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[]
46 Generate the raw format.
47 ifdef::git-diff-core[]
49 endif::git-diff-core[]
50 endif::git-format-patch[]
52 ifndef::git-format-patch[]
54 Synonym for `-p --raw`.
55 endif::git-format-patch[]
58 Spend extra time to make sure the smallest possible
62 Generate a diff using the "patience diff" algorithm.
65 Generate a diff using the "histogram diff" algorithm.
67 --diff-algorithm={patience|minimal|histogram|myers}::
68 Choose a diff algorithm. The variants are as follows:
72 The basic greedy diff algorithm. Currently, this is the default.
74 Spend extra time to make sure the smallest possible diff is
77 Use "patience diff" algorithm when generating patches.
79 This algorithm extends the patience algorithm to "support
80 low-occurrence common elements".
83 For instance, if you configured diff.algorithm variable to a
84 non-default value and want to use the default one, then you
85 have to use `--diff-algorithm=default` option.
87 --stat[=<width>[,<name-width>[,<count>]]]::
88 Generate a diffstat. By default, as much space as necessary
89 will be used for the filename part, and the rest for the graph
90 part. Maximum width defaults to terminal width, or 80 columns
91 if not connected to a terminal, and can be overridden by
92 `<width>`. The width of the filename part can be limited by
93 giving another width `<name-width>` after a comma. The width
94 of the graph part can be limited by using
95 `--stat-graph-width=<width>` (affects all commands generating
96 a stat graph) or by setting `diff.statGraphWidth=<width>`
97 (does not affect `git format-patch`).
98 By giving a third parameter `<count>`, you can limit the
99 output to the first `<count>` lines, followed by `...` if
102 These parameters can also be set individually with `--stat-width=<width>`,
103 `--stat-name-width=<name-width>` and `--stat-count=<count>`.
106 Similar to `--stat`, but shows number of added and
107 deleted lines in decimal notation and pathname without
108 abbreviation, to make it more machine friendly. For
109 binary files, outputs two `-` instead of saying
113 Output only the last line of the `--stat` format containing total
114 number of modified files, as well as number of added and deleted
117 --dirstat[=<param1,param2,...>]::
118 Output the distribution of relative amount of changes for each
119 sub-directory. The behavior of `--dirstat` can be customized by
120 passing it a comma separated list of parameters.
121 The defaults are controlled by the `diff.dirstat` configuration
122 variable (see linkgit:git-config[1]).
123 The following parameters are available:
127 Compute the dirstat numbers by counting the lines that have been
128 removed from the source, or added to the destination. This ignores
129 the amount of pure code movements within a file. In other words,
130 rearranging lines in a file is not counted as much as other changes.
131 This is the default behavior when no parameter is given.
133 Compute the dirstat numbers by doing the regular line-based diff
134 analysis, and summing the removed/added line counts. (For binary
135 files, count 64-byte chunks instead, since binary files have no
136 natural concept of lines). This is a more expensive `--dirstat`
137 behavior than the `changes` behavior, but it does count rearranged
138 lines within a file as much as other changes. The resulting output
139 is consistent with what you get from the other `--*stat` options.
141 Compute the dirstat numbers by counting the number of files changed.
142 Each changed file counts equally in the dirstat analysis. This is
143 the computationally cheapest `--dirstat` behavior, since it does
144 not have to look at the file contents at all.
146 Count changes in a child directory for the parent directory as well.
147 Note that when using `cumulative`, the sum of the percentages
148 reported may exceed 100%. The default (non-cumulative) behavior can
149 be specified with the `noncumulative` parameter.
151 An integer parameter specifies a cut-off percent (3% by default).
152 Directories contributing less than this percentage of the changes
153 are not shown in the output.
156 Example: The following will count changed files, while ignoring
157 directories with less than 10% of the total amount of changed files,
158 and accumulating child directory counts in the parent directories:
159 `--dirstat=files,10,cumulative`.
162 Output a condensed summary of extended header information
163 such as creations, renames and mode changes.
165 ifndef::git-format-patch[]
167 Synonym for `-p --stat`.
168 endif::git-format-patch[]
170 ifndef::git-format-patch[]
174 Separate the commits with NULs instead of with new newlines.
176 Also, when `--raw` or `--numstat` has been given, do not munge
177 pathnames and use NULs as output field terminators.
