6 git-range-diff - Compare two commit ranges (e.g. two versions of a branch)
11 'git range-diff' [--color=[<when>]] [--no-color] [<diff-options>]
12 [--no-dual-color] [--creation-factor=<factor>]
13 ( <range1> <range2> | <rev1>...<rev2> | <base> <rev1> <rev2> )
18 This command shows the differences between two versions of a patch
19 series, or more generally, two commit ranges (ignoring merge commits).
21 To that end, it first finds pairs of commits from both commit ranges
22 that correspond with each other. Two commits are said to correspond when
23 the diff between their patches (i.e. the author information, the commit
24 message and the commit diff) is reasonably small compared to the
25 patches' size. See ``Algorithm`` below for details.
27 Finally, the list of matching commits is shown in the order of the
28 second commit range, with unmatched commits being inserted just after
29 all of their ancestors have been shown.
35 When the commit diffs differ, `git range-diff` recreates the
36 original diffs' coloring, and adds outer -/+ diff markers with
37 the *background* being red/green to make it easier to see e.g.
38 when there was a change in what exact lines were added.
40 Additionally, the commit diff lines that are only present in the first commit
41 range are shown "dimmed" (this can be overridden using the `color.diff.<slot>`
42 config setting where `<slot>` is one of `contextDimmed`, `oldDimmed` and
43 `newDimmed`), and the commit diff lines that are only present in the second
44 commit range are shown in bold (which can be overridden using the config
45 settings `color.diff.<slot>` with `<slot>` being one of `contextBold`,
46 `oldBold` or `newBold`).
48 This is known to `range-diff` as "dual coloring". Use `--no-dual-color`
49 to revert to color all lines according to the outer diff markers
50 (and completely ignore the inner diff when it comes to color).
52 --creation-factor=<percent>::
53 Set the creation/deletion cost fudge factor to `<percent>`.
54 Defaults to 60. Try a larger value if `git range-diff` erroneously
55 considers a large change a total rewrite (deletion of one commit
56 and addition of another), and a smaller one in the reverse case.
57 See the ``Algorithm`` section below for an explanation why this is
60 --[no-]notes[=<ref>]::
61 This flag is passed to the `git log` program
62 (see linkgit:git-log[1]) that generates the patches.
65 Compare the commits specified by the two ranges, where
66 `<range1>` is considered an older version of `<range2>`.
69 Equivalent to passing `<rev2>..<rev1>` and `<rev1>..<rev2>`.
71 <base> <rev1> <rev2>::
72 Equivalent to passing `<base>..<rev1>` and `<base>..<rev2>`.
73 Note that `<base>` does not need to be the exact branch point
74 of the branches. Example: after rebasing a branch `my-topic`,
75 `git range-diff my-topic@{u} my-topic@{1} my-topic` would
76 show the differences introduced by the rebase.
78 `git range-diff` also accepts the regular diff options (see
79 linkgit:git-diff[1]), most notably the `--color=[<when>]` and
80 `--no-color` options. These options are used when generating the "diff
81 between patches", i.e. to compare the author, commit message and diff of
82 corresponding old/new commits. There is currently no means to tweak most of the
83 diff options passed to `git log` when generating those patches.
88 The output of the `range-diff` command is subject to change. It is
89 intended to be human-readable porcelain output, not something that can
90 be used across versions of Git to get a textually stable `range-diff`
91 (as opposed to something like the `--stable` option to
92 linkgit:git-patch-id[1]). There's also no equivalent of
93 linkgit:git-apply[1] for `range-diff`, the output is not intended to
96 This is particularly true when passing in diff options. Currently some
97 options like `--stat` can, as an emergent effect, produce output
98 that's quite useless in the context of `range-diff`. Future versions
99 of `range-diff` may learn to interpret such options in a manner
100 specific to `range-diff` (e.g. for `--stat` producing human-readable
101 output which summarizes how the diffstat changed).
105 This command uses the `diff.color.*` and `pager.range-diff` settings
106 (the latter is on by default).
