1GIT-LOG(1) Git Manual GIT-LOG(1)
2
6 git-log - Show commit logs
7
9 git log [<options>] [<revision range>] [[--] <path>...]
10
13 Shows the commit logs.
14 The command takes options applicable to the git rev-list command to
16 control what is shown and how, and options applicable to the git diff-*
17 commands to control how the changes each commit introduces are shown.
18
20 --follow
21 Continue listing the history of a file beyond renames (works only
22 for a single file).
23 --no-decorate, --decorate[=short|full|auto|no]
25 Print out the ref names of any commits that are shown. If short is
26 specified, the ref name prefixes refs/heads/, refs/tags/ and
27 refs/remotes/ will not be printed. If full is specified, the full
28 ref name (including prefix) will be printed. If auto is specified,
29 then if the output is going to a terminal, the ref names are shown
30 as if short were given, otherwise no ref names are shown. The
31 default option is short.
32 --decorate-refs=<pattern>, --decorate-refs-exclude=<pattern>
34 If no --decorate-refs is given, pretend as if all refs were
35 included. For each candidate, do not use it for decoration if it
36 matches any patterns given to --decorate-refs-exclude or if it
37 doesn’t match any of the patterns given to --decorate-refs.
38 --source
40 Print out the ref name given on the command line by which each
41 commit was reached.
42 --use-mailmap
44 Use mailmap file to map author and committer names and email
45 addresses to canonical real names and email addresses. See git-
46 shortlog(1).
47 --full-diff
49 Without this flag, git log -p <path>... shows commits that touch
50 the specified paths, and diffs about the same specified paths. With
51 this, the full diff is shown for commits that touch the specified
52 paths; this means that "<path>..." limits only commits, and doesn’t
53 limit diff for those commits.
54 Note that this affects all diff-based output types, e.g. those
56 produced by --stat, etc.
57 --log-size
59 Include a line “log size <number>” in the output for each commit,
60 where <number> is the length of that commit’s message in bytes.
61 Intended to speed up tools that read log messages from git log
62 output by allowing them to allocate space in advance.
63 -L <start>,<end>:<file>, -L :<funcname>:<file>
65 Trace the evolution of the line range given by "<start>,<end>" (or
66 the function name regex <funcname>) within the <file>. You may not
67 give any pathspec limiters. This is currently limited to a walk
68 starting from a single revision, i.e., you may only give zero or
69 one positive revision arguments. You can specify this option more
70 than once.
71 <start> and <end> can take one of these forms:
73 · number
75 If <start> or <end> is a number, it specifies an absolute line
77 number (lines count from 1).
78 · /regex/
80 This form will use the first line matching the given POSIX
82 regex. If <start> is a regex, it will search from the end of
83 the previous -L range, if any, otherwise from the start of
84 file. If <start> is “^/regex/”, it will search from the start
85 of file. If <end> is a regex, it will search starting at the
86 line given by <start>.
87 · +offset or -offset
89 This is only valid for <end> and will specify a number of lines
91 before or after the line given by <start>.
92 If “:<funcname>” is given in place of <start> and <end>, it is a
94 regular expression that denotes the range from the first funcname
95 line that matches <funcname>, up to the next funcname line.
96 “:<funcname>” searches from the end of the previous -L range, if
97 any, otherwise from the start of file. “^:<funcname>” searches from
98 the start of file.
99 <revision range>
101 Show only commits in the specified revision range. When no
102 <revision range> is specified, it defaults to HEAD (i.e. the whole
103 history leading to the current commit). origin..HEAD specifies all
104 the commits reachable from the current commit (i.e. HEAD), but not
105 from origin. For a complete list of ways to spell <revision range>,
106 see the Specifying Ranges section of gitrevisions(7).
107 [--] <path>...
109 Show only commits that are enough to explain how the files that
110 match the specified paths came to be. See History Simplification
111 below for details and other simplification modes.
112 Paths may need to be prefixed with -- to separate them from options
114 or the revision range, when confusion arises.
115 Commit Limiting
117 Besides specifying a range of commits that should be listed using the
118 special notations explained in the description, additional commit
119 limiting may be applied.
120 Using more options generally further limits the output (e.g.
122 --since=<date1> limits to commits newer than <date1>, and using it with
123 --grep=<pattern> further limits to commits whose log message has a line
124 that matches <pattern>), unless otherwise noted.
125 Note that these are applied before commit ordering and formatting
127 options, such as --reverse.
128 -<number>, -n <number>, --max-count=<number>
130 Limit the number of commits to output.
131 --skip=<number>
133 Skip number commits before starting to show the commit output.
134 --since=<date>, --after=<date>
136 Show commits more recent than a specific date.
137 --until=<date>, --before=<date>
139 Show commits older than a specific date.
140 --author=<pattern>, --committer=<pattern>
142 Limit the commits output to ones with author/committer header lines
143 that match the specified pattern (regular expression). With more
144 than one --author=<pattern>, commits whose author matches any of
145 the given patterns are chosen (similarly for multiple
146 --committer=<pattern>).
147 --grep-reflog=<pattern>
149 Limit the commits output to ones with reflog entries that match the
150 specified pattern (regular expression). With more than one
151 --grep-reflog, commits whose reflog message matches any of the
152 given patterns are chosen. It is an error to use this option unless
153 --walk-reflogs is in use.
154 --grep=<pattern>
156 Limit the commits output to ones with log message that matches the
157 specified pattern (regular expression). With more than one
158 --grep=<pattern>, commits whose message matches any of the given
159 patterns are chosen (but see --all-match).
160 When --show-notes is in effect, the message from the notes is
162 matched as if it were part of the log message.
163 --all-match
165 Limit the commits output to ones that match all given --grep,
166 instead of ones that match at least one.
167 --invert-grep
169 Limit the commits output to ones with log message that do not match
170 the pattern specified with --grep=<pattern>.
171 -i, --regexp-ignore-case
173 Match the regular expression limiting patterns without regard to
174 letter case.
175 --basic-regexp
177 Consider the limiting patterns to be basic regular expressions;
178 this is the default.
179 -E, --extended-regexp
181 Consider the limiting patterns to be extended regular expressions
182 instead of the default basic regular expressions.
183 -F, --fixed-strings
185 Consider the limiting patterns to be fixed strings (don’t interpret
186 pattern as a regular expression).
187 -P, --perl-regexp
189 Consider the limiting patterns to be Perl-compatible regular
190 expressions.
191 Support for these types of regular expressions is an optional
193 compile-time dependency. If Git wasn’t compiled with support for
194 them providing this option will cause it to die.
195 --remove-empty
197 Stop when a given path disappears from the tree.
198 --merges
200 Print only merge commits. This is exactly the same as
201 --min-parents=2.
202 --no-merges
204 Do not print commits with more than one parent. This is exactly the
205 same as --max-parents=1.
206 --min-parents=<number>, --max-parents=<number>, --no-min-parents,
208 --no-max-parents
209 Show only commits which have at least (or at most) that many parent
210 commits. In particular, --max-parents=1 is the same as --no-merges,
211 --min-parents=2 is the same as --merges. --max-parents=0 gives all
212 root commits and --min-parents=3 all octopus merges.
213 --no-min-parents and --no-max-parents reset these limits (to no
215 limit) again. Equivalent forms are --min-parents=0 (any commit has
216 0 or more parents) and --max-parents=-1 (negative numbers denote no
217 upper limit).
218 --first-parent
220 Follow only the first parent commit upon seeing a merge commit.
221 This option can give a better overview when viewing the evolution
222 of a particular topic branch, because merges into a topic branch
223 tend to be only about adjusting to updated upstream from time to
224 time, and this option allows you to ignore the individual commits
225 brought in to your history by such a merge. Cannot be combined with
226 --bisect.
227 --not
229 Reverses the meaning of the ^ prefix (or lack thereof) for all
230 following revision specifiers, up to the next --not.
231 --all
233 Pretend as if all the refs in refs/, along with HEAD, are listed on
234 the command line as <commit>.
235 --branches[=<pattern>]
237 Pretend as if all the refs in refs/heads are listed on the command
238 line as <commit>. If <pattern> is given, limit branches to ones
239 matching given shell glob. If pattern lacks ?, *, or [, /* at the
240 end is implied.
241 --tags[=<pattern>]
243 Pretend as if all the refs in refs/tags are listed on the command
244 line as <commit>. If <pattern> is given, limit tags to ones
245 matching given shell glob. If pattern lacks ?, *, or [, /* at the
246 end is implied.
247 --remotes[=<pattern>]
249 Pretend as if all the refs in refs/remotes are listed on the
250 command line as <commit>. If <pattern> is given, limit
251 remote-tracking branches to ones matching given shell glob. If
252 pattern lacks ?, *, or [, /* at the end is implied.
253 --glob=<glob-pattern>
255 Pretend as if all the refs matching shell glob <glob-pattern> are
256 listed on the command line as <commit>. Leading refs/, is
257 automatically prepended if missing. If pattern lacks ?, *, or [, /*
258 at the end is implied.
259 --exclude=<glob-pattern>
261 Do not include refs matching <glob-pattern> that the next --all,
262 --branches, --tags, --remotes, or --glob would otherwise consider.
263 Repetitions of this option accumulate exclusion patterns up to the
264 next --all, --branches, --tags, --remotes, or --glob option (other
265 options or arguments do not clear accumulated patterns).
