1GIT-CHECKOUT(1) Git Manual GIT-CHECKOUT(1)
2
6 git-checkout - Checkout a branch or paths to the working tree
7
9 git checkout [-q] [-f] [-m] [<branch>]
10 git checkout [-q] [-f] [-m] [--detach] [<commit>]
11 git checkout [-q] [-f] [-m] [[-b|-B|--orphan] <new_branch>] [<start_point>]
12 git checkout [-f|--ours|--theirs|-m|--conflict=<style>] [<tree-ish>] [--] <paths>...
13 git checkout [-p|--patch] [<tree-ish>] [--] [<paths>...]
14
17 Updates files in the working tree to match the version in the index or
18 the specified tree. If no paths are given, git checkout will also
19 update HEAD to set the specified branch as the current branch.
20 git checkout <branch>
22 To prepare for working on <branch>, switch to it by updating the
23 index and the files in the working tree, and by pointing HEAD at
24 the branch. Local modifications to the files in the working tree
25 are kept, so that they can be committed to the <branch>.
26 If <branch> is not found but there does exist a tracking branch in
28 exactly one remote (call it <remote>) with a matching name, treat
29 as equivalent to
30 $ git checkout -b <branch> --track <remote>/<branch>
32 You could omit <branch>, in which case the command degenerates to
34 "check out the current branch", which is a glorified no-op with a
35 rather expensive side-effects to show only the tracking
36 information, if exists, for the current branch.
37 git checkout -b|-B <new_branch> [<start point>]
39 Specifying -b causes a new branch to be created as if git-branch(1)
40 were called and then checked out. In this case you can use the
41 --track or --no-track options, which will be passed to git branch.
42 As a convenience, --track without -b implies branch creation; see
43 the description of --track below.
44 If -B is given, <new_branch> is created if it doesn’t exist;
46 otherwise, it is reset. This is the transactional equivalent of
47 $ git branch -f <branch> [<start point>]
49 $ git checkout <branch>
50 that is to say, the branch is not reset/created unless "git
52 checkout" is successful.
53 git checkout --detach [<branch>], git checkout <commit>
55 Prepare to work on top of <commit>, by detaching HEAD at it (see
56 "DETACHED HEAD" section), and updating the index and the files in
57 the working tree. Local modifications to the files in the working
58 tree are kept, so that the resulting working tree will be the state
59 recorded in the commit plus the local modifications.
60 Passing --detach forces this behavior in the case of a <branch>
62 (without the option, giving a branch name to the command would
63 check out the branch, instead of detaching HEAD at it), or the
64 current commit, if no <branch> is specified.
65 git checkout [-p|--patch] [<tree-ish>] [--] <pathspec>...
67 When <paths> or --patch are given, git checkout does not switch
68 branches. It updates the named paths in the working tree from the
69 index file or from a named <tree-ish> (most often a commit). In
70 this case, the -b and --track options are meaningless and giving
71 either of them results in an error. The <tree-ish> argument can be
72 used to specify a specific tree-ish (i.e. commit, tag or tree) to
73 update the index for the given paths before updating the working
74 tree.
75 The index may contain unmerged entries because of a previous failed
77 merge. By default, if you try to check out such an entry from the
78 index, the checkout operation will fail and nothing will be checked
79 out. Using -f will ignore these unmerged entries. The contents from
80 a specific side of the merge can be checked out of the index by
81 using --ours or --theirs. With -m, changes made to the working tree
82 file can be discarded to re-create the original conflicted merge
83 result.
84
86 -q, --quiet
87 Quiet, suppress feedback messages.
88 -f, --force
90 When switching branches, proceed even if the index or the working
91 tree differs from HEAD. This is used to throw away local changes.
92 When checking out paths from the index, do not fail upon unmerged
94 entries; instead, unmerged entries are ignored.
95 --ours, --theirs
97 When checking out paths from the index, check out stage #2 (ours)
98 or #3 (theirs) for unmerged paths.
99 -b <new_branch>
101 Create a new branch named <new_branch> and start it at
102 <start_point>; see git-branch(1) for details.
103 -B <new_branch>
105 Creates the branch <new_branch> and start it at <start_point>; if
106 it already exists, then reset it to <start_point>. This is
107 equivalent to running "git branch" with "-f"; see git-branch(1) for
108 details.
109 -t, --track
111 When creating a new branch, set up "upstream" configuration. See
112 "--track" in git-branch(1) for details.
113 If no -b option is given, the name of the new branch will be
115 derived from the remote-tracking branch. If "remotes/" or
116 "refs/remotes/" is prefixed it is stripped away, and then the part
117 up to the next slash (which would be the nickname of the remote) is
118 removed. This would tell us to use "hack" as the local branch when
119 branching off of "origin/hack" (or "remotes/origin/hack", or even
120 "refs/remotes/origin/hack"). If the given name has no slash, or the
121 above guessing results in an empty name, the guessing is aborted.
