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 [<commit>] [--] [<path>...]
11 git diff [options] <commit> <commit> [--] [<path>...]
12 git diff [options] [--no-index] [--] <path> <path>
13
16 Show changes between the working tree and the index or a tree, changes
17 between the index and a tree, changes between two trees, or changes
18 between two files on disk.
19 git diff [--options] [--] [<path>...]
21 This form is to view the changes you made relative to the index
22 (staging area for the next commit). In other words, the differences
23 are what you could tell git to further add to the index but you
24 still haven’t. You can stage these changes by using git-add(1).
25 If exactly two paths are given and at least one points outside the
27 current repository, git diff will compare the two files /
28 directories. This behavior can be forced by --no-index.
29 git diff [--options] --cached [<commit>] [--] [<path>...]
31 This form is to view the changes you staged for the next commit
32 relative to the named <commit>. Typically you would want comparison
33 with the latest commit, so if you do not give <commit>, it defaults
34 to HEAD. If HEAD does not exist (e.g. unborned branches) and
35 <commit> is not given, it shows all staged changes. --staged is a
36 synonym of --cached.
37 git diff [--options] <commit> [--] [<path>...]
39 This form is to view the changes you have in your working tree
40 relative to the named <commit>. You can use HEAD to compare it with
41 the latest commit, or a branch name to compare with the tip of a
42 different branch.
43 git diff [--options] <commit> <commit> [--] [<path>...]
45 This is to view the changes between two arbitrary <commit>.
46 git diff [--options] <commit>..<commit> [--] [<path>...]
48 This is synonymous to the previous form. If <commit> on one side is
49 omitted, it will have the same effect as using HEAD instead.
50 git diff [--options] <commit>...<commit> [--] [<path>...]
52 This form is to view the changes on the branch containing and up to
53 the second <commit>, starting at a common ancestor of both
54 <commit>. "git diff A...B" is equivalent to "git diff
55 $(git-merge-base A B) B". You can omit any one of <commit>, which
56 has the same effect as using HEAD instead.
57 Just in case if you are doing something exotic, it should be noted that
59 all of the <commit> in the above description, except in the last two
60 forms that use ".." notations, can be any <tree>. The third form (git
61 diff <commit> <commit>) can also be used to compare two <blob> objects.
62 For a more complete list of ways to spell <commit>, see "SPECIFYING
64 REVISIONS" section in gitrevisions(7). However, "diff" is about
65 comparing two endpoints, not ranges, and the range notations
66 ("<commit>..<commit>" and "<commit>...<commit>") do not mean a range as
67 defined in the "SPECIFYING RANGES" section in gitrevisions(7).
68
70 -p, -u, --patch
71 Generate patch (see section on generating patches). This is the
72 default.
73 -U<n>, --unified=<n>
75 Generate diffs with <n> lines of context instead of the usual
76 three. Implies -p.
77 --raw
79 Generate the raw format.
80 --patch-with-raw
82 Synonym for -p --raw.
83 --patience
85 Generate a diff using the "patience diff" algorithm.
86 --stat[=<width>[,<name-width>]]
88 Generate a diffstat. You can override the default output width for
89 80-column terminal by --stat=<width>. The width of the filename
90 part can be controlled by giving another width to it separated by a
91 comma.
92 --numstat
94 Similar to --stat, but shows number of added and deleted lines in
95 decimal notation and pathname without abbreviation, to make it more
96 machine friendly. For binary files, outputs two - instead of saying
97 0 0.
98 --shortstat
100 Output only the last line of the --stat format containing total
101 number of modified files, as well as number of added and deleted
102 lines.
103 --dirstat[=<limit>]
105 Output the distribution of relative amount of changes (number of
106 lines added or removed) for each sub-directory. Directories with
107 changes below a cut-off percent (3% by default) are not shown. The
108 cut-off percent can be set with --dirstat=<limit>. Changes in a
109 child directory are not counted for the parent directory, unless
110 --cumulative is used.