180 When `--raw`, `--numstat`, `--name-only` or `--name-status` has been
181 given, do not munge pathnames and use NULs as output field terminators.
184 Without this option, each pathname output will have TAB, LF, double quotes,
185 and backslash characters replaced with `\t`, `\n`, `\"`, and `\\`,
186 respectively, and the pathname will be enclosed in double quotes if
187 any of those replacements occurred.
190 Show only names of changed files.
193 Show only names and status of changed files. See the description
194 of the `--diff-filter` option on what the status letters mean.
196 --submodule[=<format>]::
197 Specify how differences in submodules are shown. When `--submodule`
198 or `--submodule=log` is given, the 'log' format is used. This format lists
199 the commits in the range like linkgit:git-submodule[1] `summary` does.
200 Omitting the `--submodule` option or specifying `--submodule=short`,
201 uses the 'short' format. This format just shows the names of the commits
202 at the beginning and end of the range. Can be tweaked via the
203 `diff.submodule` configuration variable.
207 `--color` (i.e. without '=<when>') is the same as `--color=always`.
208 '<when>' can be one of `always`, `never`, or `auto`.
210 It can be changed by the `color.ui` and `color.diff`
211 configuration settings.
215 Turn off colored diff.
217 This can be used to override configuration settings.
219 It is the same as `--color=never`.
221 --word-diff[=<mode>]::
222 Show a word diff, using the <mode> to delimit changed words.
223 By default, words are delimited by whitespace; see
224 `--word-diff-regex` below. The <mode> defaults to 'plain', and
229 Highlight changed words using only colors. Implies `--color`.
231 Show words as `[-removed-]` and `{+added+}`. Makes no
232 attempts to escape the delimiters if they appear in the input,
233 so the output may be ambiguous.
235 Use a special line-based format intended for script
236 consumption. Added/removed/unchanged runs are printed in the
237 usual unified diff format, starting with a `+`/`-`/` `
238 character at the beginning of the line and extending to the
239 end of the line. Newlines in the input are represented by a
240 tilde `~` on a line of its own.
242 Disable word diff again.
245 Note that despite the name of the first mode, color is used to
246 highlight the changed parts in all modes if enabled.
248 --word-diff-regex=<regex>::
249 Use <regex> to decide what a word is, instead of considering
250 runs of non-whitespace to be a word. Also implies
251 `--word-diff` unless it was already enabled.
253 Every non-overlapping match of the
254 <regex> is considered a word. Anything between these matches is
255 considered whitespace and ignored(!) for the purposes of finding
256 differences. You may want to append `|[^[:space:]]` to your regular
257 expression to make sure that it matches all non-whitespace characters.
258 A match that contains a newline is silently truncated(!) at the
261 The regex can also be set via a diff driver or configuration option, see
262 linkgit:gitattributes[1] or linkgit:git-config[1]. Giving it explicitly
263 overrides any diff driver or configuration setting. Diff drivers
264 override configuration settings.
266 --color-words[=<regex>]::
267 Equivalent to `--word-diff=color` plus (if a regex was
268 specified) `--word-diff-regex=<regex>`.
269 endif::git-format-patch[]
272 Turn off rename detection, even when the configuration
273 file gives the default to do so.
275 ifndef::git-format-patch[]
277 Warn if changes introduce whitespace errors. What are
278 considered whitespace errors is controlled by `core.whitespace`
279 configuration. By default, trailing whitespaces (including
280 lines that solely consist of whitespaces) and a space character
281 that is immediately followed by a tab character inside the
282 initial indent of the line are considered whitespace errors.
283 Exits with non-zero status if problems are found. Not compatible
285 endif::git-format-patch[]
288 Instead of the first handful of characters, show the full
289 pre- and post-image blob object names on the "index"
290 line when generating patch format output.
293 In addition to `--full-index`, output a binary diff that
294 can be applied with `git-apply`.
297 Instead of showing the full 40-byte hexadecimal object
298 name in diff-raw format output and diff-tree header
299 lines, show only a partial prefix. This is
300 independent of the `--full-index` option above, which controls
301 the diff-patch output format. Non default number of
302 digits can be specified with `--abbrev=<n>`.
