1Commit Limiting 2~~~~~~~~~~~~~~~ 3 4Besides specifying a range of commits that should be listed using the 5special notations explained in the description, additional commit 6limiting may be applied. 7 8Using more options generally further limits the output (e.g. 9`--since=<date1>` limits to commits newer than `<date1>`, and using it 10with `--grep=<pattern>` further limits to commits whose log message 11has a line that matches `<pattern>`), unless otherwise noted. 12 13Note that these are applied before commit 14ordering and formatting options, such as `--reverse`. 15 16-- 17 18-<number>:: 19-n <number>:: 20--max-count=<number>:: 21 Limit the number of commits to output. 22 23--skip=<number>:: 24 Skip 'number' commits before starting to show the commit output. 25 26--since=<date>:: 27--after=<date>:: 28 Show commits more recent than a specific date. 29 30--until=<date>:: 31--before=<date>:: 32 Show commits older than a specific date. 33 34ifdef::git-rev-list[] 35--max-age=<timestamp>:: 36--min-age=<timestamp>:: 37 Limit the commits output to specified time range. 38endif::git-rev-list[] 39 40--author=<pattern>:: 41--committer=<pattern>:: 42 Limit the commits output to ones with author/committer 43 header lines that match the specified pattern (regular 44 expression). With more than one `--author=<pattern>`, 45 commits whose author matches any of the given patterns are 46 chosen (similarly for multiple `--committer=<pattern>`). 47 48--grep-reflog=<pattern>:: 49 Limit the commits output to ones with reflog entries that 50 match the specified pattern (regular expression). With 51 more than one `--grep-reflog`, commits whose reflog message 52 matches any of the given patterns are chosen. It is an 53 error to use this option unless `--walk-reflogs` is in use. 54 55--grep=<pattern>:: 56 Limit the commits output to ones with log message that 57 matches the specified pattern (regular expression). With 58 more than one `--grep=<pattern>`, commits whose message 59 matches any of the given patterns are chosen (but see 60 `--all-match`). 61+ 62When `--show-notes` is in effect, the message from the notes as 63if it is part of the log message. 64 65--all-match:: 66 Limit the commits output to ones that match all given `--grep`, 67 instead of ones that match at least one. 68 69-i:: 70--regexp-ignore-case:: 71 Match the regular expression limiting patterns without regard to letter 72 case. 73 74--basic-regexp:: 75 Consider the limiting patterns to be basic regular expressions; 76 this is the default. 77 78-E:: 79--extended-regexp:: 80 Consider the limiting patterns to be extended regular expressions 81 instead of the default basic regular expressions. 82 83-F:: 84--fixed-strings:: 85 Consider the limiting patterns to be fixed strings (don't interpret 86 pattern as a regular expression). 87 88--perl-regexp:: 89 Consider the limiting patterns to be Perl-compatible regular expressions. 90 Requires libpcre to be compiled in. 91 92--remove-empty:: 93 Stop when a given path disappears from the tree. 94 95--merges:: 96 Print only merge commits. This is exactly the same as `--min-parents=2`. 97 98--no-merges:: 99 Do not print commits with more than one parent. This is 100 exactly the same as `--max-parents=1`. 101 102--min-parents=<number>:: 103--max-parents=<number>:: 104--no-min-parents:: 105--no-max-parents:: 106 Show only commits which have at least (or at most) that many parent 107 commits. In particular, `--max-parents=1` is the same as `--no-merges`, 108 `--min-parents=2` is the same as `--merges`. `--max-parents=0` 109 gives all root commits and `--min-parents=3` all octopus merges. 110+ 111`--no-min-parents` and `--no-max-parents` reset these limits (to no limit) 112again. Equivalent forms are `--min-parents=0` (any commit has 0 or more 113parents) and `--max-parents=-1` (negative numbers denote no upper limit). 114 115--first-parent:: 116 Follow only the first parent commit upon seeing a merge 117 commit. This option can give a better overview when 118 viewing the evolution of a particular topic branch, 119 because merges into a topic branch tend to be only about 120 adjusting to updated upstream from time to time, and 121 this option allows you to ignore the individual commits 122 brought in to your history by such a merge. 