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).
32 Suppress diff output. Useful for commands like `git show` that
33 show the patch by default, or to cancel the effect of `--patch`.
34 endif::git-format-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 (which is off by default) that shifts the hunk
70 boundary in an attempt to make the resulting patch easier
74 Spend extra time to make sure the smallest possible
78 Generate a diff using the "patience diff" algorithm.
81 Generate a diff using the "histogram diff" algorithm.
83 --diff-algorithm={patience|minimal|histogram|myers}::
84 Choose a diff algorithm. The variants are as follows:
88 The basic greedy diff algorithm. Currently, this is the default.
90 Spend extra time to make sure the smallest possible diff is
93 Use "patience diff" algorithm when generating patches.
95 This algorithm extends the patience algorithm to "support
96 low-occurrence common elements".
99 For instance, if you configured diff.algorithm variable to a
100 non-default value and want to use the default one, then you
101 have to use `--diff-algorithm=default` option.
103 --stat[=<width>[,<name-width>[,<count>]]]::
104 Generate a diffstat. By default, as much space as necessary
105 will be used for the filename part, and the rest for the graph
106 part. Maximum width defaults to terminal width, or 80 columns
107 if not connected to a terminal, and can be overridden by
108 `<width>`. The width of the filename part can be limited by
109 giving another width `<name-width>` after a comma. The width
110 of the graph part can be limited by using
111 `--stat-graph-width=<width>` (affects all commands generating
112 a stat graph) or by setting `diff.statGraphWidth=<width>`
113 (does not affect `git format-patch`).
114 By giving a third parameter `<count>`, you can limit the
115 output to the first `<count>` lines, followed by `...` if
118 These parameters can also be set individually with `--stat-width=<width>`,
119 `--stat-name-width=<name-width>` and `--stat-count=<count>`.
122 Similar to `--stat`, but shows number of added and
123 deleted lines in decimal notation and pathname without
124 abbreviation, to make it more machine friendly. For
125 binary files, outputs two `-` instead of saying
129 Output only the last line of the `--stat` format containing total
130 number of modified files, as well as number of added and deleted
133 --dirstat[=<param1,param2,...>]::
134 Output the distribution of relative amount of changes for each
135 sub-directory. The behavior of `--dirstat` can be customized by
136 passing it a comma separated list of parameters.
137 The defaults are controlled by the `diff.dirstat` configuration
138 variable (see linkgit:git-config[1]).
139 The following parameters are available:
143 Compute the dirstat numbers by counting the lines that have been
144 removed from the source, or added to the destination. This ignores
145 the amount of pure code movements within a file. In other words,
146 rearranging lines in a file is not counted as much as other changes.
147 This is the default behavior when no parameter is given.
149 Compute the dirstat numbers by doing the regular line-based diff
150 analysis, and summing the removed/added line counts. (For binary
151 files, count 64-byte chunks instead, since binary files have no
152 natural concept of lines). This is a more expensive `--dirstat`
153 behavior than the `changes` behavior, but it does count rearranged
154 lines within a file as much as other changes. The resulting output
155 is consistent with what you get from the other `--*stat` options.
157 Compute the dirstat numbers by counting the number of files changed.
158 Each changed file counts equally in the dirstat analysis. This is
159 the computationally cheapest `--dirstat` behavior, since it does
160 not have to look at the file contents at all.
162 Count changes in a child directory for the parent directory as well.
163 Note that when using `cumulative`, the sum of the percentages
164 reported may exceed 100%. The default (non-cumulative) behavior can
165 be specified with the `noncumulative` parameter.
167 An integer parameter specifies a cut-off percent (3% by default).
168 Directories contributing less than this percentage of the changes
169 are not shown in the output.
172 Example: The following will count changed files, while ignoring
173 directories with less than 10% of the total amount of changed files,
174 and accumulating child directory counts in the parent directories:
175 `--dirstat=files,10,cumulative`.
178 Output a condensed summary of extended header information
179 such as creations, renames and mode changes.
181 ifndef::git-format-patch[]
183 Synonym for `-p --stat`.
184 endif::git-format-patch[]
186 ifndef::git-format-patch[]
190 Separate the commits with NULs instead of with new newlines.
192 Also, when `--raw` or `--numstat` has been given, do not munge
193 pathnames and use NULs as output field terminators.
