar.c'
$ svn status
M code/foo.c
A + code/bar.c
D code/baz.c
M Makefile
$ svn diff
Index: code/foo.c
===================================================================
--- code/foo.c (revision 14)
+++ code/foo.c (working copy)
@@ -3,5 +3,5 @@
int main(int argc, char *argv[])
{
printf("I don't like being moved around!\n%s", bar());
- return 0;
+ return 1;
}
Index: code/bar.c
===================================================================
--- code/bar.c (revision 14)
+++ code/bar.c (working copy)
@@ -1,4 +1,4 @@
const char *bar(void)
{
- return "Me neither!\n";
+ return "Well, I do like being moved around!\n";
}
Index: code/baz.c
===================================================================
--- code/baz.c (revision 14)
+++ code/baz.c (working copy)
@@ -1,4 +0,0 @@
-const char *bar(void)
-{
- return "Me neither!\n";
-}
Index: Makefile
===================================================================
--- Makefile (revision 14)
+++ Makefile (working copy)
@@ -1,2 +1,2 @@
foo:
- $(CC) -o $@ code/foo.c code/baz.c
+ $(CC) -o $@ code/foo.c code/bar.c
</pre></div><p>You've now resolved your first tree conflict! You can
commit your changes and tell Harry during tea break about all
the extra work he caused for you.</p></div></div><div class="sect1" title="Summary"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.tour.summary"></a>Summary</h2></div></div></div><p>Now we've covered most of the Subversion client commands.
Notable exceptions are those dealing with branching and
merging (see <a class="xref" href="#svn.branchmerge" title="Chapter 4. Branching and Merging">Chapter 4, <em>Branching and Merging</em></a>) and properties (see
<a class="xref" href="#svn.advanced.props" title="Properties">the section called “Properties”</a>). However, you may want to
take a moment to skim through <a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a> to
get an idea of all the different commands that Subversion
has—and how you can use them to make your work
easier.</p></div><div class="footnotes"><br></br><hr width="100" align="left"></hr><div class="footnote"><p><sup>[<a id="ftn.idp6550064" href="#idp6550064" class="para">6</a>] </sup>Of course, nothing is ever
totally deleted from the repository—just from
its <code class="literal">HEAD</code> revision. You may continue
to access the deleted item in previous revisions.
Should you desire to resurrect the item so that it is
again present in <code class="literal">HEAD</code>, see
<a class="xref" href="#svn.branchmerge.basicmerging.resurrect" title="Resurrecting Deleted Items">the section called “Resurrecting Deleted Items”</a>.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp6778080" href="#idp6778080" class="para">7</a>] </sup>And if you
ask them for it, they may very well ride you out of town on
a rail.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp6969152" href="#idp6969152" class="para">8</a>] </sup>See? We told
you that Subversion was a time
machine.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp7001952" href="#idp7001952" class="para">9</a>] </sup>Well,
you <span class="emphasis"><em>could</em></span> mark files containing conflict
markers as resolved and commit them, if you really wanted to.
But this is rarely done in practice.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp7041344" href="#idp7041344" class="para">10</a>] </sup>In some cases, Subversion 1.5 and 1.6
<span class="emphasis"><em>would</em></span> actually handle this for you, but
this somewhat hit-or-miss functionality was removed in
Subversion 1.7.</p></div></div></div><div class="chapter" title="Chapter 3. Advanced Topics"><div class="titlepage"><div><div><h2 class="title"><a id="svn.advanced"></a>Chapter 3. Advanced Topics</h2></div></div></div><div class="toc"><p><strong>Table of Contents</strong></p><dl><dt><span class="sect1"><a href="#svn.tour.revs.specifiers">Revision Specifiers</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.tour.revs.keywords">Revision Keywords</a></span></dt><dt><span class="sect2"><a href="#svn.tour.revs.dates">Revision Dates</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.advanced.pegrevs">Peg and Operative Revisions</a></span></dt><dt><span class="sect1"><a href="#svn.advanced.props">Properties</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.advanced.props.why">Why Properties?</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.props.manip">Manipulating Properties</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.props.workflow">Properties and the Subversion Workflow</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.props.auto">Automatic Property Setting</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.advanced.props.file-portability">File Portability</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.advanced.props.special.mime-type">File Content Type</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.props.special.executable">File Executability</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.props.special.eol-style">End-of-Line Character Sequences</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.advanced.props.special.ignore">Ignoring Unversioned Items</a></span></dt><dt><span class="sect1"><a href="#svn.advanced.props.special.keywords">Keyword Substitution</a></span></dt><dt><span class="sect1"><a href="#svn.advanced.sparsedirs">Sparse Directories</a></span></dt><dt><span class="sect1"><a href="#svn.advanced.locking">Locking</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.advanced.locking.creation">Creating Locks</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.locking.discovery">Discovering Locks</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.locking.break-steal">Breaking and Stealing Locks</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.locking.lock-communication">Lock Communication</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.advanced.externals">Externals Definitions</a></span></dt><dt><span class="sect1"><a href="#svn.advanced.changelists">Changelists</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.advanced.changelists.creating">Creating and Modifying Changelists</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.changelists.asfilters">Changelists As Operation Filters</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.changelists.limitations">Changelist Limitations</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.serverconfig.netmodel">Network Model</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.serverconfig.netmodel.reqresp">Requests and Responses</a></span></dt><dt><span class="sect2"><a href="#svn.serverconfig.netmodel.creds">Client Credentials</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.serverconfig.netmodel.credcache">Caching credentials</a></span></dt><dt><span class="sect3"><a href="#svn.tour.initial.disabling-password-caching">Disabling password caching</a></span></dt><dt><span class="sect3"><a href="#svn.tour.initial.authn-cache-purge">Removing cached credentials</a></span></dt><dt><span class="sect3"><a href="#svn.tour.initial.different-user">Command-line authentication</a></span></dt><dt><span class="sect3"><a href="#svn.tour.initial.authn-wrapup">Authentication wrap-up</a></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.advanced.summary">Summary</a></span></dt></dl></div><p>If you've been reading this book chapter by chapter, from
start to finish, you should by now have acquired enough
knowledge to use the Subversion client to perform the most
common version control operations. You understand how to
check out a working copy from a Subversion repository. You are
comfortable with submitting and receiving changes using the
<span class="command"><strong>svn commit</strong></span> and <span class="command"><strong>svn update</strong></span>
operations. You've probably even developed a reflex that causes
you to run the <span class="command"><strong>svn status</strong></span> command almost
unconsciously. For all intents and purposes, you are ready to
use Subversion in a typical environment.</p><p>But the Subversion feature set doesn't stop at <span class="quote">“<span class="quote">common
version control operations.</span>”</span> It has other bits of
functionality besides just communicating file and
directory changes to and from a central repository.</p><p>This chapter highlights some of Subversion's features that,
while important, may not be part of the typical user's daily routine.
It assumes that you are familiar with Subversion's basic file and
directory versioning capabilities. If you aren't, you'll want to
first read <a class="xref" href="#svn.basic" title="Chapter 1. Fundamental Concepts">Chapter 1, <em>Fundamental Concepts</em></a> and <a class="xref" href="#svn.tour" title="Chapter 2. Basic Usage">Chapter 2, <em>Basic Usage</em></a>. Once you've mastered those basics and
consumed this chapter, you'll be a Subversion power user!</p><div class="sect1" title="Revision Specifiers"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.tour.revs.specifiers"></a>Revision Specifiers</h2></div></div></div><p>As we described in <a class="xref" href="#svn.basic.in-action.revs" title="Revisions">the section called “Revisions”</a>, revision numbers in Subversion are pretty
straightforward—integers that keep getting larger as you
commit more changes to your versioned data. Still, it doesn't
take long before you can no longer remember exactly what
happened in each and every revision. Fortunately, the typical
Subversion workflow doesn't often demand that you supply
arbitrary revisions to the Subversion operations you perform.
For operations that <span class="emphasis"><em>do</em></span> require a revision
specifier, you generally supply a revision number that you saw
in a commit email, in the output of some other Subversion
operation, or in some other context that would give meaning to
that particular number.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>Referring to revision numbers with
an <span class="quote">“<span class="quote"><code class="literal">r</code></span>”</span> prefix
(<code class="literal">r314</code>, for example) is an established
practice in Subversion communities, and is both supported and
encouraged by many Subversion-related tools. In most places
where you would specify a bare revision number on the command
line, you may also use
the <code class="literal">r</code><em class="replaceable"><code>NNN</code></em>
syntax.</p></td></tr></table></div><p>But occasionally, you need to pinpoint a moment in time for
which you don't already have a revision number memorized or
handy. So besides the integer revision numbers,
<span class="command"><strong>svn</strong></span> allows as input some additional forms of
revision specifiers: <em class="firstterm">revision keywords</em>
and revision dates.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>The various forms of Subversion revision specifiers can be
mixed and matched when used to specify revision ranges. For
example, you can use <code class="option">-r
<em class="replaceable"><code>REV1</code></em>:<em class="replaceable"><code>REV2</code></em></code>
where <em class="replaceable"><code>REV1</code></em> is a revision keyword
and <em class="replaceable"><code>REV2</code></em> is a revision number, or
where <em class="replaceable"><code>REV1</code></em> is a date and
<em class="replaceable"><code>REV2</code></em> is a revision keyword, and so
on. The individual revision specifiers are independently
evaluated, so you can put whatever you want on the opposite
sides of that colon.</p></td></tr></table></div><div class="sect2" title="Revision Keywords"><div class="titlepage"><div><div><h3 class="title"><a id="svn.tour.revs.keywords"></a>Revision Keywords</h3></div></div></div><a id="idp7736080" class="indexterm"></a><p>The Subversion client understands a number of revision
keywords. These keywords can be used instead of integer
arguments to the <code class="option">--revision</code>
(<code class="option">-r</code>) option, and are resolved into specific
revision numbers by Subversion:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">HEAD</code></span></dt><dd><a id="idp7740672" class="indexterm"></a><a id="idp7741744" class="indexterm"></a><p>The latest (or <span class="quote">“<span class="quote">youngest</span>”</span>) revision in
the repository.</p></dd><dt><span class="term"><code class="literal">BASE</code></span></dt><dd><a id="idp7746128" class="indexterm"></a><a id="idp7747200" class="indexterm"></a><p>The revision number of an item in a working copy.
If the item has been locally modified, this refers to
the way the item appears without those local
modifications.</p></dd><dt><span class="term"><code class="literal">COMMITTED</code></span></dt><dd><a id="idp7751296" class="indexterm"></a><a id="idp7752368" class="indexterm"></a><p>The most recent revision prior to, or equal to,
<code class="literal">BASE</code>, in which an item changed.</p></dd><dt><span class="term"><code class="literal">PREV</code></span></dt><dd><a id="idp7756976" class="indexterm"></a><a id="idp7758048" class="indexterm"></a><p>The revision immediately <span class="emphasis"><em>before</em></span>
the last revision in which an item changed.
Technically, this boils down to
<code class="literal">COMMITTED</code>-1.</p></dd></dl></div><p>As can be derived from their descriptions, the
<code class="literal">PREV</code>, <code class="literal">BASE</code>, and
<code class="literal">COMMITTED</code> revision keywords are used only
when referring to a working copy path—they don't apply
to repository URLs. <code class="literal">HEAD</code>, on the other
hand, can be used in conjunction with both of these path
types.</p><p>Here are some examples of revision keywords in
action:</p><div class="informalexample"><pre class="screen">
$ svn diff -r PREV:COMMITTED foo.c
# shows the last change committed to foo.c
$ svn log -r HEAD
# shows log message for the latest repository commit
$ svn diff -r HEAD
# compares your working copy (with all of its local changes) to the
# latest version of that tree in the repository
$ svn diff -r BASE:HEAD foo.c
# compares the unmodified version of foo.c with the latest version of
# foo.c in the repository
$ svn log -r BASE:HEAD
# shows all commit logs for the current versioned directory since you
# last updated
$ svn update -r PREV foo.c
# rewinds the last change on foo.c, decreasing foo.c's working revision
$ svn diff -r BASE:14 foo.c
# compares the unmodified version of foo.c with the way foo.c looked
# in revision 14
</pre></div></div><div class="sect2" title="Revision Dates"><div class="titlepage"><div><div><h3 class="title"><a id="svn.tour.revs.dates"></a>Revision Dates</h3></div></div></div><a id="idp7769616" class="indexterm"></a><p>Revision numbers reveal nothing about the world outside
the version control system, but sometimes you need to
correlate a moment in real time with a moment in version
history. To facilitate this, the <code class="option">--revision</code>
(<code class="option">-r</code>) option can also accept as input date
specifiers wrapped in curly braces (<code class="literal">{</code> and
<code class="literal">}</code>). Subversion accepts the standard
ISO-8601 date and time formats, plus a few others. Here are
some examples.</p><div class="informalexample"><pre class="screen">
$ svn update -r {2006-02-17}
$ svn update -r {15:30}
$ svn update -r {15:30:00.200000}
$ svn update -r {"2006-02-17 15:30"}
$ svn update -r {"2006-02-17 15:30 +0230"}
$ svn update -r {2006-02-17T15:30}
$ svn update -r {2006-02-17T15:30Z}
$ svn update -r {2006-02-17T15:30-04:00}
$ svn update -r {20060217T1530}
$ svn update -r {20060217T1530Z}
$ svn update -r {20060217T1530-0500}
…
</pre></div><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>Keep in mind that most shells will require you to, at a
minimum, quote or otherwise escape any spaces that are
included as part of revision date specifiers. Certain
shells may also take issue with the unescaped use of curly
braces, too. Consult your shell's documentation for the
requirements specific to your environment.</p></td></tr></table></div><p>When you specify a date, Subversion resolves that date to
the most recent revision of the repository as of that date,
and then continues to operate against that resolved revision
number:</p><div class="informalexample"><pre class="screen">
$ svn log -r {2006-11-28}
------------------------------------------------------------------------
r12 | ira | 2006-11-27 12:31:51 -0600 (Mon, 27 Nov 2006) | 6 lines
…
</pre></div><div class="sidebar" title="Is Subversion a Day Early?"><div class="titlepage"><div><div><p class="title"><strong>Is Subversion a Day Early?</strong></p></div></div></div><p>If you specify a single date as a revision without
specifying a time of day (for example
<code class="literal">2006-11-27</code>), you may think that Subversion
should give you the last revision that took place on the
27th of November. Instead, you'll get back a revision from
the 26th, or even earlier. Remember that Subversion will
find the <span class="emphasis"><em>most recent revision of the
repository</em></span> as of the date you give. If you give
a date without a timestamp, such as
<code class="literal">2006-11-27</code>, Subversion assumes a time of
00:00:00, so looking for the most recent revision won't
return anything on the 27th.</p><p>If you want to include the 27th in your search, you can
either specify the 27th with the time (<code class="literal">{"2006-11-27
23:59"}</code>), or just specify the next day
(<code class="literal">{2006-11-28}</code>).</p></div><p>You can also use a range of dates. Subversion will find
all revisions between both dates, inclusive:</p><div class="informalexample"><pre class="screen">
$ svn log -r {2006-11-20}:{2006-11-29}
…
</pre></div><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>Since the timestamp of a revision is stored as an
unversioned, modifiable property of the revision (see <a class="xref" href="#svn.advanced.props" title="Properties">the section called “Properties”</a>), revision timestamps can be
changed to represent complete falsifications of true
chronology, or even removed altogether. Subversion's
ability to correctly convert revision dates into real
revision numbers depends on revision datestamps maintaining
a sequential ordering—the younger the revision, the
younger its timestamp. If this ordering isn't maintained,
you will likely find that trying to use dates to specify
revision ranges in your repository doesn't always return the
data you might have expected.</p></td></tr></table></div></div></div><div class="sect1" title="Peg and Operative Revisions"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.advanced.pegrevs"></a>Peg and Operative Revisions</h2></div></div></div><p>We copy, move, rename, and completely replace files and
directories on our computers all the time. And your version
control system shouldn't get in the way of your doing these
things with your version-controlled files and directories,
either. Subversion's file management support is quite
liberating, affording almost as much flexibility for versioned
files as you'd expect when manipulating your unversioned ones.
But that flexibility means that across the lifetime of your
repository, a given versioned object might have many paths, and
a given path might represent several entirely different
versioned objects. This introduces a certain level of
complexity to your interactions with those paths and
objects.</p><p>Subversion is pretty smart about noticing when an object's
version history includes such <span class="quote">“<span class="quote">changes of address.</span>”</span>
For example, if you ask for the revision history log of a
particular file that was renamed last week, Subversion happily
provides all those logs—the revision in which the rename
itself happened, plus the logs of relevant revisions both before
and after that rename. So, most of the time, you don't even
have to think about such things. But occasionally, Subversion
needs your help to clear up ambiguities.</p><p>The simplest example of this occurs when a directory or file
is deleted from version control, and then a new directory or
file is created with the same name and added to version control.
The thing you deleted and the thing you later added aren't the
same thing. They merely happen to have had the same
path—<code class="filename">/trunk/object</code>, for example.
What, then, does it mean to ask Subversion about the history of
<code class="filename">/trunk/object</code>? Are you asking about the
thing currently at that location, or the old thing you deleted
from that location? Are you asking about the operations that
have happened to <span class="emphasis"><em>all</em></span> the objects that have
ever lived at that path? Subversion needs a hint about what you
really want.</p><p>And thanks to moves, versioned object history can get far
more twisted than even that. For example, you might have a
directory named <code class="filename">concept</code>, containing some
nascent software project you've been toying with. Eventually,
though, that project matures to the point that the idea seems to
actually have some wings, so you do the unthinkable and decide
to give the project a name.<sup>[<a id="idp7797328" href="#ftn.idp7797328" class="footnote">11</a>]</sup>
Let's say you called your software Frabnaggilywort. At this
point, it makes sense to rename the directory to reflect the
project's new name, so <code class="filename">concept</code> is renamed
to <code class="filename">frabnaggilywort</code>. Life goes on,
Frabnaggilywort releases a 1.0 version and is downloaded and
used daily by hordes of people aiming to improve their
lives.</p><p>It's a nice story, really, but it doesn't end there.
Entrepreneur that you are, you've already got another think in
the tank. So you make a new directory,
<code class="filename">concept</code>, and the cycle begins again. In
fact, the cycle begins again many times over the years, each
time starting with that old <code class="filename">concept</code>
directory, then sometimes seeing that directory renamed as the
idea cures, sometimes seeing it deleted when you scrap the idea.
Or, to get really sick, maybe you rename
<code class="filename">concept</code> to something else for a while, but
later rename the thing back to <code class="filename">concept</code> for
some reason.</p><p>In scenarios like these, attempting to instruct
Subversion to work with these reused paths can be a little like
instructing a motorist in Chicago's West Suburbs to drive east
down Roosevelt Road and turn left onto Main Street. In a mere
20 minutes, you can cross <span class="quote">“<span class="quote">Main Street</span>”</span> in
Wheaton, Glen Ellyn, and Lombard. And no, they aren't the same
street. Our motorist—and our Subversion—need a
little more detail to do the right thing.</p><p>Fortunately, Subversion allows you to tell it exactly which
Main Street you meant. The mechanism used is called a
<em class="firstterm">peg revision</em>, and you provide these to
Subversion for the sole purpose of identifying unique lines of
history. Because at most one versioned object may occupy a path
at any given time—or, more precisely, in any one
revision—the combination of a path and a peg revision is
all that is needed to unambiguously identify a specific line
of history. Peg revisions are specified to the Subversion
command-line client using <em class="firstterm">at syntax</em>, so
called because the syntax involves appending an <span class="quote">“<span class="quote">at
sign</span>”</span> (<code class="literal">@</code>) and the peg revision to the
end of the path with which the revision is associated.</p><p>But what of the <code class="option">--revision</code>
(<code class="option">-r</code>) of which we've spoken so much in this
book? That revision (or set of revisions) is called the
<em class="firstterm">operative revision</em> (or
<em class="firstterm">operative revision range</em>). Once a
particular line of history has been identified using a path and
peg revision, Subversion performs the requested operation using
the operative revision(s). To map this to our Chicagoland
streets analogy, if we are told to go to 606 N. Main Street in
Wheaton,<sup>[<a id="idp7810704" href="#ftn.idp7810704" class="footnote">12</a>]</sup> we can think
of <span class="quote">“<span class="quote">Main Street</span>”</span> as our path and
<span class="quote">“<span class="quote">Wheaton</span>”</span> as our peg revision. These two pieces of
information identify a unique path that can be traveled (north or
south on Main Street), and they keep us from traveling up and
down the wrong Main Street in search of our destination. Now we
throw in <span class="quote">“<span class="quote">606 N.</span>”</span> as our operative revision of
sorts, and we know <span class="emphasis"><em>exactly</em></span> where to
go.</p><div class="sidebar" title="The Peg Revision Algorithm"><a id="svn.advanced.pegrevs.algorithm"></a><div class="titlepage"><div><div><p class="title"><strong>The Peg Revision Algorithm</strong></p></div></div></div><p>The Subversion command-line client performs the peg revision
algorithm any time it needs to resolve possible ambiguities in
the paths and revisions provided to it. Here's an example of
such an invocation:</p><div class="informalexample"><pre class="screen">
$ svn <em class="replaceable"><code>command</code></em> -r <em class="replaceable"><code>OPERATIVE-REV</code></em> item@<em class="replaceable"><code>PEG-REV</code></em>
</pre></div><p>If <em class="replaceable"><code>OPERATIVE-REV</code></em> is older than
<em class="replaceable"><code>PEG-REV</code></em>, the algorithm is as
follows:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p>Locate <em class="replaceable"><code>item</code></em> in the revision
identified by <em class="replaceable"><code>PEG-REV</code></em>. There
can be only one such object.</p></li><li class="listitem"><p>Trace the object's history backwards (through any
possible renames) to its ancestor in the revision
<em class="replaceable"><code>OPERATIVE-REV</code></em>.</p></li><li class="listitem"><p>Perform the requested action on that ancestor,
wherever it is located, or whatever its name might
be or might have been at that time.</p></li></ol></div><p>But what if <em class="replaceable"><code>OPERATIVE-REV</code></em> is
<span class="emphasis"><em>younger</em></span> than
<em class="replaceable"><code>PEG-REV</code></em>? Well, that adds some
complexity to the theoretical problem of locating the path in
<em class="replaceable"><code>OPERATIVE-REV</code></em>, because the path's
history could have forked multiple times (thanks to copy
operations) between <em class="replaceable"><code>PEG-REV</code></em> and
<em class="replaceable"><code>OPERATIVE-REV</code></em>. And that's not
all—Subversion doesn't store enough information to
performantly trace an object's history forward, anyway. So
the algorithm is a little different:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p>Locate <em class="replaceable"><code>item</code></em> in the revision
identified by <em class="replaceable"><code>OPERATIVE-REV</code></em>. There
can be only one such object.</p></li><li class="listitem"><p>Trace the object's history backward (through any
possible renames) to its ancestor in the revision
<em class="replaceable"><code>PEG-REV</code></em>.</p></li><li class="listitem"><p>Verify that the object's location (path-wise) in
<em class="replaceable"><code>PEG-REV</code></em> is the same as it is in
<em class="replaceable"><code>OPERATIVE-REV</code></em>. If that's the
case, at least the two locations are known to be
directly related, so perform the requested action on the
location in <em class="replaceable"><code>OPERATIVE-REV</code></em>.
Otherwise, relatedness was not established, so error out
with a loud complaint that no viable location was found.
(Someday, we expect that Subversion will be able to handle
this usage scenario with more flexibility and
grace.)</p></li></ol></div><p>Note that even when you don't explicitly supply a peg
revision or operative revision, they are still present. For
your convenience, the default peg revision is
<code class="literal">BASE</code> for working copy items and
<code class="literal">HEAD</code> for repository URLs. And when no
operative revision is provided, it defaults to being the same
revision as the peg revision.</p></div><p>Say that long ago we created our repository, and in revision 1
we added our first <code class="filename">concept</code> directory, plus an
<code class="filename">IDEA</code> file in that directory talking about
the concept. After several revisions in which real code was
added and tweaked, we, in revision 20, renamed this directory to
<code class="filename">frabnaggilywort</code>. By revision 27, we had a
new concept, a new <code class="filename">concept</code> directory to
hold it, and a new <code class="filename">IDEA</code> file to describe
it. And then five years and thousands of revisions flew by,
just like they would in any good romance story.</p><p>Now, years later, we wonder what the
<code class="filename">IDEA</code> file looked like back in revision 1.
But Subversion needs to know whether we are asking about how the
<span class="emphasis"><em>current</em></span> file looked back in revision 1, or
whether we are asking for the contents of whatever file lived at
<code class="filename">concept/IDEA</code> in revision 1. Certainly
those questions have different answers, and because of peg
revisions, you can ask those questions. To find out how the
current <code class="filename">IDEA</code> file looked in that old
revision, you run:</p><div class="informalexample"><pre class="screen">
$ svn cat -r 1 concept/IDEA
svn: E195012: Unable to find repository location for 'concept/IDEA' in revision 1
</pre></div><p>Of course, in this example, the current
<code class="filename">IDEA</code> file didn't exist yet in revision 1,
so Subversion gives an error. The previous command is shorthand
for a longer notation which explicitly lists a peg revision.
The expanded notation is:</p><div class="informalexample"><pre class="screen">
$ svn cat -r 1 concept/IDEA@BASE
svn: E195012: Unable to find repository location for 'concept/IDEA' in revision 1
</pre></div><p>And when executed, it has the expected results.</p><p>The perceptive reader is probably wondering at this point whether
the peg revision syntax causes problems for working copy paths
or URLs that actually have at signs in them. After
all, how does <span class="command"><strong>svn</strong></span> know whether
<code class="literal">news@11</code> is the name of a directory in my
tree or just a syntax for <span class="quote">“<span class="quote">revision 11 of
<code class="filename">news</code></span>”</span>? Thankfully, while
<span class="command"><strong>svn</strong></span> will always assume the latter, there is a
trivial workaround. You need only append an at sign to the
end of the path, such as <code class="literal">news@11@</code>.
<span class="command"><strong>svn</strong></span> cares only about the last at sign in
the argument, and it is not considered illegal to omit a literal
peg revision specifier after that at sign. This workaround
even applies to paths that end in an at sign—you would
use <code class="literal">filename@@</code> to talk about a file named
<code class="filename">filename@</code>.</p><p>Let's ask the other question, then—in revision 1, what
were the contents of whatever file occupied the address
<code class="filename">concepts/IDEA</code> at the time? We'll use an
explicit peg revision to help us out.</p><div class="informalexample"><pre class="screen">
$ svn cat concept/IDEA@1
The idea behind this project is to come up with a piece of software
that can frab a naggily wort. Frabbing naggily worts is tricky
business, and doing it incorrectly can have serious ramifications, so
we need to employ over-the-top input validation and data verification
mechanisms.
</pre></div><p>Notice that we didn't provide an operative revision this
time. That's because when no operative revision is specified,
Subversion assumes a default operative revision that's the same
as the peg revision.</p><p>As you can see, the output from our operation appears to be
correct. The text even mentions frabbing naggily worts, so this
is almost certainly the file that describes the software now
called Frabnaggilywort. In fact, we can verify this using the
combination of an explicit peg revision and explicit operative
revision. We know that in <code class="literal">HEAD</code>, the
Frabnaggilywort project is located in the
<code class="filename">frabnaggilywort</code> directory. So we specify
that we want to see how the line of history identified in
<code class="literal">HEAD</code> as the path
<code class="filename">frabnaggilywort/IDEA</code> looked in revision
1.</p><div class="informalexample"><pre class="screen">
$ svn cat -r 1 frabnaggilywort/IDEA@HEAD
The idea behind this project is to come up with a piece of software
that can frab a naggily wort. Frabbing naggily worts is tricky
business, and doing it incorrectly can have serious ramifications, so
we need to employ over-the-top input validation and data verification
mechanisms.
</pre></div><p>And the peg and operative revisions need not be so trivial,
either. For example, say <code class="filename">frabnaggilywort</code>
had been deleted from <code class="literal">HEAD</code>, but we know it
existed in revision 20, and we want to see the diffs for its
<code class="filename">IDEA</code> file between revisions 4 and 10. We
can use peg revision 20 in conjunction with the URL that
would have held Frabnaggilywort's <code class="filename">IDEA</code> file
in revision 20, and then use 4 and 10 as our operative revision
range.</p><div class="informalexample"><pre class="screen">
$ svn diff -r 4:10 http://svn.red-bean.com/projects/frabnaggilywort/IDEA@20
Index: frabnaggilywort/IDEA
===================================================================
--- frabnaggilywort/IDEA (revision 4)
+++ frabnaggilywort/IDEA (revision 10)
@@ -1,5 +1,5 @@
-The idea behind this project is to come up with a piece of software
-that can frab a naggily wort. Frabbing naggily worts is tricky
-business, and doing it incorrectly can have serious ramifications, so
-we need to employ over-the-top input validation and data verification
-mechanisms.
+The idea behind this project is to come up with a piece of
+client-server software that can remotely frab a naggily wort.
+Frabbing naggily worts is tricky business, and doing it incorrectly
+can have serious ramifications, so we need to employ over-the-top
+input validation and data verification mechanisms.
</pre></div><p>Fortunately, most folks aren't faced with such complex
situations. But when you are, remember that peg revisions are
that extra hint Subversion needs to clear up ambiguity.</p></div><div class="sect1" title="Properties"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.advanced.props"></a>Properties</h2></div></div></div><a id="idp7872912" class="indexterm"></a><p>We've already covered in detail how Subversion stores and
retrieves various versions of files and directories in its
repository. Whole chapters have been devoted to this most
fundamental piece of functionality provided by the tool. And
if the versioning support stopped there, Subversion would still
be complete from a version control perspective.</p><p>But it doesn't stop there.</p><p>In addition to versioning your directories and files,
Subversion provides interfaces for adding, modifying, and
removing versioned metadata on each of your versioned
directories and files. We refer to this metadata as
<em class="firstterm">properties</em>, and they can be thought of as
two-column tables that map property names to arbitrary values
attached to each item in your working copy. Generally speaking,
the names and values of the properties can be whatever you want
them to be, with the constraint that the names must contain only
ASCII characters. And the best part about these properties is
that they, too, are versioned, just like the textual contents of
your files. You can modify, commit, and revert property changes
as easily as you can file content changes. And the sending and
receiving of property changes occurs as part of your typical
commit and update operations—you don't have to change your
basic processes to accommodate them.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>Subversion has reserved the set of properties whose names
begin with <code class="literal">svn:</code> as its own. While there
are only a handful of such properties in use today, you should
avoid creating custom properties for your own needs whose names
begin with this prefix. Otherwise, you run the risk that a
future release of Subversion will grow support for a feature
or behavior driven by a property of the same name but with
perhaps an entirely different interpretation.</p></td></tr></table></div><p>Properties show up elsewhere in Subversion, too. Just as
files and directories may have arbitrary property names and
values attached to them, each revision as a whole may have
arbitrary properties attached to it. The same constraints
apply—human-readable names and anything-you-want binary
values. The main difference is that revision properties are not
versioned. In other words, if you change the value of, or
delete, a revision property, there's no way, within the scope of
Subversion's functionality, to recover the previous value.</p><p>Subversion has no particular policy regarding the use of
properties. It asks only that you do not use property names that
begin with the prefix <code class="literal">svn:</code> as that's the
namespace that it sets aside for its own use. And Subversion
does, in fact, use properties—both the versioned and
unversioned variety. Certain versioned properties have special
meaning or effects when found on files and directories, or they
house a particular bit of information about the revisions on
which they are found. Certain revision properties are
automatically attached to revisions by Subversion's commit
process, and they carry information about the revision. Most of
these properties are mentioned elsewhere in this or other
chapters as part of the more general topics to which they are
related. For an exhaustive list of Subversion's predefined
properties, see <a class="xref" href="#svn.ref.properties" title="Subversion Properties">the section called “Subversion Properties”</a> in
<a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a>.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>While Subversion automatically attaches properties
(<code class="literal">svn:date</code>, <code class="literal">svn:author</code>,
<code class="literal">svn:log</code>, and so on) to revisions, it does
<span class="emphasis"><em>not</em></span> presume thereafter the existence of
those properties, and neither should you or the tools you use to
interact with your repository. Revision properties can be
deleted programmatically or via the client (if allowed by the
repository hooks) without damaging Subversion's ability to
function. So, when writing scripts which operate on your
Subversion repository data, do not make the mistake of
assuming that any particular revision property exists on a
revision.</p></td></tr></table></div><p>In this section, we will examine the utility—both to
users of Subversion and to Subversion itself—of property
support. You'll learn about the property-related
<span class="command"><strong>svn</strong></span> subcommands and how property
modifications affect your normal Subversion workflow.</p><div class="sect2" title="Why Properties?"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.props.why"></a>Why Properties?</h3></div></div></div><p>Just as Subversion uses properties to store extra
information about the files, directories, and revisions that
it contains, you might also find properties to be of similar
use. You might find it useful to have a place
close to your versioned data to hang custom metadata about
that data.</p><p>Say you wish to design a web site that houses many digital
photos and displays them with captions and a datestamp. Now,
your set of photos is constantly changing, so you'd like to
have as much of this site automated as possible. These photos
can be quite large, so as is common with sites of this nature,
you want to provide smaller thumbnail images to your site
visitors.</p><p>Now, you can get this functionality using traditional
files. That is, you can have your
<code class="filename">image123.jpg</code> and an
<code class="filename">image123-thumbnail.jpg</code> side by side in a
directory. Or if you want to keep the filenames the same, you
might have your thumbnails in a different directory, such as
<code class="filename">thumbnails/image123.jpg</code>. You can also
store your captions and datestamps in a similar fashion, again
separated from the original image file. But the problem here
is that your collection of files multiplies with each new
photo added to the site.</p><p>Now consider the same web site deployed in a way that
makes use of Subversion's file properties. Imagine having a
single image file, <code class="filename">image123.jpg</code>, with
properties set on that file that are named
<code class="literal">caption</code>, <code class="literal">datestamp</code>, and
even <code class="literal">thumbnail</code>. Now your working copy
directory looks much more manageable—in fact, it looks
to the casual browser like there are nothing but image files
in it. But your automation scripts know better. They know
that they can use <span class="command"><strong>svn</strong></span> (or better yet, they
can use the Subversion language bindings—see <a class="xref" href="#svn.developer.usingapi" title="Using the APIs">the section called “Using the APIs”</a>) to dig out the extra
information that your site needs to display without having to
read an index file or play path manipulation games.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>While Subversion places few restrictions on the names
and values you use for properties, it has not been designed
to optimally carry large property values or large sets of
properties on a given file or directory. Subversion
commonly holds all the property names and values associated
with a single item in memory at the same time, which can
cause detrimental performance or failed operations when
extremely large property sets are used.</p></td></tr></table></div><p>Custom revision properties are also frequently used. One
common such use is a property whose value contains an issue
tracker ID with which the revision is associated, perhaps
because the change made in that revision fixes a bug filed in
the tracker issue with that ID. Other uses include hanging
more friendly names on the revision—it might be hard to
remember that revision 1935 was a fully tested revision. But
if there's, say, a <code class="literal">test-results</code> property on
that revision with the value <code class="literal">all passing</code>,
that's meaningful information to have. And Subversion allows
you to easily do this via the <code class="option">--with-revprop</code>
option of the <span class="command"><strong>svn commit</strong></span> command:</p><div class="informalexample"><pre class="screen">
$ svn commit -m "Fix up the last remaining known regression bug." \
--with-revprop "test-results=all passing"
Sending lib/crit_bits.c
Transmitting file data .
Committed revision 912.
$
</pre></div><div class="sidebar" title="Searchability (or, Why Not Properties)"><div class="titlepage"><div><div><p class="title"><strong>Searchability (or, Why <span class="emphasis"><em>Not</em></span>
Properties)</strong></p></div></div></div><p>For all their utility, Subversion properties—or,
more accurately, the available interfaces to them—have
a major shortcoming: while it is a simple matter to
<span class="emphasis"><em>set</em></span> a custom property,
<span class="emphasis"><em>finding</em></span> that property later is a whole
different ball of wax.</p><p>Trying to locate a custom revision property generally
involves performing a linear walk across all the revisions
of the repository, asking of each revision, <span class="quote">“<span class="quote">Do you
have the property I'm looking for?</span>”</span> Use
the <code class="option">--with-all-revprops</code> option with
the <span class="command"><strong>svn log</strong></span> command's XML output mode to
facilitate this search. Notice the presence of the custom
revision property <code class="literal">testresults</code> in the
following output:</p><div class="informalexample"><pre class="screen">
$ svn log --with-all-revprops --xml lib/crit_bits.c
<?xml version="1.0"?>
<log>
<logentry
revision="912">
<author>harry</author>
<date>2011-07-29T14:47:41.169894Z</date>
<msg>Fix up the last remaining known regression bug.</msg>
<revprops>
<property
name="testresults">all passing</property>
</revprops>
</logentry>
…
$
</pre></div><p>Trying to find a custom versioned property is painful,
too, and often involves a recursive <span class="command"><strong>svn
propget</strong></span> across an entire working copy. In your
situation, that might not be as bad as a linear walk across
all revisions. But it certainly leaves much to be desired
in terms of both performance and likelihood of success,
especially if the scope of your search would require a
working copy from the root of your repository.</p><p>For this reason, you might choose—especially in
the revision property use case—to simply add your
metadata to the revision's log message using some
policy-driven (and perhaps programmatically enforced)
formatting that is designed to be quickly parsed from the
output of <span class="command"><strong>svn log</strong></span>. It is quite common to
see the following in Subversion log messages:</p><div class="informalexample"><pre class="programlisting">
Issue(s): IZ2376, IZ1919
Reviewed by: sally
This fixes a nasty segfault in the wort frabbing process
…
</pre></div><p>But here again lies some misfortune. Subversion doesn't
yet provide a log message templating mechanism, which would
go a long way toward helping users be consistent with the
formatting of their log-embedded revision metadata.</p></div></div><div class="sect2" title="Manipulating Properties"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.props.manip"></a>Manipulating Properties</h3></div></div></div><p>The <span class="command"><strong>svn</strong></span> program affords a few ways to
add or modify file and directory properties. For properties
with short, human-readable values, perhaps the simplest way to
add a new property is to specify the property name and value
on the command line of the <span class="command"><strong>svn propset</strong></span>
subcommand:</p><div class="informalexample"><pre class="screen">
$ svn propset copyright '(c) 2006 Red-Bean Software' calc/button.c
property 'copyright' set on 'calc/button.c'
$
</pre></div><p>But we've been touting the flexibility that Subversion
offers for your property values. And if you are planning to
have a multiline textual, or even binary, property value, you
probably do not want to supply that value on the command line.
So the <span class="command"><strong>svn propset</strong></span> subcommand takes a
<code class="option">--file</code> (<code class="option">-F</code>) option for
specifying the name of a file that contains the new property
value.</p><div class="informalexample"><pre class="screen">
$ svn propset license -F /path/to/LICENSE calc/button.c
property 'license' set on 'calc/button.c'
$
</pre></div><p>There are some restrictions on the names you can use for
properties. A property name must start with a letter, a colon
(<code class="literal">:</code>), or an underscore
(<code class="literal">_</code>); after that, you can also use digits,
hyphens (<code class="literal">-</code>), and periods
(<code class="literal">.</code>).<sup>[<a id="idp7930048" href="#ftn.idp7930048" class="footnote">13</a>]</sup></p><p>In addition to the <span class="command"><strong>propset</strong></span> command, the
<span class="command"><strong>svn</strong></span> program supplies the
<span class="command"><strong>propedit</strong></span> command. This command uses the
configured editor program (see <a class="xref" href="#svn.advanced.confarea.opts.config" title="Config">the section called “Config”</a>) to add or
modify properties. When you run the command,
<span class="command"><strong>svn</strong></span> invokes your editor program on a
temporary file that contains the current value of the property
(or that is empty, if you are adding a new property). Then,
you just modify that value in your editor program until it
represents the new value you wish to store for the property,
save the temporary file, and then exit the editor program. If
Subversion detects that you've actually changed the existing
value of the property, it will accept that as the new property
value. If you exit your editor without making any changes, no
property modification will occur:</p><div class="informalexample"><pre class="screen">
$ svn propedit copyright calc/button.c ### exit the editor without changes
No changes to property 'copyright' on 'calc/button.c'
$
</pre></div><p>We should note that, as with other <span class="command"><strong>svn</strong></span>
subcommands, those related to properties can act on multiple
paths at once. This enables you to modify properties on whole
sets of files with a single command. For example, we could
have done the following:</p><div class="informalexample"><pre class="screen">
$ svn propset copyright '(c) 2006 Red-Bean Software' calc/*
property 'copyright' set on 'calc/Makefile'
property 'copyright' set on 'calc/button.c'
property 'copyright' set on 'calc/integer.c'
…
$
</pre></div><p>All of this property adding and editing isn't really very
useful if you can't easily get the stored property value. So
the <span class="command"><strong>svn</strong></span> program supplies two subcommands
for displaying the names and values of properties stored on
files and directories. The <span class="command"><strong>svn proplist</strong></span>
command will list the names of properties that exist on a
path. Once you know the names of the properties on the node,
you can request their values individually using <span class="command"><strong>svn
propget</strong></span>. This command will, given a property name
and a path (or set of paths), print the value of the property
to the standard output stream.</p><div class="informalexample"><pre class="screen">
$ svn proplist calc/button.c
Properties on 'calc/button.c':
copyright
license
$ svn propget copyright calc/button.c
(c) 2006 Red-Bean Software
</pre></div><p>There's even a variation of the
<span class="command"><strong>proplist</strong></span> command that will list both the
name and the value for all of the properties. Simply supply the
<code class="option">--verbose</code> (<code class="option">-v</code>) option.</p><div class="informalexample"><pre class="screen">
$ svn proplist -v calc/button.c
Properties on 'calc/button.c':
copyright
(c) 2006 Red-Bean Software
license
================================================================
Copyright (c) 2006 Red-Bean Software. All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
are met:
1. Redistributions of source code must retain the above copyright
notice, this list of conditions, and the recipe for Fitz's famous
red-beans-and-rice.
…
</pre></div><p>The last property-related subcommand is
<span class="command"><strong>propdel</strong></span>. Since Subversion allows you to
store properties with empty values, you can't remove a
property altogether using <span class="command"><strong>svn propedit</strong></span> or
<span class="command"><strong>svn propset</strong></span>. For example, this command will
<span class="emphasis"><em>not</em></span> yield the desired effect:</p><div class="informalexample"><pre class="screen">
$ svn propset license "" calc/button.c
property 'license' set on 'calc/button.c'
$ svn proplist -v calc/button.c
Properties on 'calc/button.c':
copyright
(c) 2006 Red-Bean Software
license
$
</pre></div><p>You need to use the <span class="command"><strong>propdel</strong></span> subcommand
to delete properties altogether. The syntax is similar to the
other property commands:</p><div class="informalexample"><pre class="screen">
$ svn propdel license calc/button.c
property 'license' deleted from 'calc/button.c'.
$ svn proplist -v calc/button.c
Properties on 'calc/button.c':
copyright
(c) 2006 Red-Bean Software
$
</pre></div><p>Remember those unversioned revision properties? You can
modify those, too, using the same <span class="command"><strong>svn</strong></span>
subcommands that we just described. Simply add the
<code class="option">--revprop</code> command-line parameter and specify
the revision whose property you wish to modify. Since
revisions are global, you don't need to specify a target path
to these property-related commands so long as you are
positioned in a working copy of the repository whose revision
property you wish to modify. Otherwise, you can simply
provide the URL of any path in the repository of interest
(including the repository's root URL). For example, you might
want to replace the commit log message of an existing
revision.<sup>[<a id="idp7956704" href="#ftn.idp7956704" class="footnote">14</a>]</sup> If
your current working directory is part of a working copy of
your repository, you can simply run the
<span class="command"><strong>svn propset</strong></span> command with no target path:</p><div class="informalexample"><pre class="screen">
$ svn propset svn:log "* button.c: Fix a compiler warning." -r11 --revprop
property 'svn:log' set on repository revision '11'
$
</pre></div><p>But even if you haven't checked out a working copy from
that repository, you can still effect the property change by
providing the repository's root URL:</p><div class="informalexample"><pre class="screen">
$ svn propset svn:log "* button.c: Fix a compiler warning." -r11 --revprop \
http://svn.example.com/repos/project
property 'svn:log' set on repository revision '11'
$
</pre></div><p>Note that the ability to modify these unversioned
properties must be explicitly added by the repository
administrator (see <a class="xref" href="#svn.reposadmin.maint.setlog" title="Commit Log Message Correction">the section called “Commit Log Message Correction”</a>).
That's because the properties aren't versioned, so you run the risk of
losing information if you aren't careful with your edits.
The repository administrator can set up methods to protect
against this loss, and by default, modification of
unversioned properties is disabled.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>Users should, where possible, use <span class="command"><strong>svn
propedit</strong></span> instead of <span class="command"><strong>svn
propset</strong></span>. While the end result of the commands is
identical, the former will allow them to see the current
value of the property that they are about to change, which helps
them to verify that they are, in fact, making the change
they think they are making. This is especially true when
modifying unversioned revision properties. Also, it is
significantly easier to modify multiline property values in
a text editor than at the command line.</p></td></tr></table></div></div><div class="sect2" title="Properties and the Subversion Workflow"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.props.workflow"></a>Properties and the Subversion Workflow</h3></div></div></div><p>Now that you are familiar with all of the
property-related <span class="command"><strong>svn</strong></span> subcommands, let's see
how property modifications affect the usual Subversion
workflow. As we mentioned earlier, file and directory
properties are versioned, just like your file contents. As a
result, Subversion provides the same opportunities for
merging—cleanly or with conflicts—someone
else's modifications into your own.</p><p>As with file contents, your property changes are local
modifications, made permanent only when you commit them to the
repository with <span class="command"><strong>svn commit</strong></span>. Your property
changes can be easily unmade, too—the <span class="command"><strong>svn
revert</strong></span> command will restore your files and
directories to their unedited states—contents, properties,
and all. Also, you can receive interesting information about
the state of your file and directory properties by using the
<span class="command"><strong>svn status</strong></span> and <span class="command"><strong>svn diff</strong></span>
commands.</p><div class="informalexample"><pre class="screen">
$ svn status calc/button.c
M calc/button.c
$ svn diff calc/button.c
Property changes on: calc/button.c
___________________________________________________________________
Added: copyright
## -0,0 +1 ##
+(c) 2006 Red-Bean Software
$
</pre></div><p>Notice how the <span class="command"><strong>status</strong></span> subcommand
displays <code class="literal">M</code> in the second column instead of
the first. That is because we have modified the properties on
<code class="filename">calc/button.c</code>, but not its textual
contents. Had we changed both, we would have seen
<code class="literal">M</code> in the first column, too. (We cover
<span class="command"><strong>svn status</strong></span> in <a class="xref" href="#svn.tour.cycle.examine.status" title="See an overview of your changes">the section called “See an overview of your changes”</a>).</p><div class="sidebar" title="Property Conflicts"><div class="titlepage"><div><div><p class="title"><strong>Property Conflicts</strong></p></div></div></div><p>As with file contents, local property modifications can
conflict with changes committed by someone else. If you
update your working copy directory and receive property
changes on a versioned object that clash with your own,
Subversion will report that the object is in a conflicted
state.</p><div class="informalexample"><pre class="screen">
$ svn update calc
Updating 'calc':
M calc/Makefile.in
Conflict for property 'linecount' discovered on 'calc/button.c'.
Select: (p) postpone, (df) diff-full, (e) edit,
(s) show all options: p
C calc/button.c
Updated to revision 143.
Summary of conflicts:
Property conflicts: 1
$
</pre></div><p>Subversion will also create, in the same directory as
the conflicted object, a file with a
<code class="filename">.prej</code> extension that contains the
details of the conflict. You should examine the contents of
this file so you can decide how to resolve the conflict.
Until the conflict is resolved, you will see a
<code class="literal">C</code> in the second column of <span class="command"><strong>svn
status</strong></span> output for that object, and attempts to
commit your local modifications will fail.</p><div class="informalexample"><pre class="screen">
$ svn status calc
C calc/button.c
? calc/button.c.prej
$ cat calc/button.c.prej
Trying to change property 'linecount' from '1267' to '1301',
but property has been locally changed from '1267' to '1256'.
$
</pre></div><p>To resolve property conflicts, simply ensure that the
conflicting properties contain the values that they should,
and then use the <span class="command"><strong>svn resolve --accept=working</strong></span>
command to alert Subversion that you have manually resolved the
problem.</p></div><p>You might also have noticed the nonstandard way that
Subversion currently displays property differences. You can
still use <span class="command"><strong>svn diff</strong></span> and redirect its output
to create a usable patch file. The <span class="command"><strong>patch</strong></span>
program will ignore property patches—as a rule, it
ignores any noise it can't understand. This does,
unfortunately, mean that to fully apply a patch generated by
<span class="command"><strong>svn diff</strong></span> using <span class="command"><strong>patch</strong></span>,
any property modifications will need to be applied by
hand.</p><p>Subversion 1.7 improves this situation in two ways.
First, its nonstandard display of property differences is at
least machine-readable—an improvement over the display
of properties in versions prior to 1.7. But Subversion 1.7
also introduces the <span class="command"><strong>svn patch</strong></span> subcommand,
designed specifically to handle the additional information
which <span class="command"><strong>svn diff</strong></span>'s output can carry, applying
those changes to the Subversion working copy. Of specific
relevance to our topic, property differences present in patch
files generated by <span class="command"><strong>svn diff</strong></span> in Subversion
1.7 or better can be automatically applied to a working copy
by the <span class="command"><strong>svn patch</strong></span> command. For more
about <span class="command"><strong>svn patch</strong></span>, see
<a class="xref" href="#svn.ref.svn.c.patch" title="svn patch">svn patch</a> in
<a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a>.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>There's one exception to how property changes are
reported by <span class="command"><strong>svn diff</strong></span>: changes to
Subversion's special <code class="literal">svn:mergeinfo</code>
property—used to track information about merges which
have been performed in your repository—are described
in a more human-readable fashion. This is quite helpful to
the humans who have to read those descriptions. But it also
serves to cause patching programs (including <span class="command"><strong>svn
patch</strong></span>) to skip those change descriptions as noise.
This might sound like a bug, but it really isn't because
this property is intended to be managed solely by
the <span class="command"><strong>svn merge</strong></span> subcommand. For more about
merge tracking, see <a class="xref" href="#svn.branchmerge" title="Chapter 4. Branching and Merging">Chapter 4, <em>Branching and Merging</em></a>.</p></td></tr></table></div></div><div class="sect2" title="Automatic Property Setting"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.props.auto"></a>Automatic Property Setting</h3></div></div></div><p>Properties are a powerful feature of Subversion, acting as
key components of many Subversion features discussed elsewhere
in this and other chapters—textual diff and merge
support, keyword substitution, newline translation, and so on. But
to get the full benefit of properties, they must be set on the
right files and directories. Unfortunately, that
step can be easily forgotten in the routine of things, especially
since failing to set a property doesn't usually result in an
obvious error (at least compared to, say, failing to
add a file to version control). To help your properties get
applied to the places that need them, Subversion provides a
couple of simple but useful features.</p><p>Whenever you introduce a file to version control using the
<span class="command"><strong>svn add</strong></span> or <span class="command"><strong>svn import</strong></span>
commands, Subversion tries to assist by setting some common
file properties automatically. First, on operating systems
whose filesystems support an execute permission bit,
Subversion will automatically set the
<code class="literal">svn:executable</code> property on newly added or
imported files whose execute bit is enabled. (See <a class="xref" href="#svn.advanced.props.special.executable" title="File Executability">the section called “File Executability”</a> later in
this chapter for more about this property.)</p><p>Second, Subversion tries to determine the file's MIME
type. If you've configured a
<code class="literal">mime-types-files</code> runtime configuration
parameter, Subversion will try to find a MIME type mapping in
that file for your file's extension. If it finds such a
mapping, it will set your file's
<code class="literal">svn:mime-type</code> property to the MIME type it
found. If no mapping file is configured, or no mapping for
your file's extension could be found, Subversion will fall
back to heuristic algorithms to determine the file's MIME
type. Depending on how it is built, Subversion 1.7 can make
use of file scanning libraries<sup>[<a id="idp8012000" href="#ftn.idp8012000" class="footnote">15</a>]</sup> to detect a file's type based on its
content. Failing all else, Subversion will employ its own
very basic heuristic to determine whether the file contains
nontextual content. If so, it automatically sets the
<code class="literal">svn:mime-type</code> property on that file to
<code class="literal">application/octet-stream</code> (the generic
<span class="quote">“<span class="quote">this is a collection of bytes</span>”</span> MIME type). Of
course, if Subversion guesses incorrectly, or if you wish to
set the <code class="literal">svn:mime-type</code> property to something
more precise—perhaps <code class="literal">image/png</code> or
<code class="literal">application/x-shockwave-flash</code>—you can
always remove or edit that property. (For more on
Subversion's use of MIME types, see <a class="xref" href="#svn.advanced.props.special.mime-type" title="File Content Type">the section called “File Content Type”</a> later in
this chapter.)</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>UTF-16 is commonly used to encode files whose semantic
content is textual in nature, but the encoding itself makes
heavy use of bytes which are outside the typical ASCII
character byte range. As such, Subversion will tend to
classify such files as binary files, much to the chagrin of
users who desire line-based differencing and merging,
keyword substitution, and other behaviors for those
files.</p></td></tr></table></div><p>Subversion also provides, via its runtime configuration
system (see <a class="xref" href="#svn.advanced.confarea" title="Runtime Configuration Area">the section called “Runtime Configuration Area”</a>), a more
flexible automatic property setting feature that allows you
to create mappings of filename patterns to property names and
values. Once again, these mappings affect adds and imports,
and can not only override the default MIME type decision made
by Subversion during those operations, but can also set
additional Subversion or custom properties, too. For example,
you might create a mapping that says that anytime you add
JPEG files—ones whose names match the pattern
<code class="literal">*.jpg</code>—Subversion should automatically
set the <code class="literal">svn:mime-type</code> property on those
files to <code class="literal">image/jpeg</code>. Or perhaps any files
that match <code class="literal">*.cpp</code> should have
<code class="literal">svn:eol-style</code> set to
<code class="literal">native</code>, and <code class="literal">svn:keywords</code>
set to <code class="literal">Id</code>. Automatic property support is
perhaps the handiest property-related tool in the Subversion
toolbox. See <a class="xref" href="#svn.advanced.confarea.opts.config" title="Config">the section called “Config”</a> for more about
configuring that support.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>Subversion administrators commonly ask if it is possible
to configure, on the server side, a set of property
definitions which all connecting clients will automatically
consider when operating on working copies checked out from
that server. Unfortunately, Subversion doesn't offer this
feature. Administrators can use hook scripts to validate
that the properties added to and modified on files and
directories match the administrator's preferred policies,
rejecting commits which are non-compliant in this fashion.
(See <a class="xref" href="#svn.reposadmin.create.hooks" title="Implementing Repository Hooks">the section called “Implementing Repository Hooks”</a> for more
about hook scripts.) But there's no way to automatically
dictate those preferences to Subversion clients
beforehand.</p></td></tr></table></div></div></div><div class="sect1" title="File Portability"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.advanced.props.file-portability"></a>File Portability</h2></div></div></div><p>Fortunately for Subversion users who routinely find
themselves on different computers with different operating
systems, Subversion's command-line program behaves almost
identically on all those systems. If you know how to wield
<span class="command"><strong>svn</strong></span> on one platform, you know how to wield it
everywhere.</p><p>However, the same is not always true of other general classes
of software or of the actual files you keep in Subversion. For
example, on a Windows machine, the definition of a <span class="quote">“<span class="quote">text
file</span>”</span> would be similar to that used on a Linux box, but
with a key difference—the character sequences used to mark
the ends of the lines of those files. There are other
differences, too. Unix platforms have (and Subversion supports)
symbolic links; Windows does not. Unix platforms use filesystem
permission to determine executability; Windows uses filename
extensions.</p><p>Because Subversion is in no position to unite the whole
world in common definitions and implementations of all of these
things, the best it can do is to try to help make your life
simpler when you need to work with your versioned files and
directories on multiple computers and operating systems. This
section describes some of the ways Subversion does this.</p><div class="sect2" title="File Content Type"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.props.special.mime-type"></a>File Content Type</h3></div></div></div><p>Subversion joins the ranks of the many applications that
recognize and make use of Multipurpose Internet Mail
Extensions (MIME) content types. Besides being a
general-purpose storage location for a file's content type,
the value of the <code class="literal">svn:mime-type</code> file
property determines some behavioral characteristics of
Subversion itself.</p><div class="sidebar" title="Identifying File Types"><div class="titlepage"><div><div><p class="title"><strong>Identifying File Types</strong></p></div></div></div><p>Various programs on most modern operating systems make
assumptions about the type and format of the contents of a
file by the file's name, specifically its file extension.
For example, files whose names end in
<code class="filename">.txt</code> are generally assumed to be
human-readable; that is, able to be understood by simple perusal
rather than requiring complex processing to decipher. Files
whose names end in <code class="filename">.png</code>, on the other
hand, are assumed to be of the Portable Network Graphics
type—not human-readable at all, and sensible only when
interpreted by software that understands the PNG format and
can render the information in that format as a raster
image.</p><p>Unfortunately, some of those extensions have changed
their meanings over time. When personal computers first appeared,
a file named <code class="filename">README.DOC</code> would have
almost certainly been a plain-text file, just like today's
<code class="filename">.txt</code> files. But by the mid-1990s, you
could almost bet that a file of that name would not be a
plain-text file at all, but instead a Microsoft Word
document in a proprietary, non-human-readable format. But
this change didn't occur overnight—there was certainly
a period of confusion for computer users over what exactly
they had in hand when they saw a <code class="filename">.DOC</code>
file.<sup>[<a id="idp8045712" href="#ftn.idp8045712" class="footnote">16</a>]</sup></p><p>The popularity of computer networking cast still more
doubt on the mapping between a file's name and its content.
With information being served across networks and generated
dynamically by server-side scripts, there was often no real
file per se, and therefore no filename. Web
servers, for example, needed some other way to tell browsers
what they were downloading so that the browser could do something
intelligent with that information, whether that was to
display the data using a program registered to handle that
datatype or to prompt the user for where on the client
machine to store the downloaded data.</p><p>Eventually, a standard emerged for, among other things,
describing the contents of a data stream. In 1996, RFC 2045
was published. It was the first of five RFCs describing
MIME. It describes the concept of media types and subtypes
and recommends a syntax for the representation of those
types. Today, MIME media types—or <span class="quote">“<span class="quote">MIME
types</span>”</span>—are used almost universally across
email applications, web servers, and other software as the
de facto mechanism for clearing up the file content
confusion.</p></div><p>For example, one of the benefits that Subversion typically
provides is contextual, line-based merging of changes received
from the server during an update into your working file. But
for files containing nontextual data, there is often no
concept of a <span class="quote">“<span class="quote">line.</span>”</span> So, for versioned files
whose <code class="literal">svn:mime-type</code> property is set to a
nontextual MIME type (generally, something that doesn't begin
with <code class="literal">text/</code>, though there are exceptions),
Subversion does not attempt to perform contextual merges
during updates. Instead, any time you have locally modified a
binary working copy file that is also being updated, your file
is left untouched and Subversion creates two new files. One
file has a <code class="filename">.oldrev</code> extension and contains
the BASE revision of the file. The other file has a
<code class="filename">.newrev</code> extension and contains the
contents of the updated revision of the file. This behavior
is really for the protection of the user against failed
attempts at performing contextual merges on files that simply
cannot be contextually merged.</p><p>Additionally, since the acts of displaying line-based
differences and line-based change attribution are, rather
obviously, dependent on there being a meaningful definition
of <span class="quote">“<span class="quote">line</span>”</span> for a given file, files with nontextual
MIME types will by default trigger errors when used as the
targets of <span class="command"><strong>svn diff</strong></span> and <span class="command"><strong>svn
annotate</strong></span> operations. This can be especially
frustrating for users with XML files whose
<code class="literal">svn:mime-type</code> property is set to something
such as <code class="literal">application/xml</code> which is not
unambiguously human-readable and as such is treated as
nontextual by Subversion. Fortunately, those subcommands
offer a <code class="option">--force</code> option for forcing
Subversion to attempt the operations in spite of the apparent
non-human-readability of the files.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>The <code class="literal">svn:mime-type</code> property, when set
to a value that does not indicate textual file contents, can
cause some unexpected behaviors with respect to other
properties. For example, since the idea of line endings
(and therefore, line-ending conversion) makes no sense when
applied to nontextual files, Subversion will prevent you
from setting the <code class="literal">svn:eol-style</code> property
on such files. This is obvious when attempted on a single
file target—<span class="command"><strong>svn propset</strong></span> will error
out. But it might not be as clear if you perform a
recursive property set, where Subversion will silently skip
over files that it deems unsuitable for a given
property.</p></td></tr></table></div><p>Subversion provides a number of mechanisms by which to
automatically set the <code class="literal">svn:mime-type</code>
property on a versioned file. See
<a class="xref" href="#svn.advanced.props.auto" title="Automatic Property Setting">the section called “Automatic Property Setting”</a> for details.</p><p>Also, if the <code class="literal">svn:mime-type</code> property is
set, then the Subversion Apache module will use its value to
populate the <code class="literal">Content-type:</code> HTTP header when
responding to GET requests. This gives your web browser a
crucial clue about how to display a file when you use it to
peruse your Subversion repository's contents.</p></div><div class="sect2" title="File Executability"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.props.special.executable"></a>File Executability</h3></div></div></div><p>On many operating systems, the ability to execute a file
as a command is governed by the presence of an execute
permission bit. This bit usually defaults to being disabled,
and must be explicitly enabled by the user for each file that
needs it. But it would be a monumental hassle to have to
remember exactly which files in a freshly checked-out working
copy were supposed to have their executable bits toggled on,
and then to have to do that toggling. So, Subversion provides
the <code class="literal">svn:executable</code> property as a way to
specify that the executable bit for the file on which that
property is set should be enabled, and Subversion honors that
request when populating working copies with such files.</p><p>This property has no effect on filesystems that have no
concept of an executable permission bit, such as FAT32 and
NTFS.<sup>[<a id="idp7677888" href="#ftn.idp7677888" class="footnote">17</a>]</sup> Also, although it has no defined
values, Subversion will force its value
to <code class="literal">*</code> when setting this property. Finally,
this property is valid only on files, not on
directories.</p></div><div class="sect2" title="End-of-Line Character Sequences"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.props.special.eol-style"></a>End-of-Line Character Sequences</h3></div></div></div><p>Unless otherwise noted using a versioned file's
<code class="literal">svn:mime-type</code> property, Subversion
assumes the file contains human-readable data. Generally
speaking, Subversion uses this knowledge only to determine
whether contextual difference reports for that file are
possible. Otherwise, to Subversion, bytes are bytes.</p><p>This means that by default, Subversion doesn't pay any
attention to the type of <em class="firstterm">end-of-line (EOL)
markers</em> used in your files. Unfortunately,
different operating systems have different conventions about
which character sequences represent the end of a line of text
in a file. For example, the usual line-ending token used by
software on the Windows platform is a pair of ASCII control
characters—a carriage return (<code class="literal">CR</code>)
followed by a line feed (<code class="literal">LF</code>). Unix
software, however, just uses the <code class="literal">LF</code>
character to denote the end of a line.</p><p>Not all of the various tools on these operating systems
understand files that contain line endings in a format that
differs from the <em class="firstterm">native line-ending
style</em> of the operating system on which they are
running. So, typically, Unix programs treat the
<code class="literal">CR</code> character present in Windows files as a
regular character (usually rendered as <code class="literal">^M</code>),
and Windows programs combine all of the lines of a Unix file
into one giant line because no carriage return-linefeed (or
<code class="literal">CRLF</code>) character combination was found to
denote the ends of the lines.</p><p>This sensitivity to foreign EOL markers can be
frustrating for folks who share a file across different
operating systems. For example, consider a source code
file, and developers who edit this file on both Windows and
Unix systems. If all the developers always use tools that
preserve the line-ending style of the file, no problems
occur.</p><p>But in practice, many common tools either fail to
properly read a file with foreign EOL markers, or
convert the file's line endings to the native style when the
file is saved. If the former is true for a developer, he
has to use an external conversion utility (such as
<span class="command"><strong>dos2unix</strong></span> or its companion,
<span class="command"><strong>unix2dos</strong></span>) to prepare the file for
editing. The latter case requires no extra preparation.
But both cases result in a file that differs from the
original quite literally on every line! Prior to committing
his changes, the user has two choices. Either he can use a
conversion utility to restore the modified file to the same
line-ending style that it was in before his edits were made,
or he can simply commit the file—new EOL markers and
all.</p><p>The result of scenarios like these include wasted time
and unnecessary modifications to committed files. Wasted
time is painful enough. But when commits change every line
in a file, this complicates the job of determining which of
those lines were changed in a nontrivial way. Where was
that bug really fixed? On what line was a syntax error
introduced?</p><p>The solution to this problem is the
<code class="literal">svn:eol-style</code> property. When this
property is set to a valid value, Subversion uses it to
determine what special processing to perform on the file so
that the file's line-ending style isn't flip-flopping with
every commit that comes from a different operating
system. The valid values are:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">native</code></span></dt><dd><p>This causes the file to contain the EOL markers
that are native to the operating system on which
Subversion was run. In other words, if a user on a
Windows machine checks out a working copy that
contains a file with an
<code class="literal">svn:eol-style</code> property set to
<code class="literal">native</code>, that file will contain
<code class="literal">CRLF</code> EOL markers. A Unix user
checking out a working copy that contains the same
file will see <code class="literal">LF</code> EOL markers in his
copy of the file.</p><p>Note that Subversion will actually store the file
in the repository using normalized
<code class="literal">LF</code> EOL markers regardless of the
operating system. This is basically transparent to
the user, though.</p></dd><dt><span class="term"><code class="literal">CRLF</code></span></dt><dd><p>This causes the file to contain
<code class="literal">CRLF</code> sequences for EOL markers,
regardless of the operating system in use.</p></dd><dt><span class="term"><code class="literal">LF</code></span></dt><dd><p>This causes the file to contain
<code class="literal">LF</code> characters for EOL markers,
regardless of the operating system in use.</p></dd><dt><span class="term"><code class="literal">CR</code></span></dt><dd><p>This causes the file to contain
<code class="literal">CR</code> characters for EOL markers,
regardless of the operating system in use. This
line-ending style is not very common.</p></dd></dl></div></div></div><div class="sect1" title="Ignoring Unversioned Items"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.advanced.props.special.ignore"></a>Ignoring Unversioned Items</h2></div></div></div><p>In any given working copy, there is a good chance that
alongside all those versioned files and directories are other
files and directories that are neither versioned nor intended
to be. Text editors litter directories with backup files.
Software compilers generate intermediate—or even
final—files that you typically wouldn't bother to
version. And users themselves drop various other files and
directories wherever they see fit, often in version control
working copies.</p><p>It's ludicrous to expect Subversion working copies to be
somehow impervious to this kind of clutter and impurity. In
fact, Subversion counts it as a <span class="emphasis"><em>feature</em></span>
that its working copies are just typical directories, just like
unversioned trees. But these not-to-be-versioned files and
directories can cause some annoyance for Subversion users. For
example, because the <span class="command"><strong>svn add</strong></span> and <span class="command"><strong>svn
import</strong></span> commands act recursively by default and don't
know which files in a given tree you do and don't wish to
version, it's easy to accidentally add stuff to version control
that you didn't mean to. And because <span class="command"><strong>svn
status</strong></span> reports, by default, every item of interest in
a working copy—including unversioned files and
directories—its output can get quite noisy where many of
these things exist.</p><p>So Subversion provides two ways for telling it which files
you would prefer that it simply disregard. One of the ways
involves the use of Subversion's runtime configuration system
(see <a class="xref" href="#svn.advanced.confarea" title="Runtime Configuration Area">the section called “Runtime Configuration Area”</a>), and therefore
applies to all the Subversion operations that make use of that
runtime configuration—generally those performed on a particular
computer or by a particular user of a computer. The other way
makes use of Subversion's directory property support and is more
tightly bound to the versioned tree itself, and therefore
affects everyone who has a working copy of that tree. Both of
the mechanisms use <em class="firstterm">file patterns</em> (strings
of literal and special wildcard characters used to match against
filenames) to decide which files to ignore.</p><p>The Subversion runtime configuration system provides an
option, <code class="literal">global-ignores</code>, whose value is a
whitespace-delimited collection of file patterns. The
Subversion client checks these patterns against the names of the
files that are candidates for addition to version control, as
well as to unversioned files that the <span class="command"><strong>svn
status</strong></span> command notices. If any file's name matches
one of the patterns, Subversion will basically act as if the
file didn't exist at all. This is really useful for the kinds
of files that you almost never want to version, such as editor
backup files such as Emacs' <code class="literal">*~</code> and
<code class="literal">.*~</code> files.</p><div class="sidebar" title="File Patterns in Subversion"><div class="titlepage"><div><div><p class="title"><strong>File Patterns in Subversion</strong></p></div></div></div><p>File patterns (also called <em class="firstterm">globs</em> or
<em class="firstterm">shell wildcard patterns</em>) are strings of
characters that are intended to be matched against filenames,
typically for the purpose of quickly selecting some subset of
similar files from a larger grouping without having to
explicitly name each file. The patterns contain two types of
characters: regular characters, which are compared explicitly
against potential matches, and special wildcard characters,
which are interpreted differently for matching
purposes.</p><p>There are different types of file pattern syntaxes, but
Subversion uses the one most commonly found in Unix systems
implemented as the <code class="function">fnmatch</code> system
function. It supports the following wildcards, described here
simply for your convenience:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">?</code></span></dt><dd><p>Matches any single character</p></dd><dt><span class="term"><code class="literal">*</code></span></dt><dd><p>Matches any string of characters, including the
empty string</p></dd><dt><span class="term"><code class="literal">[</code></span></dt><dd><p>Begins a character class definition terminated by
<code class="literal">]</code>, used for matching a subset of
characters</p></dd></dl></div><p>You can see this same pattern matching behavior at a Unix
shell prompt. The following are some examples of patterns
being used for various things:</p><div class="informalexample"><pre class="screen">
$ ls ### the book sources
appa-quickstart.xml ch06-server-configuration.xml
appb-svn-for-cvs-users.xml ch07-customizing-svn.xml
appc-webdav.xml ch08-embedding-svn.xml
book.xml ch09-reference.xml
ch00-preface.xml ch10-world-peace-thru-svn.xml
ch01-fundamental-concepts.xml copyright.xml
ch02-basic-usage.xml foreword.xml
ch03-advanced-topics.xml images/
ch04-branching-and-merging.xml index.xml
ch05-repository-admin.xml styles.css
$ ls ch* ### the book chapters
ch00-preface.xml ch06-server-configuration.xml
ch01-fundamental-concepts.xml ch07-customizing-svn.xml
ch02-basic-usage.xml ch08-embedding-svn.xml
ch03-advanced-topics.xml ch09-reference.xml
ch04-branching-and-merging.xml ch10-world-peace-thru-svn.xml
ch05-repository-admin.xml
$ ls ch?0-* ### the book chapters whose numbers end in zero
ch00-preface.xml ch10-world-peace-thru-svn.xml
$ ls ch0[3578]-* ### the book chapters that Mike is responsible for
ch03-advanced-topics.xml ch07-customizing-svn.xml
ch05-repository-admin.xml ch08-embedding-svn.xml
$
</pre></div><p>File pattern matching is a bit more complex than what
we've described here, but this basic usage level tends to suit
the majority of Subversion users.</p></div><p>When found on a versioned directory, the
<code class="literal">svn:ignore</code> property is expected to contain a
list of newline-delimited file patterns that Subversion should
use to determine ignorable objects in that same directory.
These patterns do not override those found in the
<code class="literal">global-ignores</code> runtime configuration option,
but are instead appended to that list. And it's worth noting
again that, unlike the <code class="literal">global-ignores</code> option,
the patterns found in the <code class="literal">svn:ignore</code>
property apply only to the directory on which that property is
set, and not to any of its subdirectories. The
<code class="literal">svn:ignore</code> property is a good way to tell
Subversion to ignore files that are likely to be present in
every user's working copy of that directory, such as compiler
output or—to use an example more appropriate to this
book—the HTML, PDF, or PostScript files generated as the
result of a conversion of some source DocBook XML files to a
more legible output format.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>Subversion's support for ignorable file patterns extends
only to the one-time process of adding unversioned
files and directories to version control. Once an object is
under Subversion's control, the ignore pattern mechanisms no
longer apply to it. In other words, don't expect Subversion
to avoid committing changes you've made to a versioned file
simply because that file's name matches an ignore
pattern—Subversion <span class="emphasis"><em>always</em></span> notices
all of its versioned objects.</p></td></tr></table></div><div class="sidebar" title="Ignore Patterns for CVS Users"><div class="titlepage"><div><div><p class="title"><strong>Ignore Patterns for CVS Users</strong></p></div></div></div><p>The Subversion <code class="literal">svn:ignore</code> property is
very similar in syntax and function to the CVS
<code class="filename">.cvsignore</code> file. In fact, if you are
migrating a CVS working copy to Subversion, you can directly
migrate the ignore patterns by using the
<code class="filename">.cvsignore</code> file as input file to the
<span class="command"><strong>svn propset</strong></span> command:</p><div class="informalexample"><pre class="screen">
$ svn propset svn:ignore -F .cvsignore .
property 'svn:ignore' set on '.'
$
</pre></div><p>There are, however, some differences in the ways that CVS
and Subversion handle ignore patterns. The two systems use
the ignore patterns at some different times, and there are
slight discrepancies in what the ignore patterns apply to.
Also, Subversion does not recognize the use of the
<code class="literal">!</code> pattern as a reset back to having no
ignore patterns at all.</p></div><p>The global list of ignore patterns tends to be more a
matter of personal taste and ties more closely to a user's
particular tool chain than to the details of any particular
working copy's needs. So, the rest of this section will focus
on the <code class="literal">svn:ignore</code> property and its
uses.</p><p>Say you have the following output from <span class="command"><strong>svn
status</strong></span>:</p><div class="informalexample"><pre class="screen">
$ svn status calc
M calc/button.c
? calc/calculator
? calc/data.c
? calc/debug_log
? calc/debug_log.1
? calc/debug_log.2.gz
? calc/debug_log.3.gz
</pre></div><p>In this example, you have made some property modifications
to <code class="filename">button.c</code>, but in your working copy, you
also have some unversioned files: the latest
<code class="filename">calculator</code> program that you've compiled
from your source code, a source file named
<code class="filename">data.c</code>, and a set of debugging output
logfiles. Now, you know that your build system always results
in the <code class="filename">calculator</code> program being
generated.<sup>[<a id="idp8162720" href="#ftn.idp8162720" class="footnote">18</a>]</sup> And you know that your test suite
always leaves those debugging logfiles lying around. These
facts are true for all working copies of this project, not just
your own. And you know that you aren't interested in seeing
those things every time you run
<span class="command"><strong>svn status</strong></span>, and you are pretty sure that
nobody else is interested in them either. So you use
<strong class="userinput"><code>svn propedit svn:ignore calc</code></strong> to add some
ignore patterns to the <code class="filename">calc</code>
directory.</p><div class="informalexample"><pre class="screen">
$ svn propget svn:ignore calc
calculator
debug_log*
$
</pre></div><p>After you've added this property, you will now have a local
property modification on the <code class="filename">calc</code>
directory. But notice what else is different about your
<span class="command"><strong>svn status</strong></span> output:</p><div class="informalexample"><pre class="screen">
$ svn status
M calc
M calc/button.c
? calc/data.c
</pre></div><p>Now, all that cruft is missing from the output! Your
<code class="filename">calculator</code> compiled program and all those
logfiles are still in your working copy; Subversion just isn't
constantly reminding you that they are present and unversioned.
And now with all the uninteresting noise removed from the
display, you are left with more intriguing items—such as
that source code file <code class="filename">data.c</code> that you
probably forgot to add to version control.</p><p>Of course, this less-verbose report of your working copy
status isn't the only one available. If you actually want to
see the ignored files as part of the status report, you can pass
the <code class="option">--no-ignore</code> option to Subversion:</p><div class="informalexample"><pre class="screen">
$ svn status --no-ignore
M calc
M calc/button.c
I calc/calculator
? calc/data.c
I calc/debug_log
I calc/debug_log.1
I calc/debug_log.2.gz
I calc/debug_log.3.gz
</pre></div><p>As mentioned earlier, the list of file patterns to ignore is
also used by <span class="command"><strong>svn add</strong></span> and <span class="command"><strong>svn
import</strong></span>. Both of these operations involve asking
Subversion to begin managing some set of files and directories.
Rather than force the user to pick and choose which files in a
tree she wishes to start versioning, Subversion uses the ignore
patterns—both the global and the per-directory
lists—to determine which files should not be swept into
the version control system as part of a larger recursive
addition or import operation. And here again, you can use the
<code class="option">--no-ignore</code> option to tell Subversion to disregard
its ignores list and operate on all the files and directories
present.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>Even if <code class="literal">svn:ignore</code> is set, you may run
into problems if you use shell wildcards in a command. Shell
wildcards are expanded into an explicit list of targets before
Subversion operates on them, so running <strong class="userinput"><code>svn
<em class="replaceable"><code>SUBCOMMAND</code></em> *</code></strong> is just like
running <strong class="userinput"><code>svn <em class="replaceable"><code>SUBCOMMAND</code></em>
file1 file2 file3 …</code></strong>. In the case of the
<span class="command"><strong>svn add</strong></span> command, this has an effect similar
to passing the <code class="option">--no-ignore</code> option. So
instead of using a wildcard, use <strong class="userinput"><code>svn add --force
.</code></strong> to do a bulk scheduling of unversioned things for
addition. The explicit target will ensure that the current
directory isn't overlooked because of being already under
version control, and the <code class="option">--force</code> option will
cause Subversion to crawl through that directory, adding
unversioned files while still honoring the
<code class="literal">svn:ignore</code> property and
<code class="literal">global-ignores</code> runtime configuration
variable. Be sure to also provide the <code class="option">--depth
files</code> option to the <span class="command"><strong>svn add</strong></span>
command if you don't want a fully recursive crawl for things
to add.</p></td></tr></table></div></div><div class="sect1" title="Keyword Substitution"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.advanced.props.special.keywords"></a>Keyword Substitution</h2></div></div></div><p>Subversion has the ability to substitute
<em class="firstterm">keywords</em>—pieces of useful,
dynamic information about a versioned file—into the
contents of the file itself. Keywords generally provide
information about the last modification made to the file.
Because this information changes each time the
file changes, and more importantly, just
<span class="emphasis"><em>after</em></span> the file changes, it is a hassle
for any process except the version control system to keep
the data completely up to date. Left to human authors, the
information would inevitably grow stale.</p><p>For example, say you have a document in which you would
like to display the last date on which it was modified. You
could burden every author of that document to, just before
committing their changes, also tweak the part of the
document that describes when it was last changed. But
sooner or later, someone would forget to do that. Instead,
simply ask Subversion to perform keyword substitution on the
<code class="literal">LastChangedDate</code> keyword. You control
where the keyword is inserted into your document by placing
a <em class="firstterm">keyword anchor</em> at the desired
location in the file. This anchor is just a string of text
formatted as
<code class="literal">$</code><em class="replaceable"><code>KeywordName</code></em><code class="literal">$</code>.</p><p>All keywords are case-sensitive where they appear as
anchors in files: you must use the correct capitalization
for the keyword to be expanded. You should consider the
value of the <code class="literal">svn:keywords</code> property to be
case-sensitive, too—certain keyword names will be recognized
regardless of case, but this behavior is deprecated.</p><p>Subversion defines the list of keywords available for
substitution. That list contains the following keywords,
some of which have aliases that you can also use:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">Date</code></span></dt><dd><p>This keyword describes the last time the file was
known to have been changed in the repository, and is of
the form <code class="literal">$Date: 2006-07-22 21:42:37 -0700 (Sat,
22 Jul 2006) $</code>. It may also be specified as
<code class="literal">LastChangedDate</code>. Unlike the
<code class="literal">Id</code> keyword, which uses UTC, the
<code class="literal">Date</code> keyword displays dates using the
local time zone.</p></dd><dt><span class="term"><code class="literal">Revision</code></span></dt><dd><p>This keyword describes the last known revision in
which this file changed in the repository, and looks
something like <code class="literal">$Revision: 144 $</code>.
It may also be specified as
<code class="literal">LastChangedRevision</code> or
<code class="literal">Rev</code>.</p></dd><dt><span class="term"><code class="literal">Author</code></span></dt><dd><p>This keyword describes the last known user to
change this file in the repository, and looks
something like <code class="literal">$Author: harry $</code>.
It may also be specified as
<code class="literal">LastChangedBy</code>.</p></dd><dt><span class="term"><code class="literal">HeadURL</code></span></dt><dd><p>This keyword describes the full URL to the latest
version of the file in the repository, and looks
something like <code class="literal">$HeadURL:
http://svn.example.com/repos/trunk/calc.c $</code>.
It may be abbreviated as
<code class="literal">URL</code>.</p></dd><dt><span class="term"><code class="literal">Id</code></span></dt><dd><p>This keyword is a compressed combination of the other
keywords. Its substitution looks something like
<code class="literal">$Id: calc.c 148 2006-07-28 21:30:43Z sally
$</code>, and is interpreted to mean that the file
<code class="filename">calc.c</code> was last changed in revision
148 on the evening of July 28, 2006 by the user
<code class="literal">sally</code>. The date displayed by this
keyword is in UTC, unlike that of the
<code class="literal">Date</code> keyword (which uses the local time
zone).</p></dd><dt><span class="term"><code class="literal">Header</code></span></dt><dd><p>This keyword is similar to the <code class="literal">Id</code>
keyword but contains the full URL of the latest revision
of the item, identical to <code class="literal">HeadURL</code>.
Its substitution looks something like <code class="literal">$Header:
http://svn.example.com/repos/trunk/calc.c
148 2006-07-28 21:30:43Z sally $</code>.</p></dd></dl></div><p>Several of the preceding descriptions use the phrase
<span class="quote">“<span class="quote">last known</span>”</span> or similar wording. Keep in mind that
keyword expansion is a client-side operation, and your client
<span class="quote">“<span class="quote">knows</span>”</span> only about changes that have occurred in
the repository when you update your working copy to include
those changes. If you never update your working copy, your
keywords will never expand to different values even if those
versioned files are being changed regularly in the
repository.</p><p>Simply adding keyword anchor text to your file does nothing
special. Subversion will never attempt to perform textual
substitutions on your file contents unless explicitly asked to
do so. After all, you might be writing a
document<sup>[<a id="idp8226528" href="#ftn.idp8226528" class="footnote">19</a>]</sup> about how to use keywords, and
you don't want Subversion to substitute your beautiful examples
of unsubstituted keyword anchors!</p><p>To tell Subversion whether to substitute keywords
on a particular file, we again turn to the property-related
subcommands. The <code class="literal">svn:keywords</code> property,
when set on a versioned file, controls which keywords will
be substituted on that file. The value is a space-delimited
list of keyword names or aliases.</p><p>For example, say you have a versioned file named
<code class="filename">weather.txt</code> that looks like
this:</p><div class="informalexample"><pre class="programlisting">
Here is the latest report from the front lines.
$LastChangedDate$
$Rev$
Cumulus clouds are appearing more frequently as summer approaches.
</pre></div><p>With no <code class="literal">svn:keywords</code> property set on
that file, Subversion will do nothing special. Now, let's
enable substitution of the
<code class="literal">LastChangedDate</code> keyword.</p><div class="informalexample"><pre class="screen">
$ svn propset svn:keywords "Date Author" weather.txt
property 'svn:keywords' set on 'weather.txt'
$
</pre></div><p>Now you have made a local property modification on the
<code class="filename">weather.txt</code> file. You will see no
changes to the file's contents (unless you made some of your
own prior to setting the property). Notice that the file
contained a keyword anchor for the <code class="literal">Rev</code>
keyword, yet we did not include that keyword in the property
value we set. Subversion will happily ignore requests to
substitute keywords that are not present in the file and
will not substitute keywords that are not present in the
<code class="literal">svn:keywords</code> property value.</p><p>Immediately after you commit this property change,
Subversion will update your working file with the new
substitute text. Instead of seeing your keyword anchor
<code class="literal">$LastChangedDate$</code>, you'll see its
substituted result. That result also contains the name of
the keyword and continues to be delimited by the dollar sign
(<code class="literal">$</code>) characters. And as we predicted, the
<code class="literal">Rev</code> keyword was not substituted because
we didn't ask for it to be.</p><p>Note also that we set the <code class="literal">svn:keywords</code>
property to <code class="literal">Date Author</code>, yet the keyword
anchor used the alias <code class="literal">$LastChangedDate$</code>
and still expanded correctly:</p><div class="informalexample"><pre class="programlisting">
Here is the latest report from the front lines.
$LastChangedDate: 2006-07-22 21:42:37 -0700 (Sat, 22 Jul 2006) $
$Rev$
Cumulus clouds are appearing more frequently as summer approaches.
</pre></div><p>If someone else now commits a change to
<code class="filename">weather.txt</code>, your copy of that file
will continue to display the same substituted keyword value
as before—until you update your working copy. At that
time, the keywords in your <code class="filename">weather.txt</code>
file will be resubstituted with information that
reflects the most recent known commit to that file.</p><div class="sidebar" title="Where's $GlobalRev$?"><div class="titlepage"><div><div><p class="title"><strong>Where's $GlobalRev$?</strong></p></div></div></div><p>New users are often confused by how the
<code class="literal">$Rev$</code> keyword works. Since the repository
has a single, globally increasing revision number, many people
assume that it is this number that is reflected by the
<code class="literal">$Rev$</code> keyword's value. But
<code class="literal">$Rev$</code> expands to show the last revision in
which the file <span class="emphasis"><em>changed</em></span>, not the last
revision to which it was updated. Understanding this clears
the confusion, but frustration often remains—without the
support of a Subversion keyword to do so, how can you
automatically get the global revision number into your
files?</p><p>To do this, you need external processing. Subversion
ships with a tool called <span class="command"><strong>svnversion</strong></span>, which
was designed for just this purpose. It crawls your working
copy and generates as output the revision(s) it finds. You
can use this program, plus some additional tooling, to embed
that revision information into your files. For more
information on <span class="command"><strong>svnversion</strong></span>, see <a class="xref" href="#svn.ref.svnversion" title="svnversion—Subversion Working Copy Version Info">the section called “svnversion—Subversion Working Copy Version Info”</a> in
<a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a>.</p></div><p>You can also instruct Subversion to maintain a fixed length
(in terms of the number of bytes consumed) for the substituted
keyword. By using a double colon (<code class="literal">::</code>) after
the keyword name, followed by a number of space characters, you
define that fixed width. When Subversion goes to substitute
your keyword for the keyword and its value, it will essentially
replace only those space characters, leaving the overall width
of the keyword field unchanged. If the substituted value is
shorter than the defined field width, there will be extra
padding characters (spaces) at the end of the substituted field;
if it is too long, it is truncated with a special hash
(<code class="literal">#</code>) character just before the final dollar
sign terminator.</p><p>For example, say you have a document in which you have
some section of tabular data reflecting the document's
Subversion keywords. Using the original Subversion keyword
substitution syntax, your file might look something
like:</p><div class="informalexample"><pre class="programlisting">
$Rev$: Revision of last commit
$Author$: Author of last commit
$Date$: Date of last commit
</pre></div><p>Now, that looks nice and tabular at the start of things.
But when you then commit that file (with keyword substitution
enabled, of course), you see:</p><div class="informalexample"><pre class="programlisting">
$Rev: 12 $: Revision of last commit
$Author: harry $: Author of last commit
$Date: 2006-03-15 02:33:03 -0500 (Wed, 15 Mar 2006) $: Date of last commit
</pre></div><p>The result is not so beautiful. And you might be
tempted to then adjust the file after the substitution so
that it again looks tabular. But that holds only as long as
the keyword values are the same width. If the last
committed revision rolls into a new place value (say, from
99 to 100), or if another person with a longer username
commits the file, stuff gets all crooked again. However, if
you are using Subversion 1.2 or later, you can use the new
fixed-length keyword syntax and define some field widths that
seem sane, so your file might look like this:</p><div class="informalexample"><pre class="programlisting">
$Rev:: $: Revision of last commit
$Author:: $: Author of last commit
$Date:: $: Date of last commit
</pre></div><p>You commit this change to your file. This time,
Subversion notices the new fixed-length keyword syntax and
maintains the width of the fields as defined by the padding
you placed between the double colon and the trailing dollar
sign. After substitution, the width of the fields is
completely unchanged—the short values for
<code class="literal">Rev</code> and <code class="literal">Author</code> are
padded with spaces, and the long <code class="literal">Date</code>
field is truncated by a hash character:</p><div class="informalexample"><pre class="programlisting">
$Rev:: 13 $: Revision of last commit
$Author:: harry $: Author of last commit
$Date:: 2006-03-15 0#$: Date of last commit
</pre></div><p>The use of fixed-length keywords is especially handy when
performing substitutions into complex file formats that
themselves use fixed-length fields for data, or for which the
stored size of a given data field is overbearingly difficult to
modify from outside the format's native application. Of course,
where binary file formats are concerned, you must always take great
care that any keyword substitution you introduce—fixed-length
or otherwise—does not violate the integrity of that format.
While it might sound easy enough, this can be an astonishingly
difficult task for most of the popular binary file formats in use
today, and <span class="emphasis"><em>not</em></span> something to be undertaken by
the faint of heart!</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>Be aware that because the width of a keyword field is
measured in bytes, the potential for corruption of
multibyte values exists. For example, a username that
contains some multibyte UTF-8 characters might suffer
truncation in the middle of the string of bytes that make
up one of those characters. The result will be a mere
truncation when viewed at the byte level, but will likely
appear as a string with an incorrect or garbled final
character when viewed as UTF-8 text. It is conceivable
that certain applications, when asked to load the file,
would notice the broken UTF-8 text and deem the entire
file corrupt, refusing to operate on the file
altogether. So, when limiting keywords to a fixed size,
choose a size that allows for this type of byte-wise
expansion.</p></td></tr></table></div></div><div class="sect1" title="Sparse Directories"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.advanced.sparsedirs"></a>Sparse Directories</h2></div></div></div><p>By default, most Subversion operations on directories act in
a recursive manner. For example, <span class="command"><strong>svn
checkout</strong></span> creates a working copy with every file and
directory in the specified area of the repository, descending
recursively through the repository tree until the entire
structure is copied to your local disk. Subversion 1.5
introduces a feature called <em class="firstterm">sparse
directories</em> (or <em class="firstterm">shallow
checkouts</em>) that allows you to easily check out a
working copy—or a portion of a working copy—more
shallowly than full recursion, with the freedom to bring in
previously ignored files and subdirectories at a later
time.</p><p>For example, say we have a repository with a tree of files
and directories with names of the members of a human family with
pets. (It's an odd example, to be sure, but bear with us.) A
regular <span class="command"><strong>svn checkout</strong></span> operation will give us a
working copy of the whole tree:</p><div class="informalexample"><pre class="screen">
$ svn checkout file:///var/svn/repos mom
A mom/son
A mom/son/grandson
A mom/daughter
A mom/daughter/granddaughter1
A mom/daughter/granddaughter1/bunny1.txt
A mom/daughter/granddaughter1/bunny2.txt
A mom/daughter/granddaughter2
A mom/daughter/fishie.txt
A mom/kitty1.txt
A mom/doggie1.txt
Checked out revision 1.
$
</pre></div><p>Now, let's check out the same tree again, but this time
we'll ask Subversion to give us only the topmost directory
with none of its children at all:</p><div class="informalexample"><pre class="screen">
$ svn checkout file:///var/svn/repos mom-empty --depth empty
Checked out revision 1
$
</pre></div><p>Notice that we added to our original <span class="command"><strong>svn
checkout</strong></span> command line a new <code class="option">--depth</code>
option. This option is present on many of Subversion's
subcommands and is similar to the
<code class="option">--non-recursive</code> (<code class="option">-N</code>) and
<code class="option">--recursive</code> (<code class="option">-R</code>) options. In
fact, it combines, improves upon, supercedes, and ultimately
obsoletes these two older options. For starters, it expands the
supported degrees of depth specification available to users,
adding some previously unsupported (or inconsistently supported)
depths. Here are the depth values that you can request for a
given Subversion operation:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">--depth empty</code></span></dt><dd><p>Include only the immediate target of the operation,
not any of its file or directory children.</p></dd><dt><span class="term"><code class="literal">--depth files</code></span></dt><dd><p>Include the immediate target of the operation and any
of its immediate file children.</p></dd><dt><span class="term"><code class="literal">--depth immediates</code></span></dt><dd><p>Include the immediate target of the operation and any
of its immediate file or directory children. The directory
children will themselves be empty.</p></dd><dt><span class="term"><code class="literal">--depth infinity</code></span></dt><dd><p>Include the immediate target, its file and directory
children, its children's children, and so on to full
recursion.</p></dd></dl></div><p>Of course, merely combining two existing options into one
hardly constitutes a new feature worthy of a whole section in
our book. Fortunately, there is more to this story. This idea
of depth extends not just to the operations you perform with
your Subversion client, but also as a description of a working
copy citizen's <em class="firstterm">ambient depth</em>, which is
the depth persistently recorded by the working copy for that
item. Its key strength is this very persistence—the fact
that it is <em class="firstterm">sticky</em>. The working copy
remembers the depth you've selected for each item in it until
you later change that depth selection; by default, Subversion
commands operate on the working copy citizens present,
regardless of their selected depth settings.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>You can check the recorded ambient depth of a working copy
using the <span class="command"><strong>svn info</strong></span> command. If the ambient
depth is anything other than infinite recursion, <span class="command"><strong>svn
info</strong></span> will display a line describing that depth
value:</p><div class="informalexample"><pre class="screen">
$ svn info mom-immediates | grep "^Depth:"
Depth: immediates
$
</pre></div></td></tr></table></div><p>Our previous examples demonstrated checkouts of infinite
depth (the default for <span class="command"><strong>svn checkout</strong></span>) and
empty depth. Let's look now at examples of the other depth
values:</p><div class="informalexample"><pre class="screen">
$ svn checkout file:///var/svn/repos mom-files --depth files
A mom-files/kitty1.txt
A mom-files/doggie1.txt
Checked out revision 1.
$ svn checkout file:///var/svn/repos mom-immediates --depth immediates
A mom-immediates/son
A mom-immediates/daughter
A mom-immediates/kitty1.txt
A mom-immediates/doggie1.txt
Checked out revision 1.
$
</pre></div><p>As described, each of these depths is something more than
only the target, but something less than full recursion.</p><p>We've used <span class="command"><strong>svn checkout</strong></span> as an example
here, but you'll find the <code class="option">--depth</code> option
present on many other Subversion commands, too. In those other
commands, depth specification is a way to limit the scope of an
operation to some depth, much like the way the older
<code class="option">--non-recursive</code> (<code class="option">-N</code>) and
<code class="option">--recursive</code> (<code class="option">-R</code>) options
behave. This means that when operating on a working copy of
some depth, while requesting an operation of a shallower depth,
the operation is limited to that shallower depth. In fact, we
can make an even more general statement: given a working copy of
any arbitrary—even mixed—ambient depth, and a
Subversion command with some requested operational depth, the
command will maintain the ambient depth of the working copy
members while still limiting the scope of the operation to the
requested (or default) operational depth.</p><p>In addition to the <code class="option">--depth</code> option, the
<span class="command"><strong>svn update</strong></span> and <span class="command"><strong>svn switch</strong></span>
subcommands also accept a second depth-related option:
<code class="option">--set-depth</code>. It is with this option that you
can change the sticky depth of a working copy item. Watch what
happens as we take our empty-depth checkout and gradually
telescope it deeper using <strong class="userinput"><code>svn update
--set-depth <em class="replaceable"><code>NEW-DEPTH</code></em> <em class="replaceable"><code>TARGET</code></em></code></strong>:</p><div class="informalexample"><pre class="screen">
$ svn update --set-depth files mom-empty
Updating 'mom-empty':
A mom-empty/kittie1.txt
A mom-empty/doggie1.txt
Updated to revision 1.
$ svn update --set-depth immediates mom-empty
Updating 'mom-empty':
A mom-empty/son
A mom-empty/daughter
Updated to revision 1.
$ svn update --set-depth infinity mom-empty
Updating 'mom-empty':
A mom-empty/son/grandson
A mom-empty/daughter/granddaughter1
A mom-empty/daughter/granddaughter1/bunny1.txt
A mom-empty/daughter/granddaughter1/bunny2.txt
A mom-empty/daughter/granddaughter2
A mom-empty/daughter/fishie1.txt
Updated to revision 1.
$
</pre></div><p>As we gradually increased our depth selection, the
repository gave us more pieces of our tree.</p><p>In our example, we operated only on the root of our working
copy, changing its ambient depth value. But we can
independently change the ambient depth value of
<span class="emphasis"><em>any</em></span> subdirectory inside the working copy,
too. Careful use of this ability allows us to flesh out only
certain portions of the working copy tree, leaving other
portions absent altogether (hence the <span class="quote">“<span class="quote">sparse</span>”</span> bit
of the feature's name). Here's an example of how we might build
out a portion of one branch of our family's tree, enable full
recursion on another branch, and keep still other pieces pruned
(absent from disk).</p><div class="informalexample"><pre class="screen">
$ rm -rf mom-empty
$ svn checkout file:///var/svn/repos mom-empty --depth empty
Checked out revision 1.
$ svn update --set-depth empty mom-empty/son
Updating 'mom-empty/son':
A mom-empty/son
Updated to revision 1.
$ svn update --set-depth empty mom-empty/daughter
Updating 'mom-empty/daughter':
A mom-empty/daughter
Updated to revision 1.
$ svn update --set-depth infinity mom-empty/daughter/granddaughter1
Updating 'mom-empty/daughter/granddaughter1':
A mom-empty/daughter/granddaughter1
A mom-empty/daughter/granddaughter1/bunny1.txt
A mom-empty/daughter/granddaughter1/bunny2.txt
Updated to revision 1.
$
</pre></div><p>Fortunately, having a complex collection of ambient depths
in a single working copy doesn't complicate the way you interact
with that working copy. You can still make, revert, display,
and commit local modifications in your working copy without
providing any new options (including <code class="option">--depth</code> and
<code class="option">--set-depth</code>) to the relevant subcommands. Even
<span class="command"><strong>svn update</strong></span> works as it does elsewhere when no
specific depth is provided—it updates the working copy
targets that are present while honoring their sticky
depths.</p><p>You might at this point be wondering, <span class="quote">“<span class="quote">So what? When
would I use this?</span>”</span> One scenario where this feature
finds utility is tied to a particular repository layout,
specifically where you have many related or codependent
projects or software modules living as siblings in a single
repository location (<code class="filename">trunk/project1</code>,
<code class="filename">trunk/project2</code>,
<code class="filename">trunk/project3</code>, etc.). In such
scenarios, it might be the case that you personally care
about only a handful of those projects—maybe some primary
project and a few other modules on which it depends. You can
check out individual working copies of all of these things, but
those working copies are disjoint and, as a result, it can be
cumbersome to perform operations across several or all of them
at the same time. The alternative is to use the sparse
directories feature, building out a single working copy that
contains only the modules you care about. You'd start with an
empty-depth checkout of the common parent directory of the
projects, and then update with infinite depth only the items you
wish to have, like we demonstrated in the previous example.
Think of it like an opt-in system for working copy
citizens.</p><p>The original (Subversion 1.5) implementation of shallow
checkouts was good, but didn't support de-telescoping of working
copy items. Subversion 1.6 remedied this problem. For example,
running <strong class="userinput"><code>svn update --set-depth empty</code></strong> in
an infinite-depth working copy will discard everything but the
topmost directory.<sup>[<a id="idp8323744" href="#ftn.idp8323744" class="footnote">20</a>]</sup> Subversion
1.6 also introduced another supported value for
the <code class="option">--set-depth</code>
option: <code class="literal">exclude</code>. Using <code class="option">--set-depth
exclude</code> with <span class="command"><strong>svn update</strong></span> will cause
the update target to be removed from the working copy
entirely—a directory target won't even be left
present-but-empty. This is especially handy when there are more
things that you'd like to keep in a working copy than things
you'd like to <span class="emphasis"><em>not</em></span> keep.</p><p>Consider a directory with hundreds of subdirectories, one of
which you would like to omit from your working copy. Using
an <span class="quote">“<span class="quote">additive</span>”</span> approach to sparse directories, you
might check out the directory with an empty depth, then
explicitly telescope (using <strong class="userinput"><code>svn update --set-depth
infinity</code></strong>) each and every subdirectory of the
directory except the one you don't care about.</p><div class="informalexample"><pre class="screen">
$ svn checkout http://svn.example.com/repos/many-dirs --depth empty
…
$ svn update --set-depth infinity many-dirs/wanted-dir-1
…
$ svn update --set-depth infinity many-dirs/wanted-dir-2
…
$ svn update --set-depth infinity many-dirs/wanted-dir-3
…
### and so on, and so on, ...
</pre></div><p>This could be quite tedious, especially since you don't even
have stubs of these directories in your working copy to deal
with. Such a working copy would also have another
characteristic that you might not expect or desire: if someone
else creates any new subdirectories in this top-level directory,
you won't receive those when you update your working
copy.</p><p>Beginning with Subversion 1.6, you can take a different
approach. First, check out the directory in full. Then
run <strong class="userinput"><code>svn update --set-depth exclude</code></strong> on the
one subdirectory you don't care about.</p><div class="informalexample"><pre class="screen">
$ svn checkout http://svn.example.com/repos/many-dirs
…
$ svn update --set-depth exclude many-dirs/unwanted-dir
D many-dirs/unwanted-dir
$
</pre></div><p>This approach leaves your working copy with the same stuff
as in the first approach, but any new subdirectories which
appear in the top-level directory would also show up when you
update your working copy. The downside of this approach is that
you have to actually check out that whole subdirectory that you
don't even want just so you can tell Subversion that you don't
want it. This might not even be possible if that subdirectory
is too large to fit on your disk (which might, after all, be the
very reason you don't want it in your working copy).</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>While the functionality for excluding an existing item
from a working copy was hung off of the <span class="command"><strong>svn
update</strong></span> command, you might have noticed that the
output from <strong class="userinput"><code>svn update --set-depth
exclude</code></strong> differs from that of a normal update
operation. This output betrays the fact that, under the hood,
exclusion is a completely client-side operation, very much
unlike a typical update.</p></td></tr></table></div><p>In such a situation, you might consider a compromise
approach. First, check out the top-level directory
with <code class="option">--depth immediates</code>. Then, exclude the
directory you don't want using <strong class="userinput"><code>svn update --set-depth
exclude</code></strong>. Finally, telescope all the items that
remain to infinite depth, which should be fairly easy to do
because they are all addressable in your shell.</p><div class="informalexample"><pre class="screen">
$ svn checkout http://svn.example.com/repos/many-dirs --depth immediates
…
$ svn update --set-depth exclude many-dirs/unwanted-dir
D many-dirs/unwanted-dir
$ svn update --set-depth infinity many-dirs/*
…
$
</pre></div><p>Once again, your working copy will have the same stuff as in
the previous two scenarios. But now, any time a new file or
subdirectory is committed to the top-level directory, you'll
receive it—at an empty depth—when you update your
working copy. You can now decide what to do with such newly
appearing working copy items: expand them into infinite depth,
or exclude them altogether.</p></div><div class="sect1" title="Locking"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.advanced.locking"></a>Locking</h2></div></div></div><p>Subversion's copy-modify-merge version control model lives
and dies on its data merging algorithms—specifically on
how well those algorithms perform when trying to resolve
conflicts caused by multiple users modifying the same file
concurrently. Subversion itself provides only one such
algorithm: a three-way differencing algorithm that is smart
enough to handle data at a granularity of a single line of text.
Subversion also allows you to supplement its content merge
processing with external differencing utilities (as described in
<a class="xref" href="#svn.advanced.externaldifftools.diff3" title="External diff3">the section called “External diff3”</a> and
<a class="xref" href="#svn.advanced.externaldifftools.merge" title="External merge">the section called “External merge”</a>), some
of which may do an even better job, perhaps providing
granularity of a word or a single character of text. But common
among those algorithms is that they generally work only on text
files. The landscape starts to look pretty grim when you start
talking about content merges of nontextual file formats. And
when you can't find a tool that can handle that type of merging,
you begin to run into problems with the copy-modify-merge
model.</p><p>Let's look at a real-life example of where this model runs
aground. Harry and Sally are both graphic designers working on
the same project, a bit of marketing collateral for an
automobile mechanic. Central to the design of a particular
poster is an image of a car in need of some bodywork, stored in
a file using the PNG image format. The poster's layout is
almost finished, and both Harry and Sally are pleased with the
particular photo they chose for their damaged car—a baby
blue 1967 Ford Mustang with an unfortunate bit of crumpling on
the left front fender.</p><p>Now, as is common in graphic design work, there's a change
in plans, which causes the car's color to be a concern. So Sally
updates her working copy to <code class="literal">HEAD</code>, fires up
her photo-editing software, and sets about tweaking the image so
that the car is now cherry red. Meanwhile, Harry, feeling
particularly inspired that day, decides that the image would
have greater impact if the car also appears to have suffered
greater impact. He, too, updates to <code class="literal">HEAD</code>,
and then draws some cracks on the vehicle's windshield. He
manages to finish his work before Sally finishes hers, and after
admiring the fruits of his undeniable talent, he commits the
modified image. Shortly thereafter, Sally is finished with the
car's new finish and tries to commit her changes. But, as
expected, Subversion fails the commit, informing Sally that
her version of the image is now out of date.</p><p>Here's where the difficulty sets in. If Harry and Sally
were making changes to a text file, Sally would simply update
her working copy, receiving Harry's changes in the process. In
the worst possible case, they would have modified the same
region of the file, and Sally would have to work out by hand the
proper resolution to the conflict. But these aren't text
files—they are binary images. And while it's a simple
matter to describe what one would expect the results of this
content merge to be, there is precious little chance that any
software exists that is smart enough to examine the common
baseline image that each of these graphic artists worked
against, the changes that Harry made, and the changes that Sally
made, and then spit out an image of a busted-up red Mustang with
a cracked windshield!</p><p>Of course, things would have gone more smoothly if Harry and
Sally had serialized their modifications to the image—if,
say, Harry had waited to draw his windshield cracks on Sally's
now-red car, or if Sally had tweaked the color of a car whose
windshield was already cracked. As is discussed in
<a class="xref" href="#svn.basic.vsn-models.copy-merge" title="The copy-modify-merge solution">the section called “The copy-modify-merge solution”</a>, most of
these types of problems go away entirely where perfect
communication between Harry and Sally
exists.<sup>[<a id="idp8354048" href="#ftn.idp8354048" class="footnote">21</a>]</sup> But as one's version control
system is, in fact, one form of communication, it follows that
having that software facilitate the serialization of
nonparallelizable editing efforts is no bad thing. This is
where Subversion's implementation of the lock-modify-unlock
model steps into the spotlight. This is where we talk about
Subversion's <em class="firstterm">locking</em> feature, which is
similar to the <span class="quote">“<span class="quote">reserved checkouts</span>”</span> mechanisms of
other version control systems.</p><p>Subversion's locking feature exists ultimately to minimize
wasted time and effort. By allowing a user to programmatically
claim the exclusive right to change a file in the repository,
that user can be reasonably confident that any energy he invests
on unmergeable changes won't be wasted—his commit of those
changes will succeed. Also, because Subversion communicates to
other users that serialization is in effect for a particular
versioned object, those users can reasonably expect that the
object is about to be changed by someone else. They, too, can
then avoid wasting their time and energy on unmergeable changes
that won't be committable due to eventual
out-of-dateness.</p><p>When referring to Subversion's locking feature, one is
actually talking about a fairly diverse collection of behaviors,
which include the ability to lock a versioned
file<sup>[<a id="idp8357920" href="#ftn.idp8357920" class="footnote">22</a>]</sup> (claiming the exclusive right to
modify the file), to unlock that file (yielding that exclusive
right to modify), to see reports about which files are locked
and by whom, to annotate files for which locking before editing
is strongly advised, and so on. In this section, we'll cover
all of these facets of the larger locking feature.</p><div class="sidebar" title="The Three Meanings of “Lock”"><a id="svn.advanced.locking.meanings"></a><div class="titlepage"><div><div><p class="title"><strong>The Three Meanings of <span class="quote">“<span class="quote">Lock</span>”</span></strong></p></div></div></div><p>In this section, and almost everywhere in this book, the
words <span class="quote">“<span class="quote">lock</span>”</span> and <span class="quote">“<span class="quote">locking</span>”</span> describe
a mechanism for mutual exclusion between users to avoid
clashing commits. Unfortunately, there are two other sorts
of <span class="quote">“<span class="quote">lock</span>”</span> with which Subversion, and therefore
this book, sometimes needs to be concerned.</p><p>The second is <em class="firstterm">working copy locks</em>,
used internally by Subversion to prevent clashes between
multiple Subversion clients operating on the same working
copy. This is the sort of lock indicated by an
<code class="computeroutput">L</code> in the third column of
<span class="command"><strong>svn status</strong></span> output, and removed by the
<span class="command"><strong>svn cleanup</strong></span> command, as described in <a class="xref" href="#svn.tour.cleanup" title="Sometimes You Just Need to Clean Up">the section called “Sometimes You Just Need to Clean Up”</a>.</p><p>Third, there are <em class="firstterm">database locks</em>,
used internally by the Berkeley DB backend to prevent clashes
between multiple programs trying to access the database. This
is the sort of lock whose unwanted persistence after an error
can cause a repository to be <span class="quote">“<span class="quote">wedged,</span>”</span> as
described in <a class="xref" href="#svn.reposadmin.maint.recovery" title="Berkeley DB Recovery">the section called “Berkeley DB Recovery”</a>.</p><p>You can generally forget about these other kinds of locks
until something goes wrong that requires you to care about
them. In this book, <span class="quote">“<span class="quote">lock</span>”</span> means the first sort
unless the contrary is either clear from context or explicitly
stated.</p></div><div class="sect2" title="Creating Locks"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.locking.creation"></a>Creating Locks</h3></div></div></div><p>In the Subversion repository, a
<em class="firstterm">lock</em> is a piece of metadata that
grants exclusive access to one user to change a file. This
user is said to be the <em class="firstterm">lock owner</em>.
Each lock also has a unique identifier, typically a long
string of characters, known as the <em class="firstterm">lock
token</em>. The repository manages locks, ultimately
handling their creation, enforcement, and removal. If any
commit transaction attempts to modify or delete a locked file
(or delete one of the parent directories of the file), the
repository will demand two pieces of information—that
the client performing the commit be authenticated as the lock
owner, and that the lock token has been provided as part of
the commit process as a form of proof that the client knows which
lock it is using.</p><p>To demonstrate lock creation, let's refer back to our
example of multiple graphic designers working on the same
binary image files. Harry has decided to change a JPEG image.
To prevent other people from committing changes to the file
while he is modifying it (as well as alerting them that he is
about to change it), he locks the file in the repository using
the <span class="command"><strong>svn lock</strong></span> command.</p><div class="informalexample"><pre class="screen">
$ svn lock banana.jpg -m "Editing file for tomorrow's release."
'banana.jpg' locked by user 'harry'.
$
</pre></div><p>The preceding example demonstrates a number of new things.
First, notice that Harry passed the
<code class="option">--message</code> (<code class="option">-m</code>) option to
<span class="command"><strong>svn lock</strong></span>. Similar to <span class="command"><strong>svn
commit</strong></span>, the <span class="command"><strong>svn lock</strong></span> command can
take comments—via either <code class="option">--message</code>
(<code class="option">-m</code>) or <code class="option">--file</code>
(<code class="option">-F</code>)—to describe the reason for locking the
file. Unlike <span class="command"><strong>svn commit</strong></span>, however,
<span class="command"><strong>svn lock</strong></span> will not demand a message by
launching your preferred text editor. Lock comments are
optional, but still recommended to aid communication.</p><p>Second, the lock attempt succeeded. This means that the
file wasn't already locked, and that Harry had the latest
version of the file. If Harry's working copy of the file had
been out of date, the repository would have rejected the
request, forcing Harry to <span class="command"><strong>svn update</strong></span> and
reattempt the locking command. The locking command would also
have failed if the file had already been locked by someone
else.</p><p>As you can see, the <span class="command"><strong>svn lock</strong></span> command
prints confirmation of the successful lock. At this point,
the fact that the file is locked becomes apparent in the
output of the <span class="command"><strong>svn status</strong></span> and <span class="command"><strong>svn
info</strong></span> reporting subcommands.</p><div class="informalexample"><pre class="screen">
$ svn status
K banana.jpg
$ svn info banana.jpg
Path: banana.jpg
Name: banana.jpg
Working Copy Root Path: /home/harry/project
URL: http://svn.example.com/repos/project/banana.jpg
Repository Root: http://svn.example.com/repos/project
Repository UUID: edb2f264-5ef2-0310-a47a-87b0ce17a8ec
Revision: 2198
Node Kind: file
Schedule: normal
Last Changed Author: frank
Last Changed Rev: 1950
Last Changed Date: 2006-03-15 12:43:04 -0600 (Wed, 15 Mar 2006)
Text Last Updated: 2006-06-08 19:23:07 -0500 (Thu, 08 Jun 2006)
Properties Last Updated: 2006-06-08 19:23:07 -0500 (Thu, 08 Jun 2006)
Checksum: 3b110d3b10638f5d1f4fe0f436a5a2a5
Lock Token: opaquelocktoken:0c0f600b-88f9-0310-9e48-355b44d4a58e
Lock Owner: harry
Lock Created: 2006-06-14 17:20:31 -0500 (Wed, 14 Jun 2006)
Lock Comment (1 line):
Editing file for tomorrow's release.
$
</pre></div><p>The fact that the <span class="command"><strong>svn info</strong></span> command,
which does not contact the repository when run against working
copy paths, can display the lock token reveals an important
piece of information about those tokens: they are cached in
the working copy. The presence of the lock token is critical.
It gives the working copy authorization to make use of the
lock later on. Also, the <span class="command"><strong>svn status</strong></span>
command shows a <code class="literal">K</code> next to the file (short
for locKed), indicating that the lock token is present.</p><div class="sidebar" title="Regarding Lock Tokens"><div class="titlepage"><div><div><p class="title"><strong>Regarding Lock Tokens</strong></p></div></div></div><p>A lock token isn't an authentication token, so much as
an <span class="emphasis"><em>authorization</em></span> token. The token
isn't a protected secret. In fact, a lock's unique token is
discoverable by anyone who runs <strong class="userinput"><code>svn info
<em class="replaceable"><code>URL</code></em></code></strong>. A lock token is
special only when it lives inside a working copy. It's
proof that the lock was created in that particular working
copy, and not somewhere else by some other client. Merely
authenticating as the lock owner isn't enough to prevent
accidents.</p><p>For example, suppose you lock a file using a computer at
your office, but leave work for the day before you finish
your changes to that file. It should not be possible to
accidentally commit changes to that same file from your home
computer later that evening simply because you've
authenticated as the lock's owner. In other words, the lock
token prevents one piece of Subversion-related software from
undermining the work of another. (In our example, if you
really need to change the file from an alternative working
copy, you would need to <em class="firstterm">break</em> the
lock and relock the file.)</p></div><p>Now that Harry has locked <code class="filename">banana.jpg</code>,
Sally is unable to change or delete that file:</p><div class="informalexample"><pre class="screen">
$ svn delete banana.jpg
D banana.jpg
$ svn commit -m "Delete useless file."
Deleting banana.jpg
svn: E175002: Commit failed (details follow):
svn: E175002: Server sent unexpected return value (423 Locked) in response to
DELETE request for '/repos/project/!svn/wrk/64bad3a9-96f9-0310-818a-df4224ddc
35d/banana.jpg'
$
</pre></div><p>But Harry, after touching up the banana's shade of yellow,
is able to commit his changes to the file. That's because he
authenticates as the lock owner and also because his working
copy holds the correct lock token:</p><div class="informalexample"><pre class="screen">
$ svn status
M K banana.jpg
$ svn commit -m "Make banana more yellow"
Sending banana.jpg
Transmitting file data .
Committed revision 2201.
$ svn status
$
</pre></div><p>Notice that after the commit is finished, <span class="command"><strong>svn
status</strong></span> shows that the lock token is no longer
present in the working copy. This is the standard behavior of
<span class="command"><strong>svn commit</strong></span>—it searches the working
copy (or list of targets, if you provide such a list) for
local modifications and sends all the lock tokens it
encounters during this walk to the server as part of the
commit transaction. After the commit completes successfully,
all of the repository locks that were mentioned are
released—<span class="emphasis"><em>even on files that weren't
committed</em></span>. This is meant to discourage users from
being sloppy about locking or from holding locks for too long.
If Harry haphazardly locks 30 files in a directory named
<code class="filename">images</code> because he's unsure of which files
he needs to change, yet changes only four of those files, when he
runs <strong class="userinput"><code>svn commit images</code></strong>, the process will
still release all 30 locks.</p><p>This behavior of automatically releasing locks can be
overridden with the <code class="option">--no-unlock</code> option to
<span class="command"><strong>svn commit</strong></span>. This is best used for those
times when you want to commit changes, but still plan to make
more changes and thus need to retain existing locks. You can
also make this your default behavior by setting the
<code class="literal">no-unlock</code> runtime configuration option (see
<a class="xref" href="#svn.advanced.confarea" title="Runtime Configuration Area">the section called “Runtime Configuration Area”</a>).</p><p>Of course, locking a file doesn't oblige one to commit a
change to it. The lock can be released at any time with a
simple <span class="command"><strong>svn unlock</strong></span> command:</p><div class="informalexample"><pre class="screen">
$ svn unlock banana.c
'banana.c' unlocked.
</pre></div></div><div class="sect2" title="Discovering Locks"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.locking.discovery"></a>Discovering Locks</h3></div></div></div><p>When a commit fails due to someone else's locks, it's
fairly easy to learn about them. The easiest way is to run
<strong class="userinput"><code>svn status -u</code></strong>:</p><div class="informalexample"><pre class="screen">
$ svn status -u
M 23 bar.c
M O 32 raisin.jpg
* 72 foo.h
Status against revision: 105
$
</pre></div><p>In this example, Sally can see not only that her copy of
<code class="filename">foo.h</code> is out of date, but also that one of the
two modified files she plans to commit is locked in the
repository. The <code class="literal">O</code> symbol stands for
<span class="quote">“<span class="quote">Other,</span>”</span> meaning that a lock exists on the file
and was created by somebody else. If she were to attempt a
commit, the lock on <code class="filename">raisin.jpg</code> would
prevent it. Sally is left wondering who made the lock, when,
and why. Once again, <span class="command"><strong>svn info</strong></span> has the
answers:</p><div class="informalexample"><pre class="screen">
$ svn info ^/raisin.jpg
Path: raisin.jpg
Name: raisin.jpg
URL: http://svn.example.com/repos/project/raisin.jpg
Repository Root: http://svn.example.com/repos/project
Repository UUID: edb2f264-5ef2-0310-a47a-87b0ce17a8ec
Revision: 105
Node Kind: file
Last Changed Author: sally
Last Changed Rev: 32
Last Changed Date: 2006-01-25 12:43:04 -0600 (Sun, 25 Jan 2006)
Lock Token: opaquelocktoken:fc2b4dee-98f9-0310-abf3-653ff3226e6b
Lock Owner: harry
Lock Created: 2006-02-16 13:29:18 -0500 (Thu, 16 Feb 2006)
Lock Comment (1 line):
Need to make a quick tweak to this image.
$
</pre></div><p>Just as you can use <span class="command"><strong>svn info</strong></span> to examine
objects in the working copy, you can also use it to examine
objects in the repository. If the main argument to
<span class="command"><strong>svn info</strong></span> is a working copy path, then all
of the working copy's cached information is displayed; any
mention of a lock means that the working copy is holding a
lock token (if a file is locked by another user or in another
working copy, <span class="command"><strong>svn info</strong></span> on a working copy
path will show no lock information at all). If the main
argument to <span class="command"><strong>svn info</strong></span> is a URL, the
information reflects the latest version of an object in the
repository, and any mention of a lock describes the current
lock on the object.</p><p>So in this particular example, Sally can see that Harry
locked the file on February 16 to <span class="quote">“<span class="quote">make a quick
tweak.</span>”</span> It being June, she suspects that he probably
forgot all about the lock. She might phone Harry to complain
and ask him to release the lock. If he's unavailable, she
might try to forcibly break the lock herself or ask an
administrator to do so.</p></div><div class="sect2" title="Breaking and Stealing Locks"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.locking.break-steal"></a>Breaking and Stealing Locks</h3></div></div></div><p>A repository lock isn't sacred—in Subversion's
default configuration state, locks can be released not only by
the person who created them, but by anyone. When somebody
other than the original lock creator destroys a lock, we refer
to this as <em class="firstterm">breaking the lock</em>.</p><p>From the administrator's chair, it's simple to break
locks. The <span class="command"><strong>svnlook</strong></span>
and <span class="command"><strong>svnadmin</strong></span> programs have the ability to
display and remove locks directly from the repository. (For
more information about these tools, see
<a class="xref" href="#svn.reposadmin.maint.tk" title="An Administrator's Toolkit">the section called “An Administrator's Toolkit”</a>.)</p><div class="informalexample"><pre class="screen">
$ svnadmin lslocks /var/svn/repos
Path: /project2/images/banana.jpg
UUID Token: opaquelocktoken:c32b4d88-e8fb-2310-abb3-153ff1236923
Owner: frank
Created: 2006-06-15 13:29:18 -0500 (Thu, 15 Jun 2006)
Expires:
Comment (1 line):
Still improving the yellow color.
Path: /project/raisin.jpg
UUID Token: opaquelocktoken:fc2b4dee-98f9-0310-abf3-653ff3226e6b
Owner: harry
Created: 2006-02-16 13:29:18 -0500 (Thu, 16 Feb 2006)
Expires:
Comment (1 line):
Need to make a quick tweak to this image.
$ svnadmin rmlocks /var/svn/repos /project/raisin.jpg
Removed lock on '/project/raisin.jpg'.
$
</pre></div><p>The more interesting option is to allow users to break
each other's locks over the network. To do this, Sally simply
needs to pass the <code class="option">--force</code> to the <span class="command"><strong>svn
unlock</strong></span> command:</p><div class="informalexample"><pre class="screen">
$ svn status -u
M 23 bar.c
M O 32 raisin.jpg
* 72 foo.h
Status against revision: 105
$ svn unlock raisin.jpg
svn: E195013: 'raisin.jpg' is not locked in this working copy
$ svn info raisin.jpg | grep URL
URL: http://svn.example.com/repos/project/raisin.jpg
$ svn unlock http://svn.example.com/repos/project/raisin.jpg
svn: warning: W160039: Unlock failed on 'raisin.jpg' (403 Forbidden)
$ svn unlock --force http://svn.example.com/repos/project/raisin.jpg
'raisin.jpg' unlocked.
$
</pre></div><p>Now, Sally's initial attempt to unlock failed because she
ran <span class="command"><strong>svn unlock</strong></span> directly on her working copy
of the file, and no lock token was present. To remove the
lock directly from the repository, she needs to pass a URL
to <span class="command"><strong>svn unlock</strong></span>. Her first attempt to unlock
the URL fails, because she can't authenticate as the lock
owner (nor does she have the lock token). But when she
passes <code class="option">--force</code>, the authentication and
authorization requirements are ignored, and the remote lock is
broken.</p><p>Simply breaking a lock may not be enough. In
the running example, Sally may not only want to break Harry's
long-forgotten lock, but relock the file for her own use.
She can accomplish this by using <span class="command"><strong>svn unlock</strong></span>
with <code class="option">--force</code> and then <span class="command"><strong>svn lock</strong></span>
back-to-back, but there's a small chance that somebody else
might lock the file between the two commands. The simpler thing
to do is to <em class="firstterm">steal</em> the lock, which involves
breaking and relocking the file all in one atomic step. To
do this, Sally passes the <code class="option">--force</code> option
to <span class="command"><strong>svn lock</strong></span>:</p><div class="informalexample"><pre class="screen">
$ svn lock raisin.jpg
svn: warning: W160035: Path '/project/raisin.jpg' is already locked by user 'h
arry' in filesystem '/var/svn/repos/db'
$ svn lock --force raisin.jpg
'raisin.jpg' locked by user 'sally'.
$
</pre></div><p>In any case, whether the lock is broken or stolen, Harry
may be in for a surprise. Harry's working copy still contains
the original lock token, but that lock no longer exists. The
lock token is said to be <em class="firstterm">defunct</em>. The
lock represented by the lock token has either been broken (no
longer in the repository) or stolen (replaced with a
different lock). Either way, Harry can see this by asking
<span class="command"><strong>svn status</strong></span> to contact the
repository:</p><div class="informalexample"><pre class="screen">
$ svn status
K raisin.jpg
$ svn status -u
B 32 raisin.jpg
Status against revision: 105
$ svn update
Updating '.':
B raisin.jpg
Updated to revision 105.
$ svn status
$
</pre></div><p>If the repository lock was broken, then <strong class="userinput"><code>svn
status --show-updates</code></strong> (<code class="option">-u</code>)
displays a <code class="literal">B</code> (Broken) symbol next to the
file. If a new lock exists in place of the old one, then a
<code class="literal">T</code> (sTolen) symbol is shown. Finally,
<span class="command"><strong>svn update</strong></span> notices any defunct lock tokens
and removes them from the working copy.</p><div class="sidebar" title="Locking Policies"><div class="titlepage"><div><div><p class="title"><strong>Locking Policies</strong></p></div></div></div><p>Different systems have different notions of how strict a
lock should be. Some folks argue that locks must be
strictly enforced at all costs, releasable only by the
original creator or administrator. They argue that if
anyone can break a lock, chaos runs rampant and the
whole point of locking is defeated. The other side argues
that locks are first and foremost a communication tool. If
users are constantly breaking each other's locks, it
represents a cultural failure within the team and the
problem falls outside the scope of software enforcement.</p><p>Subversion defaults to the <span class="quote">“<span class="quote">softer</span>”</span>
approach, but still allows administrators to create stricter
enforcement policies through the use of hook scripts. In
particular, the <code class="filename">pre-lock</code> and
<code class="filename">pre-unlock</code> hooks allow administrators
to decide when lock creation and lock releases are allowed
to happen. Depending on whether a lock already exists,
these two hooks can decide whether to allow a certain user
to break or steal a lock. The
<code class="filename">post-lock</code> and
<code class="filename">post-unlock</code> hooks are also available,
and can be used to send email after locking actions. To
learn more about repository hooks, see <a class="xref" href="#svn.reposadmin.create.hooks" title="Implementing Repository Hooks">the section called “Implementing Repository Hooks”</a>.</p></div></div><div class="sect2" title="Lock Communication"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.locking.lock-communication"></a>Lock Communication</h3></div></div></div><p>We've seen how <span class="command"><strong>svn lock</strong></span>
and <span class="command"><strong>svn unlock</strong></span> can be used to create,
release, break, and steal locks. This satisfies the goal of
serializing commit access to a file. But what about the
larger problem of preventing wasted time?</p><p>For example, suppose Harry locks an image file and then
begins editing it. Meanwhile, miles away, Sally wants to do
the same thing. She doesn't think to run <strong class="userinput"><code>svn status
-u</code></strong>, so she has no idea that Harry has
already locked the file. She spends hours editing the file,
and when she tries to commit her change, she discovers that
either the file is locked or that she's out of date.
Regardless, her changes aren't mergeable with Harry's. One of
these two people has to throw away his or her work, and a lot of
time has been wasted.</p><p>Subversion's solution to this problem is to provide a
mechanism to remind users that a file ought to be locked
<span class="emphasis"><em>before</em></span> the editing begins. The mechanism
is a special property: <code class="literal">svn:needs-lock</code>. If
that property is attached to a file (regardless of its value,
which is irrelevant), Subversion will try to use
filesystem-level permissions to make the file read-only—unless,
of course, the user has explicitly locked the file.
When a lock token is present (as a result of using
<span class="command"><strong>svn lock</strong></span>), the file becomes read/write.
When the lock is released, the file becomes read-only
again.</p><p>The theory, then, is that if the image file has this
property attached, Sally would immediately notice
something is strange when she opens the file for editing:
many applications alert users immediately when a read-only
file is opened for editing, and nearly all would
prevent her from saving changes to the file. This
reminds her to lock the file before editing, whereby she
discovers the preexisting lock:</p><div class="informalexample"><pre class="screen">
$ /usr/local/bin/gimp raisin.jpg
gimp: error: file is read-only!
$ ls -l raisin.jpg
-r--r--r-- 1 sally sally 215589 Jun 8 19:23 raisin.jpg
$ svn lock raisin.jpg
svn: warning: W160035: Path '/project/raisin.jpg' is already locked by user 'h
arry' in filesystem '/var/svn/repos/db'
$ svn info http://svn.example.com/repos/project/raisin.jpg | grep Lock
Lock Token: opaquelocktoken:fc2b4dee-98f9-0310-abf3-653ff3226e6b
Lock Owner: harry
Lock Created: 2006-06-08 07:29:18 -0500 (Thu, 08 June 2006)
Lock Comment (1 line):
Making some tweaks. Locking for the next two hours.
$
</pre></div><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>Users and administrators alike are encouraged to attach
the <code class="literal">svn:needs-lock</code> property to any file
that cannot be contextually merged. This is the primary
technique for encouraging good locking habits and preventing
wasted effort.</p></td></tr></table></div><p>Note that this property is a communication tool that
works independently from the locking system. In other words,
any file can be locked, whether or not this property is
present. And conversely, the presence of this property
doesn't make the repository require a lock when
committing.</p><p>Unfortunately, the system isn't flawless. It's possible
that even when a file has the property, the read-only reminder
won't always work. Sometimes applications misbehave and
<span class="quote">“<span class="quote">hijack</span>”</span> the read-only file, silently allowing
users to edit and save the file anyway. There's not much that
Subversion can do in this situation—at the end of the
day, there's simply no substitution for good interpersonal
communication.<sup>[<a id="idp8467072" href="#ftn.idp8467072" class="footnote">23</a>]</sup></p></div></div><div class="sect1" title="Externals Definitions"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.advanced.externals"></a>Externals Definitions</h2></div></div></div><p>Sometimes it is useful to construct a working copy that is
made out of a number of different checkouts. For example, you
may want different subdirectories to come from different
locations in a repository or perhaps from different
repositories altogether. You could certainly set up such a
scenario by hand—using <span class="command"><strong>svn checkout</strong></span> to
create the sort of nested working copy structure you are trying
to achieve. But if this layout is important for everyone who
uses your repository, every other user will need to perform the
same checkout operations that you did.</p><p>Fortunately, Subversion provides support for
<em class="firstterm">externals definitions</em>. An externals
definition is a mapping of a local directory to the
URL—and ideally a particular revision—of a versioned
directory. In Subversion, you declare externals definitions in
groups using the <code class="literal">svn:externals</code> property. You
can create or modify this property using <span class="command"><strong>svn
propset</strong></span> or <span class="command"><strong>svn propedit</strong></span> (see <a class="xref" href="#svn.advanced.props.manip" title="Manipulating Properties">the section called “Manipulating Properties”</a>). It can be set on any
versioned directory, and its value describes both the external
repository location and the client-side directory to which that
location should be checked out.</p><p>The convenience of the <code class="literal">svn:externals</code>
property is that once it is set on a versioned directory,
everyone who checks out a working copy with that directory also
gets the benefit of the externals definition. In other words,
once one person has made the effort to define the nested working
copy structure, no one else has to bother—Subversion will,
after checking out the original working copy, automatically also
check out the external working copies.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>The relative target subdirectories of externals
definitions <span class="emphasis"><em>must not</em></span> already exist on
your or other users' systems—Subversion will create them
when it checks out the external working copy.</p></td></tr></table></div><p>You also get in the externals definition design all the
regular benefits of Subversion properties. The definitions are
versioned. If you need to change an externals definition, you
can do so using the regular property modification subcommands.
When you commit a change to the <code class="literal">svn:externals</code>
property, Subversion will synchronize the checked-out items
against the changed externals definition when you next run
<strong class="userinput"><code>svn update</code></strong>. The same thing will happen when
others update their working copies and receive your changes to
the externals definition.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>Because the <code class="literal">svn:externals</code> property has
a multiline value, we strongly recommend that you use
<span class="command"><strong>svn propedit</strong></span> instead of <span class="command"><strong>svn
propset</strong></span>.</p></td></tr></table></div><p>Subversion releases prior to 1.5 honor an externals
definition format that is a multiline table of subdirectories
(relative to the versioned directory on which the property is
set), optional revision flags, and fully qualified, absolute
Subversion repository URLs. An example of this might look as
follows:</p><div class="informalexample"><pre class="screen">
$ svn propget svn:externals calc
third-party/sounds http://svn.example.com/repos/sounds
third-party/skins -r148 http://svn.example.com/skinproj
third-party/skins/toolkit -r21 http://svn.example.com/skin-maker
</pre></div><p>When someone checks out a working copy of the
<code class="filename">calc</code> directory referred to in the previous
example, Subversion also continues to check out the items found
in its externals definition.</p><div class="informalexample"><pre class="screen">
$ svn checkout http://svn.example.com/repos/calc
A calc
A calc/Makefile
A calc/integer.c
A calc/button.c
Checked out revision 148.
Fetching external item into calc/third-party/sounds
A calc/third-party/sounds/ding.ogg
A calc/third-party/sounds/dong.ogg
A calc/third-party/sounds/clang.ogg
…
A calc/third-party/sounds/bang.ogg
A calc/third-party/sounds/twang.ogg
Checked out revision 14.
Fetching external item into calc/third-party/skins
…
</pre></div><p>As of Subversion 1.5, though, a new format of the
<code class="literal">svn:externals</code> property is supported.
Externals definitions are still multiline, but the order and
format of the various pieces of information have changed. The
new syntax more closely mimics the order of arguments you might
pass to <span class="command"><strong>svn checkout</strong></span>: the optional revision
flags come first, then the external Subversion repository URL,
and finally the relative local subdirectory. Notice, though,
that this time we didn't say <span class="quote">“<span class="quote">fully qualified, absolute
Subversion repository URLs.</span>”</span> That's because the new
format supports relative URLs and URLs that carry peg revisions.
The previous example of an externals definition might, in
Subversion 1.5, look like the following:</p><div class="informalexample"><pre class="screen">
$ svn propget svn:externals calc
http://svn.example.com/repos/sounds third-party/sounds
-r148 http://svn.example.com/skinproj third-party/skins
-r21 http://svn.example.com/skin-maker third-party/skins/toolkit
</pre></div><p>Or, making use of the peg revision syntax (which we describe
in detail in <a class="xref" href="#svn.advanced.pegrevs" title="Peg and Operative Revisions">the section called “Peg and Operative Revisions”</a>), it might
appear as:</p><div class="informalexample"><pre class="screen">
$ svn propget svn:externals calc
http://svn.example.com/repos/sounds third-party/sounds
http://svn.example.com/skinproj@148 third-party/skins
http://svn.example.com/skin-maker@21 third-party/skins/toolkit
</pre></div><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>You should seriously consider using explicit revision
numbers in all of your externals definitions. Doing so means
that you get to decide when to pull down a different snapshot
of external information, and exactly which snapshot to pull.
Besides avoiding the surprise of getting changes to
third-party repositories that you might not have any control
over, using explicit revision numbers also means that as you
backdate your working copy to a previous revision, your
externals definitions will also revert to the way they looked
in that previous revision, which in turn means that the
external working copies will be updated to match the way
<span class="emphasis"><em>they</em></span> looked back when your repository was
at that previous revision. For software projects, this could
be the difference between a successful and a failed build of
an older snapshot of your complex codebase.</p></td></tr></table></div><p>For most repositories, these three ways of formatting the
externals definitions have the same ultimate effect. They all
bring the same benefits. Unfortunately, they all bring the same
annoyances, too. Since the definitions shown use absolute URLs,
moving or copying a directory to which they are attached will
not affect what gets checked out as an external (though the
relative local target subdirectory will, of course, move with the
renamed directory). This can be confusing—even
frustrating—in certain situations. For example, say you
have a top-level directory named
<code class="filename">my-project</code>, and you've created an externals
definition on one of its subdirectories
(<code class="filename">my-project/some-dir</code>) that tracks the
latest revision of another of its subdirectories
(<code class="filename">my-project/external-dir</code>).</p><div class="informalexample"><pre class="screen">
$ svn checkout http://svn.example.com/projects .
A my-project
A my-project/some-dir
A my-project/external-dir
…
Fetching external item into 'my-project/some-dir/subdir'
Checked out external at revision 11.
Checked out revision 11.
$ svn propget svn:externals my-project/some-dir
subdir http://svn.example.com/projects/my-project/external-dir
$
</pre></div><p>Now you use <span class="command"><strong>svn move</strong></span> to rename the
<code class="filename">my-project</code> directory. At this point, your
externals definition will still refer to a path under the
<code class="filename">my-project</code> directory, even though that
directory no longer exists.</p><div class="informalexample"><pre class="screen">
$ svn move -q my-project renamed-project
$ svn commit -m "Rename my-project to renamed-project."
Deleting my-project
Adding renamed-project
Committed revision 12.
$ svn update
Updating '.':
svn: warning: W200000: Error handling externals definition for 'renamed-projec
t/some-dir/subdir':
svn: warning: W170000: URL 'http://svn.example.com/projects/my-project/externa
l-dir' at revision 12 doesn't exist
At revision 12.
svn: E205011: Failure occurred processing one or more externals definitions
$
</pre></div><p>Also, absolute URLs can cause problems with repositories
that are available via multiple URL schemes. For example, if
your Subversion server is configured to allow everyone to check
out the repository over <code class="literal">http://</code> or
<code class="literal">https://</code>, but only allow commits to come in
via <code class="literal">https://</code>, you have an interesting problem
on your hands. If your externals definitions use the
<code class="literal">http://</code> form of the repository URLs, you
won't be able to commit anything from the working copies created
by those externals. On the other hand, if they use the
<code class="literal">https://</code> form of the URLs, anyone who might
be checking out via <code class="literal">http://</code> because his
client doesn't support <code class="literal">https://</code> will be
unable to fetch the external items. Be aware, too, that if you
need to reparent your working copy (using <span class="command"><strong>svn
relocate</strong></span>), externals definitions will
<span class="emphasis"><em>not</em></span> also be reparented.</p><p>Subversion 1.5 takes a huge step in relieving these
frustrations. As mentioned earlier, the URLs used in the new
externals definition format can be relative, and Subversion
provides syntax magic for specifying multiple flavors of URL
relativity.</p><div class="variablelist"><a id="svn.advanced.externals.urlmagic"></a><dl><dt><span class="term"><code class="literal">../</code></span></dt><dd><p>Relative to the URL of the directory on which
the <code class="literal">svn:externals</code> property is
set</p></dd><dt><span class="term"><code class="literal">^/</code></span></dt><dd><p>Relative to the root of the repository in
which the <code class="literal">svn:externals</code> property is
versioned</p></dd><dt><span class="term"><code class="literal">//</code></span></dt><dd><p>Relative to the scheme of the URL of the
directory on which the <code class="literal">svn:externals</code>
property is set</p></dd><dt><span class="term"><code class="literal">/</code></span></dt><dd><p>Relative to the root URL of the server on
which the <code class="literal">svn:externals</code> property is
versioned</p></dd><dt><span class="term"><code class="literal">^/../<em class="replaceable"><code>REPO-NAME</code></em></code></span></dt><dd><p>Relative to a sibling repository beneath the
same <code class="literal">SVNParentPath</code> location as the
repository in which the <code class="literal">svn:externals</code> is
defined.</p></dd></dl></div><p>So, looking a fourth time at our previous externals
definition example, and making use of the new absolute URL
syntax in various ways, we might now see:</p><div class="informalexample"><pre class="screen">
$ svn propget svn:externals calc
^/sounds third-party/sounds
/skinproj@148 third-party/skins
//svn.example.com/skin-maker@21 third-party/skins/toolkit
$
</pre></div><p>Subversion 1.6 brought two more improvements to externals
definitions. First, it added a quoting and escape mechanism to
the syntax so that the path of the external working copy may
contain whitespace. This was previously problematic, of course,
because whitespace is used to delimit the fields in an externals
definition. Now you need only wrap such a path specification in
double-quote (<code class="literal">"</code>) characters or
escape the problematic characters in the path with a backslash
(<code class="literal">\</code>) character. Of course, if you have spaces
in the <span class="emphasis"><em>URL</em></span> portion of the external
definition, you should use the standard URI-encoding mechanism
to represent those.</p><div class="informalexample"><pre class="screen">
$ svn propget svn:externals paint
http://svn.thirdparty.com/repos/My%20Project "My Project"
http://svn.thirdparty.com/repos/%22Quotes%20Too%22 \"Quotes\ Too\"
$
</pre></div><p>Subversion 1.6 also introduced support for external
definitions for files. <em class="firstterm">File externals</em>
are configured just like externals for directories and appear as
a versioned file in the working copy.</p><p>For example, let's say you had the file
<code class="filename">/trunk/bikeshed/blue.html</code> in your repository,
and you wanted this file, as it appeared in revision 40,
to appear in your working copy of <code class="filename">/trunk/www/</code>
as <code class="filename">green.html</code>.</p><p>The externals definition required to achieve this should
look familiar by now:</p><div class="informalexample"><pre class="screen">
$ svn propget svn:externals www/
^/trunk/bikeshed/blue.html@40 green.html
$ svn update
Updating '.':
Fetching external item into 'www'
E www/green.html
Updated external to revision 40.
Update to revision 103.
$ svn status
X www/green.html
$
</pre></div><p>As you can see in the previous output, Subversion denotes file
externals with the letter <code class="literal">E</code> when they are
fetched into the working copy, and with the letter
<code class="literal">X</code> when showing the working copy status.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>While directory externals can place the external
directory at any depth, and any missing intermediate
directories will be created, file externals must be placed
into a working copy that is already checked out.</p></td></tr></table></div><p>When examining the file external with
<span class="command"><strong>svn info</strong></span>, you can see the URL and revision
the external is coming from:</p><div class="informalexample"><pre class="screen">
$ svn info www/green.html
Path: www/green.html
Name: green.html
Working Copy Root Path: /home/harry/projects/my-project
URL: http://svn.example.com/projects/my-project/trunk/bikeshed/blue.html
Repository Root: http://svn.example.com/projects/my-project
Repository UUID: b2a368dc-7564-11de-bb2b-113435390e17
Revision: 40
Node kind: file
Schedule: normal
Last Changed Author: harry
Last Changed Rev: 40
Last Changed Date: 2009-07-20 20:38:20 +0100 (Mon, 20 Jul 2009)
Text Last Updated: 2009-07-20 23:22:36 +0100 (Mon, 20 Jul 2009)
Checksum: 01a58b04617b92492d99662c3837b33b
$
</pre></div><p>Because file externals appear in the working copy as
versioned files, they can be modified and even committed
if they reference a file at the HEAD revision. The committed
changes will then appear in the external as well as the file
referenced by the external. However, in our example, we pinned
the external to an older revision, so attempting to commit
the external fails:</p><div class="informalexample"><pre class="screen">
$ svn status
M X www/green.html
$ svn commit -m "change the color" www/green.html
Sending www/green.html
svn: E155011: Commit failed (details follow):
svn: E155011: File '/trunk/bikeshed/blue.html' is out of date
$
</pre></div><p>Keep this in mind when defining file externals.
If you need the external to refer to a certain revision
of a file you will not be able to modify the external.
If you want to be able to modify the external, you cannot
specify a revision other than the <code class="literal">HEAD</code>
revision, which is implied if no revision is specified.</p><p>Unfortunately, the support which exists for externals definitions
in Subversion remains less than ideal. Both file and directory
externals have shortcomings. For either type of external, the
local subdirectory part of the definition cannot contain
<code class="literal">..</code> parent directory indicators (such as
<code class="filename">../../skins/myskin</code>). File externals cannot
refer to files from other repositories. A file external's URL
must always be in the same repository as the URL that the file
external will be inserted into. Also, file externals cannot be
moved or deleted. The <code class="literal">svn:externals</code> property
must be modified instead. However, file externals can be copied.</p><p>Perhaps most disappointingly, the working copies created via the
externals definition support are still disconnected from the primary
working copy (on whose versioned directories the
<code class="literal">svn:externals</code> property was actually set).
And Subversion still truly operates only on nondisjoint working
copies. So, for example, if you want to commit changes that
you've made in one or more of those external working copies, you
must run <span class="command"><strong>svn commit</strong></span> explicitly on those
working copies—committing on the primary working copy will
not recurse into any external ones.</p><p>We've already mentioned some of the additional shortcomings
of the old <code class="literal">svn:externals</code> format and how the
newer Subversion 1.5 format improves upon it. But be careful
when making use of the new format that you don't inadvertently
introduce new problems. For example, while the latest clients
will continue to recognize and support the original externals
definition format, pre-1.5 clients will <span class="emphasis"><em>not</em></span>
be able to correctly parse the new format. If you change all
your externals definitions to the newer format, you effectively
force everyone who uses those externals to upgrade their
Subversion clients to a version that can parse them. Also, be
careful to avoid naively relocating
the <code class="literal">-r<em class="replaceable"><code>NNN</code></em></code> portion
of the definition—the older format uses that revision as a
peg revision, but the newer format uses it as an operative
revision (with a peg revision of <code class="literal">HEAD</code> unless
otherwise specified; see <a class="xref" href="#svn.advanced.pegrevs" title="Peg and Operative Revisions">the section called “Peg and Operative Revisions”</a>
for a full explanation of the distinction here).</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>External working copies are still completely
self-sufficient working copies. You can operate directly on
them as you would any other working copy. This can be a handy
feature, allowing you to examine an external working copy
independently of any primary working copy
whose <code class="literal">svn:externals</code> property caused its
instantiation. Be careful, though, that you don't
inadvertently modify your external working copy in subtle ways
that cause problems. For example, while an externals
definition might specify that the external working copy should
be held at a particular revision number, if you
run <span class="command"><strong>svn update</strong></span> directly on the external
working copy, Subversion will oblige, and now your external
working copy is out of sync with its declaration in the
primary working copy. Using <span class="command"><strong>svn switch</strong></span> to
directly switch the external working copy (or some portion
thereof) to another URL could cause similar problems if the
contents of the primary working copy are expecting particular
contents in the external content.</p></td></tr></table></div><p>Besides the <span class="command"><strong>svn checkout</strong></span>, <span class="command"><strong>svn
update</strong></span>, <span class="command"><strong>svn switch</strong></span>, and
<span class="command"><strong>svn export</strong></span> commands which actually manage the
<em class="firstterm">disjoint</em> (or disconnected) subdirectories
into which externals are checked out, the <span class="command"><strong>svn
status</strong></span> command also recognizes externals definitions.
