<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE refentry PUBLIC "-//OASIS//DTD DocBook XML V4.5//EN" "http://www.oasis-open.org/docbook/xml/4.5/docbookx.dtd">
<?asciidoc-toc?>
<?asciidoc-numbered?>
<refentry>
<refmeta>
<refentrytitle>git-subtree</refentrytitle>
<manvolnum>1</manvolnum>
<refmiscinfo class="source">Git</refmiscinfo>
<refmiscinfo class="version"></refmiscinfo>
<refmiscinfo class="manual">Git Manual</refmiscinfo>
</refmeta>
<refnamediv>
  <refname>git-subtree</refname>
  <refpurpose>Merge subtrees together and split repository into subtrees</refpurpose>
</refnamediv>
<refsynopsisdiv id="_synopsis">
<blockquote>
<literallayout><emphasis>git subtree</emphasis> add   -P &lt;prefix&gt; &lt;commit&gt;
<emphasis>git subtree</emphasis> pull  -P &lt;prefix&gt; &lt;repository&gt; &lt;refspec&#8230;&gt;
<emphasis>git subtree</emphasis> push  -P &lt;prefix&gt; &lt;repository&gt; &lt;refspec&#8230;&gt;
<emphasis>git subtree</emphasis> merge -P &lt;prefix&gt; &lt;commit&gt;
<emphasis>git subtree</emphasis> split -P &lt;prefix&gt; [OPTIONS] [&lt;commit&gt;]</literallayout>
</blockquote>
</refsynopsisdiv>
<refsect1 id="_description">
<title>DESCRIPTION</title>
<simpara>Subtrees allow subprojects to be included within a subdirectory
of the main project, optionally including the subproject&#8217;s
entire history.</simpara>
<simpara>For example, you could include the source code for a library
as a subdirectory of your application.</simpara>
<simpara>Subtrees are not to be confused with submodules, which are meant for
the same task. Unlike submodules, subtrees do not need any special
constructions (like .gitmodule files or gitlinks) be present in
your repository, and do not force end-users of your
repository to do anything special or to understand how subtrees
work. A subtree is just a subdirectory that can be
committed to, branched, and merged along with your project in
any way you want.</simpara>
<simpara>They are also not to be confused with using the subtree merge
strategy. The main difference is that, besides merging
the other project as a subdirectory, you can also extract the
entire history of a subdirectory from your project and make it
into a standalone project. Unlike the subtree merge strategy
you can alternate back and forth between these
two operations. If the standalone library gets updated, you can
automatically merge the changes into your project; if you
update the library inside your project, you can "split" the
changes back out again and merge them back into the library
project.</simpara>
<simpara>For example, if a library you made for one application ends up being
useful elsewhere, you can extract its entire history and publish
that as its own git repository, without accidentally
intermingling the history of your application project.</simpara>
<tip><simpara>In order to keep your commit messages clean, we recommend that
people split their commits between the subtrees and the main
project as much as possible.  That is, if you make a change that
affects both the library and the main application, commit it in
two pieces.  That way, when you split the library commits out
later, their descriptions will still make sense.  But if this
isn&#8217;t important to you, it&#8217;s not <emphasis role="strong">necessary</emphasis>.  git subtree will
simply leave out the non-library-related parts of the commit
when it splits it out into the subproject later.</simpara></tip>
</refsect1>
<refsect1 id="_commands">
<title>COMMANDS</title>
<variablelist>
<varlistentry>
<term>
add
</term>
<listitem>
<simpara>
        Create the &lt;prefix&gt; subtree by importing its contents
        from the given &lt;refspec&gt; or &lt;repository&gt; and remote &lt;refspec&gt;.
        A new commit is created automatically, joining the imported
        project&#8217;s history with your own.  With <emphasis>--squash</emphasis>, imports
        only a single commit from the subproject, rather than its
        entire history.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
merge
</term>
<listitem>
<simpara>
        Merge recent changes up to &lt;commit&gt; into the &lt;prefix&gt;
        subtree.  As with normal <emphasis>git merge</emphasis>, this doesn&#8217;t
        remove your own local changes; it just merges those
        changes into the latest &lt;commit&gt;.  With <emphasis>--squash</emphasis>,
        creates only one commit that contains all the changes,
        rather than merging in the entire history.