196 When `--raw`, `--numstat`, `--name-only` or `--name-status` has been
197 given, do not munge pathnames and use NULs as output field terminators.
200 Without this option, each pathname output will have TAB, LF, double quotes,
201 and backslash characters replaced with `\t`, `\n`, `\"`, and `\\`,
202 respectively, and the pathname will be enclosed in double quotes if
203 any of those replacements occurred.
206 Show only names of changed files.
209 Show only names and status of changed files. See the description
210 of the `--diff-filter` option on what the status letters mean.
212 --submodule[=<format>]::
213 Specify how differences in submodules are shown. When `--submodule`
214 or `--submodule=log` is given, the 'log' format is used. This format lists
215 the commits in the range like linkgit:git-submodule[1] `summary` does.
216 Omitting the `--submodule` option or specifying `--submodule=short`,
217 uses the 'short' format. This format just shows the names of the commits
218 at the beginning and end of the range. Can be tweaked via the
219 `diff.submodule` configuration variable.
223 `--color` (i.e. without '=<when>') is the same as `--color=always`.
224 '<when>' can be one of `always`, `never`, or `auto`.
226 It can be changed by the `color.ui` and `color.diff`
227 configuration settings.
231 Turn off colored diff.
233 This can be used to override configuration settings.
235 It is the same as `--color=never`.
237 --word-diff[=<mode>]::
238 Show a word diff, using the <mode> to delimit changed words.
239 By default, words are delimited by whitespace; see
240 `--word-diff-regex` below. The <mode> defaults to 'plain', and
245 Highlight changed words using only colors. Implies `--color`.
247 Show words as `[-removed-]` and `{+added+}`. Makes no
248 attempts to escape the delimiters if they appear in the input,
249 so the output may be ambiguous.
251 Use a special line-based format intended for script
252 consumption. Added/removed/unchanged runs are printed in the
253 usual unified diff format, starting with a `+`/`-`/` `
254 character at the beginning of the line and extending to the
255 end of the line. Newlines in the input are represented by a
256 tilde `~` on a line of its own.
258 Disable word diff again.
261 Note that despite the name of the first mode, color is used to
262 highlight the changed parts in all modes if enabled.
264 --word-diff-regex=<regex>::
265 Use <regex> to decide what a word is, instead of considering
266 runs of non-whitespace to be a word. Also implies
267 `--word-diff` unless it was already enabled.
269 Every non-overlapping match of the
270 <regex> is considered a word. Anything between these matches is
271 considered whitespace and ignored(!) for the purposes of finding
272 differences. You may want to append `|[^[:space:]]` to your regular
273 expression to make sure that it matches all non-whitespace characters.
274 A match that contains a newline is silently truncated(!) at the
277 For example, `--word-diff-regex=.` will treat each character as a word
278 and, correspondingly, show differences character by character.
280 The regex can also be set via a diff driver or configuration option, see
281 linkgit:gitattributes[5] or linkgit:git-config[1]. Giving it explicitly
282 overrides any diff driver or configuration setting. Diff drivers
283 override configuration settings.
285 --color-words[=<regex>]::
286 Equivalent to `--word-diff=color` plus (if a regex was
287 specified) `--word-diff-regex=<regex>`.
288 endif::git-format-patch[]
291 Turn off rename detection, even when the configuration
292 file gives the default to do so.
294 ifndef::git-format-patch[]
296 Warn if changes introduce conflict markers or whitespace errors.
297 What are considered whitespace errors is controlled by `core.whitespace`
298 configuration. By default, trailing whitespaces (including
299 lines that solely consist of whitespaces) and a space character
300 that is immediately followed by a tab character inside the
301 initial indent of the line are considered whitespace errors.
302 Exits with non-zero status if problems are found. Not compatible
305 --ws-error-highlight=<kind>::
306 Highlight whitespace errors on lines specified by <kind>
307 in the color specified by `color.diff.whitespace`. <kind>
308 is a comma separated list of `old`, `new`, `context`. When
309 this option is not given, only whitespace errors in `new`
310 lines are highlighted. E.g. `--ws-error-highlight=new,old`
311 highlights whitespace errors on both deleted and added lines.
312 `all` can be used as a short-hand for `old,new,context`.
314 endif::git-format-patch[]
317 Instead of the first handful of characters, show the full
318 pre- and post-image blob object names on the "index"
319 line when generating patch format output.
