1 // Please don't remove this comment as asciidoc behaves badly when
2 // the first non-empty line is ifdef/ifndef. The symptom is that
3 // without this comment the <git-diff-core> attribute conditionally
4 // defined below ends up being defined unconditionally.
5 // Last checked with asciidoc 7.0.2.
7 ifndef::git-format-patch[]
13 endif::git-format-patch[]
15 ifdef::git-format-patch[]
18 Generate plain patches without any diffstats.
19 endif::git-format-patch[]
21 ifndef::git-format-patch[]
25 Generate patch (see section on generating patches).
29 endif::git-format-patch[]
33 Suppress diff output. Useful for commands like `git show` that
34 show the patch by default, or to cancel the effect of `--patch`.
38 Generate diffs with <n> lines of context instead of
40 ifndef::git-format-patch[]
42 endif::git-format-patch[]
44 ifndef::git-format-patch[]
47 Generate the diff in raw format.
48 ifdef::git-diff-core[]
50 endif::git-diff-core[]
53 For each commit, show a summary of changes using the raw diff
54 format. See the "RAW OUTPUT FORMAT" section of
55 linkgit:git-diff[1]. This is different from showing the log
56 itself in raw format, which you can achieve with
59 endif::git-format-patch[]
61 ifndef::git-format-patch[]
63 Synonym for `-p --raw`.
64 endif::git-format-patch[]
66 --compaction-heuristic::
67 --no-compaction-heuristic::
68 These are to help debugging and tuning an experimental
69 heuristic that shifts the hunk boundary in an attempt to
70 make the resulting patch easier to read.
73 Spend extra time to make sure the smallest possible
77 Generate a diff using the "patience diff" algorithm.
80 Generate a diff using the "histogram diff" algorithm.
82 --diff-algorithm={patience|minimal|histogram|myers}::
83 Choose a diff algorithm. The variants are as follows:
87 The basic greedy diff algorithm. Currently, this is the default.
89 Spend extra time to make sure the smallest possible diff is
92 Use "patience diff" algorithm when generating patches.
94 This algorithm extends the patience algorithm to "support
95 low-occurrence common elements".
98 For instance, if you configured diff.algorithm variable to a
99 non-default value and want to use the default one, then you
100 have to use `--diff-algorithm=default` option.
102 --stat[=<width>[,<name-width>[,<count>]]]::
103 Generate a diffstat. By default, as much space as necessary
104 will be used for the filename part, and the rest for the graph
105 part. Maximum width defaults to terminal width, or 80 columns
106 if not connected to a terminal, and can be overridden by
107 `<width>`. The width of the filename part can be limited by
108 giving another width `<name-width>` after a comma. The width
109 of the graph part can be limited by using
110 `--stat-graph-width=<width>` (affects all commands generating
111 a stat graph) or by setting `diff.statGraphWidth=<width>`
112 (does not affect `git format-patch`).
113 By giving a third parameter `<count>`, you can limit the
114 output to the first `<count>` lines, followed by `...` if
117 These parameters can also be set individually with `--stat-width=<width>`,
118 `--stat-name-width=<name-width>` and `--stat-count=<count>`.
121 Similar to `--stat`, but shows number of added and
122 deleted lines in decimal notation and pathname without
123 abbreviation, to make it more machine friendly. For
124 binary files, outputs two `-` instead of saying
128 Output only the last line of the `--stat` format containing total
129 number of modified files, as well as number of added and deleted
132 --dirstat[=<param1,param2,...>]::
133 Output the distribution of relative amount of changes for each
134 sub-directory. The behavior of `--dirstat` can be customized by
135 passing it a comma separated list of parameters.
136 The defaults are controlled by the `diff.dirstat` configuration
137 variable (see linkgit:git-config[1]).
138 The following parameters are available:
142 Compute the dirstat numbers by counting the lines that have been
143 removed from the source, or added to the destination. This ignores
144 the amount of pure code movements within a file. In other words,
145 rearranging lines in a file is not counted as much as other changes.
146 This is the default behavior when no parameter is given.
148 Compute the dirstat numbers by doing the regular line-based diff
149 analysis, and summing the removed/added line counts. (For binary
150 files, count 64-byte chunks instead, since binary files have no
151 natural concept of lines). This is a more expensive `--dirstat`
152 behavior than the `changes` behavior, but it does count rearranged
153 lines within a file as much as other changes. The resulting output
154 is consistent with what you get from the other `--*stat` options.