107 See linkgit:git-config[1].
113 When a rebase required merge conflicts to be resolved, compare the changes
114 introduced by the rebase directly afterwards using:
117 $ git range-diff @{u} @{1} @
121 A typical output of `git range-diff` would look like this:
124 -: ------- > 1: 0ddba11 Prepare for the inevitable!
125 1: c0debee = 2: cab005e Add a helpful message at the start
126 2: f00dbal ! 3: decafe1 Describe a bug
128 Author: A U Thor <author@example.com>
130 -TODO: Describe a bug
135 -+What is unexpected is that it will also crash.
136 ++Unexpectedly, it also crashes. This is a bug, and the jury is
137 ++still out there how to fix it best. See ticket #314 for details.
140 3: bedead < -: ------- TO-UNDO
143 In this example, there are 3 old and 3 new commits, where the developer
144 removed the 3rd, added a new one before the first two, and modified the
145 commit message of the 2nd commit as well its diff.
147 When the output goes to a terminal, it is color-coded by default, just
148 like regular `git diff`'s output. In addition, the first line (adding a
149 commit) is green, the last line (deleting a commit) is red, the second
150 line (with a perfect match) is yellow like the commit header of `git
151 show`'s output, and the third line colors the old commit red, the new
152 one green and the rest like `git show`'s commit header.
154 A naive color-coded diff of diffs is actually a bit hard to read,
155 though, as it colors the entire lines red or green. The line that added
156 "What is unexpected" in the old commit, for example, is completely red,
157 even if the intent of the old commit was to add something.
159 To help with that, `range` uses the `--dual-color` mode by default. In
160 this mode, the diff of diffs will retain the original diff colors, and
161 prefix the lines with -/+ markers that have their *background* red or
162 green, to make it more obvious that they describe how the diff itself
169 The general idea is this: we generate a cost matrix between the commits
170 in both commit ranges, then solve the least-cost assignment.
172 The cost matrix is populated thusly: for each pair of commits, both
173 diffs are generated and the "diff of diffs" is generated, with 3 context
174 lines, then the number of lines in that diff is used as cost.
176 To avoid false positives (e.g. when a patch has been removed, and an
177 unrelated patch has been added between two iterations of the same patch
178 series), the cost matrix is extended to allow for that, by adding
179 fixed-cost entries for wholesale deletes/adds.
181 Example: Let commits `1--2` be the first iteration of a patch series and
182 `A--C` the second iteration. Let's assume that `A` is a cherry-pick of
183 `2,` and `C` is a cherry-pick of `1` but with a small modification (say,
184 a fixed typo). Visualize the commits as a bipartite graph:
194 We are looking for a "best" explanation of the new series in terms of
195 the old one. We can represent an "explanation" as an edge in the graph:
206 This explanation comes for "free" because there was no change. Similarly
207 `C` could be explained using `1`, but that comes at some cost c>0
208 because of the modification:
219 In mathematical terms, what we are looking for is some sort of a minimum
220 cost bipartite matching; `1` is matched to `C` at some cost, etc. The
221 underlying graph is in fact a complete bipartite graph; the cost we
222 associate with every edge is the size of the diff between the two
223 commits' patches. To explain also new commits, we introduce dummy nodes
238 The cost of an edge `o--C` is the size of `C`'s diff, modified by a
239 fudge factor that should be smaller than 100%. The cost of an edge
240 `o--o` is free. The fudge factor is necessary because even if `1` and
241 `C` have nothing in common, they may still share a few empty lines and
242 such, possibly making the assignment `1--C`, `o--o` slightly cheaper
243 than `1--o`, `o--C` even if `1` and `C` have nothing in common. With the
244 fudge factor we require a much larger common part to consider patches as
247 The overall time needed to compute this algorithm is the time needed to
248 compute n+m commit diffs and then n*m diffs of patches, plus the time
249 needed to compute the least-cost assignment between n and m diffs. Git
250 uses an implementation of the Jonker-Volgenant algorithm to solve the
251 assignment problem, which has cubic runtime complexity. The matching
252 found in this case will look like this:
273 Part of the linkgit:git[1] suite
projects / git / blob