266 The patterns given should not begin with refs/heads, refs/tags, or
268 refs/remotes when applied to --branches, --tags, or --remotes,
269 respectively, and they must begin with refs/ when applied to --glob
270 or --all. If a trailing /* is intended, it must be given
271 explicitly.
272 --reflog
274 Pretend as if all objects mentioned by reflogs are listed on the
275 command line as <commit>.
276 --single-worktree
278 By default, all working trees will be examined by the following
279 options when there are more than one (see git-worktree(1)): --all,
280 --reflog and --indexed-objects. This option forces them to examine
281 the current working tree only.
282 --ignore-missing
284 Upon seeing an invalid object name in the input, pretend as if the
285 bad input was not given.
286 --bisect
288 Pretend as if the bad bisection ref refs/bisect/bad was listed and
289 as if it was followed by --not and the good bisection refs
290 refs/bisect/good-* on the command line. Cannot be combined with
291 --first-parent.
292 --stdin
294 In addition to the <commit> listed on the command line, read them
295 from the standard input. If a -- separator is seen, stop reading
296 commits and start reading paths to limit the result.
297 --cherry-mark
299 Like --cherry-pick (see below) but mark equivalent commits with =
300 rather than omitting them, and inequivalent ones with +.
301 --cherry-pick
303 Omit any commit that introduces the same change as another commit
304 on the “other side” when the set of commits are limited with
305 symmetric difference.
306 For example, if you have two branches, A and B, a usual way to list
308 all commits on only one side of them is with --left-right (see the
309 example below in the description of the --left-right option).
310 However, it shows the commits that were cherry-picked from the
311 other branch (for example, “3rd on b” may be cherry-picked from
312 branch A). With this option, such pairs of commits are excluded
313 from the output.
314 --left-only, --right-only
316 List only commits on the respective side of a symmetric difference,
317 i.e. only those which would be marked < resp. > by --left-right.
318 For example, --cherry-pick --right-only A...B omits those commits
320 from B which are in A or are patch-equivalent to a commit in A. In
321 other words, this lists the + commits from git cherry A B. More
322 precisely, --cherry-pick --right-only --no-merges gives the exact
323 list.
324 --cherry
326 A synonym for --right-only --cherry-mark --no-merges; useful to
327 limit the output to the commits on our side and mark those that
328 have been applied to the other side of a forked history with git
329 log --cherry upstream...mybranch, similar to git cherry upstream
330 mybranch.
331 -g, --walk-reflogs
333 Instead of walking the commit ancestry chain, walk reflog entries
334 from the most recent one to older ones. When this option is used
335 you cannot specify commits to exclude (that is, ^commit,
336 commit1..commit2, and commit1...commit2 notations cannot be used).
337 With --pretty format other than oneline (for obvious reasons), this
339 causes the output to have two extra lines of information taken from
340 the reflog. The reflog designator in the output may be shown as
341 ref@{Nth} (where Nth is the reverse-chronological index in the
342 reflog) or as ref@{timestamp} (with the timestamp for that entry),
343 depending on a few rules:
344 1. If the starting point is specified as ref@{Nth}, show the index
346 format.
347 2. If the starting point was specified as ref@{now}, show the
349 timestamp format.
350 3. If neither was used, but --date was given on the command line,
352 show the timestamp in the format requested by --date.
353 4. Otherwise, show the index format.
355 Under --pretty=oneline, the commit message is prefixed with this
357 information on the same line. This option cannot be combined with
358 --reverse. See also git-reflog(1).
359 --merge
361 After a failed merge, show refs that touch files having a conflict
362 and don’t exist on all heads to merge.
363 --boundary
365 Output excluded boundary commits. Boundary commits are prefixed
366 with -.
367 History Simplification
369 Sometimes you are only interested in parts of the history, for example
370 the commits modifying a particular <path>. But there are two parts of
371 History Simplification, one part is selecting the commits and the other
372 is how to do it, as there are various strategies to simplify the
373 history.
374 The following options select the commits to be shown:
376 <paths>
378 Commits modifying the given <paths> are selected.
379 --simplify-by-decoration
381 Commits that are referred by some branch or tag are selected.
382 Note that extra commits can be shown to give a meaningful history.
384 The following options affect the way the simplification is performed:
386 Default mode
388 Simplifies the history to the simplest history explaining the final
389 state of the tree. Simplest because it prunes some side branches if
390 the end result is the same (i.e. merging branches with the same
391 content)
392 --full-history
394 Same as the default mode, but does not prune some history.
395 --dense
397 Only the selected commits are shown, plus some to have a meaningful
398 history.
399 --sparse
401 All commits in the simplified history are shown.
402 --simplify-merges
404 Additional option to --full-history to remove some needless merges
405 from the resulting history, as there are no selected commits
406 contributing to this merge.
407 --ancestry-path
409 When given a range of commits to display (e.g. commit1..commit2 or
410 commit2 ^commit1), only display commits that exist directly on the
411 ancestry chain between the commit1 and commit2, i.e. commits that
412 are both descendants of commit1, and ancestors of commit2.
413 A more detailed explanation follows.
415 Suppose you specified foo as the <paths>. We shall call commits that
417 modify foo !TREESAME, and the rest TREESAME. (In a diff filtered for
418 foo, they look different and equal, respectively.)
419 In the following, we will always refer to the same example history to
421 illustrate the differences between simplification settings. We assume
422 that you are filtering for a file foo in this commit graph:
423 .-A---M---N---O---P---Q
425 / / / / / /
426 I B C D E Y
427 \ / / / / /
428 `-------------' X
429 The horizontal line of history A---Q is taken to be the first parent of
432 each merge. The commits are:
433 · I is the initial commit, in which foo exists with contents “asdf”,
435 and a file quux exists with contents “quux”. Initial commits are
436 compared to an empty tree, so I is !TREESAME.
437 · In A, foo contains just “foo”.
439 · B contains the same change as A. Its merge M is trivial and hence
441 TREESAME to all parents.
442 · C does not change foo, but its merge N changes it to “foobar”, so
444 it is not TREESAME to any parent.
445 · D sets foo to “baz”. Its merge O combines the strings from N and D
447 to “foobarbaz”; i.e., it is not TREESAME to any parent.
448 · E changes quux to “xyzzy”, and its merge P combines the strings to
450 “quux xyzzy”. P is TREESAME to O, but not to E.
451 · X is an independent root commit that added a new file side, and Y
453 modified it. Y is TREESAME to X. Its merge Q added side to P, and
454 Q is TREESAME to P, but not to Y.
455 rev-list walks backwards through history, including or excluding
457 commits based on whether --full-history and/or parent rewriting (via
458 --parents or --children) are used. The following settings are
459 available.
460 Default mode
462 Commits are included if they are not TREESAME to any parent (though
463 this can be changed, see --sparse below). If the commit was a
464 merge, and it was TREESAME to one parent, follow only that parent.
465 (Even if there are several TREESAME parents, follow only one of
466 them.) Otherwise, follow all parents.
467 This results in:
469 .-A---N---O
471 / / /
472 I---------D
473 Note how the rule to only follow the TREESAME parent, if one is
475 available, removed B from consideration entirely. C was considered
476 via N, but is TREESAME. Root commits are compared to an empty tree,
477 so I is !TREESAME.
478 Parent/child relations are only visible with --parents, but that
480 does not affect the commits selected in default mode, so we have
481 shown the parent lines.
482 --full-history without parent rewriting
484 This mode differs from the default in one point: always follow all
485 parents of a merge, even if it is TREESAME to one of them. Even if
486 more than one side of the merge has commits that are included, this
487 does not imply that the merge itself is! In the example, we get
488 I A B N D O P Q
490 M was excluded because it is TREESAME to both parents. E, C and B
492 were all walked, but only B was !TREESAME, so the others do not
493 appear.
494 Note that without parent rewriting, it is not really possible to
496 talk about the parent/child relationships between the commits, so
497 we show them disconnected.
498 --full-history with parent rewriting
500 Ordinary commits are only included if they are !TREESAME (though
501 this can be changed, see --sparse below).
502 Merges are always included. However, their parent list is
504 rewritten: Along each parent, prune away commits that are not
505 included themselves. This results in
506 .-A---M---N---O---P---Q
508 / / / / /
509 I B / D /
510 \ / / / /
511 `-------------'
512 Compare to --full-history without rewriting above. Note that E was
514 pruned away because it is TREESAME, but the parent list of P was
515 rewritten to contain E's parent I. The same happened for C and N,
516 and X, Y and Q.
517 In addition to the above settings, you can change whether TREESAME
519 affects inclusion:
520 --dense
522 Commits that are walked are included if they are not TREESAME to
523 any parent.
524 --sparse
526 All commits that are walked are included.
527 Note that without --full-history, this still simplifies merges: if
529 one of the parents is TREESAME, we follow only that one, so the
530 other sides of the merge are never walked.
531 --simplify-merges
533 First, build a history graph in the same way that --full-history
534 with parent rewriting does (see above).
535 Then simplify each commit C to its replacement C' in the final
537 history according to the following rules:
538 · Set C' to C.
540 · Replace each parent P of C' with its simplification P'. In the
542 process, drop parents that are ancestors of other parents or
543 that are root commits TREESAME to an empty tree, and remove
544 duplicates, but take care to never drop all parents that we are
545 TREESAME to.
546 · If after this parent rewriting, C' is a root or merge commit
548 (has zero or >1 parents), a boundary commit, or !TREESAME, it
549 remains. Otherwise, it is replaced with its only parent.