</simpara>
<literallayout class="monospaced">If you use '--squash', the merge direction doesn't
always have to be forward; you can use this command to
go back in time from v2.5 to v2.4, for example.  If your
merge introduces a conflict, you can resolve it in the
usual ways.</literallayout>
</listitem>
</varlistentry>
<varlistentry>
<term>
pull
</term>
<listitem>
<simpara>
        Exactly like <emphasis>merge</emphasis>, but parallels <emphasis>git pull</emphasis> in that
        it fetches the given commit from the specified remote
        repository.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
push
</term>
<listitem>
<simpara>
        Does a <emphasis>split</emphasis> (see below) using the &lt;prefix&gt; supplied
        and then does a <emphasis>git push</emphasis> to push the result to the
        repository and refspec. This can be used to push your
        subtree to different branches of the remote repository.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
split
</term>
<listitem>
<simpara>
        Extract a new, synthetic project history from the
        history of the &lt;prefix&gt; subtree.  The new history
        includes only the commits (including merges) that
        affected &lt;prefix&gt;, and each of those commits now has the
        contents of &lt;prefix&gt; at the root of the project instead
        of in a subdirectory.  Thus, the newly created history
        is suitable for export as a separate git repository.
</simpara>
<literallayout class="monospaced">After splitting successfully, a single commit id is
printed to stdout.  This corresponds to the HEAD of the
newly created tree, which you can manipulate however you
want.</literallayout>
<literallayout class="monospaced">Repeated splits of exactly the same history are
guaranteed to be identical (ie. to produce the same
commit ids).  Because of this, if you add new commits
and then re-split, the new commits will be attached as
commits on top of the history you generated last time,
so 'git merge' and friends will work as expected.</literallayout>
<literallayout class="monospaced">Note that if you use '--squash' when you merge, you
should usually not just '--rejoin' when you split.</literallayout>
</listitem>
</varlistentry>
</variablelist>
</refsect1>
<refsect1 id="_options">
<title>OPTIONS</title>
<variablelist>
<varlistentry>
<term>
-q
</term>
<term>
--quiet
</term>
<listitem>
<simpara>
        Suppress unnecessary output messages on stderr.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
-d
</term>
<term>
--debug
</term>
<listitem>
<simpara>
        Produce even more unnecessary output messages on stderr.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
-P &lt;prefix&gt;
</term>
<term>
--prefix=&lt;prefix&gt;
</term>
<listitem>
<simpara>
        Specify the path in the repository to the subtree you
        want to manipulate.  This option is mandatory
        for all commands.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
-m &lt;message&gt;
</term>
<term>
--message=&lt;message&gt;
</term>
<listitem>
<simpara>
        This option is only valid for add, merge and pull (unsure).
        Specify &lt;message&gt; as the commit message for the merge commit.
</simpara>
</listitem>
</varlistentry>
</variablelist>
</refsect1>
<refsect1 id="_options_for_add_merge_push_pull">
<title>OPTIONS FOR add, merge, push, pull</title>
<variablelist>
<varlistentry>
<term>
--squash
</term>
<listitem>
<simpara>
        This option is only valid for add, merge, push and pull
        commands.
</simpara>
<literallayout class="monospaced">Instead of merging the entire history from the subtree
project, produce only a single commit that contains all
the differences you want to merge, and then merge that
new commit into your project.</literallayout>
<literallayout class="monospaced">Using this option helps to reduce log clutter. People
rarely want to see every change that happened between
v1.0 and v1.1 of the library they're using, since none of the
interim versions were ever included in their application.</literallayout>
<literallayout class="monospaced">Using '--squash' also helps avoid problems when the same
subproject is included multiple times in the same
project, or is removed and then re-added.  In such a
case, it doesn't make sense to combine the histories
anyway, since it's unclear which part of the history
belongs to which subtree.</literallayout>
<literallayout class="monospaced">Furthermore, with '--squash', you can switch back and
forth between different versions of a subtree, rather
than strictly forward.  'git subtree merge --squash'
always adjusts the subtree to match the exactly
specified commit, even if getting to that commit would
require undoing some changes that were added earlier.</literallayout>
<literallayout class="monospaced">Whether or not you use '--squash', changes made in your
local repository remain intact and can be later split
and send upstream to the subproject.</literallayout>
</listitem>
</varlistentry>
</variablelist>
</refsect1>
<refsect1 id="_options_for_split">
<title>OPTIONS FOR split</title>
<variablelist>
<varlistentry>
<term>
--annotate=&lt;annotation&gt;
</term>
<listitem>
<simpara>
        This option is only valid for the split command.