111 --dirstat-by-file[=<limit>]
113 Same as --dirstat, but counts changed files instead of lines.
114 --summary
116 Output a condensed summary of extended header information such as
117 creations, renames and mode changes.
118 --patch-with-stat
120 Synonym for -p --stat.
121 -z
123 When --raw, --numstat, --name-only or --name-status has been given,
124 do not munge pathnames and use NULs as output field terminators.
125 Without this option, each pathname output will have TAB, LF, double
127 quotes, and backslash characters replaced with \t, \n, \", and \\,
128 respectively, and the pathname will be enclosed in double quotes if
129 any of those replacements occurred.
130 --name-only
132 Show only names of changed files.
133 --name-status
135 Show only names and status of changed files. See the description of
136 the --diff-filter option on what the status letters mean.
137 --submodule[=<format>]
139 Chose the output format for submodule differences. <format> can be
140 one of short and log. short just shows pairs of commit names, this
141 format is used when this option is not given. log is the default
142 value for this option and lists the commits in that commit range
143 like the summary option of git-submodule(1) does.
144 --color[=<when>]
146 Show colored diff. The value must be always (the default), never,
147 or auto.
148 --no-color
150 Turn off colored diff, even when the configuration file gives the
151 default to color output. Same as --color=never.
152 --word-diff[=<mode>]
154 Show a word diff, using the <mode> to delimit changed words. By
155 default, words are delimited by whitespace; see --word-diff-regex
156 below. The <mode> defaults to plain, and must be one of:
157 color
159 Highlight changed words using only colors. Implies --color.
160 plain
162 Show words as [-removed-] and {added}. Makes no attempts to
163 escape the delimiters if they appear in the input, so the
164 output may be ambiguous.
165 porcelain
167 Use a special line-based format intended for script
168 consumption. Added/removed/unchanged runs are printed in the
169 usual unified diff format, starting with a +/-/` ` character at
170 the beginning of the line and extending to the end of the line.
171 Newlines in the input are represented by a tilde ~ on a line of
172 its own.
173 none
175 Disable word diff again.
176 Note that despite the name of the first mode, color is used to
178 highlight the changed parts in all modes if enabled.
179 --word-diff-regex=<regex>
181 Use <regex> to decide what a word is, instead of considering runs
182 of non-whitespace to be a word. Also implies --word-diff unless it
183 was already enabled.
184 Every non-overlapping match of the <regex> is considered a word.
186 Anything between these matches is considered whitespace and
187 ignored(!) for the purposes of finding differences. You may want to
188 append |[^[:space:]] to your regular expression to make sure that
189 it matches all non-whitespace characters. A match that contains a
190 newline is silently truncated(!) at the newline.
191 The regex can also be set via a diff driver or configuration
193 option, see gitattributes(1) or git-config(1). Giving it explicitly
194 overrides any diff driver or configuration setting. Diff drivers
195 override configuration settings.
196 --color-words[=<regex>]
198 Equivalent to --word-diff=color plus (if a regex was specified)
199 --word-diff-regex=<regex>.
200 --no-renames
202 Turn off rename detection, even when the configuration file gives
203 the default to do so.
204 --check
206 Warn if changes introduce trailing whitespace or an indent that
207 uses a space before a tab. Exits with non-zero status if problems
208 are found. Not compatible with --exit-code.
209 --full-index
211 Instead of the first handful of characters, show the full pre- and
212 post-image blob object names on the "index" line when generating
213 patch format output.
214 --binary
216 In addition to --full-index, output a binary diff that can be
217 applied with git-apply.
218 --abbrev[=<n>]
220 Instead of showing the full 40-byte hexadecimal object name in
221 diff-raw format output and diff-tree header lines, show only a
222 partial prefix. This is independent of the --full-index option
223 above, which controls the diff-patch output format. Non default
224 number of digits can be specified with --abbrev=<n>.
