1GIT-COMMIT(1) Git Manual GIT-COMMIT(1)
2
6 git-commit - Record changes to the repository
7
9 git commit [-a | --interactive | --patch] [-s] [-v] [-u<mode>] [--amend]
10 [--dry-run] [(-c | -C | --fixup | --squash) <commit>]
11 [-F <file> | -m <msg>] [--reset-author] [--allow-empty]
12 [--allow-empty-message] [--no-verify] [-e] [--author=<author>]
13 [--date=<date>] [--cleanup=<mode>] [--[no-]status]
14 [-i | -o] [--pathspec-from-file=<file> [--pathspec-file-nul]]
15 [-S[<keyid>]] [--] [<pathspec>...]
16
18 Create a new commit containing the current contents of the index and
19 the given log message describing the changes. The new commit is a
20 direct child of HEAD, usually the tip of the current branch, and the
21 branch is updated to point to it (unless no branch is associated with
22 the working tree, in which case HEAD is "detached" as described in git-
23 checkout(1)).
24 The content to be committed can be specified in several ways:
26 1. by using git-add(1) to incrementally "add" changes to the index
28 before using the commit command (Note: even modified files must be
29 "added");
30 2. by using git-rm(1) to remove files from the working tree and the
32 index, again before using the commit command;
33 3. by listing files as arguments to the commit command (without
35 --interactive or --patch switch), in which case the commit will
36 ignore changes staged in the index, and instead record the current
37 content of the listed files (which must already be known to Git);
38 4. by using the -a switch with the commit command to automatically
40 "add" changes from all known files (i.e. all files that are already
41 listed in the index) and to automatically "rm" files in the index
42 that have been removed from the working tree, and then perform the
43 actual commit;
44 5. by using the --interactive or --patch switches with the commit
46 command to decide one by one which files or hunks should be part of
47 the commit in addition to contents in the index, before finalizing
48 the operation. See the “Interactive Mode” section of git-add(1) to
49 learn how to operate these modes.
50 The --dry-run option can be used to obtain a summary of what is
52 included by any of the above for the next commit by giving the same set
53 of parameters (options and paths).
54 If you make a commit and then find a mistake immediately after that,
56 you can recover from it with git reset.
57
59 -a, --all
60 Tell the command to automatically stage files that have been
61 modified and deleted, but new files you have not told Git about are
62 not affected.
63 -p, --patch
65 Use the interactive patch selection interface to chose which
66 changes to commit. See git-add(1) for details.
67 -C <commit>, --reuse-message=<commit>
69 Take an existing commit object, and reuse the log message and the
70 authorship information (including the timestamp) when creating the
71 commit.
72 -c <commit>, --reedit-message=<commit>
74 Like -C, but with -c the editor is invoked, so that the user can
75 further edit the commit message.
76 --fixup=<commit>
78 Construct a commit message for use with rebase --autosquash. The
79 commit message will be the subject line from the specified commit
80 with a prefix of "fixup! ". See git-rebase(1) for details.
81 --squash=<commit>
83 Construct a commit message for use with rebase --autosquash. The
84 commit message subject line is taken from the specified commit with
85 a prefix of "squash! ". Can be used with additional commit message
86 options (-m/-c/-C/-F). See git-rebase(1) for details.
87 --reset-author
89 When used with -C/-c/--amend options, or when committing after a
90 conflicting cherry-pick, declare that the authorship of the
91 resulting commit now belongs to the committer. This also renews the
92 author timestamp.
93 --short
95 When doing a dry-run, give the output in the short-format. See git-
96 status(1) for details. Implies --dry-run.
97 --branch
99 Show the branch and tracking info even in short-format.
100 --porcelain
102 When doing a dry-run, give the output in a porcelain-ready format.
103 See git-status(1) for details. Implies --dry-run.
104 --long
106 When doing a dry-run, give the output in the long-format. Implies
107 --dry-run.
108 -z, --null
110 When showing short or porcelain status output, print the filename
111 verbatim and terminate the entries with NUL, instead of LF. If no
112 format is given, implies the --porcelain output format. Without the
113 -z option, filenames with "unusual" characters are quoted as
114 explained for the configuration variable core.quotePath (see git-
115 config(1)).
