svn_1.5_releasenotes.html [plain text]
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<style type="text/css"> /* <![CDATA[ */
@import "branding/css/tigris.css";
@import "branding/css/inst.css";
/* ]]> */</style>
<link rel="stylesheet" type="text/css" media="print"
href="branding/css/print.css"/>
<script type="text/javascript" src="branding/scripts/tigris.js"></script>
<title>Subversion 1.5 Release Notes</title>
</head>
<body>
<div class="app">
<h1 style="text-align: center">Subversion 1.5 Release Notes</h1>
<div class="h2" id="news" title="news">
<h2>What's New in Subversion 1.5</h2>
<ul>
<li><a href="#merge-tracking"
>Merge tracking (foundational)</a></li>
<li><a href="#sparse-checkouts"
>Sparse checkouts (via new <tt>--depth</tt> option)</a></li>
<li><a href="#interactive-conflict-resolution"
>Interactive conflict resolution</a></li>
<li><a href="#changelists"
>Changelist support</a></li>
<li><a href="#externals"
>Relative URLs, peg revisions in <tt>svn:externals</tt></a></li>
<li><a href="#cyrus-sasl"
>Cyrus SASL support for ra_svn and <tt>svnserve</tt></a></li>
<li><a href="#fsfs-sharding"
>Improved support for large deployments on FSFS, via sharding</a></li>
<li><a href="#fsfs-isolate-immutable"
>Improved FSFS optimizability, via immutable file isolation</a></li>
<li><a href="#webdav-proxy"
>WebDAV transparent write-through proxy</a></li>
<li><a href="#copy-move-improvements"
>Improvements to <tt>copy</tt> and <tt>move</tt></a></li>
<li><a href="#cancellation-improvements"
>Speed improvements, cancellation response improvements</a></li>
<li><a href="#dav-modules"
>Easier to try experimental <tt>ra_serf</tt> DAV access module</a></li>
<li><a href="#apis"
>API changes, improvements, and much language bindings work</a></li>
<li><a href="#bug-fixes"
>More than 150 new bug fixes, enhancements</a></li>
</ul>
<p>Subversion 1.5 is a superset of all previous Subversion releases,
and is considered the current "best" release. Any feature or bugfix
in 1.0.x through 1.4.x is also in 1.5, but 1.5 contains features and
bugfixes not present in any earlier release. The new features will
eventually be documented in a 1.5 version of the free Subversion book,
see <a href="http://svnbook.red-bean.com" >svnbook.red-bean.com</a>.</p>
</div> <!-- news -->
<div class="h2" id="compatibility" title="compatibility">
<h2>Compatibility Concerns</h2>
<p>Older clients and servers interoperate transparently with 1.5
servers and clients. However, some of the new 1.5 features (e.g., <a
href="#merge-tracking" >merge tracking</a>) may not be available
unless both client and server are the latest version . There are also
cases (e.g., <a href="#sparse-checkouts">sparse checkouts</a>) where a
new feature will work but will run less efficiently if the client is
new and the server old.</p>
<p>There is <strong>no need</strong> to dump and reload your
repositories. Subversion 1.5 can read repositories created by earlier
versions. To upgrade an existing installation, just install the
newest libraries and binaries on top of the older ones.</p>
<p>Subversion 1.5 maintains API/ABI compatibility with earlier
releases, by only adding new functions, never removing old ones. A
program written to the 1.0, 1.1, 1.2, 1.3, or 1.4 API can both compile
and run using 1.5 libraries. However, a program written for 1.5
cannot necessarily compile or run against older libraries.</p>
<div class="h3" id="new-feature-compatibility-table"
title="new-feature-compatibility-table">
<h3>New Feature Compatibility Table</h3>
<table border="1">
<tr>
<th>New Feature</th>
<th>Minimum Client</th>
<th>Minimum Server</th>
<th>Minimum Repository</th>
<th>Notes</th></tr>
<tr>
<td><a href="#merge-tracking">Merge tracking</a></td>
<td>1.5</td>
<td>1.5</td>
<td>1.5</td>
<td></td></tr>
<tr>
<td><a href="#sparse-checkouts">Sparse checkouts</a></td>
<td>1.5</td>
<td>any</td>
<td>any</td>
<td></td></tr>
<tr>
<td><a href="#interactive-conflict-resolution">Interactive conflict resolution</a></td>
<td>1.5</td>
<td>any</td>
<td>any</td>
<td></td></tr>
<tr>
<td><a href="#changelists">Changelist support</a></td>
<td>1.5</td>
<td>any</td>
<td>any</td>
<td></td></tr>
<tr>
<td><a href="#externals">Relative URLs, peg revisions in <tt>svn:externals</tt></a></td>
<td>1.5</td>
<td>any</td>
<td>any</td>
<td></td></tr>
<tr>
<td><a href="#cyrus-sasl"> Cyrus SASL support for ra_svn and <tt>svnserve</tt></a></td>
<td>any<sup>*</sup>, 1.5</td>
<td>any<sup>*</sup>, 1.5</td>
<td>any</td>
<td><sup>*</sup>Limited, see <a href="#sasl-compatibility">SASL and <code>svn://</code> compatibility</a></td></tr>
<tr>
<td><a href="#fsfs-sharding"> Improved support for large deployments on FSFS, via sharding</a></td>
<td>any<sup>*</sup>, 1.5</td>
<td>1.5</td>
<td>1.5</td>
<td><sup>*</sup>1.5 required for <code>file://</code> compatibility</td></tr>
<tr>
<td><a href="#fsfs-isolate-immutable">Improved FSFS optimizability, via immutable file isolation</a></td>
<td>any<sup>*</sup>, 1.5</td>
<td>1.5</td>
<td>1.5</td>
<td><sup>*</sup>1.5 required for <code>file://</code> compatibility</td></tr>
<tr>
<td><a href="#webdav-proxy">WebDAV transparent write-through proxy</a></td>
<td>any</td>
<td>1.5<sup>*</sup></td>
<td>any</td>
<td><sup>*</sup>Apache 2.2 server also required</td></tr>
<tr>
<td><a href="#copy-move-improvements">Improvements to <tt>copy</tt> and <tt>move</tt></a></td>
<td>1.5</td>
<td>any<sup>*</sup>, 1.5</td>
<td>any</td>
<td><sup>*</sup>Limited, see <a href="#copy-move-improvements">Copy/move<tt>-</tt>related improvements</a></td></tr>
<tr>
<td><a href="#cancellation-improvements">Speed improvements, cancellation response improvements</a></td>
<td>1.5</td>
<td>any</td>
<td>any</td>
<td></td></tr>
<tr>
<td><a href="#dav-modules">Easier to try experimental <tt>ra_serf</tt> DAV access module</a></td>
<td>1.5</td>
<td>any</td>
<td>any</td>
<td></td></tr>
</table>
</div> <!-- new-feature-compatibility-table -->
<div class="h3" id="wc-and-repos-format-change"
title="wc-and-repos-format-change">
<h3>Working Copy and Repository Format Changes</h3>
<p>The working copy format has been upgraded. This means that 1.4 and
older Subversion clients will <em>not</em> be able to work with
working copies produced by Subversion 1.5. Working copies are <a
href="#wc-upgrades" >upgraded automatically</a>.</p>
<p>Similarly, the repository format has changed, meaning that 1.4 and
older versions of Subversion tools that normally access a repository
directly (e.g. <tt>svnserve</tt>, <tt>mod_dav_svn</tt>,
<tt>svnadmin</tt>) won't be able to read a repository created by
Subversion 1.5. But, repositories are <a href="#repos-upgrades"
><strong>not</strong> upgraded automatically</a>.</p>
<div class="h4" id="wc-upgrades" title="wc-upgrades">
<h4>Working Copy Upgrades</h4>
<p><strong>WARNING:</strong> if a Subversion 1.5 client encounters a
pre-1.5 working copy, it will <em>automatically</em> upgrade the
working copy format as soon as it touches it, making it unreadable by
older Subversion clients. If you are using several versions of
Subversion on your machine, be careful about which version you use in
which working copy, to avoid accidentally upgrading a working copy.
