1GIT-SUBTREE(1) Git Manual GIT-SUBTREE(1)
2
6 git-subtree - Merge subtrees together and split repository into
7 subtrees
8
10 git subtree add -P <prefix> <commit>
11 git subtree add -P <prefix> <repository> <ref>
12 git subtree pull -P <prefix> <repository> <ref>
13 git subtree push -P <prefix> <repository> <ref>
14 git subtree merge -P <prefix> <commit>
15 git subtree split -P <prefix> [OPTIONS] [<commit>]
16
19 Subtrees allow subprojects to be included within a subdirectory of the
20 main project, optionally including the subproject’s entire history.
21 For example, you could include the source code for a library as a
23 subdirectory of your application.
24 Subtrees are not to be confused with submodules, which are meant for
26 the same task. Unlike submodules, subtrees do not need any special
27 constructions (like .gitmodules files or gitlinks) be present in your
28 repository, and do not force end-users of your repository to do
29 anything special or to understand how subtrees work. A subtree is just
30 a subdirectory that can be committed to, branched, and merged along
31 with your project in any way you want.
32 They are also not to be confused with using the subtree merge strategy.
34 The main difference is that, besides merging the other project as a
35 subdirectory, you can also extract the entire history of a subdirectory
36 from your project and make it into a standalone project. Unlike the
37 subtree merge strategy you can alternate back and forth between these
38 two operations. If the standalone library gets updated, you can
39 automatically merge the changes into your project; if you update the
40 library inside your project, you can "split" the changes back out again
41 and merge them back into the library project.
42 For example, if a library you made for one application ends up being
44 useful elsewhere, you can extract its entire history and publish that
45 as its own git repository, without accidentally intermingling the
46 history of your application project.
47 Tip
49 In order to keep your commit messages clean, we recommend that
50 people split their commits between the subtrees and the main
51 project as much as possible. That is, if you make a change that
52 affects both the library and the main application, commit it in two
53 pieces. That way, when you split the library commits out later,
54 their descriptions will still make sense. But if this isn’t
55 important to you, it’s not necessary. git subtree will simply leave
56 out the non-library-related parts of the commit when it splits it
57 out into the subproject later.
58
60 add
61 Create the <prefix> subtree by importing its contents from the
62 given <commit> or <repository> and remote <ref>. A new commit is
63 created automatically, joining the imported project’s history with
64 your own. With --squash, imports only a single commit from the
65 subproject, rather than its entire history.
66 merge
68 Merge recent changes up to <commit> into the <prefix> subtree. As
69 with normal git merge, this doesn’t remove your own local changes;
70 it just merges those changes into the latest <commit>. With
71 --squash, creates only one commit that contains all the changes,
72 rather than merging in the entire history.
73 If you use --squash, the merge direction doesn’t always have to be
75 forward; you can use this command to go back in time from v2.5 to
76 v2.4, for example. If your merge introduces a conflict, you can
77 resolve it in the usual ways.
78 pull
80 Exactly like merge, but parallels git pull in that it fetches the
81 given ref from the specified remote repository.
82 push
84 Does a split (see below) using the <prefix> supplied and then does
85 a git push to push the result to the repository and ref. This can
86 be used to push your subtree to different branches of the remote
87 repository.
88 split
90 Extract a new, synthetic project history from the history of the
91 <prefix> subtree. The new history includes only the commits
92 (including merges) that affected <prefix>, and each of those
93 commits now has the contents of <prefix> at the root of the project
94 instead of in a subdirectory. Thus, the newly created history is
95 suitable for export as a separate git repository.
96 After splitting successfully, a single commit id is printed to
98 stdout. This corresponds to the HEAD of the newly created tree,
99 which you can manipulate however you want.
100 Repeated splits of exactly the same history are guaranteed to be
102 identical (i.e. to produce the same commit ids). Because of this,
103 if you add new commits and then re-split, the new commits will be
104 attached as commits on top of the history you generated last time,
105 so git merge and friends will work as expected.
106 Note that if you use --squash when you merge, you should usually
108 not just --rejoin when you split.
109
111 -q, --quiet
112 Suppress unnecessary output messages on stderr.
113 -d, --debug
115 Produce even more unnecessary output messages on stderr.
116 -P <prefix>, --prefix=<prefix>
118 Specify the path in the repository to the subtree you want to
119 manipulate. This option is mandatory for all commands.
120 -m <message>, --message=<message>
122 This option is only valid for add, merge and pull (unsure). Specify
123 <message> as the commit message for the merge commit.
124
126 --squash
127 This option is only valid for add, merge, and pull commands.
128 Instead of merging the entire history from the subtree project,
130 produce only a single commit that contains all the differences you
131 want to merge, and then merge that new commit into your project.
132 Using this option helps to reduce log clutter. People rarely want
134 to see every change that happened between v1.0 and v1.1 of the
135 library they’re using, since none of the interim versions were ever
136 included in their application.
137 Using --squash also helps avoid problems when the same subproject
139 is included multiple times in the same project, or is removed and
140 then re-added. In such a case, it doesn’t make sense to combine the
141 histories anyway, since it’s unclear which part of the history
142 belongs to which subtree.
143 Furthermore, with --squash, you can switch back and forth between
145 different versions of a subtree, rather than strictly forward. git
146 subtree merge --squash always adjusts the subtree to match the
147 exactly specified commit, even if getting to that commit would
148 require undoing some changes that were added earlier.
149 Whether or not you use --squash, changes made in your local
151 repository remain intact and can be later split and send upstream
152 to the subproject.