116 -F <file>, --file=<file>
118 Take the commit message from the given file. Use - to read the
119 message from the standard input.
120 --author=<author>
122 Override the commit author. Specify an explicit author using the
123 standard A U Thor <author@example.com> format. Otherwise <author>
124 is assumed to be a pattern and is used to search for an existing
125 commit by that author (i.e. rev-list --all -i --author=<author>);
126 the commit author is then copied from the first such commit found.
127 --date=<date>
129 Override the author date used in the commit.
130 -m <msg>, --message=<msg>
132 Use the given <msg> as the commit message. If multiple -m options
133 are given, their values are concatenated as separate paragraphs.
134 The -m option is mutually exclusive with -c, -C, and -F.
136 -t <file>, --template=<file>
138 When editing the commit message, start the editor with the contents
139 in the given file. The commit.template configuration variable is
140 often used to give this option implicitly to the command. This
141 mechanism can be used by projects that want to guide participants
142 with some hints on what to write in the message in what order. If
143 the user exits the editor without editing the message, the commit
144 is aborted. This has no effect when a message is given by other
145 means, e.g. with the -m or -F options.
146 -s, --signoff, --no-signoff
148 Add a Signed-off-by trailer by the committer at the end of the
149 commit log message. The meaning of a signoff depends on the project
150 to which you’re committing. For example, it may certify that the
151 committer has the rights to submit the work under the project’s
152 license or agrees to some contributor representation, such as a
153 Developer Certificate of Origin. (See
154 http://developercertificate.org for the one used by the Linux
155 kernel and Git projects.) Consult the documentation or leadership
156 of the project to which you’re contributing to understand how the
157 signoffs are used in that project.
158 The --no-signoff option can be used to countermand an earlier
160 --signoff option on the command line.
161 -n, --no-verify
163 This option bypasses the pre-commit and commit-msg hooks. See also
164 githooks(5).
165 --allow-empty
167 Usually recording a commit that has the exact same tree as its sole
168 parent commit is a mistake, and the command prevents you from
169 making such a commit. This option bypasses the safety, and is
170 primarily for use by foreign SCM interface scripts.
171 --allow-empty-message
173 Like --allow-empty this command is primarily for use by foreign SCM
174 interface scripts. It allows you to create a commit with an empty
175 commit message without using plumbing commands like git-commit-
176 tree(1).
177 --cleanup=<mode>
179 This option determines how the supplied commit message should be
180 cleaned up before committing. The <mode> can be strip, whitespace,
181 verbatim, scissors or default.
182 strip
184 Strip leading and trailing empty lines, trailing whitespace,
185 commentary and collapse consecutive empty lines.
186 whitespace
188 Same as strip except #commentary is not removed.
189 verbatim
191 Do not change the message at all.
192 scissors
194 Same as whitespace except that everything from (and including)
195 the line found below is truncated, if the message is to be
196 edited. "#" can be customized with core.commentChar.
197 # ------------------------ >8 ------------------------
199 default
201 Same as strip if the message is to be edited. Otherwise
202 whitespace.
203 The default can be changed by the commit.cleanup configuration
205 variable (see git-config(1)).
206 -e, --edit
208 The message taken from file with -F, command line with -m, and from
209 commit object with -C are usually used as the commit log message
210 unmodified. This option lets you further edit the message taken
211 from these sources.
212 --no-edit
214 Use the selected commit message without launching an editor. For
215 example, git commit --amend --no-edit amends a commit without
216 changing its commit message.
217 --amend
219 Replace the tip of the current branch by creating a new commit. The
220 recorded tree is prepared as usual (including the effect of the -i
221 and -o options and explicit pathspec), and the message from the
222 original commit is used as the starting point, instead of an empty
223 message, when no other message is specified from the command line
224 via options such as -m, -F, -c, etc. The new commit has the same
225 parents and author as the current one (the --reset-author option
226 can countermand this).
227 It is a rough equivalent for:
229 $ git reset --soft HEAD^
231 $ ... do something else to come up with the right tree ...
232 $ git commit -c ORIG_HEAD
233 but can be used to amend a merge commit.
235 You should understand the implications of rewriting history if you
237 amend a commit that has already been published. (See the
238 "RECOVERING FROM UPSTREAM REBASE" section in git-rebase(1).)
