1GIT-DIFF(1) Git Manual GIT-DIFF(1)
2
6 git-diff - Show changes between commits, commit and working tree, etc
7
9 git diff [<options>] [<commit>] [--] [<path>...]
10 git diff [<options>] --cached [--merge-base] [<commit>] [--] [<path>...]
11 git diff [<options>] [--merge-base] <commit> [<commit>...] <commit> [--] [<path>...]
12 git diff [<options>] <commit>...<commit> [--] [<path>...]
13 git diff [<options>] <blob> <blob>
14 git diff [<options>] --no-index [--] <path> <path>
15
17 Show changes between the working tree and the index or a tree, changes
18 between the index and a tree, changes between two trees, changes
19 resulting from a merge, changes between two blob objects, or changes
20 between two files on disk.
21 git diff [<options>] [--] [<path>...]
23 This form is to view the changes you made relative to the index
24 (staging area for the next commit). In other words, the differences
25 are what you could tell Git to further add to the index but you
26 still haven’t. You can stage these changes by using git-add(1).
27 git diff [<options>] --no-index [--] <path> <path>
29 This form is to compare the given two paths on the filesystem. You
30 can omit the --no-index option when running the command in a
31 working tree controlled by Git and at least one of the paths points
32 outside the working tree, or when running the command outside a
33 working tree controlled by Git. This form implies --exit-code.
34 git diff [<options>] --cached [--merge-base] [<commit>] [--] [<path>...
36 ]
37 This form is to view the changes you staged for the next commit
38 relative to the named <commit>. Typically you would want comparison
39 with the latest commit, so if you do not give <commit>, it defaults
40 to HEAD. If HEAD does not exist (e.g. unborn branches) and <commit>
41 is not given, it shows all staged changes. --staged is a synonym of
42 --cached.
43 If --merge-base is given, instead of using <commit>, use the merge
45 base of <commit> and HEAD. git diff --cached --merge-base A is
46 equivalent to git diff --cached $(git merge-base A HEAD).
47 git diff [<options>] [--merge-base] <commit> [--] [<path>...]
49 This form is to view the changes you have in your working tree
50 relative to the named <commit>. You can use HEAD to compare it with
51 the latest commit, or a branch name to compare with the tip of a
52 different branch.
53 If --merge-base is given, instead of using <commit>, use the merge
55 base of <commit> and HEAD. git diff --merge-base A is equivalent
56 to git diff $(git merge-base A HEAD).
57 git diff [<options>] [--merge-base] <commit> <commit> [--] [<path>...]
59 This is to view the changes between two arbitrary <commit>.
60 If --merge-base is given, use the merge base of the two commits for
62 the "before" side. git diff --merge-base A B is equivalent to git
63 diff $(git merge-base A B) B.
64 git diff [<options>] <commit> <commit>... <commit> [--] [<path>...]
66 This form is to view the results of a merge commit. The first
67 listed <commit> must be the merge itself; the remaining two or more
68 commits should be its parents. Convenient ways to produce the
69 desired set of revisions are to use the suffixes ^@ and ^!. If A is
70 a merge commit, then git diff A A^@, git diff A^! and git show A
71 all give the same combined diff.
72 git diff [<options>] <commit>..<commit> [--] [<path>...]
74 This is synonymous to the earlier form (without the ..) for viewing
75 the changes between two arbitrary <commit>. If <commit> on one side
76 is omitted, it will have the same effect as using HEAD instead.
77 git diff [<options>] <commit>...<commit> [--] [<path>...]
79 This form is to view the changes on the branch containing and up to
80 the second <commit>, starting at a common ancestor of both
81 <commit>. git diff A...B is equivalent to git diff $(git
82 merge-base A B) B. You can omit any one of <commit>, which has the
83 same effect as using HEAD instead.
84 Just in case you are doing something exotic, it should be noted that
86 all of the <commit> in the above description, except in the
87 --merge-base case and in the last two forms that use .. notations, can
88 be any <tree>. A tree of interest is the one pointed to by the special
89 ref AUTO_MERGE, which is written by the ort merge strategy upon hitting
90 merge conflicts (see git-merge(1)). Comparing the working tree with
91 AUTO_MERGE shows changes you’ve made so far to resolve textual
92 conflicts (see the examples below).
93 For a more complete list of ways to spell <commit>, see "SPECIFYING
95 REVISIONS" section in gitrevisions(7). However, "diff" is about
96 comparing two endpoints, not ranges, and the range notations
97 (<commit>..<commit> and <commit>...<commit>) do not mean a range as
98 defined in the "SPECIFYING RANGES" section in gitrevisions(7).
99 git diff [<options>] <blob> <blob>
101 This form is to view the differences between the raw contents of
102 two blob objects.
103
105 -p, -u, --patch
106 Generate patch (see the section called “GENERATING PATCH TEXT WITH
107 -P”). This is the default.
108 -s, --no-patch
110 Suppress all output from the diff machinery. Useful for commands
111 like git show that show the patch by default to squelch their
112 output, or to cancel the effect of options like --patch, --stat
113 earlier on the command line in an alias.
114 -U<n>, --unified=<n>
116 Generate diffs with <n> lines of context instead of the usual
117 three. Implies --patch.
118 --output=<file>
120 Output to a specific file instead of stdout.
121 --output-indicator-new=<char>, --output-indicator-old=<char>,
123 --output-indicator-context=<char>
124 Specify the character used to indicate new, old or context lines in
125 the generated patch. Normally they are +, - and ' ' respectively.
126 --raw
128 Generate the diff in raw format.
129 --patch-with-raw
131 Synonym for -p --raw.
132 --indent-heuristic
134 Enable the heuristic that shifts diff hunk boundaries to make
135 patches easier to read. This is the default.
136 --no-indent-heuristic
138 Disable the indent heuristic.
139 --minimal
141 Spend extra time to make sure the smallest possible diff is
142 produced.
143 --patience
145 Generate a diff using the "patience diff" algorithm.
146 --histogram
148 Generate a diff using the "histogram diff" algorithm.
149 --anchored=<text>
151 Generate a diff using the "anchored diff" algorithm.
152 This option may be specified more than once.
154 If a line exists in both the source and destination, exists only
156 once, and starts with this text, this algorithm attempts to prevent
157 it from appearing as a deletion or addition in the output. It uses
158 the "patience diff" algorithm internally.
159 --diff-algorithm={patience|minimal|histogram|myers}
161 Choose a diff algorithm. The variants are as follows:
162 default, myers
164 The basic greedy diff algorithm. Currently, this is the
165 default.
166 minimal
168 Spend extra time to make sure the smallest possible diff is
169 produced.
170 patience
172 Use "patience diff" algorithm when generating patches.
173 histogram
175 This algorithm extends the patience algorithm to "support
176 low-occurrence common elements".
177 For instance, if you configured the diff.algorithm variable to a
179 non-default value and want to use the default one, then you have to
180 use --diff-algorithm=default option.
181 --stat[=<width>[,<name-width>[,<count>]]]
183 Generate a diffstat. By default, as much space as necessary will be
184 used for the filename part, and the rest for the graph part.