123 124--not:: 125 Reverses the meaning of the '{caret}' prefix (or lack thereof) 126 for all following revision specifiers, up to the next `--not`. 127 128--all:: 129 Pretend as if all the refs in `refs/` are listed on the 130 command line as '<commit>'. 131 132--branches[=<pattern>]:: 133 Pretend as if all the refs in `refs/heads` are listed 134 on the command line as '<commit>'. If '<pattern>' is given, limit 135 branches to ones matching given shell glob. If pattern lacks '?', 136 '{asterisk}', or '[', '/{asterisk}' at the end is implied. 137 138--tags[=<pattern>]:: 139 Pretend as if all the refs in `refs/tags` are listed 140 on the command line as '<commit>'. If '<pattern>' is given, limit 141 tags to ones matching given shell glob. If pattern lacks '?', '{asterisk}', 142 or '[', '/{asterisk}' at the end is implied. 143 144--remotes[=<pattern>]:: 145 Pretend as if all the refs in `refs/remotes` are listed 146 on the command line as '<commit>'. If '<pattern>' is given, limit 147 remote-tracking branches to ones matching given shell glob. 148 If pattern lacks '?', '{asterisk}', or '[', '/{asterisk}' at the end is implied. 149 150--glob=<glob-pattern>:: 151 Pretend as if all the refs matching shell glob '<glob-pattern>' 152 are listed on the command line as '<commit>'. Leading 'refs/', 153 is automatically prepended if missing. If pattern lacks '?', '{asterisk}', 154 or '[', '/{asterisk}' at the end is implied. 155 156--ignore-missing:: 157 Upon seeing an invalid object name in the input, pretend as if 158 the bad input was not given. 159 160ifndef::git-rev-list[] 161--bisect:: 162 Pretend as if the bad bisection ref `refs/bisect/bad` 163 was listed and as if it was followed by `--not` and the good 164 bisection refs `refs/bisect/good-*` on the command 165 line. 166endif::git-rev-list[] 167 168--stdin:: 169 In addition to the '<commit>' listed on the command 170 line, read them from the standard input. If a '--' separator is 171 seen, stop reading commits and start reading paths to limit the 172 result. 173 174ifdef::git-rev-list[] 175--quiet:: 176 Don't print anything to standard output. This form 177 is primarily meant to allow the caller to 178 test the exit status to see if a range of objects is fully 179 connected (or not). It is faster than redirecting stdout 180 to `/dev/null` as the output does not have to be formatted. 181endif::git-rev-list[] 182 183--cherry-mark:: 184 Like `--cherry-pick` (see below) but mark equivalent commits 185 with `=` rather than omitting them, and inequivalent ones with `+`. 186 187--cherry-pick:: 188 Omit any commit that introduces the same change as 189 another commit on the ``other side'' when the set of 190 commits are limited with symmetric difference. 191+ 192For example, if you have two branches, `A` and `B`, a usual way 193to list all commits on only one side of them is with 194`--left-right` (see the example below in the description of 195the `--left-right` option). However, it shows the commits that were 196cherry-picked from the other branch (for example, ``3rd on b'' may be 197cherry-picked from branch A). With this option, such pairs of commits are 198excluded from the output. 199 200--left-only:: 201--right-only:: 202 List only commits on the respective side of a symmetric range, 203 i.e. only those which would be marked `<` resp. `>` by 204 `--left-right`. 205+ 206For example, `--cherry-pick --right-only A...B` omits those 207commits from `B` which are in `A` or are patch-equivalent to a commit in 208`A`. In other words, this lists the `+` commits from `git cherry A B`. 209More precisely, `--cherry-pick --right-only --no-merges` gives the exact 210list. 211 212--cherry:: 213 A synonym for `--right-only --cherry-mark --no-merges`; useful to 214 limit the output to the commits on our side and mark those that 215 have been applied to the other side of a forked history with 216 `git log --cherry upstream...mybranch`, similar to 217 `git cherry upstream mybranch`. 218 219-g:: 220--walk-reflogs:: 221 Instead of walking the commit ancestry chain, walk 222 reflog entries from the most recent one to older ones. 223 When this option is used you cannot specify commits to 224 exclude (that is, '{caret}commit', 'commit1..commit2', 225 nor 'commit1\...commit2' notations cannot be used). 