239 --no-post-rewrite
241 Bypass the post-rewrite hook.
242 -i, --include
244 Before making a commit out of staged contents so far, stage the
245 contents of paths given on the command line as well. This is
246 usually not what you want unless you are concluding a conflicted
247 merge.
248 -o, --only
250 Make a commit by taking the updated working tree contents of the
251 paths specified on the command line, disregarding any contents that
252 have been staged for other paths. This is the default mode of
253 operation of git commit if any paths are given on the command line,
254 in which case this option can be omitted. If this option is
255 specified together with --amend, then no paths need to be
256 specified, which can be used to amend the last commit without
257 committing changes that have already been staged. If used together
258 with --allow-empty paths are also not required, and an empty commit
259 will be created.
260 --pathspec-from-file=<file>
262 Pathspec is passed in <file> instead of commandline args. If <file>
263 is exactly - then standard input is used. Pathspec elements are
264 separated by LF or CR/LF. Pathspec elements can be quoted as
265 explained for the configuration variable core.quotePath (see git-
266 config(1)). See also --pathspec-file-nul and global
267 --literal-pathspecs.
268 --pathspec-file-nul
270 Only meaningful with --pathspec-from-file. Pathspec elements are
271 separated with NUL character and all other characters are taken
272 literally (including newlines and quotes).
273 -u[<mode>], --untracked-files[=<mode>]
275 Show untracked files.
276 The mode parameter is optional (defaults to all), and is used to
278 specify the handling of untracked files; when -u is not used, the
279 default is normal, i.e. show untracked files and directories.
280 The possible options are:
282 • no - Show no untracked files
284 • normal - Shows untracked files and directories
286 • all - Also shows individual files in untracked directories.
288 The default can be changed using the status.showUntrackedFiles
290 configuration variable documented in git-config(1).
291 -v, --verbose
293 Show unified diff between the HEAD commit and what would be
294 committed at the bottom of the commit message template to help the
295 user describe the commit by reminding what changes the commit has.
296 Note that this diff output doesn’t have its lines prefixed with #.
297 This diff will not be a part of the commit message. See the
298 commit.verbose configuration variable in git-config(1).
299 If specified twice, show in addition the unified diff between what
301 would be committed and the worktree files, i.e. the unstaged
302 changes to tracked files.
303 -q, --quiet
305 Suppress commit summary message.
306 --dry-run
308 Do not create a commit, but show a list of paths that are to be
309 committed, paths with local changes that will be left uncommitted
310 and paths that are untracked.
311 --status
313 Include the output of git-status(1) in the commit message template
314 when using an editor to prepare the commit message. Defaults to on,
315 but can be used to override configuration variable commit.status.
316 --no-status
318 Do not include the output of git-status(1) in the commit message
319 template when using an editor to prepare the default commit
320 message.
321 -S[<keyid>], --gpg-sign[=<keyid>], --no-gpg-sign
323 GPG-sign commits. The keyid argument is optional and defaults to
324 the committer identity; if specified, it must be stuck to the
325 option without a space. --no-gpg-sign is useful to countermand
326 both commit.gpgSign configuration variable, and earlier --gpg-sign.
327 --
329 Do not interpret any more arguments as options.
330 <pathspec>...
332 When pathspec is given on the command line, commit the contents of
333 the files that match the pathspec without recording the changes
334 already added to the index. The contents of these files are also
335 staged for the next commit on top of what have been staged before.
336 For more details, see the pathspec entry in gitglossary(7).
338
340 When recording your own work, the contents of modified files in your
341 working tree are temporarily stored to a staging area called the
342 "index" with git add. A file can be reverted back, only in the index
343 but not in the working tree, to that of the last commit with git
344 restore --staged <file>, which effectively reverts git add and prevents
345 the changes to this file from participating in the next commit. After
346 building the state to be committed incrementally with these commands,
347 git commit (without any pathname parameter) is used to record what has
348 been staged so far. This is the most basic form of the command. An
349 example:
350 $ edit hello.c
352 $ git rm goodbye.c
353 $ git add hello.c
354 $ git commit
355 Instead of staging files after each individual change, you can tell git
357 commit to notice the changes to the files whose contents are tracked in
358 your working tree and do corresponding git add and git rm for you. That
359 is, this example does the same as the earlier example if there is no
360 other change in your working tree:
361 $ edit hello.c
363 $ rm goodbye.c
364 $ git commit -a
365 The command git commit -a first looks at your working tree, notices
367 that you have modified hello.c and removed goodbye.c, and performs
368 necessary git add and git rm for you.