185 Maximum width defaults to terminal width, or 80 columns if not
186 connected to a terminal, and can be overridden by <width>. The
187 width of the filename part can be limited by giving another width
188 <name-width> after a comma or by setting
189 diff.statNameWidth=<width>. The width of the graph part can be
190 limited by using --stat-graph-width=<width> or by setting
191 diff.statGraphWidth=<width>. Using --stat or --stat-graph-width
192 affects all commands generating a stat graph, while setting
193 diff.statNameWidth or diff.statGraphWidth does not affect git
194 format-patch. By giving a third parameter <count>, you can limit
195 the output to the first <count> lines, followed by ... if there
196 are more.
197 These parameters can also be set individually with
199 --stat-width=<width>, --stat-name-width=<name-width> and
200 --stat-count=<count>.
201 --compact-summary
203 Output a condensed summary of extended header information such as
204 file creations or deletions ("new" or "gone", optionally "+l" if
205 it’s a symlink) and mode changes ("+x" or "-x" for adding or
206 removing executable bit respectively) in diffstat. The information
207 is put between the filename part and the graph part. Implies
208 --stat.
209 --numstat
211 Similar to --stat, but shows number of added and deleted lines in
212 decimal notation and pathname without abbreviation, to make it more
213 machine friendly. For binary files, outputs two - instead of saying
214 0 0.
215 --shortstat
217 Output only the last line of the --stat format containing total
218 number of modified files, as well as number of added and deleted
219 lines.
220 -X[<param1,param2,...>], --dirstat[=<param1,param2,...>]
222 Output the distribution of relative amount of changes for each
223 sub-directory. The behavior of --dirstat can be customized by
224 passing it a comma separated list of parameters. The defaults are
225 controlled by the diff.dirstat configuration variable (see git-
226 config(1)). The following parameters are available:
227 changes
229 Compute the dirstat numbers by counting the lines that have
230 been removed from the source, or added to the destination. This
231 ignores the amount of pure code movements within a file. In
232 other words, rearranging lines in a file is not counted as much
233 as other changes. This is the default behavior when no
234 parameter is given.
235 lines
237 Compute the dirstat numbers by doing the regular line-based
238 diff analysis, and summing the removed/added line counts. (For
239 binary files, count 64-byte chunks instead, since binary files
240 have no natural concept of lines). This is a more expensive
241 --dirstat behavior than the changes behavior, but it does count
242 rearranged lines within a file as much as other changes. The
243 resulting output is consistent with what you get from the other
244 --*stat options.
245 files
247 Compute the dirstat numbers by counting the number of files
248 changed. Each changed file counts equally in the dirstat
249 analysis. This is the computationally cheapest --dirstat
250 behavior, since it does not have to look at the file contents
251 at all.
252 cumulative
254 Count changes in a child directory for the parent directory as
255 well. Note that when using cumulative, the sum of the
256 percentages reported may exceed 100%. The default
257 (non-cumulative) behavior can be specified with the
258 noncumulative parameter.
259 <limit>
261 An integer parameter specifies a cut-off percent (3% by
262 default). Directories contributing less than this percentage of
263 the changes are not shown in the output.
264 Example: The following will count changed files, while ignoring
266 directories with less than 10% of the total amount of changed
267 files, and accumulating child directory counts in the parent
268 directories: --dirstat=files,10,cumulative.
269 --cumulative
271 Synonym for --dirstat=cumulative
272 --dirstat-by-file[=<param1,param2>...]
274 Synonym for --dirstat=files,param1,param2...
275 --summary
277 Output a condensed summary of extended header information such as
278 creations, renames and mode changes.
279 --patch-with-stat
281 Synonym for -p --stat.
282 -z
284 When --raw, --numstat, --name-only or --name-status has been given,
285 do not munge pathnames and use NULs as output field terminators.
286 Without this option, pathnames with "unusual" characters are quoted
288 as explained for the configuration variable core.quotePath (see
289 git-config(1)).
290 --name-only
292 Show only names of changed files. The file names are often encoded
293 in UTF-8. For more information see the discussion about encoding in
294 the git-log(1) manual page.
295 --name-status
297 Show only names and status of changed files. See the description of
298 the --diff-filter option on what the status letters mean. Just like
299 --name-only the file names are often encoded in UTF-8.
300 --submodule[=<format>]
302 Specify how differences in submodules are shown. When specifying
303 --submodule=short the short format is used. This format just shows
304 the names of the commits at the beginning and end of the range.
305 When --submodule or --submodule=log is specified, the log format is
306 used. This format lists the commits in the range like git-
307 submodule(1) summary does. When --submodule=diff is specified, the
308 diff format is used. This format shows an inline diff of the
309 changes in the submodule contents between the commit range.
310 Defaults to diff.submodule or the short format if the config option
311 is unset.
312 --color[=<when>]
314 Show colored diff. --color (i.e. without =<when>) is the same as
315 --color=always. <when> can be one of always, never, or auto. It
316 can be changed by the color.ui and color.diff configuration
317 settings.
318 --no-color
320 Turn off colored diff. This can be used to override configuration
321 settings. It is the same as --color=never.
322 --color-moved[=<mode>]
324 Moved lines of code are colored differently. It can be changed by
325 the diff.colorMoved configuration setting. The <mode> defaults to
326 no if the option is not given and to zebra if the option with no
327 mode is given. The mode must be one of:
328 no
330 Moved lines are not highlighted.
331 default
333 Is a synonym for zebra. This may change to a more sensible mode
334 in the future.
335 plain
337 Any line that is added in one location and was removed in
338 another location will be colored with color.diff.newMoved.
339 Similarly color.diff.oldMoved will be used for removed lines
340 that are added somewhere else in the diff. This mode picks up
341 any moved line, but it is not very useful in a review to
342 determine if a block of code was moved without permutation.
343 blocks
345 Blocks of moved text of at least 20 alphanumeric characters are
346 detected greedily. The detected blocks are painted using either
347 the color.diff.{old,new}Moved color. Adjacent blocks cannot be
348 told apart.
349 zebra
351 Blocks of moved text are detected as in blocks mode. The blocks
352 are painted using either the color.diff.{old,new}Moved color or
353 color.diff.{old,new}MovedAlternative. The change between the
354 two colors indicates that a new block was detected.
355 dimmed-zebra
357 Similar to zebra, but additional dimming of uninteresting parts
358 of moved code is performed. The bordering lines of two adjacent
359 blocks are considered interesting, the rest is uninteresting.
360 dimmed_zebra is a deprecated synonym.
361 --no-color-moved
363 Turn off move detection. This can be used to override configuration
364 settings. It is the same as --color-moved=no.
365 --color-moved-ws=<modes>
367 This configures how whitespace is ignored when performing the move
368 detection for --color-moved. It can be set by the diff.colorMovedWS
369 configuration setting. These modes can be given as a comma
370 separated list:
371 no
373 Do not ignore whitespace when performing move detection.
374 ignore-space-at-eol
376 Ignore changes in whitespace at EOL.
377 ignore-space-change
379 Ignore changes in amount of whitespace. This ignores whitespace
380 at line end, and considers all other sequences of one or more
381 whitespace characters to be equivalent.