226+ 227With `--pretty` format other than `oneline` (for obvious reasons), 228this causes the output to have two extra lines of information 229taken from the reflog. By default, 'commit@\{Nth}' notation is 230used in the output. When the starting commit is specified as 231'commit@\{now}', output also uses 'commit@\{timestamp}' notation 232instead. Under `--pretty=oneline`, the commit message is 233prefixed with this information on the same line. 234This option cannot be combined with `--reverse`. 235See also linkgit:git-reflog[1]. 236 237--merge:: 238 After a failed merge, show refs that touch files having a 239 conflict and don't exist on all heads to merge. 240 241--boundary:: 242 Output excluded boundary commits. Boundary commits are 243 prefixed with `-`. 244 245-- 246 247History Simplification 248~~~~~~~~~~~~~~~~~~~~~~ 249 250Sometimes you are only interested in parts of the history, for example the 251commits modifying a particular <path>. But there are two parts of 252'History Simplification', one part is selecting the commits and the other 253is how to do it, as there are various strategies to simplify the history. 254 255The following options select the commits to be shown: 256 257<paths>:: 258 Commits modifying the given <paths> are selected. 259 260--simplify-by-decoration:: 261 Commits that are referred by some branch or tag are selected. 262 263Note that extra commits can be shown to give a meaningful history. 264 265The following options affect the way the simplification is performed: 266 267Default mode:: 268 Simplifies the history to the simplest history explaining the 269 final state of the tree. Simplest because it prunes some side 270 branches if the end result is the same (i.e. merging branches 271 with the same content) 272 273--full-history:: 274 Same as the default mode, but does not prune some history. 275 276--dense:: 277 Only the selected commits are shown, plus some to have a 278 meaningful history. 279 280--sparse:: 281 All commits in the simplified history are shown. 282 283--simplify-merges:: 284 Additional option to `--full-history` to remove some needless 285 merges from the resulting history, as there are no selected 286 commits contributing to this merge. 287 288--ancestry-path:: 289 When given a range of commits to display (e.g. 'commit1..commit2' 290 or 'commit2 {caret}commit1'), only display commits that exist 291 directly on the ancestry chain between the 'commit1' and 292 'commit2', i.e. commits that are both descendants of 'commit1', 293 and ancestors of 'commit2'. 294 295A more detailed explanation follows. 296 297Suppose you specified `foo` as the <paths>. We shall call commits 298that modify `foo` !TREESAME, and the rest TREESAME. (In a diff 299filtered for `foo`, they look different and equal, respectively.) 300 301In the following, we will always refer to the same example history to 302illustrate the differences between simplification settings. We assume 303that you are filtering for a file `foo` in this commit graph: 304----------------------------------------------------------------------- 305 .-A---M---N---O---P---Q 306 / / / / / / 307 I B C D E Y 308 \ / / / / / 309 `-------------' X 310----------------------------------------------------------------------- 311The horizontal line of history A---Q is taken to be the first parent of 312each merge. The commits are: 313 314* `I` is the initial commit, in which `foo` exists with contents 315 ``asdf'', and a file `quux` exists with contents ``quux''. Initial 316 commits are compared to an empty tree, so `I` is !TREESAME. 317 318* In `A`, `foo` contains just ``foo''. 319 320* `B` contains the same change as `A`. Its merge `M` is trivial and 321 hence TREESAME to all parents. 322 323* `C` does not change `foo`, but its merge `N` changes it to ``foobar'', 324 so it is not TREESAME to any parent. 325 326* `D` sets `foo` to ``baz''. Its merge `O` combines the strings from 327 `N` and `D` to ``foobarbaz''; i.e., it is not TREESAME to any parent. 328 329* `E` changes `quux` to ``xyzzy'', and its merge `P` combines the 330 strings to ``quux xyzzy''. `P` is TREESAME to `O`, but not to `E`. 331 332* `X` is an independent root commit that added a new file `side`, and `Y` 333 modified it. `Y` is TREESAME to `X`. Its merge `Q` added `side` to `P`, and 334 `Q` is TREESAME to `P`, but not to `Y`. 