It displays a status code of <code class="literal">X</code> for the
disjoint external subdirectories, and then recurses into those
subdirectories to display the status of the external items
themselves. You can pass the
<code class="option">--ignore-externals</code> option to any of these
subcommands to disable externals definition processing.</p></div><div class="sect1" title="Changelists"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.advanced.changelists"></a>Changelists</h2></div></div></div><p>It is commonplace for a developer to find himself working at
any given time on multiple different, distinct changes to a
particular bit of source code. This isn't necessarily due to
poor planning or some form of digital masochism. A software
engineer often spots bugs in his peripheral vision while working
on some nearby chunk of source code. Or perhaps he's halfway
through some large change when he realizes the solution he's
working on is best committed as several smaller logical units.
Often, these logical units aren't nicely contained in some
module, safely separated from other changes. The units might
overlap, modifying different files in the same module, or even
modifying different lines in the same file.</p><p>Developers can employ various work methodologies
to keep these logical changes organized. Some use
separate working copies of the same repository to hold each
individual change in progress. Others might choose to create
short-lived feature branches in the repository and use a single
working copy that is constantly switched to point to one such
branch or another. Still others use <span class="command"><strong>diff</strong></span> and
<span class="command"><strong>patch</strong></span> tools to back up and restore uncommitted
changes to and from patch files associated with each change.
Each of these methods has its pros and cons, and to a large
degree, the details of the changes being made heavily influence
the methodology used to distinguish them.</p><p>Subversion provides a <em class="firstterm">changelists</em>
feature that adds yet another method to the mix. Changelists
are basically arbitrary labels (currently at most one per file)
applied to working copy files for the express purpose of
associating multiple files together. Users of many of Google's
software offerings are familiar with this concept already. For
example, <a class="ulink" href="http://mail.google.com/" target="_top">Gmail</a>
doesn't provide the traditional folders-based email organization
mechanism. In Gmail, you apply arbitrary labels to emails, and
multiple emails can be said to be part of the same group if they
happen to share a particular label. Viewing only a group of
similarly labeled emails then becomes a simple user interface
trick. Many other Web 2.0 sites have similar
mechanisms—consider the <span class="quote">“<span class="quote">tags</span>”</span> used by sites
such as <a class="ulink" href="http://www.youtube.com/" target="_top">YouTube</a> and
<a class="ulink" href="http://www.flickr.com/" target="_top">Flickr</a>,
<span class="quote">“<span class="quote">categories</span>”</span> applied to blog posts, and so on.
Folks understand today that organization of data is critical,
but that how that data is organized needs to be a flexible
concept. The old files-and-folders paradigm is too rigid for
some applications.</p><p>Subversion's changelist support allows you to create
changelists by applying labels to files you want to be
associated with that changelist, remove those labels, and limit
the scope of the files on which its subcommands operate to only
those bearing a particular label. In this section, we'll look
in detail at how to do these things.</p><div class="sect2" title="Creating and Modifying Changelists"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.changelists.creating"></a>Creating and Modifying Changelists</h3></div></div></div><p>You can create, modify, and delete changelists using the
<span class="command"><strong>svn changelist</strong></span> command. More accurately,
you use this command to set or unset the changelist
association of a particular working copy file. A changelist
is effectively created the first time you label a file with
that changelist; it is deleted when you remove that label from
the last file that had it. Let's examine a usage scenario
that demonstrates these concepts.</p><p>Harry is fixing some bugs in the calculator application's
mathematics logic. His work leads him to change a couple of
files:</p><div class="informalexample"><pre class="screen">
$ svn status
M integer.c
M mathops.c
$
</pre></div><p>While testing his bug fix, Harry notices that his changes
bring to light a tangentially related bug in the user
interface logic found in <code class="filename">button.c</code>. Harry
decides that he'll go ahead and fix that bug, too, as a
separate commit from his math fixes. Now, in a small working
copy with only a handful of files and few logical changes,
Harry can probably keep his two logical change groupings
mentally organized without any problem. But today he's going
to use Subversion's changelists feature as a special favor to
the authors of this book.</p><p>Harry first creates a changelist and associates with it
the two files he's already changed. He does this by using the
<span class="command"><strong>svn changelist</strong></span> command to assign the same
arbitrary changelist name to those files:</p><div class="informalexample"><pre class="screen">
$ svn changelist math-fixes integer.c mathops.c
A [math-fixes] integer.c
A [math-fixes] mathops.c
$ svn status
--- Changelist 'math-fixes':
M integer.c
M mathops.c
$
</pre></div><p>As you can see, the output of <span class="command"><strong>svn
status</strong></span> reflects this new grouping.</p><p>Harry now sets off to fix the secondary UI problem. Since
he knows which file he'll be changing, he assigns that path to
a changelist, too. Unfortunately, Harry carelessly assigns this
third file to the same changelist as the previous two files:</p><div class="informalexample"><pre class="screen">
$ svn changelist math-fixes button.c
A [math-fixes] button.c
$ svn status
--- Changelist 'math-fixes':
button.c
M integer.c
M mathops.c
$
</pre></div><p>Fortunately, Harry catches his mistake. At this point, he
has two options. He can remove the changelist association
from <code class="filename">button.c</code>, and then assign a
different changelist name:</p><div class="informalexample"><pre class="screen">
$ svn changelist --remove button.c
D [math-fixes] button.c
$ svn changelist ui-fix button.c
A [ui-fix] button.c
$
</pre></div><p>Or, he can skip the removal and just assign a new
changelist name. In this case, Subversion will first warn
Harry that <code class="filename">button.c</code> is being removed from
the first changelist:</p><div class="informalexample"><pre class="screen">
$ svn changelist ui-fix button.c
D [math-fixes] button.c
A [ui-fix] button.c
$ svn status
--- Changelist 'ui-fix':
button.c
--- Changelist 'math-fixes':
M integer.c
M mathops.c
$
</pre></div><p>Harry now has two distinct changelists present in his
working copy, and <span class="command"><strong>svn status</strong></span> will group its
output according to these changelist determinations. Notice
that even though Harry hasn't yet modified
<code class="filename">button.c</code>, it still shows up in the output
of <span class="command"><strong>svn status</strong></span> as interesting because it has
a changelist assignment. Changelists can be added to and
removed from files at any time, regardless of whether they
contain local modifications.</p><p>Harry now fixes the user interface problem in
<code class="filename">button.c</code>.</p><div class="informalexample"><pre class="screen">
$ svn status
--- Changelist 'ui-fix':
M button.c
--- Changelist 'math-fixes':
M integer.c
M mathops.c
$
</pre></div></div><div class="sect2" title="Changelists As Operation Filters"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.changelists.asfilters"></a>Changelists As Operation Filters</h3></div></div></div><p>The visual grouping that Harry sees in the output of
<span class="command"><strong>svn status</strong></span> as shown in our previous section
is nice, but not entirely useful. The
<span class="command"><strong>status</strong></span> command is but one of many
operations that he might wish to perform on his working copy.
Fortunately, many of Subversion's other operations understand
how to operate on changelists via the use of the
<code class="option">--changelist</code> option.</p><p>When provided with a <code class="option">--changelist</code> option,
Subversion commands will limit the scope of their operation to
only those files to which a particular changelist name is
assigned. If Harry now wants to see the actual changes he's
made to the files in his <code class="literal">math-fixes</code>
changelist, he <span class="emphasis"><em>could</em></span> explicitly list only
the files that make up that changelist on the <span class="command"><strong>svn
diff</strong></span> command line.</p><div class="informalexample"><pre class="screen">
$ svn diff integer.c mathops.c
Index: integer.c
===================================================================
--- integer.c (revision 1157)
+++ integer.c (working copy)
…
Index: mathops.c
===================================================================
--- mathops.c (revision 1157)
+++ mathops.c (working copy)
…
$
</pre></div><p>That works okay for a few files, but what if Harry's
change touched 20 or 30 files? That would be an annoyingly
long list of explicitly named files. Now that he's using
changelists, though, Harry can avoid explicitly listing the
set of files in his changelist from now on, and instead
provide just the changelist name:</p><div class="informalexample"><pre class="screen">
$ svn diff --changelist math-fixes
Index: integer.c
===================================================================
--- integer.c (revision 1157)
+++ integer.c (working copy)
…
Index: mathops.c
===================================================================
--- mathops.c (revision 1157)
+++ mathops.c (working copy)
…
$
</pre></div><p>And when it's time to commit, Harry can again use the
<code class="option">--changelist</code> option to limit the scope of the
commit to files in a certain changelist. He might commit his
user interface fix by doing the following:</p><div class="informalexample"><pre class="screen">
$ svn commit -m "Fix a UI bug found while working on math logic." \
--changelist ui-fix
Sending button.c
Transmitting file data .
Committed revision 1158.
$
</pre></div><p>In fact, the <span class="command"><strong>svn commit</strong></span> command
provides a second changelists-related option:
<code class="option">--keep-changelists</code>. Normally, changelist
assignments are removed from files after they are committed.
But if <code class="option">--keep-changelists</code> is provided,
Subversion will leave the changelist assignment on the
committed (and now unmodified) files. In any case, committing
files assigned to one changelist leaves other changelists
undisturbed.</p><div class="informalexample"><pre class="screen">
$ svn status
--- Changelist 'math-fixes':
M integer.c
M mathops.c
$
</pre></div><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>The <code class="option">--changelist</code> option acts only as a
filter for Subversion command targets, and will not add
targets to an operation. For example, on a commit operation
specified as <strong class="userinput"><code>svn commit /path/to/dir</code></strong>, the
target is the directory <code class="filename">/path/to/dir</code>
and its children (to infinite depth). If you then add a
changelist specifier to that command, only those files in
and under <code class="filename">/path/to/dir</code> that are
assigned that changelist name will be considered as targets
of the commit—the commit will not include files
located elsewhere (such as in
<code class="filename">/path/to/another-dir</code>), regardless of
their changelist assignment, even if they are part of the
same working copy as the operation's target(s).</p></td></tr></table></div><p>Even the <span class="command"><strong>svn changelist</strong></span> command accepts
the <code class="option">--changelist</code> option. This allows you to
quickly and easily rename or remove a changelist:</p><div class="informalexample"><pre class="screen">
$ svn changelist math-bugs --changelist math-fixes --depth infinity .
D [math-fixes] integer.c
A [math-bugs] integer.c
D [math-fixes] mathops.c
A [math-bugs] mathops.c
$ svn changelist --remove --changelist math-bugs --depth infinity .
D [math-bugs] integer.c
D [math-bugs] mathops.c
$
</pre></div><p>Finally, you can specify multiple instances of the
<code class="option">--changelist</code> option on a single command
line. Doing so limits the operation you are performing to
files found in any of the specified changesets.</p></div><div class="sect2" title="Changelist Limitations"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.changelists.limitations"></a>Changelist Limitations</h3></div></div></div><p>Subversion's changelist feature is a handy tool for
grouping working copy files, but it does have a few limitations.
Changelists are artifacts of a particular working copy, which
means that changelist assignments cannot be propagated to the
repository or otherwise shared with other users. Changelists
can be assigned only to files—Subversion doesn't
currently support the use of changelists with directories.
Finally, you can have at most one changelist assignment on a
given working copy file. Here is where the blog post category
and photo service tag analogies break down—if you find
yourself needing to assign a file to multiple changelists,
you're out of luck.</p></div></div><div class="sect1" title="Network Model"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.serverconfig.netmodel"></a>Network Model</h2></div></div></div><p>At some point, you're going to need to understand how your
Subversion client communicates with its server. Subversion's
networking layer is abstracted, meaning that Subversion clients
exhibit the same general behaviors no matter what sort of server
they are operating against. Whether speaking the HTTP protocol
(<code class="literal">http://</code>) with the Apache HTTP Server or
speaking the custom Subversion protocol
(<code class="literal">svn://</code>) with <span class="command"><strong>svnserve</strong></span>,
the basic network model is the same. In this section, we'll
explain the basics of that network model, including how
Subversion manages authentication and authorization
matters.</p><div class="sect2" title="Requests and Responses"><div class="titlepage"><div><div><h3 class="title"><a id="svn.serverconfig.netmodel.reqresp"></a>Requests and Responses</h3></div></div></div><p>The Subversion client spends most of its time managing
working copies. When it needs information from a remote
repository, however, it makes a network request, and the
server responds with an appropriate answer. The details of
the network protocol are hidden from the user—the client
attempts to access a URL, and depending on the URL scheme, a
particular protocol is used to contact the server (see <a class="xref" href="#svn.advanced.reposurls" title="Addressing the Repository">the section called “Addressing the Repository”</a>).</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>Run <strong class="userinput"><code>svn --version</code></strong> to see which
URL schemes and protocols the client knows how to use.</p></td></tr></table></div><p>When the server process receives a client request, it
often demands that the client identify itself. It issues
an authentication challenge to the client, and the client
responds by providing <em class="firstterm">credentials</em> back
to the server. Once authentication is complete, the server
responds with the original information that the client asked for.
Notice that this system is different from systems such as CVS,
where the client preemptively offers credentials (<span class="quote">“<span class="quote">logs
in</span>”</span>) to the server before ever making a request. In
Subversion, the server <span class="quote">“<span class="quote">pulls</span>”</span> credentials by
challenging the client at the appropriate moment, rather than
the client <span class="quote">“<span class="quote">pushing</span>”</span> them. This makes certain
operations more elegant. For example, if a server is
configured to allow anyone in the world to read a repository,
the server will never issue an authentication challenge
when a client attempts to <span class="command"><strong>svn checkout</strong></span>.</p><p>If the particular network requests issued by the client
result in a new revision being created in the repository
(e.g., <span class="command"><strong>svn commit</strong></span>), Subversion uses the
authenticated username associated with those requests as the
author of the revision. That is, the authenticated user's
name is stored as the value of the
<code class="literal">svn:author</code> property on the new revision
(see <a class="xref" href="#svn.ref.properties" title="Subversion Properties">the section called “Subversion Properties”</a> in
<a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a>). If the client was not
authenticated (i.e., if the server never issued an
authentication challenge), the revision's
<code class="literal">svn:author</code> property is empty.</p></div><div class="sect2" title="Client Credentials"><div class="titlepage"><div><div><h3 class="title"><a id="svn.serverconfig.netmodel.creds"></a>Client Credentials</h3></div></div></div><p>Many Subversion servers are configured to require
authentication. Sometimes anonymous read operations are
allowed, while write operations must be authenticated. In
other cases, reads and writes alike require authentication.
Subversion's different server options understand different
authentication protocols, but from the user's point of view,
authentication typically boils down to usernames and
passwords. Subversion clients offer several different ways to
retrieve and store a user's authentication credentials, from
interactive prompting for usernames and passwords to encrypted
and non-encrypted on-disk data caches.</p><p>The security-conscious reader will suspect immediately
that there is reason for concern here. <span class="quote">“<span class="quote">Caching
passwords on disk? That's terrible! You should never do
that!</span>”</span> Don't worry—it's not as bad as it sounds.
The following sections discuss the various types of credential
caches that Subversion uses, when it uses them, and how to
disable that functionality in whole or in part.</p><div class="sect3" title="Caching credentials"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.netmodel.credcache"></a>Caching credentials</h4></div></div></div><p>Subversion offers a remedy for the annoyance caused when
users are forced to type their usernames and passwords over
and over again. By default, whenever the command-line
client successfully responds to a server's authentication
challenge, credentials are cached on disk and keyed on a
combination of the server's hostname, port, and
authentication realm. This cache will then be automatically
consulted in the future, avoiding the need for the user to
re-type his or her authentication credentials. If seemingly
suitable credentials are not present in the cache, or if the
cached credentials ultimately fail to authenticate, the
client will, by default, fall back to prompting the user for
the necessary information.</p><p>The Subversion developers recognize that on-disk caches
of authentication credentials can be a security risk. To
offset this, Subversion works with available mechanisms
provided by the operating system and environment to try to
minimize the risk of leaking this information.</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>On Windows, the Subversion client stores passwords
in the <code class="filename">%APPDATA%/Subversion/auth/</code>
directory. On Windows 2000 and later, the standard
Windows cryptography services are used to encrypt the
password on disk. Because the encryption key is managed
by Windows and is tied to the user's own login
credentials, only the user can decrypt the cached
password. (Note that if the user's Windows account
password is reset by an administrator, all of the cached
passwords become undecipherable. The Subversion client
will behave as though they don't exist, prompting for
passwords when required.)</p></li><li class="listitem"><p>Similarly, on Mac OS X, the Subversion client stores
all repository passwords in the login keyring (managed
by the Keychain service), which is protected by the
user's account password. User preference settings can
impose additional policies, such as requiring that the
user's account password be entered each time the
Subversion password is used.</p></li><li class="listitem"><p>For other Unix-like operating systems, no single
standard <span class="quote">“<span class="quote">keychain</span>”</span> service exists.
However, the Subversion client knows how to store
passwords securely using the <span class="quote">“<span class="quote">GNOME Keyring</span>”</span>
and <span class="quote">“<span class="quote">KDE Wallet</span>”</span> services. Also, before
storing unencrypted passwords in the
<code class="filename">~/.subversion/auth/</code> caching area,
the Subversion client will ask the user for permission
to do so. Note that the <code class="filename">auth/</code>
caching area is still permission-protected so that only
the user (owner) can read data from it, not the world at
large. The operating system's own file permissions protect
the passwords from other non-administrative users on the
same system, provided they have no direct physical access
to the storage media of the home directory, or backups
thereof.</p></li></ul></div><p>Of course, for the truly paranoid, none of these
mechanisms meets the test of perfection. So for those folks
willing to sacrifice convenience for the ultimate in security,
Subversion provides various ways of disabling its credentials
caching system altogether.</p></div><div class="sect3" title="Disabling password caching"><div class="titlepage"><div><div><h4 class="title"><a id="svn.tour.initial.disabling-password-caching"></a>Disabling password caching</h4></div></div></div><p>When you perform a Subversion operation that requires
you to authenticate, by default Subversion tries to cache
your authentication credentials on disk in encrypted form.
On some systems, Subversion may be unable to encrypt your
authentication data. In those situations, Subversion will
ask whether you want to cache your credentials to disk in
plaintext:</p><div class="informalexample"><pre class="screen">
$ svn checkout https://host.example.com:443/svn/private-repo
-----------------------------------------------------------------------
ATTENTION! Your password for authentication realm:
<https://host.example.com:443> Subversion Repository
can only be stored to disk unencrypted! You are advised to configure
your system so that Subversion can store passwords encrypted, if
possible. See the documentation for details.
You can avoid future appearances of this warning by setting the value
of the 'store-plaintext-passwords' option to either 'yes' or 'no' in
'/tmp/servers'.
-----------------------------------------------------------------------
Store password unencrypted (yes/no)?
</pre></div><p>If you want the convenience of not having to continually
reenter your password for future operations, you can
answer <code class="literal">yes</code> to this prompt. If you're
concerned about caching your Subversion passwords in
plaintext and do not want to be asked about it again and
again, you can disable caching of plaintext passwords either
permanently, or on a server-per-server basis.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>When considering how to use Subversion's password
caching system, you'll want to consult any governing
policies that are in place for your client
computer—many companies have strict rules about the
ways that their employees' authentication credentials
should be stored.</p></td></tr></table></div><p>To permanently disable caching of passwords in
plaintext, add the line <code class="literal">store-plaintext-passwords =
no</code> to the <code class="literal">[global]</code> section in
the <code class="filename">servers</code> configuration file on the
local machine. To disable plaintext password caching for a
particular server, use the same setting in the appropriate
group section in the <code class="filename">servers</code>
configuration file. (See
<a class="xref" href="#svn.advanced.confarea.opts" title="Configuration Options">the section called “Configuration Options”</a> in
<a class="xref" href="#svn.customization" title="Chapter 7. Customizing Your Subversion Experience">Chapter 7, <em>Customizing Your Subversion Experience</em></a> for details.)</p><p>To disable password caching entirely for any single
Subversion command-line operation, pass
the <code class="option">--no-auth-cache</code> option to that command
line. To permanently disable caching entirely, add the
line <code class="literal">store-passwords = no</code> to your local
machine's Subversion configuration file.</p></div><div class="sect3" title="Removing cached credentials"><div class="titlepage"><div><div><h4 class="title"><a id="svn.tour.initial.authn-cache-purge"></a>Removing cached credentials</h4></div></div></div><p>Sometimes users will want to remove specific credentials
from the disk cache. To do this, you need to navigate into
the <code class="filename">auth/</code> area and manually delete the
appropriate cache file. Credentials are cached in individual
files; if you look inside each file, you will see keys and
values. The <code class="literal">svn:realmstring</code> key describes
the particular server realm that the file is associated
with:</p><div class="informalexample"><pre class="screen">
$ ls ~/.subversion/auth/svn.simple/
5671adf2865e267db74f09ba6f872c28
3893ed123b39500bca8a0b382839198e
5c3c22968347b390f349ff340196ed39
$ cat ~/.subversion/auth/svn.simple/5671adf2865e267db74f09ba6f872c28
K 8
username
V 3
joe
K 8
password
V 4
blah
K 15
svn:realmstring
V 45
<https://svn.domain.com:443> Joe's repository
END
</pre></div><p>Once you have located the proper cache file, just delete
it.</p></div><div class="sect3" title="Command-line authentication"><div class="titlepage"><div><div><h4 class="title"><a id="svn.tour.initial.different-user"></a>Command-line authentication</h4></div></div></div><p>All Subversion command-line operations accept
the <code class="option">--username</code>
and <code class="option">--password</code> options, which allow you to
specify your username and password, respectively, so that
Subversion isn't forced to prompt you for that information.
This is especially handy if you need to invoke Subversion
from a script and cannot rely on Subversion being able to
locate valid cached credentials for you. These options are
also helpful when Subversion has already cached
authentication credentials for you, but you know they aren't
the ones you want it to use. Perhaps several system users
share a login to the system, but each have distinct
Subversion identities. You can omit
the <code class="option">--password</code> option from this pair if
you wish Subversion to use only the provided username, but
still prompt you for that username's password.</p></div><div class="sect3" title="Authentication wrap-up"><div class="titlepage"><div><div><h4 class="title"><a id="svn.tour.initial.authn-wrapup"></a>Authentication wrap-up</h4></div></div></div><p>One last word about <span class="command"><strong>svn</strong></span>'s
authentication behavior, specifically regarding the
<code class="option">--username</code> and <code class="option">--password</code>
options. Many client subcommands accept these options, but it
is important to understand that using these options does
<span class="emphasis"><em>not</em></span> automatically send credentials to the
server. As discussed earlier, the server <span class="quote">“<span class="quote">pulls</span>”</span>
credentials from the client when it deems necessary; the
client cannot <span class="quote">“<span class="quote">push</span>”</span> them at will. If a username
and/or password are passed as options, they will be
presented to the server only if the server requests them. These
options are typically used to authenticate as a different user
than Subversion would have chosen by default (such as your
system login name) or when trying to avoid interactive
prompting (such as when calling <span class="command"><strong>svn</strong></span> from a
script).</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>A common mistake is to misconfigure a server so
that it never issues an authentication challenge. When
users pass <code class="option">--username</code> and
<code class="option">--password</code> options to the client, they're
surprised to see that they're never used; that is, new
revisions still appear to have been committed
anonymously!</p></td></tr></table></div><p>Here is a final summary that describes how a Subversion
client behaves when it receives an authentication
challenge.</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p>First, the client checks whether the user specified
any credentials as command-line options
(<code class="option">--username</code> and/or
<code class="option">--password</code>). If so, the client will try
to use those credentials to authenticate against the
server.</p></li><li class="listitem"><p>If no command-line credentials were provided, or the
provided ones were invalid, the client looks up the server's
hostname, port, and realm in the runtime configuration's
<code class="filename">auth/</code> area, to see whether appropriate
credentials are cached there. If so, it attempts to use
those credentials to authenticate.</p></li><li class="listitem"><p>Finally, if the previous mechanisms failed to
successfully authenticate the user against the server, the
client resorts to interactively prompting the user for
valid credentials (unless instructed not to do so via the
<code class="option">--non-interactive</code> option or its
client-specific equivalents).</p></li></ol></div><p>If the client successfully authenticates by any of these
methods, it will attempt to cache the credentials on disk
(unless the user has disabled this behavior, as mentioned
earlier).</p></div></div></div><div class="sect1" title="Summary"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.advanced.summary"></a>Summary</h2></div></div></div><p>After reading this chapter, you should have a firm grasp on
some of Subversion's features that, while perhaps not used
<span class="emphasis"><em>every</em></span> time you interact with your version
control system, are certainly handy to know about. But don't
stop here! Read on to the following chapter, where you'll learn
about branches, tags, and merging. Then you'll have nearly full
mastery of the Subversion client. Though our lawyers won't
allow us to promise you anything, this additional knowledge
could make you measurably more cool.<sup>[<a id="idp8722928" href="#ftn.idp8722928" class="footnote">24</a>]</sup></p></div><div class="footnotes"><br></br><hr width="100" align="left"></hr><div class="footnote"><p><sup>[<a id="ftn.idp7797328" href="#idp7797328" class="para">11</a>] </sup><span class="quote">“<span class="quote">You're not
supposed to name it. Once you name it, you start getting
attached to it.</span>”</span>—Mike Wazowski</p></div><div class="footnote"><p><sup>[<a id="ftn.idp7810704" href="#idp7810704" class="para">12</a>] </sup>606 N. Main Street, Wheaton, Illinois,
is the home of the Wheaton <span class="emphasis"><em>History</em></span> Center.
It seemed appropriate….</p></div><div class="footnote"><p><sup>[<a id="ftn.idp7930048" href="#idp7930048" class="para">13</a>] </sup>If you're familiar with
XML, this is pretty much the ASCII subset of the syntax for
XML <span class="quote">“<span class="quote">Name</span>”</span>.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp7956704" href="#idp7956704" class="para">14</a>] </sup>Fixing spelling errors, grammatical
gotchas, and <span class="quote">“<span class="quote">just-plain-wrongness</span>”</span> in commit log
messages is perhaps the most common use case for
the <code class="option">--revprop</code> option.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp8012000" href="#idp8012000" class="para">15</a>] </sup>Currently,
libmagic is the support library used to accomplish
this.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp8045712" href="#idp8045712" class="para">16</a>] </sup>You think that was rough? During that
same era, WordPerfect also used <code class="filename">.DOC</code>
for their proprietary file format's preferred
extension!</p></div><div class="footnote"><p><sup>[<a id="ftn.idp7677888" href="#idp7677888" class="para">17</a>] </sup>The Windows filesystems use file
extensions (such
as <code class="filename">.EXE</code>, <code class="filename">.BAT</code>, and
<code class="filename">.COM</code>) to denote executable
files.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp8162720" href="#idp8162720" class="para">18</a>] </sup>Isn't that the whole point of a build
system?</p></div><div class="footnote"><p><sup>[<a id="ftn.idp8226528" href="#idp8226528" class="para">19</a>] </sup>… or maybe even a section of a
book …</p></div><div class="footnote"><p><sup>[<a id="ftn.idp8323744" href="#idp8323744" class="para">20</a>] </sup>Safely, of course. As in
other situations, Subversion will leave on disk any files you've
modified or which aren't versioned.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp8354048" href="#idp8354048" class="para">21</a>] </sup>Communication wouldn't have been such bad
medicine for Harry and Sally's Hollywood namesakes, either, for
that matter.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp8357920" href="#idp8357920" class="para">22</a>] </sup>Subversion does not currently allow locks on
directories.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp8467072" href="#idp8467072" class="para">23</a>] </sup>Except, perhaps, a classic
Vulcan mind-meld.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp8722928" href="#idp8722928" class="para">24</a>] </sup>No purchase
necessary. Certains terms and conditions apply. No guarantee
of coolness—implicit or otherwise—exists. Mileage
may vary.</p></div></div></div><div class="chapter" title="Chapter 4. Branching and Merging"><div class="titlepage"><div><div><h2 class="title"><a id="svn.branchmerge"></a>Chapter 4. Branching and Merging</h2></div></div></div><div class="toc"><p><strong>Table of Contents</strong></p><dl><dt><span class="sect1"><a href="#svn.branchmerge.whatis">What's a Branch?</a></span></dt><dt><span class="sect1"><a href="#svn.branchmerge.using">Using Branches</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.branchmerge.using.create">Creating a Branch</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.using.work">Working with Your Branch</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.using.concepts">The Key Concepts Behind Branching</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.branchmerge.basicmerging">Basic Merging</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.branchmerge.changesets">Changesets</a></span></dt><dt><span class="sect2"><a href="#svn.branchemerge.basicmerging.stayinsync">Keeping a Branch in Sync</a></span></dt><dt><span class="sect2"><a href="#svn.branchemerge.basicmerging.reintegrate">Reintegrating a Branch</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.basicmerging.mergeinfo">Mergeinfo and Previews</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.basicmerging.undo">Undoing Changes</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.basicmerging.resurrect">Resurrecting Deleted Items</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.branchmerge.advanced">Advanced Merging</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.branchmerge.cherrypicking">Cherrypicking</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.advanced.advancedsyntax">Merge Syntax: Full Disclosure</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.nomergedata">Merges Without Mergeinfo</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.advanced.mergeconflicts">More on Merge Conflicts</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.advanced.blockchanges">Blocking Changes</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.advanced.reintegratetwice">Keeping a Reintegrated Branch Alive</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.advanced.logblame">Merge-Sensitive Logs and Annotations</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.advanced.ancestry">Noticing or Ignoring Ancestry</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.advanced.moves">Merges and Moves</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.advanced.pre1.5clients">Blocking Merge-Unaware Clients</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.advanced.finalword">The Final Word on Merge Tracking</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.branchmerge.switchwc">Traversing Branches</a></span></dt><dt><span class="sect1"><a href="#svn.branchmerge.tags">Tags</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.branchmerge.tags.mksimple">Creating a Simple Tag</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.tags.mkcomplex">Creating a Complex Tag</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.branchmerge.maint">Branch Maintenance</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.branchmerge.maint.layout">Repository Layout</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.maint.lifetime">Data Lifetimes</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.branchmerge.commonpatterns">Common Branching Patterns</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.branchmerge.commonpatterns.release">Release Branches</a></span></dt><dt><span class="sect2"><a href="#svn.branchmerge.commonpatterns.feature">Feature Branches</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.advanced.vendorbr">Vendor Branches</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.advanced.vendorbr.general">General Vendor Branch Management Procedure</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.vendorbr.svn_load_dirs">svn_load_dirs.pl</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.branchmerge.when">To Branch or Not to Branch?</a></span></dt><dt><span class="sect1"><a href="#svn.branchmerge.summary">Summary</a></span></dt></dl></div><div class="blockquote"><table border="0" width="100%" cellspacing="0" cellpadding="0" class="blockquote" summary="Block quote"><tr><td width="10%" valign="top"> </td><td width="80%" valign="top"><p><span class="quote">“<span class="quote">君子务本
(It is upon the Trunk that a gentleman works.)</span>”</span></p></td><td width="10%" valign="top"> </td></tr><tr><td width="10%" valign="top"> </td><td colspan="2" align="right" valign="top">--<span class="attribution">Confucius</span></td></tr></table></div><p>Branching and merging are fundamental aspects of version
control, simple enough to explain conceptually but offering just
enough complexity and nuance to merit their own chapter in this
book. Herein, we'll introduce you to the general ideas behind
these operations as well as Subversion's somewhat unique approach
to them. If you've not familiarized yourself with Subversion's
basic concepts (found in <a class="xref" href="#svn.basic" title="Chapter 1. Fundamental Concepts">Chapter 1, <em>Fundamental Concepts</em></a>), we
recommmend that you do so before reading this chapter.</p><div class="sect1" title="What's a Branch?"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.branchmerge.whatis"></a>What's a Branch?</h2></div></div></div><p>Suppose it's your job to maintain a document for a division
in your company—a handbook of some sort. One day a different
division asks you for the same handbook, but with a few parts
<span class="quote">“<span class="quote">tweaked</span>”</span> for them, since they do things slightly
differently.</p><p>What do you do in this situation? You do the obvious: make
a second copy of your document and begin maintaining the two
copies separately. As each department asks you to make small
changes, you incorporate them into one copy or the other.</p><p>You often want to make the same change to both copies. For
example, if you discover a typo in the first copy, it's very
likely that the same typo exists in the second copy. The two
documents are almost the same, after all; they differ only in
small, specific ways.</p><p>
<a id="idp9745376" class="indexterm"></a>
This is the basic concept of a branch—namely,
a line of development that exists independently of another line,
yet still shares a common history if you look far enough back in
time. A branch always begins life as a copy of something, and
moves on from there, generating its own history (see
<a class="xref" href="#svn.branchmerge.whatis.dia-1" title="Figure 4.1. Branches of development">Figure 4.1, “Branches of development”</a>).</p><div class="figure"><a id="svn.branchmerge.whatis.dia-1"></a><p class="title"><strong>Figure 4.1. Branches of development</strong></p><div class="figure-contents"><div><img src="images/ch04dia1.png" alt="Branches of development"></img></div></div></div><br class="figure-break"></br><p>Subversion has commands to help you maintain parallel
branches of your files and directories. It allows you to create
branches by copying your data, and remembers that the copies are
related to one another. It also helps you duplicate changes
from one branch to another. Finally, it can make portions of
your working copy reflect different branches so that you can
<span class="quote">“<span class="quote">mix and match</span>”</span> different lines of development in
your daily work.</p></div><div class="sect1" title="Using Branches"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.branchmerge.using"></a>Using Branches</h2></div></div></div><p>At this point, you should understand how each commit creates
a new state of the filesystem tree (called a <span class="quote">“<span class="quote">revision</span>”</span>)
in the repository. If you don't, go back and read about revisions in
<a class="xref" href="#svn.basic.in-action.revs" title="Revisions">the section called “Revisions”</a>.</p><p>Let's revisit the example from
<a class="xref" href="#svn.basic" title="Chapter 1. Fundamental Concepts">Chapter 1, <em>Fundamental Concepts</em></a>. Remember that you and your
collaborator, Sally, are sharing a repository that contains two
projects, <code class="filename">paint</code> and
<code class="filename">calc</code>. Notice that in <a class="xref" href="#svn.branchmerge.using.dia-1" title="Figure 4.2. Starting repository layout">Figure 4.2, “Starting repository layout”</a>, however, each project
directory now contains subdirectories named
<code class="filename">trunk</code> and <code class="filename">branches</code>.
The reason for this will soon become clear.</p><div class="figure"><a id="svn.branchmerge.using.dia-1"></a><p class="title"><strong>Figure 4.2. Starting repository layout</strong></p><div class="figure-contents"><div><img src="images/ch04dia2.png" alt="Starting repository layout"></img></div></div></div><br class="figure-break"></br><p>As before, assume that Sally and you both have working
copies of the <span class="quote">“<span class="quote">calc</span>”</span> project. Specifically, you
each have a working copy of <code class="filename">/calc/trunk</code>.
All the files for the project are in this subdirectory rather
than in <code class="filename">/calc</code> itself, because your team has
decided that <code class="filename">/calc/trunk</code> is where the
<span class="quote">“<span class="quote">main line</span>”</span> of development is going to take
place.</p><p>Let's say that you've been given the task of implementing a
large software feature. It will take a long time to write, and
will affect all the files in the project. The immediate problem
is that you don't want to interfere with Sally, who is in the
process of fixing small bugs here and there. She's depending on
the fact that the latest version of the project (in
<code class="filename">/calc/trunk</code>) is always usable. If you
start committing your changes bit by bit, you'll surely break
things for Sally (and other team members as well).</p><p>One strategy is to crawl into a hole: you can stop sharing
information for a week or two, gutting and reorganizing all the
files in your private working copy but not committing or
updating until you're completely finished with your task. There
are a number of problems with this, though. First, it's not
very safe. Should something bad happen to your working copy or
computer, you risk losing all your changes. Second, it's not
very flexible. Unless you manually replicate your changes
across different working copies or computers, you're stuck trying
to make your changes in a single working copy. Similarly, it's
difficult to share your work-in-progress with anyone else. A
common software development <span class="quote">“<span class="quote">best practice</span>”</span> is to
allow your peers to review your work as you go. If nobody sees
your intermediate commits, you lose potential feedback and may
end up going down the wrong path for weeks before another person
on your team notices. Finally, when you're finished with all
your changes, you might find it very difficult to merge your
completed work with the rest of the company's main body of code.
Sally (or others) may have made many other changes in the
repository that are difficult to incorporate into your working
copy when you eventually run <span class="command"><strong>svn update</strong></span> after
weeks of isolation.</p><p>The better solution is to create your own branch, or line of
development, in the repository. This allows you to save your
not-yet-completed work frequently without interfering with
others' changes and while still selectively sharing information
with your collaborators. You'll see exactly how this works as
we continue.</p><div class="sect2" title="Creating a Branch"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.using.create"></a>Creating a Branch</h3></div></div></div><p>Creating a branch is very simple—you make a copy of
your project tree in the repository using the <span class="command"><strong>svn
copy</strong></span> command. Since your project's source code is
rooted in the <code class="filename">/calc/trunk</code> directory, it's
that directory that you'll copy. Where should the new
copy live? Wherever you wish. The repository location in
which branches are stashed is left by Subversion as a matter
of project policy. Finally, your branch will need a name to
distinguish it from other branches. Once again, the name you
choose is unimportant to Subversion—you can use whatever
name works best for you and your team.</p><p>Let's assume that your team (like most) has a policy of
creating branches in the <code class="filename">branches</code>
directory that is a sibling of the project's trunk
(the <code class="filename">/calc/branches</code> directory in our
scenario). Lacking inspiration, you settle
on <code class="literal">my-calc-branch</code> as the name you wish to
give your branch. This means that you'll create a new
directory, <code class="filename">/calc/branches/my-calc-branch</code>,
which begins its life as a copy
of <code class="filename">/calc/trunk</code>.</p><p>You may already have seen <span class="command"><strong>svn copy</strong></span> used
to copy one file to another within a working copy. But it can
also be used to do a <em class="firstterm">remote
copy</em>—a copy that immediately results in a
newly committed repository revision and for which no working
copy is required at all. Just copy one URL to another:</p><div class="informalexample"><pre class="screen">
$ svn copy http://svn.example.com/repos/calc/trunk \
http://svn.example.com/repos/calc/branches/my-calc-branch \
-m "Creating a private branch of /calc/trunk."
Committed revision 341.
$
</pre></div><p>This command causes a near-instantaneous commit in the
repository, creating a new directory in revision 341. The new
directory is a copy of <code class="filename">/calc/trunk</code>. This
is shown in <a class="xref" href="#svn.branchmerge.using.create.dia-1" title="Figure 4.3. Repository with new copy">Figure 4.3, “Repository with new copy”</a>.<sup>[<a id="idp9784560" href="#ftn.idp9784560" class="footnote">25</a>]</sup> While
it's also possible to create a branch by using <span class="command"><strong>svn
copy</strong></span> to duplicate a directory within the working
copy, this technique isn't recommended. It can be quite slow,
in fact! Copying a directory on the client side is a
linear-time operation, in that it actually has to duplicate
every file and subdirectory within that working copy directory
on the local disk. Copying a directory on the server,
however, is a constant-time operation, and it's the way most
people create branches.</p><div class="figure"><a id="svn.branchmerge.using.create.dia-1"></a><p class="title"><strong>Figure 4.3. Repository with new copy</strong></p><div class="figure-contents"><div><img src="images/ch04dia3.png" alt="Repository with new copy"></img></div></div></div><br class="figure-break"></br><div class="sidebar" title="Cheap Copies"><div class="titlepage"><div><div><p class="title"><strong>Cheap Copies</strong></p></div></div></div><p>Subversion's repository has a special design. When you
copy a directory, you don't need to worry about the
repository growing huge—Subversion doesn't actually
duplicate any data. Instead, it creates a new directory
entry that points to an <span class="emphasis"><em>existing</em></span> tree.
If you're an experienced Unix user, you'll recognize this as
the same concept behind a hard link. As further changes are
made to files and directories beneath the copied directory,
Subversion continues to employ this hard link concept where
it can. It duplicates data only when it is necessary to
disambiguate different versions of objects.</p><p>This is why you'll often hear Subversion users talk
about <span class="quote">“<span class="quote">cheap copies.</span>”</span> It doesn't matter how
large the directory is—it takes a very tiny, constant
amount of time and space to make a copy of it. In fact,
this feature is the basis of how commits work in Subversion:
each revision is a <span class="quote">“<span class="quote">cheap copy</span>”</span> of the previous
revision, with a few items lazily changed within. (To read
more about this, visit Subversion's web site and read about
the <span class="quote">“<span class="quote">bubble up</span>”</span> method in Subversion's design
documents.)</p><p>Of course, these internal mechanics of copying and
sharing data are hidden from the user, who simply sees
copies of trees. The main point here is that copies are
cheap, both in time and in space. If you create a branch
entirely within the repository (by running <strong class="userinput"><code>svn copy
<em class="replaceable"><code>URL1</code></em> <em class="replaceable"><code>URL2</code></em></code></strong>),
it's a quick, constant-time operation. Make branches as
often as you want.</p></div></div><div class="sect2" title="Working with Your Branch"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.using.work"></a>Working with Your Branch</h3></div></div></div><p>Now that you've created a branch of the project, you can
check out a new working copy to start using it:</p><div class="informalexample"><pre class="screen">
$ svn checkout http://svn.example.com/repos/calc/branches/my-calc-branch
A my-calc-branch/Makefile
A my-calc-branch/integer.c
A my-calc-branch/button.c
Checked out revision 341.
$
</pre></div><p>There's nothing special about this working copy; it simply
mirrors a different directory in the repository. When you
commit changes, however, Sally won't see them when she
updates, because her working copy is of
<code class="filename">/calc/trunk</code>. (Be sure to read <a class="xref" href="#svn.branchmerge.switchwc" title="Traversing Branches">the section called “Traversing Branches”</a> later in this chapter: the
<span class="command"><strong>svn switch</strong></span> command is an alternative way of
creating a working copy of a branch.)</p><p>Let's pretend that a week goes by, and the following
commits happen:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>You make a change to
<code class="filename">/calc/branches/my-calc-branch/button.c</code>,
which creates revision 342.</p></li><li class="listitem"><p>You make a change to
<code class="filename">/calc/branches/my-calc-branch/integer.c</code>,
which creates revision 343.</p></li><li class="listitem"><p>Sally makes a change to
<code class="filename">/calc/trunk/integer.c</code>, which creates
revision 344.</p></li></ul></div><p>Now two independent lines of development (shown
in <a class="xref" href="#svn.branchmerge.using.work.dia-1" title="Figure 4.4. The branching of one file's history">Figure 4.4, “The branching of one file's history”</a>) are happening on
<code class="filename">integer.c</code>.</p><div class="figure"><a id="svn.branchmerge.using.work.dia-1"></a><p class="title"><strong>Figure 4.4. The branching of one file's history</strong></p><div class="figure-contents"><div><img src="images/ch04dia4.png" alt="The branching of one file's history"></img></div></div></div><br class="figure-break"></br><p>Things get interesting when you look at the history of
changes made to your copy of <code class="filename">integer.c</code>:</p><div class="informalexample"><pre class="screen">
$ pwd
/home/user/my-calc-branch
$ svn log -v integer.c
------------------------------------------------------------------------
r343 | user | 2002-11-07 15:27:56 -0600 (Thu, 07 Nov 2002) | 2 lines
Changed paths:
M /calc/branches/my-calc-branch/integer.c
* integer.c: frozzled the wazjub.
------------------------------------------------------------------------
r341 | user | 2002-11-03 15:27:56 -0600 (Thu, 07 Nov 2002) | 2 lines
Changed paths:
A /calc/branches/my-calc-branch (from /calc/trunk:340)
Creating a private branch of /calc/trunk.
------------------------------------------------------------------------
r303 | sally | 2002-10-29 21:14:35 -0600 (Tue, 29 Oct 2002) | 2 lines
Changed paths:
M /calc/trunk/integer.c
* integer.c: changed a docstring.
------------------------------------------------------------------------
r98 | sally | 2002-02-22 15:35:29 -0600 (Fri, 22 Feb 2002) | 2 lines
Changed paths:
A /calc/trunk/integer.c
* integer.c: adding this file to the project.
------------------------------------------------------------------------
</pre></div><p>Notice that Subversion is tracing the history of your
branch's <code class="filename">integer.c</code> all the way back
through time, even traversing the point where it was copied.
It shows the creation of the branch as an event in the
history, because <code class="filename">integer.c</code> was implicitly
copied when all of <code class="filename">/calc/trunk/</code> was
copied. Now look at what happens when Sally runs the same
command on her copy of the file:</p><div class="informalexample"><pre class="screen">
$ pwd
/home/sally/calc
$ svn log -v integer.c
------------------------------------------------------------------------
r344 | sally | 2002-11-07 15:27:56 -0600 (Thu, 07 Nov 2002) | 2 lines
Changed paths:
M /calc/trunk/integer.c
* integer.c: fix a bunch of spelling errors.
------------------------------------------------------------------------
r303 | sally | 2002-10-29 21:14:35 -0600 (Tue, 29 Oct 2002) | 2 lines
Changed paths:
M /calc/trunk/integer.c
* integer.c: changed a docstring.
------------------------------------------------------------------------
r98 | sally | 2002-02-22 15:35:29 -0600 (Fri, 22 Feb 2002) | 2 lines
Changed paths:
A /calc/trunk/integer.c
* integer.c: adding this file to the project.
------------------------------------------------------------------------
</pre></div><p>Sally sees her own revision 344 change, but not the change
you made in revision 343. As far as Subversion is concerned,
these two commits affected different files in different
repository locations. However, Subversion
<span class="emphasis"><em>does</em></span> show that the two files share a
common history. Before the branch copy was made in revision
341, the files used to be the same file. That's why you and
Sally both see the changes made in revisions 303 and
98.</p></div><div class="sect2" title="The Key Concepts Behind Branching"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.using.concepts"></a>The Key Concepts Behind Branching</h3></div></div></div><p>You should remember two important lessons
from this section. First, Subversion has no internal concept
of a branch—it knows only how to make copies. When you
copy a directory, the resultant directory is only
a <span class="quote">“<span class="quote">branch</span>”</span> because <span class="emphasis"><em>you</em></span>
attach that meaning to it. You may think of the directory
differently, or treat it differently, but to Subversion it's
just an ordinary directory that happens to carry some extra
historical information.</p><p>Second, because of this copy mechanism, Subversion's
branches exist as <span class="emphasis"><em>normal filesystem
directories</em></span> in the repository. This is different
from other version control systems, where branches are
typically defined by adding
extra-dimensional <span class="quote">“<span class="quote">labels</span>”</span> to collections of
files. The location of your branch directory doesn't matter
to Subversion. Most teams follow a convention of putting all
branches into a <code class="filename">/branches</code> directory, but
you're free to invent any policy you wish.</p></div></div><div class="sect1" title="Basic Merging"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.branchmerge.basicmerging"></a>Basic Merging</h2></div></div></div><p>Now you and Sally are working on parallel branches of the
project: you're working on a private branch, and Sally is
working on the trunk, or main line of development.</p><p>For projects that have a large number of contributors, it's
common for most people to have working copies of the trunk.
Whenever someone needs to make a long-running change that is
likely to disrupt the trunk, a standard procedure is to create a
private branch and commit changes there until all the work is
complete.</p><p>So, the good news is that you and Sally aren't interfering
with each other. The bad news is that it's very easy to drift
<span class="emphasis"><em>too</em></span> far apart. Remember that one of the
problems with the <span class="quote">“<span class="quote">crawl in a hole</span>”</span> strategy is
that by the time you're finished with your branch, it may be
near-impossible to merge your changes back into the trunk
without a huge number of conflicts.</p><p>Instead, you and Sally might continue to share changes as
you work. It's up to you to decide which changes are worth
sharing; Subversion gives you the ability to selectively
<span class="quote">“<span class="quote">copy</span>”</span> changes between branches. And when you're
completely finished with your branch, your entire set of branch
changes can be copied back into the trunk. In Subversion
terminology, the general act of replicating changes from one
branch to another is called <em class="firstterm">merging</em>, and
it is performed using various invocations of the <span class="command"><strong>svn
merge</strong></span> subcommand.</p><p>In the examples that follow, we're assuming that both your
Subversion client and server are running Subversion 1.7 (or
later). If either client or server is older than version 1.5,
things are more complicated: the system won't track changes
automatically, forcing you to use painful manual methods to
achieve similar results. That is, you'll always need to use the
detailed merge syntax to specify specific ranges of revisions to
replicate (see
<a class="xref" href="#svn.branchmerge.advanced.advancedsyntax" title="Merge Syntax: Full Disclosure">the section called “Merge Syntax: Full Disclosure”</a> later
in this chapter), and take special care to keep track of what's
already been merged and what hasn't. For this reason,
we <span class="emphasis"><em>strongly</em></span> recommend that you make sure your
client and server are at least at version 1.5.</p><div class="sidebar" title="Merge Tracking"><a id="svn.branchmerge.basicmerging.mergetracking"></a><div class="titlepage"><div><div><p class="title"><strong>Merge Tracking</strong></p></div></div></div><p>
<a id="idp9841120" class="indexterm"></a>
Subversion 1.5 introduced the
<em class="firstterm">merge tracking</em> feature to Subversion.
Prior to this feature keeping track of merges required cumbersome
manual procedures or the use of external tools. Subsequent
releases of Subversion introduced many enhancements and bug
fixes to merge tracking, which is why we recommend using the
most recent versions on both your server and client. Keep in
mind that even if your server is running 1.5 or 1.6, you can still
use a 1.7 client. This is particularly important as regards merge
tracking, because the overwhelming majority of fixes to it are on
the client side.</p></div><div class="sect2" title="Changesets"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.changesets"></a>Changesets</h3></div></div></div><p>Before we proceed further, we should warn you that there's
a lot of discussion of <span class="quote">“<span class="quote">changes</span>”</span> in
the pages ahead. A lot of people experienced with version
control systems use the terms <span class="quote">“<span class="quote">change</span>”</span>
and <span class="quote">“<span class="quote">changeset</span>”</span> interchangeably, and we should
clarify what Subversion understands as
a <em class="firstterm">changeset</em>.</p><p>Everyone seems to have a slightly different definition
of changeset, or at least a different
expectation of what it means for a version control system to
have one. For our purposes, let's say that a changeset is just
a collection of changes with a unique name. The changes might
include textual edits to file contents, modifications to tree
structure, or tweaks to metadata. In more common speak, a
changeset is just a patch with a name you can refer to.</p><p>In Subversion, a global revision
number <em class="replaceable"><code>N</code></em> names a tree in the
repository: it's the way the repository looked after the
<em class="replaceable"><code>N</code></em>th commit. It's also the name of
an implicit changeset: if you compare
tree <em class="replaceable"><code>N</code></em> with
tree <em class="replaceable"><code>N</code></em>-1, you can derive the exact
patch that was committed. For this reason, it's easy to think
of revision <em class="replaceable"><code>N</code></em> as not just a tree,
but a changeset as well. If you use an issue tracker to
manage bugs, you can use the revision numbers to refer to
particular patches that fix bugs—for example,
<span class="quote">“<span class="quote">this issue was fixed by r9238.</span>”</span> Somebody
can then run <strong class="userinput"><code>svn log -r 9238</code></strong> to read about
the exact changeset that fixed the bug, and run
<strong class="userinput"><code>svn diff -c 9238</code></strong> to see the patch itself.
And (as you'll see shortly)
Subversion's <span class="command"><strong>svn merge</strong></span> command is able to use
revision numbers. You can merge specific changesets from one
branch to another by naming them in the merge
arguments: passing <strong class="userinput"><code>-c 9238</code></strong>
to <span class="command"><strong>svn merge</strong></span> would merge changeset r9238
into your working copy.</p></div><div class="sect2" title="Keeping a Branch in Sync"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchemerge.basicmerging.stayinsync"></a>Keeping a Branch in Sync</h3></div></div></div><p>Continuing with our running example, let's suppose that a
week has passed since you started working on your private
branch. Your new feature isn't finished yet, but at the same
time you know that other people on your team continue to make
important changes in the
project's <code class="filename">/trunk</code>. It's in your best
interest to replicate those changes to your own branch, just
to make sure they mesh well with your changes. This is done
by performing a <em class="firstterm">sync merge</em>—a
merge operation designed to bring your branch up to date with
any changes made to its ancestral parent branch since your
branch was created.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>Frequently keeping your branch in sync with the main
development line helps prevent <span class="quote">“<span class="quote">surprise</span>”</span>
conflicts when the time comes for you to fold your changes
back into the trunk.</p></td></tr></table></div><p>Subversion is aware of the history of your branch and
knows when it split away from the trunk. To perform a sync
merge, first make sure your working copy of the branch
is <span class="quote">“<span class="quote">clean</span>”</span>—that it has no local
modifications reported by <span class="command"><strong>svn status</strong></span>. Then
simply run:</p><div class="informalexample"><pre class="screen">
$ pwd
/home/user/my-calc-branch
$ svn merge ^/calc/trunk
--- Merging r345 through r356 into '.':
U button.c
U integer.c
--- Recording mergeinfo for merge of r345 through r356 into '.':
U .
$
</pre></div><p>This basic syntax—<strong class="userinput"><code>svn merge
<em class="replaceable"><code>URL</code></em></code></strong>—tells
Subversion to merge all changes which have not been previously
merged from the URL to the current working directory (which is
typically the root of your working copy). Notice that we're
using the caret (<code class="literal">^</code>)
syntax<sup>[<a id="idp9866272" href="#ftn.idp9866272" class="footnote">26</a>]</sup> to avoid having to type out the
entire <code class="filename">/trunk</code> URL. Also note
the <span class="quote">“<span class="quote">Recording mergeinfo for merge…</span>”</span>
notification. This tells you that the merge is updating
the <code class="literal">svn:mergeinfo</code> property. We'll discuss
both this property and these notifications later in this
chapter, in
<a class="xref" href="#svn.branchmerge.basicmerging.mergeinfo" title="Mergeinfo and Previews">the section called “Mergeinfo and Previews”</a>.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>
<a id="idp9870384" class="indexterm"></a>
In this book and elsewhere (Subversion mailing lists, articles
on merge tracking, etc.) you will frequently come across the
term <em class="firstterm">mergeinfo</em>. This is simply shorthand
for the <code class="literal">svn:mergeinfo</code> property.</p></td></tr></table></div><div class="sidebar" title="Keeping a Branch in Sync Without Merge Tracking"><div class="titlepage"><div><div><p class="title"><strong>Keeping a Branch in Sync Without Merge Tracking</strong></p></div></div></div><p>You may not always be able to use Subversion's merge
tracking feature, perhaps because your server is running
Subversion 1.4 or earlier. In such a scenario, you can of
course still perform merges, but Subversion will need you to
manually do many of the historical calculations that it
automatically does on your behalf when the merge tracking
feature is available.</p><p>To replicate the most recent trunk changes you need to
perform sync merges the <span class="quote">“<span class="quote">old-fashioned</span>”</span>
way—by specifying ranges of revisions you wish to
merge.</p><p>Let's say you branched <code class="filename">/trunk</code> to
<code class="filename">/branches/foo-feature</code> in revision
400:</p><div class="informalexample"><pre class="screen">
$ svn log -v -r 400 ^/branches/foo-feature
------------------------------------------------------------------------
r400 | carol | 2011-11-09 10:51:27 -0500 (Wed, 09 Nov 2011) | 1 line
Changed paths:
A /branch/b2 (from /trunk:399)
Create branch for the foo feature
------------------------------------------------------------------------
</pre></div><p>When you are ready to syncronize your branch with the
ongoing changes from trunk, you specify the starting
revision as the revision of <code class="filename">/trunk</code>
which the branch was copied from and the ending revision as
<code class="literal">HEAD</code>:</p><div class="informalexample"><pre class="screen">
$ svn merge ^/trunk -r399:HEAD
--- Merging r400 through r556 into '.':
A include/foo.h
U src/main.c
A src/foo.c
…
</pre></div><p>After any conflicts have been resolved, you can commit
the merged changed to your branch. Now, to avoid
accidentally trying to merge these same changes into your
branch again in the future, you'll need to record the fact
that you've already merged them. But where should that
record be kept? One of the simplest places to record this
information is in the log message for the commit of the
merge:</p><div class="informalexample"><pre class="screen">
$ svn ci -m "Sync the foo-feature branch with ^/trunk through r556."
Sending include/foo.h
…
Transmitting file data .
Committed revision 557.
</pre></div><p>The next time you sync
<code class="filename">/branches/foo-branch</code> with
<code class="filename">/trunk</code> you repeat this process, except that
the starting revision is the <span class="emphasis"><em>youngest</em></span>
revision that's already been merged in from the trunk.
If you've been keeping good records of your merges in the
commit log messages, you should be able to determine what
that youngest revision was by reading the revision logs
associated with your branch. Once you know your starting
revision, you can perform another sync merge:</p><div class="informalexample"><pre class="screen">
$ svn merge ^/trunk -r556:HEAD
…
</pre></div></div><p>After running the prior example, your branch working copy
now contains new local modifications, and these edits are
duplications of all of the changes that have happened on the
trunk since you first created your branch:</p><div class="informalexample"><pre class="screen">
$ svn status
M .
M button.c
M integer.c
$
</pre></div><p>At this point, the wise thing to do is look at the changes
carefully with <span class="command"><strong>svn diff</strong></span>, and then build and
test your branch. Notice that the current working directory
(<span class="quote">“<span class="quote"><code class="filename">.</code></span>”</span>) has also been
modified; <span class="command"><strong>svn diff</strong></span> will show that
its <code class="literal">svn:mergeinfo</code> property has been either
created or modified. This is important merge-related metadata
that you should <span class="emphasis"><em>not</em></span> touch, since it is
needed by future <span class="command"><strong>svn merge</strong></span> commands.
(We'll learn more about this metadata later in the
chapter.)</p><p>After performing the merge, you might also need to resolve
some conflicts—just as you do with <span class="command"><strong>svn
update</strong></span>—or possibly make some small edits to get
things working properly. (Remember, just because there are
no <span class="emphasis"><em>syntactic</em></span> conflicts doesn't mean there
aren't any <span class="emphasis"><em>semantic</em></span> conflicts!) If you
encounter serious problems, you can always abort the local
changes by running <strong class="userinput"><code>svn revert . -R</code></strong> (which
will undo all local modifications) and starting a
long <span class="quote">“<span class="quote">what's going on?</span>”</span> discussion with your
collaborators. If things look good, however, you can
submit these changes into the repository:</p><div class="informalexample"><pre class="screen">
$ svn commit -m "Merged latest trunk changes to my-calc-branch."
Sending .
Sending button.c
Sending integer.c
Transmitting file data ..
Committed revision 357.
$
</pre></div><p>At this point, your private branch is now <span class="quote">“<span class="quote">in
sync</span>”</span> with the trunk, so you can rest easier knowing
that as you continue to work in isolation, you're not drifting
too far away from what everyone else is doing.</p><div class="sidebar" title="Why Not Use Patches Instead?"><div class="titlepage"><div><div><p class="title"><strong>Why Not Use Patches Instead?</strong></p></div></div></div><p>A question may be on your mind, especially if you're a
Unix user: why bother to use <span class="command"><strong>svn merge</strong></span> at
all? Why not simply use <span class="command"><strong>svn patch</strong></span> or the
operating system's <span class="command"><strong>patch</strong></span> command to
accomplish the same job? For example:</p><div class="informalexample"><pre class="screen">
$ cd my-calc-branch
$ svn diff -r 341:HEAD ^/calc/trunk > my-patch-file
$ svn patch my-patch-file
U integer.c
$
</pre></div><p>In this particular example, there really isn't much
difference. But <span class="command"><strong>svn merge</strong></span> has special
abilities that surpass the <span class="command"><strong>patch</strong></span> program.
The file format used by <span class="command"><strong>patch</strong></span> is quite
limited; it's able to tweak file contents only. There's no
way to represent changes to <span class="emphasis"><em>trees</em></span>, such
as the addition, removal, or renaming of files and
directories. Nor can the <span class="command"><strong>patch</strong></span> program
notice changes to properties. If Sally's change had,
say, added a new directory, the output of <span class="command"><strong>svn
diff</strong></span> wouldn't have mentioned it at
all. <span class="command"><strong>svn diff</strong></span> outputs only the limited
patch format, so there are some ideas it simply can't
express. Even Subversion's own <span class="command"><strong>svn patch</strong></span>
subcommand, while more flexible than patch program, still has
similar limitations.</p><p>The <span class="command"><strong>svn merge</strong></span> command, however, can
express changes in tree structure and properties by directly
applying them to your working copy. Even more important,
this command records the changes that have been duplicated
to your branch so that Subversion is aware of exactly which
changes exist in each location (see
<a class="xref" href="#svn.branchmerge.basicmerging.mergeinfo" title="Mergeinfo and Previews">the section called “Mergeinfo and Previews”</a>).
This is a critical feature that makes branch management
usable; without it, users would have to manually keep notes
on which sets of changes have or haven't been merged
yet.</p></div><p>Suppose that another week has passed. You've committed
more changes to your branch, and your comrades have continued
to improve the trunk as well. Once again, you want to
replicate the latest trunk changes to your branch and bring
yourself in sync. Just run the same merge command
again!</p><div class="informalexample"><pre class="screen">
$ svn merge ^/calc/trunk
svn: E195020: Cannot merge into mixed-revision working copy [357:378]; try up\
dating first
$
</pre></div><p>Well that was unexpected! After making changes to your
branch over the past week you now find yourself with a working
copy that contains a mixture of revisions (see
<a class="xref" href="#svn.basic.in-action.mixedrevs" title="Mixed-revision working copies">the section called “Mixed-revision working copies”</a>). With the
release of Subversion 1.7 the <span class="command"><strong>svn merge</strong></span>
subcommand disables merges into mixed-revision working copies
by default. Without going into too much detail, this is
because of limitations in the way merges are tracked by the
<code class="literal">svn:mergeinfo</code> property (see
<a class="xref" href="#svn.branchmerge.basicmerging.mergeinfo" title="Mergeinfo and Previews">the section called “Mergeinfo and Previews”</a> for
details). These limitations mean that merges into
mixed-revision working copies can result in unexpected text
and tree conflicts.<sup>[<a id="idp9920496" href="#ftn.idp9920496" class="footnote">27</a>]</sup> We don't want any needless conflicts, so
we update the working copy and then reattempt the
merge.</p><div class="informalexample"><pre class="screen">
$ svn up
Updating '.':
At revision 380.
$ svn merge ^/calc/trunk
--- Merging r357 through r380 into '.':
U integer.c
U Makefile
A README
--- Recording mergeinfo for merge of r357 through r380 into '.':
U .
$
</pre></div><p>Subversion knows which trunk changes you previously
replicated to your branch, so it carefully replicates only
those changes you don't yet have. And once again, you build,
test, and <span class="command"><strong>svn commit</strong></span> the local modifications
to your branch.</p><div class="sidebar" title="Subtree Merges and Subtree Mergeinfo"><a id="svn.branchemerge.basicmerging.stayinsync.subtree"></a><div class="titlepage"><div><div><p class="title"><strong>Subtree Merges and Subtree Mergeinfo</strong></p></div></div></div><p>
<a id="idp9926704" class="indexterm"></a>
<a id="idp9927776" class="indexterm"></a>
In most of the examples in this chapter the merge target is
the root directory of a branch (see
<a class="xref" href="#svn.branchmerge.whatis" title="What's a Branch?">the section called “What's a Branch?”</a>). While this is a
best practice, you may occasionally need to merge directly to
some child of the branch root. This type of merge is called a
<em class="firstterm">subtree merge</em> and the mergeinfo recorded
to describe it is called
<em class="firstterm">subtree mergeinfo</em>. There is nothing
special about subtree merges or subtree mergeinfo. In fact
there is really only one important point to keep in mind
about these concepts: the complete record of merges to a
branch may not be contained solely in the mergeinfo on the
branch root. You may have to look to any subtree mergeinfo
to get a full accounting. Fortunately Subversion does this
for you and rarely will you need to concern yourself with
it. A brief example will help explain:</p><div class="informalexample"><pre class="screen">
# We need to merge r958 from trunk to branches/proj-X/doc/INSTALL,
# but that revision also affects main.c, which we don't want to merge:
$ svn log --verbose --quiet -r 958 ^/
------------------------------------------------------------------------
r958 | bruce | 2011-10-20 13:28:11 -0400 (Thu, 20 Oct 2011)
Changed paths:
M /trunk/doc/INSTALL
M /trunk/src/main.c
------------------------------------------------------------------------
# No problem, we'll do a subtree merge targeting the INSTALL file
# directly, but first take a note of what mergeinfo exists on the
# root of the branch:
$ cd branches/proj-X
$ svn propget svn:mergeinfo --recursive
Properties on '.':
svn:mergeinfo
/trunk:651-652
# Now we perform the subtree merge, note that merge source
# and target both point to INSTALL:
$ svn merge ^/trunk/doc/INSTALL doc/INSTALL -c 958
--- Merging r958 into 'doc/INSTALL':
U doc/INSTALL
--- Recording mergeinfo for merge of r958 into 'doc/INSTALL':
G doc/INSTALL
# Once the merge is complete there is now subtree mergeinfo on INSTALL:
$ svn propget svn:mergeinfo --recursive
Properties on '.':
svn:mergeinfo
/trunk:651-652
Properties on 'doc/INSTALL':
svn:mergeinfo
/trunk/doc/INSTALL:651-652,958
# What if we then decide we do want all of r958? Easy, all we need do is
# repeat the merge of that revision, but this time to the root of the
# branch, Subversion notices the subtree mergeinfo on INSTALL and doesn't
# try to merge any changes to it, only the changes to main.c are merged:
$ svn merge ^/subversion/trunk . -c 958
--- Merging r958 into '.':
U src/main.c
--- Recording mergeinfo for merge of r958 into '.':
U .
--- Eliding mergeinfo from 'doc/INSTALL':
U doc/INSTALL
</pre></div><a id="idp9934288" class="indexterm"></a><p>You might be wondering why <code class="filename">INSTALL</code>
in the above example has mergeinfo for r651-652, when we
only merged r958. This is due to mergeinfo inheritance,
which we'll cover in the sidebar
<a class="xref" href="#svn.branchmerge.basicmerging.mergeinfo.inheritance" title="Mergeinfo Inheritance">Mergeinfo Inheritance</a>. Also note that the subtree mergeinfo on
<code class="filename">doc/INSTALL</code> was removed, or
<span class="quote">“<span class="quote">elided</span>”</span>. This is called
<em class="firstterm">mergeinfo elision</em> and it occurs
whenever Subversion detects redundant subtree mergeinfo.</p></div><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>Prior to Subversion 1.7, merges unconditionally updated
<span class="emphasis"><em>all</em></span> of the subtree mergeinfo under the
target to describe the merge. For users with a lot of subtree
mergeinfo this meant that relatively <span class="quote">“<span class="quote">simple</span>”</span>
merges (e.g. one which applied a diff to only a single file)
resulted in changes to every subtree with mergeinfo, even
those that were not parents of the effected path(s). This
caused some level of confusion and frustration. Subversion 1.7
addresses this problem by only updating the mergeinfo on
subtrees which are parents of the paths modified by the merge
(i.e. paths changed, added, or deleted by application of the
difference, see
<a class="xref" href="#svn.branchmerge.advanced.advancedsyntax" title="Merge Syntax: Full Disclosure">the section called “Merge Syntax: Full Disclosure”</a>).
The one exception to this behavior regards the actual merge
target; the merge target's mergeinfo is always updated to
describe the merge, even if the applied difference made no
changes.</p></td></tr></table></div></div><div class="sect2" title="Reintegrating a Branch"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchemerge.basicmerging.reintegrate"></a>Reintegrating a Branch</h3></div></div></div><p>What happens when you finally finish your work, though?
Your new feature is done, and you're ready to merge your
branch changes back to the trunk (so your team can enjoy the
bounty of your labor). The process is simple. First, bring
your branch into sync with the trunk again, just as you've been
doing all along<sup>[<a id="idp9945040" href="#ftn.idp9945040" class="footnote">28</a>]</sup>:</p><div class="informalexample"><pre class="screen">
$ svn merge ^/calc/trunk
--- Merging r381 through r385 into '.':
U button.c
U README
--- Recording mergeinfo for merge of r381 through r385 into '.':
U .
$ # build, test, ...
$ svn commit -m "Final merge of trunk changes to my-calc-branch."
Sending .
Sending button.c
Sending README
Transmitting file data ..
Committed revision 390.
</pre></div><p>Now, use <span class="command"><strong>svn merge</strong></span> with the
<code class="option">--reintegrate</code> option to replicate your branch
changes back into the trunk. You'll need a working copy
of <code class="filename">/trunk</code>. You can get one by doing
an <span class="command"><strong>svn checkout</strong></span>, dredging up an old trunk
working copy from somewhere on your disk, or
using <span class="command"><strong>svn switch</strong></span> (see
<a class="xref" href="#svn.branchmerge.switchwc" title="Traversing Branches">the section called “Traversing Branches”</a>). Your trunk
working copy cannot have any local edits or contain a mixture
of revisions (see
<a class="xref" href="#svn.basic.in-action.mixedrevs" title="Mixed-revision working copies">the section called “Mixed-revision working copies”</a>). While
these are typically best practices for merging anyway, they
are <span class="emphasis"><em>required</em></span> when using the
<code class="option">--reintegrate</code> option.</p><p>Once you have a clean working copy of the trunk, you're
ready to merge your branch back into it:</p><div class="informalexample"><pre class="screen">
$ pwd
/home/user/calc-trunk
$ svn update # (make sure the working copy is up to date)
Updating '.':
At revision 390.
$ svn merge --reintegrate ^/calc/branches/my-calc-branch
--- Merging differences between repository URLs into '.':
U button.c
U integer.c
U Makefile
--- Recording mergeinfo for merge between repository URLs into '.':
U .
$ # build, test, verify, ...
$ svn commit -m "Merge my-calc-branch back into trunk!"
Sending .
Sending button.c
Sending integer.c
Sending Makefile
Transmitting file data ..
Committed revision 391.
</pre></div><p>Congratulations, your branch-specific changes have now
been merged back into the main line of development. Notice
our use of the <code class="option">--reintegrate</code> option this time
around. The option is critical for reintegrating changes from
a branch back into its original line of
development—don't forget it! It's needed because this
sort of <span class="quote">“<span class="quote">merge back</span>”</span> is a different sort of work
than what you've done up until now. Previously, we were
asking <span class="command"><strong>svn merge</strong></span> to grab the <span class="quote">“<span class="quote">next
set</span>”</span> of changes from one line of development (the
trunk) and duplicate them to another (your branch). This is
fairly straightforward, and each time Subversion knows how to
pick up where it left off. In our prior examples, you can see
that first it merges the ranges 345:356 from trunk to branch;
later on, it continues by merging the next contiguously
available range, 356:380. When doing the final sync, it
merges the range 380:385.</p><p>When merging your branch back to the trunk, however, the
underlying mathematics are quite different. Your feature
branch is now a mishmash of both duplicated trunk changes and
private branch changes, so there's no simple contiguous range
of revisions to copy over. By specifying
the <code class="option">--reintegrate</code> option, you're asking
Subversion to carefully replicate <span class="emphasis"><em>only</em></span>
those changes unique to your branch. (And in fact, it does
this by comparing the latest trunk tree with the latest branch
tree: the resulting difference is exactly your branch
changes!)</p><p>Keep in mind that the <code class="option">--reintegrate</code>
option is quite specialized in contrast to the more general
nature of most Subversion subcommand options. It supports the
use case described above, but has little applicability outside
of that. Because of this narrow focus, in addition to
requiring an up-to-date working copy<sup>[<a id="idp9962880" href="#ftn.idp9962880" class="footnote">29</a>]</sup> with no mixed-revisions,
it will not function in combination with most of the other
<span class="command"><strong>svn merge</strong></span> options. You'll get an error if you
use any non-global options but these: <code class="option">--accept</code>,
<code class="option">--dry-run</code>, <code class="option">--diff3-cmd</code>,
<code class="option">--extensions</code>, or <code class="option">--quiet</code>.</p><p>Now that your private branch is merged to trunk, you may
wish to remove it from the repository:</p><div class="informalexample"><pre class="screen">
$ svn delete ^/calc/branches/my-calc-branch \
-m "Remove my-calc-branch, reintegrated with trunk in r391."
Committed revision 392.
</pre></div><p>But wait! Isn't the history of that branch valuable?
What if somebody wants to audit the evolution of your feature
someday and look at all of your branch changes? No need to
worry. Remember that even though your branch is no longer
visible in the <code class="filename">/branches</code> directory, its
existence is still an immutable part of the repository's
history. A simple <span class="command"><strong>svn log</strong></span> command on
the <code class="filename">/branches</code> URL will show the entire
history of your branch. Your branch can even be resurrected
at some point, should you desire (see
<a class="xref" href="#svn.branchmerge.basicmerging.resurrect" title="Resurrecting Deleted Items">the section called “Resurrecting Deleted Items”</a>).</p><p>Once a <code class="option">--reintegrate</code> merge is done from
branch to trunk, the branch is no longer usable for further
work. It's not able to correctly absorb new trunk changes,
nor can it be properly reintegrated to trunk again. For this
reason, if you want to keep working on your feature branch, we
recommend destroying it and then re-creating it from the
trunk:</p><div class="informalexample"><pre class="screen">
$ svn delete http://svn.example.com/repos/calc/branches/my-calc-branch \
-m "Remove my-calc-branch, reintegrated with trunk in r391."
Committed revision 392.
$ svn copy http://svn.example.com/repos/calc/trunk \
http://svn.example.com/repos/calc/branches/my-calc-branch \
-m "Recreate my-calc-branch from trunk@HEAD."
Committed revision 393.
</pre></div><p>There is another way of making the branch usable again
after reintegration, without deleting the branch. See
<a class="xref" href="#svn.branchmerge.advanced.reintegratetwice" title="Keeping a Reintegrated Branch Alive">the section called “Keeping a Reintegrated Branch Alive”</a>.</p></div><div class="sect2" title="Mergeinfo and Previews"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.basicmerging.mergeinfo"></a>Mergeinfo and Previews</h3></div></div></div><p>The basic mechanism Subversion uses to track
changesets—that is, which changes have been merged to
which branches—is by recording data in versioned
properties. Specifically, merge data is tracked in
the <code class="literal">svn:mergeinfo</code> property attached to
files and directories. (If you're not familiar with
Subversion properties, see <a class="xref" href="#svn.advanced.props" title="Properties">the section called “Properties”</a>.)</p><p>You can examine the property, just like any other:</p><div class="informalexample"><pre class="screen">
$ cd my-calc-branch
$ svn propget svn:mergeinfo .
/trunk:341-390
$
</pre></div><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>While it is possible to
modify <code class="literal">svn:mergeinfo</code> just as you might
any other versioned property, we strongly discourage doing
so unless you <span class="emphasis"><em>really</em></span> know what you're
doing.</p></td></tr></table></div><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>The amount of <code class="literal">svn:mergeinfo</code> on a single
path can get quite large, as can the output of a
<span class="command"><strong>svn propget --recursive</strong></span> or
<span class="command"><strong>svn proplist --recursive</strong></span> when dealing with
large amounts of subtree mergeinfo, see
<a class="xref" href="#svn.branchemerge.basicmerging.stayinsync.subtree" title="Subtree Merges and Subtree Mergeinfo">Subtree Merges and Subtree Mergeinfo</a>
. The formatted output produced by the
<code class="option">--verbose</code> option with either of these
subcommands is often very helpful in these cases.</p></td></tr></table></div><p>The <code class="literal">svn:mergeinfo</code> property is
automatically maintained by Subversion whenever you
run <span class="command"><strong>svn merge</strong></span>. Its value indicates which
changes made to a given path have been replicated into the
directory in question. In our previous example, the path
which is the source of the merged changes is
<code class="filename">/trunk</code> and the directory which has
received the changes is
<code class="filename">/branches/my-calc-branch</code>.
Earlier versions of Subversion maintained the
<code class="literal">svn:mergeinfo</code> property silently. You could
still detect the changes, after a merge completed, with the
<span class="command"><strong>svn diff</strong></span> or <span class="command"><strong>svn status</strong></span>
subcommands, but the merge itself gave no indication when it
changed the <code class="literal">svn:mergeinfo</code> property. This is no
longer true in Subversion 1.7, which has several new notifications
to alert you when a merge updates the
<code class="literal">svn:mergeinfo</code> property. These notifications
all begin with <span class="quote">“<span class="quote">--- Recording mergeinfo for</span>”</span>
and appear at the end of the merge. Unlike other merge
notifications, these don't describe the application of a
difference to a working copy
(see <a class="xref" href="#svn.branchmerge.advanced.advancedsyntax" title="Merge Syntax: Full Disclosure">the section called “Merge Syntax: Full Disclosure”</a>),
but instead describe "housekeeping" changes made to keep
track of what was merged.</p><p>Subversion also provides a subcommand, <span class="command"><strong>svn
mergeinfo</strong></span>, which is helpful in seeing not only
which changesets a directory has absorbed, but also which
changesets it's still eligible to receive. This gives a sort
of preview of which changes a subsequent <span class="command"><strong>svn
merge</strong></span> operation would replicate to your
branch.</p><div class="informalexample"><pre class="screen">
$ cd my-calc-branch
# Which changes have already been merged from trunk to branch?
$ svn mergeinfo ^/calc/trunk
r341
r342
r343
…
r388
r389
r390
# Which changes are still eligible to merge from trunk to branch?
$ svn mergeinfo ^/calc/trunk --show-revs eligible
r391
r392
r393
r394
r395
$
</pre></div><p>The <span class="command"><strong>svn mergeinfo</strong></span> command requires
a <span class="quote">“<span class="quote">source</span>”</span> URL (where the changes come
from), and takes an optional <span class="quote">“<span class="quote">target</span>”</span> URL (where
the changes merge to). If no target URL is given,
it assumes that the current working directory is the
target. In the prior example, because we're querying our
branch working copy, the command assumes we're interested in
receiving changes to <code class="filename">/branches/mybranch</code>
from the specified trunk URL.</p><div class="sidebar" title="Mergeinfo Inheritance"><a id="svn.branchmerge.basicmerging.mergeinfo.inheritance"></a><div class="titlepage"><div><div><p class="title"><strong>Mergeinfo Inheritance</strong></p></div></div></div><a id="idp10006496" class="indexterm"></a><p>When a path has the <code class="literal">svn:mergeinfo</code>
property set on it we say it has <em class="firstterm">explicit
mergeinfo</em>. This explicit mergeinfo describes
not only what changes were merged into that particular
directory, but also all the children of that directory
(because those children inherit the mergeinfo of their
parent path). For example:</p><div class="informalexample"><pre class="screen">
# What explicit mergeinfo exists on a branch?
$ svn propget svn:mergeinfo ^/branches/proj-X --recursive
/trunk:651-652
# What children does proj-X have?
$ svn list --recursive ^/branches/proj-X
doc/
doc/INSTALL
README
src/main.c
# Ask what revs were merged to a file with no explicit mergeinfo
$ svn mergeinfo ^/trunk/src/main.c ^/branches/proj-X/src/main.c
651
652
</pre></div><p>Notice from our first subcommand that only the root of
<code class="filename">/branches/proj-X</code> has any explicit
mergeinfo. However, when we use
<span class="command"><strong>svn mergeinfo</strong></span> to ask what was merged to
<code class="filename">/branches/proj-X/src/main.c</code> it reports
that the two revisions described in the explicit mergeinfo
on <code class="filename">/branches/proj-X</code> were merged. This is
because <code class="filename">/branches/proj-X/src/main.c</code>, having
no explicit mergeinfo of its own, inherits the mergeinfo from
its nearest parent with explicit mergeinfo,
<code class="filename">/branches/proj-X</code>.</p><p>There are two cases in which mergeinfo is not inherited.
First, if a path has explicit mergeinfo, then it never inherits
mergeinfo. Another way to think of this is that explicit
mergeinfo is always a complete record of the merges to a given
path, once it exists it overrides any mergeinfo that path might
otherwise inherit. The second way is when dealing with
non-inheritable mergeinfo, a special type of explicit mergeinfo
that applies <span class="emphasis"><em>only</em></span> to the directory on which
the <code class="literal">svn:mergeinfo</code> property is set (and it's
only directories, non-inheritable mergeinfo is never set on
files). For example:</p><div class="informalexample"><pre class="screen">
# The '*' decorator indicates non-inheritable mergeinfo
$ svn propget svn:mergeinfo ^/branches/proj-X
/trunk:651-652,758*
# Revision 758 is non-inheritable, but still applies to the path it is
# set on. Here the '*' decorator signals that r758 is only partially
# merged from trunk.
$ svn mergeinfo ^/trunk ^/branches/proj-X
651
652
758*
# Revision 758 is not reported as merged because it is non-inheritable
# and applies only to ^/trunk
$ svn mergeinfo ^/trunk/src/main.c ^/branches/proj-X/src/main.c
651
652
</pre></div><p>You might never have to think about mergeinfo inheritance
or encounter non-inheritable mergeinfo in your own repository.
A discussion of the full ramifications of mergeinfo inheritance
are beyond the scope of this book. If you have more questions
check out some of the references mentioned in
<a class="xref" href="#svn.branchmerge.advanced.finalword" title="The Final Word on Merge Tracking">the section called “The Final Word on Merge Tracking”</a></p></div><p>With the release of Subversion 1.7, the
<span class="command"><strong>svn mergeinfo</strong></span> subcommand can also account for
subtree mergeinfo and non-inheritable mergeinfo. It accounts for
subtree mergeinfo by use of the <code class="option">--recursive</code> or
<code class="option">--depth</code> options, while non-inheritable mergeinfo
is considered by default.</p><p>Let's say we have a branch with both subtree and
non-inheritable mergeinfo:</p><div class="informalexample"><pre class="screen">
$ svn propget svn:mergeinfo --recursive -v
# Non-inheritable mergeinfo
Properties on '.':
svn:mergeinfo
/trunk:651-652,758*
# Subtree mergeinfo
Properties on 'doc/INSTALL':
svn:mergeinfo
/trunk/doc/INSTALL:651-652,958,1060
</pre></div><p>From the above mergeinfo we see that r758 has only been
merged into the root of the branch, but not any of the root's
children. We also see that both r958 and r1060 have been
merged only to the <code class="filename">doc/INSTALL</code> file.
When we use <span class="command"><strong>svn mergeinfo</strong></span> with the
<code class="option">--recursive</code> option to see what has been merged
from <code class="filename">^/trunk</code> to this branch, we see two
revisions are flagged with the <code class="literal">*</code> marker:</p><div class="informalexample"><pre class="screen">
$ svn mergeinfo --show-revs=merged ^/trunk . --recursive
651
652
758*
958*
1060
</pre></div><p>The <code class="literal">*</code> indicates revisions that are only
<span class="emphasis"><em>partially</em></span> merged to the target in question
(the meaning is the same if we are checking for eligible
revisions). What this means in this example is that if we tried
to merge r758 or r958 from <code class="filename">^/trunk</code> then more
changes would result. Likewise, because r1060 is
<span class="emphasis"><em>not</em></span> flagged with a <code class="literal">*</code>,
we know that it only affects <code class="filename">doc/INSTALL</code>
and that trying to merge it would have no result.<sup>[<a id="idp10042976" href="#ftn.idp10042976" class="footnote">30</a>]</sup></p><p>Another way to get a more precise preview of a merge
operation is to use the <code class="option">--dry-run</code>
option:</p><div class="informalexample"><pre class="screen">
$ svn merge ^/calc/trunk --dry-run
--- Merging r391 through r395 into 'branch':
U integer.c
$ svn status
# nothing printed, working copy is still unchanged.
</pre></div><p>The <code class="option">--dry-run</code> option doesn't actually
apply any local changes to the working copy. It shows only
status codes that <span class="emphasis"><em>would</em></span> be printed in a
real merge. It's useful for getting a <span class="quote">“<span class="quote">high-level</span>”</span>
preview of the potential merge, for those times
when running <span class="command"><strong>svn diff</strong></span> gives too much
detail.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>After performing a merge operation, but before
committing the results of the merge, you can
use <strong class="userinput"><code>svn diff
--depth=empty <em class="replaceable"><code>/path/to/merge/target</code></em></code></strong>
to see only the changes to the immediate target of your
merge. If your merge target was a directory, only property
differences are displayed. This is a handy way to see
the changes to the
<code class="literal">svn:mergeinfo</code> property recorded by the
merge operation, which will remind you about what you've
just merged.</p></td></tr></table></div><p>Of course, the best way to preview a merge operation is to
just do it! Remember, running <span class="command"><strong>svn merge</strong></span>
isn't an inherently risky thing (unless you've made local
modifications to your working copy—but we already
stressed that you shouldn't merge into such an
environment). If you don't like the results of the merge,
simply run <strong class="userinput"><code>svn revert . -R</code></strong> to revert
the changes from your working copy and retry the command with
different options. The merge isn't final until you
actually <span class="command"><strong>svn commit</strong></span> the results.</p></div><div class="sect2" title="Undoing Changes"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.basicmerging.undo"></a>Undoing Changes</h3></div></div></div><p>An extremely common use for <span class="command"><strong>svn merge</strong></span>
is to roll back a change that has already been committed.
Suppose you're working away happily on a working copy of
<code class="filename">/calc/trunk</code>, and you discover that the
change made way back in revision 303, which changed
<code class="filename">integer.c</code>, is completely wrong. It never
should have been committed. You can use <span class="command"><strong>svn
merge</strong></span> to <span class="quote">“<span class="quote">undo</span>”</span> the change in your
working copy, and then commit the local modification to the
repository. All you need to do is to specify a
<span class="emphasis"><em>reverse</em></span> difference. (You can do this by
specifying <code class="option">--revision 303:302</code>, or by an
equivalent <code class="option">--change -303</code>.)</p><div class="informalexample"><pre class="screen">
$ svn merge -c -303 ^/calc/trunk
--- Reverse-merging r303 into 'integer.c':
U integer.c
--- Recording mergeinfo for reverse merge of r303 into 'integer.c':
U A-branch
$ svn status
M .
M integer.c
$ svn diff
…
# verify that the change is removed
…
$ svn commit -m "Undoing change committed in r303."
Sending integer.c
Transmitting file data .
Committed revision 350.
</pre></div><p>As we mentioned earlier, one way to think about a
repository revision is as a specific changeset. By using the
<code class="option">-r</code> option, you can ask <span class="command"><strong>svn
merge</strong></span> to apply a changeset, or a whole range of
changesets, to your working copy. In our case of undoing a
change, we're asking <span class="command"><strong>svn merge</strong></span> to apply
changeset r303 to our working copy
<span class="emphasis"><em>backward</em></span>.</p><p>Keep in mind that rolling back a change like this is just
like any other <span class="command"><strong>svn merge</strong></span> operation, so you
should use <span class="command"><strong>svn status</strong></span> and <span class="command"><strong>svn
diff</strong></span> to confirm that your work is in the state you
want it to be in, and then use <span class="command"><strong>svn commit</strong></span>
to send the final version to the repository. After
committing, this particular changeset is no longer reflected
in the <code class="literal">HEAD</code> revision.</p><p>Again, you may be thinking: well, that really didn't undo
the commit, did it? The change still exists in revision 303.
If somebody checks out a version of the
<code class="filename">calc</code> project between revisions 303 and
349, she'll still see the bad change, right?</p><p>Yes, that's true. When we talk about
<span class="quote">“<span class="quote">removing</span>”</span> a change, we're really talking about
removing it from the <code class="literal">HEAD</code> revision. The
original change still exists in the repository's history. For
most situations, this is good enough. Most people are only
interested in tracking the <code class="literal">HEAD</code> of a
project anyway. There are special cases, however, where you
really might want to destroy all evidence of the commit.
(Perhaps somebody accidentally committed a confidential
document.) This isn't so easy, it turns out, because
Subversion was deliberately designed to never lose
information. Revisions are immutable trees that build upon
one another. Removing a revision from history would cause a
domino effect, creating chaos in all subsequent revisions and
possibly invalidating all working copies.<sup>[<a id="idp10075296" href="#ftn.idp10075296" class="footnote">31</a>]</sup></p></div><div class="sect2" title="Resurrecting Deleted Items"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.basicmerging.resurrect"></a>Resurrecting Deleted Items</h3></div></div></div><p>The great thing about version control systems is that
information is never lost. Even when you delete a file or
directory, it may be gone from the <code class="literal">HEAD</code>
revision, but the object still exists in earlier revisions.
One of the most common questions new users ask is, <span class="quote">“<span class="quote">How
do I get my old file or directory back?</span>”</span></p><p>The first step is to define
exactly <span class="emphasis"><em>which</em></span> item you're
trying to resurrect. Here's a useful metaphor: you can think
of every object in the repository as existing in a sort of
two-dimensional coordinate system. The first coordinate is a
particular revision tree, and the second coordinate is a path
within that tree. So every version of your file or directory
is defined by a specific coordinate pair. (Remember the
<span class="quote">“<span class="quote">peg revision</span>”</span>
syntax—foo.c@224—mentioned back in
<a class="xref" href="#svn.advanced.pegrevs" title="Peg and Operative Revisions">the section called “Peg and Operative Revisions”</a>.)</p><p>First, you might need to use <span class="command"><strong>svn log</strong></span> to
discover the exact coordinate pair you wish to resurrect. A
good strategy is to run <strong class="userinput"><code>svn log --verbose</code></strong>
in a directory that used to contain your deleted item. The
<code class="option">--verbose</code> (<code class="option">-v</code>) option shows
a list of all changed items in each revision; all you need to
do is find the revision in which you deleted the file or
directory. You can do this visually, or by using another tool
to examine the log output (via <span class="command"><strong>grep</strong></span>, or
perhaps via an incremental search in an editor).</p><div class="informalexample"><pre class="screen">
$ cd parent-dir
$ svn log -v
…
------------------------------------------------------------------------
r808 | joe | 2003-12-26 14:29:40 -0600 (Fri, 26 Dec 2003) | 3 lines
Changed paths:
D /calc/trunk/real.c
M /calc/trunk/integer.c
Added fast fourier transform functions to integer.c.
Removed real.c because code now in double.c.
…
</pre></div><p>In the example, we're assuming that you're looking for a
deleted file <code class="filename">real.c</code>. By looking through
the logs of a parent directory, you've spotted that this file
was deleted in revision 808. Therefore, the last version of
the file to exist was in the revision right before that.
Conclusion: you want to resurrect the path
<code class="filename">/calc/trunk/real.c</code> from revision
807.</p><p>That was the hard part—the research. Now that you
know what you want to restore, you have two different
choices.</p><p>One option is to use <span class="command"><strong>svn merge</strong></span> to apply
revision 808 <span class="quote">“<span class="quote">in reverse.</span>”</span> (We already
discussed how to undo changes in
<a class="xref" href="#svn.branchmerge.basicmerging.undo" title="Undoing Changes">the section called “Undoing Changes”</a>.) This
would have the effect of re-adding <code class="filename">real.c</code>
as a local modification. The file would be scheduled for
addition, and after a commit, the file would again exist
in <code class="literal">HEAD</code>.</p><p>In this particular example, however, this is probably not
the best strategy. Reverse-applying revision 808 would not
only schedule <code class="filename">real.c</code> for addition, but
the log message indicates that it would also undo certain
changes to <code class="filename">integer.c</code>, which you don't
want. Certainly, you could reverse-merge revision 808 and
then <span class="command"><strong>svn revert</strong></span> the local modifications to
<code class="filename">integer.c</code>, but this technique doesn't
scale well. What if 90 files were changed in revision
808?</p><p>A second, more targeted strategy is not to use
<span class="command"><strong>svn merge</strong></span> at all, but rather to use the
<span class="command"><strong>svn copy</strong></span> command. Simply copy the exact
revision and path <span class="quote">“<span class="quote">coordinate pair</span>”</span> from the
repository to your working copy:</p><div class="informalexample"><pre class="screen">
$ svn copy ^/calc/trunk/real.c@807 ./real.c
$ svn status
A + real.c
$ svn commit -m "Resurrected real.c from revision 807, /calc/trunk/real.c."
Adding real.c
Transmitting file data .
Committed revision 1390.
</pre></div><p>The plus sign in the status output indicates that the item
isn't merely scheduled for addition, but scheduled for
addition <span class="quote">“<span class="quote">with history.</span>”</span> Subversion remembers
where it was copied from. In the future, running <span class="command"><strong>svn
log</strong></span> on this file will traverse back through the
file's resurrection and through all the history it had prior
to revision 807. In other words, this new
<code class="filename">real.c</code> isn't really new; it's a direct
descendant of the original, deleted file. This is usually
considered a good and useful thing. If, however, you wanted
to resurrect the file <span class="emphasis"><em>without</em></span>
maintaining a historical link to the old file, this technique
works just as well:</p><div class="informalexample"><pre class="screen">
$ svn cat ^/calc/trunk/real.c@807 > ./real.c
$ svn add real.c
A real.c
$ svn commit -m "Re-created real.c from revision 807."
Adding real.c
Transmitting file data .
Committed revision 1390.
</pre></div><p>Although our example shows us resurrecting a file, note
that these same techniques work just as well for resurrecting
deleted directories. Also note that a resurrection doesn't
have to happen in your working copy—it can happen
entirely in the repository:</p><div class="informalexample"><pre class="screen">
$ svn copy ^/calc/trunk/real.c@807 ^/calc/trunk/ \
-m "Resurrect real.c from revision 807."
Committed revision 1390.
$ svn update
Updating '.':
A real.c
Updated to revision 1390.
</pre></div></div></div><div class="sect1" title="Advanced Merging"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.branchmerge.advanced"></a>Advanced Merging</h2></div></div></div><p>Here ends the automated magic. Sooner or later, once you
get the hang of branching and merging, you're going to have to
ask Subversion to merge <span class="emphasis"><em>specific</em></span> changes
from one place to another. To do this, you're going to
have to start passing more complicated arguments to <span class="command"><strong>svn
merge</strong></span>. The next section describes the fully expanded
syntax of the command and discusses a number of common
scenarios that require it.</p><div class="sect2" title="Cherrypicking"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.cherrypicking"></a>Cherrypicking</h3></div></div></div><p>Just as the term <span class="quote">“<span class="quote">changeset</span>”</span> is often used in
version control systems, so is the term
<em class="firstterm">cherrypicking</em>. This word refers to
the act of choosing <span class="emphasis"><em>one</em></span> specific
changeset from a branch and replicating it to another.
Cherrypicking may also refer to the act of duplicating a
particular set of (not necessarily contiguous!) changesets
from one branch to another. This is in contrast to more
typical merging scenarios, where the <span class="quote">“<span class="quote">next</span>”</span>
contiguous range of revisions is duplicated
automatically.</p><p>Why would people want to replicate just a single change?
It comes up more often than you'd think. For example, let's
go back in time and imagine that you haven't yet merged your
private feature branch back to the trunk. At the
water cooler, you get word that Sally made an interesting
change to <code class="filename">integer.c</code> on the trunk.
Looking over the history of commits to the trunk, you see that
in revision 355 she fixed a critical bug that directly
impacts the feature you're working on. You might not be ready
to merge all the trunk changes to your branch just yet, but
you certainly need that particular bug fix in order to continue
your work.</p><div class="informalexample"><pre class="screen">
$ svn diff -c 355 ^/calc/trunk
Index: integer.c
===================================================================
--- integer.c (revision 354)
+++ integer.c (revision 355)
@@ -147,7 +147,7 @@
case 6: sprintf(info->operating_system, "HPFS (OS/2 or NT)"); break;
case 7: sprintf(info->operating_system, "Macintosh"); break;
case 8: sprintf(info->operating_system, "Z-System"); break;
- case 9: sprintf(info->operating_system, "CP/MM");
+ case 9: sprintf(info->operating_system, "CP/M"); break;
case 10: sprintf(info->operating_system, "TOPS-20"); break;
case 11: sprintf(info->operating_system, "NTFS (Windows NT)"); break;
case 12: sprintf(info->operating_system, "QDOS"); break;
</pre></div><p>Just as you used <span class="command"><strong>svn diff</strong></span> in the prior
example to examine revision 355, you can pass the same option
to <span class="command"><strong>svn merge</strong></span>:</p><div class="informalexample"><pre class="screen">
$ svn merge -c 355 ^/calc/trunk
--- Merging r355 into '.':
U integer.c
--- Recording mergeinfo for merge of r355 into '.':
U .
$ svn status
M integer.c
</pre></div><p>You can now go through the usual testing procedures before
committing this change to your branch. After the commit,
Subversion marks r355 as having been merged to the branch so
that future <span class="quote">“<span class="quote">magic</span>”</span> merges that synchronize your
branch with the trunk know to skip over r355. (Merging the
same change to the same branch almost always results in a
conflict!)</p><div class="informalexample"><pre class="screen">
$ cd my-calc-branch
$ svn propget svn:mergeinfo .
/trunk:341-349,355
# Notice that r355 isn't listed as "eligible" to merge, because
# it's already been merged.
$ svn mergeinfo ^/calc/trunk --show-revs eligible
r350
r351
r352
r353
r354
r356
r357
r358
r359
r360
$ svn merge ^/calc/trunk
--- Merging r350 through r354 into '.':
U .
U integer.c
U Makefile
--- Merging r356 through r360 into '.':
U .
U integer.c
U button.c
--- Recording mergeinfo for merge of r350 through r360 into '.':
U .
</pre></div><p>This use case of replicating
(or <em class="firstterm">backporting</em>) bug fixes from one
branch to another is perhaps the most popular reason for
cherrypicking changes; it comes up all the time, for example,
when a team is maintaining a <span class="quote">“<span class="quote">release branch</span>”</span> of
software. (We discuss this pattern in
<a class="xref" href="#svn.branchmerge.commonpatterns.release" title="Release Branches">the section called “Release Branches”</a>.)</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>Did you notice how, in the last example, the merge
invocation merged two distinct ranges?
The <span class="command"><strong>svn merge</strong></span> command applied
two independent patches to your working copy to skip over
changeset 355, which your branch already contained. There's
nothing inherently wrong with this, except that it has the
potential to make conflict resolution trickier. If the
first range of changes creates conflicts,
you <span class="emphasis"><em>must</em></span> resolve them interactively for
the merge process to continue and apply the second range of
changes. If you postpone a conflict from the first wave of
changes, the whole merge command will bail out with an error
message.<sup>[<a id="idp10132288" href="#ftn.idp10132288" class="footnote">32</a>]</sup></p></td></tr></table></div><p>A word of warning: while <span class="command"><strong>svn diff</strong></span> and
<span class="command"><strong>svn merge</strong></span> are very similar in concept, they
do have different syntax in many cases. Be sure to read about
them in <a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a> for details, or ask
<span class="command"><strong>svn help</strong></span>. For example, <span class="command"><strong>svn
merge</strong></span> requires a working copy path as a target, that is,
a place where it should apply the generated patch. If the
target isn't specified, it assumes you are trying to perform
one of the following common operations:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>You want to merge directory changes into your current
working directory.</p></li><li class="listitem"><p>You want to merge the changes in a specific file into
a file by the same name that exists in your current working
directory.</p></li></ul></div><p>If you are merging a directory and haven't specified a
target path, <span class="command"><strong>svn merge</strong></span> assumes the first
case and tries to apply the changes into your current
directory. If you are merging a file, and that file (or a
file by the same name) exists in your current working
directory,
<span class="command"><strong>svn merge</strong></span> assumes the second case and tries
to apply the changes to a local file with the same name.</p></div><div class="sect2" title="Merge Syntax: Full Disclosure"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.advanced.advancedsyntax"></a>Merge Syntax: Full Disclosure</h3></div></div></div><p>You've now seen some examples of the <span class="command"><strong>svn
merge</strong></span> command, and you're about to see several more.
If you're feeling confused about exactly how merging works,
you're not alone. Many users (especially those new to version
control) are initially perplexed about the proper syntax of
the command and about how and when the feature should be
used. But fear not, this command is actually much simpler
than you think! There's a very easy technique for
understanding exactly how <span class="command"><strong>svn merge</strong></span>
behaves.</p><p>The main source of confusion is the
<span class="emphasis"><em>name</em></span> of the command. The term
<span class="quote">“<span class="quote">merge</span>”</span> somehow denotes that branches are
combined together, or that some sort of mysterious
blending of data is going on. That's not the case. A better
name for the command might have been <span class="command"><strong>svn
diff-and-apply</strong></span>, because that's all that happens:
two repository trees are compared, and the differences are
applied to a working copy.</p><p>If you're using <span class="command"><strong>svn merge</strong></span> to do basic
copying of changes between branches, it will generally do the
right thing automatically. For example, a command such as the
following:</p><div class="informalexample"><pre class="screen">
$ svn merge ^/calc/branches/some-branch
</pre></div><p>will attempt to duplicate any changes made
on <code class="filename">some-branch</code> into your current working
directory, which is presumably a working copy that shares some
historical connection to the branch. The command is smart
enough to only duplicate changes that your working copy
doesn't yet have. If you repeat this command once a week, it
will only duplicate the <span class="quote">“<span class="quote">newest</span>”</span> branch changes
that happened since you last merged.</p><p>If you choose to use the <span class="command"><strong>svn merge</strong></span>
command in all its full glory by giving it specific revision
ranges to duplicate, the command takes three main
arguments:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p>An initial repository tree (often called the
<em class="firstterm">left side</em> of the comparison)</p></li><li class="listitem"><p>A final repository tree (often called the
<em class="firstterm">right side</em> of the
comparison)</p></li><li class="listitem"><p>A working copy to accept the differences as local
changes (often called the <em class="firstterm">target</em> of
the merge)</p></li></ol></div><p>Once these three arguments are specified, then the two trees
are compared and the differences applied to the
target working copy as local modifications. When the command
is done, the results are no different than if you had
hand-edited the files or run various <span class="command"><strong>svn
add</strong></span> or <span class="command"><strong>svn delete</strong></span> commands
yourself. If you like the results, you can commit them. If
you don't like the results, you can simply <span class="command"><strong>svn
revert</strong></span> all of the changes.</p><p>The syntax of <span class="command"><strong>svn merge</strong></span> allows you to
specify the three necessary arguments rather flexibly. Here
are some examples:</p><div class="informalexample"><pre class="screen">
$ svn merge http://svn.example.com/repos/branch1@150 \
http://svn.example.com/repos/branch2@212 \
my-working-copy
$ svn merge -r 100:200 http://svn.example.com/repos/trunk my-working-copy
$ svn merge -r 100:200 http://svn.example.com/repos/trunk
</pre></div><p>The first syntax lays out all three arguments explicitly,
naming each tree in the form <span class="emphasis"><em>URL@REV</em></span> and
naming the working copy target. The second syntax is used
as a shorthand for situations when you're comparing two
different revisions of the same URL. The last syntax shows
how the working copy argument is optional; if omitted, it
defaults to the current directory.</p><p>While the first example shows the <span class="quote">“<span class="quote">full</span>”</span>
syntax of <span class="command"><strong>svn merge</strong></span>, use it
very carefully; it can result in merges which do not record
any <code class="literal">svn:mergeinfo</code> metadata at all. The
next section talks a bit more about this.</p></div><div class="sect2" title="Merges Without Mergeinfo"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.nomergedata"></a>Merges Without Mergeinfo</h3></div></div></div><p>Subversion tries to generate merge metadata whenever it
can, to make future invocations of <span class="command"><strong>svn
merge</strong></span> smarter. There are still situations, however,
where <code class="literal">svn:mergeinfo</code> data is not created or
changed. Remember to be a bit wary of these scenarios:</p><div class="variablelist"><dl><dt><span class="term">Merging unrelated sources</span></dt><dd><p>If you ask <span class="command"><strong>svn merge</strong></span> to compare
two URLs that aren't related to each other, a patch is
still generated and applied to your working copy, but
no merging metadata is created. There's no common
history between the two sources, and
future <span class="quote">“<span class="quote">smart</span>”</span> merges depend on that common
history.</p></dd><dt><span class="term">Merging from foreign repositories</span></dt><dd><p>While it's possible to run a command such
as <strong class="userinput"><code>svn merge -r 100:200
<em class="replaceable"><code>http://svn.foreignproject.com/repos/trunk</code></em></code></strong>,
the resultant patch also lacks any historical merge
metadata. At the time of this writing, Subversion has no
way of representing different repository URLs within
the <code class="literal">svn:mergeinfo</code> property.</p></dd><dt><span class="term">Using <code class="option">--ignore-ancestry</code></span></dt><dd><p>If this option is passed to <span class="command"><strong>svn
merge</strong></span>, it causes the merging logic to
mindlessly generate differences the same way
that <span class="command"><strong>svn diff</strong></span> does, ignoring any
historical relationships. We discuss this later in this
chapter in
<a class="xref" href="#svn.branchmerge.advanced.ancestry" title="Noticing or Ignoring Ancestry">the section called “Noticing or Ignoring Ancestry”</a>.</p></dd><dt><span class="term">Applying reverse merges from a target's natural history</span></dt><dd><p>Earlier in this chapter
(<a class="xref" href="#svn.branchmerge.basicmerging.undo" title="Undoing Changes">the section called “Undoing Changes”</a>) we
discussed how to use <span class="command"><strong>svn merge</strong></span> to
apply a <span class="quote">“<span class="quote">reverse patch</span>”</span> as a way of rolling
back changes. If this technique is used to undo a
change to an object's personal history (e.g., commit r5
to the trunk, then immediately roll back r5
using <strong class="userinput"><code>svn merge . -c -5</code></strong>), this
sort of merge doesn't affect the recorded
mergeinfo.<sup>[<a id="idp10187360" href="#ftn.idp10187360" class="footnote">33</a>]</sup></p></dd></dl></div><div class="sidebar" title="Natural History and Implicit Mergeinfo"><a id="svn.branchmerge.nomergedata.impicit.mergeinfo"></a><div class="titlepage"><div><div><p class="title"><strong>Natural History and Implicit Mergeinfo</strong></p></div></div></div><p>As we mentioned earlier when discussing
<a class="xref" href="#svn.branchmerge.basicmerging.mergeinfo.inheritance" title="Mergeinfo Inheritance">Mergeinfo Inheritance</a>,
a path that has the
<code class="literal">svn:mergeinfo</code> property set on it is said to
have <span class="quote">“<span class="quote">explicit</span>”</span> mergeinfo. Yes, this implies a
path can have <span class="quote">“<span class="quote">implicit</span>”</span> mergeinfo, too! Implicit
mergeinfo, or <em class="firstterm">natural history</em>, is
simply a path's own history (see
<a class="xref" href="#svn.tour.history" title="Examining History">the section called “Examining History”</a>) interpreted as mergeinfo.
While implicit mergeinfo is largely
an implementation detail, it can be a useful abstraction for
understanding merge tracking behavior.</p><p>Let's say you created <code class="filename">^/trunk</code> in
revision 100 and then later, in revision 201,
created <code class="filename">^/branches/feature-branch</code> as
a copy of <code class="filename">^/trunk@200</code>. The natural
history of <code class="filename">^/branches/feature-branch</code>
contains all the repository paths and revision ranges
through which the history of the new branch has ever
passed:</p><div class="informalexample"><div class="literallayout"><p><br></br>
/trunk:100-200<br></br>
/branches/feature-branch:201<br></br>
</p></div></div><p>With each new revision added to the repository, the
natural history—and thus, implicit mergeinfo—of
the branch continues to expand to include those revisions
until the day the branch is deleted. Here's what the
implicit mergeinfo of our branch would look like when
the <code class="literal">HEAD</code> revision of the repository had
grown to 234:</p><div class="informalexample"><div class="literallayout"><p><br></br>
/trunk:100-200<br></br>
/branches/feature-branch:201-234<br></br>
</p></div></div><p>Implicit mergeinfo does not actually show up in the
<code class="literal">svn:mergeinfo</code> property, but Subversion
acts as if it does. This is why if you check out
<code class="filename">^/branches/feature-branch</code> and then
run <strong class="userinput"><code>svn merge ^/trunk -c 58</code></strong> in the
resulting working copy, nothing happens. Subversion knows
that the changes committed to <code class="filename">^/trunk</code>
in revision 58 are already present in the target's natural
history, so there's no need to try to merge them again.
After all, avoiding repeated merges of
changes <span class="emphasis"><em>is</em></span> the primary goal of
Subversion's merge tracking feature!</p></div></div><div class="sect2" title="More on Merge Conflicts"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.advanced.mergeconflicts"></a>More on Merge Conflicts</h3></div></div></div><p>Just like the <span class="command"><strong>svn update</strong></span> command,
<span class="command"><strong>svn merge</strong></span> applies changes to your working
copy. And therefore it's also capable of creating
conflicts. The conflicts produced by <span class="command"><strong>svn
merge</strong></span>, however, are sometimes different, and this
section explains those differences.</p><p>To begin with, assume that your working copy has no
local edits. When you <span class="command"><strong>svn update</strong></span> to a
particular revision, the changes sent by the server
always apply <span class="quote">“<span class="quote">cleanly</span>”</span> to your working copy.
The server produces the delta by comparing two trees: a
virtual snapshot of your working copy, and the revision tree
you're interested in. Because the left hand side of the
comparison is exactly equal to what you already have, the
delta is guaranteed to correctly convert your working copy
into the right hand tree.</p><p>But <span class="command"><strong>svn merge</strong></span> has no such guarantees
and can be much more chaotic: the advanced user can ask the
server to compare <span class="emphasis"><em>any</em></span> two trees at all,
even ones that are unrelated to the working copy! This means
there's large potential for human error. Users will sometimes
compare the wrong two trees, creating a delta that doesn't
apply cleanly. The <span class="command"><strong>svn merge</strong></span> subcommand does
its best to apply as much of the delta as possible, but some
parts may be impossible. A common sign that you merged the wrong
delta is unexpected tree conflicts:</p><div class="informalexample"><pre class="screen">
$ svn merge -r 1288:1351 http://svn.example.com/myrepos/branch
--- Merging r1289 through r1351 into '.':
C bar.c
C foo.c
C docs
--- Recording mergeinfo for merge of r1289 through r1351 into '.':
U .
Summary of conflicts:
Tree conflicts: 3
$ svn st
! C bar.c
> local missing, incoming edit upon merge
! C foo.c
> local missing, incoming edit upon merge
! C docs
> local delete, incoming edit upon merge
</pre></div><p>In the previous example, it might be the case that
<code class="filename">bar.c</code>, <code class="filename">foo.c</code>, and
<code class="filename">docs</code> all exist in both snapshots of the
branch being compared. The resultant delta wants to change
the contents of the corresponding paths in your working copy,
but those paths don't exist in the working copy. Whatever the
case, the preponderance of tree conflicts most likely means that
the user compared the wrong two trees; it's a classic
sign of user error. When this happens, it's easy to
recursively revert all the changes created by the merge
(<strong class="userinput"><code>svn revert . --recursive</code></strong>), delete any
unversioned files or directories left behind after the
revert, and rerun <span class="command"><strong>svn merge</strong></span> with the
correct arguments.</p><p>Also keep in mind that a merge into a working copy with no
local edits can still produce text conflicts.</p><div class="informalexample"><pre class="screen">
$ svn merge -c 1701 http://svn.example.com/myrepos/branchX --accept postpone
--- Merging r1701 into '.':
C glub.c
C sputter.c
--- Recording mergeinfo for merge of r1701 into '.':
U .
Summary of conflicts:
Text conflicts: 2
C:\SVN\src-branch-1.7.x>svn st
M .
? glub.c.merge-left.r1700
? glub.c.merge-right.r1701
C glub.c
? glub.c.working
? sputter.c.merge-left.r1700
? sputter.c.merge-right.r1701
C sputter.c
? sputter.c.working
Summary of conflicts:
Text conflicts: 2
</pre></div><p>How can a conflict possibly happen? Again, because the user
can request <span class="command"><strong>svn merge</strong></span> to define and apply any
old delta to the working copy, that delta may contain textual
changes that don't cleanly apply to a working file, even if
the file has no local modifications.</p><p>Another small difference between <span class="command"><strong>svn
update</strong></span> and <span class="command"><strong>svn merge</strong></span> is the names
of the full-text files created when a conflict happens. In
<a class="xref" href="#svn.tour.cycle.resolve" title="Resolve Any Conflicts">the section called “Resolve Any Conflicts”</a>, we saw that an
update produces files named
<code class="filename">filename.mine</code>,
<code class="filename">filename.rOLDREV</code>, and
<code class="filename">filename.rNEWREV</code>. When <span class="command"><strong>svn
merge</strong></span> produces a conflict, though, it creates three
files named <code class="filename">filename.working</code>,
<code class="filename">filename.merge-left.rOLDREV</code>, and
<code class="filename">filename.merge-right.rNEWREV</code>. In this case,
the terms <span class="quote">“<span class="quote">merge-left</span>”</span> and <span class="quote">“<span class="quote">merge-right</span>”</span>
are describing which side of the double-tree comparison the file
came from, <span class="quote">“<span class="quote">rOLDREV</span>”</span> describes the revision of the
left side, and <span class="quote">“<span class="quote">rNEWREV</span>”</span> the revision of the right
side. In any case, these differing names help you distinguish
between conflicts that happened as a result of an update and
ones that happened as a result of a merge.</p></div><div class="sect2" title="Blocking Changes"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.advanced.blockchanges"></a>Blocking Changes</h3></div></div></div><p>Sometimes there's a particular changeset that you don't
want automatically merged. For example, perhaps your
team's policy is to do new development work on
<code class="filename">/trunk</code>, but is more conservative about
backporting changes to a stable branch you use for releasing
to the public. On one extreme, you can manually cherrypick
single changesets from the trunk to the branch—just the
changes that are stable enough to pass muster. Maybe things
aren't quite that strict, though; perhaps most of the time
you just let <span class="command"><strong>svn merge</strong></span>
automatically merge most changes from trunk to branch. In
this case, you want a way to mask a few specific changes
out, that is, prevent them from ever being automatically
merged.</p><p>Through Subversion 1.7, the only way to block a changeset
is to make the system believe that the change has
<span class="emphasis"><em>already</em></span> been merged. To do this, invoke
the merge subcommand with the <code class="option">--record-only</code>
option:</p><div class="informalexample"><pre class="screen">
$ cd my-calc-branch
$ svn propget svn:mergeinfo .
/trunk:1680-3305
# Let's make the metadata list r3328 as already merged.
$ svn merge -c 3328 --record-only ^/calc/trunk
--- Recording mergeinfo for merge of r3328 into '.':
U .
$ svn status
M .
$ svn propget svn:mergeinfo .
/trunk:1680-3305,3328
$ svn commit -m "Block r3328 from being merged to the branch."
…
</pre></div><p>Beginning with Subversion 1.7, <code class="option">--record-only</code>
merges are transitive. This means that, in addition to recording
mergeinfo describing the blocked revision(s), any
<code class="literal">svn:mergeinfo</code> property differences in the
merge source are also applied. For example, let's say we want to
block the 'frazzle' feature from ever being merged from
<code class="filename">^/trunk</code> to our
<code class="filename">^/branches/proj-X</code> branch. We know that all
the frazzle work was done on its own branch, which was
reintegrated to <code class="filename">trunk</code> in revision 1055:</p><div class="informalexample"><pre class="screen">
$ svn log -v ^/trunk -r 1055
------------------------------------------------------------------------
r1055 | francesca | 2011-09-22 07:40:06 -0400 (Thu, 22 Sep 2011) | 3 lines
Changed paths:
M /trunk
M /trunk/src/frazzle.c
Reintegrate the frazzle-feature-branch to trunk.
</pre></div><p>Because revision 1055 was a reintegrate merge we know
that mergeinfo was recorded describing the merge:</p><div class="informalexample"><pre class="screen">
$ svn diff ^/trunk -c 1055 --depth empty
Index: .
===================================================================
--- . (revision 1054)
+++ . (revision 1055)
Property changes on: .
___________________________________________________________________
Modified: svn:mergeinfo
Merged /branches/frazzle-feature-branch:r997-1003
</pre></div><p>Now simply blocking merges of revision 1055 from
<code class="filename">^/trunk</code> isn't foolproof since someone could
merge r996:1003 directly from
<code class="filename">^/branches/frazzle-feature-branch</code>.
Fortunately the transitive nature
of <code class="option">--record-only</code> merges in Subversion 1.7
prevents this; the <code class="option">--record-only</code> merge
applies the <code class="literal">svn:mergeinfo</code> diff from
revision 1055, thus blocking merges directly from the frazzle
branch <span class="emphasis"><em>and</em></span> as it has always done prior to
Subversion 1.7, it blocks merges of revision 1055 directly
from <code class="filename">^/trunk</code>:</p><div class="informalexample"><pre class="screen">
$ cd branches/proj-X
$ svn merge ^/trunk . -c 1055 --record-only
--- Merging r1055 into '.':
G .
--- Recording mergeinfo for merge of r1055 into '.':
G .
$ svn diff --depth empty .
Index: .
===================================================================
--- . (revision 1070)
+++ . (working copy)
Property changes on: .
___________________________________________________________________
Modified: svn:mergeinfo
Merged /trunk:r1055
Merged /branches/frazzle-feature-branch:r997-1003
</pre></div><p>Blocking changes with <code class="option">--record-only</code>
works, but it's also a little bit
dangerous. The main problem is that we're not clearly
differentiating between the ideas of <span class="quote">“<span class="quote">I already have
this change</span>”</span> and <span class="quote">“<span class="quote">I don't have this change, but
don't currently want it.</span>”</span> We're effectively lying to
the system, making it think that the change was previously
merged. This puts the responsibility on you—the
user—to remember that the change wasn't actually merged,
it just wasn't wanted. There's no way to ask Subversion for a
list of <span class="quote">“<span class="quote">blocked changelists.</span>”</span> If you want to
track them (so that you can unblock them someday) you'll need
to record them in a text file somewhere, or perhaps in an
invented property.</p></div><div class="sect2" title="Keeping a Reintegrated Branch Alive"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.advanced.reintegratetwice"></a>Keeping a Reintegrated Branch Alive</h3></div></div></div><p>There is an alternative to destroying and re-creating a
branch after reintegration. To understand why it works you need
to understand why the branch is initially unfit for further use
after it has been reintegrated.</p><p>Let's assume you created your branch in revision
<span class="emphasis"><em>A</em></span>. While working on your branch, you created
one or more revisions which made changes to the branch.
Before reintegrating your branch back to trunk, you made a final
merge from trunk to your branch, and committed the result of
this merge as revision <span class="emphasis"><em>B</em></span>.</p><p>When reintegrating your branch into the trunk, you create
a new revision <span class="emphasis"><em>X</em></span> which changes the trunk.
The changes made to trunk in this revision <span class="emphasis"><em>X</em></span>
are semantically equivalent to the changes you made to your branch
between revisions <span class="emphasis"><em>A</em></span> and <span class="emphasis"><em>B</em></span>.</p><p>If you now try to merge outstanding changes from trunk to your
branch, Subversion will consider changes made in revision
<span class="emphasis"><em>X</em></span> as eligible for merging into the
branch. However, since your branch already contains all the
changes made in revision <span class="emphasis"><em>X</em></span>, merging these
changes can result in spurious conflicts! These conflicts are
often tree conflicts, especially if renames were made on the
branch or the trunk while the branch was in development.</p><p>So what can be done about this? We need to make sure that
Subversion does not try to merge revision <span class="emphasis"><em>X</em></span>
into the branch. This is done using the
<code class="option">--record-only</code> merge option, which was introduced
in <a class="xref" href="#svn.branchmerge.advanced.blockchanges" title="Blocking Changes">the section called “Blocking Changes”</a>.</p><p>To carry out the record-only merge, get a working copy of
the branch which was just reintegrated in revision
<span class="emphasis"><em>X</em></span>, and merge just revision <span class="emphasis"><em>X</em></span>
from trunk into your branch, making sure to use the
<code class="option">--record-only</code> option.</p><p>This merge uses the cherry-picking merge syntax, which was
introduced in <a class="xref" href="#svn.branchmerge.cherrypicking" title="Cherrypicking">the section called “Cherrypicking”</a>.
Continuing with the running example from
<a class="xref" href="#svn.branchemerge.basicmerging.reintegrate" title="Reintegrating a Branch">the section called “Reintegrating a Branch”</a>,
where revision <span class="emphasis"><em>X</em></span> was revision 391:</p><div class="informalexample"><pre class="screen">
$ cd my-calc-branch
$ svn update
Updating '.':
Updated to revision 393.
$ svn merge --record-only -c 391 ^/calc/trunk
--- Recording mergeinfo for merge of r391 into '.':
U .
$ svn commit -m "Block revision 391 from being merged into my-calc-branch."
Sending .
Committed revision 394.
</pre></div><p>Now your branch is ready to soak up changes from the trunk
again. After another sync of your branch to the trunk, you can
even reintegrate the branch a second time.
If necessary, you can do another record-only merge to keep the
branch alive. Rinse and repeat.</p><p>It should now also be apparent why deleting the branch and
re-creating it has the same effect as doing the above record-only
merge. Because revision <span class="emphasis"><em>X</em></span> is part of the natural
history (see the sidebar
<a class="xref" href="#svn.branchmerge.nomergedata.impicit.mergeinfo" title="Natural History and Implicit Mergeinfo">Natural History and Implicit Mergeinfo</a>)
of the newly created branch, Subversion will never attempt to
merge revision <span class="emphasis"><em>X</em></span> into the branch, avoiding
spurious conflicts.</p></div><div class="sect2" title="Merge-Sensitive Logs and Annotations"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.advanced.logblame"></a>Merge-Sensitive Logs and Annotations</h3></div></div></div><p>One of the main features of any version control system is
to keep track of who changed what, and when they did it.
The <span class="command"><strong>svn log</strong></span> and <span class="command"><strong>svn
blame</strong></span> subcommands are just the tools for this: when
invoked on individual files, they show not only the history of
changesets that affected the file, but also exactly which user
wrote which line of code, and when she did it.</p><p>When changes start getting replicated between branches,
however, things start to get complicated. For example, if you
were to ask <span class="command"><strong>svn log</strong></span> about the history of
your feature branch, it would show exactly every revision that ever
affected the branch:</p><div class="informalexample"><pre class="screen">
$ cd my-calc-branch
$ svn log -q
------------------------------------------------------------------------
r390 | user | 2002-11-22 11:01:57 -0600 (Fri, 22 Nov 2002)
------------------------------------------------------------------------
r388 | user | 2002-11-21 05:20:00 -0600 (Thu, 21 Nov 2002)
------------------------------------------------------------------------
r381 | user | 2002-11-20 15:07:06 -0600 (Wed, 20 Nov 2002)
------------------------------------------------------------------------
r359 | user | 2002-11-19 19:19:20 -0600 (Tue, 19 Nov 2002)
------------------------------------------------------------------------
r357 | user | 2002-11-15 14:29:52 -0600 (Fri, 15 Nov 2002)
------------------------------------------------------------------------
r343 | user | 2002-11-07 13:50:10 -0600 (Thu, 07 Nov 2002)
------------------------------------------------------------------------
r341 | user | 2002-11-03 07:17:16 -0600 (Sun, 03 Nov 2002)
------------------------------------------------------------------------
r303 | sally | 2002-10-29 21:14:35 -0600 (Tue, 29 Oct 2002)
------------------------------------------------------------------------
r98 | sally | 2002-02-22 15:35:29 -0600 (Fri, 22 Feb 2002)
------------------------------------------------------------------------
</pre></div><p>But is this really an accurate picture of all the changes
that happened on the branch? What's left out here is
the fact that revisions 390, 381, and 357 were actually the
results of merging changes from the trunk. If you look at one
of these logs in detail, the multiple trunk changesets that
comprised the branch change are nowhere to be seen:</p><div class="informalexample"><pre class="screen">
$ svn log -v -r 390
------------------------------------------------------------------------
r390 | user | 2002-11-22 11:01:57 -0600 (Fri, 22 Nov 2002) | 1 line
Changed paths:
M /branches/my-calc-branch/button.c
M /branches/my-calc-branch/README
Final merge of trunk changes to my-calc-branch.
</pre></div><p>We happen to know that this merge to the branch was
nothing but a merge of trunk changes. How can we see those
trunk changes as well? The answer is to use the
<code class="option">--use-merge-history</code> (<code class="option">-g</code>)
option. This option expands those <span class="quote">“<span class="quote">child</span>”</span>
changes that were part of the merge.</p><div class="informalexample"><pre class="screen">
$ svn log -v -r 390 -g
------------------------------------------------------------------------
r390 | user | 2002-11-22 11:01:57 -0600 (Fri, 22 Nov 2002) | 1 line
Changed paths:
M /branches/my-calc-branch/button.c
M /branches/my-calc-branch/README
Final merge of trunk changes to my-calc-branch.
------------------------------------------------------------------------
r383 | sally | 2002-11-21 03:19:00 -0600 (Thu, 21 Nov 2002) | 2 lines
Changed paths:
M /branches/my-calc-branch/button.c
Merged via: r390
Fix inverse graphic error on button.
------------------------------------------------------------------------
r382 | sally | 2002-11-20 16:57:06 -0600 (Wed, 20 Nov 2002) | 2 lines
Changed paths:
M /branches/my-calc-branch/README
Merged via: r390
Document my last fix in README.
</pre></div><p>By making the log operation use merge history, we see not
just the revision we queried (r390), but also the two revisions
that came along on the ride with it—a couple of changes
made by Sally to the trunk. This is a much more complete
picture of history!</p><p>The <span class="command"><strong>svn blame</strong></span> command also takes the
<code class="option">--use-merge-history</code> (<code class="option">-g</code>)
option. If this option is neglected, somebody looking at
a line-by-line annotation of <code class="filename">button.c</code> may
get the mistaken impression that you were responsible for the
lines that fixed a certain error:</p><div class="informalexample"><pre class="screen">
$ svn blame button.c
…
390 user retval = inverse_func(button, path);
390 user return retval;
390 user }
…
</pre></div><p>And while it's true that you did actually commit those
three lines in revision 390, two of them were actually written
by Sally back in revision 383:</p><div class="informalexample"><pre class="screen">
$ svn blame button.c -g
…
G 383 sally retval = inverse_func(button, path);
G 383 sally return retval;
390 user }
…
</pre></div><p>Now we know who to <span class="emphasis"><em>really</em></span> blame for
those two lines of code!</p></div><div class="sect2" title="Noticing or Ignoring Ancestry"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.advanced.ancestry"></a>Noticing or Ignoring Ancestry</h3></div></div></div><p>
<a id="idp10304080" class="indexterm"></a>
When conversing with a Subversion developer, you might
very likely hear reference to the term
<em class="firstterm">ancestry</em>. This word is used to
describe the relationship between two objects in a
repository: if they're related to each other, one
object is said to be an ancestor of the other.</p><p>For example, suppose you commit revision 100, which
includes a change to a file <code class="filename">foo.c</code>.
Then <code class="filename">foo.c@99</code> is an
<span class="quote">“<span class="quote">ancestor</span>”</span> of <code class="filename">foo.c@100</code>.
On the other hand, suppose you commit the deletion of
<code class="filename">foo.c</code> in revision 101, and then add a
new file by the same name in revision 102. In this case,
<code class="filename">foo.c@99</code> and
<code class="filename">foo.c@102</code> may appear to be related
(they have the same path), but in fact are completely
different objects in the repository. They share no history
or <span class="quote">“<span class="quote">ancestry.</span>”</span></p><p>The reason for bringing this up is to point out an
important difference between <span class="command"><strong>svn diff</strong></span> and
<span class="command"><strong>svn merge</strong></span>. The former command ignores
ancestry, while the latter command is quite sensitive to it.
For example, if you asked <span class="command"><strong>svn diff</strong></span> to
compare revisions 99 and 102 of <code class="filename">foo.c</code>,
you would see line-based diffs; the <span class="command"><strong>diff</strong></span>
command is blindly comparing two paths. But if you asked
<span class="command"><strong>svn merge</strong></span> to compare the same two objects,
it would notice that they're unrelated and first attempt to
delete the old file, then add the new file; the output would
indicate a deletion followed by an add:</p><div class="informalexample"><pre class="screen">
D foo.c
A foo.c
</pre></div><p>Most merges involve comparing trees that are ancestrally
related to one another; therefore, <span class="command"><strong>svn
merge</strong></span> defaults to this behavior. Occasionally,
however, you may want the <span class="command"><strong>merge</strong></span> command to
compare two unrelated trees. For example, you may have
imported two source-code trees representing different vendor
releases of a software project (see
<a class="xref" href="#svn.advanced.vendorbr" title="Vendor Branches">the section called “Vendor Branches”</a>). If you ask
<span class="command"><strong>svn merge</strong></span> to compare the two trees, you'd
see the entire first tree being deleted, followed by an add
of the entire second tree! In these situations, you'll want
<span class="command"><strong>svn merge</strong></span> to do a path-based comparison
only, ignoring any relations between files and directories.
Add the <code class="option">--ignore-ancestry</code> option to your
<span class="command"><strong>merge</strong></span> command, and it will behave just
like <span class="command"><strong>svn diff</strong></span>. (And conversely, the
<code class="option">--notice-ancestry</code> option will cause
<span class="command"><strong>svn diff</strong></span> to behave like the
<span class="command"><strong>svn merge</strong></span> command.)</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>The <code class="option">--ignore-ancestry</code> option also disables
<a class="xref" href="#svn.branchmerge.basicmerging.mergetracking" title="Merge Tracking">Merge Tracking</a>.
This means that <code class="literal">svn:mergeinfo</code> is not considered
when <span class="command"><strong>svn merge</strong></span> is determining what revisions
to merge, nor is <code class="literal">svn:mergeinfo</code> recorded to
describe the merge.</p></td></tr></table></div></div><div class="sect2" title="Merges and Moves"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.advanced.moves"></a>Merges and Moves</h3></div></div></div><p>A common desire is to refactor source code, especially
in Java-based software projects. Files and directories are
shuffled around and renamed, often causing great disruption
to everyone working on the project. Sounds like a perfect
case to use a branch, doesn't it? Just create a branch,
shuffle things around, and then merge the branch back to the
trunk, right?</p><p>Alas, this scenario doesn't work so well right now and
is considered one of Subversion's current weak spots. The
problem is that Subversion's <span class="command"><strong>svn update</strong></span>
command isn't as robust as it should be, particularly when
dealing with copy and move operations.</p><p>When you use <span class="command"><strong>svn copy</strong></span> to duplicate a
file, the repository remembers where the new file came from,
but it fails to transmit that information to the client which
is running <span class="command"><strong>svn update</strong></span> or <span class="command"><strong>svn
merge</strong></span>. Instead of telling the client, <span class="quote">“<span class="quote">Copy
that file you already have to this new location,</span>”</span> it
sends down an entirely new file. This can lead to
problems, especially because the same thing happens with
renamed files. A lesser-known fact about Subversion is that
it lacks <span class="quote">“<span class="quote">true renames</span>”</span>—the <span class="command"><strong>svn
move</strong></span> command is nothing more than an aggregation
of <span class="command"><strong>svn copy</strong></span> and <span class="command"><strong>svn
delete</strong></span>.</p><p>For example, suppose that while working on your private
branch, you rename <code class="filename">integer.c</code>
to <code class="filename">whole.c</code>. Effectively you've created
a new file in your branch that is a copy of the original
file, and deleted the original file. Meanwhile, back
on <code class="filename">trunk</code>, Sally has committed some
improvements to <code class="filename">integer.c</code>. Now you
decide to merge your branch to the trunk:</p><div class="informalexample"><pre class="screen">
$ cd calc/trunk
$ svn merge --reintegrate ^/calc/branches/my-calc-branch
--- Merging differences between repository URLs into '.':
D integer.c
A whole.c
U .
--- Recording mergeinfo for merge between repository URLs into '.':
U .
</pre></div><p>This doesn't look so bad at first glance, but it's also
probably not what you or Sally expected. The merge operation
has deleted the latest version of
the <code class="filename">integer.c</code> file (the one containing
Sally's latest changes), and blindly added your
new <code class="filename">whole.c</code> file—which is a
duplicate of the <span class="emphasis"><em>older</em></span> version
of <code class="filename">integer.c</code>. The net effect is that
merging your <span class="quote">“<span class="quote">rename</span>”</span> to the trunk has removed
Sally's recent changes from the latest revision!</p><p>This isn't true data loss. Sally's changes are still in
the repository's history, but it may not be immediately
obvious that this has happened. The moral of this story is
that until Subversion improves, be very careful about
merging copies and renames from one branch to
another.</p></div><div class="sect2" title="Blocking Merge-Unaware Clients"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.advanced.pre1.5clients"></a>Blocking Merge-Unaware Clients</h3></div></div></div><p>If you've just upgraded your server to Subversion 1.5 or
later, there's a risk that pre-1.5 Subversion
clients can cause problems with
<a class="xref" href="#svn.branchmerge.basicmerging.mergetracking" title="Merge Tracking">Merge Tracking</a>.
This is because pre-1.5 clients don't support this feature;
when one of these older clients performs <span class="command"><strong>svn
merge</strong></span>, it doesn't modify the value of
the <code class="literal">svn:mergeinfo</code> property at all. So the
subsequent commit, despite being the result of a merge,
doesn't tell the repository about the duplicated
changes—that information is lost. Later on,
when <span class="quote">“<span class="quote">merge-aware</span>”</span> clients attempt automatic
merging, they're likely to run into all sorts of conflicts
resulting from repeated merges.</p><p>If you and your team are relying on the merge-tracking
features of Subversion, you may want to configure your
repository to prevent older clients from committing changes.
The easy way to do this is by inspecting
the <span class="quote">“<span class="quote">capabilities</span>”</span> parameter in
the <code class="literal">start-commit</code> hook script. If the
client reports itself as having <code class="literal">mergeinfo</code>
capabilities, the hook script can allow the commit to start.
If the client doesn't report that capability, have the hook
deny the commit.
<a class="xref" href="#svn.branchmerge.advanced.hook-ex1" title="Example 4.1. Merge-tracking gatekeeper start-commit hook script">Example 4.1, “Merge-tracking gatekeeper start-commit hook script”</a> gives an
example of such a hook script:</p><div class="example"><a id="svn.branchmerge.advanced.hook-ex1"></a><p class="title"><strong>Example 4.1. Merge-tracking gatekeeper start-commit hook script</strong></p><div class="example-contents"><pre class="programlisting">
#!/usr/bin/env python
import sys
# The start-commit hook is invoked before a Subversion txn is created
# in the process of doing a commit. Subversion runs this hook
# by invoking a program (script, executable, binary, etc.) named
# 'start-commit' (for which this file is a template)
# with the following ordered arguments:
#
# [1] REPOS-PATH (the path to this repository)
# [2] USER (the authenticated user attempting to commit)
# [3] CAPABILITIES (a colon-separated list of capabilities reported
# by the client; see note below)
capabilities = sys.argv[3].split(':')
if "mergeinfo" not in capabilities:
sys.stderr.write("Commits from merge-tracking-unaware clients are "
"not permitted. Please upgrade to Subversion 1.5 "
"or newer.\n")
sys.exit(1)
sys.exit(0)
</pre></div></div><br class="example-break"></br><p>For more information about hook scripts, see
<a class="xref" href="#svn.reposadmin.create.hooks" title="Implementing Repository Hooks">the section called “Implementing Repository Hooks”</a>.</p></div><div class="sect2" title="The Final Word on Merge Tracking"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.advanced.finalword"></a>The Final Word on Merge Tracking</h3></div></div></div><p>The bottom line is that Subversion's merge-tracking
feature has an extremely complex internal implementation, and
the <code class="literal">svn:mergeinfo</code> property is the only
window the user has into the machinery.</p><p>Sometimes mergeinfo will appear on paths that you didn't
expect to be touched by an operation. Sometimes mergeinfo
won't be generated at all, when you expect it to.
Furthermore, the management of mergeinfo metadata has a whole
set of taxonomies and behaviors around it, such
as <span class="quote">“<span class="quote">explicit</span>”</span> versus <span class="quote">“<span class="quote">implicit</span>”</span>
mergeinfo, <span class="quote">“<span class="quote">operative</span>”</span>
versus <span class="quote">“<span class="quote">inoperative</span>”</span> revisions, specific
mechanisms of mergeinfo <span class="quote">“<span class="quote">elision,</span>”</span> and
even <span class="quote">“<span class="quote">inheritance</span>”</span> from parent to child
directories.</p><p>We've chosen to only briefly cover, if at all, these detailed
topics for a couple of reasons. First, the level of detail is
absolutely overwhelming for a typical user. Second, and more
importantly, the typical user <span class="emphasis"><em>shouldn't</em></span> have
to understand these concepts; they should typically remain in the
background as pesky implementation details. All that said, if you
enjoy this sort of thing, you can get a fantastic overview in a
paper posted at CollabNet's website: <a class="ulink" href="http://www.collab.net/community/subversion/articles/merge-info.html" target="_top">http://www.collab.net/community/subversion/articles/merge-info.html</a>.</p><p>For now, if you want to steer clear of the complexities of
merge tracking, we recommend that you follow these simple best
practices:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>For short-term feature branches, follow the simple
procedure described throughout
<a class="xref" href="#svn.branchmerge.basicmerging" title="Basic Merging">the section called “Basic Merging”</a>.</p></li><li class="listitem"><p>Avoid subtree merges and subtree mergeinfo, perform
merges only on the root of your branches, not on
subdirectories or files (see <a class="xref" href="#svn.branchemerge.basicmerging.stayinsync.subtree" title="Subtree Merges and Subtree Mergeinfo">Subtree Merges and Subtree Mergeinfo</a>)
.</p></li><li class="listitem"><p>Don't ever edit the <code class="literal">svn:mergeinfo</code>
property directly; use <span class="command"><strong>svn
merge</strong></span> with the <code class="option">--record-only</code> option
to effect a desired change to the metadata (as demonstrated in
<a class="xref" href="#svn.branchmerge.advanced.blockchanges" title="Blocking Changes">the section called “Blocking Changes”</a>).</p></li><li class="listitem"><p>Your merge target should be a working copy which
represents the root of a <span class="emphasis"><em>complete</em></span> tree
representing a <span class="emphasis"><em>single</em></span> location in the
repository at a single point in time:
</p><div class="itemizedlist"><ul class="itemizedlist" type="circle"><li class="listitem"><p>Don't use the <code class="option">--allow-mixed-revisions</code>
option to merge into mixed-revision working copies.</p></li><li class="listitem"><p>Don't merge to targets with <span class="quote">“<span class="quote">switched</span>”</span>
subdirectories (as described next in
<a class="xref" href="#svn.branchmerge.switchwc" title="Traversing Branches">the section called “Traversing Branches”</a>).</p></li><li class="listitem"><p>Avoid merges to targets with sparse directories.
Likewise, don't merge to depths other than
<code class="option">--depth=infinity</code></p></li><li class="listitem"><p>Be sure you have read access to all of the merge
source and read/write access to all of the merge
target.</p></li></ul></div><p>
</p></li></ul></div></div></div><div class="sect1" title="Traversing Branches"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.branchmerge.switchwc"></a>Traversing Branches</h2></div></div></div><p>The <span class="command"><strong>svn switch</strong></span> command transforms an
existing working copy to reflect a different branch. While this
command isn't strictly necessary for working with branches, it
provides a nice shortcut. In one of our earlier examples,
after creating your private branch, you checked out a fresh
working copy of the new repository directory. Instead, you can
simply ask Subversion to change your working copy of
<code class="filename">/calc/trunk</code> to mirror the new branch
location:</p><div class="informalexample"><pre class="screen">
$ cd calc
$ svn info | grep URL
URL: http://svn.example.com/repos/calc/trunk
$ svn switch ^/calc/branches/my-calc-branch
U integer.c
U button.c
U Makefile
Updated to revision 341.
$ svn info | grep URL
URL: http://svn.example.com/repos/calc/branches/my-calc-branch
</pre></div><p><span class="quote">“<span class="quote">Switching</span>”</span> a working copy that has no local
modifications to a different branch results in the working copy
looking just as it would if you'd done a fresh checkout of the
directory. It's usually more efficient to
use this command, because often branches differ by only a small
degree. The server sends only the minimal set of changes
necessary to make your working copy reflect the branch
directory.</p><p>The <span class="command"><strong>svn switch</strong></span> command also takes a
<code class="option">--revision</code> (<code class="option">-r</code>) option, so you
need not always move your working copy to the
<code class="literal">HEAD</code> of the branch.</p><p>Of course, most projects are more complicated than our
<code class="filename">calc</code> example, and contain multiple
subdirectories. Subversion users often follow a specific
algorithm when using branches:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p>Copy the project's entire <span class="quote">“<span class="quote">trunk</span>”</span> to a new
branch directory.</p></li><li class="listitem"><p>Switch only <span class="emphasis"><em>part</em></span> of the trunk
working copy to mirror the branch.</p></li></ol></div><p>In other words, if a user knows that the branch work needs
to happen on only a specific subdirectory, she uses
<span class="command"><strong>svn switch</strong></span> to move only that subdirectory to
the branch. (Or sometimes users will switch just a single
working file to the branch!) That way, the user can continue to
receive normal <span class="quote">“<span class="quote">trunk</span>”</span> updates to most of her
working copy, but the switched portions will remain immune
(unless someone commits a change to her branch). This feature
adds a whole new dimension to the concept of a <span class="quote">“<span class="quote">mixed
working copy</span>”</span>—not only can working copies contain a
mixture of working revisions, but they can also contain a
mixture of repository locations as well.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>Typically switched subdirectories share common ancestry with
the location which is switched <span class="quote">“<span class="quote">away</span>”</span> from. However
<span class="command"><strong>svn switch</strong></span> can switch a subdirectory to mirror
a repository location which it shares no common ancestry with.
To do this you need to use the
<code class="option">--ignore-ancestry</code> option.
</p></td></tr></table></div><p>If your working copy contains a number of switched subtrees
from different repository locations, it continues to function as
normal. When you update, you'll receive patches to each subtree
as appropriate. When you commit, your local changes are still
applied as a single, atomic change to the repository.</p><p>Note that while it's okay for your working copy to reflect a
mixture of repository locations, these locations must all be
within the <span class="emphasis"><em>same</em></span> repository. Subversion
repositories aren't yet able to communicate with one another;
that feature is planned for the
future.<sup>[<a id="idp10408640" href="#ftn.idp10408640" class="footnote">34</a>]</sup></p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>Administrators who need to change the URL of a repository
which is accessed via HTTP are encouraged to add to
their <code class="filename">httpd.conf</code> configuration file a
permanent redirect from the old URL location to the new one
(via the <code class="literal">RedirectPermanent</code> directive).
Subversion clients will generally display the new repository
URL in error messages generated when the user attempts to use
working copies which still reflect the old URL location. In
fact, Subversion 1.7 clients will go a step further,
automatically relocating the working copy to the new
URL.</p></td></tr></table></div><div class="sidebar" title="Switches and Updates"><div class="titlepage"><div><div><p class="title"><strong>Switches and Updates</strong></p></div></div></div><p>Have you noticed that the output of <span class="command"><strong>svn
switch</strong></span> and <span class="command"><strong>svn update</strong></span> looks the
same? The switch command is actually a superset of the update
command.</p><p>When you run <span class="command"><strong>svn update</strong></span>, you're asking
the repository to compare two trees. The repository does so,
and then sends a description of the differences back to the
client. The only difference between <span class="command"><strong>svn
switch</strong></span> and <span class="command"><strong>svn update</strong></span> is that the
latter command always compares two identical repository
paths.</p><p>That is, if your working copy is a mirror of
<code class="filename">/calc/trunk</code>, <span class="command"><strong>svn
update</strong></span> will automatically compare your working copy
of <code class="filename">/calc/trunk</code> to
<code class="filename">/calc/trunk</code> in the
<code class="literal">HEAD</code> revision. If you're switching your
working copy to a branch, <span class="command"><strong>svn switch</strong></span>
will compare your working copy of
<code class="filename">/calc/trunk</code> to some
<span class="emphasis"><em>other</em></span> branch directory in the
<code class="literal">HEAD</code> revision.</p><p>In other words, an update moves your working copy through
time. A switch moves your working copy through time
<span class="emphasis"><em>and</em></span> space.</p></div><p>Because <span class="command"><strong>svn switch</strong></span> is essentially a
variant of <span class="command"><strong>svn update</strong></span>, it shares the same
behaviors; any local modifications in your working copy are
preserved when new data arrives from the repository.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>Have you ever found yourself making some complex edits (in
your <code class="filename">/trunk</code> working copy) and suddenly
realized, <span class="quote">“<span class="quote">Hey, these changes ought to be in their own
branch?</span>”</span> There is a great two step technique to do
this:</p><div class="informalexample"><pre class="screen">
$ svn copy http://svn.example.com/repos/calc/trunk \
http://svn.example.com/repos/calc/branches/newbranch \
-m "Create branch 'newbranch'."
Committed revision 353.
$ svn switch ^/calc/branches/newbranch
At revision 353.
</pre></div><p>The <span class="command"><strong>svn switch</strong></span> command, like
<span class="command"><strong>svn update</strong></span>, preserves your local edits. At
this point, your working copy is now a reflection of the newly
created branch, and your next <span class="command"><strong>svn commit</strong></span>
invocation will send your changes there.</p></td></tr></table></div></div><div class="sect1" title="Tags"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.branchmerge.tags"></a>Tags</h2></div></div></div><p>
<a id="idp10438896" class="indexterm"></a>
Another common version control concept is a tag. A tag is
just a <span class="quote">“<span class="quote">snapshot</span>”</span> of a project in time. In
Subversion, this idea already seems to be everywhere. Each
repository revision is exactly that—a snapshot of the
filesystem after each commit.</p><p>However, people often want to give more human-friendly names
to tags, such as <code class="literal">release-1.0</code>. And they want
to make snapshots of smaller subdirectories of the filesystem.