550 The effect of this is best shown by way of comparing to
552 --full-history with parent rewriting. The example turns into:
553 .-A---M---N---O
555 / / /
556 I B D
557 \ / /
558 `---------'
559 Note the major differences in N, P, and Q over --full-history:
561 · N's parent list had I removed, because it is an ancestor of the
563 other parent M. Still, N remained because it is !TREESAME.
564 · P's parent list similarly had I removed. P was then removed
566 completely, because it had one parent and is TREESAME.
567 · Q's parent list had Y simplified to X. X was then removed,
569 because it was a TREESAME root. Q was then removed completely,
570 because it had one parent and is TREESAME.
571 Finally, there is a fifth simplification mode available:
573 --ancestry-path
575 Limit the displayed commits to those directly on the ancestry chain
576 between the “from” and “to” commits in the given commit range. I.e.
577 only display commits that are ancestor of the “to” commit and
578 descendants of the “from” commit.
579 As an example use case, consider the following commit history:
581 D---E-------F
583 / \ \
584 B---C---G---H---I---J
585 / \
586 A-------K---------------L--M
587 A regular D..M computes the set of commits that are ancestors of M,
589 but excludes the ones that are ancestors of D. This is useful to
590 see what happened to the history leading to M since D, in the sense
591 that “what does M have that did not exist in D”. The result in this
592 example would be all the commits, except A and B (and D itself, of
593 course).
594 When we want to find out what commits in M are contaminated with
596 the bug introduced by D and need fixing, however, we might want to
597 view only the subset of D..M that are actually descendants of D,
598 i.e. excluding C and K. This is exactly what the --ancestry-path
599 option does. Applied to the D..M range, it results in:
600 E-------F
602 \ \
603 G---H---I---J
604 \
605 L--M
606 The --simplify-by-decoration option allows you to view only the big
609 picture of the topology of the history, by omitting commits that are
610 not referenced by tags. Commits are marked as !TREESAME (in other
611 words, kept after history simplification rules described above) if (1)
612 they are referenced by tags, or (2) they change the contents of the
613 paths given on the command line. All other commits are marked as
614 TREESAME (subject to be simplified away).
615 Commit Ordering
617 By default, the commits are shown in reverse chronological order.
618 --date-order
620 Show no parents before all of its children are shown, but otherwise
621 show commits in the commit timestamp order.
622 --author-date-order
624 Show no parents before all of its children are shown, but otherwise
625 show commits in the author timestamp order.
626 --topo-order
628 Show no parents before all of its children are shown, and avoid
629 showing commits on multiple lines of history intermixed.
630 For example, in a commit history like this:
632 ---1----2----4----7
634 \ \
635 3----5----6----8---
636 where the numbers denote the order of commit timestamps, git
638 rev-list and friends with --date-order show the commits in the
639 timestamp order: 8 7 6 5 4 3 2 1.
640 With --topo-order, they would show 8 6 5 3 7 4 2 1 (or 8 7 4 2 6 5
642 3 1); some older commits are shown before newer ones in order to
643 avoid showing the commits from two parallel development track mixed
644 together.
645 --reverse
647 Output the commits chosen to be shown (see Commit Limiting section
648 above) in reverse order. Cannot be combined with --walk-reflogs.
649 Object Traversal
651 These options are mostly targeted for packing of Git repositories.
652 --exclude-promisor-objects
654 (For internal use only.) Prefilter object traversal at promisor
655 boundary. This is used with partial clone. This is stronger than
656 --missing=allow-promisor because it limits the traversal, rather
657 than just silencing errors about missing objects.
658 --no-walk[=(sorted|unsorted)]
660 Only show the given commits, but do not traverse their ancestors.
661 This has no effect if a range is specified. If the argument
662 unsorted is given, the commits are shown in the order they were
663 given on the command line. Otherwise (if sorted or no argument was
664 given), the commits are shown in reverse chronological order by
665 commit time. Cannot be combined with --graph.
666 --do-walk
668 Overrides a previous --no-walk.
669 Commit Formatting
671 --pretty[=<format>], --format=<format>
672 Pretty-print the contents of the commit logs in a given format,
673 where <format> can be one of oneline, short, medium, full, fuller,
674 email, raw, format:<string> and tformat:<string>. When <format> is
675 none of the above, and has %placeholder in it, it acts as if
676 --pretty=tformat:<format> were given.
677 See the "PRETTY FORMATS" section for some additional details for
679 each format. When =<format> part is omitted, it defaults to medium.
680 Note: you can specify the default pretty format in the repository
682 configuration (see git-config(1)).
683 --abbrev-commit
685 Instead of showing the full 40-byte hexadecimal commit object name,
686 show only a partial prefix. Non default number of digits can be
687 specified with "--abbrev=<n>" (which also modifies diff output, if
688 it is displayed).
689 This should make "--pretty=oneline" a whole lot more readable for
691 people using 80-column terminals.
692 --no-abbrev-commit
694 Show the full 40-byte hexadecimal commit object name. This negates
695 --abbrev-commit and those options which imply it such as
696 "--oneline". It also overrides the log.abbrevCommit variable.
697 --oneline
699 This is a shorthand for "--pretty=oneline --abbrev-commit" used
700 together.
701 --encoding=<encoding>
703 The commit objects record the encoding used for the log message in
704 their encoding header; this option can be used to tell the command
705 to re-code the commit log message in the encoding preferred by the
706 user. For non plumbing commands this defaults to UTF-8. Note that
707 if an object claims to be encoded in X and we are outputting in X,
708 we will output the object verbatim; this means that invalid
709 sequences in the original commit may be copied to the output.
710 --expand-tabs=<n>, --expand-tabs, --no-expand-tabs
712 Perform a tab expansion (replace each tab with enough spaces to
713 fill to the next display column that is multiple of <n>) in the log
714 message before showing it in the output. --expand-tabs is a
715 short-hand for --expand-tabs=8, and --no-expand-tabs is a
716 short-hand for --expand-tabs=0, which disables tab expansion.
717 By default, tabs are expanded in pretty formats that indent the log
719 message by 4 spaces (i.e. medium, which is the default, full, and
720 fuller).
721 --notes[=<treeish>]
723 Show the notes (see git-notes(1)) that annotate the commit, when
724 showing the commit log message. This is the default for git log,
725 git show and git whatchanged commands when there is no --pretty,
726 --format, or --oneline option given on the command line.
727 By default, the notes shown are from the notes refs listed in the
729 core.notesRef and notes.displayRef variables (or corresponding
730 environment overrides). See git-config(1) for more details.
731 With an optional <treeish> argument, use the treeish to find the
733 notes to display. The treeish can specify the full refname when it
734 begins with refs/notes/; when it begins with notes/, refs/ and
735 otherwise refs/notes/ is prefixed to form a full name of the ref.
736 Multiple --notes options can be combined to control which notes are
738 being displayed. Examples: "--notes=foo" will show only notes from
739 "refs/notes/foo"; "--notes=foo --notes" will show both notes from
740 "refs/notes/foo" and from the default notes ref(s).
741 --no-notes
743 Do not show notes. This negates the above --notes option, by
744 resetting the list of notes refs from which notes are shown.
745 Options are parsed in the order given on the command line, so e.g.
746 "--notes --notes=foo --no-notes --notes=bar" will only show notes
747 from "refs/notes/bar".
748 --show-notes[=<treeish>], --[no-]standard-notes
750 These options are deprecated. Use the above --notes/--no-notes
751 options instead.
752 --show-signature
754 Check the validity of a signed commit object by passing the
755 signature to gpg --verify and show the output.
756 --relative-date
758 Synonym for --date=relative.
759 --date=<format>
761 Only takes effect for dates shown in human-readable format, such as
762 when using --pretty. log.date config variable sets a default value
763 for the log command’s --date option. By default, dates are shown in
764 the original time zone (either committer’s or author’s). If -local
765 is appended to the format (e.g., iso-local), the user’s local time
766 zone is used instead.
767 --date=relative shows dates relative to the current time, e.g. “2
769 hours ago”. The -local option has no effect for --date=relative.
770 --date=local is an alias for --date=default-local.
772 --date=iso (or --date=iso8601) shows timestamps in a ISO 8601-like
774 format. The differences to the strict ISO 8601 format are:
775 · a space instead of the T date/time delimiter
777 · a space between time and time zone
779 · no colon between hours and minutes of the time zone
781 --date=iso-strict (or --date=iso8601-strict) shows timestamps in
783 strict ISO 8601 format.
784 --date=rfc (or --date=rfc2822) shows timestamps in RFC 2822 format,
786 often found in email messages.
787 --date=short shows only the date, but not the time, in YYYY-MM-DD
789 format.
790 --date=raw shows the date as seconds since the epoch (1970-01-01
792 00:00:00 UTC), followed by a space, and then the timezone as an
793 offset from UTC (a + or - with four digits; the first two are
794 hours, and the second two are minutes). I.e., as if the timestamp
795 were formatted with strftime("%s %z")). Note that the -local option
796 does not affect the seconds-since-epoch value (which is always
797 measured in UTC), but does switch the accompanying timezone value.
798 --date=unix shows the date as a Unix epoch timestamp (seconds since
800 1970). As with --raw, this is always in UTC and therefore -local
801 has no effect.
802 --date=format:... feeds the format ... to your system strftime,
804 except for %z and %Z, which are handled internally. Use
805 --date=format:%c to show the date in your system locale’s preferred
806 format. See the strftime manual for a complete list of format
807 placeholders. When using -local, the correct syntax is
808 --date=format-local:....