335 336`rev-list` walks backwards through history, including or excluding 337commits based on whether `--full-history` and/or parent rewriting 338(via `--parents` or `--children`) are used. The following settings 339are available. 340 341Default mode:: 342 Commits are included if they are not TREESAME to any parent 343 (though this can be changed, see `--sparse` below). If the 344 commit was a merge, and it was TREESAME to one parent, follow 345 only that parent. (Even if there are several TREESAME 346 parents, follow only one of them.) Otherwise, follow all 347 parents. 348+ 349This results in: 350+ 351----------------------------------------------------------------------- 352 .-A---N---O 353 / / / 354 I---------D 355----------------------------------------------------------------------- 356+ 357Note how the rule to only follow the TREESAME parent, if one is 358available, removed `B` from consideration entirely. `C` was 359considered via `N`, but is TREESAME. Root commits are compared to an 360empty tree, so `I` is !TREESAME. 361+ 362Parent/child relations are only visible with `--parents`, but that does 363not affect the commits selected in default mode, so we have shown the 364parent lines. 365 366--full-history without parent rewriting:: 367 This mode differs from the default in one point: always follow 368 all parents of a merge, even if it is TREESAME to one of them. 369 Even if more than one side of the merge has commits that are 370 included, this does not imply that the merge itself is! In 371 the example, we get 372+ 373----------------------------------------------------------------------- 374 I A B N D O P Q 375----------------------------------------------------------------------- 376+ 377`M` was excluded because it is TREESAME to both parents. `E`, 378`C` and `B` were all walked, but only `B` was !TREESAME, so the others 379do not appear. 380+ 381Note that without parent rewriting, it is not really possible to talk 382about the parent/child relationships between the commits, so we show 383them disconnected. 384 385--full-history with parent rewriting:: 386 Ordinary commits are only included if they are !TREESAME 387 (though this can be changed, see `--sparse` below). 388+ 389Merges are always included. However, their parent list is rewritten: 390Along each parent, prune away commits that are not included 391themselves. This results in 392+ 393----------------------------------------------------------------------- 394 .-A---M---N---O---P---Q 395 / / / / / 396 I B / D / 397 \ / / / / 398 `-------------' 399----------------------------------------------------------------------- 400+ 401Compare to `--full-history` without rewriting above. Note that `E` 402was pruned away because it is TREESAME, but the parent list of P was 403rewritten to contain `E`'s parent `I`. The same happened for `C` and 404`N`, and `X`, `Y` and `Q`. 405 406In addition to the above settings, you can change whether TREESAME 407affects inclusion: 408 409--dense:: 410 Commits that are walked are included if they are not TREESAME 411 to any parent. 412 413--sparse:: 414 All commits that are walked are included. 415+ 416Note that without `--full-history`, this still simplifies merges: if 417one of the parents is TREESAME, we follow only that one, so the other 418sides of the merge are never walked. 419 420--simplify-merges:: 421 First, build a history graph in the same way that 422 `--full-history` with parent rewriting does (see above). 423+ 424Then simplify each commit `C` to its replacement `C'` in the final 425history according to the following rules: 426+ 427-- 428* Set `C'` to `C`. 429+ 430* Replace each parent `P` of `C'` with its simplification `P'`. In 431 the process, drop parents that are ancestors of other parents or that are 432 root commits TREESAME to an empty tree, and remove duplicates, but take care 433 to never drop all parents that we are TREESAME to. 434+ 435* If after this parent rewriting, `C'` is a root or merge commit (has 436 zero or >1 parents), a boundary commit, or !TREESAME, it remains. 437 Otherwise, it is replaced with its only parent. 438-- 439+ 440The effect of this is best shown by way of comparing to 441`--full-history` with parent rewriting. The example turns into: 442+ 443----------------------------------------------------------------------- 444 .