1GIT-DIFF-INDEX(1) Git Manual GIT-DIFF-INDEX(1)
2
6 git-diff-index - Compare a tree to the working tree or index
7
9 git diff-index [-m] [--cached] [--merge-base] [<common diff options>] <tree-ish> [<path>...]
10
12 Compares the content and mode of the blobs found in a tree object with
13 the corresponding tracked files in the working tree, or with the
14 corresponding paths in the index. When <path> arguments are present,
15 compares only paths matching those patterns. Otherwise all tracked
16 files are compared.
17
19 -p, -u, --patch
20 Generate patch (see section on generating patches).
21 -s, --no-patch
23 Suppress diff output. Useful for commands like git show that show
24 the patch by default, or to cancel the effect of --patch.
25 -U<n>, --unified=<n>
27 Generate diffs with <n> lines of context instead of the usual
28 three. Implies --patch.
29 --output=<file>
31 Output to a specific file instead of stdout.
32 --output-indicator-new=<char>, --output-indicator-old=<char>,
34 --output-indicator-context=<char>
35 Specify the character used to indicate new, old or context lines in
36 the generated patch. Normally they are +, - and ' ' respectively.
37 --raw
39 Generate the diff in raw format. This is the default.
40 --patch-with-raw
42 Synonym for -p --raw.
43 --indent-heuristic
45 Enable the heuristic that shifts diff hunk boundaries to make
46 patches easier to read. This is the default.
47 --no-indent-heuristic
49 Disable the indent heuristic.
50 --minimal
52 Spend extra time to make sure the smallest possible diff is
53 produced.
54 --patience
56 Generate a diff using the "patience diff" algorithm.
57 --histogram
59 Generate a diff using the "histogram diff" algorithm.
60 --anchored=<text>
62 Generate a diff using the "anchored diff" algorithm.
63 This option may be specified more than once.
65 If a line exists in both the source and destination, exists only
67 once, and starts with this text, this algorithm attempts to prevent
68 it from appearing as a deletion or addition in the output. It uses
69 the "patience diff" algorithm internally.
70 --diff-algorithm={patience|minimal|histogram|myers}
72 Choose a diff algorithm. The variants are as follows:
73 default, myers
75 The basic greedy diff algorithm. Currently, this is the
76 default.
77 minimal
79 Spend extra time to make sure the smallest possible diff is
80 produced.
81 patience
83 Use "patience diff" algorithm when generating patches.
84 histogram
86 This algorithm extends the patience algorithm to "support
87 low-occurrence common elements".
88 For instance, if you configured the diff.algorithm variable to a
90 non-default value and want to use the default one, then you have to
91 use --diff-algorithm=default option.
92 --stat[=<width>[,<name-width>[,<count>]]]
94 Generate a diffstat. By default, as much space as necessary will be
95 used for the filename part, and the rest for the graph part.
96 Maximum width defaults to terminal width, or 80 columns if not
97 connected to a terminal, and can be overridden by <width>. The
98 width of the filename part can be limited by giving another width
99 <name-width> after a comma. The width of the graph part can be
100 limited by using --stat-graph-width=<width> (affects all commands
101 generating a stat graph) or by setting diff.statGraphWidth=<width>
102 (does not affect git format-patch). By giving a third parameter
103 <count>, you can limit the output to the first <count> lines,
104 followed by ... if there are more.
105 These parameters can also be set individually with
107 --stat-width=<width>, --stat-name-width=<name-width> and
108 --stat-count=<count>.
109 --compact-summary
111 Output a condensed summary of extended header information such as
112 file creations or deletions ("new" or "gone", optionally "+l" if
113 it’s a symlink) and mode changes ("+x" or "-x" for adding or
114 removing executable bit respectively) in diffstat. The information
115 is put between the filename part and the graph part. Implies
116 --stat.
117 --numstat
119 Similar to --stat, but shows number of added and deleted lines in
120 decimal notation and pathname without abbreviation, to make it more
121 machine friendly. For binary files, outputs two - instead of saying
122 0 0.
123 --shortstat
125 Output only the last line of the --stat format containing total
126 number of modified files, as well as number of added and deleted
127 lines.
128 -X[<param1,param2,...>], --dirstat[=<param1,param2,...>]
130 Output the distribution of relative amount of changes for each
131 sub-directory. The behavior of --dirstat can be customized by
132 passing it a comma separated list of parameters. The defaults are
133 controlled by the diff.dirstat configuration variable (see git-
134 config(1)). The following parameters are available:
135 changes
137 Compute the dirstat numbers by counting the lines that have
138 been removed from the source, or added to the destination. This
139 ignores the amount of pure code movements within a file. In
140 other words, rearranging lines in a file is not counted as much
141 as other changes. This is the default behavior when no
142 parameter is given.
143 lines
145 Compute the dirstat numbers by doing the regular line-based
146 diff analysis, and summing the removed/added line counts. (For
147 binary files, count 64-byte chunks instead, since binary files
148 have no natural concept of lines). This is a more expensive
149 --dirstat behavior than the changes behavior, but it does count
150 rearranged lines within a file as much as other changes. The
151 resulting output is consistent with what you get from the other
152 --*stat options.
