1 git-add(1)
2 ==========
4 NAME
5 ----
6 git-add - Add file contents to the index
8 SYNOPSIS
9 --------
10 [verse]
11 'git add' [-n] [-v] [--force | -f] [--interactive | -i] [--patch | -p]
12 [--edit | -e] [--all | [--update | -u]] [--intent-to-add | -N]
13 [--refresh] [--ignore-errors] [--ignore-missing] [--]
14 [<filepattern>...]
16 DESCRIPTION
17 -----------
18 This command updates the index using the current content found in
19 the working tree, to prepare the content staged for the next commit.
20 It typically adds the current content of existing paths as a whole,
21 but with some options it can also be used to add content with
22 only part of the changes made to the working tree files applied, or
23 remove paths that do not exist in the working tree anymore.
25 The "index" holds a snapshot of the content of the working tree, and it
26 is this snapshot that is taken as the contents of the next commit. Thus
27 after making any changes to the working directory, and before running
28 the commit command, you must use the `add` command to add any new or
29 modified files to the index.
31 This command can be performed multiple times before a commit. It only
32 adds the content of the specified file(s) at the time the add command is
33 run; if you want subsequent changes included in the next commit, then
34 you must run `git add` again to add the new content to the index.
36 The `git status` command can be used to obtain a summary of which
37 files have changes that are staged for the next commit.
39 The `git add` command will not add ignored files by default. If any
40 ignored files were explicitly specified on the command line, `git add`
41 will fail with a list of ignored files. Ignored files reached by
42 directory recursion or filename globbing performed by Git (quote your
43 globs before the shell) will be silently ignored. The 'git add' command can
44 be used to add ignored files with the `-f` (force) option.
46 Please see linkgit:git-commit[1] for alternative ways to add content to a
47 commit.
50 OPTIONS
51 -------
52 <filepattern>...::
53 Files to add content from. Fileglobs (e.g. `*.c`) can
54 be given to add all matching files. Also a
55 leading directory name (e.g. `dir` to add `dir/file1`
56 and `dir/file2`) can be given to add all files in the
57 directory, recursively.
59 -n::
60 --dry-run::
61 Don't actually add the file(s), just show if they exist and/or will
62 be ignored.
64 -v::
65 --verbose::
66 Be verbose.
68 -f::
69 --force::
70 Allow adding otherwise ignored files.
72 -i::
73 --interactive::
74 Add modified contents in the working tree interactively to
75 the index. Optional path arguments may be supplied to limit
76 operation to a subset of the working tree. See ``Interactive
77 mode'' for details.
79 -p::
80 --patch::
81 Interactively choose hunks of patch between the index and the
82 work tree and add them to the index. This gives the user a chance
83 to review the difference before adding modified contents to the
84 index.
85 +
86 This effectively runs `add --interactive`, but bypasses the
87 initial command menu and directly jumps to the `patch` subcommand.
88 See ``Interactive mode'' for details.
90 -e, \--edit::
91 Open the diff vs. the index in an editor and let the user
92 edit it. After the editor was closed, adjust the hunk headers
93 and apply the patch to the index.
94 +
95 The intent of this option is to pick and choose lines of the patch to
96 apply, or even to modify the contents of lines to be staged. This can be
97 quicker and more flexible than using the interactive hunk selector.
98 However, it is easy to confuse oneself and create a patch that does not
99 apply to the index. See EDITING PATCHES below.
101 -u::
102 --update::
103 Only match <filepattern> against already tracked files in
104 the index rather than the working tree. That means that it
105 will never stage new files, but that it will stage modified
106 new contents of tracked files and that it will remove files
107 from the index if the corresponding files in the working tree
108 have been removed.
109 +
110 If no <filepattern> is given, default to "."; in other words,
111 update all tracked files in the current directory and its
112 subdirectories.
114 -A::
115 --all::
116 Like `-u`, but match <filepattern> against files in the
117 working tree in addition to the index. That means that it
118 will find new files as well as staging modified content and
119 removing files that are no longer in the working tree.
121 -N::
122 --intent-to-add::
123 Record only the fact that the path will be added later. An entry
124 for the path is placed in the index with no content. This is
125 useful for, among other things, showing the unstaged content of
126 such files with `git diff` and committing them with `git commit
127 -a`.