After all, it's not so easy to remember that release 1.0 of a
piece of software is a particular subdirectory of revision
4822.</p><div class="sect2" title="Creating a Simple Tag"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.tags.mksimple"></a>Creating a Simple Tag</h3></div></div></div><p>Once again, <span class="command"><strong>svn copy</strong></span> comes to the
rescue. If you want to create a snapshot of
<code class="filename">/calc/trunk</code> exactly as it looks in the
<code class="literal">HEAD</code> revision, make a copy of it:</p><div class="informalexample"><pre class="screen">
$ svn copy http://svn.example.com/repos/calc/trunk \
http://svn.example.com/repos/calc/tags/release-1.0 \
-m "Tagging the 1.0 release of the 'calc' project."
Committed revision 902.
</pre></div><p>This example assumes that a
<code class="filename">/calc/tags</code> directory already exists. (If
it doesn't, you can create it using <span class="command"><strong>svn
mkdir</strong></span>.) After the copy completes, the new
<code class="filename">release-1.0</code> directory is forever a
snapshot of how the <code class="filename">/trunk</code> directory
looked in the <code class="literal">HEAD</code> revision at the time you
made the copy. Of course, you might want to be more precise
about exactly which revision you copy, in case somebody else
may have committed changes to the project when you weren't
looking. So if you know that revision 901 of
<code class="filename">/calc/trunk</code> is exactly the snapshot you
want, you can specify it by passing <code class="option">-r 901</code> to
the <span class="command"><strong>svn copy</strong></span> command.</p><p>But wait a moment: isn't this tag creation procedure the
same procedure we used to create a branch? Yes, in fact, it
is. In Subversion, there's no difference between a tag and a
branch. Both are just ordinary directories that are created
by copying. Just as with branches, the only reason a copied
directory is a <span class="quote">“<span class="quote">tag</span>”</span> is because
<span class="emphasis"><em>humans</em></span> have decided to treat it that way:
as long as nobody ever commits to the directory, it forever
remains a snapshot. If people start committing to it, it
becomes a branch.</p><p>If you are administering a repository, there are two
approaches you can take to managing tags. The first approach
is <span class="quote">“<span class="quote">hands off</span>”</span>: as a matter of project policy,
decide where your tags will live, and make sure all users know
how to treat the directories they copy. (That is, make sure
they know not to commit to them.) The second approach is more
paranoid: you can use one of the access control scripts
provided with Subversion to prevent anyone from doing anything
but creating new copies in the tags area (see
<a class="xref" href="#svn.serverconfig" title="Chapter 6. Server Configuration">Chapter 6, <em>Server Configuration</em></a>). The paranoid approach,
however, isn't usually necessary. If a user accidentally
commits a change to a tag directory, you can simply undo the
change as discussed in the previous section. This is version
control, after all!</p></div><div class="sect2" title="Creating a Complex Tag"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.tags.mkcomplex"></a>Creating a Complex Tag</h3></div></div></div><p>Sometimes you may want a <span class="quote">“<span class="quote">snapshot</span>”</span> that is
more complicated than a single directory at a single
revision.</p><p>For example, pretend your project is much larger than our
<code class="filename">calc</code> example: suppose it contains a
number of subdirectories and many more files. In the course
of your work, you may decide that you need to create a working
copy that is designed to have specific features and bug fixes.
You can accomplish this by selectively backdating files or
directories to particular revisions (using <span class="command"><strong>svn
update</strong></span> with the <code class="option">-r</code> option
liberally), by switching files and directories to particular
branches (making use of <span class="command"><strong>svn switch</strong></span>), or
even just by making a bunch of local changes. When you're
done, your working copy is a hodgepodge of repository
locations from different revisions. But after testing, you
know it's the precise combination of data you need to
tag.</p><p>Time to make a snapshot. Copying one URL to another won't
work here. In this case, you want to make a snapshot of your
exact working copy arrangement and store it in the repository.
Luckily, <span class="command"><strong>svn copy</strong></span> actually has four
different uses (which you can read about in <a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a>), including the ability to copy a
working copy tree to the repository:</p><div class="informalexample"><pre class="screen">
$ ls
my-working-copy/
$ svn copy my-working-copy \
http://svn.example.com/repos/calc/tags/mytag \
-m "Tag my existing working copy state."
Committed revision 940.
</pre></div><p>Now there is a new directory in the repository,
<code class="filename">/calc/tags/mytag</code>, which is an exact
snapshot of your working copy—mixed revisions, URLs,
local changes, and all.</p><p>Other users have found interesting uses for this feature.
Sometimes there are situations where you have a bunch of local
changes made to your working copy, and you'd like a
collaborator to see them. Instead of running <span class="command"><strong>svn
diff</strong></span> and sending a patch file (which won't capture
directory or symlink changes), you can
use <span class="command"><strong>svn copy</strong></span> to <span class="quote">“<span class="quote">upload</span>”</span> your
working copy to a private area of the repository. Your
collaborator can then either check out a verbatim copy of your
working copy or use <span class="command"><strong>svn merge</strong></span> to receive
your exact changes.</p><p>While this is a nice method for uploading a quick snapshot
of your working copy, note that this is <span class="emphasis"><em>not</em></span>
a good way to initially create a branch. Branch creation should
be an event unto itself, and this method conflates the creation
of a branch with extra changes to files, all within a single revision.
This makes it very difficult (later on) to identify a single
revision number as a branch point.</p></div></div><div class="sect1" title="Branch Maintenance"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.branchmerge.maint"></a>Branch Maintenance</h2></div></div></div><p>You may have noticed by now that Subversion is extremely
flexible. Because it implements branches and tags with the same
underlying mechanism (directory copies), and because branches
and tags appear in normal filesystem space, many people find
Subversion intimidating. It's almost <span class="emphasis"><em>too</em></span>
flexible. In this section, we'll offer some suggestions for
arranging and managing your data over time.</p><div class="sect2" title="Repository Layout"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.maint.layout"></a>Repository Layout</h3></div></div></div><p>There are some standard, recommended ways to organize the
contents of a repository. Most people create a
<code class="filename">trunk</code> directory to hold the <span class="quote">“<span class="quote">main
line</span>”</span> of development, a <code class="filename">branches</code>
directory to contain branch copies, and
a <code class="filename">tags</code> directory to contain tag copies.
If a repository holds only one project, often people create
these top-level directories:</p><div class="informalexample"><div class="literallayout"><p><br></br>
/<br></br>
trunk/<br></br>
branches/<br></br>
tags/<br></br>
</p></div></div><p>If a repository contains multiple projects, admins
typically index their layout by project. See <a class="xref" href="#svn.reposadmin.projects.chooselayout" title="Planning Your Repository Organization">the section called “Planning Your Repository Organization”</a> to read more about
<span class="quote">“<span class="quote">project roots</span>”</span>, but here's an example of such a
layout:</p><div class="informalexample"><div class="literallayout"><p><br></br>
/<br></br>
paint/<br></br>
trunk/<br></br>
branches/<br></br>
tags/<br></br>
calc/<br></br>
trunk/<br></br>
branches/<br></br>
tags/<br></br>
</p></div></div><p>Of course, you're free to ignore these common layouts.
You can create any sort of variation, whatever works best for
you or your team. Remember that whatever you choose, it's not
a permanent commitment. You can reorganize your repository at
any time. Because branches and tags are ordinary directories,
the <span class="command"><strong>svn move</strong></span> command can move or rename
them however you wish. Switching from one layout to another
is just a matter of issuing a series of server-side moves; if
you don't like the way things are organized in the repository,
just juggle the directories around.</p><p>Remember, though, that while moving directories is
easy to do, you need to be considerate of other users as well.
Your juggling can disorient users with existing
working copies. If a user has a working copy of a particular
repository directory and your <span class="command"><strong>svn move</strong></span>
subcommand removes the path from the latest revision, then
when the user next runs <span class="command"><strong>svn update</strong></span>, she is
told that her working copy represents a path that no
longer exists. She is then forced to <span class="command"><strong>svn
switch</strong></span> to the new location.</p></div><div class="sect2" title="Data Lifetimes"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.maint.lifetime"></a>Data Lifetimes</h3></div></div></div><p>Another nice feature of Subversion's model is that
branches and tags can have finite lifetimes, just like any
other versioned item. For example, suppose you eventually
finish all your work on your personal branch of the
<code class="filename">calc</code> project. After merging all of your
changes back into <code class="filename">/calc/trunk</code>, there's
no need for your private branch directory to stick around
anymore:</p><div class="informalexample"><pre class="screen">
$ svn delete http://svn.example.com/repos/calc/branches/my-calc-branch \
-m "Removing obsolete branch of calc project."
Committed revision 375.
</pre></div><p>And now your branch is gone. Of course, it's not really
gone: the directory is simply missing from the
<code class="literal">HEAD</code> revision, no longer distracting
anyone. If you use <span class="command"><strong>svn checkout</strong></span>,
<span class="command"><strong>svn switch</strong></span>, or <span class="command"><strong>svn list</strong></span>
to examine an earlier revision, you can still see
your old branch.</p><p>If browsing your deleted directory isn't enough, you can
always bring it back. Resurrecting data is very easy in
Subversion. If there's a deleted directory (or file) that
you'd like to bring back into <code class="literal">HEAD</code>, simply
use <span class="command"><strong>svn copy</strong></span> to copy it from the old
revision:</p><div class="informalexample"><pre class="screen">
$ svn copy http://svn.example.com/repos/calc/branches/my-calc-branch@374 \
http://svn.example.com/repos/calc/branches/my-calc-branch \
-m "Restore my-calc-branch."
Committed revision 376.
</pre></div><p>In our example, your personal branch had a relatively
short lifetime: you may have created it to fix a bug or
implement a new feature. When your task is done, so is the
branch. In software development, though, it's also common to
have two <span class="quote">“<span class="quote">main</span>”</span> branches running side by side for
very long periods. For example, suppose it's time to release
a stable version of the <code class="filename">calc</code> project to the
public, and you know it's going to take a couple of months to
shake bugs out of the software. You don't want people to add
new features to the project, but you don't want to tell all
developers to stop programming either. So instead, you create
a <span class="quote">“<span class="quote">stable</span>”</span> branch of the software that won't
change much:</p><div class="informalexample"><pre class="screen">
$ svn copy http://svn.example.com/repos/calc/trunk \
http://svn.example.com/repos/calc/branches/stable-1.0 \
-m "Creating stable branch of calc project."
Committed revision 377.
</pre></div><p>And now developers are free to continue adding
cutting-edge (or experimental) features to
<code class="filename">/calc/trunk</code>, and you can declare a
project policy that only bug fixes are to be committed to
<code class="filename">/calc/branches/stable-1.0</code>. That is, as
people continue to work on the trunk, a human selectively
ports bug fixes over to the stable branch. Even after the
stable branch has shipped, you'll probably continue to
maintain the branch for a long time—that is, as long
as you continue to support that release for customers. We'll
discuss this more in the next section.</p></div></div><div class="sect1" title="Common Branching Patterns"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.branchmerge.commonpatterns"></a>Common Branching Patterns</h2></div></div></div><p>There are many different uses for branching and <span class="command"><strong>svn
merge</strong></span>, and this section describes the most
common.</p><p>Version control is most often used for software
development, so here's a quick peek at two of the most common
branching/merging patterns used by teams of programmers. If
you're not using Subversion for software development, feel
free to skip this section. If you're a software developer
using version control for the first time, pay close attention,
as these patterns are often considered best practices by
experienced folk. These processes aren't specific to
Subversion; they're applicable to any version control system.
Still, it may help to see them described in Subversion
terms.</p><div class="sect2" title="Release Branches"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.commonpatterns.release"></a>Release Branches</h3></div></div></div><p>Most software has a typical life cycle: code, test,
release, repeat. There are two problems with this process.
First, developers need to keep writing new features while
quality assurance teams take time to test supposedly stable
versions of the software. New work cannot halt while the
software is tested. Second, the team almost always needs to
support older, released versions of software; if a bug is
discovered in the latest code, it most likely exists in
released versions as well, and customers will want to get
that bug fix without having to wait for a major new
release.</p><p>Here's where version control can help. The typical
procedure looks like this:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p><span class="emphasis"><em>Developers commit all new work to the
trunk.</em></span> Day-to-day changes are committed to
<code class="filename">/trunk</code>: new features, bug fixes, and
so on.</p></li><li class="listitem"><p><span class="emphasis"><em>The trunk is copied to a
<span class="quote">“<span class="quote">release</span>”</span> branch.</em></span> When the team
thinks the software is ready for release (say, a 1.0
release), <code class="filename">/trunk</code> might be copied to
<code class="filename">/branches/1.0</code>.</p></li><li class="listitem"><p><span class="emphasis"><em>Teams continue to work in
parallel.</em></span> One team begins rigorous testing of
the release branch, while another team continues new work
(say, for version 2.0) on <code class="filename">/trunk</code>. If
bugs are discovered in either location, fixes are ported
back and forth as necessary. At some point, however, even
that process stops. The branch is <span class="quote">“<span class="quote">frozen</span>”</span>
for final testing right before a release.</p></li><li class="listitem"><p><span class="emphasis"><em>The branch is tagged and
released.</em></span> When testing is complete,
<code class="filename">/branches/1.0</code> is copied to
<code class="filename">/tags/1.0.0</code> as a reference
snapshot. The tag is packaged and released to
customers.</p></li><li class="listitem"><p><span class="emphasis"><em>The branch is maintained over
time.</em></span> While work continues
on <code class="filename">/trunk</code> for version 2.0, bug fixes
continue to be ported from <code class="filename">/trunk</code> to
<code class="filename">/branches/1.0</code>. When enough
bug fixes have accumulated, management may decide to do a
1.0.1 release: <code class="filename">/branches/1.0</code> is
copied to <code class="filename">/tags/1.0.1</code>, and the tag
is packaged and released.</p></li></ol></div><p>This entire process repeats as the software matures:
when the 2.0 work is complete, a new 2.0 release branch is
created, tested, tagged, and eventually released. After
some years, the repository ends up with a number of release
branches in <span class="quote">“<span class="quote">maintenance</span>”</span> mode, and a number
of tags representing final shipped versions.</p></div><div class="sect2" title="Feature Branches"><div class="titlepage"><div><div><h3 class="title"><a id="svn.branchmerge.commonpatterns.feature"></a>Feature Branches</h3></div></div></div><p>A <em class="firstterm">feature branch</em> is the sort of
branch that's been the dominant example in this chapter (the
one you've been working on while Sally continues to work on
<code class="filename">/trunk</code>). It's a temporary branch created
to work on a complex change without interfering with the
stability of <code class="filename">/trunk</code>. Unlike release
branches (which may need to be supported forever), feature
branches are born, used for a while, merged back to the trunk,
and then ultimately deleted. They have a finite span of
usefulness.</p><p>Again, project policies vary widely concerning exactly
when it's appropriate to create a feature branch. Some
projects never use feature branches at all: commits to
<code class="filename">/trunk</code> are a free-for-all. The
advantage to this system is that it's simple—nobody
needs to learn about branching or merging. The disadvantage
is that the trunk code is often unstable or unusable. Other
projects use branches to an extreme: no change is
<span class="emphasis"><em>ever</em></span> committed to the trunk directly.
Even the most trivial changes are created on a short-lived
branch, carefully reviewed, and merged to the trunk. Then
the branch is deleted. This system guarantees an
exceptionally stable and usable trunk at all times, but at
the cost of tremendous process overhead.</p><p>Most projects take a middle-of-the-road approach. They
commonly insist that <code class="filename">/trunk</code> compile and
pass regression tests at all times. A feature branch is
required only when a change requires a large number of
destabilizing commits. A good rule of thumb is to ask this
question: if the developer worked for days in isolation and
then committed the large change all at once (so that
<code class="filename">/trunk</code> were never destabilized), would it
be too large a change to review? If the answer to that
question is <span class="quote">“<span class="quote">yes,</span>”</span> the change should be
developed on a feature branch. As the developer commits
incremental changes to the branch, they can be easily reviewed
by peers.</p><p>Finally, there's the issue of how to best keep a feature
branch in <span class="quote">“<span class="quote">sync</span>”</span> with the trunk as work
progresses. As we mentioned earlier, there's a great risk to
working on a branch for weeks or months; trunk changes may
continue to pour in, to the point where the two lines of
development differ so greatly that it may become a nightmare
trying to merge the branch back to the trunk.</p><p>This situation is best avoided by regularly merging trunk
changes to the branch. Make up a policy: once a week, merge
the last week's worth of trunk changes to the branch.</p><p>When you are eventually ready to merge the
<span class="quote">“<span class="quote">synchronized</span>”</span> feature branch back to the trunk,
begin by doing a final merge of the latest trunk
changes to the branch. When that's done, the latest versions
of branch and trunk are absolutely identical except for
your branch changes. You then merge back with
the <code class="option">--reintegrate</code> option:</p><div class="informalexample"><pre class="screen">
$ cd trunk-working-copy
$ svn update
Updating '.':
At revision 1910.
$ svn merge --reintegrate ^/calc/branches/mybranch
--- Merging differences between repository URLs into '.':
U real.c
U integer.c
A newdirectory
A newdirectory/newfile
U .
…
</pre></div><p>Another way of thinking about this pattern is that your
weekly sync of trunk to branch is analogous to running
<span class="command"><strong>svn update</strong></span> in a working copy, while the
final merge step is analogous to running <span class="command"><strong>svn
commit</strong></span> from a working copy. After all, what else
<span class="emphasis"><em>is</em></span> a working copy but a very shallow
private branch? It's a branch that's capable of
storing only one change at a time.</p></div></div><div class="sect1" title="Vendor Branches"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.advanced.vendorbr"></a>Vendor Branches</h2></div></div></div><p>As is especially the case when developing software, the data
that you maintain under version control is often closely related
to, or perhaps dependent upon, someone else's data. Generally,
the needs of your project will dictate that you stay as
up to date as possible with the data provided by that external
entity without sacrificing the stability of your own project.
This scenario plays itself out all the time—anywhere that
the information generated by one group of people has a direct
effect on that which is generated by another group.</p><p>For example, software developers might be working on an
application that makes use of a third-party library. Subversion
has just such a relationship with the Apache Portable Runtime (APR)
library (see <a class="xref" href="#svn.developer.usingapi.apr" title="The Apache Portable Runtime Library">the section called “The Apache Portable Runtime Library”</a>).
The Subversion source code depends on the APR library for all
its portability needs. In earlier stages of Subversion's
development, the project closely tracked APR's changing API,
always sticking to the <span class="quote">“<span class="quote">bleeding edge</span>”</span> of the
library's code churn. Now that both APR and Subversion have
matured, Subversion attempts to synchronize with APR's library
API only at well-tested, stable release points.</p><p>Now, if your project depends on someone else's information,
you could attempt to synchronize that information with your own
in several ways. Most painfully, you could issue oral or
written instructions to all the contributors of your project,
telling them to make sure they have the specific versions
of that third-party information that your project needs. If the
third-party information is maintained in a Subversion
repository, you could also use Subversion's externals
definitions to effectively <span class="quote">“<span class="quote">pin down</span>”</span> specific
versions of that information to some location in your own
working copy (see
<a class="xref" href="#svn.advanced.externals" title="Externals Definitions">the section called “Externals Definitions”</a>).</p><p>But sometimes you want to maintain custom modifications to
third-party code in your own version control system. Returning
to the software development example, programmers might need to
make modifications to that third-party library for their own
purposes. These modifications might include new functionality
or bug fixes, maintained internally only until they become part
of an official release of the third-party library. Or the
changes might never be relayed back to the library maintainers,
existing solely as custom tweaks to make the library further
suit the needs of the software developers.</p><p>Now you face an interesting situation. Your project could
house its custom modifications to the third-party data in some
disjointed fashion, such as using patch files or full-fledged
alternative versions of files and directories. But these quickly
become maintenance headaches, requiring some mechanism by which
to apply your custom changes to the third-party code and
necessitating regeneration of those changes with each successive
version of the third-party code that you track.</p><p>The solution to this problem is to use <em class="firstterm">vendor
branches</em>. A vendor branch is a directory tree in
your own version control system that contains information
provided by a third-party entity, or vendor. Each version of
the vendor's data that you decide to absorb into your project is
called a <em class="firstterm">vendor drop</em>.</p><p>Vendor branches provide two benefits. First, by storing
the currently supported vendor drop in your own version control
system, you ensure that the members of your project never need to question
whether they have the right version of the vendor's data. They
simply receive that correct version as part of their regular
working copy updates. Second, because the data lives in your
own Subversion repository, you can store your custom changes to
it in-place—you have no more need of an automated (or
worse, manual) method for swapping in your customizations.</p><div class="sect2" title="General Vendor Branch Management Procedure"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.vendorbr.general"></a>General Vendor Branch Management Procedure</h3></div></div></div><p>Managing vendor branches generally works like this: first,
you create a top-level directory (such as
<code class="filename">/vendor</code>) to hold the vendor branches.
Then you import the third-party code into a subdirectory of
that top-level directory. You then copy that subdirectory
into your main development branch (e.g.,
<code class="filename">/trunk</code>) at the appropriate location. You
always make your local changes in the main development branch.
With each new release of the code you are tracking, you bring
it into the vendor branch and merge the changes into
<code class="filename">/trunk</code>, resolving whatever conflicts
occur between your local changes and the upstream
changes.</p><p>An example will help to clarify this algorithm. We'll use
a scenario where your development team is creating a
calculator program that links against a third-party complex
number arithmetic library, libcomplex. We'll begin with the
initial creation of the vendor branch and the import of the
first vendor drop. We'll call our vendor branch directory
<code class="filename">libcomplex</code>, and our code drops will go
into a subdirectory of our vendor branch called
<code class="filename">current</code>. And since <span class="command"><strong>svn
import</strong></span> creates all the intermediate parent
directories it needs, we can actually accomplish both of these
steps with a single command:</p><div class="informalexample"><pre class="screen">
$ svn import /path/to/libcomplex-1.0 \
http://svn.example.com/repos/vendor/libcomplex/current \
-m "importing initial 1.0 vendor drop"
…
</pre></div><p>We now have the current version of the libcomplex source
code in <code class="filename">/vendor/libcomplex/current</code>. Now,
we tag that version (see <a class="xref" href="#svn.branchmerge.tags" title="Tags">the section called “Tags”</a>)
and then copy it into the main development branch. Our copy
will create a new directory called
<code class="filename">libcomplex</code> in our existing
<code class="filename">calc</code> project directory. It is in this
copied version of the vendor data that we will make our
customizations:</p><div class="informalexample"><pre class="screen">
$ svn copy http://svn.example.com/repos/vendor/libcomplex/current \
http://svn.example.com/repos/vendor/libcomplex/1.0 \
-m "tagging libcomplex-1.0"
…
$ svn copy http://svn.example.com/repos/vendor/libcomplex/1.0 \
http://svn.example.com/repos/calc/libcomplex \
-m "bringing libcomplex-1.0 into the main branch"
…
</pre></div><p>We check out our project's main branch—which now
includes a copy of the first vendor drop—and we get to
work customizing the libcomplex code. Before we know it, our
modified version of libcomplex is now completely integrated
into our calculator program.<sup>[<a id="idp10580176" href="#ftn.idp10580176" class="footnote">35</a>]</sup></p><p>A few weeks later, the developers of libcomplex release a
new version of their library—version 1.1—which
contains some features and functionality that we really want.
We'd like to upgrade to this new version, but without losing
the customizations we made to the existing version. What we
essentially would like to do is to replace our current
baseline version of libcomplex 1.0 with a copy of libcomplex
1.1, and then re-apply the custom modifications we previously
made to that library to the new version. But we actually
approach the problem from the other direction, applying the
changes made to libcomplex between versions 1.0 and 1.1 to our
modified copy of it.</p><p>To perform this upgrade, we check out a copy of our vendor
branch and replace the code in the
<code class="filename">current</code> directory with the new libcomplex
1.1 source code. We quite literally copy new files on top of
existing files, perhaps exploding the libcomplex 1.1 release
tarball atop our existing files and directories. The goal
here is to make our <code class="filename">current</code> directory
contain only the libcomplex 1.1 code and to ensure that all
that code is under version control. Oh, and we want to do
this with as little version control history disturbance as
possible.</p><p>After replacing the 1.0 code with 1.1 code, <span class="command"><strong>svn
status</strong></span> will show files with local modifications as
well as, perhaps, some unversioned files. If we did what we
were supposed to do, the unversioned files are only those new
files introduced in the 1.1 release of libcomplex—we
run <span class="command"><strong>svn add</strong></span> on those to get them under
version control. If the 1.1 code no longer has certain files
that were in the 1.0 tree, it may be hard to notice them;
you'd have to compare the two trees with some external tool
and then <span class="command"><strong>svn delete</strong></span> any files present in
1.0 but not in 1.1. (Although it might also be just fine to
let these same files live on in unused obscurity!) Finally,
once our <code class="filename">current</code> working copy contains
only the libcomplex 1.1 code, we commit the changes we made to
get it looking that way.</p><p>Our <code class="filename">current</code> branch now contains the
new vendor drop. We tag the new version as 1.1 (in the same
way we previously tagged the version 1.0 vendor drop), and
then merge the differences between the tag of the previous
version and the new current version into our main development
branch:</p><div class="informalexample"><pre class="screen">
$ cd working-copies/calc
$ svn merge ^/vendor/libcomplex/1.0 \
^/vendor/libcomplex/current \
libcomplex
… # resolve all the conflicts between their changes and our changes
$ svn commit -m "merging libcomplex-1.1 into the main branch"
…
</pre></div><p>In the trivial use case, the new version of our
third-party tool would look, from a files-and-directories
point of view, just like the previous version. None of the
libcomplex source files would have been deleted, renamed, or
moved to different locations—the new version would
contain only textual modifications against the previous one.
In a perfect world, our modifications would apply cleanly to
the new version of the library, with absolutely no
complications or conflicts.</p><p>But things aren't always that simple, and in fact it is
quite common for source files to get moved around between
releases of software. This complicates the process of
ensuring that our modifications are still valid for the new
version of code, and things can quickly degrade into a
situation where we have to manually re-create our
customizations in the new version. Once Subversion knows
about the history of a given source file—including all
its previous locations—the process of merging in the new
version of the library is pretty simple. But we are
responsible for telling Subversion how the source file layout
changed from vendor drop to vendor drop.</p></div><div class="sect2" title="svn_load_dirs.pl"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.vendorbr.svn_load_dirs"></a>svn_load_dirs.pl</h3></div></div></div><p>Vendor drops that contain more than a few deletes,
additions, and moves complicate the process of upgrading to
each successive version of the third-party data. So
Subversion supplies the <span class="command"><strong>svn_load_dirs.pl</strong></span>
script to assist with this process. This script automates the
importing steps we mentioned in the general vendor branch
management procedure to make sure mistakes are minimized.
You will still be responsible for using the merge commands to
merge the new versions of the third-party data into your main
development branch, but <span class="command"><strong>svn_load_dirs.pl</strong></span>
can help you more quickly and easily arrive at that
stage.</p><p>In short, <span class="command"><strong>svn_load_dirs.pl</strong></span> is an
enhancement to <span class="command"><strong>svn import</strong></span> that has several
important characteristics:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>It can be run at any point in time to bring an existing
directory in the repository to exactly match an external
directory, performing all the necessary adds and deletes,
and optionally performing moves, too.</p></li><li class="listitem"><p>It takes care of complicated series of operations between
which Subversion requires an intermediate commit—such
as before renaming a file or directory twice.</p></li><li class="listitem"><p>It will optionally tag the newly imported directory.</p></li><li class="listitem"><p>It will optionally add arbitrary properties to files and
directories that match a regular expression.</p></li></ul></div><p><span class="command"><strong>svn_load_dirs.pl</strong></span> takes three mandatory
arguments. The first argument is the URL to the base
Subversion directory to work in. This argument is followed by
the URL—relative to the first argument—into which the
current vendor drop will be imported. Finally, the third
argument is the local directory to import. Using our previous
example, a typical run of <span class="command"><strong>svn_load_dirs.pl</strong></span>
might look like this:</p><div class="informalexample"><pre class="screen">
$ svn_load_dirs.pl http://svn.example.com/repos/vendor/libcomplex \
current \
/path/to/libcomplex-1.1
…
</pre></div><p>You can indicate that you'd like
<span class="command"><strong>svn_load_dirs.pl</strong></span> to tag the new vendor drop
by passing the <code class="option">-t</code> command-line option and
specifying a tag name. This tag is another URL relative to
the first program argument.</p><div class="informalexample"><pre class="screen">
$ svn_load_dirs.pl -t libcomplex-1.1 \
http://svn.example.com/repos/vendor/libcomplex \
current \
/path/to/libcomplex-1.1
…
</pre></div><p>When you run <span class="command"><strong>svn_load_dirs.pl</strong></span>, it
examines the contents of your existing <span class="quote">“<span class="quote">current</span>”</span>
vendor drop and compares them with the proposed new vendor
drop. In the trivial case, no files will be in
one version and not the other, and the script will perform the
new import without incident. If, however, there are
discrepancies in the file layouts between versions,
<span class="command"><strong>svn_load_dirs.pl</strong></span> will ask you how
to resolve those differences. For example, you
will have the opportunity to tell the script that you know
that the file <code class="filename">math.c</code> in version 1.0 of
libcomplex was renamed to <code class="filename">arithmetic.c</code> in
libcomplex 1.1. Any discrepancies not explained by moves
are treated as regular additions and deletions.</p><p>The script also accepts a separate configuration file for
setting properties on <span class="emphasis"><em>added</em></span> files and
directories which match a regular expression. This configuration
file is specified to <span class="command"><strong>svn_load_dirs.pl</strong></span> using the
<code class="option">-p</code> command-line option. Each line of the
configuration file is a whitespace-delimited set of two or
four values: a Perl-style regular expression against which to match the
added path, a control keyword (either
<code class="literal">break</code> or <code class="literal">cont</code>), and then
optionally a property name and value.</p><div class="informalexample"><pre class="screen">
\.png$ break svn:mime-type image/png
\.jpe?g$ break svn:mime-type image/jpeg
\.m3u$ cont svn:mime-type audio/x-mpegurl
\.m3u$ break svn:eol-style LF
.* break svn:eol-style native
</pre></div><p>For each added path, the configured property changes whose
regular expression matches the path are applied in order,
unless the control specification is <code class="literal">break</code>
(which means that no more property changes should be applied
to that path). If the control specification is
<code class="literal">cont</code>—an abbreviation for
<code class="literal">continue</code>—matching will continue
with the next line of the configuration file.</p><p>Any whitespace in the regular expression, property name,
or property value must be surrounded by either single or
double quotes. You can escape quotes that
are not used for wrapping whitespace by preceding them with a
backslash (<code class="literal">\</code>) character. The backslash
escapes only quotes when parsing the configuration file, so do
not protect any other characters beyond what is necessary for
the regular expression.</p></div></div><div class="sect1" title="To Branch or Not to Branch?"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.branchmerge.when"></a>To Branch or Not to Branch?</h2></div></div></div><p>To branch or not to branch—that is an interesting
question. This chapter has provided thus far a pretty deep dive
into the waters of branching and merging, topics which have
historically been the premier source of Subversion user
confusion. As if the rote actions involved in branching and
branch management aren't sometimes tricky enough, some users get
hung up on deciding whether they need to branch at all. As
you've learned, Subversion can handle common branching and
branch management scenarios. So, the decision of whether or not
to branch a project's history is rarely a technical one.
Rather, the social impact of the decision often carries more
weight. Let's examine some of the benefits and costs of using
branches in a software project.</p><p>The most obvious benefit of working on a branch is
isolation. Changes made to the branch don't affect the other
lines of development in the project; changes made to those other
lines don't affect the branch. In this way, a branch can serve
as a great place to experiment with new features, complex bug
fixes, major code rewrites, and so on. No matter how much stuff
Sally breaks on her branch, Harry and the rest of the team can
continue with their work unhindered outside the branch.</p><p>Branches also provide a great way to organize related
changes into readily identifiable collections. For example, the
changes which comprise the complete solution to a particular bug
might be a list of non-sequential revision numbers. You might
describe them in human language as <span class="quote">“<span class="quote">revisions 1534, 1543,
1587 and 1588</span>”</span>. You'd probably reproduce those numbers
manually (or otherwise) in the issue tracker artifact which
tracks the bug. When porting the bug fix to other product
versions, you'd need to make sure to port all those revisions.
But had those changes been made on a unique branch, you'd find
yourself referring only to that branch by its name in
conversation, in issue tracker comments, and when porting
changes.</p><p>The unfortunate downside of branches, though, is that the
very isolation that makes them so
useful <span class="emphasis"><em>can</em></span> be at odds with the
collaborative needs of the project team. Depending on the work
habits of your project peers, changes made to branches might not
get the kind of constructive review, criticism, and testing that
changes made to the main line of development do. The isolation
of a branch can encourage users to forsake certain version
control <span class="quote">“<span class="quote">best practices</span>”</span>, leading to version
history which is difficult to review <span class="foreignphrase"><em class="foreignphrase">post
facto</em></span>. Developers on long-lived branches
sometimes need to work extra hard to ensure that the
evolutionary direction of their isolated copy of the codebase is
in harmony with the direction their peers are steering the main
code lines. Now, these drawbacks might be less of an issue for
true exploratory branches aimed at experimenting with the future
of a codebase with no expectation of reintegrating the results
back into the main development lines—mere policy needn't
be a vision-killer! But the simple fact remains that projects
generally benefit from an orderly approach to version control
where code and code changes enjoy the review and comprehension
of more than one team member.</p><p>That's not to say that there are no technical penalties to
branching. Pardon us while we <span class="quote">“<span class="quote">go meta</span>”</span> for a bit
here. If you think about it, every time you checkout a
Subversion working copy, you're creating a branch of sorts of
your project. It's a special sort of branch. It lives only on
your client machine; not in the repository. You synchronize
this branch with changes made in the repository
using <span class="command"><strong>svn update</strong></span>—which acts almost like
a special-cased, simplified form of an <span class="command"><strong>svn
merge</strong></span> command.<sup>[<a id="idp10637440" href="#ftn.idp10637440" class="footnote">36</a>]</sup> You effectively reintegrate your branch
each time you run <span class="command"><strong>svn commit</strong></span>. So, in that
special sense, Subversion users deal with branches and merges
all the time. Given the similarities between updating and
merging, it's no surprise, then, that the areas in which
Subversion seems to have the most shortcomings—namely,
handling file and directory renames and dealing with tree
conflicts in general—are problematic for both
the <span class="command"><strong>svn update</strong></span> and <span class="command"><strong>svn
merge</strong></span> operations. Unfortunately, <span class="command"><strong>svn
merge</strong></span> has a harder time of it precisely because of the
fact that, for every way in which <span class="command"><strong>svn update</strong></span>
is a special-cased, simplified kind of generic merge operation,
a true Subversion merge is neither special-cased nor simplified.
For this reason, merges perform much more slowly than updates,
require explicit tracking (via
the <code class="literal">svn:mergeinfo</code> property we've discussed in
this chapter) and history-crunching arithmetic, and generally
offer more opportunities for something to go awry.</p><p>To branch or not to branch? Ultimately, that depends on
what your team needs in order to find that sweet balance of
collaboration and isolation.</p></div><div class="sect1" title="Summary"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.branchmerge.summary"></a>Summary</h2></div></div></div><p>We covered a lot of ground in this chapter. We
discussed the concepts of tags and branches and demonstrated
how Subversion implements these concepts by copying directories
with the <span class="command"><strong>svn copy</strong></span> command. We showed how
to use <span class="command"><strong>svn merge</strong></span> to copy changes from one
branch to another or roll back bad changes. We went over
the use of <span class="command"><strong>svn switch</strong></span> to create
mixed-location working copies. And we talked about how one
might manage the organization and lifetimes of branches in a
repository.</p><p>Remember the Subversion mantra: branches and tags are cheap.
So don't be afraid to use them when needed!</p><p>As a helpful reminder of all the operations we discussed,
here is handy reference table you can consult as
you begin to make use of branches.</p><div class="table"><a id="svn.branchemerge.summary.tbl-1"></a><p class="title"><strong>Table 4.1. Branching and merging commands</strong></p><div class="table-contents"><table summary="Branching and merging commands" border="1"><colgroup><col></col><col></col></colgroup><thead><tr><th>Action</th><th>Command</th></tr></thead><tbody><tr><td>Create a branch or tag</td><td><strong class="userinput"><code>svn copy <em class="replaceable"><code>URL1</code></em> <em class="replaceable"><code>URL2</code></em></code></strong></td></tr><tr><td>Switch a working copy to a branch or tag</td><td><strong class="userinput"><code>svn switch <em class="replaceable"><code>URL</code></em></code></strong></td></tr><tr><td>Synchronize a branch with trunk</td><td><strong class="userinput"><code>svn merge <em class="replaceable"><code>trunkURL</code></em>; svn commit</code></strong></td></tr><tr><td>See merge history or eligible changesets</td><td><strong class="userinput"><code>svn mergeinfo <em class="replaceable"><code>SOURCE</code></em> <em class="replaceable"><code>TARGET</code></em></code></strong></td></tr><tr><td>Merge a branch back into trunk</td><td><strong class="userinput"><code>svn merge --reintegrate <em class="replaceable"><code>branchURL</code></em>; svn commit</code></strong></td></tr><tr><td>Merge one specific change</td><td><strong class="userinput"><code>svn merge -c <em class="replaceable"><code>REV</code></em> <em class="replaceable"><code>URL</code></em>; svn commit</code></strong></td></tr><tr><td>Merge a range of changes</td><td><strong class="userinput"><code>svn merge -r <em class="replaceable"><code>REV1</code></em>:<em class="replaceable"><code>REV2</code></em> <em class="replaceable"><code>URL</code></em>; svn commit</code></strong></td></tr><tr><td>Block a change from automatic merging</td><td><strong class="userinput"><code>svn merge -c <em class="replaceable"><code>REV</code></em> --record-only <em class="replaceable"><code>URL</code></em>; svn commit</code></strong></td></tr><tr><td>Preview a merge</td><td><strong class="userinput"><code>svn merge <em class="replaceable"><code>URL</code></em> --dry-run</code></strong></td></tr><tr><td>Abandon merge results</td><td><strong class="userinput"><code>svn revert -R .</code></strong></td></tr><tr><td>Resurrect something from history</td><td><strong class="userinput"><code>svn copy <em class="replaceable"><code>URL</code></em>@<em class="replaceable"><code>REV</code></em> <em class="replaceable"><code>localPATH</code></em></code></strong></td></tr><tr><td>Undo a committed change</td><td><strong class="userinput"><code>svn merge -c -<em class="replaceable"><code>REV</code></em> <em class="replaceable"><code>URL</code></em>; svn commit</code></strong></td></tr><tr><td>Examine merge-sensitive history</td><td><strong class="userinput"><code>svn log -g; svn blame -g</code></strong></td></tr><tr><td>Create a tag from a working copy</td><td><strong class="userinput"><code>svn copy . <em class="replaceable"><code>tagURL</code></em></code></strong></td></tr><tr><td>Rearrange a branch or tag</td><td><strong class="userinput"><code>svn move <em class="replaceable"><code>URL1</code></em> <em class="replaceable"><code>URL2</code></em></code></strong></td></tr><tr><td>Remove a branch or tag</td><td><strong class="userinput"><code>svn delete <em class="replaceable"><code>URL</code></em></code></strong></td></tr></tbody></table></div></div><br class="table-break"></br></div><div class="footnotes"><br></br><hr width="100" align="left"></hr><div class="footnote"><p><sup>[<a id="ftn.idp9784560" href="#idp9784560" class="para">25</a>] </sup>Subversion does not support copying between
different repositories. When using URLs with <span class="command"><strong>svn
copy</strong></span> or <span class="command"><strong>svn move</strong></span>, you can only
copy items within the same repository.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp9866272" href="#idp9866272" class="para">26</a>] </sup>This was introduced in svn
1.6.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp9920496" href="#idp9920496" class="para">27</a>] </sup>The <span class="command"><strong>svn
merge</strong></span> subcommand
option <code class="option">--allow-mixed-revisions</code> allows you to
override this prohibition, but you should only do so if you
understand the ramifications and have a good reason for
it.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp9945040" href="#idp9945040" class="para">28</a>] </sup>With Subversion 1.7 you don't
absolutely have to do all your sync merges to the root of your
branch as we do in this example. <span class="emphasis"><em>If</em></span> your
branch is effectively synced via a series of subtree
merges then the reintegrate will work, but ask yourself, if the
branch is effectively synced, then why are you doing subtree
merges? Doing so is almost always needlessly
complex.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp9962880" href="#idp9962880" class="para">29</a>] </sup>Reintegrate
merges are allowed if the target is a shallow checkout (see
<a class="xref" href="#svn.advanced.sparsedirs" title="Sparse Directories">the section called “Sparse Directories”</a>) but any paths affected
by the diff which are <span class="quote">“<span class="quote">missing</span>”</span> due to the sparse
working copy will be skipped, probably <span class="emphasis"><em>not</em></span>
what you intended!</p></div><div class="footnote"><p><sup>[<a id="ftn.idp10042976" href="#idp10042976" class="para">30</a>] </sup>
This is often termed an <span class="quote">“<span class="quote">inoperative</span>”</span> merge. Though
in this example the merge of r1060 would do something: It would
update the mergeinfo on the root of the branch, but it would be
inoperative in the sense that no diff would be
applied.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp10075296" href="#idp10075296" class="para">31</a>] </sup>The
Subversion project has plans, however, to someday implement a
command that would accomplish the task of permanently deleting
information. In the meantime, see
<a class="xref" href="#svn.reposadmin.maint.tk.svndumpfilter" title="svndumpfilter">the section called “svndumpfilter”</a> for a
possible workaround.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp10132288" href="#idp10132288" class="para">32</a>] </sup>At least, this is true in Subversion
1.7 at the time of this writing. This behavior may improve
in future versions of Subversion.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp10187360" href="#idp10187360" class="para">33</a>] </sup>Interestingly, after rolling
back a revision like this, we wouldn't be able to
reapply the revision using <strong class="userinput"><code>svn merge . -c
5</code></strong>, since the mergeinfo would already list r5
as being applied. We would have to use
the <code class="option">--ignore-ancestry</code> option to make
the merge command ignore the existing
mergeinfo!</p></div><div class="footnote"><p><sup>[<a id="ftn.idp10408640" href="#idp10408640" class="para">34</a>] </sup>You <span class="emphasis"><em>can</em></span>, however,
use <span class="command"><strong>svn relocate</strong></span> if the URL of your server
changes and you don't want to abandon an existing working copy.
See <a class="xref" href="#svn.ref.svn.c.relocate" title="svn relocate">svn relocate</a> in
<a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a> for more information and an
example.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp10580176" href="#idp10580176" class="para">35</a>] </sup>And is entirely
bug-free, of course!</p></div><div class="footnote"><p><sup>[<a id="ftn.idp10637440" href="#idp10637440" class="para">36</a>] </sup>Actually, you
<span class="emphasis"><em>could</em></span> use <strong class="userinput"><code>svn merge
-r<em class="replaceable"><code>LAST_UPDATED_REV</code></em>:HEAD .</code></strong>
in your working copy to quite literally merge in all the
repository changes since your last update if really wanted
to!</p></div></div></div><div class="chapter" title="Chapter 5. Repository Administration"><div class="titlepage"><div><div><h2 class="title"><a id="svn.reposadmin"></a>Chapter 5. Repository Administration</h2></div></div></div><div class="toc"><p><strong>Table of Contents</strong></p><dl><dt><span class="sect1"><a href="#svn.reposadmin.basics">The Subversion Repository, Defined</a></span></dt><dt><span class="sect1"><a href="#svn.reposadmin.planning">Strategies for Repository Deployment</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.reposadmin.projects.chooselayout">Planning Your Repository Organization</a></span></dt><dt><span class="sect2"><a href="#svn.reposadmin.basics.hosting">Deciding Where and How to Host Your Repository</a></span></dt><dt><span class="sect2"><a href="#svn.reposadmin.basics.backends">Choosing a Data Store</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.reposadmin.basics.backends.bdb">Berkeley DB</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.basics.backends.fsfs">FSFS</a></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.reposadmin.create">Creating and Configuring Your Repository</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.reposadmin.basics.creating">Creating the Repository</a></span></dt><dt><span class="sect2"><a href="#svn.reposadmin.create.hooks">Implementing Repository Hooks</a></span></dt><dt><span class="sect2"><a href="#svn.reposadmin.create.bdb">Berkeley DB Configuration</a></span></dt><dt><span class="sect2"><a href="#svn.reposadmin.create.fsfs">FSFS Configuration</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.reposadmin.maint">Repository Maintenance</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.reposadmin.maint.tk">An Administrator's Toolkit</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.reposadmin.maint.tk.svnadmin">svnadmin</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.tk.svnlook">svnlook</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.tk.svndumpfilter">svndumpfilter</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.tk.svnrdump">svnrdump</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.tk.svnsync">svnsync</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.tk.fsfsreshard">fsfs-reshard.py</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.tk.bdbutil">Berkeley DB utilities</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.reposadmin.maint.setlog">Commit Log Message Correction</a></span></dt><dt><span class="sect2"><a href="#svn.reposadmin.maint.diskspace">Managing Disk Space</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.reposadmin.maint.diskspace.deltas">How Subversion saves disk space</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.diskspace.deadtxns">Removing dead transactions</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.diskspace.bdblogs">Purging unused Berkeley DB logfiles</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.diskspace.fsfspacking">Packing FSFS filesystems</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.reposadmin.maint.recovery">Berkeley DB Recovery</a></span></dt><dt><span class="sect2"><a href="#svn.reposadmin.maint.migrate">Migrating Repository Data Elsewhere</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.reposadmin.maint.migrate.svnadmin">Repository data migration using svnadmin</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.migrate.svnrdump">Repository data migration using svnrdump</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.reposadmin.maint.filtering">Filtering Repository History</a></span></dt><dt><span class="sect2"><a href="#svn.reposadmin.maint.replication">Repository Replication</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.reposadmin.maint.replication.svnsync">Replication with svnsync</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.replication.svnsync-partial">Partial replication with svnsync</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.replication.svnsync-init-nonempty">A quick trick for mirror creation</a></span></dt><dt><span class="sect3"><a href="#svn.reposadmin.maint.replication.wrapup">Replication wrap-up</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.reposadmin.maint.backup">Repository Backup</a></span></dt><dt><span class="sect2"><a href="#svn.reposadmin.maint.uuids">Managing Repository UUIDs</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.reposadmin.maint.moving-and-removing">Moving and Removing Repositories</a></span></dt><dt><span class="sect1"><a href="#svn.reposadmin.summary">Summary</a></span></dt></dl></div><p>The Subversion repository is the central storehouse of all
your versioned data. As such, it becomes an obvious candidate for
all the love and attention an administrator can offer. While the
repository is generally a low-maintenance item, it is important to
understand how to properly configure and care for it so that
potential problems are avoided, and so actual problems are safely
resolved.</p><p>In this chapter, we'll discuss how to create and configure a
Subversion repository. We'll also talk about repository
maintenance, providing examples of how and when to use various
related tools provided with Subversion. We'll address some common
questions and mistakes and give some suggestions on how to arrange
the data in the repository.</p><p>If you plan to access a Subversion repository only in the role
of a user whose data is under version control (i.e., via a
Subversion client), you can skip this chapter altogether.
However, if you are, or wish to become, a Subversion repository
administrator,<sup>[<a id="idp11598224" href="#ftn.idp11598224" class="footnote">37</a>]</sup> this chapter is for you.</p><div class="sect1" title="The Subversion Repository, Defined"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.reposadmin.basics"></a>The Subversion Repository, Defined</h2></div></div></div><p>Before jumping into the broader topic of repository
administration, let's further define what a repository is. How
does it look? How does it feel? Does it take its tea hot or
iced, sweetened, and with lemon? As an administrator, you'll be
expected to understand the composition of a repository both from
a literal, OS-level perspective—how a repository looks and
acts with respect to non-Subversion tools—and from a
logical perspective—dealing with how data is represented
<span class="emphasis"><em>inside</em></span> the repository.</p><p>Seen through the eyes of a typical file browser application
(such as Windows Explorer) or command-line based filesystem
navigation tools, the Subversion repository is just another
directory full of stuff. There are some subdirectories with
human-readable configuration files in them, some subdirectories
with some not-so-human-readable data files, and so on. As in
other areas of the Subversion design, modularity is given high
regard, and hierarchical organization is preferred to cluttered
chaos. So a shallow glance into a typical repository from a
nuts-and-bolts perspective is sufficient to reveal the basic
components of the repository:</p><div class="informalexample"><pre class="screen">
$ ls repos
conf/ db/ format hooks/ locks/ README.txt
</pre></div><p>Here's a quick fly-by overview of what exactly you're seeing
in this directory listing. (Don't get bogged down in the
terminology—detailed coverage of these components exists
elsewhere in this and other chapters.)</p><div class="variablelist"><dl><dt><span class="term">conf/</span></dt><dd><p>This directory is a container for configuration
files.</p></dd><dt><span class="term">db/</span></dt><dd><p>This directory contains the data store for all of your
versioned data.<sup>[<a id="idp11652016" href="#ftn.idp11652016" class="footnote">38</a>]</sup></p></dd><dt><span class="term">format</span></dt><dd><p>This file describes the repository's internal
organizational scheme. (As it turns out,
the <code class="filename">db/</code> subdirectory sometimes also
contains a <code class="filename">format</code> file which
describes only the contents of that subdirectory and which
is not to be confused with this file.)</p></dd><dt><span class="term">hooks/</span></dt><dd><p>This directory contains hook script templates and
hook scripts, if any have been installed.</p></dd><dt><span class="term">locks/</span></dt><dd><p>Subversion uses this directory to house repository
lock files, used for managing concurrent access to the
repository.</p></dd><dt><span class="term">README.txt</span></dt><dd><p>This is a brief text file containing merely a notice
to readers that the directory they are looking in is a
Subversion repository.</p></dd></dl></div><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>Prior to Subversion 1.5, the on-disk repository structure
also always contained a <code class="filename">dav</code> subdirectory.
<code class="filename">mod_dav_svn</code> used this directory to store
information about
WebDAV <em class="firstterm">activities</em>—mappings of
high-level WebDAV protocol concepts to Subversion commit
transactions. Subversion 1.5 changed that behavior, moving
ownership of the activities directory, and the ability to
configure its location, into <code class="filename">mod_dav_svn</code>
itself. Now, new repositories will not necessarily have
a <code class="filename">dav</code> subdirectory
unless <code class="filename">mod_dav_svn</code> is in use and hasn't
been configured to store its activities database elsewhere.
See <a class="xref" href="#svn.ref.mod_dav_svn.conf.directives" title="Directives">the section called “Directives”</a> in
<a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a> for more information.</p></td></tr></table></div><p>Of course, when accessed via the Subversion libraries, this
otherwise unremarkable collection of files and directories
suddenly becomes an implementation of a virtual, versioned
filesystem, complete with customizable event triggers. This
filesystem has its own notions of directories and files, very
similar to the notions of such things held by real filesystems
(such as NTFS, FAT32, ext3, etc.). But this is a special
filesystem—it hangs these directories and files from
revisions, keeping all the changes you've ever made to them
safely stored and forever accessible. This is where the
entirety of your versioned data lives.</p></div><div class="sect1" title="Strategies for Repository Deployment"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.reposadmin.planning"></a>Strategies for Repository Deployment</h2></div></div></div><p>Due largely to the simplicity of the overall design of the
Subversion repository and the technologies on which it relies,
creating and configuring a repository are fairly straightforward
tasks. There are a few preliminary decisions you'll want to
make, but the actual work involved in any given setup of a
Subversion repository is pretty basic, tending toward
mindless repetition if you find yourself setting up multiples of
these things.</p><p>Some things you'll want to consider beforehand, though, are:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>What data do you expect to live in your repository (or
repositories), and how will that data be organized?</p></li><li class="listitem"><p>Where will your repository live, and how will it be
accessed?</p></li><li class="listitem"><p>What types of access control and repository event
reporting do you need?</p></li><li class="listitem"><p>Which of the available types of data store do you want
to use?</p></li></ul></div><p>In this section, we'll try to help you answer those
questions.</p><div class="sect2" title="Planning Your Repository Organization"><div class="titlepage"><div><div><h3 class="title"><a id="svn.reposadmin.projects.chooselayout"></a>Planning Your Repository Organization</h3></div></div></div><p>While Subversion allows you to move around versioned files
and directories without any loss of information, and even
provides ways of moving whole sets of versioned history from
one repository to another, doing so can greatly disrupt the
workflow of those who access the repository often and come to
expect things to be at certain locations. So before creating
a new repository, try to peer into the future a bit; plan
ahead before placing your data under version control. By
conscientiously <span class="quote">“<span class="quote">laying out</span>”</span> your repository or
repositories and their versioned contents ahead of time, you
can prevent many future headaches.</p><p>Let's assume that as repository administrator, you will be
responsible for supporting the version control system for
several projects. Your first decision is whether to use a
single repository for multiple projects, or to give each
project its own repository, or some compromise of these
two.</p><p>There are benefits to using a single repository for
multiple projects, most obviously the lack of duplicated
maintenance. A single repository means that there is one set
of hook programs, one thing to routinely back up, one thing to
dump and load if Subversion releases an incompatible new
version, and so on. Also, you can move data between projects
easily, without losing any historical versioning
information.</p><p>The downside of using a single repository is that
different projects may have different requirements in terms of
the repository event triggers, such as needing to send commit
notification emails to different mailing lists, or having
different definitions about what does and does not constitute
a legitimate commit. These aren't insurmountable problems, of
course—it just means that all of your hook scripts have
to be sensitive to the layout of your repository rather than
assuming that the whole repository is associated with a single
group of people. Also, remember that Subversion uses
repository-global revision numbers. While those numbers don't
have any particular magical powers, some folks still don't
like the fact that even though no changes have been made to
their project lately, the youngest revision number for the
repository keeps climbing because other projects are actively
adding new revisions.<sup>[<a id="idp11669152" href="#ftn.idp11669152" class="footnote">39</a>]</sup></p><p>A middle-ground approach can be taken, too. For example,
projects can be grouped by how well they relate to each other.
You might have a few repositories with a handful of projects
in each repository. That way, projects that are likely to
want to share data can do so easily, and as new revisions are
added to the repository, at least the developers know that
those new revisions are at least remotely related to everyone
who uses that repository.</p><p>
<a id="idp11686448" class="indexterm"></a>
After deciding how to organize your projects with respect
to repositories, you'll probably want to think about directory
hierarchies within the repositories themselves. Because
Subversion uses regular directory copies for branching and
tagging (see <a class="xref" href="#svn.branchmerge" title="Chapter 4. Branching and Merging">Chapter 4, <em>Branching and Merging</em></a>), the
Subversion community recommends that you choose a repository
location for each project
root—the <span class="quote">“<span class="quote">topmost</span>”</span> directory
that contains data related to that project—and then
create three subdirectories beneath that root:
<code class="filename">trunk</code>, meaning the directory under which
the main project development occurs;
<code class="filename">branches</code>, which is a directory in which
to create various named branches of the main development line;
and <code class="filename">tags</code>, which is a collection of tree
snapshots that are created, and perhaps destroyed, but never
changed.<sup>[<a id="idp11691600" href="#ftn.idp11691600" class="footnote">40</a>]</sup></p><p>For example, your repository might look like this:</p><div class="informalexample"><div class="literallayout"><p><br></br>
/<br></br>
calc/<br></br>
trunk/<br></br>
tags/<br></br>
branches/<br></br>
calendar/<br></br>
trunk/<br></br>
tags/<br></br>
branches/<br></br>
spreadsheet/<br></br>
trunk/<br></br>
tags/<br></br>
branches/<br></br>
…<br></br>
</p></div></div><p>Note that it doesn't matter where in your repository each
project root is. If you have only one project per repository,
the logical place to put each project root is at the root of
that project's respective repository. If you have multiple
projects, you might want to arrange them in groups inside the
repository, perhaps putting projects with similar goals or
shared code in the same subdirectory, or maybe just grouping
them alphabetically. Such an arrangement might look
like this:</p><div class="informalexample"><div class="literallayout"><p><br></br>
/<br></br>
utils/<br></br>
calc/<br></br>
trunk/<br></br>
tags/<br></br>
branches/<br></br>
calendar/<br></br>
trunk/<br></br>
tags/<br></br>
branches/<br></br>
…<br></br>
office/<br></br>
spreadsheet/<br></br>
trunk/<br></br>
tags/<br></br>
branches/<br></br>
…<br></br>
</p></div></div><p>Lay out your repository in whatever way you see fit.
Subversion does not expect or enforce a particular layout—in
its eyes, a directory is a directory is a directory.
Ultimately, you should choose the repository arrangement that
meets the needs of the people who work on the projects that
live there.</p><p>In the name of full disclosure, though, we'll mention
another very common layout. In this layout, the
<code class="filename">trunk</code>, <code class="filename">tags</code>, and
<code class="filename">branches</code> directories live in the root
directory of your repository, and your projects are in
subdirectories beneath those, like so:</p><div class="informalexample"><div class="literallayout"><p><br></br>
/<br></br>
trunk/<br></br>
calc/<br></br>
calendar/<br></br>
spreadsheet/<br></br>
…<br></br>
tags/<br></br>
calc/<br></br>
calendar/<br></br>
spreadsheet/<br></br>
…<br></br>
branches/<br></br>
calc/<br></br>
calendar/<br></br>
spreadsheet/<br></br>
…<br></br>
</p></div></div><p>There's nothing particularly incorrect about such a
layout, but it may or may not seem as intuitive for your
users. Especially in large, multiproject situations with
many users, those users may tend to be familiar with only one
or two of the projects in the repository. But the
projects-as-branch-siblings approach tends to deemphasize project
individuality and focus on the entire set of projects as a
single entity. That's a social issue, though. We like our
originally suggested arrangement for purely practical
reasons—it's easier to ask about (or modify, or migrate
elsewhere) the entire history of a single project when there's
a single repository path that holds the entire
history—past, present, tagged, and branched—for
that project and that project alone.</p></div><div class="sect2" title="Deciding Where and How to Host Your Repository"><div class="titlepage"><div><div><h3 class="title"><a id="svn.reposadmin.basics.hosting"></a>Deciding Where and How to Host Your Repository</h3></div></div></div><p>Before creating your Subversion repository, an obvious
question you'll need to answer is where the thing is going to
live. This is strongly connected to myriad other
questions involving how the repository will be accessed (via a
Subversion server or directly), by whom (users behind your
corporate firewall or the whole world out on the open
Internet), what other services you'll be providing around
Subversion (repository browsing interfaces, email-based
commit notification, etc.), your data backup strategy, and so
on.</p><p>We cover server choice and configuration in <a class="xref" href="#svn.serverconfig" title="Chapter 6. Server Configuration">Chapter 6, <em>Server Configuration</em></a>, but the point we'd like to
briefly make here is simply that the answers to some of these
other questions might have implications that force your hand
when deciding where your repository will live. For example,
certain deployment scenarios might require accessing the
repository via a remote filesystem from multiple computers, in
which case (as you'll read in the next section) your choice of
a repository backend data store turns out not to be a choice
at all because only one of the available backends will work
in this scenario.</p><p>Addressing each possible way to deploy
Subversion is both impossible and outside the scope of this
book. We simply encourage you to evaluate your options using
these pages and other sources as your reference material and to
plan ahead.</p></div><div class="sect2" title="Choosing a Data Store"><div class="titlepage"><div><div><h3 class="title"><a id="svn.reposadmin.basics.backends"></a>Choosing a Data Store</h3></div></div></div><p>Subversion provides two options for the
type of underlying data store—often referred to as
<span class="quote">“<span class="quote">the backend</span>”</span> or, somewhat confusingly,
<span class="quote">“<span class="quote">the (versioned) filesystem</span>”</span>—that each
repository uses. One type of data store keeps everything in a
Berkeley DB (or BDB) database environment; repositories that
use this type are often referred to as being
<span class="quote">“<span class="quote">BDB-backed.</span>”</span> The other type stores data in
ordinary flat files, using a custom format. Subversion
developers have adopted the habit of referring to this latter
data storage mechanism
as <em class="firstterm">FSFS</em><sup>[<a id="idp11714608" href="#ftn.idp11714608" class="footnote">41</a>]</sup>—a
versioned filesystem implementation that uses the native OS
filesystem directly—rather than via a database library
or some other abstraction layer—to store data.</p><p><a class="xref" href="#svn.reposadmin.basics.backends.tbl-1" title="Table 5.1. Repository data store comparison">Table 5.1, “Repository data store comparison”</a>
gives a comparative overview of Berkeley DB and FSFS
repositories.</p><div class="table"><a id="svn.reposadmin.basics.backends.tbl-1"></a><p class="title"><strong>Table 5.1. Repository data store comparison</strong></p><div class="table-contents"><table summary="Repository data store comparison" border="1"><colgroup><col></col><col></col><col></col><col></col></colgroup><thead><tr><th>Category</th><th>Feature</th><th>Berkeley DB</th><th>FSFS</th></tr></thead><tbody><tr><td rowspan="2">Reliability</td><td>Data integrity</td><td>When properly deployed, extremely reliable;
Berkeley DB 4.4 brings auto-recovery</td><td>Older versions had some rarely demonstrated, but
data-destroying bugs</td></tr><tr><td>Sensitivity to interruptions</td><td>Very; crashes and permission problems can leave the
database <span class="quote">“<span class="quote">wedged,</span>”</span> requiring journaled
recovery procedures</td><td>Quite insensitive</td></tr><tr><td rowspan="4">Accessibility</td><td>Usable from a read-only mount</td><td>No</td><td>Yes</td></tr><tr><td>Platform-independent storage</td><td>No</td><td>Yes</td></tr><tr><td>Usable over network filesystems</td><td>Generally, no</td><td>Yes</td></tr><tr><td>Group permissions handling</td><td>Sensitive to user umask problems; best if accessed
by only one user</td><td>Works around umask problems</td></tr><tr><td rowspan="3">Scalability</td><td>Repository disk usage</td><td>Larger (especially if logfiles aren't purged)</td><td>Smaller</td></tr><tr><td>Number of revision trees</td><td>Database; no problems</td><td>Some older native filesystems don't scale well with
thousands of entries in a single directory</td></tr><tr><td>Directories with many files</td><td>Slower</td><td>Faster</td></tr><tr><td rowspan="2">Performance</td><td>Checking out latest revision</td><td>No meaningful difference</td><td>No meaningful difference</td></tr><tr><td>Large commits</td><td>Slower overall, but cost is amortized across the
lifetime of the commit</td><td>Faster overall, but finalization delay may cause
client timeouts</td></tr></tbody></table></div></div><br class="table-break"></br><p>There are advantages and disadvantages to each of these
two backend types. Neither of them is more
<span class="quote">“<span class="quote">official</span>”</span> than the other, though the newer FSFS
is the default data store as of Subversion 1.2. Both are
reliable enough to trust with your versioned data. But as you
can see in <a class="xref" href="#svn.reposadmin.basics.backends.tbl-1" title="Table 5.1. Repository data store comparison">Table 5.1, “Repository data store comparison”</a>, the FSFS
backend provides quite a bit more flexibility in terms of its
supported deployment scenarios. More flexibility means you
have to work a little harder to find ways to deploy it
incorrectly. Those reasons—plus the fact that not using
Berkeley DB means there's one fewer component in the
system—largely explain why today almost everyone uses
the FSFS backend when creating new repositories.</p><p>Fortunately, most programs that access Subversion
repositories are blissfully ignorant of which backend data
store is in use. And you aren't even necessarily stuck with
your first choice of a data store—in the event that you
change your mind later, Subversion provides ways of migrating
your repository's data into another repository that uses a
different backend data store. We talk more about that later
in this chapter.</p><p>The following subsections provide a more detailed look at
the available backend data store types.</p><div class="sect3" title="Berkeley DB"><div class="titlepage"><div><div><h4 class="title"><a id="svn.reposadmin.basics.backends.bdb"></a>Berkeley DB</h4></div></div></div><p>When the initial design phase of Subversion was in
progress, the developers decided to use Berkeley DB for a
variety of reasons, including its open source license,
transaction support, reliability, performance, API
simplicity, thread safety, support for cursors, and so
on.</p><p>Berkeley DB provides real transaction
support—perhaps its most powerful feature. Multiple
processes accessing your Subversion repositories don't have
to worry about accidentally clobbering each other's data.
The isolation provided by the transaction system is such
that for any given operation, the Subversion repository code
sees a static view of the database—not a database that
is constantly changing at the hand of some other
process—and can make decisions based on that view. If
the decision made happens to conflict with what another
process is doing, the entire operation is rolled back as though
it never happened, and Subversion gracefully retries the
operation against a new, updated (and yet still static) view
of the database.</p><p>Another great feature of Berkeley DB is <em class="firstterm">hot
backups</em>—the ability to back up the
database environment without taking it
<span class="quote">“<span class="quote">offline.</span>”</span> We'll discuss how to back up your
repository later in this chapter (in <a class="xref" href="#svn.reposadmin.maint.backup" title="Repository Backup">the section called “Repository Backup”</a>), but the benefits
of being able to make fully functional copies of your
repositories without any downtime should be obvious.</p><p>Berkeley DB is also a very reliable database system when
properly used. Subversion uses Berkeley DB's logging
facilities, which means that the database first writes to
on-disk logfiles a description of any modifications it is
about to make, and then makes the modification itself. This
is to ensure that if anything goes wrong, the database
system can back up to a previous
<em class="firstterm">checkpoint</em>—a location in the
logfiles known not to be corrupt—and replay
transactions until the data is restored to a usable state.
See <a class="xref" href="#svn.reposadmin.maint.diskspace" title="Managing Disk Space">the section called “Managing Disk Space”</a> later
in this chapter for more about Berkeley DB logfiles.</p><p>But every rose has its thorn, and so we must note some
known limitations of Berkeley DB. First, Berkeley DB
environments are not portable. You cannot simply copy a
Subversion repository that was created on a Unix system onto
a Windows system and expect it to work. While much of the
Berkeley DB database format is architecture-independent,
other aspects of the environment are not.
Second, Subversion uses Berkeley DB in a way that will not
operate on Windows 95/98 systems—if you need to house
a BDB-backed repository on a Windows machine, stick with
Windows 2000 or later.</p><p>While Berkeley DB promises to behave correctly on
network shares that meet a particular set of
specifications,<sup>[<a id="idp11756656" href="#ftn.idp11756656" class="footnote">42</a>]</sup> most
networked filesystem types and appliances do
<span class="emphasis"><em>not</em></span> actually meet those requirements.
And in no case can you allow a BDB-backed repository that
resides on a network share to be accessed by multiple
clients of that share at once (which quite often is the
whole point of having the repository live on a network share
in the first place).</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>If you attempt to use Berkeley DB on a noncompliant
remote filesystem, the results are unpredictable—you
may see mysterious errors right away, or it may be months
before you discover that your repository database is
subtly corrupted. You should strongly consider using the
FSFS data store for repositories that need to live on a
network share.</p></td></tr></table></div><p>Finally, because Berkeley DB is a library linked
directly into Subversion, it's more sensitive to
interruptions than a typical relational database system.
Most SQL systems, for example, have a dedicated server
process that mediates all access to tables. If a program
accessing the database crashes for some reason, the database
daemon notices the lost connection and cleans up any mess
left behind. And because the database daemon is the only
process accessing the tables, applications don't need to
worry about permission conflicts. These things are not the
case with Berkeley DB, however. Subversion (and programs
using Subversion libraries) access the database tables
directly, which means that a program crash can leave the
database in a temporarily inconsistent, inaccessible state.
When this happens, an administrator needs to ask Berkeley DB
to restore to a checkpoint, which is a bit of an annoyance.
Other things can cause a repository to <span class="quote">“<span class="quote">wedge</span>”</span>
besides crashed processes, such as programs conflicting over
ownership and permissions on the database files.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>Berkeley DB 4.4 brings (to Subversion 1.4 and later)
the ability for Subversion to automatically and
transparently recover Berkeley DB environments in need of
such recovery. When a Subversion process attaches to a
repository's Berkeley DB environment, it uses some process
accounting mechanisms to detect any unclean disconnections
by previous processes, performs any necessary recovery,
and then continues on as though nothing happened. This
doesn't completely eliminate instances of repository
wedging, but it does drastically reduce the amount of
human interaction required to recover from them.</p></td></tr></table></div><p>So while a Berkeley DB repository is quite fast and
scalable, it's best used by a single server process running
as one user—such as Apache's <span class="command"><strong>httpd</strong></span>
or <span class="command"><strong>svnserve</strong></span> (see <a class="xref" href="#svn.serverconfig" title="Chapter 6. Server Configuration">Chapter 6, <em>Server Configuration</em></a>)—rather than accessing it
as many different users via <code class="literal">file://</code> or
<code class="literal">svn+ssh://</code> URLs. If you're accessing a Berkeley
DB repository directly as multiple users, be sure to read
<a class="xref" href="#svn.serverconfig.multimethod" title="Supporting Multiple Repository Access Methods">the section called “Supporting Multiple Repository Access Methods”</a> later in this
chapter.</p></div><div class="sect3" title="FSFS"><div class="titlepage"><div><div><h4 class="title"><a id="svn.reposadmin.basics.backends.fsfs"></a>FSFS</h4></div></div></div><p>In mid-2004, a second type of repository storage
system—one that doesn't use a database at
all—came into being. An FSFS repository stores the
changes associated with a revision in a single file, and so
all of a repository's revisions can be found in a single
subdirectory full of numbered files. Transactions are
created in separate subdirectories as individual files.
When complete, the transaction file is renamed and moved
into the revisions directory, thus guaranteeing that commits
are atomic. And because a revision file is permanent and
unchanging, the repository also can be backed up while
<span class="quote">“<span class="quote">hot,</span>”</span> just like a BDB-backed
repository.</p><div class="sidebar" title="Revision files and shards"><a id="svn.reposadmin.basics.backends.fsfs.revfiles"></a><div class="titlepage"><div><div><p class="title"><strong>Revision files and shards</strong></p></div></div></div><p>FSFS repositories contain files that describe the
changes made in a single revision, and files that contain
the revision properties associated with a single revision.
Repositories created in versions of Subversion prior to
1.5 keep these files in two directories—one for each
type of file. As new revisions are committed to the
repository, Subversion drops more files into these two
directories—over time, the number of these files in
each directory can grow to be quite large. This has been
observed to cause performance problems on certain
network-based filesystems.</p><p>Subversion 1.5 creates FSFS-backed repositories using
a slightly modified layout in which the contents of these
two directories are <em class="firstterm">sharded</em>, or
scattered across several subdirectories. This can greatly
reduce the time it takes the system to locate any one of
these files, and therefore increases the overall
performance of Subversion when reading from the
repository.</p><p>Subversion 1.6 and later takes the sharded layout one
step further, allowing administrators to
optionally <em class="firstterm">pack</em> each of their
repository shards up into a single multi-revision file.
This can have both performance and disk usage benefits.
See
<a class="xref" href="#svn.reposadmin.maint.diskspace.fsfspacking" title="Packing FSFS filesystems">the section called “Packing FSFS filesystems”</a>
for more information.</p></div><p>The FSFS revision files describe a revision's
directory structure, file contents, and deltas against files
in other revision trees. Unlike a Berkeley DB database,
this storage format is portable across different operating
systems and isn't sensitive to CPU architecture. Because
no journaling or shared-memory files are being used, the
repository can be safely accessed over a network filesystem
and examined in a read-only environment. The lack of
database overhead also means the overall repository
size is a bit smaller.</p><p>FSFS has different performance characteristics, too.
When committing a directory with a huge number of files,
FSFS is able to more quickly append directory entries. On
the other hand, FSFS has a longer delay when finalizing a
commit while it performs tasks that the BDB backend
amortizes across the lifetime of the commit, which could in
extreme cases cause clients to time out while waiting for a
response.</p><p>The most important distinction, however, is FSFS's
imperviousness to wedging when something goes wrong. If a
process using a Berkeley DB database runs into a permissions
problem or suddenly crashes, the database can be left in an
unusable state until an administrator recovers it. If the
same scenarios happen to a process using an FSFS repository,
the repository isn't affected at all. At worst, some
transaction data is left behind.</p></div></div></div><div class="sect1" title="Creating and Configuring Your Repository"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.reposadmin.create"></a>Creating and Configuring Your Repository</h2></div></div></div><p>Earlier in this chapter (in
<a class="xref" href="#svn.reposadmin.planning" title="Strategies for Repository Deployment">the section called “Strategies for Repository Deployment”</a>), we looked at some
of the important decisions that should be made before creating
and configuring your Subversion repository. Now, we finally get
to get our hands dirty! In this section, we'll see how to
actually create a Subversion repository and configure it to
perform custom actions when special repository events
occur.</p><div class="sect2" title="Creating the Repository"><div class="titlepage"><div><div><h3 class="title"><a id="svn.reposadmin.basics.creating"></a>Creating the Repository</h3></div></div></div><p>Subversion repository creation is an incredibly simple
task. The <span class="command"><strong>svnadmin</strong></span> utility that comes with
Subversion provides a subcommand (<span class="command"><strong>svnadmin
create</strong></span>) for doing just that.</p><div class="informalexample"><pre class="screen">
$ # Create a repository
$ svnadmin create /var/svn/repos
$
</pre></div><p>Assuming that the parent directory
<code class="filename">/var/svn</code> exists and that you have
sufficient permissions to modify that directory, the previous
command creates a new repository in the directory
<code class="filename">/var/svn/repos</code>, and with the default
filesystem data store (FSFS). You can explicitly choose the
filesystem type using the <code class="option">--fs-type</code> argument,
which accepts as a parameter either <code class="literal">fsfs</code> or
<code class="literal">bdb</code>.</p><div class="informalexample"><pre class="screen">
$ # Create an FSFS-backed repository
$ svnadmin create --fs-type fsfs /var/svn/repos
$
</pre></div><div class="informalexample"><pre class="screen">
# Create a Berkeley-DB-backed repository
$ svnadmin create --fs-type bdb /var/svn/repos
$
</pre></div><p>After running this simple command, you have a Subversion
repository. Depending on how users will access this new
repository, you might need to fiddle with its filesystem
permissions. But since basic system administration is rather
outside the scope of this text, we'll leave further
exploration of that topic as an exercise to the reader.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>The path argument to <span class="command"><strong>svnadmin</strong></span> is just
a regular filesystem path and not a URL like the
<span class="command"><strong>svn</strong></span> client program uses when referring to
repositories. Both <span class="command"><strong>svnadmin</strong></span> and
<span class="command"><strong>svnlook</strong></span> are considered server-side
utilities—they are used on the machine where the
repository resides to examine or modify aspects of the
repository, and are in fact unable to perform tasks across a
network. A common mistake made by Subversion newcomers is
trying to pass URLs (even <span class="quote">“<span class="quote">local</span>”</span>
<code class="literal">file://</code> ones) to these two programs.</p></td></tr></table></div><p>Present in the <code class="filename">db/</code> subdirectory of
your repository is the implementation of the versioned
filesystem. Your new repository's versioned filesystem begins
life at revision 0, which is defined to consist of nothing but
the top-level root (<code class="filename">/</code>) directory.
Initially, revision 0 also has a single revision property,
<code class="literal">svn:date</code>, set to the time at which the
repository was created.</p><p>Now that you have a repository, it's time to customize
it.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>While some parts of a Subversion repository—such
as the configuration files and hook scripts—are meant
to be examined and modified manually, you shouldn't (and
shouldn't need to) tamper with the other parts of the
repository <span class="quote">“<span class="quote">by hand.</span>”</span> The
<span class="command"><strong>svnadmin</strong></span> tool should be sufficient for
any changes necessary to your repository, or you can look to
third-party tools (such as Berkeley DB's tool suite) for
tweaking relevant subsections of the repository. Do
<span class="emphasis"><em>not</em></span> attempt manual manipulation of your
version control history by poking and prodding around in
your repository's data store files!</p></td></tr></table></div></div><div class="sect2" title="Implementing Repository Hooks"><div class="titlepage"><div><div><h3 class="title"><a id="svn.reposadmin.create.hooks"></a>Implementing Repository Hooks</h3></div></div></div><p>A <em class="firstterm">hook</em> is a program triggered by
some repository event, such as the creation of a new revision
or the modification of an unversioned property. Some hooks
(the so-called <span class="quote">“<span class="quote">pre hooks</span>”</span>) run in advance of a
repository operation and provide a means by which to both
report what is about to happen and prevent it from
happening at all. Other hooks (the <span class="quote">“<span class="quote">post hooks</span>”</span>)
run after the completion of a repository event and are useful
for performing tasks that examine—but don't
modify—the repository. Each hook is handed enough
information to tell what that event is (or was), the specific
repository changes proposed (or completed), and the username
of the person who triggered the event.</p><p>The <code class="filename">hooks</code> subdirectory is, by
default, filled with templates for various repository
hooks:</p><div class="informalexample"><pre class="screen">
$ ls repos/hooks/
post-commit.tmpl post-unlock.tmpl pre-revprop-change.tmpl
post-lock.tmpl pre-commit.tmpl pre-unlock.tmpl
post-revprop-change.tmpl pre-lock.tmpl start-commit.tmpl
$
</pre></div><p>There is one template for each hook that the Subversion
repository supports; by examining the contents of those
template scripts, you can see what triggers each script
to run and what data is passed to that script. Also present
in many of these templates are examples of how one might use
that script, in conjunction with other Subversion-supplied
programs, to perform common useful tasks. To actually install
a working hook, you need only place some executable program or
script into the <code class="filename">repos/hooks</code> directory,
which can be executed as the name (such as
<span class="command"><strong>start-commit</strong></span> or
<span class="command"><strong>post-commit</strong></span>) of the hook.</p><p>On Unix platforms, this means supplying a script or
program (which could be a shell script, a Python program, a
compiled C binary, or any number of other things) named
exactly like the name of the hook. Of course, the template
files are present for more than just informational
purposes—the easiest way to install a hook on Unix
platforms is to simply copy the appropriate template file to a
new file that lacks the <code class="filename">.tmpl</code> extension,
customize the hook's contents, and ensure that the script is
executable. Windows, however, uses file extensions to
determine whether a program is executable, so you would
need to supply a program whose basename is the name of the
hook and whose extension is one of the special extensions
recognized by Windows for executable programs, such as
<code class="filename">.exe</code> for programs and
<code class="filename">.bat</code> for batch files.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>For security reasons, the Subversion repository executes
hook programs with an empty environment—that is, no
environment variables are set at all, not even
<code class="literal">$PATH</code> (or <code class="literal">%PATH%</code>,
under Windows). Because of this, many administrators
are baffled when their hook program runs fine by hand, but
doesn't work when run by Subversion. Be sure to explicitly
set any necessary environment variables in your hook program
and/or use absolute paths to programs.</p></td></tr></table></div><p>Subversion executes hooks as the same user who owns the
process that is accessing the Subversion repository. In most
cases, the repository is being accessed via a Subversion
server, so this user is the same user as whom the server
runs on the system. The hooks themselves will need to be
configured with OS-level permissions that allow that user to
execute them. Also, this means that any programs or files
(including the Subversion repository) accessed directly
or indirectly by the hook will be accessed as the same user.
In other words, be alert to potential permission-related
problems that could prevent the hook from performing the tasks
it is designed to perform.</p><p>There are several hooks implemented by the Subversion
repository, and you can get details about each of them in
<a class="xref" href="#svn.ref.reposhooks" title="Repository Hooks">the section called “Repository Hooks”</a> in
<a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a>. As a repository
administrator, you'll need to decide which hooks you wish
to implement (by way of providing an appropriately named and
permissioned hook program), and how. When you make this
decision, keep in mind
the big picture of how your repository is deployed.
For example, if you are using server configuration
to determine which users are permitted to commit
changes to your repository, you don't need to do this
sort of access control via the hook system.</p><p>There is no shortage of Subversion hook programs and
scripts that are freely available either from the Subversion community
itself or elsewhere. These scripts cover a wide range of
utility—basic access control, policy adherence checking,
issue tracker integration, email- or syndication-based commit
notification, and beyond. Or, if you wish to write your own,
see <a class="xref" href="#svn.developer" title="Chapter 8. Embedding Subversion">Chapter 8, <em>Embedding Subversion</em></a>.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>While hook scripts can do almost
anything, there is one dimension in which hook script
authors should show restraint: do <span class="emphasis"><em>not</em></span>
modify a commit transaction using hook scripts. While it
might be tempting to use hook scripts to automatically
correct errors, shortcomings, or policy violations present
in the files being committed, doing so can cause problems.
Subversion keeps client-side caches of certain bits of
repository data, and if you change a commit transaction in
this way, those caches become indetectably stale. This
inconsistency can lead to surprising and unexpected
behavior. Instead of modifying the transaction, you should
simply <span class="emphasis"><em>validate</em></span> the transaction in the
<code class="filename">pre-commit</code> hook and reject the commit
if it does not meet the desired requirements. As a
bonus, your users will learn the value of careful,
compliance-minded work habits.</p></td></tr></table></div></div><div class="sect2" title="Berkeley DB Configuration"><div class="titlepage"><div><div><h3 class="title"><a id="svn.reposadmin.create.bdb"></a>Berkeley DB Configuration</h3></div></div></div><p>A Berkeley DB environment is an encapsulation of one or
more databases, logfiles, region files, and configuration
files. The Berkeley DB environment has its own set of default
configuration values for things such as the number of database
locks allowed to be taken out at any given time, the maximum
size of the journaling logfiles, and so on. Subversion's
filesystem logic additionally chooses default values for some
of the Berkeley DB configuration options. However, sometimes
your particular repository, with its unique collection of data
and access patterns, might require a different set of
configuration option values.</p><p>The producers of Berkeley DB understand that different
applications and database environments have different
requirements, so they have provided a mechanism for overriding
at runtime many of the configuration values for the Berkeley
DB environment. BDB checks for the presence of a file named
<code class="filename">DB_CONFIG</code> in the environment directory
(namely, the repository's <code class="filename">db</code>
subdirectory), and parses the options found in that file.
Subversion itself creates this file when it creates the rest
of the repository. The file initially contains some default
options, as well as pointers to the Berkeley DB online
documentation so that you can read about what those options do. Of
course, you are free to add any of the supported Berkeley DB
options to your <code class="filename">DB_CONFIG</code> file. Just be
aware that while Subversion never attempts to read or
interpret the contents of the file and makes no direct use of
the option settings in it, you'll want to avoid any
configuration changes that may cause Berkeley DB to behave in
a fashion that is at odds with what Subversion might expect.
Also, changes made to <code class="filename">DB_CONFIG</code> won't
take effect until you recover the database environment (using
<span class="command"><strong>svnadmin recover</strong></span>).</p></div><div class="sect2" title="FSFS Configuration"><div class="titlepage"><div><div><h3 class="title"><a id="svn.reposadmin.create.fsfs"></a>FSFS Configuration</h3></div></div></div><p>As of Subversion 1.6, FSFS filesystems have several
configurable parameters which an administrator can use to
fine-tune the performance or disk usage of their repositories.
You can find these options—and the documentation for
them—in the <code class="filename">db/fsfs.conf</code> file in
the repository.</p></div></div><div class="sect1" title="Repository Maintenance"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.reposadmin.maint"></a>Repository Maintenance</h2></div></div></div><p>Maintaining a Subversion repository can be daunting, mostly
due to the complexities inherent in systems that have a database
backend. Doing the task well is all about knowing the
tools—what they are, when to use them, and how. This
section will introduce you to the repository administration
tools provided by Subversion and discuss how to wield them to
accomplish tasks such as repository data migration, upgrades,
backups, and cleanups.</p><div class="sect2" title="An Administrator's Toolkit"><div class="titlepage"><div><div><h3 class="title"><a id="svn.reposadmin.maint.tk"></a>An Administrator's Toolkit</h3></div></div></div><p>Subversion provides a handful of utilities useful for
creating, inspecting, modifying, and repairing your repository.
Let's look more closely at each of those tools. Afterward,
we'll briefly examine some of the utilities included in the
Berkeley DB distribution that provide functionality specific
to your repository's database backend not otherwise provided
by Subversion's own tools.</p><div class="sect3" title="svnadmin"><div class="titlepage"><div><div><h4 class="title"><a id="svn.reposadmin.maint.tk.svnadmin"></a>svnadmin</h4></div></div></div><p>The <span class="command"><strong>svnadmin</strong></span> program is the
repository administrator's best friend. Besides providing
the ability to create Subversion repositories, this program
allows you to perform several maintenance operations on
those repositories. The syntax of
<span class="command"><strong>svnadmin</strong></span> is similar to that of other
Subversion command-line programs:</p><div class="informalexample"><pre class="screen">
$ svnadmin help
general usage: svnadmin SUBCOMMAND REPOS_PATH [ARGS & OPTIONS ...]
Type 'svnadmin help <subcommand>' for help on a specific subcommand.
Type 'svnadmin --version' to see the program version and FS modules.
Available subcommands:
crashtest
create
deltify
…
</pre></div><p>Previously in this chapter (in <a class="xref" href="#svn.reposadmin.basics.creating" title="Creating the Repository">the section called “Creating the Repository”</a>), we were
introduced to the <span class="command"><strong>svnadmin create</strong></span>
subcommand. Most of the other <span class="command"><strong>svnadmin</strong></span>
subcommands we will cover later in this chapter. And you
can consult <a class="xref" href="#svn.ref.svnadmin" title="svnadmin—Subversion Repository Administration">the section called “svnadmin—Subversion Repository Administration”</a> in
<a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a> for a full rundown of subcommands
and what each of them offers.</p></div><div class="sect3" title="svnlook"><div class="titlepage"><div><div><h4 class="title"><a id="svn.reposadmin.maint.tk.svnlook"></a>svnlook</h4></div></div></div><p><span class="command"><strong>svnlook</strong></span> is a tool provided by
Subversion for examining the various revisions and
<em class="firstterm">transactions</em> (which are revisions
in the making) in a repository. No part of this program
attempts to change the repository. <span class="command"><strong>svnlook</strong></span>
is typically used by the repository hooks for reporting the
changes that are about to be committed (in the case of the
<span class="command"><strong>pre-commit</strong></span> hook) or that were just
committed (in the case of the <span class="command"><strong>post-commit</strong></span>
hook) to the repository. A repository administrator may use
this tool for diagnostic purposes.</p><p><span class="command"><strong>svnlook</strong></span> has a straightforward
syntax:</p><div class="informalexample"><pre class="screen">
$ svnlook help
general usage: svnlook SUBCOMMAND REPOS_PATH [ARGS & OPTIONS ...]
Note: any subcommand which takes the '--revision' and '--transaction'
options will, if invoked without one of those options, act on
the repository's youngest revision.
Type 'svnlook help <subcommand>' for help on a specific subcommand.
Type 'svnlook --version' to see the program version and FS modules.
…
</pre></div><p>Most of <span class="command"><strong>svnlook</strong></span>'s
subcommands can operate on either a revision or a
transaction tree, printing information about the tree
itself, or how it differs from the previous revision of the
repository. You use the <code class="option">--revision</code>
(<code class="option">-r</code>) and <code class="option">--transaction</code>
(<code class="option">-t</code>) options to specify which revision or
transaction, respectively, to examine. In the absence of
both the <code class="option">--revision</code> (<code class="option">-r</code>)
and <code class="option">--transaction</code> (<code class="option">-t</code>)
options, <span class="command"><strong>svnlook</strong></span> will examine the
youngest (or <code class="literal">HEAD</code>) revision in the
repository. So the following two commands do exactly the
same thing when 19 is the youngest revision in the
repository located at
<code class="filename">/var/svn/repos</code>:</p><div class="informalexample"><pre class="screen">
$ svnlook info /var/svn/repos
$ svnlook info /var/svn/repos -r 19
</pre></div><p>One exception to these rules about subcommands is
the <span class="command"><strong>svnlook youngest</strong></span> subcommand, which
takes no options and simply prints out the repository's
youngest revision number:</p><div class="informalexample"><pre class="screen">
$ svnlook youngest /var/svn/repos
19
$
</pre></div><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>Keep in mind that the only transactions you can browse
are uncommitted ones. Most repositories will have no such
transactions because transactions are usually either
committed (in which case, you should access them as
revision with the <code class="option">--revision</code>
(<code class="option">-r</code>) option) or aborted and
removed.</p></td></tr></table></div><p>Output from <span class="command"><strong>svnlook</strong></span> is designed to be
both human- and machine-parsable. Take, as an example, the
output of the <span class="command"><strong>svnlook info</strong></span> subcommand:</p><div class="informalexample"><pre class="screen">
$ svnlook info /var/svn/repos
sally
2002-11-04 09:29:13 -0600 (Mon, 04 Nov 2002)
27
Added the usual
Greek tree.
$
</pre></div><p>The output of <span class="command"><strong>svnlook info</strong></span> consists
of the following, in the order given:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p>The author, followed by a newline</p></li><li class="listitem"><p>The date, followed by a newline</p></li><li class="listitem"><p>The number of characters in the log message,
followed by a newline</p></li><li class="listitem"><p>The log message itself, followed by a newline</p></li></ol></div><p>This output is human-readable, meaning items such as the
datestamp are displayed using a textual representation
instead of something more obscure (such as the number of
nanoseconds since the Tastee Freez guy drove by). But the
output is also machine-parsable—because the log
message can contain multiple lines and be unbounded in
length, <span class="command"><strong>svnlook</strong></span> provides the length of
that message before the message itself. This allows scripts
and other wrappers around this command to make intelligent
decisions about the log message, such as how much memory to
allocate for the message, or at least how many bytes to skip
in the event that this output is not the last bit of data in
the stream.</p><p><span class="command"><strong>svnlook</strong></span> can perform a variety of
other queries: displaying subsets of bits of information
we've mentioned previously, recursively listing versioned
directory trees, reporting which paths were modified in a
given revision or transaction, showing textual and property
differences made to files and directories, and so on. See
<a class="xref" href="#svn.ref.svnlook" title="svnlook—Subversion Repository Examination">the section called “svnlook—Subversion Repository Examination”</a> in
<a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a> for a full reference of
<span class="command"><strong>svnlook</strong></span>'s features.</p></div><div class="sect3" title="svndumpfilter"><div class="titlepage"><div><div><h4 class="title"><a id="svn.reposadmin.maint.tk.svndumpfilter"></a>svndumpfilter</h4></div></div></div><p>While it won't be the most commonly used tool at the
administrator's disposal, <span class="command"><strong>svndumpfilter</strong></span>
provides a very particular brand of useful
functionality—the ability to quickly and easily modify
streams of Subversion repository history data by acting as a
path-based filter.</p><p>The syntax of <span class="command"><strong>svndumpfilter</strong></span> is as
follows:</p><div class="informalexample"><pre class="screen">
$ svndumpfilter help
general usage: svndumpfilter SUBCOMMAND [ARGS & OPTIONS ...]
Type 'svndumpfilter help <subcommand>' for help on a specific subcommand.
Type 'svndumpfilter --version' to see the program version.
Available subcommands:
exclude
include
help (?, h)
</pre></div><p>There are only two interesting subcommands:
<span class="command"><strong>svndumpfilter exclude</strong></span> and
<span class="command"><strong>svndumpfilter include</strong></span>. They allow you to
make the choice between implicit or explicit inclusion of
paths in the stream. You can learn more about these
subcommands and <span class="command"><strong>svndumpfilter</strong></span>'s unique
purpose later in this chapter, in <a class="xref" href="#svn.reposadmin.maint.filtering" title="Filtering Repository History">the section called “Filtering Repository History”</a>.</p></div><div class="sect3" title="svnrdump"><div class="titlepage"><div><div><h4 class="title"><a id="svn.reposadmin.maint.tk.svnrdump"></a>svnrdump</h4></div></div></div><p>The <span class="command"><strong>svnrdump</strong></span> program is, to put it
simply, essentially just network-aware flavors of
the <span class="command"><strong>svnadmin dump</strong></span> and <span class="command"><strong>svnadmin
load</strong></span> subcommands, rolled up into a separate
program.</p><div class="informalexample"><pre class="screen">
$ svnrdump help
general usage: svnrdump SUBCOMMAND URL [-r LOWER[:UPPER]]
Type 'svnrdump help <subcommand>' for help on a specific subcommand.
Type 'svnrdump --version' to see the program version and RA modules.
Available subcommands:
dump
load
help (?, h)
$
</pre></div><p>We discuss the use of <span class="command"><strong>svnrdump</strong></span> and
the aforementioned <span class="command"><strong>svnadmin</strong></span> commands
later in this chapter (see
<a class="xref" href="#svn.reposadmin.maint.migrate" title="Migrating Repository Data Elsewhere">the section called “Migrating Repository Data Elsewhere”</a>).</p></div><div class="sect3" title="svnsync"><div class="titlepage"><div><div><h4 class="title"><a id="svn.reposadmin.maint.tk.svnsync"></a>svnsync</h4></div></div></div><p>The <span class="command"><strong>svnsync</strong></span> program provides all the
functionality required for maintaining a read-only mirror of
a Subversion repository. The program really has one
job—to transfer one repository's versioned history
into another repository. And while there are few ways to do
that, its primary strength is that it can operate
remotely—the <span class="quote">“<span class="quote">source</span>”</span> and
<span class="quote">“<span class="quote">sink</span>”</span><sup>[<a id="idp11924736" href="#ftn.idp11924736" class="footnote">43</a>]</sup> repositories may
be on different computers from each other and
from <span class="command"><strong>svnsync</strong></span> itself.</p><p>As you might expect, <span class="command"><strong>svnsync</strong></span> has a
syntax that looks very much like every other program we've
mentioned in this chapter:</p><div class="informalexample"><pre class="screen">
$ svnsync help
general usage: svnsync SUBCOMMAND DEST_URL [ARGS & OPTIONS ...]
Type 'svnsync help <subcommand>' for help on a specific subcommand.
Type 'svnsync --version' to see the program version and RA modules.
Available subcommands:
initialize (init)
synchronize (sync)
copy-revprops
info
help (?, h)
$
</pre></div><p>We talk more about replicating repositories with
<span class="command"><strong>svnsync</strong></span> later in this chapter (see <a class="xref" href="#svn.reposadmin.maint.replication" title="Repository Replication">the section called “Repository Replication”</a>).</p></div><div class="sect3" title="fsfs-reshard.py"><div class="titlepage"><div><div><h4 class="title"><a id="svn.reposadmin.maint.tk.fsfsreshard"></a>fsfs-reshard.py</h4></div></div></div><p>While not an official member of the Subversion
toolchain, the <span class="command"><strong>fsfs-reshard.py</strong></span> script
(found in the <code class="filename">tools/server-side</code>
directory of the Subversion source distribution) is a useful
performance tuning tool for administrators of FSFS-backed
Subversion repositories. As described in the sidebar
<a class="xref" href="#svn.reposadmin.basics.backends.fsfs.revfiles" title="Revision files and shards">Revision files and shards</a>,
FSFS repositories use individual files to house information
about each revision. Sometimes these files all live in a
single directory; sometimes they are sharded across many
directories. But the neat thing is that the number of
directories used to house these files is configurable.
That's where <span class="command"><strong>fsfs-reshard.py</strong></span> comes
in.</p><p><span class="command"><strong>fsfs-reshard.py</strong></span> reshuffles the
repository's file structure into a new arrangement that
reflects the requested number of sharding subdirectories and
updates the repository configuration to preserve this
change. When used in conjunction with the <span class="command"><strong>svnadmin
upgrade</strong></span> command, this is especially useful for
upgrading a pre-1.5 Subversion (unsharded) repository to the
latest filesystem format and sharding its data files (which
Subversion will not automatically do for you). This script
can also be used for fine-tuning an already sharded
repository.</p></div><div class="sect3" title="Berkeley DB utilities"><div class="titlepage"><div><div><h4 class="title"><a id="svn.reposadmin.maint.tk.bdbutil"></a>Berkeley DB utilities</h4></div></div></div><p>If you're using a Berkeley DB repository, all of
your versioned filesystem's structure and data live in a set
of database tables within the <code class="filename">db/</code>
subdirectory of your repository. This subdirectory is a
regular Berkeley DB environment directory and can therefore
be used in conjunction with any of the Berkeley database
tools, typically provided as part of the Berkeley DB
distribution.</p><p>For day-to-day Subversion use, these tools are
unnecessary. Most of the functionality typically needed for
Subversion repositories has been duplicated in the
<span class="command"><strong>svnadmin</strong></span> tool. For example,
<span class="command"><strong>svnadmin list-unused-dblogs</strong></span> and
<span class="command"><strong>svnadmin list-dblogs</strong></span> perform a
subset of what is provided by the Berkeley
<span class="command"><strong>db_archive</strong></span> utility, and <span class="command"><strong>svnadmin
recover</strong></span> reflects the common use cases of the
<span class="command"><strong>db_recover</strong></span> utility.</p><p>However, there are still a few Berkeley DB utilities
that you might find useful. The <span class="command"><strong>db_dump</strong></span>
and <span class="command"><strong>db_load</strong></span> programs write and read,
respectively, a custom file format that describes the keys
and values in a Berkeley DB database. Since Berkeley
databases are not portable across machine architectures,
this format is a useful way to transfer those databases from
machine to machine, irrespective of architecture or
operating system. As we describe later in this chapter, you
can also use <span class="command"><strong>svnadmin dump</strong></span> and
<span class="command"><strong>svnadmin load</strong></span> for similar purposes, but
<span class="command"><strong>db_dump</strong></span> and <span class="command"><strong>db_load</strong></span>
can do certain jobs just as well and much faster. They can
also be useful if the experienced Berkeley DB hacker needs
to do in-place tweaking of the data in a BDB-backed
repository for some reason, which is something Subversion's
utilities won't allow. Also, the <span class="command"><strong>db_stat</strong></span>
utility can provide useful information about the status of
your Berkeley DB environment, including detailed statistics
about the locking and storage subsystems.</p><p>For more information on the Berkeley DB tool chain,
visit the documentation section of the Berkeley DB section
of Oracle's web site, located at <a class="ulink" href="http://www.oracle.com/technology/documentation/berkeley-db/db/" target="_top">http://www.oracle.com/technology/documentation/berkeley-db/db/</a>.</p></div></div><div class="sect2" title="Commit Log Message Correction"><div class="titlepage"><div><div><h3 class="title"><a id="svn.reposadmin.maint.setlog"></a>Commit Log Message Correction</h3></div></div></div><p>Sometimes a user will have an error in her log message (a
misspelling or some misinformation, perhaps). If the
repository is configured (using the
<code class="literal">pre-revprop-change</code> hook; see <a class="xref" href="#svn.reposadmin.create.hooks" title="Implementing Repository Hooks">the section called “Implementing Repository Hooks”</a>) to accept changes to
this log message after the commit is finished, the user
can <span class="quote">“<span class="quote">fix</span>”</span> her log message remotely using
<span class="command"><strong>svn propset</strong></span> (see <a class="xref" href="#svn.ref.svn.c.propset" title="svn propset (pset, ps)">svn propset (pset, ps)</a> in
<a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a>). However, because of the
potential to lose information forever, Subversion repositories
are not, by default, configured to allow changes to
unversioned properties—except by an
administrator.</p><p>If a log message needs to be changed by an administrator,
this can be done using <span class="command"><strong>svnadmin setlog</strong></span>.
This command changes the log message (the
<code class="literal">svn:log</code> property) on a given revision of a
repository, reading the new value from a provided file.</p><div class="informalexample"><pre class="screen">
$ echo "Here is the new, correct log message" > newlog.txt
$ svnadmin setlog myrepos newlog.txt -r 388
</pre></div><p>The <span class="command"><strong>svnadmin setlog</strong></span> command, by
default, is still bound by the same protections against
modifying unversioned properties as a remote client
is—the <code class="literal">pre-revprop-change</code> and
<code class="literal">post-revprop-change</code> hooks are still
triggered, and therefore must be set up to accept changes of
this nature. But an administrator can get around these
protections by passing the <code class="option">--bypass-hooks</code>
option to the <span class="command"><strong>svnadmin setlog</strong></span> command.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>Remember, though, that by bypassing the hooks, you are
likely avoiding such things as email notifications of
property changes, backup systems that track unversioned
property changes, and so on. In other words, be very
careful about what you are changing, and how you change
it.</p></td></tr></table></div></div><div class="sect2" title="Managing Disk Space"><div class="titlepage"><div><div><h3 class="title"><a id="svn.reposadmin.maint.diskspace"></a>Managing Disk Space</h3></div></div></div><p>While the cost of storage has dropped incredibly in the
past few years, disk usage is still a valid concern for
administrators seeking to version large amounts of data.
Every bit of version history information stored in the live
repository needs to be backed up
elsewhere, perhaps multiple times as part of rotating backup
schedules. It is useful to know what pieces of Subversion's
repository data need to remain on the live site, which need to
be backed up, and which can be safely removed.</p><div class="sect3" title="How Subversion saves disk space"><div class="titlepage"><div><div><h4 class="title"><a id="svn.reposadmin.maint.diskspace.deltas"></a>How Subversion saves disk space</h4></div></div></div><p>To keep the repository small, Subversion uses
<em class="firstterm">deltification</em> (or delta-based storage)
within the repository itself. Deltification involves
encoding the representation of a chunk of data as a
collection of differences against some other chunk of data.
If the two pieces of data are very similar, this
deltification results in storage savings for the deltified
chunk—rather than taking up space equal to the size of
the original data, it takes up only enough space to
say, <span class="quote">“<span class="quote">I look just like this other piece of data over
here, except for the following couple of changes.</span>”</span>
The result is that most of the repository data that tends to
be bulky—namely, the contents of versioned
files—is stored at a much smaller size than the
original full-text representation of that data.</p><p>While deltified storage has been a part of Subversion's
design since the very beginning, there have been additional
improvements made over the years. Subversion repositories
created with Subversion 1.4 or later benefit from
compression of the full-text representations of file
contents. Repositories created with Subversion 1.6 or later
further enjoy the disk space savings afforded by
<em class="firstterm">representation sharing</em>, a feature
which allows multiple files or file revisions with identical
file content to refer to a single shared instance of that data
rather than each having their own distinct copy thereof.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>Because all of the data that is subject to
deltification in a BDB-backed repository is stored in a
single Berkeley DB database file, reducing the size of the
stored values will not immediately reduce the size of the
database file itself. Berkeley DB will, however, keep
internal records of unused areas of the database file and
consume those areas first before growing the size of the
database file. So while deltification doesn't produce
immediate space savings, it can drastically slow future
growth of the database.</p></td></tr></table></div></div><div class="sect3" title="Removing dead transactions"><div class="titlepage"><div><div><h4 class="title"><a id="svn.reposadmin.maint.diskspace.deadtxns"></a>Removing dead transactions</h4></div></div></div><p>Though they are uncommon, there are circumstances in
which a Subversion commit process might fail, leaving behind
in the repository the remnants of the revision-to-be that
wasn't—an uncommitted transaction and all the file and
directory changes associated with it. This could happen for
several reasons: perhaps the client operation was
inelegantly terminated by the user, or a network failure
occurred in the middle of an operation.
Regardless of the reason, dead transactions can happen.
They don't do any real harm, other than consuming disk
space. A fastidious administrator may nonetheless wish to
remove them.</p><p>You can use the <span class="command"><strong>svnadmin lstxns</strong></span>
command to list the names of the currently outstanding
transactions:</p><div class="informalexample"><pre class="screen">
$ svnadmin lstxns myrepos
19
3a1
a45
$
</pre></div><p>Each item in the resultant output can then be used with
<span class="command"><strong>svnlook</strong></span> (and its
<code class="option">--transaction</code> (<code class="option">-t</code>) option)
to determine who created the transaction, when it was
created, what types of changes were made in the
transaction—information that is helpful in determining
whether the transaction is a safe candidate for
removal! If you do indeed want to remove a transaction, its
name can be passed to <span class="command"><strong>svnadmin rmtxns</strong></span>,
which will perform the cleanup of the transaction. In fact,
<span class="command"><strong>svnadmin rmtxns</strong></span> can take its input
directly from the output of
<span class="command"><strong>svnadmin lstxns</strong></span>!</p><div class="informalexample"><pre class="screen">
$ svnadmin rmtxns myrepos `svnadmin lstxns myrepos`
$
</pre></div><p>If you use these two subcommands like this, you should
consider making your repository temporarily inaccessible to
clients. That way, no one can begin a legitimate
transaction before you start your cleanup. <a class="xref" href="#svn.reposadmin.maint.diskspace.deadtxns.ex-1" title="Example 5.1. txn-info.sh (reporting outstanding transactions)">Example 5.1, “txn-info.sh (reporting outstanding transactions)”</a>
contains a bit of shell-scripting that can quickly generate
information about each outstanding transaction in your
repository.</p><div class="example"><a id="svn.reposadmin.maint.diskspace.deadtxns.ex-1"></a><p class="title"><strong>Example 5.1. txn-info.sh (reporting outstanding transactions)</strong></p><div class="example-contents"><pre class="programlisting">
#!/bin/sh
### Generate informational output for all outstanding transactions in
### a Subversion repository.
REPOS="${1}"
if [ "x$REPOS" = x ] ; then
echo "usage: $0 REPOS_PATH"
exit
fi
for TXN in `svnadmin lstxns ${REPOS}`; do
echo "---[ Transaction ${TXN} ]-------------------------------------------"
svnlook info "${REPOS}" -t "${TXN}"
done
</pre></div></div><br class="example-break"></br><p>The output of the script is basically a concatenation of
several chunks of <span class="command"><strong>svnlook info</strong></span> output
(see <a class="xref" href="#svn.reposadmin.maint.tk.svnlook" title="svnlook">the section called “svnlook”</a>) and
will look something like this:</p><div class="informalexample"><pre class="screen">
$ txn-info.sh myrepos
---[ Transaction 19 ]-------------------------------------------
sally
2001-09-04 11:57:19 -0500 (Tue, 04 Sep 2001)
0
---[ Transaction 3a1 ]-------------------------------------------
harry
2001-09-10 16:50:30 -0500 (Mon, 10 Sep 2001)
39
Trying to commit over a faulty network.
---[ Transaction a45 ]-------------------------------------------
sally
2001-09-12 11:09:28 -0500 (Wed, 12 Sep 2001)
0
$
</pre></div><p>A long-abandoned transaction usually represents some
sort of failed or interrupted commit. A transaction's
datestamp can provide interesting information—for
example, how likely is it that an operation begun nine
months ago is still active?</p><p>In short, transaction cleanup decisions need not be made
unwisely. Various sources of information—including
Apache's error and access logs, Subversion's operational
logs, Subversion revision history, and so on—can be
employed in the decision-making process. And of course, an
administrator can often simply communicate with a seemingly
dead transaction's owner (via email, e.g.) to verify
that the transaction is, in fact, in a zombie state.</p></div><div class="sect3" title="Purging unused Berkeley DB logfiles"><div class="titlepage"><div><div><h4 class="title"><a id="svn.reposadmin.maint.diskspace.bdblogs"></a>Purging unused Berkeley DB logfiles</h4></div></div></div><p>Until recently, the largest offender of disk space usage
with respect to BDB-backed Subversion repositories were the
logfiles in which Berkeley DB performs its prewrites before
modifying the actual database files. These files capture
all the actions taken along the route of changing the
database from one state to another—while the database
files, at any given time, reflect a particular state, the
logfiles contain all of the many changes along the way
<span class="emphasis"><em>between</em></span> states. Thus, they can grow
and accumulate quite rapidly.</p><p>Fortunately, beginning with the 4.2 release of Berkeley
DB, the database environment has the ability to remove its
own unused logfiles automatically. Any
repositories created using <span class="command"><strong>svnadmin</strong></span>
when compiled against Berkeley DB version 4.2 or later
will be configured for this automatic logfile removal. If
you don't want this feature enabled, simply pass the
<code class="option">--bdb-log-keep</code> option to the
<span class="command"><strong>svnadmin create</strong></span> command. If you forget
to do this or change your mind at a later time, simply edit
the <code class="filename">DB_CONFIG</code> file found in your
repository's <code class="filename">db</code> directory, comment out
the line that contains the <code class="literal">set_flags
DB_LOG_AUTOREMOVE</code> directive, and then run
<span class="command"><strong>svnadmin recover</strong></span> on your repository to
force the configuration changes to take effect. See <a class="xref" href="#svn.reposadmin.create.bdb" title="Berkeley DB Configuration">the section called “Berkeley DB Configuration”</a> for more information about
database configuration.</p><p>Without some sort of automatic logfile removal in
place, logfiles will accumulate as you use your repository.
This is actually somewhat of a feature of the database
system—you should be able to recreate your entire
database using nothing but the logfiles, so these files can
be useful for catastrophic database recovery. But
typically, you'll want to archive the logfiles that are no
longer in use by Berkeley DB, and then remove them from disk
to conserve space. Use the <span class="command"><strong>svnadmin
list-unused-dblogs</strong></span> command to list the unused
logfiles:</p><div class="informalexample"><pre class="screen">
$ svnadmin list-unused-dblogs /var/svn/repos
/var/svn/repos/log.0000000031
/var/svn/repos/log.0000000032
/var/svn/repos/log.0000000033
…
$ rm `svnadmin list-unused-dblogs /var/svn/repos`
## disk space reclaimed!
</pre></div><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>BDB-backed repositories whose logfiles are used as
part of a backup or disaster recovery plan should
<span class="emphasis"><em>not</em></span> make use of the logfile
autoremoval feature. Reconstruction of a repository's
data from logfiles can only be accomplished only when
<span class="emphasis"><em>all</em></span> the logfiles are available. If
some of the logfiles are removed from disk before the
backup system has a chance to copy them elsewhere, the
incomplete set of backed-up logfiles is essentially
useless.</p></td></tr></table></div></div><div class="sect3" title="Packing FSFS filesystems"><div class="titlepage"><div><div><h4 class="title"><a id="svn.reposadmin.maint.diskspace.fsfspacking"></a>Packing FSFS filesystems</h4></div></div></div><p>As described in the sidebar
<a class="xref" href="#svn.reposadmin.basics.backends.fsfs.revfiles" title="Revision files and shards">Revision files and shards</a>,
FSFS-backed Subversion repositories create, by default, a
new on-disk file for each revision added to the repository.
Having thousands of these files present on your Subversion
server—even when housed in separate shard
directories—can lead to inefficiencies.</p><p>The first problem is that the operating system has to
reference many different files over a short period of time.
This leads to inefficient use of disk caches and, as a
result, more time spent seeking across large disks. Because
of this, Subversion pays a performance penalty when
accessing your versioned data.</p><p>The second problem is a bit more subtle. Because of the
ways that most filesystems allocate disk space, each file
claims more space on the disk than it actually uses. The
amount of extra space required to house a single file can
average anywhere from 2 to 16 kilobytes <span class="emphasis"><em>per
file</em></span>, depending on the underlying
filesystem in use. This translates directly
into a per-revision disk usage penalty for FSFS-backed
repositories. The effect is most pronounced in repositories
which have many small revisions, since the overhead involved
in storing the revision file quickly outgrows the size of
the actual data being stored.</p><p>To solve these problems, Subversion 1.6 introduced the
<span class="command"><strong>svnadmin pack</strong></span> command. By concatenating
all the files of a completed shard into a single <span class="quote">“<span class="quote">pack</span>”</span> file
and then removing the original per-revision
files, <span class="command"><strong>svnadmin pack</strong></span> reduces the file
count within a given shard down to just a single file. In
doing so, it aids filesystem caches and reduces (to one) the
number of times a file storage overhead penalty is
paid.</p><p>Subversion can pack existing sharded repositories which
have been upgraded to the 1.6 filesystem format or later (see
<a class="xref" href="#svn.ref.svnadmin.c.upgrade" title="svnadmin upgrade">svnadmin upgrade</a>) in
<a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a>. To do so, just
run <span class="command"><strong>svnadmin pack</strong></span> on the
repository:</p><div class="informalexample"><pre class="screen">
$ svnadmin pack /var/svn/repos
Packing shard 0...done.
Packing shard 1...done.
Packing shard 2...done.
…
Packing shard 34...done.
Packing shard 35...done.
Packing shard 36...done.
$
</pre></div><p>Because the packing process obtains the required locks
before doing its work, you can run it on live repositories,
or even as part of a post-commit hook. Repacking packed
shards is legal, but will have no effect on the disk usage
of the repository.</p><p><span class="command"><strong>svnadmin pack</strong></span> has no effect on
BDB-backed Subversion repositories.</p></div></div><div class="sect2" title="Berkeley DB Recovery"><div class="titlepage"><div><div><h3 class="title"><a id="svn.reposadmin.maint.recovery"></a>Berkeley DB Recovery</h3></div></div></div><p>As mentioned in <a class="xref" href="#svn.reposadmin.basics.backends.bdb" title="Berkeley DB">the section called “Berkeley DB”</a>, a Berkeley DB
repository can sometimes be left in a frozen state if not closed
properly. When this happens, an administrator needs to rewind
the database back into a consistent state. This is unique to
BDB-backed repositories, though—if you are using
FSFS-backed ones instead, this won't apply to you. And for
those of you using Subversion 1.4 with Berkeley DB 4.4 or
later, you should find that Subversion has become much more
resilient in these types of situations. Still, wedged
Berkeley DB repositories do occur, and an administrator needs
to know how to safely deal with this circumstance.</p><p>To protect the data in your repository, Berkeley
DB uses a locking mechanism. This mechanism ensures that
portions of the database are not simultaneously modified by
multiple database accessors, and that each process sees the
data in the correct state when that data is being read from
the database. When a process needs to change something in the
database, it first checks for the existence of a lock on the
target data. If the data is not locked, the process locks the
data, makes the change it wants to make, and then unlocks the
data. Other processes are forced to wait until that lock is
removed before they are permitted to continue accessing that
section of the database. (This has nothing to do with the
locks that you, as a user, can apply to versioned files within
the repository; we try to clear up the confusion caused by
this terminology collision in the sidebar <a class="xref" href="#svn.advanced.locking.meanings" title="The Three Meanings of “Lock”">The Three Meanings of <span class="quote">“<span class="quote">Lock</span>”</span></a>.)</p><p>In the course of using your Subversion repository, fatal
errors or interruptions can prevent a process from having the
chance to remove the locks it has placed in the database. The
result is that the backend database system gets
<span class="quote">“<span class="quote">wedged.</span>”</span> When this happens, any attempts to
access the repository hang indefinitely (since each new
accessor is waiting for a lock to go away—which isn't
going to happen).</p><p>If this happens to your repository, don't panic. The
Berkeley DB filesystem takes advantage of database
transactions, checkpoints, and prewrite journaling to ensure
that only the most catastrophic of events<sup>[<a id="idp12036112" href="#ftn.idp12036112" class="footnote">44</a>]</sup> can permanently destroy a database
environment. A sufficiently paranoid repository administrator
will have made off-site backups of the repository data in some
fashion, but don't head off to the tape backup storage closet
just yet.</p><p>Instead, use the following recipe to attempt to
<span class="quote">“<span class="quote">unwedge</span>”</span> your repository:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p>Make sure no processes are accessing (or
attempting to access) the repository. For networked
repositories, this also means shutting down the Apache HTTP
Server or svnserve daemon.</p></li><li class="listitem"><p>Become the user who owns and manages the repository.
This is important, as recovering a repository while
running as the wrong user can tweak the permissions of the
repository's files in such a way that your repository will
still be inaccessible even after it is
<span class="quote">“<span class="quote">unwedged.</span>”</span></p></li><li class="listitem"><p>Run the command <strong class="userinput"><code>svnadmin recover
/var/svn/repos</code></strong>. You should see output such as
this:</p><div class="informalexample"><pre class="screen">
Repository lock acquired.
Please wait; recovering the repository may take some time...
Recovery completed.
The latest repos revision is 19.
</pre></div><p>This command may take many minutes to complete.</p></li><li class="listitem"><p>Restart the server process.</p></li></ol></div><p>This procedure fixes almost every case of repository
wedging. Make sure that you run this command as the user that
owns and manages the database, not just as
<code class="literal">root</code>. Part of the recovery process might
involve re-creating from scratch various database files (shared
memory regions, e.g.). Recovering as
<code class="literal">root</code> will create those files such that they
are owned by <code class="literal">root</code>, which means that even
after you restore connectivity to your repository, regular
users will be unable to access it.</p><p>If the previous procedure, for some reason, does not
successfully unwedge your repository, you should do two
things. First, move your broken repository directory aside
(perhaps by renaming it to something like
<code class="filename">repos.BROKEN</code>) and then restore your
latest backup of it. Then, send an email to the Subversion
users mailing list (at <code class="email"><<a class="email" href="mailto:users@subversion.apache.org">users@subversion.apache.org</a>></code>)
describing your problem in detail. Data integrity is an
extremely high priority to the Subversion developers.</p></div><div class="sect2" title="Migrating Repository Data Elsewhere"><div class="titlepage"><div><div><h3 class="title"><a id="svn.reposadmin.maint.migrate"></a>Migrating Repository Data Elsewhere</h3></div></div></div><p>A Subversion filesystem has its data spread throughout
files in the repository, in a fashion generally
understood by (and of interest to) only the Subversion
developers themselves. However, circumstances may arise that
call for all, or some subset, of that data to be copied or
moved into another repository.</p><p>Subversion provides such functionality by way of
<em class="firstterm">repository dump streams</em>. A repository
dump stream (often referred to as a <span class="quote">“<span class="quote">dump file</span>”</span>
when stored as a file on disk) is a portable, flat file format
that describes the various revisions in your
repository—what was changed, by whom, when, and so on.
This dump stream is the primary mechanism used to marshal
versioned history—in whole or in part, with or without
modification—between repositories. And Subversion
provides the tools necessary for creating and loading these
dump streams: the <span class="command"><strong>svnadmin dump</strong></span> and
<span class="command"><strong>svnadmin load</strong></span> subcommands, respectively,
and the <span class="command"><strong>svnrdump</strong></span> program.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>While the Subversion repository dump format contains
human-readable portions and a familiar structure (it
resembles an RFC 822 format, the same type of format used
for most email), it is <span class="emphasis"><em>not</em></span> a plain-text
file format. It is a binary file format, highly sensitive
to meddling. For example, many text editors will corrupt
the file by automatically converting line endings.</p></td></tr></table></div><p>There are many reasons for dumping and loading Subversion
repository data. Early in Subversion's life, the most common
reason was due to the evolution of Subversion itself. As
Subversion matured, there were times when changes made to the
backend database schema caused compatibility issues with
previous versions of the repository, so users had to dump
their repository data using the previous version of
Subversion and load it into a freshly created repository with
the new version of Subversion. Now, these types of schema
changes haven't occurred since Subversion's 1.0 release, and
the Subversion developers promise not to force users to dump
and load their repositories when upgrading between minor
versions (such as from 1.3 to 1.4) of Subversion. But there
are still other reasons for dumping and loading, including
re-deploying a Berkeley DB repository on a new OS or CPU
architecture, switching between the Berkeley DB and FSFS
backends, or (as we'll cover later in this chapter in <a class="xref" href="#svn.reposadmin.maint.filtering" title="Filtering Repository History">the section called “Filtering Repository History”</a>) purging versioned
data from repository history.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>The Subversion repository dump format describes
versioned repository changes only. It will not carry any
information about uncommitted transactions, user locks on
filesystem paths, repository or server configuration
customizations (including hook scripts), and so on.</p></td></tr></table></div><p>The Subversion repository dump format also enables
conversion from a different storage mechanism or version
control system altogether. Because the dump file format is,
for the most part, human-readable, it should be relatively
easy to describe generic sets of changes—each of which
should be treated as a new revision—using this file
format. In fact, the <span class="command"><strong>cvs2svn</strong></span> utility (see
<a class="xref" href="#svn.forcvs.convert" title="Converting a Repository from CVS to Subversion">the section called “Converting a Repository from CVS to Subversion”</a>) uses the dump format to
represent the contents of a CVS repository so that those
contents can be copied into a Subversion repository.</p><p>For now, we'll concern ourselves only with migration of
repository data between Subversion repositories, which we'll
describe in detail in the sections which follow.</p><div class="sect3" title="Repository data migration using svnadmin"><div class="titlepage"><div><div><h4 class="title"><a id="svn.reposadmin.maint.migrate.svnadmin"></a>Repository data migration using svnadmin</h4></div></div></div><p>Whatever your reason for migrating repository history,
using the <span class="command"><strong>svnadmin dump</strong></span> and
<span class="command"><strong>svnadmin load</strong></span> subcommands is
straightforward. <span class="command"><strong>svnadmin dump</strong></span> will output
a range of repository revisions that are formatted using
Subversion's custom filesystem dump format. The dump format
is printed to the standard output stream, while informative
messages are printed to the standard error stream. This
allows you to redirect the output stream to a file while
watching the status output in your terminal window. For
example:</p><div class="informalexample"><pre class="screen">
$ svnlook youngest myrepos
26
$ svnadmin dump myrepos > dumpfile
* Dumped revision 0.
* Dumped revision 1.
* Dumped revision 2.
…
* Dumped revision 25.
* Dumped revision 26.
</pre></div><p>At the end of the process, you will have a single file
(<code class="filename">dumpfile</code> in the previous example) that
contains all the data stored in your repository in the
requested range of revisions. Note that <span class="command"><strong>svnadmin
dump</strong></span> is reading revision trees from the repository
just like any other <span class="quote">“<span class="quote">reader</span>”</span> process would
(e.g., <span class="command"><strong>svn checkout</strong></span>), so it's safe
to run this command at any time.</p><p>The other subcommand in the pair, <span class="command"><strong>svnadmin
load</strong></span>, parses the standard input stream as a
Subversion repository dump file and effectively replays those
dumped revisions into the target repository for that
operation. It also gives informative feedback, this time
using the standard output stream:</p><div class="informalexample"><pre class="screen">
$ svnadmin load newrepos < dumpfile
<<< Started new txn, based on original revision 1
* adding path : A ... done.
* adding path : A/B ... done.
…
------- Committed new rev 1 (loaded from original rev 1) >>>
<<< Started new txn, based on original revision 2
* editing path : A/mu ... done.
* editing path : A/D/G/rho ... done.
------- Committed new rev 2 (loaded from original rev 2) >>>
…
<<< Started new txn, based on original revision 25
* editing path : A/D/gamma ... done.
------- Committed new rev 25 (loaded from original rev 25) >>>
<<< Started new txn, based on original revision 26
* adding path : A/Z/zeta ... done.
* editing path : A/mu ... done.
------- Committed new rev 26 (loaded from original rev 26) >>>
</pre></div><p>The result of a load is new revisions added to a
repository—the same thing you get by making commits
against that repository from a regular Subversion client.
Just as in a commit, you can use hook programs to perform
actions before and after each of the commits made during a
load process. By passing the
<code class="option">--use-pre-commit-hook</code> and
<code class="option">--use-post-commit-hook</code> options to
<span class="command"><strong>svnadmin load</strong></span>, you can instruct Subversion
to execute the pre-commit and post-commit hook programs,
respectively, for each loaded revision. You might use these,
for example, to ensure that loaded revisions pass through the
same validation steps that regular commits pass through. Of
course, you should use these options with care—if your
post-commit hook sends emails to a mailing list for each new
commit, you might not want to spew hundreds or thousands of
commit emails in rapid succession at that list! You can read
more about the use of hook scripts in
<a class="xref" href="#svn.reposadmin.create.hooks" title="Implementing Repository Hooks">the section called “Implementing Repository Hooks”</a>.</p><p>Note that because <span class="command"><strong>svnadmin</strong></span> uses
standard input and output streams for the repository dump and
load processes, people who are feeling especially saucy can try
things such as this (perhaps even using different versions of
<span class="command"><strong>svnadmin</strong></span> on each side of the pipe):</p><div class="informalexample"><pre class="screen">
$ svnadmin create newrepos
$ svnadmin dump oldrepos | svnadmin load newrepos
</pre></div><p>By default, the dump file will be quite large—much
larger than the repository itself. That's because by default
every version of every file is expressed as a full text in the
dump file. This is the fastest and simplest behavior, and
it's nice if you're piping the dump data directly into some other
process (such as a compression program, filtering program, or
loading process). But if you're creating a dump file
for longer-term storage, you'll likely want to save disk space
by using the <code class="option">--deltas</code> option. With this
option, successive revisions of files will be output as
compressed, binary differences—just as file revisions
are stored in a repository. This option is slower, but it
results in a dump file much closer in size to the original
repository.</p><p>We mentioned previously that <span class="command"><strong>svnadmin
dump</strong></span> outputs a range of revisions. Use the
<code class="option">--revision</code> (<code class="option">-r</code>) option to
specify a single revision, or a range of revisions, to dump.
If you omit this option, all the existing repository revisions
will be dumped.</p><div class="informalexample"><pre class="screen">
$ svnadmin dump myrepos -r 23 > rev-23.dumpfile
$ svnadmin dump myrepos -r 100:200 > revs-100-200.dumpfile
</pre></div><p>As Subversion dumps each new revision, it outputs only
enough information to allow a future loader to re-create that
revision based on the previous one. In other words, for any
given revision in the dump file, only the items that were
changed in that revision will appear in the dump. The only
exception to this rule is the first revision that is dumped
with the current <span class="command"><strong>svnadmin dump</strong></span>
command.</p><p>By default, Subversion will not express the first dumped
revision as merely differences to be applied to the previous
revision. For one thing, there is no previous revision in the
dump file! And second, Subversion cannot know the state of
the repository into which the dump data will be loaded (if it
ever is). To ensure that the output of each
execution of <span class="command"><strong>svnadmin dump</strong></span> is
self-sufficient, the first dumped revision is, by default, a
full representation of every directory, file, and property in
that revision of the repository.</p><p>However, you can change this default behavior. If you add
the <code class="option">--incremental</code> option when you dump your
repository, <span class="command"><strong>svnadmin</strong></span> will compare the first
dumped revision against the previous revision in the
repository—the same way it treats every other revision that
gets dumped. It will then output the first revision exactly
as it does the rest of the revisions in the dump
range—mentioning only the changes that occurred in that
revision. The benefit of this is that you can create several
small dump files that can be loaded in succession, instead of
one large one, like so:</p><div class="informalexample"><pre class="screen">
$ svnadmin dump myrepos -r 0:1000 > dumpfile1
$ svnadmin dump myrepos -r 1001:2000 --incremental > dumpfile2
$ svnadmin dump myrepos -r 2001:3000 --incremental > dumpfile3
</pre></div><p>These dump files could be loaded into a new repository
with the following command sequence:</p><div class="informalexample"><pre class="screen">
$ svnadmin load newrepos < dumpfile1
$ svnadmin load newrepos < dumpfile2
$ svnadmin load newrepos < dumpfile3
</pre></div><p>Another neat trick you can perform with this
<code class="option">--incremental</code> option involves appending to an
existing dump file a new range of dumped revisions. For
example, you might have a <code class="literal">post-commit</code> hook
that simply appends the repository dump of the single revision
that triggered the hook. Or you might have a script that runs
nightly to append dump file data for all the revisions that
were added to the repository since the last time the script
ran. Used like this, <span class="command"><strong>svnadmin dump</strong></span> can be
one way to back up changes to your repository over time in case
of a system crash or some other catastrophic event.</p><p>The dump format can also be used to merge the contents of
several different repositories into a single repository. By
using the <code class="option">--parent-dir</code> option of
<span class="command"><strong>svnadmin load</strong></span>, you can specify a new
virtual root directory for the load process. That means if
you have dump files for three repositories—say
<code class="filename">calc-dumpfile</code>,
<code class="filename">cal-dumpfile</code>, and
<code class="filename">ss-dumpfile</code>—you can first create a new
repository to hold them all:</p><div class="informalexample"><pre class="screen">
$ svnadmin create /var/svn/projects
$
</pre></div><p>Then, make new directories in the repository that will
encapsulate the contents of each of the three previous
repositories:</p><div class="informalexample"><pre class="screen">
$ svn mkdir -m "Initial project roots" \
file:///var/svn/projects/calc \
file:///var/svn/projects/calendar \
file:///var/svn/projects/spreadsheet
Committed revision 1.
$
</pre></div><p>Lastly, load the individual dump files into their
respective locations in the new repository:</p><div class="informalexample"><pre class="screen">
$ svnadmin load /var/svn/projects --parent-dir calc < calc-dumpfile
…
$ svnadmin load /var/svn/projects --parent-dir calendar < cal-dumpfile
…
$ svnadmin load /var/svn/projects --parent-dir spreadsheet < ss-dumpfile
…
$
</pre></div></div><div class="sect3" title="Repository data migration using svnrdump"><div class="titlepage"><div><div><h4 class="title"><a id="svn.reposadmin.maint.migrate.svnrdump"></a>Repository data migration using svnrdump</h4></div></div></div><p>In Subversion 1.7, <span class="command"><strong>svnrdump</strong></span> joined
the set of stock Subversion tools. It offers fairly
specialized functionality, essentially as a network-aware
version of the <span class="command"><strong>svnadmin dump</strong></span>
and <span class="command"><strong>svnadmin load</strong></span> commands which we
discuss in depth in
<a class="xref" href="#svn.reposadmin.maint.migrate.svnadmin" title="Repository data migration using svnadmin">the section called “Repository data migration using svnadmin”</a>. <span class="command"><strong>svnrdump dump</strong></span> will generate a dump
stream from a remote repository, spewing it to standard
output; <span class="command"><strong>svnrdump load</strong></span> will read a dump
stream from standard input and load it into a remote
repository. Using <span class="command"><strong>svnrdump</strong></span>, you can
generate incremental dumps just as you might
with <span class="command"><strong>svnadmin dump</strong></span>. You can even dump a
subtree of the repository—something
that <span class="command"><strong>svnadmin dump</strong></span> cannot do.</p><p>The primary difference is that instead of requiring
direct access to the repository, <span class="command"><strong>svnrdump</strong></span>
operates remotely, using the very same Repository Access
(RA) protocols that the Subversion client does. As such,
you might need to provide authentication credentials. Also,
your remote interations are subject to any authorization
limitations configured on the Subversion server.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p><span class="command"><strong>svnrdump dump</strong></span> requires that the
remote server be running Subversion 1.4 or newer. It
currently generates dump streams only of the sort which
are created when you pass the <code class="option">--deltas</code>
option to <span class="command"><strong>svnadmin dump</strong></span>. This isn't
interesting in the typical use-cases, but might impact
specific types of custom transformations you might wish to
apply to the resulting dump stream.</p></td></tr></table></div><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>Because it modifies revision properties after
committing new revisions, <span class="command"><strong>svnrdump load</strong></span>
requires that the target repository have revision property
changes enabled via the pre-revprop-change hook. See
<a class="xref" href="#svn.ref.reposhooks.pre-revprop-change" title="pre-revprop-change">pre-revprop-change</a> in
<a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a> for details.</p></td></tr></table></div><p>As you might expect, you can use
<span class="command"><strong>svnadmin</strong></span> and <span class="command"><strong>svnrdump</strong></span>
in concert. You can, for example, use <span class="command"><strong>svnrdump
dump</strong></span> to generate a dump stream from a remote
repository, and pipe the results thereof through
<span class="command"><strong>svnadmin load</strong></span> to copy all that repository
history into a local repository. Or you can do the reverse,
copying history from a local repository into a remote
one.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>By using <code class="literal">file://</code>
URLs, <span class="command"><strong>svnrdump</strong></span> can also access local
repositories, but it will be doing so via Subversion's
Repository Access (RA) abstraction layer—you'll get
better performance out of <span class="command"><strong>svnadmin</strong></span> in
such situations.</p></td></tr></table></div></div></div><div class="sect2" title="Filtering Repository History"><div class="titlepage"><div><div><h3 class="title"><a id="svn.reposadmin.maint.filtering"></a>Filtering Repository History</h3></div></div></div><p>Since Subversion stores your versioned history using, at
the very least, binary differencing algorithms and data
compression (optionally in a completely opaque database
system), attempting manual tweaks is unwise if not quite
difficult, and at any rate strongly discouraged. And once
data has been stored in your repository, Subversion generally
doesn't provide an easy way to remove that
data.<sup>[<a id="idp12136880" href="#ftn.idp12136880" class="footnote">45</a>]</sup> But inevitably, there
will be times when you would like to manipulate the history of
your repository. You might need to strip out all instances of
a file that was accidentally added to the repository (and
shouldn't be there for whatever
reason).<sup>[<a id="idp12137696" href="#ftn.idp12137696" class="footnote">46</a>]</sup> Or, perhaps you have multiple
projects sharing a single repository, and you decide to split
them up into their own repositories. To accomplish tasks such
as these, administrators need a more manageable and malleable
representation of the data in their repositories—the
Subversion repository dump format.</p><p>As we described earlier in <a class="xref" href="#svn.reposadmin.maint.migrate" title="Migrating Repository Data Elsewhere">the section called “Migrating Repository Data Elsewhere”</a>, the Subversion
repository dump format is a human-readable representation of
the changes that you've made to your versioned data over time.
Use the <span class="command"><strong>svnadmin dump</strong></span> or <span class="command"><strong>svnrdump
dump</strong></span> command to generate the dump data,
and <span class="command"><strong>svnadmin load</strong></span> or <span class="command"><strong>svnrdump
load</strong></span> to populate a new repository with it. The
great thing about the human-readability aspect of the dump
format is that, if you aren't careless about it, you can
manually inspect and modify it. Of course, the downside is
that if you have three years' worth of repository activity
encapsulated in what is likely to be a very large dump file,
it could take you a long, long time to manually inspect and
modify it.</p><p>That's where <span class="command"><strong>svndumpfilter</strong></span> becomes
useful. This program acts as a path-based filter for
repository dump streams. Simply give it either a list of
paths you wish to keep or a list of paths you wish to not
keep, and then pipe your repository dump data through this
filter. The result will be a modified stream of dump data
that contains only the versioned paths you (explicitly or
implicitly) requested.</p><p>Let's look at a realistic example of how you might use this
program. Earlier in this chapter (see <a class="xref" href="#svn.reposadmin.projects.chooselayout" title="Planning Your Repository Organization">the section called “Planning Your Repository Organization”</a>), we discussed the
process of deciding how to choose a layout for the data in
your repositories—using one repository per project or
combining them, arranging stuff within your repository, and
so on. But sometimes after new revisions start flying in,
you rethink your layout and would like to make some changes.
A common change is the decision to move multiple projects
that are sharing a single repository into separate
repositories for each project.</p><p>Our imaginary repository contains three projects:
<code class="literal">calc</code>, <code class="literal">calendar</code>, and
<code class="literal">spreadsheet</code>. They have been living
side-by-side in a layout like this:</p><div class="informalexample"><div class="literallayout"><p><br></br>
/<br></br>
calc/<br></br>
trunk/<br></br>
branches/<br></br>
tags/<br></br>
calendar/<br></br>
trunk/<br></br>
branches/<br></br>
tags/<br></br>
spreadsheet/<br></br>
trunk/<br></br>
branches/<br></br>
tags/<br></br>
</p></div></div><p>To get these three projects into their own repositories,
we first dump the whole repository:</p><div class="informalexample"><pre class="screen">
$ svnadmin dump /var/svn/repos > repos-dumpfile
* Dumped revision 0.
* Dumped revision 1.
* Dumped revision 2.
* Dumped revision 3.
…
$
</pre></div><p>Next, run that dump file through the filter, each time
including only one of our top-level directories. This results
in three new dump files:</p><div class="informalexample"><pre class="screen">
$ svndumpfilter include calc < repos-dumpfile > calc-dumpfile
…
$ svndumpfilter include calendar < repos-dumpfile > cal-dumpfile
…
$ svndumpfilter include spreadsheet < repos-dumpfile > ss-dumpfile
…
$
</pre></div><p>At this point, you have to make a decision. Each of your
dump files will create a valid repository, but will preserve
the paths exactly as they were in the original repository.
This means that even though you would have a repository solely
for your <code class="literal">calc</code> project, that repository
would still have a top-level directory named
<code class="filename">calc</code>. If you want your
<code class="filename">trunk</code>, <code class="filename">tags</code>, and
<code class="filename">branches</code> directories to live in the root
of your repository, you might wish to edit your dump files,
tweaking the <code class="literal">Node-path</code> and
<code class="literal">Node-copyfrom-path</code> headers so that they no
longer have that first <code class="filename">calc/</code> path
component. Also, you'll want to remove the section of dump
data that creates the <code class="filename">calc</code> directory. It
will look something like the following:</p><div class="informalexample"><pre class="programlisting">
Node-path: calc
Node-action: add
Node-kind: dir
Content-length: 0
</pre></div><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>If you do plan on manually editing the dump file to
remove a top-level directory, make sure your editor is
not set to automatically convert end-of-line characters to
the native format (e.g., <code class="literal">\r\n</code> to
<code class="literal">\n</code>), as the content will then not agree
with the metadata. This will render the dump file
useless.</p></td></tr></table></div><p>All that remains now is to create your three new
repositories, and load each dump file into the right
repository, ignoring the UUID found in the dump stream:</p><div class="informalexample"><pre class="screen">
$ svnadmin create calc
$ svnadmin load --ignore-uuid calc < calc-dumpfile
<<< Started new transaction, based on original revision 1
* adding path : Makefile ... done.
* adding path : button.c ... done.
…
$ svnadmin create calendar
$ svnadmin load --ignore-uuid calendar < cal-dumpfile
<<< Started new transaction, based on original revision 1
* adding path : Makefile ... done.
* adding path : cal.c ... done.
…
$ svnadmin create spreadsheet
$ svnadmin load --ignore-uuid spreadsheet < ss-dumpfile
<<< Started new transaction, based on original revision 1
* adding path : Makefile ... done.
* adding path : ss.c ... done.
…
$
</pre></div><p>Both of <span class="command"><strong>svndumpfilter</strong></span>'s subcommands
accept options for deciding how to deal with
<span class="quote">“<span class="quote">empty</span>”</span> revisions. If a given revision
contains only changes to paths that were filtered out, that
now-empty revision could be considered uninteresting or even
unwanted. So to give the user control over what to do with
those revisions, <span class="command"><strong>svndumpfilter</strong></span> provides
the following command-line options:</p><div class="variablelist"><dl><dt><span class="term"><code class="option">--drop-empty-revs</code></span></dt><dd><p>Do not generate empty revisions at all—just
omit them.</p></dd><dt><span class="term"><code class="option">--renumber-revs</code></span></dt><dd><p>If empty revisions are dropped (using the
<code class="option">--drop-empty-revs</code> option), change the
revision numbers of the remaining revisions so that
there are no gaps in the numeric sequence.</p></dd><dt><span class="term"><code class="option">--preserve-revprops</code></span></dt><dd><p>If empty revisions are not dropped, preserve the
revision properties (log message, author, date, custom
properties, etc.) for those empty revisions.
Otherwise, empty revisions will contain only the
original datestamp, and a generated log message that
indicates that this revision was emptied by
<span class="command"><strong>svndumpfilter</strong></span>.</p></dd></dl></div><p>While <span class="command"><strong>svndumpfilter</strong></span> can be very
useful and a huge timesaver, there are unfortunately a
couple of gotchas. First, this utility is overly sensitive
to path semantics. Pay attention to whether paths in your
dump file are specified with or without leading slashes.
You'll want to look at the <code class="literal">Node-path</code> and
<code class="literal">Node-copyfrom-path</code> headers.</p><div class="informalexample"><pre class="programlisting">
…
Node-path: spreadsheet/Makefile
…
</pre></div><p>If the paths have leading slashes, you should
include leading slashes in the paths you pass to
<span class="command"><strong>svndumpfilter include</strong></span> and
<span class="command"><strong>svndumpfilter exclude</strong></span> (and if they don't,
you shouldn't). Further, if your dump file has an
inconsistent usage of leading slashes for some
reason,<sup>[<a id="idp12184400" href="#ftn.idp12184400" class="footnote">47</a>]</sup> you should probably normalize
those paths so that they all have, or all lack, leading
slashes.</p><p>Also, copied paths can give you some trouble.
Subversion supports copy operations in the repository, where
a new path is created by copying some already existing path.
It is possible that at some point in the lifetime of your
repository, you might have copied a file or directory from
some location that <span class="command"><strong>svndumpfilter</strong></span> is
excluding, to a location that it is including. To
make the dump data self-sufficient,
<span class="command"><strong>svndumpfilter</strong></span> needs to still show the
addition of the new path—including the contents of any
files created by the copy—and not represent that
addition as a copy from a source that won't exist in your
filtered dump data stream. But because the Subversion
repository dump format shows only what was changed in each
revision, the contents of the copy source might not be
readily available. If you suspect that you have any copies
of this sort in your repository, you might want to rethink
your set of included/excluded paths, perhaps including the
paths that served as sources of your troublesome copy
operations, too.</p><p>Finally, <span class="command"><strong>svndumpfilter</strong></span> takes path
filtering quite literally. If you are trying to copy the
history of a project rooted at
<code class="filename">trunk/my-project</code> and move it into a
repository of its own, you would, of course, use the
<span class="command"><strong>svndumpfilter include</strong></span> command to keep all
the changes in and under
<code class="filename">trunk/my-project</code>. But the resultant
dump file makes no assumptions about the repository into
which you plan to load this data. Specifically, the dump
data might begin with the revision that added the
<code class="filename">trunk/my-project</code> directory, but it will
<span class="emphasis"><em>not</em></span> contain directives that would
create the <code class="filename">trunk</code> directory itself
(because <code class="filename">trunk</code> doesn't match the
include filter). You'll need to make sure that any
directories that the new dump stream expects to exist
actually do exist in the target repository before trying to
load the stream into that repository.</p></div><div class="sect2" title="Repository Replication"><div class="titlepage"><div><div><h3 class="title"><a id="svn.reposadmin.maint.replication"></a>Repository Replication</h3></div></div></div><p>There are several scenarios in which it is quite handy to
have a Subversion repository whose version history is exactly
the same as some other repository's. Perhaps the most obvious
one is the maintenance of a simple backup repository, used
when the primary repository has become inaccessible due to a
hardware failure, network outage, or other such annoyance.
Other scenarios include deploying mirror repositories to
distribute heavy Subversion load across multiple servers, use
as a soft-upgrade mechanism, and so on.</p><p>Subversion provides a program for managing scenarios such
as these. <span class="command"><strong>svnsync</strong></span> works by essentially
asking the Subversion server to <span class="quote">“<span class="quote">replay</span>”</span>
revisions, one at a time. It then uses that revision
information to mimic a commit of the same to another
repository. Neither repository needs to be locally accessible
to the machine on which <span class="command"><strong>svnsync</strong></span> is
running—its parameters are repository URLs, and it does
all its work through Subversion's Repository Access (RA)
interfaces. All it requires is read access to the source
repository and read/write access to the destination
repository.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>When using <span class="command"><strong>svnsync</strong></span> against a remote
source repository, the Subversion server for that repository
must be running Subversion version 1.4 or later.</p></td></tr></table></div><div class="sect3" title="Replication with svnsync"><div class="titlepage"><div><div><h4 class="title"><a id="svn.reposadmin.maint.replication.svnsync"></a>Replication with svnsync</h4></div></div></div><p>Assuming you already have a source repository that you'd
like to mirror, the next thing you need is a target repository
that will actually serve as that mirror. This target
repository can use either of the available filesystem
data-store backends (see
<a class="xref" href="#svn.reposadmin.basics.backends" title="Choosing a Data Store">the section called “Choosing a Data Store”</a>)—Subversion's abstraction layers ensure that such
details don't matter. But by default, it must
not yet have any version history in it. (We'll discuss an
exception to this later in this section.)</p><p>The protocol that <span class="command"><strong>svnsync</strong></span> uses to
communicate revision information is highly sensitive to
mismatches between the versioned histories contained in the
source and target repositories. For this reason,
while <span class="command"><strong>svnsync</strong></span>
cannot <span class="emphasis"><em>demand</em></span> that the target repository
be read-only,<sup>[<a id="idp12208992" href="#ftn.idp12208992" class="footnote">48</a>]</sup>
allowing the revision history in the target repository to
change by any mechanism other than the mirroring process is a
recipe for disaster.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>Do <span class="emphasis"><em>not</em></span> modify a mirror repository
in such a way as to cause its version history to deviate
from that of the repository it mirrors. The only commits
and revision property modifications that ever occur on that
mirror repository should be those performed by the
<span class="command"><strong>svnsync</strong></span> tool.</p></td></tr></table></div><p>Another requirement of the target repository is that the
<span class="command"><strong>svnsync</strong></span> process be allowed to modify
revision properties. Because <span class="command"><strong>svnsync</strong></span> works
within the framework of that repository's hook system, the
default state of the repository (which is to disallow revision
property changes; see <a class="xref" href="#svn.ref.reposhooks.pre-revprop-change" title="pre-revprop-change">pre-revprop-change</a> in
<a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a>) is insufficient. You'll need to
explicitly implement the pre-revprop-change hook, and your
script must allow <span class="command"><strong>svnsync</strong></span> to set and
change revision properties. With those provisions in place,
you are ready to start mirroring repository
revisions.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>It's a good idea to implement authorization measures
that allow your repository replication process to perform
its tasks while preventing other users from modifying the
contents of your mirror repository at all.</p></td></tr></table></div><p>Let's walk through the use of <span class="command"><strong>svnsync</strong></span>
in a somewhat typical mirroring scenario. We'll pepper this
discourse with practical recommendations, which you are free to
disregard if they aren't required by or suitable for your
environment.</p><p>We will be mirroring the public Subversion repository
which houses the source code for this very book and exposing
that mirror publicly on the Internet, hosted on a different
machine than the one on which the original Subversion source
code repository lives. This remote host has a global
configuration that permits anonymous users to read the
contents of repositories on the host, but requires users to
authenticate to modify those repositories. (Please forgive
us for glossing over the details of Subversion server
configuration for the moment—those are covered
thoroughly in <a class="xref" href="#svn.serverconfig" title="Chapter 6. Server Configuration">Chapter 6, <em>Server Configuration</em></a>.) And for
no other reason than that it makes for a more interesting
example, we'll be driving the replication process from a
third machine—the one that we currently find ourselves
using.</p><p>First, we'll create the repository which will be our
mirror. This and the next couple of steps do require shell
access to the machine on which the mirror repository will
live. Once the repository is all configured, though, we
shouldn't need to touch it directly again.</p><div class="informalexample"><pre class="screen">
$ ssh admin@svn.example.com "svnadmin create /var/svn/svn-mirror"
admin@svn.example.com's password: ********
$
</pre></div><p>At this point, we have our repository, and due to our
server's configuration, that repository is now
<span class="quote">“<span class="quote">live</span>”</span> on the Internet. Now, because we don't
want anything modifying the repository except our replication
process, we need a way to distinguish that process from other
would-be committers. To do so, we use a dedicated username
for our process. Only commits and revision property
modifications performed by the special username
<code class="literal">syncuser</code> will be allowed.</p><p>We'll use the repository's hook system both to allow the
replication process to do what it needs to do and to enforce
that only it is doing those things. We accomplish this by
implementing two of the repository event
hooks—pre-revprop-change and start-commit. Our
<code class="filename">pre-revprop-change</code> hook script is found
in <a class="xref" href="#svn.reposadmin.maint.replication.pre-revprop-change" title="Example 5.2. Mirror repository's pre-revprop-change hook script">Example 5.2, “Mirror repository's pre-revprop-change hook script”</a>, and basically verifies that the user attempting the
property changes is our <code class="literal">syncuser</code> user. If
so, the change is allowed; otherwise, it is denied.</p><div class="example"><a id="svn.reposadmin.maint.replication.pre-revprop-change"></a><p class="title"><strong>Example 5.2. Mirror repository's pre-revprop-change hook script</strong></p><div class="example-contents"><pre class="programlisting">
#!/bin/sh
USER="$3"
if [ "$USER" = "syncuser" ]; then exit 0; fi
echo "Only the syncuser user may change revision properties" >&2
exit 1
</pre></div></div><br class="example-break"></br><p>That covers revision property changes. Now we need to
ensure that only the <code class="literal">syncuser</code> user is
permitted to commit new revisions to the repository. We do
this using a <code class="filename">start-commit</code> hook script
such as the one in <a class="xref" href="#svn.reposadmin.maint.replication.start-commit" title="Example 5.3. Mirror repository's start-commit hook script">Example 5.3, “Mirror repository's start-commit hook script”</a>.</p><div class="example"><a id="svn.reposadmin.maint.replication.start-commit"></a><p class="title"><strong>Example 5.3. Mirror repository's start-commit hook script</strong></p><div class="example-contents"><pre class="programlisting">
#!/bin/sh
USER="$2"
if [ "$USER" = "syncuser" ]; then exit 0; fi
echo "Only the syncuser user may commit new revisions" >&2
exit 1
</pre></div></div><br class="example-break"></br><p>After installing our hook scripts and ensuring that they
are executable by the Subversion server, we're finished with
the setup of the mirror repository. Now, we get to actually
do the mirroring.</p><p>The first thing we need to do with
<span class="command"><strong>svnsync</strong></span> is to register in our target
repository the fact that it will be a mirror of the source
repository. We do this using the <span class="command"><strong>svnsync
initialize</strong></span> subcommand. The URLs we provide point
to the root directories of the target and source
repositories, respectively. In Subversion 1.4, this is
required—only full mirroring of repositories is
permitted. Beginning with Subversion 1.5, though, you can
use <span class="command"><strong>svnsync</strong></span> to mirror only some subtree
of the repository, too.</p><div class="informalexample"><pre class="screen">
$ svnsync help init
initialize (init): usage: svnsync initialize DEST_URL SOURCE_URL
Initialize a destination repository for synchronization from
another repository.