-A---M---N---O 445 / / / 446 I B D 447 \ / / 448 `---------' 449----------------------------------------------------------------------- 450+ 451Note the major differences in `N`, `P`, and `Q` over `--full-history`: 452+ 453-- 454* `N`'s parent list had `I` removed, because it is an ancestor of the 455 other parent `M`. Still, `N` remained because it is !TREESAME. 456+ 457* `P`'s parent list similarly had `I` removed. `P` was then 458 removed completely, because it had one parent and is TREESAME. 459+ 460* `Q`'s parent list had `Y` simplified to `X`. `X` was then removed, because it 461 was a TREESAME root. `Q` was then removed completely, because it had one 462 parent and is TREESAME. 463-- 464 465Finally, there is a fifth simplification mode available: 466 467--ancestry-path:: 468 Limit the displayed commits to those directly on the ancestry 469 chain between the ``from'' and ``to'' commits in the given commit 470 range. I.e. only display commits that are ancestor of the ``to'' 471 commit and descendants of the ``from'' commit. 472+ 473As an example use case, consider the following commit history: 474+ 475----------------------------------------------------------------------- 476 D---E-------F 477 / \ \ 478 B---C---G---H---I---J 479 / \ 480 A-------K---------------L--M 481----------------------------------------------------------------------- 482+ 483A regular 'D..M' computes the set of commits that are ancestors of `M`, 484but excludes the ones that are ancestors of `D`. This is useful to see 485what happened to the history leading to `M` since `D`, in the sense 486that ``what does `M` have that did not exist in `D`''. The result in this 487example would be all the commits, except `A` and `B` (and `D` itself, 488of course). 489+ 490When we want to find out what commits in `M` are contaminated with the 491bug introduced by `D` and need fixing, however, we might want to view 492only the subset of 'D..M' that are actually descendants of `D`, i.e. 493excluding `C` and `K`. This is exactly what the `--ancestry-path` 494option does. Applied to the 'D..M' range, it results in: 495+ 496----------------------------------------------------------------------- 497 E-------F 498 \ \ 499 G---H---I---J 500 \ 501 L--M 502----------------------------------------------------------------------- 503 504The `--simplify-by-decoration` option allows you to view only the 505big picture of the topology of the history, by omitting commits 506that are not referenced by tags. Commits are marked as !TREESAME 507(in other words, kept after history simplification rules described 508above) if (1) they are referenced by tags, or (2) they change the 509contents of the paths given on the command line. All other 510commits are marked as TREESAME (subject to be simplified away). 511 512ifdef::git-rev-list[] 513Bisection Helpers 514~~~~~~~~~~~~~~~~~ 515 516--bisect:: 517 Limit output to the one commit object which is roughly halfway between 518 included and excluded commits. Note that the bad bisection ref 519 `refs/bisect/bad` is added to the included commits (if it 520 exists) and the good bisection refs `refs/bisect/good-*` are 521 added to the excluded commits (if they exist). Thus, supposing there 522 are no refs in `refs/bisect/`, if 523+ 524----------------------------------------------------------------------- 525 $ git rev-list --bisect foo ^bar ^baz 526----------------------------------------------------------------------- 527+ 528outputs 'midpoint', the output of the two commands 529+ 530----------------------------------------------------------------------- 531 $ git rev-list foo ^midpoint 532 $ git rev-list midpoint ^bar ^baz 533----------------------------------------------------------------------- 534+ 535would be of roughly the same length. Finding the change which 536introduces a regression is thus reduced to a binary search: repeatedly 537generate and test new 'midpoint's until the commit chain is of length 538one. 539 540--bisect-vars:: 541 This calculates the same as `--bisect`, except that refs in 542 `refs/bisect/` are not used, and except that this outputs 543 text ready to be eval'ed by the shell. These lines will assign the 544 name of the midpoint revision to the variable `bisect_rev`, and the 545 expected number of commits to be tested after `bisect_rev` is tested 546 to `bisect_nr`, the expected number of commits to be tested if 547 `bisect_rev` turns out to be good to `bisect_good`, the expected 548 number of commits to be tested if `bisect_rev` turns out to be bad to 549 `bisect_bad`, and the number of commits we are bisecting right now to 550 `bisect_all`. 