6 git-blame - Show what revision and author last modified each line of a file
11 'git-blame' [-c] [-l] [-t] [-f] [-n] [-p] [--incremental] [-L n,m] [-S <revs-file>]
12 [-M] [-C] [-C] [--since=<date>] [<rev>] [--] <file>
17 Annotates each line in the given file with information from the revision which
18 last modified the line. Optionally, start annotating from the given revision.
20 Also it can limit the range of lines annotated.
22 This report doesn't tell you anything about lines which have been deleted or
23 replaced; you need to use a tool such as gitlink:git-diff[1] or the "pickaxe"
24 interface briefly mentioned in the following paragraph.
26 Apart from supporting file annotation, git also supports searching the
27 development history for when a code snippet occurred in a change. This makes it
28 possible to track when a code snippet was added to a file, moved or copied
29 between files, and eventually deleted or replaced. It works by searching for
30 a text string in the diff. A small example:
32 -----------------------------------------------------------------------------
33 $ git log --pretty=oneline -S'blame_usage'
34 5040f17eba15504bad66b14a645bddd9b015ebb7 blame -S <ancestry-file>
35 ea4c7f9bf69e781dd0cd88d2bccb2bf5cc15c9a7 git-blame: Make the output
36 -----------------------------------------------------------------------------
41 Use the same output mode as gitlink:git-annotate[1] (Default: off).
44 Annotate only the specified line range (lines count from 1).
47 Show long rev (Default: off).
50 Show raw timestamp (Default: off).
52 -S, --rev-file <revs-file>::
53 Use revs from revs-file instead of calling gitlink:git-rev-list[1].
56 Show filename in the original commit. By default
57 filename is shown if there is any line that came from a
58 file with different name, due to rename detection.
61 Show line number in the original commit (Default: off).
64 Show in a format designed for machine consumption.
67 Show the result incrementally in a format designed for
71 Detect moving lines in the file as well. When a commit
72 moves a block of lines in a file (e.g. the original file
73 has A and then B, and the commit changes it to B and
74 then A), traditional 'blame' algorithm typically blames
75 the lines that were moved up (i.e. B) to the parent and
76 assigns blame to the lines that were moved down (i.e. A)
77 to the child commit. With this option, both groups of
78 lines are blamed on the parent.
81 In addition to `-M`, detect lines copied from other
82 files that were modified in the same commit. This is
83 useful when you reorganize your program and move code
84 around across files. When this option is given twice,
85 the command looks for copies from all other files in the
86 parent for the commit that creates the file in addition.
95 In this format, each line is output after a header; the
96 header at the minimum has the first line which has:
98 - 40-byte SHA-1 of the commit the line is attributed to;
99 - the line number of the line in the original file;
100 - the line number of the line in the final file;
101 - on a line that starts a group of line from a different
102 commit than the previous one, the number of lines in this
103 group. On subsequent lines this field is absent.
105 This header line is followed by the following information
106 at least once for each commit:
108 - author name ("author"), email ("author-mail"), time
109 ("author-time"), and timezone ("author-tz"); similarly
111 - filename in the commit the line is attributed to.
112 - the first line of the commit log message ("summary").
114 The contents of the actual line is output after the above
115 header, prefixed by a TAB. This is to allow adding more
116 header elements later.
122 Unlike `git-blame` and `git-annotate` in older git, the extent
123 of annotation can be limited to both line ranges and revision
124 ranges. When you are interested in finding the origin for
125 ll. 40-60 for file `foo`, you can use `-L` option like these
126 (they mean the same thing -- both ask for 21 lines starting at
129 git blame -L 40,60 foo
130 git blame -L 40,+21 foo
132 Also you can use regular expression to specify the line range.
134 git blame -L '/^sub hello {/,/^}$/' foo
136 would limit the annotation to the body of `hello` subroutine.
138 When you are not interested in changes older than the version
139 v2.6.18, or changes older than 3 weeks, you can use revision
140 range specifiers similar to `git-rev-list`:
142 git blame v2.6.18.. -- foo
143 git blame --since=3.weeks -- foo
145 When revision range specifiers are used to limit the annotation,
146 lines that have not changed since the range boundary (either the
147 commit v2.6.18 or the most recent commit that is more than 3
148 weeks old in the above example) are blamed for that range
151 A particularly useful way is to see if an added file have lines
152 created by copy-and-paste from existing files. Sometimes this
153 indicates that the developer was being sloppy and did not
154 refactor the code properly. You can first find the commit that
155 introduced the file with:
157 git log --diff-filter=A --pretty=short -- foo
159 and then annotate the change between the commit and its
160 parents, using `commit{caret}!` notation:
162 git blame -C -C -f $commit^! -- foo
168 When called with `--incremental` option, the command outputs the
169 result as it is built. The output generally will talk about
170 lines touched by more recent commits first (i.e. the lines will
171 be annotated out of order) and is meant to be used by
174 The output format is similar to the Porcelain format, but it
175 does not contain the actual lines from the file that is being
178 . Each blame entry always starts with a line of:
180 <40-byte hex sha1> <sourceline> <resultline> <num_lines>
182 Line numbers count from 1.
184 . The first time that commit shows up in the stream, it has various
185 other information about it printed out with a one-word tag at the
186 beginning of each line about that "extended commit info" (author,
187 email, committer, dates, summary etc).
189 . Unlike Porcelain format, the filename information is always
190 given and terminates the entry:
192 "filename" <whitespace-quoted-filename-goes-here>
194 and thus it's really quite easy to parse for some line- and word-oriented
195 parser (which should be quite natural for most scripting languages).
198 For people who do parsing: to make it more robust, just ignore any
199 lines in between the first and last one ("<sha1>" and "filename" lines)
200 where you don't recognize the tag-words (or care about that particular
201 one) at the beginning of the "extended information" lines. That way, if
202 there is ever added information (like the commit encoding or extended
203 commit commentary), a blame viewer won't ever care.
208 gitlink:git-annotate[1]
212 Written by Junio C Hamano <junkio@cox.net>
216 Part of the gitlink:git[7] suite