</simpara>
<literallayout class="monospaced">When generating synthetic history, add &lt;annotation&gt; as a
prefix to each commit message.  Since we're creating new
commits with the same commit message, but possibly
different content, from the original commits, this can help
to differentiate them and avoid confusion.</literallayout>
<literallayout class="monospaced">Whenever you split, you need to use the same
&lt;annotation&gt;, or else you don't have a guarantee that
the new re-created history will be identical to the old
one.  That will prevent merging from working correctly.
git subtree tries to make it work anyway, particularly
if you use --rejoin, but it may not always be effective.</literallayout>
</listitem>
</varlistentry>
<varlistentry>
<term>
-b &lt;branch&gt;
</term>
<term>
--branch=&lt;branch&gt;
</term>
<listitem>
<simpara>
        This option is only valid for the split command.
</simpara>
<literallayout class="monospaced">After generating the synthetic history, create a new
branch called &lt;branch&gt; that contains the new history.
This is suitable for immediate pushing upstream.
&lt;branch&gt; must not already exist.</literallayout>
</listitem>
</varlistentry>
<varlistentry>
<term>
--ignore-joins
</term>
<listitem>
<simpara>
        This option is only valid for the split command.
</simpara>
<literallayout class="monospaced">If you use '--rejoin', git subtree attempts to optimize
its history reconstruction to generate only the new
commits since the last '--rejoin'.  '--ignore-join'
disables this behaviour, forcing it to regenerate the
entire history.  In a large project, this can take a
long time.</literallayout>
</listitem>
</varlistentry>
<varlistentry>
<term>
--onto=&lt;onto&gt;
</term>
<listitem>
<simpara>
        This option is only valid for the split command.
</simpara>
<literallayout class="monospaced">If your subtree was originally imported using something
other than git subtree, its history may not match what
git subtree is expecting.  In that case, you can specify
the commit id &lt;onto&gt; that corresponds to the first
revision of the subproject's history that was imported
into your project, and git subtree will attempt to build
its history from there.</literallayout>
<literallayout class="monospaced">If you used 'git subtree add', you should never need
this option.</literallayout>
</listitem>
</varlistentry>
<varlistentry>
<term>
--rejoin
</term>
<listitem>
<simpara>
        This option is only valid for the split command.
</simpara>
<literallayout class="monospaced">After splitting, merge the newly created synthetic
history back into your main project.  That way, future
splits can search only the part of history that has
been added since the most recent --rejoin.</literallayout>
<literallayout class="monospaced">If your split commits end up merged into the upstream
subproject, and then you want to get the latest upstream
version, this will allow git's merge algorithm to more
intelligently avoid conflicts (since it knows these
synthetic commits are already part of the upstream
repository).</literallayout>
<literallayout class="monospaced">Unfortunately, using this option results in 'git log'
showing an extra copy of every new commit that was
created (the original, and the synthetic one).</literallayout>
<literallayout class="monospaced">If you do all your merges with '--squash', don't use
'--rejoin' when you split, because you don't want the
subproject's history to be part of your project anyway.</literallayout>
</listitem>
</varlistentry>
</variablelist>
</refsect1>
<refsect1 id="_example_1_add_command">
<title>EXAMPLE 1. Add command</title>
<simpara>Let&#8217;s assume that you have a local repository that you would like
to add an external vendor library to. In this case we will add the
git-subtree repository as a subdirectory of your already existing
git-extensions repository in ~/git-extensions/:</simpara>
<literallayout class="monospaced">$ git subtree add --prefix=git-subtree --squash \
        git://github.com/apenwarr/git-subtree.git master</literallayout>
<simpara><emphasis>master</emphasis> needs to be a valid remote ref and can be a different branch
name</simpara>
<simpara>You can omit the --squash flag, but doing so will increase the number
of commits that are incldued in your local repository.</simpara>
<simpara>We now have a ~/git-extensions/git-subtree directory containing code
from the master branch of git://github.com/apenwarr/git-subtree.git
in our git-extensions repository.</simpara>
</refsect1>
<refsect1 id="_example_2_extract_a_subtree_using_commit_merge_and_pull">
<title>EXAMPLE 2. Extract a subtree using commit, merge and pull</title>
<simpara>Let&#8217;s use the repository for the git source code as an example.