809 --date=default is the default format, and is similar to
811 --date=rfc2822, with a few exceptions:
812 · there is no comma after the day-of-week
814 · the time zone is omitted when the local time zone is used
816 --parents
818 Print also the parents of the commit (in the form "commit
819 parent..."). Also enables parent rewriting, see History
820 Simplification above.
821 --children
823 Print also the children of the commit (in the form "commit
824 child..."). Also enables parent rewriting, see History
825 Simplification above.
826 --left-right
828 Mark which side of a symmetric difference a commit is reachable
829 from. Commits from the left side are prefixed with < and those from
830 the right with >. If combined with --boundary, those commits are
831 prefixed with -.
832 For example, if you have this topology:
834 y---b---b branch B
836 / \ /
837 / .
838 / / \
839 o---x---a---a branch A
840 you would get an output like this:
842 $ git rev-list --left-right --boundary --pretty=oneline A...B
844 >bbbbbbb... 3rd on b
846 >bbbbbbb... 2nd on b
847 <aaaaaaa... 3rd on a
848 <aaaaaaa... 2nd on a
849 -yyyyyyy... 1st on b
850 -xxxxxxx... 1st on a
851 --graph
854 Draw a text-based graphical representation of the commit history on
855 the left hand side of the output. This may cause extra lines to be
856 printed in between commits, in order for the graph history to be
857 drawn properly. Cannot be combined with --no-walk.
858 This enables parent rewriting, see History Simplification above.
860 This implies the --topo-order option by default, but the
862 --date-order option may also be specified.
863 --show-linear-break[=<barrier>]
865 When --graph is not used, all history branches are flattened which
866 can make it hard to see that the two consecutive commits do not
867 belong to a linear branch. This option puts a barrier in between
868 them in that case. If <barrier> is specified, it is the string that
869 will be shown instead of the default one.
870 Diff Formatting
872 Listed below are options that control the formatting of diff output.
873 Some of them are specific to git-rev-list(1), however other diff
874 options may be given. See git-diff-files(1) for more options.
875 -c
877 With this option, diff output for a merge commit shows the
878 differences from each of the parents to the merge result
879 simultaneously instead of showing pairwise diff between a parent
880 and the result one at a time. Furthermore, it lists only files
881 which were modified from all parents.
882 --cc
884 This flag implies the -c option and further compresses the patch
885 output by omitting uninteresting hunks whose contents in the
886 parents have only two variants and the merge result picks one of
887 them without modification.
888 -m
890 This flag makes the merge commits show the full diff like regular
891 commits; for each merge parent, a separate log entry and diff is
892 generated. An exception is that only diff against the first parent
893 is shown when --first-parent option is given; in that case, the
894 output represents the changes the merge brought into the
895 then-current branch.
896 -r
898 Show recursive diffs.
899 -t
901 Show the tree objects in the diff output. This implies -r.
902
904 If the commit is a merge, and if the pretty-format is not oneline,
905 email or raw, an additional line is inserted before the Author: line.
906 This line begins with "Merge: " and the sha1s of ancestral commits are
907 printed, separated by spaces. Note that the listed commits may not
908 necessarily be the list of the direct parent commits if you have
909 limited your view of history: for example, if you are only interested
910 in changes related to a certain directory or file.
911 There are several built-in formats, and you can define additional
913 formats by setting a pretty.<name> config option to either another
914 format name, or a format: string, as described below (see git-
915 config(1)). Here are the details of the built-in formats:
916 · oneline
918 <sha1> <title line>
920 This is designed to be as compact as possible.
922 · short
924 commit <sha1>
926 Author: <author>
927 <title line>
929 · medium
931 commit <sha1>
933 Author: <author>
934 Date: <author date>
935 <title line>
937 <full commit message>
939 · full
941 commit <sha1>
943 Author: <author>
944 Commit: <committer>
945 <title line>
947 <full commit message>
949 · fuller
951 commit <sha1>
953 Author: <author>
954 AuthorDate: <author date>
955 Commit: <committer>
956 CommitDate: <committer date>
957 <title line>
959 <full commit message>
961 · email
963 From <sha1> <date>
965 From: <author>
966 Date: <author date>
967 Subject: [PATCH] <title line>
968 <full commit message>
970 · raw
972 The raw format shows the entire commit exactly as stored in the
974 commit object. Notably, the SHA-1s are displayed in full,
975 regardless of whether --abbrev or --no-abbrev are used, and parents
976 information show the true parent commits, without taking grafts or
977 history simplification into account. Note that this format affects
978 the way commits are displayed, but not the way the diff is shown
979 e.g. with git log --raw. To get full object names in a raw diff
980 format, use --no-abbrev.
981 · format:<string>
983 The format:<string> format allows you to specify which information
985 you want to show. It works a little bit like printf format, with
986 the notable exception that you get a newline with %n instead of \n.
987 E.g, format:"The author of %h was %an, %ar%nThe title was >>%s<<%n"
989 would show something like this:
990 The author of fe6e0ee was Junio C Hamano, 23 hours ago
992 The title was >>t4119: test autocomputing -p<n> for traditional diff input.<<
993 The placeholders are:
995 · %H: commit hash
997 · %h: abbreviated commit hash
999 · %T: tree hash
1001 · %t: abbreviated tree hash
1003 · %P: parent hashes
1005 · %p: abbreviated parent hashes
1007 · %an: author name
1009 · %aN: author name (respecting .mailmap, see git-shortlog(1) or
1011 git-blame(1))
1012 · %ae: author email
1014 · %aE: author email (respecting .mailmap, see git-shortlog(1) or
1016 git-blame(1))
1017 · %ad: author date (format respects --date= option)
1019 · %aD: author date, RFC2822 style
1021 · %ar: author date, relative
1023 · %at: author date, UNIX timestamp
1025 · %ai: author date, ISO 8601-like format
1027 · %aI: author date, strict ISO 8601 format
1029 · %cn: committer name
1031 · %cN: committer name (respecting .mailmap, see git-shortlog(1)
1033 or git-blame(1))
1034 · %ce: committer email
1036 · %cE: committer email (respecting .mailmap, see git-shortlog(1)
1038 or git-blame(1))
1039 · %cd: committer date (format respects --date= option)
1041 · %cD: committer date, RFC2822 style
1043 · %cr: committer date, relative
1045 · %ct: committer date, UNIX timestamp
1047 · %ci: committer date, ISO 8601-like format
1049 · %cI: committer date, strict ISO 8601 format
1051 · %d: ref names, like the --decorate option of git-log(1)
1053 · %D: ref names without the " (", ")" wrapping.
1055 · %e: encoding
1057 · %s: subject
1059 · %f: sanitized subject line, suitable for a filename
1061 · %b: body
1063 · %B: raw body (unwrapped subject and body)
1065 · %N: commit notes
1067 · %GG: raw verification message from GPG for a signed commit
1069 · %G?: show "G" for a good (valid) signature, "B" for a bad
1071 signature, "U" for a good signature with unknown validity, "X"
1072 for a good signature that has expired, "Y" for a good signature
1073 made by an expired key, "R" for a good signature made by a
1074 revoked key, "E" if the signature cannot be checked (e.g.
1075 missing key) and "N" for no signature
1076 · %GS: show the name of the signer for a signed commit
1078 · %GK: show the key used to sign a signed commit
1080 · %gD: reflog selector, e.g., refs/stash@{1} or refs/stash@{2
1082 minutes ago}; the format follows the rules described for the -g
1083 option. The portion before the @ is the refname as given on the
1084 command line (so git log -g refs/heads/master would yield
1085 refs/heads/master@{0}).
1086 · %gd: shortened reflog selector; same as %gD, but the refname
1088 portion is shortened for human readability (so
1089 refs/heads/master becomes just master).
1090 · %gn: reflog identity name
1092 · %gN: reflog identity name (respecting .mailmap, see git-
1094 shortlog(1) or git-blame(1))
1095 · %ge: reflog identity email
1097 · %gE: reflog identity email (respecting .mailmap, see git-
1099 shortlog(1) or git-blame(1))
1100 · %gs: reflog subject
1102 · %Cred: switch color to red
1104 · %Cgreen: switch color to green
1106 · %Cblue: switch color to blue
1108 · %Creset: reset color
1110 · %C(...): color specification, as described under Values in the
1112 "CONFIGURATION FILE" section of git-config(1). By default,
1113 colors are shown only when enabled for log output (by
1114 color.diff, color.ui, or --color, and respecting the auto
1115 settings of the former if we are going to a terminal).
1116 %C(auto,...) is accepted as a historical synonym for the
1117 default (e.g., %C(auto,red)). Specifying %C(always,...) will
1118 show the colors even when color is not otherwise enabled
1119 (though consider just using `--color=always to enable color for
1120 the whole output, including this format and anything else git
1121 might color). auto alone (i.e. %C(auto)) will turn on auto
1122 coloring on the next placeholders until the color is switched
1123 again.
1124 · %m: left (<), right (>) or boundary (-) mark
1126 · %n: newline
1128 · %%: a raw %
1130 · %x00: print a byte from a hex code
1132 · %w([<w>[,<i1>[,<i2>]]]): switch line wrapping, like the -w
1134 option of git-shortlog(1).
1135 · %<(<N>[,trunc|ltrunc|mtrunc]): make the next placeholder take
1137 at least N columns, padding spaces on the right if necessary.