156 Compute the dirstat numbers by counting the number of files changed.
157 Each changed file counts equally in the dirstat analysis. This is
158 the computationally cheapest `--dirstat` behavior, since it does
159 not have to look at the file contents at all.
161 Count changes in a child directory for the parent directory as well.
162 Note that when using `cumulative`, the sum of the percentages
163 reported may exceed 100%. The default (non-cumulative) behavior can
164 be specified with the `noncumulative` parameter.
166 An integer parameter specifies a cut-off percent (3% by default).
167 Directories contributing less than this percentage of the changes
168 are not shown in the output.
171 Example: The following will count changed files, while ignoring
172 directories with less than 10% of the total amount of changed files,
173 and accumulating child directory counts in the parent directories:
174 `--dirstat=files,10,cumulative`.
177 Output a condensed summary of extended header information
178 such as creations, renames and mode changes.
180 ifndef::git-format-patch[]
182 Synonym for `-p --stat`.
183 endif::git-format-patch[]
185 ifndef::git-format-patch[]
189 Separate the commits with NULs instead of with new newlines.
191 Also, when `--raw` or `--numstat` has been given, do not munge
192 pathnames and use NULs as output field terminators.
195 When `--raw`, `--numstat`, `--name-only` or `--name-status` has been
196 given, do not munge pathnames and use NULs as output field terminators.
199 Without this option, each pathname output will have TAB, LF, double quotes,
200 and backslash characters replaced with `\t`, `\n`, `\"`, and `\\`,
201 respectively, and the pathname will be enclosed in double quotes if
202 any of those replacements occurred.
205 Show only names of changed files.
208 Show only names and status of changed files. See the description
209 of the `--diff-filter` option on what the status letters mean.
211 --submodule[=<format>]::
212 Specify how differences in submodules are shown. When `--submodule`
213 or `--submodule=log` is given, the 'log' format is used. This format lists
214 the commits in the range like linkgit:git-submodule[1] `summary` does.
215 Omitting the `--submodule` option or specifying `--submodule=short`,
216 uses the 'short' format. This format just shows the names of the commits
217 at the beginning and end of the range. Can be tweaked via the
218 `diff.submodule` configuration variable.
222 `--color` (i.e. without '=<when>') is the same as `--color=always`.
223 '<when>' can be one of `always`, `never`, or `auto`.
225 It can be changed by the `color.ui` and `color.diff`
226 configuration settings.
230 Turn off colored diff.
232 This can be used to override configuration settings.
234 It is the same as `--color=never`.
236 --word-diff[=<mode>]::
237 Show a word diff, using the <mode> to delimit changed words.
238 By default, words are delimited by whitespace; see
239 `--word-diff-regex` below. The <mode> defaults to 'plain', and
244 Highlight changed words using only colors. Implies `--color`.
246 Show words as `[-removed-]` and `{+added+}`. Makes no
247 attempts to escape the delimiters if they appear in the input,
248 so the output may be ambiguous.
250 Use a special line-based format intended for script
251 consumption. Added/removed/unchanged runs are printed in the
252 usual unified diff format, starting with a `+`/`-`/` `
253 character at the beginning of the line and extending to the
254 end of the line. Newlines in the input are represented by a
255 tilde `~` on a line of its own.
257 Disable word diff again.
260 Note that despite the name of the first mode, color is used to
261 highlight the changed parts in all modes if enabled.
263 --word-diff-regex=<regex>::
264 Use <regex> to decide what a word is, instead of considering
265 runs of non-whitespace to be a word. Also implies
266 `--word-diff` unless it was already enabled.
268 Every non-overlapping match of the
269 <regex> is considered a word. Anything between these matches is
270 considered whitespace and ignored(!) for the purposes of finding
271 differences. You may want to append `|[^[:space:]]` to your regular
272 expression to make sure that it matches all non-whitespace characters.
273 A match that contains a newline is silently truncated(!) at the
276 The regex can also be set via a diff driver or configuration option, see
277 linkgit:gitattributes[1] or linkgit:git-config[1]. Giving it explicitly
278 overrides any diff driver or configuration setting. Diff drivers
279 override configuration settings.
281 --color-words[=<regex>]::
282 Equivalent to `--word-diff=color` plus (if a regex was
283 specified) `--word-diff-regex=<regex>`.
284 endif::git-format-patch[]
287 Turn off rename detection, even when the configuration
288 file gives the default to do so.