382 ignore-all-space
384 Ignore whitespace when comparing lines. This ignores
385 differences even if one line has whitespace where the other
386 line has none.
387 allow-indentation-change
389 Initially ignore any whitespace in the move detection, then
390 group the moved code blocks only into a block if the change in
391 whitespace is the same per line. This is incompatible with the
392 other modes.
393 --no-color-moved-ws
395 Do not ignore whitespace when performing move detection. This can
396 be used to override configuration settings. It is the same as
397 --color-moved-ws=no.
398 --word-diff[=<mode>]
400 Show a word diff, using the <mode> to delimit changed words. By
401 default, words are delimited by whitespace; see --word-diff-regex
402 below. The <mode> defaults to plain, and must be one of:
403 color
405 Highlight changed words using only colors. Implies --color.
406 plain
408 Show words as [-removed-] and {+added+}. Makes no attempts to
409 escape the delimiters if they appear in the input, so the
410 output may be ambiguous.
411 porcelain
413 Use a special line-based format intended for script
414 consumption. Added/removed/unchanged runs are printed in the
415 usual unified diff format, starting with a +/-/` ` character at
416 the beginning of the line and extending to the end of the line.
417 Newlines in the input are represented by a tilde ~ on a line of
418 its own.
419 none
421 Disable word diff again.
422 Note that despite the name of the first mode, color is used to
424 highlight the changed parts in all modes if enabled.
425 --word-diff-regex=<regex>
427 Use <regex> to decide what a word is, instead of considering runs
428 of non-whitespace to be a word. Also implies --word-diff unless it
429 was already enabled.
430 Every non-overlapping match of the <regex> is considered a word.
432 Anything between these matches is considered whitespace and
433 ignored(!) for the purposes of finding differences. You may want to
434 append |[^[:space:]] to your regular expression to make sure that
435 it matches all non-whitespace characters. A match that contains a
436 newline is silently truncated(!) at the newline.
437 For example, --word-diff-regex=. will treat each character as a
439 word and, correspondingly, show differences character by character.
440 The regex can also be set via a diff driver or configuration
442 option, see gitattributes(5) or git-config(1). Giving it explicitly
443 overrides any diff driver or configuration setting. Diff drivers
444 override configuration settings.
445 --color-words[=<regex>]
447 Equivalent to --word-diff=color plus (if a regex was specified)
448 --word-diff-regex=<regex>.
449 --no-renames
451 Turn off rename detection, even when the configuration file gives
452 the default to do so.
453 --[no-]rename-empty
455 Whether to use empty blobs as rename source.
456 --check
458 Warn if changes introduce conflict markers or whitespace errors.
459 What are considered whitespace errors is controlled by
460 core.whitespace configuration. By default, trailing whitespaces
461 (including lines that consist solely of whitespaces) and a space
462 character that is immediately followed by a tab character inside
463 the initial indent of the line are considered whitespace errors.
464 Exits with non-zero status if problems are found. Not compatible
465 with --exit-code.
466 --ws-error-highlight=<kind>
468 Highlight whitespace errors in the context, old or new lines of the
469 diff. Multiple values are separated by comma, none resets previous
470 values, default reset the list to new and all is a shorthand for
471 old,new,context. When this option is not given, and the
472 configuration variable diff.wsErrorHighlight is not set, only
473 whitespace errors in new lines are highlighted. The whitespace
474 errors are colored with color.diff.whitespace.
475 --full-index
477 Instead of the first handful of characters, show the full pre- and
478 post-image blob object names on the "index" line when generating
479 patch format output.
480 --binary
482 In addition to --full-index, output a binary diff that can be
483 applied with git-apply. Implies --patch.
484 --abbrev[=<n>]
486 Instead of showing the full 40-byte hexadecimal object name in
487 diff-raw format output and diff-tree header lines, show the
488 shortest prefix that is at least <n> hexdigits long that uniquely
489 refers the object. In diff-patch output format, --full-index takes
490 higher precedence, i.e. if --full-index is specified, full blob
491 names will be shown regardless of --abbrev. Non default number of
492 digits can be specified with --abbrev=<n>.
493 -B[<n>][/<m>], --break-rewrites[=[<n>][/<m>]]
495 Break complete rewrite changes into pairs of delete and create.
496 This serves two purposes:
497 It affects the way a change that amounts to a total rewrite of a
499 file not as a series of deletion and insertion mixed together with
500 a very few lines that happen to match textually as the context, but
501 as a single deletion of everything old followed by a single
502 insertion of everything new, and the number m controls this aspect
503 of the -B option (defaults to 60%). -B/70% specifies that less
504 than 30% of the original should remain in the result for Git to
505 consider it a total rewrite (i.e. otherwise the resulting patch
506 will be a series of deletion and insertion mixed together with
507 context lines).
508 When used with -M, a totally-rewritten file is also considered as
510 the source of a rename (usually -M only considers a file that
511 disappeared as the source of a rename), and the number n controls
512 this aspect of the -B option (defaults to 50%). -B20% specifies
513 that a change with addition and deletion compared to 20% or more of
514 the file’s size are eligible for being picked up as a possible
515 source of a rename to another file.
516 -M[<n>], --find-renames[=<n>]
518 Detect renames. If n is specified, it is a threshold on the
519 similarity index (i.e. amount of addition/deletions compared to the
520 file’s size). For example, -M90% means Git should consider a
521 delete/add pair to be a rename if more than 90% of the file hasn’t
522 changed. Without a % sign, the number is to be read as a fraction,
523 with a decimal point before it. I.e., -M5 becomes 0.5, and is thus
524 the same as -M50%. Similarly, -M05 is the same as -M5%. To limit
525 detection to exact renames, use -M100%. The default similarity
526 index is 50%.
527 -C[<n>], --find-copies[=<n>]
529 Detect copies as well as renames. See also --find-copies-harder. If
530 n is specified, it has the same meaning as for -M<n>.
531 --find-copies-harder
533 For performance reasons, by default, -C option finds copies only if
534 the original file of the copy was modified in the same changeset.
535 This flag makes the command inspect unmodified files as candidates
536 for the source of copy. This is a very expensive operation for
537 large projects, so use it with caution. Giving more than one -C
538 option has the same effect.
539 -D, --irreversible-delete
541 Omit the preimage for deletes, i.e. print only the header but not
542 the diff between the preimage and /dev/null. The resulting patch is
543 not meant to be applied with patch or git apply; this is solely for
544 people who want to just concentrate on reviewing the text after the
545 change. In addition, the output obviously lacks enough information
546 to apply such a patch in reverse, even manually, hence the name of
547 the option.
548 When used together with -B, omit also the preimage in the deletion
550 part of a delete/create pair.