1138 Optionally truncate at the beginning (ltrunc), the middle
1139 (mtrunc) or the end (trunc) if the output is longer than N
1140 columns. Note that truncating only works correctly with N >= 2.
1141 · %<|(<N>): make the next placeholder take at least until Nth
1143 columns, padding spaces on the right if necessary
1144 · %>(<N>), %>|(<N>): similar to %<(<N>), %<|(<N>) respectively,
1146 but padding spaces on the left
1147 · %>>(<N>), %>>|(<N>): similar to %>(<N>), %>|(<N>) respectively,
1149 except that if the next placeholder takes more spaces than
1150 given and there are spaces on its left, use those spaces
1151 · %><(<N>), %><|(<N>): similar to %<(<N>), %<|(<N>) respectively,
1153 but padding both sides (i.e. the text is centered)
1154 · %(trailers[:options]): display the trailers of the body as
1156 interpreted by git-interpret-trailers(1). The trailers string
1157 may be followed by a colon and zero or more comma-separated
1158 options. If the only option is given, omit non-trailer lines
1159 from the trailer block. If the unfold option is given, behave
1160 as if interpret-trailer’s --unfold option was given. E.g.,
1161 %(trailers:only,unfold) to do both.
1162 Note
1164 Some placeholders may depend on other options given to the revision
1165 traversal engine. For example, the %g* reflog options will insert
1166 an empty string unless we are traversing reflog entries (e.g., by
1167 git log -g). The %d and %D placeholders will use the "short"
1168 decoration format if --decorate was not already provided on the
1169 command line.
1170 If you add a + (plus sign) after % of a placeholder, a line-feed is
1172 inserted immediately before the expansion if and only if the
1173 placeholder expands to a non-empty string.
1174 If you add a - (minus sign) after % of a placeholder, all consecutive
1176 line-feeds immediately preceding the expansion are deleted if and only
1177 if the placeholder expands to an empty string.
1178 If you add a ` ` (space) after % of a placeholder, a space is inserted
1180 immediately before the expansion if and only if the placeholder expands
1181 to a non-empty string.
1182 · tformat:
1184 The tformat: format works exactly like format:, except that it
1186 provides "terminator" semantics instead of "separator" semantics.
1187 In other words, each commit has the message terminator character
1188 (usually a newline) appended, rather than a separator placed
1189 between entries. This means that the final entry of a single-line
1190 format will be properly terminated with a new line, just as the
1191 "oneline" format does. For example:
1192 $ git log -2 --pretty=format:%h 4da45bef \
1194 | perl -pe '$_ .= " -- NO NEWLINE\n" unless /\n/'
1195 4da45be
1196 7134973 -- NO NEWLINE
1197 $ git log -2 --pretty=tformat:%h 4da45bef \
1199 | perl -pe '$_ .= " -- NO NEWLINE\n" unless /\n/'
1200 4da45be
1201 7134973
1202 In addition, any unrecognized string that has a % in it is
1204 interpreted as if it has tformat: in front of it. For example,
1205 these two are equivalent:
1206 $ git log -2 --pretty=tformat:%h 4da45bef
1208 $ git log -2 --pretty=%h 4da45bef
1209
1212 -p, -u, --patch
1213 Generate patch (see section on generating patches).
1214 -s, --no-patch
1216 Suppress diff output. Useful for commands like git show that show
1217 the patch by default, or to cancel the effect of --patch.
1218 -U<n>, --unified=<n>
1220 Generate diffs with <n> lines of context instead of the usual
1221 three. Implies -p.
1222 --raw
1224 For each commit, show a summary of changes using the raw diff
1225 format. See the "RAW OUTPUT FORMAT" section of git-diff(1). This is
1226 different from showing the log itself in raw format, which you can
1227 achieve with --format=raw.
1228 --patch-with-raw
1230 Synonym for -p --raw.
1231 --indent-heuristic
1233 Enable the heuristic that shifts diff hunk boundaries to make
1234 patches easier to read. This is the default.
1235 --no-indent-heuristic
1237 Disable the indent heuristic.
1238 --minimal
1240 Spend extra time to make sure the smallest possible diff is
1241 produced.
1242 --patience
1244 Generate a diff using the "patience diff" algorithm.
1245 --histogram
1247 Generate a diff using the "histogram diff" algorithm.
1248 --anchored=<text>
1250 Generate a diff using the "anchored diff" algorithm.
1251 This option may be specified more than once.
1253 If a line exists in both the source and destination, exists only
1255 once, and starts with this text, this algorithm attempts to prevent
1256 it from appearing as a deletion or addition in the output. It uses
1257 the "patience diff" algorithm internally.
1258 --diff-algorithm={patience|minimal|histogram|myers}
1260 Choose a diff algorithm. The variants are as follows:
1261 default, myers
1263 The basic greedy diff algorithm. Currently, this is the
1264 default.
1265 minimal
1267 Spend extra time to make sure the smallest possible diff is
1268 produced.
1269 patience
1271 Use "patience diff" algorithm when generating patches.
1272 histogram
1274 This algorithm extends the patience algorithm to "support
1275 low-occurrence common elements".
1276 For instance, if you configured the diff.algorithm variable to a
1278 non-default value and want to use the default one, then you have to
1279 use --diff-algorithm=default option.
1280 --stat[=<width>[,<name-width>[,<count>]]]
1282 Generate a diffstat. By default, as much space as necessary will be
1283 used for the filename part, and the rest for the graph part.
1284 Maximum width defaults to terminal width, or 80 columns if not
1285 connected to a terminal, and can be overridden by <width>. The
1286 width of the filename part can be limited by giving another width
1287 <name-width> after a comma. The width of the graph part can be
1288 limited by using --stat-graph-width=<width> (affects all commands
1289 generating a stat graph) or by setting diff.statGraphWidth=<width>
1290 (does not affect git format-patch). By giving a third parameter
1291 <count>, you can limit the output to the first <count> lines,
1292 followed by ... if there are more.
1293 These parameters can also be set individually with
1295 --stat-width=<width>, --stat-name-width=<name-width> and
1296 --stat-count=<count>.
1297 --compact-summary
1299 Output a condensed summary of extended header information such as
1300 file creations or deletions ("new" or "gone", optionally "+l" if
1301 it’s a symlink) and mode changes ("+x" or "-x" for adding or
1302 removing executable bit respectively) in diffstat. The information
1303 is put between the filename part and the graph part. Implies
1304 --stat.
1305 --numstat
1307 Similar to --stat, but shows number of added and deleted lines in
1308 decimal notation and pathname without abbreviation, to make it more
1309 machine friendly. For binary files, outputs two - instead of saying
1310 0 0.
1311 --shortstat
1313 Output only the last line of the --stat format containing total
1314 number of modified files, as well as number of added and deleted
1315 lines.
1316 --dirstat[=<param1,param2,...>]
1318 Output the distribution of relative amount of changes for each
1319 sub-directory. The behavior of --dirstat can be customized by
1320 passing it a comma separated list of parameters. The defaults are
1321 controlled by the diff.dirstat configuration variable (see git-
1322 config(1)). The following parameters are available:
1323 changes
1325 Compute the dirstat numbers by counting the lines that have
1326 been removed from the source, or added to the destination. This
1327 ignores the amount of pure code movements within a file. In
1328 other words, rearranging lines in a file is not counted as much
1329 as other changes. This is the default behavior when no
1330 parameter is given.
1331 lines
1333 Compute the dirstat numbers by doing the regular line-based
1334 diff analysis, and summing the removed/added line counts. (For
1335 binary files, count 64-byte chunks instead, since binary files
1336 have no natural concept of lines). This is a more expensive
1337 --dirstat behavior than the changes behavior, but it does count
1338 rearranged lines within a file as much as other changes. The
1339 resulting output is consistent with what you get from the other
1340 --*stat options.
1341 files
1343 Compute the dirstat numbers by counting the number of files
1344 changed. Each changed file counts equally in the dirstat
1345 analysis. This is the computationally cheapest --dirstat
1346 behavior, since it does not have to look at the file contents
1347 at all.
1348 cumulative
1350 Count changes in a child directory for the parent directory as
1351 well. Note that when using cumulative, the sum of the
1352 percentages reported may exceed 100%. The default
1353 (non-cumulative) behavior can be specified with the
1354 noncumulative parameter.
1355 <limit>
1357 An integer parameter specifies a cut-off percent (3% by
1358 default). Directories contributing less than this percentage of
1359 the changes are not shown in the output.
1360 Example: The following will count changed files, while ignoring
1362 directories with less than 10% of the total amount of changed
1363 files, and accumulating child directory counts in the parent
1364 directories: --dirstat=files,10,cumulative.
1365 --summary
1367 Output a condensed summary of extended header information such as
1368 creations, renames and mode changes.
1369 --patch-with-stat
1371 Synonym for -p --stat.
1372 -z
1374 Separate the commits with NULs instead of with new newlines.
1375 Also, when --raw or --numstat has been given, do not munge
1377 pathnames and use NULs as output field terminators.
1378 Without this option, pathnames with "unusual" characters are quoted
1380 as explained for the configuration variable core.quotePath (see
1381 git-config(1)).
1382 --name-only
1384 Show only names of changed files.
1385 --name-status
1387 Show only names and status of changed files. See the description of
1388 the --diff-filter option on what the status letters mean.
1389 --submodule[=<format>]
1391 Specify how differences in submodules are shown. When specifying
1392 --submodule=short the short format is used. This format just shows
1393 the names of the commits at the beginning and end of the range.
