1GIT-COMMIT(1) Git Manual GIT-COMMIT(1)
2
6 git-commit - Record changes to the repository
7
9 git commit [-a | --interactive] [-s] [-v] [-u<mode>] [--amend] [--dry-run]
10 [(-c | -C | --fixup | --squash) <commit>] [-F <file> | -m <msg>]
11 [--reset-author] [--allow-empty] [--allow-empty-message] [--no-verify]
12 [-e] [--author=<author>] [--date=<date>] [--cleanup=<mode>]
13 [--status | --no-status] [-i | -o] [--] [<file>...]
14
17 Stores the current contents of the index in a new commit along with a
18 log message from the user describing the changes.
19 The content to be added can be specified in several ways:
21 1. by using git add to incrementally "add" changes to the index before
23 using the commit command (Note: even modified files must be
24 "added");
25 2. by using git rm to remove files from the working tree and the
27 index, again before using the commit command;
28 3. by listing files as arguments to the commit command, in which case
30 the commit will ignore changes staged in the index, and instead
31 record the current content of the listed files (which must already
32 be known to git);
33 4. by using the -a switch with the commit command to automatically
35 "add" changes from all known files (i.e. all files that are already
36 listed in the index) and to automatically "rm" files in the index
37 that have been removed from the working tree, and then perform the
38 actual commit;
39 5. by using the --interactive switch with the commit command to decide
41 one by one which files should be part of the commit, before
42 finalizing the operation. Currently, this is done by invoking git
43 add --interactive.
44 The --dry-run option can be used to obtain a summary of what is
46 included by any of the above for the next commit by giving the same set
47 of parameters (options and paths).
48 If you make a commit and then find a mistake immediately after that,
50 you can recover from it with git reset.
51
53 -a, --all
54 Tell the command to automatically stage files that have been
55 modified and deleted, but new files you have not told git about are
56 not affected.
57 -C <commit>, --reuse-message=<commit>
59 Take an existing commit object, and reuse the log message and the
60 authorship information (including the timestamp) when creating the
61 commit.
62 -c <commit>, --reedit-message=<commit>
64 Like -C, but with -c the editor is invoked, so that the user can
65 further edit the commit message.
66 --fixup=<commit>
68 Construct a commit message for use with rebase --autosquash. The
69 commit message will be the subject line from the specified commit
70 with a prefix of "fixup! ". See git-rebase(1) for details.
71 --squash=<commit>
73 Construct a commit message for use with rebase --autosquash. The
74 commit message subject line is taken from the specified commit with
75 a prefix of "squash! ". Can be used with additional commit message
76 options (-m/-c/-C/-F). See git-rebase(1) for details.
77 --reset-author
79 When used with -C/-c/--amend options, declare that the authorship
80 of the resulting commit now belongs of the committer. This also
81 renews the author timestamp.
82 --short
84 When doing a dry-run, give the output in the short-format. See git-
85 status(1) for details. Implies --dry-run.
86 --porcelain
88 When doing a dry-run, give the output in a porcelain-ready format.
89 See git-status(1) for details. Implies --dry-run.
90 -z
92 When showing short or porcelain status output, terminate entries in
93 the status output with NUL, instead of LF. If no format is given,
94 implies the --porcelain output format.
95 -F <file>, --file=<file>
97 Take the commit message from the given file. Use - to read the
98 message from the standard input.
99 --author=<author>
101 Override the commit author. Specify an explicit author using the
102 standard A U Thor <author@example.com[1]> format. Otherwise
103 <author> is assumed to be a pattern and is used to search for an
104 existing commit by that author (i.e. rev-list --all -i
105 --author=<author>); the commit author is then copied from the first
106 such commit found.
107 --date=<date>
109 Override the author date used in the commit.
110 -m <msg>, --message=<msg>
112 Use the given <msg> as the commit message.
113 -t <file>, --template=<file>
115 Use the contents of the given file as the initial version of the
116 commit message. The editor is invoked and you can make subsequent
117 changes. If a message is specified using the -m or -F options, this
118 option has no effect. This overrides the commit.template
119 configuration variable.
120 -s, --signoff
122 Add Signed-off-by line by the committer at the end of the commit
123 log message.
124 -n, --no-verify
126 This option bypasses the pre-commit and commit-msg hooks. See also
127 githooks(5).
128 --allow-empty
130 Usually recording a commit that has the exact same tree as its sole
131 parent commit is a mistake, and the command prevents you from
132 making such a commit. This option bypasses the safety, and is
133 primarily for use by foreign SCM interface scripts.
134 --allow-empty-message
136 Like --allow-empty this command is primarily for use by foreign SCM
137 interface scripts. It allows you to create a commit with an empty
138 commit message without using plumbing commands like git-commit-
139 tree(1).
140 --cleanup=<mode>
142 This option sets how the commit message is cleaned up. The <mode>
143 can be one of verbatim, whitespace, strip, and default. The default
144 mode will strip leading and trailing empty lines and #commentary
145 from the commit message only if the message is to be edited.