551 -l<num>
553 The -M and -C options involve some preliminary steps that can
554 detect subsets of renames/copies cheaply, followed by an exhaustive
555 fallback portion that compares all remaining unpaired destinations
556 to all relevant sources. (For renames, only remaining unpaired
557 sources are relevant; for copies, all original sources are
558 relevant.) For N sources and destinations, this exhaustive check is
559 O(N^2). This option prevents the exhaustive portion of rename/copy
560 detection from running if the number of source/destination files
561 involved exceeds the specified number. Defaults to
562 diff.renameLimit. Note that a value of 0 is treated as unlimited.
563 --diff-filter=[(A|C|D|M|R|T|U|X|B)...[*]]
565 Select only files that are Added (A), Copied (C), Deleted (D),
566 Modified (M), Renamed (R), have their type (i.e. regular file,
567 symlink, submodule, ...) changed (T), are Unmerged (U), are Unknown
568 (X), or have had their pairing Broken (B). Any combination of the
569 filter characters (including none) can be used. When *
570 (All-or-none) is added to the combination, all paths are selected
571 if there is any file that matches other criteria in the comparison;
572 if there is no file that matches other criteria, nothing is
573 selected.
574 Also, these upper-case letters can be downcased to exclude. E.g.
576 --diff-filter=ad excludes added and deleted paths.
577 Note that not all diffs can feature all types. For instance, copied
579 and renamed entries cannot appear if detection for those types is
580 disabled.
581 -S<string>
583 Look for differences that change the number of occurrences of the
584 specified string (i.e. addition/deletion) in a file. Intended for
585 the scripter’s use.
586 It is useful when you’re looking for an exact block of code (like a
588 struct), and want to know the history of that block since it first
589 came into being: use the feature iteratively to feed the
590 interesting block in the preimage back into -S, and keep going
591 until you get the very first version of the block.
592 Binary files are searched as well.
594 -G<regex>
596 Look for differences whose patch text contains added/removed lines
597 that match <regex>.
598 To illustrate the difference between -S<regex> --pickaxe-regex and
600 -G<regex>, consider a commit with the following diff in the same
601 file:
602 + return frotz(nitfol, two->ptr, 1, 0);
604 ...
605 - hit = frotz(nitfol, mf2.ptr, 1, 0);
606 While git log -G"frotz\(nitfol" will show this commit, git log
608 -S"frotz\(nitfol" --pickaxe-regex will not (because the number of
609 occurrences of that string did not change).
610 Unless --text is supplied patches of binary files without a
612 textconv filter will be ignored.
613 See the pickaxe entry in gitdiffcore(7) for more information.
615 --find-object=<object-id>
617 Look for differences that change the number of occurrences of the
618 specified object. Similar to -S, just the argument is different in
619 that it doesn’t search for a specific string but for a specific
620 object id.
621 The object can be a blob or a submodule commit. It implies the -t
623 option in git-log to also find trees.
624 --pickaxe-all
626 When -S or -G finds a change, show all the changes in that
627 changeset, not just the files that contain the change in <string>.
628 --pickaxe-regex
630 Treat the <string> given to -S as an extended POSIX regular
631 expression to match.
632 -O<orderfile>
634 Control the order in which files appear in the output. This
635 overrides the diff.orderFile configuration variable (see git-
636 config(1)). To cancel diff.orderFile, use -O/dev/null.
637 The output order is determined by the order of glob patterns in
639 <orderfile>. All files with pathnames that match the first pattern
640 are output first, all files with pathnames that match the second
641 pattern (but not the first) are output next, and so on. All files
642 with pathnames that do not match any pattern are output last, as if
643 there was an implicit match-all pattern at the end of the file. If
644 multiple pathnames have the same rank (they match the same pattern
645 but no earlier patterns), their output order relative to each other
646 is the normal order.
647 <orderfile> is parsed as follows:
649 • Blank lines are ignored, so they can be used as separators for
651 readability.
652 • Lines starting with a hash ("#") are ignored, so they can be
654 used for comments. Add a backslash ("\") to the beginning of
655 the pattern if it starts with a hash.
656 • Each other line contains a single pattern.
658 Patterns have the same syntax and semantics as patterns used for
660 fnmatch(3) without the FNM_PATHNAME flag, except a pathname also
661 matches a pattern if removing any number of the final pathname
662 components matches the pattern. For example, the pattern "foo*bar"
663 matches "fooasdfbar" and "foo/bar/baz/asdf" but not "foobarx".
664 --skip-to=<file>, --rotate-to=<file>
666 Discard the files before the named <file> from the output (i.e.
667 skip to), or move them to the end of the output (i.e. rotate to).
668 These options were invented primarily for the use of the git
669 difftool command, and may not be very useful otherwise.
670 -R
672 Swap two inputs; that is, show differences from index or on-disk
673 file to tree contents.
674 --relative[=<path>], --no-relative
676 When run from a subdirectory of the project, it can be told to
677 exclude changes outside the directory and show pathnames relative
678 to it with this option. When you are not in a subdirectory (e.g. in
679 a bare repository), you can name which subdirectory to make the
680 output relative to by giving a <path> as an argument.
681 --no-relative can be used to countermand both diff.relative config
682 option and previous --relative.
683 -a, --text
685 Treat all files as text.
686 --ignore-cr-at-eol
688 Ignore carriage-return at the end of line when doing a comparison.
689 --ignore-space-at-eol
691 Ignore changes in whitespace at EOL.
692 -b, --ignore-space-change
694 Ignore changes in amount of whitespace. This ignores whitespace at
695 line end, and considers all other sequences of one or more
696 whitespace characters to be equivalent.
697 -w, --ignore-all-space
699 Ignore whitespace when comparing lines. This ignores differences
700 even if one line has whitespace where the other line has none.
701 --ignore-blank-lines
703 Ignore changes whose lines are all blank.
704 -I<regex>, --ignore-matching-lines=<regex>
706 Ignore changes whose all lines match <regex>. This option may be
707 specified more than once.
708 --inter-hunk-context=<lines>
710 Show the context between diff hunks, up to the specified number of
711 lines, thereby fusing hunks that are close to each other. Defaults
712 to diff.interHunkContext or 0 if the config option is unset.
713 -W, --function-context
715 Show whole function as context lines for each change. The function
716 names are determined in the same way as git diff works out patch
717 hunk headers (see Defining a custom hunk-header in
718 gitattributes(5)).
719 --exit-code
721 Make the program exit with codes similar to diff(1). That is, it
722 exits with 1 if there were differences and 0 means no differences.
723 --quiet
725 Disable all output of the program. Implies --exit-code.
726 --ext-diff
728 Allow an external diff helper to be executed. If you set an
729 external diff driver with gitattributes(5), you need to use this
730 option with git-log(1) and friends.
731 --no-ext-diff
733 Disallow external diff drivers.
734 --textconv, --no-textconv
736 Allow (or disallow) external text conversion filters to be run when
737 comparing binary files. See gitattributes(5) for details. Because
738 textconv filters are typically a one-way conversion, the resulting
739 diff is suitable for human consumption, but cannot be applied. For
740 this reason, textconv filters are enabled by default only for git-
741 diff(1) and git-log(1), but not for git-format-patch(1) or diff
742 plumbing commands.
