compat/fnmatch: update old-style definition to ANSI
[git/git.git] / Documentation / git-commit.txt
1 git-commit(1)
2 =============
5 ----
6 git-commit - Record changes to the repository
9 --------
10 [verse]
11 'git commit' [-a | --interactive | --patch] [-s] [-v] [-u<mode>] [--amend]
12 [--dry-run] [(-c | -C | --fixup | --squash) <commit>]
13 [-F <file> | -m <msg>] [--reset-author] [--allow-empty]
14 [--allow-empty-message] [--no-verify] [-e] [--author=<author>]
15 [--date=<date>] [--cleanup=<mode>] [--status | --no-status]
16 [-i | -o] [--] [<file>...]
19 -----------
20 Stores the current contents of the index in a new commit along
21 with a log message from the user describing the changes.
23 The content to be added can be specified in several ways:
25 1. by using 'git add' to incrementally "add" changes to the
26 index before using the 'commit' command (Note: even modified
27 files must be "added");
29 2. by using 'git rm' to remove files from the working tree
30 and the index, again before using the 'commit' command;
32 3. by listing files as arguments to the 'commit' command, in which
33 case the commit will ignore changes staged in the index, and instead
34 record the current content of the listed files (which must already
35 be known to git);
37 4. by using the -a switch with the 'commit' command to automatically
38 "add" changes from all known files (i.e. all files that are already
39 listed in the index) and to automatically "rm" files in the index
40 that have been removed from the working tree, and then perform the
41 actual commit;
43 5. by using the --interactive or --patch switches with the 'commit' command
44 to decide one by one which files or hunks should be part of the commit,
45 before finalizing the operation. See the ``Interactive Mode'' section of
46 linkgit:git-add[1] to learn how to operate these modes.
48 The `--dry-run` option can be used to obtain a
49 summary of what is included by any of the above for the next
50 commit by giving the same set of parameters (options and paths).
52 If you make a commit and then find a mistake immediately after
53 that, you can recover from it with 'git reset'.
57 -------
58 -a::
59 --all::
60 Tell the command to automatically stage files that have
61 been modified and deleted, but new files you have not
62 told git about are not affected.
64 -p::
65 --patch::
66 Use the interactive patch selection interface to chose
67 which changes to commit. See linkgit:git-add[1] for
68 details.
70 -C <commit>::
71 --reuse-message=<commit>::
72 Take an existing commit object, and reuse the log message
73 and the authorship information (including the timestamp)
74 when creating the commit.
76 -c <commit>::
77 --reedit-message=<commit>::
78 Like '-C', but with '-c' the editor is invoked, so that
79 the user can further edit the commit message.
81 --fixup=<commit>::
82 Construct a commit message for use with `rebase --autosquash`.
83 The commit message will be the subject line from the specified
84 commit with a prefix of "fixup! ". See linkgit:git-rebase[1]
85 for details.
87 --squash=<commit>::
88 Construct a commit message for use with `rebase --autosquash`.
89 The commit message subject line is taken from the specified
90 commit with a prefix of "squash! ". Can be used with additional
91 commit message options (`-m`/`-c`/`-C`/`-F`). See
92 linkgit:git-rebase[1] for details.
94 --reset-author::
95 When used with -C/-c/--amend options, or when committing after a
96 a conflicting cherry-pick, declare that the authorship of the
97 resulting commit now belongs of the committer. This also renews
98 the author timestamp.
100 --short::
101 When doing a dry-run, give the output in the short-format. See
102 linkgit:git-status[1] for details. Implies `--dry-run`.
104 --branch::
105 Show the branch and tracking info even in short-format.
107 --porcelain::
108 When doing a dry-run, give the output in a porcelain-ready
109 format. See linkgit:git-status[1] for details. Implies
110 `--dry-run`.
112 -z::
113 --null::
114 When showing `short` or `porcelain` status output, terminate
115 entries in the status output with NUL, instead of LF. If no
116 format is given, implies the `--porcelain` output format.
118 -F <file>::
119 --file=<file>::
120 Take the commit message from the given file. Use '-' to
121 read the message from the standard input.
123 --author=<author>::
124 Override the commit author. Specify an explicit author using the
125 standard `A U Thor <>` format. Otherwise <author>
126 is assumed to be a pattern and is used to search for an existing
127 commit by that author (i.e. rev-list --all -i --author=<author>);
128 the commit author is then copied from the first such commit found.
130 --date=<date>::
131 Override the author date used in the commit.
133 -m <msg>::
134 --message=<msg>::
135 Use the given <msg> as the commit message.
137 -t <file>::
138 --template=<file>::
139 When editing the commit message, start the editor with the
140 contents in the given file. The `commit.template` configuration
141 variable is often used to give this option implicitly to the
142 command. This mechanism can be used by projects that want to
143 guide participants with some hints on what to write in the message
144 in what order. If the user exits the editor without editing the
145 message, the commit is aborted. This has no effect when a message
146 is given by other means, e.g. with the `-m` or `-F` options.