(But note that this "auto upgrade" behavior does <em>not</em> occur
with the <a href="#repos-upgrades" >repositories</a>, only working
copies.)</p>
<p>If you accidentally upgrade a 1.4 working copy to 1.5, and wish to
downgrade back to 1.4, use the <a
href="http://svn.collab.net/repos/svn/trunk/tools/client-side/change-svn-wc-format.py"
><tt>change-svn-wc-format.py</tt></a> script. See <a
href="http://subversion.tigris.org/faq.html#working-copy-format-change"
>this FAQ entry</a> for details, and run the script with the
<code>--help</code> option for usage instructions.</p>
</div> <!-- wc-upgrades -->
<div class="h4" id="repos-upgrades" title="repos-upgrades">
<h4>Repository Upgrades</h4>
<p>The Subversion 1.5 server works with 1.4 and older repositories,
and it will <em>not</em> upgrade such repositories to 1.5 unless
specifically requested to via the
<strong><code>svnadmin upgrade</code></strong> command. This means
that some of the new 1.5 features will not become available simply by
upgrading your server: you will also have to upgrade your
repositories. (We decided not to auto-upgrade repositories because we
didn't want 1.5 to silently make repositories unusable by
1.4 — that step should be a conscious decision on the
part of the repository admin.)</p>
<p>After running <strong><code>svnadmin upgrade</code></strong>,
you may wish to also run the <a
href="http://svn.collab.net/repos/svn/trunk/tools/server-side/svn-populate-node-origins-index.c"
>svn-populate-node-origins-index</a> program on the repository.
Subversion 1.5 maintains a node-origins index for each repository, and
builds the index lazily as the information is needed. But for old
repositories with lots of revisions, it's better to create the index
in one step, using the aforementioned tool, than to have live queries
be slower until the index has built itself. See <a
href="http://subversion.tigris.org/issues/show_bug.cgi?id=3024" >issue
#3024</a> for details.</p>
</div> <!-- repos-upgrades -->
</div> <!-- wc-and-repos-format-change -->
<div class="h3" id="output-changes" title="output-changes">
<h3>Command Line Output Changes</h3>
<p>Although we try hard to keep output from the command line programs
compatible between releases, new information sometimes has to be
added. This can break scripts that rely on the exact format of the
output. Unfortunately, we are not able to enumerate all of the output
changes in 1.5, but one of them is that conflict markers in files now
match the file's defined eol-style.</p>
</div> <!-- output-changes -->
<div class="h3" id="sasl-compatibility" title="sasl-compatibility">
<h3>SASL and <code>svn://</code> compatibility</h3>
<p>All 1.x clients, with or without Cyrus SASL support, will be able to
authenticate against all 1.x servers that do not have Cyrus SASL enabled.
Note that the <tt>CRAM-MD5</tt> and <tt>ANONYMOUS</tt> mechanisms are
built into Subversion, so you'll be able to use them even if the
corresponding Cyrus SASL plugins are missing.</p>
<p>1.x clients without Cyrus SASL support will be able to authenticate
against 1.5+ servers with SASL enabled, provided the server allows the
<tt>CRAM-MD5</tt> and/or <tt>ANONYMOUS</tt> mechanisms.</p>
<p>1.5+ clients with Cyrus SASL support will be able to authenticate against
1.5+ servers with SASL enabled, provided at least one of the mechanisms
supported by the server is also supported by the client.</p>
<p>See <a href="#cyrus-sasl" >this section</a> for more information on
using Cyrus SASL with Subversion.</p>
</div> <!-- sasl-compatibility -->
<div class="h3" id="revprops-compatibility" title="revprops-compatibility">
<h3>Custom revprops may now need to be examined by <tt>pre-commit</tt>
hooks</h3>
<p>Each revision in Subversion has a set of <q>revision properties</q>
associated with it; in addition to the standard log message, author,
and date, arbitrary user-defined properties can be set. These
properties are not themselves versioned, so there is no way to revert
changes to them. Because of this, you cannot modify revprops on a
committed revision without setting up a <tt>pre-revprop-change</tt>
hook on the server. Before Subversion 1.5, only the three standard
revprops could be set at commit time; administrators using hooks to
control the contents of the three standard revprops needed to check
during <tt>pre-commit</tt> and <tt>pre-revprop-change</tt>, but for
custom revprops, only <tt>pre-revprop-change</tt> appeared
relevant.</p>
<p>Subversion 1.5 adds
a <a href="#commit-revprops"><tt>--with-revprop</tt> option to <tt>svn
commit</tt></a> which allows users to specify revprops for revisions
as they commit them. Thus, in Subversion 1.5, it is possible for
revproperties to be set on revisions without
any <tt>pre-revprop-change</tt>. The <tt>pre-commit</tt> hook can
examine the transaction's revision properties before it is committed
(just like many servers already use <tt>pre-commit</tt> to check for
properly formatted log messages, e.g.).</p>
<p>This means that if you previously relied
on <tt>pre-revprop-change</tt> to control the ability to write to
revprops (as opposed to just keeping track of their history), you may
want to add similar controls to <tt>pre-commit</tt>.</p>
</div> <!-- revprops-compatibility -->
</div> <!-- compatibility -->
<div class="h2" id="new-features" title="new-features">
<h2>New Features</h2>
<div class="h3" id="merge-tracking" title="merge-tracking">
<h3>Merge tracking (foundational) (<em>client and server</em>)</h3>
<p><em>Merge tracking</em> means Subversion keeps track of what
changes have been merged where. This reduces the overhead involved in
maintaining branches, and gives users a way to inquire what changes
are merged — or are available to be
merged — on different lines of development.</p>
<p>The merge tracking support in Subversion 1.5 is "foundational": its
basic functionality is implemented, but there are still parts of our
<a href="http://subversion.tigris.org/merge-tracking/" >original
spec</a> that remain to be done, and merging is sometimes too slow.
