Documentation/git-commit.txt

   1 git-commit(1)
   2 =============
   3
   4 NAME
   5 ----
   6 git-commit - Record changes to the repository
   7
   8 SYNOPSIS
   9 --------
  10 [verse]
  11 'git-commit' [-a | --interactive] [-s] [-v] [-u]
  12            [(-c | -C) <commit> | -F <file> | -m <msg> | --amend]
  13            [--allow-empty] [--no-verify] [-e] [--author <author>]
  14            [--cleanup=<mode>] [--] [[-i | -o ]<file>...]
  15
  16 DESCRIPTION
  17 -----------
  18 Use 'git commit' to store the current contents of the index in a new
  19 commit along with a log message describing the changes you have made.
  20
  21 The content to be added can be specified in several ways:
  22
  23 1. by using linkgit:git-add[1] to incrementally "add" changes to the
  24    index before using the 'commit' command (Note: even modified
  25    files must be "added");
  26
  27 2. by using linkgit:git-rm[1] to remove files from the working tree
  28    and the index, again before using the 'commit' command;
  29
  30 3. by listing files as arguments to the 'commit' command, in which
  31    case the commit will ignore changes staged in the index, and instead
  32    record the current content of the listed files;
  33
  34 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
  40 5. by using the --interactive switch with the 'commit' command to decide one
  41    by one which files should be part of the commit, before finalizing the
  42    operation.  Currently, this is done by invoking `git-add --interactive`.
  43
  44 The linkgit:git-status[1] command can be used to obtain a
  45 summary of what is included by any of the above for the next
  46 commit by giving the same set of parameters you would give to
  47 this command.
  48
  49 If you make a commit and then found a mistake immediately after
  50 that, you can recover from it with linkgit:git-reset[1].
  51
  52
  53 OPTIONS
  54 -------
  55 -a|--all::
  56         Tell the command to automatically stage files that have
  57         been modified and deleted, but new files you have not
  58         told git about are not affected.
  59
  60 -c or -C <commit>::
  61         Take existing commit object, and reuse the log message
  62         and the authorship information (including the timestamp)
  63         when creating the commit.  With '-C', the editor is not
  64         invoked; with '-c' the user can further edit the commit
  65         message.
  66
  67 -F <file>::
  68         Take the commit message from the given file.  Use '-' to
  69         read the message from the standard input.
  70
  71 --author <author>::
  72         Override the author name used in the commit.  Use
  73         `A U Thor <author@example.com>` format.
  74
  75 -m <msg>|--message=<msg>::
  76         Use the given <msg> as the commit message.
  77
  78 -t <file>|--template=<file>::
  79         Use the contents of the given file as the initial version
  80         of the commit message. The editor is invoked and you can
  81         make subsequent changes. If a message is specified using
  82         the `-m` or `-F` options, this option has no effect. This
  83         overrides the `commit.template` configuration variable.
  84
  85 -s|--signoff::
  86         Add Signed-off-by line at the end of the commit message.
  87
  88 --no-verify::
  89         This option bypasses the pre-commit and commit-msg hooks.
  90         See also link:hooks.html[hooks].
  91
  92 --allow-empty::
  93         Usually recording a commit that has the exact same tree as its
  94         sole parent commit is a mistake, and the command prevents you
  95         from making such a commit.  This option bypasses the safety, and
  96         is primarily for use by foreign scm interface scripts.
  97
  98 --cleanup=<mode>::
  99         This option sets how the commit message is cleaned up.
 100         The  '<mode>' can be one of 'verbatim', 'whitespace', 'strip',
 101         and 'default'. The 'default' mode will strip leading and
 102         trailing empty lines and #commentary from the commit message
 103         only if the message is to be edited. Otherwise only whitespace
 104         removed. The 'verbatim' mode does not change message at all,
 105         'whitespace' removes just leading/trailing whitespace lines
 106         and 'strip' removes both whitespace and commentary.
 107
 108 -e|--edit::
 109         The message taken from file with `-F`, command line with
 110         `-m`, and from file with `-C` are usually used as the
 111         commit log message unmodified.  This option lets you
 112         further edit the message taken from these sources.
 113
 114 --amend::
 115
 116         Used to amend the tip of the current branch. Prepare the tree
 117         object you would want to replace the latest commit as usual
 118         (this includes the usual -i/-o and explicit paths), and the
 119         commit log editor is seeded with the commit message from the
 120         tip of the current branch. The commit you create replaces the
 121         current tip -- if it was a merge, it will have the parents of
 122         the current tip as parents -- so the current top commit is
 123         discarded.
 124 +
 125 --
 126 It is a rough equivalent for:
 127 ------
 128         $ git reset --soft HEAD^
 129         $ ... do something else to come up with the right tree ...
 130         $ git commit -c ORIG_HEAD
 131
 132 ------
 133 but can be used to amend a merge commit.
 134 --
 135
 136 -i|--include::
 137         Before making a commit out of staged contents so far,
 138         stage the contents of paths given on the command line
 139         as well.  This is usually not what you want unless you
 140         are concluding a conflicted merge.
 141
 142 -u|--untracked-files::
 143         Show all untracked files, also those in uninteresting
 144         directories, in the "Untracked files:" section of commit
 145         message template.  Without this option only its name and
 146         a trailing slash are displayed for each untracked
 147         directory.