146 Otherwise only whitespace removed. The verbatim mode does not
147 change message at all, whitespace removes just leading/trailing
148 whitespace lines and strip removes both whitespace and commentary.
149 -e, --edit
151 The message taken from file with -F, command line with -m, and from
152 file with -C are usually used as the commit log message unmodified.
153 This option lets you further edit the message taken from these
154 sources.
155 --amend
157 Used to amend the tip of the current branch. Prepare the tree
158 object you would want to replace the latest commit as usual (this
159 includes the usual -i/-o and explicit paths), and the commit log
160 editor is seeded with the commit message from the tip of the
161 current branch. The commit you create replaces the current tip — if
162 it was a merge, it will have the parents of the current tip as
163 parents — so the current top commit is discarded.
164 It is a rough equivalent for:
166 $ git reset --soft HEAD^
168 $ ... do something else to come up with the right tree ...
169 $ git commit -c ORIG_HEAD
170 but can be used to amend a merge commit.
172 You should understand the implications of rewriting history if you
174 amend a commit that has already been published. (See the
175 "RECOVERING FROM UPSTREAM REBASE" section in git-rebase(1).)
176 -i, --include
178 Before making a commit out of staged contents so far, stage the
179 contents of paths given on the command line as well. This is
180 usually not what you want unless you are concluding a conflicted
181 merge.
182 -o, --only
184 Make a commit only from the paths specified on the command line,
185 disregarding any contents that have been staged so far. This is the
186 default mode of operation of git commit if any paths are given on
187 the command line, in which case this option can be omitted. If this
188 option is specified together with --amend, then no paths need to be
189 specified, which can be used to amend the last commit without
190 committing changes that have already been staged.
191 -u[<mode>], --untracked-files[=<mode>]
193 Show untracked files.
194 The mode parameter is optional (defaults to all), and is used to
196 specify the handling of untracked files; when -u is not used, the
197 default is normal, i.e. show untracked files and directories.
198 The possible options are:
200 · no - Show no untracked files
202 · normal - Shows untracked files and directories
204 · all - Also shows individual files in untracked directories.
206 The default can be changed using the status.showUntrackedFiles
208 configuration variable documented in git-config(1).
209 -v, --verbose
211 Show unified diff between the HEAD commit and what would be
212 committed at the bottom of the commit message template. Note that
213 this diff output doesn’t have its lines prefixed with #.
214 -q, --quiet
216 Suppress commit summary message.
217 --dry-run
219 Do not create a commit, but show a list of paths that are to be
220 committed, paths with local changes that will be left uncommitted
221 and paths that are untracked.
222 --status
224 Include the output of git-status(1) in the commit message template
225 when using an editor to prepare the commit message. Defaults to on,
226 but can be used to override configuration variable commit.status.
227 --no-status
229 Do not include the output of git-status(1) in the commit message
230 template when using an editor to prepare the default commit
231 message.
232 --
234 Do not interpret any more arguments as options.
235 <file>...
237 When files are given on the command line, the command commits the
238 contents of the named files, without recording the changes already
239 staged. The contents of these files are also staged for the next
240 commit on top of what have been staged before.
241
243 The GIT_AUTHOR_DATE, GIT_COMMITTER_DATE environment variables and the
244 --date option support the following date formats:
245 Git internal format
247 It is <unix timestamp> <timezone offset>, where <unix timestamp> is
248 the number of seconds since the UNIX epoch. <timezone offset> is a
249 positive or negative offset from UTC. For example CET (which is 2
250 hours ahead UTC) is +0200.
251 RFC 2822
253 The standard email format as described by RFC 2822, for example
254 Thu, 07 Apr 2005 22:13:13 +0200.
255 ISO 8601
257 Time and date specified by the ISO 8601 standard, for example
258 2005-04-07T22:13:13. The parser accepts a space instead of the T
259 character as well.
260 Note
262 In addition, the date part is accepted in the following
263 formats: YYYY.MM.DD, MM/DD/YYYY and DD.MM.YYYY.
264
266 When recording your own work, the contents of modified files in your
267 working tree are temporarily stored to a staging area called the
268 "index" with git add. A file can be reverted back, only in the index
269 but not in the working tree, to that of the last commit with git reset
270 HEAD — <file>, which effectively reverts git add and prevents the
271 changes to this file from participating in the next commit. After
272 building the state to be committed incrementally with these commands,
273 git commit (without any pathname parameter) is used to record what has
274 been staged so far. This is the most basic form of the command. An
275 example:
276 $ edit hello.c
278 $ git rm goodbye.c
279 $ git add hello.c
280 $ git commit
281 Instead of staging files after each individual change, you can tell git
284 commit to notice the changes to the files whose contents are tracked in
285 your working tree and do corresponding git add and git rm for you. That
286 is, this example does the same as the earlier example if there is no
287 other change in your working tree:
288 $ edit hello.c
290 $ rm goodbye.c
291 $ git commit -a
292 The command git commit -a first looks at your working tree, notices
295 that you have modified hello.c and removed goodbye.c, and performs
296 necessary git add and git rm for you.