There will be merge tracking improvements in Subversion 1.5.1 and
afterwards. In the meantime:</p>
<ul>
<li><p>Take a look at <a href="#mt-known-issues"
>the list of known issues</a></p></li>
<li><p>Read <a
href="http://svnbook.red-bean.com/nightly/en/svn.branchmerge.advanced.html#svn.branchmerge.advanced.finalword"
>"The Final Word on Merge Tracking"</a> in the Subversion Book</p></li>
<li><p>Read <a
href="http://www.collab.net/community/subversion/articles/merge-info.html"
>"Understanding the internals of Subversion's merge tracking feature"</a>
by Paul Burba</p></li>
</ul>
<div class="h4" id="mt-overview" title="mt-overview">
<h4>Overview</h4>
<p>Subversion's merge tracking is designed to:</p>
<ul>
<li>Reduce the bookkeeping overhead for branch maintenance</li>
<li>Avoid common cases of the "repeated merge" problem</li>
<li>Allow for cherry-picking of changes</li>
</ul>
<p>Each changeset is identified by its revision number, and merged
changesets are recorded in the new
<strong><code>svn:mergeinfo</code></strong> property (known colloquially
as "mergeinfo") set on the destination of the merge. Subversion keeps
mergeinfo up-to-date automatically, but there is also a way to <a
href="#mt-record-only" >record/unrecord merges manually</a>, since
there will always be cases where a human knows something that
Subversion doesn't.</p>
</div> <!-- mt-overview -->
<div class="h4" id="mt-ui" title="mt-ui">
<h4>User interface</h4>
<p>Merging changes from (say) trunk to a branch no longer requires
that you specify the revision range. Instead, each time you want to
sync up with trunk, you can just do:</p>
<pre> $ cd BRANCH_WORKING_COPY
$ svn merge URL_TO_TRUNK</pre>
<p>Subversion will figure out what changes from URL_TO_TRUNK have not
yet been merged and pull in just those changes. When it's time to
merge the branch back to trunk, do this:</p>
<pre> $ cd TRUNK_WORKING_COPY
$ svn merge --reintegrate URL_TO_BRANCH</pre>
<p>Below is a more formal description of all the
merge-tracking-related interface changes.</p>
<p>The <strong><code>svn merge</code></strong> command takes two new
options: <strong><code>--record-only</code></strong> and
<strong><code>--reintegrate</code></strong>.</p>
<div id="mt-record-only" title="mt-record-only">
<p>The --record-only option works with -r and does exactly what you
think it does: it marks a revision as merged (or unmerged, if using
the "-" revision number negation syntax), without actually changing
anything besides the mergeinfo. For example, this would be useful
when someone has hand-edited a file in such a way as to effectively
incorporate a change that was already made somewhere else. Rather
than have the original change be ported over in the next
synchronization merge, thus risking textual conflicts wherever the two
versions trivially differ, you can just record the change as already
merged. (See <a
href="http://subversion.tigris.org/merge-tracking/requirements.html#manual-merge"
>the merge-tracking requirements</a>, and <a
href="http://svnbook.red-bean.com/nightly/en/svn.branchmerge.advanced.html#svn.branchmerge.advanced.blockchanges"
>Blocking Changes</a> in the Subversion book, for more details.)</p>
</div> <!-- mt-record-only -->
<div id="mt-reintegrate" title="mt-reintegrate">
<p>The --reintegrate option is used when merging a branch back to
trunk; it checks for some common safeguard conditions and then does
the merge in a fast and efficient way. See <a
href="http://svnbook.red-bean.com/nightly/en/svn.branchmerge.basicmerging.html#svn.branchemerge.basicmerging.stayinsync"
>Keeping a branch in sync</a> in the Subversion book for more.</p>
</div> <!-- mt-reintegrate -->
<p>The new <strong><code>svn mergeinfo</code></strong> command can
show which changesets a directory has absorbed and which changesets
it's still eligible to receive. See <a
href="http://svnbook.red-bean.com/nightly/en/svn.branchmerge.basicmerging.html#svn.branchmerge.basicmerging.mergeinfo"
>Mergeinfo and Previews</a> in the Subversion book for more information.</p>
<p>The <strong><code>svn log</code></strong> and
<strong><code>svn blame</code></strong> commands take a new
<strong><code>-g</code></strong>
(<strong><code>--use-merge-history</code></strong>) option, which tells
them to take mergeinfo into account. Without this option, they won't.</p>
<p>The reason for the -g option is that it is sometimes useful to
ignore merge history. In blame output you sometimes want to see the
person B who merged a change, though other times you want to see the
person A who originally wrote the change that B later merged; use -g
to get the latter information. In log output, you sometimes want to
see exactly the revisions that were committed on a given line of
development, though other times you want to see, as part of the same
output, the original changes that were later ported as merge
revisions; again, -g includes the latter information (tagged with
lines that say "<tt>Merged via: </tt>" followed by the revision number
in which the merge took place). See <a
href="http://svnbook.red-bean.com/nightly/en/svn.branchmerge.advanced.html#svn.branchmerge.advanced.logblame"
>Merge-Sensitive Logs and Annotations</a> in the Subversion book for
more details.</p>
</div> <!-- mt-ui -->
<div class="h4" id="mt-compatibility" title="mt-compatibility">
<h4>Merge tracking and compatibility</h4>
<p>As described earlier, merge-tracking is not supported unless you <a
href="#repos-upgrades" >upgrade the repository</a> as well as the
server.</p>
<p>If you were using the <a
href="http://svn.collab.net/repos/svn/trunk/contrib/client-side/svnmerge/"
>svnmerge</a> wrapper program to do merging, and now want to switch to
using Subversion 1.5's native merge-tracking, then you should use the
<a
href="http://svn.collab.net/repos/svn/trunk/contrib/client-side/svnmerge/svnmerge-migrate-history.py"
>svnmerge-migrate-history.py</a> script to convert svnmerge's custom
properties to the <tt>svn:mergeinfo</tt> properties that Subversion
itself now uses.</p>
</div> <!-- mt-compatibility -->
<div class="h4" id="mt-known-issues" title="mt-known-issues">
<h4>Known Issues</h4>
<p>There are still known issues with merge tracking in 1.5.0. We're
working on the most important ones first, which are:</p>
<ul>
<li><a href="http://subversion.tigris.org/issues/show_bug.cgi?id=3128"
>issue #3128</a>: <code>merge --reintegrate</code> should handle
renames better</li>
<li><a href="http://subversion.tigris.org/issues/show_bug.cgi?id=2897"
>issue #2897</a>: proper support for reflective merges</li>
<li><a href="http://subversion.tigris.org/issues/show_bug.cgi?id=3126"
>issue #3126</a>: better change-availability reporting
(<code>svn mergeinfo</code> shows too few or too many eligible
revisions)</li>
<li><a href="http://subversion.tigris.org/issues/show_bug.cgi?id=2837"
>issue #2837</a>: allow merge tracking to be useful when performing
cyclic merging</li>
<li><a href="http://subversion.tigris.org/issues/show_bug.cgi?id=3056"
>issue #3056</a>: avoid repeat merge in an intelligent way</li>
<li><a href="http://subversion.tigris.org/issues/show_bug.cgi?id=2898"
>issue #2898</a>: handle mergeinfo when deleting/adding children
you don't have access to</li>
<li><a href="http://subversion.tigris.org/issues/show_bug.cgi?id=3067"
>issue #3067</a>: subtrees that don't exist at the start or end of
a merge range shouldn't break the merge</li>
<li><a href="http://subversion.tigris.org/issues/show_bug.cgi?id=3157"
>issue #3157</a>: merging a change from a path's natural history
creates self-referential mergeinfo</li>
<li><a href="http://subversion.tigris.org/issues/show_bug.cgi?id=3174"
>issue #3174</a>: merge algorithm chokes on subtrees needing
special attention that have been renamed</li>
</ul>
<p>(See also the <a
href="http://subversion.tigris.org/issues/buglist.cgi?issue_status=UNCONFIRMED&issue_status=NEW&issue_status=STARTED&issue_status=REOPENED&keywords=merge-tracking&keywords_type=anytokens"
>list of merge-tracking-related open issues</a> and the <a
href="http://subversion.tigris.org/issues/buglist.cgi?issue_status=UNCONFIRMED&issue_status=NEW&issue_status=STARTED&issue_status=REOPENED&short_desc=merg&short_desc_type=substring"
>list of all merge-related open issues</a>, though note that not all
of merge-related issues are about merge-tracking and many predate it.)</p>
<p><a href="http://subversion.tigris.org/issues/show_bug.cgi?id=3157"
>Issue #3157</a> is particularly easy to run into, so here's a more
detailed explanation of what the issue is and how to work around
it:</p>
<p>Merge tracking information currently can include the "natural
history" of a path. One situation where this happens is when creating
a <tt>branch</tt> from <tt>trunk</tt> at a certain revision <tt>X</tt>
and later merging all changes from <tt>branch</tt> back to
<tt>trunk</tt>, without explicitly limiting the merge range and
without using the <a href="#mt-reintegrate"
><code>--reintegrate</code> option</a>. Afterwards, the merge
tracking information on <tt>trunk</tt> will list revisions before
<tt>X</tt>. This does no harm to future merge operations, but
thereafter <code>svn log -g</code> on <tt>trunk</tt> will report these
revisions from the "natural history", which is typically not expected.