…
$ svnsync initialize http://svn.example.com/svn-mirror \
http://svnbook.googlecode.com/svn \
--sync-username syncuser --sync-password syncpass
Copied properties for revision 0 (svn:sync-* properties skipped).
NOTE: Normalized svn:* properties to LF line endings (1 rev-props, 0 node-props).
$
</pre></div><p>Our target repository will now remember that it is a
mirror of the public Subversion source code repository.
Notice that we provided a username and password as arguments
to <span class="command"><strong>svnsync</strong></span>—that was required by the
pre-revprop-change hook on our mirror repository.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>In Subversion 1.4, the values given to
<span class="command"><strong>svnsync</strong></span>'s <code class="option">--username</code> and
<code class="option">--password</code> command-line options were used
for authentication against both the source and destination
repositories. This caused problems when a user's
credentials weren't exactly the same for both repositories,
especially when running in noninteractive mode (with the
<code class="option">--non-interactive</code> option). This was
fixed in Subversion 1.5 with the introduction of two new
pairs of options. Use
<code class="option">--source-username</code> and
<code class="option">--source-password</code> to provide authentication
credentials for the source repository; use
<code class="option">--sync-username</code> and
<code class="option">--sync-password</code> to provide credentials for
the destination repository. (The old
<code class="option">--username</code> and <code class="option">--password</code>
options still exist for compatibility, but we advise against
using them.)</p></td></tr></table></div><p>And now comes the fun part. With a single subcommand, we
can tell <span class="command"><strong>svnsync</strong></span> to copy all the
as-yet-unmirrored revisions from the source repository to the
target.<sup>[<a id="idp12250256" href="#ftn.idp12250256" class="footnote">49</a>]</sup> The
<span class="command"><strong>svnsync synchronize</strong></span> subcommand will peek
into the special revision properties previously stored on the
target repository and determine how much of the source
repository has been previously mirrored—in this case,
the most recently mirrored revision is r0. Then it will query
the source repository and determine what the latest revision
in that repository is. Finally, it asks the source
repository's server to start replaying all the revisions
between 0 and that latest revision. As
<span class="command"><strong>svnsync</strong></span> gets the resultant response from
the source repository's server, it begins forwarding those
revisions to the target repository's server as new
commits.</p><div class="informalexample"><pre class="screen">
$ svnsync help synchronize
synchronize (sync): usage: svnsync synchronize DEST_URL [SOURCE_URL]
Transfer all pending revisions to the destination from the source
with which it was initialized.
…
$ svnsync synchronize http://svn.example.com/svn-mirror \
http://svnbook.googlecode.com/svn
Committed revision 1.
Copied properties for revision 1.
Committed revision 2.
Copied properties for revision 2.
Transmitting file data .
Committed revision 3.
Copied properties for revision 3.
…
Transmitting file data .
Committed revision 4063.
Copied properties for revision 4063.
Transmitting file data .
Committed revision 4064.
Copied properties for revision 4064.
Transmitting file data ....
Committed revision 4065.
Copied properties for revision 4065.
$
</pre></div><p>Of particular interest here is that for each mirrored
revision, there is first a commit of that revision to the
target repository, and then property changes follow. This
two-phase replication is required because the initial commit
is performed by (and attributed to) the user
<code class="literal">syncuser</code> and is datestamped with the time
as of that revision's creation. <span class="command"><strong>svnsync</strong></span>
has to follow up with an immediate series of property
modifications that copy into the target repository all the
original revision properties found for that revision in the
source repository, which also has the effect of fixing the
author and datestamp of the revision to match that of the
source repository.</p><p>Also noteworthy is that <span class="command"><strong>svnsync</strong></span>
performs careful bookkeeping that allows it to be safely
interrupted and restarted without ruining the integrity of the
mirrored data. If a network glitch occurs while mirroring a
repository, simply repeat the <span class="command"><strong>svnsync
synchronize</strong></span> command, and it will happily pick up
right where it left off. In fact, as new revisions appear in
the source repository, this is exactly what you do
to keep your mirror up to date.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>As part of its bookkeeping, <span class="command"><strong>svnsync</strong></span>
records in the mirror repository the URL with which the
mirror was initialized. Because of this, invocations of
<span class="command"><strong>svnsync</strong></span> which follow the initialization
step do not <span class="emphasis"><em>require</em></span> that you provide
the source URL on the command line again. However, for
security purposes, we recommend that you continue to do so.
Depending on how it is deployed, it may not be safe for
<span class="command"><strong>svnsync</strong></span> to trust the source URL which it
retrieves from the mirror repository, and from which it
pulls versioned data.</p></td></tr></table></div><div class="sidebar" title="svnsync Bookkeeping"><div class="titlepage"><div><div><p class="title"><strong>svnsync Bookkeeping</strong></p></div></div></div><p><span class="command"><strong>svnsync</strong></span> needs to be able to set and
modify revision properties on the mirror repository because
those properties are part of the data it is tasked with
mirroring. As those properties change in the source
repository, those changes need to be reflected in the mirror
repository, too. But <span class="command"><strong>svnsync</strong></span> also uses a
set of custom revision properties—stored in revision 0
of the mirror repository—for its own internal
bookkeeping. These properties contain information such as
the URL and UUID of the source repository, plus some
additional state-tracking information.</p><p>One of those pieces of state-tracking information is a
flag that essentially just means <span class="quote">“<span class="quote">there's a
synchronization in progress right now.</span>”</span> This is used
to prevent multiple <span class="command"><strong>svnsync</strong></span> processes
from colliding with each other while trying to mirror data
to the same destination repository. Now, generally you
won't need to pay any attention whatsoever to
<span class="emphasis"><em>any</em></span> of these special properties (all of
which begin with the prefix <code class="literal">svn:sync-</code>).
Occasionally, though, if a synchronization fails
unexpectedly, Subversion never has a chance to remove this
particular state flag. This causes all future
synchronization attempts to fail because it appears that a
synchronization is still in progress when, in fact, none is.
Fortunately, recovering from this situation is easy to do.
In Subversion 1.7, you can use the newly introduced
<code class="option">--steal-lock</code> option with
<span class="command"><strong>svnsync</strong></span>'s commands. In previous
Subversion versions, you need only to remove the
<code class="literal">svn:sync-lock</code> property which serves as
this flag from revision 0 of the mirror repository:</p><div class="informalexample"><pre class="screen">
$ svn propdel --revprop -r0 svn:sync-lock http://svn.example.com/svn-mirror
property 'svn:sync-lock' deleted from repository revision 0
$
</pre></div><p>Also, <span class="command"><strong>svnsync</strong></span> stores the source
repository URL provided at mirror initialization time in a
bookkeeping property on the mirror repository. Future
synchronization operations against that mirror which omit
the source URL at the command line will consult the
special <code class="literal">svn:sync-from-url</code> property
stored on the mirror itself to know where to synchronize
from. This value is used literally by the synchronization
process, though. Be wary of using non-fully-qualified
domain names (such as referring
to <code class="literal">svnbook.red-bean.com</code> as
simply <code class="literal">svnbook</code> because that happens to
work when you are connected directly to
the <code class="literal">red-bean.com</code> network), domain names
which don't resolve or resolve differently depending on
where you happen to be operating from, or IP addresses
(which can change over time). But here again, if you need
an existing mirror to start referring to a different URL
for the same source repository, you can change the
bookkeeping property which houses that information. Users
of Subversion 1.7 or better can use <span class="command"><strong>svnsync init
--allow-non-empty</strong></span> to reinitialize their mirrors
with new source URL:</p><div class="informalexample"><pre class="screen">
$ svnsync initialize --allow-non-empty http://svn.example.com/svn-mirror \
<em class="replaceable"><code>NEW-SOURCE-URL</code></em>
Copied properties for revision 4065.
$
</pre></div><p>If you are running an older version of Subversion,
you'll need to manually tweak
the <code class="literal">svn:sync-from-url</code> bookkeeping
property:</p><div class="informalexample"><pre class="screen">
$ svn propset --revprop -r0 svn:sync-from-url <em class="replaceable"><code>NEW-SOURCE-URL</code></em> \
http://svn.example.com/svn-mirror
property 'svn:sync-from-url' set on repository revision 0
$
</pre></div><p>Another interesting thing about these special
bookkeeping properties is that <span class="command"><strong>svnsync</strong></span>
will not attempt to mirror any of those properties when they
are found in the source repository. The reason is probably
obvious, but basically boils down to
<span class="command"><strong>svnsync</strong></span> not being able to distinguish the
special properties it has merely copied from the source
repository from those it needs to consult and maintain for
its own bookkeeping needs. This situation could occur if,
for example, you were maintaining a mirror of a mirror of a
third repository. When <span class="command"><strong>svnsync</strong></span> sees its
own special properties in revision 0 of the source
repository, it simply ignores them.</p><p>An <span class="command"><strong>svnsync info</strong></span> subcommand was
added in Subversion 1.6 to easily display the special
bookkeeping properties in the destination
repository.</p><div class="informalexample"><pre class="screen">
$ svnsync help info
info: usage: svnsync info DEST_URL
Print information about the synchronization destination repository
located at DEST_URL.
…
$ svnsync info http://svn.example.com/svn-mirror
Source URL: http://svnbook.googlecode.com/svn
Source Repository UUID: 931749d0-5854-0410-9456-f14be4d6b398
Last Merged Revision: 4065
$
</pre></div></div><p>There is, however, one bit of inelegance in the process.
Because Subversion revision properties can be changed at any
time throughout the lifetime of the repository, and because
they don't leave an audit trail that indicates when they were
changed, replication processes have to pay special attention
to them. If you've already mirrored the first 15 revisions of
a repository and someone then changes a revision property on
revision 12, <span class="command"><strong>svnsync</strong></span> won't know to go back
and patch up its copy of revision 12. You'll need to tell it
to do so manually by using (or with some additional tooling
around) the <span class="command"><strong>svnsync copy-revprops</strong></span>
subcommand, which simply rereplicates all the revision
properties for a particular revision or range thereof.</p><div class="informalexample"><pre class="screen">
$ svnsync help copy-revprops
copy-revprops: usage:
1. svnsync copy-revprops DEST_URL [SOURCE_URL]
2. svnsync copy-revprops DEST_URL REV[:REV2]
…
$ svnsync copy-revprops http://svn.example.com/svn-mirror 12
Copied properties for revision 12.
$
</pre></div><p>That's repository replication
via <span class="command"><strong>svnsync</strong></span> in a nutshell. You'll likely
want some automation around such a process. For example,
while our example was a pull-and-push setup, you might wish to
have your primary repository push changes to one or more
blessed mirrors as part of its post-commit and
post-revprop-change hook implementations. This would enable
the mirror to be up to date in as near to real time as is
likely possible.</p></div><div class="sect3" title="Partial replication with svnsync"><div class="titlepage"><div><div><h4 class="title"><a id="svn.reposadmin.maint.replication.svnsync-partial"></a>Partial replication with svnsync</h4></div></div></div><p><span class="command"><strong>svnsync</strong></span> isn't limited to full copies
of everything which lives in a repository. It can handle
various shades of partial replication, too. For example,
while it isn't very commonplace to do so,
<span class="command"><strong>svnsync</strong></span> does gracefully mirror repositories
in which the user as whom it authenticates has only partial
read access. It simply copies only the bits of the repository
that it is permitted to see. Obviously, such a mirror is not
useful as a backup solution.</p><p>As of Subversion 1.5, <span class="command"><strong>svnsync</strong></span> also
has the ability to mirror a subset of a repository rather than
the whole thing. The process of setting up and maintaining
such a mirror is exactly the same as when mirroring a whole
repository, except that instead of specifying the source
repository's root URL when running <span class="command"><strong>svnsync
init</strong></span>, you specify the URL of some subdirectory
within that repository. Synchronization to that mirror will
now copy only the bits that changed under that source
repository subdirectory. There are some limitations to this
support, though. First, you can't mirror multiple disjoint
subdirectories of the source repository into a single mirror
repository—you'd need to instead mirror some parent
directory that is common to both. Second, the filtering
logic is entirely path-based, so if the subdirectory you are
mirroring was renamed at some point in the past, your mirror
would contain only the revisions since the directory appeared
at the URL you specified. And likewise, if the source
subdirectory is renamed in the future, your synchronization
processes will stop mirroring data at the point that the
source URL you specified is no longer valid.</p></div><div class="sect3" title="A quick trick for mirror creation"><div class="titlepage"><div><div><h4 class="title"><a id="svn.reposadmin.maint.replication.svnsync-init-nonempty"></a>A quick trick for mirror creation</h4></div></div></div><p>We mentioned previously the cost of setting up an
initial mirror of an existing repository. For many folks,
the sheer cost of transmitting thousands—or
millions—of revisions of history to a new mirror
repository via <span class="command"><strong>svnsync</strong></span> is a show-stopper.
Fortunately, Subversion 1.7 provides a workaround by way of
a new <code class="option">--allow-non-empty</code> option to
<span class="command"><strong>svnsync initialize</strong></span>. This option allows
you to initialize one repository as a mirror of another
while bypassing the verification that the to-be-initialized
mirror has no version history present in it. Per our
previous warnings about the sensitivity of this whole
replication process, you should rightly discern that this is
an option to be used only with great caution. But it's
wonderfully handy when you have administrative access to the
source repository, where you can simply make a physical copy
of the repository and then initialize that copy as a new
mirror:</p><div class="informalexample"><pre class="screen">
$ svnadmin hotcopy /path/to/repos /path/to/mirror-repos
$ ### create /path/to/mirror-repos/hooks/pre-revprop-change
$ svnsync initialize file:///path/to/mirror-repos \
file:///path/to/repos
svnsync: E000022: Destination repository already contains revision history; co
nsider using --allow-non-empty if the repository's revisions are known to mirr
or their respective revisions in the source repository
$ svnsync initialize --allow-non-empty file:///path/to/mirror-repos \
file:///path/to/repos
Copied properties for revision 32042.
$
</pre></div><p>Admins who are running a version of Subversion prior to
1.7 (and thus do not have access to <span class="command"><strong>svnsync
initialize</strong></span>'s <code class="option">--allow-non-empty</code>
feature) can accomplish effectively the same thing that that
feature does through <span class="emphasis"><em>careful</em></span>
manipulation of the r0 revision properties on the copy of
the repository which is slated to become a mirror of the
original. Use <span class="command"><strong>svnadmin setrevprop</strong></span> to
create the same bookkeeping properties
that <span class="command"><strong>svnsync</strong></span> would have created
there.</p></div><div class="sect3" title="Replication wrap-up"><div class="titlepage"><div><div><h4 class="title"><a id="svn.reposadmin.maint.replication.wrapup"></a>Replication wrap-up</h4></div></div></div><p>We've discussed a couple of ways to replicate revision
history from one repository to another. So let's look now
at the user end of these operations. How does replication
and the various situations which call for it affect
Subversion clients?</p><p>As far as user interaction with repositories and mirrors
goes, it <span class="emphasis"><em>is</em></span> possible to have a single
working copy that interacts with both, but you'll have to
jump through some hoops to make it happen. First, you need
to ensure that both the primary and mirror repositories have
the same repository UUID (which is not the case by default).
See <a class="xref" href="#svn.reposadmin.maint.uuids" title="Managing Repository UUIDs">the section called “Managing Repository UUIDs”</a> later in
this chapter for more about this.</p><p>Once the two repositories have the same UUID, you can use
<span class="command"><strong>svn relocate</strong></span> to point your working
copy to whichever of the repositories you wish to operate
against, a process that is described in
<a class="xref" href="#svn.ref.svn.c.relocate" title="svn relocate">svn relocate</a> in
<a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a>. There is a possible danger here,
though, in that if the primary and mirror repositories
aren't in close synchronization, a working copy up to date
with, and pointing to, the primary repository will, if
relocated to point to an out-of-date mirror, become confused
about the apparent sudden loss of revisions it fully expects
to be present, and it will throw errors to that effect. If
this occurs, you can relocate your working copy back to the
primary repository and then either wait until the mirror
repository is up to date, or backdate your working copy to a
revision you know is present in the sync repository, and
then retry the relocation.</p><p>Finally, be aware that the revision-based replication
provided by <span class="command"><strong>svnsync</strong></span> is only
that—replication of revisions. Only the kinds of
information carried by the Subversion repository dump file
format are available for replication. As such, tools such
as <span class="command"><strong>svnsync</strong></span>
(and <span class="command"><strong>svnrdump</strong></span>, which we discuss in
<a class="xref" href="#svn.reposadmin.maint.migrate.svnrdump" title="Repository data migration using svnrdump">the section called “Repository data migration using svnrdump”</a>)
are limited in ways similar to that of the repository dump
stream. They do not include in their replicated information
such things as the hook implementations, repository or
server configuration data, uncommitted transactions, or
information about user locks on repository paths.</p></div></div><div class="sect2" title="Repository Backup"><div class="titlepage"><div><div><h3 class="title"><a id="svn.reposadmin.maint.backup"></a>Repository Backup</h3></div></div></div><p>Despite numerous advances in technology since the birth of
the modern computer, one thing unfortunately rings true with
crystalline clarity—sometimes things go very, very
awry. Power outages, network connectivity dropouts, corrupt
RAM, and crashed hard drives are but a taste of the evil that
Fate is poised to unleash on even the most conscientious
administrator. And so we arrive at a very important
topic—how to make backup copies of your repository
data.</p><p>There are two types of backup methods available for
Subversion repository administrators—full and
incremental. A full backup of the repository involves
squirreling away in one sweeping action all the information
required to fully reconstruct that repository in the event of
a catastrophe. Usually, it means, quite literally, the
duplication of the entire repository directory (which includes
either a Berkeley DB or FSFS environment). Incremental
backups are lesser things: backups of only the portion of the
repository data that has changed since the previous
backup.</p><p>As far as full backups go, the naïve approach might seem
like a sane one, but unless you temporarily disable all other
access to your repository, simply doing a recursive directory
copy runs the risk of generating a faulty backup. In the case
of Berkeley DB, the documentation describes a certain order in
which database files can be copied that will guarantee a valid
backup copy. A similar ordering exists for FSFS data. But
you don't have to implement these algorithms yourself, because
the Subversion development team has already done so. The
<span class="command"><strong>svnadmin hotcopy</strong></span> command takes care of the
minutia involved in making a hot backup of your repository.
And its invocation is as trivial as the Unix
<span class="command"><strong>cp</strong></span> or Windows <span class="command"><strong>copy</strong></span>
operations:</p><div class="informalexample"><pre class="screen">
$ svnadmin hotcopy /var/svn/repos /var/svn/repos-backup
</pre></div><p>The resultant backup is a fully functional Subversion
repository, able to be dropped in as a replacement for your
live repository should something go horribly wrong.</p><p>When making copies of a Berkeley DB repository, you can
even instruct <span class="command"><strong>svnadmin hotcopy</strong></span> to purge any
unused Berkeley DB logfiles (see <a class="xref" href="#svn.reposadmin.maint.diskspace.bdblogs" title="Purging unused Berkeley DB logfiles">the section called “Purging unused Berkeley DB logfiles”</a>) from the
original repository upon completion of the copy. Simply
provide the <code class="option">--clean-logs</code> option on the
command line.</p><div class="informalexample"><pre class="screen">
$ svnadmin hotcopy --clean-logs /var/svn/bdb-repos /var/svn/bdb-repos-backup
</pre></div><p>Additional tooling around this command is available, too.
The <code class="filename">tools/backup/</code> directory of the
Subversion source distribution holds the
<span class="command"><strong>hot-backup.py</strong></span> script. This script adds a
bit of backup management atop <span class="command"><strong>svnadmin
hotcopy</strong></span>, allowing you to keep only the most recent
configured number of backups of each repository. It will
automatically manage the names of the backed-up repository
directories to avoid collisions with previous backups and
will <span class="quote">“<span class="quote">rotate off</span>”</span> older backups, deleting them so
that only the most recent ones remain. Even if you also have an
incremental backup, you might want to run this program on a
regular basis. For example, you might consider using
<span class="command"><strong>hot-backup.py</strong></span> from a program scheduler
(such as <span class="command"><strong>cron</strong></span> on Unix systems), which can
cause it to run nightly (or at whatever granularity of time
you deem safe).</p><p>Some administrators use a different backup mechanism built
around generating and storing repository dump data. We
described in <a class="xref" href="#svn.reposadmin.maint.migrate" title="Migrating Repository Data Elsewhere">the section called “Migrating Repository Data Elsewhere”</a>
how to use <span class="command"><strong>svnadmin dump</strong></span> with
the <code class="option">--incremental</code> option to perform an
incremental backup of a given revision or range of revisions.
And of course, you can achieve a full backup variation of this
by omitting the <code class="option">--incremental</code> option to that
command. There is some value in these methods, in that the
format of your backed-up information is flexible—it's
not tied to a particular platform, versioned filesystem type,
or release of Subversion or Berkeley DB. But that flexibility
comes at a cost, namely that restoring that data can take a
long time—longer with each new revision committed to
your repository. Also, as is the case with so many of the
various backup methods, revision property changes that are
made to already backed-up revisions won't get picked up by a
nonoverlapping, incremental dump generation. For these
reasons, we recommend against relying solely on dump-based
backup approaches.</p><p>As you can see, each of the various backup types and
methods has its advantages and disadvantages. The easiest is
by far the full hot backup, which will always result in a
perfect working replica of your repository. Should something
bad happen to your live repository, you can restore from the
backup with a simple recursive directory copy. Unfortunately,
if you are maintaining multiple backups of your repository,
these full copies will each eat up just as much disk space as
your live repository. Incremental backups, by contrast, tend
to be quicker to generate and smaller to store. But the
restoration process can be a pain, often involving applying
multiple incremental backups. And other methods have their
own peculiarities. Administrators need to find the balance
between the cost of making the backup and the cost of
restoring it.</p><p>The <span class="command"><strong>svnsync</strong></span> program (see <a class="xref" href="#svn.reposadmin.maint.replication" title="Repository Replication">the section called “Repository Replication”</a>) actually
provides a rather handy middle-ground approach. If you are
regularly synchronizing a read-only mirror with your main
repository, in a pinch your read-only mirror is probably
a good candidate for replacing that main repository if it
falls over. The primary disadvantage of this method is that
only the versioned repository data gets
synchronized—repository configuration files,
user-specified repository path locks, and other items that
might live in the physical repository directory but not
<span class="emphasis"><em>inside</em></span> the repository's virtual versioned
filesystem are not handled by <span class="command"><strong>svnsync</strong></span>.</p><p>In any backup scenario, repository administrators need to
be aware of how modifications to unversioned revision
properties affect their backups. Since these changes do not
themselves generate new revisions, they will not trigger
post-commit hooks, and may not even trigger the
pre-revprop-change and post-revprop-change
hooks.<sup>[<a id="idp12352528" href="#ftn.idp12352528" class="footnote">50</a>]</sup> And since you can change
revision properties without respect to chronological
order—you can change any revision's properties at any
time—an incremental backup of the latest few revisions
might not catch a property modification to a revision that was
included as part of a previous backup.</p><p>Generally speaking, only the truly paranoid would need to
back up their entire repository, say, every time a commit
occurred. However, assuming that a given repository has some
other redundancy mechanism in place with relatively fine
granularity (such as per-commit emails or incremental dumps), a
hot backup of the database might be something that a
repository administrator would want to include as part of a
system-wide nightly backup. It's your data—protect it
as much as you'd like.</p><p>Often, the best approach to repository backups is a
diversified one that leverages combinations of the methods
described here. The Subversion developers, for example, back
up the Subversion source code repository nightly using
<span class="command"><strong>hot-backup.py</strong></span> and an off-site
<span class="command"><strong>rsync</strong></span> of those full backups; keep multiple
archives of all the commit and property change notification
emails; and have repository mirrors maintained by various
volunteers using <span class="command"><strong>svnsync</strong></span>. Your solution
might be similar, but should be catered to your needs and that
delicate balance of convenience with paranoia. And whatever
you do, validate your backups from time to time—what
good is a spare tire that has a hole in it? While all of this
might not save your hardware from the iron fist of
Fate,<sup>[<a id="idp12355104" href="#ftn.idp12355104" class="footnote">51</a>]</sup> it
should certainly help you recover from those trying
times.</p></div><div class="sect2" title="Managing Repository UUIDs"><div class="titlepage"><div><div><h3 class="title"><a id="svn.reposadmin.maint.uuids"></a>Managing Repository UUIDs</h3></div></div></div><p>Subversion repositories have a universally unique
identifier (UUID) associated with them. This is used by
Subversion clients to verify the identity of a repository when
other forms of verification aren't good enough (such as
checking the repository URL, which can change over time).
Most Subversion repository administrators rarely, if ever,
need to think about repository UUIDs as anything more than a
trivial implementation detail of Subversion. Sometimes,
however, there is cause for attention to this detail.</p><p>As a general rule, you want the UUIDs of your live
repositories to be unique. That is, after all, the point of
having UUIDs. But there are times when you want the
repository UUIDs of two repositories to be exactly the same.
For example, if you make a copy of a repository for backup
purposes, you want the backup to be a perfect replica of the
original so that, in the event that you have to restore that
backup and replace the live repository, users don't suddenly
see what looks like a different repository. When dumping and
loading repository history (as described earlier in <a class="xref" href="#svn.reposadmin.maint.migrate" title="Migrating Repository Data Elsewhere">the section called “Migrating Repository Data Elsewhere”</a>), you get to decide
whether to apply the UUID encapsulated in the data dump
stream to the repository in which you are loading the data. The
particular circumstance will dictate the correct
behavior.</p><p>There are a couple of ways to set (or reset) a
repository's UUID, should you need to. As of Subversion 1.5,
this is as simple as using the <span class="command"><strong>svnadmin
setuuid</strong></span> command. If you provide this subcommand
with an explicit UUID, it will validate that the UUID is
well-formed and then set the repository UUID to that value.
If you omit the UUID, a brand-new UUID will be generated for
your repository.</p><div class="informalexample"><pre class="screen">
$ svnlook uuid /var/svn/repos
cf2b9d22-acb5-11dc-bc8c-05e83ce5dbec
$ svnadmin setuuid /var/svn/repos # generate a new UUID
$ svnlook uuid /var/svn/repos
3c3c38fe-acc0-11dc-acbc-1b37ff1c8e7c
$ svnadmin setuuid /var/svn/repos \
cf2b9d22-acb5-11dc-bc8c-05e83ce5dbec # restore the old UUID
$ svnlook uuid /var/svn/repos
cf2b9d22-acb5-11dc-bc8c-05e83ce5dbec
$
</pre></div><p>For folks using versions of Subversion earlier than 1.5,
these tasks are a little more complicated. You can explicitly
set a repository's UUID by piping a repository dump file stub
that carries the new UUID specification through
<strong class="userinput"><code>svnadmin load --force-uuid
<em class="replaceable"><code>REPOS-PATH</code></em></code></strong>.</p><div class="informalexample"><pre class="screen">
$ svnadmin load --force-uuid /var/svn/repos <<EOF
SVN-fs-dump-format-version: 2
UUID: cf2b9d22-acb5-11dc-bc8c-05e83ce5dbec
EOF
$ svnlook uuid /var/svn/repos
cf2b9d22-acb5-11dc-bc8c-05e83ce5dbec
$
</pre></div><p>Having older versions of Subversion generate a brand-new
UUID is not quite as simple to do, though. Your best bet here
is to find some other way to generate a UUID, and then
explicitly set the repository's UUID to that value.</p></div></div><div class="sect1" title="Moving and Removing Repositories"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.reposadmin.maint.moving-and-removing"></a>Moving and Removing Repositories</h2></div></div></div><p>Subversion repository data is wholly contained within the
repository directory. As such, you can move a Subversion
repository to some other location on disk, rename a repository,
copy a repository, or delete a repository altogether using the
tools provided by your operating system for manipulating
directories—<span class="command"><strong>mv</strong></span>, <span class="command"><strong>cp
-a</strong></span>, and <span class="command"><strong>rm -r</strong></span> on Unix platforms;
<span class="command"><strong>copy</strong></span>, <span class="command"><strong>move</strong></span>, and
<span class="command"><strong>rmdir /s /q</strong></span> on Windows; vast numbers of mouse
and menu gyrations in various graphical file explorer
applications, and so on.</p><p>Of course, there's often still more to be done when trying
to cleanly affect changes such as this. For example, you might
need to update your Subversion server configuration to point to
the new location of a relocated repository or to remove
configuration bits for a now-deleted repository. If you have
automated processes that publish information from or about your
repositories, they may need to be updated. Hook scripts might
need to be reconfigured. Users may need to be notified. The list
can go on indefinitely, or at least to the extent that you've
built processes and procedures around your Subversion
repository.</p><p>In the case of a copied repository, you should also consider
the fact that Subversion uses repository UUIDs to distinguish
repositories. If you copy a Subversion repository using a
typical shell recursive copy command, you'll wind up with two
repositories that are identical in every way—including
their UUIDs. In some circumstances, this might be desirable.
But in the instances where it is not, you'll need to generate a
new UUID for one of these identical repositories. See
<a class="xref" href="#svn.reposadmin.maint.uuids" title="Managing Repository UUIDs">the section called “Managing Repository UUIDs”</a> for more about
managing repository UUIDs.</p></div><div class="sect1" title="Summary"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.reposadmin.summary"></a>Summary</h2></div></div></div><p>By now you should have a basic understanding of how to
create, configure, and maintain Subversion repositories. We
introduced you to the various tools that will assist you with
this task. Throughout the chapter, we noted common
administration pitfalls and offered suggestions for avoiding
them.</p><p>All that remains is for you to decide what exciting data to
store in your repository, and finally, how to make it available
over a network. The next chapter is all about networking.</p></div><div class="footnotes"><br></br><hr width="100" align="left"></hr><div class="footnote"><p><sup>[<a id="ftn.idp11598224" href="#idp11598224" class="para">37</a>] </sup>This may sound really prestigious
and lofty, but we're just talking about anyone who is interested
in that mysterious realm beyond the working copy where everyone's
data hangs out.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp11652016" href="#idp11652016" class="para">38</a>] </sup>Strictly speaking,
Subversion doesn't dictate that the versioned data live
here, and there are known (albeit proprietary) alternative
backend storage implementations which do not, in fact,
store data in this directory.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp11669152" href="#idp11669152" class="para">39</a>] </sup>Whether founded in
ignorance or in poorly considered concepts about how to derive
legitimate software development metrics, global revision
numbers are a silly thing to fear,
and <span class="emphasis"><em>not</em></span> the kind of thing you should
weigh when deciding how to arrange your projects and
repositories.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp11691600" href="#idp11691600" class="para">40</a>] </sup>The <code class="filename">trunk</code>,
<code class="filename">tags</code>, and <code class="filename">branches</code>
trio is sometimes referred to as <span class="quote">“<span class="quote">the TTB
directories.</span>”</span></p></div><div class="footnote"><p><sup>[<a id="ftn.idp11714608" href="#idp11714608" class="para">41</a>] </sup>Often
pronounced <span class="quote">“<span class="quote">fuzz-fuzz,</span>”</span> if Jack Repenning has
anything to say about it. (This book, however, assumes that
the reader is
thinking <span class="quote">“<span class="quote">eff-ess-eff-ess.</span>”</span>)</p></div><div class="footnote"><p><sup>[<a id="ftn.idp11756656" href="#idp11756656" class="para">42</a>] </sup>Berkeley DB requires that the
underlying filesystem implement strict POSIX locking
semantics, and more importantly, the ability to map files
directly into process memory.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp11924736" href="#idp11924736" class="para">43</a>] </sup>Or is that,
the <span class="quote">“<span class="quote">sync</span>”</span>?</p></div><div class="footnote"><p><sup>[<a id="ftn.idp12036112" href="#idp12036112" class="para">44</a>] </sup>For
example, hard drive + huge electromagnet =
disaster.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp12136880" href="#idp12136880" class="para">45</a>] </sup>That's rather the reason you use version
control at all, right?</p></div><div class="footnote"><p><sup>[<a id="ftn.idp12137696" href="#idp12137696" class="para">46</a>] </sup>Conscious, cautious removal of certain
bits of versioned data is actually supported by real use
cases. That's why an <span class="quote">“<span class="quote">obliterate</span>”</span> feature has
been one of the most highly requested Subversion features, and
one which the Subversion developers hope to soon
provide.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp12184400" href="#idp12184400" class="para">47</a>] </sup>While <span class="command"><strong>svnadmin dump</strong></span>
has a consistent leading slash policy (to not include them),
other programs that generate dump data might not be so
consistent.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp12208992" href="#idp12208992" class="para">48</a>] </sup>In fact, it can't truly be
read-only, or <span class="command"><strong>svnsync</strong></span> itself would have a
tough time copying revision history into it.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp12250256" href="#idp12250256" class="para">49</a>] </sup>Be forewarned that while it will take
only a few seconds for the average reader to parse this
paragraph and the sample output that follows it, the actual
time required to complete such a mirroring operation is, shall
we say, quite a bit longer.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp12352528" href="#idp12352528" class="para">50</a>] </sup><span class="command"><strong>svnadmin setlog</strong></span> can
be called in a way that bypasses the hook interface
altogether.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp12355104" href="#idp12355104" class="para">51</a>] </sup>You know—the collective term for
all of her <span class="quote">“<span class="quote">fickle fingers.</span>”</span></p></div></div></div><div class="chapter" title="Chapter 6. Server Configuration"><div class="titlepage"><div><div><h2 class="title"><a id="svn.serverconfig"></a>Chapter 6. Server Configuration</h2></div></div></div><div class="toc"><p><strong>Table of Contents</strong></p><dl><dt><span class="sect1"><a href="#svn.serverconfig.overview">Overview</a></span></dt><dt><span class="sect1"><a href="#svn.serverconfig.choosing">Choosing a Server Configuration</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.serverconfig.choosing.svnserve">The svnserve Server</a></span></dt><dt><span class="sect2"><a href="#svn.serverconfig.choosing.svn-ssh">svnserve over SSH</a></span></dt><dt><span class="sect2"><a href="#svn.serverconfig.choosing.apache">The Apache HTTP Server</a></span></dt><dt><span class="sect2"><a href="#svn.serverconfig.choosing.recommendations">Recommendations</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.serverconfig.svnserve">svnserve, a Custom Server</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.serverconfig.svnserve.invoking">Invoking the Server</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.invoking.daemon">svnserve as daemon</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.invoking.inetd">svnserve via inetd</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.invoking.tunnel">svnserve over a tunnel</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.invoking.winservice">svnserve as a Windows service</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.invoking.launchd">svnserve as a launchd job</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.serverconfig.svnserve.auth">Built-in Authentication and Authorization</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.auth.users">Create a users file and realm</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.auth.general">Set access controls</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.serverconfig.svnserve.sasl">Using <span class="command"><strong>svnserve</strong></span> with SASL</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.sasl.authn">Authenticating with SASL</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.sasl.encryption">SASL encryption</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.serverconfig.svnserve.sshauth">Tunneling over SSH</a></span></dt><dt><span class="sect2"><a href="#svn.serverconfig.svnserve.sshtricks">SSH Configuration Tricks</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.sshtricks.setup">Initial setup</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.svnserve.sshtricks.fixedcmd">Controlling the invoked command</a></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.serverconfig.httpd">httpd, the Apache HTTP Server</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.serverconfig.httpd.prereqs">Prerequisites</a></span></dt><dt><span class="sect2"><a href="#svn.serverconfig.httpd.basic">Basic Apache Configuration</a></span></dt><dt><span class="sect2"><a href="#svn.serverconfig.httpd.authn">Authentication Options</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.serverconfig.httpd.authn.basic">Basic authentication</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.httpd.authn.digest">Digest authentication</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.serverconfig.httpd.authz">Authorization Options</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.serverconfig.httpd.authz.blanket">Blanket access control</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.httpd.authz.perdir">Per-directory access control</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.httpd.authz.pathauthzoff">Disabling path-based checks</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.serverconfig.httpd.ssl">Protecting network traffic with SSL</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.serverconfig.httpd.ssl.server">Subversion server SSL certificate configuration</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.httpd.ssl.client">Subversion client SSL certificate management</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.serverconfig.httpd.extra">Extra Goodies</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.serverconfig.httpd.extra.browsing">Repository browsing</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.httpd.extra.logging">Apache logging</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.httpd.extra.writethruproxy">Write-through proxying</a></span></dt><dt><span class="sect3"><a href="#svn.serverconfig.httpd.extra.other">Other Apache features</a></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.serverconfig.pathbasedauthz">Path-Based Authorization</a></span></dt><dt><span class="sect1"><a href="#svn.serverconfig.operational-logging">High-level Logging</a></span></dt><dt><span class="sect1"><a href="#svn.serverconfig.optimization">Server Optimization</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.serverconfig.optimization.caching">Data Caching</a></span></dt><dt><span class="sect2"><a href="#svn.serverconfig.optimization.compression">Network Compression of Data</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.serverconfig.multimethod">Supporting Multiple Repository Access Methods</a></span></dt></dl></div><p>A Subversion repository can be accessed simultaneously by
clients running on the same machine on which the repository
resides using URLs carrying the <code class="literal">file://</code> scheme.
But the typical Subversion setup involves a single server machine
being accessed from clients on computers all over the
office—or, perhaps, all over the world.</p><p>This chapter describes how to get your Subversion repository
exposed outside its host machine for use by remote clients. We
will cover Subversion's currently available server mechanisms,
discussing the configuration and use of each. After reading this
chapter, you should be able to decide which networking setup is
right for your needs, as well as understand how to enable such a
setup on your host computer.</p><div class="sect1" title="Overview"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.serverconfig.overview"></a>Overview</h2></div></div></div><p>Subversion was designed with an abstract repository access layer.
This means that a repository can be programmatically accessed by
any sort of server process, and the client <span class="quote">“<span class="quote">repository
access</span>”</span> API allows programmers to write plug-ins that
speak relevant network protocols. In theory, Subversion can use
an infinite number of network implementations. In practice,
there are only two Subversion servers in widespread use today.</p><p>Apache is an extremely popular web server; using the
<span class="command"><strong>mod_dav_svn</strong></span> module, Apache can access a
repository and make it available to clients via the
WebDAV/DeltaV protocol, which is an extension of HTTP. Because
Apache is an extremely extensible server, it provides a number
of features <span class="quote">“<span class="quote">for free,</span>”</span> such as encrypted SSL
communication, logging, integration with a number of third-party
authentication systems, and limited built-in web browsing of
repositories.</p><p>In the other corner is <span class="command"><strong>svnserve</strong></span>: a small,
lightweight server program that speaks a custom protocol with
clients. Because its protocol is explicitly designed for
Subversion and is stateful (unlike HTTP), it provides
significantly faster network operations—but at the cost of
some features as well. While it can use SASL to provide a
variety of authentication and encryption options, it has no
logging or built-in web browsing. It is, however, extremely
easy to set up and is often the best option for small teams just
starting out with Subversion.</p><p>The network protocol which <span class="command"><strong>svnserve</strong></span>
speaks may also be tunneled over an SSH connection. This
deployment option for <span class="command"><strong>svnserve</strong></span> differs quite
a bit in features from a traditional <span class="command"><strong>svnserve</strong></span>
deployment. SSH is used to encrypt all communication. SSH is
also used exclusively to authenticate, so real system accounts
are required on the server host (unlike
vanilla <span class="command"><strong>svnserve</strong></span>, which has its own private
user accounts). Finally, because this setup requires that each
user spawn a private, temporary <span class="command"><strong>svnserve</strong></span>
process, it's equivalent (from a permissions point of view) to
allowing a group of local users to all access the repository
via <code class="literal">file://</code> URLs. Path-based access control
has no meaning, since each user is accessing the repository
database files directly.</p><p><a class="xref" href="#svn.serverconfig.overview.tbl-1" title="Table 6.1. Comparison of subversion server options">Table 6.1, “Comparison of subversion server options”</a> provides a
quick summary of the three typical server deployments.</p><div class="table"><a id="svn.serverconfig.overview.tbl-1"></a><p class="title"><strong>Table 6.1. Comparison of subversion server options</strong></p><div class="table-contents"><table summary="Comparison of subversion server options" border="1"><colgroup><col></col><col></col><col></col><col></col></colgroup><thead><tr><th>Feature</th><th>Apache + mod_dav_svn</th><th>svnserve</th><th>svnserve over SSH</th></tr></thead><tbody><tr><td>Authentication options</td><td>HTTP Basic or Digest auth, X.509 certificates, LDAP,
NTLM, or any other mechanism available to Apache httpd</td><td>CRAM-MD5 by default; LDAP, NTLM, or any other mechanism
available to SASL</td><td>SSH</td></tr><tr><td>User account options</td><td>Private <span class="quote">“<span class="quote">users</span>”</span> file, or other mechanisms
available to Apache httpd (LDAP, SQL, etc.)</td><td>Private <span class="quote">“<span class="quote">users</span>”</span> file, or other
mechanisms available to SASL (LDAP, SQL, etc.)</td><td>System accounts</td></tr><tr><td>Authorization options</td><td>Read/write access can be granted over the whole
repository, or specified per path</td><td>Read/write access can be granted over the whole
repository, or specified per path</td><td>Read/write access only grantable over the whole
repository</td></tr><tr><td>Encryption</td><td>Available via optional SSL (https)</td><td>Available via optional SASL features</td><td>Inherent in SSH connection</td></tr><tr><td>Logging</td><td>High-level operational logging of Subversion
operations plus detailed logging at the
per-HTTP-request level</td><td>High-level operational logging only</td><td>High-level operational logging only</td></tr><tr><td>Interoperability</td><td>Accessible by other WebDAV clients</td><td>Talks only to svn clients</td><td>Talks only to svn clients</td></tr><tr><td>Web viewing</td><td>Limited built-in support, or via third-party tools
such as ViewVC</td><td>Only via third-party tools such as ViewVC</td><td>Only via third-party tools such as ViewVC</td></tr><tr><td>Master-slave server replication</td><td>Transparent write-proxying available from slave to
master</td><td>Can only create read-only slave servers</td><td>Can only create read-only slave servers</td></tr><tr><td>Speed</td><td>Somewhat slower</td><td>Somewhat faster</td><td>Somewhat faster</td></tr><tr><td>Initial setup</td><td>Somewhat complex</td><td>Extremely simple</td><td>Moderately simple</td></tr></tbody></table></div></div><br class="table-break"></br></div><div class="sect1" title="Choosing a Server Configuration"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.serverconfig.choosing"></a>Choosing a Server Configuration</h2></div></div></div><p>So, which server should you use? Which is best?</p><p>Obviously, there's no right answer to that question. Every
team has different needs, and the different servers all
represent different sets of trade-offs. The Subversion project
itself doesn't endorse one server or another, or consider either
server more <span class="quote">“<span class="quote">official</span>”</span> than another.</p><p>Here are some reasons why you might choose one deployment
over another, as well as reasons you
might <span class="emphasis"><em>not</em></span> choose one.</p><div class="sect2" title="The svnserve Server"><div class="titlepage"><div><div><h3 class="title"><a id="svn.serverconfig.choosing.svnserve"></a>The svnserve Server</h3></div></div></div><div class="variablelist"><dl><dt><span class="term">Why you might want to use it:</span></dt><dd><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>Quick and easy to set up.</p></li><li class="listitem"><p>Network protocol is stateful and noticeably
faster than WebDAV.</p></li><li class="listitem"><p>No need to create system accounts on
server.</p></li><li class="listitem"><p>Password is not passed over the network.</p></li></ul></div></dd><dt><span class="term">Why you might want to avoid it:</span></dt><dd><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>By default, only one authentication method is
available, the network protocol is not encrypted,
and the server stores clear text passwords. (All
these things can be changed by configuring SASL, but
it's a bit more work to do.)</p></li><li class="listitem"><p>No advanced logging facilities.</p></li><li class="listitem"><p>No built-in web browsing. (You'd have to
install a separate web server and repository
browsing software to add this.)</p></li></ul></div></dd></dl></div></div><div class="sect2" title="svnserve over SSH"><div class="titlepage"><div><div><h3 class="title"><a id="svn.serverconfig.choosing.svn-ssh"></a>svnserve over SSH</h3></div></div></div><div class="variablelist"><dl><dt><span class="term">Why you might want to use it:</span></dt><dd><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>The network protocol is stateful and noticeably
faster than WebDAV.</p></li><li class="listitem"><p>You can take advantage of existing SSH accounts
and user infrastructure.</p></li><li class="listitem"><p>All network traffic is encrypted.</p></li></ul></div></dd><dt><span class="term">Why you might want to avoid it:</span></dt><dd><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>Only one choice of authentication method is
available.</p></li><li class="listitem"><p>No advanced logging facilities.</p></li><li class="listitem"><p>It requires users to be in the same system
group, or use a shared SSH key.</p></li><li class="listitem"><p>If used improperly, it can lead to file
permission problems.</p></li></ul></div></dd></dl></div></div><div class="sect2" title="The Apache HTTP Server"><div class="titlepage"><div><div><h3 class="title"><a id="svn.serverconfig.choosing.apache"></a>The Apache HTTP Server</h3></div></div></div><div class="variablelist"><dl><dt><span class="term">Why you might want to use it:</span></dt><dd><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>It allows Subversion to use any of the
numerous authentication systems already integrated
with Apache.</p></li><li class="listitem"><p>There is no need to create system accounts on
the server.</p></li><li class="listitem"><p>Full Apache logging is available.</p></li><li class="listitem"><p>Network traffic can be encrypted via SSL.</p></li><li class="listitem"><p>HTTP(S) can usually go through corporate
firewalls.</p></li><li class="listitem"><p>Built-in repository browsing is available via
web browser.</p></li><li class="listitem"><p>The repository can be mounted as a network
drive for transparent version control (see <a class="xref" href="#svn.webdav.autoversioning" title="Autoversioning">the section called “Autoversioning”</a>).</p></li></ul></div></dd><dt><span class="term">Why you might want to avoid it:</span></dt><dd><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>Noticeably slower than <span class="command"><strong>svnserve</strong></span>,
because HTTP is a stateless protocol and requires
more network turnarounds.</p></li><li class="listitem"><p>Initial setup can be complex.</p></li></ul></div></dd></dl></div></div><div class="sect2" title="Recommendations"><div class="titlepage"><div><div><h3 class="title"><a id="svn.serverconfig.choosing.recommendations"></a>Recommendations</h3></div></div></div><p>In general, the authors of this book recommend a vanilla
<span class="command"><strong>svnserve</strong></span> installation for small teams just
trying to get started with a Subversion server; it's the
simplest to set up and has the fewest maintenance issues.
You can always switch to a more complex server
deployment as your needs change.</p><p>Here are some general recommendations and tips, based on
years of supporting users:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>If you're trying to set up the simplest possible server
for your group, a vanilla <span class="command"><strong>svnserve</strong></span>
installation is the easiest, fastest route. Note,
however, that your repository data will be transmitted in
the clear over the network. If your deployment is
entirely within your company's LAN or VPN, this isn't an
issue. If the repository is exposed to the wide-open
Internet, you might want to make sure that either the
repository's contents aren't sensitive (e.g., it contains
only open source code), or that you go the extra mile in
configuring SASL to encrypt network communications.</p></li><li class="listitem"><p>If you need to integrate with existing legacy identity
systems (LDAP, Active Directory, NTLM, X.509, etc.),
you must use either the Apache-based server
or <span class="command"><strong>svnserve</strong></span> configured with SASL.</p></li><li class="listitem"><p>If you've decided to use either Apache or stock
<span class="command"><strong>svnserve</strong></span>, create a single
<span class="command"><strong>svn</strong></span> user on your system and run the
server process as that user. Be sure to make the
repository directory wholly owned by the
<span class="command"><strong>svn</strong></span> user as well. From a security
point of view, this keeps the repository data nicely
siloed and protected by operating system filesystem
permissions, changeable by only the Subversion server
process itself.</p></li><li class="listitem"><p>If you have an existing infrastructure that is heavily based
on SSH accounts, and if your users already have system
accounts on your server machine, it makes sense to
deploy an <span class="command"><strong>svnserve</strong></span>-over-SSH solution.
Otherwise, we don't widely recommend this option to the
public. It's generally considered safer to have your
users access the repository via (imaginary) accounts
managed by <span class="command"><strong>svnserve</strong></span> or Apache, rather
than by full-blown system accounts. If your deep desire
for encrypted communication still draws you to this
option, we recommend using Apache with SSL or
<span class="command"><strong>svnserve</strong></span> with SASL encryption
instead.</p></li><li class="listitem"><p>Do <span class="emphasis"><em>not</em></span> be seduced by the simple
idea of having all of your users access a repository
directly via <code class="literal">file://</code> URLs. Even if the
repository is readily available to everyone via a network
share, this is a bad idea. It removes any layers of
protection between the users and the repository: users can
accidentally (or intentionally) corrupt the repository
database, it becomes hard to take the repository offline
for inspection or upgrade, and it can lead to a mess of
file permission problems (see <a class="xref" href="#svn.serverconfig.multimethod" title="Supporting Multiple Repository Access Methods">the section called “Supporting Multiple Repository Access Methods”</a>). Note that this
is also one of the reasons we warn against accessing
repositories via <code class="literal">svn+ssh://</code>
URLs—from a security standpoint, it's effectively
the same as local users accessing via
<code class="literal">file://</code>, and it can entail all the same
problems if the administrator isn't careful.</p></li></ul></div></div></div><div class="sect1" title="svnserve, a Custom Server"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.serverconfig.svnserve"></a>svnserve, a Custom Server</h2></div></div></div><p>The <span class="command"><strong>svnserve</strong></span> program is a lightweight
server, capable of speaking to clients over TCP/IP using a
custom, stateful protocol. Clients contact an
<span class="command"><strong>svnserve</strong></span> server by using URLs that begin with
the <code class="literal">svn://</code> or <code class="literal">svn+ssh://</code>
scheme. This section will explain the different ways of running
<span class="command"><strong>svnserve</strong></span>, how clients authenticate themselves
to the server, and how to configure appropriate access control
to your repositories.</p><div class="sect2" title="Invoking the Server"><div class="titlepage"><div><div><h3 class="title"><a id="svn.serverconfig.svnserve.invoking"></a>Invoking the Server</h3></div></div></div><p>There are a few different ways to run the
<span class="command"><strong>svnserve</strong></span> program:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>Run <span class="command"><strong>svnserve</strong></span> as a standalone
daemon, listening for requests.</p></li><li class="listitem"><p>Have the Unix <span class="command"><strong>inetd</strong></span> daemon
temporarily spawn <span class="command"><strong>svnserve</strong></span> whenever a
request comes in on a certain port.</p></li><li class="listitem"><p>Have SSH invoke a temporary <span class="command"><strong>svnserve</strong></span>
over an encrypted tunnel.</p></li><li class="listitem"><p>Run <span class="command"><strong>svnserve</strong></span> as a Microsoft Windows
service.</p></li><li class="listitem"><p>Run <span class="command"><strong>svnserve</strong></span> as a launchd job.</p></li></ul></div><p>The following sections will cover in detail these various
deployment options for <span class="command"><strong>svnserve</strong></span>.</p><div class="sect3" title="svnserve as daemon"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.svnserve.invoking.daemon"></a>svnserve as daemon</h4></div></div></div><p>The easiest option is to run <span class="command"><strong>svnserve</strong></span>
as a standalone <span class="quote">“<span class="quote">daemon</span>”</span> process. Use the
<code class="option">-d</code> option for this:</p><div class="informalexample"><pre class="screen">
$ svnserve -d
$ # svnserve is now running, listening on port 3690
</pre></div><p>When running <span class="command"><strong>svnserve</strong></span> in daemon mode,
you can use the <code class="option">--listen-port</code> and
<code class="option">--listen-host</code> options to customize the
exact port and hostname to <span class="quote">“<span class="quote">bind</span>”</span> to.</p><p>Once we successfully start <span class="command"><strong>svnserve</strong></span>
as explained previously, it makes every repository on your
system available to the network. A client needs to specify
an <span class="emphasis"><em>absolute</em></span> path in the repository URL.
For example, if a repository is located at
<code class="filename">/var/svn/project1</code>, a client would reach
it via <code class="uri">svn://host.example.com/var/svn/project1</code>.
To increase security, you can pass the <code class="option">-r</code>
option to <span class="command"><strong>svnserve</strong></span>, which restricts it to
exporting only repositories below that path. For
example:</p><div class="informalexample"><pre class="screen">
$ svnserve -d -r /var/svn
…
</pre></div><p>Using the <code class="option">-r</code> option effectively
modifies the location that the program treats as the root of
the remote filesystem space. Clients then use URLs that
have that path portion removed from them, leaving much
shorter (and much less revealing) URLs:</p><div class="informalexample"><pre class="screen">
$ svn checkout svn://host.example.com/project1
…
</pre></div></div><div class="sect3" title="svnserve via inetd"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.svnserve.invoking.inetd"></a>svnserve via inetd</h4></div></div></div><p>If you want <span class="command"><strong>inetd</strong></span> to launch the
process, you need to pass the <code class="option">-i</code>
(<code class="option">--inetd</code>) option. In the following
example, we've shown the output from running
<code class="literal">svnserve -i</code> at the command line, but note
that this isn't how you actually start the daemon; see the
paragraphs following the example for how to configure
<span class="command"><strong>inetd</strong></span> to start
<span class="command"><strong>svnserve</strong></span>.</p><div class="informalexample"><pre class="screen">
$ svnserve -i
( success ( 2 2 ( ) ( edit-pipeline svndiff1 absent-entries commit-revprops d\
epth log-revprops atomic-revprops partial-replay ) ) )
</pre></div><p>When invoked with the <code class="option">--inetd</code> option,
<span class="command"><strong>svnserve</strong></span> attempts to speak with a
Subversion client via <code class="filename">stdin</code> and
<code class="filename">stdout</code> using a custom protocol. This
is the standard behavior for a program being run via
<span class="command"><strong>inetd</strong></span>. The IANA has reserved port 3690
for the Subversion protocol, so on a Unix-like system you
can add lines to <code class="filename">/etc/services</code> such as
these (if they don't already exist):</p><div class="informalexample"><pre class="programlisting">
svn 3690/tcp # Subversion
svn 3690/udp # Subversion
</pre></div><p>If your system is using a classic Unix-like
<span class="command"><strong>inetd</strong></span> daemon, you can add this line to
<code class="filename">/etc/inetd.conf</code>:</p><div class="informalexample"><pre class="programlisting">
svn stream tcp nowait svnowner /usr/bin/svnserve svnserve -i
</pre></div><p>Make sure <span class="quote">“<span class="quote">svnowner</span>”</span> is a user that has
appropriate permissions to access your repositories. Now,
when a client connection comes into your server on port
3690, <span class="command"><strong>inetd</strong></span> will spawn an
<span class="command"><strong>svnserve</strong></span> process to service it. Of
course, you may also want to add <code class="option">-r</code> to the
configuration line as well, to restrict which repositories
are exported.</p></div><div class="sect3" title="svnserve over a tunnel"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.svnserve.invoking.tunnel"></a>svnserve over a tunnel</h4></div></div></div><p>Another way to invoke <span class="command"><strong>svnserve</strong></span> is in
tunnel mode, using the <code class="option">-t</code> option. This
mode assumes that a remote-service program such as
<span class="command"><strong>rsh</strong></span> or <span class="command"><strong>ssh</strong></span> has
successfully authenticated a user and is now invoking a
private <span class="command"><strong>svnserve</strong></span> process <span class="emphasis"><em>as
that user</em></span>. (Note that you, the user, will
rarely, if ever, have reason to invoke
<span class="command"><strong>svnserve</strong></span> with the <code class="option">-t</code> at
the command line; instead, the SSH daemon
does so for you.) The <span class="command"><strong>svnserve</strong></span> program
behaves normally (communicating via
<code class="filename">stdin</code> and <code class="filename">stdout</code>)
and assumes that the traffic is being automatically
redirected over some sort of tunnel back to the client.
When <span class="command"><strong>svnserve</strong></span> is invoked by a tunnel
agent like this, be sure that the authenticated user has
full read and write access to the repository database files.
It's essentially the same as a local user accessing the
repository via <code class="literal">file://</code> URLs.</p><p>This option is described in much more detail later in
this chapter in <a class="xref" href="#svn.serverconfig.svnserve.sshauth" title="Tunneling over SSH">the section called “Tunneling over SSH”</a>.</p></div><div class="sect3" title="svnserve as a Windows service"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.svnserve.invoking.winservice"></a>svnserve as a Windows service</h4></div></div></div><p>If your Windows system is a descendant of Windows NT
(Windows 2000 or newer), you can
run <span class="command"><strong>svnserve</strong></span> as a standard Windows
service. This is typically a much nicer experience than
running it as a standalone daemon with
the <code class="option">--daemon</code> (<code class="option">-d</code>) option.
Using daemon mode requires launching a console, typing a
command, and then leaving the console window running
indefinitely. A Windows service, however, runs in the
background, can start at boot time automatically, and can be
started and stopped using the same consistent administration
interface as other Windows services.</p><p>You'll need to define the new service using the
command-line tool <span class="command"><strong>SC.EXE</strong></span>. Much like
the <span class="command"><strong>inetd</strong></span> configuration line, you must
specify an exact invocation of <span class="command"><strong>svnserve</strong></span>
for Windows to run at startup time:</p><div class="informalexample"><pre class="screen">
C:\> sc create svn
binpath= "C:\svn\bin\svnserve.exe --service -r C:\repos"
displayname= "Subversion Server"
depend= Tcpip
start= auto
</pre></div><p>This defines a new Windows service named
<code class="literal">svn</code> which executes a particular
<span class="command"><strong>svnserve.exe</strong></span> command when started (in
this case, rooted at <code class="filename">C:\repos</code>). There
are a number of caveats in the prior example,
however.</p><p>First, notice that the <span class="command"><strong>svnserve.exe</strong></span>
program must always be invoked with the
<code class="option">--service</code> option. Any other options to
<span class="command"><strong>svnserve</strong></span> must then be specified on the
same line, but you cannot add conflicting options such as
<code class="option">--daemon</code>
(<code class="option">-d</code>), <code class="option">--tunnel</code>,
or <code class="option">--inetd</code> (<code class="option">-i</code>). Options
such as <code class="option">-r</code>
or <code class="option">--listen-port</code> are fine, though. Second,
be careful about spaces when invoking
the <span class="command"><strong>SC.EXE</strong></span> command: the <code class="literal">key=
value</code> patterns must have no spaces between
<code class="literal">key=</code> and must have exactly one space
before the <code class="literal">value</code>. Lastly, be careful
about spaces in your command line to be invoked. If a
directory name contains spaces (or other characters that
need escaping), place the entire inner value of
<code class="literal">binpath</code> in double quotes, by escaping
them:</p><div class="informalexample"><pre class="screen">
C:\> sc create svn
binpath= "\"C:\program files\svn\bin\svnserve.exe\" --service -r C:\repos"
displayname= "Subversion Server"
depend= Tcpip
start= auto
</pre></div><p>Also note that the word <code class="literal">binpath</code> is
misleading—its value is a <span class="emphasis"><em>command
line</em></span>, not the path to an executable. That's why
you need to surround it with quotes if it contains
embedded spaces.</p><p>Once the service is defined, it can be stopped, started,
or queried using standard GUI tools (the Services
administrative control panel), or at the command
line:</p><div class="informalexample"><pre class="screen">
C:\> net stop svn
C:\> net start svn
</pre></div><p>The service can also be uninstalled (i.e., undefined) by
deleting its definition: <strong class="userinput"><code>sc delete svn</code></strong>.
Just be sure to stop the service first!
The <span class="command"><strong>SC.EXE</strong></span> program has many other
subcommands and options; run <strong class="userinput"><code>sc /?</code></strong> to
learn more about it.</p></div><div class="sect3" title="svnserve as a launchd job"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.svnserve.invoking.launchd"></a>svnserve as a launchd job</h4></div></div></div><p>Mac OS X (10.4 and higher) uses <span class="command"><strong>launchd</strong></span>
to manage processes (including daemons) both system-wide and
per-user. A <span class="command"><strong>launchd</strong></span> job is specified by
parameters in an XML property list file, and
the <span class="command"><strong>launchctl</strong></span> command is used to manage
the lifecycle of those jobs.</p><p>When configured to run as a <span class="command"><strong>launchd</strong></span>
job, <span class="command"><strong>svnserve</strong></span> is automatically launched
on demand whenever incoming Subversion
<code class="literal">svn://</code> network traffic needs to be
handled. This is far more convenient than a configuration
which requires you to manually invoke
<span class="command"><strong>svnserve</strong></span> as a long-running
background process.</p><p>To configure <span class="command"><strong>svnserve</strong></span> as
a <span class="command"><strong>launchd</strong></span> job, first create a job
definition file named
<code class="filename">/Library/LaunchDaemons/org.apache.subversion.svnserve.plist</code>.
<a class="xref" href="#svn.serverconfig.svnserve.invoking.launchd.ex-1" title="Example 6.1. A sample svnserve launchd job definition">Example 6.1, “A sample svnserve launchd job definition”</a>
provides an example of such a file.</p><div class="example"><a id="svn.serverconfig.svnserve.invoking.launchd.ex-1"></a><p class="title"><strong>Example 6.1. A sample svnserve launchd job definition</strong></p><div class="example-contents"><pre class="programlisting">
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN"
"http://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0">
<dict>
<key>Label</key>
<string>org.apache.subversion.svnserve</string>
<key>ServiceDescription</key>
<string>Host Subversion repositories using svn:// scheme</string>
<key>ProgramArguments</key>
<array>
<string>/usr/bin/svnserve</string>
<string>--inetd</string>
<string>--root=/var/svn</string>
</array>
<key>UserName</key>
<string>svn</string>
<key>GroupName</key>
<string>svn</string>
<key>inetdCompatibility</key>
<dict>
<key>Wait</key>
<false/>
</dict>
<key>Sockets</key>
<dict>
<key>Listeners</key>
<array>
<dict>
<key>SockServiceName</key>
<string>svn</string>
<key>Bonjour</key>
<true/>
</dict>
</array>
</dict>
</dict>
</plist>
</pre></div></div><br class="example-break"></br><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>The <span class="command"><strong>launchd</strong></span> system can be somewhat
challenging to learn. Fortunately, documentation exists
for the commands described in this section. For example,
run <strong class="userinput"><code>man launchd</code></strong> from the command
line to see the manual page for <span class="command"><strong>launchd</strong></span>
itself, <strong class="userinput"><code>man launchd.plist</code></strong> to read
about the job definition format, etc.</p></td></tr></table></div><p>Once your job definition file is created, you can
activate the job using <span class="command"><strong>launchctl load</strong></span>:</p><div class="informalexample"><pre class="screen">
$ sudo launchctl load \
-w /Library/LaunchDaemons/org.apache.subversion.svnserve.plist
</pre></div><p>To be clear, this action doesn't actually launch
<span class="command"><strong>svnserve</strong></span> yet. It simply tells
<span class="command"><strong>launchd</strong></span> how to fire up
<span class="command"><strong>svnserve</strong></span> when incoming networking traffic
arrives on the <code class="literal">svn</code> network port; it will
be terminated it after the traffic has been handled.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>Because we want <span class="command"><strong>svnserve</strong></span> to be a
system-wide daemon process, we need to
use <span class="command"><strong>sudo</strong></span> to manage this job as an
administrator. Note also that the
<code class="literal">UserName</code>
and <code class="literal">GroupName</code> keys in the definition
file are optional—if omitted, the job will run as
the user who loaded the job.</p></td></tr></table></div><p>Deactivating the job is just as easy to do—use
<span class="command"><strong>launchctl unload</strong></span>:</p><div class="informalexample"><pre class="screen">
$ sudo launchctl unload \
-w /Library/LaunchDaemons/org.apache.subversion.svnserve.plist
</pre></div><p><span class="command"><strong>launchctl</strong></span> also provides a way for you
to query the status of jobs. If the job is loaded, there
will be line which matches the <code class="literal">Label</code>
specified in the job definition file:</p><div class="informalexample"><pre class="screen">
$ sudo launchctl list | grep org.apache.subversion.svnserve
- 0 org.apache.subversion.svnserve
$
</pre></div></div></div><div class="sect2" title="Built-in Authentication and Authorization"><div class="titlepage"><div><div><h3 class="title"><a id="svn.serverconfig.svnserve.auth"></a>Built-in Authentication and Authorization</h3></div></div></div><p>When a client connects to an <span class="command"><strong>svnserve</strong></span>
process, the following things happen:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>The client selects a specific repository.</p></li><li class="listitem"><p>The server processes the repository's
<code class="filename">conf/svnserve.conf</code> file and begins to
enforce any authentication and authorization policies it
describes.</p></li><li class="listitem"><p>Depending on the defined policies, one of the
following may occur:</p><div class="itemizedlist"><ul class="itemizedlist" type="circle"><li class="listitem"><p>The client may be allowed to make requests
anonymously, without ever receiving an authentication
challenge.</p></li><li class="listitem"><p>The client may be challenged for authentication at
any time.</p></li><li class="listitem"><p>If operating in tunnel mode, the client will
declare itself to be already externally authenticated
(typically by SSH).</p></li></ul></div></li></ul></div><p>The <span class="command"><strong>svnserve</strong></span> server, by default, knows
only how to send a CRAM-MD5<sup>[<a id="idp13371568" href="#ftn.idp13371568" class="footnote">52</a>]</sup> authentication challenge. In essence,
the server sends a small amount of data to the client. The
client uses the MD5 hash algorithm to create a fingerprint of
the data and password combined, and then sends the fingerprint
as a response. The server performs the same computation with
the stored password to verify that the result is identical.
<span class="emphasis"><em>At no point does the actual password travel over the
network.</em></span></p><p>If your <span class="command"><strong>svnserve</strong></span> server was built with
SASL support, it not only knows how to send CRAM-MD5 challenges,
but also likely knows a whole host of other authentication
mechanisms. See <a class="xref" href="#svn.serverconfig.svnserve.sasl" title="Using svnserve with SASL">the section called “Using <span class="command"><strong>svnserve</strong></span> with SASL”</a> later in this
chapter to learn how to configure SASL authentication and
encryption.</p><p>It's also possible, of course, for the client to be
externally authenticated via a tunnel agent, such as
<span class="command"><strong>ssh</strong></span>. In that case, the server simply
examines the user it's running as, and uses this name as the
authenticated username. For more on this, see the later
section, <a class="xref" href="#svn.serverconfig.svnserve.sshauth" title="Tunneling over SSH">the section called “Tunneling over SSH”</a>.</p><p>As you've already guessed, a repository's
<code class="filename">svnserve.conf</code> file is the central
mechanism for controlling authentication and authorization
policies. The file has the same format as other configuration
files (see <a class="xref" href="#svn.advanced.confarea" title="Runtime Configuration Area">the section called “Runtime Configuration Area”</a>):
section names are marked by square brackets
(<code class="literal">[</code> and <code class="literal">]</code>), comments
begin with hashes (<code class="literal">#</code>), and each section
contains specific variables that can be set (<code class="literal">variable
= value</code>). Let's walk through these files and learn
how to use them.</p><div class="sect3" title="Create a users file and realm"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.svnserve.auth.users"></a>Create a users file and realm</h4></div></div></div><p>For now, the <code class="literal">[general]</code> section of
<code class="filename">svnserve.conf</code> has all the variables you
need. Begin by changing the values of those variables:
choose a name for a file that will contain your usernames
and passwords and choose an authentication realm:</p><div class="informalexample"><pre class="programlisting">
[general]
password-db = userfile
realm = example realm
</pre></div><p>The <code class="literal">realm</code> is a name that you define.
It tells clients which sort of <span class="quote">“<span class="quote">authentication
namespace</span>”</span> they're connecting to; the Subversion
client displays it in the authentication prompt and uses it
as a key (along with the server's hostname and port) for
caching credentials on disk (see <a class="xref" href="#svn.serverconfig.netmodel.credcache" title="Caching credentials">the section called “Caching credentials”</a>). The
<code class="literal">password-db</code> variable points to a separate
file that contains a list of usernames and passwords, using
the same familiar format. For example:</p><div class="informalexample"><pre class="programlisting">
[users]
harry = foopassword
sally = barpassword
</pre></div><p>The value of <code class="literal">password-db</code> can be an
absolute or relative path to the users file. For many
admins, it's easy to keep the file right in the
<code class="filename">conf/</code> area of the repository, alongside
<code class="filename">svnserve.conf</code>. On the other hand, it's
possible you may want to have two or more repositories share
the same users file; in that case, the file should probably
live in a more public place. The repositories sharing the
users file should also be configured to have the same realm,
since the list of users essentially defines an
authentication realm. Wherever the file lives, be sure to
set the file's read and write permissions appropriately. If
you know which user(s) <span class="command"><strong>svnserve</strong></span> will run
as, restrict read access to the users file as necessary.</p></div><div class="sect3" title="Set access controls"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.svnserve.auth.general"></a>Set access controls</h4></div></div></div><p>There are two more variables to set in the
<code class="filename">svnserve.conf</code> file: they determine what
unauthenticated (anonymous) and authenticated users are
allowed to do. The variables <code class="literal">anon-access</code>
and <code class="literal">auth-access</code> can be set to the value
<code class="literal">none</code>, <code class="literal">read</code>, or
<code class="literal">write</code>. Setting the value to
<code class="literal">none</code> prohibits both reading and writing;
<code class="literal">read</code> allows read-only access to the
repository, and <code class="literal">write</code> allows complete
read/write access to the repository. For example:</p><div class="informalexample"><pre class="programlisting">
[general]
password-db = userfile
realm = example realm
# anonymous users can only read the repository
anon-access = read
# authenticated users can both read and write
auth-access = write
</pre></div><p>The example settings are, in fact, the default values of
the variables, should you forget to define them. If you
want to be even more conservative, you can block anonymous
access completely:</p><div class="informalexample"><pre class="programlisting">
[general]
password-db = userfile
realm = example realm
# anonymous users aren't allowed
anon-access = none
# authenticated users can both read and write
auth-access = write
</pre></div><p>The server process understands not only
these <span class="quote">“<span class="quote">blanket</span>”</span> access controls to the
repository, but also finer-grained access restrictions
placed on specific files and directories within the
repository. To make use of this feature, you need to define
a file containing more detailed rules, and then set
the <code class="literal">authz-db</code> variable to point to
it:</p><div class="informalexample"><pre class="programlisting">
[general]
password-db = userfile
realm = example realm
# Specific access rules for specific locations
authz-db = authzfile
</pre></div><p>We discuss the syntax of the <code class="filename">authzfile</code> file
in detail later in this chapter, in
<a class="xref" href="#svn.serverconfig.pathbasedauthz" title="Path-Based Authorization">the section called “Path-Based Authorization”</a>. Note
that the <code class="literal">authz-db</code> variable isn't mutually
exclusive with the <code class="literal">anon-access</code>
and <code class="literal">auth-access</code> variables; if all the
variables are defined at once, <span class="emphasis"><em>all</em></span>
of the rules must be satisfied before access is allowed.</p></div></div><div class="sect2" title="Using svnserve with SASL"><div class="titlepage"><div><div><h3 class="title"><a id="svn.serverconfig.svnserve.sasl"></a>Using <span class="command"><strong>svnserve</strong></span> with SASL</h3></div></div></div><p>For many teams, the built-in CRAM-MD5 authentication is
all they need from <span class="command"><strong>svnserve</strong></span>. However, if
your server (and your Subversion clients) were built with the
Cyrus Simple Authentication and Security Layer (SASL) library,
you have a number of authentication and encryption
options available to you.</p><div class="sidebar" title="What Is SASL?"><div class="titlepage"><div><div><p class="title"><strong>What Is SASL?</strong></p></div></div></div><p>The Cyrus Simple Authentication and Security Layer is
open source software written by Carnegie Mellon University.
It adds generic authentication and encryption capabilities
to any network protocol, and as of Subversion 1.5 and later,
both the <span class="command"><strong>svnserve</strong></span> server
and <span class="command"><strong>svn</strong></span> client know how to make use of
this library. It may or may not be available to you: if
you're building Subversion yourself, you'll need to have at
least version 2.1 of SASL installed on your system, and
you'll need to make sure that it's detected during
Subversion's build process. The Subversion command-line
client will report the availability of Cyrus SASL when you
run <strong class="userinput"><code>svn --version</code></strong>; if you're using
some other Subversion client, you might need to check with
the package maintainer as to whether SASL support was
compiled in.</p><p>SASL comes with a number of pluggable modules that
represent different authentication systems: Kerberos
(GSSAPI), NTLM, One-Time-Passwords (OTP), DIGEST-MD5, LDAP,
Secure-Remote-Password (SRP), and others. Certain
mechanisms may or may not be available to you; be sure to
check which modules are provided.</p><p>You can download Cyrus SASL (both code and
documentation) from
<a class="ulink" href="http://asg.web.cmu.edu/sasl/sasl-library.html" target="_top">http://asg.web.cmu.edu/sasl/sasl-library.html</a>.</p></div><p>Normally, when a subversion client connects to
<span class="command"><strong>svnserve</strong></span>, the server sends a greeting that
advertises a list of the capabilities it supports, and the
client responds with a similar list of capabilities. If the
server is configured to require authentication, it then sends
a challenge that lists the authentication mechanisms
available; the client responds by choosing one of the
mechanisms, and then authentication is carried out in some
number of round-trip messages. Even when SASL capabilities
aren't present, the client and server inherently know how to
use the CRAM-MD5 and ANONYMOUS mechanisms (see
<a class="xref" href="#svn.serverconfig.svnserve.auth" title="Built-in Authentication and Authorization">the section called “Built-in Authentication and Authorization”</a>). If server
and client were linked against SASL, a number of other
authentication mechanisms may also be available. However,
you'll need to explicitly configure SASL on the server side to
advertise them.</p><div class="sect3" title="Authenticating with SASL"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.svnserve.sasl.authn"></a>Authenticating with SASL</h4></div></div></div><p>To activate specific SASL mechanisms on the server,
you'll need to do two things. First, create
a <code class="literal">[sasl]</code> section in your
repository's <code class="filename">svnserve.conf</code> file with an
initial key-value pair:</p><div class="informalexample"><pre class="programlisting">
[sasl]
use-sasl = true
</pre></div><p>Second, create a main SASL configuration file
called <code class="filename">svn.conf</code> in a place where the
SASL library can find it—typically in the directory
where SASL plug-ins are located. You'll have to locate the
plug-in directory on your particular system, such
as <code class="filename">/usr/lib/sasl2/</code>
or <code class="filename">/etc/sasl2/</code>. (Note that this
is <span class="emphasis"><em>not</em></span>
the <code class="filename">svnserve.conf</code> file that lives
within a repository!)</p><p>On a Windows server, you'll also have to edit the system
registry (using a tool such as <span class="command"><strong>regedit</strong></span>)
to tell SASL where to find things. Create a registry key
named <code class="literal">[HKEY_LOCAL_MACHINE\SOFTWARE\Carnegie
Mellon\Project Cyrus\SASL Library]</code>, and place two
keys inside it: a key called <code class="literal">SearchPath</code>
(whose value is a path to the directory containing the SASL
<code class="filename">sasl*.dll</code> plug-in libraries), and a key
called
<code class="literal">ConfFile</code> (whose value is a path to the
parent directory containing
the <code class="filename">svn.conf</code> file you created).</p><p>Because SASL provides so many different kinds of
authentication mechanisms, it would be foolish (and far
beyond the scope of this book) to try to describe every
possible server-side configuration. Instead, we recommend
that you read the documentation supplied in the
<code class="filename">doc/</code> subdirectory of the SASL source
code. It goes into great detail about every mechanism and
how to configure the server appropriately for each. For the
purposes of this discussion, we'll just demonstrate a simple
example of configuring the DIGEST-MD5 mechanism. For
example, if your <code class="filename">subversion.conf</code>
(or <code class="filename">svn.conf</code>) file contains the
following:</p><div class="informalexample"><pre class="programlisting">
pwcheck_method: auxprop
auxprop_plugin: sasldb
sasldb_path: /etc/my_sasldb
mech_list: DIGEST-MD5
</pre></div><p>you've told SASL to advertise the DIGEST-MD5
mechanism to clients and to check user passwords against a
private password database located
at <code class="filename">/etc/my_sasldb</code>. A system
administrator can then use
the <span class="command"><strong>saslpasswd2</strong></span> program to add or modify
usernames and passwords in the database:</p><div class="informalexample"><pre class="screen">
$ saslpasswd2 -c -f /etc/my_sasldb -u realm username
</pre></div><p>A few words of warning: first, make sure the
<span class="quote">“<span class="quote">realm</span>”</span> argument
to <span class="command"><strong>saslpasswd2</strong></span> matches the same realm
you've defined in your
repository's <code class="filename">svnserve.conf</code> file; if
they don't match, authentication will fail. Also, due to a
shortcoming in SASL, the common realm must be a string with
no space characters. Finally, if you decide to go with the
standard SASL password database, make sure
the <span class="command"><strong>svnserve</strong></span> program has read access to
the file (and possibly write access as well, if you're using
a mechanism such as OTP).</p><p>This is just one simple way of configuring SASL. Many
other authentication mechanisms are available, and passwords
can be stored in other places such as in LDAP or a SQL
database. Consult the full SASL documentation for
details.</p><p>Remember that if you configure your server to only allow
certain SASL authentication mechanisms, this forces all
connecting clients to have SASL support as well. Any
Subversion client built without SASL support (which includes
all pre-1.5 clients) will be unable to authenticate. On the
one hand, this sort of restriction may be exactly what you
want (<span class="quote">“<span class="quote">My clients must all use Kerberos!</span>”</span>).
However, if you still want non-SASL clients to be able to
authenticate, be sure to advertise the CRAM-MD5 mechanism as
an option. All clients are able to use CRAM-MD5, whether
they have SASL capabilities or not.</p></div><div class="sect3" title="SASL encryption"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.svnserve.sasl.encryption"></a>SASL encryption</h4></div></div></div><p>SASL is also able to perform data encryption if a
particular mechanism supports it. The built-in CRAM-MD5
mechanism doesn't support encryption, but DIGEST-MD5 does,
and mechanisms such as SRP actually require use of the
OpenSSL library. To enable or disable different levels of
encryption, you can set two values in your repository's
<code class="filename">svnserve.conf</code> file:</p><div class="informalexample"><pre class="programlisting">
[sasl]
use-sasl = true
min-encryption = 128
max-encryption = 256
</pre></div><p>The <code class="literal">min-encryption</code> and
<code class="literal">max-encryption</code> variables control the
level of encryption demanded by the server. To disable
encryption completely, set both values to 0. To enable
simple checksumming of data (i.e., prevent tampering and
guarantee data integrity without encryption), set both
values to 1. If you wish to allow—but not
require—encryption, set the minimum value to 0, and
the maximum value to some bit length. To require encryption
unconditionally, set both values to numbers greater than 1.
In our previous example, we require clients to do at least
128-bit encryption, but no more than 256-bit
encryption.</p></div></div><div class="sect2" title="Tunneling over SSH"><div class="titlepage"><div><div><h3 class="title"><a id="svn.serverconfig.svnserve.sshauth"></a>Tunneling over SSH</h3></div></div></div><p><span class="command"><strong>svnserve</strong></span>'s built-in authentication (and
SASL support) can be very handy, because it avoids the need to
create real system accounts. On the other hand, some
administrators already have well-established SSH
authentication frameworks in place. In these situations, all
of the project's users already have system accounts and the
ability to <span class="quote">“<span class="quote">SSH into</span>”</span> the server machine.</p><p>It's easy to use SSH in conjunction with
<span class="command"><strong>svnserve</strong></span>. The client simply uses the
<code class="literal">svn+ssh://</code> URL scheme to connect:</p><div class="informalexample"><pre class="screen">
$ whoami
harry
$ svn list svn+ssh://host.example.com/repos/project
harryssh@host.example.com's password: *****
foo
bar
baz
…
</pre></div><p>In this example, the Subversion client is invoking a local
<span class="command"><strong>ssh</strong></span> process, connecting to
<code class="literal">host.example.com</code>, authenticating as the
user <code class="literal">harryssh</code> (according to SSH user
configuration), then spawning a private
<span class="command"><strong>svnserve</strong></span> process on the remote machine
running as the user <code class="literal">harryssh</code>. The
<span class="command"><strong>svnserve</strong></span> command is being invoked in tunnel
mode (<code class="option">-t</code>), and its network protocol is being
<span class="quote">“<span class="quote">tunneled</span>”</span> over the encrypted connection by
<span class="command"><strong>ssh</strong></span>, the tunnel agent.
If the client performs a commit, the authenticated username
<code class="literal">harryssh</code> will be used as the author
of the new revision.</p><p>The important thing to understand here is that the
Subversion client is <span class="emphasis"><em>not</em></span> connecting to a
running <span class="command"><strong>svnserve</strong></span> daemon. This method of
access doesn't require a daemon, nor does it notice one if
present. It relies wholly on the ability of
<span class="command"><strong>ssh</strong></span> to spawn a temporary
<span class="command"><strong>svnserve</strong></span> process, which then terminates
when the network connection is closed.</p><p>When using <code class="literal">svn+ssh://</code> URLs to access a
repository, remember that it's the <span class="command"><strong>ssh</strong></span>
program prompting for authentication, and
<span class="emphasis"><em>not</em></span> the <span class="command"><strong>svn</strong></span> client
program. That means there's no automatic password-caching
going on (see <a class="xref" href="#svn.serverconfig.netmodel.credcache" title="Caching credentials">the section called “Caching credentials”</a>).
The Subversion client often makes multiple connections to the
repository, though users don't normally notice this due to the
password caching feature. When using
<code class="literal">svn+ssh://</code> URLs, however, users may be
annoyed by <span class="command"><strong>ssh</strong></span> repeatedly asking for a
password for every outbound connection. The solution is to
use a separate SSH password-caching tool such as
<span class="command"><strong>ssh-agent</strong></span> on a Unix-like system, or
<span class="command"><strong>pageant</strong></span> on Windows.</p><p>When running over a tunnel, authorization is primarily
controlled by operating system permissions to the repository's
database files; it's very much the same as if Harry were
accessing the repository directly via a
<code class="literal">file://</code> URL. If multiple system users are
going to be accessing the repository directly, you may want to
place them into a common group, and you'll need to be careful
about umasks (be sure to read <a class="xref" href="#svn.serverconfig.multimethod" title="Supporting Multiple Repository Access Methods">the section called “Supporting Multiple Repository Access Methods”</a> later in this
chapter). But even in the case of tunneling, you can still use the
<code class="filename">svnserve.conf</code> file to block access, by
simply setting <code class="literal">auth-access = read</code>
or <code class="literal">auth-access = none</code>.<sup>[<a id="idp13490048" href="#ftn.idp13490048" class="footnote">53</a>]</sup></p><p>You'd think that the story of SSH tunneling would end
here, but it doesn't. Subversion allows you to create custom
tunnel behaviors in your runtime <code class="filename">config</code>
file (see <a class="xref" href="#svn.advanced.confarea" title="Runtime Configuration Area">the section called “Runtime Configuration Area”</a>). For
example, suppose you want to use RSH instead of
SSH.<sup>[<a id="idp13493328" href="#ftn.idp13493328" class="footnote">54</a>]</sup> In
the <code class="literal">[tunnels]</code> section of your
<code class="filename">config</code> file, simply define it like
this:</p><div class="informalexample"><pre class="programlisting">
[tunnels]
rsh = rsh
</pre></div><p>And now, you can use this new tunnel definition by using a
URL scheme that matches the name of your new variable:
<code class="literal">svn+rsh://host/path</code>. When using the new
URL scheme, the Subversion client will actually be running the
command <strong class="userinput"><code>rsh host svnserve -t</code></strong> behind the
scenes. If you include a username in the URL (e.g.,
<code class="literal">svn+rsh://username@host/path</code>), the client
will also include that in its command (<strong class="userinput"><code>rsh
username@host svnserve -t</code></strong>). But you can define new
tunneling schemes to be much more clever than that:</p><div class="informalexample"><pre class="programlisting">
[tunnels]
joessh = $JOESSH /opt/alternate/ssh -p 29934
</pre></div><p>This example demonstrates a couple of things. First, it
shows how to make the Subversion client launch a very specific
tunneling binary (the one located at
<code class="filename">/opt/alternate/ssh</code>) with specific
options. In this case, accessing an
<code class="literal">svn+joessh://</code> URL would invoke the
particular SSH binary with <code class="option">-p 29934</code> as
arguments—useful if you want the tunnel program to
connect to a nonstandard port.</p><p>Second, it shows how to define a custom environment
variable that can override the name of the tunneling program.
Setting the <code class="literal">SVN_SSH</code> environment variable is
a convenient way to override the default SSH tunnel agent.
But if you need to have several different overrides for
different servers, each perhaps contacting a different port or
passing a different set of options to SSH, you can use the
mechanism demonstrated in this example. Now if we were to set
the <code class="literal">JOESSH</code> environment variable, its value
would override the entire value of the tunnel
variable—<span class="command"><strong>$JOESSH</strong></span> would be executed
instead of <strong class="userinput"><code>/opt/alternate/ssh -p
29934</code></strong>.</p></div><div class="sect2" title="SSH Configuration Tricks"><div class="titlepage"><div><div><h3 class="title"><a id="svn.serverconfig.svnserve.sshtricks"></a>SSH Configuration Tricks</h3></div></div></div><p>It's possible to control not only the way in which the
client invokes <span class="command"><strong>ssh</strong></span>, but also to control
the behavior of <span class="command"><strong>sshd</strong></span> on your server
machine. In this section, we'll show how to control the
exact <span class="command"><strong>svnserve</strong></span> command executed
by <span class="command"><strong>sshd</strong></span>, as well as how to have multiple
users share a single system account.</p><div class="sect3" title="Initial setup"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.svnserve.sshtricks.setup"></a>Initial setup</h4></div></div></div><p>To begin, locate the home directory of the account
you'll be using to launch <span class="command"><strong>svnserve</strong></span>. Make
sure the account has an SSH public/private keypair
installed, and that the user can log in via public-key
authentication. Password authentication will not work,
since all of the following SSH tricks revolve around using
the SSH <code class="filename">authorized_keys</code> file.</p><p>If it doesn't already exist, create the
<code class="filename">authorized_keys</code> file (on Unix,
typically <code class="filename">~/.ssh/authorized_keys</code>).
Each line in this file describes a public key that is
allowed to connect. The lines are typically of the
form:</p><div class="informalexample"><pre class="programlisting">
ssh-dsa AAAABtce9euch… user@example.com
</pre></div><p>The first field describes the type of key, the second
field is the base64-encoded key itself, and the third field
is a comment. However, it's a lesser known fact that the
entire line can be preceded by a <code class="literal">command</code>
field:</p><div class="informalexample"><pre class="programlisting">
command="program" ssh-dsa AAAABtce9euch… user@example.com
</pre></div><p>When the <code class="literal">command</code> field is set, the
SSH daemon will run the named program instead of the
typical tunnel-mode <span class="command"><strong>svnserve</strong></span> invocation that the
Subversion client asks for. This opens the door to a number
of server-side tricks. In the following examples, we
abbreviate the lines of the file as:</p><div class="informalexample"><pre class="programlisting">
command="program" TYPE KEY COMMENT
</pre></div></div><div class="sect3" title="Controlling the invoked command"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.svnserve.sshtricks.fixedcmd"></a>Controlling the invoked command</h4></div></div></div><p>Because we can specify the executed server-side command,
it's easy to name a specific <span class="command"><strong>svnserve</strong></span>
binary to run and to pass it extra arguments:</p><div class="informalexample"><pre class="programlisting">
command="/path/to/svnserve -t -r /virtual/root" TYPE KEY COMMENT
</pre></div><p>In this example, <code class="filename">/path/to/svnserve</code>
might be a custom wrapper script
around <span class="command"><strong>svnserve</strong></span> which sets the umask (see
<a class="xref" href="#svn.serverconfig.multimethod" title="Supporting Multiple Repository Access Methods">the section called “Supporting Multiple Repository Access Methods”</a>). It also
shows how to anchor <span class="command"><strong>svnserve</strong></span> in a virtual
root directory, just as one often does when
running <span class="command"><strong>svnserve</strong></span> as a daemon process.
This might be done either to restrict access to parts of the
system, or simply to relieve the user of having to type an
absolute path in the <code class="literal">svn+ssh://</code>
URL.</p><p>It's also possible to have multiple users share a single
account. Instead of creating a separate system account for
each user, generate a public/private key pair for each
person. Then place each public key into
the <code class="filename">authorized_keys</code> file, one per
line, and use the <code class="option">--tunnel-user</code>
option:</p><div class="informalexample"><pre class="programlisting">
command="svnserve -t --tunnel-user=harry" TYPE1 KEY1 harry@example.com
command="svnserve -t --tunnel-user=sally" TYPE2 KEY2 sally@example.com
</pre></div><p>This example allows both Harry and Sally to connect to
the same account via public key authentication. Each of
them has a custom command that will be executed;
the <code class="option">--tunnel-user</code> option
tells <span class="command"><strong>svnserve</strong></span> to assume that the named
argument is the authenticated user. Without
<code class="option">--tunnel-user</code>, it would appear as though
all commits were coming from the one shared system
account.</p><p>A final word of caution: giving a user access to the
server via public-key in a shared account might still allow
other forms of SSH access, even if you've set
the <code class="literal">command</code> value
in <code class="filename">authorized_keys</code>. For example, the
user may still get shell access through SSH or be able to
perform X11 or general port forwarding through your server.
To give the user as little permission as possible, you may
want to specify a number of restrictive options immediately
after the <code class="literal">command</code>:</p><div class="informalexample"><pre class="programlisting">
command="svnserve -t --tunnel-user=harry",no-port-forwarding,no-agent-forw
arding,no-X11-forwarding,no-pty TYPE1 KEY1 harry@example.com
</pre></div><p>Note that this all must be on one line—truly on
one line—since SSH <code class="filename">authorized_keys</code>
files do not even allow the conventional backslash character
(<code class="literal">\</code>) for line continuation. The only
reason we've shown it with a line break is to fit it on
the physical page of a book.</p></div></div></div><div class="sect1" title="httpd, the Apache HTTP Server"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.serverconfig.httpd"></a>httpd, the Apache HTTP Server</h2></div></div></div><p>The Apache HTTP Server is a <span class="quote">“<span class="quote">heavy-duty</span>”</span>
network server that Subversion can leverage. Via a custom
module, <span class="command"><strong>httpd</strong></span> makes Subversion repositories
available to clients via the WebDAV/DeltaV<sup>[<a id="idp13550992" href="#ftn.idp13550992" class="footnote">55</a>]</sup>
protocol, which is an extension to HTTP 1.1. This protocol
takes the ubiquitous HTTP protocol that is the core of the World
Wide Web, and adds writing—specifically, versioned
writing—capabilities. The result is a standardized,
robust system that is conveniently packaged as part of the
Apache 2.0 software, supported by numerous operating systems and
third-party products, and doesn't require network administrators
to open up yet another custom port.<sup>[<a id="idp13552256" href="#ftn.idp13552256" class="footnote">56</a>]</sup> While an Apache-Subversion
server has more features than <span class="command"><strong>svnserve</strong></span>, it's
also a bit more difficult to set up. With flexibility often
comes more complexity.</p><p>Much of the following discussion includes references to
Apache configuration directives. While some examples are given
of the use of these directives, describing them in full is
outside the scope of this chapter. The Apache team maintains
excellent documentation, publicly available on their web site at
<a class="ulink" href="http://httpd.apache.org" target="_top">http://httpd.apache.org</a>. For example, a general
reference for the configuration directives is located at
<a class="ulink" href="http://httpd.apache.org/docs-2.0/mod/directives.html" target="_top">http://httpd.apache.org/docs-2.0/mod/directives.html</a>.</p><p>Also, as you make changes to your Apache setup, it is likely
that somewhere along the way a mistake will be made. If you are
not already familiar with Apache's logging subsystem, you should
become aware of it. In your <code class="filename">httpd.conf</code>
file are directives that specify the on-disk locations of the
access and error logs generated by Apache (the
<code class="literal">CustomLog</code> and <code class="literal">ErrorLog</code>
directives, respectively).
Subversion's <span class="command"><strong>mod_dav_svn</strong></span> uses Apache's error
logging interface as well. You can always browse the contents
of those files for information that might reveal the source of a
problem that is not clearly noticeable otherwise.</p><div class="sect2" title="Prerequisites"><div class="titlepage"><div><div><h3 class="title"><a id="svn.serverconfig.httpd.prereqs"></a>Prerequisites</h3></div></div></div><p>To network your repository over HTTP, you basically need
four components, available in two packages. You'll need
Apache <span class="command"><strong>httpd</strong></span> 2.0 or newer, the
<span class="command"><strong>mod_dav</strong></span> DAV module that comes with it,
Subversion, and the <span class="command"><strong>mod_dav_svn</strong></span>
filesystem provider module distributed with Subversion.
Once you have all of those components, the process of
networking your repository is as simple as:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>Getting httpd up and running with
the <span class="command"><strong>mod_dav</strong></span> module</p></li><li class="listitem"><p>Installing the <span class="command"><strong>mod_dav_svn</strong></span> backend
to <span class="command"><strong>mod_dav</strong></span>, which uses Subversion's
libraries to access the repository</p></li><li class="listitem"><p>Configuring your <code class="filename">httpd.conf</code>
file to export (or expose) the repository</p></li></ul></div><p>You can accomplish the first two items either by
compiling <span class="command"><strong>httpd</strong></span> and Subversion from
source code or by installing prebuilt binary packages of
them on your system. For the most up-to-date information on
how to compile Subversion for use with the Apache HTTP Server,
as well as how to compile and configure Apache itself for
this purpose, see the <code class="filename">INSTALL</code> file in
the top level of the Subversion source code tree.</p></div><div class="sect2" title="Basic Apache Configuration"><div class="titlepage"><div><div><h3 class="title"><a id="svn.serverconfig.httpd.basic"></a>Basic Apache Configuration</h3></div></div></div><p>Once you have all the necessary components installed on
your system, all that remains is the configuration of Apache
via its <code class="filename">httpd.conf</code> file. Instruct Apache
to load the <span class="command"><strong>mod_dav_svn</strong></span> module using the
<code class="literal">LoadModule</code> directive. This directive must
precede any other Subversion-related configuration items. If
your Apache was installed using the default layout, your
<span class="command"><strong>mod_dav_svn</strong></span> module should have been
installed in the <code class="filename">modules</code> subdirectory of
the Apache install location (often
<code class="filename">/usr/local/apache2</code>). The
<code class="literal">LoadModule</code> directive has a simple syntax,
mapping a named module to the location of a shared library on
disk:</p><div class="informalexample"><pre class="programlisting">
LoadModule dav_svn_module modules/mod_dav_svn.so
</pre></div><p>Apache interprets the <code class="literal">LoadModule</code>
configuration item's library path as relative to its own
server root. If configured as previously shown, Apache will
look for the Subversion DAV module shared library in its
own <code class="filename">modules/</code> subdirectory. Depending on
how Subversion was installed on your system, you might need to
specify a different path for this library altogether, perhaps
even an absolute path such as in the following example:</p><div class="informalexample"><pre class="programlisting">
LoadModule dav_svn_module C:/Subversion/lib/mod_dav_svn.so
</pre></div><p>Note that if <span class="command"><strong>mod_dav</strong></span> was compiled as a
shared object (instead of statically linked directly to the
<span class="command"><strong>httpd</strong></span> binary), you'll need a similar
<code class="literal">LoadModule</code> statement for it, too. Be sure
that it comes before the <span class="command"><strong>mod_dav_svn</strong></span> line:</p><div class="informalexample"><pre class="programlisting">
LoadModule dav_module modules/mod_dav.so
LoadModule dav_svn_module modules/mod_dav_svn.so
</pre></div><p>At a later location in your configuration file, you now
need to tell Apache where you keep your Subversion repository
(or repositories). The <code class="literal">Location</code> directive
has an XML-like notation, starting with an opening tag and
ending with a closing tag, with various other configuration
directives in the middle. The purpose of the
<code class="literal">Location</code> directive is to instruct Apache to
do something special when handling requests that are directed
at a given URL or one of its children. In the case of
Subversion, you want Apache to simply hand off support for
URLs that point at versioned resources to the DAV layer. You
can instruct Apache to delegate the handling of all URLs whose
path portions (the part of the URL that follows the server's
name and the optional port number) begin with
<code class="filename">/repos/</code> to a DAV provider whose
repository is located at
<code class="filename">/var/svn/repository</code> using the
following <code class="filename">httpd.conf</code> syntax:</p><div class="informalexample"><pre class="programlisting">
<Location /repos>
DAV svn
SVNPath /var/svn/repository
</Location>
</pre></div><p>If you plan to support multiple Subversion repositories
that will reside in the same parent directory on your local
disk, you can use an alternative
directive—<code class="literal">SVNParentPath</code>—to
indicate that common parent directory. For example, if you
know you will be creating multiple Subversion repositories in
a directory <code class="filename">/var/svn</code> that would be
accessed via URLs such as
<code class="uri">http://my.server.com/svn/repos1</code>,
<code class="uri">http://my.server.com/svn/repos2</code>, and so on, you
could use the <code class="filename">httpd.conf</code> configuration
syntax in the following example:</p><div class="informalexample"><pre class="programlisting">
<Location /svn>
DAV svn
# Automatically map any "/svn/foo" URL to repository /var/svn/foo
SVNParentPath /var/svn
</Location>
</pre></div><p>Using this syntax, Apache will delegate the
handling of all URLs whose path portions begin with
<code class="filename">/svn/</code> to the Subversion DAV provider,
which will then assume that any items in the directory
specified by the <code class="literal">SVNParentPath</code> directive
are actually Subversion repositories. This is a particularly
convenient syntax in that, unlike the use of the
<code class="literal">SVNPath</code> directive, you don't have to
restart Apache to add or remove hosted repositories.</p><p>Be sure that when you define your new
<code class="literal">Location</code>, it doesn't overlap with other
exported locations. For example, if your main
<code class="literal">DocumentRoot</code> is exported to
<code class="filename">/www</code>, do not export a Subversion
repository in <code class="literal"><Location /www/repos></code>.
If a request comes in for the URI
<code class="filename">/www/repos/foo.c</code>, Apache won't know
whether to look for a file <code class="filename">repos/foo.c</code> in
the <code class="literal">DocumentRoot</code>, or whether to delegate
<span class="command"><strong>mod_dav_svn</strong></span> to return
<code class="filename">foo.c</code> from the Subversion repository.
The result is often an error from the server of the form
<code class="literal">301 Moved Permanently</code>.</p><div class="sidebar" title="Server Names and the COPY Request"><div class="titlepage"><div><div><p class="title"><strong>Server Names and the COPY Request</strong></p></div></div></div><p>Subversion makes use of the <code class="literal">COPY</code>
request type to perform server-side copies of files and
directories. As part of the sanity checking done by the
Apache modules, the source of the copy is expected to be
located on the same machine as the destination of the copy.
To satisfy this requirement, you might need to
tell <span class="command"><strong>mod_dav</strong></span> the name you use as the
hostname of your server. Generally, you can use
the <code class="literal">ServerName</code> directive in
<code class="filename">httpd.conf</code> to accomplish this.</p><div class="informalexample"><pre class="programlisting">
ServerName svn.example.com
</pre></div><p>If you are using Apache's virtual hosting support via
the <code class="literal">NameVirtualHost</code> directive, you may
need to use the <code class="literal">ServerAlias</code> directive to
specify additional names by which your server is known.
Again, refer to the Apache documentation for full
details.</p></div><p>At this stage, you should strongly consider the question
of permissions. If you've been running Apache for some time
now as your regular web server, you probably already have a
collection of content—web pages, scripts, and such.
These items have already been configured with a set of
permissions that allows them to work with Apache, or more
appropriately, that allows Apache to work with those files.
Apache, when used as a Subversion server, will also need the
correct permissions to read and write to your Subversion
repository.</p><p>You will need to determine a permission system setup that
satisfies Subversion's requirements without messing up any
previously existing web page or script installations. This
might mean changing the permissions on your Subversion
repository to match those in use by other things that Apache
serves for you, or it could mean using the
<code class="literal">User</code> and <code class="literal">Group</code>
directives in <code class="filename">httpd.conf</code> to specify that
Apache should run as the user and group that owns your
Subversion repository. There is no single correct way to set
up your permissions, and each administrator will have
different reasons for doing things a certain way. Just be
aware that permission-related problems are perhaps the most
common oversight when configuring a Subversion repository for
use with Apache.</p></div><div class="sect2" title="Authentication Options"><div class="titlepage"><div><div><h3 class="title"><a id="svn.serverconfig.httpd.authn"></a>Authentication Options</h3></div></div></div><p>At this point, if you configured
<code class="filename">httpd.conf</code> to contain something such as the
following:</p><div class="informalexample"><pre class="programlisting">
<Location /svn>
DAV svn
SVNParentPath /var/svn
</Location>
</pre></div><p>your repository is <span class="quote">“<span class="quote">anonymously</span>”</span>
accessible to the world. Until you configure some
authentication and authorization policies, the Subversion
repositories that you make available via the
<code class="literal">Location</code> directive will be generally
accessible to everyone. In other words:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>Anyone can use a Subversion client to check out a
working copy of a repository URL (or any of its
subdirectories).</p></li><li class="listitem"><p>Anyone can interactively browse the repository's
latest revision simply by pointing a web browser to
the repository URL.</p></li><li class="listitem"><p>Anyone can commit to the repository.</p></li></ul></div><p>Of course, you might have already set up
a <code class="filename">pre-commit</code> hook script to prevent
commits (see <a class="xref" href="#svn.reposadmin.create.hooks" title="Implementing Repository Hooks">the section called “Implementing Repository Hooks”</a>).
But as you read on, you'll see that it's also possible to use
Apache's built-in methods to restrict access in specific
ways.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>Requiring authentication defends against invalid users
directly accessing the repository, but does not guard the
privacy of valid users' network activity. See
<a class="xref" href="#svn.serverconfig.httpd.ssl" title="Protecting network traffic with SSL">the section called “Protecting network traffic with SSL”</a> for how to
configure your server to support SSL encryption, which can
provide that extra layer of protection.</p></td></tr></table></div><div class="sect3" title="Basic authentication"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.httpd.authn.basic"></a>Basic authentication</h4></div></div></div><p>The easiest way to authenticate a client is via the HTTP
Basic authentication mechanism, which simply uses a username
and password to verify a user's identity. Apache provides
the <span class="command"><strong>htpasswd</strong></span> utility<sup>[<a id="idp13639200" href="#ftn.idp13639200" class="footnote">57</a>]</sup> for managing files containing
usernames and passwords.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>Basic authentication is <span class="emphasis"><em>extremely</em></span>
insecure, because it sends passwords over the network
in very nearly plain text. See
<a class="xref" href="#svn.serverconfig.httpd.authn.digest" title="Digest authentication">the section called “Digest authentication”</a> for
details on using the much safer Digest mechanism.</p></td></tr></table></div><p>First, create a password file and grant access to
users Harry and Sally:</p><div class="informalexample"><pre class="screen">
$ ### First time: use -c to create the file
$ ### Use -m to use MD5 encryption of the password, which is more secure
$ htpasswd -c -m /etc/svn-auth.htpasswd harry
New password: *****
Re-type new password: *****
Adding password for user harry
$ htpasswd -m /etc/svn-auth.htpasswd sally
New password: *******
Re-type new password: *******
Adding password for user sally
$
</pre></div><p>Next, ensure that Apache has access to the modules which
provide the Basic authentication and related
functionality: <code class="literal">mod_auth_basic</code>,
<code class="literal">mod_authn_file</code>, and
<code class="literal">mod_authz_user</code>. In many cases, these
modules are compiled into <span class="command"><strong>httpd</strong></span> itself,
but if not, you might need to explicitly load one or more of
them using the <code class="literal">LoadModule</code> directive:</p><div class="informalexample"><pre class="programlisting">
LoadModule auth_basic_module modules/mod_auth_basic.so
LoadModule authn_file_module modules/mod_authn_file.so
LoadModule authz_user_module moduels/mod_authz_user.so
</pre></div><p>After ensuring the Apache has access to the required
functionality, you'll need to add some more directives
inside the <code class="literal"><Location></code> block to tell
Apache what type of authentication you wish to use, and just
how to to so:</p><div class="informalexample"><pre class="programlisting">
<Location /svn>
DAV svn
SVNParentPath /var/svn
# Authentication: Basic
AuthName "Subversion repository"
AuthType Basic
AuthBasicProvider file
AuthUserFile /etc/svn-auth.htpasswd
</Location>
</pre></div><p>These directives work as follows:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p><code class="literal">AuthName</code> is an arbitrary name
that you choose for the authentication domain. Most
browsers display this name in the dialog box when
prompting for username and password.</p></li><li class="listitem"><p><code class="literal">AuthType</code> specifies the type of
authentication to use.</p></li><li class="listitem"><p><code class="literal">AuthBasicProvider</code> specifies the
Basic authentication provider to use for the location.
In our example, we wish to consult a local password
file.</p></li><li class="listitem"><p><code class="literal">AuthUserFile</code> specifies the
location of the password file to use.</p></li></ul></div><p>However, this <code class="literal"><Location></code> block
doesn't yet do anything useful. It merely tells Apache that
<span class="emphasis"><em>if</em></span> authorization were required, it
should challenge the Subversion client for a username and
password. (When authorization is required, Apache requires
authentication as well.) What's missing here, however, are
directives that tell Apache <span class="emphasis"><em>which sorts</em></span>
of client requests require authorization; currently, none do.
The simplest thing is to specify that <span class="emphasis"><em>all</em></span>
requests require authorization by adding
<code class="literal">Require valid-user</code> to the block:</p><div class="informalexample"><pre class="programlisting">
<Location /svn>
DAV svn
SVNParentPath /var/svn
# Authentication: Basic
AuthName "Subversion repository"
AuthType Basic
AuthBasicProvider file
AuthUserFile /etc/svn-auth.htpasswd
# Authorization: Authenticated users only
Require valid-user
</Location>
</pre></div><p>Refer to <a class="xref" href="#svn.serverconfig.httpd.authz" title="Authorization Options">the section called “Authorization Options”</a>
for more detail on the <code class="literal">Require</code> directive
and other ways to set authorization policies.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>The default value of the
<code class="literal">AuthBasicProvider</code> option is
<code class="literal">file</code>, so we won't bother including
it in future examples. Just know that if in a broader
context you've set this value to something else, you'll
need to explicitly reset it to <code class="literal">file</code>
within your Subversion <code class="literal"><Location></code>
block in order to get that behavior.</p></td></tr></table></div></div><div class="sect3" title="Digest authentication"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.httpd.authn.digest"></a>Digest authentication</h4></div></div></div><p>Digest authentication is an improvement on Basic
authentication which allows the server to verify a client's
identity without sending the password over the network
unprotected. Both client and server create a non-reversible
MD5 hash of the username, password, requested URI, and a
<em class="firstterm">nonce</em> (number used once) provided by
the server and changed each time authentication is required.
The client sends its hash to the server, and the server then
verifies that the hashes match.</p><p>Configuring Apache to use Digest authentication is
straightforward. You'll need to ensure that
the <code class="literal">mod_auth_digest</code> module is available
(instead of <code class="literal">mod_auth_basic</code>), and then
make a few small variations on our prior example:</p><div class="informalexample"><pre class="programlisting">
<Location /svn>
DAV svn
SVNParentPath /var/svn
# Authentication: Digest
AuthName "Subversion repository"
AuthType Digest
AuthDigestProvider file
AuthUserFile /etc/svn-auth.htdigest
# Authorization: Authenticated users only
Require valid-user
</Location>
</pre></div><p>Notice that <code class="literal">AuthType</code> is now set to
<code class="literal">Digest</code>, and we specify a different path
for <code class="literal">AuthUserFile</code>. Digest authentication
uses a different file format than Basic authentication,
created and managed using Apache's <span class="command"><strong>htdigest</strong></span>
utility<sup>[<a id="idp13681040" href="#ftn.idp13681040" class="footnote">58</a>]</sup> rather than <span class="command"><strong>htpasswd</strong></span>.
Digest authentication also has the additional concept of a
<span class="quote">“<span class="quote">realm</span>”</span>, which must match the value of the
<code class="literal">AuthName</code> directive.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>For digest authentication, the authentication provider
is selected using the <code class="literal">AuthDigestProvider</code>
as shown in the previous example. As was the case with
the <code class="literal">AuthBasicProvider</code> directive,
<code class="literal">file</code> is the default value of the
<code class="literal">AuthDigestProvider</code> option, so this
line is not strictly required unless you need to override
a different value thereof inherited from a broader
configuration context.</p></td></tr></table></div><p>The password file can be created as follows:</p><div class="informalexample"><pre class="screen">
$ ### First time: use -c to create the file
$ htdigest -c /etc/svn-auth.htdigest "Subversion repository" harry
Adding password for harry in realm Subversion repository.
New password: *****
Re-type new password: *****
$ htdigest /etc/svn-auth.htdigest "Subversion repository" sally
Adding user sally in realm Subversion repository
New password: *******
Re-type new password: *******
$
</pre></div></div></div><div class="sect2" title="Authorization Options"><div class="titlepage"><div><div><h3 class="title"><a id="svn.serverconfig.httpd.authz"></a>Authorization Options</h3></div></div></div><p>At this point, you've configured authentication, but not
authorization. Apache is able to challenge clients and
confirm identities, but it has not been told how to allow or
restrict access to the clients bearing those identities. This
section describes two strategies for controlling access to
your repositories.</p><div class="sect3" title="Blanket access control"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.httpd.authz.blanket"></a>Blanket access control</h4></div></div></div><p>The simplest form of access control is to authorize
certain users for either read-only access to a repository or
read/write access to a repository.</p><p>You can restrict access on all repository operations by
adding <code class="literal">Require valid-user</code> directly inside
the <code class="literal"><Location></code> block. The example
from <a class="xref" href="#svn.serverconfig.httpd.authn.digest" title="Digest authentication">the section called “Digest authentication”</a>
allows only clients that successfully authenticate to do
anything with the Subversion repository:</p><div class="informalexample"><pre class="programlisting">
<Location /svn>
DAV svn
SVNParentPath /var/svn
# Authentication: Digest
AuthName "Subversion repository"
AuthType Digest
AuthUserFile /etc/svn-auth.htdigest
# Authorization: Authenticated users only
Require valid-user
</Location>
</pre></div><p>Sometimes you don't need to run such a tight ship. For
example, Subversion's own source code repository at
<a class="ulink" href="http://svn.collab.net/repos/svn" target="_top">http://svn.collab.net/repos/svn</a> allows anyone
in the world to perform read-only repository tasks (such as
checking out working copies and browsing the repository),
but restricts write operations to authenticated users. The
<code class="literal">Limit</code> and <code class="literal">LimitExcept</code>
directives allow for this type of selective restriction.
Like the <code class="literal">Location</code> directive, these blocks
have starting and ending tags, and you would nest them
inside your <code class="literal"><Location></code>
block.</p><p>The parameters present on the <code class="literal">Limit</code>
and <code class="literal">LimitExcept</code> directives are HTTP
request types that are affected by that block. For example,
to allow anonymous read-only operations, you would use the
<code class="literal">LimitExcept</code> directive (passing the
<code class="literal">GET</code>, <code class="literal">PROPFIND</code>,
<code class="literal">OPTIONS</code>, and <code class="literal">REPORT</code>
request type parameters) and place the previously mentioned
<code class="literal">Require valid-user</code> directive inside the
<code class="literal"><LimitExcept></code> block instead of just
inside the <code class="literal"><Location></code> block.</p><div class="informalexample"><pre class="programlisting">
<Location /svn>
DAV svn
SVNParentPath /var/svn
# Authentication: Digest
AuthName "Subversion repository"
AuthType Digest
AuthUserFile /etc/svn-auth.htdigest
# Authorization: Authenticated users only for non-read-only
# (write) operations; allow anonymous reads
<LimitExcept GET PROPFIND OPTIONS REPORT>
Require valid-user
</LimitExcept>
</Location>
</pre></div><p>These are only a few simple examples. For more in-depth
information about Apache access control and the
<code class="literal">Require</code> directive, take a look at the
<code class="literal">Security</code> section of the Apache
documentation's tutorials collection at <a class="ulink" href="http://httpd.apache.org/docs-2.0/misc/tutorials.html" target="_top">http://httpd.apache.org/docs-2.0/misc/tutorials.html</a>.</p></div><div class="sect3" title="Per-directory access control"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.httpd.authz.perdir"></a>Per-directory access control</h4></div></div></div><p>It's possible to set up finer-grained permissions using
<span class="command"><strong>mod_authz_svn</strong></span>. This Apache module grabs
the various opaque URLs passing from client to server, asks
<span class="command"><strong>mod_dav_svn</strong></span> to decode them, and then
possibly vetoes requests based on access policies defined in
a configuration file.</p><p>If you've built Subversion from source code,
<span class="command"><strong>mod_authz_svn</strong></span> is automatically built
and installed alongside <span class="command"><strong>mod_dav_svn</strong></span>.
Many binary distributions install it automatically as well.
To verify that it's installed correctly, make sure it comes
right after <span class="command"><strong>mod_dav_svn</strong></span>'s
<code class="literal">LoadModule</code> directive in
<code class="filename">httpd.conf</code>:</p><div class="informalexample"><pre class="programlisting">
LoadModule dav_module modules/mod_dav.so
LoadModule dav_svn_module modules/mod_dav_svn.so
LoadModule authz_svn_module modules/mod_authz_svn.so
</pre></div><p>To activate this module, you need to configure your
<code class="literal"><Location></code> block to use the
<code class="literal">AuthzSVNAccessFile</code> directive, which
specifies a file containing the permissions policy for paths
within your repositories. (In a moment, we'll discuss the
format of that file.)</p><p>Apache is flexible, so you have the option to configure
your block in one of three general patterns. To begin,
choose one of these basic configuration patterns. (The
following examples are very simple; look at Apache's own
documentation for much more detail on Apache authentication
and authorization options.)</p><p>The most open approach is to allow access to everyone.
This means Apache never sends authentication challenges, and
all users are treated as <span class="quote">“<span class="quote">anonymous</span>”</span>. (See
<a class="xref" href="#svn.serverconfig.httpd.authz.perdir.ex-1" title="Example 6.2. A sample configuration for anonymous access">Example 6.2, “A sample configuration for anonymous access”</a>.)</p><div class="example"><a id="svn.serverconfig.httpd.authz.perdir.ex-1"></a><p class="title"><strong>Example 6.2. A sample configuration for anonymous access</strong></p><div class="example-contents"><pre class="programlisting">
<Location /repos>
DAV svn
SVNParentPath /var/svn
# Authentication: None
# Authorization: Path-based access control
AuthzSVNAccessFile /path/to/access/file
</Location>
</pre></div></div><br class="example-break"></br><p>On the opposite end of the paranoia scale, you can
configure Apache to authenticate all clients.
This block unconditionally requires authentication via the
<code class="literal">Require valid-user</code> directive, and defines
a means to authenticate valid users. (See
<a class="xref" href="#svn.serverconfig.httpd.authz.perdir.ex-2" title="Example 6.3. A sample configuration for authenticated access">Example 6.3, “A sample configuration for authenticated access”</a>.)</p><div class="example"><a id="svn.serverconfig.httpd.authz.perdir.ex-2"></a><p class="title"><strong>Example 6.3. A sample configuration for authenticated access</strong></p><div class="example-contents"><pre class="programlisting">
<Location /repos>
DAV svn
SVNParentPath /var/svn
# Authentication: Digest
AuthName "Subversion repository"
AuthType Digest
AuthUserFile /etc/svn-auth.htdigest
# Authorization: Path-based access control; authenticated users only
AuthzSVNAccessFile /path/to/access/file
Require valid-user
</Location>
</pre></div></div><br class="example-break"></br><p>A third very popular pattern is to allow a combination
of authenticated and anonymous access. For example, many
administrators want to allow anonymous users to read certain
repository directories, but restrict access to more
sensitive areas to authenticated users. In this setup, all
users start out accessing the repository anonymously. If
your access control policy demands a real username at any
point, Apache will demand authentication from the client.
To do this, use both the <code class="literal">Satisfy Any</code>
and <code class="literal">Require valid-user</code> directives. (See
<a class="xref" href="#svn.serverconfig.httpd.authz.perdir.ex-3" title="Example 6.4. A sample configuration for mixed authenticated/anonymous access">Example 6.4, “A sample configuration for mixed
authenticated/anonymous access”</a>.)</p><div class="example"><a id="svn.serverconfig.httpd.authz.perdir.ex-3"></a><p class="title"><strong>Example 6.4. A sample configuration for mixed
authenticated/anonymous access</strong></p><div class="example-contents"><pre class="programlisting">
<Location /repos>
DAV svn
SVNParentPath /var/svn
# Authentication: Digest
AuthName "Subversion repository"
AuthType Digest
AuthUserFile /etc/svn-auth.htdigest
# Authorization: Path-based access control; try anonymous access
# first, but authenticate if necessary
AuthzSVNAccessFile /path/to/access/file
Satisfy Any
Require valid-user
</Location>
</pre></div></div><br class="example-break"></br><p>The next step is to create the authorization file
containing access rules for particular paths within the
repository. We describe how later in this chapter, in
<a class="xref" href="#svn.serverconfig.pathbasedauthz" title="Path-Based Authorization">the section called “Path-Based Authorization”</a>.</p></div><div class="sect3" title="Disabling path-based checks"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.httpd.authz.pathauthzoff"></a>Disabling path-based checks</h4></div></div></div><p>The <span class="command"><strong>mod_dav_svn</strong></span> module goes through a
lot of work to make sure that data you've marked
<span class="quote">“<span class="quote">unreadable</span>”</span> doesn't get accidentally leaked.
This means it needs to closely monitor all of the paths
and file-contents returned by commands such as <span class="command"><strong>svn
checkout</strong></span> and <span class="command"><strong>svn update</strong></span>.
If these commands encounter a path that isn't
readable according to some authorization policy, the
path is typically omitted altogether. In the case of
history or rename tracing—for example, running a command such
as <strong class="userinput"><code>svn cat -r OLD foo.c</code></strong> on a file that
was renamed long ago—the rename tracking will simply
halt if one of the object's former names is determined to be
read-restricted.</p><p>All of this path checking can sometimes be quite
expensive, especially in the case of <span class="command"><strong>svn
log</strong></span>. When retrieving a list of revisions, the
server looks at every changed path in each revision and
checks it for readability. If an unreadable path is
discovered, it's omitted from the list of the revision's
changed paths (normally seen with
the <code class="option">--verbose</code> (<code class="option">-v</code>) option),
and the whole log message is suppressed. Needless to say,
this can be time-consuming on revisions that affect a large
number of files. This is the cost of security: even if you
haven't configured a module such as
<span class="command"><strong>mod_authz_svn</strong></span> at all, the
<span class="command"><strong>mod_dav_svn</strong></span> module is still asking Apache
<span class="command"><strong>httpd</strong></span> to run authorization checks on
every path. The <span class="command"><strong>mod_dav_svn</strong></span> module has
no idea what authorization modules have been installed, so
all it can do is ask Apache to invoke whatever might be
present.</p><p>On the other hand, there's also an escape hatch of
sorts, which allows you to trade security features for
speed. If you're not enforcing any sort of per-directory
authorization (i.e., not using
<span class="command"><strong>mod_authz_svn</strong></span> or similar module),
you can disable all of this path checking. In your
<code class="filename">httpd.conf</code> file, use the
<code class="literal">SVNPathAuthz</code> directive as shown in
<a class="xref" href="#svn.serverconfig.httpd.authz.pathauthzoff.ex-1" title="Example 6.5. Disabling path checks altogether">Example 6.5, “Disabling path checks altogether”</a>.</p><div class="example"><a id="svn.serverconfig.httpd.authz.pathauthzoff.ex-1"></a><p class="title"><strong>Example 6.5. Disabling path checks altogether</strong></p><div class="example-contents"><pre class="programlisting">
<Location /repos>
DAV svn
SVNParentPath /var/svn
SVNPathAuthz off
</Location>
</pre></div></div><br class="example-break"></br><p>The <code class="literal">SVNPathAuthz</code> directive
is <span class="quote">“<span class="quote">on</span>”</span> by default. When
set to <span class="quote">“<span class="quote">off,</span>”</span> all path-based authorization
checking is disabled;
<span class="command"><strong>mod_dav_svn</strong></span> stops invoking authorization
checks on every path it discovers.</p></div></div><div class="sect2" title="Protecting network traffic with SSL"><div class="titlepage"><div><div><h3 class="title"><a id="svn.serverconfig.httpd.ssl"></a>Protecting network traffic with SSL</h3></div></div></div><p>Connecting to a repository via <code class="literal">http://</code>
means that all Subversion activity is sent across the network
in the clear. This means that actions such as checkouts,
commits, and updates could potentially be intercepted by an
unauthorized party <span class="quote">“<span class="quote">sniffing</span>”</span> network traffic.
Encrypting traffic using SSL is a good way to protect
potentially sensitive information over the network.</p><p>If a Subversion client is compiled to use OpenSSL,
it gains the ability to speak to an Apache server via
<code class="literal">https://</code> URLs, so all traffic is encrypted
with a per-connection session key. The WebDAV library used by
the Subversion client is not only able to verify server
certificates, but can also supply client certificates when
challenged by the server.</p><div class="sect3" title="Subversion server SSL certificate configuration"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.httpd.ssl.server"></a>Subversion server SSL certificate configuration</h4></div></div></div><p>It's beyond the scope of this book to describe how to
generate client and server SSL certificates and how to
configure Apache to use them. Many other references,
including Apache's own documentation, describe the process.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>SSL certificates from well-known entities generally
cost money, but at a bare minimum, you can configure
Apache to use a self-signed certificate generated with a
tool such as OpenSSL (<a class="ulink" href="http://openssl.org" target="_top">http://openssl.org</a>).<sup>[<a id="idp13773568" href="#ftn.idp13773568" class="footnote">59</a>]</sup></p></td></tr></table></div></div><div class="sect3" title="Subversion client SSL certificate management"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.httpd.ssl.client"></a>Subversion client SSL certificate management</h4></div></div></div><p>When connecting to Apache via <code class="literal">https://</code>,
a Subversion client can receive two different types of
responses:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>A server certificate</p></li><li class="listitem"><p>A challenge for a client certificate</p></li></ul></div><div class="sect4" title="Server certificate"><div class="titlepage"><div><div><h5 class="title"><a id="svn.serverconfig.httpd.ssl.client.servercert"></a>Server certificate</h5></div></div></div><p>When the client receives a server certificate, it needs
to verify that the server is who it claims to be. OpenSSL
does this by examining the signer of the server certificate,
or <em class="firstterm">certificate authority</em> (CA). If
OpenSSL is unable to automatically trust the CA, or if some
other problem occurs (such as an expired certificate or
hostname mismatch), the Subversion command-line client will
ask you whether you want to trust the server certificate
anyway:</p><div class="informalexample"><pre class="screen">
$ svn list https://host.example.com/repos/project
Error validating server certificate for 'https://host.example.com:443':
- The certificate is not issued by a trusted authority. Use the
fingerprint to validate the certificate manually!
Certificate information:
- Hostname: host.example.com
- Valid: from Jan 30 19:23:56 2004 GMT until Jan 30 19:23:56 2006 GMT
- Issuer: CA, example.com, Sometown, California, US
- Fingerprint: 7d:e1:a9:34:33:39:ba:6a:e9:a5:c4:22:98:7b:76:5c:92:a0:9c:7b
(R)eject, accept (t)emporarily or accept (p)ermanently?
</pre></div><p>This dialogue is essentially the same question you may
have seen coming from your web browser (which is just
another HTTP client like Subversion). If you choose the
<code class="literal">(p)</code>ermanent option, Subversion will
cache the server certificate in your private runtime
<code class="filename">auth/</code> area, just as your username and
password are cached (see <a class="xref" href="#svn.serverconfig.netmodel.credcache" title="Caching credentials">the section called “Caching credentials”</a>), and will
automatically trust the certificate in the future.</p><p>Your runtime <code class="filename">servers</code> file also gives
you the ability to make your Subversion client automatically
trust specific CAs, either globally or on a per-host basis.
Simply set the <code class="literal">ssl-authority-files</code>
variable to a semicolon-separated list of PEM-encoded CA
certificates:</p><div class="informalexample"><pre class="programlisting">
[global]
ssl-authority-files = /path/to/CAcert1.pem;/path/to/CAcert2.pem
</pre></div><p>Many OpenSSL installations also have a predefined set
of <span class="quote">“<span class="quote">default</span>”</span> CAs that are nearly universally
trusted. To make the Subversion client automatically trust
these standard authorities, set the
<code class="literal">ssl-trust-default-ca</code> variable to
<code class="literal">true</code>.</p></div><div class="sect4" title="Client certificate challenge"><div class="titlepage"><div><div><h5 class="title"><a id="svn.serverconfig.httpd.ssl.client.clientcert"></a>Client certificate challenge</h5></div></div></div><p>If the client receives a challenge for a certificate,
the server is asking the client to prove its identity.
The client must send back a certificate signed by a CA
that the server trusts, along with a <em class="firstterm">challenge
response</em> which proves that the client owns the
private key associated with the certificate. The private
key and certificate are usually stored in an encrypted
format on disk, protected by a passphrase. When Subversion
receives this challenge, it will ask you for the path to the
encrypted file and the passphrase that protects it:</p><div class="informalexample"><pre class="screen">
$ svn list https://host.example.com/repos/project
Authentication realm: https://host.example.com:443
Client certificate filename: /path/to/my/cert.p12
Passphrase for '/path/to/my/cert.p12': ********
</pre></div><p>Notice that the client credentials are stored in a
<code class="literal">.p12</code> file. To use a client certificate
with Subversion, it must be in PKCS#12 format, which is a
portable standard. Most web browsers are able to import
and export certificates in that format. Another option
is to use the OpenSSL command-line tools to convert
existing certificates into PKCS#12.</p><p>The runtime <code class="filename">servers</code> file also
allows you to automate this challenge on a per-host basis.
If you set the <code class="literal">ssl-client-cert-file</code>
and <code class="literal">ssl-client-cert-password</code> variables,
Subversion can automatically respond to a client
certificate challenge without prompting you:</p><div class="informalexample"><pre class="programlisting">
[groups]
examplehost = host.example.com
[examplehost]
ssl-client-cert-file = /path/to/my/cert.p12
ssl-client-cert-password = somepassword
</pre></div><p>More security-conscious folk might want to exclude
<code class="literal">ssl-client-cert-password</code> to avoid
storing the passphrase in the clear on disk.</p></div></div></div><div class="sect2" title="Extra Goodies"><div class="titlepage"><div><div><h3 class="title"><a id="svn.serverconfig.httpd.extra"></a>Extra Goodies</h3></div></div></div><p>We've covered most of the authentication and authorization
options for Apache and <span class="command"><strong>mod_dav_svn</strong></span>. But
there are a few other nice features that Apache
provides.</p><div class="sect3" title="Repository browsing"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.httpd.extra.browsing"></a>Repository browsing</h4></div></div></div><p>One of the most useful benefits of an Apache/WebDAV
configuration for your Subversion repository is that your
versioned files and directories are immediately available
for viewing via a regular web browser. Since Subversion
uses URLs to identify versioned resources, those URLs used
for HTTP-based repository access can be typed directly into
a web browser. Your browser will issue an
HTTP <code class="literal">GET</code> request for that URL; based on
whether that URL represents a versioned directory or
file, <span class="command"><strong>mod_dav_svn</strong></span> will respond with a
directory listing or with file contents.</p><div class="sect4" title="URL syntax"><div class="titlepage"><div><div><h5 class="title"><a id="svn.serverconfig.httpd.extra.browsing.syntax"></a>URL syntax</h5></div></div></div><p>If the URLs do not contain any information about which
version of the resource you wish to
see, <span class="command"><strong>mod_dav_svn</strong></span> will answer with the
youngest version. This functionality has the wonderful
side effect that you can pass around Subversion URLs to
your peers as references to documents, and those URLs will
always point at the latest manifestation of that document.
Of course, you can even use the URLs as hyperlinks from
other web sites, too.</p><p>As of Subversion 1.6, <span class="command"><strong>mod_dav_svn</strong></span>
supports a public URI syntax for examining older revisions
of both files and directories. The syntax uses the query
string portion of the URL to specify either or both of a
peg revision and operative revision, which Subversion will
then use to determine which version of the file or
directory to display to your web browser. Add the query
string name/value pair
<code class="literal">p=<em class="replaceable"><code>PEGREV</code></em></code>,
where <em class="replaceable"><code>PEGREV</code></em> is a revision
number, to specify the peg revision you wish to apply to the
request. Use
<code class="literal">r=<em class="replaceable"><code>REV</code></em></code>,
where <em class="replaceable"><code>REV</code></em> is a revision number, to
specify an operative revision.</p><p>For example, if you wish to see the latest version of a
<code class="filename">README.txt</code> file located in your
project's <code class="filename">/trunk</code>, point your web
browser to that file's repository URL, which might look
something like the following:</p><div class="informalexample"><pre class="programlisting">
http://host.example.com/repos/project/trunk/README.txt
</pre></div><p>If you now wish to see some older version of that
file, add an operative revision to the URL's query
string:</p><div class="informalexample"><pre class="programlisting">
http://host.example.com/repos/project/trunk/README.txt?r=1234
</pre></div><p>What if the thing you're trying to view no longer exists
in the youngest revision of the repository? That's where a
peg revision is handy:</p><div class="informalexample"><pre class="programlisting">
http://host.example.com/repos/project/trunk/deleted-thing.txt?p=321
</pre></div><p>And of course, you can combine peg revision and
operative revision specifiers to fine-tune the exact item
you wish to view:</p><div class="informalexample"><pre class="programlisting">
http://host.example.com/repos/project/trunk/renamed-thing.txt?p=123&r=21
</pre></div><p>The previous URL would display revision 21 of the
object which, in revision 123, was located
at <code class="filename">/trunk/renamed-thing.txt</code> in the
repository. See <a class="xref" href="#svn.advanced.pegrevs" title="Peg and Operative Revisions">the section called “Peg and Operative Revisions”</a> for a
detailed explanation of these <span class="quote">“<span class="quote">peg revision</span>”</span>
and <span class="quote">“<span class="quote">operative revision</span>”</span> concepts. They can
be a bit tricky to wrap your head around.</p><p>As a reminder, this feature
of <span class="command"><strong>mod_dav_svn</strong></span> offers only a limited
repository browsing experience. You can see directory
listings and file contents, but no revision properties
(such as commit log messages) or file/directory
properties. For folks who require more extensive browsing
of repositories and their history, there are several
third-party software packages which offer this. Some
examples include ViewVC
(<a class="ulink" href="http://viewvc.tigris.org" target="_top">http://viewvc.tigris.org</a>), Trac
(<a class="ulink" href="http://trac.edgewall.org" target="_top">http://trac.edgewall.org</a>) and WebSVN
(<a class="ulink" href="http://websvn.info" target="_top">http://websvn.info</a>). These third-party
tools don't affect <span class="command"><strong>mod_dav_svn</strong></span>'s
built-in <span class="quote">“<span class="quote">browseability</span>”</span>, and generally offer
a much wider set of features, including the display of the
aforementioned property sets, display of content
differences between file revisions, and so on.</p></div><div class="sect4" title="Proper MIME type"><div class="titlepage"><div><div><h5 class="title"><a id="svn.serverconfig.httpd.extra.browsing.mimetype"></a>Proper MIME type</h5></div></div></div><p>When browsing a Subversion repository, the web browser
gets a clue about how to render a file's contents by
looking at the <code class="literal">Content-Type:</code> header
returned in Apache's response to the
HTTP <code class="literal">GET</code> request. The value of this
header is some sort of MIME type. By default, Apache will
tell the web browsers that all repository files are of
the <span class="quote">“<span class="quote">default</span>”</span> MIME type,
typically <code class="literal">text/plain</code>. This can be
frustrating, however, if a user wishes repository files to
render as something more meaningful—for example,
it might be nice to have a <code class="filename">foo.html</code> file
in the repository actually render as HTML when
browsing.</p><p>To make this happen, you need only to make sure that
your files have the proper <code class="literal">svn:mime-type</code>
set. We discuss this in more detail in
<a class="xref" href="#svn.advanced.props.special.mime-type" title="File Content Type">the section called “File Content Type”</a>,
and you can even configure your client to automatically
attach proper <code class="literal">svn:mime-type</code> properties
to files entering the repository for the first time; see
<a class="xref" href="#svn.advanced.props.auto" title="Automatic Property Setting">the section called “Automatic Property Setting”</a>.</p><p>Continuing our example, if one were to set
the <code class="literal">svn:mime-type</code> property
to <code class="literal">text/html</code> on
file <code class="filename">foo.html</code>, Apache would properly
tell your web browser to render the file as HTML. One
could also attach proper <code class="literal">image/*</code>
MIME-type properties to image files and ultimately get an
entire web site to be viewable directly from a repository!
There's generally no problem with this, as long as the web
site doesn't contain any dynamically generated
content.</p></div><div class="sect4" title="Customizing the look"><div class="titlepage"><div><div><h5 class="title"><a id="svn.serverconfig.httpd.extra.browsing.xslt"></a>Customizing the look</h5></div></div></div><p>You generally will get more use out of URLs to
versioned files—after all, that's where the
interesting content tends to lie. But you might have
occasion to browse a Subversion directory listing, where
you'll quickly note that the generated HTML used to
display that listing is very basic, and certainly not
intended to be aesthetically pleasing (or even
interesting). To enable customization of these directory
displays, Subversion provides an XML index feature. A
single <code class="literal">SVNIndexXSLT</code> directive in your
repository's <code class="literal">Location</code> block of
<code class="filename">httpd.conf</code> will
instruct <span class="command"><strong>mod_dav_svn</strong></span> to generate XML
output when displaying a directory listing, and to
reference the XSLT stylesheet of your choice:</p><div class="informalexample"><pre class="programlisting">
<Location /svn>
DAV svn
SVNParentPath /var/svn
SVNIndexXSLT "/svnindex.xsl"
…
</Location>
</pre></div><p>Using the <code class="literal">SVNIndexXSLT</code> directive
and a creative XSLT stylesheet, you can make your
directory listings match the color schemes and imagery
used in other parts of your web site. Or, if you'd
prefer, you can use the sample stylesheets provided in the
Subversion source distribution's
<code class="filename">tools/xslt/</code> directory.
Keep in mind that the path provided to the
<code class="literal">SVNIndexXSLT</code> directory is actually a
URL path—browsers need to be able to read your
stylesheets to make use of them!</p></div><div class="sect4" title="Listing repositories"><div class="titlepage"><div><div><h5 class="title"><a id="svn.serverconfig.httpd.extra.browsing.reposlisting"></a>Listing repositories</h5></div></div></div><p>If you're serving a collection of repositories from a
single URL via the <code class="literal">SVNParentPath</code>
directive, then it's also possible to have Apache display
all available repositories to a web browser. Just
activate the <code class="literal">SVNListParentPath</code>
directive:</p><div class="informalexample"><pre class="programlisting">
<Location /svn>
DAV svn
SVNParentPath /var/svn
SVNListParentPath on
…
</Location>
</pre></div><p>If a user now points her web browser to the
URL <code class="literal">http://host.example.com/svn/</code>,
she'll see a list of all Subversion repositories sitting
in <code class="filename">/var/svn</code>. Obviously, this can be
a security problem, so this feature is turned off by
default.</p></div></div><div class="sect3" title="Apache logging"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.httpd.extra.logging"></a>Apache logging</h4></div></div></div><p>Because Apache is an HTTP server at heart, it contains
fantastically flexible logging features. It's beyond the
scope of this book to discuss all of the ways logging can be
configured, but we should point out that even the most
generic <code class="filename">httpd.conf</code> file will cause
Apache to produce two logs:
<code class="filename">error_log</code>
and <code class="filename">access_log</code>. These logs may appear
in different places, but are typically created in the
logging area of your Apache installation. (On Unix, they
often live
in <code class="filename">/usr/local/apache2/logs/</code>.)</p><p>The <code class="filename">error_log</code> describes any internal
errors that Apache runs into as it works.
The <code class="filename">access_log</code> file records every
incoming HTTP request received by Apache. This makes it
easy to see, for example, which IP addresses Subversion
clients are coming from, how often particular clients use
the server, which users are authenticating properly, and
which requests succeed or fail.</p><p>Unfortunately, because HTTP is a stateless protocol,
even the simplest Subversion client operation generates
multiple network requests. It's very difficult to look at
the <code class="filename">access_log</code> and deduce what the
client was doing—most operations look like a series
of cryptic <code class="literal">PROPPATCH</code>, <code class="literal">GET</code>,
<code class="literal">PUT</code>, and <code class="literal">REPORT</code>
requests. To make things worse, many client operations send
nearly identical series of requests, so it's even harder to
tell them apart.</p><p><span class="command"><strong>mod_dav_svn</strong></span>, however, can come to
your aid. By activating an <span class="quote">“<span class="quote">operational
logging</span>”</span> feature, you can
ask <span class="command"><strong>mod_dav_svn</strong></span> to create a separate log
file describing what sort of high-level operations your
clients are performing.</p><p>To do this, you need to make use of
Apache's <code class="literal">CustomLog</code> directive (which is
explained in more detail in Apache's own documentation).
Be sure to invoke this
directive <span class="emphasis"><em>outside</em></span> your
Subversion <code class="literal">Location</code> block:</p><div class="informalexample"><pre class="programlisting">
<Location /svn>
DAV svn
…
</Location>
CustomLog logs/svn_logfile "%t %u %{SVN-ACTION}e" env=SVN-ACTION
</pre></div><p>In this example, we're asking Apache to create a special
logfile, <code class="filename">svn_logfile</code>, in the standard
Apache <code class="filename">logs</code> directory.
The <code class="literal">%t</code> and <code class="literal">%u</code>
variables are replaced by the time and username of the
request, respectively. The really important parts are the
two instances of <code class="literal">SVN-ACTION</code>.
When Apache sees that variable, it substitutes the value of
the <code class="literal">SVN-ACTION</code> environment variable,
which is automatically set by <span class="command"><strong>mod_dav_svn</strong></span>
whenever it detects a high-level client action.</p><p>So, instead of having to interpret a
traditional <code class="filename">access_log</code> like
this:</p><div class="informalexample"><pre class="programlisting">
[26/Jan/2007:22:25:29 -0600] "PROPFIND /svn/calc/!svn/vcc/default HTTP/1.1" 207 398
[26/Jan/2007:22:25:29 -0600] "PROPFIND /svn/calc/!svn/bln/59 HTTP/1.1" 207 449
[26/Jan/2007:22:25:29 -0600] "PROPFIND /svn/calc HTTP/1.1" 207 647
[26/Jan/2007:22:25:29 -0600] "REPORT /svn/calc/!svn/vcc/default HTTP/1.1" 200 607
[26/Jan/2007:22:25:31 -0600] "OPTIONS /svn/calc HTTP/1.1" 200 188
[26/Jan/2007:22:25:31 -0600] "MKACTIVITY /svn/calc/!svn/act/e6035ef7-5df0-4ac0-b811-4be7c823f998 HTTP/1.1" 201 227
…
</pre></div><p>you can peruse a much more
intelligible <code class="filename">svn_logfile</code> like
this:</p><div class="informalexample"><pre class="programlisting">
[26/Jan/2007:22:24:20 -0600] - get-dir /tags r1729 props
[26/Jan/2007:22:24:27 -0600] - update /trunk r1729 depth=infinity
[26/Jan/2007:22:25:29 -0600] - status /trunk/foo r1729 depth=infinity
[26/Jan/2007:22:25:31 -0600] sally commit r1730
</pre></div><p>In addition to the <code class="literal">SVN-ACTION</code>
environment variable, <span class="command"><strong>mod_dav_svn</strong></span> also
populates the <code class="literal">SVN-REPOS</code>
and <code class="literal">SVN-REPOS-NAME</code> variables, which carry
the filesystem path to the repository and the basename
thereof, respectively. You might wish to include references
to one or both of these variables in
your <code class="literal">CustomLog</code> format string, too,
especially if you are combining usage information from
multiple repositories into a single log file.</p><p>For an exhaustive list of all actions logged, see <a class="xref" href="#svn.serverconfig.operational-logging" title="High-level Logging">the section called “High-level Logging”</a>.</p></div><div class="sect3" title="Write-through proxying"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.httpd.extra.writethruproxy"></a>Write-through proxying</h4></div></div></div><p>One of the nice advantages of using Apache as a
Subversion server is that it can be set up for simple
replication. For example, suppose that your team is
distributed across four offices around the globe. The
Subversion repository can exist only in one of those
offices, which means the other three offices will not enjoy
accessing it—they're likely to experience
significantly slower traffic and response times when
updating and committing code. A powerful solution is to set
up a system consisting of one <em class="firstterm">master</em>
Apache server and several <em class="firstterm">slave</em>
Apache servers. If you place a slave server in each office,
users can check out a working copy from whichever slave
is closest to them. All read requests go to their local
slave. Write requests get automatically routed to the
single master server. When the commit completes, the master
then automatically <span class="quote">“<span class="quote">pushes</span>”</span> the new revision to
each slave server using the <span class="command"><strong>svnsync</strong></span>
replication tool.</p><p>This configuration creates a huge perceptual speed
increase for your users, because Subversion client traffic
is typically 80–90% read requests. And if those
requests are coming from a <span class="emphasis"><em>local</em></span>
server, it's a huge win.</p><p>In this section, we'll walk you through a standard setup
of this single-master/multiple-slave system. However, keep
in mind that your servers must be running at least Apache
2.2.0 (with <span class="command"><strong>mod_proxy</strong></span> loaded) and
Subversion 1.5 (<span class="command"><strong>mod_dav_svn</strong></span>).</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>Ours is just one example of how you might setup a
Subversion write-through proxy configuration. There are
other approaches. For example, rather than having the
master server push changes out to every slave server, the
slaves could periodically pull those changes from the
master. Or perhaps the master could push changes to a
single slave, which then pushes the same change to the
next slave, and so on down the line. Administrators are
encouraged to use this section for basic understanding
of what happens in a Subversion WebDAV proxy deployment
scenario, and then implement the specific approach that
works best for their organization.</p></td></tr></table></div><div class="sect4" title="Configure the servers"><div class="titlepage"><div><div><h5 class="title"><a id="svn.serverconfig.httpd.extra.writethruproxy.configure"></a>Configure the servers</h5></div></div></div><p>First, configure your master server's
<code class="filename">httpd.conf</code> file in the usual way.
Make the repository available at a certain URI location,
and configure authentication and authorization however
you'd like. After that's done, configure each of your
<span class="quote">“<span class="quote">slave</span>”</span> servers in the exact same way, but
add the special <code class="literal">SVNMasterURI</code> directive
to the block:</p><div class="informalexample"><pre class="programlisting">
<Location /svn>
DAV svn
SVNPath /var/svn/repos
SVNMasterURI http://master.example.com/svn
…
</Location>
</pre></div><p>This new directive tells a slave server to redirect
all write requests to the master. (This is done
automatically via Apache's <span class="command"><strong>mod_proxy</strong></span>
module.) Ordinary read requests, however, are still
serviced by the slaves. Be sure that your master and
slave servers all have matching authentication and
authorization configurations; if they fall out of sync,
it can lead to big headaches.</p><p>Next, we need to deal with the problem of infinite
recursion. With the current configuration, imagine what
will happen when a Subversion client performs a commit to
the master server. After the commit completes, the server
uses <span class="command"><strong>svnsync</strong></span> to replicate the new
revision to each slave. But because
<span class="command"><strong>svnsync</strong></span> appears to be just another
Subversion client performing a commit, the slave will
immediately attempt to proxy the incoming write request
back to the master! Hilarity ensues.</p><p>The solution to this problem is to have the master
push revisions to a different
<code class="literal"><Location></code> on the slaves. This
location is configured to <span class="emphasis"><em>not</em></span> proxy
write requests at all, but to accept normal commits from
(and only from) the master's IP address:</p><div class="informalexample"><pre class="programlisting">
<Location /svn-proxy-sync>
DAV svn
SVNPath /var/svn/repos
Order deny,allow
Deny from all
# Only let the server's IP address access this Location:
Allow from 10.20.30.40
…
</Location>
</pre></div></div><div class="sect4" title="Set up replication"><div class="titlepage"><div><div><h5 class="title"><a id="svn.serverconfig.httpd.extra.writethruproxy.replicate"></a>Set up replication</h5></div></div></div><p>Now that you've configured
your <code class="literal">Location</code> blocks on master and
slaves, you need to configure the master to replicate to
the slaves. Our walkthough uses <span class="command"><strong>svnsync</strong></span>,
which is covered in more detail in
<a class="xref" href="#svn.reposadmin.maint.replication.svnsync" title="Replication with svnsync">the section called “Replication with svnsync”</a>.</p><p>First, make sure that each slave repository has a
<code class="filename">pre-revprop-change</code> hook script which
allows remote revision property changes. (This is
standard procedure for being on the receiving end of
<span class="command"><strong>svnsync</strong></span>.) Then log into the master
server and configure each of the slave repository URIs to
receive data from the master repository on the local
disk:</p><div class="informalexample"><pre class="screen">
$ svnsync init http://slave1.example.com/svn-proxy-sync \
file:///var/svn/repos
Copied properties for revision 0.