1394 When --submodule or --submodule=log is specified, the log format is
1395 used. This format lists the commits in the range like git-
1396 submodule(1) summary does. When --submodule=diff is specified, the
1397 diff format is used. This format shows an inline diff of the
1398 changes in the submodule contents between the commit range.
1399 Defaults to diff.submodule or the short format if the config option
1400 is unset.
1401 --color[=<when>]
1403 Show colored diff. --color (i.e. without =<when>) is the same as
1404 --color=always. <when> can be one of always, never, or auto.
1405 --no-color
1407 Turn off colored diff. It is the same as --color=never.
1408 --color-moved[=<mode>]
1410 Moved lines of code are colored differently. The <mode> defaults to
1411 no if the option is not given and to zebra if the option with no
1412 mode is given. The mode must be one of:
1413 no
1415 Moved lines are not highlighted.
1416 default
1418 Is a synonym for zebra. This may change to a more sensible mode
1419 in the future.
1420 plain
1422 Any line that is added in one location and was removed in
1423 another location will be colored with color.diff.newMoved.
1424 Similarly color.diff.oldMoved will be used for removed lines
1425 that are added somewhere else in the diff. This mode picks up
1426 any moved line, but it is not very useful in a review to
1427 determine if a block of code was moved without permutation.
1428 zebra
1430 Blocks of moved text of at least 20 alphanumeric characters are
1431 detected greedily. The detected blocks are painted using either
1432 the color.diff.{old,new}Moved color or
1433 color.diff.{old,new}MovedAlternative. The change between the
1434 two colors indicates that a new block was detected.
1435 dimmed_zebra
1437 Similar to zebra, but additional dimming of uninteresting parts
1438 of moved code is performed. The bordering lines of two adjacent
1439 blocks are considered interesting, the rest is uninteresting.
1440 --word-diff[=<mode>]
1442 Show a word diff, using the <mode> to delimit changed words. By
1443 default, words are delimited by whitespace; see --word-diff-regex
1444 below. The <mode> defaults to plain, and must be one of:
1445 color
1447 Highlight changed words using only colors. Implies --color.
1448 plain
1450 Show words as [-removed-] and {+added+}. Makes no attempts to
1451 escape the delimiters if they appear in the input, so the
1452 output may be ambiguous.
1453 porcelain
1455 Use a special line-based format intended for script
1456 consumption. Added/removed/unchanged runs are printed in the
1457 usual unified diff format, starting with a +/-/` ` character at
1458 the beginning of the line and extending to the end of the line.
1459 Newlines in the input are represented by a tilde ~ on a line of
1460 its own.
1461 none
1463 Disable word diff again.
1464 Note that despite the name of the first mode, color is used to
1466 highlight the changed parts in all modes if enabled.
1467 --word-diff-regex=<regex>
1469 Use <regex> to decide what a word is, instead of considering runs
1470 of non-whitespace to be a word. Also implies --word-diff unless it
1471 was already enabled.
1472 Every non-overlapping match of the <regex> is considered a word.
1474 Anything between these matches is considered whitespace and
1475 ignored(!) for the purposes of finding differences. You may want to
1476 append |[^[:space:]] to your regular expression to make sure that
1477 it matches all non-whitespace characters. A match that contains a
1478 newline is silently truncated(!) at the newline.
1479 For example, --word-diff-regex=. will treat each character as a
1481 word and, correspondingly, show differences character by character.
1482 The regex can also be set via a diff driver or configuration
1484 option, see gitattributes(5) or git-config(1). Giving it explicitly
1485 overrides any diff driver or configuration setting. Diff drivers
1486 override configuration settings.
1487 --color-words[=<regex>]
1489 Equivalent to --word-diff=color plus (if a regex was specified)
1490 --word-diff-regex=<regex>.
1491 --no-renames
1493 Turn off rename detection, even when the configuration file gives
1494 the default to do so.
1495 --check
1497 Warn if changes introduce conflict markers or whitespace errors.
1498 What are considered whitespace errors is controlled by
1499 core.whitespace configuration. By default, trailing whitespaces
1500 (including lines that consist solely of whitespaces) and a space
1501 character that is immediately followed by a tab character inside
1502 the initial indent of the line are considered whitespace errors.
1503 Exits with non-zero status if problems are found. Not compatible
1504 with --exit-code.
1505 --ws-error-highlight=<kind>
1507 Highlight whitespace errors in the context, old or new lines of the
1508 diff. Multiple values are separated by comma, none resets previous
1509 values, default reset the list to new and all is a shorthand for
1510 old,new,context. When this option is not given, and the
1511 configuration variable diff.wsErrorHighlight is not set, only
1512 whitespace errors in new lines are highlighted. The whitespace
1513 errors are colored with color.diff.whitespace.
1514 --full-index
1516 Instead of the first handful of characters, show the full pre- and
1517 post-image blob object names on the "index" line when generating
1518 patch format output.
1519 --binary
1521 In addition to --full-index, output a binary diff that can be
1522 applied with git-apply.
1523 --abbrev[=<n>]
1525 Instead of showing the full 40-byte hexadecimal object name in
1526 diff-raw format output and diff-tree header lines, show only a
1527 partial prefix. This is independent of the --full-index option
1528 above, which controls the diff-patch output format. Non default
1529 number of digits can be specified with --abbrev=<n>.
1530 -B[<n>][/<m>], --break-rewrites[=[<n>][/<m>]]
1532 Break complete rewrite changes into pairs of delete and create.
1533 This serves two purposes:
1534 It affects the way a change that amounts to a total rewrite of a
1536 file not as a series of deletion and insertion mixed together with
1537 a very few lines that happen to match textually as the context, but
1538 as a single deletion of everything old followed by a single
1539 insertion of everything new, and the number m controls this aspect
1540 of the -B option (defaults to 60%). -B/70% specifies that less
1541 than 30% of the original should remain in the result for Git to
1542 consider it a total rewrite (i.e. otherwise the resulting patch
1543 will be a series of deletion and insertion mixed together with
1544 context lines).
1545 When used with -M, a totally-rewritten file is also considered as
1547 the source of a rename (usually -M only considers a file that
1548 disappeared as the source of a rename), and the number n controls
1549 this aspect of the -B option (defaults to 50%). -B20% specifies
1550 that a change with addition and deletion compared to 20% or more of
1551 the file’s size are eligible for being picked up as a possible
1552 source of a rename to another file.
1553 -M[<n>], --find-renames[=<n>]
1555 If generating diffs, detect and report renames for each commit. For
1556 following files across renames while traversing history, see
1557 --follow. If n is specified, it is a threshold on the similarity
1558 index (i.e. amount of addition/deletions compared to the file’s
1559 size). For example, -M90% means Git should consider a delete/add
1560 pair to be a rename if more than 90% of the file hasn’t changed.
1561 Without a % sign, the number is to be read as a fraction, with a
1562 decimal point before it. I.e., -M5 becomes 0.5, and is thus the
1563 same as -M50%. Similarly, -M05 is the same as -M5%. To limit
1564 detection to exact renames, use -M100%. The default similarity
1565 index is 50%.
1566 -C[<n>], --find-copies[=<n>]
1568 Detect copies as well as renames. See also --find-copies-harder. If
1569 n is specified, it has the same meaning as for -M<n>.
1570 --find-copies-harder
1572 For performance reasons, by default, -C option finds copies only if
1573 the original file of the copy was modified in the same changeset.
1574 This flag makes the command inspect unmodified files as candidates
1575 for the source of copy. This is a very expensive operation for
1576 large projects, so use it with caution. Giving more than one -C
1577 option has the same effect.
1578 -D, --irreversible-delete
1580 Omit the preimage for deletes, i.e. print only the header but not
1581 the diff between the preimage and /dev/null. The resulting patch is
1582 not meant to be applied with patch or git apply; this is solely for
1583 people who want to just concentrate on reviewing the text after the
1584 change. In addition, the output obviously lacks enough information
1585 to apply such a patch in reverse, even manually, hence the name of
1586 the option.
1587 When used together with -B, omit also the preimage in the deletion
1589 part of a delete/create pair.
1590 -l<num>
1592 The -M and -C options require O(n^2) processing time where n is the
1593 number of potential rename/copy targets. This option prevents
1594 rename/copy detection from running if the number of rename/copy
1595 targets exceeds the specified number.
1596 --diff-filter=[(A|C|D|M|R|T|U|X|B)...[*]]
1598 Select only files that are Added (A), Copied (C), Deleted (D),
1599 Modified (M), Renamed (R), have their type (i.e. regular file,
1600 symlink, submodule, ...) changed (T), are Unmerged (U), are Unknown
1601 (X), or have had their pairing Broken (B). Any combination of the
1602 filter characters (including none) can be used. When *
1603 (All-or-none) is added to the combination, all paths are selected
1604 if there is any file that matches other criteria in the comparison;
1605 if there is no file that matches other criteria, nothing is
1606 selected.
1607 Also, these upper-case letters can be downcased to exclude. E.g.
1609 --diff-filter=ad excludes added and deleted paths.
1610 Note that not all diffs can feature all types. For instance, diffs
1612 from the index to the working tree can never have Added entries
1613 (because the set of paths included in the diff is limited by what
1614 is in the index). Similarly, copied and renamed entries cannot
1615 appear if detection for those types is disabled.
1616 -S<string>
1618 Look for differences that change the number of occurrences of the
1619 specified string (i.e. addition/deletion) in a file. Intended for
1620 the scripter’s use.