153 files
155 Compute the dirstat numbers by counting the number of files
156 changed. Each changed file counts equally in the dirstat
157 analysis. This is the computationally cheapest --dirstat
158 behavior, since it does not have to look at the file contents
159 at all.
160 cumulative
162 Count changes in a child directory for the parent directory as
163 well. Note that when using cumulative, the sum of the
164 percentages reported may exceed 100%. The default
165 (non-cumulative) behavior can be specified with the
166 noncumulative parameter.
167 <limit>
169 An integer parameter specifies a cut-off percent (3% by
170 default). Directories contributing less than this percentage of
171 the changes are not shown in the output.
172 Example: The following will count changed files, while ignoring
174 directories with less than 10% of the total amount of changed
175 files, and accumulating child directory counts in the parent
176 directories: --dirstat=files,10,cumulative.
177 --cumulative
179 Synonym for --dirstat=cumulative
180 --dirstat-by-file[=<param1,param2>...]
182 Synonym for --dirstat=files,param1,param2...
183 --summary
185 Output a condensed summary of extended header information such as
186 creations, renames and mode changes.
187 --patch-with-stat
189 Synonym for -p --stat.
190 -z
192 When --raw, --numstat, --name-only or --name-status has been given,
193 do not munge pathnames and use NULs as output field terminators.
194 Without this option, pathnames with "unusual" characters are quoted
196 as explained for the configuration variable core.quotePath (see
197 git-config(1)).
198 --name-only
200 Show only names of changed files.
201 --name-status
203 Show only names and status of changed files. See the description of
204 the --diff-filter option on what the status letters mean.
205 --submodule[=<format>]
207 Specify how differences in submodules are shown. When specifying
208 --submodule=short the short format is used. This format just shows
209 the names of the commits at the beginning and end of the range.
210 When --submodule or --submodule=log is specified, the log format is
211 used. This format lists the commits in the range like git-
212 submodule(1) summary does. When --submodule=diff is specified, the
213 diff format is used. This format shows an inline diff of the
214 changes in the submodule contents between the commit range.
215 Defaults to diff.submodule or the short format if the config option
216 is unset.
217 --color[=<when>]
219 Show colored diff. --color (i.e. without =<when>) is the same as
220 --color=always. <when> can be one of always, never, or auto.
221 --no-color
223 Turn off colored diff. It is the same as --color=never.
224 --color-moved[=<mode>]
226 Moved lines of code are colored differently. The <mode> defaults to
227 no if the option is not given and to zebra if the option with no
228 mode is given. The mode must be one of:
229 no
231 Moved lines are not highlighted.
232 default
234 Is a synonym for zebra. This may change to a more sensible mode
235 in the future.
236 plain
238 Any line that is added in one location and was removed in
239 another location will be colored with color.diff.newMoved.
240 Similarly color.diff.oldMoved will be used for removed lines
241 that are added somewhere else in the diff. This mode picks up
242 any moved line, but it is not very useful in a review to
243 determine if a block of code was moved without permutation.
244 blocks
246 Blocks of moved text of at least 20 alphanumeric characters are
247 detected greedily. The detected blocks are painted using either
248 the color.diff.{old,new}Moved color. Adjacent blocks cannot be
249 told apart.
250 zebra
252 Blocks of moved text are detected as in blocks mode. The blocks
253 are painted using either the color.diff.{old,new}Moved color or
254 color.diff.{old,new}MovedAlternative. The change between the
255 two colors indicates that a new block was detected.
256 dimmed-zebra
258 Similar to zebra, but additional dimming of uninteresting parts
259 of moved code is performed. The bordering lines of two adjacent
260 blocks are considered interesting, the rest is uninteresting.
261 dimmed_zebra is a deprecated synonym.
262 --no-color-moved
264 Turn off move detection. This can be used to override configuration
265 settings. It is the same as --color-moved=no.
266 --color-moved-ws=<modes>
268 This configures how whitespace is ignored when performing the move
269 detection for --color-moved. These modes can be given as a comma
270 separated list:
271 no
273 Do not ignore whitespace when performing move detection.
274 ignore-space-at-eol
276 Ignore changes in whitespace at EOL.
277 ignore-space-change
279 Ignore changes in amount of whitespace. This ignores whitespace
280 at line end, and considers all other sequences of one or more
281 whitespace characters to be equivalent.
282 ignore-all-space
284 Ignore whitespace when comparing lines. This ignores
285 differences even if one line has whitespace where the other
286 line has none.
287 allow-indentation-change
289 Initially ignore any whitespace in the move detection, then
290 group the moved code blocks only into a block if the change in
291 whitespace is the same per line. This is incompatible with the
292 other modes.
293 --no-color-moved-ws
295 Do not ignore whitespace when performing move detection. This can
296 be used to override configuration settings. It is the same as
297 --color-moved-ws=no.
298 --word-diff[=<mode>]
300 Show a word diff, using the <mode> to delimit changed words. By
301 default, words are delimited by whitespace; see --word-diff-regex
302 below. The <mode> defaults to plain, and must be one of:
303 color
305 Highlight changed words using only colors. Implies --color.
306 plain
308 Show words as [-removed-] and {+added+}. Makes no attempts to
309 escape the delimiters if they appear in the input, so the
310 output may be ambiguous.