This is because <code>svn log -g</code> uses the merge tracking
information from previous revisions (not only from HEAD).</p>
<p>Note that this should not be a problem with release branches (also
called "maintenance branches"), because one typically does not merge
them back to trunk. Instead, recommended practice is for new changes
to enter on trunk and be ported outward to the branches that need the
changes. However, you could run into this problem with a feature
branch, which would be merged back to trunk once, at the end of the
branch's lifetime.</p>
<p>One workaround is to correct the merge tracking information before
committing the merged revision, by reverse merging the "natural
history" information: <code>svn merge --record-only -rX:1</code>.
Another solution is to avoid this situation entirely, by explicitly
specifying merge ranges in the first place, e.g.: <tt>-rX:HEAD</tt>.
See <a
href="http://subversion.tigris.org/servlets/ReadMsg?listName=dev&msgNo=139379"
>this mail</a> for more information.</p>
<p>When merging a feature branch, the best solution is to <a
href="#mt-reintegrate" >use the <code>--reintegrate</code> option</a>;
then this problem won't happen in the first place.</p>
</div> <!-- mt-known-issues -->
<div class="h4" id="mt-further-reading" title="mt-further-reading">
<h4>Further reading</h4>
<p>For more on merge-tracking in Subversion 1.5, see:</p>
<ul>
<li><a href="http://svnbook.red-bean.com/nightly/en/svn.branchmerge.html"
>Chapter 4: Branching and Merging</a> in the Subversion book
</li>
<li><a href="http://blog.red-bean.com/sussman/?p=92" >Subversion 1.5
merge-tracking in a nutshell</a> (blog post by Ben Collins-Sussman)
</li>
<li><a href="http://subversion.tigris.org/merge-tracking/" >The merge
tracking design documents</a>
(requirements, use cases, functional spec, design)
</li>
</ul>
</div> <!-- mt-further-reading -->
</div> <!-- merge-tracking -->
<div class="h3" id="sparse-checkouts" title="sparse-checkouts">
<h3>Sparse checkouts (<em>client and server</em>)</h3>
<p>Many users have very large trees of which they only want to
checkout certain parts. In previous versions of Subversion,
<code>checkout -N</code> was not really up to this task. Subversion
1.5 introduces the <strong><code>--depth</code></strong> option to the
<strong><code>checkout</code></strong>,
<strong><code>switch</code></strong>, and
<strong><code>update</code></strong> subcommands. This option
replaces <code>-N</code>, and allows users to construct working
copies containing just what's needed, leaving out everything else.</p>
<div class="h4" id="sc-depth" title="sc-depth">
<h4>Depth overview</h4>
<p>Each directory now understands the notion of <em>depth</em>, which has
four possible values: <tt>empty</tt>, <tt>files</tt>, <tt>immediates</tt>,
<tt>infinity</tt>. The values are defined as follows:</p>
<table border="1">
<tr>
<td><tt>empty</tt></td>
<td>Updates will not pull in any files or subdirectories not already
present.</td>
</tr>
<tr>
<td><tt>files</tt></td>
<td>Updates will pull in any files not already present, but not
subdirectories.</td>
</tr>
<tr>
<td><tt>immediates</tt></td>
<td>Updates will pull in any files or subdirectories not already present;
the new subdirectories will have depth-empty.</td>
</tr>
<tr>
<td><tt>infinity</tt></td>
<td>Updates will pull in any files or subdirectories not already present;
the new subdirectories will have depth-infinity. Equivalent to today's
default update behavior.</td>
</tr>
</table>
<p>The <code>--depth</code> option sets depth values as it updates the working
copy, tweaking new subdirectories' depth values as described above.</p>
</div> <!-- sc-depth -->
<div class="h4" id="sc-ui" title="sc-ui">
<h4>User Interface</h4>
<p>Affected commands:</p>
<ul>
<li><code>checkout</code></li>
<li><code>switch</code></li>
<li><code>update</code></li>
<li><code>status</code></li>
<li><code>info</code></li>
</ul>
<p>The <code>-N</code> option becomes a synonym for <code>--depth=files</code>
for these commands. This changes the existing <code>-N</code> behavior for
these commands, but in a trivial way (see below).</p>
<p><code>checkout</code> without <code>--depth</code> or
<code>-N</code> behaves the same as it does today.
<code>switch</code> and <code>update</code> without
<code>--depth</code> or <code>-N</code> behave the same way as today
if and only if the working copy is fully depth-infinity.
<code>switch</code> and <code>update</code> without
<code>--depth</code> or <code>-N</code> will <em>not</em> change depth
values (exception: a missing directory specified on the command line
will be pulled in).</p>
<p>Thus, <code>checkout</code> is identical to <code>checkout
--depth=infinity</code>, but <code>switch</code> and <code>update</code> are
not the same as <code>switch --depth=infinity</code> and <code>update
--depth=infinity</code>. The former update entries according to existing depth
values, while the latter pull in everything.</p>
<p>To get started, run <code>checkout</code> with
<code>--depth=empty</code> or <code>--depth=files</code>. If
additional files or directories are desired, pull them in with
<code>update</code> commands using appropriate <code>--depth</code>
options. The <code>svn status</code> command should list the ambient
depths of directories, in addition to whatever other statuses are
listed. The <code>svn info</code> command also lists ambient depth,
when invoked on a directory whose depth is not the default
(depth-infinity)</p>
<p><strong>Note:</strong> There is currently no <em>deselection</em>
interface for sparse checkouts — that is, there is no
command to unselect or "fold up" a subdirectory after you've brought
it into a sparse working copy (although there are some fairly easy
workarounds to achieve the same effect). See <a
href="http://subversion.tigris.org/issues/show_bug.cgi?id=2843" >issue
#2843</a> for details.</p>
</div> <!-- sc-ui -->
<div class="h4" id="sc-compat" title="sc-compat">
<h4>Compatibility with older servers</h4>
<p>The new <tt>--depth</tt> feature naturally requires the client to
be 1.5+, and will work most efficiently if the server is also 1.5+.
However, the client will still behave correctly if the server is 1.4.x
or lower; things will just be less efficient.</p>
<p>This is because older servers do not
understand — and therefore
ignore — what the client tells them about "depth". So
when a client requests a depth shallower than depth-infinity, older
servers will send back more data than the client wants. However, a
1.5+ client will know it's talking to an older server and filter out
this extra data. Thus, operations may take a while, because the
server sends a lot of data over the network that the client then
ignores, but the final result on the client side will be the same.
(Note that older servers understand a <tt>recurse</tt> flag in the
network protocols, and 1.5+ clients send that flag based on the depth;
this alleviates some of the extra network traffic penalty.)</p>
</div> <!-- sc-compat -->
<div class="h4" id="sc-further-reading" title="sc-further-reading">
<h4>Further reading</h4>
<ul>
<li>The <a
href="http://svnbook.red-bean.com/nightly/en/svn.advanced.sparsedirs.html"
>Sparse Directories</a> section in the Subversion book.
</li>
<li>The <a
href="http://svn.collab.net/repos/svn/trunk/notes/sparse-directories.txt"
>design/implementation document</a> in the repository.</li>
</ul>
</div> <!-- sc-further-reading -->
</div> <!-- sparse-checkouts -->
<div class="h3" id="interactive-conflict-resolution"
title="interactive-conflict-resolution">
<h3>Interactive Conflict Resolution (<em>client</em>)</h3>
<p>Conflict resolution is now done interactively by the command-line
client for the
<code>update</code>/<code>switch</code>/<code>merge</code>
subcommands, and the client library offers a callback function so
other clients can do similarly.</p>
<p>Here's an example using the command-line client:</p>
<pre>
$ svn up
U contrib/client-side/svnmerge/svnmerge_test.py
Conflict discovered in 'contrib/client-side/svnmerge/svnmerge.py'.