297 After staging changes to many files, you can alter the order the
299 changes are recorded in, by giving pathnames to git commit. When
300 pathnames are given, the command makes a commit that only records the
301 changes made to the named paths:
302 $ edit hello.c hello.h
304 $ git add hello.c hello.h
305 $ edit Makefile
306 $ git commit Makefile
307 This makes a commit that records the modification to Makefile. The
310 changes staged for hello.c and hello.h are not included in the
311 resulting commit. However, their changes are not lost — they are still
312 staged and merely held back. After the above sequence, if you do:
313 $ git commit
315 this second commit would record the changes to hello.c and hello.h as
318 expected.
319 After a merge (initiated by git merge or git pull) stops because of
321 conflicts, cleanly merged paths are already staged to be committed for
322 you, and paths that conflicted are left in unmerged state. You would
323 have to first check which paths are conflicting with git status and
324 after fixing them manually in your working tree, you would stage the
325 result as usual with git add:
326 $ git status | grep unmerged
328 unmerged: hello.c
329 $ edit hello.c
330 $ git add hello.c
331 After resolving conflicts and staging the result, git ls-files -u would
334 stop mentioning the conflicted path. When you are done, run git commit
335 to finally record the merge:
336 $ git commit
338 As with the case to record your own changes, you can use -a option to
341 save typing. One difference is that during a merge resolution, you
342 cannot use git commit with pathnames to alter the order the changes are
343 committed, because the merge should be recorded as a single commit. In
344 fact, the command refuses to run when given pathnames (but see -i
345 option).
346
348 Though not required, it’s a good idea to begin the commit message with
349 a single short (less than 50 character) line summarizing the change,
350 followed by a blank line and then a more thorough description. Tools
351 that turn commits into email, for example, use the first line on the
352 Subject: line and the rest of the commit in the body.
353 At the core level, git is character encoding agnostic.
355 · The pathnames recorded in the index and in the tree objects are
357 treated as uninterpreted sequences of non-NUL bytes. What
358 readdir(2) returns are what are recorded and compared with the data
359 git keeps track of, which in turn are expected to be what lstat(2)
360 and creat(2) accepts. There is no such thing as pathname encoding
361 translation.
362 · The contents of the blob objects are uninterpreted sequences of
364 bytes. There is no encoding translation at the core level.
365 · The commit log messages are uninterpreted sequences of non-NUL
367 bytes.
368 Although we encourage that the commit log messages are encoded in
370 UTF-8, both the core and git Porcelain are designed not to force UTF-8
371 on projects. If all participants of a particular project find it more
372 convenient to use legacy encodings, git does not forbid it. However,
373 there are a few things to keep in mind.
374 1. git commit and git commit-tree issues a warning if the commit log
376 message given to it does not look like a valid UTF-8 string, unless
377 you explicitly say your project uses a legacy encoding. The way to
378 say this is to have i18n.commitencoding in .git/config file, like
379 this:
380 [i18n]
382 commitencoding = ISO-8859-1
383 Commit objects created with the above setting record the value of
385 i18n.commitencoding in its encoding header. This is to help other
386 people who look at them later. Lack of this header implies that the
387 commit log message is encoded in UTF-8.
388 2. git log, git show, git blame and friends look at the encoding
390 header of a commit object, and try to re-code the log message into
391 UTF-8 unless otherwise specified. You can specify the desired
392 output encoding with i18n.logoutputencoding in .git/config file,
393 like this:
394 [i18n]
396 logoutputencoding = ISO-8859-1
397 If you do not have this configuration variable, the value of
399 i18n.commitencoding is used instead.
400 Note that we deliberately chose not to re-code the commit log message
402 when a commit is made to force UTF-8 at the commit object level,
403 because re-coding to UTF-8 is not necessarily a reversible operation.
404
406 The editor used to edit the commit log message will be chosen from the
407 GIT_EDITOR environment variable, the core.editor configuration
408 variable, the VISUAL environment variable, or the EDITOR environment
409 variable (in that order). See git-var(1) for details.
410
412 This command can run commit-msg, prepare-commit-msg, pre-commit, and
413 post-commit hooks. See githooks(5) for more information.
414
416 git-add(1), git-rm(1), git-mv(1), git-merge(1), git-commit-tree(1)
417
419 Written by Linus Torvalds <torvalds@osdl.org[2]> and Junio C Hamano
420 <gitster@pobox.com[3]>
421
423 Part of the git(1) suite
424
426 1. author@example.com
427 mailto:author@example.com
428 2. torvalds@osdl.org
430 mailto:torvalds@osdl.org
431 3. gitster@pobox.com
433 mailto:gitster@pobox.com
434Git 1.7.4.4 04/11/2011 GIT-COMMIT(1)