311 porcelain
313 Use a special line-based format intended for script
314 consumption. Added/removed/unchanged runs are printed in the
315 usual unified diff format, starting with a +/-/` ` character at
316 the beginning of the line and extending to the end of the line.
317 Newlines in the input are represented by a tilde ~ on a line of
318 its own.
319 none
321 Disable word diff again.
322 Note that despite the name of the first mode, color is used to
324 highlight the changed parts in all modes if enabled.
325 --word-diff-regex=<regex>
327 Use <regex> to decide what a word is, instead of considering runs
328 of non-whitespace to be a word. Also implies --word-diff unless it
329 was already enabled.
330 Every non-overlapping match of the <regex> is considered a word.
332 Anything between these matches is considered whitespace and
333 ignored(!) for the purposes of finding differences. You may want to
334 append |[^[:space:]] to your regular expression to make sure that
335 it matches all non-whitespace characters. A match that contains a
336 newline is silently truncated(!) at the newline.
337 For example, --word-diff-regex=. will treat each character as a
339 word and, correspondingly, show differences character by character.
340 The regex can also be set via a diff driver or configuration
342 option, see gitattributes(5) or git-config(1). Giving it explicitly
343 overrides any diff driver or configuration setting. Diff drivers
344 override configuration settings.
345 --color-words[=<regex>]
347 Equivalent to --word-diff=color plus (if a regex was specified)
348 --word-diff-regex=<regex>.
349 --no-renames
351 Turn off rename detection, even when the configuration file gives
352 the default to do so.
353 --[no-]rename-empty
355 Whether to use empty blobs as rename source.
356 --check
358 Warn if changes introduce conflict markers or whitespace errors.
359 What are considered whitespace errors is controlled by
360 core.whitespace configuration. By default, trailing whitespaces
361 (including lines that consist solely of whitespaces) and a space
362 character that is immediately followed by a tab character inside
363 the initial indent of the line are considered whitespace errors.
364 Exits with non-zero status if problems are found. Not compatible
365 with --exit-code.
366 --ws-error-highlight=<kind>
368 Highlight whitespace errors in the context, old or new lines of the
369 diff. Multiple values are separated by comma, none resets previous
370 values, default reset the list to new and all is a shorthand for
371 old,new,context. When this option is not given, and the
372 configuration variable diff.wsErrorHighlight is not set, only
373 whitespace errors in new lines are highlighted. The whitespace
374 errors are colored with color.diff.whitespace.
375 --full-index
377 Instead of the first handful of characters, show the full pre- and
378 post-image blob object names on the "index" line when generating
379 patch format output.
380 --binary
382 In addition to --full-index, output a binary diff that can be
383 applied with git-apply. Implies --patch.
384 --abbrev[=<n>]
386 Instead of showing the full 40-byte hexadecimal object name in
387 diff-raw format output and diff-tree header lines, show the
388 shortest prefix that is at least <n> hexdigits long that uniquely
389 refers the object. In diff-patch output format, --full-index takes
390 higher precedence, i.e. if --full-index is specified, full blob
391 names will be shown regardless of --abbrev. Non default number of
392 digits can be specified with --abbrev=<n>.
393 -B[<n>][/<m>], --break-rewrites[=[<n>][/<m>]]
395 Break complete rewrite changes into pairs of delete and create.
396 This serves two purposes:
397 It affects the way a change that amounts to a total rewrite of a
399 file not as a series of deletion and insertion mixed together with
400 a very few lines that happen to match textually as the context, but
401 as a single deletion of everything old followed by a single
402 insertion of everything new, and the number m controls this aspect
403 of the -B option (defaults to 60%). -B/70% specifies that less
404 than 30% of the original should remain in the result for Git to
405 consider it a total rewrite (i.e. otherwise the resulting patch
406 will be a series of deletion and insertion mixed together with
407 context lines).
408 When used with -M, a totally-rewritten file is also considered as
410 the source of a rename (usually -M only considers a file that
411 disappeared as the source of a rename), and the number n controls
412 this aspect of the -B option (defaults to 50%). -B20% specifies
413 that a change with addition and deletion compared to 20% or more of
414 the file’s size are eligible for being picked up as a possible
415 source of a rename to another file.
416 -M[<n>], --find-renames[=<n>]
418 Detect renames. If n is specified, it is a threshold on the
419 similarity index (i.e. amount of addition/deletions compared to the
420 file’s size). For example, -M90% means Git should consider a
421 delete/add pair to be a rename if more than 90% of the file hasn’t
422 changed. Without a % sign, the number is to be read as a fraction,
423 with a decimal point before it. I.e., -M5 becomes 0.5, and is thus
424 the same as -M50%. Similarly, -M05 is the same as -M5%. To limit
425 detection to exact renames, use -M100%. The default similarity
426 index is 50%.
427 -C[<n>], --find-copies[=<n>]
429 Detect copies as well as renames. See also --find-copies-harder. If
430 n is specified, it has the same meaning as for -M<n>.