290 ifndef::git-format-patch[]
292 Warn if changes introduce whitespace errors. What are
293 considered whitespace errors is controlled by `core.whitespace`
294 configuration. By default, trailing whitespaces (including
295 lines that solely consist of whitespaces) and a space character
296 that is immediately followed by a tab character inside the
297 initial indent of the line are considered whitespace errors.
298 Exits with non-zero status if problems are found. Not compatible
300 endif::git-format-patch[]
303 Instead of the first handful of characters, show the full
304 pre- and post-image blob object names on the "index"
305 line when generating patch format output.
308 In addition to `--full-index`, output a binary diff that
309 can be applied with `git-apply`.
312 Instead of showing the full 40-byte hexadecimal object
313 name in diff-raw format output and diff-tree header
314 lines, show only a partial prefix. This is
315 independent of the `--full-index` option above, which controls
316 the diff-patch output format. Non default number of
317 digits can be specified with `--abbrev=<n>`.
320 --break-rewrites[=[<n>][/<m>]]::
321 Break complete rewrite changes into pairs of delete and
322 create. This serves two purposes:
324 It affects the way a change that amounts to a total rewrite of a file
325 not as a series of deletion and insertion mixed together with a very
326 few lines that happen to match textually as the context, but as a
327 single deletion of everything old followed by a single insertion of
328 everything new, and the number `m` controls this aspect of the -B
329 option (defaults to 60%). `-B/70%` specifies that less than 30% of the
330 original should remain in the result for Git to consider it a total
331 rewrite (i.e. otherwise the resulting patch will be a series of
332 deletion and insertion mixed together with context lines).
334 When used with -M, a totally-rewritten file is also considered as the
335 source of a rename (usually -M only considers a file that disappeared
336 as the source of a rename), and the number `n` controls this aspect of
337 the -B option (defaults to 50%). `-B20%` specifies that a change with
338 addition and deletion compared to 20% or more of the file's size are
339 eligible for being picked up as a possible source of a rename to
343 --find-renames[=<n>]::
348 If generating diffs, detect and report renames for each commit.
349 For following files across renames while traversing history, see
352 If `n` is specified, it is a threshold on the similarity
353 index (i.e. amount of addition/deletions compared to the
354 file's size). For example, `-M90%` means Git should consider a
355 delete/add pair to be a rename if more than 90% of the file
356 hasn't changed. Without a `%` sign, the number is to be read as
357 a fraction, with a decimal point before it. I.e., `-M5` becomes
358 0.5, and is thus the same as `-M50%`. Similarly, `-M05` is
359 the same as `-M5%`. To limit detection to exact renames, use
360 `-M100%`. The default similarity index is 50%.
363 --find-copies[=<n>]::
364 Detect copies as well as renames. See also `--find-copies-harder`.
365 If `n` is specified, it has the same meaning as for `-M<n>`.
367 --find-copies-harder::
368 For performance reasons, by default, `-C` option finds copies only
369 if the original file of the copy was modified in the same
370 changeset. This flag makes the command
371 inspect unmodified files as candidates for the source of
372 copy. This is a very expensive operation for large
373 projects, so use it with caution. Giving more than one
374 `-C` option has the same effect.
377 --irreversible-delete::
378 Omit the preimage for deletes, i.e. print only the header but not
379 the diff between the preimage and `/dev/null`. The resulting patch
380 is not meant to be applied with `patch` or `git apply`; this is
381 solely for people who want to just concentrate on reviewing the
382 text after the change. In addition, the output obviously lack
383 enough information to apply such a patch in reverse, even manually,
384 hence the name of the option.
386 When used together with `-B`, omit also the preimage in the deletion part
387 of a delete/create pair.
390 The `-M` and `-C` options require O(n^2) processing time where n
391 is the number of potential rename/copy targets. This
392 option prevents rename/copy detection from running if
393 the number of rename/copy targets exceeds the specified
396 ifndef::git-format-patch[]
397 --diff-filter=[(A|C|D|M|R|T|U|X|B)...[*]]::
398 Select only files that are Added (`A`), Copied (`C`),
399 Deleted (`D`), Modified (`M`), Renamed (`R`), have their
400 type (i.e. regular file, symlink, submodule, ...) changed (`T`),
401 are Unmerged (`U`), are
402 Unknown (`X`), or have had their pairing Broken (`B`).
403 Any combination of the filter characters (including none) can be used.
404 When `*` (All-or-none) is added to the combination, all
405 paths are selected if there is any file that matches
406 other criteria in the comparison; if there is no file
407 that matches other criteria, nothing is selected.
410 Look for differences that change the number of occurrences of
411 the specified string (i.e. addition/deletion) in a file.