322 In addition to `--full-index`, output a binary diff that
323 can be applied with `git-apply`.
326 Instead of showing the full 40-byte hexadecimal object
327 name in diff-raw format output and diff-tree header
328 lines, show only a partial prefix. This is
329 independent of the `--full-index` option above, which controls
330 the diff-patch output format. Non default number of
331 digits can be specified with `--abbrev=<n>`.
334 --break-rewrites[=[<n>][/<m>]]::
335 Break complete rewrite changes into pairs of delete and
336 create. This serves two purposes:
338 It affects the way a change that amounts to a total rewrite of a file
339 not as a series of deletion and insertion mixed together with a very
340 few lines that happen to match textually as the context, but as a
341 single deletion of everything old followed by a single insertion of
342 everything new, and the number `m` controls this aspect of the -B
343 option (defaults to 60%). `-B/70%` specifies that less than 30% of the
344 original should remain in the result for Git to consider it a total
345 rewrite (i.e. otherwise the resulting patch will be a series of
346 deletion and insertion mixed together with context lines).
348 When used with -M, a totally-rewritten file is also considered as the
349 source of a rename (usually -M only considers a file that disappeared
350 as the source of a rename), and the number `n` controls this aspect of
351 the -B option (defaults to 50%). `-B20%` specifies that a change with
352 addition and deletion compared to 20% or more of the file's size are
353 eligible for being picked up as a possible source of a rename to
357 --find-renames[=<n>]::
362 If generating diffs, detect and report renames for each commit.
363 For following files across renames while traversing history, see
366 If `n` is specified, it is a threshold on the similarity
367 index (i.e. amount of addition/deletions compared to the
368 file's size). For example, `-M90%` means Git should consider a
369 delete/add pair to be a rename if more than 90% of the file
370 hasn't changed. Without a `%` sign, the number is to be read as
371 a fraction, with a decimal point before it. I.e., `-M5` becomes
372 0.5, and is thus the same as `-M50%`. Similarly, `-M05` is
373 the same as `-M5%`. To limit detection to exact renames, use
374 `-M100%`. The default similarity index is 50%.
377 --find-copies[=<n>]::
378 Detect copies as well as renames. See also `--find-copies-harder`.
379 If `n` is specified, it has the same meaning as for `-M<n>`.
381 --find-copies-harder::
382 For performance reasons, by default, `-C` option finds copies only
383 if the original file of the copy was modified in the same
384 changeset. This flag makes the command
385 inspect unmodified files as candidates for the source of
386 copy. This is a very expensive operation for large
387 projects, so use it with caution. Giving more than one
388 `-C` option has the same effect.
391 --irreversible-delete::
392 Omit the preimage for deletes, i.e. print only the header but not
393 the diff between the preimage and `/dev/null`. The resulting patch
394 is not meant to be applied with `patch` or `git apply`; this is
395 solely for people who want to just concentrate on reviewing the
396 text after the change. In addition, the output obviously lack
397 enough information to apply such a patch in reverse, even manually,
398 hence the name of the option.
400 When used together with `-B`, omit also the preimage in the deletion part
401 of a delete/create pair.
404 The `-M` and `-C` options require O(n^2) processing time where n
405 is the number of potential rename/copy targets. This
406 option prevents rename/copy detection from running if
407 the number of rename/copy targets exceeds the specified
410 ifndef::git-format-patch[]
411 --diff-filter=[(A|C|D|M|R|T|U|X|B)...[*]]::
412 Select only files that are Added (`A`), Copied (`C`),
413 Deleted (`D`), Modified (`M`), Renamed (`R`), have their
414 type (i.e. regular file, symlink, submodule, ...) changed (`T`),
415 are Unmerged (`U`), are
416 Unknown (`X`), or have had their pairing Broken (`B`).
417 Any combination of the filter characters (including none) can be used.
418 When `*` (All-or-none) is added to the combination, all
419 paths are selected if there is any file that matches
420 other criteria in the comparison; if there is no file
421 that matches other criteria, nothing is selected.
424 Look for differences that change the number of occurrences of
425 the specified string (i.e. addition/deletion) in a file.
426 Intended for the scripter's use.
428 It is useful when you're looking for an exact block of code (like a
429 struct), and want to know the history of that block since it first
430 came into being: use the feature iteratively to feed the interesting
431 block in the preimage back into `-S`, and keep going until you get the
432 very first version of the block.