129 --refresh::
130 Don't add the file(s), but only refresh their stat()
131 information in the index.
133 --ignore-errors::
134 If some files could not be added because of errors indexing
135 them, do not abort the operation, but continue adding the
136 others. The command shall still exit with non-zero status.
137 The configuration variable `add.ignoreErrors` can be set to
138 true to make this the default behaviour.
140 --ignore-missing::
141 This option can only be used together with --dry-run. By using
142 this option the user can check if any of the given files would
143 be ignored, no matter if they are already present in the work
144 tree or not.
146 \--::
147 This option can be used to separate command-line options from
148 the list of files, (useful when filenames might be mistaken
149 for command-line options).
152 Configuration
153 -------------
155 The optional configuration variable `core.excludesfile` indicates a path to a
156 file containing patterns of file names to exclude from git-add, similar to
157 $GIT_DIR/info/exclude. Patterns in the exclude file are used in addition to
158 those in info/exclude. See linkgit:gitrepository-layout[5].
161 EXAMPLES
162 --------
164 * Adds content from all `*.txt` files under `Documentation` directory
165 and its subdirectories:
166 +
167 ------------
168 $ git add Documentation/\*.txt
169 ------------
170 +
171 Note that the asterisk `*` is quoted from the shell in this
172 example; this lets the command include the files from
173 subdirectories of `Documentation/` directory.
175 * Considers adding content from all git-*.sh scripts:
176 +
177 ------------
178 $ git add git-*.sh
179 ------------
180 +
181 Because this example lets the shell expand the asterisk (i.e. you are
182 listing the files explicitly), it does not consider
183 `subdir/git-foo.sh`.
185 Interactive mode
186 ----------------
187 When the command enters the interactive mode, it shows the
188 output of the 'status' subcommand, and then goes into its
189 interactive command loop.
191 The command loop shows the list of subcommands available, and
192 gives a prompt "What now> ". In general, when the prompt ends
193 with a single '>', you can pick only one of the choices given
194 and type return, like this:
196 ------------
197 *** Commands ***
198 1: status 2: update 3: revert 4: add untracked
199 5: patch 6: diff 7: quit 8: help
200 What now> 1
201 ------------
203 You also could say `s` or `sta` or `status` above as long as the
204 choice is unique.
206 The main command loop has 6 subcommands (plus help and quit).
208 status::
210 This shows the change between HEAD and index (i.e. what will be
211 committed if you say `git commit`), and between index and
212 working tree files (i.e. what you could stage further before
213 `git commit` using `git add`) for each path. A sample output
214 looks like this:
215 +
216 ------------
217 staged unstaged path
218 1: binary nothing foo.png
219 2: +403/-35 +1/-1 git-add--interactive.perl
220 ------------
221 +
222 It shows that foo.png has differences from HEAD (but that is
223 binary so line count cannot be shown) and there is no
224 difference between indexed copy and the working tree
225 version (if the working tree version were also different,
226 'binary' would have been shown in place of 'nothing'). The
227 other file, git-add{litdd}interactive.perl, has 403 lines added
228 and 35 lines deleted if you commit what is in the index, but
229 working tree file has further modifications (one addition and
230 one deletion).
232 update::
234 This shows the status information and issues an "Update>>"
235 prompt. When the prompt ends with double '>>', you can
236 make more than one selection, concatenated with whitespace or
237 comma. Also you can say ranges. E.g. "2-5 7,9" to choose
238 2,3,4,5,7,9 from the list. If the second number in a range is
239 omitted, all remaining patches are taken. E.g. "7-" to choose
240 7,8,9 from the list. You can say '*' to choose everything.
241 +
242 What you chose are then highlighted with '*',
243 like this:
244 +
245 ------------
246 staged unstaged path
247 1: binary nothing foo.png
248 * 2: +403/-35 +1/-1 git-add--interactive.perl
249 ------------
250 +
251 To remove selection, prefix the input with `-`
252 like this:
253 +
254 ------------
255 Update>> -2
256 ------------
257 +
258 After making the selection, answer with an empty line to stage the
259 contents of working tree files for selected paths in the index.