305 --break-rewrites[=[<n>][/<m>]]::
306 Break complete rewrite changes into pairs of delete and
307 create. This serves two purposes:
309 It affects the way a change that amounts to a total rewrite of a file
310 not as a series of deletion and insertion mixed together with a very
311 few lines that happen to match textually as the context, but as a
312 single deletion of everything old followed by a single insertion of
313 everything new, and the number `m` controls this aspect of the -B
314 option (defaults to 60%). `-B/70%` specifies that less than 30% of the
315 original should remain in the result for Git to consider it a total
316 rewrite (i.e. otherwise the resulting patch will be a series of
317 deletion and insertion mixed together with context lines).
319 When used with -M, a totally-rewritten file is also considered as the
320 source of a rename (usually -M only considers a file that disappeared
321 as the source of a rename), and the number `n` controls this aspect of
322 the -B option (defaults to 50%). `-B20%` specifies that a change with
323 addition and deletion compared to 20% or more of the file's size are
324 eligible for being picked up as a possible source of a rename to
328 --find-renames[=<n>]::
333 If generating diffs, detect and report renames for each commit.
334 For following files across renames while traversing history, see
337 If `n` is specified, it is a threshold on the similarity
338 index (i.e. amount of addition/deletions compared to the
339 file's size). For example, `-M90%` means Git should consider a
340 delete/add pair to be a rename if more than 90% of the file
341 hasn't changed. Without a `%` sign, the number is to be read as
342 a fraction, with a decimal point before it. I.e., `-M5` becomes
343 0.5, and is thus the same as `-M50%`. Similarly, `-M05` is
344 the same as `-M5%`. To limit detection to exact renames, use
345 `-M100%`. The default similarity index is 50%.
348 --find-copies[=<n>]::
349 Detect copies as well as renames. See also `--find-copies-harder`.
350 If `n` is specified, it has the same meaning as for `-M<n>`.
352 --find-copies-harder::
353 For performance reasons, by default, `-C` option finds copies only
354 if the original file of the copy was modified in the same
355 changeset. This flag makes the command
356 inspect unmodified files as candidates for the source of
357 copy. This is a very expensive operation for large
358 projects, so use it with caution. Giving more than one
359 `-C` option has the same effect.
362 --irreversible-delete::
363 Omit the preimage for deletes, i.e. print only the header but not
364 the diff between the preimage and `/dev/null`. The resulting patch
365 is not meant to be applied with `patch` or `git apply`; this is
366 solely for people who want to just concentrate on reviewing the
367 text after the change. In addition, the output obviously lack
368 enough information to apply such a patch in reverse, even manually,
369 hence the name of the option.
371 When used together with `-B`, omit also the preimage in the deletion part
372 of a delete/create pair.
375 The `-M` and `-C` options require O(n^2) processing time where n
376 is the number of potential rename/copy targets. This
377 option prevents rename/copy detection from running if
378 the number of rename/copy targets exceeds the specified
381 ifndef::git-format-patch[]
382 --diff-filter=[(A|C|D|M|R|T|U|X|B)...[*]]::
383 Select only files that are Added (`A`), Copied (`C`),
384 Deleted (`D`), Modified (`M`), Renamed (`R`), have their
385 type (i.e. regular file, symlink, submodule, ...) changed (`T`),
386 are Unmerged (`U`), are
387 Unknown (`X`), or have had their pairing Broken (`B`).
388 Any combination of the filter characters (including none) can be used.
389 When `*` (All-or-none) is added to the combination, all
390 paths are selected if there is any file that matches
391 other criteria in the comparison; if there is no file
392 that matches other criteria, nothing is selected.
395 Look for differences that change the number of occurrences of
396 the specified string (i.e. addition/deletion) in a file.
397 Intended for the scripter's use.
399 It is useful when you're looking for an exact block of code (like a
400 struct), and want to know the history of that block since it first
401 came into being: use the feature iteratively to feed the interesting
402 block in the preimage back into `-S`, and keep going until you get the
403 very first version of the block.