369 After staging changes to many files, you can alter the order the
371 changes are recorded in, by giving pathnames to git commit. When
372 pathnames are given, the command makes a commit that only records the
373 changes made to the named paths:
374 $ edit hello.c hello.h
376 $ git add hello.c hello.h
377 $ edit Makefile
378 $ git commit Makefile
379 This makes a commit that records the modification to Makefile. The
381 changes staged for hello.c and hello.h are not included in the
382 resulting commit. However, their changes are not lost — they are still
383 staged and merely held back. After the above sequence, if you do:
384 $ git commit
386 this second commit would record the changes to hello.c and hello.h as
388 expected.
389 After a merge (initiated by git merge or git pull) stops because of
391 conflicts, cleanly merged paths are already staged to be committed for
392 you, and paths that conflicted are left in unmerged state. You would
393 have to first check which paths are conflicting with git status and
394 after fixing them manually in your working tree, you would stage the
395 result as usual with git add:
396 $ git status | grep unmerged
398 unmerged: hello.c
399 $ edit hello.c
400 $ git add hello.c
401 After resolving conflicts and staging the result, git ls-files -u would
403 stop mentioning the conflicted path. When you are done, run git commit
404 to finally record the merge:
405 $ git commit
407 As with the case to record your own changes, you can use -a option to
409 save typing. One difference is that during a merge resolution, you
410 cannot use git commit with pathnames to alter the order the changes are
411 committed, because the merge should be recorded as a single commit. In
412 fact, the command refuses to run when given pathnames (but see -i
413 option).
414
416 Author and committer information is taken from the following
417 environment variables, if set:
418 GIT_AUTHOR_NAME
420 GIT_AUTHOR_EMAIL
421 GIT_AUTHOR_DATE
422 GIT_COMMITTER_NAME
423 GIT_COMMITTER_EMAIL
424 GIT_COMMITTER_DATE
425 (nb "<", ">" and "\n"s are stripped)
427 The author and committer names are by convention some form of a
429 personal name (that is, the name by which other humans refer to you),
430 although Git does not enforce or require any particular form. Arbitrary
431 Unicode may be used, subject to the constraints listed above. This name
432 has no effect on authentication; for that, see the credential.username
433 variable in git-config(1).
434 In case (some of) these environment variables are not set, the
436 information is taken from the configuration items user.name and
437 user.email, or, if not present, the environment variable EMAIL, or, if
438 that is not set, system user name and the hostname used for outgoing
439 mail (taken from /etc/mailname and falling back to the fully qualified
440 hostname when that file does not exist).
441 The author.name and committer.name and their corresponding email
443 options override user.name and user.email if set and are overridden
444 themselves by the environment variables.
445 The typical usage is to set just the user.name and user.email
447 variables; the other options are provided for more complex use cases.
448
450 The GIT_AUTHOR_DATE and GIT_COMMITTER_DATE environment variables
451 support the following date formats:
452 Git internal format
454 It is <unix timestamp> <time zone offset>, where <unix timestamp>
455 is the number of seconds since the UNIX epoch. <time zone offset>
456 is a positive or negative offset from UTC. For example CET (which
457 is 1 hour ahead of UTC) is +0100.
458 RFC 2822
460 The standard email format as described by RFC 2822, for example
461 Thu, 07 Apr 2005 22:13:13 +0200.
462 ISO 8601
464 Time and date specified by the ISO 8601 standard, for example
465 2005-04-07T22:13:13. The parser accepts a space instead of the T
466 character as well. Fractional parts of a second will be ignored,
467 for example 2005-04-07T22:13:13.019 will be treated as
468 2005-04-07T22:13:13.
469 Note
471 In addition, the date part is accepted in the following
472 formats: YYYY.MM.DD, MM/DD/YYYY and DD.MM.YYYY.