225 -B[<n>][/<m>], --break-rewrites[=[<n>][/<m>]]
227 Break complete rewrite changes into pairs of delete and create.
228 This serves two purposes:
229 It affects the way a change that amounts to a total rewrite of a
231 file not as a series of deletion and insertion mixed together with
232 a very few lines that happen to match textually as the context, but
233 as a single deletion of everything old followed by a single
234 insertion of everything new, and the number m controls this aspect
235 of the -B option (defaults to 60%). -B/70% specifies that less
236 than 30% of the original should remain in the result for git to
237 consider it a total rewrite (i.e. otherwise the resulting patch
238 will be a series of deletion and insertion mixed together with
239 context lines).
240 When used with -M, a totally-rewritten file is also considered as
242 the source of a rename (usually -M only considers a file that
243 disappeared as the source of a rename), and the number n controls
244 this aspect of the -B option (defaults to 50%). -B20% specifies
245 that a change with addition and deletion compared to 20% or more of
246 the file’s size are eligible for being picked up as a possible
247 source of a rename to another file.
248 -M[<n>], --find-renames[=<n>]
250 Detect renames. If n is specified, it is a is a threshold on the
251 similarity index (i.e. amount of addition/deletions compared to the
252 file’s size). For example, -M90% means git should consider a
253 delete/add pair to be a rename if more than 90% of the file hasn’t
254 changed.
255 -C[<n>], --find-copies[=<n>]
257 Detect copies as well as renames. See also --find-copies-harder. If
258 n is specified, it has the same meaning as for -M<n>.
259 --find-copies-harder
261 For performance reasons, by default, -C option finds copies only if
262 the original file of the copy was modified in the same changeset.
263 This flag makes the command inspect unmodified files as candidates
264 for the source of copy. This is a very expensive operation for
265 large projects, so use it with caution. Giving more than one -C
266 option has the same effect.
267 -l<num>
269 The -M and -C options require O(n^2) processing time where n is the
270 number of potential rename/copy targets. This option prevents
271 rename/copy detection from running if the number of rename/copy
272 targets exceeds the specified number.
273 --diff-filter=[(A|C|D|M|R|T|U|X|B)...[*]]
275 Select only files that are Added (A), Copied (C), Deleted (D),
276 Modified (M), Renamed (R), have their type (i.e. regular file,
277 symlink, submodule, ...) changed (T), are Unmerged (U), are Unknown
278 (X), or have had their pairing Broken (B). Any combination of the
279 filter characters (including none) can be used. When *
280 (All-or-none) is added to the combination, all paths are selected
281 if there is any file that matches other criteria in the comparison;
282 if there is no file that matches other criteria, nothing is
283 selected.
284 -S<string>
286 Look for differences that introduce or remove an instance of
287 <string>. Note that this is different than the string simply
288 appearing in diff output; see the pickaxe entry in gitdiffcore(7)
289 for more details.
290 -G<regex>
292 Look for differences whose added or removed line matches the given
293 <regex>.
294 --pickaxe-all
296 When -S or -G finds a change, show all the changes in that
297 changeset, not just the files that contain the change in <string>.
298 --pickaxe-regex
300 Make the <string> not a plain string but an extended POSIX regex to
301 match.
302 -O<orderfile>
304 Output the patch in the order specified in the <orderfile>, which
305 has one shell glob pattern per line.
306 -R
308 Swap two inputs; that is, show differences from index or on-disk
309 file to tree contents.
310 --relative[=<path>]
312 When run from a subdirectory of the project, it can be told to
313 exclude changes outside the directory and show pathnames relative
314 to it with this option. When you are not in a subdirectory (e.g. in
315 a bare repository), you can name which subdirectory to make the
316 output relative to by giving a <path> as an argument.
317 -a, --text
319 Treat all files as text.
320 --ignore-space-at-eol
322 Ignore changes in whitespace at EOL.
323 -b, --ignore-space-change
325 Ignore changes in amount of whitespace. This ignores whitespace at
326 line end, and considers all other sequences of one or more
327 whitespace characters to be equivalent.