406 Look for differences whose patch text contains added/removed
407 lines that match <regex>.
409 To illustrate the difference between `-S<regex> --pickaxe-regex` and
410 `-G<regex>`, consider a commit with the following diff in the same
414 + return !regexec(regexp, two->ptr, 1, ®match, 0);
416 - hit = !regexec(regexp, mf2.ptr, 1, ®match, 0);
419 While `git log -G"regexec\(regexp"` will show this commit, `git log
420 -S"regexec\(regexp" --pickaxe-regex` will not (because the number of
421 occurrences of that string did not change).
423 See the 'pickaxe' entry in linkgit:gitdiffcore[7] for more
427 When `-S` or `-G` finds a change, show all the changes in that
428 changeset, not just the files that contain the change
432 Treat the <string> given to `-S` as an extended POSIX regular
434 endif::git-format-patch[]
437 Output the patch in the order specified in the
438 <orderfile>, which has one shell glob pattern per line.
439 This overrides the `diff.orderFile` configuration variable
440 (see linkgit:git-config[1]). To cancel `diff.orderFile`,
443 ifndef::git-format-patch[]
445 Swap two inputs; that is, show differences from index or
446 on-disk file to tree contents.
448 --relative[=<path>]::
449 When run from a subdirectory of the project, it can be
450 told to exclude changes outside the directory and show
451 pathnames relative to it with this option. When you are
452 not in a subdirectory (e.g. in a bare repository), you
453 can name which subdirectory to make the output relative
454 to by giving a <path> as an argument.
455 endif::git-format-patch[]
459 Treat all files as text.
461 --ignore-space-at-eol::
462 Ignore changes in whitespace at EOL.
465 --ignore-space-change::
466 Ignore changes in amount of whitespace. This ignores whitespace
467 at line end, and considers all other sequences of one or
468 more whitespace characters to be equivalent.
472 Ignore whitespace when comparing lines. This ignores
473 differences even if one line has whitespace where the other
476 --ignore-blank-lines::
477 Ignore changes whose lines are all blank.
479 --inter-hunk-context=<lines>::
480 Show the context between diff hunks, up to the specified number
481 of lines, thereby fusing hunks that are close to each other.
485 Show whole surrounding functions of changes.
487 ifndef::git-format-patch[]
490 Make the program exit with codes similar to diff(1).
491 That is, it exits with 1 if there were differences and
492 0 means no differences.
495 Disable all output of the program. Implies `--exit-code`.
497 endif::git-format-patch[]
500 Allow an external diff helper to be executed. If you set an
501 external diff driver with linkgit:gitattributes[5], you need
502 to use this option with linkgit:git-log[1] and friends.
505 Disallow external diff drivers.
509 Allow (or disallow) external text conversion filters to be run
510 when comparing binary files. See linkgit:gitattributes[5] for
511 details. Because textconv filters are typically a one-way
512 conversion, the resulting diff is suitable for human
513 consumption, but cannot be applied. For this reason, textconv
514 filters are enabled by default only for linkgit:git-diff[1] and
515 linkgit:git-log[1], but not for linkgit:git-format-patch[1] or
516 diff plumbing commands.
518 --ignore-submodules[=<when>]::
519 Ignore changes to submodules in the diff generation. <when> can be
520 either "none", "untracked", "dirty" or "all", which is the default.
521 Using "none" will consider the submodule modified when it either contains
522 untracked or modified files or its HEAD differs from the commit recorded
523 in the superproject and can be used to override any settings of the
524 'ignore' option in linkgit:git-config[1] or linkgit:gitmodules[5]. When
525 "untracked" is used submodules are not considered dirty when they only
526 contain untracked content (but they are still scanned for modified
527 content). Using "dirty" ignores all changes to the work tree of submodules,
528 only changes to the commits stored in the superproject are shown (this was
529 the behavior until 1.7.0). Using "all" hides all changes to submodules.
531 --src-prefix=<prefix>::
532 Show the given source prefix instead of "a/".
534 --dst-prefix=<prefix>::
535 Show the given destination prefix instead of "b/".
538 Do not show any source or destination prefix.
540 For more detailed explanation on these common options, see also
541 linkgit:gitdiffcore[7].