Select: (p) postpone, (df) diff-full, (e) edit,
(s) show all options: s
(p) postpone - mark the conflict to be resolved later
(df) diff-full - show all changes made to merged file
(e) edit - change merged file in an editor
(r) resolved - accept merged version of file
(mf) mine-full - accept my version of entire file (ignore their changes)
(tf) theirs-full - accept their version of entire file (lose my changes)
(l) launch - launch external tool to resolve conflict
(s) show all - show this list
Select: (p) postpone, (df) diff-full, (e) edit,
(s) show all options: tf
G contrib/client-side/svnmerge/svnmerge.py
Updated to revision 25685.
$
</pre>
<p>This feature can be selectively disabled by using the --non-interactive
option, or disabled permanently by setting '[miscellany] interactive-conflicts
= no' in your run-time config file.</p>
<p>The API for interactive conflict resolution is exposed via a
callback function and the following new data types:</p>
<ul>
<li><code>svn_wc_conflict_resolver_func_t</code>, the callback API
itself</li>
<li><code>svn_wc_conflict_description_t</code>, a description of the
conflict passed to the callback</li>
<li><code>svn_wc_conflict_action_t</code>, the part of the conflict
description indicating what the merge was trying to do</li>
<li><code>svn_wc_conflict_reason_t</code>, the part of the conflict
description indicating the type of conflict</li>
<li><code>svn_wc_conflict_result_t</code>, returned by the callback
as the result of any conflict resolution attempt</li>
<li><code>svn_wc_conflict_choice_t</code>, an enum indicating what
course of action the user chose</li>
</ul>
<p>Clients provide their callback function to Subversion's libraries
by setting it on the (new) <code>conflict_func</code> field of their
<code>svn_client_ctx_t</code>, and may provide additional state to the
callback via the corresponding <code>conflict_baton</code> field.</p>
</div> <!-- interactive-conflict-resolution -->
<div class="h3" id="changelists" title="changelists"> <h3>Changelist
support (<em>client</em>)</h3> <p>The Subversion client now contains
the notion of a <em>changelist</em>: a group of files which are
associated with a chosen name. This becomes especially useful when
working on several different set of files within the same working
copy. Instead of having to remember each file in each set, Subversion
1.5 will allow you to associate a changelist with each set of files.
Most commands which take a set of files as targets will now also
accept the <strong><code>--changelist</code></strong> option, which
filters those targets based upon the members of the changelist.
Changelist membership can be edited using the new
<strong><code>changelist</code></strong> subcommand.</p>
<p>Changelists are handled entirely by the client. They are never sent
to the server, and aren't visible to other users of the same repository.
Also, the <code>--changelist</code> option is never additive; if a file
wouldn't have been included in the list of targets without
<code>--changelist</code>, it will not be added to it, regardless of membership
in the changelist. Currently, a file may only be in one changelist at a
time, and directories can not be members of a changelist.</p>
<p>The <code>--changelist</code> option is supported by the following
subcommands:</p>
<ul>
<li><code>changelist</code></li>
<li><code>commit</code></li>
<li><code>diff</code> (only wc-wc and wc-repos cases)</li>
<li><code>info</code></li>
<li><code>propget</code></li>
<li><code>proplist</code></li>
<li><code>propset</code></li>
<li><code>propdel</code></li>
<li><code>revert</code></li>
<li><code>status</code></li>
<li><code>update</code></li>
</ul>
<div class="h4" id="cl-further-reading" title="cl-further-reading">
<h4>Further reading</h4>
<ul>
<li>The <a
href="http://svnbook.red-bean.com/nightly/en/svn.advanced.changelists.html"
>Changelists</a> section in the Subversion book.
</li>
<li>The <a
href="http://svn.collab.net/repos/svn/trunk/notes/changelist-design.txt"
>design/implementation document</a> in the repository.</li>
</ul>
</div> <!-- cl-further-reading -->
</div> <!-- changelists -->
<div class="h3" id="externals" title="externals">
<h3>Relative URLs, peg revisions in <tt>svn:externals</tt>
(<em>client</em>)</h3>
<p>Two additions to the <tt>svn:externals</tt> feature</p>
<div class="h4" id="externals-peg-revs" title="externals-peg-revs">
<h4>Peg revision syntax for URLs</h4>
<p><em>Compatibility and the new syntax:</em> For compatibility
reasons, the pre-1.5 svn:externals syntax continues to not understand
peg revisions. A new format has been introduced to allow peg
revisions in URLs.</p>
<p>The old format of</p>
<pre>
foo http://example.com/repos/zig
foo/bar -r 1234 http://example.com/repos/zag
</pre>
<p>does not support peg revisions, and the following externals will
not work (unless there are directories named <tt>zig@HEAD</tt> and
<tt>zag@HEAD</tt>):</p>
<pre>
foo http://example.com/repos/zig@HEAD
foo/bar -r 1234 http://example.com/repos/zag@HEAD
</pre>
<p>The new format moves the URL first followed by the directory the
external is checked out or exported into; if there is an operative
(-r) revision, it precedes the URL. Here are four externals lines:</p>
<pre>
http://example.com/repos/zig foo1
-r 1234 http://example.com/repos/zag foo/bar1
http://example.com/repos/zig@HEAD foo2
-r 1234 http://example.com/repos/zag@HEAD foo/bar2
</pre>
<p>In other words, both operative ("-r") and peg ("@") revisions are
allowed, but neither is required.</p>
</div> <!-- externals-peg-revisions -->
<div class="h4" id="externals-relative-urls" title="externals-relative-urls">
<h4>Support for relative URLs</h4>
<p>Prior to Subversion 1.5, the URLs in an <tt>svn:externals</tt>
specification must be absolute. Now they can be relative. Four
different relative externals specifications are supported. (In the
following examples, assume we have two repositories: one at
<tt>http://example.com/svn/repos-1</tt> and another at
<tt>http://example.com/svn/repos-2</tt>. We have a checkout of
<tt>http://example.com/svn/repos-1/project1/trunk</tt> and the
<tt>svn:externals</tt> property is set on <tt>trunk</tt>.)</p>
<dl><dt><strong>../</strong></dt>
<dd><p>Relative to the directory with the <tt>svn:externals</tt>
property. These URLs always begin with the string <tt>../</tt>, for
example:</p>
<pre>
../../project2/trunk common/project2/trunk
</pre>
<p>This will extract
<tt>http://example.com/svn/repos-1/project2/trunk</tt> into
<tt>common/project2/trunk</tt>. The external's URL is relative to the
URL of the directory with the <tt>svn:externals</tt> property, not the
directory where the external is written to disk.</p></dd>
<dt><strong>^/</strong></dt>
<dd><p>Relative to the repository root.</p>
<pre>
^/project2/trunk common/project2/trunk
</pre>
<p>This will extract
<tt>http://example.com/svn/repos-1/project2/trunk</tt> into
<tt>common/project2/trunk</tt>.</p>
<p>You can also refer to other repositories easily using repository
root relative URLs:</p>
<pre>
^/../repos-2/foo/trunk common/foo/trunk
</pre>
<p>This will extract <tt>http://example.com/svn/repos-2/foo/trunk</tt>
into <tt>common/foo/trunk</tt>.</p>
</dd>
<dt><strong>//</strong></dt>
<dd><p>Relative to the scheme. This copies the scheme of the checkout
or export URL into the URL in <tt>svn:externals</tt>. It is useful
when the same hostname must the accessed with different schemes
depending upon network location; i.e. clients in the intranet use
<tt>http://</tt> while external clients use <tt>svn+ssh://</tt>.</p>
<pre>
//example.com/svn/repos-1/project2/trunk common/project2/trunk
</pre>
<p>This will extract
<tt>http://example.com/svn/repos-1/project2/trunk</tt> into