551 552--bisect-all:: 553 This outputs all the commit objects between the included and excluded 554 commits, ordered by their distance to the included and excluded 555 commits. Refs in `refs/bisect/` are not used. The farthest 556 from them is displayed first. (This is the only one displayed by 557 `--bisect`.) 558+ 559This is useful because it makes it easy to choose a good commit to 560test when you want to avoid to test some of them for some reason (they 561may not compile for example). 562+ 563This option can be used along with `--bisect-vars`, in this case, 564after all the sorted commit objects, there will be the same text as if 565`--bisect-vars` had been used alone. 566endif::git-rev-list[] 567 568 569Commit Ordering 570~~~~~~~~~~~~~~~ 571 572By default, the commits are shown in reverse chronological order. 573 574--date-order:: 575 Show no parents before all of its children are shown, but 576 otherwise show commits in the commit timestamp order. 577 578--author-date-order:: 579 Show no parents before all of its children are shown, but 580 otherwise show commits in the author timestamp order. 581 582--topo-order:: 583 Show no parents before all of its children are shown, and 584 avoid showing commits on multiple lines of history 585 intermixed. 586+ 587For example, in a commit history like this: 588+ 589---------------------------------------------------------------- 590 591 ---1----2----4----7 592 \ \ 593 3----5----6----8--- 594 595---------------------------------------------------------------- 596+ 597where the numbers denote the order of commit timestamps, `git 598rev-list` and friends with `--date-order` show the commits in the 599timestamp order: 8 7 6 5 4 3 2 1. 600+ 601With `--topo-order`, they would show 8 6 5 3 7 4 2 1 (or 8 7 4 2 6 5 6023 1); some older commits are shown before newer ones in order to 603avoid showing the commits from two parallel development track mixed 604together. 605 606--reverse:: 607 Output the commits in reverse order. 608 Cannot be combined with `--walk-reflogs`. 609 610Object Traversal 611~~~~~~~~~~~~~~~~ 612 613These options are mostly targeted for packing of Git repositories. 614 615--objects:: 616 Print the object IDs of any object referenced by the listed 617 commits. `--objects foo ^bar` thus means ``send me 618 all object IDs which I need to download if I have the commit 619 object _bar_ but not _foo_''. 620 621--objects-edge:: 622 Similar to `--objects`, but also print the IDs of excluded 623 commits prefixed with a ``-'' character. This is used by 624 linkgit:git-pack-objects[1] to build ``thin'' pack, which records 625 objects in deltified form based on objects contained in these 626 excluded commits to reduce network traffic. 627 628--unpacked:: 629 Only useful with `--objects`; print the object IDs that are not 630 in packs. 631 632--no-walk[=(sorted|unsorted)]:: 633 Only show the given commits, but do not traverse their ancestors. 634 This has no effect if a range is specified. If the argument 635 `unsorted` is given, the commits are shown in the order they were 636 given on the command line. Otherwise (if `sorted` or no argument 637 was given), the commits are shown in reverse chronological order 638 by commit time. 639 640--do-walk:: 641 Overrides a previous `--no-walk`. 642 643Commit Formatting 644~~~~~~~~~~~~~~~~~ 645 646ifdef::git-rev-list[] 647Using these options, linkgit:git-rev-list[1] will act similar to the 648more specialized family of commit log tools: linkgit:git-log[1], 649linkgit:git-show[1], and linkgit:git-whatchanged[1] 650endif::git-rev-list[] 651 652include::pretty-options.txt[] 653 654--relative-date:: 655 Synonym for `--date=relative`. 656 657--date=(relative|local|default|iso|rfc|short|raw):: 658 Only takes effect for dates shown in human-readable format, such 659 as when using `--pretty`. `log.date` config variable sets a default 660 value for the log command's `--date` option. 661+ 662`--date=relative` shows dates relative to the current time, 663e.g. ``2 hours ago''. 