431 --find-copies-harder
433 For performance reasons, by default, -C option finds copies only if
434 the original file of the copy was modified in the same changeset.
435 This flag makes the command inspect unmodified files as candidates
436 for the source of copy. This is a very expensive operation for
437 large projects, so use it with caution. Giving more than one -C
438 option has the same effect.
439 -D, --irreversible-delete
441 Omit the preimage for deletes, i.e. print only the header but not
442 the diff between the preimage and /dev/null. The resulting patch is
443 not meant to be applied with patch or git apply; this is solely for
444 people who want to just concentrate on reviewing the text after the
445 change. In addition, the output obviously lacks enough information
446 to apply such a patch in reverse, even manually, hence the name of
447 the option.
448 When used together with -B, omit also the preimage in the deletion
450 part of a delete/create pair.
451 -l<num>
453 The -M and -C options require O(n^2) processing time where n is the
454 number of potential rename/copy targets. This option prevents
455 rename/copy detection from running if the number of rename/copy
456 targets exceeds the specified number.
457 --diff-filter=[(A|C|D|M|R|T|U|X|B)...[*]]
459 Select only files that are Added (A), Copied (C), Deleted (D),
460 Modified (M), Renamed (R), have their type (i.e. regular file,
461 symlink, submodule, ...) changed (T), are Unmerged (U), are Unknown
462 (X), or have had their pairing Broken (B). Any combination of the
463 filter characters (including none) can be used. When *
464 (All-or-none) is added to the combination, all paths are selected
465 if there is any file that matches other criteria in the comparison;
466 if there is no file that matches other criteria, nothing is
467 selected.
468 Also, these upper-case letters can be downcased to exclude. E.g.
470 --diff-filter=ad excludes added and deleted paths.
471 Note that not all diffs can feature all types. For instance, diffs
473 from the index to the working tree can never have Added entries
474 (because the set of paths included in the diff is limited by what
475 is in the index). Similarly, copied and renamed entries cannot
476 appear if detection for those types is disabled.
477 -S<string>
479 Look for differences that change the number of occurrences of the
480 specified string (i.e. addition/deletion) in a file. Intended for
481 the scripter’s use.
482 It is useful when you’re looking for an exact block of code (like a
484 struct), and want to know the history of that block since it first
485 came into being: use the feature iteratively to feed the
486 interesting block in the preimage back into -S, and keep going
487 until you get the very first version of the block.
488 Binary files are searched as well.
490 -G<regex>
492 Look for differences whose patch text contains added/removed lines
493 that match <regex>.
494 To illustrate the difference between -S<regex> --pickaxe-regex and
496 -G<regex>, consider a commit with the following diff in the same
497 file:
498 + return frotz(nitfol, two->ptr, 1, 0);
500 ...
501 - hit = frotz(nitfol, mf2.ptr, 1, 0);
502 While git log -G"frotz\(nitfol" will show this commit, git log
504 -S"frotz\(nitfol" --pickaxe-regex will not (because the number of
505 occurrences of that string did not change).
506 Unless --text is supplied patches of binary files without a
508 textconv filter will be ignored.
509 See the pickaxe entry in gitdiffcore(7) for more information.
511 --find-object=<object-id>
513 Look for differences that change the number of occurrences of the
514 specified object. Similar to -S, just the argument is different in
515 that it doesn’t search for a specific string but for a specific
516 object id.
517 The object can be a blob or a submodule commit. It implies the -t
519 option in git-log to also find trees.
520 --pickaxe-all
522 When -S or -G finds a change, show all the changes in that
523 changeset, not just the files that contain the change in <string>.
524 --pickaxe-regex
526 Treat the <string> given to -S as an extended POSIX regular
527 expression to match.
528 -O<orderfile>
530 Control the order in which files appear in the output. This
531 overrides the diff.orderFile configuration variable (see git-
532 config(1)). To cancel diff.orderFile, use -O/dev/null.
533 The output order is determined by the order of glob patterns in
535 <orderfile>. All files with pathnames that match the first pattern
536 are output first, all files with pathnames that match the second
537 pattern (but not the first) are output next, and so on. All files
538 with pathnames that do not match any pattern are output last, as if
539 there was an implicit match-all pattern at the end of the file. If
540 multiple pathnames have the same rank (they match the same pattern
541 but no earlier patterns), their output order relative to each other
542 is the normal order.
543 <orderfile> is parsed as follows:
545 • Blank lines are ignored, so they can be used as separators for
547 readability.
548 • Lines starting with a hash ("#") are ignored, so they can be
550 used for comments. Add a backslash ("\") to the beginning of
551 the pattern if it starts with a hash.
552 • Each other line contains a single pattern.
554 Patterns have the same syntax and semantics as patterns used for
556 fnmatch(3) without the FNM_PATHNAME flag, except a pathname also
557 matches a pattern if removing any number of the final pathname
558 components matches the pattern. For example, the pattern "foo*bar"
559 matches "fooasdfbar" and "foo/bar/baz/asdf" but not "foobarx".
560 --skip-to=<file>, --rotate-to=<file>
562 Discard the files before the named <file> from the output (i.e.
563 skip to), or move them to the end of the output (i.e. rotate to).