<tt>common/project2/trunk</tt>. If the working copy was checked out
from <tt>svn+ssh://example.com/svn/repos-1/project1/trunk</tt> then
this URL would be
<tt>svn+ssh://example.com/svn/repos-1/project1/trunk</tt>.</p>
</dd>
<dt><strong>/</strong></dt>
<dd><p>Server root relative URLs. This copies the scheme and hostname
from the checkout or export URL into the <tt>svn:externals</tt>
URL.</p>
<pre>
/svn/repos-1/project2/trunk common/project2/trunk
</pre>
<p>This will extract
<tt>http://example.com/svn/repos-1/project2/trunk</tt> into
<tt>common/project2/trunk</tt>. If the working copy was checked out
from <tt>svn+ssh://example.com/svn/repos-1/project1/trunk</tt> then
this URL would be
<tt>svn+ssh://example.com/svn/repos-1/project1/trunk</tt>.</p>
</dd>
</dl>
<p>Relative URLs are still supported in the old <tt>svn:externals</tt>
format (that does not support peg revisions).</p>
<p>When Subversion sees an <tt>svn:externals</tt> without an absolute
URL, it takes the first argument as a relative URL and the second as
the target directory.</p>
</div> <!-- externals-relative-urls -->
<div class="h4" id="externals-further-reading"
title="externals-further-reading">
<h4>Further reading</h4>
<p>See The <a
href="http://svnbook.red-bean.com/nightly/en/svn.advanced.externals.html"
>svn:externals</a> section of the Subversion Book.</p>
</div> <!-- externals-further-reading -->
</div> <!-- externals -->
<div class="h3" id="cyrus-sasl" title="cyrus-sasl">
<h3>Cyrus SASL support for ra_svn and <tt>svnserve</tt> (<em>client
and server</em>)</h3>
<p>From <a
href="http://en.wikipedia.org/wiki/Simple_Authentication_and_Security_Layer">
Wikipedia</a>: "SASL is a framework for authentication and data security in
Internet protocols. It decouples authentication mechanisms from application
protocols, in theory allowing any authentication mechanism supported by SASL to
be used in any application protocol that uses SASL."</p>
<p>In practice, the server sends a list of authentication mechanisms that it
supports. The client then selects one of these mechanisms based on what the
client supports, and informs the server of its decision. After that, a
number of messages are exchanged until either authentication succeeds or an
error occurs. In the latter case, the client is allowed to restart
authentication.</p>
<p>The <tt>svn://</tt> protocol has always supported this type of negotiation.
However, only the <tt>CRAM-MD5</tt> and <tt>ANONYMOUS</tt> mechanisms were
implemented. <a href="http://asg.web.cmu.edu/sasl/">Cyrus SASL</a> supports
all these, and, in addition, provides a host of other mechanisms such as
<tt>DIGEST-MD5</tt>, OTP (One-Time Passwords), GSSAPI (used for Kerberos
authentication), NTLM (NT LAN Manager), SRP (Secure Remote Password), and
others. The exact list of available mechanisms depends on how SASL was
compiled, as many of them either have external dependencies, or are not
built by default. Also, because each mechanism is actually a shared library
that is dynamically loaded at runtime, many distributions package these
mechanisms separately from the core library.</p>
<p>Please see the <a href="#sasl-compatibility">compatibility
section</a> for information regarding using a 1.5 SASL-enabled server
with pre-1.5 clients. More information about Subversion's SASL
support can be found in <a
href="http://svn.collab.net/repos/svn/trunk/notes/sasl.txt"
>sasl.txt</a>.</p>
</div> <!-- cyrus-sasl -->
<div class="h3" id="fsfs-sharding" title="fsfs-sharding">
<h3>Improved support for large deployments on FSFS, via sharding
(<em>server</em>)</h3>
<p>The FSFS filesystem backend stores each revision in its own file, and prior
to Subversion 1.5, all of these files were stored in a common directory in
the repository. Now, <em>newly created</em> FSFS repositories will use a
two-level directory tree with up to (by default) 1000 files per directory.
These repositories will only be compatible with other Subversion 1.5 clients,
but of course, Subversion 1.5 will be able to continue using repositories
employing the older scheme without any problem.</p>
<p>The primary reason for the change is to allow the revision count to
grow beyond the filesystem's (efficient) directory entry limit. While
modern filesystems can support millions of entries per directory, they
become slower and common administrative tools (e.g. directory
listings, backups) become unwieldy or fail completely.</p>
<p>For more information about the technical underpinnings of FSFS sharding,
see this <a href="http://www.farside.org.uk/200704/tree_structured_fsfs">blog
entry</a>.
<!-- ### Reshard script is not quite finished and currently
### aborts if you try to run it.
<a
href="http://svn.collab.net/repos/svn/trunk/tools/server-side/fsfs-reshard.py"
>Migration script</a> provided.
-->
</p>
<p>The shard size can by adjusted by editing the
"<code>layout sharded</code>" line in "db/format" after
<code>svnadmin create</code> but before populating the
repository.</p>
</div> <!-- fsfs-sharding -->
<div class="h3" id="fsfs-isolate-immutable" title="fsfs-isolate-immutable">
<h3>Improved FSFS optimizability, via immutable file isolation
(<em>server</em>)</h3>
<p>The FSFS repositories never change a revision after it is written
to the disk. Although this should allow the operating system to cache
these files forever, certain filesystems (e.g. NFS) prohibit such
caching by default. Now the FSFS repository layout has been changed
such that the immutable files are confined to the subdirectories
"db/revs" and "db/txn-protorevs". This allows these directories to be
on a mount point where caching is enabled (on Linux look at the
"nocto" option to nfs(5)).</p>
<p>Since commit transactions are now built up in "db/txn-protorevs"
instead of "db/transactions", the latter directory no longer needs to
be on the same mount point. If the repository is stored on a slower
filesystem (e.g. NFS), then commit performance can be improved by
moving the transactions directory to local disk (using a symbolic
link). If you are using multiple svn servers behind a network load
scaler, you must configure the load scaler to direct to the same
server for the duration of the transaction. This is typically called
"client affinity".</p>
</div> <!-- fsfs-isolate-immutable -->
<div class="h3" id="webdav-proxy" title="webdav-proxy">
<h3>WebDAV transparent write-through proxy (<em>server</em>)</h3>
<p>Subversion 1.4 <a
href="http://subversion.tigris.org/svn_1.4_releasenotes.html#svnsync">
introduced <tt>svnsync</tt></a> — a tool which
provided the ability to replicate repository history from one
repository to another. Though useful, <tt>svnsync</tt> could only
pull revision history from a repository, not push additional commits
back to the master. Subversion 1.5 adds WebDAV proxy support to
<tt>mod_dav_svn</tt>, effectively allowing bidirectional revision
history replication between master servers and slave servers using
<tt>mod_dav_svn</tt>.</p>
<p>All clients interact with a slave server, but the slave transparently
passes all of the write-oriented activites to the master (rewriting the
content as necessary). The slaves are essentially read-only, but they
do have a complete copy of the repository locally. This serves to
alleviate read traffic from the master server which may be desirable
in certain circumstances.</p>
<p>This model has several advantages to using a straight HTTP DAV-aware
caching proxy, in that each slave can respond to all read-only requests
without ever having to relay them to the master backend.</p>
<div class="h4" id="webdav-proxy-requirements"
title="webdav-proxy-requirements">
<h4>Requirements</h4>
<ul>
<li>Subversion 1.5 or newer.</li>
<li>Apache HTTP Server 2.2.0 or higher with mod_proxy enabled.