435 Look for differences whose patch text contains added/removed
436 lines that match <regex>.
438 To illustrate the difference between `-S<regex> --pickaxe-regex` and
439 `-G<regex>`, consider a commit with the following diff in the same
443 + return !regexec(regexp, two->ptr, 1, ®match, 0);
445 - hit = !regexec(regexp, mf2.ptr, 1, ®match, 0);
448 While `git log -G"regexec\(regexp"` will show this commit, `git log
449 -S"regexec\(regexp" --pickaxe-regex` will not (because the number of
450 occurrences of that string did not change).
452 See the 'pickaxe' entry in linkgit:gitdiffcore[7] for more
456 When `-S` or `-G` finds a change, show all the changes in that
457 changeset, not just the files that contain the change
461 Treat the <string> given to `-S` as an extended POSIX regular
463 endif::git-format-patch[]
466 Output the patch in the order specified in the
467 <orderfile>, which has one shell glob pattern per line.
468 This overrides the `diff.orderFile` configuration variable
469 (see linkgit:git-config[1]). To cancel `diff.orderFile`,
472 ifndef::git-format-patch[]
474 Swap two inputs; that is, show differences from index or
475 on-disk file to tree contents.
477 --relative[=<path>]::
478 When run from a subdirectory of the project, it can be
479 told to exclude changes outside the directory and show
480 pathnames relative to it with this option. When you are
481 not in a subdirectory (e.g. in a bare repository), you
482 can name which subdirectory to make the output relative
483 to by giving a <path> as an argument.
484 endif::git-format-patch[]
488 Treat all files as text.
490 --ignore-space-at-eol::
491 Ignore changes in whitespace at EOL.
494 --ignore-space-change::
495 Ignore changes in amount of whitespace. This ignores whitespace
496 at line end, and considers all other sequences of one or
497 more whitespace characters to be equivalent.
501 Ignore whitespace when comparing lines. This ignores
502 differences even if one line has whitespace where the other
505 --ignore-blank-lines::
506 Ignore changes whose lines are all blank.
508 --inter-hunk-context=<lines>::
509 Show the context between diff hunks, up to the specified number
510 of lines, thereby fusing hunks that are close to each other.
514 Show whole surrounding functions of changes.
516 ifndef::git-format-patch[]
519 Make the program exit with codes similar to diff(1).
520 That is, it exits with 1 if there were differences and
521 0 means no differences.
524 Disable all output of the program. Implies `--exit-code`.
526 endif::git-format-patch[]
529 Allow an external diff helper to be executed. If you set an
530 external diff driver with linkgit:gitattributes[5], you need
531 to use this option with linkgit:git-log[1] and friends.
534 Disallow external diff drivers.
538 Allow (or disallow) external text conversion filters to be run
539 when comparing binary files. See linkgit:gitattributes[5] for
540 details. Because textconv filters are typically a one-way
541 conversion, the resulting diff is suitable for human
542 consumption, but cannot be applied. For this reason, textconv
543 filters are enabled by default only for linkgit:git-diff[1] and
544 linkgit:git-log[1], but not for linkgit:git-format-patch[1] or
545 diff plumbing commands.
547 --ignore-submodules[=<when>]::
548 Ignore changes to submodules in the diff generation. <when> can be
549 either "none", "untracked", "dirty" or "all", which is the default.
550 Using "none" will consider the submodule modified when it either contains
551 untracked or modified files or its HEAD differs from the commit recorded
552 in the superproject and can be used to override any settings of the
553 'ignore' option in linkgit:git-config[1] or linkgit:gitmodules[5]. When
554 "untracked" is used submodules are not considered dirty when they only
555 contain untracked content (but they are still scanned for modified
556 content). Using "dirty" ignores all changes to the work tree of submodules,
557 only changes to the commits stored in the superproject are shown (this was
558 the behavior until 1.7.0). Using "all" hides all changes to submodules.
560 --src-prefix=<prefix>::
561 Show the given source prefix instead of "a/".
563 --dst-prefix=<prefix>::
564 Show the given destination prefix instead of "b/".
567 Do not show any source or destination prefix.
569 For more detailed explanation on these common options, see also
570 linkgit:gitdiffcore[7].