328 -w, --ignore-all-space
330 Ignore whitespace when comparing lines. This ignores differences
331 even if one line has whitespace where the other line has none.
332 --inter-hunk-context=<lines>
334 Show the context between diff hunks, up to the specified number of
335 lines, thereby fusing hunks that are close to each other.
336 --exit-code
338 Make the program exit with codes similar to diff(1). That is, it
339 exits with 1 if there were differences and 0 means no differences.
340 --quiet
342 Disable all output of the program. Implies --exit-code.
343 --ext-diff
345 Allow an external diff helper to be executed. If you set an
346 external diff driver with gitattributes(5), you need to use this
347 option with git-log(1) and friends.
348 --no-ext-diff
350 Disallow external diff drivers.
351 --ignore-submodules[=<when>]
353 Ignore changes to submodules in the diff generation. <when> can be
354 either "none", "untracked", "dirty" or "all", which is the default
355 Using "none" will consider the submodule modified when it either
356 contains untracked or modified files or its HEAD differs from the
357 commit recorded in the superproject and can be used to override any
358 settings of the ignore option in git-config(1) or gitmodules(5).
359 When "untracked" is used submodules are not considered dirty when
360 they only contain untracked content (but they are still scanned for
361 modified content). Using "dirty" ignores all changes to the work
362 tree of submodules, only changes to the commits stored in the
363 superproject are shown (this was the behavior until 1.7.0). Using
364 "all" hides all changes to submodules.
365 --src-prefix=<prefix>
367 Show the given source prefix instead of "a/".
368 --dst-prefix=<prefix>
370 Show the given destination prefix instead of "b/".
371 --no-prefix
373 Do not show any source or destination prefix.
374 For more detailed explanation on these common options, see also
376 gitdiffcore(7).
377 <path>...
379 The <paths> parameters, when given, are used to limit the diff to
380 the named paths (you can give directory names and get diff for all
381 files under them).
382
384 The raw output format from "git-diff-index", "git-diff-tree",
385 "git-diff-files" and "git diff --raw" are very similar.
386 These commands all compare two sets of things; what is compared
388 differs:
389 git-diff-index <tree-ish>
391 compares the <tree-ish> and the files on the filesystem.
392 git-diff-index --cached <tree-ish>
394 compares the <tree-ish> and the index.
395 git-diff-tree [-r] <tree-ish-1> <tree-ish-2> [<pattern>...]
397 compares the trees named by the two arguments.
398 git-diff-files [<pattern>...]
400 compares the index and the files on the filesystem.
401 The "git-diff-tree" command begins its output by printing the hash of
403 what is being compared. After that, all the commands print one output
404 line per changed file.
405 An output line is formatted this way:
407 in-place edit :100644 100644 bcd1234... 0123456... M file0
409 copy-edit :100644 100644 abcd123... 1234567... C68 file1 file2
410 rename-edit :100644 100644 abcd123... 1234567... R86 file1 file3
411 create :000000 100644 0000000... 1234567... A file4
412 delete :100644 000000 1234567... 0000000... D file5
413 unmerged :000000 000000 0000000... 0000000... U file6
414 That is, from the left to the right:
417 1. a colon.
419 2. mode for "src"; 000000 if creation or unmerged.
421 3. a space.
423 4. mode for "dst"; 000000 if deletion or unmerged.
425 5. a space.
427 6. sha1 for "src"; 0{40} if creation or unmerged.
429 7. a space.
431 8. sha1 for "dst"; 0{40} if creation, unmerged or "look at work tree".
433 9. a space.
435 10. status, followed by optional "score" number.
437 11. a tab or a NUL when -z option is used.
439 12. path for "src"
441 13. a tab or a NUL when -z option is used; only exists for C or R.
443 14. path for "dst"; only exists for C or R.