153
155 --annotate=<annotation>
156 This option is only valid for the split command.
157 When generating synthetic history, add <annotation> as a prefix to
159 each commit message. Since we’re creating new commits with the same
160 commit message, but possibly different content, from the original
161 commits, this can help to differentiate them and avoid confusion.
162 Whenever you split, you need to use the same <annotation>, or else
164 you don’t have a guarantee that the new re-created history will be
165 identical to the old one. That will prevent merging from working
166 correctly. git subtree tries to make it work anyway, particularly
167 if you use --rejoin, but it may not always be effective.
168 -b <branch>, --branch=<branch>
170 This option is only valid for the split command.
171 After generating the synthetic history, create a new branch called
173 <branch> that contains the new history. This is suitable for
174 immediate pushing upstream. <branch> must not already exist.
175 --ignore-joins
177 This option is only valid for the split command.
178 If you use --rejoin, git subtree attempts to optimize its history
180 reconstruction to generate only the new commits since the last
181 --rejoin. --ignore-join disables this behaviour, forcing it to
182 regenerate the entire history. In a large project, this can take a
183 long time.
184 --onto=<onto>
186 This option is only valid for the split command.
187 If your subtree was originally imported using something other than
189 git subtree, its history may not match what git subtree is
190 expecting. In that case, you can specify the commit id <onto> that
191 corresponds to the first revision of the subproject’s history that
192 was imported into your project, and git subtree will attempt to
193 build its history from there.
194 If you used git subtree add, you should never need this option.
196 --rejoin
198 This option is only valid for the split command.
199 After splitting, merge the newly created synthetic history back
201 into your main project. That way, future splits can search only the
202 part of history that has been added since the most recent --rejoin.
203 If your split commits end up merged into the upstream subproject,
205 and then you want to get the latest upstream version, this will
206 allow git’s merge algorithm to more intelligently avoid conflicts
207 (since it knows these synthetic commits are already part of the
208 upstream repository).
209 Unfortunately, using this option results in git log showing an
211 extra copy of every new commit that was created (the original, and
212 the synthetic one).
213 If you do all your merges with --squash, don’t use --rejoin when
215 you split, because you don’t want the subproject’s history to be
216 part of your project anyway.
217
219 Let’s assume that you have a local repository that you would like to
220 add an external vendor library to. In this case we will add the
221 git-subtree repository as a subdirectory of your already existing
222 git-extensions repository in ~/git-extensions/:
223 $ git subtree add --prefix=git-subtree --squash \
225 git://github.com/apenwarr/git-subtree.git master
226 master needs to be a valid remote ref and can be a different branch
228 name
229 You can omit the --squash flag, but doing so will increase the number
231 of commits that are included in your local repository.
232 We now have a ~/git-extensions/git-subtree directory containing code
234 from the master branch of git://github.com/apenwarr/git-subtree.git in
235 our git-extensions repository.
236
238 Let’s use the repository for the git source code as an example. First,
239 get your own copy of the git.git repository:
240 $ git clone git://git.kernel.org/pub/scm/git/git.git test-git
242 $ cd test-git
243 gitweb (commit 1130ef3) was merged into git as of commit 0a8f4f0, after
245 which it was no longer maintained separately. But imagine it had been
246 maintained separately, and we wanted to extract git’s changes to gitweb
247 since that time, to share with the upstream. You could do this:
248 $ git subtree split --prefix=gitweb --annotate='(split) ' \
250 0a8f4f0^.. --onto=1130ef3 --rejoin \
251 --branch gitweb-latest
252 $ gitk gitweb-latest
253 $ git push git@github.com:whatever/gitweb.git gitweb-latest:master
254 (We use 0a8f4f0^.. because that means "all the changes from 0a8f4f0 to
256 the current version, including 0a8f4f0 itself.")
257 If gitweb had originally been merged using git subtree add (or a
259 previous split had already been done with --rejoin specified) then you
260 can do all your splits without having to remember any weird commit ids:
261 $ git subtree split --prefix=gitweb --annotate='(split) ' --rejoin \
263 --branch gitweb-latest2
264 And you can merge changes back in from the upstream project just as
266 easily:
267 $ git subtree pull --prefix=gitweb \
269 git@github.com:whatever/gitweb.git master
270 Or, using --squash, you can actually rewind to an earlier version of
272 gitweb:
273 $ git subtree merge --prefix=gitweb --squash gitweb-latest~10
275 Then make some changes:
277 $ date >gitweb/myfile
279 $ git add gitweb/myfile
280 $ git commit -m 'created myfile'
281 And fast forward again:
283 $ git subtree merge --prefix=gitweb --squash gitweb-latest
285 And notice that your change is still intact:
287 $ ls -l gitweb/myfile
289 And you can split it out and look at your changes versus the standard
291 gitweb:
292 git log gitweb-latest..$(git subtree split --prefix=gitweb)
294
296 Suppose you have a source directory with many files and subdirectories,
297 and you want to extract the lib directory to its own git project.
298 Here’s a short way to do it:
299 First, make the new repository wherever you want:
301 $ <go to the new location>
303 $ git init --bare
304 Back in your original directory:
306 $ git subtree split --prefix=lib --annotate="(split)" -b split
308 Then push the new branch onto the new empty repository:
310 $ git push <new-repo> split:master
312
314 Written by Avery Pennarun <apenwarr@gmail.com[1]>
315
317 Part of the git(1) suite
318
320 1. apenwarr@gmail.com
321 mailto:apenwarr@gmail.com
322Git 2.21.0 02/24/2019 GIT-SUBTREE(1)