First, get your own copy of the git.git repository:</simpara>
<literallayout class="monospaced">$ git clone git://git.kernel.org/pub/scm/git/git.git test-git
$ cd test-git</literallayout>
<simpara>gitweb (commit 1130ef3) was merged into git as of commit
0a8f4f0, after which it was no longer maintained separately.
But imagine it had been maintained separately, and we wanted to
extract git&#8217;s changes to gitweb since that time, to share with
the upstream.  You could do this:</simpara>
<literallayout class="monospaced">$ git subtree split --prefix=gitweb --annotate='(split) ' \
        0a8f4f0^.. --onto=1130ef3 --rejoin \
        --branch gitweb-latest
$ gitk gitweb-latest
$ git push git@github.com:whatever/gitweb.git gitweb-latest:master</literallayout>
<simpara>(We use <emphasis>0a8f4f0^..</emphasis> because that means "all the changes from
0a8f4f0 to the current version, including 0a8f4f0 itself.")</simpara>
<simpara>If gitweb had originally been merged using <emphasis>git subtree add</emphasis> (or
a previous split had already been done with --rejoin specified)
then you can do all your splits without having to remember any
weird commit ids:</simpara>
<literallayout class="monospaced">$ git subtree split --prefix=gitweb --annotate='(split) ' --rejoin \
        --branch gitweb-latest2</literallayout>
<simpara>And you can merge changes back in from the upstream project just
as easily:</simpara>
<literallayout class="monospaced">$ git subtree pull --prefix=gitweb \
        git@github.com:whatever/gitweb.git master</literallayout>
<simpara>Or, using <emphasis>--squash</emphasis>, you can actually rewind to an earlier
version of gitweb:</simpara>
<literallayout class="monospaced">$ git subtree merge --prefix=gitweb --squash gitweb-latest~10</literallayout>
<simpara>Then make some changes:</simpara>
<literallayout class="monospaced">$ date &gt;gitweb/myfile
$ git add gitweb/myfile
$ git commit -m 'created myfile'</literallayout>
<simpara>And fast forward again:</simpara>
<literallayout class="monospaced">$ git subtree merge --prefix=gitweb --squash gitweb-latest</literallayout>
<simpara>And notice that your change is still intact:</simpara>
<literallayout class="monospaced">$ ls -l gitweb/myfile</literallayout>
<simpara>And you can split it out and look at your changes versus
the standard gitweb:</simpara>
<literallayout class="monospaced">git log gitweb-latest..$(git subtree split --prefix=gitweb)</literallayout>
</refsect1>
<refsect1 id="_example_3_extract_a_subtree_using_branch">
<title>EXAMPLE 3. Extract a subtree using branch</title>
<simpara>Suppose you have a source directory with many files and
subdirectories, and you want to extract the lib directory to its own
git project. Here&#8217;s a short way to do it:</simpara>
<simpara>First, make the new repository wherever you want:</simpara>
<literallayout class="monospaced">$ &lt;go to the new location&gt;
$ git init --bare</literallayout>
<simpara>Back in your original directory:</simpara>
<literallayout class="monospaced">$ git subtree split --prefix=lib --annotate="(split)" -b split</literallayout>
<simpara>Then push the new branch onto the new empty repository:</simpara>
<literallayout class="monospaced">$ git push &lt;new-repo&gt; split:master</literallayout>
</refsect1>
<refsect1 id="_author">
<title>AUTHOR</title>
<simpara>Written by Avery Pennarun &lt;<ulink url="mailto:apenwarr@gmail.com">apenwarr@gmail.com</ulink>&gt;</simpara>
</refsect1>
<refsect1 id="_git">
<title>GIT</title>
<simpara>Part of the <citerefentry>
<refentrytitle>git</refentrytitle><manvolnum>1</manvolnum>
</citerefentry> suite</simpara>
</refsect1>
</refentry>