564 These were invented primarily for use of the git difftool command,
565 and may not be very useful otherwise.
566 -R
568 Swap two inputs; that is, show differences from index or on-disk
569 file to tree contents.
570 --relative[=<path>], --no-relative
572 When run from a subdirectory of the project, it can be told to
573 exclude changes outside the directory and show pathnames relative
574 to it with this option. When you are not in a subdirectory (e.g. in
575 a bare repository), you can name which subdirectory to make the
576 output relative to by giving a <path> as an argument.
577 --no-relative can be used to countermand both diff.relative config
578 option and previous --relative.
579 -a, --text
581 Treat all files as text.
582 --ignore-cr-at-eol
584 Ignore carriage-return at the end of line when doing a comparison.
585 --ignore-space-at-eol
587 Ignore changes in whitespace at EOL.
588 -b, --ignore-space-change
590 Ignore changes in amount of whitespace. This ignores whitespace at
591 line end, and considers all other sequences of one or more
592 whitespace characters to be equivalent.
593 -w, --ignore-all-space
595 Ignore whitespace when comparing lines. This ignores differences
596 even if one line has whitespace where the other line has none.
597 --ignore-blank-lines
599 Ignore changes whose lines are all blank.
600 -I<regex>, --ignore-matching-lines=<regex>
602 Ignore changes whose all lines match <regex>. This option may be
603 specified more than once.
604 --inter-hunk-context=<lines>
606 Show the context between diff hunks, up to the specified number of
607 lines, thereby fusing hunks that are close to each other. Defaults
608 to diff.interHunkContext or 0 if the config option is unset.
609 -W, --function-context
611 Show whole function as context lines for each change. The function
612 names are determined in the same way as git diff works out patch
613 hunk headers (see Defining a custom hunk-header in
614 gitattributes(5)).
615 --exit-code
617 Make the program exit with codes similar to diff(1). That is, it
618 exits with 1 if there were differences and 0 means no differences.
619 --quiet
621 Disable all output of the program. Implies --exit-code.
622 --ext-diff
624 Allow an external diff helper to be executed. If you set an
625 external diff driver with gitattributes(5), you need to use this
626 option with git-log(1) and friends.
627 --no-ext-diff
629 Disallow external diff drivers.
630 --textconv, --no-textconv
632 Allow (or disallow) external text conversion filters to be run when
633 comparing binary files. See gitattributes(5) for details. Because
634 textconv filters are typically a one-way conversion, the resulting
635 diff is suitable for human consumption, but cannot be applied. For
636 this reason, textconv filters are enabled by default only for git-
637 diff(1) and git-log(1), but not for git-format-patch(1) or diff
638 plumbing commands.
639 --ignore-submodules[=<when>]
641 Ignore changes to submodules in the diff generation. <when> can be
642 either "none", "untracked", "dirty" or "all", which is the default.
643 Using "none" will consider the submodule modified when it either
644 contains untracked or modified files or its HEAD differs from the
645 commit recorded in the superproject and can be used to override any
646 settings of the ignore option in git-config(1) or gitmodules(5).
647 When "untracked" is used submodules are not considered dirty when
648 they only contain untracked content (but they are still scanned for
649 modified content). Using "dirty" ignores all changes to the work
650 tree of submodules, only changes to the commits stored in the
651 superproject are shown (this was the behavior until 1.7.0). Using
652 "all" hides all changes to submodules.
653 --src-prefix=<prefix>
655 Show the given source prefix instead of "a/".
656 --dst-prefix=<prefix>
658 Show the given destination prefix instead of "b/".
659 --no-prefix
661 Do not show any source or destination prefix.
662 --line-prefix=<prefix>
664 Prepend an additional prefix to every line of output.
665 --ita-invisible-in-index
667 By default entries added by "git add -N" appear as an existing
668 empty file in "git diff" and a new file in "git diff --cached".
669 This option makes the entry appear as a new file in "git diff" and
670 non-existent in "git diff --cached". This option could be reverted
671 with --ita-visible-in-index. Both options are experimental and
672 could be removed in future.
673 For more detailed explanation on these common options, see also
675 gitdiffcore(7).
676 <tree-ish>
678 The id of a tree object to diff against.
679 --cached
681 Do not consider the on-disk file at all.
682 --merge-base
684 Instead of comparing <tree-ish> directly, use the merge base
685 between <tree-ish> and HEAD instead. <tree-ish> must be a commit.
686 -m
688 By default, files recorded in the index but not checked out are
689 reported as deleted. This flag makes git diff-index say that all
690 non-checked-out files are up to date.
691
693 The raw output format from "git-diff-index", "git-diff-tree",
694 "git-diff-files" and "git diff --raw" are very similar.
695 These commands all compare two sets of things; what is compared
697 differs:
698 git-diff-index <tree-ish>
700 compares the <tree-ish> and the files on the filesystem.
701 git-diff-index --cached <tree-ish>
703 compares the <tree-ish> and the index.
704 git-diff-tree [-r] <tree-ish-1> <tree-ish-2> [<pattern>...]
706 compares the trees named by the two arguments.
707 git-diff-files [<pattern>...]
709 compares the index and the files on the filesystem.