(Several fixes to mod_proxy were committed for this patch that were not
backported to the 2.0 branch of httpd.)</li>
</ul>
</div> <!-- webdav-proxy-requirements -->
<div class="h4" id="webdav-proxy-example" title="webdav-proxy-example">
<h4>Example configuration</h4>
<p>Participants:</p>
<ul>
<li>Slave → <tt>slave.example.com</tt> (there can be many!)</li>
<li>Master → <tt>master.example.com</tt> (there can only be one!)</li>
<li>A WebDAV client (<tt>ra_neon</tt>, <tt>ra_serf</tt>, other WebDAV
clients)</li>
</ul>
<p>Each client does:</p>
<pre><code>
% svn co http://slave.example.com/repos/slave/
...
% svn ci
% ...etc...
</code></pre>
<p>(The client can perform all operations as normal.)</p>
<p>Each slave has:</p>
<pre><code>
<Location /repos/slave>
DAV svn
SVNPath /my/local/copy/of/repos
SVNMasterURI http://master.example.com/repos/master/
</Location>
</code></pre>
<p>The master MUST have a post-commit hook that updates all of the slaves. An
example that does this using <code>svnadmin dump</code>/<code>svnadmin
load</code> and ssh is provided below. <tt>svnsync</tt> can probably do the
same thing.</p>
<p>Additionally, if locks are permitted on the master repository, lock
databases need to kept in sync via post-lock and post-unlock hooks on
the master pushing the lock state to the slaves. (Username
preservation is left as an exercise to the reader. Translation: <a
href="hacking.html#patches">patches to these notes are most
welcome</a>.) If the lock database is not propagated, users will not
be able to accurately determine whether a lock is
held — but locking will still work.</p>
<p>A sample synchronization script may look like this:</p>
<pre><code>
#!/bin/sh
REPOS="$1"
REV="$2"
SLAVE_HOST=slave.example.com
SLAVE_PATH=/my/local/copy/of/repos
# Ensure svnadmin is in $PATH on both this machine and the remote server!
svnadmin dump --incremental -r$2 $1 > /tmp/$2.dump
scp /tmp/$2.dump $SLAVE_HOST:$SLAVE_PATH
ssh $SLAVE_HOST "svnadmin load $SLAVE_PATH < $SLAVE_PATH/$2.dump"
ssh $SLAVE_HOST "rm $SLAVE_PATH/$2.dump"
rm /tmp/$2.dump
</code></pre>
</div> <!-- webdav-proxy-example -->
<div class="h4" id="webdav-proxy-further-reading"
title="webdav-proxy-further-reading">
<h4>Further reading</h4>
<p>Additional information about WebDAV proxy support is available <a
href="http://svn.collab.net/repos/svn/trunk/notes/webdav-proxy">in the
repository</a>.</p>
</div> <!-- webdav-proxy-further-reading -->
</div> <!-- webdav-proxy -->
</div> <!-- new-features -->
<div class="h2" id="enhancements" title="enhancements">
<h2>Enhancements and Bugfixes</h2>
<div class="h3" id="copy-move-improvements" title="copy-move-improvements">
<h3>Copy/move-related improvements (<em>client and server</em>)</h3>
<p>The abilities and behavior of <code>copy</code> and
<code>move</code> operations are significantly improved in 1.5+.</p>
<div class="h4" id="leverage-copyfrom" title="leverage-copyfrom">
<h4>Improved copy-handling during updates (<em>client and server</em>)</h4>
<p>A common problem in older versions of Subversion was the way in
which <code>svn update</code> handled incoming copies and moves.</p>
<p>Consider this scenario: Harry runs <code>svn move foo bar; svn
commit</code>, and meanwhile Sally makes local changes to 'foo', and
then runs <code>svn update</code>. In earlier versions of Subversion,
the server would send down a completely new file 'bar', and unversion
the file 'foo' (if it had no uncommitted changes, Subversion would
remove it entirely.) From Sally's point of view, her changes seem to
be lost; the newly added 'bar' file has the older content, and the
file 'foo' has been taken out of version control.</p>
<p>In Subversion 1.5, the client and server both attempt to be smarter
about this. The server doesn't send a whole new file during the
update, but rather instructions to copy something that likely already
exists in the working copy. So Sally's 'foo' file is copied to 'bar'
(with local edits intact!).</p>
<p>In theory, this is the best-case scenario. There are a few
caveats: this "proper copying" of existing working-copy resources only
works on files, not (yet) on directories. Also, if an incoming
move-operation deletes 'foo' before it attempts to copy it to 'bar',
then the copy will fail, and the client reverts to the old behavior of
fetching a pristine copy of the file from the repository. We hope to
address this in svn 1.6.</p>
<p>See <a
href="http://subversion.tigris.org/issues/show_bug.cgi?id=503" >issue
#503</a> for more.</p>
</div> <!-- leverage-copyfrom -->
<div class="h4" id="copy-peg-revs" title="copy-peg-revs">
<h4>Peg revisions (<em>client</em>)</h4>
<p>Copy and move operations now accept sources with peg ("@")
revisions.</p>
<p>See <a
href="http://subversion.tigris.org/issues/show_bug.cgi?id=2546" >issue
#2546</a> for more.</p>
</div> <!-- copy-peg-revs -->
<div class="h4" id="multi-op-copy-move" title="multi-op-copy-move">
<h4>Multiple working copy copy/move operations (<em>client</em>)</h4>
<p>Clients may now perform chained copy/move operations locally on a
single object in a working copy:</p>
<pre>
svn mv path1 path2
svn mv path2 path3
</pre>
<p>See <a
href="http://subversion.tigris.org/issues/show_bug.cgi?id=756" >issue
#756</a> for more.</p>
</div> <!-- multi-op-copy-move -->
<div class="h4" id="multi-src-copy" title="multi-src-copy">
<h4>Improved handling multiple copy/move sources (<em>client</em>)</h4>
<p>Clients now accept multiple sources for copy and move operations, with
the ability to copy/move each of the sources to the specified directory.