743 --ignore-submodules[=<when>]
745 Ignore changes to submodules in the diff generation. <when> can be
746 either "none", "untracked", "dirty" or "all", which is the default.
747 Using "none" will consider the submodule modified when it either
748 contains untracked or modified files or its HEAD differs from the
749 commit recorded in the superproject and can be used to override any
750 settings of the ignore option in git-config(1) or gitmodules(5).
751 When "untracked" is used submodules are not considered dirty when
752 they only contain untracked content (but they are still scanned for
753 modified content). Using "dirty" ignores all changes to the work
754 tree of submodules, only changes to the commits stored in the
755 superproject are shown (this was the behavior until 1.7.0). Using
756 "all" hides all changes to submodules.
757 --src-prefix=<prefix>
759 Show the given source prefix instead of "a/".
760 --dst-prefix=<prefix>
762 Show the given destination prefix instead of "b/".
763 --no-prefix
765 Do not show any source or destination prefix.
766 --default-prefix
768 Use the default source and destination prefixes ("a/" and "b/").
769 This is usually the default already, but may be used to override
770 config such as diff.noprefix.
771 --line-prefix=<prefix>
773 Prepend an additional prefix to every line of output.
774 --ita-invisible-in-index
776 By default entries added by "git add -N" appear as an existing
777 empty file in "git diff" and a new file in "git diff --cached".
778 This option makes the entry appear as a new file in "git diff" and
779 non-existent in "git diff --cached". This option could be reverted
780 with --ita-visible-in-index. Both options are experimental and
781 could be removed in future.
782 For more detailed explanation on these common options, see also
784 gitdiffcore(7).
785 -1 --base, -2 --ours, -3 --theirs
787 Compare the working tree with the "base" version (stage #1), "our
788 branch" (stage #2) or "their branch" (stage #3). The index contains
789 these stages only for unmerged entries i.e. while resolving
790 conflicts. See git-read-tree(1) section "3-Way Merge" for detailed
791 information.
792 -0
794 Omit diff output for unmerged entries and just show "Unmerged". Can
795 be used only when comparing the working tree with the index.
796 <path>...
798 The <paths> parameters, when given, are used to limit the diff to
799 the named paths (you can give directory names and get diff for all
800 files under them).
801
803 The raw output format from "git-diff-index", "git-diff-tree",
804 "git-diff-files" and "git diff --raw" are very similar.
805 These commands all compare two sets of things; what is compared
807 differs:
808 git-diff-index <tree-ish>
810 compares the <tree-ish> and the files on the filesystem.
811 git-diff-index --cached <tree-ish>
813 compares the <tree-ish> and the index.
814 git-diff-tree [-r] <tree-ish-1> <tree-ish-2> [<pattern>...]
816 compares the trees named by the two arguments.
817 git-diff-files [<pattern>...]
819 compares the index and the files on the filesystem.
820 The "git-diff-tree" command begins its output by printing the hash of
822 what is being compared. After that, all the commands print one output
823 line per changed file.
824 An output line is formatted this way:
826 in-place edit :100644 100644 bcd1234 0123456 M file0
828 copy-edit :100644 100644 abcd123 1234567 C68 file1 file2
829 rename-edit :100644 100644 abcd123 1234567 R86 file1 file3
830 create :000000 100644 0000000 1234567 A file4
831 delete :100644 000000 1234567 0000000 D file5
832 unmerged :000000 000000 0000000 0000000 U file6
833 That is, from the left to the right:
835 1. a colon.
837 2. mode for "src"; 000000 if creation or unmerged.
839 3. a space.
841 4. mode for "dst"; 000000 if deletion or unmerged.
843 5. a space.
845 6. sha1 for "src"; 0{40} if creation or unmerged.
847 7. a space.
849 8. sha1 for "dst"; 0{40} if deletion, unmerged or "work tree out of
851 sync with the index".
852 9. a space.
854 10. status, followed by optional "score" number.
856 11. a tab or a NUL when -z option is used.
858 12. path for "src"
860 13. a tab or a NUL when -z option is used; only exists for C or R.
862 14. path for "dst"; only exists for C or R.
864 15. an LF or a NUL when -z option is used, to terminate the record.
866 Possible status letters are:
868 • A: addition of a file
870 • C: copy of a file into a new one
872 • D: deletion of a file
874 • M: modification of the contents or mode of a file
876 • R: renaming of a file
878 • T: change in the type of the file (regular file, symbolic link or
880 submodule)
881 • U: file is unmerged (you must complete the merge before it can be
883 committed)
884 • X: "unknown" change type (most probably a bug, please report it)
886 Status letters C and R are always followed by a score (denoting the
888 percentage of similarity between the source and target of the move or
889 copy). Status letter M may be followed by a score (denoting the
890 percentage of dissimilarity) for file rewrites.
891 The sha1 for "dst" is shown as all 0’s if a file on the filesystem is
893 out of sync with the index.
894 Example:
896 :100644 100644 5be4a4a 0000000 M file.c
898 Without the -z option, pathnames with "unusual" characters are quoted
900 as explained for the configuration variable core.quotePath (see git-
901 config(1)). Using -z the filename is output verbatim and the line is
902 terminated by a NUL byte.
903
905 "git-diff-tree", "git-diff-files" and "git-diff --raw" can take -c or
906 --cc option to generate diff output also for merge commits. The output
907 differs from the format described above in the following way:
908 1. there is a colon for each parent
910 2. there are more "src" modes and "src" sha1
912 3. status is concatenated status characters for each parent
914 4. no optional "score" number
916 5. tab-separated pathname(s) of the file
918 For -c and --cc, only the destination or final path is shown even if
920 the file was renamed on any side of history. With --combined-all-paths,
921 the name of the path in each parent is shown followed by the name of
922 the path in the merge commit.
923 Examples for -c and --cc without --combined-all-paths:
925 ::100644 100644 100644 fabadb8 cc95eb0 4866510 MM desc.c
927 ::100755 100755 100755 52b7a2d 6d1ac04 d2ac7d7 RM bar.sh
928 ::100644 100644 100644 e07d6c5 9042e82 ee91881 RR phooey.c
929 Examples when --combined-all-paths added to either -c or --cc:
931 ::100644 100644 100644 fabadb8 cc95eb0 4866510 MM desc.c desc.c desc.c
933 ::100755 100755 100755 52b7a2d 6d1ac04 d2ac7d7 RM foo.sh bar.sh bar.sh
934 ::100644 100644 100644 e07d6c5 9042e82 ee91881 RR fooey.c fuey.c phooey.c
935 Note that combined diff lists only files which were modified from all
937 parents.
938
940 Running git-diff(1), git-log(1), git-show(1), git-diff-index(1), git-
941 diff-tree(1), or git-diff-files(1) with the -p option produces patch
942 text. You can customize the creation of patch text via the
943 GIT_EXTERNAL_DIFF and the GIT_DIFF_OPTS environment variables (see
944 git(1)), and the diff attribute (see gitattributes(5)).