1621 It is useful when you’re looking for an exact block of code (like a
1623 struct), and want to know the history of that block since it first
1624 came into being: use the feature iteratively to feed the
1625 interesting block in the preimage back into -S, and keep going
1626 until you get the very first version of the block.
1627 -G<regex>
1629 Look for differences whose patch text contains added/removed lines
1630 that match <regex>.
1631 To illustrate the difference between -S<regex> --pickaxe-regex and
1633 -G<regex>, consider a commit with the following diff in the same
1634 file:
1635 + return !regexec(regexp, two->ptr, 1, ®match, 0);
1637 ...
1638 - hit = !regexec(regexp, mf2.ptr, 1, ®match, 0);
1639 While git log -G"regexec\(regexp" will show this commit, git log
1641 -S"regexec\(regexp" --pickaxe-regex will not (because the number of
1642 occurrences of that string did not change).
1643 See the pickaxe entry in gitdiffcore(7) for more information.
1645 --find-object=<object-id>
1647 Look for differences that change the number of occurrences of the
1648 specified object. Similar to -S, just the argument is different in
1649 that it doesn’t search for a specific string but for a specific
1650 object id.
1651 The object can be a blob or a submodule commit. It implies the -t
1653 option in git-log to also find trees.
1654 --pickaxe-all
1656 When -S or -G finds a change, show all the changes in that
1657 changeset, not just the files that contain the change in <string>.
1658 --pickaxe-regex
1660 Treat the <string> given to -S as an extended POSIX regular
1661 expression to match.
1662 -O<orderfile>
1664 Control the order in which files appear in the output. This
1665 overrides the diff.orderFile configuration variable (see git-
1666 config(1)). To cancel diff.orderFile, use -O/dev/null.
1667 The output order is determined by the order of glob patterns in
1669 <orderfile>. All files with pathnames that match the first pattern
1670 are output first, all files with pathnames that match the second
1671 pattern (but not the first) are output next, and so on. All files
1672 with pathnames that do not match any pattern are output last, as if
1673 there was an implicit match-all pattern at the end of the file. If
1674 multiple pathnames have the same rank (they match the same pattern
1675 but no earlier patterns), their output order relative to each other
1676 is the normal order.
1677 <orderfile> is parsed as follows:
1679 · Blank lines are ignored, so they can be used as separators for
1681 readability.
1682 · Lines starting with a hash ("#") are ignored, so they can be
1684 used for comments. Add a backslash ("\") to the beginning of
1685 the pattern if it starts with a hash.
1686 · Each other line contains a single pattern.
1688 Patterns have the same syntax and semantics as patterns used for
1690 fnmatch(3) without the FNM_PATHNAME flag, except a pathname also
1691 matches a pattern if removing any number of the final pathname
1692 components matches the pattern. For example, the pattern "foo*bar"
1693 matches "fooasdfbar" and "foo/bar/baz/asdf" but not "foobarx".
1694 -R
1696 Swap two inputs; that is, show differences from index or on-disk
1697 file to tree contents.
1698 --relative[=<path>]
1700 When run from a subdirectory of the project, it can be told to
1701 exclude changes outside the directory and show pathnames relative
1702 to it with this option. When you are not in a subdirectory (e.g. in
1703 a bare repository), you can name which subdirectory to make the
1704 output relative to by giving a <path> as an argument.
1705 -a, --text
1707 Treat all files as text.
1708 --ignore-cr-at-eol
1710 Ignore carriage-return at the end of line when doing a comparison.
1711 --ignore-space-at-eol
1713 Ignore changes in whitespace at EOL.
1714 -b, --ignore-space-change
1716 Ignore changes in amount of whitespace. This ignores whitespace at
1717 line end, and considers all other sequences of one or more
1718 whitespace characters to be equivalent.
1719 -w, --ignore-all-space
1721 Ignore whitespace when comparing lines. This ignores differences
1722 even if one line has whitespace where the other line has none.
1723 --ignore-blank-lines
1725 Ignore changes whose lines are all blank.
1726 --inter-hunk-context=<lines>
1728 Show the context between diff hunks, up to the specified number of
1729 lines, thereby fusing hunks that are close to each other. Defaults
1730 to diff.interHunkContext or 0 if the config option is unset.
1731 -W, --function-context
1733 Show whole surrounding functions of changes.
1734 --ext-diff
1736 Allow an external diff helper to be executed. If you set an
1737 external diff driver with gitattributes(5), you need to use this
1738 option with git-log(1) and friends.
1739 --no-ext-diff
1741 Disallow external diff drivers.
1742 --textconv, --no-textconv
1744 Allow (or disallow) external text conversion filters to be run when
1745 comparing binary files. See gitattributes(5) for details. Because
1746 textconv filters are typically a one-way conversion, the resulting
1747 diff is suitable for human consumption, but cannot be applied. For
1748 this reason, textconv filters are enabled by default only for git-
1749 diff(1) and git-log(1), but not for git-format-patch(1) or diff
1750 plumbing commands.
1751 --ignore-submodules[=<when>]
1753 Ignore changes to submodules in the diff generation. <when> can be
1754 either "none", "untracked", "dirty" or "all", which is the default.
1755 Using "none" will consider the submodule modified when it either
1756 contains untracked or modified files or its HEAD differs from the
1757 commit recorded in the superproject and can be used to override any
1758 settings of the ignore option in git-config(1) or gitmodules(5).
1759 When "untracked" is used submodules are not considered dirty when
1760 they only contain untracked content (but they are still scanned for
1761 modified content). Using "dirty" ignores all changes to the work
1762 tree of submodules, only changes to the commits stored in the
1763 superproject are shown (this was the behavior until 1.7.0). Using
1764 "all" hides all changes to submodules.
1765 --src-prefix=<prefix>
1767 Show the given source prefix instead of "a/".
1768 --dst-prefix=<prefix>
1770 Show the given destination prefix instead of "b/".
1771 --no-prefix
1773 Do not show any source or destination prefix.
1774 --line-prefix=<prefix>
1776 Prepend an additional prefix to every line of output.
1777 --ita-invisible-in-index
1779 By default entries added by "git add -N" appear as an existing
1780 empty file in "git diff" and a new file in "git diff --cached".
1781 This option makes the entry appear as a new file in "git diff" and
1782 non-existent in "git diff --cached". This option could be reverted
1783 with --ita-visible-in-index. Both options are experimental and
1784 could be removed in future.
1785 For more detailed explanation on these common options, see also
1787 gitdiffcore(7).
1788
1790 When "git-diff-index", "git-diff-tree", or "git-diff-files" are run
1791 with a -p option, "git diff" without the --raw option, or "git log"
1792 with the "-p" option, they do not produce the output described above;
1793 instead they produce a patch file. You can customize the creation of
1794 such patches via the GIT_EXTERNAL_DIFF and the GIT_DIFF_OPTS
1795 environment variables.
1796 What the -p option produces is slightly different from the traditional
1798 diff format:
1799 1. It is preceded with a "git diff" header that looks like this:
1801 diff --git a/file1 b/file2
1803 The a/ and b/ filenames are the same unless rename/copy is
1805 involved. Especially, even for a creation or a deletion, /dev/null
1806 is not used in place of the a/ or b/ filenames.
1807 When rename/copy is involved, file1 and file2 show the name of the
1809 source file of the rename/copy and the name of the file that
1810 rename/copy produces, respectively.
1811 2. It is followed by one or more extended header lines:
1813 old mode <mode>
1815 new mode <mode>
1816 deleted file mode <mode>
1817 new file mode <mode>
1818 copy from <path>
1819 copy to <path>
1820 rename from <path>
1821 rename to <path>
1822 similarity index <number>
1823 dissimilarity index <number>
1824 index <hash>..<hash> <mode>
1825 File modes are printed as 6-digit octal numbers including the file
1827 type and file permission bits.
1828 Path names in extended headers do not include the a/ and b/
1830 prefixes.
1831 The similarity index is the percentage of unchanged lines, and the
1833 dissimilarity index is the percentage of changed lines. It is a
1834 rounded down integer, followed by a percent sign. The similarity
1835 index value of 100% is thus reserved for two equal files, while
1836 100% dissimilarity means that no line from the old file made it
1837 into the new one.
1838 The index line includes the SHA-1 checksum before and after the
1840 change. The <mode> is included if the file mode does not change;
1841 otherwise, separate lines indicate the old and the new mode.
1842 3. Pathnames with "unusual" characters are quoted as explained for the
1844 configuration variable core.quotePath (see git-config(1)).
1845 4. All the file1 files in the output refer to files before the commit,
1847 and all the file2 files refer to files after the commit. It is
1848 incorrect to apply each change to each file sequentially. For
1849 example, this patch will swap a and b:
1850 diff --git a/a b/b
1852 rename from a
1853 rename to b
1854 diff --git a/b b/a
1855 rename from b
1856 rename to a
1857
1859 Any diff-generating command can take the -c or --cc option to produce a
1860 combined diff when showing a merge. This is the default format when
1861 showing merges with git-diff(1) or git-show(1). Note also that you can
1862 give the -m option to any of these commands to force generation of
1863 diffs with individual parents of a merge.