445 15. an LF or a NUL when -z option is used, to terminate the record.
447 Possible status letters are:
449 · A: addition of a file
451 · C: copy of a file into a new one
453 · D: deletion of a file
455 · M: modification of the contents or mode of a file
457 · R: renaming of a file
459 · T: change in the type of the file
461 · U: file is unmerged (you must complete the merge before it can be
463 committed)
464 · X: "unknown" change type (most probably a bug, please report it)
466 Status letters C and R are always followed by a score (denoting the
468 percentage of similarity between the source and target of the move or
469 copy), and are the only ones to be so.
470 <sha1> is shown as all 0’s if a file is new on the filesystem and it is
472 out of sync with the index.
473 Example:
475 :100644 100644 5be4a4...... 000000...... M file.c
477 When -z option is not used, TAB, LF, and backslash characters in
480 pathnames are represented as \t, \n, and \\, respectively.
481
483 "git-diff-tree", "git-diff-files" and "git-diff --raw" can take -c or
484 --cc option to generate diff output also for merge commits. The output
485 differs from the format described above in the following way:
486 1. there is a colon for each parent
488 2. there are more "src" modes and "src" sha1
490 3. status is concatenated status characters for each parent
492 4. no optional "score" number
494 5. single path, only for "dst"
496 Example:
498 ::100644 100644 100644 fabadb8... cc95eb0... 4866510... MM describe.c
500 Note that combined diff lists only files which were modified from all
503 parents.
504
506 When "git-diff-index", "git-diff-tree", or "git-diff-files" are run
507 with a -p option, "git diff" without the --raw option, or "git log"
508 with the "-p" option, they do not produce the output described above;
509 instead they produce a patch file. You can customize the creation of
510 such patches via the GIT_EXTERNAL_DIFF and the GIT_DIFF_OPTS
511 environment variables.
512 What the -p option produces is slightly different from the traditional
514 diff format:
515 1. It is preceded with a "git diff" header that looks like this:
517 diff --git a/file1 b/file2
519 The a/ and b/ filenames are the same unless rename/copy is
521 involved. Especially, even for a creation or a deletion, /dev/null
522 is not used in place of the a/ or b/ filenames.
523 When rename/copy is involved, file1 and file2 show the name of the
525 source file of the rename/copy and the name of the file that
526 rename/copy produces, respectively.
527 2. It is followed by one or more extended header lines:
529 old mode <mode>
531 new mode <mode>
532 deleted file mode <mode>
533 new file mode <mode>
534 copy from <path>
535 copy to <path>
536 rename from <path>
537 rename to <path>
538 similarity index <number>
539 dissimilarity index <number>
540 index <hash>..<hash> <mode>
541 File modes are printed as 6-digit octal numbers including the file
543 type and file permission bits.
544 Path names in extended headers do not include the a/ and b/
546 prefixes.
547 The similarity index is the percentage of unchanged lines, and the
549 dissimilarity index is the percentage of changed lines. It is a
550 rounded down integer, followed by a percent sign. The similarity
551 index value of 100% is thus reserved for two equal files, while
552 100% dissimilarity means that no line from the old file made it
553 into the new one.
554 The index line includes the SHA-1 checksum before and after the
556 change. The <mode> is included if the file mode does not change;
557 otherwise, separate lines indicate the old and the new mode.
558 3. TAB, LF, double quote and backslash characters in pathnames are
560 represented as \t, \n, \" and \\, respectively. If there is need
561 for such substitution then the whole pathname is put in double
562 quotes.
563 4. All the file1 files in the output refer to files before the commit,
565 and all the file2 files refer to files after the commit. It is
566 incorrect to apply each change to each file sequentially. For
567 example, this patch will swap a and b:
568 diff --git a/a b/b
570 rename from a
571 rename to b
572 diff --git a/b b/a
573 rename from b
574 rename to a
575
577 Any diff-generating command can take the ‘-c` or --cc option to produce
578 a combined diff when showing a merge. This is the default format when
579 showing merges with git-diff(1) or git-show(1). Note also that you can
580 give the `-m’ option to any of these commands to force generation of
581 diffs with individual parents of a merge.