945 What the -p option produces is slightly different from the traditional
947 diff format:
948 1. It is preceded by a "git diff" header that looks like this:
950 diff --git a/file1 b/file2
952 The a/ and b/ filenames are the same unless rename/copy is
954 involved. Especially, even for a creation or a deletion, /dev/null
955 is not used in place of the a/ or b/ filenames.
956 When a rename/copy is involved, file1 and file2 show the name of
958 the source file of the rename/copy and the name of the file that
959 the rename/copy produces, respectively.
960 2. It is followed by one or more extended header lines:
962 old mode <mode>
964 new mode <mode>
965 deleted file mode <mode>
966 new file mode <mode>
967 copy from <path>
968 copy to <path>
969 rename from <path>
970 rename to <path>
971 similarity index <number>
972 dissimilarity index <number>
973 index <hash>..<hash> <mode>
974 File modes are printed as 6-digit octal numbers including the file
976 type and file permission bits.
977 Path names in extended headers do not include the a/ and b/
979 prefixes.
980 The similarity index is the percentage of unchanged lines, and the
982 dissimilarity index is the percentage of changed lines. It is a
983 rounded down integer, followed by a percent sign. The similarity
984 index value of 100% is thus reserved for two equal files, while
985 100% dissimilarity means that no line from the old file made it
986 into the new one.
987 The index line includes the blob object names before and after the
989 change. The <mode> is included if the file mode does not change;
990 otherwise, separate lines indicate the old and the new mode.
991 3. Pathnames with "unusual" characters are quoted as explained for the
993 configuration variable core.quotePath (see git-config(1)).
994 4. All the file1 files in the output refer to files before the commit,
996 and all the file2 files refer to files after the commit. It is
997 incorrect to apply each change to each file sequentially. For
998 example, this patch will swap a and b:
999 diff --git a/a b/b
1001 rename from a
1002 rename to b
1003 diff --git a/b b/a
1004 rename from b
1005 rename to a
1006 5. Hunk headers mention the name of the function to which the hunk
1008 applies. See "Defining a custom hunk-header" in gitattributes(5)
1009 for details of how to tailor this to specific languages.
1010
1012 Any diff-generating command can take the -c or --cc option to produce a
1013 combined diff when showing a merge. This is the default format when
1014 showing merges with git-diff(1) or git-show(1). Note also that you can
1015 give suitable --diff-merges option to any of these commands to force
1016 generation of diffs in a specific format.
1017 A "combined diff" format looks like this:
1019 diff --combined describe.c
1021 index fabadb8,cc95eb0..4866510
1022 --- a/describe.c
1023 +++ b/describe.c
1024 @@@ -98,20 -98,12 +98,20 @@@
1025 return (a_date > b_date) ? -1 : (a_date == b_date) ? 0 : 1;
1026 }
1027 - static void describe(char *arg)
1029 -static void describe(struct commit *cmit, int last_one)
1030 ++static void describe(char *arg, int last_one)
1031 {
1032 + unsigned char sha1[20];
1033 + struct commit *cmit;
1034 struct commit_list *list;
1035 static int initialized = 0;
1036 struct commit_name *n;
1037 + if (get_sha1(arg, sha1) < 0)
1039 + usage(describe_usage);
1040 + cmit = lookup_commit_reference(sha1);
1041 + if (!cmit)
1042 + usage(describe_usage);
1043 +
1044 if (!initialized) {
1045 initialized = 1;
1046 for_each_ref(get_name);
1047 1. It is preceded by a "git diff" header, that looks like this (when
1049 the -c option is used):
1050 diff --combined file
1052 or like this (when the --cc option is used):
1054 diff --cc file
1056 2. It is followed by one or more extended header lines (this example
1058 shows a merge with two parents):
1059 index <hash>,<hash>..<hash>
1061 mode <mode>,<mode>..<mode>
1062 new file mode <mode>
1063 deleted file mode <mode>,<mode>
1064 The mode <mode>,<mode>..<mode> line appears only if at least one of
1066 the <mode> is different from the rest. Extended headers with
1067 information about detected content movement (renames and copying
1068 detection) are designed to work with the diff of two <tree-ish> and
1069 are not used by combined diff format.
1070 3. It is followed by a two-line from-file/to-file header:
1072 --- a/file
1074 +++ b/file
1075 Similar to the two-line header for the traditional unified diff
1077 format, /dev/null is used to signal created or deleted files.
1078 However, if the --combined-all-paths option is provided, instead of
1080 a two-line from-file/to-file, you get an N+1 line from-file/to-file
1081 header, where N is the number of parents in the merge commit:
1082 --- a/file
1084 --- a/file
1085 --- a/file
1086 +++ b/file
1087 This extended format can be useful if rename or copy detection is
1089 active, to allow you to see the original name of the file in
1090 different parents.
1091 4. Chunk header format is modified to prevent people from accidentally
1093 feeding it to patch -p1. Combined diff format was created for
1094 review of merge commit changes, and was not meant to be applied.
1095 The change is similar to the change in the extended index header:
1096 @@@ <from-file-range> <from-file-range> <to-file-range> @@@
1098 There are (number of parents + 1) @ characters in the chunk header
1100 for combined diff format.
1101 Unlike the traditional unified diff format, which shows two files A and
1103 B with a single column that has - (minus — appears in A but removed in
1104 B), + (plus — missing in A but added to B), or " " (space — unchanged)
1105 prefix, this format compares two or more files file1, file2,... with
1106 one file X, and shows how X differs from each of fileN. One column for
1107 each of fileN is prepended to the output line to note how X’s line is
1108 different from it.
1109 A - character in the column N means that the line appears in fileN but
1111 it does not appear in the result. A + character in the column N means
1112 that the line appears in the result, and fileN does not have that line
1113 (in other words, the line was added, from the point of view of that
1114 parent).
1115 In the above example output, the function signature was changed from
1117 both files (hence two - removals from both file1 and file2, plus ++ to
1118 mean one line that was added does not appear in either file1 or file2).
1119 Also, eight other lines are the same from file1 but do not appear in
1120 file2 (hence prefixed with +).
1121 When shown by git diff-tree -c, it compares the parents of a merge
1123 commit with the merge result (i.e. file1..fileN are the parents). When
1124 shown by git diff-files -c, it compares the two unresolved merge
1125 parents with the working tree file (i.e. file1 is stage 2 aka "our
1126 version", file2 is stage 3 aka "their version").
1127
1129 The --summary option describes newly added, deleted, renamed and copied
1130 files. The --stat option adds diffstat(1) graph to the output. These
1131 options can be combined with other options, such as -p, and are meant
1132 for human consumption.
1133 When showing a change that involves a rename or a copy, --stat output
1135 formats the pathnames compactly by combining common prefix and suffix
1136 of the pathnames. For example, a change that moves arch/i386/Makefile
1137 to arch/x86/Makefile while modifying 4 lines will be shown like this:
1138 arch/{i386 => x86}/Makefile | 4 +--
1140 The --numstat option gives the diffstat(1) information but is designed
1142 for easier machine consumption. An entry in --numstat output looks like
1143 this:
1144 1 2 README
1146 3 1 arch/{i386 => x86}/Makefile
1147 That is, from left to right:
1149 1. the number of added lines;
1151 2. a tab;
1153 3. the number of deleted lines;
1155 4. a tab;
1157 5. pathname (possibly with rename/copy information);
1159 6. a newline.