1864 A combined diff format looks like this:
1866 diff --combined describe.c
1868 index fabadb8,cc95eb0..4866510
1869 --- a/describe.c
1870 +++ b/describe.c
1871 @@@ -98,20 -98,12 +98,20 @@@
1872 return (a_date > b_date) ? -1 : (a_date == b_date) ? 0 : 1;
1873 }
1874 - static void describe(char *arg)
1876 -static void describe(struct commit *cmit, int last_one)
1877 ++static void describe(char *arg, int last_one)
1878 {
1879 + unsigned char sha1[20];
1880 + struct commit *cmit;
1881 struct commit_list *list;
1882 static int initialized = 0;
1883 struct commit_name *n;
1884 + if (get_sha1(arg, sha1) < 0)
1886 + usage(describe_usage);
1887 + cmit = lookup_commit_reference(sha1);
1888 + if (!cmit)
1889 + usage(describe_usage);
1890 +
1891 if (!initialized) {
1892 initialized = 1;
1893 for_each_ref(get_name);
1894 1. It is preceded with a "git diff" header, that looks like this (when
1898 -c option is used):
1899 diff --combined file
1901 or like this (when --cc option is used):
1903 diff --cc file
1905 2. It is followed by one or more extended header lines (this example
1907 shows a merge with two parents):
1908 index <hash>,<hash>..<hash>
1910 mode <mode>,<mode>..<mode>
1911 new file mode <mode>
1912 deleted file mode <mode>,<mode>
1913 The mode <mode>,<mode>..<mode> line appears only if at least one of
1915 the <mode> is different from the rest. Extended headers with
1916 information about detected contents movement (renames and copying
1917 detection) are designed to work with diff of two <tree-ish> and are
1918 not used by combined diff format.
1919 3. It is followed by two-line from-file/to-file header
1921 --- a/file
1923 +++ b/file
1924 Similar to two-line header for traditional unified diff format,
1926 /dev/null is used to signal created or deleted files.
1927 4. Chunk header format is modified to prevent people from accidentally
1929 feeding it to patch -p1. Combined diff format was created for
1930 review of merge commit changes, and was not meant for apply. The
1931 change is similar to the change in the extended index header:
1932 @@@ <from-file-range> <from-file-range> <to-file-range> @@@
1934 There are (number of parents + 1) @ characters in the chunk header
1936 for combined diff format.
1937 Unlike the traditional unified diff format, which shows two files A and
1939 B with a single column that has - (minus — appears in A but removed in
1940 B), + (plus — missing in A but added to B), or " " (space — unchanged)
1941 prefix, this format compares two or more files file1, file2,... with
1942 one file X, and shows how X differs from each of fileN. One column for
1943 each of fileN is prepended to the output line to note how X’s line is
1944 different from it.
1945 A - character in the column N means that the line appears in fileN but
1947 it does not appear in the result. A + character in the column N means
1948 that the line appears in the result, and fileN does not have that line
1949 (in other words, the line was added, from the point of view of that
1950 parent).
1951 In the above example output, the function signature was changed from
1953 both files (hence two - removals from both file1 and file2, plus ++ to
1954 mean one line that was added does not appear in either file1 or file2).
1955 Also eight other lines are the same from file1 but do not appear in
1956 file2 (hence prefixed with +).
1957 When shown by git diff-tree -c, it compares the parents of a merge
1959 commit with the merge result (i.e. file1..fileN are the parents). When
1960 shown by git diff-files -c, it compares the two unresolved merge
1961 parents with the working tree file (i.e. file1 is stage 2 aka "our
1962 version", file2 is stage 3 aka "their version").
1963
1965 git log --no-merges
1966 Show the whole commit history, but skip any merges
1967 git log v2.6.12.. include/scsi drivers/scsi
1969 Show all commits since version v2.6.12 that changed any file in the
1970 include/scsi or drivers/scsi subdirectories
1971 git log --since="2 weeks ago" -- gitk
1973 Show the changes during the last two weeks to the file gitk. The --
1974 is necessary to avoid confusion with the branch named gitk
1975 git log --name-status release..test
1977 Show the commits that are in the "test" branch but not yet in the
1978 "release" branch, along with the list of paths each commit
1979 modifies.
1980 git log --follow builtin/rev-list.c
1982 Shows the commits that changed builtin/rev-list.c, including those
1983 commits that occurred before the file was given its present name.
1984 git log --branches --not --remotes=origin
1986 Shows all commits that are in any of local branches but not in any
1987 of remote-tracking branches for origin (what you have that origin
1988 doesn’t).
1989 git log master --not --remotes=*/master
1991 Shows all commits that are in local master but not in any remote
1992 repository master branches.
1993 git log -p -m --first-parent
1995 Shows the history including change diffs, but only from the “main
1996 branch” perspective, skipping commits that come from merged
1997 branches, and showing full diffs of changes introduced by the
1998 merges. This makes sense only when following a strict policy of
1999 merging all topic branches when staying on a single integration
2000 branch.
2001 git log -L '/int main/',/^}/:main.c
2003 Shows how the function main() in the file main.c evolved over time.
2004 git log -3
2006 Limits the number of commits to show to 3.
2007
2009 Git is to some extent character encoding agnostic.
2010 · The contents of the blob objects are uninterpreted sequences of
2012 bytes. There is no encoding translation at the core level.
2013 · Path names are encoded in UTF-8 normalization form C. This applies
2015 to tree objects, the index file, ref names, as well as path names
2016 in command line arguments, environment variables and config files
2017 (.git/config (see git-config(1)), gitignore(5), gitattributes(5)
2018 and gitmodules(5)).
2019 Note that Git at the core level treats path names simply as
2021 sequences of non-NUL bytes, there are no path name encoding
2022 conversions (except on Mac and Windows). Therefore, using non-ASCII
2023 path names will mostly work even on platforms and file systems that
2024 use legacy extended ASCII encodings. However, repositories created
2025 on such systems will not work properly on UTF-8-based systems (e.g.
2026 Linux, Mac, Windows) and vice versa. Additionally, many Git-based
2027 tools simply assume path names to be UTF-8 and will fail to display
2028 other encodings correctly.
2029 · Commit log messages are typically encoded in UTF-8, but other
2031 extended ASCII encodings are also supported. This includes
2032 ISO-8859-x, CP125x and many others, but not UTF-16/32, EBCDIC and
2033 CJK multi-byte encodings (GBK, Shift-JIS, Big5, EUC-x, CP9xx etc.).
2034 Although we encourage that the commit log messages are encoded in
2036 UTF-8, both the core and Git Porcelain are designed not to force UTF-8
2037 on projects. If all participants of a particular project find it more
2038 convenient to use legacy encodings, Git does not forbid it. However,
2039 there are a few things to keep in mind.
2040 1. git commit and git commit-tree issues a warning if the commit log
2042 message given to it does not look like a valid UTF-8 string, unless
2043 you explicitly say your project uses a legacy encoding. The way to
2044 say this is to have i18n.commitencoding in .git/config file, like
2045 this:
2046 [i18n]
2048 commitEncoding = ISO-8859-1
2049 Commit objects created with the above setting record the value of
2051 i18n.commitEncoding in its encoding header. This is to help other
2052 people who look at them later. Lack of this header implies that the
2053 commit log message is encoded in UTF-8.
2054 2. git log, git show, git blame and friends look at the encoding
2056 header of a commit object, and try to re-code the log message into
2057 UTF-8 unless otherwise specified. You can specify the desired
2058 output encoding with i18n.logOutputEncoding in .git/config file,
2059 like this:
2060 [i18n]
2062 logOutputEncoding = ISO-8859-1
2063 If you do not have this configuration variable, the value of
2065 i18n.commitEncoding is used instead.
2066 Note that we deliberately chose not to re-code the commit log message
2068 when a commit is made to force UTF-8 at the commit object level,
2069 because re-coding to UTF-8 is not necessarily a reversible operation.
2070
2072 See git-config(1) for core variables and git-diff(1) for settings
2073 related to diff generation.
2074 format.pretty
2076 Default for the --format option. (See Pretty Formats above.)
2077 Defaults to medium.
2078 i18n.logOutputEncoding
2080 Encoding to use when displaying logs. (See Discussion above.)
2081 Defaults to the value of i18n.commitEncoding if set, and UTF-8
2082 otherwise.
2083 log.date
2085 Default format for human-readable dates. (Compare the --date
2086 option.) Defaults to "default", which means to write dates like Sat
2087 May 8 19:35:34 2010 -0500.
2088 log.follow
2090 If true, git log will act as if the --follow option was used when a
2091 single <path> is given. This has the same limitations as --follow,
2092 i.e. it cannot be used to follow multiple files and does not work
2093 well on non-linear history.
2094 log.showRoot
2096 If false, git log and related commands will not treat the initial
2097 commit as a big creation event. Any root commits in git log -p
2098 output would be shown without a diff attached. The default is true.
2099 log.showSignature
2101 If true, git log and related commands will act as if the
2102 --show-signature option was passed to them.
2103 mailmap.*
2105 See git-shortlog(1).
2106 notes.displayRef
2108 Which refs, in addition to the default set by core.notesRef or
2109 GIT_NOTES_REF, to read notes from when showing commit messages with
2110 the log family of commands. See git-notes(1).
2111 May be an unabbreviated ref name or a glob and may be specified
2113 multiple times. A warning will be issued for refs that do not
2114 exist, but a glob that does not match any refs is silently ignored.
2115 This setting can be disabled by the --no-notes option, overridden
2117 by the GIT_NOTES_DISPLAY_REF environment variable, and overridden
2118 by the --notes=<ref> option.
2119
2121 Part of the git(1) suite
2122Git 2.18.1 05/14/2019 GIT-LOG(1)