664+ 665`--date=local` shows timestamps in user's local time zone. 666+ 667`--date=iso` (or `--date=iso8601`) shows timestamps in ISO 8601 format. 668+ 669`--date=rfc` (or `--date=rfc2822`) shows timestamps in RFC 2822 670format, often found in email messages. 671+ 672`--date=short` shows only the date, but not the time, in `YYYY-MM-DD` format. 673+ 674`--date=raw` shows the date in the internal raw Git format `%s %z` format. 675+ 676`--date=default` shows timestamps in the original time zone 677(either committer's or author's). 678 679ifdef::git-rev-list[] 680--header:: 681 Print the contents of the commit in raw-format; each record is 682 separated with a NUL character. 683endif::git-rev-list[] 684 685--parents:: 686 Print also the parents of the commit (in the form "commit parent..."). 687 Also enables parent rewriting, see 'History Simplification' below. 688 689--children:: 690 Print also the children of the commit (in the form "commit child..."). 691 Also enables parent rewriting, see 'History Simplification' below. 692 693ifdef::git-rev-list[] 694--timestamp:: 695 Print the raw commit timestamp. 696endif::git-rev-list[] 697 698--left-right:: 699 Mark which side of a symmetric diff a commit is reachable from. 700 Commits from the left side are prefixed with `<` and those from 701 the right with `>`. If combined with `--boundary`, those 702 commits are prefixed with `-`. 703+ 704For example, if you have this topology: 705+ 706----------------------------------------------------------------------- 707 y---b---b branch B 708 / \ / 709 / . 710 / / \ 711 o---x---a---a branch A 712----------------------------------------------------------------------- 713+ 714you would get an output like this: 715+ 716----------------------------------------------------------------------- 717 $ git rev-list --left-right --boundary --pretty=oneline A...B 718 719 >bbbbbbb... 3rd on b 720 >bbbbbbb... 2nd on b 721 <aaaaaaa... 3rd on a 722 <aaaaaaa... 2nd on a 723 -yyyyyyy... 1st on b 724 -xxxxxxx... 1st on a 725----------------------------------------------------------------------- 726 727--graph:: 728 Draw a text-based graphical representation of the commit history 729 on the left hand side of the output. This may cause extra lines 730 to be printed in between commits, in order for the graph history 731 to be drawn properly. 732+ 733This enables parent rewriting, see 'History Simplification' below. 734+ 735This implies the `--topo-order` option by default, but the 736`--date-order` option may also be specified. 737 738ifdef::git-rev-list[] 739--count:: 740 Print a number stating how many commits would have been 741 listed, and suppress all other output. When used together 742 with `--left-right`, instead print the counts for left and 743 right commits, separated by a tab. When used together with 744 `--cherry-mark`, omit patch equivalent commits from these 745 counts and print the count for equivalent commits separated 746 by a tab. 747endif::git-rev-list[] 748 749ifndef::git-rev-list[] 750Diff Formatting 751~~~~~~~~~~~~~~~ 752 753Listed below are options that control the formatting of diff output. 754Some of them are specific to linkgit:git-rev-list[1], however other diff 755options may be given. See linkgit:git-diff-files[1] for more options. 756 757-c:: 758 With this option, diff output for a merge commit 759 shows the differences from each of the parents to the merge result 760 simultaneously instead of showing pairwise diff between a parent 761 and the result one at a time. Furthermore, it lists only files 762 which were modified from all parents. 763 764--cc:: 765 This flag implies the `-c` option and further compresses the 766 patch output by omitting uninteresting hunks whose contents in 767 the parents have only two variants and the merge result picks 768 one of them without modification. 769 770-m:: 771 This flag makes the merge commits show the full diff like 772 regular commits; for each merge parent, a separate log entry 773 and diff is generated. An exception is that only diff against 774 the first parent is shown when `--first-parent` option is given; 775 in that case, the output represents the changes the merge 776 brought _into_ the then-current branch. 777 778-r:: 779 Show recursive diffs. 780 781-t:: 782 Show the tree objects in the diff output. This implies `-r`. 783endif::git-rev-list[]