$ svnsync init http://slave2.example.com/svn-proxy-sync \
file:///var/svn/repos
Copied properties for revision 0.
$ svnsync init http://slave3.example.com/svn-proxy-sync \
file:///var/svn/repos
Copied properties for revision 0.
# Perform the initial replication
$ svnsync sync http://slave1.example.com/svn-proxy-sync \
file:///var/svn/repos
Transmitting file data ....
Committed revision 1.
Copied properties for revision 1.
Transmitting file data .......
Committed revision 2.
Copied properties for revision 2.
…
$ svnsync sync http://slave2.example.com/svn-proxy-sync \
file:///var/svn/repos
Transmitting file data ....
Committed revision 1.
Copied properties for revision 1.
Transmitting file data .......
Committed revision 2.
Copied properties for revision 2.
…
$ svnsync sync http://slave3.example.com/svn-proxy-sync \
file:///var/svn/repos
Transmitting file data ....
Committed revision 1.
Copied properties for revision 1.
Transmitting file data .......
Committed revision 2.
Copied properties for revision 2.
…
</pre></div><p>After this is done, we configure the master server's
<code class="literal">post-commit</code> hook script to invoke
<span class="command"><strong>svnsync</strong></span> on each slave server:</p><div class="informalexample"><pre class="programlisting">
#!/bin/sh
# Post-commit script to replicate newly committed revision to slaves
svnsync sync http://slave1.example.com/svn-proxy-sync \
file:///var/svn/repos > /dev/null 2>&1 &
svnsync sync http://slave2.example.com/svn-proxy-sync \
file:///var/svn/repos > /dev/null 2>&1 &
svnsync sync http://slave3.example.com/svn-proxy-sync \
file:///var/svn/repos > /dev/null 2>&1 &
</pre></div><p>The extra bits on the end of each line aren't
necessary, but they're a sneaky way to allow the sync
commands to run in the background so that the Subversion
client isn't left waiting forever for the commit to
finish. In addition to this
<code class="literal">post-commit</code> hook, you'll need a
<code class="literal">post-revprop-change</code> hook as well so
that when a user, say, modifies a log message, the slave
servers get that change also:</p><div class="informalexample"><pre class="programlisting">
#!/bin/sh
# Post-revprop-change script to replicate revprop-changes to slaves
REV=${2}
svnsync copy-revprops http://slave1.example.com/svn-proxy-sync \
file:///var/svn/repos \
-r ${REV} > /dev/null 2>&1 &
svnsync copy-revprops http://slave2.example.com/svn-proxy-sync \
file:///var/svn/repos \
-r ${REV} > /dev/null 2>&1 &
svnsync copy-revprops http://slave3.example.com/svn-proxy-sync \
file:///var/svn/repos \
-r ${REV} > /dev/null 2>&1 &
</pre></div><p>The only thing we've left out here is what to do about
user-level locks (of the <span class="command"><strong>svn lock</strong></span>
variety). Locks are enforced by the master server during
commit operations; but all the information about locks is
distributed during read operations such as <span class="command"><strong>svn
update</strong></span> and <span class="command"><strong>svn status</strong></span> by the
slave server. As such, a fully functioning proxy setup
needs to perfectly replicate lock information from the
master server to the slave servers. Unfortunately, most
of the mechanisms that one might employ to accomplish this
replication fall short in one way or
another<sup>[<a id="idp13940608" href="#ftn.idp13940608" class="footnote">60</a>]</sup>. Many teams
don't use Subversion's locking features at all, so this
may be a nonissue for you. Sadly, for those teams which
do use locks, we have no recommendations on how to
gracefully work around this shortcoming.</p></div><div class="sect4" title="Caveats"><div class="titlepage"><div><div><h5 class="title"><a id="svn.serverconfig.httpd.extra.writethruproxy.caveats"></a>Caveats</h5></div></div></div><p>Your master/slave replication system should now be
ready to use. A couple of words of warning are in order,
however. Remember that this replication isn't entirely
robust in the face of computer or network crashes. For
example, if one of the automated
<span class="command"><strong>svnsync</strong></span> commands fails to complete for
some reason, the slaves will begin to fall behind. For
example, your remote users will see that they've committed
revision 100, but then when they run <span class="command"><strong>svn
update</strong></span>, their local server will tell them that
revision 100 doesn't yet exist! Of course, the problem
will be automatically fixed the next time another commit
happens and the subsequent <span class="command"><strong>svnsync</strong></span> is
successful—the sync will replicate all waiting
revisions. But still, you may want to set up some sort of
out-of-band monitoring to notice synchronization failures
and force <span class="command"><strong>svnsync</strong></span> to run when things go
wrong.</p><p>Another limitation of the write-through proxy
deployment model involves version mismatches—of the
version of Subversion which is installed, that
is—between the master and slave servers. Each new
release of Subversion may (and often does) add new
features to the network protocol used between the clients
and servers. Since feature negotiation happens against
the slave, it is the slave's protocol version and feature
set which is used. But write operations are passed
through to the master server quite literally. Therefore,
there is always a risk that the slave server will answer a
feature negotiation request from the client in way that is
true for the slave, but untrue for the master if the
master is running an older version of Subversion. This
could result in the client trying to use a new feature
that the master doesn't understand, and failing. There
are a couple of known problems of this sort in Subversion
1.7, which introduced a major revision of its HTTP
protocol. If you are deploying a Subversion 1.7 slave
server in front of a pre-1.7 master, you'll want to
configure your slave server's
Subversion <code class="literal"><Location></code> block with
the <code class="literal">SVNAdvertiseV2Protocol Off</code>
directive.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>For the best results possible, try to run the same
version of Subversion on your master and slave
servers.</p></td></tr></table></div><div class="sidebar" title="Can We Set Up Replication with svnserve?"><div class="titlepage"><div><div><p class="title"><strong>Can We Set Up Replication with svnserve?</strong></p></div></div></div><p>If you're using <span class="command"><strong>svnserve</strong></span> instead
of Apache as your server, you can certainly configure
your repository's hook scripts to invoke
<span class="command"><strong>svnsync</strong></span> as we've shown here, thereby
causing automatic replication from master to slaves.
Unfortunately, at the time of this writing there is no way to
make slave <span class="command"><strong>svnserve</strong></span> servers
automatically proxy write requests back to the master
server. This means your users would only be able to
check out read-only working copies from the slave
servers. You'd have to configure your slave servers to
disallow write access completely. This might be useful
for creating read-only <span class="quote">“<span class="quote">mirrors</span>”</span> of popular
open source projects, but it's not a transparent
proxying system.</p></div></div></div><div class="sect3" title="Other Apache features"><div class="titlepage"><div><div><h4 class="title"><a id="svn.serverconfig.httpd.extra.other"></a>Other Apache features</h4></div></div></div><p>Several of the features already provided by Apache in
its role as a robust web server can be leveraged for
increased functionality or security in Subversion as well.
The Subversion client is able to use SSL (the Secure Sockets
Layer, discussed earlier). If your Subversion client is
built to support SSL, it can access your Apache server
using <code class="literal">https://</code> and enjoy a high-quality
encrypted network session.</p><p>Equally useful are other features of the Apache and
Subversion relationship, such as the ability to specify a
custom port (instead of the default HTTP port 80) or a
virtual domain name by which the Subversion repository
should be accessed, or the ability to access the repository
through an HTTP proxy.</p><p>Finally, because <span class="command"><strong>mod_dav_svn</strong></span> is
speaking a subset of the WebDAV/DeltaV protocol, it's
possible to access the repository via third-party DAV
clients. Most modern operating systems (Win32, OS X, and
Linux) have the built-in ability to mount a DAV server as a
standard network <span class="quote">“<span class="quote">shared folder.</span>”</span> This is a
complicated topic, but also wondrous when implemented. For
details, read <a class="xref" href="#svn.webdav" title="Appendix C. WebDAV and Autoversioning">Appendix C, <em>WebDAV and Autoversioning</em></a>.</p><p>Note that there are a number of other small tweaks one can
make to <span class="command"><strong>mod_dav_svn</strong></span> that are too obscure
to mention in this chapter. For a complete list of
all <code class="filename">httpd.conf</code> directives
that <span class="command"><strong>mod_dav_svn</strong></span> responds to, see
<a class="xref" href="#svn.ref.mod_dav_svn.conf.directives" title="Directives">the section called “Directives”</a> in
<a class="xref" href="#svn.ref" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a>.</p></div></div></div><div class="sect1" title="Path-Based Authorization"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.serverconfig.pathbasedauthz"></a>Path-Based Authorization</h2></div></div></div><p>Both Apache and <span class="command"><strong>svnserve</strong></span> are capable of
granting (or denying) permissions to users. Typically this is
done over the entire repository: a user can read the repository
(or not), and she can write to the repository (or not). It's
also possible, however, to define finer-grained access rules.
One set of users may have permission to write to a certain
directory in the repository, but not others; another directory
might not even be readable by all but a few special
people. As files are paths, too, it's even possible to restrict
access on a per file basis.</p><p>Both servers use a common file format to describe these
path-based access rules. In the case of Apache, one needs to
load the <span class="command"><strong>mod_authz_svn</strong></span> module and then add
the <code class="literal">AuthzSVNAccessFile</code> directive (within
the <code class="filename">httpd.conf</code> file) pointing to your own
access rules file. (For a full explanation, see
<a class="xref" href="#svn.serverconfig.httpd.authz.perdir" title="Per-directory access control">the section called “Per-directory access control”</a>.) If
you're using <span class="command"><strong>svnserve</strong></span>, you need to make
the <code class="literal">authz-db</code> variable
(within <code class="filename">svnserve.conf</code>) point to your access
rules file.</p><div class="sidebar" title="Do You Really Need Path-Based Access Control?"><div class="titlepage"><div><div><p class="title"><strong>Do You Really Need Path-Based Access Control?</strong></p></div></div></div><p>A lot of administrators setting up Subversion for the
first time tend to jump into path-based access control without
giving it a lot of thought. The administrator usually knows
which teams of people are working on which projects, so it's
easy to jump in and grant certain teams access to certain
directories and not others. It seems like a natural thing,
and it appeases the administrator's desire to maintain tight
control of the repository.</p><p>Note, though, that there are often invisible (and
visible!) costs associated with this feature. In the visible
category, the server needs to do a lot more work to ensure
that the user has the right to read or write each specific
path; in certain situations, there's very noticeable
performance loss. In the invisible category, consider the
culture you're creating. Most of the time, while certain
users <span class="emphasis"><em>shouldn't</em></span> be committing changes to
certain parts of the repository, that social contract doesn't
need to be technologically enforced. Teams can sometimes
spontaneously collaborate with each other; someone may want to
help someone else out by committing to an area she doesn't
normally work on. By preventing this sort of thing at the
server level, you're setting up barriers to unexpected
collaboration. You're also creating a bunch of rules that
need to be maintained as projects develop, new users are
added, and so on. It's a bunch of extra work to
maintain.</p><p>Remember that this is a version control system! Even if
somebody accidentally commits a change to something she
shouldn't, it's easy to undo the change. And if a user
commits to the wrong place with deliberate malice, it's a
social problem anyway, and that the problem needs to be dealt
with outside Subversion.</p><p>So, before you begin restricting users' access rights, ask
yourself whether there's a real, honest need for this, or
whether it's just something that <span class="quote">“<span class="quote">sounds good</span>”</span> to
an administrator. Decide whether it's worth sacrificing some
server speed, and remember that there's very little risk
involved; it's bad to become dependent on technology as a
crutch for social problems.<sup>[<a id="idp13983120" href="#ftn.idp13983120" class="footnote">61</a>]</sup></p><p>As an example to ponder, consider that the Subversion
project itself has always had a notion of who is allowed to
commit where, but it's always been enforced socially. This is
a good model of community trust, especially for open source
projects. Of course, sometimes there <span class="emphasis"><em>are</em></span>
truly legitimate needs for path-based access control; within
corporations, for example, certain types of data really can be
sensitive, and access needs to be genuinely restricted to
small groups of people.</p></div><p>Once your server knows where to find your access file, it's
time to define the rules.</p><p>The syntax of the file is the same familiar one used
by <code class="filename">svnserve.conf</code> and the runtime
configuration files. Lines that start with a hash
(<code class="literal">#</code>) are ignored. In its simplest form, each
section names a repository and path within it, as well as the
authenticated usernames are the option names within each
section. The value of each option describes the user's level of
access to the repository path: either
<code class="literal">r</code> (read-only) or <code class="literal">rw</code>
(read/write). If the user is not mentioned at all, no access is
allowed.</p><p>To be more specific: the value of the section names is
either of the form <code class="literal">[repos-name:path]</code> or of the
form <code class="literal">[path]</code>.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>Prior to version 1.7, Subversion treated repository names
and paths in a case-insensitive fashion for the purposes of
access control, converting them to lower case internally
before comparing them against the contents of your access
file. It now does these comparisons case-sensitively. If you
upgraded to Subversion 1.7 from an older version, you should
review your access files for case correctness.</p></td></tr></table></div><p>If you're using the
<code class="literal">SVNParentPath</code> directive, it's important
to specify the repository names in your sections. If you omit
them, a section such as
<code class="literal">[/some/dir]</code> will match the path
<code class="filename">/some/dir</code> in <span class="emphasis"><em>every</em></span>
repository. If you're using the <code class="literal">SVNPath</code>
directive, however, it's fine to only define paths in your
sections—after all, there's only one repository.</p><div class="informalexample"><pre class="programlisting">
[calc:/branches/calc/bug-142]
harry = rw
sally = r
</pre></div><p>In this first example, the user <code class="literal">harry</code> has
full read and write access on the
<code class="filename">/branches/calc/bug-142</code> directory in the
<code class="literal">calc</code> repository, but the user
<code class="literal">sally</code> has read-only access. Any other users
are blocked from accessing this directory.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p><span class="command"><strong>mod_dav_svn</strong></span> offers a directive,
<code class="literal">SVNReposName</code>, which allows administrators
to define a more human-friendly name, of sorts, for a
repository:</p><div class="informalexample"><pre class="programlisting">
<Location /svn/calc>
SVNPath /var/svn/calc
SVNReposName "Calculator Application"
…
</pre></div><p>This allows <span class="command"><strong>mod_dav_svn</strong></span> to identify the
repository by something other than merely its server directory
basename—<code class="filename">calc</code>, in the previous
example—when providing directory listings of repository
content. Be aware, however, that when consulting the access
file for authorization rules, Subversion uses this repository
basename for comparison, <span class="emphasis"><em>not</em></span> any
configured human-friendly name.</p></td></tr></table></div><p>Of course, permissions are inherited from parent to child
directory. That means we can specify a subdirectory with a
different access policy for Sally:</p><div class="informalexample"><pre class="programlisting">
[calc:/branches/calc/bug-142]
harry = rw
sally = r
# give sally write access only to the 'testing' subdir
[calc:/branches/calc/bug-142/testing]
sally = rw
</pre></div><p>Now Sally can write to the <code class="filename">testing</code>
subdirectory of the branch, but can still only read other parts.
Harry, meanwhile, continues to have complete read/write access
to the whole branch.</p><p>It's also possible to explicitly deny permission to someone
via inheritance rules, by setting the username variable to
nothing:</p><div class="informalexample"><pre class="programlisting">
[calc:/branches/calc/bug-142]
harry = rw
sally = r
[calc:/branches/calc/bug-142/secret]
harry =
</pre></div><p>In this example, Harry has read/write access to the
entire <code class="filename">bug-142</code> tree, but has absolutely no
access at all to the <code class="filename">secret</code> subdirectory
within it.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>The thing to remember is that the most specific path
always matches first. The server tries to match the path
itself, and then the parent of the path, then the parent of
that, and so on. The net effect is that mentioning a specific
path in the access file will always override any permissions
inherited from parent directories.</p></td></tr></table></div><p>By default, nobody has any access to the repository at all.
That means that if you're starting with an empty file, you'll
probably want to give at least read permission to all users at
the root of the repository. You can do this by using the
asterisk variable (<code class="literal">*</code>), which means <span class="quote">“<span class="quote">all
users</span>”</span>:</p><div class="informalexample"><pre class="programlisting">
[/]
* = r
</pre></div><p>This is a common setup; notice that no repository
name is mentioned in the section name. This makes all repositories
world-readable to all users. Once all users have read access to
the repositories, you can give explicit
<code class="literal">rw</code> permission to certain users on specific
subdirectories within specific repositories.</p><p>The access file also allows you to define whole groups of
users, much like the Unix <code class="filename">/etc/group</code>
file:</p><div class="informalexample"><pre class="programlisting">
[groups]
calc-developers = harry, sally, joe
paint-developers = frank, sally, jane
everyone = harry, sally, joe, frank, jane
</pre></div><p>Groups can be granted access control just like users.
Distinguish them with an <span class="quote">“<span class="quote">at</span>”</span>
(<code class="literal">@</code>) prefix:</p><div class="informalexample"><pre class="programlisting">
[calc:/projects/calc]
@calc-developers = rw
[paint:/projects/paint]
jane = r
@paint-developers = rw
</pre></div><p>Another important fact is that group permissions are not
overridden by individual user permissions. Rather, the
<span class="emphasis"><em>combination</em></span> of all matching permissions is
granted. In the prior example, Jane is a member of the
<code class="literal">paint-developers</code> group, which has read/write
access. Combined with the <code class="literal">jane = r</code> rule,
this still gives Jane read/write access. Permissions for group
members can only be extended beyond the permissions the group
already has. Restricting users who are part of a group to less
than their group's permissions is impossible.</p><p>Groups can also be defined to contain other groups:</p><div class="informalexample"><pre class="programlisting">
[groups]
calc-developers = harry, sally, joe
paint-developers = frank, sally, jane
everyone = @calc-developers, @paint-developers
</pre></div><p>Subversion 1.5 brought several useful features to the access
file syntax—username aliases, authentication class tokens,
and a new rule exclusion mechanism—all of which further
simplify the maintenance of the access file. We'll describe
first the username aliases feature.</p><p>Some authentication systems expect and carry relatively
short usernames of the sorts we've been describing
here—<code class="literal">harry</code>,
<code class="literal">sally</code>, <code class="literal">joe</code>, and so on. But
other authentication systems—such as those which use LDAP
stores or SSL client certificates—may carry much more
complex usernames. For example, Harry's username in an
LDAP-protected system might be <code class="literal">CN=Harold
Hacker,OU=Engineers,DC=red-bean,DC=com</code>. With
usernames like that, the access file can become quite bloated
with long or obscure usernames that are easy to mistype.
Fortunately, username aliases allow you to have to type the
correct complex username only once, in a statement which assigns to
it a more easily digestable alias.</p><div class="informalexample"><pre class="programlisting">
[aliases]
harry = CN=Harold Hacker,OU=Engineers,DC=red-bean,DC=com
sally = CN=Sally Swatterbug,OU=Engineers,DC=red-bean,DC=com
joe = CN=Gerald I. Joseph,OU=Engineers,DC=red-bean,DC=com
…
</pre></div><p>Once you've defined a set of aliases, you can refer to the
users elsewhere in the access file via their aliases in all the
same places you could have instead used their actual usernames.
Simply prepend an ampersand to the alias to distinguish it from
a regular username:</p><div class="informalexample"><pre class="programlisting">
[groups]
calc-developers = &harry, &sally, &joe
paint-developers = &frank, &sally, &jane
everyone = @calc-developers, @paint-developers
</pre></div><p>You might also choose to use aliases if your users'
usernames change frequently. Doing so allows you to need to
update only the aliases table when these username changes occur,
instead of doing global-search-and-replace operations on the
whole access file.</p><p>Subversion also supports some <span class="quote">“<span class="quote">magic</span>”</span> tokens
for helping you to make rule assignments based on the user's
authentication class. One such token is
the <code class="literal">$authenticated</code> token. Use this token
where you would otherwise specify a username, alias, or group
name in your authorization rules to declare the permissions
granted to any user who has authenticated with any username at
all. Similarly employed is the <code class="literal">$anonymous</code>
token, except that it matches everyone who has
<span class="emphasis"><em>not</em></span> authenticated with a username.</p><div class="informalexample"><pre class="programlisting">
[calendar:/projects/calendar]
$anonymous = r
$authenticated = rw
</pre></div><p>Finally, another handy bit of access file syntax magic is
the use of the tilde (<code class="literal">~</code>) character as an
exclusion marker. In your authorization rules, prefixing a
username, alias, group name, or authentication class token with
a tilde character will cause Subversion to apply the rule to
users who do <span class="emphasis"><em>not</em></span> match the rule. Though
somewhat unnecessarily obfuscated, the following block is
equivalent to the one in the previous example:</p><div class="informalexample"><pre class="programlisting">
[calendar:/projects/calendar]
~$authenticated = r
~$anonymous = rw
</pre></div><p>A less obvious example might be as follows:</p><div class="informalexample"><pre class="programlisting">
[groups]
calc-developers = &harry, &sally, &joe
calc-owners = &hewlett, &packard
calc = @calc-developers, @calc-owners
# Any calc participant has read-write access...
[calc:/projects/calc]
@calc = rw
# ...but only allow the owners to make and modify release tags.
[calc:/projects/calc/tags]
~@calc-owners = r
</pre></div><p>All of the above examples use directories, because defining
access rules on them is the most common case. But is similarly
able to restrict access on file paths, too.</p><div class="informalexample"><pre class="programlisting">
[calendar:/projects/calendar/manager.ics]
harry = rw
sally = r
</pre></div><div class="sidebar" title="Partial Readability and Checkouts"><div class="titlepage"><div><div><p class="title"><strong>Partial Readability and Checkouts</strong></p></div></div></div><p>If you're using Apache as your Subversion server and have
made certain subdirectories of your repository unreadable to
certain users, you need to be aware of a possible nonoptimal
behavior with <span class="command"><strong>svn checkout</strong></span>.</p><p>When the client requests a checkout or update over HTTP,
it makes a single server request and receives a single (often
large) server response. When the server receives the request,
that is the <span class="emphasis"><em>only</em></span> opportunity Apache has
to demand user authentication. This has some odd side
effects. For example, if a certain subdirectory of the
repository is readable only by user Sally, and user Harry
checks out a parent directory, his client will respond to the
initial authentication challenge as Harry. As the server
generates the large response, there's no way it can resend an
authentication challenge when it reaches the special
subdirectory; thus the subdirectory is skipped altogether,
rather than asking the user to reauthenticate as Sally at the
right moment. In a similar way, if the root of the repository
is anonymously world-readable, the entire checkout will be
done without authentication—again, skipping the
unreadable directory, rather than asking for authentication
partway through.</p></div></div><div class="sect1" title="High-level Logging"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.serverconfig.operational-logging"></a>High-level Logging</h2></div></div></div><p>Both the Apache <span class="command"><strong>httpd</strong></span>
and <span class="command"><strong>svnserve</strong></span> Subversion servers provide
support for high-level logging of Subversion operations.
Configuring each of the server options to provide this level of
logging is done differently, of course, but the output from each
is designed to conform to a uniform syntax.</p><p>To enable high-level logging in <span class="command"><strong>svnserve</strong></span>,
you need only use the <code class="option">--log-file</code> command-line
option when starting the server, passing as the value to the
option the file to which <span class="command"><strong>svnserve</strong></span> should
write its log output.</p><div class="informalexample"><pre class="screen">
$ svnserve -d -r /path/to/repositories --log-file /var/log/svn.log
</pre></div><p>Enabling the same in Apache is a bit more involved, but is
essentially an extension of Apache's stock log output
configuration mechanisms—see
<a class="xref" href="#svn.serverconfig.httpd.extra.logging" title="Apache logging">the section called “Apache logging”</a> for
details.</p><p>The following is a list of Subversion action log messages
produced by its high-level logging mechanism, followed by one or
more examples of the log message as it appears in the log
output.</p><div class="variablelist"><dl><dt><span class="term">Checkout or export</span></dt><dd><pre class="programlisting">
checkout-or-export /path r62 depth=infinity
</pre></dd><dt><span class="term">Commit</span></dt><dd><pre class="programlisting">
commit harry r100
</pre></dd><dt><span class="term">Diffs</span></dt><dd><pre class="programlisting">
diff /path r15:20 depth=infinity ignore-ancestry
diff /path1@15 /path2@20 depth=infinity ignore-ancestry
</pre></dd><dt><span class="term">Fetch a directory</span></dt><dd><pre class="programlisting">
get-dir /trunk r17 text
</pre></dd><dt><span class="term">Fetch a file</span></dt><dd><pre class="programlisting">
get-file /path r20 props
</pre></dd><dt><span class="term">Fetch a file revision</span></dt><dd><pre class="programlisting">
get-file-revs /path r12:15 include-merged-revisions
</pre></dd><dt><span class="term">Fetch merge information</span></dt><dd><pre class="programlisting">
get-mergeinfo (/path1 /path2)
</pre></dd><dt><span class="term">Lock</span></dt><dd><pre class="programlisting">
lock /path steal
</pre></dd><dt><span class="term">Log</span></dt><dd><pre class="programlisting">
log (/path1,/path2,/path3) r20:90 discover-changed-paths revprops=()
</pre></dd><dt><span class="term">Replay revisions (svnsync)</span></dt><dd><pre class="programlisting">
replay /path r19
</pre></dd><dt><span class="term">Revision property change</span></dt><dd><pre class="programlisting">
change-rev-prop r50 propertyname
</pre></dd><dt><span class="term">Revision property list</span></dt><dd><pre class="programlisting">
rev-proplist r34
</pre></dd><dt><span class="term">Status</span></dt><dd><pre class="programlisting">
status /path r62 depth=infinity
</pre></dd><dt><span class="term">Switch</span></dt><dd><pre class="programlisting">
switch /pathA /pathB@50 depth=infinity
</pre></dd><dt><span class="term">Unlock</span></dt><dd><pre class="programlisting">
unlock /path break
</pre></dd><dt><span class="term">Update</span></dt><dd><pre class="programlisting">
update /path r17 send-copyfrom-args
</pre></dd></dl></div><p>As a convenience to administrators who wish to post-process
their Subversion high-level logging output (perhaps for
reporting or analysis purposes), Subversion source code
distributions provide a Python module (located at
<code class="filename">tools/server-side/svn_server_log_parse.py</code>)
which can be used to parse Subversion's log output.</p></div><div class="sect1" title="Server Optimization"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.serverconfig.optimization"></a>Server Optimization</h2></div></div></div><p>Part of the due diligence when offering a service such as a
Subversion server involves capacity planning and performance
tuning. Subversion doesn't tend to be particularly greedy in
terms of server resources such as CPU cycles and memory, but any
service can benefit from optimizations, especially when usage of
the service skyrockets<sup>[<a id="idp14098352" href="#ftn.idp14098352" class="footnote">62</a>]</sup>. In this section, we'll discuss
some ways you can tweak your Subversion server configuration
to offer even better performance and scalability.</p><div class="sect2" title="Data Caching"><div class="titlepage"><div><div><h3 class="title"><a id="svn.serverconfig.optimization.caching"></a>Data Caching</h3></div></div></div><p>Generally speaking, the most expensive part of a
Subversion server's job is fetching data from the repository.
Subversion 1.6 attempted to offset this cost by introducing
some in-memory caching of certain classes of data read from
the repository. But Subversion 1.7 takes this a step further,
not only caching the results of some of the more costly
operations, but also by providing in each of the available
servers the means by which fine-tune the size and some
behaviors of the cache.</p><p>For <span class="command"><strong>svnserve</strong></span>, you can specify the size
of the cache using the <code class="option">--memory-cache-size</code>
(<code class="option">-M</code>) command-line option. You can also
dictate whether <span class="command"><strong>svnserve</strong></span> should attempt to
cache content fulltexts and deltas via the
boolean <code class="option">--cache-fulltexts</code>
and <code class="option">--cache-txdeltas</code> options,
respectively.</p><div class="informalexample"><pre class="screen">
$ svnserve -d -r /path/to/repositories \
--memory-cache-size 1024 \
--cache-txdeltas yes \
--cache-fulltexts yes
…
$
</pre></div><p><span class="command"><strong>mod_dav_svn</strong></span> provides the same degree of
cache configurability via <code class="filename">httpd.conf</code>
directives.
The <code class="literal">SVNInMemoryCacheSize</code>,
<code class="literal">SVNCacheFullTexts</code>,
and <code class="literal">SVNCacheTextDeltas</code> directives may be
used at the server configuration level to control Subversion's
data cache characteristics:</p><div class="informalexample"><pre class="programlisting">
<IfModule dav_svn_module>
# Enable a 1 Gb Subversion data cache for both fulltext and deltas.
SVNInMemoryCacheSize 1048576
SVNCacheTextDeltas On
SVNCacheFullTexts On
</IfModule>
</pre></div><p>So what settings should you use? Certainly you need to
consider what resources are available on your server. To get
any benefit out of the cache at all, you'll probably want to
let the cache be at least large enough to hold all the files
which are most commonly accessed in your repository (for
example, your project's <code class="filename">trunk</code> directory
tree).</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>Setting the memory cache size to <code class="literal">0</code>
will disable this enhanced caching mechanism and cause
Subversion to fall back to using the older cache mechanisms
introduced in Subversion 1.6.</p></td></tr></table></div><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>Currently, only repositories which make use of the FSFS
backend data store make use of this data caching
functionality.</p></td></tr></table></div></div><div class="sect2" title="Network Compression of Data"><div class="titlepage"><div><div><h3 class="title"><a id="svn.serverconfig.optimization.compression"></a>Network Compression of Data</h3></div></div></div><p>Compressing the data transmitted across the wire can
greatly reduce the size of those network transmissions, but
comes at the cost of server (and client) CPU cycles.
Depending on your server's CPU capacity, the typical access
patterns of the clients who use your servers, and the
bandwidth of the networks between them, you might wish to fine
tune just how hard your server will work to compress the data
it sends across the wire. To assist with this fine tuning
process, Subversion 1.7 offers
the <code class="option">--compression</code> (<code class="option">-c</code>)
option to <span class="command"><strong>svnserve</strong></span> and
the <code class="literal">SVNCompressionLevel</code> directive
for <span class="command"><strong>mod_dav_svn</strong></span>. Both accept a value which
is an integer between 0 and 9 (inclusive), where 9 offers the
best compression of wire data, and 0 disables compression
altogether.</p><p>For example, on a local area network (LAN) with 1-Gigabit
connections, it might not make sense to have the server
compress its network transmissions (which also forces the
clients to decompress them), as the network itself is so fast
that users won't really benefit from the smaller overall
network payload. On the other hand, servers which are
accessed primarily by clients with low-bandwidth connections
would be doing those clients a favor by minimizing the overall
size of its network communications.</p></div></div><div class="sect1" title="Supporting Multiple Repository Access Methods"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.serverconfig.multimethod"></a>Supporting Multiple Repository Access Methods</h2></div></div></div><p>You've seen how a repository can be accessed in many
different ways. But is it possible—or safe—for your
repository to be accessed by multiple methods simultaneously?
The answer is yes, provided you use a bit of foresight.</p><p>At any given time, these processes may require read and
write access to your repository:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>Regular system users using a Subversion client (as
themselves) to access the repository directly via
<code class="literal">file://</code> URLs</p></li><li class="listitem"><p>Regular system users connecting to SSH-spawned private
<span class="command"><strong>svnserve</strong></span> processes (running as
themselves), which access the repository</p></li><li class="listitem"><p>An <span class="command"><strong>svnserve</strong></span> process—either a
daemon or one launched by
<span class="command"><strong>inetd</strong></span>—running as a particular fixed
user</p></li><li class="listitem"><p>An Apache <span class="command"><strong>httpd</strong></span> process, running as a
particular fixed user</p></li></ul></div><p>The most common problem administrators run into is
repository ownership and permissions. Does every process (or
user) in the preceding list have the rights to read and write the
repository's underlying data files? Assuming you have a
Unix-like operating system, a straightforward approach might be
to place every potential repository user into a
new <code class="literal">svn</code> group, and make the repository wholly
owned by that group. But even that's not enough, because a
process may write to the database files using an unfriendly
umask—one that prevents access by other users.</p><p>So the next step beyond setting up a common group for
repository users is to force every repository-accessing process
to use a sane umask. For users accessing the repository
directly, you can make the <span class="command"><strong>svn</strong></span> program into a
wrapper script that first runs <strong class="userinput"><code>umask 002</code></strong> and
then runs the real <span class="command"><strong>svn</strong></span> client program. You
can write a similar wrapper script for the
<span class="command"><strong>svnserve</strong></span> program, and add a <strong class="userinput"><code>umask
002</code></strong> command to Apache's own startup script,
<code class="filename">apachectl</code>. For example:</p><div class="informalexample"><pre class="screen">
$ cat /usr/bin/svn
#!/bin/sh
umask 002
/usr/bin/svn-real "$@"
</pre></div><p>Another common problem is often encountered on Unix-like
systems. If your repository is backed by Berkeley DB, for
example, it occasionally creates new log files to journal its
actions. Even if the Berkeley DB repository is wholly owned by
the <span class="command"><strong>svn</strong></span> group, these newly created log files
won't necessarily be owned by that same group, which then
creates more permissions problems for your users. A good
workaround is to set the group SUID bit on the
repository's <code class="filename">db</code> directory. This causes all
newly created log files to have the same group owner as the
parent directory.</p><p>Once you've jumped through these hoops, your repository
should be accessible by all the necessary processes. It may
seem a bit messy and complicated, but the problems of having
multiple users sharing write access to common files are classic
ones that are not often elegantly solved.</p><p>Fortunately, most repository administrators will never
<span class="emphasis"><em>need</em></span> to have such a complex configuration.
Users who wish to access repositories that live on the same
machine are not limited to using <code class="literal">file://</code>
access URLs—they can typically contact the Apache HTTP
server or <span class="command"><strong>svnserve</strong></span> using
<code class="literal">localhost</code> for the server name in their
<code class="literal">http://</code> or <code class="literal">svn://</code> URL.
And maintaining multiple server processes for your Subversion
repositories is likely to be more of a headache than necessary.
We recommend that you choose a single server that best meets your
needs and stick with it!</p><div class="sidebar" title="The svn+ssh:// Server Checklist"><div class="titlepage"><div><div><p class="title"><strong>The svn+ssh:// Server Checklist</strong></p></div></div></div><p>It can be quite tricky to get a bunch of users with
existing SSH accounts to share a repository without
permissions problems. If you're confused about all the things
that you (as an administrator) need to do on a Unix-like
system, here's a quick checklist that resummarizes some of the
topics discussed in this section:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>All of your SSH users need to be able to read and
write to the repository, so put all the SSH users into a
single group.</p></li><li class="listitem"><p>Make the repository wholly owned by that group.</p></li><li class="listitem"><p>Set the group permissions to read/write.</p></li><li class="listitem"><p>Your users need to use a sane umask when accessing the
repository, so make sure <span class="command"><strong>svnserve</strong></span>
(<code class="filename">/usr/bin/svnserve</code>, or wherever it
lives in <code class="literal">$PATH</code>) is actually a wrapper
script that runs <strong class="userinput"><code>umask 002</code></strong> and
executes the real <span class="command"><strong>svnserve</strong></span>
binary.</p></li><li class="listitem"><p>Take similar measures when using
<span class="command"><strong>svnlook</strong></span> and
<span class="command"><strong>svnadmin</strong></span>. Either run them with a sane
umask or wrap them as just described.</p></li></ul></div></div></div><div class="footnotes"><br></br><hr width="100" align="left"></hr><div class="footnote"><p><sup>[<a id="ftn.idp13371568" href="#idp13371568" class="para">52</a>] </sup>See RFC
2195.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp13490048" href="#idp13490048" class="para">53</a>] </sup>Note
that using any sort of <span class="command"><strong>svnserve</strong></span>-enforced
access control at all is a bit pointless; the user already has
direct access to the repository
database.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp13493328" href="#idp13493328" class="para">54</a>] </sup>We don't actually recommend this, since
RSH is notably less secure than SSH.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp13550992" href="#idp13550992" class="para">55</a>] </sup>See
<a class="ulink" href="http://www.webdav.org/" target="_top">http://www.webdav.org/</a>.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp13552256" href="#idp13552256" class="para">56</a>] </sup>They really
hate doing that.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp13639200" href="#idp13639200" class="para">57</a>] </sup>See
<a class="ulink" href="http://httpd.apache.org/docs/current/programs/htpasswd.html" target="_top">http://httpd.apache.org/docs/current/programs/htpasswd.html</a>.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp13681040" href="#idp13681040" class="para">58</a>] </sup>See
<a class="ulink" href="http://httpd.apache.org/docs/current/programs/htdigest.html" target="_top">http://httpd.apache.org/docs/current/programs/htdigest.html</a>.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp13773568" href="#idp13773568" class="para">59</a>] </sup>While self-signed certificates are
still vulnerable to a <span class="quote">“<span class="quote">man-in-the-middle</span>”</span>
attack, such an attack is much more difficult for a casual
observer to pull off, compared to sniffing unprotected
passwords.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp13940608" href="#idp13940608" class="para">60</a>] </sup><a class="ulink" href="http://subversion.tigris.org/issues/show_bug.cgi?id=3457" target="_top">http://subversion.tigris.org/issues/show_bug.cgi?id=3457</a> tracks these problems.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp13983120" href="#idp13983120" class="para">61</a>] </sup>A common theme in
this book!</p></div><div class="footnote"><p><sup>[<a id="ftn.idp14098352" href="#idp14098352" class="para">62</a>] </sup>In Subversion's case, the
skyrocketing affect is, of course, due to its cool name. Well,
that and its popularity, reliability, ease of
use….</p></div></div></div><div class="chapter" title="Chapter 7. Customizing Your Subversion Experience"><div class="titlepage"><div><div><h2 class="title"><a id="svn.customization"></a>Chapter 7. Customizing Your Subversion Experience</h2></div></div></div><div class="toc"><p><strong>Table of Contents</strong></p><dl><dt><span class="sect1"><a href="#svn.advanced.confarea">Runtime Configuration Area</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.advanced.confarea.layout">Configuration Area Layout</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.confarea.windows-registry">Configuration and the Windows Registry</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.confarea.opts">Configuration Options</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.advanced.confarea.opts.servers">Servers</a></span></dt><dt><span class="sect3"><a href="#svn.advanced.confarea.opts.config">Config</a></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.advanced.l10n">Localization</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.advanced.l10n.understanding">Understanding Locales</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.l10n.svnuse">Subversion's Use of Locales</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.advanced.externaleditors">Using External Editors</a></span></dt><dt><span class="sect1"><a href="#svn.advanced.externaldifftools">Using External Differencing and Merge Tools</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.advanced.externaldifftools.diff">External diff</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.externaldifftools.diff3">External diff3</a></span></dt><dt><span class="sect2"><a href="#svn.advanced.externaldifftools.merge">External merge</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.customization.summary">Summary</a></span></dt></dl></div><p>Version control can be a complex subject, as much art as
science, that offers myriad ways of getting stuff done.
Throughout this book, you've read of the various Subversion
command-line client subcommands and the options that modify their
behavior. In this chapter, we'll look into still more ways to
customize the way Subversion works for you—setting up the
Subversion runtime configuration, using external helper
applications, Subversion's interaction with the operating system's
configured locale, and so on.</p><div class="sect1" title="Runtime Configuration Area"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.advanced.confarea"></a>Runtime Configuration Area</h2></div></div></div><p>Subversion provides many optional behaviors that the user can
control. Many of these options are of the kind
that a user would wish to apply to all Subversion operations.
So, rather than forcing users to remember command-line arguments
for specifying these options and to use them for every
operation they perform, Subversion uses configuration files,
segregated into a Subversion configuration area.</p><p>The Subversion <em class="firstterm">configuration area</em> is
a two-tiered hierarchy of option names and their values.
Usually, this boils down to a special directory that contains
<em class="firstterm">configuration files</em> (the first tier),
which are just text files in standard INI format where
<span class="quote">“<span class="quote">sections</span>”</span> provide the second tier. You can
easily edit these files using your favorite text editor (such as
Emacs or vi), and they contain directives read by the client to
determine which of several optional behaviors the user
prefers.</p><div class="sect2" title="Configuration Area Layout"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.confarea.layout"></a>Configuration Area Layout</h3></div></div></div><p>The first time the <span class="command"><strong>svn</strong></span> command-line
client is executed, it creates a per-user configuration area.
On Unix-like systems, this area appears as a directory
named <code class="filename">.subversion</code> in the user's home
directory. On Win32 systems, Subversion creates a folder
named <code class="filename">Subversion</code>, typically inside
the <code class="filename">Application Data</code> area of the user's
profile directory (which, by the way, is usually a hidden
directory). However, on this platform, the exact location
differs from system to system and is dictated by the Windows
Registry.<sup>[<a id="idp15216960" href="#ftn.idp15216960" class="footnote">63</a>]</sup>
We will refer to the per-user configuration area using its
Unix name, <code class="filename">.subversion</code>.</p><p>In addition to the per-user configuration area, Subversion
also recognizes the existence of a system-wide configuration
area. This gives system administrators the ability to
establish defaults for all users on a given machine. Note
that the system-wide configuration area alone does not dictate
mandatory policy—the settings in the per-user
configuration area override those in the system-wide one, and
command-line arguments supplied to the <span class="command"><strong>svn</strong></span>
program have the final word on behavior. On Unix-like
platforms, the system-wide configuration area is
expected to be the <code class="filename">/etc/subversion</code>
directory; on Windows machines, it looks for a
<code class="filename">Subversion</code> directory inside the common
<code class="filename">Application Data</code> location (again, as
specified by the Windows Registry). Unlike the per-user
case, the <span class="command"><strong>svn</strong></span> program does not attempt
to create the system-wide configuration area.</p><p>The per-user configuration area currently contains three
files—two configuration files (<code class="filename">config</code> and
<code class="filename">servers</code>), and a <code class="filename">README.txt</code>
file, which describes the INI format. At the time of their
creation, the files contain default values for each of the
supported Subversion options, mostly commented out and grouped
with textual descriptions about how the values for the key
affect Subversion's behavior. To change a certain behavior,
you need only to load the appropriate configuration file into
a text editor, and to modify the desired option's value. If at
any time you wish to have the default configuration settings
restored, you can simply remove (or rename) your configuration
directory and then run some innocuous <span class="command"><strong>svn</strong></span>
command, such as <strong class="userinput"><code>svn --version</code></strong>. A new
configuration directory with the default contents will be
created.</p><p>Subversion also allows you to override individual
configuration option values at the command line via
the <code class="option">--config-option</code> option, which is
especially useful if you need to make a (very) temporary
change in behavior. For more about this option's proper
usage, see <a class="xref" href="#svn.ref.svn.sw" title="svn Options">the section called “svn Options”</a>.</p><p>The per-user configuration area also contains a cache of
authentication data. The <code class="filename">auth</code> directory
holds a set of subdirectories that contain pieces of cached
information used by Subversion's various supported
authentication methods. This directory is created in such a
way that only the user herself has permission to read its
contents.</p></div><div class="sect2" title="Configuration and the Windows Registry"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.confarea.windows-registry"></a>Configuration and the Windows Registry</h3></div></div></div><p>In addition to the usual INI-based configuration area,
Subversion clients running on Windows platforms may also use
the Windows Registry to hold the configuration data. The
option names and their values are the same as in the INI
files. The <span class="quote">“<span class="quote">file/section</span>”</span> hierarchy is
preserved as well, though addressed in a slightly different
fashion—in this schema, files and sections are just
levels in the Registry key tree.</p><p>Subversion looks for system-wide configuration values
under the
<code class="literal">HKEY_LOCAL_MACHINE\Software\Tigris.org\Subversion</code>
key. For example, the <code class="literal">global-ignores</code> option,
which is in the <code class="literal">miscellany</code> section of the
<code class="filename">config</code> file, would be found at
<code class="literal">HKEY_LOCAL_MACHINE\Software\Tigris.org\Subversion\Config\Miscellany\global-ignores</code>.
Per-user configuration values should be stored under
<code class="literal">HKEY_CURRENT_USER\Software\Tigris.org\Subversion</code>.</p><p>Registry-based configuration options are parsed
<span class="emphasis"><em>before</em></span> their file-based counterparts, so
they are overridden by values found in the configuration files. In
other words, Subversion looks for configuration information in
the following locations on a Windows system; lower-numbered
locations take precedence over higher-numbered locations:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p>Command-line options</p></li><li class="listitem"><p>The per-user INI files</p></li><li class="listitem"><p>The per-user Registry values</p></li><li class="listitem"><p>The system-wide INI files</p></li><li class="listitem"><p>The system-wide Registry values</p></li></ol></div><p>Also, the Windows Registry doesn't really support the
notion of something being <span class="quote">“<span class="quote">commented out.</span>”</span>
However, Subversion will ignore any option key whose name
begins with a hash (<code class="literal">#</code>) character. This
allows you to effectively comment out a Subversion option
without deleting the entire key from the Registry, obviously
simplifying the process of restoring that option.</p><p>The <span class="command"><strong>svn</strong></span> command-line client never
attempts to write to the Windows Registry and will not attempt
to create a default configuration area there. You can create
the keys you need using the <span class="command"><strong>REGEDIT</strong></span>
program. Alternatively, you can create a
<code class="filename">.reg</code> file (such as the one in <a class="xref" href="#svn.advanced.confarea.windows-registry.ex-1" title="Example 7.1. Sample registration entries (.reg) file">Example 7.1, “Sample registration entries (.reg) file”</a>), and
then double-click on that file's icon in the Explorer shell,
which will cause the data to be merged into your
Registry.</p><div class="example"><a id="svn.advanced.confarea.windows-registry.ex-1"></a><p class="title"><strong>Example 7.1. Sample registration entries (.reg) file</strong></p><div class="example-contents"><pre class="programlisting">
REGEDIT4
[HKEY_LOCAL_MACHINE\Software\Tigris.org\Subversion\Servers\groups]
[HKEY_LOCAL_MACHINE\Software\Tigris.org\Subversion\Servers\global]
"#http-auth-types"="basic;digest;negotiate"
"#http-compression"="yes"
"#http-library"=""
"#http-proxy-exceptions"=""
"#http-proxy-host"=""
"#http-proxy-password"=""
"#http-proxy-port"=""
"#http-proxy-username"=""
"#http-timeout"="0"
"#neon-debug-mask"=""
"#ssl-authority-files"=""
"#ssl-client-cert-file"=""
"#ssl-client-cert-password"=""
"#ssl-pkcs11-provider"=""
"#ssl-trust-default-ca"=""
"#store-auth-creds"="yes"
"#store-passwords"="yes"
"#store-plaintext-passwords"="ask"
"#store-ssl-client-cert-pp"="yes"
"#store-ssl-client-cert-pp-plaintext"="ask"
"#username"=""
[HKEY_CURRENT_USER\Software\Tigris.org\Subversion\Config\auth]
"#password-stores"="windows-cryptoapi"
[HKEY_CURRENT_USER\Software\Tigris.org\Subversion\Config\helpers]
"#diff-cmd"=""
"#diff-extensions"="-u"
"#diff3-cmd"=""
"#diff3-has-program-arg"=""
"#editor-cmd"="notepad"
"#merge-tool-cmd"=""
[HKEY_CURRENT_USER\Software\Tigris.org\Subversion\Config\tunnels]
[HKEY_CURRENT_USER\Software\Tigris.org\Subversion\Config\miscellany]
"#enable-auto-props"="no"
"#global-ignores"="*.o *.lo *.la *.al .libs *.so *.so.[0-9]* *.a *.pyc *.pyo *.rej *~ #*# .#* .*.swp .DS_Store"
"#interactive-conflicts"="yes"
"#log-encoding"=""
"#mime-types-file"=""
"#no-unlock"="no"
"#preserved-conflict-file-exts"="doc ppt xls od?"
"#use-commit-times"="no"
[HKEY_CURRENT_USER\Software\Tigris.org\Subversion\Config\auto-props]
</pre></div></div><br class="example-break"></br><p><a class="xref" href="#svn.advanced.confarea.windows-registry.ex-1" title="Example 7.1. Sample registration entries (.reg) file">Example 7.1, “Sample registration entries (.reg) file”</a>
shows the contents of a <code class="filename">.reg</code> file, which
contains some of the most commonly used configuration options
and their default values. Note the presence of both
system-wide (for network proxy-related options) and per-user
settings (editor programs and password storage, among others).
Also note that all the options are effectively commented out.
You need only to remove the hash (<code class="literal">#</code>)
character from the beginning of the option names and set the
values as you desire.</p></div><div class="sect2" title="Configuration Options"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.confarea.opts"></a>Configuration Options</h3></div></div></div><p>In this section, we will discuss the specific
runtime configuration options that Subversion currently
supports.</p><div class="sect3" title="Servers"><div class="titlepage"><div><div><h4 class="title"><a id="svn.advanced.confarea.opts.servers"></a>Servers</h4></div></div></div><p>The <code class="filename">servers</code> file contains
Subversion configuration options related to the network
layers. There are two special sections in this
file—<code class="literal">[groups]</code> and
<code class="literal">[global]</code>. The <code class="literal">[groups]</code>
section is essentially a cross-reference table. The keys in
this section are the names of other sections in the file;
their values are <em class="firstterm">globs</em>—textual
tokens that possibly contain wildcard
characters—that are compared against the hostnames of
the machine to which Subversion requests are sent.</p><div class="informalexample"><pre class="programlisting">
[groups]
beanie-babies = *.red-bean.com
collabnet = svn.collab.net
[beanie-babies]
…
[collabnet]
…
</pre></div><p>When Subversion is used over a network, it attempts to
match the name of the server it is trying to reach with a
group name under the <code class="literal">[groups]</code> section. If
a match is made, Subversion then looks for a section in the
<code class="filename">servers</code> file whose name is the matched
group's name. From that section, it reads the actual network
configuration settings.</p><p>The <code class="literal">[global]</code> section contains the
settings that are meant for all of the servers not matched
by one of the globs under the <code class="literal">[groups]</code>
section. The options available in this section are
exactly the same as those that are valid for the other server
sections in the file (except, of course, the special
<code class="literal">[groups]</code> section), and are as
follows:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">http-auth-types</code></span></dt><dd><p>This is a semicolon-delimited list of HTTP
authentication types which the client will deem
acceptable. Valid types
are <code class="literal">basic</code>, <code class="literal">digest</code>,
and <code class="literal">negotiate</code>, with the default
behavior being acceptance of any these authentication
types. A client which insists on not transmitting
authentication credentials in cleartext might, for
example, be configured such that the value of this
option is
<code class="literal">digest;negotiate</code>—omitting
<code class="literal">basic</code> from the list. (Note that
this setting is only honored by Subversion's
Neon-based HTTP provider module.)</p></dd><dt><span class="term"><code class="literal">http-compression</code></span></dt><dd><p>This specifies whether Subversion should
attempt to compress network requests made to DAV-ready
servers. The default value is <code class="literal">yes</code>
(though compression will occur only if that capability
is compiled into the network layer). Set this to
<code class="literal">no</code> to disable compression, such as
when debugging network transmissions.</p></dd><dt><span class="term"><code class="literal">http-library</code></span></dt><dd><p>Subversion provides a pair of repository access
modules that understand its WebDAV network protocol.
The original one, which shipped with Subversion 1.0, is
<code class="literal">libsvn_ra_neon</code> (though back then it
was called <code class="literal">libsvn_ra_dav</code>). Newer
Subversion versions also provide
<code class="literal">libsvn_ra_serf</code>, which uses a
different underlying implementation and aims to
support some of the newer HTTP concepts.</p><p>At this point, <code class="literal">libsvn_ra_serf</code>
is still considered experimental, though it appears to
work in the common cases quite well. To encourage
experimentation, Subversion provides the
<code class="literal">http-library</code> runtime configuration
option to allow users to specify (generally, or in a
per-server-group fashion) which WebDAV access module
they'd prefer to use—<code class="literal">neon</code> or
<code class="literal">serf</code>.</p></dd><dt><span class="term"><code class="literal">http-proxy-exceptions</code></span></dt><dd><p>This specifies a comma-separated list of patterns
for repository hostnames that should be accessed
directly, without using the proxy machine. The
pattern syntax is the same as is used in the Unix
shell for filenames. A repository hostname matching
any of these patterns will not be proxied.</p></dd><dt><span class="term"><code class="literal">http-proxy-host</code></span></dt><dd><p>This specifies the hostname of the proxy computer
through which your HTTP-based Subversion requests must
pass. It defaults to an empty value, which means that
Subversion will not attempt to route HTTP requests
through a proxy computer, and will instead attempt to
contact the destination machine directly.</p></dd><dt><span class="term"><code class="literal">http-proxy-password</code></span></dt><dd><p>This specifies the password to supply to the proxy
machine. It defaults to an empty value.</p></dd><dt><span class="term"><code class="literal">http-proxy-port</code></span></dt><dd><p>This specifies the port number on the proxy host
to use. It defaults to an empty value.</p></dd><dt><span class="term"><code class="literal">http-proxy-username</code></span></dt><dd><p>This specifies the username to supply to the proxy
machine. It defaults to an empty value.</p></dd><dt><span class="term"><code class="literal">http-timeout</code></span></dt><dd><p>This specifies the amount of time, in seconds, to
wait for a server response. If you experience
problems with a slow network connection causing
Subversion operations to time out, you should increase
the value of this option. The default value is
<code class="literal">0</code>, which instructs the underlying
HTTP library, Neon, to use its default timeout
setting.</p></dd><dt><span class="term"><code class="literal">neon-debug-mask</code></span></dt><dd><p>This is an integer mask that the underlying HTTP
library, Neon, uses for choosing what type of
debugging output to yield. The default value is
<code class="literal">0</code>, which will silence all debugging
output. For more information about how Subversion
makes use of Neon, see <a class="xref" href="#svn.developer" title="Chapter 8. Embedding Subversion">Chapter 8, <em>Embedding Subversion</em></a>.</p></dd><dt><span class="term"><code class="literal">ssl-authority-files</code></span></dt><dd><p>This is a semicolon-delimited list of paths to files
containing certificates of the certificate authorities
(or CAs) that
are accepted by the Subversion client when accessing the
repository over HTTPS.</p></dd><dt><span class="term"><code class="literal">ssl-client-cert-file</code></span></dt><dd><p>If a host (or set of hosts) requires an SSL client
certificate, you'll normally be prompted for a path to
your certificate. By setting this variable to that
same path, Subversion will be able to find your client
certificate automatically without prompting you.
There's no standard place to store your certificate on
disk; Subversion will grab it from any path you
specify.</p></dd><dt><span class="term"><code class="literal">ssl-client-cert-password</code></span></dt><dd><p>If your SSL client certificate file is encrypted
by a passphrase, Subversion will prompt you for the
passphrase whenever the certificate is used. If you
find this annoying (and don't mind storing the
password in the <code class="filename">servers</code> file),
you can set this variable to the certificate's
passphrase. You won't be prompted anymore.</p></dd><dt><span class="term"><code class="literal">ssl-pkcs11-provider</code></span></dt><dd><p>The value of this option is the name of the
PKCS#11 provider from which an SSL client certificate
will be drawn (if the server asks for one). This
setting is only honored by Subversion's Neon-based
HTTP provider module.</p></dd><dt><span class="term"><code class="literal">ssl-trust-default-ca</code></span></dt><dd><p>Set this variable to <code class="literal">yes</code> if you
want Subversion to automatically trust the set of
default CAs that ship with OpenSSL.</p></dd><dt><span class="term"><code class="literal">store-auth-creds</code></span></dt><dd><p>This setting is the same as
<code class="literal">store-passwords</code>, except that it
enables or disables on-disk caching of
<span class="emphasis"><em>all</em></span> authentication information:
usernames, passwords, server certificates, and any
other types of cacheable credentials.</p></dd><dt><span class="term"><code class="literal">store-passwords</code></span></dt><dd><p>This instructs Subversion to cache, or not to
cache, passwords that are supplied by the user in
response to server authentication challenges. The
default value is <code class="literal">yes</code>. Set this to
<code class="literal">no</code> to disable this on-disk password
caching. You can override this option for a single
instance of the <span class="command"><strong>svn</strong></span> command using
the <code class="option">--no-auth-cache</code> command-line
parameter (for those subcommands that support it).
For more information regarding that, see
<a class="xref" href="#svn.serverconfig.netmodel.credcache" title="Caching credentials">the section called “Caching credentials”</a>.
Note that regardless of how this option is configured,
Subversion will not store passwords in plaintext
unless the <code class="literal">store-plaintext-passwords</code>
option is also set to <code class="literal">yes</code>.</p></dd><dt><span class="term"><code class="literal">store-plaintext-passwords</code></span></dt><dd><p>This variable is only important on UNIX-like systems.
It controls what the Subversion client does in case
the password for the current authentication realm can
only be cached on disk in unencrypted form, in the
<code class="filename">~/.subversion/auth/</code> caching area.
You can set it to <code class="literal">yes</code> or
<code class="literal">no</code> to enable or disable caching of
passwords in unencrypted form, respectively.
The default setting is <code class="literal">ask</code>, which causes
the Subversion client to ask you each time a
<span class="emphasis"><em>new</em></span> password is about to be added to
the <code class="filename">~/.subversion/auth/</code> caching area.</p></dd><dt><span class="term"><code class="literal">store-ssl-client-cert-pp</code></span></dt><dd><p>This option controls whether Subversion will cache
SSL client certificate passphrases provided by the
user. Its value defaults to <code class="literal">yes</code>.
Set this to <code class="literal">no</code> to disable this
passphrase caching.</p></dd><dt><span class="term"><code class="literal">store-ssl-client-cert-pp-plaintext</code></span></dt><dd><p>This option controls whether Subversion, when
attempting to cache an SSL client certificate
passphrase, will be allowed to do so using its on-disk
plaintext storage mechanism. The default value of
this option is <code class="literal">ask</code>, which causes
the Subversion client to ask you each time a
<span class="emphasis"><em>new</em></span> client certificate passphrase
is about to be added to
the <code class="filename">~/.subversion/auth/</code> caching
area. Set this option's value
to <code class="literal">yes</code> or <code class="literal">no</code> to
indicate your preference and avoid related
prompts.</p></dd></dl></div></div><div class="sect3" title="Config"><div class="titlepage"><div><div><h4 class="title"><a id="svn.advanced.confarea.opts.config"></a>Config</h4></div></div></div><p>The <code class="filename">config</code> file contains the rest
of the currently available Subversion runtime
options—those not related to networking. There are
only a few options in use as of this writing, but they are
again grouped into sections in expectation of future
additions.</p><p>The <code class="literal">[auth]</code> section contains settings
related to Subversion's authentication and authorization
against the repository. It contains the following:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">password-stores</code></span></dt><dd><p>This comma-delimited list specifies which (if any)
system-provided password stores Subversion should
attempt to use when saving and retrieving cached
authentication credentials, and in what order
Subversion should prefer them. The default value is
<code class="literal">gnome-keyring, kwallet, keychain,
windows-crypto-api</code>, representing the GNOME
Keyring, KDE Wallet, Mac OS X Keychain, and Microsoft
Windows cryptography API, respectively. Listed stores
which are not available on the system are
ignored.</p></dd><dt><span class="term"><code class="literal">store-passwords</code></span></dt><dd><p>This option has been deprecated from
the <code class="filename">config</code> file. It now lives as
a per-server configuration item in
the <code class="filename">servers</code> configuration area.
See <a class="xref" href="#svn.advanced.confarea.opts.servers" title="Servers">the section called “Servers”</a>
for details.</p></dd><dt><span class="term"><code class="literal">store-auth-creds</code></span></dt><dd><p>This option has been deprecated from
the <code class="filename">config</code> file. It now lives as
a per-server configuration item in
the <code class="filename">servers</code> configuration area.
See <a class="xref" href="#svn.advanced.confarea.opts.servers" title="Servers">the section called “Servers”</a>
for details.</p></dd></dl></div><p>The <code class="literal">[helpers]</code> section controls which
external applications Subversion uses to accomplish its
tasks. Valid options in this section are:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">diff-cmd</code></span></dt><dd><p>This specifies the absolute path of a differencing
program, used when Subversion generates
<span class="quote">“<span class="quote">diff</span>”</span> output (such as when using the
<span class="command"><strong>svn diff</strong></span> command). By default,
Subversion uses an internal differencing
library—setting this option will cause it to
perform this task using an external program. See
<a class="xref" href="#svn.advanced.externaldifftools" title="Using External Differencing and Merge Tools">the section called “Using External Differencing and Merge Tools”</a> for
more details on using such programs.</p></dd><dt><span class="term"><code class="literal">diff-extensions</code></span></dt><dd><p>Like the <code class="option">--extensions</code>
(<code class="option">-x</code>) command-line option, this
specifies additional options passed to the file
content differencing engine. The set of meaningful
extension options differs depending on whether the
client is using Subversion's internal differencing
engine or an external mechanism. See the output
of <strong class="userinput"><code>svn help diff</code></strong> for details.
The default value for this option
is <code class="literal">-u</code>.</p></dd><dt><span class="term"><code class="literal">diff3-cmd</code></span></dt><dd><p>This specifies the absolute path of a three-way
differencing program. Subversion uses this program to
merge changes made by the user with those received
from the repository. By default, Subversion uses an
internal differencing library—setting this
option will cause it to perform this task using an
external program. See <a class="xref" href="#svn.advanced.externaldifftools" title="Using External Differencing and Merge Tools">the section called “Using External Differencing and Merge Tools”</a> for more
details on using such programs.</p></dd><dt><span class="term"><code class="literal">diff3-has-program-arg</code></span></dt><dd><p>This flag should be set to <code class="literal">true</code>
if the program specified by the
<code class="literal">diff3-cmd</code> option accepts a
<code class="option">--diff-program</code> command-line
parameter.</p></dd><dt><span class="term"><code class="literal">editor-cmd</code></span></dt><dd><p>This specifies the program Subversion will use to
query the user for certain types of textual metadata
or when interactively resolving conflicts. See
<a class="xref" href="#svn.advanced.externaleditors" title="Using External Editors">the section called “Using External Editors”</a> for
more details on using external text editors with
Subversion.</p></dd><dt><span class="term"><code class="literal">merge-tool-cmd</code></span></dt><dd><p>This specifies the program that Subversion will
use to perform three-way merge operations on your
versioned files. See <a class="xref" href="#svn.advanced.externaldifftools" title="Using External Differencing and Merge Tools">the section called “Using External Differencing and Merge Tools”</a> for more
details on using such programs.</p></dd></dl></div><p>The <code class="literal">[tunnels]</code> section allows you to
define new tunnel schemes for use with
<span class="command"><strong>svnserve</strong></span> and <code class="literal">svn://</code>
client connections. For more details, see <a class="xref" href="#svn.serverconfig.svnserve.sshauth" title="Tunneling over SSH">the section called “Tunneling over SSH”</a>.</p><p>The <code class="literal">miscellany</code> section is where
everything that doesn't belong elsewhere winds
up.<sup>[<a id="idp15396240" href="#ftn.idp15396240" class="footnote">64</a>]</sup> In this section, you can
find:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">enable-auto-props</code></span></dt><dd><p>This instructs Subversion to automatically set
properties on newly added or imported files. The
default value is <code class="literal">no</code>, so set this to
<code class="literal">yes</code> to enable this feature.
The <code class="literal">[auto-props]</code> section of this file
specifies which properties are to be set on which files.</p></dd><dt><span class="term"><code class="literal">global-ignores</code></span></dt><dd><p>When running the <span class="command"><strong>svn status</strong></span>
command, Subversion lists unversioned files and
directories along with the versioned ones, annotating
them with a <code class="literal">?</code> character (see <a class="xref" href="#svn.tour.cycle.examine.status" title="See an overview of your changes">the section called “See an overview of your changes”</a>). Sometimes it can
be annoying to see uninteresting, unversioned
items—for example, object files that result from
a program's compilation—in this display. The
<code class="literal">global-ignores</code> option is a list of
whitespace-delimited globs that describe the names of
files and directories that Subversion should not
display unless they are versioned. The default value
is <code class="literal">*.o *.lo *.la *.al .libs *.so *.so.[0-9]*
*.a *.pyc *.pyo *.rej *~ #*# .#* .*.swp
.DS_Store</code>.</p><p>As well as <span class="command"><strong>svn status</strong></span>, the
<span class="command"><strong>svn add</strong></span> and <span class="command"><strong>svn import</strong></span>
commands also ignore files that match the list
when they are scanning a directory. You can override this
behavior for a single instance of any of these commands
by explicitly specifying the filename, or by using
the <code class="option">--no-ignore</code> command-line flag.</p><p>For information on finer-grained control of
ignored items, see
<a class="xref" href="#svn.advanced.props.special.ignore" title="Ignoring Unversioned Items">the section called “Ignoring Unversioned Items”</a>.</p></dd><dt><span class="term"><code class="literal">interactive-conflicts</code></span></dt><dd><p>This is a Boolean option that specifies whether
Subversion should try to resolve conflicts
interactively. If its value is <code class="literal">yes</code>
(which is the default value), Subversion will prompt
the user for how to handle conflicts in the manner
demonstrated in <a class="xref" href="#svn.tour.cycle.resolve" title="Resolve Any Conflicts">the section called “Resolve Any Conflicts”</a>. Otherwise, it will simply flag the conflict and
continue its operation, postponing resolution to a later
time.</p></dd><dt><span class="term"><code class="literal">log-encoding</code></span></dt><dd><p>This variable sets the default character set
encoding for commit log messages. It's a permanent
form of the <code class="option">--encoding</code> option (see
<a class="xref" href="#svn.ref.svn.sw" title="svn Options">the section called “svn Options”</a>). The Subversion
repository stores log messages in UTF-8 and assumes
that your log message is written using your operating
system's native locale. You should specify a
different encoding if your commit messages are written
in any other encoding.</p></dd><dt><span class="term"><code class="literal">mime-types-file</code></span></dt><dd><p>This option, new to Subversion 1.5, specifies the
path of a MIME types mapping file, such as the
<code class="filename">mime.types</code> file provided by the
Apache HTTP Server. Subversion uses this file to
assign MIME types to newly added or imported files.
See <a class="xref" href="#svn.advanced.props.auto" title="Automatic Property Setting">the section called “Automatic Property Setting”</a> and
<a class="xref" href="#svn.advanced.props.special.mime-type" title="File Content Type">the section called “File Content Type”</a> for more about Subversion's detection and use of
file content types.</p></dd><dt><span class="term"><code class="literal">no-unlock</code></span></dt><dd><p>This Boolean option corresponds to <span class="command"><strong>svn
commit</strong></span>'s <code class="option">--no-unlock</code>
option, which tells Subversion not to release locks on
files you've just committed. If this runtime option
is set to <code class="literal">yes</code>, Subversion will
never release locks automatically, leaving you to run
<span class="command"><strong>svn unlock</strong></span> explicitly. It defaults
to <code class="literal">no</code>.</p></dd><dt><span class="term"><code class="literal">preserved-conflict-file-exts</code></span></dt><dd><p>The value of this option is a space-delimited list
of file extensions that Subversion should preserve
when generating conflict filenames. By default, the
list is empty. This option is new to Subversion
1.5.</p><p>When Subversion detects conflicting file content
changes, it defers resolution of those conflicts to the
user. To assist in the resolution, Subversion keeps
pristine copies of the various competing versions of
the file in the working copy. By default, those
conflict files have names constructed by appending to
the original filename a custom extension such as
<code class="filename">.mine</code> or
<code class="filename">.<em class="replaceable"><code>REV</code></em></code>
(where <em class="replaceable"><code>REV</code></em> is a revision
number). A mild annoyance with this naming scheme is
that on operating systems where a file's extension
determines the default application used to open and
edit that file, appending a custom extension prevents
the file from being easily opened by its native
application. For example, if the file
<code class="filename">ReleaseNotes.pdf</code> was conflicted,
the conflict files might be named
<code class="filename">ReleaseNotes.pdf.mine</code> or
<code class="filename">ReleaseNotes.pdf.r4231</code>. While
your system might be configured to use Adobe's Acrobat
Reader to open files whose extensions are
<code class="filename">.pdf</code>, there probably isn't an
application configured on your system to open all
files whose extensions are
<code class="filename">.r4231</code>.</p><p>You can fix this annoyance by using this
configuration option, though. For files with one of
the specified extensions, Subversion will append to
the conflict file names the custom extension just as
before, but then also reappend the file's original
extension. Using the previous example, and assuming
that <code class="literal">pdf</code> is one of the extensions
configured in this list thereof, the conflict files
generated for <code class="filename">ReleaseNotes.pdf</code>
would instead be named
<code class="filename">ReleaseNotes.pdf.mine.pdf</code> and
<code class="filename">ReleaseNotes.pdf.r4231.pdf</code>.
Because each file ends in
<code class="filename">.pdf</code>, the correct default
application will be used to view them.</p></dd><dt><span class="term"><code class="literal">use-commit-times</code></span></dt><dd><p>Normally your working copy files have timestamps
that reflect the last time they were touched by any
process, whether your own editor or some
<span class="command"><strong>svn</strong></span> subcommand. This is generally
convenient for people developing software, because
build systems often look at timestamps as a way of
deciding which files need to be recompiled.</p><p>In other situations, however, it's sometimes nice
for the working copy files to have timestamps that
reflect the last time they were changed in the
repository. The <span class="command"><strong>svn export</strong></span> command
always places these <span class="quote">“<span class="quote">last-commit
timestamps</span>”</span> on trees that it produces. By
setting this config variable to
<code class="literal">yes</code>, the <span class="command"><strong>svn
checkout</strong></span>, <span class="command"><strong>svn update</strong></span>,
<span class="command"><strong>svn switch</strong></span>, and <span class="command"><strong>svn
revert</strong></span> commands will also set last-commit
timestamps on files that they touch.</p></dd></dl></div><p>The <code class="literal">[auto-props]</code> section controls the
Subversion client's ability to automatically set properties
on files when they are added or imported. It contains any
number of key-value pairs in the
format <code class="literal"><em class="replaceable"><code>PATTERN</code></em>
= <em class="replaceable"><code>PROPNAME</code></em>=<em class="replaceable"><code>VALUE</code></em>[;<em class="replaceable"><code>PROPNAME</code></em>=<em class="replaceable"><code>VALUE</code></em>
...]</code>, where <em class="replaceable"><code>PATTERN</code></em> is
a file pattern that matches one or more filenames and the
rest of the line is a semicolon-delimited set of property
assignments. (If you need to use a semicolon in your
property's name or value, you can escape it by doubling
it.)</p><div class="informalexample"><pre class="screen">
$ cat ~/.subversion/config
…
[auto-props]
*.c = svn:eol-style=native
*.html = svn:eol-style=native;svn:mime-type=text/html;; charset=UTF8
*.sh = svn:eol-style=native;svn:executable
…
$ cd projects/myproject
$ svn status
? www/index.html
$ svn add www/index.html
A www/index.html
$ svn diff www/index.html
…
Property changes on: www/index.html
___________________________________________________________________
Added: svn:mime-type
## -0,0 +1 ##
+text/html; charset=UTF8
Added: svn:eol-style
## -0,0 +1 ##
+native
$
</pre></div><p>Multiple matches on a file will result in
multiple propsets for that file; however, there is no
guarantee that auto-props will be applied in the order in
which they are listed in the config file, so you can't have
one rule <span class="quote">“<span class="quote">override</span>”</span> another. You can find
several examples of auto-props usage in the
<code class="filename">config</code> file. Lastly, don't
forget to set <code class="literal">enable-auto-props</code> to
<code class="literal">yes</code> in the <code class="literal">miscellany</code>
section if you want to enable auto-props.</p></div></div></div><div class="sect1" title="Localization"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.advanced.l10n"></a>Localization</h2></div></div></div><p><em class="firstterm">Localization</em> is the act of making
programs behave in a region-specific way. When a program
formats numbers or dates in a way specific to your part of the
world or prints messages (or accepts input) in your native
language, the program is said to
be <em class="firstterm">localized</em>. This section describes
steps Subversion has made toward localization.</p><div class="sect2" title="Understanding Locales"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.l10n.understanding"></a>Understanding Locales</h3></div></div></div><p>Most modern operating systems have a notion of the
<span class="quote">“<span class="quote">current locale</span>”</span>—that is, the region or
country whose localization conventions are honored. These
conventions—typically chosen by some runtime
configuration mechanism on the computer—affect the way
in which programs present data to the user, as well as the way
in which they accept user input.</p><p>On most Unix-like systems, you can check the values of the
locale-related runtime configuration options by running the
<span class="command"><strong>locale</strong></span> command:</p><div class="informalexample"><pre class="screen">
$ locale
LANG=
LC_COLLATE="C"
LC_CTYPE="C"
LC_MESSAGES="C"
LC_MONETARY="C"
LC_NUMERIC="C"
LC_TIME="C"
LC_ALL="C"
$
</pre></div><p>The output is a list of locale-related environment
variables and their current values. In this example, the
variables are all set to the default <code class="literal">C</code>
locale, but users can set these variables to specific
country/language code combinations. For example, if one were
to set the <code class="literal">LC_TIME</code> variable to
<code class="literal">fr_CA</code>, programs would know to present
time and date information formatted according to a
French-speaking Canadian's expectations. And if one were to
set the <code class="literal">LC_MESSAGES</code> variable to
<code class="literal">zh_TW</code>, programs would know to present
human-readable messages in Traditional Chinese. Setting the
<code class="literal">LC_ALL</code> variable has the effect of changing
every locale variable to the same value. The value of
<code class="literal">LANG</code> is used as a default value for any
locale variable that is unset. To see the list of available
locales on a Unix system, run the command <strong class="userinput"><code>locale
-a</code></strong>.</p><p>On Windows, locale configuration is done via the
<span class="quote">“<span class="quote">Regional and Language Options</span>”</span> control panel
item. There you can view and select the values of individual
settings from the available locales, and even customize (at a
sickening level of detail) several of the display formatting
conventions.</p></div><div class="sect2" title="Subversion's Use of Locales"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.l10n.svnuse"></a>Subversion's Use of Locales</h3></div></div></div><p>The Subversion client, <span class="command"><strong>svn</strong></span>, honors the
current locale configuration in two ways. First, it notices
the value of the <code class="literal">LC_MESSAGES</code> variable and
attempts to print all messages in the specified language. For
example:</p><div class="informalexample"><pre class="screen">
$ export LC_MESSAGES=de_DE
$ svn help cat
cat: Gibt den Inhalt der angegebenen Dateien oder URLs aus.
Aufruf: cat ZIEL[@REV]...
…
</pre></div><p>This behavior works identically on both Unix and Windows
systems. Note, though, that while your operating system might
have support for a certain locale, the Subversion client still
may not be able to speak the particular language. In order to
produce localized messages, human volunteers must provide
translations for each language. The translations are written
using the GNU gettext package, which results in translation
modules that end with the <code class="filename">.mo</code> filename
extension. For example, the German translation file is named
<code class="filename">de.mo</code>. These translation files are
installed somewhere on your system. On Unix, they typically
live in <code class="filename">/usr/share/locale/</code>, while
on Windows they're often found in the
<code class="filename">share\locale\</code> folder in Subversion's
installation area. Once installed, a module is named after
the program for which it provides translations. For example, the
<code class="filename">de.mo</code> file may ultimately end up
installed as
<code class="filename">/usr/share/locale/de/LC_MESSAGES/subversion.mo</code>.
By browsing the installed <code class="filename">.mo</code> files, you
can see which languages the Subversion client is able to
speak.</p><p>The second way in which the locale is honored involves how
<span class="command"><strong>svn</strong></span> interprets your input. The repository
stores all paths, filenames, and log messages in Unicode,
encoded as UTF-8. In that sense, the repository is
<em class="firstterm">internationalized</em>—that is, the
repository is ready to accept input in any human language.
This means, however, that the Subversion client is responsible
for sending only UTF-8 filenames and log messages into the
repository. To do this, it must convert the data
from the native locale into UTF-8.</p><p>For example, suppose you create a file named
<code class="filename">caffè.txt</code>, and then when committing the
file, you write the log message as <span class="quote">“<span class="quote">Adesso il caffè è
più forte.</span>”</span> Both the filename and the log message contain
non-ASCII characters, but because your locale is set to
<code class="literal">it_IT</code>, the Subversion client knows to
interpret them as Italian. It uses an Italian character set
to convert the data to UTF-8 before sending it off to the
repository.</p><p>Note that while the repository demands UTF-8 filenames and
log messages, it <span class="emphasis"><em>does not</em></span> pay attention
to file contents. Subversion treats file contents as opaque
strings of bytes, and neither client nor server makes an
attempt to understand the character set or encoding of the
contents.</p><div class="sidebar" title="Character Set Conversion Errors"><div class="titlepage"><div><div><p class="title"><strong>Character Set Conversion Errors</strong></p></div></div></div><p>While using Subversion, you might get hit with an error
related to character set conversions:</p><div class="informalexample"><pre class="screen">
svn: E000022: Can't convert string from native encoding to 'UTF-8':
…
svn: E000022: Can't convert string from 'UTF-8' to native encoding:
…
</pre></div><p>Errors such as this typically occur when the Subversion
client has received a UTF-8 string from the repository, but
not all of the characters in that string can be represented
using the encoding of the current locale. For example, if
your locale is <code class="literal">en_US</code> but a collaborator
has committed a Japanese filename, you're likely to see this
error when you receive the file during an <span class="command"><strong>svn
update</strong></span>.</p><p>The solution is either to set your locale to something
that <span class="emphasis"><em>can</em></span> represent the incoming UTF-8
data, or to change the filename or log message in the
repository. (And don't forget to slap your collaborator's
hand—projects should decide on common languages ahead of
time so that all participants are using the same
locale.)</p></div></div></div><div class="sect1" title="Using External Editors"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.advanced.externaleditors"></a>Using External Editors</h2></div></div></div><p>The most obvious way to get data into Subversion is through
the addition of files to version control, committing changes to
those files, and so on. But other pieces of
information besides merely versioned file data live in your
Subversion repository. Some of these bits of
information—commit log messages, lock comments, and some
property values—tend to be textual in nature and are
provided explicitly by users. Most of this information can be
provided to the Subversion command-line client using the
<code class="option">--message</code> (<code class="option">-m</code>) and
<code class="option">--file</code> (<code class="option">-F</code>) options with the
appropriate subcommands.</p><p>Each of these options has its pros and cons. For example,
when performing a commit, <code class="option">--file</code>
(<code class="option">-F</code>) works well if you've already prepared a
text file that holds your commit log message. If you didn't,
though, you can use <code class="option">--message</code>
(<code class="option">-m</code>) to provide a log message on the command
line. Unfortunately, it can be tricky to compose anything more
than a simple one-line message on the command line. Users want
more flexibility—multiline, free-form log message editing
on demand.</p><p>Subversion supports this by allowing you to specify an
external text editor that it will launch as necessary
to give you a more powerful input mechanism for this textual
metadata. There are several ways to tell Subversion which
editor you'd like use. Subversion checks the following things,
in the order specified, when it wants to launch such an
editor:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p><code class="literal">--editor-cmd</code> command-line option</p></li><li class="listitem"><p><code class="literal">SVN_EDITOR</code> environment variable</p></li><li class="listitem"><p><code class="literal">editor-cmd</code> runtime configuration option</p></li><li class="listitem"><p><code class="literal">VISUAL</code> environment variable</p></li><li class="listitem"><p><code class="literal">EDITOR</code> environment variable</p></li><li class="listitem"><p>Possibly, a fallback value built into the Subversion
libraries (not present in the official builds)</p></li></ol></div><p>The value of any of these options or variables is the
beginning of a command line to be executed by the shell.
Subversion appends to that command line a space and the pathname
of a temporary file to be edited. So, to be used with
Subversion, the configured or specified editor needs to support
an invocation in which its last command-line parameter is a file
to be edited, and it should be able to save the file in place
and return a zero exit code to indicate success.</p><p>As noted, external editors can be used to provide commit log
messages to any of the committing subcommands (such as
<span class="command"><strong>svn commit</strong></span> or <span class="command"><strong>import</strong></span>,
<span class="command"><strong>svn mkdir</strong></span> or <span class="command"><strong>delete</strong></span> when
provided a URL target, etc.), and Subversion will try to
launch the editor automatically if you don't specify either of
the <code class="option">--message</code> (<code class="option">-m</code>) or
<code class="option">--file</code> (<code class="option">-F</code>) options. The
<span class="command"><strong>svn propedit</strong></span> command is built almost entirely
around the use of an external editor. And beginning in version
1.5, Subversion will also use the configured external text
editor when the user asks it to launch an editor during
interactive conflict resolution. Oddly, there doesn't appear to
be a way to use external editors to interactively provide lock
comments.</p></div><div class="sect1" title="Using External Differencing and Merge Tools"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.advanced.externaldifftools"></a>Using External Differencing and Merge Tools</h2></div></div></div><p>The interface between Subversion and external two- and three-way
differencing tools harkens back to a time when Subversion's only
contextual differencing capabilities were built around
invocations of the GNU diffutils toolchain, specifically the
<span class="command"><strong>diff</strong></span> and <span class="command"><strong>diff3</strong></span> utilities.
To get the kind of behavior Subversion needed, it called these
utilities with more than a handful of options and parameters,
most of which were quite specific to the utilities. Some time
later, Subversion grew its own internal differencing library,
and as a failover mechanism, the <code class="option">--diff-cmd</code> and
<code class="option">--diff3-cmd</code> options were added to the
Subversion command-line client so that users could more easily
indicate that they preferred to use the GNU diff and diff3
utilities instead of the newfangled internal diff library. If
those options were used, Subversion would simply ignore the
internal diff library, and fall back to running those external
programs, lengthy argument lists and all. And that's where
things remain today.</p><p>It didn't take long for folks to realize that having such
easy configuration mechanisms for specifying that Subversion
should use the external GNU diff and diff3 utilities located at
a particular place on the system could be applied toward the use
of other differencing tools, too. After all, Subversion didn't
actually verify that the things it was being told to run were
members of the GNU diffutils toolchain. But the only
configurable aspect of using those external tools is their
location on the system—not the option set, parameter
order, and so on. Subversion continues to throw all those GNU utility
options at your external diff tool regardless of whether
that program can understand those options. And that's where
things get unintuitive for most users.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>The decision on when to fire off a contextual two- or three-way
diff as part of a larger Subversion operation is made entirely
by Subversion and is affected by, among other things, whether
the files being operated on are human-readable as
determined by their <code class="literal">svn:mime-type</code> property.
This means, for example, that even if you had the niftiest
Microsoft Word-aware differencing or merging tool in the
universe, it would never be invoked by Subversion as long as
your versioned Word documents had a configured MIME type that
denoted that they were not human-readable (such as
<code class="literal">application/msword</code>). For more about MIME
type settings, see <a class="xref" href="#svn.advanced.props.special.mime-type" title="File Content Type">the section called “File Content Type”</a></p></td></tr></table></div><p>Much later, Subversion 1.5 introduced interactive resolution
of conflicts (described in
<a class="xref" href="#svn.tour.cycle.resolve" title="Resolve Any Conflicts">the section called “Resolve Any Conflicts”</a>). One of the options
that this feature provides to users is the ability to
interactively launch a third-party merge tool. If this action
is taken, Subversion will check to see if the user has specified
such a tool for use in this way. Subversion will first check
the <code class="literal">SVN_MERGE</code> environment variable for the
name of an external merge tool. If that variable is not set, it
will look for the same information in the value of
the <code class="literal">merge-tool-cmd</code> runtime configuration
option. Upon finding a configured external merge tool, it will
invoke that tool.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>While the general purposes of the three-way differencing
and merge tools are roughly the same (finding a way to make
separate-but-overlapping file changes live in harmony),
Subversion exercises each of these options at different times
and for different reasons. The internal three-way
differencing engine and its optional external replacement are
used when interaction with the user
is <span class="emphasis"><em>not</em></span> expected. In fact, significant
delay introduced by such a tool can actually result in the
failure of some time-sensitive Subversion operations. It's
the external merge tool that is intended to be invoked
interactively.</p></td></tr></table></div><p>Now, while the interface between Subversion and an external
merge tool is significantly less convoluted than that between
Subversion and the diff and diff3 tools, the likelihood of
finding such a tool whose calling conventions exactly match what
Subversion expects is still quite low. The key to using
external differencing and merge tools with Subversion is to use
wrapper scripts, which convert the input from Subversion into
something that your specific differencing tool can understand,
and then convert the output of your tool back into a format that
Subversion expects. The following sections cover the specifics
of those expectations.</p><div class="sect2" title="External diff"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.externaldifftools.diff"></a>External diff</h3></div></div></div><p>Subversion calls external diff programs with parameters
suitable for the GNU diff utility, and expects only that the
external program will return with a successful error code per
the GNU diff definition thereof. For most alternative diff
programs, only the sixth and seventh arguments—the paths
of the files that represent the left and right sides of the
diff, respectively—are of interest. Note that
Subversion runs the diff program once per modified file
covered by the Subversion operation, so if your program runs
in an asynchronous fashion (or is <span class="quote">“<span class="quote">backgrounded</span>”</span>),
you might have several instances of it all running
simultaneously. Finally, Subversion expects that your program
return an error code of 1 if your program detected
differences, or 0 if it did not—any other error code is
considered a fatal error.<sup>[<a id="idp15535360" href="#ftn.idp15535360" class="footnote">65</a>]</sup></p><p><a class="xref" href="#svn.advanced.externaldifftools.diff.ex-1" title="Example 7.2. diffwrap.py">Example 7.2, “diffwrap.py”</a>
and <a class="xref" href="#svn.advanced.externaldifftools.diff.ex-2" title="Example 7.3. diffwrap.bat">Example 7.3, “diffwrap.bat”</a>
are templates for external diff tool wrappers in the Python
and Windows batch scripting languages, respectively.</p><div class="example"><a id="svn.advanced.externaldifftools.diff.ex-1"></a><p class="title"><strong>Example 7.2. diffwrap.py</strong></p><div class="example-contents"><pre class="programlisting">
#!/usr/bin/env python
import sys
import os
# Configure your favorite diff program here.
DIFF = "/usr/local/bin/my-diff-tool"
# Subversion provides the paths we need as the last two parameters.
LEFT = sys.argv[-2]
RIGHT = sys.argv[-1]
# Call the diff command (change the following line to make sense for
# your diff program).
cmd = [DIFF, '--left', LEFT, '--right', RIGHT]
os.execv(cmd[0], cmd)
# Return an errorcode of 0 if no differences were detected, 1 if some were.
# Any other errorcode will be treated as fatal.
</pre></div></div><br class="example-break"></br><div class="example"><a id="svn.advanced.externaldifftools.diff.ex-2"></a><p class="title"><strong>Example 7.3. diffwrap.bat</strong></p><div class="example-contents"><pre class="programlisting">
@ECHO OFF
REM Configure your favorite diff program here.
SET DIFF="C:\Program Files\Funky Stuff\My Diff Tool.exe"
REM Subversion provides the paths we need as the last two parameters.
REM These are parameters 6 and 7 (unless you use svn diff -x, in
REM which case, all bets are off).
SET LEFT=%6
SET RIGHT=%7
REM Call the diff command (change the following line to make sense for
REM your diff program).
%DIFF% --left %LEFT% --right %RIGHT%
REM Return an errorcode of 0 if no differences were detected, 1 if some were.
REM Any other errorcode will be treated as fatal.
</pre></div></div><br class="example-break"></br></div><div class="sect2" title="External diff3"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.externaldifftools.diff3"></a>External diff3</h3></div></div></div><p>Subversion invokes three-way differencing programs to
perform non-interactive merges. When configured to use an
external three-way differencing program, it executes that
program with parameters suitable for the GNU diff3 utility,
expecting that the external program will return with a
successful error code and that the full file contents that
result from the completed merge operation are printed on the
standard output stream (so that Subversion can redirect them
into the appropriate version-controlled file). For most
alternative merge programs, only the ninth, tenth, and
eleventh arguments, the paths of the files which represent
the <span class="quote">“<span class="quote">mine</span>”</span>, <span class="quote">“<span class="quote">older</span>”</span>,
and <span class="quote">“<span class="quote">yours</span>”</span> inputs, respectively, are of
interest. Note that because Subversion depends on the output
of your merge program, your wrapper script must not exit
before that output has been delivered to Subversion. When it
finally does exit, it should return an error code of 0 if the
merge was successful, or 1 if unresolved conflicts remain in
the output—any other error code is considered a fatal
error.</p><p><a class="xref" href="#svn.advanced.externaldifftools.diff3.ex-1" title="Example 7.4. diff3wrap.py">Example 7.4, “diff3wrap.py”</a>
and <a class="xref" href="#svn.advanced.externaldifftools.diff3.ex-2" title="Example 7.5. diff3wrap.bat">Example 7.5, “diff3wrap.bat”</a> are
templates for external three-way differencing tool wrappers in
the Python and Windows batch scripting languages,
respectively.</p><div class="example"><a id="svn.advanced.externaldifftools.diff3.ex-1"></a><p class="title"><strong>Example 7.4. diff3wrap.py</strong></p><div class="example-contents"><pre class="programlisting">
#!/usr/bin/env python
import sys
import os
# Configure your favorite three-way diff program here.
DIFF3 = "/usr/local/bin/my-diff3-tool"
# Subversion provides the paths we need as the last three parameters.
MINE = sys.argv[-3]
OLDER = sys.argv[-2]
YOURS = sys.argv[-1]
# Call the three-way diff command (change the following line to make
# sense for your three-way diff program).
cmd = [DIFF3, '--older', OLDER, '--mine', MINE, '--yours', YOURS]
os.execv(cmd[0], cmd)
# After performing the merge, this script needs to print the contents
# of the merged file to stdout. Do that in whatever way you see fit.
# Return an errorcode of 0 on successful merge, 1 if unresolved conflicts
# remain in the result. Any other errorcode will be treated as fatal.
</pre></div></div><br class="example-break"></br><div class="example"><a id="svn.advanced.externaldifftools.diff3.ex-2"></a><p class="title"><strong>Example 7.5. diff3wrap.bat</strong></p><div class="example-contents"><pre class="programlisting">
@ECHO OFF
REM Configure your favorite three-way diff program here.
SET DIFF3="C:\Program Files\Funky Stuff\My Diff3 Tool.exe"
REM Subversion provides the paths we need as the last three parameters.
REM These are parameters 9, 10, and 11. But we have access to only
REM nine parameters at a time, so we shift our nine-parameter window
REM twice to let us get to what we need.
SHIFT
SHIFT
SET MINE=%7
SET OLDER=%8
SET YOURS=%9
REM Call the three-way diff command (change the following line to make
REM sense for your three-way diff program).
%DIFF3% --older %OLDER% --mine %MINE% --yours %YOURS%
REM After performing the merge, this script needs to print the contents
REM of the merged file to stdout. Do that in whatever way you see fit.
REM Return an errorcode of 0 on successful merge, 1 if unresolved conflicts
REM remain in the result. Any other errorcode will be treated as fatal.
</pre></div></div><br class="example-break"></br></div><div class="sect2" title="External merge"><div class="titlepage"><div><div><h3 class="title"><a id="svn.advanced.externaldifftools.merge"></a>External merge</h3></div></div></div><p>Subversion optionally invokes an external merge tool as
part of its support for interactive conflict resolution. It
provides as arguments to the merge tool the following: the
path of the unmodified base file, the path of
the <span class="quote">“<span class="quote">theirs</span>”</span> file (which contains upstream
changes), the path of the <span class="quote">“<span class="quote">mine</span>”</span> file (which
contains local modifications), the path of the file into which
the final resolved contents should be stored by the merge
tool, and the working copy path of the conflicted file
(relative to the original target of the merge operation). The
merge tool is expected to return an error code of 0 to
indicate success, or 1 to indicate failure.</p><p><a class="xref" href="#svn.advanced.externaldifftools.merge.ex-1" title="Example 7.6. mergewrap.py">Example 7.6, “mergewrap.py”</a>
and <a class="xref" href="#svn.advanced.externaldifftools.merge.ex-2" title="Example 7.7. mergewrap.bat">Example 7.7, “mergewrap.bat”</a> are
templates for external merge tool wrappers in the Python
and Windows batch scripting languages, respectively.</p><div class="example"><a id="svn.advanced.externaldifftools.merge.ex-1"></a><p class="title"><strong>Example 7.6. mergewrap.py</strong></p><div class="example-contents"><pre class="programlisting">
#!/usr/bin/env python
import sys
import os
# Configure your favorite merge program here.
MERGE = "/usr/local/bin/my-merge-tool"
# Get the paths provided by Subversion.
BASE = sys.argv[1]
THEIRS = sys.argv[2]
MINE = sys.argv[3]
MERGED = sys.argv[4]
WCPATH = sys.argv[5]
# Call the merge command (change the following line to make sense for
# your merge program).
cmd = [MERGE, '--base', BASE, '--mine', MINE, '--theirs', THEIRS,
'--outfile', MERGED]
os.execv(cmd[0], cmd)
# Return an errorcode of 0 if the conflict was resolved; 1 otherwise.
# Any other errorcode will be treated as fatal.
</pre></div></div><br class="example-break"></br><div class="example"><a id="svn.advanced.externaldifftools.merge.ex-2"></a><p class="title"><strong>Example 7.7. mergewrap.bat</strong></p><div class="example-contents"><pre class="programlisting">
@ECHO OFF
REM Configure your favorite merge program here.
SET MERGE="C:\Program Files\Funky Stuff\My Merge Tool.exe"
REM Get the paths provided by Subversion.
SET BASE=%1
SET THEIRS=%2
SET MINE=%3
SET MERGED=%4
SET WCPATH=%5
REM Call the merge command (change the following line to make sense for
REM your merge program).
%MERGE% --base %BASE% --mine %MINE% --theirs %THEIRS% --outfile %MERGED%
REM Return an errorcode of 0 if the conflict was resolved; 1 otherwise.
REM Any other errorcode will be treated as fatal.
</pre></div></div><br class="example-break"></br></div></div><div class="sect1" title="Summary"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.customization.summary"></a>Summary</h2></div></div></div><p>Sometimes there's a single right way to do things; sometimes
there are many. Subversion's developers understand that while
the majority of its exact behaviors are acceptable to most of
its users, there are some corners of its functionality where
such a universally pleasing approach doesn't exist. In those
places, Subversion offers users the opportunity to tell it how
<span class="emphasis"><em>they</em></span> want it to behave.</p><p>In this chapter, we explored Subversion's runtime
configuration system and other mechanisms by which users can
control those configurable behaviors. If you are a developer,
though, the next chapter will take you one step further. It
describes how you can further customize your Subversion
experience by writing your own software against Subversion's
libraries.</p></div><div class="footnotes"><br></br><hr width="100" align="left"></hr><div class="footnote"><p><sup>[<a id="ftn.idp15216960" href="#idp15216960" class="para">63</a>] </sup>The <code class="literal">APPDATA</code>
environment variable points to the <code class="filename">Application
Data</code> area, so you can always refer to this folder
as <code class="filename">%APPDATA%\Subversion</code>.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp15396240" href="#idp15396240" class="para">64</a>] </sup>Anyone for potluck
dinner?</p></div><div class="footnote"><p><sup>[<a id="ftn.idp15535360" href="#idp15535360" class="para">65</a>] </sup>The GNU diff manual
page puts it this way: <span class="quote">“<span class="quote">An exit status of 0 means no
differences were found, 1 means some differences were found,
and 2 means trouble.</span>”</span></p></div></div></div><div class="chapter" title="Chapter 8. Embedding Subversion"><div class="titlepage"><div><div><h2 class="title"><a id="svn.developer"></a>Chapter 8. Embedding Subversion</h2></div></div></div><div class="toc"><p><strong>Table of Contents</strong></p><dl><dt><span class="sect1"><a href="#svn.developer.layerlib">Layered Library Design</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.developer.layerlib.repos">Repository Layer</a></span></dt><dt><span class="sect2"><a href="#svn.developer.layerlib.ra">Repository Access Layer</a></span></dt><dt><span class="sect2"><a href="#svn.developer.layerlib.client">Client Layer</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.developer.usingapi">Using the APIs</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.developer.usingapi.apr">The Apache Portable Runtime Library</a></span></dt><dt><span class="sect2"><a href="#svn.developer.usingapi.funcsbatons">Functions and Batons</a></span></dt><dt><span class="sect2"><a href="#svn.developer.usingapi.urlpath">URL and Path Requirements</a></span></dt><dt><span class="sect2"><a href="#svn.developer.usingapi.otherlangs">Using Languages Other Than C and C++</a></span></dt><dt><span class="sect2"><a href="#svn.developer.usingapi.codesamples">Code Samples</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.developer.summary">Summary</a></span></dt></dl></div><p>Subversion has a modular design: it's implemented as a
collection of libraries written in C. Each library has a
well-defined purpose and application programming interface (API),
and that interface is available not only for Subversion itself to
use, but for any software that wishes to embed or otherwise
programmatically control Subversion. Additionally, Subversion's
API is available not only to other C programs, but also to
programs written in higher-level languages such as Python, Perl,
Java, and Ruby.</p><p>This chapter is for those who wish to interact with Subversion
through its public API or its various language bindings. If you
wish to write robust wrapper scripts around Subversion
functionality to simplify your own life, are trying to develop
more complex integrations between Subversion and other pieces of
software, or just have an interest in Subversion's various library
modules and what they offer, this chapter is for you. If,
however, you don't foresee yourself participating with Subversion
at such a level, feel free to skip this chapter with the
confidence that your experience as a Subversion user will not be
affected.</p><div class="sect1" title="Layered Library Design"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.developer.layerlib"></a>Layered Library Design</h2></div></div></div><p>Each of Subversion's core libraries can be said to exist in
one of three main layers—the Repository layer, the
Repository Access (RA) layer, or the Client layer (see
<a class="xref" href="#svn.intro.architecture.dia-1" title="Figure 1. Subversion's architecture">Figure 1, “Subversion's architecture”</a> in the Preface).
We will examine these layers shortly, but first, let's briefly
summarize Subversion's various libraries. For the sake of
consistency, we will refer to the libraries by their
extensionless Unix library names
(<code class="filename">libsvn_fs</code>, <code class="filename">libsvn_wc</code>,
<code class="filename">mod_dav_svn</code>, etc.).</p><div class="variablelist"><dl><dt><span class="term">libsvn_client</span></dt><dd><p>Primary interface for client
programs</p></dd><dt><span class="term">libsvn_delta</span></dt><dd><p>Tree and byte-stream differencing
routines</p></dd><dt><span class="term">libsvn_diff</span></dt><dd><p>Contextual differencing and merging
routines</p></dd><dt><span class="term">libsvn_fs</span></dt><dd><p>Filesystem commons and module
loader</p></dd><dt><span class="term">libsvn_fs_base</span></dt><dd><p>The Berkeley DB filesystem
backend</p></dd><dt><span class="term">libsvn_fs_fs</span></dt><dd><p>The native filesystem (FSFS)
backend</p></dd><dt><span class="term">libsvn_ra</span></dt><dd><p>Repository Access commons and module
loader</p></dd><dt><span class="term">libsvn_ra_local</span></dt><dd><p>The local Repository Access
module</p></dd><dt><span class="term">libsvn_ra_neon</span></dt><dd><p>The WebDAV Repository Access
module</p></dd><dt><span class="term">libsvn_ra_serf</span></dt><dd><p>Another (experimental) WebDAV Repository
Access module</p></dd><dt><span class="term">libsvn_ra_svn</span></dt><dd><p>The custom protocol Repository Access
module</p></dd><dt><span class="term">libsvn_repos</span></dt><dd><p>Repository interface</p></dd><dt><span class="term">libsvn_subr</span></dt><dd><p>Miscellaneous helpful
subroutines</p></dd><dt><span class="term">libsvn_wc</span></dt><dd><p>The working copy management
library</p></dd><dt><span class="term">mod_authz_svn</span></dt><dd><p>Apache authorization module for Subversion
repositories access via WebDAV</p></dd><dt><span class="term">mod_dav_svn</span></dt><dd><p>Apache module for mapping WebDAV operations to
Subversion ones</p></dd></dl></div><p>The fact that the word <span class="quote">“<span class="quote">miscellaneous</span>”</span>
appears only once in the previous list is a good sign. The
Subversion development team is serious about making sure that
functionality lives in the right layer and libraries. Perhaps
the greatest advantage of the modular design is its lack of
complexity from a developer's point of view. As a developer,
you can quickly formulate that kind of <span class="quote">“<span class="quote">big
picture</span>”</span> that allows you to pinpoint the location of
certain pieces of functionality with relative ease.</p><p>Another benefit of modularity is the ability to replace a
given module with a whole new library that implements the same
API without affecting the rest of the code base. In some sense,
this happens within Subversion already. The
<code class="filename">libsvn_ra_local</code>,
<code class="filename">libsvn_ra_neon</code>,
<code class="filename">libsvn_ra_serf</code>, and
<code class="filename">libsvn_ra_svn</code> libraries each implement the
same interface, all working as plug-ins to
<code class="filename">libsvn_ra</code>. And all four communicate with
the Repository layer—<code class="filename">libsvn_ra_local</code> connects to the
repository directly; the other three do so over a network. The
<code class="filename">libsvn_fs_base</code> and
<code class="filename">libsvn_fs_fs</code> libraries are another pair of
libraries that implement the same functionality in different
ways—both are plug-ins to the common
<code class="filename">libsvn_fs</code> library.</p><p>The client itself also highlights the benefits of modularity
in the Subversion design. Subversion's
<code class="filename">libsvn_client</code> library is a one-stop shop
for most of the functionality necessary for designing a working
Subversion client (see <a class="xref" href="#svn.developer.layerlib.client" title="Client Layer">the section called “Client Layer”</a>). So while the
Subversion distribution provides only the <span class="command"><strong>svn</strong></span>
command-line client program, several third-party
programs provide various forms of graphical client UIs.
These GUIs use the same APIs that the stock command-line client
does. This type of modularity has played a large role in the
proliferation of available Subversion clients and IDE
integrations and, by extension, to the tremendous adoption rate
of Subversion itself.</p><div class="sect2" title="Repository Layer"><div class="titlepage"><div><div><h3 class="title"><a id="svn.developer.layerlib.repos"></a>Repository Layer</h3></div></div></div><p>When referring to Subversion's Repository layer, we're
generally talking about two basic concepts—the versioned
filesystem implementation (accessed via
<code class="filename">libsvn_fs</code>, and supported by its
<code class="filename">libsvn_fs_base</code> and
<code class="filename">libsvn_fs_fs</code> plug-ins), and the repository
logic that wraps it (as implemented in
<code class="filename">libsvn_repos</code>). These libraries provide
the storage and reporting mechanisms for the various revisions
of your version-controlled data. This layer is connected to
the Client layer via the Repository Access layer, and is, from
the perspective of the Subversion user, the stuff at the
<span class="quote">“<span class="quote">other end of the line.</span>”</span></p><p>The Subversion filesystem is not a kernel-level filesystem
that one would install in an operating system (such as the
Linux ext2 or NTFS), but instead is a virtual filesystem.
Rather than storing <span class="quote">“<span class="quote">files</span>”</span> and
<span class="quote">“<span class="quote">directories</span>”</span> as real files and directories (the
kind you can navigate through using your favorite shell
program), it uses one of two available abstract storage
backends—either a Berkeley DB database environment or a
flat-file representation. (To learn more about the two
repository backends, see <a class="xref" href="#svn.reposadmin.basics.backends" title="Choosing a Data Store">the section called “Choosing a Data Store”</a>.) There has even
been considerable interest by the development community in
giving future releases of Subversion the ability to use other
backend database systems, perhaps through a mechanism such as
Open Database Connectivity (ODBC). In fact, Google did
something similar to this before launching the Google Code
Project Hosting service: they announced in mid-2006 that
members of its open source team had written a new proprietary
Subversion filesystem plug-in that used Google's ultra-scalable
Bigtable database for its storage.</p><p>The filesystem API exported by
<code class="filename">libsvn_fs</code> contains the kinds of
functionality you would expect from any other filesystem
API—you can create and remove files and directories,
copy and move them around, modify file contents, and so on.
It also has features that are not quite as common, such as the
ability to add, modify, and remove metadata
(<span class="quote">“<span class="quote">properties</span>”</span>) on each file or directory.
Furthermore, the Subversion filesystem is a versioning
filesystem, which means that as you make changes to your
directory tree, Subversion remembers what your tree looked
like before those changes. And before the previous changes.
And the previous ones. And so on, all the way back through
versioning time to (and just beyond) the moment you first
started adding things to the filesystem.</p><p>All the modifications you make to your tree are done
within the context of a Subversion commit transaction. The
following is a simplified general routine for modifying your
filesystem:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p>Begin a Subversion commit transaction.</p></li><li class="listitem"><p>Make your changes (adds, deletes, property
modifications, etc.).</p></li><li class="listitem"><p>Commit your transaction.</p></li></ol></div><p>Once you have committed your transaction, your filesystem
modifications are permanently stored as historical artifacts.
Each of these cycles generates a single new revision of your
tree, and each revision is forever accessible as an immutable
snapshot of <span class="quote">“<span class="quote">the way things were.</span>”</span></p><div class="sidebar" title="The Transaction Distraction"><div class="titlepage"><div><div><p class="title"><strong>The Transaction Distraction</strong></p></div></div></div><p>The notion of a Subversion transaction can become easily
confused with the transaction support provided by the
underlying database itself, especially given the former's
close proximity to the Berkeley DB database code in
<code class="filename">libsvn_fs_base</code>. Both types of
transaction exist to provide atomicity and isolation. In
other words, transactions give you the ability to perform a
set of actions in an all-or-nothing fashion—either all
the actions in the set complete with success, or they all
get treated as though <span class="emphasis"><em>none</em></span> of them ever
happened—and in a way that does not interfere with
other processes acting on the data.</p><p>Database transactions generally encompass small
operations related specifically to the modification of data
in the database itself (such as changing the contents of a
table row). Subversion transactions are larger in scope,
encompassing higher-level operations such as making
modifications to a set of files and directories that are
intended to be stored as the next revision of the filesystem
tree. If that isn't confusing enough, consider the fact
that Subversion uses a database transaction during the
creation of a Subversion transaction (so that if the
creation of a Subversion transaction fails, the database will
look as though we had never attempted that creation in the first
place)!</p><p>Fortunately for users of the filesystem API, the
transaction support provided by the database system itself
is hidden almost entirely from view (as should be expected
from a properly modularized library scheme). It is only
when you start digging into the implementation of the
filesystem itself that such things become visible (or
interesting).</p></div><p>Most of the functionality the filesystem
interface provides deals with actions that occur on individual
filesystem paths. That is, from outside the filesystem, the
primary mechanism for describing and accessing the individual
revisions of files and directories comes through the use of
path strings such as <code class="filename">/foo/bar</code>, just as though
you were addressing files and directories through your
favorite shell program. You add new files and directories by
passing their paths-to-be to the right API functions. You
query for information about them by the same mechanism.</p><p>Unlike most filesystems, though, a path alone is not
enough information to identify a file or directory in
Subversion. Think of a directory tree as a two-dimensional
system, where a node's siblings represent a sort of
left-and-right motion, and navigating into the node's
subdirectories represents a downward motion. <a class="xref" href="#svn.developer.layerlib.repos.dia-1" title="Figure 8.1. Files and directories in two dimensions">Figure 8.1, “Files and directories in two dimensions”</a> shows a typical
representation of a tree as exactly that.</p><div class="figure"><a id="svn.developer.layerlib.repos.dia-1"></a><p class="title"><strong>Figure 8.1. Files and directories in two dimensions</strong></p><div class="figure-contents"><div><img src="images/ch08dia1.png" alt="Files and directories in two dimensions"></img></div></div></div><br class="figure-break"></br><p>The difference here is that the Subversion filesystem has
a nifty third dimension that most filesystems do not
have—Time!<sup>[<a id="idp16034992" href="#ftn.idp16034992" class="footnote">66</a>]</sup> In the filesystem
interface, nearly every function that has a
<em class="parameter"><code>path</code></em> argument also expects a
<em class="parameter"><code>root</code></em> argument. This
<code class="literal">svn_fs_root_t</code> argument describes
either a revision or a Subversion transaction (which is simply
a revision in the making) and provides that third dimension
of context needed to understand the difference between
<code class="filename">/foo/bar</code> in revision 32, and the same
path as it exists in revision 98. <a class="xref" href="#svn.developer.layerlib.repos.dia-2" title="Figure 8.2. Versioning time—the third dimension!">Figure 8.2, “Versioning time—the third dimension!”</a> shows revision
history as an added dimension to the Subversion filesystem
universe.</p><div class="figure"><a id="svn.developer.layerlib.repos.dia-2"></a><p class="title"><strong>Figure 8.2. Versioning time—the third dimension!</strong></p><div class="figure-contents"><div><img src="images/ch08dia2.png" alt="Versioning time—the third dimension!"></img></div></div></div><br class="figure-break"></br><p>As we mentioned earlier, the
<code class="filename">libsvn_fs</code> API looks and feels like any
other filesystem, except that it has this wonderful versioning
capability. It was designed to be usable by any program
interested in a versioning filesystem. Not coincidentally,
Subversion itself is interested in that functionality. But
while the filesystem API should be sufficient for basic file
and directory versioning support, Subversion wants
more—and that is where <code class="filename">libsvn_repos</code>
comes in.</p><p>The Subversion repository library
(<code class="filename">libsvn_repos</code>) sits (logically speaking)
atop the <code class="filename">libsvn_fs</code> API, providing
additional functionality beyond that of the underlying
versioned filesystem logic. It does not completely wrap each
and every filesystem function—only certain major steps
in the general cycle of filesystem activity are wrapped by the
repository interface. Some of these include the creation and
commit of Subversion transactions and the modification of
revision properties. These particular events are wrapped by
the repository layer because they have hooks associated with
them. A repository hook system is not strictly related to
implementing a versioning filesystem, so it lives in the
repository wrapper library.</p><p>The hooks mechanism is but one of the reasons for the
abstraction of a separate repository library from the rest of
the filesystem code. The <code class="filename">libsvn_repos</code>
API provides several other important utilities to Subversion.
These include the abilities to:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>Create, open, destroy, and perform recovery steps on a
Subversion repository and the filesystem included in that
repository.</p></li><li class="listitem"><p>Describe the differences between two filesystem
trees.</p></li><li class="listitem"><p>Query for the commit log messages associated with all
(or some) of the revisions in which a set of files was
modified in the filesystem.</p></li><li class="listitem"><p>Generate a human-readable <span class="quote">“<span class="quote">dump</span>”</span> of the
filesystem—a complete representation of the revisions in
the filesystem.</p></li><li class="listitem"><p>Parse that dump format, loading the dumped revisions
into a different Subversion repository.</p></li></ul></div><p>As Subversion continues to evolve, the repository library
will grow with the filesystem library to offer increased
functionality and configurable option support.</p></div><div class="sect2" title="Repository Access Layer"><div class="titlepage"><div><div><h3 class="title"><a id="svn.developer.layerlib.ra"></a>Repository Access Layer</h3></div></div></div><p>If the Subversion Repository layer is at <span class="quote">“<span class="quote">the other
end of the line,</span>”</span> the Repository Access (RA) layer is
the line itself. Charged with marshaling data between the
client libraries and the repository, this layer includes the
<code class="filename">libsvn_ra</code> module loader library, the RA
modules themselves (which currently includes
<code class="filename">libsvn_ra_neon</code>,
<code class="filename">libsvn_ra_local</code>,
<code class="filename">libsvn_ra_serf</code>, and
<code class="filename">libsvn_ra_svn</code>), and any additional
libraries needed by one or more of those RA modules (such as
the <code class="filename">mod_dav_svn</code> Apache module or
<code class="filename">libsvn_ra_svn</code>'s server,
<span class="command"><strong>svnserve</strong></span>).</p><p>Since Subversion uses URLs to identify its repository
resources, the protocol portion of the URL scheme (usually
<code class="literal">file://</code>, <code class="literal">http://</code>,
<code class="literal">https://</code>, <code class="literal">svn://</code>, or
<code class="literal">svn+ssh://</code>) is used to determine which RA
module will handle the communications. Each module registers
a list of the protocols it knows how to <span class="quote">“<span class="quote">speak</span>”</span>
so that the RA loader can, at runtime, determine which module
to use for the task at hand. You can determine which RA
modules are available to the Subversion command-line client,
and what protocols they claim to support, by running
<strong class="userinput"><code>svn --version</code></strong>:</p><div class="informalexample"><pre class="screen">
$ svn --version
svn, version 1.7.0
compiled Nov 15 2011, 10:10:24
Copyright (C) 2011 The Apache Software Foundation.
This software consists of contributions made by many people; see the NOTICE
file for more information.
Subversion is open source software, see http://subversion.apache.org/
The following repository access (RA) modules are available:
* ra_neon : Module for accessing a repository via WebDAV protocol using Neon.
- handles 'http' scheme
- handles 'https' scheme
* ra_svn : Module for accessing a repository using the svn network protocol.
- with Cyrus SASL authentication
- handles 'svn' scheme
* ra_local : Module for accessing a repository on local disk.
- handles 'file' scheme
* ra_serf : Module for accessing a repository via WebDAV protocol using serf.
- handles 'http' scheme
- handles 'https' scheme
$
</pre></div><p>The public API exported by the RA layer contains
functionality necessary for sending and receiving versioned
data to and from the repository. And each of the available RA
plug-ins is able to perform that task using a specific
protocol—<code class="filename">libsvn_ra_neon</code>
and <code class="filename">libsvn_ra_serf</code> speak HTTP/WebDAV
(optionally using SSL encryption) with an Apache HTTP Server
that is running the <code class="filename">mod_dav_svn</code>
Subversion server module; <code class="filename">libsvn_ra_svn</code>
speaks a custom network protocol with the
<span class="command"><strong>svnserve</strong></span> program; and so on.</p><p>For those who wish to access a Subversion repository
using still another protocol, that is precisely why the
Repository Access layer is modularized! Developers can simply
write a new library that implements the RA interface on one
side and communicates with the repository on the other. Your
new library can use existing network protocols or you can
invent your own. You could use interprocess communication
(IPC) calls, or—let's get crazy, shall we?—you
could even implement an email-based protocol. Subversion
supplies the APIs; you supply the creativity.</p></div><div class="sect2" title="Client Layer"><div class="titlepage"><div><div><h3 class="title"><a id="svn.developer.layerlib.client"></a>Client Layer</h3></div></div></div><p>On the client side, the Subversion working copy is where
all the action takes place. The bulk of functionality
implemented by the client-side libraries exists for the sole
purpose of managing working copies—directories full of
files and other subdirectories that serve as a sort of local,
editable <span class="quote">“<span class="quote">reflection</span>”</span> of one or more repository
locations—and propagating changes to and from the
Repository Access layer.</p><p>Subversion's working copy library,
<code class="filename">libsvn_wc</code>, is directly responsible for
managing the data in the working copies. To accomplish this,
the library stores administrative information about the
working copy within a special subdirectory. This
subdirectory, named <code class="filename">.svn</code>, is present in
each working copy and contains various other files
and directories that record state and provide a private
workspace for administrative action. For those familiar with
CVS, this <code class="filename">.svn</code> subdirectory is similar in
purpose to the <code class="filename">CVS</code> administrative
directories found in CVS working copies.</p><p>The Subversion client library,
<code class="filename">libsvn_client</code>, has the broadest
responsibility; its job is to mingle the functionality of the
working copy library with that of the Repository Access layer,
and then to provide the highest-level API to any application
that wishes to perform general revision control actions. For
example, the function
<code class="function">svn_client_checkout()</code> takes a URL as an
argument. It passes this URL to the RA layer and opens an
authenticated session with a particular repository. It then
asks the repository for a certain tree, and sends this tree
into the working copy library, which then writes a full
working copy to disk (<code class="filename">.svn</code> directories
and all).</p><p>The client library is designed to be used by any
application. While the Subversion source code includes a
standard command-line client, it should be very easy to write
any number of GUI clients on top of the client library. New
GUIs (or any new client, really) for Subversion need not be
clunky wrappers around the included command-line
client—they have full access via the
<code class="filename">libsvn_client</code> API to the same functionality,
data, and callback mechanisms that the command-line client
uses. In fact, the Subversion source code tree contains a
small C program (which you can find at
<code class="filename">tools/examples/minimal_client.c</code>) that
exemplifies how to wield the Subversion API to create a simple
client program.</p><div class="sidebar" title="Binding Directly—A Word About Correctness"><div class="titlepage"><div><div><p class="title"><strong>Binding Directly—A Word About Correctness</strong></p></div></div></div><p>Why should your GUI program bind directly with a
<code class="filename">libsvn_client</code> instead of acting as a
wrapper around a command-line program? Besides simply being
more efficient, it can be more correct as well. A
command-line program (such as the one supplied with
Subversion) that binds to the client library needs to
effectively translate feedback and requested data bits from
C types to some form of human-readable output. This type of
translation can be lossy. That is, the program may not
display all of the information harvested from the API or may
combine bits of information for compact
representation.</p><p>If you wrap such a command-line program with yet another
program, the second program has access only to
already interpreted (and as we mentioned, likely incomplete)
information, which it must <span class="emphasis"><em>again</em></span>
translate into <span class="emphasis"><em>its</em></span> representation
format. With each layer of wrapping, the integrity of the
original data is potentially tainted more and more, much
like the result of making a copy of a copy (of a copy…)
of a favorite audio or video cassette.</p><p>But the most compelling argument for binding directly to
the APIs instead of wrapping other programs is that the
Subversion project makes compatibility promises regarding
its APIs. Across minor versions of those APIs (such as
between 1.3 and 1.4), no function's prototype will change.
In other words, you aren't forced to update your program's
source code simply because you've upgraded to a new version
of Subversion. Certain functions might be deprecated, but
they still work, and this gives you a buffer of time to
eventually embrace the newer APIs. These kinds of
compatibility promises do not exist for Subversion
command-line program output, which is subject to change from
release to release.</p></div></div></div><div class="sect1" title="Using the APIs"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.developer.usingapi"></a>Using the APIs</h2></div></div></div><p>Developing applications against the Subversion library APIs
is fairly straightforward. Subversion is primarily a set of C
libraries, with header (<code class="filename">.h</code>) files that live
in the <code class="filename">subversion/include</code> directory of the
source tree. These headers are copied into your system
locations (e.g., <code class="filename">/usr/local/include</code>)
when you build and install Subversion itself from source. These
headers represent the entirety of the functions and types meant
to be accessible by users of the Subversion libraries. The
Subversion developer community is meticulous about ensuring that
the public API is well documented—refer directly to the
header files for that documentation.</p><p>When examining the public header files, the first thing you
might notice is that Subversion's datatypes and functions are
namespace-protected. That is, every public Subversion symbol
name begins with <code class="literal">svn_</code>, followed by a short
code for the library in which the symbol is defined (such as
<code class="literal">wc</code>, <code class="literal">client</code>,
<code class="literal">fs</code>, etc.), followed by a single underscore
(<code class="literal">_</code>), and then the rest of the symbol name.
Semipublic functions (used among source files of a given
library but not by code outside that library, and found inside
the library directories themselves) differ from this naming
scheme in that instead of a single underscore after the library
code, they use a double underscore
(<code class="literal">_ _</code>). Functions that are private to
a given source file have no special prefixing and are declared
<code class="literal">static</code>. Of course, a compiler isn't
interested in these naming conventions, but they help to clarify
the scope of a given function or datatype.</p><p>Another good source of information about programming against
the Subversion APIs is the project's own hacking guidelines,
which you can find at
<a class="ulink" href="http://subversion.apache.org/docs/community-guide/" target="_top">http://subversion.apache.org/docs/community-guide/</a>. This document contains useful information, which, while
aimed at developers and would-be developers of Subversion
itself, is equally applicable to folks developing against
Subversion as a set of third-party
libraries.<sup>[<a id="idp16116384" href="#ftn.idp16116384" class="footnote">67</a>]</sup></p><div class="sect2" title="The Apache Portable Runtime Library"><div class="titlepage"><div><div><h3 class="title"><a id="svn.developer.usingapi.apr"></a>The Apache Portable Runtime Library</h3></div></div></div><p>Along with Subversion's own datatypes, you will see many
references to datatypes that begin with
<code class="literal">apr_</code>—symbols from the Apache Portable
Runtime (APR) library. APR is Apache's portability library,
originally carved out of its server code as an attempt to
separate the OS-specific bits from the OS-independent portions
of the code. The result was a library that provides a generic
API for performing operations that differ mildly—or
wildly—from OS to OS. While the Apache HTTP Server was
obviously the first user of the APR library, the Subversion
developers immediately recognized the value of using APR as
well. This means that there is practically no OS-specific
code in Subversion itself. Also, it means that the Subversion
client compiles and runs anywhere that the Apache HTTP Server
does. Currently, this list includes all flavors of Unix,
Win32, BeOS, OS/2, and Mac OS X.</p><p>In addition to providing consistent implementations of
system calls that differ across operating
systems,<sup>[<a id="idp16119920" href="#ftn.idp16119920" class="footnote">68</a>]</sup> APR gives
Subversion immediate access to many custom datatypes, such as
dynamic arrays and hash tables. Subversion uses these types
extensively. But perhaps the most pervasive APR datatype,
found in nearly every Subversion API prototype, is the
<code class="literal">apr_pool_t</code>—the APR memory pool.
Subversion uses pools internally for all its memory allocation
needs (unless an external library requires a different memory
management mechanism for data passed through its
API),<sup>[<a id="idp16122864" href="#ftn.idp16122864" class="footnote">69</a>]</sup> and while a person coding against
the Subversion APIs is not required to do the same,
she <span class="emphasis"><em>is</em></span> required to provide pools to the
API functions that need them. This means that users of the
Subversion API must also link against APR, must
call <code class="function">apr_initialize()</code> to initialize the
APR subsystem, and then must create and manage pools for use
with Subversion API calls, typically by
using <code class="function">svn_pool_create()</code>,
<code class="function">svn_pool_clear()</code>, and
<code class="function">svn_pool_destroy()</code>.</p><div class="sidebar" title="Programming with Memory Pools"><div class="titlepage"><div><div><p class="title"><strong>Programming with Memory Pools</strong></p></div></div></div><p>Almost every developer who has used the C programming
language has at some point sighed at the daunting task of
managing memory usage. Allocating enough memory to use,
keeping track of those allocations, freeing the memory when
you no longer need it—these tasks can be quite
complex. And of course, failure to do those things properly
can result in a program that crashes itself, or worse,
crashes the computer.</p><p>Higher-level languages, on the other hand, either take
the job of memory management away from you completely or
make it something you toy with only when doing extremely
tight program optimization. Languages such as Java and
Python use <em class="firstterm">garbage collection</em>,
allocating memory for objects when needed, and automatically
freeing that memory when the object is no longer in
use.</p><p>APR provides a middle-ground approach called
<em class="firstterm">pool-based memory management</em>. It
allows the developer to control memory usage at a lower
resolution—per chunk (or <span class="quote">“<span class="quote">pool</span>”</span>) of
memory, instead of per allocated object. Rather than using
<code class="function">malloc()</code> and friends to allocate enough
memory for a given object, you ask APR to allocate the
memory from a memory pool. When you're finished using the
objects you've created in the pool, you destroy the entire
pool, effectively de-allocating the memory consumed by
<span class="emphasis"><em>all</em></span> the objects you allocated from it.
Thus, rather than keeping track of individual objects that
need to be de-allocated, your program simply considers the
general lifetimes of those objects and allocates the objects
in a pool whose lifetime (the time between the pool's
creation and its deletion) matches the object's
needs.</p></div></div><div class="sect2" title="Functions and Batons"><div class="titlepage"><div><div><h3 class="title"><a id="svn.developer.usingapi.funcsbatons"></a>Functions and Batons</h3></div></div></div><p>To facilitate <span class="quote">“<span class="quote">streamy</span>”</span> (asynchronous) behavior
and provide consumers of the Subversion C API with hooks for
handling information in customizable ways, many functions in
the API accept pairs of parameters: a pointer to a callback
function, and a pointer to a blob of memory called
a <em class="firstterm">baton</em> that carries context
information for that callback function. Batons are typically
C structures with additional information that the callback
function needs but which is not given directly to the callback
function by the driving API function.</p></div><div class="sect2" title="URL and Path Requirements"><div class="titlepage"><div><div><h3 class="title"><a id="svn.developer.usingapi.urlpath"></a>URL and Path Requirements</h3></div></div></div><p>With remote version control operation as the whole point
of Subversion's existence, it makes sense that some attention
has been paid to internationalization (i18n) support. After
all, while <span class="quote">“<span class="quote">remote</span>”</span> might mean <span class="quote">“<span class="quote">across the
office,</span>”</span> it could just as well mean <span class="quote">“<span class="quote">across the
globe.</span>”</span> To facilitate this, all of Subversion's public
interfaces that accept path arguments expect those paths to be
canonicalized—which is most easily accomplished by
passing them through <code class="function">svn_dirent_canonicalize()</code>
or <code class="function">svn_uri_canonicalize()</code> (depending on
whether you are canonicalizing a local system path or a URL,
respectively)—and encoded in UTF-8. This means, for
example, that any new client binary that drives the
<code class="filename">libsvn_client</code> interface needs to first
convert paths from the locale-specific encoding to UTF-8
before passing those paths to the Subversion libraries, and
then reconvert any resultant output paths from Subversion
back into the locale's encoding before using those paths for
non-Subversion purposes. Fortunately, Subversion provides a
suite of functions (see
<code class="filename">subversion/include/svn_utf.h</code>) that
any program can use to do these conversions.</p><p>Also, Subversion APIs require all URL parameters to be
properly URI-encoded. So, instead of passing
<code class="uri">file:///home/username/My File.txt</code> as the URL of a
file named <code class="filename">My File.txt</code>, you need to pass
<code class="uri">file:///home/username/My%20File.txt</code>. Again,
Subversion supplies helper functions that your application can
use—<code class="function">svn_path_uri_encode()</code> and
<code class="function">svn_path_uri_decode()</code>, for URI encoding
and decoding, respectively.</p></div><div class="sect2" title="Using Languages Other Than C and C++"><div class="titlepage"><div><div><h3 class="title"><a id="svn.developer.usingapi.otherlangs"></a>Using Languages Other Than C and C++</h3></div></div></div><p>If you are interested in using the Subversion libraries in
conjunction with something other than a C program—say, a
Python or Perl script—Subversion has some support for this
via the Simplified Wrapper and Interface Generator (SWIG). The
SWIG bindings for Subversion are located in
<code class="filename">subversion/bindings/swig</code>. They are still
maturing, but they are usable. These bindings allow you
to call Subversion API functions indirectly, using wrappers that
translate the datatypes native to your scripting language into
the datatypes needed by Subversion's C libraries.</p><p>Significant efforts have been made toward creating
functional SWIG-generated bindings for Python, Perl, and Ruby.
To some extent, the work done preparing the SWIG interface
files for these languages is reusable in efforts to generate
bindings for other languages supported by SWIG (which include
versions of C#, Guile, Java, MzScheme, OCaml, PHP, and Tcl,
among others). However, some extra programming is required to
compensate for complex APIs that SWIG needs some help
translating between languages. For more information on SWIG
itself, see the project's web site at <a class="ulink" href="http://www.swig.org/" target="_top">http://www.swig.org/</a>.</p><p>Subversion also has language bindings for Java. The
javahl bindings (located in
<code class="filename">subversion/bindings/java</code> in the
Subversion source tree) aren't SWIG-based, but are instead a
mixture of Java and hand-coded JNI. Javahl covers most
Subversion client-side APIs and is specifically targeted at
implementors of Java-based Subversion clients and IDE
integrations.</p><p>Subversion's language bindings tend to lack the level of
developer attention given to the core Subversion modules, but
can generally be trusted as production-ready. A number of
scripts and applications, alternative Subversion GUI clients,
and other third-party tools are successfully using
Subversion's language bindings today to accomplish their
Subversion integrations.</p><p>It's worth noting here that there are other options for
interfacing with Subversion using other languages: alternative
bindings for Subversion that aren't provided by the
Subversion development community at all. There
are a couple of popular ones we feel are especially
noteworthy. First, Barry Scott's PySVN bindings (<a class="ulink" href="http://pysvn.tigris.org/" target="_top">http://pysvn.tigris.org/</a>) are a popular option for
binding with Python. PySVN boasts of a more Pythonic
interface than the more C-like APIs provided by Subversion's
own Python bindings. And if you're looking for a pure Java
implementation of Subversion, check out SVNKit (<a class="ulink" href="http://svnkit.com/" target="_top">http://svnkit.com/</a>), which is Subversion rewritten
from the ground up in Java.</p><div class="sidebar" title="SVNKit Versus javahl"><div class="titlepage"><div><div><p class="title"><strong>SVNKit Versus javahl</strong></p></div></div></div><p>In 2005, a small company called TMate announced the
1.0.0 release of JavaSVN—a pure Java implementation of
Subversion. Since then, the project has been renamed to
SVNKit (available at <a class="ulink" href="http://svnkit.com/" target="_top">http://svnkit.com/</a>)
and has seen great success as a provider of Subversion
functionality to various Subversion clients, IDE
integrations, and other third-party tools.</p><p>The SVNKit library is interesting in that, unlike the
javahl library, it is not merely a wrapper around the
official Subversion core libraries. In fact, it shares no
code with Subversion at all. But while it is easy to
confuse SVNKit with javahl, and easier still to not even
realize which of these libraries you are using, folks should
be aware that SVNKit differs from javahl in some significant
ways. First, while SVNKit is developed as open source
software just like Subversion, SVNKit's license is more
restrictive than that of
Subversion.<sup>[<a id="idp16160688" href="#ftn.idp16160688" class="footnote">70</a>]</sup> Finally, by aiming to be a
pure Java Subversion library, SVNKit is limited in which
portions of Subversion can be reasonably cloned while still
keeping up with Subversion's releases. This has already
happened once—SVNKit cannot access BDB-backed
Subversion repositories via the <code class="literal">file://</code>
protocol because there's no pure Java implementation of
Berkeley DB that is file-format-compatible with the native
implementation of that library.</p><p>That said, SVNKit has a well-established track record of
reliability. And a pure Java solution is much more robust
in the face of programming errors—a bug in SVNKit
might raise a catchable Java Exception, but a bug in the
Subversion core libraries as accessed via javahl can bring
down your entire Java Runtime Environment. So, weigh the
costs when choosing a Java-based Subversion
implementation.</p></div></div><div class="sect2" title="Code Samples"><div class="titlepage"><div><div><h3 class="title"><a id="svn.developer.usingapi.codesamples"></a>Code Samples</h3></div></div></div><p><a class="xref" href="#svn.developer.layerlib.repos.ex-1" title="Example 8.1. Using the repository layer">Example 8.1, “Using the repository layer”</a>
contains a code segment (written in C) that illustrates some
of the concepts we've been discussing. It uses both the
repository and filesystem interfaces (as can be determined by
the prefixes <code class="literal">svn_repos_</code> and
<code class="literal">svn_fs_</code> of the function names,
respectively) to create a new revision in which a directory is
added. You can see the use of an APR pool, which is passed
around for memory allocation purposes. Also, the code reveals
a somewhat obscure fact about Subversion error
handling—all Subversion errors must be explicitly
handled to avoid memory leakage (and in some cases,
application failure).</p><div class="example"><a id="svn.developer.layerlib.repos.ex-1"></a><p class="title"><strong>Example 8.1. Using the repository layer</strong></p><div class="example-contents"><pre class="programlisting">
/* Convert a Subversion error into a simple boolean error code.
*
* NOTE: Subversion errors must be cleared (using svn_error_clear())
* because they are allocated from the global pool, else memory
* leaking occurs.
*/
#define INT_ERR(expr) \
do { \
svn_error_t *__temperr = (expr); \
if (__temperr) \
{ \
svn_error_clear(__temperr); \
return 1; \
} \
return 0; \
} while (0)
/* Create a new directory at the path NEW_DIRECTORY in the Subversion
* repository located at REPOS_PATH. Perform all memory allocation in
* POOL. This function will create a new revision for the addition of
* NEW_DIRECTORY. Return zero if the operation completes
* successfully, nonzero otherwise.
*/
static int
make_new_directory(const char *repos_path,
const char *new_directory,
apr_pool_t *pool)
{
svn_error_t *err;
svn_repos_t *repos;
svn_fs_t *fs;
svn_revnum_t youngest_rev;
svn_fs_txn_t *txn;
svn_fs_root_t *txn_root;
const char *conflict_str;
/* Open the repository located at REPOS_PATH.
*/
INT_ERR(svn_repos_open(&repos, repos_path, pool));
/* Get a pointer to the filesystem object that is stored in REPOS.
*/
fs = svn_repos_fs(repos);
/* Ask the filesystem to tell us the youngest revision that
* currently exists.
*/
INT_ERR(svn_fs_youngest_rev(&youngest_rev, fs, pool));
/* Begin a new transaction that is based on YOUNGEST_REV. We are
* less likely to have our later commit rejected as conflicting if we
* always try to make our changes against a copy of the latest snapshot
* of the filesystem tree.
*/
INT_ERR(svn_repos_fs_begin_txn_for_commit2(&txn, repos, youngest_rev,
apr_hash_make(pool), pool));
/* Now that we have started a new Subversion transaction, get a root
* object that represents that transaction.
*/
INT_ERR(svn_fs_txn_root(&txn_root, txn, pool));
/* Create our new directory under the transaction root, at the path
* NEW_DIRECTORY.
*/
INT_ERR(svn_fs_make_dir(txn_root, new_directory, pool));
/* Commit the transaction, creating a new revision of the filesystem
* which includes our added directory path.
*/
err = svn_repos_fs_commit_txn(&conflict_str, repos,
&youngest_rev, txn, pool);
if (! err)
{
/* No error? Excellent! Print a brief report of our success.
*/
printf("Directory '%s' was successfully added as new revision "
"'%ld'.\n", new_directory, youngest_rev);
}
else if (err->apr_err == SVN_ERR_FS_CONFLICT)
{
/* Uh-oh. Our commit failed as the result of a conflict
* (someone else seems to have made changes to the same area
* of the filesystem that we tried to modify). Print an error
* message.
*/
printf("A conflict occurred at path '%s' while attempting "
"to add directory '%s' to the repository at '%s'.\n",
conflict_str, new_directory, repos_path);
}
else
{
/* Some other error has occurred. Print an error message.
*/
printf("An error occurred while attempting to add directory '%s' "
"to the repository at '%s'.\n",
new_directory, repos_path);
}
INT_ERR(err);
}
</pre></div></div><br class="example-break"></br><p>Note that in <a class="xref" href="#svn.developer.layerlib.repos.ex-1" title="Example 8.1. Using the repository layer">Example 8.1, “Using the repository layer”</a>, the code could
just as easily have committed the transaction using
<code class="function">svn_fs_commit_txn()</code>. But the filesystem
API knows nothing about the repository library's hook
mechanism. If you want your Subversion repository to
automatically perform some set of non-Subversion tasks every
time you commit a transaction (e.g., sending an
email that describes all the changes made in that transaction
to your developer mailing list), you need to use the
<code class="filename">libsvn_repos</code>-wrapped version of that
function, which adds the hook triggering
functionality—in this case,
<code class="function">svn_repos_fs_commit_txn()</code>. (For more
information regarding Subversion's repository hooks, see <a class="xref" href="#svn.reposadmin.create.hooks" title="Implementing Repository Hooks">the section called “Implementing Repository Hooks”</a>.)</p><p>Now let's switch languages. <a class="xref" href="#svn.developer.usingapi.otherlangs.ex-1" title="Example 8.2. Using the repository layer with Python">Example 8.2, “Using the repository layer with Python”</a> is a
sample program that uses Subversion's SWIG Python bindings to
recursively crawl the youngest repository revision, and to
print the various paths reached during the crawl.</p><div class="example"><a id="svn.developer.usingapi.otherlangs.ex-1"></a><p class="title"><strong>Example 8.2. Using the repository layer with Python</strong></p><div class="example-contents"><pre class="programlisting">
#!/usr/bin/python
"""Crawl a repository, printing versioned object path names."""
import sys
import os.path
import svn.fs, svn.core, svn.repos
def crawl_filesystem_dir(root, directory):
"""Recursively crawl DIRECTORY under ROOT in the filesystem, and return
a list of all the paths at or below DIRECTORY."""
# Print the name of this path.
print directory + "/"
# Get the directory entries for DIRECTORY.
entries = svn.fs.svn_fs_dir_entries(root, directory)
# Loop over the entries.
names = entries.keys()
for name in names:
# Calculate the entry's full path.
full_path = directory + '/' + name
# If the entry is a directory, recurse. The recursion will return
# a list with the entry and all its children, which we will add to
# our running list of paths.
if svn.fs.svn_fs_is_dir(root, full_path):
crawl_filesystem_dir(root, full_path)
else:
# Else it's a file, so print its path here.
print full_path
def crawl_youngest(repos_path):
"""Open the repository at REPOS_PATH, and recursively crawl its
youngest revision."""
# Open the repository at REPOS_PATH, and get a reference to its
# versioning filesystem.
repos_obj = svn.repos.svn_repos_open(repos_path)
fs_obj = svn.repos.svn_repos_fs(repos_obj)
# Query the current youngest revision.
youngest_rev = svn.fs.svn_fs_youngest_rev(fs_obj)
# Open a root object representing the youngest (HEAD) revision.
root_obj = svn.fs.svn_fs_revision_root(fs_obj, youngest_rev)
# Do the recursive crawl.
crawl_filesystem_dir(root_obj, "")
if __name__ == "__main__":
# Check for sane usage.
if len(sys.argv) != 2:
sys.stderr.write("Usage: %s REPOS_PATH\n"
% (os.path.basename(sys.argv[0])))
sys.exit(1)
# Canonicalize the repository path.
repos_path = svn.core.svn_dirent_canonicalize(sys.argv[1])
# Do the real work.
crawl_youngest(repos_path)
</pre></div></div><br class="example-break"></br><p>This same program in C would need to deal with APR's
memory pool system. But Python handles memory usage
automatically, and Subversion's Python bindings adhere to that
convention. In C, you'd be working with custom datatypes
(such as those provided by the APR library) for representing
the hash of entries and the list of paths, but Python has
hashes (called <span class="quote">“<span class="quote">dictionaries</span>”</span>) and lists as
built-in datatypes, and it provides a rich collection of
functions for operating on those types. So SWIG (with the
help of some customizations in Subversion's language bindings
layer) takes care of mapping those custom datatypes into the
native datatypes of the target language. This provides a more
intuitive interface for users of that language.</p><p>The Subversion Python bindings can be used for working
copy operations, too. In the previous section of this
chapter, we mentioned the <code class="filename">libsvn_client</code>
interface and how it exists for the sole purpose of
simplifying the process of writing a Subversion client. <a class="xref" href="#svn.developer.usingapi.otherlangs.ex-2" title="Example 8.3. A Python status crawler">Example 8.3, “A Python status crawler”</a> is a brief
example of how that library can be accessed via the SWIG
Python bindings to re-create a scaled-down version of the
<span class="command"><strong>svn status</strong></span> command.</p><div class="example"><a id="svn.developer.usingapi.otherlangs.ex-2"></a><p class="title"><strong>Example 8.3. A Python status crawler</strong></p><div class="example-contents"><pre class="programlisting">
#!/usr/bin/env python
"""Crawl a working copy directory, printing status information."""
import sys
import os.path
import getopt
import svn.core, svn.client, svn.wc
def generate_status_code(status):
"""Translate a status value into a single-character status code,
using the same logic as the Subversion command-line client."""
code_map = { svn.wc.svn_wc_status_none : ' ',
svn.wc.svn_wc_status_normal : ' ',
svn.wc.svn_wc_status_added : 'A',
svn.wc.svn_wc_status_missing : '!',
svn.wc.svn_wc_status_incomplete : '!',
svn.wc.svn_wc_status_deleted : 'D',
svn.wc.svn_wc_status_replaced : 'R',
svn.wc.svn_wc_status_modified : 'M',
svn.wc.svn_wc_status_conflicted : 'C',
svn.wc.svn_wc_status_obstructed : '~',
svn.wc.svn_wc_status_ignored : 'I',
svn.wc.svn_wc_status_external : 'X',
svn.wc.svn_wc_status_unversioned : '?',
}
return code_map.get(status, '?')
def do_status(wc_path, verbose, prefix):
# Build a client context baton.
ctx = svn.client.svn_client_create_context()
def _status_callback(path, status):
"""A callback function for svn_client_status."""
# Print the path, minus the bit that overlaps with the root of
# the status crawl
text_status = generate_status_code(status.text_status)
prop_status = generate_status_code(status.prop_status)
prefix_text = ''
if prefix is not None:
prefix_text = prefix + " "
print '%s%s%s %s' % (prefix_text, text_status, prop_status, path)
# Do the status crawl, using _status_callback() as our callback function.
revision = svn.core.svn_opt_revision_t()
revision.type = svn.core.svn_opt_revision_head
svn.client.svn_client_status2(wc_path, revision, _status_callback,
svn.core.svn_depth_infinity, verbose,
0, 0, 1, ctx)
def usage_and_exit(errorcode):
"""Print usage message, and exit with ERRORCODE."""
stream = errorcode and sys.stderr or sys.stdout
stream.write("""Usage: %s OPTIONS WC-PATH
Print working copy status, optionally with a bit of prefix text.
Options:
--help, -h : Show this usage message
--prefix ARG : Print ARG, followed by a space, before each line of output
--verbose, -v : Show all statuses, even uninteresting ones
""" % (os.path.basename(sys.argv[0])))
sys.exit(errorcode)
if __name__ == '__main__':
# Parse command-line options.
try:
opts, args = getopt.getopt(sys.argv[1:], "hv",
["help", "prefix=", "verbose"])
except getopt.GetoptError:
usage_and_exit(1)
verbose = 0
prefix = None
for opt, arg in opts:
if opt in ("-h", "--help"):
usage_and_exit(0)
if opt in ("--prefix"):
prefix = arg
if opt in ("-v", "--verbose"):
verbose = 1
if len(args) != 1:
usage_and_exit(2)
# Canonicalize the working copy path.
wc_path = svn.core.svn_dirent_canonicalize(args[0])
# Do the real work.
try:
do_status(wc_path, verbose, prefix)
except svn.core.SubversionException, e:
sys.stderr.write("Error (%d): %s\n" % (e.apr_err, e.message))
sys.exit(1)
</pre></div></div><br class="example-break"></br><p>As was the case in
<a class="xref" href="#svn.developer.usingapi.otherlangs.ex-1" title="Example 8.2. Using the repository layer with Python">Example 8.2, “Using the repository layer with Python”</a>,
this program is pool-free and uses, for the most part, normal
Python datatypes.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>Run user-provided paths
through the appropriate canonicalization function
(<code class="function">svn_dirent_canonicalize()</code> or
<code class="function">svn_uri_canonicalize()</code>) before passing
them to other API functions. Failure to do so can trigger
assertions in the underlying Subversion C library which
translate into rather immediate and unceremonious program
abortion.</p></td></tr></table></div><p>Of particular interest to users of the Python flavor of
Subversion's API is the implementation of callback functions.
As previously mentioned, Subversion's C API makes liberal use
of the callback function/baton paradigm. API functions which
in C accept a function and baton pair only accept a callback
function parameter in Python. How, then, does the caller pass
arbitrary context information to the callback function? In
Python, this is done by taking advantage of Python's scoping
rules and default argument values. You can see this in action
in <a class="xref" href="#svn.developer.usingapi.otherlangs.ex-2" title="Example 8.3. A Python status crawler">Example 8.3, “A Python status crawler”</a>.
The <code class="function">svn_client_status2()</code> function is
given a callback function
(<code class="function">_status_callback()</code>) but no
baton—<code class="function">_status_callback()</code> gets
access to the user-provided prefix string because that
variable falls into the scope of the function
automatically.</p></div></div><div class="sect1" title="Summary"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.developer.summary"></a>Summary</h2></div></div></div><p>One of Subversion's greatest features isn't something you
get from running its command-line client or other tools. It's
the fact that Subversion was designed modularly and provides a
stable, public API so that others—like yourself,
perhaps—can write custom software that drives Subversion's
core logic.</p><p>In this chapter, we took a closer look at Subversion's
architecture, examining its logical layers and describing that
public API, the very same API that Subversion's own layers use
to communicate with each other. Many developers have found
interesting uses for the Subversion API, from simple repository
hook scripts, to integrations between Subversion and some other
application, to completely different version control systems.