1161 When -z output option is in effect, the output is formatted this way:
1163 1 2 README NUL
1165 3 1 NUL arch/i386/Makefile NUL arch/x86/Makefile NUL
1166 That is:
1168 1. the number of added lines;
1170 2. a tab;
1172 3. the number of deleted lines;
1174 4. a tab;
1176 5. a NUL (only exists if renamed/copied);
1178 6. pathname in preimage;
1180 7. a NUL (only exists if renamed/copied);
1182 8. pathname in postimage (only exists if renamed/copied);
1184 9. a NUL.
1186 The extra NUL before the preimage path in renamed case is to allow
1188 scripts that read the output to tell if the current record being read
1189 is a single-path record or a rename/copy record without reading ahead.
1190 After reading added and deleted lines, reading up to NUL would yield
1191 the pathname, but if that is NUL, the record will show two paths.
1192
1194 Various ways to check your working tree
1195 $ git diff [1m(1)
1197 $ git diff --cached [1m(2)
1198 $ git diff HEAD [1m(3)
1199 $ git diff AUTO_MERGE [1m(4)
1200 1. Changes in the working tree not yet staged for the next
1202 commit.
1203 2. Changes between the index and your last commit; what you
1204 would be committing if you run git commit without -a
1205 option.
1206 3. Changes in the working tree since your last commit; what
1207 you would be committing if you run git commit -a
1208 4. Changes in the working tree you’ve made to resolve textual
1209 conflicts so far.
1210 Comparing with arbitrary commits
1212 $ git diff test [1m(1)
1214 $ git diff HEAD -- ./test [1m(2)
1215 $ git diff HEAD^ HEAD [1m(3)
1216 1. Instead of using the tip of the current branch, compare
1218 with the tip of "test" branch.
1219 2. Instead of comparing with the tip of "test" branch,
1220 compare with the tip of the current branch, but limit the
1221 comparison to the file "test".
1222 3. Compare the version before the last commit and the last
1223 commit.
1224 Comparing branches
1226 $ git diff topic master [1m(1)
1228 $ git diff topic..master [1m(2)
1229 $ git diff topic...master [1m(3)
1230 1. Changes between the tips of the topic and the master
1232 branches.
1233 2. Same as above.
1234 3. Changes that occurred on the master branch since when the
1235 topic branch was started off it.
1236 Limiting the diff output
1238 $ git diff --diff-filter=MRC [1m(1)
1240 $ git diff --name-status [1m(2)
1241 $ git diff arch/i386 include/asm-i386 [1m(3)
1242 1. Show only modification, rename, and copy, but not addition
1244 or deletion.
1245 2. Show only names and the nature of change, but not actual
1246 diff output.
1247 3. Limit diff output to named subtrees.
1248 Munging the diff output
1250 $ git diff --find-copies-harder -B -C [1m(1)
1252 $ git diff -R [1m(2)
1253 1. Spend extra cycles to find renames, copies and complete
1256 rewrites (very expensive).
1257 2. Output diff in reverse.
1258
1260 Everything below this line in this section is selectively included from
1261 the git-config(1) documentation. The content is the same as what’s
1262 found there:
1263 diff.autoRefreshIndex
1265 When using git diff to compare with work tree files, do not
1266 consider stat-only changes as changed. Instead, silently run git
1267 update-index --refresh to update the cached stat information for
1268 paths whose contents in the work tree match the contents in the
1269 index. This option defaults to true. Note that this affects only
1270 git diff Porcelain, and not lower level diff commands such as git
1271 diff-files.
1272 diff.dirstat
1274 A comma separated list of --dirstat parameters specifying the
1275 default behavior of the --dirstat option to git-diff(1) and
1276 friends. The defaults can be overridden on the command line (using
1277 --dirstat=<param1,param2,...>). The fallback defaults (when not
1278 changed by diff.dirstat) are changes,noncumulative,3. The following
1279 parameters are available:
1280 changes
1282 Compute the dirstat numbers by counting the lines that have
1283 been removed from the source, or added to the destination. This
1284 ignores the amount of pure code movements within a file. In
1285 other words, rearranging lines in a file is not counted as much
1286 as other changes. This is the default behavior when no
1287 parameter is given.
1288 lines
1290 Compute the dirstat numbers by doing the regular line-based
1291 diff analysis, and summing the removed/added line counts. (For
1292 binary files, count 64-byte chunks instead, since binary files
1293 have no natural concept of lines). This is a more expensive
1294 --dirstat behavior than the changes behavior, but it does count
1295 rearranged lines within a file as much as other changes. The
1296 resulting output is consistent with what you get from the other
1297 --*stat options.
1298 files
1300 Compute the dirstat numbers by counting the number of files
1301 changed. Each changed file counts equally in the dirstat
1302 analysis. This is the computationally cheapest --dirstat
1303 behavior, since it does not have to look at the file contents
1304 at all.
1305 cumulative
1307 Count changes in a child directory for the parent directory as
1308 well. Note that when using cumulative, the sum of the
1309 percentages reported may exceed 100%. The default
1310 (non-cumulative) behavior can be specified with the
1311 noncumulative parameter.
1312 <limit>
1314 An integer parameter specifies a cut-off percent (3% by
1315 default). Directories contributing less than this percentage of
1316 the changes are not shown in the output.
1317 Example: The following will count changed files, while ignoring
1319 directories with less than 10% of the total amount of changed
1320 files, and accumulating child directory counts in the parent
1321 directories: files,10,cumulative.
1322 diff.statNameWidth
1324 Limit the width of the filename part in --stat output. If set,
1325 applies to all commands generating --stat output except
1326 format-patch.
1327 diff.statGraphWidth
1329 Limit the width of the graph part in --stat output. If set, applies
1330 to all commands generating --stat output except format-patch.
1331 diff.context
1333 Generate diffs with <n> lines of context instead of the default of
1334 3. This value is overridden by the -U option.
1335 diff.interHunkContext
1337 Show the context between diff hunks, up to the specified number of
1338 lines, thereby fusing the hunks that are close to each other. This
1339 value serves as the default for the --inter-hunk-context command
1340 line option.
1341 diff.external
1343 If this config variable is set, diff generation is not performed
1344 using the internal diff machinery, but using the given command. Can
1345 be overridden with the “GIT_EXTERNAL_DIFF” environment variable.
1346 The command is called with parameters as described under "git
1347 Diffs" in git(1). Note: if you want to use an external diff program
1348 only on a subset of your files, you might want to use
1349 gitattributes(5) instead.
1350 diff.ignoreSubmodules
1352 Sets the default value of --ignore-submodules. Note that this
1353 affects only git diff Porcelain, and not lower level diff commands
1354 such as git diff-files. git checkout and git switch also honor
1355 this setting when reporting uncommitted changes. Setting it to all
1356 disables the submodule summary normally shown by git commit and git
1357 status when status.submoduleSummary is set unless it is overridden
1358 by using the --ignore-submodules command-line option. The git
1359 submodule commands are not affected by this setting. By default
1360 this is set to untracked so that any untracked submodules are
1361 ignored.