This mirrors the behavior of standard command-line copy and move tools,
such as <code>cp</code> and <code>mv</code>. For example:</p>
<pre>
svn mkdir new_subdir
svn mv foo.txt bar.txt baz.txt new_subdir
</pre>
<p>In practice, this means users can take advantage of shell globbing
when doing a local copy or move:</p>
<pre>
svn cp *.c dir
</pre>
<p>Multiple source copy/move also works for all previously defined
copy/move working copy and repository combinations.</p>
<p>See <a
href="http://subversion.tigris.org/issues/show_bug.cgi?id=747" >issue
#747</a> for more.</p>
</div> <!-- multi-src-copy -->
<div class="h4" id="cp-takes-base" title="cp-takes-base">
<h4>Copy takes -rBASE (<em>client</em>)</h4>
<p>Copy now understands the special revision "BASE" in a working copy
(as in: "<tt>-rBASE</tt>").</p>
<p>See <a
href="http://subversion.tigris.org/issues/show_bug.cgi?id=1643" >issue
#1643</a> for more.</p>
</div> <!-- cp-takes-base -->
<div class="h4" id="cp-mv-intermediate-dirs" title="cp-mv-intermediate-dirs">
<h4>Creation of intermediate directories with copy/move
(<em>client and server</em>)</h4>
<p>See the section on the new <a href="#intermediate-directories"
>--parents option</a> for more about this.</p>
</div> <!-- cp-mv-intermediate-dirs -->
</div> <!-- copy-move-improvements -->
<div class="h3" id="cancellation-improvements"
title="cancellation-improvements">
<h3>Cancellation improvements (<em>client</em>)</h3>
<p>Clients operations are now significantly more responsive to
cancellation (e.g. via <tt>control-c</tt>). In pre-1.5 releases,
after directing an operation to stop, one sometimes had to wait for
some time (e.g. while I/O occurred) before the operation would
actually stop.</p>
</div> <!-- cancellation-improvements -->
<div class="h3" id="cmdline" title="cmdline">
<h3>Command-line client improvements (<em>client</em>)</h3>
<p>There are far too many enhancements and new options to the
command-line client to list them all here. Aside from all the ones
mentioned already in these release notes, below are a few more that we
consider important, but please see the 1.5.0 section in the <a
href="http://svn.collab.net/repos/svn/trunk/CHANGES">CHANGES</a> file
for a complete list.</p>
<div class="h4" id="resolve" title="resolve">
<h4>New "resolve" subcommand replaces "resolved"</h4>
<p>A new <strong><tt>resolve</tt></strong> subcommand replaces the
"<tt>resolved</tt>" subcommand (the latter is deprecated, but still
present for compatibility). The new subcommand takes a
<strong><tt>--accept=orig|mine|repo</tt></strong> option to select
which version of a file to retain (which means Subversion now supports
batch-style conflict resolution).</p>
<p>See <a
href="http://subversion.tigris.org/issues/show_bug.cgi?id=2784" >issue
#2784</a> for more.</p>
</div> <!-- resolve -->
<div class="h4" id="intermediate-directories"
title="intermediate-directories">
<h4>Create intermediate directories if asked</h4>
<p>Add, mkdir, copy, and move take a new
<strong><tt>--parents</tt></strong> option, which makes intermediate
directories as necessary to create the destination path.</p>
<p>See <a
href="http://subversion.tigris.org/issues/show_bug.cgi?id=1776" >issue
#1776</a> for more</p>
</div> <!-- intermediate-directories -->
<div class="h4" id="keep-local" title="keep-local">
<h4>New --keep-local option retains path after delete.</h4>
<p>Delete (remove) now takes a
<strong><code>--keep-local</code></strong> option to retain its
targets locally, so paths will not be removed even if unmodified.</p>
</div> <!-- keep-local -->
<div class="h4" id="commit-revprops" title="commit-revprops">
<h4>Commit now takes a --with-revprop option.</h4>
Commit now takes a <strong><tt>--with-revprop</tt></strong> option
allowing one to set revision properties
<p>See <a
href="http://subversion.tigris.org/issues/show_bug.cgi?id=1976" >issue
#1976</a> for more.</p>
</div> <!-- commit-revprops -->
</div> <!-- cmdline -->
<div class="h3" id="dav-modules" title="dav-modules">
<h3>Easier to try out experimental <tt>ra_serf</tt> DAV access module
(<em>client</em>)</h3>
<p>Subversion 1.4 introduced the experimental <tt>ra_serf</tt>
repository access module for accessing HTTP[S] DAV Subversion servers.
This uses the <a href="http://code.google.com/p/serf/">serf</a>
library instead of the Neon library which the original DAV support
uses. serf supports pipelined requests which may lead to better
performance. However, Subversion 1.4 required you to choose which
module to use for accessing DAV servers at build time, which made it
difficult to find out which module performs better for your usage
patterns.</p>
<p>Subversion 1.5 allows you to build both modules at the same time;
you can choose which library to use on a global or host-by-host basis
by setting the <tt>http-library</tt> variable in your run-time server
configuration file (<tt>~/.subversion/servers</tt>). In recognition
of the fact that both libraries are DAV clients, we have
renamed <tt>ra_dav</tt> to <tt>ra_neon</tt>.</p>
</div> <!-- dav-modules -->
<div class="h3" id="apis" title="apis">
<h3>API changes, improvements and language bindings
(<em>client and server</em>)</h3>
<p>There are too many new and revised APIs in Subversion 1.5.0 to even
begin to list them all here. See the <a
href="http://svn.collab.net/svn-doxygen/" >Subversion API
Documentation</a> page for general API information. If you develop a
3rd-party client application that uses Subversion APIs, you should
probably look at the header files for the interfaces you use and see
what's changed.</p>
<p>One general change is that most APIs that formerly took a
<tt>recurse</tt> parameter have been upgraded to accept a
<tt>depth</tt> parameter instead, to enable the new <a
href="#sparse-checkouts">sparse checkouts</a> feature.</p>
<p>Language bindings have mostly been updated for the new APIs, though
some may lag more than others.</p>
</div> <!-- apis -->
<div class="h3" id="bug-fixes" title="bug-fixes">
<h3>Bug fixes (<em>client and server</em>)</h3>
<p>A great many bugs have been fixed. See the 1.5.0 section in the <a
href="http://svn.collab.net/repos/svn/trunk/CHANGES">CHANGES</a> file
for details.</p>
</div> <!-- bug-fixes -->
</div> <!-- enhancements -->
<div class="h2" id="svn-1.3-deprecation" title="svn-1.3-deprecation">
<h2>Subversion 1.3.x series no longer supported</h2>
<p>The Subversion 1.3.x line is no longer supported. This doesn't
mean that your 1.3 installation is doomed; if it works well and is all
you need, that's fine. "No longer supported" just means we've stopped
accepting bug reports against 1.3.x versions, and will not make any
more 1.3.x bugfix releases, except perhaps for absolutely critical
security or data-loss bugs.</p>
</div> <!-- svn-1.3-deprecation -->
<div class="h2" id="deps" title="deps">
<h2>Subversion Dependencies Distribution: Binary Compatibility Issues</h2>
<div class="h3" id="deps-background" title="deps-background">
<h3>Background</h3>
<p>APR 0.9.x and 1.x are binary-incompatible. This means if you are
already using Subversion with APR 0.9.x, and then upgrade your libapr
to 1.X <em>without rebuilding Subversion</em>, things will break.
Things will also break if your Subversion server libraries are linked
to one version of APR but your Apache HTTPD server is linked to a
different version.</p>
<p>For a long time, Subversion's main source distribution included APR
0.9.x, which was the latest available at the time, along with a few
other things (e.g., Neon, zlib) that weren't yet widespread on
installation systems.</p>
</div> <!-- deps-background -->
<div class="h3" id="deps-now" title="deps-now">
<h3>Subversion 1.5.0 Dependencies Distribution</h3>
<p>Today, these dependencies are no longer exotic, so our source
distribution contains just Subversion itself. Those building
Subversion are expected to have the necessary libraries already
installed, or to be able to fetch them easily. But for convenience,
we still offer a "deps" distribution: it doesn't contain Subversion,
it just has source code for those third-party libraries.</p>
<p>Until Subversion 1.5.0, the deps distribution contained APR 0.9.x,
but as of 1.5.0, we're finally upgrading it to APR 1.x. This is
because by now there are very few systems that will have binary
compatibility issues, and of those, few are likely to build using the
"deps" dist.</p>
<p>If you already have a Subversion installation using APR 0.9.x, it's
still possible to move to APR 1.x safely (although you are not
required to, unless you use the deps dist). Just be sure to recompile
Subversion, and Apache httpd if necessary, after upgrading APR.</p>
<p>Note that it's perfectly safe to use APR 1.x from the beginning.
In fact, we recommend it. If you're building Subversion for the first
time, there's no compatibility issue to worry about, so just grab the
latest version of APR (or use our deps dist).</p>
</div> <!-- deps-now -->
</div> <!-- deps -->
</div> <!-- app -->
</body>
</html>