582 A combined diff format looks like this:
584 diff --combined describe.c
586 index fabadb8,cc95eb0..4866510
587 --- a/describe.c
588 +++ b/describe.c
589 @@@ -98,20 -98,12 +98,20 @@@
590 return (a_date > b_date) ? -1 : (a_date == b_date) ? 0 : 1;
591 }
592 - static void describe(char *arg)
594 -static void describe(struct commit *cmit, int last_one)
595 ++static void describe(char *arg, int last_one)
596 {
597 + unsigned char sha1[20];
598 + struct commit *cmit;
599 struct commit_list *list;
600 static int initialized = 0;
601 struct commit_name *n;
602 + if (get_sha1(arg, sha1) < 0)
604 + usage(describe_usage);
605 + cmit = lookup_commit_reference(sha1);
606 + if (!cmit)
607 + usage(describe_usage);
608 +
609 if (!initialized) {
610 initialized = 1;
611 for_each_ref(get_name);
612 1. It is preceded with a "git diff" header, that looks like this (when
616 -c option is used):
617 diff --combined file
619 or like this (when --cc option is used):
621 diff --cc file
623 2. It is followed by one or more extended header lines (this example
625 shows a merge with two parents):
626 index <hash>,<hash>..<hash>
628 mode <mode>,<mode>..<mode>
629 new file mode <mode>
630 deleted file mode <mode>,<mode>
631 The mode <mode>,<mode>..<mode> line appears only if at least one of
633 the <mode> is different from the rest. Extended headers with
634 information about detected contents movement (renames and copying
635 detection) are designed to work with diff of two <tree-ish> and are
636 not used by combined diff format.
637 3. It is followed by two-line from-file/to-file header
639 --- a/file
641 +++ b/file
642 Similar to two-line header for traditional unified diff format,
644 /dev/null is used to signal created or deleted files.
645 4. Chunk header format is modified to prevent people from accidentally
647 feeding it to patch -p1. Combined diff format was created for
648 review of merge commit changes, and was not meant for apply. The
649 change is similar to the change in the extended index header:
650 @@@ <from-file-range> <from-file-range> <to-file-range> @@@
652 There are (number of parents + 1) @ characters in the chunk header
654 for combined diff format.
655 Unlike the traditional unified diff format, which shows two files A and
657 B with a single column that has - (minus — appears in A but removed in
658 B), + (plus — missing in A but added to B), or " " (space — unchanged)
659 prefix, this format compares two or more files file1, file2,... with
660 one file X, and shows how X differs from each of fileN. One column for
661 each of fileN is prepended to the output line to note how X’s line is
662 different from it.
663 A - character in the column N means that the line appears in fileN but
665 it does not appear in the result. A + character in the column N means
666 that the line appears in the result, and fileN does not have that line
667 (in other words, the line was added, from the point of view of that
668 parent).
669 In the above example output, the function signature was changed from
671 both files (hence two - removals from both file1 and file2, plus ++ to
672 mean one line that was added does not appear in either file1 nor
673 file2). Also eight other lines are the same from file1 but do not
674 appear in file2 (hence prefixed with +).
675 When shown by git diff-tree -c, it compares the parents of a merge
677 commit with the merge result (i.e. file1..fileN are the parents). When
678 shown by git diff-files -c, it compares the two unresolved merge
679 parents with the working tree file (i.e. file1 is stage 2 aka "our
680 version", file2 is stage 3 aka "their version").
681
683 The --summary option describes newly added, deleted, renamed and copied
684 files. The --stat option adds diffstat(1) graph to the output. These
685 options can be combined with other options, such as -p, and are meant
686 for human consumption.