148 -s::
149 --signoff::
150 Add Signed-off-by line by the committer at the end of the commit
151 log message.
153 -n::
154 --no-verify::
155 This option bypasses the pre-commit and commit-msg hooks.
156 See also linkgit:githooks[5].
158 --allow-empty::
159 Usually recording a commit that has the exact same tree as its
160 sole parent commit is a mistake, and the command prevents you
161 from making such a commit. This option bypasses the safety, and
162 is primarily for use by foreign SCM interface scripts.
164 --allow-empty-message::
165 Like --allow-empty this command is primarily for use by foreign
166 SCM interface scripts. It allows you to create a commit with an
167 empty commit message without using plumbing commands like
168 linkgit:git-commit-tree[1].
170 --cleanup=<mode>::
171 This option sets how the commit message is cleaned up.
172 The '<mode>' can be one of 'verbatim', 'whitespace', 'strip',
173 and 'default'. The 'default' mode will strip leading and
174 trailing empty lines and #commentary from the commit message
175 only if the message is to be edited. Otherwise only whitespace
176 removed. The 'verbatim' mode does not change message at all,
177 'whitespace' removes just leading/trailing whitespace lines
178 and 'strip' removes both whitespace and commentary.
180 -e::
181 --edit::
182 The message taken from file with `-F`, command line with
183 `-m`, and from file with `-C` are usually used as the
184 commit log message unmodified. This option lets you
185 further edit the message taken from these sources.
187 --no-edit::
188 Use the selected commit message without launching an editor.
189 For example, `git commit --amend --no-edit` amends a commit
190 without changing its commit message.
192 --amend::
193 Used to amend the tip of the current branch. Prepare the tree
194 object you would want to replace the latest commit as usual
195 (this includes the usual -i/-o and explicit paths), and the
196 commit log editor is seeded with the commit message from the
197 tip of the current branch. The commit you create replaces the
198 current tip -- if it was a merge, it will have the parents of
199 the current tip as parents -- so the current top commit is
200 discarded.
201 +
202 --
203 It is a rough equivalent for:
204 ------
205 $ git reset --soft HEAD^
206 $ ... do something else to come up with the right tree ...
207 $ git commit -c ORIG_HEAD
209 ------
210 but can be used to amend a merge commit.
211 --
212 +
213 You should understand the implications of rewriting history if you
214 amend a commit that has already been published. (See the "RECOVERING
215 FROM UPSTREAM REBASE" section in linkgit:git-rebase[1].)
217 --no-post-rewrite::
218 Bypass the post-rewrite hook.
220 -i::
221 --include::
222 Before making a commit out of staged contents so far,
223 stage the contents of paths given on the command line
224 as well. This is usually not what you want unless you
225 are concluding a conflicted merge.
227 -o::
228 --only::
229 Make a commit only from the paths specified on the
230 command line, disregarding any contents that have been
231 staged so far. This is the default mode of operation of
232 'git commit' if any paths are given on the command line,
233 in which case this option can be omitted.
234 If this option is specified together with '--amend', then
235 no paths need to be specified, which can be used to amend
236 the last commit without committing changes that have
237 already been staged.
239 -u[<mode>]::
240 --untracked-files[=<mode>]::
241 Show untracked files.
242 +
243 The mode parameter is optional (defaults to 'all'), and is used to
244 specify the handling of untracked files; when -u is not used, the
245 default is 'normal', i.e. show untracked files and directories.
246 +
247 The possible options are:
248 +
249 - 'no' - Show no untracked files
250 - 'normal' - Shows untracked files and directories
251 - 'all' - Also shows individual files in untracked directories.
252 +
253 The default can be changed using the status.showUntrackedFiles
254 configuration variable documented in linkgit:git-config[1].
256 -v::
257 --verbose::
258 Show unified diff between the HEAD commit and what
259 would be committed at the bottom of the commit message
260 template. Note that this diff output doesn't have its
261 lines prefixed with '#'.
263 -q::
264 --quiet::
265 Suppress commit summary message.
267 --dry-run::
268 Do not create a commit, but show a list of paths that are
269 to be committed, paths with local changes that will be left
270 uncommitted and paths that are untracked.
272 --status::
273 Include the output of linkgit:git-status[1] in the commit
274 message template when using an editor to prepare the commit
275 message. Defaults to on, but can be used to override
276 configuration variable commit.status.
278 --no-status::
279 Do not include the output of linkgit:git-status[1] in the
280 commit message template when using an editor to prepare the
281 default commit message.