 148
 149 -v|--verbose::
 150         Show unified diff between the HEAD commit and what
 151         would be committed at the bottom of the commit message
 152         template.  Note that this diff output doesn't have its
 153         lines prefixed with '#'.
 154
 155 -q|--quiet::
 156         Suppress commit summary message.
 157
 158 \--::
 159         Do not interpret any more arguments as options.
 160
 161 <file>...::
 162         When files are given on the command line, the command
 163         commits the contents of the named files, without
 164         recording the changes already staged.  The contents of
 165         these files are also staged for the next commit on top
 166         of what have been staged before.
 167
 168
 169 EXAMPLES
 170 --------
 171 When recording your own work, the contents of modified files in
 172 your working tree are temporarily stored to a staging area
 173 called the "index" with linkgit:git-add[1].  A file can be
 174 reverted back, only in the index but not in the working tree,
 175 to that of the last commit with `git-reset HEAD -- <file>`,
 176 which effectively reverts `git-add` and prevents the changes to
 177 this file from participating in the next commit.  After building
 178 the state to be committed incrementally with these commands,
 179 `git commit` (without any pathname parameter) is used to record what
 180 has been staged so far.  This is the most basic form of the
 181 command.  An example:
 182
 183 ------------
 184 $ edit hello.c
 185 $ git rm goodbye.c
 186 $ git add hello.c
 187 $ git commit
 188 ------------
 189
 190 Instead of staging files after each individual change, you can
 191 tell `git commit` to notice the changes to the files whose
 192 contents are tracked in
 193 your working tree and do corresponding `git add` and `git rm`
 194 for you.  That is, this example does the same as the earlier
 195 example if there is no other change in your working tree:
 196
 197 ------------
 198 $ edit hello.c
 199 $ rm goodbye.c
 200 $ git commit -a
 201 ------------
 202
 203 The command `git commit -a` first looks at your working tree,
 204 notices that you have modified hello.c and removed goodbye.c,
 205 and performs necessary `git add` and `git rm` for you.
 206
 207 After staging changes to many files, you can alter the order the
 208 changes are recorded in, by giving pathnames to `git commit`.
 209 When pathnames are given, the command makes a commit that
 210 only records the changes made to the named paths:
 211
 212 ------------
 213 $ edit hello.c hello.h
 214 $ git add hello.c hello.h
 215 $ edit Makefile
 216 $ git commit Makefile
 217 ------------
 218
 219 This makes a commit that records the modification to `Makefile`.
 220 The changes staged for `hello.c` and `hello.h` are not included
 221 in the resulting commit.  However, their changes are not lost --
 222 they are still staged and merely held back.  After the above
 223 sequence, if you do:
 224
 225 ------------
 226 $ git commit
 227 ------------
 228
 229 this second commit would record the changes to `hello.c` and
 230 `hello.h` as expected.
 231
 232 After a merge (initiated by either linkgit:git-merge[1] or
 233 linkgit:git-pull[1]) stops because of conflicts, cleanly merged
 234 paths are already staged to be committed for you, and paths that
 235 conflicted are left in unmerged state.  You would have to first
 236 check which paths are conflicting with linkgit:git-status[1]
 237 and after fixing them manually in your working tree, you would
 238 stage the result as usual with linkgit:git-add[1]:
 239
 240 ------------
 241 $ git status | grep unmerged
 242 unmerged: hello.c
 243 $ edit hello.c
 244 $ git add hello.c
 245 ------------
 246
 247 After resolving conflicts and staging the result, `git ls-files -u`
 248 would stop mentioning the conflicted path.  When you are done,
 249 run `git commit` to finally record the merge:
 250
 251 ------------
 252 $ git commit
 253 ------------
 254
 255 As with the case to record your own changes, you can use `-a`
 256 option to save typing.  One difference is that during a merge
 257 resolution, you cannot use `git commit` with pathnames to
 258 alter the order the changes are committed, because the merge
 259 should be recorded as a single commit.  In fact, the command
 260 refuses to run when given pathnames (but see `-i` option).
 261
 262
 263 DISCUSSION
 264 ----------
 265
 266 Though not required, it's a good idea to begin the commit message
 267 with a single short (less than 50 character) line summarizing the
 268 change, followed by a blank line and then a more thorough description.
 269 Tools that turn commits into email, for example, use the first line
 270 on the Subject: line and the rest of the commit in the body.
 271
 272 include::i18n.txt[]
 273
 274 ENVIRONMENT AND CONFIGURATION VARIABLES
 275 ---------------------------------------
 276 The editor used to edit the commit log message will be chosen from the
 277 GIT_EDITOR environment variable, the core.editor configuration variable, the
 278 VISUAL environment variable, or the EDITOR environment variable (in that
 279 order).
 280
 281 HOOKS
 282 -----
 283 This command can run `commit-msg`, `prepare-commit-msg`, `pre-commit`,
 284 and `post-commit` hooks.  See link:hooks.html[hooks] for more
 285 information.
 286
 287
 288 SEE ALSO
 289 --------
 290 linkgit:git-add[1],
 291 linkgit:git-rm[1],
 292 linkgit:git-mv[1],
 293 linkgit:git-merge[1],
 294 linkgit:git-commit-tree[1]
 295
 296 Author
 297 ------
 298 Written by Linus Torvalds <torvalds@osdl.org> and
 299 Junio C Hamano <junkio@cox.net>
 300
 301
 302 GIT
 303 ---
 304 Part of the linkgit:git[7] suite