687 When showing a change that involves a rename or a copy, --stat output
689 formats the pathnames compactly by combining common prefix and suffix
690 of the pathnames. For example, a change that moves arch/i386/Makefile
691 to arch/x86/Makefile while modifying 4 lines will be shown like this:
692 arch/{i386 => x86}/Makefile | 4 +--
694 The --numstat option gives the diffstat(1) information but is designed
697 for easier machine consumption. An entry in --numstat output looks like
698 this:
699 1 2 README
701 3 1 arch/{i386 => x86}/Makefile
702 That is, from left to right:
705 1. the number of added lines;
707 2. a tab;
709 3. the number of deleted lines;
711 4. a tab;
713 5. pathname (possibly with rename/copy information);
715 6. a newline.
717 When -z output option is in effect, the output is formatted this way:
719 1 2 README NUL
721 3 1 NUL arch/i386/Makefile NUL arch/x86/Makefile NUL
722 That is:
725 1. the number of added lines;
727 2. a tab;
729 3. the number of deleted lines;
731 4. a tab;
733 5. a NUL (only exists if renamed/copied);
735 6. pathname in preimage;
737 7. a NUL (only exists if renamed/copied);
739 8. pathname in postimage (only exists if renamed/copied);
741 9. a NUL.
743 The extra NUL before the preimage path in renamed case is to allow
745 scripts that read the output to tell if the current record being read
746 is a single-path record or a rename/copy record without reading ahead.
747 After reading added and deleted lines, reading up to NUL would yield
748 the pathname, but if that is NUL, the record will show two paths.
749
751 Various ways to check your working tree
752 $ git diff [1m(1)
754 $ git diff --cached [1m(2)
755 $ git diff HEAD [1m(3)
756 1. Changes in the working tree not yet staged for the next commit.
758 2. Changes between the index and your last commit; what you would
759 be committing if you run "git commit" without "-a" option.
760 3. Changes in the working tree since your last commit; what you
761 would be committing if you run "git commit -a"
762 Comparing with arbitrary commits
764 $ git diff test [1m(1)
766 $ git diff HEAD -- ./test [1m(2)
767 $ git diff HEAD^ HEAD [1m(3)
768 1. Instead of using the tip of the current branch, compare with the
770 tip of "test" branch.
771 2. Instead of comparing with the tip of "test" branch, compare with
772 the tip of the current branch, but limit the comparison to the file
773 "test".
774 3. Compare the version before the last commit and the last commit.
775 Comparing branches
777 $ git diff topic master [1m(1)
779 $ git diff topic..master [1m(2)
780 $ git diff topic...master [1m(3)
781 1. Changes between the tips of the topic and the master branches.
783 2. Same as above.
784 3. Changes that occurred on the master branch since when the topic
785 branch was started off it.
786 Limiting the diff output
788 $ git diff --diff-filter=MRC [1m(1)
790 $ git diff --name-status [1m(2)
791 $ git diff arch/i386 include/asm-i386 [1m(3)
792 1. Show only modification, rename and copy, but not addition nor
794 deletion.
795 2. Show only names and the nature of change, but not actual diff
796 output.
797 3. Limit diff output to named subtrees.
798 Munging the diff output
800 $ git diff --find-copies-harder -B -C [1m(1)
802 $ git diff -R [1m(2)
803 1. Spend extra cycles to find renames, copies and complete rewrites
805 (very expensive).
806 2. Output diff in reverse.
807
809 diff(1), git-difftool(1), git-log(1), gitdiffcore(7), git-format-
810 patch(1), git-apply(1)
811
813 Written by Linus Torvalds <torvalds@osdl.org[1]>
814
816 Documentation by Junio C Hamano and the git-list
817 <git@vger.kernel.org[2]>.
818
820 Part of the git(1) suite
821
823 1. torvalds@osdl.org
824 mailto:torvalds@osdl.org
825 2. git@vger.kernel.org
827 mailto:git@vger.kernel.org
828Git 1.7.4.4 04/11/2011 GIT-DIFF(1)