412 Intended for the scripter's use.
414 It is useful when you're looking for an exact block of code (like a
415 struct), and want to know the history of that block since it first
416 came into being: use the feature iteratively to feed the interesting
417 block in the preimage back into `-S`, and keep going until you get the
418 very first version of the block.
421 Look for differences whose patch text contains added/removed
422 lines that match <regex>.
424 To illustrate the difference between `-S<regex> --pickaxe-regex` and
425 `-G<regex>`, consider a commit with the following diff in the same
429 + return !regexec(regexp, two->ptr, 1, ®match, 0);
431 - hit = !regexec(regexp, mf2.ptr, 1, ®match, 0);
434 While `git log -G"regexec\(regexp"` will show this commit, `git log
435 -S"regexec\(regexp" --pickaxe-regex` will not (because the number of
436 occurrences of that string did not change).
438 See the 'pickaxe' entry in linkgit:gitdiffcore[7] for more
442 When `-S` or `-G` finds a change, show all the changes in that
443 changeset, not just the files that contain the change
447 Treat the <string> given to `-S` as an extended POSIX regular
449 endif::git-format-patch[]
452 Output the patch in the order specified in the
453 <orderfile>, which has one shell glob pattern per line.
454 This overrides the `diff.orderFile` configuration variable
455 (see linkgit:git-config[1]). To cancel `diff.orderFile`,
458 ifndef::git-format-patch[]
460 Swap two inputs; that is, show differences from index or
461 on-disk file to tree contents.
463 --relative[=<path>]::
464 When run from a subdirectory of the project, it can be
465 told to exclude changes outside the directory and show
466 pathnames relative to it with this option. When you are
467 not in a subdirectory (e.g. in a bare repository), you
468 can name which subdirectory to make the output relative
469 to by giving a <path> as an argument.
470 endif::git-format-patch[]
474 Treat all files as text.
476 --ignore-space-at-eol::
477 Ignore changes in whitespace at EOL.
480 --ignore-space-change::
481 Ignore changes in amount of whitespace. This ignores whitespace
482 at line end, and considers all other sequences of one or
483 more whitespace characters to be equivalent.
487 Ignore whitespace when comparing lines. This ignores
488 differences even if one line has whitespace where the other
491 --ignore-blank-lines::
492 Ignore changes whose lines are all blank.
494 --inter-hunk-context=<lines>::
495 Show the context between diff hunks, up to the specified number
496 of lines, thereby fusing hunks that are close to each other.
500 Show whole surrounding functions of changes.
502 ifndef::git-format-patch[]
505 Make the program exit with codes similar to diff(1).
506 That is, it exits with 1 if there were differences and
507 0 means no differences.
510 Disable all output of the program. Implies `--exit-code`.
512 endif::git-format-patch[]
515 Allow an external diff helper to be executed. If you set an
516 external diff driver with linkgit:gitattributes[5], you need
517 to use this option with linkgit:git-log[1] and friends.
520 Disallow external diff drivers.
524 Allow (or disallow) external text conversion filters to be run
525 when comparing binary files. See linkgit:gitattributes[5] for
526 details. Because textconv filters are typically a one-way
527 conversion, the resulting diff is suitable for human
528 consumption, but cannot be applied. For this reason, textconv
529 filters are enabled by default only for linkgit:git-diff[1] and
530 linkgit:git-log[1], but not for linkgit:git-format-patch[1] or
531 diff plumbing commands.
533 --ignore-submodules[=<when>]::
534 Ignore changes to submodules in the diff generation. <when> can be
535 either "none", "untracked", "dirty" or "all", which is the default.
536 Using "none" will consider the submodule modified when it either contains
537 untracked or modified files or its HEAD differs from the commit recorded
538 in the superproject and can be used to override any settings of the
539 'ignore' option in linkgit:git-config[1] or linkgit:gitmodules[5]. When
540 "untracked" is used submodules are not considered dirty when they only
541 contain untracked content (but they are still scanned for modified
542 content). Using "dirty" ignores all changes to the work tree of submodules,
543 only changes to the commits stored in the superproject are shown (this was
544 the behavior until 1.7.0). Using "all" hides all changes to submodules.
546 --src-prefix=<prefix>::
547 Show the given source prefix instead of "a/".
549 --dst-prefix=<prefix>::
550 Show the given destination prefix instead of "b/".
553 Do not show any source or destination prefix.
555 For more detailed explanation on these common options, see also
556 linkgit:gitdiffcore[7].