122 You can explicitly give a name with -b in such a case.
123 --no-track
125 Do not set up "upstream" configuration, even if the
126 branch.autosetupmerge configuration variable is true.
127 -l
129 Create the new branch’s reflog; see git-branch(1) for details.
130 --detach
132 Rather than checking out a branch to work on it, check out a commit
133 for inspection and discardable experiments. This is the default
134 behavior of "git checkout <commit>" when <commit> is not a branch
135 name. See the "DETACHED HEAD" section below for details.
136 --orphan <new_branch>
138 Create a new orphan branch, named <new_branch>, started from
139 <start_point> and switch to it. The first commit made on this new
140 branch will have no parents and it will be the root of a new
141 history totally disconnected from all the other branches and
142 commits.
143 The index and the working tree are adjusted as if you had
145 previously run "git checkout <start_point>". This allows you to
146 start a new history that records a set of paths similar to
147 <start_point> by easily running "git commit -a" to make the root
148 commit.
149 This can be useful when you want to publish the tree from a commit
151 without exposing its full history. You might want to do this to
152 publish an open source branch of a project whose current tree is
153 "clean", but whose full history contains proprietary or otherwise
154 encumbered bits of code.
155 If you want to start a disconnected history that records a set of
157 paths that is totally different from the one of <start_point>, then
158 you should clear the index and the working tree right after
159 creating the orphan branch by running "git rm -rf ." from the top
160 level of the working tree. Afterwards you will be ready to prepare
161 your new files, repopulating the working tree, by copying them from
162 elsewhere, extracting a tarball, etc.
163 --ignore-skip-worktree-bits
165 In sparse checkout mode, git checkout -- <paths> would update only
166 entries matched by <paths> and sparse patterns in
167 $GIT_DIR/info/sparse-checkout. This option ignores the sparse
168 patterns and adds back any files in <paths>.
169 -m, --merge
171 When switching branches, if you have local modifications to one or
172 more files that are different between the current branch and the
173 branch to which you are switching, the command refuses to switch
174 branches in order to preserve your modifications in context.
175 However, with this option, a three-way merge between the current
176 branch, your working tree contents, and the new branch is done, and
177 you will be on the new branch.
178 When a merge conflict happens, the index entries for conflicting
180 paths are left unmerged, and you need to resolve the conflicts and
181 mark the resolved paths with git add (or git rm if the merge should
182 result in deletion of the path).
183 When checking out paths from the index, this option lets you
185 recreate the conflicted merge in the specified paths.
186 --conflict=<style>
188 The same as --merge option above, but changes the way the
189 conflicting hunks are presented, overriding the merge.conflictstyle
190 configuration variable. Possible values are "merge" (default) and
191 "diff3" (in addition to what is shown by "merge" style, shows the
192 original contents).
193 -p, --patch
195 Interactively select hunks in the difference between the <tree-ish>
196 (or the index, if unspecified) and the working tree. The chosen
197 hunks are then applied in reverse to the working tree (and if a
198 <tree-ish> was specified, the index).
199 This means that you can use git checkout -p to selectively discard
201 edits from your current working tree. See the “Interactive Mode”
202 section of git-add(1) to learn how to operate the --patch mode.
203 <branch>
205 Branch to checkout; if it refers to a branch (i.e., a name that,
206 when prepended with "refs/heads/", is a valid ref), then that
207 branch is checked out. Otherwise, if it refers to a valid commit,
208 your HEAD becomes "detached" and you are no longer on any branch
209 (see below for details).
210 As a special case, the "@{-N}" syntax for the N-th last branch
212 checks out the branch (instead of detaching). You may also specify
213 - which is synonymous with "@{-1}".
214 As a further special case, you may use "A...B" as a shortcut for
216 the merge base of A and B if there is exactly one merge base. You
217 can leave out at most one of A and B, in which case it defaults to
218 HEAD.
219 <new_branch>
221 Name for the new branch.
222 <start_point>
224 The name of a commit at which to start the new branch; see git-
225 branch(1) for details. Defaults to HEAD.
226 <tree-ish>
228 Tree to checkout from (when paths are given). If not specified, the
229 index will be used.