283 \--::
284 Do not interpret any more arguments as options.
286 <file>...::
287 When files are given on the command line, the command
288 commits the contents of the named files, without
289 recording the changes already staged. The contents of
290 these files are also staged for the next commit on top
291 of what have been staged before.
293 :git-commit: 1
294 include::date-formats.txt[]
297 --------
298 When recording your own work, the contents of modified files in
299 your working tree are temporarily stored to a staging area
300 called the "index" with 'git add'. A file can be
301 reverted back, only in the index but not in the working tree,
302 to that of the last commit with `git reset HEAD -- <file>`,
303 which effectively reverts 'git add' and prevents the changes to
304 this file from participating in the next commit. After building
305 the state to be committed incrementally with these commands,
306 `git commit` (without any pathname parameter) is used to record what
307 has been staged so far. This is the most basic form of the
308 command. An example:
310 ------------
311 $ edit hello.c
312 $ git rm goodbye.c
313 $ git add hello.c
314 $ git commit
315 ------------
317 Instead of staging files after each individual change, you can
318 tell `git commit` to notice the changes to the files whose
319 contents are tracked in
320 your working tree and do corresponding `git add` and `git rm`
321 for you. That is, this example does the same as the earlier
322 example if there is no other change in your working tree:
324 ------------
325 $ edit hello.c
326 $ rm goodbye.c
327 $ git commit -a
328 ------------
330 The command `git commit -a` first looks at your working tree,
331 notices that you have modified hello.c and removed goodbye.c,
332 and performs necessary `git add` and `git rm` for you.
334 After staging changes to many files, you can alter the order the
335 changes are recorded in, by giving pathnames to `git commit`.
336 When pathnames are given, the command makes a commit that
337 only records the changes made to the named paths:
339 ------------
340 $ edit hello.c hello.h
341 $ git add hello.c hello.h
342 $ edit Makefile
343 $ git commit Makefile
344 ------------
346 This makes a commit that records the modification to `Makefile`.
347 The changes staged for `hello.c` and `hello.h` are not included
348 in the resulting commit. However, their changes are not lost --
349 they are still staged and merely held back. After the above
350 sequence, if you do:
352 ------------
353 $ git commit
354 ------------
356 this second commit would record the changes to `hello.c` and
357 `hello.h` as expected.
359 After a merge (initiated by 'git merge' or 'git pull') stops
360 because of conflicts, cleanly merged
361 paths are already staged to be committed for you, and paths that
362 conflicted are left in unmerged state. You would have to first
363 check which paths are conflicting with 'git status'
364 and after fixing them manually in your working tree, you would
365 stage the result as usual with 'git add':
367 ------------
368 $ git status | grep unmerged
369 unmerged: hello.c
370 $ edit hello.c
371 $ git add hello.c
372 ------------
374 After resolving conflicts and staging the result, `git ls-files -u`
375 would stop mentioning the conflicted path. When you are done,
376 run `git commit` to finally record the merge:
378 ------------
379 $ git commit
380 ------------
382 As with the case to record your own changes, you can use `-a`
383 option to save typing. One difference is that during a merge
384 resolution, you cannot use `git commit` with pathnames to
385 alter the order the changes are committed, because the merge
386 should be recorded as a single commit. In fact, the command
387 refuses to run when given pathnames (but see `-i` option).
391 ----------
393 Though not required, it's a good idea to begin the commit message
394 with a single short (less than 50 character) line summarizing the
395 change, followed by a blank line and then a more thorough description.
396 The text up to the first blank line in a commit message is treated
397 as the commit title, and that title is used throughout git.
398 For example, linkgit:git-format-patch[1] turns a commit into email, and it uses
399 the title on the Subject line and the rest of the commit in the body.
401 include::i18n.txt[]
404 ---------------------------------------
405 The editor used to edit the commit log message will be chosen from the
406 GIT_EDITOR environment variable, the core.editor configuration variable, the
407 VISUAL environment variable, or the EDITOR environment variable (in that
408 order). See linkgit:git-var[1] for details.
411 -----
412 This command can run `commit-msg`, `prepare-commit-msg`, `pre-commit`,
413 and `post-commit` hooks. See linkgit:githooks[5] for more
414 information.
417 -----
420 This file contains the commit message of a commit in progress.
421 If `git commit` exits due to an error before creating a commit,
422 any commit message that has been provided by the user (e.g., in
423 an editor session) will be available in this file, but will be
424 overwritten by the next invocation of `git commit`.
427 --------
428 linkgit:git-add[1],
429 linkgit:git-rm[1],
430 linkgit:git-mv[1],
431 linkgit:git-merge[1],
432 linkgit:git-commit-tree[1]
434 GIT
435 ---
436 Part of the linkgit:git[1] suite