What unique itch will <span class="emphasis"><em>you</em></span> scratch with
it?</p></div><div class="footnotes"><br></br><hr width="100" align="left"></hr><div class="footnote"><p><sup>[<a id="ftn.idp16034992" href="#idp16034992" class="para">66</a>] </sup>We understand that this may
come as a shock to sci-fi fans who have long been under the
impression that Time was actually
the <span class="emphasis"><em>fourth</em></span> dimension, and we apologize
for any emotional trauma induced by our assertion of a
different theory.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp16116384" href="#idp16116384" class="para">67</a>] </sup>After all, Subversion uses
Subversion's APIs, too.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp16119920" href="#idp16119920" class="para">68</a>] </sup>Subversion uses ANSI system calls and
datatypes as much as possible.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp16122864" href="#idp16122864" class="para">69</a>] </sup>Neon and Berkeley DB are examples of such
libraries.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp16160688" href="#idp16160688" class="para">70</a>] </sup>Redistributions in any form must
be accompanied by information on how to obtain complete
source code for the software that uses SVNKit and any
accompanying software that uses the software that uses
SVNKit. See <a class="ulink" href="http://svnkit.com/license.html" target="_top">http://svnkit.com/license.html</a>
for details.</p></div></div></div><div class="chapter" title="Chapter 9. Subversion Complete Reference"><div class="titlepage"><div><div><h2 class="title"><a id="svn.ref"></a>Chapter 9. Subversion Complete Reference</h2></div></div></div><div class="toc"><p><strong>Table of Contents</strong></p><dl><dt><span class="sect1"><a href="#svn.ref.svn">svn—Subversion Command-Line Client</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.ref.svn.sw">svn Options</a></span></dt><dt><span class="sect2"><a href="#svn.ref.svn.c">svn Subcommands</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.add">svn add</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.blame">svn blame (praise, annotate, ann)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.cat">svn cat</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.changelist">svn changelist (cl)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.checkout">svn checkout (co)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.cleanup">svn cleanup</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.commit">svn commit (ci)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.copy">svn copy (cp)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.delete">svn delete (del, remove, rm)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.diff">svn diff (di)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.export">svn export</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.help">svn help (h, ?)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.import">svn import</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.info">svn info</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.list">svn list (ls)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.lock">svn lock</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.log">svn log</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.merge">svn merge</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.mergeinfo">svn mergeinfo</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.mkdir">svn mkdir</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.move">svn move (mv)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.patch">svn patch</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.propdel">svn propdel (pdel, pd)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.propedit">svn propedit (pedit, pe)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.propget">svn propget (pget, pg)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.proplist">svn proplist (plist, pl)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.propset">svn propset (pset, ps)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.relocate">svn relocate</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.resolve">svn resolve</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.resolved">svn resolved</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.revert">svn revert</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.status">svn status (stat, st)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.switch">svn switch (sw)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.unlock">svn unlock</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.update">svn update (up)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svn.c.upgrade">svn upgrade</a></span><span class="refpurpose"></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.ref.svnadmin">svnadmin—Subversion Repository Administration</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.ref.svnadmin.sw">svnadmin Options</a></span></dt><dt><span class="sect2"><a href="#svn.ref.svnadmin.c">svnadmin Subcommands</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.crashtest">svnadmin crashtest</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.create">svnadmin create</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.deltify">svnadmin deltify</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.dump">svnadmin dump</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.help">svnadmin help (h, ?)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.hotcopy">svnadmin hotcopy</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.list-dblogs">svnadmin list-dblogs</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.list-unused-dblogs">svnadmin list-unused-dblogs</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.load">svnadmin load</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.lslocks">svnadmin lslocks</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.lstxns">svnadmin lstxns</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.pack">svnadmin pack</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.recover">svnadmin recover</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.rmlocks">svnadmin rmlocks</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.rmtxns">svnadmin rmtxns</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.setlog">svnadmin setlog</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.setrevprop">svnadmin setrevprop</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.setuuid">svnadmin setuuid</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.upgrade">svnadmin upgrade</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnadmin.c.verify">svnadmin verify</a></span><span class="refpurpose"></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.ref.svnlook">svnlook—Subversion Repository Examination</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.ref.svnlook.sw">svnlook Options</a></span></dt><dt><span class="sect2"><a href="#svn.ref.svnlook.c">svnlook Subcommands</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.author">svnlook author</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.cat">svnlook cat</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.changed">svnlook changed</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.date">svnlook date</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.diff">svnlook diff</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.dirs-changed">svnlook dirs-changed</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.filesize">svnlook filesize</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.help">svnlook help (h, ?)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.history">svnlook history</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.info">svnlook info</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.lock">svnlook lock</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.log">svnlook log</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.propget">svnlook propget (pget, pg)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.proplist">svnlook proplist (plist, pl)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.tree">svnlook tree</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.uuid">svnlook uuid</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnlook.c.youngest">svnlook youngest</a></span><span class="refpurpose"></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.ref.svnsync">svnsync—Subversion Repository Mirroring</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.ref.svnsync.sw">svnsync Options</a></span></dt><dt><span class="sect2"><a href="#svn.ref.svnsync.c">svnsync Subcommands</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.svnsync.c.copy-revprops">svnsync copy-revprops</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnsync.c.help">svnsync help</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnsync.c.info">svnsync info</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnsync.c.init">svnsync initialize (init)</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnsync.c.sync">svnsync synchronize (sync)</a></span><span class="refpurpose"></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.ref.svnrdump">svnrdump—Remote Subversion Repository Data Migration</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.ref.svnrdump.sw">svnrdump Options</a></span></dt><dt><span class="sect2"><a href="#svn.ref.svnrdump.c">svnrdump Subcommands</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.svnrdump.c.dump">svnrdump dump</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnrdump.c.help">svnrdump help</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svnrdump.c.load">svnrdump load</a></span><span class="refpurpose"></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.ref.svnserve">svnserve—Custom Subversion Server</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.ref.svnserve.sw">svnserve Options</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.ref.svndumpfilter">svndumpfilter—Subversion History Filtering</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.ref.svndumpfilter.options">svndumpfilter Options</a></span></dt><dt><span class="sect2"><a href="#svn.ref.svndumpfilter.c">svndumpfilter Subcommands</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.svndumpfilter.commands.c.exclude">svndumpfilter exclude</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svndumpfilter.commands.c.include">svndumpfilter include</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.svndumpfilter.commands.c.help">svndumpfilter help</a></span><span class="refpurpose"></span></dt></dl></dd></dl></dd><dt><span class="sect1"><a href="#svn.ref.svnversion">svnversion—Subversion Working Copy Version Info</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.svnversion.re">svnversion</a></span><span class="refpurpose"></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.ref.mod_dav_svn">mod_dav_svn—Subversion Apache HTTP Server Module</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.mod_dav_svn.conf">mod_dav_svn Configuration Directives</a></span><span class="refpurpose"></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.ref.mod_authz_svn">mod_authz_svn—Subversion Apache HTTP Authorization Module</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.mod_authz_svn.conf">mod_authz_svn Configuration Directives</a></span><span class="refpurpose"></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.ref.properties">Subversion Properties</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.ref.properties.versioned-props">Versioned Properties</a></span></dt><dt><span class="sect2"><a href="#svn.ref.properties.unversioned-props">Unversioned Properties</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.ref.reposhooks">Repository Hooks</a></span></dt><dd><dl><dt><span class="refentrytitle"><a href="#svn.ref.reposhooks.start-commit">start-commit</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.reposhooks.pre-commit">pre-commit</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.reposhooks.post-commit">post-commit</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.reposhooks.pre-revprop-change">pre-revprop-change</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.reposhooks.post-revprop-change">post-revprop-change</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.reposhooks.pre-lock">pre-lock</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.reposhooks.post-lock">post-lock</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.reposhooks.pre-unlock">pre-unlock</a></span><span class="refpurpose"></span></dt><dt><span class="refentrytitle"><a href="#svn.ref.reposhooks.post-unlock">post-unlock</a></span><span class="refpurpose"></span></dt></dl></dd></dl></div><p>This chapter is intended to be a complete reference to using
Subversion. It includes command summaries and examples for all
the command-line tools provided as part of the stock Subversion
distribution, configuration information for the Subversion server
modules, and other information that lends itself to a reference
format.</p><div class="sect1" title="svn—Subversion Command-Line Client"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.ref.svn"></a>svn—Subversion Command-Line Client</h2></div></div></div><p><span class="command"><strong>svn</strong></span> is the official command-line client
of Subversion. Its functionality is offered via a collection of
task-specific subcommands, most of which accept a number of
options for fine-grained control of the program's
behavior.</p><p>When using the <span class="command"><strong>svn</strong></span> program, subcommands
and other non-option arguments must appear in a specified order
on the command line. Options, on the other hand, may appear
anywhere on the command line (after the program name, of
course), and in general, their order is irrelevant. For
example, all of the following are valid ways to use <span class="command"><strong>svn
status</strong></span>, and are interpreted in exactly the same
way:</p><div class="informalexample"><pre class="screen">
$ svn -vq status myfile
$ svn status -v -q myfile
$ svn -q status -v myfile
$ svn status -vq myfile
$ svn status myfile -qv
</pre></div><p>The following sections describe each of the various
subcommands and options provided by the <span class="command"><strong>svn</strong></span>
command-line client program, including some examples of each
subcommand's typical uses.</p><div class="sect2" title="svn Options"><div class="titlepage"><div><div><h3 class="title"><a id="svn.ref.svn.sw"></a>svn Options</h3></div></div></div><p>While Subversion has different options for its
subcommands, all options exist in a single
namespace—that is, each option is guaranteed to mean the
roughly same thing regardless of the subcommand you use it
with. For example, <code class="option">--verbose</code>
(<code class="option">-v</code>) always means <span class="quote">“<span class="quote">verbose
output,</span>”</span> regardless of the subcommand you use it
with.</p><p>The <span class="command"><strong>svn</strong></span> command-line client usually
exits quickly with an error if you pass it an option which
does not apply to the specified subcommand. But as of
Subversion 1.5, several of the options which apply to
all—or nearly all—of the subcommands have been
deemed acceptable by all subcommands, even if they have no
effect on some of them. (This change was made primarily to
improve the client's ability to called from custom wrapping
scripts.) These options appear grouped together in the
command-line client's usage messages as global options, as can
be seen in the following bit of output:</p><div class="informalexample"><pre class="screen">
$ svn help upgrade
upgrade: Upgrade the metadata storage format for a working copy.
usage: upgrade [WCPATH...]
Local modifications are preserved.
Valid options:
-q [--quiet] : print nothing, or only summary information
Global options:
--username ARG : specify a username ARG
--password ARG : specify a password ARG
--no-auth-cache : do not cache authentication tokens
--non-interactive : do no interactive prompting
--trust-server-cert : accept SSL server certificates from unknown
certificate authorities without prompting (but only
with '--non-interactive')
--config-dir ARG : read user configuration files from directory ARG
--config-option ARG : set user configuration option in the format:
FILE:SECTION:OPTION=[VALUE]
For example:
servers:global:http-library=serf
$
</pre></div><p><span class="command"><strong>svn</strong></span> subcommands recognize the following
global options:</p><div class="variablelist"><dl><dt><a id="svn.ref.svn.sw.config_dir"></a><span class="term"><code class="option">--config-dir</code> <em class="replaceable"><code>DIR</code></em></span></dt><dd><p>Instructs Subversion to read configuration
information from the specified directory instead of the
default location (<code class="filename">.subversion</code> in
the user's home directory).</p></dd><dt><a id="svn.ref.svn.sw.config_option"></a><span class="term"><code class="option">--config-option</code> <em class="replaceable"><code>CONFSPEC</code></em></span></dt><dd><p>Sets, for the duration of the command, the value of
a runtime configuration
option. <em class="replaceable"><code>CONFSPEC</code></em> is a string
which specifies the configuration option namespace, name
and value that you'd like to assign, formatted as
<em class="replaceable"><code>FILE</code></em>:<em class="replaceable"><code>SECTION</code></em>:<em class="replaceable"><code>OPTION</code></em>=[<em class="replaceable"><code>VALUE</code></em>].
In this syntax, <em class="replaceable"><code>FILE</code></em>
and <em class="replaceable"><code>SECTION</code></em> are the runtime
configuration file (either <code class="literal">config</code>
or <code class="literal">servers</code>) and the section thereof,
respectively, which contain the option whose value you
wish to change. <em class="replaceable"><code>OPTION</code></em> is,
of course, the option itself,
and <em class="replaceable"><code>VALUE</code></em> the value (if any)
you wish to assign to the option. For example, to
temporarily disable the use of the compression in the
HTTP protocol,
use <code class="option">--config-option=servers:global:http-compression=no</code>.
You can use this option multiple times to change
multiple option values simultaneously.</p></dd><dt><a id="svn.ref.svn.sw.no_auth_cache"></a><span class="term"><code class="option">--no-auth-cache</code></span></dt><dd><p>Prevents caching of authentication information
(e.g., username and password) in the Subversion
runtime configuration directories.</p></dd><dt><a id="svn.ref.svn.sw.non_interactive"></a><span class="term"><code class="option">--non-interactive</code></span></dt><dd><p>Disables all interactive prompting. Some examples
of interactive prompting include requests for
authentication credentials and conflict resolution
decisions. This is useful if you're running Subversion
inside an automated script and it's more appropriate
to have Subversion fail than to prompt for more
information.</p></dd><dt><a id="svn.ref.svn.sw.password"></a><span class="term"><code class="option">--password</code> <em class="replaceable"><code>PASSWD</code></em></span></dt><dd><p>Specifies the password to use when authenticating
against a Subversion server. If not provided, or if
incorrect, Subversion will prompt you for this
information as needed.</p></dd><dt><a id="svn.ref.svn.sw.trust_server_cert"></a><span class="term"><code class="option">--trust-server-cert</code></span></dt><dd><p>When used with <code class="option">--non-interactive</code>,
instructs Subversion to accept SSL server certificates
issued by unknown certificate authorities without first
prompting the user. For security's sake, you should use
this option only when the integrity of the remote server
and the network path between it and your client is known
to be trustworthy.</p></dd><dt><a id="svn.ref.svn.sw.username"></a><span class="term"><code class="option">--username</code> <em class="replaceable"><code>NAME</code></em></span></dt><dd><p>Specifies the username to use when authenticating
against a Subversion server. If not provided, or if
incorrect, Subversion will prompt you for this
information as needed.</p></dd></dl></div><p>The rest of the options apply and are accepted by only a
subset of the subcommand. They are as follows:</p><div class="variablelist"><dl><dt><a id="svn.ref.svn.sw.accept"></a><span class="term"><code class="option">--accept</code> <em class="replaceable"><code>ACTION</code></em></span></dt><dd><p>Specifies an action for automatic conflict
resolution, disabling the interactive prompts which ask
the user how to handle each conflict as it is noticed.
Though which of the specific actions are applicable
differs depending on which subcommand is in use,
Subversion supports the following long (and short)
values for <em class="replaceable"><code>ACTION</code></em>:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">postpone</code>
(<code class="literal">p</code>)</span></dt><dd><p>Take no resolution action at all and instead
allow the conflicts to be recorded for future
resolution.</p></dd><dt><span class="term"><code class="literal">edit</code>
(<code class="literal">e</code>)</span></dt><dd><p>Open each conflicted file in a text editor for
manual resolution of line-based conflicts.</p></dd><dt><span class="term"><code class="literal">launch</code>
(<code class="literal">l</code>)</span></dt><dd><p>Launch an interactive merge conflict
resolution tool for each conflicted file.</p></dd><dt><span class="term"><code class="literal">base</code></span></dt><dd><p>Choose the file that was the (unmodified)
<code class="literal">BASE</code> revision before you tried
to integrate changes from the server into
your working copy.</p></dd><dt><span class="term"><code class="literal">working</code></span></dt><dd><p>Assuming that you've manually handled the
conflict resolution, choose the version of the file
as it currently stands in your working copy.</p></dd><dt><span class="term"><code class="literal">mine-full</code>
(<code class="literal">mf</code>)</span></dt><dd><p>Resolve conflicted files by preserving all
local modifications and discarding all changes
fetched from the server during the operation which
caused the conflict.</p></dd><dt><span class="term"><code class="literal">theirs-full</code>
(<code class="literal">tf</code>)</span></dt><dd><p>Resolve conflicted files by discarding all
local modifications and integrating all changes
fetched from the server during the operation which
caused the conflict.</p></dd><dt><span class="term"><code class="literal">mine-conflict</code>
(<code class="literal">mc</code>)</span></dt><dd><p>Resolve conflicted files by preferring local
modifications over the changes fetched from the
server in conflicting regions of each file's
content.</p></dd><dt><span class="term"><code class="literal">theirs-conflict</code>
(<code class="literal">tc</code>)</span></dt><dd><p>Resolve conflicted files by preferring the
changes fetched from the server over local
modifications in conflicting regions of each file's
content.</p></dd></dl></div><p>Consult the output of <span class="command"><strong>svn help
<em class="replaceable"><code>SUBCOMMAND</code></em></strong></span> to see
exactly which actions are supported by the specific
subcommand of interest.</p></dd><dt><a id="svn.ref.svn.sw.allow_mixed_revisions"></a><span class="term"><code class="option">--allow-mixed-revisions</code></span></dt><dd><p>Disables the verification—performed by default
by <span class="command"><strong>svn merge</strong></span> as of Subversion
1.7—that the target of a merge operation and all
of its children are at a uniform revision. While
merging into a single-revision working copy target is
the recommended best practice, this option may be used
to permit merges into mixed-revision working copies as
necessary.</p></dd><dt><a id="svn.ref.svn.sw.auto_props"></a><span class="term"><code class="option">--auto-props</code></span></dt><dd><p>Enables automatic property assignment (per runtime
configuration rules), overriding the
<code class="literal">enable-auto-props</code> runtime
configuration directive.</p></dd><dt><a id="svn.ref.svn.sw.change"></a><span class="term"><code class="option">--change</code> (<code class="option">-c</code>) <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>Perform the requested operation using a specific
<span class="quote">“<span class="quote">change</span>”</span>. Generally speaking, this option
is syntactic sugar for <strong class="userinput"><code>-r
<em class="replaceable"><code>ARG-1</code></em>:<em class="replaceable"><code>ARG</code></em></code></strong>.
Some subcommands permit a comma-separated list of
revision number arguments (e.g., <strong class="userinput"><code>-c
<em class="replaceable"><code>ARG1</code></em>,<em class="replaceable"><code>ARG2</code></em>,<em class="replaceable"><code>ARG3</code></em></code></strong>).
Alternatively, you can provide two arguments separated
by a dash (as
in <strong class="userinput"><code>-c <em class="replaceable"><code>ARG1</code></em>-<em class="replaceable"><code>ARG2</code></em></code></strong>)
to identify the range of revisions
between <em class="replaceable"><code>ARG1</code></em>
and <em class="replaceable"><code>ARG2</code></em>, inclusive.
Finally, if the revision argument is negated, the
implied revision range is reversed: <strong class="userinput"><code>-c
-45</code></strong> is equivalent to <strong class="userinput"><code>-r
45:44</code></strong>.</p></dd><dt><a id="svn.ref.svn.sw.changelist"></a><span class="term"><code class="option">--changelist</code> (<code class="option">--cl</code>) <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>Instructs Subversion to operate only on members of the changelist named
<em class="replaceable"><code>ARG</code></em>. You can use this option
multiple times to specify sets of changelists.</p></dd><dt><a id="svn.ref.svn.sw.depth"></a><span class="term"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>Instructs Subversion to limit the scope of an
operation to a particular tree
depth. <em class="replaceable"><code>ARG</code></em> is one of
<code class="literal">empty</code> (only the target
itself), <code class="literal">files</code> (the target and any
immediate file children thereof),
<code class="literal">immediates</code> (the target and any
immediate children thereof), or
<code class="literal">infinity</code> (the target and all of its
descendants—full recursion).</p></dd><dt><a id="svn.ref.svn.sw.diff"></a><span class="term"><code class="option">--diff</code></span></dt><dd><p>Enables a special output mode for <span class="command"><strong>svn
log</strong></span> which includes a difference report (a
la <span class="command"><strong>svn diff</strong></span>) as part of each
revision's information.</p></dd><dt><a id="svn.ref.svn.sw.diff_cmd"></a><span class="term"><code class="option">--diff-cmd</code> <em class="replaceable"><code>CMD</code></em></span></dt><dd><p>Specifies an external program to use to show
differences between files. When <span class="command"><strong>svn
diff</strong></span> is invoked without this option, it uses
Subversion's internal differencing engine, which provides
unified diffs by default. If you want to use an
external differencing program, use <code class="option">--diff-cmd</code>.
You can then pass options to the specified program using the
<code class="option">--extensions</code> (<code class="option">-x</code>)
option.</p></dd><dt><a id="svn.ref.svn.sw.diff3_cmd"></a><span class="term"><code class="option">--diff3-cmd</code> <em class="replaceable"><code>CMD</code></em></span></dt><dd><p>Specifies an external 3-way differencing program
(used to merge line-based changes into files).</p></dd><dt><a id="svn.ref.svn.sw.dry_run"></a><span class="term"><code class="option">--dry-run</code></span></dt><dd><p>Goes through all the motions of running a command,
but makes no actual changes—either on disk or in
the repository.</p></dd><dt><a id="svn.ref.svn.sw.editor_cmd"></a><span class="term"><code class="option">--editor-cmd</code> <em class="replaceable"><code>CMD</code></em></span></dt><dd><p>Specifies an external program to use to edit a log message
or a property value. See the <code class="literal">editor-cmd</code>
section in <a class="xref" href="#svn.advanced.confarea.opts.config" title="Config">the section called “Config”</a>
for ways to specify a default editor.</p></dd><dt><a id="svn.ref.svn.sw.encoding"></a><span class="term"><code class="option">--encoding</code> <em class="replaceable"><code>ENC</code></em></span></dt><dd><p>Tells Subversion that your commit message is
composed using the character encoding provided. The
default character encoding is derived from your
operating system's native locale; use this option if
your commit message is composed using any other
encoding.</p></dd><dt><a id="svn.ref.svn.sw.extensions"></a><span class="term"><code class="option">--extensions</code> (<code class="option">-x</code>) <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>Specifies customizations which Subversion should
make when performing difference calculations. Valid
extensions include:</p><div class="variablelist"><dl><dt><span class="term"><code class="option">--ignore-space-change</code> (<code class="option">-b</code>)</span></dt><dd><p>Ignore changes in the amount of white space.</p></dd><dt><span class="term"><code class="option">--ignore-all-space</code> (<code class="option">-w</code>)</span></dt><dd><p>Ignore all white space.</p></dd><dt><span class="term"><code class="option">--ignore-eol-style</code></span></dt><dd><p>Ignore changes in EOL (end-of-line) style.</p></dd><dt><span class="term"><code class="option">--show-c-function</code> (<code class="option">-p</code>)</span></dt><dd><p>Show C function names in the diff output.</p></dd><dt><span class="term"><code class="option">--unified</code> (<code class="option">-u</code>)</span></dt><dd><p>Show three lines of unified diff context.</p></dd></dl></div><p>The default value of <em class="replaceable"><code>ARG</code></em>
is <code class="literal">-u</code>. If you wish to pass multiple
arguments, you must enclose all of them in quotes.</p><p>Note that when Subversion is configured to invoke an
external diff command, the value of
the <code class="option">--extension</code> (<code class="option">-x</code>)
option isn't restricted to the previously mentioned
options, but may be <span class="emphasis"><em>any</em></span> additional
arguments which Subversion should pass to that command.</p></dd><dt><a id="svn.ref.svn.sw.file"></a><span class="term"><code class="option">--file</code> (<code class="option">-F</code>) <em class="replaceable"><code>FILENAME</code></em></span></dt><dd><p>Uses the contents of the named file for the
specified subcommand. Different subcommands do
different things with this content. For example,
<span class="command"><strong>svn commit</strong></span> uses the content as a
commit log message, whereas <span class="command"><strong>svn
propset</strong></span> uses it as a property value.</p></dd><dt><a id="svn.ref.svn.sw.force"></a><span class="term"><code class="option">--force</code></span></dt><dd><p>Forces a particular command or operation to run.
Subversion will prevent you from performing some
operations in normal usage, but you can pass this option
to tell Subversion <span class="quote">“<span class="quote">I know what I'm doing as well
as the possible repercussions of doing it, so let me at
'em.</span>”</span> This option is the programmatic equivalent
of doing your own electrical work with the power
on—if you don't know what you're doing, you're
likely to get a nasty shock.</p></dd><dt><a id="svn.ref.svn.sw.force_log"></a><span class="term"><code class="option">--force-log</code></span></dt><dd><p>Forces a suspicious parameter passed to the
<code class="option">--message</code> (<code class="option">-m</code>) or
<code class="option">--file</code> (<code class="option">-F</code>) option to
be accepted as valid. By default, Subversion will
produce an error if parameters to these options look
like they might instead be targets of the subcommand.
For example, if you pass a versioned file's path to the
<code class="option">--file</code> (<code class="option">-F</code>) option,
Subversion will assume you've made a mistake, that the
path was instead intended as the target of the
operation, and that you simply failed to provide some
other—unversioned—file as the source of your
log message. To assert your intent and override these
types of errors, pass the <code class="option">--force-log</code>
option to subcommands that accept log messages.</p></dd><dt><a id="svn.ref.svn.sw.git"></a><span class="term"><code class="option">--git</code></span></dt><dd><p>Enables a special output mode for <span class="command"><strong>svn
diff</strong></span> designed for cross-compatibility with the
popular Git distributed version control system.</p></dd><dt><a id="svn.ref.svn.sw.help"></a><span class="term"><code class="option">--help</code> (<code class="option">-h</code>, <code class="option">-?</code>)</span></dt><dd><p>If used with one or more subcommands, shows the
built-in help text for each. If used alone, it displays
the general client help text.</p></dd><dt><a id="svn.ref.svn.sw.ignore_ancestry"></a><span class="term"><code class="option">--ignore-ancestry</code></span></dt><dd><p>Tells Subversion to ignore ancestry when calculating
differences (rely on path contents alone). Also disables
<a class="xref" href="#svn.branchmerge.basicmerging.mergetracking" title="Merge Tracking">Merge Tracking</a>
when used with the <span class="command"><strong>svn merge</strong></span> subcommand.
</p></dd><dt><a id="svn.ref.svn.sw.ignore_externals"></a><span class="term"><code class="option">--ignore-externals</code></span></dt><dd><p>Tells Subversion to ignore externals definitions and
the external working copies managed by them.</p></dd><dt><a id="svn.ref.svn.sw.ignore_keywords"></a><span class="term"><code class="option">--ignore-keywords</code></span></dt><dd><p>Disables keyword expansion.</p></dd><dt><a id="svn.ref.svn.sw.ignore_whitespace"></a><span class="term"><code class="option">--ignore-whitespace</code></span></dt><dd><p>Instructs <span class="command"><strong>svn patch</strong></span> to ignore
whitespace when attempting to identify patch
context.</p></dd><dt><a id="svn.ref.svn.sw.incremental"></a><span class="term"><code class="option">--incremental</code></span></dt><dd><p>Prints output in a format suitable for
concatenation to prior similar output.</p></dd><dt><a id="svn.ref.svn.sw.internal_diff"></a><span class="term"><code class="option">--internal-diff</code></span></dt><dd><p>Instructs Subversion to use its built-in
differencing engine despite any external differencing
mechanism that may be specified for use in the user's
runtime configuration.</p></dd><dt><a id="svn.ref.svn.sw.keep_changelists"></a><span class="term"><code class="option">--keep-changelists</code></span></dt><dd><p>Tells Subversion not to remove the changelist assigments
from working copy items after committing.</p></dd><dt><a id="svn.ref.svn.sw.keep_local"></a><span class="term"><code class="option">--keep-local</code></span></dt><dd><p>Keeps the local copy of a file or directory (used
with the <span class="command"><strong>svn delete</strong></span> command).</p></dd><dt><a id="svn.ref.svn.sw.limit"></a><span class="term"><code class="option">--limit</code> (<code class="option">-l</code>) <em class="replaceable"><code>NUM</code></em></span></dt><dd><p>Shows only the first <em class="replaceable"><code>NUM</code></em>
log messages.</p></dd><dt><a id="svn.ref.svn.sw.message"></a><span class="term"><code class="option">--message</code> (<code class="option">-m</code>) <em class="replaceable"><code>MESSAGE</code></em></span></dt><dd><p>Indicates that you will specify either a log message
or a lock comment on the command line, following this
option. For example:</p><div class="informalexample"><pre class="screen">
$ svn commit -m "They don't make Sunday."
</pre></div></dd><dt><a id="svn.ref.svn.sw.native_eol"></a><span class="term"><code class="option">--native-eol</code> <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>Causes <span class="command"><strong>svn export</strong></span> to use a
specific end-of-line sequence as if it was the native
sequence for the client platform.
<em class="replaceable"><code>ARG</code></em> may be one
of <code class="literal">CR</code>, <code class="literal">LF</code>,
or <code class="literal">CRLF</code>.</p></dd><dt><a id="svn.ref.svn.sw.new"></a><span class="term"><code class="option">--new</code> <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>Uses <em class="replaceable"><code>ARG</code></em> as the newer
target (for use with <span class="command"><strong>svn diff</strong></span>).</p></dd><dt><a id="svn.ref.svn.sw.no_auto_props"></a><span class="term"><code class="option">--no-auto-props</code></span></dt><dd><p>Disables automatic property setting, overriding the
<code class="literal">enable-auto-props</code> runtime
configuration directive.</p></dd><dt><a id="svn.ref.svn.sw.no_diff_deleted"></a><span class="term"><code class="option">--no-diff-deleted</code></span></dt><dd><p>Prevents Subversion from printing differences for
deleted files. The default behavior when you remove a
file is for <span class="command"><strong>svn diff</strong></span> to print the
same differences that you would see if you had kept
the file but removed all of its content.</p></dd><dt><a id="svn.ref.svn.sw.no_ignore"></a><span class="term"><code class="option">--no-ignore</code></span></dt><dd><p>Shows files in the status listing that would
normally be omitted since they match a pattern in the
<code class="literal">global-ignores</code> configuration option
or the <code class="literal">svn:ignore</code> property. See <a class="xref" href="#svn.advanced.confarea.opts.config" title="Config">the section called “Config”</a> and <a class="xref" href="#svn.advanced.props.special.ignore" title="Ignoring Unversioned Items">the section called “Ignoring Unversioned Items”</a> for more
information.</p></dd><dt><a id="svn.ref.svn.sw.no_unlock"></a><span class="term"><code class="option">--no-unlock</code></span></dt><dd><p>Tells Subversion not to automatically unlock files.
(The default commit behavior is to unlock all files
listed as part of the commit.) See
<a class="xref" href="#svn.advanced.locking" title="Locking">the section called “Locking”</a> for more
information.</p></dd><dt><a id="svn.ref.svn.sw.non_recursive"></a><span class="term"><code class="option">--non-recursive</code> (<code class="option">-N</code>)</span></dt><dd><p><span class="emphasis"><em>Deprecated</em></span>. Stops a subcommand
from recursing into subdirectories. Most subcommands
recurse by default, but some do not. Users should avoid
this option and use the more precise <code class="option">--depth</code>
option instead. For most subcommands, specifying
<code class="option">--non-recursive</code> produces behavior which
is the same as if you'd specified
<code class="option">--depth=files</code>, but there are exceptions:
non-recursive <span class="command"><strong>svn status</strong></span> operates at the
<code class="literal">immediates</code> depth, and the non-recursive
forms of <span class="command"><strong>svn revert</strong></span>,
<span class="command"><strong>svn add</strong></span>, and <span class="command"><strong>svn commit</strong></span>
operate at an <code class="literal">empty</code> depth.</p></dd><dt><a id="svn.ref.svn.sw.notice_ancestry"></a><span class="term"><code class="option">--notice-ancestry</code></span></dt><dd><p>Pays attention to ancestry when calculating
differences.</p></dd><dt><a id="svn.ref.svn.sw.old"></a><span class="term"><code class="option">--old</code> <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>Uses <em class="replaceable"><code>ARG</code></em> as the older
target (for use with <span class="command"><strong>svn diff</strong></span>).</p></dd><dt><a id="svn.ref.svn.sw.parents"></a><span class="term"><code class="option">--parents</code></span></dt><dd><p>Creates and adds nonexistent or nonversioned parent
subdirectories to the working copy or repository as part
of an operation. This is useful for automatically
creating multiple subdirectories where none currently
exist. If performed on a URL, all the directories will
be created in a single commit.</p></dd><dt><a id="svn.ref.svn.sw.quiet"></a><span class="term"><code class="option">--quiet</code> (<code class="option">-q</code>)</span></dt><dd><p>Requests that the client print only essential
information while performing an operation.</p></dd><dt><a id="svn.ref.svn.sw.record_only"></a><span class="term"><code class="option">--record-only</code></span></dt><dd><p>Enables a special mode of <span class="command"><strong>svn
merge</strong></span> in which the specified merge operation is
recorded in the local merge tracking information, but is
not actually performed.</p></dd><dt><a id="svn.ref.svn.sw.recursive"></a><span class="term"><code class="option">--recursive</code> (<code class="option">-R</code>)</span></dt><dd><p>Makes a subcommand recurse into subdirectories.
(Most subcommands recurse by default.)</p></dd><dt><a id="svn.ref.svn.sw.reintegrate"></a><span class="term"><code class="option">--reintegrate</code></span></dt><dd><p>Used with the <span class="command"><strong>svn merge</strong></span>
subcommand to merge all of the source URL's changes into
the working copy. See
<a class="xref" href="#svn.branchemerge.basicmerging.stayinsync" title="Keeping a Branch in Sync">the section called “Keeping a Branch in Sync”</a>
for details.</p></dd><dt><a id="svn.ref.svn.sw.relocate"></a><span class="term"><code class="option">--relocate</code></span></dt><dd><p>Deprecated. When used with the <span class="command"><strong>svn
switch</strong></span> subcommand, changes the location of the
repository that your working copy references. The
preferred approach as of Subversion 1.7, however, is to
use the <span class="command"><strong>svn relocate</strong></span> subcommand. See
<a class="xref" href="#svn.ref.svn.c.relocate" title="svn relocate">svn relocate</a> for more
details and an example.</p></dd><dt><a id="svn.ref.svn.sw.remove"></a><span class="term"><code class="option">--remove</code></span></dt><dd><p>Used with <span class="command"><strong>svn changelist</strong></span> to
disassociate—rather than associate (which is the
default operation)—the target(s) from a
changelist.</p></dd><dt><a id="svn.ref.svn.sw.reverse_diff"></a><span class="term"><code class="option">--reverse-diff</code></span></dt><dd><p>Causes <span class="command"><strong>svn patch</strong></span> to interpret the
input patch instructions in reverse—treating added
lines as removed ones and vice-versa.</p></dd><dt><a id="svn.ref.svn.sw.revision"></a><span class="term"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></span></dt><dd><p>Specifies a revision (or range of revisions) on with
which to operate. You can provide revision numbers,
keywords, or dates (in curly braces) as arguments to the
revision option. If you wish to offer a range of
revisions, you can provide two revisions separated by a
colon. For example:</p><div class="informalexample"><pre class="screen">
$ svn log -r 1729
$ svn log -r 1729:HEAD
$ svn log -r 1729:1744
$ svn log -r {2001-12-04}:{2002-02-17}
$ svn log -r 1729:{2002-02-17}
</pre></div><p>See <a class="xref" href="#svn.tour.revs.keywords" title="Revision Keywords">the section called “Revision Keywords”</a> for more
information.</p></dd><dt><a id="svn.ref.svn.sw.revprop"></a><span class="term"><code class="option">--revprop</code></span></dt><dd><p>Operates on a revision property instead of a
property specific to a file or directory. This option
requires that you also pass a revision with the
<code class="option">--revision</code> (<code class="option">-r</code>)
option.</p></dd><dt><a id="svn.ref.svn.sw.set_depth"></a><span class="term"><code class="option">--set-depth</code> <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>Sets the sticky depth on a directory in a working
copy to one of <code class="literal">exclude</code>, <code class="literal">empty</code>,
<code class="literal">files</code>, <code class="literal">immediates</code>,
or <code class="literal">infinity</code>. For detailed coverage
of what these mean and how to use this option, see
<a class="xref" href="#svn.advanced.sparsedirs" title="Sparse Directories">the section called “Sparse Directories”</a>.</p></dd><dt><a id="svn.ref.svn.sw.show_copies_as_adds"></a><span class="term"><code class="option">--show-copies-as-adds</code></span></dt><dd><p>Enables a special output mode for <span class="command"><strong>svn
diff</strong></span> in which the content difference for a
file created via a copy operation appears as it would
for a brand new file (with each line therein appearing
as an addition to an empty file) rather than as a delta
against the original file from which the copy was
created.</p></dd><dt><a id="svn.ref.svn.sw.show_revs"></a><span class="term"><code class="option">--show-revs</code> <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>Used to make <span class="command"><strong>svn mergeinfo</strong></span>
display certain classes of merge tracking
information. <em class="replaceable"><code>ARG</code></em> may be
either <code class="literal">merged</code>
or <code class="literal">eligible</code>, indicating a desire to
see revisions either already merged or eligible for future
merge from the specified source URL, respectively.</p></dd><dt><a id="svn.ref.svn.sw.show_updates"></a><span class="term"><code class="option">--show-updates</code> (<code class="option">-u</code>)</span></dt><dd><p>Causes the client to display information about
which files in your working copy are out of date.
This doesn't actually update any of your
files—it just shows you which files will be
updated if you then use <span class="command"><strong>svn update</strong></span>.</p></dd><dt><a id="svn.ref.svn.sw.stop_on_copy"></a><span class="term"><code class="option">--stop-on-copy</code></span></dt><dd><p>Causes a Subversion subcommand that traverses
the history of a versioned resource to stop harvesting
that historical information when a copy—that is, a
location in history where that resource was copied from
another location in the repository—is
encountered.</p></dd><dt><a id="svn.ref.svn.sw.strict"></a><span class="term"><code class="option">--strict</code></span></dt><dd><p>Causes Subversion to use strict semantics, a notion
that is rather vague unless talking about specific
subcommands (namely, <span class="command"><strong>svn propget</strong></span>).</p></dd><dt><a id="svn.ref.svn.sw.strip"></a><span class="term"><code class="option">--strip</code> <em class="replaceable"><code>NUM</code></em></span></dt><dd><p>Used by <span class="command"><strong>svn patch</strong></span> to
ignore <em class="replaceable"><code>NUM</code></em> leading path
components found on paths specified in the patch input
file.</p></dd><dt><a id="svn.ref.svn.sw.summarize"></a><span class="term"><code class="option">--summarize</code></span></dt><dd><p>Display only high-level summary notifications about
the operation instead of its detailed output.</p></dd><dt><a id="svn.ref.svn.sw.targets"></a><span class="term"><code class="option">--targets</code> <em class="replaceable"><code>FILENAME</code></em></span></dt><dd><p>Tells Subversion to read additional target paths for
the operation from <em class="replaceable"><code>FILENAME</code></em>.
<em class="replaceable"><code>FILENAME</code></em> should contain one
path per line, with each path expected to use the same
encoding and formatting that it would if you had
specified it directly as an argument on the command
line.</p></dd><dt><a id="svn.ref.svn.sw.use_merge_history"></a><span class="term"><code class="option">--use-merge-history</code> (<code class="option">-g</code>)</span></dt><dd><p>Uses or displays additional information from merge
history.</p></dd><dt><a id="svn.ref.svn.sw.verbose"></a><span class="term"><code class="option">--verbose</code> (<code class="option">-v</code>)</span></dt><dd><p>Requests that the client print out as much
information as it can while running any subcommand.
This may result in Subversion printing out additional
fields, detailed information about every file, or
additional information regarding its actions.</p></dd><dt><a id="svn.ref.svn.sw.version"></a><span class="term"><code class="option">--version</code></span></dt><dd><p>Prints the client version info. This information
includes not only the version number of the client, but
also a listing of all repository access modules that the
client can use to access a Subversion repository.
With <code class="option">--quiet</code> (<code class="option">-q</code>) it
prints only the version number in a compact form.</p></dd><dt><a id="svn.ref.svn.sw.with_all_revprops"></a><span class="term"><code class="option">--with-all-revprops</code></span></dt><dd><p>Used with the <code class="option">--xml</code> option
to <span class="command"><strong>svn log</strong></span>, instructs Subversion to
retrieve and display all revision properties—the
standard ones used internally by Subversion as well as any
user-defined ones—in the log output.</p></dd><dt><a id="svn.ref.svn.sw.with_no_revprops"></a><span class="term"><code class="option">--with-no-revprops</code></span></dt><dd><p>Used with the <code class="option">--xml</code> option
to <span class="command"><strong>svn log</strong></span>, instructs Subversion to
omit all revision properties—including the standard
log message, author, and revision datestamp—from the
log output.</p></dd><dt><a id="svn.ref.svn.sw.with_revprop"></a><span class="term"><code class="option">--with-revprop</code> <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>When used with any command that writes to the
repository, sets the revision property, using the
<em class="replaceable"><code>NAME=VALUE</code></em> format,
<em class="replaceable"><code>NAME</code></em> to
<em class="replaceable"><code>VALUE</code></em>. When used with
<span class="command"><strong>svn log</strong></span> in <code class="option">--xml</code> mode, this displays the value of
<em class="replaceable"><code>ARG</code></em> in the log output.</p></dd><dt><a id="svn.ref.svn.sw.xml"></a><span class="term"><code class="option">--xml</code></span></dt><dd><p>Prints output in XML format.</p></dd></dl></div></div><div class="sect2" title="svn Subcommands"><div class="titlepage"><div><div><h3 class="title"><a id="svn.ref.svn.c"></a>svn Subcommands</h3></div></div></div><p>Here are the various subcommands for
the <span class="command"><strong>svn</strong></span> program. For the sake of brevity,
we omit the global options (described in
<a class="xref" href="#svn.ref.svn.sw" title="svn Options">the section called “svn Options”</a>) from the subcommand
descriptions which follow.</p><div class="refentry" title="svn add"><a id="svn.ref.svn.c.add"></a><div class="titlepage"></div><a id="idp16778656" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn add — Add files, directories, or symbolic links.</p></div><div class="refsect1" title="Synopsis"><a id="idp16781712"></a><h2>Synopsis</h2><p><code class="literal">svn add PATH...</code></p></div><div class="refsect1" title="Description"><a id="idp16783328"></a><h2>Description</h2><p>Schedule files, directories, or symbolic links in your
working copy for addition to the repository. They will be
uploaded and added to the repository on your next commit.
If you add something and change your mind before
committing, you can unschedule the addition using
<span class="command"><strong>svn revert</strong></span>.</p></div><div class="refsect1" title="Options"><a id="idp16785552"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.auto_props"><code class="option">--auto-props</code></a>
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.force"><code class="option">--force</code></a>
<a class="xref" href="#svn.ref.svn.sw.no_auto_props"><code class="option">--no-auto-props</code></a>
<a class="xref" href="#svn.ref.svn.sw.no_ignore"><code class="option">--no-ignore</code></a>
<a class="xref" href="#svn.ref.svn.sw.parents"><code class="option">--parents</code></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.targets"><code class="option">--targets</code> <em class="replaceable"><code>FILENAME</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp16793344"></a><h2>Examples</h2><p>To add a file to your working copy:</p><div class="informalexample"><pre class="screen">
$ svn add foo.c
A foo.c
</pre></div><p>When adding a directory, the default behavior of
<span class="command"><strong>svn add</strong></span> is to recurse:</p><div class="informalexample"><pre class="screen">
$ svn add testdir
A testdir
A testdir/a
A testdir/b
A testdir/c
A testdir/d
</pre></div><p>You can add a directory without adding its
contents:</p><div class="informalexample"><pre class="screen">
$ svn add --depth=empty otherdir
A otherdir
</pre></div><p>Attempts to schedule the addition of an item which is
already versioned will fail by default. This behavior
foils the most common scenario under which users attempt
this: when trying to get to Subversion to recursively
examine a versioned directory and add any unversioned
items inside of it. To override the default behavior and
force Subversion to recurse into already-versioned
directories, pass the <code class="option">--force</code>
option:</p><div class="informalexample"><pre class="screen">
$ svn add versioned-dir
svn: warning: W150002: '/home/cmpilato/projects/subversion/site' is already un\
der version control
$ svn add versioned-dir --force
A versioned-dir/foo.c
A versioned-dir/somedir/bar.c
A (bin) versioned-dir/otherdir/docs/baz.doc
…
</pre></div></div></div><div class="refentry" title="svn blame (praise, annotate, ann)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.blame"></a><div class="titlepage"></div><a id="idp16803440" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn blame (praise, annotate, ann) — Show author and revision information inline
for the specified files or URLs.</p></div><div class="refsect1" title="Synopsis"><a id="idp16806592"></a><h2>Synopsis</h2><p><code class="literal">svn blame TARGET[@REV]...</code></p></div><div class="refsect1" title="Description"><a id="idp16808224"></a><h2>Description</h2><p>Show author and revision information inline for the
specified files or URLs. Each line of text is annotated
at the beginning with the author (username) and the
revision number for the last change to that line.</p></div><div class="refsect1" title="Options"><a id="idp16809664"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.extensions"><code class="option">--extensions</code> (<code class="option">-x</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.force"><code class="option">--force</code></a>
<a class="xref" href="#svn.ref.svn.sw.incremental"><code class="option">--incremental</code></a>
<a class="xref" href="#svn.ref.svn.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.use_merge_history"><code class="option">--use-merge-history</code> (<code class="option">-g</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.verbose"><code class="option">--verbose</code> (<code class="option">-v</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.xml"><code class="option">--xml</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp16816624"></a><h2>Examples</h2><p>If you want to see blame-annotated source for
<code class="filename">readme.txt</code> in your test
repository:</p><div class="informalexample"><pre class="screen">
$ svn blame http://svn.red-bean.com/repos/test/readme.txt
3 sally This is a README file.
5 harry Don't bother reading it. The boss is a knucklehead.
3 sally
…
</pre></div><p>Now, just because <span class="command"><strong>svn blame</strong></span> says
that Harry last modified <code class="filename">readme.txt</code>
in revision 5, understand that this subcommand is by
default very picky about what constitutes a change.
Before clubbing Harry over the head for what appears to be
insubordination, first consider that perhaps the change he
made to the file might have been only to its specific
character content, not to its overall semantic meaning.
Perhaps his changes were the result of blindly running a
whitespace cleanup script on this file. You might need
to examine the specific differences and related log
message to understand exactly what Harry did to this file
in revision 5.</p><div class="informalexample"><pre class="screen">
$ svn log -c 5 http://svn.red-bean.com/repos/test/readme.txt
------------------------------------------------------------------------
r5 | harry | 2008-05-29 07:26:12 -0600 (Thu, 29 May 2008) | 1 line
Commit the results of 'double-space-after-period.sh'.
------------------------------------------------------------------------
$ svn diff -c 5 http://svn.red-bean.com/repos/test/readme.txt
Index: http://svn.red-bean.com/repos/test/readme.txt
===================================================================
--- http://svn.red-bean.com/repos/test/readme.txt (revision 4)
+++ http://svn.red-bean.com/repos/test/readme.txt (revision 5)
@@ -1,5 +1,5 @@
This is a README file.
-Don't bother reading it. The boss is a knucklehead.
+Don't bother reading it. The boss is a knucklehead.
INSTRUCTIONS
============
$
</pre></div><p>Sure enough, Harry only changed the whitespace in that
line. Fortunately, the <code class="option">--extensions</code>
(<code class="option">-x</code>) option can help you better determine
the last time that a <span class="emphasis"><em>meaningful</em></span>
change was made to a given line of text. For example,
here's how you can see the annotation information while
disregarding mere whitespace changes:</p><div class="informalexample"><pre class="screen">
$ svn blame -x -b http://svn.red-bean.com/repos/test/readme.txt
3 sally This is a README file.
4 jess Don't bother reading it. The boss is a knucklehead.
3 sally
…
</pre></div><p>If you use the <code class="option">--xml</code> option, you can
get XML output describing the blame annotations, but not
the contents of the lines themselves:</p><div class="informalexample"><pre class="screen">
$ svn blame --xml http://svn.red-bean.com/repos/test/readme.txt
<?xml version="1.0"?>
<blame>
<target
path="readme.txt">
<entry
line-number="1">
<commit
revision="3">
<author>sally</author>
<date>2008-05-25T19:12:31.428953Z</date>
</commit>
</entry>
<entry
line-number="2">
<commit
revision="5">
<author>harry</author>
<date>2008-05-29T13:26:12.293121Z</date>
</commit>
</entry>
<entry
line-number="3">
…
</entry>
</target>
</blame>
$
</pre></div></div></div><div class="refentry" title="svn cat"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.cat"></a><div class="titlepage"></div><a id="idp16831664" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn cat — Output the contents of the specified files or
URLs.</p></div><div class="refsect1" title="Synopsis"><a id="idp16834784"></a><h2>Synopsis</h2><p><code class="literal">svn cat TARGET[@REV]...</code></p></div><div class="refsect1" title="Description"><a id="idp16836400"></a><h2>Description</h2><p>Output the contents of the specified files or URLs.
For listing the contents of directories, see <span class="command"><strong>svn
list</strong></span> later in this chapter.</p></div><div class="refsect1" title="Options"><a id="idp16838432"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp16840944"></a><h2>Examples</h2><p>If you want to view <code class="filename">readme.txt</code> in
your repository without checking it out:</p><div class="informalexample"><pre class="screen">
$ svn cat http://svn.red-bean.com/repos/test/readme.txt
This is a README file.
Don't bother reading it. The boss is a knucklehead.
INSTRUCTIONS
============
Step 1: Do this.
Step 2: Do that.
$
</pre></div><p>You can view specific versions of files, too.</p><div class="informalexample"><pre class="screen">
$ svn cat -r 3 http://svn.red-bean.com/repos/test/readme.txt
This is a README file.
INSTRUCTIONS
============
Step 1: Do this.
Step 2: Do that.
$
</pre></div><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>You might develop a reflex action of
using <span class="command"><strong>svn cat</strong></span> to view your working
file contents. But keep in mind that the default peg
revision for <span class="command"><strong>svn cat</strong></span> when used on a
working copy file target is <code class="literal">BASE</code>, the
unmodified base revision of that file. Don't be
surprised when a simple <strong class="userinput"><code>svn cat
/path/to/file</code></strong> invocation fails to display
your local modifications to that file!</p></td></tr></table></div><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>If your working copy is out of date (or you have
local modifications) and you want to see the
<code class="literal">HEAD</code> revision of a file in your
working copy, use the <code class="option">--revision</code>
(<code class="option">-r</code>) option: <strong class="userinput"><code>svn cat -r
HEAD <em class="replaceable"><code>FILENAME</code></em></code></strong></p></td></tr></table></div></div></div><div class="refentry" title="svn changelist (cl)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.changelist"></a><div class="titlepage"></div><a id="idp16854544" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn changelist (cl) — Associate (or deassociate) local paths with a
changelist.</p></div><div class="refsect1" title="Synopsis"><a id="idp16857680"></a><h2>Synopsis</h2><p><code class="literal">changelist CLNAME TARGET...</code></p><p><code class="literal">changelist --remove TARGET...</code></p></div><div class="refsect1" title="Description"><a id="idp16860096"></a><h2>Description</h2><p>Used for dividing files in a working copy into a
changelist (logical named grouping) in order to allow
users to easily work on multiple file collections within a
single working copy.</p></div><div class="refsect1" title="Options"><a id="idp16861520"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.changelist"><code class="option">--changelist</code> (<code class="option">--cl</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.recursive"><code class="option">--recursive</code> (<code class="option">-R</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.remove"><code class="option">--remove</code></a>
<a class="xref" href="#svn.ref.svn.sw.targets"><code class="option">--targets</code> <em class="replaceable"><code>FILENAME</code></em></a>
</pre></div></div><div class="refsect1" title="Example"><a id="idp16867552"></a><h2>Example</h2><p>Edit three files, add them to a changelist, then
commit only files in that changelist:</p><div class="informalexample"><pre class="screen">
$ svn changelist issue1729 foo.c bar.c baz.c
A [issue1729] foo.c
A [issue1729] bar.c
A [issue1729] baz.c
$ svn status
A someotherfile.c
A test/sometest.c
--- Changelist 'issue1729':
A foo.c
A bar.c
A baz.c
$ svn commit --changelist issue1729 -m "Fixing Issue 1729."
Adding bar.c
Adding baz.c
Adding foo.c
Transmitting file data ...
Committed revision 2.
$ svn status
A someotherfile.c
A test/sometest.c
$
</pre></div><p>Note that in the previous example, only the files in
changelist <code class="literal">issue1729</code> were
committed.</p></div></div><div class="refentry" title="svn checkout (co)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.checkout"></a><div class="titlepage"></div><a id="idp16872784" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn checkout (co) — Check out a working copy from a repository.</p></div><div class="refsect1" title="Synopsis"><a id="idp16875872"></a><h2>Synopsis</h2><p><code class="literal">svn checkout URL[@REV]... [PATH]</code></p></div><div class="refsect1" title="Description"><a id="idp16877504"></a><h2>Description</h2><p>Check out a working copy from a repository. If
<em class="replaceable"><code>PATH</code></em> is omitted, the
basename of the URL will be used as the destination.
If multiple URLs are given, each will be checked out into a
subdirectory of <em class="replaceable"><code>PATH</code></em>, with the
name of the subdirectory being the basename of the
URL.</p></div><div class="refsect1" title="Options"><a id="idp16879904"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.force"><code class="option">--force</code></a>
<a class="xref" href="#svn.ref.svn.sw.ignore_externals"><code class="option">--ignore-externals</code></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp16885136"></a><h2>Examples</h2><p>Check out a working copy into a directory called
<code class="filename">mine</code>:</p><div class="informalexample"><pre class="screen">
$ svn checkout file:///var/svn/repos/test mine
A mine/a
A mine/b
A mine/c
A mine/d
Checked out revision 20.
$ ls
mine
$
</pre></div><p>Check out two different directories into two separate
working copies:</p><div class="informalexample"><pre class="screen">
$ svn checkout file:///var/svn/repos/test \
file:///var/svn/repos/quiz
A test/a
A test/b
A test/c
A test/d
Checked out revision 20.
A quiz/l
A quiz/m
Checked out revision 13.
$ ls
quiz test
$
</pre></div><p>Check out two different directories into two separate
working copies, but place both into a directory called
<code class="filename">working-copies</code>:</p><div class="informalexample"><pre class="screen">
$ svn checkout file:///var/svn/repos/test \
file:///var/svn/repos/quiz \
working-copies
A working-copies/test/a
A working-copies/test/b
A working-copies/test/c
A working-copies/test/d
Checked out revision 20.
A working-copies/quiz/l
A working-copies/quiz/m
Checked out revision 13.
$ ls
working-copies
</pre></div><p>If you interrupt a checkout (or something else
interrupts your checkout, such as loss of connectivity,
etc.), you can restart it either by issuing the identical
checkout command again or by updating the incomplete
working copy:</p><div class="informalexample"><pre class="screen">
$ svn checkout file:///var/svn/repos/test mine
A mine/a
A mine/b
^C
svn: E200015: Caught signal
$ svn checkout file:///var/svn/repos/test mine
A mine/c
^C
svn: E200015: Caught signal
$ svn update mine
Updating 'mine':
A mine/d
Updated to revision 20.
$
</pre></div><p>If you wish to check out some revision other than the
most recent one, you can do so by providing the
<code class="option">--revision</code> (<code class="option">-r</code>) option
to the <span class="command"><strong>svn checkout</strong></span> command:</p><div class="informalexample"><pre class="screen">
$ svn checkout -r 2 file:///var/svn/repos/test mine
A mine/a
Checked out revision 2.
$
</pre></div><p>Prior to version 1.7, Subversion would complain by
default if you try to check out a directory atop an
existing directory which contains files or subdirectories
that the checkout itself would have created. Subversion
1.7 handles this situation differently, allowing the
checkout to proceed but marking any obstructing objects as
tree conflicts. Use the <code class="option">--force</code> option
to override this safeguard. When you check out with
the <code class="option">--force</code> option, any unversioned file
in the checkout target tree which ordinarily would
obstruct the checkout will still become versioned, but
Subversion will preserve its contents as-is. If those
contents differ from the repository file at that path
(which was downloaded as part of the checkout), the file
will appear to have local modifications—the changes
required to transform the versioned file you checked out
into the unversioned file you had before checking
out—when the checkout completes.</p><div class="informalexample"><pre class="screen">
$ mkdir project
$ mkdir project/lib
$ touch project/lib/file.c
$ svn checkout file:///var/svn/repos/project/trunk project --force
E project/lib
A project/lib/subdir
E project/lib/file.c
A project/lib/anotherfile.c
A project/include/header.h
Checked out revision 21.
$ svn status wc
M project/lib/file.c
$ svn diff wc
Index: project/lib/file.c
===================================================================
--- project/lib/file.c (revision 1)
+++ project/lib/file.c (working copy)
@@ -3 +0,0 @@
-/* file.c: Code for acting file-ishly. */
-#include <stdio.h>
-/* Not feeling particularly creative today. */
$
</pre></div><p>As in another other working copy, you have the choices
typically available: reverting some or all of those
local <span class="quote">“<span class="quote">modifications</span>”</span>, committing them, or
continuing to modify your working copy.</p><p>This feature is especially useful for performing
in-place imports of unversioned directory trees. By first
importing the tree into the repository, and then checking
out new repository location atop the unversioned tree with
the <code class="option">--force</code> option, you effectively
transform the unversioned tree into a working
copy.</p><div class="informalexample"><pre class="screen">
$ svn mkdir -m "Create newproject project root." \
file://var/svn/repos/newproject
$ svn import -m "Import initial newproject codebase." newproject \
file://var/svn/repos/newproject/trunk
Adding newproject/include
Adding newproject/include/newproject.h
Adding newproject/lib
Adding newproject/lib/helpers.c
Adding newproject/lib/base.c
Adding newproject/notes
Adding newproject/notes/README
Committed revision 22.
$ svn checkout file://`pwd`/repos-1.6/newproject/trunk newproject --force
E newproject/include
E newproject/include/newproject.h
E newproject/lib
E newproject/lib/helpers.c
E newproject/lib/base.c
E newproject/notes
E newproject/notes/README
Checked out revision 2.
$ svn status newproject
$
</pre></div></div></div><div class="refentry" title="svn cleanup"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.cleanup"></a><div class="titlepage"></div><a id="idp16417776" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn cleanup — Recursively clean up the working copy</p></div><div class="refsect1" title="Synopsis"><a id="idp16919120"></a><h2>Synopsis</h2><p><code class="literal">svn cleanup [PATH...]</code></p></div><div class="refsect1" title="Description"><a id="idp16920704"></a><h2>Description</h2><p>Recursively clean up the working copy, removing
working copy locks and resuming unfinished operations. If
you ever get a <code class="literal">working copy locked</code>
error, run this command to remove stale locks and get your
working copy into a usable state again.</p><p>If, for some reason, an <span class="command"><strong>svn update</strong></span>
fails due to a problem running an external diff program
(e.g., user input or network failure), pass the
<code class="option">--diff3-cmd</code> to allow the cleanup process
to complete any required merging using your external diff
program. You can also specify any configuration directory
with the <code class="option">--config-dir</code> option, but you
should need these options extremely infrequently.</p></div><div class="refsect1" title="Options"><a id="idp16925296"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.diff3_cmd"><code class="option">--diff3-cmd</code> <em class="replaceable"><code>CMD</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp16927984"></a><h2>Examples</h2><p>Well, there's not much to the examples here, as
<span class="command"><strong>svn cleanup</strong></span> generates no output. If
you pass no <em class="replaceable"><code>PATH</code></em>, then
<span class="quote">“<span class="quote"><code class="filename">.</code></span>”</span> is used:</p><div class="informalexample"><pre class="screen">
$ svn cleanup
$ svn cleanup /var/svn/working-copy
</pre></div></div></div><div class="refentry" title="svn commit (ci)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.commit"></a><div class="titlepage"></div><a id="idp16933584" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn commit (ci) — Send changes from your working copy to the repository.</p></div><div class="refsect1" title="Synopsis"><a id="idp16936672"></a><h2>Synopsis</h2><p><code class="literal">svn commit [PATH...]</code></p></div><div class="refsect1" title="Description"><a id="idp16938288"></a><h2>Description</h2><p>Send changes from your working copy to the repository.
If you do not supply a log message with your commit by
using either the <code class="option">--file</code>
(<code class="option">-F</code>) or <code class="option">--message</code>
(<code class="option">-m</code>) option,
<span class="command"><strong>svn</strong></span> will launch your editor for you
to compose a commit message. See the
<code class="literal">editor-cmd</code> list entry in <a class="xref" href="#svn.advanced.confarea.opts.config" title="Config">the section called “Config”</a>.</p><p><span class="command"><strong>svn commit</strong></span> will send any lock
tokens that it finds and will release locks on all
<em class="replaceable"><code>PATH</code></em>s committed (recursively)
unless <code class="option">--no-unlock</code> is passed.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>If you begin a commit and Subversion launches your
editor to compose the commit message, you can still
abort without committing your changes. If you want to
cancel your commit, just quit your editor without saving
your commit message and Subversion will prompt you to
either abort the commit, continue with no message, or
edit the message again.</p></td></tr></table></div></div><div class="refsect1" title="Options"><a id="idp16946768"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.changelist"><code class="option">--changelist</code> (<code class="option">--cl</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.editor_cmd"><code class="option">--editor-cmd</code> <em class="replaceable"><code>CMD</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.encoding"><code class="option">--encoding</code> <em class="replaceable"><code>ENC</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.file"><code class="option">--file</code> (<code class="option">-F</code>) <em class="replaceable"><code>FILENAME</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.force_log"><code class="option">--force-log</code></a>
<a class="xref" href="#svn.ref.svn.sw.keep_changelists"><code class="option">--keep-changelists</code></a>
<a class="xref" href="#svn.ref.svn.sw.message"><code class="option">--message</code> (<code class="option">-m</code>) <em class="replaceable"><code>MESSAGE</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.no_unlock"><code class="option">--no-unlock</code></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.targets"><code class="option">--targets</code> <em class="replaceable"><code>FILENAME</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.with_revprop"><code class="option">--with-revprop</code> <em class="replaceable"><code>ARG</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp16957248"></a><h2>Examples</h2><p>Commit a simple modification to a file with the
commit message on the command line and an implicit
target of your current directory
(<span class="quote">“<span class="quote"><code class="filename">.</code></span>”</span>):</p><div class="informalexample"><pre class="screen">
$ svn commit -m "added howto section."
Sending a
Transmitting file data .
Committed revision 3.
</pre></div><p>Commit a modification to the file
<code class="filename">foo.c</code> (explicitly specified on the
command line) with the commit message in a file named
<code class="filename">msg</code>:</p><div class="informalexample"><pre class="screen">
$ svn commit -F msg foo.c
Sending foo.c
Transmitting file data .
Committed revision 5.
</pre></div><p>If you want to use a file that's under version control
for your commit message with <code class="option">--file</code>
(<code class="option">-F</code>), you need to pass the
<code class="option">--force-log</code> option:</p><div class="informalexample"><pre class="screen">
$ svn commit -F file_under_vc.txt foo.c
svn: E205004: Log message file is a versioned file; use '--force-log' to override
$ svn commit --force-log -F file_under_vc.txt foo.c
Sending foo.c
Transmitting file data .
Committed revision 6.
</pre></div><p>To commit a file scheduled for deletion:</p><div class="informalexample"><pre class="screen">
$ svn commit -m "removed file 'c'."
Deleting c
Committed revision 7.
</pre></div></div></div><div class="refentry" title="svn copy (cp)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.copy"></a><div class="titlepage"></div><a id="idp16969616" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn copy (cp) — Copy a file or directory in a working copy or
in the repository.</p></div><div class="refsect1" title="Synopsis"><a id="idp16972736"></a><h2>Synopsis</h2><p><code class="literal">svn copy SRC[@REV]... DST</code></p></div><div class="refsect1" title="Description"><a id="idp16974368"></a><h2>Description</h2><p>Copy one or more files in a working copy or in the
repository. <em class="replaceable"><code>SRC</code></em> and
<em class="replaceable"><code>DST</code></em> can each be either a
working copy (WC) path or URL. When copying multiple
sources, add the copies as immediate children
of <em class="replaceable"><code>DST</code></em> (which, of course, must
be a directory).</p><div class="variablelist"><dl><dt><span class="term">WC → WC</span></dt><dd><p>Copy and schedule an item for
addition (with history).</p></dd><dt><span class="term">WC → URL</span></dt><dd><p>Immediately commit a copy of WC to URL.</p></dd><dt><span class="term">URL → WC</span></dt><dd><p>Check out URL into WC and schedule it for
addition.</p></dd><dt><span class="term">URL → URL</span></dt><dd><p>Complete server-side copy. This is
usually used to branch and tag.</p></dd></dl></div><p>If no peg revision (i.e.,
<em class="replaceable"><code>@REV</code></em>) is supplied, by default
the <code class="literal">BASE</code> revision will be used for
files copied from the working copy, while the
<code class="literal">HEAD</code> revision will be used for files
copied from a URL.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>You can only copy files within a single repository.
Subversion does not support cross-repository copying.</p></td></tr></table></div></div><div class="refsect1" title="Options"><a id="idp16987760"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.editor_cmd"><code class="option">--editor-cmd</code> <em class="replaceable"><code>CMD</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.encoding"><code class="option">--encoding</code> <em class="replaceable"><code>ENC</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.file"><code class="option">--file</code> (<code class="option">-F</code>) <em class="replaceable"><code>FILENAME</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.force_log"><code class="option">--force-log</code></a>
<a class="xref" href="#svn.ref.svn.sw.ignore_externals"><code class="option">--ignore-externals</code></a>
<a class="xref" href="#svn.ref.svn.sw.message"><code class="option">--message</code> (<code class="option">-m</code>) <em class="replaceable"><code>MESSAGE</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.parents"><code class="option">--parents</code></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.with_revprop"><code class="option">--with-revprop</code> <em class="replaceable"><code>ARG</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp16996160"></a><h2>Examples</h2><p>Copy an item within your working copy (this
schedules the copy—nothing goes into the repository
until you commit):</p><div class="informalexample"><pre class="screen">
$ svn copy foo.txt bar.txt
A bar.txt
$ svn status
A + bar.txt
</pre></div><p>Copy several files in a working copy into a
subdirectory:</p><div class="informalexample"><pre class="screen">
$ svn copy bat.c baz.c qux.c src
A src/bat.c
A src/baz.c
A src/qux.c
</pre></div><p>Copy revision 8 of <code class="filename">bat.c</code> into your
working copy under a different name:</p><div class="informalexample"><pre class="screen">
$ svn copy -r 8 bat.c ya-old-bat.c
A ya-old-bat.c
</pre></div><p>Copy an item in your working copy to a URL in the
repository (this is an immediate commit, so you must supply a
commit message):</p><div class="informalexample"><pre class="screen">
$ svn copy near.txt file:///var/svn/repos/test/far-away.txt -m "Remote copy."
Committed revision 8.
</pre></div><p>Copy an item from the repository to your working
copy (this just schedules the copy—nothing goes into the
repository until you commit):</p><div class="informalexample"><pre class="screen">
$ svn copy file:///var/svn/repos/test/far-away -r 6 near-here
A near-here
</pre></div><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>This is the recommended way to resurrect a dead
file in your repository!</p></td></tr></table></div><p>And finally, copy between two URLs:</p><div class="informalexample"><pre class="screen">
$ svn copy file:///var/svn/repos/test/far-away \
file:///var/svn/repos/test/over-there -m "remote copy."
Committed revision 9.
</pre></div><div class="informalexample"><pre class="screen">
$ svn copy file:///var/svn/repos/test/trunk \
file:///var/svn/repos/test/tags/0.6.32-prerelease -m "tag tree"
Committed revision 12.
</pre></div><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>This is the easiest way to <span class="quote">“<span class="quote">tag</span>”</span> a
revision in your repository—just <span class="command"><strong>svn
copy</strong></span> that revision (usually
<code class="literal">HEAD</code>) into your <code class="filename">tags</code> directory.</p></td></tr></table></div><p>And don't worry if you forgot to tag—you can
always specify an older revision and tag anytime:</p><div class="informalexample"><pre class="screen">
$ svn copy -r 11 file:///var/svn/repos/test/trunk \
file:///var/svn/repos/test/tags/0.6.32-prerelease \
-m "Forgot to tag at rev 11"
Committed revision 13.
</pre></div></div></div><div class="refentry" title="svn delete (del, remove, rm)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.delete"></a><div class="titlepage"></div><a id="idp17016064" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn delete (del, remove, rm) — Delete an item from a working copy
or the repository.</p></div><div class="refsect1" title="Synopsis"><a id="idp17019136"></a><h2>Synopsis</h2><p><code class="literal">svn delete PATH...</code></p><p><code class="literal">svn delete URL...</code></p></div><div class="refsect1" title="Description"><a id="idp17021568"></a><h2>Description</h2><p>Items specified by <em class="replaceable"><code>PATH</code></em> are
scheduled for deletion upon the next commit. Files (and
directories that have not been committed) are immediately
removed from the working copy unless the
<code class="option">--keep-local</code> option is given. The
command will not remove any unversioned or modified items;
use the <code class="option">--force</code> option to override this
behavior.</p><p>Items specified by URL are deleted from
the repository via an immediate commit. Multiple URLs are
committed atomically.</p></div><div class="refsect1" title="Options"><a id="idp17024992"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.editor_cmd"><code class="option">--editor-cmd</code> <em class="replaceable"><code>CMD</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.encoding"><code class="option">--encoding</code> <em class="replaceable"><code>ENC</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.file"><code class="option">--file</code> (<code class="option">-F</code>) <em class="replaceable"><code>FILENAME</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.force"><code class="option">--force</code></a>
<a class="xref" href="#svn.ref.svn.sw.force_log"><code class="option">--force-log</code></a>
<a class="xref" href="#svn.ref.svn.sw.keep_local"><code class="option">--keep-local</code></a>
<a class="xref" href="#svn.ref.svn.sw.message"><code class="option">--message</code> (<code class="option">-m</code>) <em class="replaceable"><code>MESSAGE</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.targets"><code class="option">--targets</code> <em class="replaceable"><code>FILENAME</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.with_revprop"><code class="option">--with-revprop</code> <em class="replaceable"><code>ARG</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17033536"></a><h2>Examples</h2><p>Using <span class="command"><strong>svn</strong></span> to delete a file from
your working copy deletes your local copy of the file, but
it merely schedules the file to be deleted from the
repository. When you commit, the file is deleted in the
repository.</p><div class="informalexample"><pre class="screen">
$ svn delete myfile
D myfile
$ svn commit -m "Deleted file 'myfile'."
Deleting myfile
Transmitting file data .
Committed revision 14.
</pre></div><p>Deleting a URL, however, is immediate, so you have
to supply a log message:</p><div class="informalexample"><pre class="screen">
$ svn delete -m "Deleting file 'yourfile'" \
file:///var/svn/repos/test/yourfile
Committed revision 15.
</pre></div><p>Here's an example of how to force deletion of a file
that has local mods:</p><div class="informalexample"><pre class="screen">
$ svn delete over-there
svn: E195006: Use --force to override this restriction (local modifications m\
ay be lost)
svn: E195006: '/home/sally/project/over-there' has local modifications -- com\
mit or revert them first
$ svn delete --force over-there
D over-there
$
</pre></div><p>Use the <code class="option">--keep-local</code> option to
override the default <span class="command"><strong>svn delete</strong></span>
behavior of also removing the target file that was
scheduled for versioned deletion. This is helpful when
you realize that you've accidentally committed the
addition of a file that you need to keep around in your
working copy, but which shouldn't have been added to
version control.</p><div class="informalexample"><pre class="screen">
$ svn delete --keep-local conf/program.conf
D conf/program.conf
$ svn commit -m "Remove accidentally-added configuration file."
Deleting conf/program.conf
Transmitting file data .
Committed revision 21.
$ svn status
? conf/program.conf
$
</pre></div></div></div><div class="refentry" title="svn diff (di)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.diff"></a><div class="titlepage"></div><a id="idp17044752" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn diff (di) — This displays the differences between two revisions or paths.</p></div><div class="refsect1" title="Synopsis"><a id="idp17047856"></a><h2>Synopsis</h2><p><code class="literal">diff [-c M | -r N[:M]] [TARGET[@REV]...]</code></p><p><code class="literal">diff [-r N[:M]] --old=OLD-TGT[@OLDREV] [--new=NEW-TGT[@NEWREV]] [PATH...]</code></p><p><code class="literal">diff OLD-URL[@OLDREV] NEW-URL[@NEWREV]</code></p></div><div class="refsect1" title="Description"><a id="idp17051216"></a><h2>Description</h2><p>Display the differences between two paths. You can
use <span class="command"><strong>svn diff</strong></span> in the following ways:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>Use just <span class="command"><strong>svn diff</strong></span> to display local
modifications in a working copy.</p></li><li class="listitem"><p>Display the changes made to
<em class="replaceable"><code>TARGET</code></em>s as they are seen in
<em class="replaceable"><code>REV</code></em> between two revisions.
<em class="replaceable"><code>TARGET</code></em>s may be all working copy
paths or all <em class="replaceable"><code>URL</code></em>s. If
<em class="replaceable"><code>TARGET</code></em>s are working copy paths,
<em class="replaceable"><code>N</code></em> defaults to
<code class="literal">BASE</code> and <em class="replaceable"><code>M</code></em>
to the working copy; if <em class="replaceable"><code>TARGET</code></em>s
are <em class="replaceable"><code>URL</code></em>s,
<em class="replaceable"><code>N</code></em> must be specified and
<em class="replaceable"><code>M</code></em> defaults to
<code class="literal">HEAD</code>. The <code class="literal">-c M</code> option
is equivalent to <code class="literal">-r N:M</code> where <code class="literal">N =
M-1</code>. Using <code class="literal">-c -M</code> does the
reverse: <code class="literal">-r M:N</code> where <code class="literal">N =
M-1</code>.</p></li><li class="listitem"><p>Display the differences between
<em class="replaceable"><code>OLD-TGT</code></em> as it was seen in
<em class="replaceable"><code>OLDREV</code></em> and
<em class="replaceable"><code>NEW-TGT</code></em> as it was seen in
<em class="replaceable"><code>NEWREV</code></em>.
<em class="replaceable"><code>PATH</code></em>s, if given, are relative
to <em class="replaceable"><code>OLD-TGT</code></em> and
<em class="replaceable"><code>NEW-TGT</code></em> and restrict the output
to differences for those paths.
<em class="replaceable"><code>OLD-TGT</code></em> and
<em class="replaceable"><code>NEW-TGT</code></em> may be working copy
paths or <em class="replaceable"><code>URL[@REV]</code></em>.
<em class="replaceable"><code>NEW-TGT</code></em> defaults to
<em class="replaceable"><code>OLD-TGT</code></em> if not specified.
<code class="literal">-r N</code>
makes <em class="replaceable"><code>OLDREV</code></em> default to
<code class="literal">N</code>; <code class="literal">-r N:M</code>
makes <em class="replaceable"><code>OLDREV</code></em> default to
<em class="replaceable"><code>N</code></em> and
<em class="replaceable"><code>NEWREV</code></em> default to
<em class="replaceable"><code>M</code></em>.</p></li></ul></div><p><strong class="userinput"><code>svn diff OLD-URL[@OLDREV]
NEW-URL[@NEWREV]</code></strong> is shorthand for <strong class="userinput"><code>svn
diff --old=OLD-URL[@OLDREV]
--new=NEW-URL[@NEWREV].</code></strong></p><p><strong class="userinput"><code>svn diff -r N:M URL</code></strong> is shorthand
for <strong class="userinput"><code>svn diff -r N:M --old=URL
--new=URL</code></strong>.</p><p><strong class="userinput"><code>svn diff [-r N[:M]] URL1[@N]
URL2[@M]</code></strong> is shorthand for <strong class="userinput"><code>svn diff [-r
N[:M]] --old=URL1 --new=URL2</code></strong>.</p><p>If <em class="replaceable"><code>TARGET</code></em> is a URL, then
revs <code class="literal">N</code> and <code class="literal">M</code> can be
given either via the
<code class="option">--revision</code> (<code class="option">-r</code>) option
or by using the
<span class="quote">“<span class="quote">@</span>”</span> notation as described earlier.</p><p>If <em class="replaceable"><code>TARGET</code></em> is a working copy
path, the default behavior (when no
<code class="option">--revision</code> (<code class="option">-r</code>) option
is provided) is to display the differences between the
base and working copies
of <em class="replaceable"><code>TARGET</code></em>. If a
<code class="option">--revision</code> (<code class="option">-r</code>) option
is specified in this scenario, though, it means:</p><div class="variablelist"><dl><dt><span class="term"><code class="option">--revision N:M</code></span></dt><dd><p>The server compares <em class="replaceable"><code>TARGET@N</code></em>
and <em class="replaceable"><code>TARGET@M</code></em>.</p></dd><dt><span class="term"><code class="option">--revision N</code></span></dt><dd><p>The client compares
<em class="replaceable"><code>TARGET@N</code></em>
against the working copy.</p></dd></dl></div><p>If the alternate syntax is used, the server compares
<em class="replaceable"><code>URL1</code></em> and
<em class="replaceable"><code>URL2</code></em> at revisions
<em class="replaceable"><code>N</code></em> and
<em class="replaceable"><code>M</code></em>, respectively. If either
<em class="replaceable"><code>N</code></em> or
<em class="replaceable"><code>M</code></em> is omitted, a value of
<code class="literal">HEAD</code> is assumed.</p><p>By default, <span class="command"><strong>svn diff</strong></span> ignores the
ancestry of files and merely compares the contents of the
two files being compared. If you use
<code class="option">--notice-ancestry</code>, the ancestry of the
paths in question will be taken into consideration when
comparing revisions (i.e., if you run <span class="command"><strong>svn
diff</strong></span> on two files with identical contents but
different ancestry, you will see the entire contents of
the file as having been removed and added again).</p></div><div class="refsect1" title="Options"><a id="idp17100912"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.change"><code class="option">--change</code> (<code class="option">-c</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.changelist"><code class="option">--changelist</code> (<code class="option">--cl</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.diff_cmd"><code class="option">--diff-cmd</code> <em class="replaceable"><code>CMD</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.extensions"><code class="option">--extensions</code> (<code class="option">-x</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.force"><code class="option">--force</code></a>
<a class="xref" href="#svn.ref.svn.sw.git"><code class="option">--git</code></a>
<a class="xref" href="#svn.ref.svn.sw.internal_diff"><code class="option">--internal-diff</code></a>
<a class="xref" href="#svn.ref.svn.sw.new"><code class="option">--new</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.no_diff_deleted"><code class="option">--no-diff-deleted</code></a>
<a class="xref" href="#svn.ref.svn.sw.notice_ancestry"><code class="option">--notice-ancestry</code></a>
<a class="xref" href="#svn.ref.svn.sw.old"><code class="option">--old</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.show_copies_as_adds"><code class="option">--show-copies-as-adds</code></a>
<a class="xref" href="#svn.ref.svn.sw.summarize"><code class="option">--summarize</code></a>
<a class="xref" href="#svn.ref.svn.sw.xml"><code class="option">--xml</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17114272"></a><h2>Examples</h2><p>Compare <code class="literal">BASE</code> and your working copy
(one of the most popular uses of <span class="command"><strong>svn
diff</strong></span>):</p><div class="informalexample"><pre class="screen">
$ svn diff COMMITTERS
Index: COMMITTERS
===================================================================
--- COMMITTERS (revision 4404)
+++ COMMITTERS (working copy)
…
</pre></div><p>See what changed in the file
<code class="filename">COMMITTERS</code> revision 9115:</p><div class="informalexample"><pre class="screen">
$ svn diff -c 9115 COMMITTERS
Index: COMMITTERS
===================================================================
--- COMMITTERS (revision 3900)
+++ COMMITTERS (working copy)
…
</pre></div><p>See how your working copy's modifications compare
against an older revision:</p><div class="informalexample"><pre class="screen">
$ svn diff -r 3900 COMMITTERS
Index: COMMITTERS
===================================================================
--- COMMITTERS (revision 3900)
+++ COMMITTERS (working copy)
…
</pre></div><p>Compare revision 3000 to revision 3500 using <span class="quote">“<span class="quote">@</span>”</span>
syntax:</p><div class="informalexample"><pre class="screen">
$ svn diff http://svn.collab.net/repos/svn/trunk/COMMITTERS@3000 \
http://svn.collab.net/repos/svn/trunk/COMMITTERS@3500
Index: COMMITTERS
===================================================================
--- COMMITTERS (revision 3000)
+++ COMMITTERS (revision 3500)
…
</pre></div><p>Compare revision 3000 to revision 3500 using range
notation (pass only the one URL in this
case):</p><div class="informalexample"><pre class="screen">
$ svn diff -r 3000:3500 http://svn.collab.net/repos/svn/trunk/COMMITTERS
Index: COMMITTERS
===================================================================
--- COMMITTERS (revision 3000)
+++ COMMITTERS (revision 3500)
…
</pre></div><p>Compare revision 3000 to revision 3500 of all the files in
<code class="filename">trunk</code> using range notation:</p><div class="informalexample"><pre class="screen">
$ svn diff -r 3000:3500 http://svn.collab.net/repos/svn/trunk
</pre></div><p>Compare revision 3000 to revision 3500 of only three
files in <code class="filename">trunk</code> using range
notation:</p><div class="informalexample"><pre class="screen">
$ svn diff -r 3000:3500 --old http://svn.collab.net/repos/svn/trunk \
COMMITTERS README HACKING
</pre></div><p>If you have a working copy, you can obtain the
differences without typing in the long URLs:</p><div class="informalexample"><pre class="screen">
$ svn diff -r 3000:3500 COMMITTERS
Index: COMMITTERS
===================================================================
--- COMMITTERS (revision 3000)
+++ COMMITTERS (revision 3500)
…
</pre></div><p>Use <code class="option">--diff-cmd</code>
<em class="replaceable"><code>CMD</code></em> <code class="option">--extensions</code>
(<code class="option">-x</code>) to pass arguments directly to the
external diff program:</p><div class="informalexample"><pre class="screen">
$ svn diff --diff-cmd /usr/bin/diff -x "-i -b" COMMITTERS
Index: COMMITTERS
===================================================================
0a1,2
> This is a test
>
$
</pre></div><p>Lastly, you can use the <code class="option">--xml</code> option
along with the <code class="option">--summarize</code> option to view
XML describing the changes that occurred between
revisions, but not the contents of the diff itself:</p><div class="informalexample"><pre class="screen">
$ svn diff --summarize --xml http://svn.red-bean.com/repos/test@r2 \
http://svn.red-bean.com/repos/test
<?xml version="1.0"?>
<diff>
<paths>
<path
props="none"
kind="file"
item="modified">http://svn.red-bean.com/repos/test/sandwich.txt</path>
<path
props="none"
kind="file"
item="deleted">http://svn.red-bean.com/repos/test/burrito.txt</path>
<path
props="none"
kind="dir"
item="added">http://svn.red-bean.com/repos/test/snacks</path>
</paths>
</diff>
</pre></div></div></div><div class="refentry" title="svn export"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.export"></a><div class="titlepage"></div><a id="idp17140608" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn export — Export a clean directory tree.</p></div><div class="refsect1" title="Synopsis"><a id="idp17143680"></a><h2>Synopsis</h2><p><code class="literal">svn export [-r REV] URL[@PEGREV] [PATH]</code></p><p><code class="literal">svn export [-r REV] PATH1[@PEGREV] [PATH2]</code></p></div><div class="refsect1" title="Description"><a id="idp17146160"></a><h2>Description</h2><p>The first form exports a clean directory tree from the
repository specified by <em class="replaceable"><code>URL</code></em>—at revision
<em class="replaceable"><code>REV</code></em> if it is given; otherwise,
at <code class="literal">HEAD</code>, into
<em class="replaceable"><code>PATH</code></em>. If
<em class="replaceable"><code>PATH</code></em> is omitted, the last
component of the <em class="replaceable"><code>URL</code></em> is used
for the local directory name.</p><p>The second form exports a clean directory tree from
the working copy specified by
<em class="replaceable"><code>PATH1</code></em> into
<em class="replaceable"><code>PATH2</code></em>. All local changes will
be preserved, but files not under version control will not
be copied.</p></div><div class="refsect1" title="Options"><a id="idp17151936"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.force"><code class="option">--force</code></a>
<a class="xref" href="#svn.ref.svn.sw.ignore_externals"><code class="option">--ignore-externals</code></a>
<a class="xref" href="#svn.ref.svn.sw.ignore_keywords"><code class="option">--ignore-keywords</code></a>
<a class="xref" href="#svn.ref.svn.sw.native_eol"><code class="option">--native-eol</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17158640"></a><h2>Examples</h2><p>Export from your working copy (doesn't print every
file and directory):</p><div class="informalexample"><pre class="screen">
$ svn export a-wc my-export
Export complete.
</pre></div><p>Export directly from the repository (prints every
file and directory):</p><div class="informalexample"><pre class="screen">
$ svn export file:///var/svn/repos my-export
A my-export/test
A my-export/quiz
…
Exported revision 15.
</pre></div><p>When rolling operating-system-specific release
packages, it can be useful to export a tree that uses a
specific EOL character for line endings. The
<code class="option">--native-eol</code> option will do this, but it
affects only files that have <code class="literal">svn:eol-style =
native</code> properties attached to them. For
example, to export a tree with all CRLF line endings
(possibly for a Windows <code class="filename">.zip</code> file
distribution):</p><div class="informalexample"><pre class="screen">
$ svn export file:///var/svn/repos my-export --native-eol CRLF
A my-export/test
A my-export/quiz
…
Exported revision 15.
</pre></div><p>You can specify <code class="literal">LR</code>,
<code class="literal">CR</code>, or <code class="literal">CRLF</code> as a
line-ending type with the <code class="option">--native-eol</code>
option.</p></div></div><div class="refentry" title="svn help (h, ?)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.help"></a><div class="titlepage"></div><a id="idp17170912" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn help (h, ?) — Help!</p></div><div class="refsect1" title="Synopsis"><a id="idp17173968"></a><h2>Synopsis</h2><p><code class="literal">svn help [SUBCOMMAND...]</code></p></div><div class="refsect1" title="Description"><a id="idp17175600"></a><h2>Description</h2><p>This is your best friend when you're using
Subversion and this book isn't within reach!</p></div><div class="refsect1" title="Options"><a id="idp17176896"></a><h2>Options</h2><p>None</p></div></div><div class="refentry" title="svn import"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.import"></a><div class="titlepage"></div><a id="idp17179312" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn import — Commit an unversioned file or tree into the
repository.</p></div><div class="refsect1" title="Synopsis"><a id="idp17182416"></a><h2>Synopsis</h2><p><code class="literal">svn import [PATH] URL</code></p></div><div class="refsect1" title="Description"><a id="idp17184032"></a><h2>Description</h2><p>Recursively commit a copy of
<em class="replaceable"><code>PATH</code></em> to
<em class="replaceable"><code>URL</code></em>. If
<em class="replaceable"><code>PATH</code></em> is omitted,
<span class="quote">“<span class="quote"><code class="filename">.</code></span>”</span> is assumed. Parent
directories are created in the repository as necessary.
Unversionable items such as device files and pipes are
ignored even if <code class="option">--force</code> is
specified.</p></div><div class="refsect1" title="Options"><a id="idp17188064"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.auto_props"><code class="option">--auto-props</code></a>
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.editor_cmd"><code class="option">--editor-cmd</code> <em class="replaceable"><code>CMD</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.encoding"><code class="option">--encoding</code> <em class="replaceable"><code>ENC</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.file"><code class="option">--file</code> (<code class="option">-F</code>) <em class="replaceable"><code>FILENAME</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.force"><code class="option">--force</code></a>
<a class="xref" href="#svn.ref.svn.sw.force_log"><code class="option">--force-log</code></a>
<a class="xref" href="#svn.ref.svn.sw.message"><code class="option">--message</code> (<code class="option">-m</code>) <em class="replaceable"><code>MESSAGE</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.no_auto_props"><code class="option">--no-auto-props</code></a>
<a class="xref" href="#svn.ref.svn.sw.no_ignore"><code class="option">--no-ignore</code></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.with_revprop"><code class="option">--with-revprop</code> <em class="replaceable"><code>ARG</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17197824"></a><h2>Examples</h2><p>This imports the local directory
<code class="filename">myproj</code> into
<code class="filename">trunk/misc</code> in your repository. The
directory <code class="filename">trunk/misc</code> need not exist
before you import into it—<span class="command"><strong>svn
import</strong></span> will recursively create directories for
you.</p><div class="informalexample"><pre class="screen">
$ svn import -m "New import" myproj \
http://svn.red-bean.com/repos/trunk/misc
Adding myproj/sample.txt
…
Transmitting file data .........
Committed revision 16.
</pre></div><p>Be aware that this will <span class="emphasis"><em>not</em></span>
create a directory named <code class="filename">myproj</code> in
the repository. If that's what you want, simply add
<code class="filename">myproj</code> to the end of the URL:</p><div class="informalexample"><pre class="screen">
$ svn import -m "New import" myproj \
http://svn.red-bean.com/repos/trunk/misc/myproj
Adding myproj/sample.txt
…
Transmitting file data .........
Committed revision 16.
</pre></div><p>After importing data, note that the original tree is
<span class="emphasis"><em>not</em></span> under version control. To start
working, you still need to <span class="command"><strong>svn checkout</strong></span>
a fresh working copy of the tree.</p></div></div><div class="refentry" title="svn info"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.info"></a><div class="titlepage"></div><a id="idp17209792" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn info — Display information about a local or remote
item.</p></div><div class="refsect1" title="Synopsis"><a id="idp17212896"></a><h2>Synopsis</h2><p><code class="literal">svn info [TARGET[@REV]...]</code></p></div><div class="refsect1" title="Description"><a id="idp17214528"></a><h2>Description</h2><p>Print information about the working copy paths or URLs
specified. The information displayed for each path may
include (as pertinent to the object at that path):</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>information about the repository in which
the object is versioned</p></li><li class="listitem"><p>the most recent commit made to the
specified version of the object</p></li><li class="listitem"><p>any user-level locks held on the
object</p></li><li class="listitem"><p>local scheduling information (added,
deleted, copied, etc.)</p></li><li class="listitem"><p>local conflict information</p></li></ul></div></div><div class="refsect1" title="Options"><a id="idp17219264"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.changelist"><code class="option">--changelist</code> (<code class="option">--cl</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.incremental"><code class="option">--incremental</code></a>
<a class="xref" href="#svn.ref.svn.sw.recursive"><code class="option">--recursive</code> (<code class="option">-R</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.targets"><code class="option">--targets</code> <em class="replaceable"><code>FILENAME</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.xml"><code class="option">--xml</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17225712"></a><h2>Examples</h2><p><span class="command"><strong>svn info</strong></span> will show you all the
useful information that it has for items in your working
copy. It will show information for files:</p><div class="informalexample"><pre class="screen">
$ svn info foo.c
Path: foo.c
Name: foo.c
Working Copy Root Path: /home/sally/projects/test
URL: http://svn.red-bean.com/repos/test/foo.c
Repository Root: http://svn.red-bean.com/repos/test
Repository UUID: 5e7d134a-54fb-0310-bd04-b611643e5c25
Revision: 4417
Node Kind: file
Schedule: normal
Last Changed Author: sally
Last Changed Rev: 20
Last Changed Date: 2003-01-13 16:43:13 -0600 (Mon, 13 Jan 2003)
Text Last Updated: 2003-01-16 21:18:16 -0600 (Thu, 16 Jan 2003)
Properties Last Updated: 2003-01-13 21:50:19 -0600 (Mon, 13 Jan 2003)
Checksum: d6aeb60b0662ccceb6bce4bac344cb66
</pre></div><p>It will also show information for directories:</p><div class="informalexample"><pre class="screen">
$ svn info vendors
Path: vendors
Working Copy Root Path: /home/sally/projects/test
URL: http://svn.red-bean.com/repos/test/vendors
Repository Root: http://svn.red-bean.com/repos/test
Repository UUID: 5e7d134a-54fb-0310-bd04-b611643e5c25
Revision: 19
Node Kind: directory
Schedule: normal
Last Changed Author: harry
Last Changed Rev: 19
Last Changed Date: 2003-01-16 23:21:19 -0600 (Thu, 16 Jan 2003)
Properties Last Updated: 2003-01-16 23:39:02 -0600 (Thu, 16 Jan 2003)
</pre></div><p><span class="command"><strong>svn info</strong></span> also acts on URLs (also
note that the file <code class="filename">readme.doc</code> in
this example is locked, so lock information is also
provided):</p><div class="informalexample"><pre class="screen">
$ svn info http://svn.red-bean.com/repos/test/readme.doc
Path: readme.doc
Name: readme.doc
URL: http://svn.red-bean.com/repos/test/readme.doc
Repository Root: http://svn.red-bean.com/repos/test
Repository UUID: 5e7d134a-54fb-0310-bd04-b611643e5c25
Revision: 1
Node Kind: file
Schedule: normal
Last Changed Author: sally
Last Changed Rev: 42
Last Changed Date: 2003-01-14 23:21:19 -0600 (Tue, 14 Jan 2003)
Lock Token: opaquelocktoken:14011d4b-54fb-0310-8541-dbd16bd471b2
Lock Owner: harry
Lock Created: 2003-01-15 17:35:12 -0600 (Wed, 15 Jan 2003)
Lock Comment (1 line):
My test lock comment
</pre></div><p>Lastly, <span class="command"><strong>svn info</strong></span> output is
available in XML format by passing
the <code class="option">--xml</code> option:</p><div class="informalexample"><pre class="screen">
$ svn info --xml http://svn.red-bean.com/repos/test
<?xml version="1.0"?>
<info>
<entry
kind="dir"
path="."
revision="1">
<url>http://svn.red-bean.com/repos/test</url>
<repository>
<root>http://svn.red-bean.com/repos/test</root>
<uuid>5e7d134a-54fb-0310-bd04-b611643e5c25</uuid>
</repository>
<wc-info>
<schedule>normal</schedule>
<depth>infinity</depth>
</wc-info>
<commit
revision="1">
<author>sally</author>
<date>2003-01-15T23:35:12.847647Z</date>
</commit>
</entry>
</info>
</pre></div></div></div><div class="refentry" title="svn list (ls)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.list"></a><div class="titlepage"></div><a id="idp17239232" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn list (ls) — List directory entries in the repository.</p></div><div class="refsect1" title="Synopsis"><a id="idp17242320"></a><h2>Synopsis</h2><p><code class="literal">svn list [TARGET[@REV]...]</code></p></div><div class="refsect1" title="Description"><a id="idp17243952"></a><h2>Description</h2><p>List each <em class="replaceable"><code>TARGET</code></em> file and
the contents of each <em class="replaceable"><code>TARGET</code></em>
directory as they exist in the repository. If
<em class="replaceable"><code>TARGET</code></em> is a working copy path,
the corresponding repository URL will be used.</p><p>The default <em class="replaceable"><code>TARGET</code></em> is
<span class="quote">“<span class="quote"><code class="filename">.</code></span>”</span>, meaning the
repository URL of the current working copy
directory.</p><p>With <code class="option">--verbose</code>
(<code class="option">-v</code>), <span class="command"><strong>svn list</strong></span> shows
the following fields for each item:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>Revision number of the last
commit</p></li><li class="listitem"><p>Author of the last commit</p></li><li class="listitem"><p>If locked, the letter <span class="quote">“<span class="quote">O</span>”</span> (see the preceding section on <a class="xref" href="#svn.ref.svn.c.info" title="svn info">svn info</a> for details).</p></li><li class="listitem"><p>Size (in bytes)</p></li><li class="listitem"><p>Date and time of the last
commit</p></li></ul></div><p>With <code class="option">--xml</code>, output is in XML format
(with a header and an enclosing document element unless
<code class="option">--incremental</code> is also specified). All of
the information is present; the <code class="option">--verbose</code>
(<code class="option">-v</code>) option is not accepted.</p></div><div class="refsect1" title="Options"><a id="idp17257200"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.incremental"><code class="option">--incremental</code></a>
<a class="xref" href="#svn.ref.svn.sw.recursive"><code class="option">--recursive</code> (<code class="option">-R</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.verbose"><code class="option">--verbose</code> (<code class="option">-v</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.xml"><code class="option">--xml</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17262976"></a><h2>Examples</h2><p><span class="command"><strong>svn list</strong></span> is most useful if you
want to see what files a repository has without
downloading a working copy:</p><div class="informalexample"><pre class="screen">
$ svn list http://svn.red-bean.com/repos/test/support
README.txt
INSTALL
examples/
…
</pre></div><p>You can pass the <code class="option">--verbose</code>
(<code class="option">-v</code>) option for additional information,
rather like the Unix command
<span class="command"><strong>ls -l</strong></span>:</p><div class="informalexample"><pre class="screen">
$ svn list -v file:///var/svn/repos
16 sally 28361 Jan 16 23:18 README.txt
27 sally 0 Jan 18 15:27 INSTALL
24 harry Jan 18 11:27 examples/
</pre></div><p>You can also get <span class="command"><strong>svn list</strong></span> output in
XML format with the <code class="option">--xml</code> option:</p><div class="informalexample"><pre class="screen">
$ svn list --xml http://svn.red-bean.com/repos/test
<?xml version="1.0"?>
<lists>
<list
path="http://svn.red-bean.com/repos/test">
<entry
kind="dir">
<name>examples</name>
<size>0</size>
<commit
revision="24">
<author>harry</author>
<date>2008-01-18T06:35:53.048870Z</date>
</commit>
</entry>
...
</list>
</lists>
</pre></div><p>For further details, see the earlier section <a class="xref" href="#svn.tour.history.browsing.list" title="Listing versioned directories">the section called “Listing versioned directories”</a>.</p></div></div><div class="refentry" title="svn lock"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.lock"></a><div class="titlepage"></div><a id="idp17275136" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn lock — Lock working copy paths or URLs in the
repository so that no other user can commit changes to
them.</p></div><div class="refsect1" title="Synopsis"><a id="idp17278304"></a><h2>Synopsis</h2><p><code class="literal">svn lock TARGET...</code></p></div><div class="refsect1" title="Description"><a id="idp17279920"></a><h2>Description</h2><p>Lock each <em class="replaceable"><code>TARGET</code></em>. If any
<em class="replaceable"><code>TARGET</code></em> is already locked by
another user, print a warning and continue locking the
rest of the <em class="replaceable"><code>TARGET</code></em>s. Use
<code class="option">--force</code> to steal a lock from another user
or working copy.</p></div><div class="refsect1" title="Options"><a id="idp17283072"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.encoding"><code class="option">--encoding</code> <em class="replaceable"><code>ENC</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.file"><code class="option">--file</code> (<code class="option">-F</code>) <em class="replaceable"><code>FILENAME</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.force"><code class="option">--force</code></a>
<a class="xref" href="#svn.ref.svn.sw.force_log"><code class="option">--force-log</code></a>
<a class="xref" href="#svn.ref.svn.sw.message"><code class="option">--message</code> (<code class="option">-m</code>) <em class="replaceable"><code>MESSAGE</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.targets"><code class="option">--targets</code> <em class="replaceable"><code>FILENAME</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17288816"></a><h2>Examples</h2><p>Lock two files in your working copy:</p><div class="informalexample"><pre class="screen">
$ svn lock tree.jpg house.jpg
'tree.jpg' locked by user 'harry'.
'house.jpg' locked by user 'harry'.
</pre></div><p>Lock a file in your working copy that is currently
locked by another user:</p><div class="informalexample"><pre class="screen">
$ svn lock tree.jpg
svn: warning: W160035: Path '/tree.jpg is already locked by user 'sally' in fi
lesystem '/var/svn/repos/db'
$ svn lock --force tree.jpg
'tree.jpg' locked by user 'harry'.
</pre></div><p>Lock a file without a working copy:</p><div class="informalexample"><pre class="screen">
$ svn lock http://svn.red-bean.com/repos/test/tree.jpg
'tree.jpg' locked by user 'harry'.
</pre></div><p>For further details, see <a class="xref" href="#svn.advanced.locking" title="Locking">the section called “Locking”</a>.</p></div></div><div class="refentry" title="svn log"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.log"></a><div class="titlepage"></div><a id="idp17296720" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn log — Display commit log messages.</p></div><div class="refsect1" title="Synopsis"><a id="idp17299808"></a><h2>Synopsis</h2><p><code class="literal">svn log [PATH]</code></p><p><code class="literal">svn log URL[@REV] [PATH...]</code></p></div><div class="refsect1" title="Description"><a id="idp17302256"></a><h2>Description</h2><p>Shows log messages from the repository.
If no arguments are supplied, <span class="command"><strong>svn
log</strong></span> shows the log messages for all files and
directories inside (and including) the current working
directory of your working copy. You can refine the
results by specifying a path, one or more revisions, or
any combination of the two. The default revision range
for a local path is <code class="literal">BASE:1</code>.</p><p>If you specify a URL alone, it prints log
messages for everything the URL contains. If you
add paths past the URL, only messages for those paths
under that URL will be printed. The default revision range
for a URL is <code class="literal">HEAD:1</code>.</p><p>With <code class="option">--verbose</code>
(<code class="option">-v</code>), <span class="command"><strong>svn log</strong></span> will
also print all affected paths with each log message.
With <code class="option">--quiet</code>
(<code class="option">-q</code>), <span class="command"><strong>svn log</strong></span> will not
print the log message body itself, this is compatible
with <code class="option">--verbose</code>
(<code class="option">-v</code>).</p><p>Each log message is printed just once, even if more
than one of the affected paths for that revision were
explicitly requested. Logs follow copy history by
default. Use <code class="option">--stop-on-copy</code> to disable
this behavior, which can be useful for determining branch
points.</p></div><div class="refsect1" title="Options"><a id="idp17312272"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.change"><code class="option">--change</code> (<code class="option">-c</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.diff"><code class="option">--diff</code></a>
<a class="xref" href="#svn.ref.svn.sw.diff_cmd"><code class="option">--diff-cmd</code> <em class="replaceable"><code>CMD</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.extensions"><code class="option">--extensions</code> (<code class="option">-x</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.incremental"><code class="option">--incremental</code></a>
<a class="xref" href="#svn.ref.svn.sw.internal_diff"><code class="option">--internal-diff</code></a>
<a class="xref" href="#svn.ref.svn.sw.limit"><code class="option">--limit</code> (<code class="option">-l</code>) <em class="replaceable"><code>NUM</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.stop_on_copy"><code class="option">--stop-on-copy</code></a>
<a class="xref" href="#svn.ref.svn.sw.targets"><code class="option">--targets</code> <em class="replaceable"><code>FILENAME</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.use_merge_history"><code class="option">--use-merge-history</code> (<code class="option">-g</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.verbose"><code class="option">--verbose</code> (<code class="option">-v</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.with_all_revprops"><code class="option">--with-all-revprops</code></a>
<a class="xref" href="#svn.ref.svn.sw.with_no_revprops"><code class="option">--with-no-revprops</code></a>
<a class="xref" href="#svn.ref.svn.sw.with_revprop"><code class="option">--with-revprop</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.xml"><code class="option">--xml</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17326432"></a><h2>Examples</h2><p>You can see the log messages for all the paths that
changed in your working copy by running <strong class="userinput"><code>svn
log</code></strong> from the top:</p><div class="informalexample"><pre class="screen">
$ svn log
------------------------------------------------------------------------
r20 | harry | 2003-01-17 22:56:19 -0600 (Fri, 17 Jan 2003) | 1 line
Tweak.
------------------------------------------------------------------------
r17 | sally | 2003-01-16 23:21:19 -0600 (Thu, 16 Jan 2003) | 2 lines
…
</pre></div><p>Examine all log messages for a particular file in
your working copy:</p><div class="informalexample"><pre class="screen">
$ svn log foo.c
------------------------------------------------------------------------
r32 | sally | 2003-01-13 00:43:13 -0600 (Mon, 13 Jan 2003) | 1 line
Added defines.
------------------------------------------------------------------------
r28 | sally | 2003-01-07 21:48:33 -0600 (Tue, 07 Jan 2003) | 3 lines
…
</pre></div><p>If you don't have a working copy handy, you can log
a URL:</p><div class="informalexample"><pre class="screen">
$ svn log http://svn.red-bean.com/repos/test/foo.c
------------------------------------------------------------------------
r32 | sally | 2003-01-13 00:43:13 -0600 (Mon, 13 Jan 2003) | 1 line
Added defines.
------------------------------------------------------------------------
r28 | sally | 2003-01-07 21:48:33 -0600 (Tue, 07 Jan 2003) | 3 lines
…
</pre></div><p>If you want several distinct paths underneath the
same URL, you can use the <code class="literal">URL [PATH...]</code>
syntax:</p><div class="informalexample"><pre class="screen">
$ svn log http://svn.red-bean.com/repos/test/ foo.c bar.c
------------------------------------------------------------------------
r32 | sally | 2003-01-13 00:43:13 -0600 (Mon, 13 Jan 2003) | 1 line
Added defines.
------------------------------------------------------------------------
r31 | harry | 2003-01-10 12:25:08 -0600 (Fri, 10 Jan 2003) | 1 line
Added new file bar.c
------------------------------------------------------------------------
r28 | sally | 2003-01-07 21:48:33 -0600 (Tue, 07 Jan 2003) | 3 lines
…
</pre></div><p>The <code class="option">--verbose</code> (<code class="option">-v</code>)
option causes <span class="command"><strong>svn log</strong></span> to include
information about the paths that were changed in each
displayed revision. These paths appear, one path per line
of output, with action codes that indicate what type of
change was made to the path.</p><div class="informalexample"><pre class="screen">
$ svn log -v http://svn.red-bean.com/repos/test/ foo.c bar.c
------------------------------------------------------------------------
r32 | sally | 2003-01-13 00:43:13 -0600 (Mon, 13 Jan 2003) | 1 line
Changed paths:
M /foo.c
Added defines.
------------------------------------------------------------------------
r31 | harry | 2003-01-10 12:25:08 -0600 (Fri, 10 Jan 2003) | 1 line
Changed paths:
A /bar.c
Added new file bar.c
------------------------------------------------------------------------
r28 | sally | 2003-01-07 21:48:33 -0600 (Tue, 07 Jan 2003) | 3 lines
…
</pre></div><p><span class="command"><strong>svn log</strong></span> uses just a handful of action codes,
and they are similar to the
ones the <span class="command"><strong>svn update</strong></span>
command uses:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">A</code></span></dt><dd><p>The item was added.</p></dd><dt><span class="term"><code class="literal">D</code></span></dt><dd><p>The item was deleted.</p></dd><dt><span class="term"><code class="literal">M</code></span></dt><dd><p>Properties or textual contents on the
item were changed.</p></dd><dt><span class="term"><code class="literal">R</code></span></dt><dd><p>The item was replaced by a different one
at the same location.</p></dd></dl></div><p>In addition to the action codes which precede the
changed paths, <span class="command"><strong>svn log</strong></span> with
the <code class="option">--verbose</code> (<code class="option">-v</code>)
option will note whether a path was added or replaced as
the result of a copy operation. It does so by
printing <code class="literal">(from
<em class="replaceable"><code>COPY-FROM-PATH</code></em>:<em class="replaceable"><code>COPY-FROM-REV</code></em>)</code>
after such paths.</p><p>When you're concatenating the results of multiple
calls to the log command, you may want to use the
<code class="option">--incremental</code> option. <span class="command"><strong>svn
log</strong></span> normally prints out a dashed line at the
beginning of a log message, after each subsequent log
message, and following the final log message. If you
ran <span class="command"><strong>svn log</strong></span> on a range of two
revisions, you would get this:</p><div class="informalexample"><pre class="screen">
$ svn log -r 14:15
------------------------------------------------------------------------
r14 | …
------------------------------------------------------------------------
r15 | …
------------------------------------------------------------------------
</pre></div><p>However, if you wanted to gather two nonsequential log
messages into a file, you might do something like
this:</p><div class="informalexample"><pre class="screen">
$ svn log -r 14 > mylog
$ svn log -r 19 >> mylog
$ svn log -r 27 >> mylog
$ cat mylog
------------------------------------------------------------------------
r14 | …
------------------------------------------------------------------------
------------------------------------------------------------------------
r19 | …
------------------------------------------------------------------------
------------------------------------------------------------------------
r27 | …
------------------------------------------------------------------------
</pre></div><p>You can avoid the clutter of the double dashed lines
in your output by using the <code class="option">--incremental</code>
option:</p><div class="informalexample"><pre class="screen">
$ svn log --incremental -r 14 > mylog
$ svn log --incremental -r 19 >> mylog
$ svn log --incremental -r 27 >> mylog
$ cat mylog
------------------------------------------------------------------------
r14 | …
------------------------------------------------------------------------
r19 | …
------------------------------------------------------------------------
r27 | …
</pre></div><p>The <code class="option">--incremental</code> option provides
similar output control when using the
<code class="option">--xml</code> option:</p><div class="informalexample"><pre class="screen">
$ svn log --xml --incremental -r 1 sandwich.txt
<logentry
revision="1">
<author>harry</author>
<date>2008-06-03T06:35:53.048870Z</date>
<msg>Initial Import.</msg>
</logentry>
</pre></div><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>Sometimes when you run <span class="command"><strong>svn log</strong></span> on
a specific path and a specific revision, you see no log
information output at all, as in the following:</p><div class="informalexample"><pre class="screen">
$ svn log -r 20 http://svn.red-bean.com/untouched.txt
------------------------------------------------------------------------
</pre></div><p>That just means the path wasn't modified in that
revision. To get log information for that revision,
either run the log operation against the repository's
root URL, or specify a path that you happen to know was
changed in that revision:</p><div class="informalexample"><pre class="screen">
$ svn log -r 20 touched.txt
------------------------------------------------------------------------
r20 | sally | 2003-01-17 22:56:19 -0600 (Fri, 17 Jan 2003) | 1 line
Made a change.
------------------------------------------------------------------------
</pre></div></td></tr></table></div><p>Beginning with Subversion 1.7, users can take
advantage of a special output mode which combines the
information from <span class="command"><strong>svn log</strong></span> with what you
would see when running <span class="command"><strong>svn diff</strong></span> on the
same targets for each revision of the log. Simply
invoke <span class="command"><strong>svn log</strong></span> with
the <code class="option">--diff</code> option to trigger this output
mode.</p><div class="informalexample"><pre class="screen">
$ svn log -r 20 touched.txt --diff
------------------------------------------------------------------------
r20 | sally | 2003-01-17 22:56:19 -0600 (Fri, 17 Jan 2003) | 1 line
Made a change.
Index: touched.txt
===================================================================
--- touched.txt (revision 19)
+++ touched.txt (revision 20)
@@ -1 +1,2 @@
This is the file 'touched.txt'.
+We add such exciting text to files around here!
------------------------------------------------------------------------
$
</pre></div><p>As with <span class="command"><strong>svn diff</strong></span>, you may also make
use of many of the various options which control the way
the difference is generated,
including <code class="option">--depth</code>, <code class="option">--diff-cmd</code>,
and <code class="option">--extensions</code>
(<code class="option">-x</code>).</p></div></div><div class="refentry" title="svn merge"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.merge"></a><div class="titlepage"></div><a id="idp17378160" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn merge — Apply the differences between two sources to a
working copy path.</p></div><div class="refsect1" title="Synopsis"><a id="idp17381280"></a><h2>Synopsis</h2><p><code class="literal">svn merge [-c M[,N...] | -r N:M ...] SOURCE[@REV] [TARGET_WCPATH]</code></p><p><code class="literal">svn merge --reintegrate SOURCE[@REV] [TARGET_WCPATH]</code></p><p><code class="literal">svn merge SOURCE1[@N] SOURCE2[@M] [TARGET_WCPATH]</code></p></div><div class="refsect1" title="Description"><a id="idp17384640"></a><h2>Description</h2><p>In all three forms
<em class="replaceable"><code>TARGET_WCPATH</code></em> is
the working copy path that will receive the differences. If
<em class="replaceable"><code>TARGET_WCPATH</code></em> is omitted, the
changes are applied to the current working directory,
unless the sources have identical basenames that match a
file within the current working directory. In
this case, the differences will be applied to that
file.</p><p>In the first two forms, <em class="replaceable"><code>SOURCE</code></em>
can be either a URL or a working copy path (in which case its
corresponding URL is used). If the peg revision
<em class="replaceable"><code>REV</code></em> is not specified, then
<code class="literal">HEAD</code> is assumed. In the third form the
same rules apply for
<em class="replaceable"><code>SOURCE1</code></em>,
<em class="replaceable"><code>SOURCE2</code></em>,
<em class="replaceable"><code>M</code></em>, and <em class="replaceable"><code>N</code></em>
with the only difference being that if either source is a
working copy path, then the peg revisions
<span class="emphasis"><em>must</em></span> be explicitly stated.</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>Sync and Cherrypick Merges</p><p>The first form, when used without either the
<code class="option">-c</code> or <code class="option">-r</code> options, is
called a <span class="quote">“<span class="quote">sync</span>”</span> merge and
<code class="literal">-r 1:REV</code> is implied. This variant is
used to merge all eligible changes to a branch from its
immediate ancestor branch, see
<a class="xref" href="#svn.branchemerge.basicmerging.stayinsync" title="Keeping a Branch in Sync">the section called “Keeping a Branch in Sync”</a>.
</p><p>When the first form is used with the
<code class="option">-c</code> or <code class="option">-r</code> options, this
is called a <span class="quote">“<span class="quote">cherrypick</span>”</span> merge and is used
to merge an explicitly defined set of changes from one
branch to another, see
<a class="xref" href="#svn.branchmerge.cherrypicking" title="Cherrypicking">the section called “Cherrypicking”</a></p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>Multiple <code class="literal">-c</code> and/or
<code class="literal">-r</code> instances may be specified, and
mixing of forward and reverse ranges is allowed—
the ranges are internally compacted to their minimum
representation before merging begins (which may result
in a no-op merge or conflicts that cause the merge to
stop before merging all of the requested revisions).</p></td></tr></table></div><p>In both variants of the first form,
<em class="replaceable"><code>SOURCE</code></em> in revision
<em class="replaceable"><code>REV</code></em> is compared as it existed
between revisions <em class="replaceable"><code>N</code></em> and
<em class="replaceable"><code>M</code></em> for each revision range
provided.</p></li><li class="listitem"><p>Reintegrate Merges</p><p>The second form is called a <span class="quote">“<span class="quote">reintegrate
merge</span>”</span> and is used to bring changes from a feature
branch (<em class="replaceable"><code>SOURCE</code></em>)
back into the feature branch's immediate ancestor branch
(<em class="replaceable"><code>TARGET_WCPATH</code></em>).</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>Reintegrate merges support only this specialized
use case and as such have a number of special
requirements and limitations that the other two merge
forms do not posses. See
<a class="xref" href="#svn.branchemerge.basicmerging.stayinsync" title="Keeping a Branch in Sync">the section called “Keeping a Branch in Sync”</a>,
<a class="xref" href="#svn.branchemerge.basicmerging.reintegrate" title="Reintegrating a Branch">the section called “Reintegrating a Branch”</a>,
<a class="xref" href="#svn.branchmerge.advanced.reintegratetwice" title="Keeping a Reintegrated Branch Alive">the section called “Keeping a Reintegrated Branch Alive”</a>,
and
<a class="xref" href="#svn.branchmerge.commonpatterns.feature" title="Feature Branches">the section called “Feature Branches”</a>.
</p></td></tr></table></div></li><li class="listitem"><p>2-URL Merges</p><p>In the third form, called a <span class="quote">“<span class="quote">2-URL Merge</span>”</span>,
the difference between <em class="replaceable"><code>SOURCE1</code></em>
at revision <em class="replaceable"><code>N</code></em> and
<em class="replaceable"><code>SOURCE2</code></em> at revision
<em class="replaceable"><code>M</code></em> is generated and applied to
<em class="replaceable"><code>TARGET_WCPATH</code></em>. The revisions
default to <code class="literal">HEAD</code> if omitted.</p></li></ul></div><p>If
<a class="xref" href="#svn.branchmerge.basicmerging.mergetracking" title="Merge Tracking">Merge Tracking</a>
is active, then Subversion will internally track metadata
(i.e. the <code class="literal">svn:mergeinfo</code> property) about
merge operations when the two merge sources are ancestrally
related—if the first source is an ancestor of the
second or vice versa—this is guaranteed to be the case
when using the first two forms. Subversion will also take
preexisting merge metadata on the working copy target into
account when determining what revisions to merge and in an
effort to avoid repeat merges and needless conflicts it may
only merge a subset of the requested ranges.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p><a class="xref" href="#svn.branchmerge.basicmerging.mergetracking" title="Merge Tracking">Merge Tracking</a>
can be disabled by using the <code class="option">--ignore-ancestry</code>
option.</p></td></tr></table></div><p>Unlike <span class="command"><strong>svn diff</strong></span>, the merge command
takes the ancestry of a file into consideration when
performing a merge operation. This is very important when
you're merging changes from one branch into another and you've
renamed a file on one branch but not the other.</p></div><div class="refsect1" title="Options"><a id="idp17420640"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.accept"><code class="option">--accept</code> <em class="replaceable"><code>ACTION</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.allow_mixed_revisions"><code class="option">--allow-mixed-revisions</code></a>
<a class="xref" href="#svn.ref.svn.sw.change"><code class="option">--change</code> (<code class="option">-c</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.diff3_cmd"><code class="option">--diff3-cmd</code> <em class="replaceable"><code>CMD</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.dry_run"><code class="option">--dry-run</code></a>
<a class="xref" href="#svn.ref.svn.sw.extensions"><code class="option">--extensions</code> (<code class="option">-x</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.force"><code class="option">--force</code></a>
<a class="xref" href="#svn.ref.svn.sw.ignore_ancestry"><code class="option">--ignore-ancestry</code></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.record_only"><code class="option">--record-only</code></a>
<a class="xref" href="#svn.ref.svn.sw.reintegrate"><code class="option">--reintegrate</code></a>
<a class="xref" href="#svn.ref.svn.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17431600"></a><h2>Examples</h2><p>Merge a branch back into the trunk (assuming that
you have an up-to-date working copy of the trunk):</p><div class="informalexample"><pre class="screen">
$ svn merge --reintegrate \
http://svn.example.com/repos/calc/branches/my-calc-branch
--- Merging differences between repository URLs into '.':
U button.c
U integer.c
U Makefile
U .
--- Recording mergeinfo for merge between repository URLs into '.':
U .
$ # build, test, verify, ...
$ svn commit -m "Merge my-calc-branch back into trunk!"
Sending .
Sending button.c
Sending integer.c
Sending Makefile
Transmitting file data ..
Committed revision 391.
</pre></div><p>To merge changes to a single file:</p><div class="informalexample"><pre class="screen">
$ svn merge -c 31 ^/trunk/thhgttg.txt thhgttg.txt
--- Merging r31 into 'thhgttg.txt':
U thhgttg.txt
--- Recording mergeinfo for merge of r31 into 'thhgttg.txt':
U thhgttg.txt
</pre></div></div></div><div class="refentry" title="svn mergeinfo"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.mergeinfo"></a><div class="titlepage"></div><a id="idp17437376" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn mergeinfo — Query merge-related information. See <a class="xref" href="#svn.branchmerge.basicmerging.mergeinfo" title="Mergeinfo and Previews">the section called “Mergeinfo and Previews”</a> for
details.</p></div><div class="refsect1" title="Synopsis"><a id="idp17441168"></a><h2>Synopsis</h2><p><code class="literal">svn mergeinfo SOURCE_URL[@REV] [TARGET[@REV]]</code></p></div><div class="refsect1" title="Description"><a id="idp17442816"></a><h2>Description</h2><p>Query information related to merges (or potential
merges) between <em class="replaceable"><code>SOURCE-URL</code></em> and
<em class="replaceable"><code>TARGET</code></em>. If
the <code class="option">--show-revs</code> option is not provided,
display revisions which have been merged
from <em class="replaceable"><code>SOURCE-URL</code></em>
to <em class="replaceable"><code>TARGET</code></em>. Otherwise, display
either <code class="literal">merged</code>
or <code class="literal">eligible</code> revisions as specified by
the <code class="option">--show-revs</code> option.</p></div><div class="refsect1" title="Options"><a id="idp17448368"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.recursive"><code class="option">--recursive</code> (<code class="option">-R</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.show_revs"><code class="option">--show-revs</code> <em class="replaceable"><code>ARG</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17453008"></a><h2>Examples</h2><p>Find out which changesets your have been merged from
your trunk directory into your test branch:</p><div class="informalexample"><pre class="screen">
$ svn propget svn:mergeinfo ^/branches/test
/branches/other:3-4
/trunk:11-13,14,16
$ svn mergeinfo --show-revs merged ^/trunk ^/branches/test
r11
r12
r13
r14
r16
$
</pre></div><p>Note that the default output from the <span class="command"><strong>svn
mergeinfo</strong></span> command is to display merged revisions, so
the <code class="option">--show-revs</code> option shown in the
command line of the previous example is not strictly
required.</p><p>Find out which changesets from your trunk directory
have not yet been merged into your test branch:</p><div class="informalexample"><pre class="screen">
$ svn mergeinfo --show-revs eligible ^/trunk ^/branches/test
r15
r17
r20
r21
r22
$
</pre></div></div></div><div class="refentry" title="svn mkdir"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.mkdir"></a><div class="titlepage"></div><a id="idp17460160" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn mkdir — Create a new directory under version control.</p></div><div class="refsect1" title="Synopsis"><a id="idp17463248"></a><h2>Synopsis</h2><p><code class="literal">svn mkdir PATH...</code></p><p><code class="literal">svn mkdir URL...</code></p></div><div class="refsect1" title="Description"><a id="idp17465680"></a><h2>Description</h2><p>Create a directory with a name given by the final
component of the <em class="replaceable"><code>PATH</code></em> or <em class="replaceable"><code>URL</code></em>.
A directory specified by a working copy
<em class="replaceable"><code>PATH</code></em> is scheduled for addition
in the working copy. A directory specified by a URL is
created in the repository via an immediate commit.
Multiple directory URLs are committed atomically. In both
cases, all the intermediate directories must already exist
unless the <code class="option">--parents</code> option is
used.</p></div><div class="refsect1" title="Options"><a id="idp17469088"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.editor_cmd"><code class="option">--editor-cmd</code> <em class="replaceable"><code>CMD</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.encoding"><code class="option">--encoding</code> <em class="replaceable"><code>ENC</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.file"><code class="option">--file</code> (<code class="option">-F</code>) <em class="replaceable"><code>FILENAME</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.force_log"><code class="option">--force-log</code></a>
<a class="xref" href="#svn.ref.svn.sw.message"><code class="option">--message</code> (<code class="option">-m</code>) <em class="replaceable"><code>MESSAGE</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.parents"><code class="option">--parents</code></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.with_revprop"><code class="option">--with-revprop</code> <em class="replaceable"><code>ARG</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17476192"></a><h2>Examples</h2><p>Create a directory in your working copy:</p><div class="informalexample"><pre class="screen">
$ svn mkdir newdir
A newdir
</pre></div><p>Create one in the repository (this is an instant commit, so a
log message is required):</p><div class="informalexample"><pre class="screen">
$ svn mkdir -m "Making a new dir." http://svn.red-bean.com/repos/newdir
Committed revision 26.
</pre></div></div></div><div class="refentry" title="svn move (mv)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.move"></a><div class="titlepage"></div><a id="idp17481328" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn move (mv) — Move a file or directory.</p></div><div class="refsect1" title="Synopsis"><a id="idp17484400"></a><h2>Synopsis</h2><p><code class="literal">svn move SRC... DST</code></p></div><div class="refsect1" title="Description"><a id="idp17486016"></a><h2>Description</h2><p>This command moves files or directories in your
working copy or in the repository.</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>This command is equivalent to an <span class="command"><strong>svn
copy</strong></span> followed by <span class="command"><strong>svn
delete</strong></span>.</p></td></tr></table></div><p>When moving multiple sources, they will be added as
children of <em class="replaceable"><code>DST</code></em>, which must be
a directory.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>Subversion does not support moving between working
copies and URLs. In addition, you can only move files
within a single repository—Subversion does not
support cross-repository moving. Subversion supports
the following types of moves within a single repository:</p></td></tr></table></div><div class="variablelist"><dl><dt><span class="term">WC → WC</span></dt><dd><p>Move and schedule a file or directory for
addition (with history).</p></dd><dt><span class="term">URL → URL</span></dt><dd><p>Complete server-side rename.</p></dd></dl></div></div><div class="refsect1" title="Options"><a id="idp17495248"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.editor_cmd"><code class="option">--editor-cmd</code> <em class="replaceable"><code>CMD</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.encoding"><code class="option">--encoding</code> <em class="replaceable"><code>ENC</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.file"><code class="option">--file</code> (<code class="option">-F</code>) <em class="replaceable"><code>FILENAME</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.force"><code class="option">--force</code></a>
<a class="xref" href="#svn.ref.svn.sw.force_log"><code class="option">--force-log</code></a>
<a class="xref" href="#svn.ref.svn.sw.message"><code class="option">--message</code> (<code class="option">-m</code>) <em class="replaceable"><code>MESSAGE</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.parents"><code class="option">--parents</code></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.with_revprop"><code class="option">--with-revprop</code> <em class="replaceable"><code>ARG</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17503632"></a><h2>Examples</h2><p>Move a file in your working copy:</p><div class="informalexample"><pre class="screen">
$ svn move foo.c bar.c
A bar.c
D foo.c
</pre></div><p>Move several files in your working copy into a
subdirectory:</p><div class="informalexample"><pre class="screen">
$ svn move baz.c bat.c qux.c src
A src/baz.c
D baz.c
A src/bat.c
D bat.c
A src/qux.c
D qux.c
</pre></div><p>Move a file in the repository (this is an immediate commit,
so it requires a commit message):</p><div class="informalexample"><pre class="screen">
$ svn move -m "Move a file" http://svn.red-bean.com/repos/foo.c \
http://svn.red-bean.com/repos/bar.c
Committed revision 27.
</pre></div></div></div><div class="refentry" title="svn patch"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.patch"></a><div class="titlepage"></div><a id="idp17510560" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn patch — Apply changes represented in a unidiff patch to
the working copy.</p></div><div class="refsect1" title="Synopsis"><a id="idp17513680"></a><h2>Synopsis</h2><p><code class="literal">svn patch PATCHFILE [WCPATH]</code></p></div><div class="refsect1" title="Description"><a id="idp17515312"></a><h2>Description</h2><p>This subcommand will apply changes described a
unidiff-formatted patch
file <em class="replaceable"><code>PATCHFILE</code></em> to the working
copy <em class="replaceable"><code>WCPATH</code></em>. As with most
other working copy subcommands,
if <em class="replaceable"><code>WCPATH</code></em> is omitted, the
changes are applied to the current working directory. A
unidiff patch suitable for application to a working copy
can be produced with the <span class="command"><strong>svn diff</strong></span>
command or third-party differencing tools. Any
non-unidiff content found in the patch file is
ignored.</p><p>Changes listed in the patch file will either be
applied or rejected. If a change does not match at its
exact line offset, it may be applied earlier or later in
the file if a match is found elsewhere for the surrounding
lines of context provided by the patch. A change may also
be applied with <em class="firstterm">fuzz</em>—meaning,
one or more lines of context are ignored when attempting
to match the change location. If no matching context can
be found for a change, the change conflicts and will be
written to a reject file which bears the extension
<code class="filename">.svnpatch.rej</code>.</p><p><span class="command"><strong>svn patch</strong></span> reports a status line for
patched file or directory using letter codes, very similar
to the way that <span class="command"><strong>svn update</strong></span> provides
notification. The letter codes have the following
meanings:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">A</code></span></dt><dd><p>Added</p></dd><dt><span class="term"><code class="literal">D</code></span></dt><dd><p>Deleted</p></dd><dt><span class="term"><code class="literal">C</code></span></dt><dd><p>Conflicted</p></dd><dt><span class="term"><code class="literal">G</code></span></dt><dd><p>Merged</p></dd><dt><span class="term"><code class="literal">U</code></span></dt><dd><p>Updated</p></dd></dl></div><p>Changes applied with an offset or fuzz are reported on
lines starting with the '<code class="literal">></code>'
symbol. You should review such changes carefully.</p><p>If the patch removes all content from a file, that
file is automatically scheduled for deletion. Likewise,
if the patch creates a new file, that file is
automatically scheduled for addition. Use <span class="command"><strong>svn
revert</strong></span> to undo undesired deletions and
additions.</p></div><div class="refsect1" title="Options"><a id="idp17535984"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.dry_run"><code class="option">--dry-run</code></a>
<a class="xref" href="#svn.ref.svn.sw.ignore_whitespace"><code class="option">--ignore-whitespace</code></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.reverse_diff"><code class="option">--reverse-diff</code></a>
<a class="xref" href="#svn.ref.svn.sw.strip"><code class="option">--strip</code> <em class="replaceable"><code>NUM</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17541440"></a><h2>Examples</h2><p>Apply a simple patch file generated by
the <span class="command"><strong>svn diff</strong></span> command. Our patch file
will create a new file, delete another file, and modify a
third's contents and properties. Here's the patch file
itself (which we'll assume is creatively named
<code class="filename">PATCH</code>):</p><div class="informalexample"><pre class="programlisting">
Index: deleted-file
===================================================================
--- deleted-file (revision 3)
+++ deleted-file (working copy)
@@ -1 +0,0 @@
-This file will be deleted.
Index: changed-file
===================================================================
--- changed-file (revision 4)
+++ changed-file (working copy)
@@ -1,6 +1,6 @@
The letters in a line of text
Could make your day much better.
But expanded into paragraphs,
-I'd tell of kangaroos and calves
+I'd tell of monkeys and giraffes
Until you were all smiles and laughs
From my letter made of letters.
Property changes on: changed-file
___________________________________________________________________
Added: propname
## -0,0 +1 ##
+propvalue
Index: added-file
===================================================================
--- added-file (revision 0)
+++ added-file (working copy)
@@ -0,0 +1 @@
+This is an added file.
</pre></div><p>We can apply the previous patch file to another
working copy from our repository using <span class="command"><strong>svn
patch</strong></span>, and verify that it did the right thing by
using <span class="command"><strong>svn diff</strong></span>:</p><div class="informalexample"><pre class="screen">
$ cd /some/other/workingcopy
$ svn patch /path/to/PATCH
D deleted-file
UU changed-file
A added-file
$ svn diff
Index: deleted-file
===================================================================
--- deleted-file (revision 3)
+++ deleted-file (working copy)
@@ -1 +0,0 @@
-This file will be deleted.
Index: changed-file
===================================================================
--- changed-file (revision 4)
+++ changed-file (working copy)
@@ -1,6 +1,6 @@
The letters in a line of text
Could make your day much better.
But expanded into paragraphs,
-I'd tell of kangaroos and calves
+I'd tell of monkeys and giraffes
Until you were all smiles and laughs
From my letter made of letters.
Property changes on: changed-file
___________________________________________________________________
Added: propname
## -0,0 +1 ##
+propvalue
Index: added-file
===================================================================
--- added-file (revision 0)
+++ added-file (working copy)
@@ -0,0 +1 @@
+This is an added file.
$
</pre></div><p>Sometimes you might need Subversion to interpret a
patch <span class="quote">“<span class="quote">in reverse</span>”</span>—where added things
get treated as removed things, and vice-versa. Use
the <code class="option">--reverse-diff</code> option for this
purpose. In the following example, we'll squirrel away
a patch file which describes the changes in our working
copy, and then use a reverse patch operation to undo those
changes.</p><div class="informalexample"><pre class="screen">
$ svn status
M foo.c
$ svn diff > PATCH
$ cat PATCH
Index: foo.c
===================================================================
--- foo.c (revision 128)
+++ foo.c (working copy)
@@ -1003,7 +1003,7 @@
return ERROR_ON_THE_G_STRING;
/* Do something in a loop. */
- for (i = 0; i < txns->nelts; i++)
+ for (i = 0; i < txns->nelts; i--)
{
status = do_something(i);
if (status)
$ svn patch --reverse-diff PATCH
U foo.c
$ svn status
$
</pre></div></div></div><div class="refentry" title="svn propdel (pdel, pd)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.propdel"></a><div class="titlepage"></div><a id="idp17554976" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn propdel (pdel, pd) — Remove a property from an item.</p></div><div class="refsect1" title="Synopsis"><a id="idp17558048"></a><h2>Synopsis</h2><p><code class="literal">svn propdel PROPNAME [PATH...]</code></p><p><code class="literal">svn propdel PROPNAME --revprop -r REV [TARGET]</code></p></div><div class="refsect1" title="Description"><a id="idp17560528"></a><h2>Description</h2><p>This removes properties from files, directories, or
revisions. The first form removes versioned properties in
your working copy, and the second removes unversioned
remote properties on a repository revision
(<em class="replaceable"><code>TARGET</code></em> determines only which
repository to access).</p></div><div class="refsect1" title="Options"><a id="idp17562480"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.changelist"><code class="option">--changelist</code> (<code class="option">--cl</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.recursive"><code class="option">--recursive</code> (<code class="option">-R</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.revprop"><code class="option">--revprop</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17568384"></a><h2>Examples</h2><p>Delete a property from a file in your working
copy:</p><div class="informalexample"><pre class="screen">
$ svn propdel svn:mime-type some-script
property 'svn:mime-type' deleted from 'some-script'.
</pre></div><p>Delete a revision property:</p><div class="informalexample"><pre class="screen">
$ svn propdel --revprop -r 26 release-date
property 'release-date' deleted from repository revision '26'
</pre></div></div></div><div class="refentry" title="svn propedit (pedit, pe)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.propedit"></a><div class="titlepage"></div><a id="idp17573632" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn propedit (pedit, pe) — Edit the property of one or more items under
version control. See <a class="xref" href="#svn.ref.svn.c.propset" title="svn propset (pset, ps)">svn propset (pset, ps)</a> later in this chapter.</p></div><div class="refsect1" title="Synopsis"><a id="idp17577440"></a><h2>Synopsis</h2><p><code class="literal">svn propedit PROPNAME TARGET...</code></p><p><code class="literal">svn propedit PROPNAME --revprop -r REV [TARGET]</code></p></div><div class="refsect1" title="Description"><a id="idp17579920"></a><h2>Description</h2><p>Edit one or more properties using your favorite
editor. The first form edits versioned properties in
your working copy, and the second edits unversioned
remote properties on a repository revision
(<em class="replaceable"><code>TARGET</code></em> determines only which
repository to access).</p></div><div class="refsect1" title="Options"><a id="idp17581856"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.editor_cmd"><code class="option">--editor-cmd</code> <em class="replaceable"><code>CMD</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.encoding"><code class="option">--encoding</code> <em class="replaceable"><code>ENC</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.file"><code class="option">--file</code> (<code class="option">-F</code>) <em class="replaceable"><code>FILENAME</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.force"><code class="option">--force</code></a>
<a class="xref" href="#svn.ref.svn.sw.force_log"><code class="option">--force-log</code></a>
<a class="xref" href="#svn.ref.svn.sw.message"><code class="option">--message</code> (<code class="option">-m</code>) <em class="replaceable"><code>MESSAGE</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.revprop"><code class="option">--revprop</code></a>
<a class="xref" href="#svn.ref.svn.sw.with_revprop"><code class="option">--with-revprop</code> <em class="replaceable"><code>ARG</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17589600"></a><h2>Examples</h2><p><span class="command"><strong>svn propedit</strong></span> makes it easy to
modify properties that have multiple values:</p><div class="informalexample"><pre class="screen">
$ svn propedit svn:keywords foo.c
# svn will open in your favorite text editor a temporary file
# containing the current contents of the svn:keywords property. You
# can add multiple values to a property easily here by entering one
# value per line. When you save the temporary file and exit,
# Subversion will re-read the temporary file and use its updated
# contents as the new value of the property.
Set new value for property 'svn:keywords' on 'foo.c'
$
</pre></div></div></div><div class="refentry" title="svn propget (pget, pg)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.propget"></a><div class="titlepage"></div><a id="idp17594144" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn propget (pget, pg) — Print the value of a property.</p></div><div class="refsect1" title="Synopsis"><a id="idp17597216"></a><h2>Synopsis</h2><p><code class="literal">svn propget PROPNAME [TARGET[@REV]...]</code></p><p><code class="literal">svn propget PROPNAME --revprop -r REV [URL]</code></p></div><div class="refsect1" title="Description"><a id="idp17599696"></a><h2>Description</h2><p>Print the value of a property on files,
directories, or revisions. The first form prints the
versioned property of an item or items in your working
copy, and the second prints unversioned remote
properties on a repository revision. See <a class="xref" href="#svn.advanced.props" title="Properties">the section called “Properties”</a> for more information on
properties.</p></div><div class="refsect1" title="Options"><a id="idp17601920"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.changelist"><code class="option">--changelist</code> (<code class="option">--cl</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.recursive"><code class="option">--recursive</code> (<code class="option">-R</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.revprop"><code class="option">--revprop</code></a>
<a class="xref" href="#svn.ref.svn.sw.strict"><code class="option">--strict</code></a>
<a class="xref" href="#svn.ref.svn.sw.verbose"><code class="option">--verbose</code> (<code class="option">-v</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.xml"><code class="option">--xml</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17609008"></a><h2>Examples</h2><p>Examine a property of a file in your working
copy:</p><div class="informalexample"><pre class="screen">
$ svn propget svn:keywords foo.c
Author
Date
Rev
</pre></div><p>The same goes for a revision property:</p><div class="informalexample"><pre class="screen">
$ svn propget svn:log --revprop -r 20
Began journal.
</pre></div><p>For a more structured display of properties, use
the <code class="option">--verbose</code> (<code class="option">-v</code>)
option:</p><div class="informalexample"><pre class="screen">
$ svn propget svn:keywords foo.c --verbose
Properties on 'foo.c':
svn:keywords
Author
Date
Rev
</pre></div><p>By default, <span class="command"><strong>svn propget</strong></span> will append
a trailing end-of-line sequence to the property value it
prints. Most of the time, this is a desirable feature
that has a positive effect on the printed output. But
there are times when you might wish to capture the precise
property value, perhaps because that value is not textual
in nature, but of some binary format (such as a JPEG
thumbnail stored as a property value, for example). To
disable pretty-printing of property values, use
the <code class="option">--strict</code> option.</p><p>Lastly, you can get <span class="command"><strong>svn propget</strong></span>
output in XML format with the <code class="option">--xml</code>
option:</p><div class="informalexample"><pre class="screen">
$ svn propget --xml svn:ignore .
<?xml version="1.0"?>
<properties>
<target
path="">
<property
name="svn:ignore">*.o
</property>
</target>
</properties>
</pre></div></div></div><div class="refentry" title="svn proplist (plist, pl)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.proplist"></a><div class="titlepage"></div><a id="idp17621472" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn proplist (plist, pl) — List all properties.</p></div><div class="refsect1" title="Synopsis"><a id="idp17624544"></a><h2>Synopsis</h2><p><code class="literal">svn proplist [TARGET[@REV]...]</code></p><p><code class="literal">svn proplist --revprop -r REV [TARGET]</code></p></div><div class="refsect1" title="Description"><a id="idp17627008"></a><h2>Description</h2><p>List all properties on files, directories, or
revisions. The first form lists versioned properties in
your working copy, and the second lists unversioned remote
properties on a repository revision
(<em class="replaceable"><code>TARGET</code></em> determines only which
repository to access).</p></div><div class="refsect1" title="Options"><a id="idp17628944"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.changelist"><code class="option">--changelist</code> (<code class="option">--cl</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.recursive"><code class="option">--recursive</code> (<code class="option">-R</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.revprop"><code class="option">--revprop</code></a>
<a class="xref" href="#svn.ref.svn.sw.verbose"><code class="option">--verbose</code> (<code class="option">-v</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.xml"><code class="option">--xml</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17636000"></a><h2>Examples</h2><p>You can use <span class="command"><strong>proplist</strong></span> to see the properties on an
item in your working copy:</p><div class="informalexample"><pre class="screen">
$ svn proplist foo.c
Properties on 'foo.c':
svn:mime-type
svn:keywords
owner
</pre></div><p>But with the <code class="option">--verbose</code>
(<code class="option">-v</code>) flag, <span class="command"><strong>svn
proplist</strong></span> is extremely handy as it also shows you
the values for the properties:</p><div class="informalexample"><pre class="screen">
$ svn proplist -v foo.c
Properties on 'foo.c':
svn:mime-type
text/plain
svn:keywords
Author Date Rev
owner
sally
</pre></div><p>Lastly, you can get <span class="command"><strong>svn proplist</strong></span>
output in xml format with the <code class="option">--xml</code>
option:</p><div class="informalexample"><pre class="screen">
$ svn proplist --xml
<?xml version="1.0"?>
<properties>
<target
path=".">
<property
name="svn:ignore"/>
</target>
</properties>
</pre></div></div></div><div class="refentry" title="svn propset (pset, ps)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.propset"></a><div class="titlepage"></div><a id="idp17646496" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn propset (pset, ps) — Set <em class="replaceable"><code>PROPNAME</code></em>
to <em class="replaceable"><code>PROPVAL</code></em> on files, directories,
or revisions.</p></div><div class="refsect1" title="Synopsis"><a id="idp17650480"></a><h2>Synopsis</h2><p><code class="literal">svn propset PROPNAME [PROPVAL | -F VALFILE] PATH...</code></p><p><code class="literal">svn propset PROPNAME --revprop -r REV [PROPVAL | -F VALFILE] [TARGET]</code></p></div><div class="refsect1" title="Description"><a id="idp17652992"></a><h2>Description</h2><p>Set <em class="replaceable"><code>PROPNAME</code></em> to
<em class="replaceable"><code>PROPVAL</code></em> on files, directories,
or revisions. The first example creates a versioned, local
property change in the working copy, and the second
creates an unversioned, remote property change on a
repository revision
(<em class="replaceable"><code>TARGET</code></em> determines only which
repository to access).</p><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>Subversion has a number of <span class="quote">“<span class="quote">special</span>”</span>
properties that affect its behavior. See <a class="xref" href="#svn.ref.properties" title="Subversion Properties">the section called “Subversion Properties”</a> later in this chapter for more on these
properties.</p></td></tr></table></div></div><div class="refsect1" title="Options"><a id="idp17657920"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.changelist"><code class="option">--changelist</code> (<code class="option">--cl</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.encoding"><code class="option">--encoding</code> <em class="replaceable"><code>ENC</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.file"><code class="option">--file</code> (<code class="option">-F</code>) <em class="replaceable"><code>FILENAME</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.force"><code class="option">--force</code></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.recursive"><code class="option">--recursive</code> (<code class="option">-R</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.revprop"><code class="option">--revprop</code></a>
<a class="xref" href="#svn.ref.svn.sw.targets"><code class="option">--targets</code> <em class="replaceable"><code>FILENAME</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17666224"></a><h2>Examples</h2><p>Set the MIME type for a file:</p><div class="informalexample"><pre class="screen">
$ svn propset svn:mime-type image/jpeg foo.jpg
property 'svn:mime-type' set on 'foo.jpg'
</pre></div><p>On a Unix system, if you want a file to have the
executable permission set:</p><div class="informalexample"><pre class="screen">
$ svn propset svn:executable ON somescript
property 'svn:executable' set on 'somescript'
</pre></div><p>Perhaps you have an internal policy to set certain
properties for the benefit of your coworkers:</p><div class="informalexample"><pre class="screen">
$ svn propset owner sally foo.c
property 'owner' set on 'foo.c'
</pre></div><p>If you made a mistake in a log message for a
particular revision and want to change it, use
<code class="option">--revprop</code> and set <code class="literal">svn:log</code>
to the new log message:</p><div class="informalexample"><pre class="screen">
$ svn propset --revprop -r 25 svn:log "Journaled about trip to New York."
property 'svn:log' set on repository revision '25'
</pre></div><p>Or, if you don't have a working copy, you can
provide a URL:</p><div class="informalexample"><pre class="screen">
$ svn propset --revprop -r 26 svn:log "Document nap." \
http://svn.red-bean.com/repos
property 'svn:log' set on repository revision '25'
</pre></div><p>Lastly, you can tell <span class="command"><strong>propset</strong></span> to take its input from
a file. You could even use this to set the contents of a
property to something binary:</p><div class="informalexample"><pre class="screen">
$ svn propset owner-pic -F sally.jpg moo.c
property 'owner-pic' set on 'moo.c'
</pre></div><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>By default, you cannot modify revision properties
in a Subversion repository. Your repository
administrator must explicitly enable revision property
modifications by creating a hook named
<code class="literal">pre-revprop-change</code>. See <a class="xref" href="#svn.reposadmin.create.hooks" title="Implementing Repository Hooks">the section called “Implementing Repository Hooks”</a> for more information on
hook scripts.</p></td></tr></table></div></div></div><div class="refentry" title="svn relocate"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.relocate"></a><div class="titlepage"></div><a id="idp17682480" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn relocate — Relocate the working copy to point to a
different repository root URL.</p></div><div class="refsect1" title="Synopsis"><a id="idp17685600"></a><h2>Synopsis</h2><p><code class="literal">svn relocate FROM-PREFIX TO-PREFIX [PATH...]</code></p><p><code class="literal">svn relocate TO-URL [PATH]</code></p></div><div class="refsect1" title="Description"><a id="idp17688032"></a><h2>Description</h2><p>Sometimes an administrator might change the location
(or apparent location, from the client's perspective) of a
repository. The content of the repository doesn't change,
but the repository's root URL does. The hostname may
change because the repository is now being served from a
different computer. Or, perhaps the URL scheme changes
because the repository is now being served via SSL
(using <code class="literal">https://</code>) instead of over plain
HTTP. There are many different reasons for these types of
repository relocations. But ideally, a <span class="quote">“<span class="quote">change of
address</span>”</span> for a repository shouldn't suddently cause
all the working copies which point to that repository to
become forever unusable. And fortunately, that's not the
case. Rather than force users to check out a new working
copy when a repository is relocated, Subversion provides
the <span class="command"><strong>svn relocate</strong></span> command, which
<span class="quote">“<span class="quote">rewrites</span>”</span> the working copy's administrative
metadata to refer to the new repository location.</p><p>The first <span class="command"><strong>svn relocate</strong></span> syntax
allows you to update one or more working copies by what
essentially amounts to a find-and-replace within the
repository root URLs recorded in those working copies.
Subversion will replace the initial substring
<em class="replaceable"><code>FROM-PREFIX</code></em> with the
string <em class="replaceable"><code>TO-PREFIX</code></em> in those URLs.
These initial URL substrings can be as long or as short as
is necessary to differentiate between them. Obviously, to
use this syntax form, you need to know both the current
root URL of the repository to which the working copy is
pointing, and the new URL of that repository.