261 revert::
263 This has a very similar UI to 'update', and the staged
264 information for selected paths are reverted to that of the
265 HEAD version. Reverting new paths makes them untracked.
267 add untracked::
269 This has a very similar UI to 'update' and
270 'revert', and lets you add untracked paths to the index.
272 patch::
274 This lets you choose one path out of a 'status' like selection.
275 After choosing the path, it presents the diff between the index
276 and the working tree file and asks you if you want to stage
277 the change of each hunk. You can select one of the following
278 options and type return:
280 y - stage this hunk
281 n - do not stage this hunk
282 q - quit; do not stage this hunk nor any of the remaining ones
283 a - stage this hunk and all later hunks in the file
284 d - do not stage this hunk nor any of the later hunks in the file
285 g - select a hunk to go to
286 / - search for a hunk matching the given regex
287 j - leave this hunk undecided, see next undecided hunk
288 J - leave this hunk undecided, see next hunk
289 k - leave this hunk undecided, see previous undecided hunk
290 K - leave this hunk undecided, see previous hunk
291 s - split the current hunk into smaller hunks
292 e - manually edit the current hunk
293 ? - print help
294 +
295 After deciding the fate for all hunks, if there is any hunk
296 that was chosen, the index is updated with the selected hunks.
297 +
298 You can omit having to type return here, by setting the configuration
299 variable `interactive.singlekey` to `true`.
301 diff::
303 This lets you review what will be committed (i.e. between
304 HEAD and index).
307 EDITING PATCHES
308 ---------------
310 Invoking `git add -e` or selecting `e` from the interactive hunk
311 selector will open a patch in your editor; after the editor exits, the
312 result is applied to the index. You are free to make arbitrary changes
313 to the patch, but note that some changes may have confusing results, or
314 even result in a patch that cannot be applied. If you want to abort the
315 operation entirely (i.e., stage nothing new in the index), simply delete
316 all lines of the patch. The list below describes some common things you
317 may see in a patch, and which editing operations make sense on them.
319 --
320 added content::
322 Added content is represented by lines beginning with "{plus}". You can
323 prevent staging any addition lines by deleting them.
325 removed content::
327 Removed content is represented by lines beginning with "-". You can
328 prevent staging their removal by converting the "-" to a " " (space).
330 modified content::
332 Modified content is represented by "-" lines (removing the old content)
333 followed by "{plus}" lines (adding the replacement content). You can
334 prevent staging the modification by converting "-" lines to " ", and
335 removing "{plus}" lines. Beware that modifying only half of the pair is
336 likely to introduce confusing changes to the index.
337 --
339 There are also more complex operations that can be performed. But beware
340 that because the patch is applied only to the index and not the working
341 tree, the working tree will appear to "undo" the change in the index.
342 For example, introducing a new line into the index that is in neither
343 the HEAD nor the working tree will stage the new line for commit, but
344 the line will appear to be reverted in the working tree.
346 Avoid using these constructs, or do so with extreme caution.
348 --
349 removing untouched content::
351 Content which does not differ between the index and working tree may be
352 shown on context lines, beginning with a " " (space). You can stage
353 context lines for removal by converting the space to a "-". The
354 resulting working tree file will appear to re-add the content.
356 modifying existing content::
358 One can also modify context lines by staging them for removal (by
359 converting " " to "-") and adding a "{plus}" line with the new content.
360 Similarly, one can modify "{plus}" lines for existing additions or
361 modifications. In all cases, the new modification will appear reverted
362 in the working tree.
364 new content::
366 You may also add new content that does not exist in the patch; simply
367 add new lines, each starting with "{plus}". The addition will appear
368 reverted in the working tree.
369 --
371 There are also several operations which should be avoided entirely, as
372 they will make the patch impossible to apply:
374 * adding context (" ") or removal ("-") lines
375 * deleting context or removal lines
376 * modifying the contents of context or removal lines
378 SEE ALSO
379 --------
380 linkgit:git-status[1]
381 linkgit:git-rm[1]
382 linkgit:git-reset[1]
383 linkgit:git-mv[1]
384 linkgit:git-commit[1]
385 linkgit:git-update-index[1]
387 GIT
388 ---
389 Part of the linkgit:git[1] suite