230
232 HEAD normally refers to a named branch (e.g. master). Meanwhile, each
233 branch refers to a specific commit. Let’s look at a repo with three
234 commits, one of them tagged, and with branch master checked out:
235 HEAD (refers to branch 'master')
237 |
238 v
239 a---b---c branch 'master' (refers to commit 'c')
240 ^
241 |
242 tag 'v2.0' (refers to commit 'b')
243 When a commit is created in this state, the branch is updated to refer
246 to the new commit. Specifically, git commit creates a new commit d,
247 whose parent is commit c, and then updates branch master to refer to
248 new commit d. HEAD still refers to branch master and so indirectly now
249 refers to commit d:
250 $ edit; git add; git commit
252 HEAD (refers to branch 'master')
254 |
255 v
256 a---b---c---d branch 'master' (refers to commit 'd')
257 ^
258 |
259 tag 'v2.0' (refers to commit 'b')
260 It is sometimes useful to be able to checkout a commit that is not at
263 the tip of any named branch, or even to create a new commit that is not
264 referenced by a named branch. Let’s look at what happens when we
265 checkout commit b (here we show two ways this may be done):
266 $ git checkout v2.0 # or
268 $ git checkout master^^
269 HEAD (refers to commit 'b')
271 |
272 v
273 a---b---c---d branch 'master' (refers to commit 'd')
274 ^
275 |
276 tag 'v2.0' (refers to commit 'b')
277 Notice that regardless of which checkout command we use, HEAD now
280 refers directly to commit b. This is known as being in detached HEAD
281 state. It means simply that HEAD refers to a specific commit, as
282 opposed to referring to a named branch. Let’s see what happens when we
283 create a commit:
284 $ edit; git add; git commit
286 HEAD (refers to commit 'e')
288 |
289 v
290 e
291 /
292 a---b---c---d branch 'master' (refers to commit 'd')
293 ^
294 |
295 tag 'v2.0' (refers to commit 'b')
296 There is now a new commit e, but it is referenced only by HEAD. We can
299 of course add yet another commit in this state:
300 $ edit; git add; git commit
302 HEAD (refers to commit 'f')
304 |
305 v
306 e---f
307 /
308 a---b---c---d branch 'master' (refers to commit 'd')
309 ^
310 |
311 tag 'v2.0' (refers to commit 'b')
312 In fact, we can perform all the normal Git operations. But, let’s look
315 at what happens when we then checkout master:
316 $ git checkout master
318 HEAD (refers to branch 'master')
320 e---f |
321 / v
322 a---b---c---d branch 'master' (refers to commit 'd')
323 ^
324 |
325 tag 'v2.0' (refers to commit 'b')
326 It is important to realize that at this point nothing refers to commit
329 f. Eventually commit f (and by extension commit e) will be deleted by
330 the routine Git garbage collection process, unless we create a
331 reference before that happens. If we have not yet moved away from
332 commit f, any of these will create a reference to it:
333 $ git checkout -b foo [1m(1)
335 $ git branch foo [1m(2)
336 $ git tag foo [1m(3)
337 1. creates a new branch foo, which refers to commit f, and then updates
340 HEAD to refer to branch foo. In other words, we’ll no longer be in
341 detached HEAD state after this command.
342 2. similarly creates a new branch foo, which refers to commit f, but
343 leaves HEAD detached.
344 3. creates a new tag foo, which refers to commit f, leaving HEAD
345 detached.
346 If we have moved away from commit f, then we must first recover its
348 object name (typically by using git reflog), and then we can create a
349 reference to it. For example, to see the last two commits to which HEAD
350 referred, we can use either of these commands:
351 $ git reflog -2 HEAD # or
353 $ git log -g -2 HEAD
354
357 1. The following sequence checks out the master branch, reverts the
358 Makefile to two revisions back, deletes hello.c by mistake, and
359 gets it back from the index.
360 $ git checkout master [1m(1)
362 $ git checkout master~2 Makefile [1m(2)
363 $ rm -f hello.c
364 $ git checkout hello.c [1m(3)
365 1. switch branch
367 2. take a file out of another commit
368 3. restore hello.c from the index
369 If you want to check out all C source files out of the index, you
371 can say
372 $ git checkout -- '*.c'
374 Note the quotes around *.c. The file hello.c will also be checked
376 out, even though it is no longer in the working tree, because the
377 file globbing is used to match entries in the index (not in the
378 working tree by the shell).
379 If you have an unfortunate branch that is named hello.c, this step
381 would be confused as an instruction to switch to that branch. You
382 should instead write:
383 $ git checkout -- hello.c
385 2. After working in the wrong branch, switching to the correct branch
388 would be done using:
389 $ git checkout mytopic
391 However, your "wrong" branch and correct "mytopic" branch may
393 differ in files that you have modified locally, in which case the
394 above checkout would fail like this:
395 $ git checkout mytopic
397 error: You have local changes to 'frotz'; not switching branches.
398 You can give the -m flag to the command, which would try a
400 three-way merge:
401 $ git checkout -m mytopic
403 Auto-merging frotz
404 After this three-way merge, the local modifications are not
406 registered in your index file, so git diff would show you what
407 changes you made since the tip of the new branch.
408 3. When a merge conflict happens during switching branches with the -m
410 option, you would see something like this:
411 $ git checkout -m mytopic
413 Auto-merging frotz
414 ERROR: Merge conflict in frotz
415 fatal: merge program failed
416 At this point, git diff shows the changes cleanly merged as in the
418 previous example, as well as the changes in the conflicted files.
419 Edit and resolve the conflict and mark it resolved with git add as
420 usual:
421 $ edit frotz
423 $ git add frotz
424
427 Part of the git(1) suite
428Git 1.8.3.1 11/19/2018 GIT-CHECKOUT(1)