(You can use <span class="command"><strong>svn info</strong></span> to determine
the former.)</p><p>The second syntax does not require that you know the
current repository root URL with which the working copy is
associated at all—only the new repository URL
(<em class="replaceable"><code>TO-URL</code></em>) to which it should be
pointing. In this syntax form, only one working copy may
be relocated at a time.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>Users often get confused about the difference
between <span class="command"><strong>svn switch</strong></span> and <span class="command"><strong>svn
relocate</strong></span>. Here's the rule of thumb:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>If the working copy needs to reflect a new
directory <span class="emphasis"><em>within</em></span> the
repository, use <span class="command"><strong>svn
switch</strong></span>.</p></li><li class="listitem"><p>If the working copy still reflects the
same repository directory, but the location of the
repository itself has changed, use <span class="command"><strong>svn
relocate</strong></span>.</p></li></ul></div></td></tr></table></div></div><div class="refsect1" title="Options"><a id="idp17703552"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.ignore_externals"><code class="option">--ignore-externals</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17706112"></a><h2>Examples</h2><p>Let's start with a working copy that reflects a local
repository URL:</p><div class="informalexample"><pre class="screen">
$ svn info | grep URL:
URL: file:///var/svn/repos/trunk
$
</pre></div><p>One day the administrator decides to rename the
on-disk repository directory. We missed the memo, so we
see an error the next time we try to update our working
copy.</p><div class="informalexample"><pre class="screen">
$ svn up
Updating '.':
svn: E180001: Unable to connect to a repository at URL 'file:///var/svn/repos/trunk'
</pre></div><p>After cornering the administrator over by the vending
machines, we learn about the repository being moved and
are told the new URL. Rather than checkout a new working
copy, though, we simply ask Subversion to rewrite the
working copy metadata to point to the new repository
location.</p><div class="informalexample"><pre class="screen">
$ svn relocate file:///var/svn/new-repos/trunk
$
</pre></div><p>Subversion doesn't tell us much about what it did, but
hey—error-free operation is really all we need,
right? Our working copy is functional for online
operations again.</p><div class="informalexample"><pre class="screen">
$ svn up
Updating '.':
A lib/new.c
M src/code.h
M src/headers.h
…
</pre></div><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>Once again, this type of relocation
affects <span class="emphasis"><em>working copy metadata only</em></span>.
It will not change your versioned or unversioned file
contents, perform any version control operations (such
as commits or updates), and so on.</p></td></tr></table></div><p>A few months later, we're told that the company is
moving development to separate machines and that we'll be
using HTTP to access the repository. So we relocate our
working copy again.</p><div class="informalexample"><pre class="screen">
$ svn relocate http://svn.company.com/repos/trunk
$
</pre></div><p>Now, each time we perform a relocation of this sort,
Subversion contacts the repository—at its new URL,
of course—to verify a few things.</p><p>First, it wants to compare the UUID of the repository
against what is stored in the working copy. If these UUIDs
don't match, the working copy relocation is disallowed.
Maybe this isn't the same repository (just in a new
location) after all?</p><p>Secondly, Subversion wants to ensure that the updated
working copy metadata jives with respect to the directory
location <span class="emphasis"><em>inside</em></span> the repository.
Subversion won't let you accidentally relocate a working
copy of a branch in your repository to the URL of a
different branch in the same repository. (That's
what <span class="command"><strong>svn switch</strong></span>, described in
<a class="xref" href="#svn.ref.svn.c.switch" title="svn switch (sw)">svn switch (sw)</a>, is for.)</p><p>Also, Subversion will not allow you to relocate a
subtree of the working copy. If you're going to relocate
the working copy at all, you must relocate the whole
thing. This is to protect the integrity of the working
copy metadata and behavior as a whole. (And really, you'd
be hard pressed to come up with a compelling reason to
relocate only a piece of your working copy anyway.)</p><p>Let's look at one final relocation opportunity. After
using HTTP access for some time, the company moves to
SSL-only access. Now, the only change to the repository
URL is that the scheme goes from
being <code class="literal">http://</code> to
being <code class="literal">https://</code>. There are two
different ways that we could make our working copy reflect
ths change. The first is to do exactly as we've done
before and relocate to the new repository URL.</p><div class="informalexample"><pre class="screen">
$ svn relocate https://svn.company.com/repos/trunk
$
</pre></div><p>But we have another option here, too. We could simply
ask Subversion to swap out the changed prefixes of the
URL.</p><div class="informalexample"><pre class="screen">
$ svn relocate http https
$
</pre></div><p>Either approach leaves us a working copy whose
metadata has been updated to point to the right repository
location.</p><p>By default, <span class="command"><strong>svn relocate</strong></span> will
traverse any external working copies nested within your
working copy and attempt relocation of those working
copies, too. Use the <code class="option">--ignore-externals</code>
option to disable this behavior.</p></div></div><div class="refentry" title="svn resolve"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.resolve"></a><div class="titlepage"></div><a id="idp17730448" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn resolve — Resolve conflicts on working copy files or
directories.</p></div><div class="refsect1" title="Synopsis"><a id="idp17733552"></a><h2>Synopsis</h2><p><code class="literal">svn resolve PATH...</code></p></div><div class="refsect1" title="Description"><a id="idp17735168"></a><h2>Description</h2><p>Resolve <span class="quote">“<span class="quote">conflicted</span>”</span> state on working
copy files or directories. This routine does not
semantically resolve conflict markers; however, it
replaces <em class="replaceable"><code>PATH</code></em> with the version
specified by the <code class="option">--accept</code> argument and
then removes conflict-related artifact files. This allows
<em class="replaceable"><code>PATH</code></em> to be committed
again—that is, it tells Subversion that the
conflicts have been <span class="quote">“<span class="quote">resolved.</span>”</span></p><p>See <a class="xref" href="#svn.tour.cycle.resolve" title="Resolve Any Conflicts">the section called “Resolve Any Conflicts”</a> for an in-depth look at
resolving conflicts.</p></div><div class="refsect1" title="Options"><a id="idp17739984"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.accept"><code class="option">--accept</code> <em class="replaceable"><code>ACTION</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.recursive"><code class="option">--recursive</code> (<code class="option">-R</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.targets"><code class="option">--targets</code> <em class="replaceable"><code>FILENAME</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17745056"></a><h2>Examples</h2><p>Here's an example where, after a postponed conflict
resolution during update, <span class="command"><strong>svn resolve</strong></span>
replaces the all conflicts in
file <code class="filename">foo.c</code> with your edits:</p><div class="informalexample"><pre class="screen">
$ svn update
Updating '.':
Conflict discovered in 'foo.c'.
Select: (p) postpone, (df) diff-full, (e) edit,
(mc) mine-conflict, (tc) theirs-conflict,
(s) show all options: p
C foo.c
Updated to revision 5.
Summary of conflicts:
Text conflicts: 1
$ svn resolve --accept mine-full foo.c
Resolved conflicted state of 'foo.c'
$
</pre></div></div></div><div class="refentry" title="svn resolved"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.resolved"></a><div class="titlepage"></div><a id="idp17750432" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn resolved — <span class="emphasis"><em>Deprecated</em></span>.
Remove <span class="quote">“<span class="quote">conflicted</span>”</span> state on working copy
files or directories.</p></div><div class="refsect1" title="Synopsis"><a id="idp17754272"></a><h2>Synopsis</h2><p><code class="literal">svn resolved PATH...</code></p></div><div class="refsect1" title="Description"><a id="idp17755888"></a><h2>Description</h2><p>This command has been deprecated in favor of
running <strong class="userinput"><code>svn resolve --accept working <em class="replaceable"><code>PATH</code></em></code></strong>.
See <a class="xref" href="#svn.ref.svn.c.resolve" title="svn resolve">svn resolve</a> in the preceding section for
details.</p><p>Remove <span class="quote">“<span class="quote">conflicted</span>”</span> state on working copy
files or directories. This routine does not semantically
resolve conflict markers; it merely removes
conflict-related artifact files and allows
<em class="replaceable"><code>PATH</code></em> to be committed again;
that is, it tells Subversion that the conflicts have been
<span class="quote">“<span class="quote">resolved.</span>”</span> See <a class="xref" href="#svn.tour.cycle.resolve" title="Resolve Any Conflicts">the section called “Resolve Any Conflicts”</a> for an in-depth look at
resolving conflicts.</p></div><div class="refsect1" title="Options"><a id="idp17761824"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.recursive"><code class="option">--recursive</code> (<code class="option">-R</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.targets"><code class="option">--targets</code> <em class="replaceable"><code>FILENAME</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17766256"></a><h2>Examples</h2><p>If you get a conflict on an update, your working
copy will sprout three new files:</p><div class="informalexample"><pre class="screen">
$ svn update
Updating '.':
C foo.c
Updated to revision 31.
Summary of conflicts:
Text conflicts: 1
$ ls foo.c*
foo.c
foo.c.mine
foo.c.r30
foo.c.r31
$
</pre></div><p>Once you've resolved the conflict and
<code class="filename">foo.c</code> is ready to be committed, run
<span class="command"><strong>svn resolved</strong></span> to let your working copy
know you've taken care of everything.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>You <span class="emphasis"><em>can</em></span> just remove the
conflict files and commit, but <span class="command"><strong>svn
resolved</strong></span> fixes up some bookkeeping data
in the working copy administrative area in addition
to removing the conflict files, so we recommend that
you use this command.</p></td></tr></table></div></div></div><div class="refentry" title="svn revert"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.revert"></a><div class="titlepage"></div><a id="idp17774112" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn revert — Undo all local edits.</p></div><div class="refsect1" title="Synopsis"><a id="idp17777168"></a><h2>Synopsis</h2><p><code class="literal">svn revert PATH...</code></p></div><div class="refsect1" title="Description"><a id="idp17778784"></a><h2>Description</h2><p>Reverts any local changes to a file or directory and
resolves any conflicted states. <span class="command"><strong>svn
revert</strong></span> will revert not only the contents of an
item in your working copy, but also any property
changes. Finally, you can use it to undo any scheduling
operations that you may have performed (e.g., files scheduled
for addition or deletion can be
<span class="quote">“<span class="quote">unscheduled</span>”</span>).</p></div><div class="refsect1" title="Options"><a id="idp17781504"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.changelist"><code class="option">--changelist</code> (<code class="option">--cl</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.recursive"><code class="option">--recursive</code> (<code class="option">-R</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.targets"><code class="option">--targets</code> <em class="replaceable"><code>FILENAME</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17786640"></a><h2>Examples</h2><p>Discard changes to a file:</p><div class="informalexample"><pre class="screen">
$ svn revert foo.c
Reverted foo.c
</pre></div><p>If you want to revert a whole directory of files,
use the <code class="option">--depth=infinity</code> option:</p><div class="informalexample"><pre class="screen">
$ svn revert --depth=infinity .
Reverted newdir/afile
Reverted foo.c
Reverted bar.txt
</pre></div><p>Lastly, you can undo any scheduling
operations:</p><div class="informalexample"><pre class="screen">
$ svn add mistake.txt whoops
A mistake.txt
A whoops
A whoops/oopsie.c
$ svn revert mistake.txt whoops
Reverted mistake.txt
Reverted whoops
$ svn status
? mistake.txt
? whoops
</pre></div><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p><span class="command"><strong>svn revert</strong></span> is inherently
dangerous, since its entire purpose is to throw away
data—namely, your uncommitted changes. Once
you've reverted, Subversion provides <span class="emphasis"><em>no
way</em></span> to get back those uncommitted
changes.</p><p>If you provide no targets to <span class="command"><strong>svn
revert</strong></span>, it will do nothing. To protect you
from accidentally losing changes in your working
copy, <span class="command"><strong>svn revert</strong></span> requires you to
explicitly provide at least one target.</p></td></tr></table></div></div></div><div class="refentry" title="svn status (stat, st)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.status"></a><div class="titlepage"></div><a id="idp17797984" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn status (stat, st) — Print the status of working copy files and directories.</p></div><div class="refsect1" title="Synopsis"><a id="idp17801072"></a><h2>Synopsis</h2><p><code class="literal">svn status [PATH...]</code></p></div><div class="refsect1" title="Description"><a id="idp17802688"></a><h2>Description</h2><p>Print the status of working copy files and
directories. With no arguments, it prints only locally
modified items (no repository access). With
<code class="option">--show-updates</code> (<code class="option">-u</code>), it
adds working revision and server out-of-date information.
With <code class="option">--verbose</code> (<code class="option">-v</code>), it
prints full revision information on every item.
With <code class="option">--quiet</code> (<code class="option">-q</code>), it
prints only summary information about locally modified
items.</p><p>The first seven columns in the output are each one
character wide, and each column gives you information
about a different aspect of each working copy item.</p><p>The first column indicates that an item
was added, deleted, or otherwise changed:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">' '</code></span></dt><dd><p>No modifications.</p></dd><dt><span class="term"><code class="literal">'A'</code></span></dt><dd><p>Item is scheduled for addition.</p></dd><dt><span class="term"><code class="literal">'D'</code></span></dt><dd><p>Item is scheduled for deletion.</p></dd><dt><span class="term"><code class="literal">'M'</code></span></dt><dd><p>Item has been modified.</p></dd><dt><span class="term"><code class="literal">'R'</code></span></dt><dd><p>Item has been replaced in your working copy.
This means the file was scheduled for deletion, and
then a new file with the same name was scheduled for
addition in its place.</p></dd><dt><span class="term"><code class="literal">'C'</code></span></dt><dd><p>The contents (as opposed to the properties) of
the item conflict with updates received from the
repository.</p></dd><dt><span class="term"><code class="literal">'X'</code></span></dt><dd><p>Item is present because of an externals definition.</p></dd><dt><span class="term"><code class="literal">'I'</code></span></dt><dd><p>Item is being ignored (e.g., with the
<code class="literal">svn:ignore</code> property).</p></dd><dt><span class="term"><code class="literal">'?'</code></span></dt><dd><p>Item is not under version control.</p></dd><dt><span class="term"><code class="literal">'!'</code></span></dt><dd><p>Item is missing (e.g., you moved or deleted it
without using <span class="command"><strong>svn</strong></span>). This also
indicates that a directory is incomplete (a checkout
or update was interrupted).</p></dd><dt><span class="term"><code class="literal">'~'</code></span></dt><dd><p>Item is versioned as one kind of object (file,
directory, link), but has been replaced by a different
kind of object.</p></dd></dl></div><p>The second column tells the status of a file's or
directory's properties:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">' '</code></span></dt><dd><p>No modifications.</p></dd><dt><span class="term"><code class="literal">'M'</code></span></dt><dd><p>Properties for this item have been
modified.</p></dd><dt><span class="term"><code class="literal">'C'</code></span></dt><dd><p>Properties for this item are in conflict with
property updates received from the
repository.</p></dd></dl></div><p>The third column is populated only if the
working copy directory is locked (see
<a class="xref" href="#svn.tour.cleanup" title="Sometimes You Just Need to Clean Up">the section called “Sometimes You Just Need to Clean Up”</a>):</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">' '</code></span></dt><dd><p>Item is not locked.</p></dd><dt><span class="term"><code class="literal">'L'</code></span></dt><dd><p>Item is locked.</p></dd></dl></div><p>The fourth column is populated only if the item is
scheduled for addition-with-history:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">' '</code></span></dt><dd><p>No history scheduled with commit.</p></dd><dt><span class="term"><code class="literal">'+'</code></span></dt><dd><p>History scheduled with commit.</p></dd></dl></div><p>The fifth column is populated only if the item is
switched relative to its parent (see <a class="xref" href="#svn.branchmerge.switchwc" title="Traversing Branches">the section called “Traversing Branches”</a>):</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">' '</code></span></dt><dd><p>Item is a child of its parent directory.</p></dd><dt><span class="term"><code class="literal">'S'</code></span></dt><dd><p>Item is switched.</p></dd></dl></div><p>The sixth column is populated with lock information:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">' '</code></span></dt><dd><p>When <code class="option">--show-updates</code>
(<code class="option">-u</code>) is used, this means the file is not
locked. If <code class="option">--show-updates</code>
(<code class="option">-u</code>) is <span class="emphasis"><em>not</em></span>
used, this merely means that the file is not locked
in this working copy.</p></dd><dt><span class="term"><code class="literal">'K'</code></span></dt><dd><p>File is locked in this working copy.</p></dd><dt><span class="term"><code class="literal">'O'</code></span></dt><dd><p>File is locked either by another user or in
another working copy. This appears only when
<code class="option">--show-updates</code>
(<code class="option">-u</code>) is used.</p></dd><dt><span class="term"><code class="literal">'T'</code></span></dt><dd><p>File was locked in this working copy, but the
lock has been <span class="quote">“<span class="quote">stolen</span>”</span> and is invalid.
The file is currently locked in the repository. This
appears only when <code class="option">--show-updates</code>
(<code class="option">-u</code>) is used.</p></dd><dt><span class="term"><code class="literal">'B'</code></span></dt><dd><p>File was locked in this working copy, but the
lock has been <span class="quote">“<span class="quote">broken</span>”</span> and is invalid.
The file is no longer locked. This appears only
when <code class="option">--show-updates</code>
(<code class="option">-u</code>) is used.</p></dd></dl></div><p>The seventh column is populated only if the item is
the victim of a tree conflict:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">' '</code></span></dt><dd><p>Item is not the victim of a tree conflict.</p></dd><dt><span class="term"><code class="literal">'C'</code></span></dt><dd><p>Item is the victim of a tree conflict.</p></dd></dl></div><p>The eighth column is always blank.</p><p>The out-of-date information appears in the ninth
column (only if you pass the
<code class="option">--show-updates</code> (<code class="option">-u</code>)
option):</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">' '</code></span></dt><dd><p>The item in your working copy is up to date.</p></dd><dt><span class="term"><code class="literal">'*'</code></span></dt><dd><p>A newer revision of the item exists on the
server.</p></dd></dl></div><p>The remaining fields are variable width and delimited
by spaces. The working revision is the next field if
the <code class="option">--show-updates</code> (<code class="option">-u</code>)
or <code class="option">--verbose</code> (<code class="option">-v</code>) option
is passed.</p><p>If the <code class="option">--verbose</code>
(<code class="option">-v</code>) option is passed, the last committed
revision and last committed author are displayed
next.</p><p>The working copy path is always the final field, so
it can include spaces.</p></div><div class="refsect1" title="Options"><a id="idp17887632"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.changelist"><code class="option">--changelist</code> (<code class="option">--cl</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.ignore_externals"><code class="option">--ignore-externals</code></a>
<a class="xref" href="#svn.ref.svn.sw.incremental"><code class="option">--incremental</code></a>
<a class="xref" href="#svn.ref.svn.sw.no_ignore"><code class="option">--no-ignore</code></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.show_updates"><code class="option">--show-updates</code> (<code class="option">-u</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.verbose"><code class="option">--verbose</code> (<code class="option">-v</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.xml"><code class="option">--xml</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17895488"></a><h2>Examples</h2><p>This is the easiest way to find out what changes you
have made to your working copy:</p><div class="informalexample"><pre class="screen">
$ svn status wc
M wc/bar.c
A + wc/qax.c
</pre></div><p>If you want to find out what files in your working
copy are out of date, pass
the <code class="option">--show-updates</code> (<code class="option">-u</code>)
option (this will <span class="emphasis"><em>not</em></span> make any
changes to your working copy). Here you can see that
<code class="filename">wc/foo.c</code> has changed in the
repository since we last updated our working
copy:</p><div class="informalexample"><pre class="screen">
$ svn status -u wc
M 965 wc/bar.c
* 965 wc/foo.c
A + 965 wc/qax.c
Status against revision: 981
</pre></div><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p><code class="option">--show-updates</code>
(<code class="option">-u</code>) <span class="emphasis"><em>only</em></span> places
an asterisk next to items that are out of date (i.e.,
items that will be updated from the repository if you
later use <span class="command"><strong>svn update</strong></span>).
<code class="option">--show-updates</code> (<code class="option">-u</code>)
does <span class="emphasis"><em>not</em></span> cause the status listing
to reflect the repository's version of the item
(although you can see the revision number in the
repository by passing the <code class="option">--verbose</code>
(<code class="option">-v</code>) option).</p></td></tr></table></div><p>The most information you can get out of
the status subcommand is as follows:</p><div class="informalexample"><pre class="screen">
$ svn status -u -v wc
M 965 938 sally wc/bar.c
* 965 922 harry wc/foo.c
A + 965 687 harry wc/qax.c
965 687 harry wc/zig.c
Status against revision: 981
</pre></div><p>Lastly, you can get <span class="command"><strong>svn status</strong></span>
output in XML format with the <code class="option">--xml</code>
option:</p><div class="informalexample"><pre class="screen">
$ svn status --xml wc
<?xml version="1.0"?>
<status>
<target
path="wc">
<entry
path="qax.c">
<wc-status
props="none"
item="added"
revision="0">
</wc-status>
</entry>
<entry
path="bar.c">
<wc-status
props="normal"
item="modified"
revision="965">
<commit
revision="965">
<author>sally</author>
<date>2008-05-28T06:35:53.048870Z</date>
</commit>
</wc-status>
</entry>
</target>
</status>
</pre></div><p>For many more examples of <span class="command"><strong>svn status</strong></span>,
see <a class="xref" href="#svn.tour.cycle.examine.status" title="See an overview of your changes">the section called “See an overview of your changes”</a>.</p></div></div><div class="refentry" title="svn switch (sw)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.switch"></a><div class="titlepage"></div><a id="idp17915104" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn switch (sw) — Update working copy to a different URL.</p></div><div class="refsect1" title="Synopsis"><a id="idp17918128"></a><h2>Synopsis</h2><p><code class="literal">svn switch URL[@PEGREV] [PATH]</code></p><p><code class="literal">switch --relocate FROM TO [PATH...]</code></p></div><div class="refsect1" title="Description"><a id="idp17920592"></a><h2>Description</h2><p>The first variant of this subcommand (without the
<code class="option">--relocate</code> option) updates your working
copy to point to a new URL. This is the Subversion way to
make a working copy begin tracking a new branch. If
specified, <em class="replaceable"><code>PEGREV</code></em> determines in
which revision the target is first looked up. See
<a class="xref" href="#svn.branchmerge.switchwc" title="Traversing Branches">the section called “Traversing Branches”</a> for an in-depth
look at switching.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>Beginning with Subversion 1.7, the <span class="command"><strong>svn
switch</strong></span> command will demand by default that the
URL to which you are switching your working copy shares
a common ancestry with item that the working copy
currently reflects. You can override this behavior by
specifying the <code class="option">--ignore-ancestry</code>
option.</p></td></tr></table></div><p>If <code class="option">--force</code> is used, unversioned
obstructing paths in the working copy do not automatically
cause a failure if the switch attempts to add the same
path. If the obstructing path is the same type (file or
directory) as the corresponding path in the repository, it
becomes versioned but its contents are left untouched in
the working copy. This means that an obstructing
directory's unversioned children may also obstruct and
become versioned. For files, any content differences
between the obstruction and the repository are treated
like a local modification to the working copy. All
properties from the repository are applied to the
obstructing path.</p><p>As with most subcommands, you can limit the scope of
the switch operation to a particular tree depth using the
<code class="option">--depth</code> option. Alternatively, you can
use the <code class="option">--set-depth</code> option to set a new
<span class="quote">“<span class="quote">sticky</span>”</span> working copy depth on the switch
target.</p><p>The <code class="option">--relocate</code> option is deprecated
as of Subversion 1.7. Use <span class="command"><strong>svn relocate</strong></span>
(described in <a class="xref" href="#svn.ref.svn.c.relocate" title="svn relocate">svn relocate</a>)
to perform working copy relocation instead.</p></div><div class="refsect1" title="Options"><a id="idp17931952"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.accept"><code class="option">--accept</code> <em class="replaceable"><code>ACTION</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.diff3_cmd"><code class="option">--diff3-cmd</code> <em class="replaceable"><code>CMD</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.force"><code class="option">--force</code></a>
<a class="xref" href="#svn.ref.svn.sw.ignore_ancestry"><code class="option">--ignore-ancestry</code></a>
<a class="xref" href="#svn.ref.svn.sw.ignore_externals"><code class="option">--ignore-externals</code></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.relocate"><code class="option">--relocate</code></a>
<a class="xref" href="#svn.ref.svn.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.set_depth"><code class="option">--set-depth</code> <em class="replaceable"><code>ARG</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17940592"></a><h2>Examples</h2><p>If you're currently inside the directory
<code class="filename">vendors</code>, which was branched to
<code class="filename">vendors-with-fix</code>, and you'd like to
switch your working copy to that branch:</p><div class="informalexample"><pre class="screen">
$ svn switch http://svn.red-bean.com/repos/branches/vendors-with-fix .
U myproj/foo.txt
U myproj/bar.txt
U myproj/baz.c
U myproj/qux.c
Updated to revision 31.
</pre></div><p>To switch back, just provide the URL to the
location in the repository from which you originally
checked out your working copy:</p><div class="informalexample"><pre class="screen">
$ svn switch http://svn.red-bean.com/repos/trunk/vendors .
U myproj/foo.txt
U myproj/bar.txt
U myproj/baz.c
U myproj/qux.c
Updated to revision 31.
</pre></div><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>You <span class="emphasis"><em>can</em></span> switch just part of
your working copy to a branch if you don't want to
switch your entire working copy, but this is not
generally recommended. It's too easy to forget that
you've done this and wind up accidentally making and
committing changes both to the switched and unswitched
portions of your tree.</p></td></tr></table></div></div></div><div class="refentry" title="svn unlock"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.unlock"></a><div class="titlepage"></div><a id="idp17949104" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn unlock — Unlock working copy paths or URLs.</p></div><div class="refsect1" title="Synopsis"><a id="idp17952176"></a><h2>Synopsis</h2><p><code class="literal">svn unlock TARGET...</code></p></div><div class="refsect1" title="Description"><a id="idp17953792"></a><h2>Description</h2><p>Unlock each <em class="replaceable"><code>TARGET</code></em>. If any
<em class="replaceable"><code>TARGET</code></em> is locked by
another user or no valid lock token exists in the working
copy, print a warning and continue unlocking the rest of
the <em class="replaceable"><code>TARGET</code></em>s. Use
<code class="option">--force</code> to break a lock belonging to
another user or working copy.</p></div><div class="refsect1" title="Options"><a id="idp17957008"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.force"><code class="option">--force</code></a>
<a class="xref" href="#svn.ref.svn.sw.targets"><code class="option">--targets</code> <em class="replaceable"><code>FILENAME</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp17960160"></a><h2>Examples</h2><p>Unlock two files in your working copy:</p><div class="informalexample"><pre class="screen">
$ svn unlock tree.jpg house.jpg
'tree.jpg' unlocked.
'house.jpg' unlocked.
</pre></div><p>Unlock a file in your working copy that is currently
locked by another user:</p><div class="informalexample"><pre class="screen">
$ svn unlock tree.jpg
svn: E195013: 'tree.jpg' is not locked in this working copy
$ svn unlock --force tree.jpg
'tree.jpg' unlocked.
</pre></div><p>Unlock a file without a working copy:</p><div class="informalexample"><pre class="screen">
$ svn unlock http://svn.red-bean.com/repos/test/tree.jpg
'tree.jpg unlocked.
</pre></div><p>For further details, see <a class="xref" href="#svn.advanced.locking" title="Locking">the section called “Locking”</a>.</p></div></div><div class="refentry" title="svn update (up)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.update"></a><div class="titlepage"></div><a id="idp17968032" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn update (up) — Update your working copy.</p></div><div class="refsect1" title="Synopsis"><a id="idp17971104"></a><h2>Synopsis</h2><p><code class="literal">svn update [PATH...]</code></p></div><div class="refsect1" title="Description"><a id="idp17972720"></a><h2>Description</h2><p><span class="command"><strong>svn update</strong></span> brings changes from the
repository into your working copy. If no revision is
given, it brings your working copy up to date with the
<code class="literal">HEAD</code> revision. Otherwise, it
synchronizes the working copy to the revision given by the
<code class="option">--revision</code> (<code class="option">-r</code>) option.
As part of the synchronization, <span class="command"><strong>svn
update</strong></span> also removes any stale locks (see
<a class="xref" href="#svn.tour.cleanup" title="Sometimes You Just Need to Clean Up">the section called “Sometimes You Just Need to Clean Up”</a>) found in the working
copy.</p><p>For each updated item, it prints a line that starts
with a character reporting the action taken. These
characters have the following meaning:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">A</code></span></dt><dd><p>Added</p></dd><dt><span class="term"><code class="literal">B</code></span></dt><dd><p>Broken lock (third column only)</p></dd><dt><span class="term"><code class="literal">D</code></span></dt><dd><p>Deleted</p></dd><dt><span class="term"><code class="literal">U</code></span></dt><dd><p>Updated</p></dd><dt><span class="term"><code class="literal">C</code></span></dt><dd><p>Conflicted</p></dd><dt><span class="term"><code class="literal">G</code></span></dt><dd><p>Merged</p></dd><dt><span class="term"><code class="literal">E</code></span></dt><dd><p>Existed</p></dd></dl></div><p>A character in the first column signifies an update to
the actual file, whereas updates to the file's properties
are shown in the second column. Lock information is
printed in the third column.</p><p>As with most subcommands, you can limit the scope of
the update operation to a particular tree depth using the
<code class="option">--depth</code> option. Alternatively, you can
use the <code class="option">--set-depth</code> option to set a new
<span class="quote">“<span class="quote">sticky</span>”</span> working copy depth on the update
target.</p></div><div class="refsect1" title="Options"><a id="idp17995104"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.accept"><code class="option">--accept</code> <em class="replaceable"><code>ACTION</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.changelist"><code class="option">--changelist</code> (<code class="option">--cl</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.depth"><code class="option">--depth</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.diff3_cmd"><code class="option">--diff3-cmd</code> <em class="replaceable"><code>CMD</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.editor_cmd"><code class="option">--editor-cmd</code> <em class="replaceable"><code>CMD</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.force"><code class="option">--force</code></a>
<a class="xref" href="#svn.ref.svn.sw.ignore_externals"><code class="option">--ignore-externals</code></a>
<a class="xref" href="#svn.ref.svn.sw.parents"><code class="option">--parents</code></a>
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svn.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>REV</code></em></a>
<a class="xref" href="#svn.ref.svn.sw.set_depth"><code class="option">--set-depth</code> <em class="replaceable"><code>ARG</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18004192"></a><h2>Examples</h2><p>Pick up repository changes that have happened since
your last update:</p><div class="informalexample"><pre class="screen">
$ svn update
Updating '.':
A newdir/toggle.c
A newdir/disclose.c
A newdir/launch.c
D newdir/README
Updated to revision 32.
</pre></div><p>You can also <span class="quote">“<span class="quote">update</span>”</span> your working copy
to an older revision (Subversion doesn't have the concept
of <span class="quote">“<span class="quote">sticky</span>”</span> files like CVS does; see <a class="xref" href="#svn.forcvs" title="Appendix B. Subversion for CVS Users">Appendix B, <em>Subversion for CVS Users</em></a>):</p><div class="informalexample"><pre class="screen">
$ svn update -r30
Updating '.':
A newdir/README
D newdir/toggle.c
D newdir/disclose.c
D newdir/launch.c
U foo.c
Updated to revision 30.
</pre></div><div class="tip" title="Tip" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Tip"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="images/tip.png"></img></td><th align="left">Tip</th></tr><tr><td align="left" valign="top"><p>If you want to examine an older revision of a
single file, you may want to use <span class="command"><strong>svn
cat</strong></span> instead—it won't change your working
copy.</p></td></tr></table></div><p><span class="command"><strong>svn update</strong></span> is also the primary
mechanism used to configure sparse working copies. When
used with the <code class="option">--set-depth</code>, the update
operation will omit or reenlist individual working copy
members by modifying their recorded ambient depth to the
depth you specify (fetching information from the
repository as necessary). See
<a class="xref" href="#svn.advanced.sparsedirs" title="Sparse Directories">the section called “Sparse Directories”</a> for more about
sparse directories.</p><p>You can update multiple targets with a single
invocation, and Subversion will not only gracefully skip
any unversioned targets you provide it, but as of
Subversion 1.7 will also include a post-update summary of
all the updates it performed:</p><div class="informalexample"><pre class="screen">
$ cd my-projects
$ svn update *
Updating 'calc':
U button.c
U integer.c
Updated to revision 394.
Skipped 'tempfile.tmp'
Updating 'paint':
A palettes.c
U brushes.c
Updated to revision 60.
Updating 'ziptastic':
At revision 43.
Summary of updates:
Updated 'calc' to r394.
Updated 'paint' to r60.
Updated 'ziptastic' to r43.
Summary of conflicts:
Skipped paths: 1
$
</pre></div></div></div><div class="refentry" title="svn upgrade"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svn.c.upgrade"></a><div class="titlepage"></div><a id="idp18017376" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svn upgrade — Upgrade the metadata storage format for a
working copy.</p></div><div class="refsect1" title="Synopsis"><a id="idp18020480"></a><h2>Synopsis</h2><p><code class="literal">svn upgrade [PATH...]</code></p></div><div class="refsect1" title="Description"><a id="idp18022096"></a><h2>Description</h2><p>As new versions of Subversion are released, the format
used for the working copy metadata changes to accomodate
new features or fix bugs. Older versions of Subversion
would automatically upgrade working copies to the new
format the first time the working copy was used by the new
version of the software. Beginning with Subversion 1.7,
working copy upgrades must be explicitly performed at the
user's request. <span class="command"><strong>svn upgrade</strong></span> is the
subcommand used to trigger that upgrade process.</p></div><div class="refsect1" title="Options"><a id="idp18024560"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svn.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18027072"></a><h2>Examples</h2><p>If you attempt to use Subversion 1.7 on a working copy
created with an older version of Subversion, you will see
an error:</p><div class="informalexample"><pre class="screen">
$ svn status
svn: E155036: Please see the 'svn upgrade' command
svn: E155036: Working copy '/home/sally/project' is too old (format 10, create
d by Subversion 1.6)
$
</pre></div><p>Use the <span class="command"><strong>svn upgrade</strong></span> command to
upgrade the working copy to the most recent metadata
format supported by your version of Subversion.</p><div class="informalexample"><pre class="screen">
$ svn upgrade
Upgraded '.'
Upgraded 'A'
Upgraded 'A/B'
Upgraded 'A/B/E'
Upgraded 'A/B/F'
Upgraded 'A/C'
Upgraded 'A/D'
Upgraded 'A/D/G'
Upgraded 'A/D/H'
$ svn status
D A/B/E/alpha
M A/D/gamma
A A/newfile
$
</pre></div><p>Notice that <span class="command"><strong>svn upgrade</strong></span> preserved
the local modifications present in the working copy at the
time of the upgrade, which were introduced by the version
of Subversion previously used to manipulate this working
copy.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>As was the case with automatically upgraded working
copies in the past, explicitly upgraded working copies
will be unusable by older versions of Subversion,
too.</p></td></tr></table></div></div></div></div></div><div class="sect1" title="svnadmin—Subversion Repository Administration"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.ref.svnadmin"></a>svnadmin—Subversion Repository Administration</h2></div></div></div><p><span class="command"><strong>svnadmin</strong></span> is the administrative tool for
monitoring and repairing your Subversion repository. For
detailed information on repository administration, see the maintenance section for <a class="xref" href="#svn.reposadmin.maint.tk.svnadmin" title="svnadmin">the section called “svnadmin”</a>.</p><p>Since <span class="command"><strong>svnadmin</strong></span> works via direct repository access
(and thus can only be used on the machine that holds the
repository), it refers to the repository with a path, not a URL.</p><div class="sect2" title="svnadmin Options"><div class="titlepage"><div><div><h3 class="title"><a id="svn.ref.svnadmin.sw"></a>svnadmin Options</h3></div></div></div><p>Options in <span class="command"><strong>svnadmin</strong></span> are global, just
as they are in <span class="command"><strong>svn</strong></span>:</p><div class="variablelist"><dl><dt><a id="svn.ref.svnadmin.sw.bdb_log_keep"></a><span class="term"><code class="option">--bdb-log-keep</code></span></dt><dd><p>(Berkeley DB-specific.) Disable automatic log removal
of database logfiles. Having these logfiles around
can be convenient if you need to restore from a
catastrophic repository failure.</p></dd><dt><a id="svn.ref.svnadmin.sw.bdb_txn_nosync"></a><span class="term"><code class="option">--bdb-txn-nosync</code></span></dt><dd><p>(Berkeley DB-specific.) Disables fsync when
committing database transactions. Used with the
<span class="command"><strong>svnadmin create</strong></span> command to create a
Berkeley DB-backed repository with
<code class="literal">DB_TXN_NOSYNC</code> enabled (which improves
speed but has some risks associated with it).</p></dd><dt><a id="svn.ref.svnadmin.sw.bypass_hooks"></a><span class="term"><code class="option">--bypass-hooks</code></span></dt><dd><p>Bypass the repository hook system.</p></dd><dt><a id="svn.ref.svnadmin.sw.bypass_prop_validation"></a><span class="term"><code class="option">--bypass-prop-validation</code></span></dt><dd><p>When loading a dump file, disable the logic which
validates property values.</p></dd><dt><a id="svn.ref.svnadmin.sw.clean_logs"></a><span class="term"><code class="option">--clean-logs</code></span></dt><dd><p>Remove unused Berkeley DB logs.</p></dd><dt><a id="svn.ref.svnadmin.sw.config_dir"></a><span class="term"><code class="option">--config-dir</code> <em class="replaceable"><code>DIR</code></em></span></dt><dd><p>Instructs Subversion to read configuration
information from the specified directory instead of the
default location (<code class="filename">.subversion</code> in
the user's home directory).</p></dd><dt><a id="svn.ref.svnadmin.sw.deltas"></a><span class="term"><code class="option">--deltas</code></span></dt><dd><p>When creating a repository dump file, specify
changes in versioned properties and file contents as
deltas against their previous state.</p></dd><dt><a id="svn.ref.svnadmin.sw.fs_type"></a><span class="term"><code class="option">--fs-type</code> <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>When creating a repository,
use <em class="replaceable"><code>ARG</code></em> as the requested
filesyste type. <em class="replaceable"><code>ARG</code></em> may be
either <code class="literal">bdb</code>
or <code class="literal">fsfs</code>.</p></dd><dt><a id="svn.ref.svnadmin.sw.force_uuid"></a><span class="term"><code class="option">--force-uuid</code></span></dt><dd><p>By default, when loading data into a repository that
already contains revisions, <span class="command"><strong>svnadmin</strong></span>
will ignore the UUID from the dump stream. This option
will cause the repository's UUID to be set to the UUID
from the stream.</p></dd><dt><a id="svn.ref.svnadmin.sw.ignore_uuid"></a><span class="term"><code class="option">--ignore-uuid</code></span></dt><dd><p>By default, when loading data into an empty repository,
<span class="command"><strong>svnadmin</strong></span> will set the repository's UUID
to the UUID from the dump stream. This option will cause
the UUID from the stream to be ignored.</p></dd><dt><a id="svn.ref.svnadmin.sw.incremental"></a><span class="term"><code class="option">--incremental</code></span></dt><dd><p>Dump a revision only as a diff against the
previous revision, instead of the usual
fulltext.</p></dd><dt><a id="svn.ref.svnadmin.sw.memory_cache_size"></a><span class="term"><code class="option">--memory-cache-size</code> (<code class="option">-M</code>) <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>Configures the size (in Megabytes) of the extra
in-memory cache used to minimize redundant operations.
The default value is <code class="literal">16</code>. (This cache
is used for FSFS-backed repositories only.)</p></dd><dt><a id="svn.ref.svnadmin.sw.parent_dir"></a><span class="term"><code class="option">--parent-dir</code> <em class="replaceable"><code>DIR</code></em></span></dt><dd><p>When loading a dump file, root paths at
<em class="replaceable"><code>DIR</code></em> instead of
<code class="filename">/</code>.</p></dd><dt><a id="svn.ref.svnadmin.sw.pre_1.4_compatible"></a><span class="term"><code class="option">--pre-1.4-compatible</code></span></dt><dd><p>When creating a new repository, use a format that is
compatible with versions of Subversion earlier than
Subversion 1.4.</p></dd><dt><a id="svn.ref.svnadmin.sw.pre_1.5_compatible"></a><span class="term"><code class="option">--pre-1.5-compatible</code></span></dt><dd><p>When creating a new repository, use a format that is
compatible with versions of Subversion earlier than
Subversion 1.5.</p></dd><dt><a id="svn.ref.svnadmin.sw.pre_1.6_compatible"></a><span class="term"><code class="option">--pre-1.6-compatible</code></span></dt><dd><p>When creating a new repository, use a format that is
compatible with versions of Subversion earlier than
Subversion 1.6.</p></dd><dt><a id="svn.ref.svnadmin.sw.revision"></a><span class="term"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>Specify a particular revision to operate
on.</p></dd><dt><a id="svn.ref.svnadmin.sw.quiet"></a><span class="term"><code class="option">--quiet</code> (<code class="option">-q</code>)</span></dt><dd><p>Do not show normal progress—show only
errors.</p></dd><dt><a id="svn.ref.svnadmin.sw.use_post_commit_hook"></a><span class="term"><code class="option">--use-post-commit-hook</code></span></dt><dd><p>When loading a dump file, runs the repository's
<code class="literal">post-commit</code> hook after finalizing each newly loaded
revision.</p></dd><dt><a id="svn.ref.svnadmin.sw.use_post_revprop_change_hook"></a><span class="term"><code class="option">--use-post-revprop-change-hook</code></span></dt><dd><p>When changing a revision property, runs the
repository's <code class="literal">post-revprop-change</code> hook after changing the
revision property.</p></dd><dt><a id="svn.ref.svnadmin.sw.use_pre_commit_hook"></a><span class="term"><code class="option">--use-pre-commit-hook</code></span></dt><dd><p>When loading a dump file, runs the repository's
<code class="literal">pre-commit</code> hook before finalizing each newly loaded
revision. If the hook fails, aborts the commit and
terminates the load process.</p></dd><dt><a id="svn.ref.svnadmin.sw.use_pre_revprop_change_hook"></a><span class="term"><code class="option">--use-pre-revprop-change-hook</code></span></dt><dd><p>When changing a revision property, runs the
repository's <code class="literal">pre-revprop-change</code> hook before changing the
revision property. If the hook fails, aborts the
modification and terminates.</p></dd><dt><a id="svn.ref.svnadmin.sw.wait"></a><span class="term"><code class="option">--wait</code></span></dt><dd><p>For operations which require exclusive repository
access, wait until the requisite repository lock has
been obtained instead of immediately erroring out when
it cannot be.</p></dd></dl></div></div><div class="sect2" title="svnadmin Subcommands"><div class="titlepage"><div><div><h3 class="title"><a id="svn.ref.svnadmin.c"></a>svnadmin Subcommands</h3></div></div></div><p>Here are the various subcommands
for the <span class="command"><strong>svnadmin</strong></span> program.</p><div class="refentry" title="svnadmin crashtest"><a id="svn.ref.svnadmin.c.crashtest"></a><div class="titlepage"></div><a id="idp18116000" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin crashtest — Simulate a process that crashes.</p></div><div class="refsect1" title="Synopsis"><a id="idp18119088"></a><h2>Synopsis</h2><p><code class="literal">svnadmin crashtest REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18120720"></a><h2>Description</h2><p>Open the repository at <em class="replaceable"><code>REPOS_PATH</code></em>, then abort, thus
simulating a process that crashes while holding an open
repository handle. This is used for testing automatic
repository recovery (a new feature in Berkeley DB 4.4).
It's unlikely that you'll need to run this
command.</p></div><div class="refsect1" title="Options"><a id="idp18122672"></a><h2>Options</h2><p>None</p></div><div class="refsect1" title="Examples"><a id="idp18123888"></a><h2>Examples</h2><div class="informalexample"><pre class="screen">
$ svnadmin crashtest /var/svn/repos
Aborted
</pre></div><p>Exciting, isn't it?</p></div></div><div class="refentry" title="svnadmin create"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnadmin.c.create"></a><div class="titlepage"></div><a id="idp18127408" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin create — Create a new, empty repository.</p></div><div class="refsect1" title="Synopsis"><a id="idp18130496"></a><h2>Synopsis</h2><p><code class="literal">svnadmin create REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18132128"></a><h2>Description</h2><p>Create a new, empty repository at the path provided.
If the provided directory does not exist, it will be
created for
you.<sup>[<a id="idp18133232" href="#ftn.idp18133232" class="footnote">71</a>]</sup> As of
Subversion 1.2, <span class="command"><strong>svnadmin</strong></span> creates new
repositories with the
<code class="literal">FSFS</code> filesystem backend by
default.</p><p>While <span class="command"><strong>svnadmin create</strong></span> will create
the base directory for a new repository, it will not
create intermediate directories. For example, if you
have an empty directory named
<code class="filename">/var/svn</code>, creating
<code class="filename">/var/svn/repos</code> will work, while
attempting to create
<code class="filename">/var/svn/subdirectory/repos</code> will
fail with an error. Also, keep in mind that, depending
on where on your system you are creating your
repository, you might need to run <span class="command"><strong>svnadmin
create</strong></span> as a user with elevated privileges
(such as the <code class="literal">root</code> user).</p></div><div class="refsect1" title="Options"><a id="idp18142448"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnadmin.sw.bdb_log_keep"><code class="option">--bdb-log-keep</code></a>
<a class="xref" href="#svn.ref.svnadmin.sw.bdb_txn_nosync"><code class="option">--bdb-txn-nosync</code></a>
<a class="xref" href="#svn.ref.svnadmin.sw.config_dir"><code class="option">--config-dir</code> <em class="replaceable"><code>DIR</code></em></a>
<a class="xref" href="#svn.ref.svnadmin.sw.fs_type"><code class="option">--fs-type</code> <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svnadmin.sw.pre_1.4_compatible"><code class="option">--pre-1.4-compatible</code></a>
<a class="xref" href="#svn.ref.svnadmin.sw.pre_1.5_compatible"><code class="option">--pre-1.5-compatible</code></a>
<a class="xref" href="#svn.ref.svnadmin.sw.pre_1.6_compatible"><code class="option">--pre-1.6-compatible</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18149744"></a><h2>Examples</h2><p>Creating a new repository is this easy:</p><div class="informalexample"><pre class="screen">
$ cd /var/svn
$ svnadmin create repos
$
</pre></div><p>In Subversion 1.0, a Berkeley DB repository is always
created. In Subversion 1.1, a Berkeley DB repository is
the default repository type, but an FSFS repository can be
created using the <code class="option">--fs-type</code>
option:</p><div class="informalexample"><pre class="screen">
$ cd /var/svn
$ svnadmin create repos --fs-type fsfs
$
</pre></div></div><div class="footnotes"><br></br><hr width="100" align="left"></hr><div class="footnote"><p><sup>[<a id="ftn.idp18133232" href="#idp18133232" class="para">71</a>] </sup>Remember, <span class="command"><strong>svnadmin</strong></span>
works only with local <span class="emphasis"><em>paths</em></span>, not
<span class="emphasis"><em>URLs</em></span>.</p></div></div></div><div class="refentry" title="svnadmin deltify"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnadmin.c.deltify"></a><div class="titlepage"></div><a id="idp18155536" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin deltify — Deltify changed paths in a revision range.</p></div><div class="refsect1" title="Synopsis"><a id="idp18158688"></a><h2>Synopsis</h2><p><code class="literal">svnadmin deltify [-r LOWER[:UPPER]] REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18160336"></a><h2>Description</h2><p><span class="command"><strong>svnadmin deltify</strong></span> exists in
current versions of Subversion only for historical reasons.
This command is deprecated and no longer needed.</p><p>It dates from a time when Subversion offered
administrators greater control over compression strategies
in the repository. This turned out to be a lot of
complexity for <span class="emphasis"><em>very</em></span> little gain, and
this <span class="quote">“<span class="quote">feature</span>”</span> was deprecated.</p></div><div class="refsect1" title="Options"><a id="idp18163712"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnadmin.sw.memory_cache_size"><code class="option">--memory-cache-size</code> (<code class="option">-M</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svnadmin.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svnadmin.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>ARG</code></em></a>
</pre></div></div></div><div class="refentry" title="svnadmin dump"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnadmin.c.dump"></a><div class="titlepage"></div><a id="idp18169152" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin dump — Dump the contents of the filesystem to <code class="filename">stdout</code>.</p></div><div class="refsect1" title="Synopsis"><a id="idp18172896"></a><h2>Synopsis</h2><p><code class="literal">svnadmin dump REPOS_PATH [-r LOWER[:UPPER]] [--incremental] [--deltas]</code></p></div><div class="refsect1" title="Description"><a id="idp18174560"></a><h2>Description</h2><p>Dump the contents of the filesystem to <code class="filename">stdout</code> in a
<span class="quote">“<span class="quote">dump file</span>”</span> portable format, sending feedback
to <code class="filename">stderr</code>. Dump revisions
<em class="replaceable"><code>LOWER</code></em> rev through
<em class="replaceable"><code>UPPER</code></em> rev. If no revisions are
given, dump all revision trees. If only
<em class="replaceable"><code>LOWER</code></em> is given, dump that one
revision tree. See <a class="xref" href="#svn.reposadmin.maint.migrate" title="Migrating Repository Data Elsewhere">the section called “Migrating Repository Data Elsewhere”</a>
for a practical use.</p><p>By default, the Subversion dump stream contains a
single revision (the first revision in the requested
revision range) in which every file and directory in the
repository in that revision is presented as though that whole
tree was added at once, followed by other revisions (the
remainder of the revisions in the requested range), which
contain only the files and directories that were modified
in those revisions. For a modified file, the complete
full-text representation of its contents, as well as all of
its properties, are presented in the dump file; for a
directory, all of its properties are presented.</p><p>Two useful options modify the dump file
generator's behavior. The first is the
<code class="option">--incremental</code> option, which simply causes
that first revision in the dump stream to contain only
the files and directories modified in that revision,
instead of being presented as the addition of a new tree,
and in exactly the same way that every other revision in
the dump file is presented. This is useful for generating
a relatively small dump file to be loaded into another
repository that already has the files and directories
that exist in the original repository.</p><p>The second useful option is <code class="option">--deltas</code>.
This option causes <span class="command"><strong>svnadmin dump</strong></span> to,
instead of emitting full-text representations of file
contents and property lists, emit only deltas of those
items against their previous versions. This reduces (in
some cases, drastically) the size of the dump file that
<span class="command"><strong>svnadmin dump</strong></span> creates. There are, however,
disadvantages to using this option—deltified
dump files are more CPU-intensive to create, cannot be
operated on by <span class="command"><strong>svndumpfilter</strong></span>, and tend
not to compress as well as their nondeltified counterparts
when using third-party tools such as <span class="command"><strong>gzip</strong></span>
and <span class="command"><strong>bzip2</strong></span>.</p></div><div class="refsect1" title="Options"><a id="idp18187984"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnadmin.sw.deltas"><code class="option">--deltas</code></a>
<a class="xref" href="#svn.ref.svnadmin.sw.incremental"><code class="option">--incremental</code></a>
<a class="xref" href="#svn.ref.svnadmin.sw.memory_cache_size"><code class="option">--memory-cache-size</code> (<code class="option">-M</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svnadmin.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svnadmin.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>ARG</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18193488"></a><h2>Examples</h2><p>Dump your whole repository:</p><div class="informalexample"><pre class="screen">
$ svnadmin dump /var/svn/repos > full.dump
* Dumped revision 0.
* Dumped revision 1.
* Dumped revision 2.
…
</pre></div><p>Incrementally dump a single transaction from your
repository:</p><div class="informalexample"><pre class="screen">
$ svnadmin dump /var/svn/repos -r 21 --incremental > incr.dump
* Dumped revision 21.
</pre></div></div></div><div class="refentry" title="svnadmin help (h, ?)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnadmin.c.help"></a><div class="titlepage"></div><a id="idp18198672" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin help (h, ?) — Help!</p></div><div class="refsect1" title="Synopsis"><a id="idp18201760"></a><h2>Synopsis</h2><p><code class="literal">svnadmin help [SUBCOMMAND...]</code></p></div><div class="refsect1" title="Description"><a id="idp18203392"></a><h2>Description</h2><p>This subcommand is useful when you're trapped on a
desert island with neither a Net connection nor a copy of
this book.</p></div></div><div class="refentry" title="svnadmin hotcopy"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnadmin.c.hotcopy"></a><div class="titlepage"></div><a id="idp18206000" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin hotcopy — Make a hot copy of a repository.</p></div><div class="refsect1" title="Synopsis"><a id="idp18209088"></a><h2>Synopsis</h2><p><code class="literal">svnadmin hotcopy REPOS_PATH NEW_REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18210736"></a><h2>Description</h2><p>This subcommand makes a full <span class="quote">“<span class="quote">hot</span>”</span> backup
of your repository, including all hooks, configuration
files, and, of course, database files. If you pass the
<code class="option">--clean-logs</code> option,
<span class="command"><strong>svnadmin</strong></span> will perform a hot copy of your
repository, and then remove unused Berkeley DB logs from
the original repository. You can run this command at any
time and make a safe copy of the repository, regardless of
whether other processes are using the repository.</p></div><div class="refsect1" title="Options"><a id="idp18214032"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnadmin.sw.clean_logs"><code class="option">--clean-logs</code></a>
</pre></div><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>As described in <a class="xref" href="#svn.reposadmin.basics.backends.bdb" title="Berkeley DB">the section called “Berkeley DB”</a>, hot-copied
Berkeley DB repositories are <span class="emphasis"><em>not</em></span>
portable across operating systems, nor will they work on
machines with a different <span class="quote">“<span class="quote">endianness</span>”</span> than
the machine where they were created.</p></td></tr></table></div></div></div><div class="refentry" title="svnadmin list-dblogs"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnadmin.c.list-dblogs"></a><div class="titlepage"></div><a id="idp18220480" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin list-dblogs — Ask Berkeley DB which logfiles exist for a
given Subversion repository (applies only to repositories
using the <code class="literal">bdb</code> backend).</p></div><div class="refsect1" title="Synopsis"><a id="idp18224400"></a><h2>Synopsis</h2><p><code class="literal">svnadmin list-dblogs REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18225984"></a><h2>Description</h2><p>Berkeley DB creates logs of all changes to the
repository, which allow it to recover in the face of
catastrophe. Unless you enable
<code class="literal">DB_LOG_AUTOREMOVE</code>, the logfiles
accumulate, although most are no longer used and can be
deleted to reclaim disk space. See <a class="xref" href="#svn.reposadmin.maint.diskspace" title="Managing Disk Space">the section called “Managing Disk Space”</a> for more
information.</p></div></div><div class="refentry" title="svnadmin list-unused-dblogs"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnadmin.c.list-unused-dblogs"></a><div class="titlepage"></div><a id="idp18230160" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin list-unused-dblogs — Ask Berkeley DB which logfiles can be safely
deleted (applies only to repositories using the
<code class="literal">bdb</code> backend).</p></div><div class="refsect1" title="Synopsis"><a id="idp18234032"></a><h2>Synopsis</h2><p><code class="literal">svnadmin list-unused-dblogs REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18235664"></a><h2>Description</h2><p>Berkeley DB creates logs of all changes to the
repository, which allow it to recover in the face of
catastrophe. Unless you enable
<code class="literal">DB_LOG_AUTOREMOVE</code>, the logfiles
accumulate, although most are no longer used and can be
deleted to reclaim disk space. See <a class="xref" href="#svn.reposadmin.maint.diskspace" title="Managing Disk Space">the section called “Managing Disk Space”</a> for more
information.</p></div><div class="refsect1" title="Examples"><a id="idp18238592"></a><h2>Examples</h2><p>Remove all unused logfiles from the repository:</p><div class="informalexample"><pre class="screen">
$ svnadmin list-unused-dblogs /var/svn/repos
/var/svn/repos/log.0000000031
/var/svn/repos/log.0000000032
/var/svn/repos/log.0000000033
$ svnadmin list-unused-dblogs /var/svn/repos | xargs rm
## disk space reclaimed!
</pre></div></div></div><div class="refentry" title="svnadmin load"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnadmin.c.load"></a><div class="titlepage"></div><a id="idp18242384" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin load — Read a
repository dump stream from
<code class="filename">stdin</code>.</p></div><div class="refsect1" title="Synopsis"><a id="idp18246208"></a><h2>Synopsis</h2><p><code class="literal">svnadmin load REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18247840"></a><h2>Description</h2><p>Read a repository dump stream from
<code class="filename">stdin</code>, committing new revisions into the repository's
filesystem. Send progress feedback to <code class="filename">stdout</code>.</p></div><div class="refsect1" title="Options"><a id="idp18250576"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnadmin.sw.bypass_prop_validation"><code class="option">--bypass-prop-validation</code></a>
<a class="xref" href="#svn.ref.svnadmin.sw.force_uuid"><code class="option">--force-uuid</code></a>
<a class="xref" href="#svn.ref.svnadmin.sw.ignore_uuid"><code class="option">--ignore-uuid</code></a>
<a class="xref" href="#svn.ref.svnadmin.sw.memory_cache_size"><code class="option">--memory-cache-size</code> (<code class="option">-M</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svnadmin.sw.parent_dir"><code class="option">--parent-dir</code> <em class="replaceable"><code>DIR</code></em></a>
<a class="xref" href="#svn.ref.svnadmin.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svnadmin.sw.use_post_commit_hook"><code class="option">--use-post-commit-hook</code></a>
<a class="xref" href="#svn.ref.svnadmin.sw.use_pre_commit_hook"><code class="option">--use-pre-commit-hook</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18258544"></a><h2>Examples</h2><p>This shows the beginning of loading a repository from
a backup file (made, of course, with <span class="command"><strong>svnadmin
dump</strong></span>):</p><div class="informalexample"><pre class="screen">
$ svnadmin load /var/svn/restored < repos-backup
<<< Started new txn, based on original revision 1
* adding path : test ... done.
* adding path : test/a ... done.
…
</pre></div><p>Or if you want to load into a subdirectory:</p><div class="informalexample"><pre class="screen">
$ svnadmin load --parent-dir new/subdir/for/project \
/var/svn/restored < repos-backup
<<< Started new txn, based on original revision 1
* adding path : test ... done.
* adding path : test/a ... done.
…
</pre></div><p>Newer versions of Subversion have grown more strict
regarding the format of the values of Subversion's own
built-in properties. Of course, properties created with
older versions of Subversion wouldn't have benefitted from
that strictness, and as such might be improperly
formatted. Dump streams carry property values as-is, so
using Subversion 1.7 to load dump streams created from
repositories with ill-formatted property values will, by
default, trigger a validation error. There are several
workaround for this problem. First, you can manually
repair the problematic property values in the source
repository and recreate the dump stream. Or, you can
manually tweak the dump stream itself to fix those
property values. Finally, if you'd rather not deal with
the problem right now, use the
<code class="option">--bypass-prop-validation</code> option
with <span class="command"><strong>svnadmin load</strong></span>.</p></div></div><div class="refentry" title="svnadmin lslocks"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnadmin.c.lslocks"></a><div class="titlepage"></div><a id="idp18267360" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin lslocks — Print descriptions of all locks.</p></div><div class="refsect1" title="Synopsis"><a id="idp18270448"></a><h2>Synopsis</h2><p><code class="literal">svnadmin lslocks REPOS_PATH [PATH-IN-REPOS]</code></p></div><div class="refsect1" title="Description"><a id="idp18272096"></a><h2>Description</h2><p>Print descriptions of all locks in
repository <em class="replaceable"><code>REPOS_PATH</code></em>
underneath the
path <em class="replaceable"><code>PATH-IN-REPOS</code></em>.
If <em class="replaceable"><code>PATH-IN-REPOS</code></em> is not
provided, it defaults to the root directory of the
repository.</p></div><div class="refsect1" title="Options"><a id="idp18274816"></a><h2>Options</h2><p>None</p></div><div class="refsect1" title="Examples"><a id="idp18276032"></a><h2>Examples</h2><p>This lists the one locked file in the repository at
<code class="filename">/var/svn/repos</code>:</p><div class="informalexample"><pre class="screen">
$ svnadmin lslocks /var/svn/repos
Path: /tree.jpg
UUID Token: opaquelocktoken:ab00ddf0-6afb-0310-9cd0-dda813329753
Owner: harry
Created: 2005-07-08 17:27:36 -0500 (Fri, 08 Jul 2005)
Expires:
Comment (1 line):
Rework the uppermost branches on the bald cypress in the foreground.
</pre></div></div></div><div class="refentry" title="svnadmin lstxns"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnadmin.c.lstxns"></a><div class="titlepage"></div><a id="idp18280592" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin lstxns — Print the names of all uncommitted transactions.</p></div><div class="refsect1" title="Synopsis"><a id="idp18283744"></a><h2>Synopsis</h2><p><code class="literal">svnadmin lstxns REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18285328"></a><h2>Description</h2><p>Print the names of all uncommitted transactions. See
<a class="xref" href="#svn.reposadmin.maint.diskspace.deadtxns" title="Removing dead transactions">the section called “Removing dead transactions”</a>
for information on how uncommitted transactions are
created and what you should do with them.</p></div><div class="refsect1" title="Examples"><a id="idp18287568"></a><h2>Examples</h2><p>List all outstanding transactions in a
repository:</p><div class="informalexample"><pre class="screen">
$ svnadmin lstxns /var/svn/repos/
1w
1x
</pre></div></div></div><div class="refentry" title="svnadmin pack"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnadmin.c.pack"></a><div class="titlepage"></div><a id="idp18291152" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin pack — Possibly compact the repository into a more
efficient storage model.</p></div><div class="refsect1" title="Synopsis"><a id="idp18294304"></a><h2>Synopsis</h2><p><code class="literal">svnadmin pack REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18295888"></a><h2>Description</h2><p>See
<a class="xref" href="#svn.reposadmin.maint.diskspace.fsfspacking" title="Packing FSFS filesystems">the section called “Packing FSFS filesystems”</a> for more information.</p></div><div class="refsect1" title="Options"><a id="idp18297904"></a><h2>Options</h2><p>None</p></div></div><div class="refentry" title="svnadmin recover"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnadmin.c.recover"></a><div class="titlepage"></div><a id="idp18300368" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin recover — Bring a repository database back into a
consistent state (applies only to repositories using the
<code class="literal">bdb</code> backend). In addition, if
<code class="filename">repos/conf/passwd</code> does not exist, it
will create a default passwordfile .</p></div><div class="refsect1" title="Synopsis"><a id="idp18305008"></a><h2>Synopsis</h2><p><code class="literal">svnadmin recover REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18306640"></a><h2>Description</h2><p>Run this command if you get an error indicating that
your repository needs to be recovered.</p></div><div class="refsect1" title="Options"><a id="idp18307936"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnadmin.sw.wait"><code class="option">--wait</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18310624"></a><h2>Examples</h2><p>Recover a hung repository:</p><div class="informalexample"><pre class="screen">
$ svnadmin recover /var/svn/repos/
Repository lock acquired.
Please wait; recovering the repository may take some time...
Recovery completed.
The latest repos revision is 34.
</pre></div><p>Recovering the database requires an exclusive lock on
the repository. (This is a <span class="quote">“<span class="quote">database lock</span>”</span>;
see the sidebar <a class="xref" href="#svn.advanced.locking.meanings" title="The Three Meanings of “Lock”">The Three Meanings of <span class="quote">“<span class="quote">Lock</span>”</span></a>.)
If another process is accessing the repository,
then <span class="command"><strong>svnadmin recover</strong></span> will error:</p><div class="informalexample"><pre class="screen">
$ svnadmin recover /var/svn/repos
svn: E165000: Failed to get exclusive repository access; perhaps another proce
ss such as httpd, svnserve or svn has it open?
$
</pre></div><p>The <code class="option">--wait</code> option, however, will
cause <span class="command"><strong>svnadmin recover</strong></span> to wait
indefinitely for other processes to disconnect:</p><div class="informalexample"><pre class="screen">
$ svnadmin recover /var/svn/repos --wait
Waiting on repository lock; perhaps another process has it open?
### time goes by…
Repository lock acquired.
Please wait; recovering the repository may take some time...
Recovery completed.
The latest repos revision is 34.
</pre></div></div></div><div class="refentry" title="svnadmin rmlocks"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnadmin.c.rmlocks"></a><div class="titlepage"></div><a id="idp18320912" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin rmlocks — Unconditionally remove one or more locks from a
repository.</p></div><div class="refsect1" title="Synopsis"><a id="idp18324080"></a><h2>Synopsis</h2><p><code class="literal">svnadmin rmlocks REPOS_PATH LOCKED_PATH...</code></p></div><div class="refsect1" title="Description"><a id="idp18325728"></a><h2>Description</h2><p>Remove one or more locks from each <em class="replaceable"><code>LOCKED_PATH</code></em>.</p></div><div class="refsect1" title="Options"><a id="idp18327328"></a><h2>Options</h2><p>None</p></div><div class="refsect1" title="Examples"><a id="idp18328592"></a><h2>Examples</h2><p>This deletes the locks on
<code class="filename">tree.jpg</code> and
<code class="filename">house.jpg</code> in the repository at
<code class="filename">/var/svn/repos</code>:</p><div class="informalexample"><pre class="screen">
$ svnadmin rmlocks /var/svn/repos tree.jpg house.jpg
Removed lock on '/tree.jpg.
Removed lock on '/house.jpg.
</pre></div></div></div><div class="refentry" title="svnadmin rmtxns"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnadmin.c.rmtxns"></a><div class="titlepage"></div><a id="idp18334400" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin rmtxns — Delete transactions from a repository.</p></div><div class="refsect1" title="Synopsis"><a id="idp18337488"></a><h2>Synopsis</h2><p><code class="literal">svnadmin rmtxns REPOS_PATH TXN_NAME...</code></p></div><div class="refsect1" title="Description"><a id="idp18339120"></a><h2>Description</h2><p>Delete outstanding transactions from a repository.
This is covered in detail in <a class="xref" href="#svn.reposadmin.maint.diskspace.deadtxns" title="Removing dead transactions">the section called “Removing dead transactions”</a>.</p></div><div class="refsect1" title="Options"><a id="idp18341136"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnadmin.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18343648"></a><h2>Examples</h2><p>Remove named transactions:</p><div class="informalexample"><pre class="screen">
$ svnadmin rmtxns /var/svn/repos/ 1w 1x
</pre></div><p>Fortunately, the output of <span class="command"><strong>lstxns</strong></span>
works great as the input for <span class="command"><strong>rmtxns</strong></span>:</p><div class="informalexample"><pre class="screen">
$ svnadmin rmtxns /var/svn/repos/ `svnadmin lstxns /var/svn/repos/`
</pre></div><p>This removes all uncommitted transactions from
your repository.</p></div></div><div class="refentry" title="svnadmin setlog"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnadmin.c.setlog"></a><div class="titlepage"></div><a id="idp18350672" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin setlog — Set the log message on a revision.</p></div><div class="refsect1" title="Synopsis"><a id="idp18353760"></a><h2>Synopsis</h2><p><code class="literal">svnadmin setlog REPOS_PATH -r REVISION FILE</code></p></div><div class="refsect1" title="Description"><a id="idp18355408"></a><h2>Description</h2><p>Set the log message on
revision <em class="replaceable"><code>REVISION</code></em> to the
contents of <em class="replaceable"><code>FILE</code></em>.</p><p>This is similar to using <span class="command"><strong>svn propset</strong></span>
with the <code class="option">--revprop</code> option to set the <code class="literal">svn:log</code> property
on a revision, except that you can also use the option
<code class="option">--bypass-hooks</code> to avoid running any pre-
or post-commit hooks, which is useful if the modification
of revision properties has not been enabled in
the <code class="literal">pre-revprop-change</code> hook.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>Revision properties are not under version control,
so this command will permanently overwrite the previous
log message.</p></td></tr></table></div></div><div class="refsect1" title="Options"><a id="idp18362192"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnadmin.sw.bypass_hooks"><code class="option">--bypass-hooks</code></a>
<a class="xref" href="#svn.ref.svnadmin.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>ARG</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18365552"></a><h2>Examples</h2><p>Set the log message for revision 19 to the contents of
the file <code class="filename">msg</code>:</p><div class="informalexample"><pre class="screen">
$ svnadmin setlog /var/svn/repos/ -r 19 msg
</pre></div></div></div><div class="refentry" title="svnadmin setrevprop"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnadmin.c.setrevprop"></a><div class="titlepage"></div><a id="idp18369856" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin setrevprop — Set a property on a revision.</p></div><div class="refsect1" title="Synopsis"><a id="idp18372944"></a><h2>Synopsis</h2><p><code class="literal">svnadmin setrevprop REPOS_PATH -r REVISION NAME FILE</code></p></div><div class="refsect1" title="Description"><a id="idp18374592"></a><h2>Description</h2><p>Set the property <em class="replaceable"><code>NAME</code></em> on
revision <em class="replaceable"><code>REVISION</code></em> to the
contents of <em class="replaceable"><code>FILE</code></em>. Use
<code class="option">--use-pre-revprop-change-hook</code>
or <code class="option">--use-post-revprop-change-hook</code> to
trigger the revision property-related hooks (e.g., if
you want an email notification sent from your <code class="literal">post-revprop-change-hook</code>).</p></div><div class="refsect1" title="Options"><a id="idp18378928"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnadmin.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svnadmin.sw.use_post_revprop_change_hook"><code class="option">--use-post-revprop-change-hook</code></a>
<a class="xref" href="#svn.ref.svnadmin.sw.use_pre_revprop_change_hook"><code class="option">--use-pre-revprop-change-hook</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18383168"></a><h2>Examples</h2><p>The following sets the revision
property <code class="literal">repository-photo</code> to the contents
of the file <code class="filename">sandwich.png</code>:</p><div class="informalexample"><pre class="screen">
$ svnadmin setrevprop /var/svn/repos -r 0 repository-photo sandwich.png
</pre></div><p>As you can see, <span class="command"><strong>svnadmin setrevprop</strong></span>
has no output upon success.</p></div></div><div class="refentry" title="svnadmin setuuid"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnadmin.c.setuuid"></a><div class="titlepage"></div><a id="idp18389392" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin setuuid — Reset the repository UUID.</p></div><div class="refsect1" title="Synopsis"><a id="idp18392480"></a><h2>Synopsis</h2><p><code class="literal">svnadmin setuuid REPOS_PATH [NEW_UUID]</code></p></div><div class="refsect1" title="Description"><a id="idp18394112"></a><h2>Description</h2><p>Reset the repository UUID for the repository located
at <em class="replaceable"><code>REPOS_PATH</code></em>.
If <em class="replaceable"><code>NEW_UUID</code></em> is provided, use
that as the new repository UUID; otherwise, generate a
brand-new UUID for the repository.</p></div><div class="refsect1" title="Options"><a id="idp18396384"></a><h2>Options</h2><p>None</p></div><div class="refsect1" title="Examples"><a id="idp18397600"></a><h2>Examples</h2><p>If you've <span class="command"><strong>svnsync</strong></span>ed <code class="filename">/var/svn/repos</code>
to <code class="filename">/var/svn/repos-new</code> and intend to use <code class="filename">repos-new</code> as your
canonical repository, you may want to change the UUID for
<code class="filename">repos-new</code> to the UUID of <code class="filename">repos</code> so that your users don't
have to check out a new working copy to accommodate the
change:</p><div class="informalexample"><pre class="screen">
$ svnadmin setuuid /var/svn/repos-new 2109a8dd-854f-0410-ad31-d604008985ab
</pre></div><p>As you can see, <span class="command"><strong>svnadmin setuuid</strong></span>
has no output upon success.</p></div></div><div class="refentry" title="svnadmin upgrade"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnadmin.c.upgrade"></a><div class="titlepage"></div><a id="idp18406896" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin upgrade — Upgrade a repository to the latest supported
schema version.</p></div><div class="refsect1" title="Synopsis"><a id="idp18410080"></a><h2>Synopsis</h2><p><code class="literal">svnadmin upgrade REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18411664"></a><h2>Description</h2><p>Upgrade the repository located
at <em class="replaceable"><code>REPOS_PATH</code></em> to the latest
supported schema version.</p><p>This functionality is provided as a convenience for
repository administrators who wish to make use of new
Subversion functionality without having to undertake a
potentially costly full repository dump and load
operation. As such, the upgrade performs only the minimum
amount of work needed to accomplish this while still
maintaining the integrity of the repository. While a dump
and subsequent load guarantee the most optimized
repository state, <span class="command"><strong>svnadmin upgrade</strong></span> does
not.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>You should <span class="emphasis"><em>always</em></span> back up your
repository before upgrading.</p></td></tr></table></div></div><div class="refsect1" title="Options"><a id="idp18416352"></a><h2>Options</h2><p>None</p></div><div class="refsect1" title="Examples"><a id="idp18417568"></a><h2>Examples</h2><p>Upgrade the repository at
path <code class="filename">/var/repos/svn</code>:</p><div class="informalexample"><pre class="screen">
$ svnadmin upgrade /var/repos/svn
Repository lock acquired.
Please wait; upgrading the repository may take some time...
Upgrade completed.
</pre></div></div></div><div class="refentry" title="svnadmin verify"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnadmin.c.verify"></a><div class="titlepage"></div><a id="idp18422432" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnadmin verify — Verify the data stored in the repository.</p></div><div class="refsect1" title="Synopsis"><a id="idp18425584"></a><h2>Synopsis</h2><p><code class="literal">svnadmin verify REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18427168"></a><h2>Description</h2><p>Run this command if you wish to verify the integrity
of your repository. This basically iterates through all
revisions in the repository by internally dumping all
revisions and discarding the output—it's a good idea
to run this on a regular basis to guard against latent
hard disk failures and <span class="quote">“<span class="quote">bitrot.</span>”</span> If this
command fails—which it will do at the first sign of
a problem—that means your repository has at
least one corrupted revision, and you should restore the
corrupted revision from a backup (you did make a backup,
didn't you?).</p></div><div class="refsect1" title="Options"><a id="idp18429536"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnadmin.sw.memory_cache_size"><code class="option">--memory-cache-size</code> (<code class="option">-M</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svnadmin.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svnadmin.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>ARG</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18433360"></a><h2>Examples</h2><p>Verify a hung repository:</p><div class="informalexample"><pre class="screen">
$ svnadmin verify /var/svn/repos/
* Verified revision 1729.
</pre></div></div></div></div></div><div class="sect1" title="svnlook—Subversion Repository Examination"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.ref.svnlook"></a>svnlook—Subversion Repository Examination</h2></div></div></div><p><span class="command"><strong>svnlook</strong></span> is a command-line utility for
examining different aspects of a Subversion repository. It
does not make any changes to the repository—it's just
used for <span class="quote">“<span class="quote">peeking.</span>”</span>
<span class="command"><strong>svnlook</strong></span> is typically used by the
repository hooks, but a repository administrator might find it
useful for diagnostic purposes.</p><p>Since <span class="command"><strong>svnlook</strong></span> works via direct repository access
(and thus can be used only on the machine that holds the
repository), it refers to the repository with a path, not a URL.</p><p>If no revision or transaction is specified,
<span class="command"><strong>svnlook</strong></span> defaults to the youngest (most
recent) revision of the repository.</p><div class="sect2" title="svnlook Options"><div class="titlepage"><div><div><h3 class="title"><a id="svn.ref.svnlook.sw"></a>svnlook Options</h3></div></div></div><p>Options in <span class="command"><strong>svnlook</strong></span> are global, just
as they are in <span class="command"><strong>svn</strong></span> and
<span class="command"><strong>svnadmin</strong></span>; however, most options apply
to only one subcommand since the functionality of
<span class="command"><strong>svnlook</strong></span> is (intentionally) limited in
scope:</p><div class="variablelist"><dl><dt><a id="svn.ref.svnlook.sw.copy_info"></a><span class="term"><code class="option">--copy-info</code></span></dt><dd><p>Causes <span class="command"><strong>svnlook changed</strong></span> to show
detailed copy source information.</p></dd><dt><a id="svn.ref.svnlook.sw.diff_copy_from"></a><span class="term"><code class="option">--diff-copy-from</code></span></dt><dd><p>Print differences for copied items against the copy
source.</p></dd><dt><a id="svn.ref.svnlook.sw.extensions"></a><span class="term"><code class="option">--extensions</code> (<code class="option">-x</code>) <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>Specifies customizations which Subversion should
make when performing difference calculations. Valid
extensions include:</p><div class="variablelist"><dl><dt><span class="term"><code class="option">--ignore-space-change</code> (<code class="option">-b</code>)</span></dt><dd><p>Ignore changes in the amount of white space.</p></dd><dt><span class="term"><code class="option">--ignore-all-space</code> (<code class="option">-w</code>)</span></dt><dd><p>Ignore all white space.</p></dd><dt><span class="term"><code class="option">--ignore-eol-style</code></span></dt><dd><p>Ignore changes in EOL (end-of-line) style.</p></dd><dt><span class="term"><code class="option">--show-c-function</code> (<code class="option">-p</code>)</span></dt><dd><p>Show C function names in the diff output.</p></dd><dt><span class="term"><code class="option">--unified</code> (<code class="option">-u</code>)</span></dt><dd><p>Show three lines of unified diff context.</p></dd></dl></div><p>The default value is <code class="literal">-u</code>.</p><p>Note that when Subversion is configured to invoke an
external diff command, the value of
the <code class="option">--extension</code> (<code class="option">-x</code>)
option isn't restricted to the previously mentioned
options, but may be <span class="emphasis"><em>any</em></span> additional
arguments which Subversion should pass to that command.
If you wish to pass multiple arguments, you must enclose
all of them in quotes.</p></dd><dt><a id="svn.ref.svnlook.sw.full_paths"></a><span class="term"><code class="option">--full-paths</code></span></dt><dd><p>Causes <span class="command"><strong>svnlook tree</strong></span> to display
full paths instead of hierarchical, indented path
components.</p></dd><dt><a id="svn.ref.svnlook.sw.limit"></a><span class="term"><code class="option">--limit</code> (<code class="option">-l</code>) <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>Limit output to a maximum number of
<em class="replaceable"><code>ARG</code></em> items.</p></dd><dt><a id="svn.ref.svnlook.sw.no_diff_deleted"></a><span class="term"><code class="option">--no-diff-deleted</code></span></dt><dd><p>Prevents <span class="command"><strong>svnlook diff</strong></span> from printing
differences for deleted files. The default behavior
when a file is deleted in a transaction/revision is
to print the same differences that you would see if
you had left the file but removed all the content.</p></dd><dt><a id="svn.ref.svnlook.sw.no_diff_added"></a><span class="term"><code class="option">--no-diff-added</code></span></dt><dd><p>Prevents <span class="command"><strong>svnlook diff</strong></span> from printing
differences for added files. The default behavior when
you add a file is to print the same differences that you
would see if you had added the entire contents of an
existing (empty) file.</p></dd><dt><a id="svn.ref.svnlook.sw.non_recursive"></a><span class="term"><code class="option">--non-recursive</code> (<code class="option">-N</code>)</span></dt><dd><p>Operate on a single directory only.</p></dd><dt><a id="svn.ref.svnlook.sw.revision"></a><span class="term"><code class="option">--revision</code> (<code class="option">-r</code>)</span></dt><dd><p>Specifies a particular revision number that you wish
to examine.</p></dd><dt><a id="svn.ref.svnlook.sw.revprop"></a><span class="term"><code class="option">--revprop</code></span></dt><dd><p>Operates on a revision property instead of a
property specific to a file or directory. This option
requires that you also pass a revision with the
<code class="option">--revision</code> (<code class="option">-r</code>)
option.</p></dd><dt><a id="svn.ref.svnlook.sw.transaction"></a><span class="term"><code class="option">--transaction</code> (<code class="option">-t</code>)</span></dt><dd><p>Specifies a particular transaction ID that you wish to
examine.</p></dd><dt><a id="svn.ref.svnlook.sw.show_ids"></a><span class="term"><code class="option">--show-ids</code></span></dt><dd><p>Shows the filesystem node revision IDs for each path in the filesystem
tree.</p></dd><dt><a id="svn.ref.svnlook.sw.verbose"></a><span class="term"><code class="option">--verbose</code> (<code class="option">-v</code>)</span></dt><dd><p>Be verbose. When used with <span class="command"><strong>svnlook
proplist</strong></span>, for example, this causes Subversion
to display not just the list of properties, but their
values also.</p></dd><dt><a id="svn.ref.svnlook.sw.xml"></a><span class="term"><code class="option">--xml</code></span></dt><dd><p></p></dd></dl></div></div><div class="sect2" title="svnlook Subcommands"><div class="titlepage"><div><div><h3 class="title"><a id="svn.ref.svnlook.c"></a>svnlook Subcommands</h3></div></div></div><p>Here are the various subcommands
for the <span class="command"><strong>svnlook</strong></span> program.</p><div class="refentry" title="svnlook author"><a id="svn.ref.svnlook.c.author"></a><div class="titlepage"></div><a id="idp18507472" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnlook author — Print the author.</p></div><div class="refsect1" title="Synopsis"><a id="idp18510592"></a><h2>Synopsis</h2><p><code class="literal">svnlook author REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18512176"></a><h2>Description</h2><p>Print the author of a revision or transaction in
the repository.</p></div><div class="refsect1" title="Options"><a id="idp18513456"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnlook.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>)</a>
<a class="xref" href="#svn.ref.svnlook.sw.transaction"><code class="option">--transaction</code> (<code class="option">-t</code>)</a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18516880"></a><h2>Examples</h2><p><span class="command"><strong>svnlook author</strong></span> is handy, but not
very exciting:</p><div class="informalexample"><pre class="screen">
$ svnlook author -r 40 /var/svn/repos
sally
</pre></div></div></div><div class="refentry" title="svnlook cat"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnlook.c.cat"></a><div class="titlepage"></div><a id="idp18521008" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnlook cat — Print the contents of a file.</p></div><div class="refsect1" title="Synopsis"><a id="idp18524080"></a><h2>Synopsis</h2><p><code class="literal">svnlook cat REPOS_PATH PATH_IN_REPOS</code></p></div><div class="refsect1" title="Description"><a id="idp18525712"></a><h2>Description</h2><p>Print the contents of a file.</p></div><div class="refsect1" title="Options"><a id="idp18526944"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnlook.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>)</a>
<a class="xref" href="#svn.ref.svnlook.sw.transaction"><code class="option">--transaction</code> (<code class="option">-t</code>)</a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18530176"></a><h2>Examples</h2><p>This shows the contents of a file in transaction
<code class="literal">ax8</code>, located at
<code class="filename">/trunk/README</code>:</p><div class="informalexample"><pre class="screen">
$ svnlook cat -t ax8 /var/svn/repos /trunk/README
Subversion, a version control system.
=====================================
$LastChangedDate: 2003-07-17 10:45:25 -0500 (Thu, 17 Jul 2003) $
Contents:
I. A FEW POINTERS
II. DOCUMENTATION
III. PARTICIPATING IN THE SUBVERSION COMMUNITY
…
</pre></div></div></div><div class="refentry" title="svnlook changed"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnlook.c.changed"></a><div class="titlepage"></div><a id="idp18535488" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnlook changed — Print the paths that were changed.</p></div><div class="refsect1" title="Synopsis"><a id="idp18538576"></a><h2>Synopsis</h2><p><code class="literal">svnlook changed REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18540208"></a><h2>Description</h2><p>Print the paths that were changed in a particular
revision or transaction, as well as <span class="quote">“<span class="quote">svn
update-style</span>”</span> status letters in the first two
columns:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">'A '</code></span></dt><dd><p>Item added to repository</p></dd><dt><span class="term"><code class="literal">'D '</code></span></dt><dd><p>Item deleted from repository</p></dd><dt><span class="term"><code class="literal">'U '</code></span></dt><dd><p>File contents changed</p></dd><dt><span class="term"><code class="literal">'_U'</code></span></dt><dd><p>Properties of item changed; note the leading underscore</p></dd><dt><span class="term"><code class="literal">'UU'</code></span></dt><dd><p>File contents and properties changed</p></dd></dl></div><p>Files and directories can be distinguished, as
directory paths are displayed with a
trailing <span class="quote">“<span class="quote"><code class="literal">/</code></span>”</span>
character.</p></div><div class="refsect1" title="Options"><a id="idp18553792"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnlook.sw.copy_info"><code class="option">--copy-info</code></a>
<a class="xref" href="#svn.ref.svnlook.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>)</a>
<a class="xref" href="#svn.ref.svnlook.sw.transaction"><code class="option">--transaction</code> (<code class="option">-t</code>)</a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18557824"></a><h2>Examples</h2><p>This shows a list of all the changed files and
directories in revision 39 of a test repository. Note
that the first changed item is a directory, as evidenced
by the trailing <code class="literal">/</code>:</p><div class="informalexample"><pre class="screen">
$ svnlook changed -r 39 /var/svn/repos
A trunk/vendors/deli/
A trunk/vendors/deli/chips.txt
A trunk/vendors/deli/sandwich.txt
A trunk/vendors/deli/pickle.txt
U trunk/vendors/baker/bagel.txt
_U trunk/vendors/baker/croissant.txt
UU trunk/vendors/baker/pretzel.txt
D trunk/vendors/baker/baguette.txt
</pre></div><p>Here's an example that shows a revision in which a
file was renamed:</p><div class="informalexample"><pre class="screen">
$ svnlook changed -r 64 /var/svn/repos
A trunk/vendors/baker/toast.txt
D trunk/vendors/baker/bread.txt
</pre></div><p>Unfortunately, nothing in the preceding output reveals
the connection between the deleted and added files. Use
the <code class="option">--copy-info</code> option to make this
relationship more apparent:</p><div class="informalexample"><pre class="screen">
$ svnlook changed -r 64 --copy-info /var/svn/repos
A + trunk/vendors/baker/toast.txt
(from trunk/vendors/baker/bread.txt:r63)
D trunk/vendors/baker/bread.txt
</pre></div></div></div><div class="refentry" title="svnlook date"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnlook.c.date"></a><div class="titlepage"></div><a id="idp18566384" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnlook date — Print the datestamp.</p></div><div class="refsect1" title="Synopsis"><a id="idp18569472"></a><h2>Synopsis</h2><p><code class="literal">svnlook date REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18571088"></a><h2>Description</h2><p>Print the datestamp of a revision or transaction in
a repository.</p></div><div class="refsect1" title="Options"><a id="idp18572368"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnlook.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>)</a>
<a class="xref" href="#svn.ref.svnlook.sw.transaction"><code class="option">--transaction</code> (<code class="option">-t</code>)</a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18575600"></a><h2>Examples</h2><p>This shows the date of revision 40 of a test
repository:</p><div class="informalexample"><pre class="screen">
$ svnlook date -r 40 /var/svn/repos/
2003-02-22 17:44:49 -0600 (Sat, 22 Feb 2003)
</pre></div></div></div><div class="refentry" title="svnlook diff"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnlook.c.diff"></a><div class="titlepage"></div><a id="idp18579216" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnlook diff — Print differences of changed files and properties.</p></div><div class="refsect1" title="Synopsis"><a id="idp18582336"></a><h2>Synopsis</h2><p><code class="literal">svnlook diff REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18583952"></a><h2>Description</h2><p>Print GNU-style differences of changed files and
properties in a repository.</p></div><div class="refsect1" title="Options"><a id="idp18585248"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnlook.sw.diff_copy_from"><code class="option">--diff-copy-from</code></a>
<a class="xref" href="#svn.ref.svnlook.sw.no_diff_added"><code class="option">--no-diff-added</code></a>
<a class="xref" href="#svn.ref.svnlook.sw.no_diff_deleted"><code class="option">--no-diff-deleted</code></a>
<a class="xref" href="#svn.ref.svnlook.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>)</a>
<a class="xref" href="#svn.ref.svnlook.sw.transaction"><code class="option">--transaction</code> (<code class="option">-t</code>)</a>
<a class="xref" href="#svn.ref.svnlook.sw.extensions"><code class="option">--extensions</code> (<code class="option">-x</code>) <em class="replaceable"><code>ARG</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18591568"></a><h2>Examples</h2><p>This shows a newly added (empty) file, a modified
binary file, and a renamed (that is, copied and deleted)
file with modifications:</p><div class="informalexample"><pre class="screen">
$ svnlook diff -r 40 /var/svn/repos
Copied: trunk/relish.txt (from rev 39, trunk/vendors/deli/pickle.txt)
===================================================================
--- trunk/relish.txt (rev 0)
+++ trunk/relish.txt 2013-01-29 20:39:17 UTC (rev 40)
@@ -0,0 +1 @@
+Pickle relish is mostly made from cucumbers.
Deleted: trunk/vendors/deli/pickle.txt
===================================================================
--- trunk/vendors/deli/pickle.txt (rev 39)
+++ trunk/vendors/deli/pickle.txt 2013-01-29 20:39:17 UTC (rev 49)
@@ -1 +0,0 @@
-Pickles are mostly made from cucumbers.
Modified: trunk/vendors/deli/logo.jpg
===================================================================
(Binary files differ)
Added: trunk/vendors/deli/soda.txt
===================================================================
$
</pre></div><p>By default, <span class="command"><strong>svnlook diff</strong></span> will treat
copied files very much like any other added file,
displaying in their entirety the contents of the new file
and merely using a different label to draw the copy/add
distinction. However, you can use the
<code class="option">--diff-copy-from</code> option to cause
<span class="command"><strong>svnlook diff</strong></span> to consider a copied
file as worthy of mention only if it differs from the file
from which it was copied, and to actually describe those
differences.</p><div class="informalexample"><pre class="screen">
$ svnlook diff -r 40 /var/svn/repos --diff-copy-from
Copied: trunk/relish.txt (from rev 39, trunk/vendors/deli/pickle.txt)
===================================================================
--- trunk/vendors/deli/pickle.txt 2013-01-29 20:39:17 UTC (rev 39)
+++ trunk/relish.txt 2013-01-29 20:47:40 UTC (rev 3)
@@ -1 +1 @@
-Pickles are mostly made from cucumbers.
+Pickle relish is mostly made from cucumbers.
Deleted: trunk/vendors/deli/pickle.txt
===================================================================
--- trunk/vendors/deli/pickle.txt (rev 39)
+++ trunk/vendors/deli/pickle.txt 2013-01-29 20:39:17 UTC (rev 40)
@@ -1 +0,0 @@
-Pickles are mostly made from cucumbers.
Modified: trunk/vendors/deli/logo.jpg
===================================================================
(Binary files differ)
Added: trunk/vendors/deli/soda.txt
==============================================================================
$
</pre></div><p>Use the <code class="option">--no-diff-deleted</code> option to
silence output regarding deleted files.</p><div class="informalexample"><pre class="screen">
$ svnlook diff -r 40 /var/svn/repos --no-diff-deleted
Copied: trunk/relish.txt (from rev 39, trunk/vendors/deli/pickle.txt)
===================================================================
--- trunk/relish.txt (rev 0)
+++ trunk/relish.txt 2013-01-29 20:39:17 UTC (rev 40)
@@ -0,0 +1 @@
+Pickle relish is mostly made from cucumbers.
Modified: trunk/vendors/deli/logo.jpg
===================================================================
(Binary files differ)
Added: trunk/vendors/deli/soda.txt
==============================================================================
$
</pre></div><p>Note that in each of the previous examples, when a
file has a nontextual <code class="literal">svn:mime-type</code>
property, the differences are not explicitly shown.</p></div></div><div class="refentry" title="svnlook dirs-changed"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnlook.c.dirs-changed"></a><div class="titlepage"></div><a id="idp18604688" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnlook dirs-changed — Print the directories that were themselves changed.</p></div><div class="refsect1" title="Synopsis"><a id="idp18607840"></a><h2>Synopsis</h2><p><code class="literal">svnlook dirs-changed REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18609424"></a><h2>Description</h2><p>Print the directories that were themselves changed
(property edits) or whose file children were
changed.</p></div><div class="refsect1" title="Options"><a id="idp18610752"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnlook.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>)</a>
<a class="xref" href="#svn.ref.svnlook.sw.transaction"><code class="option">--transaction</code> (<code class="option">-t</code>)</a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18613984"></a><h2>Examples</h2><p>This shows the directories that changed in
revision 40 in our sample repository:</p><div class="informalexample"><pre class="screen">
$ svnlook dirs-changed -r 40 /var/svn/repos
trunk/vendors/deli/
</pre></div></div></div><div class="refentry" title="svnlook filesize"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnlook.c.filesize"></a><div class="titlepage"></div><a id="idp18617664" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnlook filesize — Print the size (in bytes) of a versioned
file.</p></div><div class="refsect1" title="Synopsis"><a id="idp18620832"></a><h2>Synopsis</h2><p><code class="literal">svnlook filesize REPOS_PATH
PATH_IN_REPOS</code></p></div><div class="refsect1" title="Description"><a id="idp18622480"></a><h2>Description</h2><p>Print the file size (in bytes) of the file located
at <em class="replaceable"><code>PATH_IN_REPOS</code></em> in the HEAD
revision of the repository identified
by <em class="replaceable"><code>REPOS_PATH</code></em> as a base-10
integer followed by an end-of-line character. Use
the <code class="option">--revision</code> (<code class="option">-r</code>)
and <code class="option">--transaction</code> (<code class="option">-t</code>)
options to specify a version of the file other than HEAD
whose file size you wish to display.</p></div><div class="refsect1" title="Options"><a id="idp18626544"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnlook.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>)</a>
<a class="xref" href="#svn.ref.svnlook.sw.transaction"><code class="option">--transaction</code> (<code class="option">-t</code>)</a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18629728"></a><h2>Examples</h2><p>The following demonstrates how to display the size of
the <code class="filename">trunk/vendors/deli/soda.txt</code> file
as it appeared in revision 40 of our sample
repository:</p><div class="informalexample"><pre class="screen">
$ svnlook filesize -r 40 /var/svn/repos trunk/vendors/deli/soda.txt
23
$
</pre></div></div></div><div class="refentry" title="svnlook help (h, ?)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnlook.c.help"></a><div class="titlepage"></div><a id="idp18634144" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnlook help (h, ?) — Help!</p></div><div class="refsect1" title="Synopsis"><a id="idp18637232"></a><h2>Synopsis</h2><p><code class="literal">Also svnlook -h and svnlook -?.</code></p></div><div class="refsect1" title="Description"><a id="idp18638864"></a><h2>Description</h2><p>Displays the help message
for <span class="command"><strong>svnlook</strong></span>. This command, like its
brother, <span class="command"><strong>svn help</strong></span>, is also your friend,
even though you never call it anymore and forgot to invite
it to your last party.</p></div><div class="refsect1" title="Options"><a id="idp18641696"></a><h2>Options</h2><p>None</p></div></div><div class="refentry" title="svnlook history"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnlook.c.history"></a><div class="titlepage"></div><a id="idp18644224" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnlook history — Print information about the history of a path in
the repository (or the root directory if no path is
supplied).</p></div><div class="refsect1" title="Synopsis"><a id="idp18647456"></a><h2>Synopsis</h2><p><code class="literal">svnlook history REPOS_PATH [PATH_IN_REPOS]</code></p></div><div class="refsect1" title="Description"><a id="idp18649104"></a><h2>Description</h2><p>Print information about the history of a path in
the repository (or the root directory if no path is
supplied).</p></div><div class="refsect1" title="Options"><a id="idp18650432"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnlook.sw.limit"><code class="option">--limit</code> (<code class="option">-l</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svnlook.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>)</a>
<a class="xref" href="#svn.ref.svnlook.sw.show_ids"><code class="option">--show-ids</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18654480"></a><h2>Examples</h2><p>This shows the history output for the path
<code class="filename">/branches/bookstore</code> as of revision 13 in our
sample repository:</p><div class="informalexample"><pre class="screen">
$ svnlook history -r 13 /var/svn/repos /branches/bookstore --show-ids
REVISION PATH <ID>
-------- ---------
13 /branches/bookstore <1.1.r13/390>
12 /branches/bookstore <1.1.r12/413>
11 /branches/bookstore <1.1.r11/0>
9 /trunk <1.0.r9/551>
8 /trunk <1.0.r8/131357096>
7 /trunk <1.0.r7/294>
6 /trunk <1.0.r6/353>
5 /trunk <1.0.r5/349>
4 /trunk <1.0.r4/332>
3 /trunk <1.0.r3/335>
2 /trunk <1.0.r2/295>
1 /trunk <1.0.r1/532>
</pre></div></div></div><div class="refentry" title="svnlook info"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnlook.c.info"></a><div class="titlepage"></div><a id="idp18659648" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnlook info — Print the author, datestamp, log message size,
and log message.</p></div><div class="refsect1" title="Synopsis"><a id="idp18662800"></a><h2>Synopsis</h2><p><code class="literal">svnlook info REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18664416"></a><h2>Description</h2><p>Print the author, datestamp, log message size (in bytes),
and log message, followed by a newline character.</p></div><div class="refsect1" title="Options"><a id="idp18665728"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnlook.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>)</a>
<a class="xref" href="#svn.ref.svnlook.sw.transaction"><code class="option">--transaction</code> (<code class="option">-t</code>)</a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18668912"></a><h2>Examples</h2><p>This shows the info output for revision 40 in our
sample repository:</p><div class="informalexample"><pre class="screen">
$ svnlook info -r 40 /var/svn/repos
sally
2003-02-22 17:44:49 -0600 (Sat, 22 Feb 2003)
16
Rearrange lunch.
</pre></div></div></div><div class="refentry" title="svnlook lock"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnlook.c.lock"></a><div class="titlepage"></div><a id="idp18672576" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnlook lock — If a lock exists on a path in the repository,
describe it.</p></div><div class="refsect1" title="Synopsis"><a id="idp18675712"></a><h2>Synopsis</h2><p><code class="literal">svnlook lock REPOS_PATH PATH_IN_REPOS</code></p></div><div class="refsect1" title="Description"><a id="idp18677344"></a><h2>Description</h2><p>Print all information available for the lock at
<em class="replaceable"><code>PATH_IN_REPOS</code></em>. If
<em class="replaceable"><code>PATH_IN_REPOS</code></em> is not locked,
print nothing.</p></div><div class="refsect1" title="Options"><a id="idp18679536"></a><h2>Options</h2><p>None</p></div><div class="refsect1" title="Examples"><a id="idp18680800"></a><h2>Examples</h2><p>This describes the lock on the file
<code class="filename">tree.jpg</code>:</p><div class="informalexample"><pre class="screen">
$ svnlook lock /var/svn/repos tree.jpg
UUID Token: opaquelocktoken:ab00ddf0-6afb-0310-9cd0-dda813329753
Owner: harry
Created: 2005-07-08 17:27:36 -0500 (Fri, 08 Jul 2005)
Expires:
Comment (1 line):
Rework the uppermost branches on the bald cypress in the foreground.
</pre></div></div></div><div class="refentry" title="svnlook log"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnlook.c.log"></a><div class="titlepage"></div><a id="idp18685328" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnlook log — Print the log message, followed by a newline
character.</p></div><div class="refsect1" title="Synopsis"><a id="idp18688432"></a><h2>Synopsis</h2><p><code class="literal">svnlook log REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18690048"></a><h2>Description</h2><p>Print the log message.</p></div><div class="refsect1" title="Options"><a id="idp18691264"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnlook.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>)</a>
<a class="xref" href="#svn.ref.svnlook.sw.transaction"><code class="option">--transaction</code> (<code class="option">-t</code>)</a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18694496"></a><h2>Examples</h2><p>This shows the log output for revision 40 in our sample
repository:</p><div class="informalexample"><pre class="screen">
$ svnlook log /var/svn/repos/
Rearrange lunch.
</pre></div></div></div><div class="refentry" title="svnlook propget (pget, pg)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnlook.c.propget"></a><div class="titlepage"></div><a id="idp18698144" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnlook propget (pget, pg) — Print the raw value of a property on a path in
the repository.</p></div><div class="refsect1" title="Synopsis"><a id="idp18701296"></a><h2>Synopsis</h2><p><code class="literal">svnlook propget REPOS_PATH PROPNAME [PATH_IN_REPOS]</code></p></div><div class="refsect1" title="Description"><a id="idp18702944"></a><h2>Description</h2><p>List the value of a property on a path in the
repository.</p></div><div class="refsect1" title="Options"><a id="idp18704208"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnlook.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>)</a>
<a class="xref" href="#svn.ref.svnlook.sw.revprop"><code class="option">--revprop</code></a>
<a class="xref" href="#svn.ref.svnlook.sw.transaction"><code class="option">--transaction</code> (<code class="option">-t</code>)</a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18708240"></a><h2>Examples</h2><p>This shows the value of the <span class="quote">“<span class="quote">seasonings</span>”</span> property
on the file <code class="filename">/trunk/sandwich</code> in the
<code class="literal">HEAD</code> revision:</p><div class="informalexample"><pre class="screen">
$ svnlook pg /var/svn/repos seasonings /trunk/sandwich
mustard
</pre></div></div></div><div class="refentry" title="svnlook proplist (plist, pl)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnlook.c.proplist"></a><div class="titlepage"></div><a id="idp18713760" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnlook proplist (plist, pl) — Print the names and values of versioned file and
directory properties.</p></div><div class="refsect1" title="Synopsis"><a id="idp18716912"></a><h2>Synopsis</h2><p><code class="literal">svnlook proplist REPOS_PATH [PATH_IN_REPOS]</code></p></div><div class="refsect1" title="Description"><a id="idp18718560"></a><h2>Description</h2><p>List the properties of a path in the repository. With
<code class="option">--verbose</code> (<code class="option">-v</code>), show the
property values too.</p></div><div class="refsect1" title="Options"><a id="idp18720688"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnlook.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>)</a>
<a class="xref" href="#svn.ref.svnlook.sw.revprop"><code class="option">--revprop</code></a>
<a class="xref" href="#svn.ref.svnlook.sw.transaction"><code class="option">--transaction</code> (<code class="option">-t</code>)</a>
<a class="xref" href="#svn.ref.svnlook.sw.verbose"><code class="option">--verbose</code> (<code class="option">-v</code>)</a>
<a class="xref" href="#svn.ref.svnlook.sw.xml"><code class="option">--xml</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18726144"></a><h2>Examples</h2><p>This shows the names of properties set on the file
<code class="filename">/trunk/README</code> in the
<code class="literal">HEAD</code> revision:</p><div class="informalexample"><pre class="screen">
$ svnlook proplist /var/svn/repos /trunk/README
original-author
svn:mime-type
</pre></div><p>This is the same command as in the preceding example,
but this time showing the property values as well:</p><div class="informalexample"><pre class="screen">
$ svnlook -v proplist /var/svn/repos /trunk/README
original-author : harry
svn:mime-type : text/plain
</pre></div></div></div><div class="refentry" title="svnlook tree"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnlook.c.tree"></a><div class="titlepage"></div><a id="idp18732896" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnlook tree — Print the tree.</p></div><div class="refsect1" title="Synopsis"><a id="idp18735984"></a><h2>Synopsis</h2><p><code class="literal">svnlook tree REPOS_PATH [PATH_IN_REPOS]</code></p></div><div class="refsect1" title="Description"><a id="idp18737616"></a><h2>Description</h2><p>Print the tree, starting at
<em class="replaceable"><code>PATH_IN_REPOS</code></em> (if supplied; at
the root of the tree otherwise), optionally showing node
revision IDs.</p></div><div class="refsect1" title="Options"><a id="idp18739408"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnlook.sw.full_paths"><code class="option">--full-paths</code></a>
<a class="xref" href="#svn.ref.svnlook.sw.non_recursive"><code class="option">--non-recursive</code> (<code class="option">-N</code>)</a>
<a class="xref" href="#svn.ref.svnlook.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>)</a>
<a class="xref" href="#svn.ref.svnlook.sw.show_ids"><code class="option">--show-ids</code></a>
<a class="xref" href="#svn.ref.svnlook.sw.transaction"><code class="option">--transaction</code> (<code class="option">-t</code>)</a>
</pre></div></div><div class="refsect1" title="Example"><a id="idp18744848"></a><h2>Example</h2><p>This shows the tree output for revision 13 in our
sample repository:</p><div class="informalexample"><pre class="screen">
$ svnlook tree -r 13 /var/svn/repos
/
trunk/
button.c
Makefile
integer.c
branches/
bookstore/
button.c
Makefile
integer.c
…
</pre></div><p>Use the <code class="option">--show-ids</code> option to include
node revision IDs (unique internal identifiers for
specific nodes in Subversion's versioned filesystem
implementation):</p><div class="informalexample"><pre class="screen">
$ svnlook tree -r 13 /var/svn/repos --show-ids
/ <0.0.r13/811>
trunk/ <1.0.r9/551>
button.c <2.0.r9/238>
Makefile <3.0.r7/41>
integer.c <4.0.r6/98>
branches/ <5.0.r13/593>
bookstore/ <1.1.r13/390>
button.c <2.1.r12/85>
Makefile <3.0.r7/41>
integer.c <4.1.r13/109>
…
</pre></div><p>For output which lends itself more readily to being
parsed by scripts, use the <code class="option">--full-paths</code>
option, which causes <span class="command"><strong>svnlook</strong></span> to print
the full repository path of each tree item and to not use
indentation to indicate hierarchy:</p><div class="informalexample"><pre class="screen">
$ svnlook tree -r 13 /var/svn/repos --show-ids --full-paths
/ <0.0.r13/811>
trunk/ <1.0.r9/551>
trunk/button.c <2.0.r9/238>
trunk/Makefile <3.0.r7/41>
trunk/integer.c <4.0.r6/98>
branches/ <5.0.r13/593>
branches/bookstore/ <1.1.r13/390>
branches/bookstore/button.c <2.1.r12/85>
branches/bookstore/Makefile <3.0.r7/41>
branches/bookstore/integer.c <4.1.r13/109>
…
</pre></div></div></div><div class="refentry" title="svnlook uuid"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnlook.c.uuid"></a><div class="titlepage"></div><a id="idp18754176" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnlook uuid — Print the repository's
<code class="literal">UUID</code>.</p></div><div class="refsect1" title="Synopsis"><a id="idp18757968"></a><h2>Synopsis</h2><p><code class="literal">svnlook uuid REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18759584"></a><h2>Description</h2><p>Print the <code class="literal">UUID</code> for the repository.
The <code class="literal">UUID</code> is the repository's
<span class="emphasis"><em>u</em></span>niversal <span class="emphasis"><em>u</em></span>nique
<span class="emphasis"><em>id</em></span>entifier. The Subversion client
uses this identifier to differentiate between one
repository and another.</p></div><div class="refsect1" title="Options"><a id="idp18763680"></a><h2>Options</h2><p>None</p></div><div class="refsect1" title="Examples"><a id="idp18764896"></a><h2>Examples</h2><div class="informalexample"><pre class="screen">
$ svnlook uuid /var/svn/repos
e7fe1b91-8cd5-0310-98dd-2f12e793c5e8
</pre></div></div></div><div class="refentry" title="svnlook youngest"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnlook.c.youngest"></a><div class="titlepage"></div><a id="idp18768016" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnlook youngest — Print the youngest revision number.</p></div><div class="refsect1" title="Synopsis"><a id="idp18771104"></a><h2>Synopsis</h2><p><code class="literal">svnlook youngest REPOS_PATH</code></p></div><div class="refsect1" title="Description"><a id="idp18772736"></a><h2>Description</h2><p>Print the youngest revision number of a repository.</p></div><div class="refsect1" title="Options"><a id="idp18773984"></a><h2>Options</h2><p>None</p></div><div class="refsect1" title="Examples"><a id="idp18775200"></a><h2>Examples</h2><p>This shows the youngest revision of our sample
repository:</p><div class="informalexample"><pre class="screen">
$ svnlook youngest /var/svn/repos/
42
</pre></div></div></div></div></div><div class="sect1" title="svnsync—Subversion Repository Mirroring"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.ref.svnsync"></a>svnsync—Subversion Repository Mirroring</h2></div></div></div><p><span class="command"><strong>svnsync</strong></span> is the Subversion remote
repository mirroring tool. Put simply, it allows you to replay
the revisions of one repository into another one.</p><p>In any mirroring scenario, there are two repositories: the
source repository, and the mirror (or <span class="quote">“<span class="quote">sink</span>”</span>)
repository. The source repository is the repository from which
<span class="command"><strong>svnsync</strong></span> pulls revisions. The mirror
repository is the destination for the revisions pulled from the
source repository. Each of the repositories may be local or
remote—they are only ever addressed by their URLs.</p><p>The <span class="command"><strong>svnsync</strong></span> process requires only read
access to the source repository; it never attempts to modify it.
But obviously, <span class="command"><strong>svnsync</strong></span> requires both read and
write access to the mirror repository.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p><span class="command"><strong>svnsync</strong></span> is very sensitive to changes
made in the mirror repository that weren't made as part of a
mirroring operation. To prevent this from happening, it's
best if the <span class="command"><strong>svnsync</strong></span> process is the only
process permitted to modify the mirror repository.</p></td></tr></table></div><div class="sect2" title="svnsync Options"><div class="titlepage"><div><div><h3 class="title"><a id="svn.ref.svnsync.sw"></a>svnsync Options</h3></div></div></div><p>Options in <span class="command"><strong>svnsync</strong></span> are global, just
as they are in <span class="command"><strong>svn</strong></span> and
<span class="command"><strong>svnadmin</strong></span>:</p><div class="variablelist"><dl><dt><a id="svn.ref.svnsync.sw.allow_non_empty"></a><span class="term"><code class="option">--allow-non-empty</code></span></dt><dd><p>Disables the verification (which <span class="command"><strong>svnsync
initialize</strong></span> performs by default) that the
repository being initialized is empty of history
version.</p></dd><dt><a id="svn.ref.svnsync.sw.config_dir"></a><span class="term"><code class="option">--config-dir</code> <em class="replaceable"><code>DIR</code></em></span></dt><dd><p>Instructs Subversion to read configuration
information from the specified directory instead of the
default location (<code class="filename">.subversion</code> in
the user's home directory).</p></dd><dt><a id="svn.ref.svnsync.sw.config_option"></a><span class="term"><code class="option">--config-option</code> <em class="replaceable"><code>CONFSPEC</code></em></span></dt><dd><p>Sets, for the duration of the command, the value of
a runtime configuration
option. <em class="replaceable"><code>CONFSPEC</code></em> is a string
which specifies the configuration option namespace, name
and value that you'd like to assign, formatted as
<em class="replaceable"><code>FILE</code></em>:<em class="replaceable"><code>SECTION</code></em>:<em class="replaceable"><code>OPTION</code></em>=[<em class="replaceable"><code>VALUE</code></em>].
In this syntax, <em class="replaceable"><code>FILE</code></em>
and <em class="replaceable"><code>SECTION</code></em> are the runtime
configuration file (either <code class="literal">config</code>
or <code class="literal">servers</code>) and the section thereof,
respectively, which contain the option whose value you
wish to change. <em class="replaceable"><code>OPTION</code></em> is,
of course, the option itself,
and <em class="replaceable"><code>VALUE</code></em> the value (if any)
you wish to assign to the option. For example, to
temporarily disable the use of the compression in the
HTTP protocol,
use <code class="option">--config-option=servers:global:http-compression=no</code>.
You can use this option multiple times to change
multiple option values simultaneously.</p></dd><dt><a id="svn.ref.svnsync.sw.disable_locking"></a><span class="term"><code class="option">--disable-locking</code></span></dt><dd><p>Causes <span class="command"><strong>svnsync</strong></span> to bypass its own
exclusive access mechanisms and operate on the
assumption that its exclusive access to the mirror
repository is being guaranteed through some other,
out-of-band mechanism.</p></dd><dt><a id="svn.ref.svnsync.sw.no_auth_cache"></a><span class="term"><code class="option">--no-auth-cache</code></span></dt><dd><p>Prevents caching of authentication information
(e.g., username and password) in the Subversion
runtime configuration directories.</p></dd><dt><a id="svn.ref.svnsync.sw.non_interactive"></a><span class="term"><code class="option">--non-interactive</code></span></dt><dd><p>In the case of an authentication failure or
insufficient credentials, prevents prompting for
credentials (e.g., username or password). This is useful
if you're running Subversion inside an automated
script and it's more appropriate to have Subversion fail
than to prompt for more information.</p></dd><dt><a id="svn.ref.svnsync.sw.quiet"></a><span class="term"><code class="option">--quiet</code> (<code class="option">-q</code>)</span></dt><dd><p>Requests that the client print only essential
information while performing an operation.</p></dd><dt><a id="svn.ref.svnsync.sw.revision"></a><span class="term"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>Used by <span class="command"><strong>svnsync copy-revprops</strong></span> to
specify a particular revision or revision range on which
to operate.</p></dd><dt><a id="svn.ref.svnsync.sw.source_password"></a><span class="term"><code class="option">--source-password</code> <em class="replaceable"><code>PASSWD</code></em></span></dt><dd><p>Specifies the password for the Subversion server
from which you are syncing. If not provided, or if
incorrect, Subversion will prompt you for this
information as needed.</p></dd><dt><a id="svn.ref.svnsync.sw.source_prop_encoding"></a><span class="term"><code class="option">--source-prop-encoding <em class="replaceable"><code>ARG</code></em></code></span></dt><dd><p>Instructs <span class="command"><strong>svnsync</strong></span> to assume that
translatable Subversion revision properties found in the
source repository are stored using the character
encoding <em class="replaceable"><code>ARG</code></em> and to transcode
those into UTF-8 when copying them into the mirror
repository.</p></dd><dt><a id="svn.ref.svnsync.sw.source_username"></a><span class="term"><code class="option">--source-username</code> <em class="replaceable"><code>NAME</code></em></span></dt><dd><p>Specifies the username for the Subversion server
from which you are syncing. If not provided, or if
incorrect, Subversion will prompt you for this
information as needed.</p></dd><dt><a id="svn.ref.svnsync.sw.steal_lock"></a><span class="term"><code class="option">--steal-lock</code></span></dt><dd><p>Causes <span class="command"><strong>svnsync</strong></span> to steal, as
necessary, the lock which it uses on the mirror
repository to ensure exclusive repository access. (This
option should only be used when a lock exists in the
mirror repository and is known to be stale—that
is, when you are certain that there are no
other <span class="command"><strong>svnsync</strong></span> processes accessing
that repository.)</p></dd><dt><a id="svn.ref.svnsync.sw.sync_password"></a><span class="term"><code class="option">--sync-password</code> <em class="replaceable"><code>PASSWD</code></em></span></dt><dd><p>Specifies the password for the Subversion server to
which you are syncing. If not provided, or if
incorrect, Subversion will prompt you for this
information as needed.</p></dd><dt><a id="svn.ref.svnsync.sw.sync_username"></a><span class="term"><code class="option">--sync-username</code> <em class="replaceable"><code>NAME</code></em></span></dt><dd><p>Specifies the username for the Subversion server to
which you are syncing. If not provided, or if
incorrect, Subversion will prompt you for this
information as needed.</p></dd><dt><a id="svn.ref.svnsync.sw.trust_server_cert"></a><span class="term"><code class="option">--trust-server-cert</code></span></dt><dd><p>Used with <code class="option">--non-interactive</code> to accept
any unknown SSL server certificates without prompting.</p></dd></dl></div></div><div class="sect2" title="svnsync Subcommands"><div class="titlepage"><div><div><h3 class="title"><a id="svn.ref.svnsync.c"></a>svnsync Subcommands</h3></div></div></div><p>Here are the various subcommands
for the <span class="command"><strong>svnsync</strong></span> program.</p><div class="refentry" title="svnsync copy-revprops"><a id="svn.ref.svnsync.c.copy-revprops"></a><div class="titlepage"></div><a id="idp18848352" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnsync copy-revprops — Copy all revision properties for a particular
revision (or range of revisions) from the source
repository to the mirror repository.</p></div><div class="refsect1" title="Synopsis"><a id="idp18851616"></a><h2>Synopsis</h2><p><code class="literal">svnsync copy-revprops DEST_URL
[SOURCE_URL]</code></p><p><code class="literal">svnsync copy-revprops DEST_URL
REV[:REV2]</code></p></div><div class="refsect1" title="Description"><a id="idp18854112"></a><h2>Description</h2><p>Because Subversion revision properties can be changed
at any time, it's possible that the properties for some
revision might be changed after that revision has already
been synchronized to another repository. Because the
<span class="command"><strong>svnsync synchronize</strong></span> command operates
only on the range of revisions that have not yet been
synchronized, it won't notice a revision property change
outside that range. Left as is, this causes a deviation
in the values of that revision's properties between the
source and mirror repositories. <span class="command"><strong>svnsync
copy-revprops</strong></span> is the answer to this problem.
Use it to resynchronize the revision properties for a
particular revision or range of revisions.</p><p>When <em class="replaceable"><code>SOURCE_URL</code></em> is
provided, <span class="command"><strong>svnsync</strong></span> will use it as the
repository URL which the destination repository is
mirroring. Generally, <em class="replaceable"><code>SOURCE_URL</code></em>
will be exactly the same source URL as was used with
the <span class="command"><strong>svnsync initialize</strong></span> command when the
mirror was first set up. You may choose, however, to
omit <em class="replaceable"><code>SOURCE_URL</code></em>, in which
case <span class="command"><strong>svnsync</strong></span> will consult the mirror
repository's records to determine the source URL which
should be used.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>We strongly recommend that you specify the source
URL on the command-line, especially when untrusted users
have write access to the revision 0 properties
which <span class="command"><strong>svnsync</strong></span> uses to coordinate its
efforts.</p></td></tr></table></div></div><div class="refsect1" title="Options"><a id="idp18863488"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnsync.sw.config_dir"><code class="option">--config-dir</code> <em class="replaceable"><code>DIR</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.config_option"><code class="option">--config-option</code> <em class="replaceable"><code>CONFSPEC</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.disable_locking"><code class="option">--disable-locking</code></a>
<a class="xref" href="#svn.ref.svnsync.sw.no_auth_cache"><code class="option">--no-auth-cache</code></a>
<a class="xref" href="#svn.ref.svnsync.sw.non_interactive"><code class="option">--non-interactive</code></a>
<a class="xref" href="#svn.ref.svnsync.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svnsync.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.source_password"><code class="option">--source-password</code> <em class="replaceable"><code>PASSWD</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.source_prop_encoding"><code class="option">--source-prop-encoding <em class="replaceable"><code>ARG</code></em></code></a>
<a class="xref" href="#svn.ref.svnsync.sw.source_username"><code class="option">--source-username</code> <em class="replaceable"><code>NAME</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.steal_lock"><code class="option">--steal-lock</code></a>
<a class="xref" href="#svn.ref.svnsync.sw.sync_password"><code class="option">--sync-password</code> <em class="replaceable"><code>PASSWD</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.sync_username"><code class="option">--sync-username</code> <em class="replaceable"><code>NAME</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.trust_server_cert"><code class="option">--trust-server-cert</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18876320"></a><h2>Examples</h2><p>Resynchronize the revision properties associated with
a single revision (r6):</p><div class="informalexample"><pre class="screen">
$ svnsync copy-revprops -r 6 file:///var/svn/repos-mirror \
http://svn.example.com/repos
Copied properties for revision 6.
$
</pre></div></div></div><div class="refentry" title="svnsync help"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnsync.c.help"></a><div class="titlepage"></div><a id="idp18880048" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnsync help — Help!</p></div><div class="refsect1" title="Synopsis"><a id="idp18883136"></a><h2>Synopsis</h2><p><code class="literal">svnsync help</code></p></div><div class="refsect1" title="Description"><a id="idp18884752"></a><h2>Description</h2><p>This subcommand is useful when you're trapped in a
foreign prison with neither a Net connection nor a copy of
this book, but you do have a local Wi-Fi network running
and you'd like to sync a copy of your repository over to
the backup server that Ira The Knife is running over in
cell block D.</p></div><div class="refsect1" title="Options"><a id="idp18886304"></a><h2>Options</h2><p>None</p></div></div><div class="refentry" title="svnsync info"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnsync.c.info"></a><div class="titlepage"></div><a id="idp18888720" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnsync info — Print information about the synchronization
of a destination repository.</p></div><div class="refsect1" title="Synopsis"><a id="idp18891872"></a><h2>Synopsis</h2><p><code class="literal">svnsync info DEST_URL</code></p></div><div class="refsect1" title="Description"><a id="idp18893488"></a><h2>Description</h2><p>Print the synchronization source URL, source repository
UUID and the last revision merged from the source to the
destination repository at <em class="replaceable"><code>DEST_URL</code></em>.</p></div><div class="refsect1" title="Options"><a id="idp18895264"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnsync.sw.config_dir"><code class="option">--config-dir</code> <em class="replaceable"><code>DIR</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.config_option"><code class="option">--config-option</code> <em class="replaceable"><code>CONFSPEC</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.no_auth_cache"><code class="option">--no-auth-cache</code></a>
<a class="xref" href="#svn.ref.svnsync.sw.non_interactive"><code class="option">--non-interactive</code></a>
<a class="xref" href="#svn.ref.svnsync.sw.source_password"><code class="option">--source-password</code> <em class="replaceable"><code>PASSWD</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.source_username"><code class="option">--source-username</code> <em class="replaceable"><code>NAME</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.sync_password"><code class="option">--sync-password</code> <em class="replaceable"><code>PASSWD</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.sync_username"><code class="option">--sync-username</code> <em class="replaceable"><code>NAME</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.trust_server_cert"><code class="option">--trust-server-cert</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18903200"></a><h2>Examples</h2><p>Print the synchronization information of a mirror
repository:</p><div class="informalexample"><pre class="screen">
$ svnsync info file:///var/svn/repos-mirror
Source URL: http://svn.example.com/repos
Source Repository UUID: e7fe1b91-8cd5-0310-98dd-2f12e793c5e8
Last Merged Revision: 47
$
</pre></div></div></div><div class="refentry" title="svnsync initialize (init)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnsync.c.init"></a><div class="titlepage"></div><a id="idp18906928" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnsync initialize (init) — Initialize a mirror repository for
synchronization from the source repository.</p></div><div class="refsect1" title="Synopsis"><a id="idp18910112"></a><h2>Synopsis</h2><p><code class="literal">svnsync initialize MIRROR_URL
SOURCE_URL</code></p></div><div class="refsect1" title="Description"><a id="idp18911760"></a><h2>Description</h2><p><span class="command"><strong>svnsync initialize</strong></span> verifies that a
repository meets the basic requirements of a new mirror
repository and records the initial administrative
information that associates the mirror repository with the
source repository (specified
by <em class="replaceable"><code>SOURCE_URL</code></em>). This is the
first <span class="command"><strong>svnsync</strong></span> operation you run on a
would-be mirror repository.</p><p>Ordinarily, <em class="replaceable"><code>SOURCE_URL</code></em> is
the URL of the root directory of the Subversion repository
you wish to mirror. Subversion 1.5 and newer allow you to
use <span class="command"><strong>svnsync</strong></span> for partial repository
mirroring, though — simply specify the URL of the
source repository subdirectory you wish to mirror
as <em class="replaceable"><code>SOURCE_URL</code></em>.</p><p>By default, the aforementioned basic requirements of a
mirror are that it allows revision property modifications
and that it contains no version history. However, as of
Subversion 1.7, you may now optionally disable the
verification that the target repository is empty using
the <code class="option">--allow-non-empty</code> option. While the
use of this option should not become habitual (as it
bypasses a valuable safeguard mechanism), it does aid in
one very common use-case: initializing a copy of a
repository as a mirror of the original. This is
especially handy when setting up new mirrors of
repositories which contain a large amount of version
history. Rather than initialize a brand new repository as
a mirror and then syncronize all of the history into it,
administrators will find it <span class="emphasis"><em>significantly</em></span>
faster to first make a copy of the mature repository
(perhaps using <span class="command"><strong>svnadmin hotcopy</strong></span>) and
then use <span class="command"><strong>svnsync initialize
--allow-non-empty</strong></span> to initialize that copy as a
mirror which is now already up-to-date with the
original.</p></div><div class="refsect1" title="Options"><a id="idp18921168"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnsync.sw.allow_non_empty"><code class="option">--allow-non-empty</code></a>
<a class="xref" href="#svn.ref.svnsync.sw.config_dir"><code class="option">--config-dir</code> <em class="replaceable"><code>DIR</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.config_option"><code class="option">--config-option</code> <em class="replaceable"><code>CONFSPEC</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.disable_locking"><code class="option">--disable-locking</code></a>
<a class="xref" href="#svn.ref.svnsync.sw.no_auth_cache"><code class="option">--no-auth-cache</code></a>
<a class="xref" href="#svn.ref.svnsync.sw.non_interactive"><code class="option">--non-interactive</code></a>
<a class="xref" href="#svn.ref.svnsync.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svnsync.sw.source_password"><code class="option">--source-password</code> <em class="replaceable"><code>PASSWD</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.source_prop_encoding"><code class="option">--source-prop-encoding <em class="replaceable"><code>ARG</code></em></code></a>
<a class="xref" href="#svn.ref.svnsync.sw.source_username"><code class="option">--source-username</code> <em class="replaceable"><code>NAME</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.steal_lock"><code class="option">--steal-lock</code></a>
<a class="xref" href="#svn.ref.svnsync.sw.sync_password"><code class="option">--sync-password</code> <em class="replaceable"><code>PASSWD</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.sync_username"><code class="option">--sync-username</code> <em class="replaceable"><code>NAME</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.trust_server_cert"><code class="option">--trust-server-cert</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18932592"></a><h2>Examples</h2><p>Fail to initialize a mirror repository due to
inability to modify revision properties:</p><div class="informalexample"><pre class="screen">
$ svnsync initialize file:///var/svn/repos-mirror \
http://svn.example.com/repos
svnsync: Repository has not been enabled to accept revision propchanges;
ask the administrator to create a pre-revprop-change hook
$
</pre></div><p>Initialize a repository as a mirror, having already
created a <code class="literal">pre-revprop-change</code> hook that
permits all revision property changes:</p><div class="informalexample"><pre class="screen">
$ svnsync initialize file:///var/svn/repos-mirror \
http://svn.example.com/repos
Copied properties for revision 0.
$
</pre></div></div></div><div class="refentry" title="svnsync synchronize (sync)"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnsync.c.sync"></a><div class="titlepage"></div><a id="idp18938848" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnsync synchronize (sync) — Transfer all pending revisions from the source
repository to the mirror repository.</p></div><div class="refsect1" title="Synopsis"><a id="idp18941984"></a><h2>Synopsis</h2><p><code class="literal">svnsync synchronize DEST_URL
[SOURCE_URL]</code></p></div><div class="refsect1" title="Description"><a id="idp18943632"></a><h2>Description</h2><p>The <span class="command"><strong>svnsync synchronize</strong></span> command
does all the heavy lifting of a repository mirroring
operation. After consulting with the mirror repository to
see which revisions have already been copied into it, it
then begins to copy any not-yet-mirrored revisions from
the source repository.</p><p><span class="command"><strong>svnsync synchronize</strong></span> can be
gracefully canceled and restarted.</p><p>When <em class="replaceable"><code>SOURCE_URL</code></em> is
provided, <span class="command"><strong>svnsync</strong></span> will use it as the
repository URL which the destination repository is
mirroring. Generally, <em class="replaceable"><code>SOURCE_URL</code></em>
will be exactly the same source URL as was used with
the <span class="command"><strong>svnsync initialize</strong></span> command when the
mirror was first set up. You may choose, however, to
omit <em class="replaceable"><code>SOURCE_URL</code></em>, in which
case <span class="command"><strong>svnsync</strong></span> will consult the mirror
repository's records to determine the source URL which
should be used.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>We strongly recommend that you specify the source
URL on the command-line, especially when untrusted users
have write access to the revision 0 properties
which <span class="command"><strong>svnsync</strong></span> uses to coordinate its
efforts.</p></td></tr></table></div></div><div class="refsect1" title="Options"><a id="idp18952912"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnsync.sw.config_dir"><code class="option">--config-dir</code> <em class="replaceable"><code>DIR</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.config_option"><code class="option">--config-option</code> <em class="replaceable"><code>CONFSPEC</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.disable_locking"><code class="option">--disable-locking</code></a>
<a class="xref" href="#svn.ref.svnsync.sw.no_auth_cache"><code class="option">--no-auth-cache</code></a>
<a class="xref" href="#svn.ref.svnsync.sw.non_interactive"><code class="option">--non-interactive</code></a>
<a class="xref" href="#svn.ref.svnsync.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svnsync.sw.source_password"><code class="option">--source-password</code> <em class="replaceable"><code>PASSWD</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.source_prop_encoding"><code class="option">--source-prop-encoding <em class="replaceable"><code>ARG</code></em></code></a>
<a class="xref" href="#svn.ref.svnsync.sw.source_username"><code class="option">--source-username</code> <em class="replaceable"><code>NAME</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.steal_lock"><code class="option">--steal-lock</code></a>
<a class="xref" href="#svn.ref.svnsync.sw.sync_password"><code class="option">--sync-password</code> <em class="replaceable"><code>PASSWD</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.sync_username"><code class="option">--sync-username</code> <em class="replaceable"><code>NAME</code></em></a>
<a class="xref" href="#svn.ref.svnsync.sw.trust_server_cert"><code class="option">--trust-server-cert</code></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp18963536"></a><h2>Examples</h2><p>Copy unsynchronized revisions from the source
repository to the mirror repository:</p><div class="informalexample"><pre class="screen">
$ svnsync synchronize file:///var/svn/repos-mirror \
http://svn.example.com/repos
Committed revision 1.
Copied properties for revision 1.
Committed revision 2.
Copied properties for revision 2.
Committed revision 3.
Copied properties for revision 3.
…
Committed revision 45.
Copied properties for revision 45.
Committed revision 46.
Copied properties for revision 46.
Committed revision 47.
Copied properties for revision 47.
$
</pre></div></div></div></div></div><div class="sect1" title="svnrdump—Remote Subversion Repository Data Migration"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.ref.svnrdump"></a>svnrdump—Remote Subversion Repository Data Migration</h2></div></div></div><p><span class="command"><strong>svnrdump</strong></span> joined the Subversion tool chain
in the Subversion 1.7 release. It is best described as a
network-aware version of the <span class="command"><strong>svnadmin dump</strong></span>
and <span class="command"><strong>svnadmin load</strong></span> commands, paired together
and released as a separate standalone program. We discuss the
process of dumping and loading repository data—using
both <span class="command"><strong>svnadmin</strong></span>
and <span class="command"><strong>svnrdump</strong></span>— in
<a class="xref" href="#svn.reposadmin.maint.migrate" title="Migrating Repository Data Elsewhere">the section called “Migrating Repository Data Elsewhere”</a>.</p><div class="sect2" title="svnrdump Options"><div class="titlepage"><div><div><h3 class="title"><a id="svn.ref.svnrdump.sw"></a>svnrdump Options</h3></div></div></div><p>Options in <span class="command"><strong>svnrdump</strong></span> are global, just
as they are in <span class="command"><strong>svn</strong></span> and
<span class="command"><strong>svnadmin</strong></span>:</p><div class="variablelist"><dl><dt><a id="svn.ref.svnrdump.sw.config_dir"></a><span class="term"><code class="option">--config-dir</code> <em class="replaceable"><code>DIR</code></em></span></dt><dd><p>Instructs Subversion to read configuration
information from the specified directory instead of the
default location (<code class="filename">.subversion</code> in
the user's home directory).</p></dd><dt><a id="svn.ref.svnrdump.sw.config_option"></a><span class="term"><code class="option">--config-option</code> <em class="replaceable"><code>FILE</code></em>:<em class="replaceable"><code>SECTION</code></em>:<em class="replaceable"><code>OPTION</code></em>=[<em class="replaceable"><code>VALUE</code></em>]</span></dt><dd><p>Sets, for the duration of the command, the value of
a runtime configuration
option. <em class="replaceable"><code>FILE</code></em>
and <em class="replaceable"><code>SECTION</code></em> are the runtime
configuration file (either <code class="literal">config</code>
or <code class="literal">servers</code>) and the section thereof,
respectively, which contain the option whose value you
wish to change. <em class="replaceable"><code>OPTION</code></em> is,
of course, the option itself,
and <em class="replaceable"><code>VALUE</code></em> the value (if any)
you wish to assign to the option. For example, to
temporarily disable the use of the compression in the
HTTP protocol,
use <code class="option">--config-option=servers:global:http-compression=no</code>.
You can use this option multiple times to change
multiple option values simultaneously.</p></dd><dt><a id="svn.ref.svnrdump.sw.incremental"></a><span class="term"><code class="option">--incremental</code></span></dt><dd><p>Dump a revision or revision range only as a diff
against the previous revision, instead of the default,
which is begin a dumped revision range with a complete
expansion of all contents of the repository as of that
revision.</p></dd><dt><a id="svn.ref.svnrdump.sw.no_auth_cache"></a><span class="term"><code class="option">--no-auth-cache</code></span></dt><dd><p>Prevents caching of authentication information
(e.g., username and password) in the Subversion
runtime configuration directories.</p></dd><dt><a id="svn.ref.svnrdump.sw.non_interactive"></a><span class="term"><code class="option">--non-interactive</code></span></dt><dd><p>In the case of an authentication failure or
insufficient credentials, prevents prompting for
credentials (e.g., username or password). This is useful
if you're running Subversion inside an automated
script and it's more appropriate to have Subversion fail
than to prompt for more information.</p></dd><dt><a id="svn.ref.svnrdump.sw.password"></a><span class="term"><code class="option">--password</code> <em class="replaceable"><code>PASSWD</code></em></span></dt><dd><p>Specifies the password to use when authenticating
against a Subversion server. If not provided, or if
incorrect, Subversion will prompt you for this
information as needed.</p></dd><dt><a id="svn.ref.svnrdump.sw.quiet"></a><span class="term"><code class="option">--quiet</code> (<code class="option">-q</code>)</span></dt><dd><p>Requests that the client print only essential
information while performing an operation.</p></dd><dt><a id="svn.ref.svnrdump.sw.revision"></a><span class="term"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>Used by <span class="command"><strong>svnsync copy-revprops</strong></span> to
specify a particular revision or revision range on which
to operate.</p></dd><dt><a id="svn.ref.svnrdump.sw.trust_server_cert"></a><span class="term"><code class="option">--trust-server-cert</code></span></dt><dd><p>Used with <code class="option">--non-interactive</code> to accept
any unknown SSL server certificates without prompting.</p></dd><dt><a id="svn.ref.svnrdump.sw.username"></a><span class="term"><code class="option">--username</code> <em class="replaceable"><code>NAME</code></em></span></dt><dd><p>Specifies the username to use when authenticating
against a Subversion server. If not provided, or if
incorrect, Subversion will prompt you for this
information as needed.</p></dd></dl></div></div><div class="sect2" title="svnrdump Subcommands"><div class="titlepage"><div><div><h3 class="title"><a id="svn.ref.svnrdump.c"></a>svnrdump Subcommands</h3></div></div></div><p>Here are the various subcommands
for the <span class="command"><strong>svnrdump</strong></span> program.</p><div class="refentry" title="svnrdump dump"><a id="svn.ref.svnrdump.c.dump"></a><div class="titlepage"></div><a id="idp19016224" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnrdump dump</p></div><div class="refsect1" title="Synopsis"><a id="idp19019120"></a><h2>Synopsis</h2><p><code class="literal">svnrdump dump SOURCE_URL</code></p></div><div class="refsect1" title="Description"><a id="idp19020752"></a><h2>Description</h2><p>Dump—that is, generate a repository dump stream
of—revisions of the repository item located
at <em class="replaceable"><code>SOURCE_URL</code></em>, printing the
information to standard output. By default, the entire
history will be included in the dump stream, but the scope
of the operation can be limited via the use of
the <code class="option">--revision</code> (<code class="option">-r</code>)
option.</p></div><div class="refsect1" title="Options"><a id="idp19023568"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnrdump.sw.config_dir"><code class="option">--config-dir</code> <em class="replaceable"><code>DIR</code></em></a>
<a class="xref" href="#svn.ref.svnrdump.sw.config_option"><code class="option">--config-option</code> <em class="replaceable"><code>FILE</code></em>:<em class="replaceable"><code>SECTION</code></em>:<em class="replaceable"><code>OPTION</code></em>=[<em class="replaceable"><code>VALUE</code></em>]</a>
<a class="xref" href="#svn.ref.svnrdump.sw.incremental"><code class="option">--incremental</code></a>
<a class="xref" href="#svn.ref.svnrdump.sw.no_auth_cache"><code class="option">--no-auth-cache</code></a>
<a class="xref" href="#svn.ref.svnrdump.sw.non_interactive"><code class="option">--non-interactive</code></a>
<a class="xref" href="#svn.ref.svnrdump.sw.password"><code class="option">--password</code> <em class="replaceable"><code>PASSWD</code></em></a>
<a class="xref" href="#svn.ref.svnrdump.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svnrdump.sw.revision"><code class="option">--revision</code> (<code class="option">-r</code>) <em class="replaceable"><code>ARG</code></em></a>
<a class="xref" href="#svn.ref.svnrdump.sw.trust_server_cert"><code class="option">--trust-server-cert</code></a>
<a class="xref" href="#svn.ref.svnrdump.sw.username"><code class="option">--username</code> <em class="replaceable"><code>NAME</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp19033200"></a><h2>Examples</h2><p>Generate a dump stream of the full history of a remote
repository (assuming that the user as who this runs has
authorization to read all paths in the repository).</p><div class="informalexample"><pre class="screen">
$ svnrdump dump http://svn.example.com/repos/calc > full.dump
* Dumped revision 0.
* Dumped revision 1.
* Dumped revision 2.
…
</pre></div><p>Incrementally dump a single revision from that same
repository:</p><div class="informalexample"><pre class="screen">
$ svnrdump dump http://svn.example.com/repos/calc \
-r 21 --incremental > inc.dump
* Dumped revision 21.
$
</pre></div></div></div><div class="refentry" title="svnrdump help"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnrdump.c.help"></a><div class="titlepage"></div><a id="idp19038640" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnrdump help — Help!</p></div><div class="refsect1" title="Synopsis"><a id="idp19041728"></a><h2>Synopsis</h2><p><code class="literal">svnrdump help</code></p></div><div class="refsect1" title="Description"><a id="idp19043344"></a><h2>Description</h2><p>Use this to get help. Well, certain kinds of help.
Need white lab coat and straightjackets kind of help?
There's a whole different protocol for that sort of
thing.</p></div><div class="refsect1" title="Options"><a id="idp19044752"></a><h2>Options</h2><p>None</p></div></div><div class="refentry" title="svnrdump load"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svnrdump.c.load"></a><div class="titlepage"></div><a id="idp19047184" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnrdump load</p></div><div class="refsect1" title="Synopsis"><a id="idp19050112"></a><h2>Synopsis</h2><p><code class="literal">svnrdump load DEST_URL</code></p></div><div class="refsect1" title="Description"><a id="idp19051728"></a><h2>Description</h2><p>Read from standard input revision information
described in a Subversion repository dump stream, and load
that information into the repository located
at <em class="replaceable"><code>DEST_URL</code></em>.</p></div><div class="refsect1" title="Options"><a id="idp19053536"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svnrdump.sw.config_dir"><code class="option">--config-dir</code> <em class="replaceable"><code>DIR</code></em></a>
<a class="xref" href="#svn.ref.svnrdump.sw.config_option"><code class="option">--config-option</code> <em class="replaceable"><code>FILE</code></em>:<em class="replaceable"><code>SECTION</code></em>:<em class="replaceable"><code>OPTION</code></em>=[<em class="replaceable"><code>VALUE</code></em>]</a>
<a class="xref" href="#svn.ref.svnrdump.sw.no_auth_cache"><code class="option">--no-auth-cache</code></a>
<a class="xref" href="#svn.ref.svnrdump.sw.non_interactive"><code class="option">--non-interactive</code></a>
<a class="xref" href="#svn.ref.svnrdump.sw.password"><code class="option">--password</code> <em class="replaceable"><code>PASSWD</code></em></a>
<a class="xref" href="#svn.ref.svnrdump.sw.quiet"><code class="option">--quiet</code> (<code class="option">-q</code>)</a>
<a class="xref" href="#svn.ref.svnrdump.sw.trust_server_cert"><code class="option">--trust-server-cert</code></a>
<a class="xref" href="#svn.ref.svnrdump.sw.username"><code class="option">--username</code> <em class="replaceable"><code>NAME</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp19060800"></a><h2>Examples</h2><p>Dump the contents of a local repository, and
use <span class="command"><strong>svnrdump load</strong></span> to transfer that
revision information into an existing remote
repository:</p><div class="informalexample"><pre class="screen">
$ svnadmin dump -q /var/svn/repos/new-project | \
svnrdump load http://svn.example.com/repos/new-project
* Loaded revision 0
* Loaded revision 1.
* Loaded revision 2.
…
</pre></div><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>To operate properly <span class="command"><strong>svnrdump load</strong></span>
requires that the target repository have revision
property modification enabled via the pre-revprop-change
hook. For details about that hook, see
<a class="xref" href="#svn.ref.reposhooks.pre-revprop-change" title="pre-revprop-change">pre-revprop-change</a>.</p></td></tr></table></div></div></div></div></div><div class="sect1" title="svnserve—Custom Subversion Server"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.ref.svnserve"></a>svnserve—Custom Subversion Server</h2></div></div></div><p><span class="command"><strong>svnserve</strong></span> allows access to Subversion
repositories using Subversion's custom network protocol.</p><p>You can run <span class="command"><strong>svnserve</strong></span> as a standalone
server process (for clients that are using the
<code class="literal">svn://</code> access method); you can have a daemon
such as <span class="command"><strong>inetd</strong></span> or <span class="command"><strong>xinetd</strong></span>
launch it for you on demand (also for
<code class="literal">svn://</code>), or you can have
<span class="command"><strong>sshd</strong></span> launch it on demand for the
<code class="literal">svn+ssh://</code> access method.</p><p>Regardless of the access method, once the client has
selected a repository by transmitting its URL,
<span class="command"><strong>svnserve</strong></span> reads a file named
<code class="filename">conf/svnserve.conf</code> in the repository
directory to determine repository-specific settings such as
what authentication database to use and what authorization
policies to apply. See <a class="xref" href="#svn.serverconfig.svnserve" title="svnserve, a Custom Server">the section called “svnserve, a Custom Server”</a> for details of the
<code class="filename">svnserve.conf</code> file.</p><div class="sect2" title="svnserve Options"><div class="titlepage"><div><div><h3 class="title"><a id="svn.ref.svnserve.sw"></a>svnserve Options</h3></div></div></div><p>Unlike the previous commands we've
described, <span class="command"><strong>svnserve</strong></span> has no
subcommands—it is controlled exclusively by
options.</p><div class="variablelist"><dl><dt><a id="svn.ref.svnserve.sw.cache_fulltexts"></a><span class="term"><code class="option">--cache-fulltexts</code> <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>Toggles support for fulltext file content caching (in
FSFS repositories only).</p></dd><dt><a id="svn.ref.svnserve.sw.cache_txdeltas"></a><span class="term"><code class="option">--cache-txdeltas</code> <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>Toggles support for file content delta caching (in
FSFS repositories only).</p></dd><dt><a id="svn.ref.svnserve.sw.compression"></a><span class="term"><code class="option">--compression</code> <em class="replaceable"><code>LEVEL</code></em></span></dt><dd><p>Specifies the level of compression used for wire
transmissions as an integer beween 0 and 9, inclusive.
A value of <code class="literal">9</code> offers the best
compression, <code class="literal">5</code> is the default value,
and <code class="literal">0</code> disables compression
altogether.</p></dd><dt><a id="svn.ref.svnserve.sw.daemon"></a><span class="term"><code class="option">--daemon</code> (<code class="option">-d</code>)</span></dt><dd><p>Causes <span class="command"><strong>svnserve</strong></span> to run in daemon
mode. <span class="command"><strong>svnserve</strong></span> backgrounds itself
and accepts and serves TCP/IP connections on
the <code class="literal">svn</code> port (3690, by
default).</p></dd><dt><a id="svn.ref.svnserve.sw.foreground"></a><span class="term"><code class="option">--foreground</code></span></dt><dd><p>When used together with <code class="option">-d</code>,
causes <span class="command"><strong>svnserve</strong></span> to stay in the
foreground. This is mainly useful for debugging.</p></dd><dt><a id="svn.ref.svnserve.sw.inetd"></a><span class="term"><code class="option">--inetd</code> (<code class="option">-i</code>)</span></dt><dd><p>Causes <span class="command"><strong>svnserve</strong></span> to use the
<code class="filename">stdin</code> and <code class="filename">stdout</code> file descriptors, as is appropriate for a
daemon running out of <span class="command"><strong>inetd</strong></span>.</p></dd><dt><a id="svn.ref.svnserve.sw.help"></a><span class="term"><code class="option">--help</code> (<code class="option">-h</code>)</span></dt><dd><p>Displays a usage summary and exits.</p></dd><dt><a id="svn.ref.svnserve.sw.listen_host"></a><span class="term"><code class="option">--listen-host</code> <em class="replaceable"><code>HOST</code></em></span></dt><dd><p>Causes <span class="command"><strong>svnserve</strong></span> to listen on the
interface specified by <em class="replaceable"><code>HOST</code></em>,
which may be either a hostname or an IP address.</p></dd><dt><a id="svn.ref.svnserve.sw.listen_once"></a><span class="term"><code class="option">--listen-once</code> (<code class="option">-X</code>)</span></dt><dd><p>Causes <span class="command"><strong>svnserve</strong></span> to accept one
connection on the <code class="literal">svn</code> port, serve it,
and exit. This option is mainly useful for
debugging.</p></dd><dt><a id="svn.ref.svnserve.sw.listen_port"></a><span class="term"><code class="option">--listen-port</code> <em class="replaceable"><code>PORT</code></em></span></dt><dd><p>Causes <span class="command"><strong>svnserve</strong></span> to listen on
<em class="replaceable"><code>PORT</code></em> when run in daemon mode.
(FreeBSD daemons listen only on tcp6 by default—this
option tells them to also listen on tcp4.)</p></dd><dt><a id="svn.ref.svnserve.sw.log_file"></a><span class="term"><code class="option">--log-file</code> <em class="replaceable"><code>FILENAME</code></em></span></dt><dd><p>Instructs <span class="command"><strong>svnserve</strong></span> to create (if
necessary) and use the file located
at <em class="replaceable"><code>FILENAME</code></em> for Subversion
operational log output of the same sort
that <span class="command"><strong>mod_dav_svn</strong></span> generates. See
<a class="xref" href="#svn.serverconfig.operational-logging" title="High-level Logging">the section called “High-level Logging”</a>
for details.</p></dd><dt><a id="svn.ref.svnserve.sw.memory_cache_size"></a><span class="term"><code class="option">--memory-cache-size</code> (<code class="option">-M</code>) <em class="replaceable"><code>ARG</code></em></span></dt><dd><p>Configures the size (in Megabytes) of the extra
in-memory cache used to minimize redundant operations.