710 The "git-diff-tree" command begins its output by printing the hash of
712 what is being compared. After that, all the commands print one output
713 line per changed file.
714 An output line is formatted this way:
716 in-place edit :100644 100644 bcd1234 0123456 M file0
718 copy-edit :100644 100644 abcd123 1234567 C68 file1 file2
719 rename-edit :100644 100644 abcd123 1234567 R86 file1 file3
720 create :000000 100644 0000000 1234567 A file4
721 delete :100644 000000 1234567 0000000 D file5
722 unmerged :000000 000000 0000000 0000000 U file6
723 That is, from the left to the right:
725 1. a colon.
727 2. mode for "src"; 000000 if creation or unmerged.
729 3. a space.
731 4. mode for "dst"; 000000 if deletion or unmerged.
733 5. a space.
735 6. sha1 for "src"; 0{40} if creation or unmerged.
737 7. a space.
739 8. sha1 for "dst"; 0{40} if creation, unmerged or "look at work tree".
741 9. a space.
743 10. status, followed by optional "score" number.
745 11. a tab or a NUL when -z option is used.
747 12. path for "src"
749 13. a tab or a NUL when -z option is used; only exists for C or R.
751 14. path for "dst"; only exists for C or R.
753 15. an LF or a NUL when -z option is used, to terminate the record.
755 Possible status letters are:
757 • A: addition of a file
759 • C: copy of a file into a new one
761 • D: deletion of a file
763 • M: modification of the contents or mode of a file
765 • R: renaming of a file
767 • T: change in the type of the file
769 • U: file is unmerged (you must complete the merge before it can be
771 committed)
772 • X: "unknown" change type (most probably a bug, please report it)
774 Status letters C and R are always followed by a score (denoting the
776 percentage of similarity between the source and target of the move or
777 copy). Status letter M may be followed by a score (denoting the
778 percentage of dissimilarity) for file rewrites.
779 <sha1> is shown as all 0’s if a file is new on the filesystem and it is
781 out of sync with the index.
782 Example:
784 :100644 100644 5be4a4a 0000000 M file.c
786 Without the -z option, pathnames with "unusual" characters are quoted
788 as explained for the configuration variable core.quotePath (see git-
789 config(1)). Using -z the filename is output verbatim and the line is
790 terminated by a NUL byte.
791
793 "git-diff-tree", "git-diff-files" and "git-diff --raw" can take -c or
794 --cc option to generate diff output also for merge commits. The output
795 differs from the format described above in the following way:
796 1. there is a colon for each parent
798 2. there are more "src" modes and "src" sha1
800 3. status is concatenated status characters for each parent
802 4. no optional "score" number
804 5. tab-separated pathname(s) of the file
806 For -c and --cc, only the destination or final path is shown even if
808 the file was renamed on any side of history. With --combined-all-paths,
809 the name of the path in each parent is shown followed by the name of
810 the path in the merge commit.
811 Examples for -c and --cc without --combined-all-paths:
813 ::100644 100644 100644 fabadb8 cc95eb0 4866510 MM desc.c
815 ::100755 100755 100755 52b7a2d 6d1ac04 d2ac7d7 RM bar.sh
816 ::100644 100644 100644 e07d6c5 9042e82 ee91881 RR phooey.c
817 Examples when --combined-all-paths added to either -c or --cc:
819 ::100644 100644 100644 fabadb8 cc95eb0 4866510 MM desc.c desc.c desc.c
821 ::100755 100755 100755 52b7a2d 6d1ac04 d2ac7d7 RM foo.sh bar.sh bar.sh
822 ::100644 100644 100644 e07d6c5 9042e82 ee91881 RR fooey.c fuey.c phooey.c
823 Note that combined diff lists only files which were modified from all
825 parents.
826
828 Running git-diff(1), git-log(1), git-show(1), git-diff-index(1), git-
829 diff-tree(1), or git-diff-files(1) with the -p option produces patch
830 text. You can customize the creation of patch text via the
831 GIT_EXTERNAL_DIFF and the GIT_DIFF_OPTS environment variables (see
832 git(1)).
833 What the -p option produces is slightly different from the traditional
835 diff format:
836 1. It is preceded with a "git diff" header that looks like this:
838 diff --git a/file1 b/file2
840 The a/ and b/ filenames are the same unless rename/copy is
842 involved. Especially, even for a creation or a deletion, /dev/null
843 is not used in place of the a/ or b/ filenames.
844 When rename/copy is involved, file1 and file2 show the name of the
846 source file of the rename/copy and the name of the file that
847 rename/copy produces, respectively.
848 2. It is followed by one or more extended header lines:
850 old mode <mode>
852 new mode <mode>
853 deleted file mode <mode>
854 new file mode <mode>
855 copy from <path>
856 copy to <path>
857 rename from <path>
858 rename to <path>
859 similarity index <number>
860 dissimilarity index <number>
861 index <hash>..<hash> <mode>
862 File modes are printed as 6-digit octal numbers including the file
864 type and file permission bits.
865 Path names in extended headers do not include the a/ and b/
867 prefixes.