473 In addition to recognizing all date formats above, the --date option
475 will also try to make sense of other, more human-centric date formats,
476 such as relative dates like "yesterday" or "last Friday at noon".
477
479 Though not required, it’s a good idea to begin the commit message with
480 a single short (less than 50 character) line summarizing the change,
481 followed by a blank line and then a more thorough description. The text
482 up to the first blank line in a commit message is treated as the commit
483 title, and that title is used throughout Git. For example, git-format-
484 patch(1) turns a commit into email, and it uses the title on the
485 Subject line and the rest of the commit in the body.
486 Git is to some extent character encoding agnostic.
488 • The contents of the blob objects are uninterpreted sequences of
490 bytes. There is no encoding translation at the core level.
491 • Path names are encoded in UTF-8 normalization form C. This applies
493 to tree objects, the index file, ref names, as well as path names
494 in command line arguments, environment variables and config files
495 (.git/config (see git-config(1)), gitignore(5), gitattributes(5)
496 and gitmodules(5)).
497 Note that Git at the core level treats path names simply as
499 sequences of non-NUL bytes, there are no path name encoding
500 conversions (except on Mac and Windows). Therefore, using non-ASCII
501 path names will mostly work even on platforms and file systems that
502 use legacy extended ASCII encodings. However, repositories created
503 on such systems will not work properly on UTF-8-based systems (e.g.
504 Linux, Mac, Windows) and vice versa. Additionally, many Git-based
505 tools simply assume path names to be UTF-8 and will fail to display
506 other encodings correctly.
507 • Commit log messages are typically encoded in UTF-8, but other
509 extended ASCII encodings are also supported. This includes
510 ISO-8859-x, CP125x and many others, but not UTF-16/32, EBCDIC and
511 CJK multi-byte encodings (GBK, Shift-JIS, Big5, EUC-x, CP9xx etc.).
512 Although we encourage that the commit log messages are encoded in
514 UTF-8, both the core and Git Porcelain are designed not to force UTF-8
515 on projects. If all participants of a particular project find it more
516 convenient to use legacy encodings, Git does not forbid it. However,
517 there are a few things to keep in mind.
518 1. git commit and git commit-tree issues a warning if the commit log
520 message given to it does not look like a valid UTF-8 string, unless
521 you explicitly say your project uses a legacy encoding. The way to
522 say this is to have i18n.commitEncoding in .git/config file, like
523 this:
524 [i18n]
526 commitEncoding = ISO-8859-1
527 Commit objects created with the above setting record the value of
529 i18n.commitEncoding in its encoding header. This is to help other
530 people who look at them later. Lack of this header implies that the
531 commit log message is encoded in UTF-8.
532 2. git log, git show, git blame and friends look at the encoding
534 header of a commit object, and try to re-code the log message into
535 UTF-8 unless otherwise specified. You can specify the desired
536 output encoding with i18n.logOutputEncoding in .git/config file,
537 like this:
538 [i18n]
540 logOutputEncoding = ISO-8859-1
541 If you do not have this configuration variable, the value of
543 i18n.commitEncoding is used instead.
544 Note that we deliberately chose not to re-code the commit log message
546 when a commit is made to force UTF-8 at the commit object level,
547 because re-coding to UTF-8 is not necessarily a reversible operation.
548
550 The editor used to edit the commit log message will be chosen from the
551 GIT_EDITOR environment variable, the core.editor configuration
552 variable, the VISUAL environment variable, or the EDITOR environment
553 variable (in that order). See git-var(1) for details.
554
556 This command can run commit-msg, prepare-commit-msg, pre-commit,
557 post-commit and post-rewrite hooks. See githooks(5) for more
558 information.
559
561 $GIT_DIR/COMMIT_EDITMSG
562 This file contains the commit message of a commit in progress. If
563 git commit exits due to an error before creating a commit, any
564 commit message that has been provided by the user (e.g., in an
565 editor session) will be available in this file, but will be
566 overwritten by the next invocation of git commit.
567
569 git-add(1), git-rm(1), git-mv(1), git-merge(1), git-commit-tree(1)
570
572 Part of the git(1) suite
573Git 2.31.1 2021-03-26 GIT-COMMIT(1)