The default value is <code class="literal">16</code>. (This cache
is used for FSFS-backed repositories only.)</p></dd><dt><a id="svn.ref.svnserve.sw.pid_file"></a><span class="term"><code class="option">--pid-file</code> <em class="replaceable"><code>FILENAME</code></em></span></dt><dd><p>Causes <span class="command"><strong>svnserve</strong></span> to write its
process ID to <em class="replaceable"><code>FILENAME</code></em>, which
must be writable by the user under
which <span class="command"><strong>svnserve</strong></span> is running.</p></dd><dt><a id="svn.ref.svnserve.sw.prefer_ipv6"></a><span class="term"><code class="option">--prefer-ipv6</code> (<code class="option">-6</code>)</span></dt><dd><p>When resolving the listen hostname, prever an IPv6
answer over an IPv4 one. IPv4 is preferred by
default.</p></dd><dt><a id="svn.ref.svnserve.sw.quiet"></a><span class="term"><code class="option">--quiet</code></span></dt><dd><p>Disables progress notifications. Error output will
still be printed.</p></dd><dt><a id="svn.ref.svnserve.sw.root"></a><span class="term"><code class="option">--root</code> (<code class="option">-r</code>) <em class="replaceable"><code>ROOT</code></em></span></dt><dd><p>Sets the virtual root for repositories served by
<span class="command"><strong>svnserve</strong></span>. The pathname in URLs
provided by the client will be interpreted relative to
this root and will not be allowed to escape this
root.</p></dd><dt><a id="svn.ref.svnserve.sw.threads"></a><span class="term"><code class="option">--threads</code> (<code class="option">-T</code>)</span></dt><dd><p>When running in daemon mode, causes
<span class="command"><strong>svnserve</strong></span> to spawn a thread instead of
a process for each connection (e.g., for when running on
Windows). The <span class="command"><strong>svnserve</strong></span> process still
backgrounds itself at startup time.</p></dd><dt><a id="svn.ref.svnserve.sw.tunnel"></a><span class="term"><code class="option">--tunnel</code> (<code class="option">-t</code>)</span></dt><dd><p>Causes <span class="command"><strong>svnserve</strong></span> to run in tunnel
mode, which is just like the <span class="command"><strong>inetd</strong></span>
mode of operation (both modes serve one connection over
<code class="filename">stdin</code>/<code class="filename">stdout</code>, and then exit), except that the connection
is considered to be preauthenticated with the username
of the current UID. This flag is automatically passed
for you by the client when running over a tunnel agent
such as <span class="command"><strong>ssh</strong></span>. That means there's
rarely any need for <span class="emphasis"><em>you</em></span> to pass
this option to <span class="command"><strong>svnserve</strong></span>. So, if you
find yourself
typing <code class="literal">svnserve --tunnel</code> on the
command line and wondering what to do next, see
<a class="xref" href="#svn.serverconfig.svnserve.sshauth" title="Tunneling over SSH">the section called “Tunneling over SSH”</a>.</p></dd><dt><a id="svn.ref.svnserve.sw.tunnel_user"></a><span class="term"><code class="option">--tunnel-user</code> <em class="replaceable"><code>NAME</code></em></span></dt><dd><p>Used in conjunction with the <code class="option">--tunnel</code>
option, tells <span class="command"><strong>svnserve</strong></span> to assume that
<em class="replaceable"><code>NAME</code></em> is the authenticated
user, rather than the UID of the <span class="command"><strong>svnserve</strong></span>
process. This is useful for users wishing to share a single
system account over SSH, but to maintain separate commit
identities.</p></dd><dt><a id="svn.ref.svnserve.sw.version"></a><span class="term"><code class="option">--version</code></span></dt><dd><p>Displays version information and a list of repository
backend modules available, and then exits.</p></dd></dl></div></div></div><div class="sect1" title="svndumpfilter—Subversion History Filtering"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.ref.svndumpfilter"></a>svndumpfilter—Subversion History Filtering</h2></div></div></div><p><span class="command"><strong>svndumpfilter</strong></span> is a command-line utility
for removing history from a Subversion dump file by either
excluding or including paths beginning with one or more named
prefixes. For details, see
<a class="xref" href="#svn.reposadmin.maint.tk.svndumpfilter" title="svndumpfilter">the section called “svndumpfilter”</a>.</p><div class="sect2" title="svndumpfilter Options"><div class="titlepage"><div><div><h3 class="title"><a id="svn.ref.svndumpfilter.options"></a>svndumpfilter Options</h3></div></div></div><p>Options in <span class="command"><strong>svndumpfilter</strong></span> are global, just
as they are in <span class="command"><strong>svn</strong></span> and
<span class="command"><strong>svnadmin</strong></span>:</p><div class="variablelist"><dl><dt><a id="svn.ref.svndumpfilter.sw.drop_empty_revs"></a><span class="term"><code class="option">--drop-empty-revs</code></span></dt><dd><p>If filtering causes any revision to be empty (i.e.,
causes no change to the repository), removes these
revisions from the final dump file.</p></dd><dt><a id="svn.ref.svndumpfilter.sw.pattern"></a><span class="term"><code class="option">--pattern</code></span></dt><dd><p>Treat the path prefixes provided to the filtering
commands as file glob patterns rather than explicit path
substrings.</p></dd><dt><a id="svn.ref.svndumpfilter.sw.renumber_revs"></a><span class="term"><code class="option">--renumber-revs</code></span></dt><dd><p>Renumbers revisions that remain after
filtering.</p></dd><dt><a id="svn.ref.svndumpfilter.sw.skip_missing_merge_sources"></a><span class="term"><code class="option">--skip-missing-merge-sources</code></span></dt><dd><p>Skips merge sources that have been removed as part of
the filtering. Without this
option, <span class="command"><strong>svndumpfilter</strong></span> will exit with
an error if the merge source for a retained path is
removed by filtering.</p></dd><dt><a id="svn.ref.svndumpfilter.sw.preserve_revprops"></a><span class="term"><code class="option">--preserve-revprops</code></span></dt><dd><p>If all nodes in a revision are removed by filtering
and <code class="option">--drop-empty-revs</code> is not passed,
the default behavior of <span class="command"><strong>svndumpfilter</strong></span>
is to remove all revision properties except for the date
and the log message (which will merely indicate that the
revision is empty). Passing this option will preserve
existing revision properties (which may or may not make
sense since the related content is no longer present in
the dump file).</p></dd><dt><a id="svn.ref.svndumpfilter.sw.targets"></a><span class="term"><code class="option">--targets</code> <em class="replaceable"><code>FILENAME</code></em></span></dt><dd><p>Instructs <span class="command"><strong>svndumpfilter</strong></span> to read
additional path prefixes—one per line—from
the file located at <em class="replaceable"><code>FILENAME</code></em>.
This is especially useful for complex filtering
operations which require more prefixes than the
operating system allows to be specified on a single
command line.</p></dd><dt><a id="svn.ref.svndumpfilter.sw.quiet"></a><span class="term"><code class="option">--quiet</code></span></dt><dd><p>Does not display filtering statistics.</p></dd></dl></div></div><div class="sect2" title="svndumpfilter Subcommands"><div class="titlepage"><div><div><h3 class="title"><a id="svn.ref.svndumpfilter.c"></a>svndumpfilter Subcommands</h3></div></div></div><p>Here are the various subcommands
for the <span class="command"><strong>svndumpfilter</strong></span> program.</p><div class="refentry" title="svndumpfilter exclude"><a id="svn.ref.svndumpfilter.commands.c.exclude"></a><div class="titlepage"></div><a id="idp19199168" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svndumpfilter exclude — Filter out nodes with given prefixes from the dump stream.</p></div><div class="refsect1" title="Synopsis"><a id="idp19202336"></a><h2>Synopsis</h2><p><code class="literal">svndumpfilter exclude PATH_PREFIX...</code></p></div><div class="refsect1" title="Description"><a id="idp19203968"></a><h2>Description</h2><p>This can be used to exclude nodes that begin with one or
more <em class="replaceable"><code>PATH_PREFIX</code></em>es from a
filtered dump file.</p></div><div class="refsect1" title="Options"><a id="idp19205664"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svndumpfilter.sw.drop_empty_revs"><code class="option">--drop-empty-revs</code></a>
<a class="xref" href="#svn.ref.svndumpfilter.sw.pattern"><code class="option">--pattern</code></a>
<a class="xref" href="#svn.ref.svndumpfilter.sw.preserve_revprops"><code class="option">--preserve-revprops</code></a>
<a class="xref" href="#svn.ref.svndumpfilter.sw.quiet"><code class="option">--quiet</code></a>
<a class="xref" href="#svn.ref.svndumpfilter.sw.renumber_revs"><code class="option">--renumber-revs</code></a>
<a class="xref" href="#svn.ref.svndumpfilter.sw.skip_missing_merge_sources"><code class="option">--skip-missing-merge-sources</code></a>
<a class="xref" href="#svn.ref.svndumpfilter.sw.targets"><code class="option">--targets</code> <em class="replaceable"><code>FILENAME</code></em></a>
</pre></div></div><div class="refsect1" title="Examples"><a id="idp19213184"></a><h2>Examples</h2><p>If we have a dump file from a repository with a number
of different picnic-related directories in it, but we want
to keep everything <span class="emphasis"><em>except</em></span>
the <code class="filename">sandwiches</code> part of the
repository, we'll exclude only that path:</p><div class="informalexample"><pre class="screen">
$ svndumpfilter exclude sandwiches < dumpfile > filtered-dumpfile
Excluding prefixes:
'/sandwiches'
Revision 0 committed as 0.
Revision 1 committed as 1.
Revision 2 committed as 2.
Revision 3 committed as 3.
Revision 4 committed as 4.
Dropped 1 node(s):
'/sandwiches'
$
</pre></div><p>Beginning in Subversion 1.7,
<span class="command"><strong>svndumpfilter</strong></span> can optionally treat
the <em class="replaceable"><code>PATH_PREFIX</code></em>s not merely as
explicit substrings, but as file patterns instead. So,
for example, if you wished to filter out paths which ended
with <code class="filename">.OLD</code>, you would do the
following:</p><div class="informalexample"><pre class="screen">
$ svndumpfilter exclude --pattern "*.OLD" < dumpfile > filtered-dumpfile
Excluding prefix patterns:
'/*.OLD'
Revision 0 committed as 0.
Revision 1 committed as 1.
Revision 2 committed as 2.
Revision 3 committed as 3.
Revision 4 committed as 4.
Dropped 3 node(s):
'/condiments/salt.OLD'
'/condiments/pepper.OLD'
'/toppings/cheese.OLD'
$
</pre></div></div></div><div class="refentry" title="svndumpfilter include"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svndumpfilter.commands.c.include"></a><div class="titlepage"></div><a id="idp19223104" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svndumpfilter include — Filter out nodes without given prefixes from
dump stream.</p></div><div class="refsect1" title="Synopsis"><a id="idp19226272"></a><h2>Synopsis</h2><p><code class="literal">svndumpfilter include PATH_PREFIX...</code></p></div><div class="refsect1" title="Description"><a id="idp19227904"></a><h2>Description</h2><p>Can be used to include nodes that begin with one or
more <em class="replaceable"><code>PATH_PREFIX</code></em>es in a
filtered dump file (thus excluding all other paths).</p></div><div class="refsect1" title="Options"><a id="idp19229680"></a><h2>Options</h2><div class="informalexample"><pre class="screen">
<a class="xref" href="#svn.ref.svndumpfilter.sw.drop_empty_revs"><code class="option">--drop-empty-revs</code></a>
<a class="xref" href="#svn.ref.svndumpfilter.sw.pattern"><code class="option">--pattern</code></a>
<a class="xref" href="#svn.ref.svndumpfilter.sw.preserve_revprops"><code class="option">--preserve-revprops</code></a>
<a class="xref" href="#svn.ref.svndumpfilter.sw.quiet"><code class="option">--quiet</code></a>
<a class="xref" href="#svn.ref.svndumpfilter.sw.renumber_revs"><code class="option">--renumber-revs</code></a>
<a class="xref" href="#svn.ref.svndumpfilter.sw.skip_missing_merge_sources"><code class="option">--skip-missing-merge-sources</code></a>
<a class="xref" href="#svn.ref.svndumpfilter.sw.targets"><code class="option">--targets</code> <em class="replaceable"><code>FILENAME</code></em></a>
</pre></div></div><div class="refsect1" title="Example"><a id="idp19236368"></a><h2>Example</h2><p>If we have a dump file from a repository with a number of
different picnic-related directories in it, but want to
keep only the <code class="filename">sandwiches</code> part of the
repository, we'll include only that path:</p><div class="informalexample"><pre class="screen">
$ svndumpfilter include sandwiches < dumpfile > filtered-dumpfile
Including prefixes:
'/sandwiches'
Revision 0 committed as 0.
Revision 1 committed as 1.
Revision 2 committed as 2.
Revision 3 committed as 3.
Revision 4 committed as 4.
Dropped 12 node(s):
'/condiments'
'/condiments/pepper'
'/condiments/pepper.OLD'
'/condiments/salt'
'/condiments/salt.OLD'
'/drinks'
'/snacks'
'/supplies'
'/toppings'
'/toppings/cheese'
'/toppings/cheese.OLD'
'/toppings/lettuce'
$
</pre></div><p>Beginning in Subversion 1.7,
<span class="command"><strong>svndumpfilter</strong></span> can optionally treat
the <em class="replaceable"><code>PATH_PREFIX</code></em>s not merely as
explicit substrings, but as file patterns instead. So,
for example, if you wished to include only paths which ended
with <code class="filename">ks</code>, you would do the
following:</p><div class="informalexample"><pre class="screen">
$ svndumpfilter include --pattern "*ks" < dumpfile > filtered-dumpfile
Including prefix patterns:
'/*ks'
Revision 0 committed as 0.
Revision 1 committed as 1.
Revision 2 committed as 2.
Revision 3 committed as 3.
Revision 4 committed as 4.
Dropped 11 node(s):
'/condiments'
'/condiments/pepper'
'/condiments/pepper.OLD'
'/condiments/salt'
'/condiments/salt.OLD'
'/sandwiches'
'/supplies'
'/toppings'
'/toppings/cheese'
'/toppings/cheese.OLD'
'/toppings/lettuce'
$
</pre></div></div></div><div class="refentry" title="svndumpfilter help"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.svndumpfilter.commands.c.help"></a><div class="titlepage"></div><a id="idp19246496" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svndumpfilter help — Help!</p></div><div class="refsect1" title="Synopsis"><a id="idp19249616"></a><h2>Synopsis</h2><p><code class="literal">svndumpfilter help [SUBCOMMAND...]</code></p></div><div class="refsect1" title="Description"><a id="idp19251200"></a><h2>Description</h2><p>Displays the help message
for <span class="command"><strong>svndumpfilter</strong></span>. Unlike other help
commands documented in this chapter, there is no witty
commentary for this help command. The authors of this
book deeply regret the omission.</p></div><div class="refsect1" title="Options"><a id="idp19253344"></a><h2>Options</h2><p>None</p></div></div></div></div><div class="sect1" title="svnversion—Subversion Working Copy Version Info"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.ref.svnversion"></a>svnversion—Subversion Working Copy Version Info</h2></div></div></div><div class="refentry" title="svnversion"><a id="svn.ref.svnversion.re"></a><div class="titlepage"></div><a id="idp19258352" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>svnversion — Summarize the local revision(s) of a working
copy.</p></div><div class="refsect1" title="Synopsis"><a id="svn.ref.svnversion.re.syn"></a><h2>Synopsis</h2><p><code class="literal">svnversion [OPTIONS] [WC_PATH [TRAIL_URL]]</code></p></div><div class="refsect1" title="Description"><a id="svn.ref.svnversion.re.desc"></a><h2>Description</h2><p><span class="command"><strong>svnversion</strong></span> is a program for
summarizing the revision mixture of a working copy. The
resultant revision number, or revision range, is written to
standard output.</p><p>It's common to use this output in your build process
when defining the version number of your program.</p><p><em class="replaceable"><code>TRAIL_URL</code></em>, if present, is the
trailing portion of the URL used to determine whether
<em class="replaceable"><code>WC_PATH</code></em> itself is switched
(detection of switches within
<em class="replaceable"><code>WC_PATH</code></em> does not rely on
<em class="replaceable"><code>TRAIL_URL</code></em>).</p><p>When <em class="replaceable"><code>WC_PATH</code></em> is not defined,
the current directory will be used as the working copy path.
<em class="replaceable"><code>TRAIL_URL</code></em> cannot be defined if
<em class="replaceable"><code>WC_PATH</code></em> is not explicitly
given.</p></div><div class="refsect1" title="Options"><a id="svn.ref.svnversion.re.sw"></a><h2>Options</h2><p>Like <span class="command"><strong>svnserve</strong></span>,
<span class="command"><strong>svnversion</strong></span> has no subcommands—only
options:</p><div class="variablelist"><dl><dt><a id="svn.ref.svnversion.sw.no_newline"></a><span class="term"><code class="option">--no-newline</code> (<code class="option">-n</code>)</span></dt><dd><p>Omits the usual trailing newline from the output.</p></dd><dt><a id="svn.ref.svnversion.sw.committed"></a><span class="term"><code class="option">--committed</code> (<code class="option">-c</code>)</span></dt><dd><p>Uses the last-changed revisions rather than the
current (i.e., highest locally available) revisions.</p></dd><dt><a id="svn.ref.svnversion.sw.help"></a><span class="term"><code class="option">--help</code> (<code class="option">-h</code>)</span></dt><dd><p>Prints a help summary.</p></dd><dt><a id="svn.ref.svnversion.sw.quiet"></a><span class="term"><code class="option">--quiet</code> (<code class="option">-q</code>)</span></dt><dd><p>Requests that the program print only essential
information while performing an operation.</p></dd><dt><a id="svn.ref.svnversion.sw.version"></a><span class="term"><code class="option">--version</code></span></dt><dd><p>Prints the version of <span class="command"><strong>svnversion</strong></span>
and exit with no error.</p></dd></dl></div></div><div class="refsect1" title="Examples"><a id="svn.ref.svnversion.re.examples"></a><h2>Examples</h2><p>If the working copy is all at the same revision (e.g.,
immediately after an update), then that revision is
printed out:</p><div class="informalexample"><pre class="screen">
$ svnversion
4168
</pre></div><p>You can add <em class="replaceable"><code>TRAIL_URL</code></em> to make
sure the working copy is not switched from what you
expect. Note that the <em class="replaceable"><code>WC_PATH</code></em>
is required in this command:</p><div class="informalexample"><pre class="screen">
$ svnversion . /var/svn/trunk
4168
</pre></div><p>For a mixed-revision working copy, the range of
revisions present is printed:</p><div class="informalexample"><pre class="screen">
$ svnversion
4123:4168
</pre></div><p>If the working copy contains modifications, a trailing
'<code class="literal">M</code>' is added:</p><div class="informalexample"><pre class="screen">
$ svnversion
4168M
</pre></div><p>If the working copy is switched, a trailing
'<code class="literal">S</code>' is added:</p><div class="informalexample"><pre class="screen">
$ svnversion
4168S
</pre></div><p><span class="command"><strong>svnversion</strong></span> will also inform you if
the target working copy is sparsely populated (see
<a class="xref" href="#svn.advanced.sparsedirs" title="Sparse Directories">the section called “Sparse Directories”</a>) by attaching the
'<code class="literal">P</code>' code to its output:</p><div class="informalexample"><pre class="screen">
$ svnversion
4168P
</pre></div><p>Thus, here is a mixed-revision, sparsely populated and
switched working copy containing some local
modifications:</p><div class="informalexample"><pre class="screen">
$ svnversion
4123:4168MSP
</pre></div></div></div></div><div class="sect1" title="mod_dav_svn—Subversion Apache HTTP Server Module"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.ref.mod_dav_svn"></a>mod_dav_svn—Subversion Apache HTTP Server Module</h2></div></div></div><div class="refentry" title="mod_dav_svn Configuration Directives"><a id="svn.ref.mod_dav_svn.conf"></a><div class="titlepage"></div><div class="refnamediv"><h2>Name</h2><p>mod_dav_svn Configuration Directives — Apache configuration
directives for serving Subversion repositories through the
Apache HTTP Server.</p></div><div class="refsect1" title="Description"><a id="svn.ref.mod_dav_svn.conf.desc"></a><h2>Description</h2><p>This section briefly describes each Subversion
Apache configuration directive. For an in-depth
description of configuring Apache with Subversion, see <a class="xref" href="#svn.serverconfig.httpd" title="httpd, the Apache HTTP Server">the section called “httpd, the Apache HTTP Server”</a>.</p></div><div class="refsect1" title="Directives"><a id="svn.ref.mod_dav_svn.conf.directives"></a><h2>Directives</h2><p>These are the <code class="filename">httpd.conf</code> directives
that apply to <span class="command"><strong>mod_dav_svn</strong></span>:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">DAV svn</code></span></dt><dd><p>Must be included in any
<code class="literal">Directory</code> or
<code class="literal">Location</code> block for a Subversion
repository. It tells <span class="command"><strong>httpd</strong></span> to use the Subversion
backend for <code class="literal">mod_dav</code> to handle all
requests.</p></dd><dt><span class="term"><code class="literal">SVNActivitiesDB
<em class="replaceable"><code>directory-path</code></em></code></span></dt><dd><p>Specifies the location in the filesystem where the
activities database should be stored. By default,
<span class="command"><strong>mod_dav_svn</strong></span> creates and uses a directory in the
repository
called <code class="filename">dav/activities.d</code>. The
path specified with this option must be an absolute
path.</p><p>If specified for
an <code class="literal">SVNParentPath</code> area, <span class="command"><strong>mod_dav_svn</strong></span>
appends the basename of the repository to the path
specified here. For example:</p><div class="informalexample"><pre class="screen">
<Location /svn>
DAV svn
# any "/svn/foo" URL will map to a repository in
# /net/svn.nfs/repositories/foo
SVNParentPath "/net/svn.nfs/repositories"
# any "/svn/foo" URL will map to an activities db in
# /var/db/svn/activities/foo
SVNActivitiesDB "/var/db/svn/activities"
</Location>
</pre></div></dd><dt><span class="term"><code class="literal">SVNAdvertiseV2Protocol On|Off</code></span></dt><dd><p>New to Subversion 1.7, this toggles
whether <span class="command"><strong>mod_dav_svn</strong></span> advertises its
support for the new version of its HTTP protocol also
introduced in that version. Most admins will not wish
to use this directive (which is <code class="literal">On</code>
by default), choosing instead to enjoy the performance
benefits that the new protocol offers. However, whena
configuring a server as a write-through proxy to
another server which does not support the new
protocol, set this directive's value
to <code class="literal">Off</code>.</p></dd><dt><span class="term"><code class="literal">SVNAllowBulkUpdates On|Off</code></span></dt><dd><p>Toggles support for all-inclusive
responses to update-style <code class="literal">REPORT</code>
requests. Subversion clients use
<code class="literal">REPORT</code> requests to get information
about directory tree checkouts and updates from
<span class="command"><strong>mod_dav_svn</strong></span>. They can ask the
server to send that information in one of two ways:
with the entirety of the tree's information in one
massive response, or with a
<em class="firstterm">skelta</em> (a skeletal
representation of a tree delta) which contains just
enough information for the client to know what
<span class="emphasis"><em>additional</em></span> data to request from
the server. When this directive is included with a
value of <code class="literal">Off</code>,
<span class="command"><strong>mod_dav_svn</strong></span> will only ever respond
to these <code class="literal">REPORT</code> requests with
skelta responses, regardless of the type of responses
requested by the client.</p><p>Most folks won't need to use this directive at
all. It primarily exists for administrators who
wish—for security or auditing reasons—to
force Subversion clients to fetch individually all the
files and directories needed for updates and
checkouts, thus leaving an audit trail of
<code class="literal">GET</code> and <code class="literal">PROPFIND</code>
requests in Apache's logs. The default value of this
directive is <code class="literal">On</code>.</p></dd><dt><span class="term"><code class="literal">SVNAutoversioning On|Off</code></span></dt><dd><p>When its value is
<code class="literal">On</code>, allows write requests from
WebDAV clients to result in automatic commits. A
generic log message is auto-generated and attached to
each revision. If you enable autoversioning, you'll
likely want to set <code class="literal">ModMimeUsePathInfo
On</code> so that <code class="literal">mod_mime</code> can
set <code class="literal">svn:mime-type</code> to the correct
MIME type automatically (as best as
<code class="literal">mod_mime</code> is able to, of course).
For more information, see <a class="xref" href="#svn.webdav" title="Appendix C. WebDAV and Autoversioning">Appendix C, <em>WebDAV and Autoversioning</em></a>. The default value of this
directive is <code class="literal">Off</code>.</p></dd><dt><span class="term"><code class="literal">SVNCacheFullTexts On|Off</code></span></dt><dd><p>When set to <code class="literal">On</code>, this tells
Subversion to cache content fulltexts if sufficient
in-memory cache is available, which could offer a
significant performance benefit to the server. (See
also the <code class="literal">SVNInMemoryCacheSize</code>
directive.) The default value of this directive
is <code class="literal">Off</code>.</p></dd><dt><span class="term"><code class="literal">SVNCacheTextDeltas On|Off</code></span></dt><dd><p>When set to <code class="literal">On</code>, this tells
Subversion to cache content deltas if sufficient
in-memory cache is available, which could offer a
significant performance benefit to the server. (See
also the <code class="literal">SVNInMemoryCacheSize</code>
directive.) The default value of this directive
is <code class="literal">Off</code>.</p></dd><dt><span class="term"><code class="literal">SVNCompressionLevel
<em class="replaceable"><code>level</code></em></code></span></dt><dd><p>Specifies the compression level used when sending
file content over the network. A value
of <code class="literal">0</code> disables compression
altogether, and <code class="literal">9</code> is the maximum
value. <code class="literal">5</code> is the default
value.</p></dd><dt><span class="term"><code class="literal">SVNIndexXSLT
<em class="replaceable"><code>directory-path</code></em></code></span></dt><dd><p>Specifies the URI of an XSL transformation for
directory indexes. This directive is optional.</p></dd><dt><span class="term"><code class="literal">SVNInMemoryCacheSize
<em class="replaceable"><code>size</code></em></code></span></dt><dd><p>Specifies the maximum size (in kbytes) per process
of Subversion's in-memory object cache. The default
value is <code class="literal">16384</code>; use a value
of <code class="literal">0</code> to deactivate this cache
altogether.</p></dd><dt><span class="term"><code class="literal">SVNListParentPath On|Off</code></span></dt><dd><p>When set to <code class="literal">On</code>, allows
a <code class="literal">GET</code>
of <code class="literal">SVNParentPath</code>, which results in
a listing of all repositories under that path. The
default setting is
<code class="literal">Off</code>.</p></dd><dt><span class="term"><code class="literal">SVNMasterURI
<em class="replaceable"><code>url</code></em></code></span></dt><dd><p>Specifies a URI to the master Subversion
repository (used for a write-through proxy).</p></dd><dt><span class="term"><code class="literal">SVNParentPath
<em class="replaceable"><code>directory-path</code></em></code></span></dt><dd><p>Specifies the location in the filesystem of a
parent directory whose child directories are
Subversion repositories. In a configuration block for
a Subversion repository, either this directive or
<code class="literal">SVNPath</code> must be present, but not
both.</p></dd><dt><span class="term"><code class="literal">SVNPath
<em class="replaceable"><code>directory-path</code></em></code></span></dt><dd><p>Specifies the location in the
filesystem for a Subversion repository's files. In a
configuration block for a Subversion repository,
either this directive or
<code class="literal">SVNParentPath</code> must be present, but
not both.</p></dd><dt><span class="term"><code class="literal">SVNPathAuthz On|Off|short_circuit</code></span></dt><dd><p>Controls path-based authorization by enabling
subrequests (<code class="literal">On</code>), disabling
subrequests (<code class="literal">Off</code>; see <a class="xref" href="#svn.serverconfig.httpd.authz.pathauthzoff" title="Disabling path-based checks">the section called “Disabling path-based checks”</a>),
or querying <span class="command"><strong>mod_authz_svn</strong></span> directly
(<code class="literal">short_circuit</code>). The default
value of this directive is <code class="literal">On</code>.</p></dd><dt><span class="term"><code class="literal">SVNReposName
<em class="replaceable"><code>name</code></em></code></span></dt><dd><p>Specifies the name of a Subversion repository for
use in <code class="literal">HTTP GET</code> responses. This
value will be prepended to the title of all directory
listings (which are served when you navigate to a
Subversion repository with a web browser). This
directive is optional.</p></dd><dt><span class="term"><code class="literal">SVNSpecialURI
<em class="replaceable"><code>component</code></em></code></span></dt><dd><p>Specifies the URI component (namespace) for
special Subversion resources. The default is
<code class="literal">!svn</code>, and most
administrators will never use this directive. Set
this only if there is a pressing need to have a file
named <code class="filename">!svn</code> in your repository.
If you change this on a server already in use, it will
break all of the outstanding working copies, and your
users will hunt you down with pitchforks and flaming
torches.</p></dd></dl></div></div></div></div><div class="sect1" title="mod_authz_svn—Subversion Apache HTTP Authorization Module"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.ref.mod_authz_svn"></a>mod_authz_svn—Subversion Apache HTTP Authorization Module</h2></div></div></div><div class="refentry" title="mod_authz_svn Configuration Directives"><a id="svn.ref.mod_authz_svn.conf"></a><div class="titlepage"></div><div class="refnamediv"><h2>Name</h2><p>mod_authz_svn Configuration Directives — Apache configuration directives for configuring
path-based authorization for Subversion repositories served
through the Apache HTTP Server.</p></div><div class="refsect1" title="Description"><a id="svn.ref.mod_authz_svn.conf.desc"></a><h2>Description</h2><p>This section briefly describes each Apache configuration
directive offered by <span class="command"><strong>mod_authz_svn</strong></span>. For
an in-depth description of using path-based authorization in
Subversion, see <a class="xref" href="#svn.serverconfig.pathbasedauthz" title="Path-Based Authorization">the section called “Path-Based Authorization”</a>.</p></div><div class="refsect1" title="Directives"><a id="svn.ref.mod_authz_svn.conf.directives"></a><h2>Directives</h2><p>These are the <code class="filename">httpd.conf</code> directives
that apply to <span class="command"><strong>mod_authz_svn</strong></span>:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">AuthzForceUsernameCase Upper|Lower</code></span></dt><dd><p>Set to <code class="literal">Upper</code>
or <code class="literal">Lower</code> to perform case conversion
of the specified sort on the authenticated username
before checking it for authorization. While usernames
are compared in a case-sensitive fashion against those
referenced in the authorization rules file, this
directive can at least normalize variably-cased
usernames into something consistent.</p></dd><dt><span class="term"><code class="literal">AuthzSVNAccessFile
<em class="replaceable"><code>file-path</code></em></code></span></dt><dd><p>Consult <em class="replaceable"><code>file-path</code></em> for
access rules describing the permissions for paths in
Subversion repository.</p></dd><dt><span class="term"><code class="literal">AuthzSVNAnonymous On|Off</code></span></dt><dd><p>Set to <code class="literal">Off</code> to disable two
special-case behaviours of this module: interaction
with the <code class="literal">Satisfy Any</code> directive and
enforcement of the authorization policy even when no
<code class="literal">Require</code> directives are present.
The default value of this directive is
<code class="literal">On</code>.</p></dd><dt><span class="term"><code class="literal">AuthzSVNAuthoritative On|Off</code></span></dt><dd><p>Set to <code class="literal">Off</code> to allow access
control to be passed along to lower modules. The
default value of this directive is
<code class="literal">On</code>.</p></dd><dt><span class="term"><code class="literal">AuthzSVNNoAuthWhenAnonymousAllowed On|Off</code></span></dt><dd><p>Set to <code class="literal">On</code> to suppress
authentication and authorization for requests which
anonymous users are allowed to perform. The default
value of this directive is
<code class="literal">On</code>.</p></dd></dl></div></div></div></div><div class="sect1" title="Subversion Properties"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.ref.properties"></a>Subversion Properties</h2></div></div></div><p>Subversion allows users to invent arbitrarily named
versioned properties on files and directories, as well as
unversioned properties on revisions. The only restriction is on
properties whose names begin with <code class="literal">svn:</code> (those
are reserved for Subversion's own use). While these properties
may be set by users to control Subversion's behavior, users may
not invent new <code class="literal">svn:</code> properties.</p><div class="sect2" title="Versioned Properties"><div class="titlepage"><div><div><h3 class="title"><a id="svn.ref.properties.versioned-props"></a>Versioned Properties</h3></div></div></div><p>These are the versioned properties that Subversion reserves
for its own use:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">svn:executable</code></span></dt><dd><p>If present on a file, the client will make the
file executable in Unix-hosted working copies. See
<a class="xref" href="#svn.advanced.props.special.executable" title="File Executability">the section called “File Executability”</a>.</p></dd><dt><span class="term"><code class="literal">svn:mime-type</code></span></dt><dd><p>If present on a file, the value indicates the
file's MIME type. This allows the client to decide
whether line-based contextual merging is safe to
perform during updates, and can also affect how the
file behaves when fetched via a web browser. See
<a class="xref" href="#svn.advanced.props.special.mime-type" title="File Content Type">the section called “File Content Type”</a>.</p></dd><dt><span class="term"><code class="literal">svn:ignore</code></span></dt><dd><p>If present on a directory, the value is a list of
<span class="emphasis"><em>unversioned</em></span> file patterns to be ignored
by <span class="command"><strong>svn status</strong></span> and other
subcommands. See
<a class="xref" href="#svn.advanced.props.special.ignore" title="Ignoring Unversioned Items">the section called “Ignoring Unversioned Items”</a>.</p></dd><dt><span class="term"><code class="literal">svn:keywords</code></span></dt><dd><p>If present on a file, the value tells the client
how to expand particular keywords within the file.
See
<a class="xref" href="#svn.advanced.props.special.keywords" title="Keyword Substitution">the section called “Keyword Substitution”</a>.</p></dd><dt><span class="term"><code class="literal">svn:eol-style</code></span></dt><dd><p>If present on a file, the value tells the client how
to manipulate the file's line-endings in the working
copy and in exported trees. See <a class="xref" href="#svn.advanced.props.special.eol-style" title="End-of-Line Character Sequences">the section called “End-of-Line Character Sequences”</a> and
<a class="xref" href="#svn.ref.svn.c.export" title="svn export">svn export</a> earlier in this chapter.</p></dd><dt><span class="term"><code class="literal">svn:externals</code></span></dt><dd><p>If present on a directory, the value is a multiline
list of other paths and URLs the client should check
out. See
<a class="xref" href="#svn.advanced.externals" title="Externals Definitions">the section called “Externals Definitions”</a>.</p></dd><dt><span class="term"><code class="literal">svn:special</code></span></dt><dd><p>If present on a file, indicates that the file is not
an ordinary file, but a symbolic link or other special
object.<sup>[<a id="idp19456880" href="#ftn.idp19456880" class="footnote">72</a>]</sup></p></dd><dt><span class="term"><code class="literal">svn:needs-lock</code></span></dt><dd><p>If present on a file, tells the client to make the
file read-only in the working copy, as a reminder that
the file should be locked before editing begins. See
<a class="xref" href="#svn.advanced.locking.lock-communication" title="Lock Communication">the section called “Lock Communication”</a>.</p></dd><dt><span class="term"><code class="literal">svn:mergeinfo</code></span></dt><dd><p>Used by Subversion to track merge data. See
<a class="xref" href="#svn.branchmerge.basicmerging.mergeinfo" title="Mergeinfo and Previews">the section called “Mergeinfo and Previews”</a>
for details, but you should never edit this property
unless you <span class="emphasis"><em>really</em></span> know what you're
doing.</p></dd></dl></div></div><div class="sect2" title="Unversioned Properties"><div class="titlepage"><div><div><h3 class="title"><a id="svn.ref.properties.unversioned-props"></a>Unversioned Properties</h3></div></div></div><p>These are the unversioned properties that Subversion reserves
for its own use:</p><div class="variablelist"><dl><dt><span class="term"><code class="literal">svn:author</code></span></dt><dd><p>If present, contains the authenticated username of
the person who created the revision. (If not present,
the revision was committed anonymously.)</p></dd><dt><span class="term"><code class="literal">svn:autoversioned</code></span></dt><dd><p>If present, the revision was created via the
autoversioning feature. See
<a class="xref" href="#svn.webdav.autoversioning" title="Autoversioning">the section called “Autoversioning”</a>.</p></dd><dt><span class="term"><code class="literal">svn:date</code></span></dt><dd><p>Contains the UTC time the revision was created, in
ISO 8601 format. The value comes from the
<span class="emphasis"><em>server</em></span> machine's clock, not the
client's.</p></dd><dt><span class="term"><code class="literal">svn:log</code></span></dt><dd><p>Contains the log message describing the
revision.</p></dd><dt><span class="term"><code class="literal">svn:rdump-lock</code></span></dt><dd><p>Used to temporarily enforce mutually exclusive
access to the repository by <span class="command"><strong>svnrdump
load</strong></span>. This property is generally only
observed when such an operation is active—or when
an <span class="command"><strong>svnrdump</strong></span> command failed to cleanly
disconnect from the repository. (This property is only
relevant when it appears on revision 0.)</p></dd><dt><span class="term"><code class="literal">svn:sync-currently-copying</code></span></dt><dd><p>Contains the revision number from the source
repository which is currently being mirrored to this one
by the <span class="command"><strong>svnsync</strong></span> tool. (This property
is only relevant when it appears on revision 0.)</p></dd><dt><span class="term"><code class="literal">svn:sync-from-uuid</code></span></dt><dd><p>Contains the UUID of the repository of which this
repository has been initialized as a mirror by
the <span class="command"><strong>svnsync</strong></span> tool. (This property is
only relevant when it appears on revision 0.)</p></dd><dt><span class="term"><code class="literal">svn:sync-from-url</code></span></dt><dd><p>Contains the URL of the repository directory of
which this repository has been initialized as a mirror
by the <span class="command"><strong>svnsync</strong></span> tool. (This property
is only relevant when it appears on revision 0.)</p></dd><dt><span class="term"><code class="literal">svn:sync-last-merged-rev</code></span></dt><dd><p>Contains the revision of the source repository which
was most recently and successfully mirrored to this one.
(This property is only relevant when it appears on
revision 0.)</p></dd><dt><span class="term"><code class="literal">svn:sync-lock</code></span></dt><dd><p>Used to temporarily enforce mutually exclusive
access to the repository by <span class="command"><strong>svnsync</strong></span>
mirroring operations. This property is generally only
observed when such an operation is active—or when
an <span class="command"><strong>svnsync</strong></span> command failed to cleanly
disconnect from the repository. (This property is
only relevant when it appears on revision 0.)</p></dd></dl></div></div></div><div class="sect1" title="Repository Hooks"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.ref.reposhooks"></a>Repository Hooks</h2></div></div></div><p>These are the repository hooks that Subversion provides:</p><div class="refentry" title="start-commit"><a id="svn.ref.reposhooks.start-commit"></a><div class="titlepage"></div><a id="idp19499392" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>start-commit — Notification of the beginning of a commit.</p></div><div class="refsect1" title="Description"><a id="idp19502544"></a><h2>Description</h2><p>The start-commit hook is run before the commit
transaction is even created. It is typically used to
decide whether the user has commit privileges at all.</p><p>If the start-commit hook program returns a nonzero exit
value, the commit is stopped before the commit transaction
is even created, and anything printed to <code class="filename">stderr</code> is
marshalled back to the client.</p></div><div class="refsect1" title="Input Parameter(s)"><a id="idp19505248"></a><h2>Input Parameter(s)</h2><p>The command-line arguments passed to the hook program,
in order, are:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p>Repository path</p></li><li class="listitem"><p>Authenticated username attempting the commit</p></li><li class="listitem"><p>Colon-separated list of
capabilities that a client passes to the server,
including <code class="literal">depth</code>,
<code class="literal">mergeinfo</code>, and
<code class="literal">log-revprops</code> (new in Subversion 1.5).</p></li></ol></div></div><div class="refsect1" title="Common uses"><a id="idp19512112"></a><h2>Common uses</h2><p>Access control (e.g., temporarily lock out commits for
some reason).</p><p>A means to allow access only from clients that have
certain capabilities.</p></div></div><div class="refentry" title="pre-commit"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.reposhooks.pre-commit"></a><div class="titlepage"></div><a id="idp19515184" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>pre-commit — Notification just prior to commit completion.</p></div><div class="refsect1" title="Description"><a id="idp19518336"></a><h2>Description</h2><p>The <code class="literal">pre-commit</code> hook is run just
before a commit transaction is promoted to a new revision.
Typically, this hook is used to protect against commits that
are disallowed due to content or location (e.g., your
site might require that all commits to a certain branch
include a ticket number from the bug tracker, or that the
incoming log message is nonempty).</p><p>If the <code class="literal">pre-commit</code> hook program
returns a nonzero exit value, the commit is aborted, the
commit transaction is removed, and anything printed to
<code class="filename">stderr</code> is marshalled back to the client.</p></div><div class="refsect1" title="Input parameter(s)"><a id="idp19522624"></a><h2>Input parameter(s)</h2><p>The command-line arguments passed to the hook program,
in order, are:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p>Repository path</p></li><li class="listitem"><p>Commit transaction name</p></li></ol></div><p>Additionally, Subversion passes any lock tokens provided
by the committing client to the hook script via standard
input. When present, these are formatted as a single line
containing the string <code class="literal">LOCK-TOKENS:</code>,
followed by additional lines—one per lock
token—which contain the lock token information. Each
lock token information line consists of the URI-escaped
repository filesystem path associated with the lock,
followed by the pipe (<code class="literal">|</code>) separator
character, and finally the lock token string.</p></div><div class="refsect1" title="Common uses"><a id="idp19528768"></a><h2>Common uses</h2><p>Change validation and control</p></div></div><div class="refentry" title="post-commit"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.reposhooks.post-commit"></a><div class="titlepage"></div><a id="idp19531280" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>post-commit — Notification of a successful commit.</p></div><div class="refsect1" title="Description"><a id="idp19534368"></a><h2>Description</h2><p>The <code class="literal">post-commit</code> hook is run after the
transaction is committed and a new revision is created. Most
people use this hook to send out descriptive emails about
the commit or to notify some other tool (such as an issue
tracker) that a commit has happened. Some configurations
also use this hook to trigger backup processes.</p><p>If the <code class="literal">post-commit</code> hook returns a
nonzero exit status, the commit <span class="emphasis"><em>will
not</em></span> be aborted since it has already
completed. However, anything that the hook printed
to <code class="filename">stderr</code> will be marshalled back to the
client, making it easier to diagnose hook failures.</p></div><div class="refsect1" title="Input parameter(s)"><a id="idp19539088"></a><h2>Input parameter(s)</h2><p>The command-line arguments passed to the hook program,
in order, are:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p>Repository path</p></li><li class="listitem"><p>Revision number created by the commit</p></li></ol></div></div><div class="refsect1" title="Common uses"><a id="idp19542912"></a><h2>Common uses</h2><p>Commit notification; tool integration</p></div></div><div class="refentry" title="pre-revprop-change"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.reposhooks.pre-revprop-change"></a><div class="titlepage"></div><a id="idp19545488" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>pre-revprop-change — Notification of a revision property change
attempt.</p></div><div class="refsect1" title="Description"><a id="idp19548576"></a><h2>Description</h2><p>The <code class="literal">pre-revprop-change</code> hook is run
immediately prior to the modification of a revision property
when performed outside the scope of a normal commit. Unlike
the other hooks, the default state of this one is to deny
the proposed action. The hook must actually exist and
return a zero exit value before a revision property
modification can happen.</p><p>If the <code class="literal">pre-revprop-change</code> hook
doesn't exist, isn't executable, or returns a nonzero exit
value, no change to the property will be made, and anything
printed to <code class="filename">stderr</code> is marshalled back to the client.</p></div><div class="refsect1" title="Input parameter(s)"><a id="idp19552848"></a><h2>Input parameter(s)</h2><p>The command-line arguments passed to the hook program,
in order, are:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p>Repository path</p></li><li class="listitem"><p>Revision whose property is about to be modified</p></li><li class="listitem"><p>Authenticated username attempting the property change</p></li><li class="listitem"><p>Name of the property changed</p></li><li class="listitem"><p>Change description: <code class="literal">A</code> (added),
<code class="literal">D</code> (deleted), or <code class="literal">M</code>
(modified)</p></li></ol></div><p>Additionally, Subversion passes the intended new value
of the property to the hook program via standard
input.</p></div><div class="refsect1" title="Common uses"><a id="idp19561760"></a><h2>Common uses</h2><p>Access control; change validation and control</p></div></div><div class="refentry" title="post-revprop-change"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.reposhooks.post-revprop-change"></a><div class="titlepage"></div><a id="idp19564400" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>post-revprop-change — Notification of a successful revision property
change.</p></div><div class="refsect1" title="Description"><a id="idp19567488"></a><h2>Description</h2><p>The <code class="literal">post-revprop-change</code> hook is run
immediately after the modification of a revision property
when performed outside the scope of a normal commit. As you can
derive from the description of its counterpart, the
<code class="literal">pre-revprop-change</code> hook, this hook will
not run at all unless
the <code class="literal">pre-revprop-change</code> hook is
implemented. It is typically used to send email
notification of the property change.</p><p>If the <code class="literal">post-revprop-change</code> hook returns a
nonzero exit status, the change <span class="emphasis"><em>will
not</em></span> be aborted since it has already
completed. However, anything that the hook printed
to <code class="filename">stderr</code> will be marshalled back to the
client, making it easier to diagnose hook failures.</p></div><div class="refsect1" title="Input parameter(s)"><a id="idp19573744"></a><h2>Input parameter(s)</h2><p>The command-line arguments passed to the hook program,
in order, are:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p>Repository path</p></li><li class="listitem"><p>Revision whose property was modified</p></li><li class="listitem"><p>Authenticated username of the person making the change</p></li><li class="listitem"><p>Name of the property changed</p></li><li class="listitem"><p>Change description: <code class="literal">A</code> (added),
<code class="literal">D</code> (deleted), or <code class="literal">M</code>
(modified)</p></li></ol></div><p>Additionally, Subversion passes to the hook program, via
standard input, the previous value of the property.</p></div><div class="refsect1" title="Common uses"><a id="idp19582576"></a><h2>Common uses</h2><p>Property change notification</p></div></div><div class="refentry" title="pre-lock"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.reposhooks.pre-lock"></a><div class="titlepage"></div><a id="idp19585200" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>pre-lock — Notification of a path lock attempt.</p></div><div class="refsect1" title="Description"><a id="idp19588288"></a><h2>Description</h2><p>The <code class="literal">pre-lock</code> hook runs whenever
someone attempts to lock a path. It can be used to prevent
locks altogether or to create a more complex policy
specifying exactly which users are allowed to lock
particular paths. If the hook notices a preexisting lock,
it can also decide whether a user is allowed
to <span class="quote">“<span class="quote">steal</span>”</span> the existing lock.</p><p>If the <code class="literal">pre-lock</code> hook program returns
a nonzero exit value, the lock action is aborted and
anything printed to <code class="filename">stderr</code> is
marshalled back to the client.</p><p>The hook program may optionally dictate the lock token
which will be assigned to the lock by printing the desired
lock token to standard output. Because of this,
implementations of this hook should carefully avoid
unexpected output sent to standard output.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>If the <code class="literal">pre-lock</code> script takes
advantage of lock token dictation feature, the
responsibility of generating a <span class="emphasis"><em>unique</em></span>
lock token falls to the script itself. Failure to
generate unique lock tokens may result in
undefined—and very likely,
undesired—behavior.</p></td></tr></table></div></div><div class="refsect1" title="Input parameter(s)"><a id="idp19596352"></a><h2>Input parameter(s)</h2><p>The command-line arguments passed to the hook program,
in order, are:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p>Repository path</p></li><li class="listitem"><p>Versioned path that is to be locked</p></li><li class="listitem"><p>Authenticated username of the person attempting the
lock</p></li><li class="listitem"><p>Comment provided when the lock was created</p></li><li class="listitem"><p><code class="literal">1</code> if the user is attempting to
steal an existing lock; <code class="literal">0</code>
otherwise</p></li></ol></div></div><div class="refsect1" title="Common uses"><a id="idp19603792"></a><h2>Common uses</h2><p>Access control</p></div></div><div class="refentry" title="post-lock"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.reposhooks.post-lock"></a><div class="titlepage"></div><a id="idp19606400" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>post-lock — Notification of a successful path lock.</p></div><div class="refsect1" title="Description"><a id="idp19609488"></a><h2>Description</h2><p>The <code class="literal">post-lock</code> hook runs after one or
more paths have been locked. It is typically used to send
email notification of the lock event.</p><p>If the <code class="literal">post-lock</code> hook returns a
nonzero exit status, the lock <span class="emphasis"><em>will
not</em></span> be aborted since it has already
completed. However, anything that the hook printed
to <code class="filename">stderr</code> will be marshalled back to the
client, making it easier to diagnose hook failures.</p></div><div class="refsect1" title="Input parameter(s)"><a id="idp19614000"></a><h2>Input parameter(s)</h2><p>The command-line arguments passed to the hook program,
in order, are:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p>Repository path</p></li><li class="listitem"><p>Authenticated username of the person who locked the
paths</p></li></ol></div><p>Additionally, the list of paths locked is passed to the
hook program via standard input, one path per line.</p></div><div class="refsect1" title="Common uses"><a id="idp19618384"></a><h2>Common uses</h2><p>Lock notification</p></div></div><div class="refentry" title="pre-unlock"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.reposhooks.pre-unlock"></a><div class="titlepage"></div><a id="idp19620992" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>pre-unlock — Notification of a path unlock attempt.</p></div><div class="refsect1" title="Description"><a id="idp19624080"></a><h2>Description</h2><p>The <code class="literal">pre-unlock</code> hook runs whenever
someone attempts to remove a lock on a file. It can be used
to create policies that specify which users are allowed to
unlock particular paths. It's particularly important for
determining policies about lock breakage. If user A locks a
file, is user B allowed to break the lock? What if the lock
is more than a week old? These sorts of things can be
decided and enforced by the hook.</p><p>If the <code class="literal">pre-unlock</code> hook program
returns a nonzero exit value, the unlock action is aborted
and anything printed to <code class="filename">stderr</code> is marshalled back to the
client.</p></div><div class="refsect1" title="Input parameter(s)"><a id="idp19628464"></a><h2>Input parameter(s)</h2><p>The command-line arguments passed to the hook program,
in order, are:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p>Repository path</p></li><li class="listitem"><p>Versioned path which is to be unlocked</p></li><li class="listitem"><p>Authenticated username of the person attempting the
unlock</p></li><li class="listitem"><p>Lock token associated with the lock which is to be
removed</p></li><li class="listitem"><p><code class="literal">1</code> if the user is attempting to
break the lock; <code class="literal">0</code> otherwise</p></li></ol></div></div><div class="refsect1" title="Common uses"><a id="idp19636000"></a><h2>Common uses</h2><p>Access control</p></div></div><div class="refentry" title="post-unlock"><div class="refentry.separator"><hr></hr></div><a id="svn.ref.reposhooks.post-unlock"></a><div class="titlepage"></div><a id="idp19638560" class="indexterm"></a><div class="refnamediv"><h2>Name</h2><p>post-unlock — Notification of a successful path unlock.</p></div><div class="refsect1" title="Description"><a id="idp19641712"></a><h2>Description</h2><p>The <code class="literal">post-unlock</code> hook runs after one
or more paths have been unlocked. It is typically used to
send email notification of the unlock event.</p><p>If the <code class="literal">post-unlock</code> hook returns a
nonzero exit status, the unlock <span class="emphasis"><em>will
not</em></span> be aborted since it has already
completed. However, anything that the hook printed
to <code class="filename">stderr</code> will be marshalled back to the
client, making it easier to diagnose hook failures.</p></div><div class="refsect1" title="Input parameter(s)"><a id="idp19646224"></a><h2>Input parameter(s)</h2><p>The command-line arguments passed to the hook program,
in order, are:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p>Repository path</p></li><li class="listitem"><p>Authenticated username of the person who unlocked the
paths</p></li></ol></div><p>Additionally, the list of paths unlocked is passed to the
hook program via standard input, one path per line.</p></div><div class="refsect1" title="Common uses"><a id="idp19650608"></a><h2>Common uses</h2><p>Unlock notification</p></div></div></div><div class="footnotes"><br></br><hr width="100" align="left"></hr><div class="footnote"><p><sup>[<a id="ftn.idp18133232" href="#idp18133232" class="para">71</a>] </sup>Remember, <span class="command"><strong>svnadmin</strong></span>
works only with local <span class="emphasis"><em>paths</em></span>, not
<span class="emphasis"><em>URLs</em></span>.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp19456880" href="#idp19456880" class="para">72</a>] </sup>As of this writing, symbolic
links are indeed the only <span class="quote">“<span class="quote">special</span>”</span>
objects. But there might be more in future releases of
Subversion.</p></div></div></div><div class="appendix" title="Appendix A. Subversion Quick-Start Guide"><div class="titlepage"><div><div><h2 class="title"><a id="svn.intro"></a>Appendix A. Subversion Quick-Start Guide</h2></div></div></div><div class="toc"><p><strong>Table of Contents</strong></p><dl><dt><span class="sect1"><a href="#svn.intro.install">Installing Subversion</a></span></dt><dt><span class="sect1"><a href="#svn.intro.quickstart">High-Speed Tutorial</a></span></dt></dl></div><p>If you're eager to get Subversion up and running (and you
enjoy learning by experimentation), this appendix will show you how
to create a repository, import code, and then check it back out
again as a working copy. Along the way, we give links to the
relevant chapters of this book.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>If you're new to the entire concept of version control or to
the <span class="quote">“<span class="quote">copy-modify-merge</span>”</span> model used by both CVS and
Subversion, you should read <a class="xref" href="#svn.basic" title="Chapter 1. Fundamental Concepts">Chapter 1, <em>Fundamental Concepts</em></a>
before going any further.</p></td></tr></table></div><div class="sect1" title="Installing Subversion"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.intro.install"></a>Installing Subversion</h2></div></div></div><p>Subversion is built on a portability layer called
APR—the Apache Portable Runtime library. The APR library
provides all the interfaces that Subversion needs to function on
different operating systems: disk access, network access, memory
management, and so on. While Subversion is able to use Apache
HTTP Server (or, <span class="command"><strong>httpd</strong></span>) as one of its network
server programs, its dependence on APR <span class="emphasis"><em>does
not</em></span> mean that <span class="command"><strong>httpd</strong></span> is a required
component. APR is a standalone library usable by any
application. It does mean, however, that Subversion clients and
servers run on any operating system
that <span class="command"><strong>httpd</strong></span> runs on: Windows, Linux, all
flavors of BSD, Mac OS X, NetWare, and others.</p><p>The easiest way to get Subversion is to download a binary
package built for your operating system. Subversion's web site
(<a class="ulink" href="http://subversion.apache.org" target="_top">http://subversion.apache.org</a>) often has these
packages available for download, posted by volunteers. The site
usually contains graphical installer packages for users of
Microsoft operating systems. If you run a Unix-like operating
system, you can use your system's native package distribution
system (RPMs, DEBs, the ports tree, etc.) to get
Subversion.</p><p>Alternatively, you can build Subversion directly from source
code, though it's not always an easy task. (If you're not
experienced at building open source software packages, you're
probably better off downloading a binary distribution instead!)
From the Subversion web site, download the latest source code
release. After unpacking it, follow the instructions in
the <code class="filename">INSTALL</code> file to build it.</p><p>If you're one of those folks that likes to use bleeding-edge
software, you can also get the Subversion source code from the
Subversion repository in which it lives. Obviously, you'll need
to already have a Subversion client on hand to do this. But
once you do, you can check out a working copy from
<a class="ulink" href="http://svn.apache.org/repos/asf/subversion" target="_top">http://svn.apache.org/repos/asf/subversion</a><sup>[<a id="idp22928704" href="#ftn.idp22928704" class="footnote">73</a>]</sup>:</p><div class="informalexample"><pre class="screen">
$ svn checkout http://svn.apache.org/repos/asf/subversion/trunk subversion
A subversion/HACKING
A subversion/INSTALL
A subversion/README
A subversion/autogen.sh
A subversion/build.conf
…
</pre></div><p>The preceding command will create a working copy of the
latest (unreleased) Subversion source code into a subdirectory
named <code class="filename">subversion</code> in your current working
directory. You can adjust that last argument as you see fit.
Regardless of what you call the new working copy directory,
though, after this operation completes, you will now have the
Subversion source code. Of course, you will still need to fetch
a few helper libraries (apr, apr-util, etc.)—see the
<code class="filename">INSTALL</code> file in the top level of the
working copy for details.</p></div><div class="sect1" title="High-Speed Tutorial"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.intro.quickstart"></a>High-Speed Tutorial</h2></div></div></div><div class="blockquote"><blockquote class="blockquote"><p><span class="quote">“<span class="quote">Please make sure your seat backs are in their full,
upright position and that your tray tables are stored. Flight
attendants, prepare for take-off….</span>”</span></p></blockquote></div><p>What follows is a quick tutorial that walks you through some
basic Subversion configuration and operation. When you finish
it, you should have a general understanding of Subversion's
typical usage.</p><div class="note" title="Note" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Note"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="images/note.png"></img></td><th align="left">Note</th></tr><tr><td align="left" valign="top"><p>The examples used in this appendix assume that you have
<span class="command"><strong>svn</strong></span>, the Subversion command-line client,
and <span class="command"><strong>svnadmin</strong></span>, the administrative tool,
ready to go on a Unix-like operating system. (This tutorial
also works at the Windows command-line prompt, assuming you
make some obvious tweaks.) We also assume you are using
Subversion 1.2 or later (run <strong class="userinput"><code>svn --version</code></strong>
to check).</p></td></tr></table></div><p>Subversion stores all versioned data in a central
repository. To begin, create a new repository:</p><div class="informalexample"><pre class="screen">
$ cd /var/svn
$ svnadmin create repos
$ ls repos
conf/ dav/ db/ format hooks/ locks/ README.txt
$
</pre></div><p>This command creates a Subversion repository in the directory
<code class="filename">/var/svn/repos</code>, creating
the <code class="filename">repos</code> directory itself if it doesn't
already exist. This directory contains (among other things) a
collection of database files. You won't see your versioned
files if you peek inside. For more information about repository
creation and maintenance, see
<a class="xref" href="#svn.reposadmin" title="Chapter 5. Repository Administration">Chapter 5, <em>Repository Administration</em></a>.</p><p>Subversion has no concept of a <span class="quote">“<span class="quote">project.</span>”</span> The
repository is just a virtual versioned filesystem, a large tree
that can hold anything you wish. Some administrators prefer to
store only one project in a repository, and others prefer to
store multiple projects in a repository by placing them into
separate directories. We discuss the merits of each approach
in <a class="xref" href="#svn.reposadmin.projects.chooselayout" title="Planning Your Repository Organization">the section called “Planning Your Repository Organization”</a>.
Either way, the repository manages only files and directories,
so it's up to humans to interpret particular directories as
<span class="quote">“<span class="quote">projects.</span>”</span> So while you might see references to
projects throughout this book, keep in mind that we're only ever
talking about some directory (or collection of directories) in
the repository.</p><p>In this example, we assume you already have some sort
of project (a collection of files and directories) that you wish
to import into your newly created Subversion repository. Begin
by organizing your data into a single directory called
<code class="filename">myproject</code> (or whatever you wish). For
reasons explained in <a class="xref" href="#svn.branchmerge" title="Chapter 4. Branching and Merging">Chapter 4, <em>Branching and Merging</em></a>, your
project's tree structure should contain three top-level
directories named <code class="filename">branches</code>,
<code class="filename">tags</code>, and <code class="filename">trunk</code>. The
<code class="filename">trunk</code> directory should contain all of your
data, and the <code class="filename">branches</code> and
<code class="filename">tags</code> directories should be empty:</p><div class="informalexample"><div class="literallayout"><p><br></br>
/tmp/<br></br>
myproject/<br></br>
branches/<br></br>
tags/<br></br>
trunk/<br></br>
foo.c<br></br>
bar.c<br></br>
Makefile<br></br>
…<br></br>
</p></div></div><p>The <code class="filename">branches</code>, <code class="filename">tags</code>,
and <code class="filename">trunk</code> subdirectories aren't actually
required by Subversion. They're merely a popular convention
that you'll most likely want to use later on.</p><p>Once you have your tree of data ready to go, import it into
the repository with the <span class="command"><strong>svn import</strong></span> command
(see <a class="xref" href="#svn.tour.importing" title="Getting Data into Your Repository">the section called “Getting Data into Your Repository”</a>):</p><div class="informalexample"><pre class="screen">
$ svn import /tmp/myproject file:///var/svn/repos/myproject \
-m "initial import"
Adding /tmp/myproject/branches
Adding /tmp/myproject/tags
Adding /tmp/myproject/trunk
Adding /tmp/myproject/trunk/foo.c
Adding /tmp/myproject/trunk/bar.c
Adding /tmp/myproject/trunk/Makefile
…
Committed revision 1.
$
</pre></div><p>Now the repository contains this tree of data. As mentioned
earlier, you won't see your files by directly peeking into the
repository; they're all stored within a database. But the
repository's imaginary filesystem now contains a top-level
directory named <code class="filename">myproject</code>, which in turn
contains your data.</p><p>Note that the original <code class="filename">/tmp/myproject</code>
directory is unchanged; Subversion is unaware of it. (In fact,
you can even delete that directory if you wish.) To
start manipulating repository data, you need to create a new
<span class="quote">“<span class="quote">working copy</span>”</span> of the data, a sort of private
workspace. Ask Subversion to <span class="quote">“<span class="quote">check out</span>”</span> a working
copy of the <code class="filename">myproject/trunk</code> directory in
the repository:</p><div class="informalexample"><pre class="screen">
$ svn checkout file:///var/svn/repos/myproject/trunk myproject
A myproject/foo.c
A myproject/bar.c
A myproject/Makefile
…
Checked out revision 1.
$
</pre></div><p>Now you have a personal copy of part of the repository in a
new directory named <code class="filename">myproject</code>. You can edit
the files in your working copy and then commit those changes
back into the repository.</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>Enter your working copy and edit a file's
contents.</p></li><li class="listitem"><p>Run <strong class="userinput"><code>svn diff</code></strong> to see unified diff
output of your changes.</p></li><li class="listitem"><p>Run <strong class="userinput"><code>svn commit</code></strong> to commit the new
version of your file to the repository.</p></li><li class="listitem"><p>Run <strong class="userinput"><code>svn update</code></strong> to bring your working
copy <span class="quote">“<span class="quote">up to date</span>”</span> with the repository.</p></li></ul></div><p>For a full tour of all the things you can do with your
working copy, read <a class="xref" href="#svn.tour" title="Chapter 2. Basic Usage">Chapter 2, <em>Basic Usage</em></a>.</p><p>At this point, you have the option of making your repository
available to others over a network. See <a class="xref" href="#svn.serverconfig" title="Chapter 6. Server Configuration">Chapter 6, <em>Server Configuration</em></a> to learn about the different sorts of
server processes available and how to configure them.</p></div><div class="footnotes"><br></br><hr width="100" align="left"></hr><div class="footnote"><p><sup>[<a id="ftn.idp22928704" href="#idp22928704" class="para">73</a>] </sup>Note that the URL checked out in the example
ends not with <code class="literal">subversion</code>, but with a
subdirectory thereof called <code class="literal">trunk</code>. See our
discussion of Subversion's branching and tagging model for the
reasoning behind this.</p></div></div></div><div class="appendix" title="Appendix B. Subversion for CVS Users"><div class="titlepage"><div><div><h2 class="title"><a id="svn.forcvs"></a>Appendix B. Subversion for CVS Users</h2></div></div></div><div class="toc"><p><strong>Table of Contents</strong></p><dl><dt><span class="sect1"><a href="#svn.forcvs.revnums">Revision Numbers Are Different Now</a></span></dt><dt><span class="sect1"><a href="#svn.forcvs.directories">Directory Versions</a></span></dt><dt><span class="sect1"><a href="#svn.forcvs.disconnected">More Disconnected Operations</a></span></dt><dt><span class="sect1"><a href="#svn.forcvs.status-vs-update">Distinction Between Status and Update</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.forcvs.status-vs-update.status">Status</a></span></dt><dt><span class="sect2"><a href="#svn.forcvs.status-vs-update.update">Update</a></span></dt></dl></dd><dt><span class="sect1"><a href="#svn.forcvs.branches-and-tags">Branches and Tags</a></span></dt><dt><span class="sect1"><a href="#svn.forcvs.properties">Metadata Properties</a></span></dt><dt><span class="sect1"><a href="#svn.forcvs.conflicts">Conflict Resolution</a></span></dt><dt><span class="sect1"><a href="#svn.forcvs.binary-and-trans">Binary Files and Translation</a></span></dt><dt><span class="sect1"><a href="#svn.forcvs.modules">Versioned Modules</a></span></dt><dt><span class="sect1"><a href="#svn.forcvs.auth">Authentication</a></span></dt><dt><span class="sect1"><a href="#svn.forcvs.convert">Converting a Repository from CVS to Subversion</a></span></dt></dl></div><p>This appendix is a guide for CVS users new to Subversion.
It's essentially a list of differences between the two systems
as <span class="quote">“<span class="quote">viewed from 10,000 feet.</span>”</span> For each section, we
provide references to relevant chapters when
possible.</p><p>Although the goal of Subversion is to take over the current
and future CVS user base, some new features and design changes
were required to fix certain <span class="quote">“<span class="quote">broken</span>”</span> behaviors
that CVS had. This means that, as a CVS user, you may need to
break habits—ones that you forgot were odd to begin
with.</p><div class="sect1" title="Revision Numbers Are Different Now"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.forcvs.revnums"></a>Revision Numbers Are Different Now</h2></div></div></div><p>In CVS, revision numbers are per file. This is because CVS
stores its data in RCS files; each file has a corresponding RCS
file in the repository, and the repository is roughly laid out
according to the structure of your project tree.</p><p>In Subversion, the repository looks like a single
filesystem. Each commit results in an entirely new filesystem
tree; in essence, the repository is an array of trees. Each of
these trees is labeled with a single revision number. When
someone talks about <span class="quote">“<span class="quote">revision 54</span>”</span>, he's talking
about a particular tree (and indirectly, the way the filesystem
looked after the 54th commit).</p><p>Technically, it's not valid to talk about <span class="quote">“<span class="quote">revision 5
of <code class="filename">foo.c</code>.</span>”</span> Instead, one would say
<span class="quote">“<span class="quote"><code class="filename">foo.c</code> as it appears in revision
5.</span>”</span> Also, be careful when making assumptions about the
evolution of a file. In CVS, revisions 5 and 6 of
<code class="filename">foo.c</code> are always different. In Subversion,
it's most likely that <code class="filename">foo.c</code> did
<span class="emphasis"><em>not</em></span> change between revisions 5 and
6.</p><p>Similarly, in CVS, a tag or branch is an annotation on the
file or on the version information for that individual file,
whereas in Subversion, a tag or branch is a copy of an entire
tree (by convention, into the <code class="filename">/branches</code>
or <code class="filename">/tags</code> directories that appear at the top
level of the repository, beside <code class="filename">/trunk</code>). In
the repository as a whole, many versions of each file may be
visible: the latest version on each branch, every tagged
version, and of course the latest version on the trunk
itself. So, to refine the terms even further, one would often
say <span class="quote">“<span class="quote"><code class="filename">foo.c</code> as it appears in
<code class="filename">/branches/REL1</code> in revision
5.</span>”</span></p><p>For more details on this topic, see <a class="xref" href="#svn.basic.in-action.revs" title="Revisions">the section called “Revisions”</a>.</p></div><div class="sect1" title="Directory Versions"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.forcvs.directories"></a>Directory Versions</h2></div></div></div><p>Subversion tracks tree structures, not just file contents.
It's one of the biggest reasons Subversion was written to
replace CVS.</p><p>Here's what this means to you, as a former CVS user:</p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>The <span class="command"><strong>svn add</strong></span> and <span class="command"><strong>svn
delete</strong></span> commands work on directories now, just as
they work on files. So do <span class="command"><strong>svn copy</strong></span> and
<span class="command"><strong>svn move</strong></span>. However, these commands do
<span class="emphasis"><em>not</em></span> cause any kind of immediate change
in the repository. Instead, the working items are simply
<span class="quote">“<span class="quote">scheduled</span>”</span> for addition or deletion. No
repository changes happen until you run <strong class="userinput"><code>svn
commit</code></strong>.</p></li><li class="listitem"><p>Directories aren't dumb containers anymore; they have
revision numbers like files. (Or more properly, it's
correct to talk about <span class="quote">“<span class="quote">directory
<code class="filename">foo/</code> in revision 5.</span>”</span>)</p></li></ul></div><p>Let's talk more about that last point. Directory versioning
is a hard problem; because we want to allow mixed-revision
working copies, there are some limitations on how far we can
abuse this model.</p><p>From a theoretical point of view, we define <span class="quote">“<span class="quote">revision
5 of directory <code class="filename">foo</code></span>”</span> to mean a
specific collection of directory entries and properties. Now
suppose we start adding and removing files from
<code class="filename">foo</code>, and then commit. It would be a lie
to say that we still have revision 5 of
<code class="filename">foo</code>. However, if we bumped
<code class="filename">foo</code>'s revision number after the commit,
that would be a lie too; there may be other changes to
<code class="filename">foo</code> we haven't yet received, because we
haven't updated yet.</p><p>Subversion deals with this problem by quietly tracking
committed adds and deletes in the <code class="filename">.svn</code>
area. When you eventually run <strong class="userinput"><code>svn update</code></strong>,
all accounts are settled with the repository, and the
directory's new revision number is set correctly.
<span class="emphasis"><em>Therefore, only after an update is it truly safe to
say that you have a <span class="quote">“<span class="quote">perfect</span>”</span> revision of a
directory.</em></span> Most of the time, your working copy will
contain <span class="quote">“<span class="quote">imperfect</span>”</span> directory revisions.</p><p>Similarly, a problem arises if you attempt to commit
property changes on a directory. Normally, the commit would
bump the working directory's local revision number. But again,
that would be a lie, as there may be adds or deletes that
the directory doesn't yet have, because no update has happened.
<span class="emphasis"><em>Therefore, you are not allowed to commit
property changes on a directory unless the directory is
up to date.</em></span></p><p>For more discussion about the limitations of directory
versioning, see <a class="xref" href="#svn.basic.in-action.mixedrevs" title="Mixed-revision working copies">the section called “Mixed-revision working copies”</a>.</p></div><div class="sect1" title="More Disconnected Operations"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.forcvs.disconnected"></a>More Disconnected Operations</h2></div></div></div><p>In recent years, disk space has become outrageously cheap
and abundant, but network bandwidth has not. Therefore, the
Subversion working copy has been optimized around the scarcer
resource.</p><p>The <code class="filename">.svn</code> administrative directory
serves the same purpose as the <code class="filename">CVS</code>
directory, except that it also stores read-only,
<span class="quote">“<span class="quote">pristine</span>”</span> copies of your files. This allows you
to do many things offline:</p><div class="variablelist"><dl><dt><span class="term"><span class="command"><strong>svn status</strong></span></span></dt><dd><p>Shows you any local changes you've made (see <a class="xref" href="#svn.tour.cycle.examine.status" title="See an overview of your changes">the section called “See an overview of your changes”</a>)</p></dd><dt><span class="term"><span class="command"><strong>svn diff</strong></span></span></dt><dd><p>Shows you the details of your changes (see <a class="xref" href="#svn.tour.cycle.examine.diff" title="Examine the details of your local modifications">the section called “Examine the details of your local modifications”</a>)</p></dd><dt><span class="term"><span class="command"><strong>svn revert</strong></span></span></dt><dd><p>Removes your local changes (see <a class="xref" href="#svn.tour.cycle.revert" title="Fix Your Mistakes">the section called “Fix Your Mistakes”</a>)</p></dd></dl></div><p>Also, the cached pristine files allow the Subversion client
to send differences when committing, which CVS cannot do.</p><p>The last subcommand in the list—<span class="command"><strong>svn
revert</strong></span>—is new. It will not only remove local
changes, but also unschedule operations such as adds and
deletes. Although deleting the file and then running <strong class="userinput"><code>svn
update</code></strong> will still work, doing so distorts the true
purpose of updating. And, while we're on this subject…</p></div><div class="sect1" title="Distinction Between Status and Update"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.forcvs.status-vs-update"></a>Distinction Between Status and Update</h2></div></div></div><p>Subversion attempts to erase a lot of the confusion
between the <span class="command"><strong>cvs status</strong></span> and
<span class="command"><strong>cvs update</strong></span> commands.</p><p>The <span class="command"><strong>cvs status</strong></span> command has two purposes:
first, to show the user any local modifications in the working
copy, and second, to show the user which files are out of date.
Unfortunately, because of CVS's hard-to-read status output, many
CVS users don't take advantage of this command at all. Instead,
they've developed a habit of running <strong class="userinput"><code>cvs
update</code></strong> or <strong class="userinput"><code>cvs -n update</code></strong> to quickly
see their changes. If users forget to use
the <code class="option">-n</code> option, this has the side effect of
merging repository changes they may not be ready to deal
with.</p><p>Subversion removes this muddle by making the output of
<span class="command"><strong>svn status</strong></span> easy to read for both humans and
parsers. Also, <span class="command"><strong>svn update</strong></span> prints only
information about files that are updated,
<span class="emphasis"><em>not</em></span> local modifications.</p><div class="sect2" title="Status"><div class="titlepage"><div><div><h3 class="title"><a id="svn.forcvs.status-vs-update.status"></a>Status</h3></div></div></div><p><span class="command"><strong>svn status</strong></span> prints all files that have
local modifications. By default, the repository is not
contacted. While this subcommand accepts a fair number of
options, the following are the most commonly used ones:</p><div class="variablelist"><dl><dt><span class="term"><code class="option">-u</code></span></dt><dd><p>Contact the repository to determine, and then display,
out-of-dateness information.</p></dd><dt><span class="term"><code class="option">-v</code></span></dt><dd><p>Show <span class="emphasis"><em>all</em></span> entries under
version control.</p></dd><dt><span class="term"><code class="option">-N</code></span></dt><dd><p>Run nonrecursively (do not descend into
subdirectories).</p></dd></dl></div><p>The <span class="command"><strong>svn status</strong></span> command has two output
formats. In the default <span class="quote">“<span class="quote">short</span>”</span> format, local
modifications look like this:</p><div class="informalexample"><pre class="screen">
$ svn status
M foo.c
M bar/baz.c
</pre></div><p>If you specify the <code class="option">--show-updates</code>
(<code class="option">-u</code>) option, a longer output format is
used:</p><div class="informalexample"><pre class="screen">
$ svn status -u
M 1047 foo.c
* 1045 faces.html
* bloo.png
M 1050 bar/baz.c
Status against revision: 1066
</pre></div><p>In this case, two new columns appear. The second column
contains an asterisk if the file or directory is out of date.
The third column shows the working copy's revision number of the
item. In the previous example, the asterisk indicates that
<code class="filename">faces.html</code> would be patched if we updated,
and that <code class="filename">bloo.png</code> is a newly added file in
the repository. (The absence of any revision number next to
<code class="filename">bloo.png</code> means that it doesn't yet exist in
the working copy.)</p><p>For a more detailed discussion of <span class="command"><strong>svn
status</strong></span>, including an explanation of the status codes
shown in the previous example, see
<a class="xref" href="#svn.tour.cycle.examine.status" title="See an overview of your changes">the section called “See an overview of your changes”</a>.</p></div><div class="sect2" title="Update"><div class="titlepage"><div><div><h3 class="title"><a id="svn.forcvs.status-vs-update.update"></a>Update</h3></div></div></div><p><span class="command"><strong>svn update</strong></span> updates your working copy,
and prints only information about files that it updates.</p><p>Subversion has combined CVS's <code class="literal">P</code> and
<code class="literal">U</code> codes into just <code class="literal">U</code>. When
a merge or conflict occurs, Subversion simply prints
<code class="literal">G</code> or <code class="literal">C</code>, rather than a
whole sentence about it.</p><p>For a more detailed discussion of <span class="command"><strong>svn
update</strong></span>, see <a class="xref" href="#svn.tour.cycle.update" title="Update Your Working Copy">the section called “Update Your Working Copy”</a>.</p></div></div><div class="sect1" title="Branches and Tags"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.forcvs.branches-and-tags"></a>Branches and Tags</h2></div></div></div><p>Subversion doesn't distinguish between filesystem space and
<span class="quote">“<span class="quote">branch</span>”</span> space; branches and tags are ordinary
directories within the filesystem. This is probably the single
biggest mental hurdle that a CVS user will need to cross. Read
all about it in <a class="xref" href="#svn.branchmerge" title="Chapter 4. Branching and Merging">Chapter 4, <em>Branching and Merging</em></a>.</p><div class="warning" title="Warning" style="margin-left: 0.5in; margin-right: 0.5in;"><table border="0" summary="Warning"><tr><td rowspan="2" align="center" valign="top" width="25"><img alt="[Warning]" src="images/warning.png"></img></td><th align="left">Warning</th></tr><tr><td align="left" valign="top"><p>Since Subversion treats branches and tags as ordinary
directories, your project's various lines of development
probably live in subdirectories of the main project directory.
So remember to check out using the URL of the subdirectory
that contains the particular line of development you want, not
the project's root URL. If you make the mistake of checking
out the root of the project, you may very well wind up with a
working copy that contains a complete copy of your project's
content for each and every one of its branches and tags.
<sup>[<a id="idp23145216" href="#ftn.idp23145216" class="footnote">74</a>]</sup></p></td></tr></table></div></div><div class="sect1" title="Metadata Properties"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.forcvs.properties"></a>Metadata Properties</h2></div></div></div><p>A new feature of Subversion is that you can attach arbitrary
metadata (or <span class="quote">“<span class="quote">properties</span>”</span>) to files and
directories. Properties are arbitrary name/value pairs
associated with files and directories in your working
copy.</p><p>To set or get a property name, use the <span class="command"><strong>svn
propset</strong></span> and <span class="command"><strong>svn propget</strong></span>
subcommands. To list all properties on an object, use
<span class="command"><strong>svn proplist</strong></span>.</p><p>For more information, see <a class="xref" href="#svn.advanced.props" title="Properties">the section called “Properties”</a>.</p></div><div class="sect1" title="Conflict Resolution"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.forcvs.conflicts"></a>Conflict Resolution</h2></div></div></div><p>CVS marks conflicts with inline <span class="quote">“<span class="quote">conflict
markers,</span>”</span> and then prints a <code class="literal">C</code> during
an update or merge operation. Historically, this has caused
problems, because CVS isn't doing enough. Many users forget
about (or don't see) the <code class="literal">C</code> after it whizzes
by on their terminal. They often forget that the conflict
markers are even present, and then accidentally commit files
containing those conflict markers.</p><p>Subversion solves this problem in a pair of ways. First,
when a conflict occurs in a file, Subversion records the fact
that the file is in a state of conflict, and won't allow you to
commit changes to that file until you explicitly resolve the
conflict. Second, Subversion provides interactive conflict
resolution, which allows you to resolve conflicts as they happen
instead of having to go back and do so after the update or merge
operation completes. See <a class="xref" href="#svn.tour.cycle.resolve" title="Resolve Any Conflicts">the section called “Resolve Any Conflicts”</a> for more about conflict resolution in Subversion.</p></div><div class="sect1" title="Binary Files and Translation"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.forcvs.binary-and-trans"></a>Binary Files and Translation</h2></div></div></div><p>In the most general sense, Subversion handles binary files
more gracefully than CVS does. Because CVS uses RCS, it can
only store successive full copies of a changing binary file.
Subversion, however, expresses differences between files using a
binary differencing algorithm, regardless of whether they
contain textual or binary data. That means all files are
stored differentially (compressed) in the repository.</p><p>CVS users have to mark binary files with
<code class="option">-kb</code> flags to prevent data from being
garbled (due to keyword expansion and line-ending translations).
They sometimes forget to do this.</p><p>Subversion takes the more paranoid route. First, it never
performs any kind of keyword or line-ending translation unless
you explicitly ask it to do so (see
<a class="xref" href="#svn.advanced.props.special.keywords" title="Keyword Substitution">the section called “Keyword Substitution”</a> and
<a class="xref" href="#svn.advanced.props.special.eol-style" title="End-of-Line Character Sequences">the section called “End-of-Line Character Sequences”</a> for more
details). By default, Subversion treats all file data as
literal byte strings, and files are always stored in the
repository in an untranslated state.</p><p>Second, Subversion maintains an internal notion of whether a
file is <span class="quote">“<span class="quote">text</span>”</span> or <span class="quote">“<span class="quote">binary</span>”</span> data, but
this notion is <span class="emphasis"><em>only</em></span> extant in the working
copy. During an <span class="command"><strong>svn update</strong></span>, Subversion will
perform contextual merges on locally modified text files, but
will not attempt to do so for binary files.</p><p>To determine whether a contextual merge is possible,
Subversion examines the <code class="literal">svn:mime-type</code>
property. If the file has no <code class="literal">svn:mime-type</code>
property, or has a MIME type that is textual (e.g.,
<code class="literal">text/*</code>),
Subversion assumes it is text. Otherwise, Subversion assumes
the file is binary. Subversion also helps users by running a
binary-detection algorithm in the <span class="command"><strong>svn import</strong></span>
and <span class="command"><strong>svn add</strong></span> commands. These commands will
make a good guess and then (possibly) set a binary
<code class="literal">svn:mime-type</code> property on the file being
added. (If Subversion guesses wrong, the user can always remove
or hand-edit the property.)</p></div><div class="sect1" title="Versioned Modules"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.forcvs.modules"></a>Versioned Modules</h2></div></div></div><p>Unlike CVS, a Subversion working copy is aware that it has
checked out a module. That means if somebody changes the
definition of a module (e.g., adds or removes components), a
call to <span class="command"><strong>svn update</strong></span> will update the working
copy appropriately, adding and removing components.</p><p>Subversion defines modules as a list of directories within a
directory property; see <a class="xref" href="#svn.advanced.externals" title="Externals Definitions">the section called “Externals Definitions”</a>.</p></div><div class="sect1" title="Authentication"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.forcvs.auth"></a>Authentication</h2></div></div></div><p>With CVS's pserver, you are required to log in to the server
(using the <span class="command"><strong>cvs login</strong></span> command) before
performing any read or write operation—you sometimes even
have to log in for anonymous operations. With a Subversion
repository using Apache <span class="command"><strong>httpd</strong></span> or
<span class="command"><strong>svnserve</strong></span> as the server, you don't provide any
authentication credentials at the outset—if an operation
that you perform requires authentication, the server will
challenge you for your credentials (whether those credentials
are username and password, a client certificate, or even both).
So if your repository is world-readable, you will not be
required to authenticate at all for read operations.</p><p>As with CVS, Subversion still caches your credentials on
disk (in your <code class="filename">~/.subversion/auth/</code>
directory) unless you tell it not to by using the
<code class="option">--no-auth-cache</code> option.</p><p>The exception to this behavior, however, is in the case of
accessing an <span class="command"><strong>svnserve</strong></span> server over an SSH
tunnel, using the <code class="literal">svn+ssh://</code> URL scheme. In
that case, the <span class="command"><strong>ssh</strong></span> program unconditionally
demands authentication just to start the tunnel.</p></div><div class="sect1" title="Converting a Repository from CVS to Subversion"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.forcvs.convert"></a>Converting a Repository from CVS to Subversion</h2></div></div></div><p>Perhaps the most important way to familiarize CVS users with
Subversion is to let them continue to work on their projects
using the new system. And while that can be somewhat
accomplished using a flat import into a Subversion repository of
an exported CVS repository, the more thorough solution involves
transferring not just the latest snapshot of their data, but all
the history behind it as well, from one system to another. This
is an extremely difficult problem to solve; it involves
deducing changesets in the absence of atomicity and translating
between the systems' completely orthogonal branching policies,
among other complications. Still, a handful of tools claim
to at least partially support the ability to convert
existing CVS repositories into Subversion ones.</p><p>The most popular (and mature) conversion tool is
cvs2svn (<a class="ulink" href="http://cvs2svn.tigris.org/" target="_top">http://cvs2svn.tigris.org/</a>), a Python
program originally created by members of Subversion's own
development community. This tool is meant to run exactly once:
it scans your CVS repository multiple times and attempts to
deduce commits, branches, and tags as best it can. When it
finishes, the result is either a Subversion repository or a
portable Subversion dump file representing your code's history.
See the web site for detailed instructions and caveats.</p></div><div class="footnotes"><br></br><hr width="100" align="left"></hr><div class="footnote"><p><sup>[<a id="ftn.idp23145216" href="#idp23145216" class="para">74</a>] </sup>That is, providing you don't run out of disk
space before your checkout finishes.</p></div></div></div><div class="appendix" title="Appendix C. WebDAV and Autoversioning"><div class="titlepage"><div><div><h2 class="title"><a id="svn.webdav"></a>Appendix C. WebDAV and Autoversioning</h2></div></div></div><div class="toc"><p><strong>Table of Contents</strong></p><dl><dt><span class="sect1"><a href="#svn.webdav.basic">What Is WebDAV?</a></span></dt><dt><span class="sect1"><a href="#svn.webdav.autoversioning">Autoversioning</a></span></dt><dt><span class="sect1"><a href="#svn.webdav.clients">Client Interoperability</a></span></dt><dd><dl><dt><span class="sect2"><a href="#svn.webdav.clients.standalone">Standalone WebDAV Applications</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.webdav.clients.standalone.windows">Microsoft Office, Dreamweaver, Photoshop</a></span></dt><dt><span class="sect3"><a href="#svn.webdav.clients.standalone.free">cadaver, DAV Explorer</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.webdav.clients.file-explorer-extensions">File-Explorer WebDAV Extensions</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.webdav.clients.file-explorer-extensions.windows">Microsoft Web Folders</a></span></dt><dt><span class="sect3"><a href="#svn.webdav.clients.file-explorer-extensions.linux-de">Nautilus, Konqueror</a></span></dt></dl></dd><dt><span class="sect2"><a href="#svn.webdav.clients.fs-impl">WebDAV Filesystem Implementation</a></span></dt><dd><dl><dt><span class="sect3"><a href="#svn.webdav.clients.fs-impl.windows">WebDrive, NetDrive</a></span></dt><dt><span class="sect3"><a href="#svn.webdav.clients.fs-impl.macosx">Mac OS X</a></span></dt><dt><span class="sect3"><a href="#svn.webdav.clients.fs-impl.linux">Linux davfs2</a></span></dt></dl></dd></dl></dd></dl></div><p>WebDAV is an extension to HTTP, and it is growing more and more
popular as a standard for file sharing. Today's operating systems
are becoming extremely web-aware, and many now have built-in
support for mounting <span class="quote">“<span class="quote">shares</span>”</span> exported by WebDAV
servers.</p><p>If you use Apache as your Subversion network server, to
some extent you are also running a WebDAV server. This appendix
gives some background on the nature of this protocol, how
Subversion uses it, and how well Subversion interoperates with
other software that is WebDAV-aware.</p><div class="sect1" title="What Is WebDAV?"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.webdav.basic"></a>What Is WebDAV?</h2></div></div></div><p><em class="firstterm">DAV</em> stands for <span class="quote">“<span class="quote">Distributed
Authoring and Versioning.</span>”</span> RFC 2518 defines a set of
concepts and accompanying extension methods to HTTP 1.1 that
make the Web a more universal read/write medium. The basic
idea is that a WebDAV-compliant web server can act like a
generic file server; clients can <span class="quote">“<span class="quote">mount</span>”</span> shared
folders over HTTP that behave much like other network
filesystems (such as NFS or SMB).</p><p>The tragedy, though, is that despite the acronym, the RFC
specification doesn't actually describe any sort of version
control. Basic WebDAV clients and servers assume that only one
version of each file or directory exists, and that it can be
repeatedly overwritten.</p><p>Because RFC 2518 left out versioning concepts, another
committee was left with the responsibility of writing RFC 3253 a
few years later. The new RFC adds versioning concepts to
WebDAV, placing the <span class="quote">“<span class="quote">V</span>”</span> back in
<span class="quote">“<span class="quote">DAV</span>”</span>—hence the term <span class="quote">“<span class="quote">DeltaV.</span>”</span>
WebDAV/DeltaV clients and servers are often called just
<span class="quote">“<span class="quote">DeltaV</span>”</span> programs, since DeltaV implies the
existence of basic WebDAV.</p><p>The original WebDAV standard has been widely successful.
Every modern computer operating system has a general WebDAV
client built in (details to follow), and a number of popular
standalone applications are also able to speak
WebDAV—Microsoft Office, Dreamweaver, and Photoshop, to
name a few. On the server end, Apache HTTP Server has been
able to provide WebDAV services since 1998 and is considered the
de facto open source standard. Several other
commercial WebDAV servers are available, including Microsoft's own
IIS.</p><p>DeltaV, unfortunately, has not been so successful. It's
very difficult to find any DeltaV clients or servers. The few
that do exist are relatively unknown commercial products, and
thus it's very difficult to test interoperability. It's not
entirely clear as to why DeltaV has remained stagnant. Some
opine that the specification is just too complex. Others argue
that while WebDAV's features have mass appeal (even the least
technical users appreciate network file sharing), its version
control features just aren't interesting or necessary for most
users. Finally, some believe that DeltaV remains unpopular
because there's still no open source server product that
implements it well.</p><p>When Subversion was still in its design phase, it seemed
like a great idea to use Apache as a network server. It already
had a module to provide WebDAV services. DeltaV was a
relatively new specification. The hope was that the Subversion
server module (<span class="command"><strong>mod_dav_svn</strong></span>) would eventually
evolve into an open source DeltaV reference implementation.
Unfortunately, DeltaV has a very specific versioning model that
doesn't quite line up with Subversion's model. Some concepts
were mappable; others were not.</p><p>What does this mean, then?</p><p>First, the Subversion client is not a fully implemented
DeltaV client. It needs certain types of things from the server
that DeltaV itself cannot provide, and thus is largely dependent
on a number of Subversion-specific
HTTP <code class="literal">REPORT</code> requests that
only <span class="command"><strong>mod_dav_svn</strong></span> understands.</p><p>Second, <span class="command"><strong>mod_dav_svn</strong></span> is not a
fully realized DeltaV server. Many portions of the DeltaV
specification were irrelevant to Subversion, and thus were left
unimplemented.</p><p>A long-held debate in the Subversion developer community
about whether it was worthfile to remedy either of these
situations eventually reached closure, with the Subversion
developers officially deciding to abandon plans to fully support
DeltaV. As of Subversion 1.7, Subversion clients and servers
introduce numerous non-standard simplifications of the DeltaV
standards<sup>[<a id="idp23364224" href="#ftn.idp23364224" class="footnote">75</a>]</sup>, with more
customizations of this sort likely to come. Those versions of
Subversion will, of course, continue to provide the same DeltaV
feature support already present in older releases, but no new
work will be done to increase coverage of the
specification—Subversion is intentionally moving away from
strict DeltaV as its primary HTTP-based protocol.</p></div><div class="sect1" title="Autoversioning"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.webdav.autoversioning"></a>Autoversioning</h2></div></div></div><p>While the Subversion client is not a full DeltaV client, and
the Subversion server is not a full DeltaV server, there's still a
glimmer of WebDAV interoperability to be happy about:
<em class="firstterm">autoversioning</em>.</p><p>Autoversioning is an optional feature defined in the DeltaV
standard. A typical DeltaV server will reject an ignorant
WebDAV client attempting to do a <code class="literal">PUT</code> to a
file that's under version control. To change a
version-controlled file, the server expects a series of proper
versioning requests: something like
<code class="literal">MKACTIVITY</code>, <code class="literal">CHECKOUT</code>,
<code class="literal">PUT</code>, <code class="literal">CHECKIN</code>. But if the
DeltaV server supports autoversioning, write requests from
basic WebDAV clients are accepted. The server behaves as though the
client <span class="emphasis"><em>had</em></span> issued the proper series of
versioning requests, performing a commit under the hood. In
other words, it allows a DeltaV server to interoperate with
ordinary WebDAV clients that don't understand versioning.</p><p>Because so many operating systems already have integrated
WebDAV clients, the use case for this feature can be incredibly
appealing to administrators working with non-technical users.
Imagine an office of ordinary users running Microsoft Windows or
Mac OS. Each user <span class="quote">“<span class="quote">mounts</span>”</span> the Subversion
repository, which appears to be an ordinary network folder.
They use the shared folder as they always do: open files, edit
them, and save them. Meanwhile, the server is automatically
versioning everything. Any administrator (or knowledgeable
user) can still use a Subversion client to search history and
retrieve older versions of data.</p><p>This scenario isn't fiction—it's real and it works.
To activate autoversioning in <span class="command"><strong>mod_dav_svn</strong></span>,
use the <code class="literal">SVNAutoversioning</code> directive within
the <code class="filename">httpd.conf</code> <code class="literal">Location</code>
block, like so:</p><div class="informalexample"><pre class="programlisting">
<Location /repos>
DAV svn
SVNPath /var/svn/repository
SVNAutoversioning on
</Location>
</pre></div><p>When Subversion autoversioning is active, write requests
from WebDAV clients result in automatic commits. A generic log
message is automatically generated and attached to each
revision.</p><p>Before activating this feature, however, understand what
you're getting into. WebDAV clients tend to do
<span class="emphasis"><em>many</em></span> write requests, resulting in a huge
number of automatically committed revisions. For example, when
saving data, many clients will do a <code class="literal">PUT</code> of a
0-byte file (as a way of reserving a name) followed by another
<code class="literal">PUT</code> with the real file data. The single
file-write results in two separate commits. Also consider that
many applications auto-save every few minutes, resulting in even
more commits.</p><p>If you have a post-commit hook program that sends email, you
may want to disable email generation either altogether or on
certain sections of the repository; it depends on whether you
think the influx of emails will still prove to be valuable
notifications or not. Also, a smart post-commit hook program
can distinguish between a transaction created via autoversioning
and one created through a normal Subversion commit operation.
The trick is to look for a revision property
named <code class="literal">svn:autoversioned</code>. If present, the
commit was made by a generic WebDAV client.</p><p>Another feature that may be a useful complement for
Subversion's autoversioning comes from Apache's
<code class="literal">mod_mime</code> module. If a WebDAV client adds a
new file to the repository, there's no opportunity for the user
to set the the <code class="literal">svn:mime-type</code> property. This
might cause the file to appear as a generic icon when viewed
within a WebDAV shared folder, not having an association with
any application. One remedy is to have a sysadmin (or other
Subversion-knowledgeable person) check out a working copy and
manually set the <code class="literal">svn:mime-type</code> property on
necessary files. But there's potentially no end to such cleanup
tasks. Instead, you can use the
<code class="literal">ModMimeUsePathInfo</code> directive in your
Subversion <code class="literal"><Location></code> block:</p><div class="informalexample"><pre class="programlisting">
<Location /repos>
DAV svn
SVNPath /var/svn/repository
SVNAutoversioning on
ModMimeUsePathInfo on
</Location>
</pre></div><p>This directive allows <code class="literal">mod_mime</code> to attempt
automatic deduction of the MIME type on new files that enter the
repository via autoversioning. The module looks at the file's
named extension and possibly the contents as well; if the file
matches some common patterns, the
file's <code class="literal">svn:mime-type</code> property will be set
automatically.</p></div><div class="sect1" title="Client Interoperability"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="svn.webdav.clients"></a>Client Interoperability</h2></div></div></div><p>All WebDAV clients fall into one of three
categories—standalone applications, file-explorer
extensions, or filesystem implementations. These categories
broadly define the types of WebDAV functionality available to
users. <a class="xref" href="#svn.webdav.clients.tbl-1" title="Table C.1. Common WebDAV clients">Table C.1, “Common WebDAV clients”</a> gives our
categorization as well as a quick description of some common pieces of
WebDAV-enabled software. You can find more details about these software
offerings, as well as their general category, in
the sections that follow.</p><div class="table"><a id="svn.webdav.clients.tbl-1"></a><p class="title"><strong>Table C.1. Common WebDAV clients</strong></p><div class="table-contents"><table summary="Common WebDAV clients" border="1"><colgroup><col></col><col></col><col></col><col></col><col></col><col></col></colgroup><thead><tr><th>Software</th><th>Type</th><th>Windows</th><th>Mac</th><th>Linux</th><th>Description</th></tr></thead><tbody><tr><td>Adobe Photoshop</td><td>Standalone WebDAV application</td><td>X</td><td> </td><td> </td><td>Image editing software, allowing direct opening
from, and writing to, WebDAV URLs</td></tr><tr><td>cadaver</td><td>Standalone WebDAV application</td><td> </td><td>X</td><td>X</td><td>Command-line WebDAV client supporting file
transfer, tree, and locking operations</td></tr><tr><td>DAV Explorer</td><td>Standalone WebDAV application</td><td>X</td><td>X</td><td>X</td><td>Java GUI tool for exploring WebDAV shares</td></tr><tr><td>Adobe Dreamweaver</td><td>Standalone WebDAV application</td><td>X</td><td> </td><td> </td><td>Web production software able to directly read from
and write to WebDAV URLs</td></tr><tr><td>Microsoft Office</td><td>Standalone WebDAV application</td><td>X</td><td> </td><td> </td><td>Office productivity suite with several components
able to directly read from and write to WebDAV
URLs</td></tr><tr><td>Microsoft Web Folders</td><td>File-explorer WebDAV extension</td><td>X</td><td> </td><td> </td><td>GUI file explorer program able to perform tree
operations on a WebDAV share</td></tr><tr><td>GNOME Nautilus</td><td>File-explorer WebDAV extension</td><td> </td><td> </td><td>X</td><td>GUI file explorer able to perform tree
operations on a WebDAV share</td></tr><tr><td>KDE Konqueror</td><td>File-explorer WebDAV extension</td><td> </td><td> </td><td>X</td><td>GUI file explorer able to perform tree
operations on a WebDAV share</td></tr><tr><td>Mac OS X</td><td>WebDAV filesystem implementation</td><td> </td><td>X</td><td> </td><td>Operating system that has built-in support for mounting
WebDAV shares.</td></tr><tr><td>Novell NetDrive</td><td>WebDAV filesystem implementation</td><td>X</td><td> </td><td> </td><td>Drive-mapping program for assigning Windows drive
letters to a mounted remote WebDAV share</td></tr><tr><td>SRT WebDrive</td><td>WebDAV filesystem implementation</td><td>X</td><td> </td><td> </td><td>File transfer software, which, among other things,
allows the assignment of Windows drive letters to a
mounted remote WebDAV share</td></tr><tr><td>davfs2</td><td>WebDAV filesystem implementation</td><td> </td><td> </td><td>X</td><td>Linux filesystem driver that allows you to mount a
WebDAV share</td></tr></tbody></table></div></div><br class="table-break"></br><div class="sect2" title="Standalone WebDAV Applications"><div class="titlepage"><div><div><h3 class="title"><a id="svn.webdav.clients.standalone"></a>Standalone WebDAV Applications</h3></div></div></div><p>A WebDAV application is a program that speaks WebDAV
protocols with a WebDAV server. We'll cover some of the most
popular programs with this kind of WebDAV support.</p><div class="sect3" title="Microsoft Office, Dreamweaver, Photoshop"><div class="titlepage"><div><div><h4 class="title"><a id="svn.webdav.clients.standalone.windows"></a>Microsoft Office, Dreamweaver, Photoshop</h4></div></div></div><p>On Windows, several well-known applications contain
integrated WebDAV client functionality, such as Microsoft's
Office,<sup>[<a id="idp23439472" href="#ftn.idp23439472" class="footnote">76</a>]</sup> Adobe's Photoshop and
Dreamweaver programs. They're able to directly open and
save to URLs, and tend to make heavy use of WebDAV locks
when editing a file.</p><p>Note that while many of these programs also exist for
Mac OS X, they do not appear to support WebDAV directly
on that platform. In fact, on Mac OS X, the
<span class="guimenu">File→Open</span> dialog box doesn't allow
one to type a path or URL at all. It's likely that the
WebDAV features were deliberately left out of Macintosh
versions of these programs, since OS X already provides such
excellent low-level filesystem support for WebDAV.</p></div><div class="sect3" title="cadaver, DAV Explorer"><div class="titlepage"><div><div><h4 class="title"><a id="svn.webdav.clients.standalone.free"></a>cadaver, DAV Explorer</h4></div></div></div><p>cadaver is a bare-bones Unix command-line program for
browsing and changing WebDAV shares. Like the Subversion
client, it uses the neon HTTP library—not
surprisingly, since both neon and cadaver are written by the
same author. cadaver is free software (GPL license) and is
available at
<a class="ulink" href="http://www.webdav.org/cadaver/" target="_top">http://www.webdav.org/cadaver/</a>.</p><p>Using cadaver is similar to using a command-line FTP
program, and thus it's extremely useful for basic WebDAV
debugging. It can be used to upload or download files in a
pinch, to examine properties, and to copy, move,
lock, or unlock files:</p><div class="informalexample"><pre class="screen">
$ cadaver http://host/repos
dav:/repos/> ls
Listing collection `/repos/': succeeded.
Coll: > foobar 0 May 10 16:19
> playwright.el 2864 May 4 16:18
> proofbypoem.txt 1461 May 5 15:09
> westcoast.jpg 66737 May 5 15:09
dav:/repos/> put README
Uploading README to `/repos/README':
Progress: [=============================>] 100.0% of 357 bytes succeeded.
dav:/repos/> get proofbypoem.txt
Downloading `/repos/proofbypoem.txt' to proofbypoem.txt:
Progress: [=============================>] 100.0% of 1461 bytes succeeded.
</pre></div><p>DAV Explorer is another standalone WebDAV client, written
in Java. It's under a free Apache-like license and is
available at <a class="ulink" href="http://www.ics.uci.edu/~webdav/" target="_top">http://www.ics.uci.edu/~webdav/</a>.
It does everything cadaver does, but has the
advantages of being portable and being a more user-friendly GUI
application. It's also one of the first clients to support
the new WebDAV Access Control Protocol (RFC 3744).</p><p>Of course, DAV Explorer's ACL support is useless in this
case, since <span class="command"><strong>mod_dav_svn</strong></span> doesn't support
it. The fact that both cadaver and DAV Explorer support
some limited DeltaV commands isn't particularly useful
either, since they don't allow <code class="literal">MKACTIVITY</code>
requests. But it's not relevant anyway; we're assuming all
of these clients are operating against an autoversioning
repository.</p></div></div><div class="sect2" title="File-Explorer WebDAV Extensions"><div class="titlepage"><div><div><h3 class="title"><a id="svn.webdav.clients.file-explorer-extensions"></a>File-Explorer WebDAV Extensions</h3></div></div></div><p>Some popular file explorer GUI programs support WebDAV
extensions that allow a user to browse a DAV share as though it
was just another directory on the local computer, and to
perform basic tree editing operations on the items in that
share. For example, Windows Explorer is able to browse a
WebDAV server as a <span class="quote">“<span class="quote">network place.</span>”</span> Users can
drag files to and from the desktop, or can rename, copy, or
delete files in the usual way. But because it's only a
feature of the file explorer, the DAV share isn't visible to
ordinary applications. All DAV interaction must happen
through the explorer interface.</p><div class="sect3" title="Microsoft Web Folders"><div class="titlepage"><div><div><h4 class="title"><a id="svn.webdav.clients.file-explorer-extensions.windows"></a>Microsoft Web Folders</h4></div></div></div><p>Microsoft was one of the original backers of the WebDAV
specification, and first started shipping a client in
Windows 98, which was known as Web Folders. This client was
also shipped in Windows NT 4.0 and Windows 2000.</p><p>The original Web Folders client was an extension to
Explorer, the main GUI program used to browse filesystems. It
works well enough. In Windows 98, the feature might need to
be explicitly installed if Web Folders aren't already visible
inside My Computer. In Windows 2000, simply
add a new <span class="quote">“<span class="quote">network place,</span>”</span> enter the URL, and the
WebDAV share will pop up for browsing.</p><p>With the release of Windows XP, Microsoft started shipping
a new implementation of Web Folders, known as the WebDAV
Mini-Redirector. The new implementation is a
filesystem-level client, allowing WebDAV shares to be mounted
as drive letters. Unfortunately, this implementation is
incredibly buggy. The client usually tries to convert HTTP
URLs (<code class="literal">http://host/repos</code>) into UNC share
notation (<code class="literal">\\host\repos</code>); it also often
tries to use Windows Domain authentication to respond to
basic-auth HTTP challenges, sending usernames as
<code class="literal">HOST\username</code>. These interoperability
problems are severe and are documented in numerous places around
the Web, to the frustration of many users. Even Greg Stein,
the original author of Apache's WebDAV module, bluntly
states that XP Web Folders simply can't operate against an Apache
server.</p><p>Windows Vista's initial implementation of Web Folders seems to
be almost the same as XP's, so it has the same sort of
problems. With luck, Microsoft will remedy these issues in
a Vista Service Pack.</p><p>However, there seem to be workarounds for both XP and
Vista that allow Web Folders to work against Apache. Users
have mostly reported success with these techniques, so we'll
relay them here.</p><p>On Windows XP, you have two options. First, search
Microsoft's web site for update KB907306, <span class="quote">“<span class="quote">Software
Update for Web Folders.</span>”</span> This may fix all your
problems. If it doesn't, it seems that the original pre-XP
Web Folders implementation is still buried within the
system. You can unearth it by going to Network
Places and adding a new network place. When prompted,
enter the URL of the repository, but <span class="emphasis"><em>include a
port number</em></span> in the URL. For
example, you should enter <strong class="userinput"><code>http://host/repos</code></strong>
as <strong class="userinput"><code>http://host:80/repos</code></strong> instead.
Respond to any authentication prompts with your Subversion
credentials.</p><p>On Windows Vista, the same KB907306 update may clear
everything up. But there may still be other issues. Some
users have reported that Vista considers
all <code class="literal">http://</code> connections insecure, and thus
will always fail any authentication challenges from Apache
unless the connection happens
over <code class="literal">https://</code>. If you're unable to connect
to the Subversion repository via SSL, you can tweak the
system registry to turn off this behavior. Just change the
value of the <code class="literal">HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\WebClient\Parameters\BasicAuthLevel</code>
key from <strong class="userinput"><code>1</code></strong> to <strong class="userinput"><code>2</code></strong>. A final warning: be sure to set up the Web
Folder to point to the repository's root directory
(<code class="filename">/</code>), rather than some subdirectory
such as <code class="filename">/trunk</code>. Vista Web Folders
seems to work only against repository roots.</p><p>In general, while these workarounds may function for
you, you might get a better overall experience using a
third-party WebDAV client such as WebDrive or NetDrive.</p></div><div class="sect3" title="Nautilus, Konqueror"><div class="titlepage"><div><div><h4 class="title"><a id="svn.webdav.clients.file-explorer-extensions.linux-de"></a>Nautilus, Konqueror</h4></div></div></div><p>Nautilus is the official file manager/browser for the
GNOME desktop (<a class="ulink" href="http://www.gnome.org" target="_top">http://www.gnome.org</a>), and
Konqueror is the manager/browser for the KDE desktop (<a class="ulink" href="http://www.kde.org" target="_top">http://www.kde.org</a>). Both of these applications
have an explorer-level WebDAV client built in, and they
operate just fine against an autoversioning repository.</p><p>In GNOME's Nautilus, select the <span class="guimenuitem">File→Open
location</span> menu item and enter the URL in the
dialog box presented. The repository should then be
displayed like any other filesystem.</p><p>In KDE's Konqueror, you need to use the
<code class="literal">webdav://</code> scheme when entering the URL in
the location bar. If you enter
an <code class="literal">http://</code> URL, Konqueror will behave
like an ordinary web browser. You'll likely see the generic
HTML directory listing produced
by <span class="command"><strong>mod_dav_svn</strong></span>. When you enter
<strong class="userinput"><code>webdav://host/repos</code></strong> instead of
<strong class="userinput"><code>http://host/repos</code></strong>, Konqueror becomes a
WebDAV client and displays the repository as a
filesystem.</p></div></div><div class="sect2" title="WebDAV Filesystem Implementation"><div class="titlepage"><div><div><h3 class="title"><a id="svn.webdav.clients.fs-impl"></a>WebDAV Filesystem Implementation</h3></div></div></div><p>The WebDAV filesystem implementation is arguably the best
sort of WebDAV client. It's implemented as a low-level
filesystem module, typically within the operating system's
kernel. This means that the DAV share is mounted like any
other network filesystem, similar to mounting an NFS share on
Unix or attaching an SMB share as a drive letter in Windows.
As a result, this sort of client provides completely
transparent read/write WebDAV access to all programs.
Applications aren't even aware that WebDAV requests are
happening.</p><div class="sect3" title="WebDrive, NetDrive"><div class="titlepage"><div><div><h4 class="title"><a id="svn.webdav.clients.fs-impl.windows"></a>WebDrive, NetDrive</h4></div></div></div><p>Both WebDrive and NetDrive are excellent commercial
products that allow a WebDAV share to be attached as drive
letters in Windows. As a result, you can operate on the
contents of these WebDAV-backed pseudodrives as easily as
you can against real local hard drives, and in the same
ways. You can purchase WebDrive from South River
Technologies (<a class="ulink" href="http://www.southrivertech.com" target="_top">http://www.southrivertech.com</a>).
Novell's NetDrive is freely available online, but requires
users to have a NetWare license.</p></div><div class="sect3" title="Mac OS X"><div class="titlepage"><div><div><h4 class="title"><a id="svn.webdav.clients.fs-impl.macosx"></a>Mac OS X</h4></div></div></div><p>Apple's OS X operating system has an integrated
filesystem-level WebDAV client. From the Finder, select the
<span class="guimenuitem">Go→Connect to Server</span> menu
item. Enter a WebDAV URL, and it appears as a disk on the
desktop, just like any other mounted volume. You can also
mount a WebDAV share from the Darwin terminal by using the
<code class="literal">webdav</code> filesystem type with the
<span class="command"><strong>mount</strong></span> command:</p><div class="informalexample"><pre class="screen">
$ mount -t webdav http://svn.example.com/repos/project /some/mountpoint
$
</pre></div><p>Note that if your <span class="command"><strong>mod_dav_svn</strong></span> is
older than version 1.2, OS X will refuse to mount the share
as read/write; it will appear as read-only. This is because
OS X insists on locking support for read/write shares, and
the ability to lock files first appeared in Subversion
1.2.</p><p>Also, OS X's WebDAV client can sometimes be overly
sensitive to HTTP redirects. If OS X is unable to mount the
repository at all, you may need to enable
the <code class="literal">BrowserMatch</code> directive in the Apache
server's <code class="filename">httpd.conf</code>:</p><div class="informalexample"><pre class="programlisting">
BrowserMatch "^WebDAVFS/1.[012]" redirect-carefully
</pre></div></div><div class="sect3" title="Linux davfs2"><div class="titlepage"><div><div><h4 class="title"><a id="svn.webdav.clients.fs-impl.linux"></a>Linux davfs2</h4></div></div></div><p>Linux davfs2 is a filesystem module for the Linux
kernel, whose development is organized at
<a class="ulink" href="http://dav.sourceforge.net/" target="_top">http://dav.sourceforge.net/</a>. Once you
install davfs2, you can mount a WebDAV network share using
the usual Linux mount command:</p><div class="informalexample"><pre class="screen">
$ mount.davfs http://host/repos /mnt/dav
</pre></div></div></div></div><div class="footnotes"><br></br><hr width="100" align="left"></hr><div class="footnote"><p><sup>[<a id="ftn.idp23364224" href="#idp23364224" class="para">75</a>] </sup>The Subversion developers colloquially
refer to this deviation from the DeltaV standard
as <span class="quote">“<span class="quote">HTTPv2</span>”</span>.</p></div><div class="footnote"><p><sup>[<a id="ftn.idp23439472" href="#idp23439472" class="para">76</a>] </sup>WebDAV support was removed from
Microsoft Access for some reason, but it exists in the rest
of the Office suite.</p></div></div></div><div class="appendix" title="Appendix D. Copyright"><div class="titlepage"><div><div><h2 class="title"><a id="svn.copyright"></a>Appendix D. Copyright</h2></div></div></div><p>Copyright (c) 2002-2013 Ben Collins-Sussman, Brian
W. Fitzpatrick, C. Michael Pilato.</p><p>This work is licensed under the Creative Commons Attribution
License. To view a copy of this license, visit
<a class="ulink" href="http://creativecommons.org/licenses/by/2.0/" target="_top">http://creativecommons.org/licenses/by/2.0/</a> or
send a letter to Creative Commons, 559 Nathan Abbott Way,
Stanford, California 94305, USA.</p><p>A summary of the license is given below, followed by the full
legal text.</p><pre class="programlisting">
You are free:
* to copy, distribute, display, and perform the work
* to make derivative works
* to make commercial use of the work
Under the following conditions:
Attribution. You must give the original author credit.
* For any reuse or distribution, you must make clear to others the
license terms of this work.
* Any of these conditions can be waived if you get permission from
the author.
Your fair use and other rights are in no way affected by the above.
The above is a summary of the full license below.
====================================================================
Creative Commons Legal Code
Attribution 2.0
CREATIVE COMMONS CORPORATION IS NOT A LAW FIRM AND DOES NOT PROVIDE
LEGAL SERVICES. DISTRIBUTION OF THIS LICENSE DOES NOT CREATE AN
ATTORNEY-CLIENT RELATIONSHIP. CREATIVE COMMONS PROVIDES THIS
INFORMATION ON AN "AS-IS" BASIS. CREATIVE COMMONS MAKES NO WARRANTIES
REGARDING THE INFORMATION PROVIDED, AND DISCLAIMS LIABILITY FOR
DAMAGES RESULTING FROM ITS USE.
License
THE WORK (AS DEFINED BELOW) IS PROVIDED UNDER THE TERMS OF THIS
CREATIVE COMMONS PUBLIC LICENSE ("CCPL" OR "LICENSE"). THE WORK IS
PROTECTED BY COPYRIGHT AND/OR OTHER APPLICABLE LAW. ANY USE OF THE
WORK OTHER THAN AS AUTHORIZED UNDER THIS LICENSE OR COPYRIGHT LAW IS
PROHIBITED.
BY EXERCISING ANY RIGHTS TO THE WORK PROVIDED HERE, YOU ACCEPT AND
AGREE TO BE BOUND BY THE TERMS OF THIS LICENSE. THE LICENSOR GRANTS
YOU THE RIGHTS CONTAINED HERE IN CONSIDERATION OF YOUR ACCEPTANCE OF
SUCH TERMS AND CONDITIONS.
1. Definitions
a. "Collective Work" means a work, such as a periodical issue,
anthology or encyclopedia, in which the Work in its entirety in
unmodified form, along with a number of other contributions,
constituting separate and independent works in themselves, are
assembled into a collective whole. A work that constitutes a
Collective Work will not be considered a Derivative Work (as
defined below) for the purposes of this License.
b. "Derivative Work" means a work based upon the Work or upon the
Work and other pre-existing works, such as a translation,
musical arrangement, dramatization, fictionalization, motion
picture version, sound recording, art reproduction, abridgment,
condensation, or any other form in which the Work may be recast,
transformed, or adapted, except that a work that constitutes a
Collective Work will not be considered a Derivative Work for the
purpose of this License. For the avoidance of doubt, where the
Work is a musical composition or sound recording, the
synchronization of the Work in timed-relation with a moving
image ("synching") will be considered a Derivative Work for the
purpose of this License.
c. "Licensor" means the individual or entity that offers the Work
under the terms of this License.
d. "Original Author" means the individual or entity who created the Work.
e. "Work" means the copyrightable work of authorship offered under
the terms of this License.
f. "You" means an individual or entity exercising rights under this
License who has not previously violated the terms of this
License with respect to the Work, or who has received express
permission from the Licensor to exercise rights under this
License despite a previous violation.
2. Fair Use Rights. Nothing in this license is intended to reduce,
limit, or restrict any rights arising from fair use, first sale or
other limitations on the exclusive rights of the copyright owner
under copyright law or other applicable laws.
3. License Grant. Subject to the terms and conditions of this License,
Licensor hereby grants You a worldwide, royalty-free,
non-exclusive, perpetual (for the duration of the applicable
copyright) license to exercise the rights in the Work as stated
below:
a. to reproduce the Work, to incorporate the Work into one or more
Collective Works, and to reproduce the Work as incorporated in
the Collective Works;
b. to create and reproduce Derivative Works;
c. to distribute copies or phonorecords of, display publicly,
perform publicly, and perform publicly by means of a digital
audio transmission the Work including as incorporated in
Collective Works;
d. to distribute copies or phonorecords of, display publicly,
perform publicly, and perform publicly by means of a digital
audio transmission Derivative Works.
e.
For the avoidance of doubt, where the work is a musical composition:
i. Performance Royalties Under Blanket Licenses. Licensor
waives the exclusive right to collect, whether
individually or via a performance rights society
(e.g. ASCAP, BMI, SESAC), royalties for the public
performance or public digital performance (e.g. webcast)
of the Work.
ii. Mechanical Rights and Statutory Royalties. Licensor waives
the exclusive right to collect, whether individually or
via a music rights agency or designated agent (e.g. Harry
Fox Agency), royalties for any phonorecord You create from
the Work ("cover version") and distribute, subject to the
compulsory license created by 17 USC Section 115 of the US
Copyright Act (or the equivalent in other jurisdictions).
f. Webcasting Rights and Statutory Royalties. For the avoidance of
doubt, where the Work is a sound recording, Licensor waives the
exclusive right to collect, whether individually or via a
performance-rights society (e.g. SoundExchange), royalties for
the public digital performance (e.g. webcast) of the Work,
subject to the compulsory license created by 17 USC Section 114
of the US Copyright Act (or the equivalent in other
jurisdictions).
The above rights may be exercised in all media and formats whether now
known or hereafter devised. The above rights include the right to make
such modifications as are technically necessary to exercise the rights
in other media and formats. All rights not expressly granted by
Licensor are hereby reserved.
4. Restrictions.The license granted in Section 3 above is expressly
made subject to and limited by the following restrictions:
a. You may distribute, publicly display, publicly perform, or
publicly digitally perform the Work only under the terms of this
License, and You must include a copy of, or the Uniform Resource
Identifier for, this License with every copy or phonorecord of
the Work You distribute, publicly display, publicly perform, or
publicly digitally perform. You may not offer or impose any
terms on the Work that alter or restrict the terms of this
License or the recipients' exercise of the rights granted
hereunder. You may not sublicense the Work. You must keep intact
all notices that refer to this License and to the disclaimer of
warranties. You may not distribute, publicly display, publicly
perform, or publicly digitally perform the Work with any
technological measures that control access or use of the Work in
a manner inconsistent with the terms of this License
Agreement. The above applies to the Work as incorporated in a
Collective Work, but this does not require the Collective Work
apart from the Work itself to be made subject to the terms of
this License. If You create a Collective Work, upon notice from
any Licensor You must, to the extent practicable, remove from
the Collective Work any reference to such Licensor or the
Original Author, as requested. If You create a Derivative Work,
upon notice from any Licensor You must, to the extent
practicable, remove from the Derivative Work any reference to
such Licensor or the Original Author, as requested.
b. If you distribute, publicly display, publicly perform, or
publicly digitally perform the Work or any Derivative Works or
Collective Works, You must keep intact all copyright notices for
the Work and give the Original Author credit reasonable to the
medium or means You are utilizing by conveying the name (or
pseudonym if applicable) of the Original Author if supplied; the
title of the Work if supplied; to the extent reasonably
practicable, the Uniform Resource Identifier, if any, that
Licensor specifies to be associated with the Work, unless such
URI does not refer to the copyright notice or licensing
information for the Work; and in the case of a Derivative Work,
a credit identifying the use of the Work in the Derivative Work
(e.g., "French translation of the Work by Original Author," or
"Screenplay based on original Work by Original Author"). Such
credit may be implemented in any reasonable manner; provided,
however, that in the case of a Derivative Work or Collective
Work, at a minimum such credit will appear where any other
comparable authorship credit appears and in a manner at least as
prominent as such other comparable authorship credit.
5. Representations, Warranties and Disclaimer
UNLESS OTHERWISE MUTUALLY AGREED TO BY THE PARTIES IN WRITING,
LICENSOR OFFERS THE WORK AS-IS AND MAKES NO REPRESENTATIONS OR
WARRANTIES OF ANY KIND CONCERNING THE WORK, EXPRESS, IMPLIED,
STATUTORY OR OTHERWISE, INCLUDING, WITHOUT LIMITATION, WARRANTIES OF
TITLE, MERCHANTIBILITY, FITNESS FOR A PARTICULAR PURPOSE,
NONINFRINGEMENT, OR THE ABSENCE OF LATENT OR OTHER DEFECTS, ACCURACY,
OR THE PRESENCE OF ABSENCE OF ERRORS, WHETHER OR NOT
DISCOVERABLE. SOME JURISDICTIONS DO NOT ALLOW THE EXCLUSION OF IMPLIED
WARRANTIES, SO SUCH EXCLUSION MAY NOT APPLY TO YOU.
6. Limitation on Liability. EXCEPT TO THE EXTENT REQUIRED BY
APPLICABLE LAW, IN NO EVENT WILL LICENSOR BE LIABLE TO YOU ON ANY
LEGAL THEORY FOR ANY SPECIAL, INCIDENTAL, CONSEQUENTIAL, PUNITIVE
OR EXEMPLARY DAMAGES ARISING OUT OF THIS LICENSE OR THE USE OF THE
WORK, EVEN IF LICENSOR HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH
DAMAGES.
7. Termination
a. This License and the rights granted hereunder will terminate
automatically upon any breach by You of the terms of this
License. Individuals or entities who have received Derivative
Works or Collective Works from You under this License, however,
will not have their licenses terminated provided such
individuals or entities remain in full compliance with those
licenses. Sections 1, 2, 5, 6, 7, and 8 will survive any
termination of this License.
b. Subject to the above terms and conditions, the license granted
here is perpetual (for the duration of the applicable copyright
in the Work). Notwithstanding the above, Licensor reserves the
right to release the Work under different license terms or to
stop distributing the Work at any time; provided, however that
any such election will not serve to withdraw this License (or
any other license that has been, or is required to be, granted
under the terms of this License), and this License will continue
in full force and effect unless terminated as stated above.
8. Miscellaneous
a. Each time You distribute or publicly digitally perform the Work
or a Collective Work, the Licensor offers to the recipient a
license to the Work on the same terms and conditions as the
license granted to You under this License.
b. Each time You distribute or publicly digitally perform a
Derivative Work, Licensor offers to the recipient a license to
the original Work on the same terms and conditions as the
license granted to You under this License.
c. If any provision of this License is invalid or unenforceable
under applicable law, it shall not affect the validity or
enforceability of the remainder of the terms of this License,
and without further action by the parties to this agreement,
such provision shall be reformed to the minimum extent necessary
to make such provision valid and enforceable.
d. No term or provision of this License shall be deemed waived and
no breach consented to unless such waiver or consent shall be in
writing and signed by the party to be charged with such waiver
or consent.
e. This License constitutes the entire agreement between the
parties with respect to the Work licensed here. There are no
understandings, agreements or representations with respect to
the Work not specified here. Licensor shall not be bound by any
additional provisions that may appear in any communication from
You. This License may not be modified without the mutual written
agreement of the Licensor and You.
Creative Commons is not a party to this License, and makes no warranty
whatsoever in connection with the Work. Creative Commons will not be
liable to You or any party on any legal theory for any damages
whatsoever, including without limitation any general, special,
incidental or consequential damages arising in connection to this
license. Notwithstanding the foregoing two (2) sentences, if Creative
Commons has expressly identified itself as the Licensor hereunder, it
shall have all rights and obligations of Licensor.
Except for the limited purpose of indicating to the public that the
Work is licensed under the CCPL, neither party will use the trademark
"Creative Commons" or any related trademark or logo of Creative
Commons without the prior written consent of Creative Commons. Any
permitted use will be in compliance with Creative Commons'
then-current trademark usage guidelines, as may be published on its
website or otherwise made available upon request from time to time.
Creative Commons may be contacted at http://creativecommons.org/.
====================================================================
</pre></div><div class="index" title="Index"><div class="titlepage"><div><div><h2 class="title"><a id="svn.index"></a>Index</h2></div></div></div><div class="index"><div class="indexdiv"><h3>A</h3><dl><dt>ancestry, <a class="indexterm" href="#svn.branchmerge.advanced.ancestry">Noticing or Ignoring Ancestry</a></dt></dl></div><div class="indexdiv"><h3>B</h3><dl><dt>BASE, <a class="indexterm" href="#svn.tour.revs.keywords">Revision Keywords</a></dt><dt>branches, <a class="indexterm" href="#svn.tour.importing.layout">Recommended Repository Layout</a>, <a class="indexterm" href="#svn.branchmerge.whatis">What's a Branch?</a></dt></dl></div><div class="indexdiv"><h3>C</h3><dl><dt>checkout (see working copy, creation)</dt><dt>CollabNet, <a class="indexterm" href="#svn.intro.history">Subversion's History</a></dt><dt>COMMITTED, <a class="indexterm" href="#svn.tour.revs.keywords">Revision Keywords</a></dt><dt>committing (see working copy, commit)</dt><dt>Concurrent Versions System, <a class="indexterm" href="#svn.preface">Preface</a></dt><dt>conflicts, <a class="indexterm" href="#svn.basic.vsn-models.copy-merge">The copy-modify-merge solution</a></dt><dd><dl><dt>conflict markers, <a class="indexterm" href="#svn.tour.cycle.resolve.pending">Postponing conflict resolution</a></dt><dt>resolution, <a class="indexterm" href="#svn.tour.cycle.resolve.theirsfull">Discarding your changes in favor of a newly fetched
revision</a></dt><dd><dl><dt>interactive, <a class="indexterm" href="#svn.tour.cycle.resolve.resolve">Resolving conflict differences interactively</a></dt><dt>manual, <a class="indexterm" href="#svn.tour.cycle.resolve.byhand">Merging conflicts by hand</a></dt><dt>postponing, <a class="indexterm" href="#svn.tour.cycle.resolve.pending">Postponing conflict resolution</a></dt></dl></dd><dt>resolving, <a class="indexterm" href="#svn.tour.cycle.resolve">Resolve Any Conflicts</a></dt><dt>reviewing, <a class="indexterm" href="#svn.tour.cycle.resolve.diff">Viewing conflict differences interactively</a></dt></dl></dd><dt>CVS (see Concurrent Versions System)</dt></dl></div><div class="indexdiv"><h3>D</h3><dl><dt>delta, <a class="indexterm" href="#svn.tour.cycle.examine">Review Your Changes</a></dt></dl></div><div class="indexdiv"><h3>H</h3><dl><dt>HEAD, <a class="indexterm" href="#svn.tour.revs.keywords">Revision Keywords</a></dt></dl></div><div class="indexdiv"><h3>L</h3><dl><dt>log message, <a class="indexterm" href="#svn.tour.cycle.examine">Review Your Changes</a></dt></dl></div><div class="indexdiv"><h3>M</h3><dl><dt>merge tracking, <a class="indexterm" href="#svn.branchmerge.basicmerging">Basic Merging</a></dt><dt>mergeinfo, <a class="indexterm" href="#svn.branchemerge.basicmerging.stayinsync">Keeping a Branch in Sync</a></dt><dt>mergeinfo elision, <a class="indexterm" href="#svn.branchemerge.basicmerging.stayinsync">Keeping a Branch in Sync</a></dt><dt>mergeinfo inheritance, <a class="indexterm" href="#svn.branchmerge.basicmerging.mergeinfo">Mergeinfo and Previews</a></dt><dt>mod_dav_svn, <a class="indexterm" href="#svn.intro.components">Subversion's Components</a></dt></dl></div><div class="indexdiv"><h3>P</h3><dl><dt>patches, <a class="indexterm" href="#svn.tour.cycle.examine.diff">Examine the details of your local modifications</a></dt><dt>PREV, <a class="indexterm" href="#svn.tour.revs.keywords">Revision Keywords</a></dt><dt>project root, <a class="indexterm" href="#svn.tour.importing.layout">Recommended Repository Layout</a>, <a class="indexterm" href="#svn.reposadmin.projects.chooselayout">Planning Your Repository Organization</a></dt><dt>properties, <a class="indexterm" href="#svn.advanced.props">Properties</a></dt></dl></div><div class="indexdiv"><h3>R</h3><dl><dt>repository</dt><dd><dl><dt>defined, <a class="indexterm" href="#svn.basic.repository">The Repository</a></dt><dt>hooks</dt><dd><dl><dt>post-commit, <a class="indexterm" href="#svn.ref.reposhooks.post-commit">post-commit</a></dt><dt>post-lock, <a class="indexterm" href="#svn.ref.reposhooks.post-lock">post-lock</a></dt><dt>post-revprop-change, <a class="indexterm" href="#svn.ref.reposhooks.post-revprop-change">post-revprop-change</a></dt><dt>post-unlock, <a class="indexterm" href="#svn.ref.reposhooks.post-unlock">post-unlock</a></dt><dt>pre-commit, <a class="indexterm" href="#svn.ref.reposhooks.pre-commit">pre-commit</a></dt><dt>pre-lock, <a class="indexterm" href="#svn.ref.reposhooks.pre-lock">pre-lock</a></dt><dt>pre-revprop-change, <a class="indexterm" href="#svn.ref.reposhooks.pre-revprop-change">pre-revprop-change</a></dt><dt>pre-unlock, <a class="indexterm" href="#svn.ref.reposhooks.pre-unlock">pre-unlock</a></dt><dt>start-commit, <a class="indexterm" href="#svn.ref.reposhooks.start-commit">start-commit</a></dt></dl></dd></dl></dd><dt>revisions</dt><dd><dl><dt>as dates, <a class="indexterm" href="#svn.tour.revs.dates">Revision Dates</a></dt><dt>defined, <a class="indexterm" href="#svn.basic.in-action.revs">Revisions</a></dt><dt>global, <a class="indexterm" href="#svn.basic.in-action.revs">Revisions</a></dt><dt>keywords, <a class="indexterm" href="#svn.tour.revs.keywords">Revision Keywords</a></dt><dd><dl><dt>BASE, <a class="indexterm" href="#svn.tour.revs.keywords">Revision Keywords</a></dt><dt>COMMITTED, <a class="indexterm" href="#svn.tour.revs.keywords">Revision Keywords</a></dt><dt>HEAD, <a class="indexterm" href="#svn.tour.revs.keywords">Revision Keywords</a></dt><dt>PREV, <a class="indexterm" href="#svn.tour.revs.keywords">Revision Keywords</a></dt></dl></dd></dl></dd></dl></div><div class="indexdiv"><h3>S</h3><dl><dt>SCM (see software configuration management)</dt><dt>software configuration management, <a class="indexterm" href="#svn.intro.whatis">What Is Subversion?</a></dt><dt>subtree merge, <a class="indexterm" href="#svn.branchemerge.basicmerging.stayinsync">Keeping a Branch in Sync</a></dt><dt>subtree mergeinfo, <a class="indexterm" href="#svn.branchemerge.basicmerging.stayinsync">Keeping a Branch in Sync</a></dt><dt>Subversion</dt><dd><dl><dt>architecture, <a class="indexterm" href="#svn.intro.architecture">Subversion's Architecture</a></dt><dt>components, <a class="indexterm" href="#svn.intro.components">Subversion's Components</a></dt><dt>defined, <a class="indexterm" href="#svn.intro.whatis">What Is Subversion?</a></dt><dt>history of, <a class="indexterm" href="#svn.intro.history">Subversion's History</a>, <a class="indexterm" href="#svn.intro.whatsnew">What's New in Subversion</a></dt></dl></dd><dt>svn, <a class="indexterm" href="#svn.intro.components">Subversion's Components</a></dt><dd><dl><dt>options, <a class="indexterm" href="#svn.tour.help">Help!</a></dt><dt>subcommands</dt><dd><dl><dt>add, <a class="indexterm" href="#svn.tour.cycle.edit">Make Your Changes </a>, <a class="indexterm" href="#svn.ref.svn.c.add">svn add</a></dt><dt>blame, <a class="indexterm" href="#svn.ref.svn.c.blame">svn blame (praise, annotate, ann)</a></dt><dt>cat, <a class="indexterm" href="#svn.ref.svn.c.cat">svn cat</a></dt><dt>changelist, <a class="indexterm" href="#svn.ref.svn.c.changelist">svn changelist (cl)</a></dt><dt>checkout, <a class="indexterm" href="#svn.basic.in-action.wc-funcdamentals">Fundamental working copy interactions</a>, <a class="indexterm" href="#svn.tour.initial">Creating a Working Copy</a>, <a class="indexterm" href="#svn.ref.svn.c.checkout">svn checkout (co)</a></dt><dt>cleanup, <a class="indexterm" href="#svn.ref.svn.c.cleanup">svn cleanup</a></dt><dt>commit, <a class="indexterm" href="#svn.basic.in-action.wc-funcdamentals">Fundamental working copy interactions</a>, <a class="indexterm" href="#svn.ref.svn.c.commit">svn commit (ci)</a></dt><dt>copy, <a class="indexterm" href="#svn.tour.cycle.edit">Make Your Changes </a>, <a class="indexterm" href="#svn.ref.svn.c.copy">svn copy (cp)</a></dt><dt>delete, <a class="indexterm" href="#svn.tour.cycle.edit">Make Your Changes </a>, <a class="indexterm" href="#svn.ref.svn.c.delete">svn delete (del, remove, rm)</a></dt><dt>diff, <a class="indexterm" href="#svn.tour.cycle.examine.diff">Examine the details of your local modifications</a>, <a class="indexterm" href="#svn.ref.svn.c.diff">svn diff (di)</a></dt><dt>export, <a class="indexterm" href="#svn.ref.svn.c.export">svn export</a></dt><dt>help, <a class="indexterm" href="#svn.tour.help">Help!</a>, <a class="indexterm" href="#svn.ref.svn.c.help">svn help (h, ?)</a></dt><dt>import, <a class="indexterm" href="#svn.tour.importing.import">Importing Files and Directories</a>, <a class="indexterm" href="#svn.ref.svn.c.import">svn import</a></dt><dt>info, <a class="indexterm" href="#svn.ref.svn.c.info">svn info</a></dt><dt>list, <a class="indexterm" href="#svn.ref.svn.c.list">svn list (ls)</a></dt><dt>lock, <a class="indexterm" href="#svn.ref.svn.c.lock">svn lock</a></dt><dt>log, <a class="indexterm" href="#svn.ref.svn.c.log">svn log</a></dt><dt>merge, <a class="indexterm" href="#svn.ref.svn.c.merge">svn merge</a></dt><dt>mergeinfo, <a class="indexterm" href="#svn.ref.svn.c.mergeinfo">svn mergeinfo</a></dt><dt>mkdir, <a class="indexterm" href="#svn.tour.cycle.edit">Make Your Changes </a>, <a class="indexterm" href="#svn.ref.svn.c.mkdir">svn mkdir</a></dt><dt>move, <a class="indexterm" href="#svn.tour.cycle.edit">Make Your Changes </a>, <a class="indexterm" href="#svn.ref.svn.c.move">svn move (mv)</a></dt><dt>patch, <a class="indexterm" href="#svn.ref.svn.c.patch">svn patch</a></dt><dt>propdel, <a class="indexterm" href="#svn.ref.svn.c.propdel">svn propdel (pdel, pd)</a></dt><dt>propedit, <a class="indexterm" href="#svn.ref.svn.c.propedit">svn propedit (pedit, pe)</a></dt><dt>propget, <a class="indexterm" href="#svn.ref.svn.c.propget">svn propget (pget, pg)</a></dt><dt>proplist, <a class="indexterm" href="#svn.ref.svn.c.proplist">svn proplist (plist, pl)</a></dt><dt>propset, <a class="indexterm" href="#svn.ref.svn.c.propset">svn propset (pset, ps)</a></dt><dt>relocate, <a class="indexterm" href="#svn.ref.svn.c.relocate">svn relocate</a></dt><dt>resolve, <a class="indexterm" href="#svn.ref.svn.c.resolve">svn resolve</a></dt><dt>resolved, <a class="indexterm" href="#svn.ref.svn.c.resolved">svn resolved</a></dt><dt>revert, <a class="indexterm" href="#svn.tour.cycle.revert">Fix Your Mistakes</a>, <a class="indexterm" href="#svn.ref.svn.c.revert">svn revert</a></dt><dt>status, <a class="indexterm" href="#svn.tour.cycle.examine.status">See an overview of your changes</a>, <a class="indexterm" href="#svn.ref.svn.c.status">svn status (stat, st)</a></dt><dt>switch, <a class="indexterm" href="#svn.ref.svn.c.switch">svn switch (sw)</a></dt><dt>unlock, <a class="indexterm" href="#svn.ref.svn.c.unlock">svn unlock</a></dt><dt>update, <a class="indexterm" href="#svn.basic.in-action.wc-funcdamentals">Fundamental working copy interactions</a>, <a class="indexterm" href="#svn.tour.cycle.update">Update Your Working Copy</a>, <a class="indexterm" href="#svn.ref.svn.c.update">svn update (up)</a></dt><dt>upgrade, <a class="indexterm" href="#svn.ref.svn.c.upgrade">svn upgrade</a></dt></dl></dd><dt>syntax</dt><dd><dl><dt>URLs, <a class="indexterm" href="#svn.advanced.reposurls">Addressing the Repository</a></dt></dl></dd></dl></dd><dt>svnadmin, <a class="indexterm" href="#svn.intro.components">Subversion's Components</a></dt><dd><dl><dt>subcommands</dt><dd><dl><dt>crashtest, <a class="indexterm" href="#svn.ref.svnadmin.c.crashtest">svnadmin crashtest</a></dt><dt>create, <a class="indexterm" href="#svn.ref.svnadmin.c.create">svnadmin create</a></dt><dt>deltify, <a class="indexterm" href="#svn.ref.svnadmin.c.deltify">svnadmin deltify</a></dt><dt>dump, <a class="indexterm" href="#svn.ref.svnadmin.c.dump">svnadmin dump</a></dt><dt>help, <a class="indexterm" href="#svn.ref.svnadmin.c.help">svnadmin help (h, ?)</a></dt><dt>hotcopy, <a class="indexterm" href="#svn.ref.svnadmin.c.hotcopy">svnadmin hotcopy</a></dt><dt>list-dblogs, <a class="indexterm" href="#svn.ref.svnadmin.c.list-dblogs">svnadmin list-dblogs</a></dt><dt>list-unused-dblogs, <a class="indexterm" href="#svn.ref.svnadmin.c.list-unused-dblogs">svnadmin list-unused-dblogs</a></dt><dt>load, <a class="indexterm" href="#svn.ref.svnadmin.c.load">svnadmin load</a></dt><dt>lslocks, <a class="indexterm" href="#svn.ref.svnadmin.c.lslocks">svnadmin lslocks</a></dt><dt>lstxns, <a class="indexterm" href="#svn.ref.svnadmin.c.lstxns">svnadmin lstxns</a></dt><dt>pack, <a class="indexterm" href="#svn.ref.svnadmin.c.pack">svnadmin pack</a></dt><dt>recover, <a class="indexterm" href="#svn.ref.svnadmin.c.recover">svnadmin recover</a></dt><dt>rmlocks, <a class="indexterm" href="#svn.ref.svnadmin.c.rmlocks">svnadmin rmlocks</a></dt><dt>rmtxns, <a class="indexterm" href="#svn.ref.svnadmin.c.rmtxns">svnadmin rmtxns</a></dt><dt>setlog, <a class="indexterm" href="#svn.ref.svnadmin.c.setlog">svnadmin setlog</a></dt><dt>setrevprop, <a class="indexterm" href="#svn.ref.svnadmin.c.setrevprop">svnadmin setrevprop</a></dt><dt>setuuid, <a class="indexterm" href="#svn.ref.svnadmin.c.setuuid">svnadmin setuuid</a></dt><dt>upgrade, <a class="indexterm" href="#svn.ref.svnadmin.c.upgrade">svnadmin upgrade</a></dt><dt>verify, <a class="indexterm" href="#svn.ref.svnadmin.c.verify">svnadmin verify</a></dt></dl></dd></dl></dd><dt>svndumpfilter, <a class="indexterm" href="#svn.intro.components">Subversion's Components</a></dt><dd><dl><dt>subcommands</dt><dd><dl><dt>exclude, <a class="indexterm" href="#svn.ref.svndumpfilter.commands.c.exclude">svndumpfilter exclude</a></dt><dt>help, <a class="indexterm" href="#svn.ref.svndumpfilter.commands.c.help">svndumpfilter help</a></dt><dt>include, <a class="indexterm" href="#svn.ref.svndumpfilter.commands.c.include">svndumpfilter include</a></dt></dl></dd></dl></dd><dt>svnlook, <a class="indexterm" href="#svn.intro.components">Subversion's Components</a></dt><dd><dl><dt>subcommands</dt><dd><dl><dt>author, <a class="indexterm" href="#svn.ref.svnlook.c.author">svnlook author</a></dt><dt>cat, <a class="indexterm" href="#svn.ref.svnlook.c.cat">svnlook cat</a></dt><dt>changed, <a class="indexterm" href="#svn.ref.svnlook.c.changed">svnlook changed</a></dt><dt>date, <a class="indexterm" href="#svn.ref.svnlook.c.date">svnlook date</a></dt><dt>diff, <a class="indexterm" href="#svn.ref.svnlook.c.diff">svnlook diff</a></dt><dt>dirs-changed, <a class="indexterm" href="#svn.ref.svnlook.c.dirs-changed">svnlook dirs-changed</a></dt><dt>filesize, <a class="indexterm" href="#svn.ref.svnlook.c.filesize">svnlook filesize</a></dt><dt>help, <a class="indexterm" href="#svn.ref.svnlook.c.help">svnlook help (h, ?)</a></dt><dt>history, <a class="indexterm" href="#svn.ref.svnlook.c.history">svnlook history</a></dt><dt>info, <a class="indexterm" href="#svn.ref.svnlook.c.info">svnlook info</a></dt><dt>lock, <a class="indexterm" href="#svn.ref.svnlook.c.lock">svnlook lock</a></dt><dt>log, <a class="indexterm" href="#svn.ref.svnlook.c.log">svnlook log</a></dt><dt>propget, <a class="indexterm" href="#svn.ref.svnlook.c.propget">svnlook propget (pget, pg)</a></dt><dt>proplist, <a class="indexterm" href="#svn.ref.svnlook.c.proplist">svnlook proplist (plist, pl)</a></dt><dt>tree, <a class="indexterm" href="#svn.ref.svnlook.c.tree">svnlook tree</a></dt><dt>uuid, <a class="indexterm" href="#svn.ref.svnlook.c.uuid">svnlook uuid</a></dt><dt>youngest, <a class="indexterm" href="#svn.ref.svnlook.c.youngest">svnlook youngest</a></dt></dl></dd></dl></dd><dt>svnrdump</dt><dd><dl><dt>subcommands</dt><dd><dl><dt>dump, <a class="indexterm" href="#svn.ref.svnrdump.c.dump">svnrdump dump</a></dt><dt>help, <a class="indexterm" href="#svn.ref.svnrdump.c.help">svnrdump help</a></dt><dt>load, <a class="indexterm" href="#svn.ref.svnrdump.c.load">svnrdump load</a></dt></dl></dd></dl></dd><dt>svnserve, <a class="indexterm" href="#svn.intro.components">Subversion's Components</a></dt><dt>svnsync, <a class="indexterm" href="#svn.intro.components">Subversion's Components</a></dt><dd><dl><dt>subcommands</dt><dd><dl><dt>copy-revprops, <a class="indexterm" href="#svn.ref.svnsync.c.copy-revprops">svnsync copy-revprops</a></dt><dt>help, <a class="indexterm" href="#svn.ref.svnsync.c.help">svnsync help</a></dt><dt>info, <a class="indexterm" href="#svn.ref.svnsync.c.info">svnsync info</a></dt><dt>initialize, <a class="indexterm" href="#svn.ref.svnsync.c.init">svnsync initialize (init)</a></dt><dt>synchronize, <a class="indexterm" href="#svn.ref.svnsync.c.sync">svnsync synchronize (sync)</a></dt></dl></dd><dt>syntax</dt><dd><dl><dt>URLs, <a class="indexterm" href="#svn.advanced.reposurls">Addressing the Repository</a></dt></dl></dd></dl></dd><dt>svnversion, <a class="indexterm" href="#svn.intro.components">Subversion's Components</a>, <a class="indexterm" href="#svn.ref.svnversion.re">svnversion</a></dt></dl></div><div class="indexdiv"><h3>T</h3><dl><dt>tags, <a class="indexterm" href="#svn.tour.importing.layout">Recommended Repository Layout</a>, <a class="indexterm" href="#svn.branchmerge.tags">Tags</a></dt><dt>text-base, <a class="indexterm" href="#svn.tour.cycle.examine">Review Your Changes</a></dt><dt>trunk, <a class="indexterm" href="#svn.tour.importing.layout">Recommended Repository Layout</a></dt></dl></div><div class="indexdiv"><h3>U</h3><dl><dt>unified diff, <a class="indexterm" href="#svn.tour.cycle.examine.diff">Examine the details of your local modifications</a></dt><dt>updating (see working copy, update)</dt></dl></div><div class="indexdiv"><h3>V</h3><dl><dt>VCS (see version control systems)</dt><dt>version control</dt><dd><dl><dt>models</dt><dd><dl><dt>copy-modify-merge, <a class="indexterm" href="#svn.basic.vsn-models.copy-merge">The copy-modify-merge solution</a></dt><dt>lock-modify-unlock, <a class="indexterm" href="#svn.basic.vsn-models.lock-unlock">The lock-modify-unlock solution</a></dt></dl></dd></dl></dd><dt>version control systems, <a class="indexterm" href="#svn.intro.whatis">What Is Subversion?</a>, <a class="indexterm" href="#svn.basic.version-control-basics">Version Control Basics</a></dt></dl></div><div class="indexdiv"><h3>W</h3><dl><dt>working copy</dt><dd><dl><dt>commit, <a class="indexterm" href="#svn.basic.in-action.wc-funcdamentals">Fundamental working copy interactions</a></dt><dt>creation, <a class="indexterm" href="#svn.basic.in-action.wc-funcdamentals">Fundamental working copy interactions</a></dt><dt>defined, <a class="indexterm" href="#svn.basic.working-copy">The Working Copy</a>, <a class="indexterm" href="#svn.basic.in-action.wc">Subversion Working Copies</a></dt><dt>mixed-revision, <a class="indexterm" href="#svn.basic.in-action.mixedrevs">Mixed-revision working copies</a></dt><dt>update, <a class="indexterm" href="#svn.basic.in-action.wc-funcdamentals">Fundamental working copy interactions</a></dt></dl></dd></dl></div></div></div></div></body></html>