868 The similarity index is the percentage of unchanged lines, and the
870 dissimilarity index is the percentage of changed lines. It is a
871 rounded down integer, followed by a percent sign. The similarity
872 index value of 100% is thus reserved for two equal files, while
873 100% dissimilarity means that no line from the old file made it
874 into the new one.
875 The index line includes the blob object names before and after the
877 change. The <mode> is included if the file mode does not change;
878 otherwise, separate lines indicate the old and the new mode.
879 3. Pathnames with "unusual" characters are quoted as explained for the
881 configuration variable core.quotePath (see git-config(1)).
882 4. All the file1 files in the output refer to files before the commit,
884 and all the file2 files refer to files after the commit. It is
885 incorrect to apply each change to each file sequentially. For
886 example, this patch will swap a and b:
887 diff --git a/a b/b
889 rename from a
890 rename to b
891 diff --git a/b b/a
892 rename from b
893 rename to a
894
896 Any diff-generating command can take the -c or --cc option to produce a
897 combined diff when showing a merge. This is the default format when
898 showing merges with git-diff(1) or git-show(1). Note also that you can
899 give suitable --diff-merges option to any of these commands to force
900 generation of diffs in specific format.
901 A "combined diff" format looks like this:
903 diff --combined describe.c
905 index fabadb8,cc95eb0..4866510
906 --- a/describe.c
907 +++ b/describe.c
908 @@@ -98,20 -98,12 +98,20 @@@
909 return (a_date > b_date) ? -1 : (a_date == b_date) ? 0 : 1;
910 }
911 - static void describe(char *arg)
913 -static void describe(struct commit *cmit, int last_one)
914 ++static void describe(char *arg, int last_one)
915 {
916 + unsigned char sha1[20];
917 + struct commit *cmit;
918 struct commit_list *list;
919 static int initialized = 0;
920 struct commit_name *n;
921 + if (get_sha1(arg, sha1) < 0)
923 + usage(describe_usage);
924 + cmit = lookup_commit_reference(sha1);
925 + if (!cmit)
926 + usage(describe_usage);
927 +
928 if (!initialized) {
929 initialized = 1;
930 for_each_ref(get_name);
931 1. It is preceded with a "git diff" header, that looks like this (when
933 the -c option is used):
934 diff --combined file
936 or like this (when the --cc option is used):
938 diff --cc file
940 2. It is followed by one or more extended header lines (this example
942 shows a merge with two parents):
943 index <hash>,<hash>..<hash>
945 mode <mode>,<mode>..<mode>
946 new file mode <mode>
947 deleted file mode <mode>,<mode>
948 The mode <mode>,<mode>..<mode> line appears only if at least one of
950 the <mode> is different from the rest. Extended headers with
951 information about detected contents movement (renames and copying
952 detection) are designed to work with diff of two <tree-ish> and are
953 not used by combined diff format.
954 3. It is followed by two-line from-file/to-file header
956 --- a/file
958 +++ b/file
959 Similar to two-line header for traditional unified diff format,
961 /dev/null is used to signal created or deleted files.
962 However, if the --combined-all-paths option is provided, instead of
964 a two-line from-file/to-file you get a N+1 line from-file/to-file
965 header, where N is the number of parents in the merge commit
966 --- a/file
968 --- a/file
969 --- a/file
970 +++ b/file
971 This extended format can be useful if rename or copy detection is
973 active, to allow you to see the original name of the file in
974 different parents.
975 4. Chunk header format is modified to prevent people from accidentally
977 feeding it to patch -p1. Combined diff format was created for
978 review of merge commit changes, and was not meant to be applied.
979 The change is similar to the change in the extended index header:
980 @@@ <from-file-range> <from-file-range> <to-file-range> @@@
982 There are (number of parents + 1) @ characters in the chunk header
984 for combined diff format.
985 Unlike the traditional unified diff format, which shows two files A and
987 B with a single column that has - (minus — appears in A but removed in
988 B), + (plus — missing in A but added to B), or " " (space — unchanged)
989 prefix, this format compares two or more files file1, file2,... with
990 one file X, and shows how X differs from each of fileN. One column for
991 each of fileN is prepended to the output line to note how X’s line is
992 different from it.
993 A - character in the column N means that the line appears in fileN but
995 it does not appear in the result. A + character in the column N means
996 that the line appears in the result, and fileN does not have that line
997 (in other words, the line was added, from the point of view of that
998 parent).
999 In the above example output, the function signature was changed from
1001 both files (hence two - removals from both file1 and file2, plus ++ to
1002 mean one line that was added does not appear in either file1 or file2).
1003 Also eight other lines are the same from file1 but do not appear in
1004 file2 (hence prefixed with +).
1005 When shown by git diff-tree -c, it compares the parents of a merge
1007 commit with the merge result (i.e. file1..fileN are the parents). When
1008 shown by git diff-files -c, it compares the two unresolved merge
1009 parents with the working tree file (i.e. file1 is stage 2 aka "our
1010 version", file2 is stage 3 aka "their version").
1011
1013 The --summary option describes newly added, deleted, renamed and copied
1014 files. The --stat option adds diffstat(1) graph to the output. These
1015 options can be combined with other options, such as -p, and are meant
1016 for human consumption.