1362 diff.mnemonicPrefix
1364 If set, git diff uses a prefix pair that is different from the
1365 standard "a/" and "b/" depending on what is being compared. When
1366 this configuration is in effect, reverse diff output also swaps the
1367 order of the prefixes:
1368 git diff
1370 compares the (i)ndex and the (w)ork tree;
1371 git diff HEAD
1373 compares a (c)ommit and the (w)ork tree;
1374 git diff --cached
1376 compares a (c)ommit and the (i)ndex;
1377 git diff HEAD:file1 file2
1379 compares an (o)bject and a (w)ork tree entity;
1380 git diff --no-index a b
1382 compares two non-git things (1) and (2).
1383 diff.noprefix
1385 If set, git diff does not show any source or destination prefix.
1386 diff.relative
1388 If set to true, git diff does not show changes outside of the
1389 directory and show pathnames relative to the current directory.
1390 diff.orderFile
1392 File indicating how to order files within a diff. See the -O option
1393 to git-diff(1) for details. If diff.orderFile is a relative
1394 pathname, it is treated as relative to the top of the working tree.
1395 diff.renameLimit
1397 The number of files to consider in the exhaustive portion of
1398 copy/rename detection; equivalent to the git diff option -l. If not
1399 set, the default value is currently 1000. This setting has no
1400 effect if rename detection is turned off.
1401 diff.renames
1403 Whether and how Git detects renames. If set to "false", rename
1404 detection is disabled. If set to "true", basic rename detection is
1405 enabled. If set to "copies" or "copy", Git will detect copies, as
1406 well. Defaults to true. Note that this affects only git diff
1407 Porcelain like git-diff(1) and git-log(1), and not lower level
1408 commands such as git-diff-files(1).
1409 diff.suppressBlankEmpty
1411 A boolean to inhibit the standard behavior of printing a space
1412 before each empty output line. Defaults to false.
1413 diff.submodule
1415 Specify the format in which differences in submodules are shown.
1416 The "short" format just shows the names of the commits at the
1417 beginning and end of the range. The "log" format lists the commits
1418 in the range like git-submodule(1) summary does. The "diff" format
1419 shows an inline diff of the changed contents of the submodule.
1420 Defaults to "short".
1421 diff.wordRegex
1423 A POSIX Extended Regular Expression used to determine what is a
1424 "word" when performing word-by-word difference calculations.
1425 Character sequences that match the regular expression are "words",
1426 all other characters are ignorable whitespace.
1427 diff.<driver>.command
1429 The custom diff driver command. See gitattributes(5) for details.
1430 diff.<driver>.xfuncname
1432 The regular expression that the diff driver should use to recognize
1433 the hunk header. A built-in pattern may also be used. See
1434 gitattributes(5) for details.
1435 diff.<driver>.binary
1437 Set this option to true to make the diff driver treat files as
1438 binary. See gitattributes(5) for details.
1439 diff.<driver>.textconv
1441 The command that the diff driver should call to generate the
1442 text-converted version of a file. The result of the conversion is
1443 used to generate a human-readable diff. See gitattributes(5) for
1444 details.
1445 diff.<driver>.wordRegex
1447 The regular expression that the diff driver should use to split
1448 words in a line. See gitattributes(5) for details.
1449 diff.<driver>.cachetextconv
1451 Set this option to true to make the diff driver cache the text
1452 conversion outputs. See gitattributes(5) for details.
1453 araxis
1455 Use Araxis Merge (requires a graphical session)
1456 bc
1458 Use Beyond Compare (requires a graphical session)
1459 bc3
1461 Use Beyond Compare (requires a graphical session)
1462 bc4
1464 Use Beyond Compare (requires a graphical session)
1465 codecompare
1467 Use Code Compare (requires a graphical session)
1468 deltawalker
1470 Use DeltaWalker (requires a graphical session)
1471 diffmerge
1473 Use DiffMerge (requires a graphical session)
1474 diffuse
1476 Use Diffuse (requires a graphical session)
1477 ecmerge
1479 Use ECMerge (requires a graphical session)
1480 emerge
1482 Use Emacs' Emerge
1483 examdiff
1485 Use ExamDiff Pro (requires a graphical session)
1486 guiffy
1488 Use Guiffy’s Diff Tool (requires a graphical session)
1489 gvimdiff
1491 Use gVim (requires a graphical session)
1492 kdiff3
1494 Use KDiff3 (requires a graphical session)
1495 kompare
1497 Use Kompare (requires a graphical session)
1498 meld
1500 Use Meld (requires a graphical session)
1501 nvimdiff
1503 Use Neovim
1504 opendiff
1506 Use FileMerge (requires a graphical session)
1507 p4merge
1509 Use HelixCore P4Merge (requires a graphical session)
1510 smerge
1512 Use Sublime Merge (requires a graphical session)
1513 tkdiff
1515 Use TkDiff (requires a graphical session)
1516 vimdiff
1518 Use Vim
1519 winmerge
1521 Use WinMerge (requires a graphical session)
1522 xxdiff
1524 Use xxdiff (requires a graphical session)
1525 diff.indentHeuristic
1527 Set this option to false to disable the default heuristics that
1528 shift diff hunk boundaries to make patches easier to read.
1529 diff.algorithm
1531 Choose a diff algorithm. The variants are as follows:
1532 default, myers
1534 The basic greedy diff algorithm. Currently, this is the
1535 default.
1536 minimal
1538 Spend extra time to make sure the smallest possible diff is
1539 produced.
1540 patience
1542 Use "patience diff" algorithm when generating patches.
1543 histogram
1545 This algorithm extends the patience algorithm to "support
1546 low-occurrence common elements".
1547 diff.wsErrorHighlight
1549 Highlight whitespace errors in the context, old or new lines of the
1550 diff. Multiple values are separated by comma, none resets previous
1551 values, default reset the list to new and all is a shorthand for
1552 old,new,context. The whitespace errors are colored with
1553 color.diff.whitespace. The command line option
1554 --ws-error-highlight=<kind> overrides this setting.
1555 diff.colorMoved
1557 If set to either a valid <mode> or a true value, moved lines in a
1558 diff are colored differently, for details of valid modes see
1559 --color-moved in git-diff(1). If simply set to true the default
1560 color mode will be used. When set to false, moved lines are not
1561 colored.
1562 diff.colorMovedWS
1564 When moved lines are colored using e.g. the diff.colorMoved
1565 setting, this option controls the <mode> how spaces are treated for
1566 details of valid modes see --color-moved-ws in git-diff(1).
1567
1569 diff(1), git-difftool(1), git-log(1), gitdiffcore(7), git-format-
1570 patch(1), git-apply(1), git-show(1)
1571
1573 Part of the git(1) suite
1574Git 2.43.0 11/20/2023 GIT-DIFF(1)