1017 When showing a change that involves a rename or a copy, --stat output
1019 formats the pathnames compactly by combining common prefix and suffix
1020 of the pathnames. For example, a change that moves arch/i386/Makefile
1021 to arch/x86/Makefile while modifying 4 lines will be shown like this:
1022 arch/{i386 => x86}/Makefile | 4 +--
1024 The --numstat option gives the diffstat(1) information but is designed
1026 for easier machine consumption. An entry in --numstat output looks like
1027 this:
1028 1 2 README
1030 3 1 arch/{i386 => x86}/Makefile
1031 That is, from left to right:
1033 1. the number of added lines;
1035 2. a tab;
1037 3. the number of deleted lines;
1039 4. a tab;
1041 5. pathname (possibly with rename/copy information);
1043 6. a newline.
1045 When -z output option is in effect, the output is formatted this way:
1047 1 2 README NUL
1049 3 1 NUL arch/i386/Makefile NUL arch/x86/Makefile NUL
1050 That is:
1052 1. the number of added lines;
1054 2. a tab;
1056 3. the number of deleted lines;
1058 4. a tab;
1060 5. a NUL (only exists if renamed/copied);
1062 6. pathname in preimage;
1064 7. a NUL (only exists if renamed/copied);
1066 8. pathname in postimage (only exists if renamed/copied);
1068 9. a NUL.
1070 The extra NUL before the preimage path in renamed case is to allow
1072 scripts that read the output to tell if the current record being read
1073 is a single-path record or a rename/copy record without reading ahead.
1074 After reading added and deleted lines, reading up to NUL would yield
1075 the pathname, but if that is NUL, the record will show two paths.
1076
1078 You can choose whether you want to trust the index file entirely (using
1079 the --cached flag) or ask the diff logic to show any files that don’t
1080 match the stat state as being "tentatively changed". Both of these
1081 operations are very useful indeed.
1082
1084 If --cached is specified, it allows you to ask:
1085 show me the differences between HEAD and the current index
1087 contents (the ones I'd write using 'git write-tree')
1088 For example, let’s say that you have worked on your working directory,
1090 updated some files in the index and are ready to commit. You want to
1091 see exactly what you are going to commit, without having to write a new
1092 tree object and compare it that way, and to do that, you just do
1093 git diff-index --cached HEAD
1095 Example: let’s say I had renamed commit.c to git-commit.c, and I had
1097 done an update-index to make that effective in the index file. git
1098 diff-files wouldn’t show anything at all, since the index file matches
1099 my working directory. But doing a git diff-index does:
1100 torvalds@ppc970:~/git> git diff-index --cached HEAD
1102 -100644 blob 4161aecc6700a2eb579e842af0b7f22b98443f74 commit.c
1103 +100644 blob 4161aecc6700a2eb579e842af0b7f22b98443f74 git-commit.c
1104 You can see easily that the above is a rename.
1106 In fact, git diff-index --cached should always be entirely equivalent
1108 to actually doing a git write-tree and comparing that. Except this one
1109 is much nicer for the case where you just want to check where you are.
1110 So doing a git diff-index --cached is basically very useful when you
1112 are asking yourself "what have I already marked for being committed,
1113 and what’s the difference to a previous tree".
1114
1116 The "non-cached" mode takes a different approach, and is potentially
1117 the more useful of the two in that what it does can’t be emulated with
1118 a git write-tree + git diff-tree. Thus that’s the default mode. The
1119 non-cached version asks the question:
1120 show me the differences between HEAD and the currently checked out
1122 tree - index contents _and_ files that aren't up to date
1123 which is obviously a very useful question too, since that tells you
1125 what you could commit. Again, the output matches the git diff-tree -r
1126 output to a tee, but with a twist.
1127 The twist is that if some file doesn’t match the index, we don’t have a
1129 backing store thing for it, and we use the magic "all-zero" sha1 to
1130 show that. So let’s say that you have edited kernel/sched.c, but have
1131 not actually done a git update-index on it yet - there is no "object"
1132 associated with the new state, and you get:
1133 torvalds@ppc970:~/v2.6/linux> git diff-index --abbrev HEAD
1135 :100644 100664 7476bb... 000000... kernel/sched.c
1136 i.e., it shows that the tree has changed, and that kernel/sched.c is
1138 not up to date and may contain new stuff. The all-zero sha1 means that
1139 to get the real diff, you need to look at the object in the working
1140 directory directly rather than do an object-to-object diff.
1141 Note
1143 As with other commands of this type, git diff-index does not
1144 actually look at the contents of the file at all. So maybe
1145 kernel/sched.c hasn’t actually changed, and it’s just that you
1146 touched it. In either case, it’s a note that you need to git
1147 update-index it to make the index be in sync.
1148 Note
1150 You can have a mixture of files show up as "has been updated" and
1151 "is still dirty in the working directory" together. You can always
1152 tell which file is in which state, since the "has been updated"
1153 ones show a valid sha1, and the "not in sync with the index" ones
1154 will always have the special all-zero sha1.
1155
1157 Part of the git(1) suite
1158Git 2.31.1 2021-03-26 GIT-DIFF-INDEX(1)