]>
Commit | Line | Data |
---|---|---|
8f2b72a9 JF |
1 | git-blame(1) |
2 | ============ | |
3 | ||
4 | NAME | |
5 | ---- | |
26e8c5d3 | 6 | git-blame - Show what revision and author last modified each line of a file |
8f2b72a9 JF |
7 | |
8 | SYNOPSIS | |
9 | -------- | |
acca687f | 10 | [verse] |
13b8f68c TR |
11 | 'git blame' [-c] [-b] [-l] [--root] [-t] [-f] [-n] [-s] [-e] [-p] [-w] [--incremental] |
12 | [-L n,m | -L :fn] [-S <revs-file>] [-M] [-C] [-C] [-C] [--since=<date>] | |
13 | [--abbrev=<n>] [<rev> | --contents <file> | --reverse <rev>] [--] <file> | |
8f2b72a9 JF |
14 | |
15 | DESCRIPTION | |
16 | ----------- | |
26e8c5d3 JF |
17 | |
18 | Annotates each line in the given file with information from the revision which | |
19 | last modified the line. Optionally, start annotating from the given revision. | |
20 | ||
b89510f0 | 21 | The command can also limit the range of lines annotated. |
acca687f | 22 | |
e5dce96e JH |
23 | The origin of lines is automatically followed across whole-file |
24 | renames (currently there is no option to turn the rename-following | |
25 | off). To follow lines moved from one file to another, or to follow | |
26 | lines that were copied and pasted from another file, etc., see the | |
27 | `-C` and `-M` options. | |
28 | ||
b89510f0 | 29 | The report does not tell you anything about lines which have been deleted or |
0b444cdb | 30 | replaced; you need to use a tool such as 'git diff' or the "pickaxe" |
26e8c5d3 JF |
31 | interface briefly mentioned in the following paragraph. |
32 | ||
2de9b711 | 33 | Apart from supporting file annotation, Git also supports searching the |
23bfbb81 | 34 | development history for when a code snippet occurred in a change. This makes it |
26e8c5d3 JF |
35 | possible to track when a code snippet was added to a file, moved or copied |
36 | between files, and eventually deleted or replaced. It works by searching for | |
37 | a text string in the diff. A small example: | |
38 | ||
39 | ----------------------------------------------------------------------------- | |
40 | $ git log --pretty=oneline -S'blame_usage' | |
41 | 5040f17eba15504bad66b14a645bddd9b015ebb7 blame -S <ancestry-file> | |
42 | ea4c7f9bf69e781dd0cd88d2bccb2bf5cc15c9a7 git-blame: Make the output | |
43 | ----------------------------------------------------------------------------- | |
8f2b72a9 JF |
44 | |
45 | OPTIONS | |
46 | ------- | |
635f4a30 | 47 | include::blame-options.txt[] |
b19ee24b | 48 | |
635f4a30 | 49 | -c:: |
5162e697 | 50 | Use the same output mode as linkgit:git-annotate[1] (Default: off). |
8f2b72a9 | 51 | |
635f4a30 AR |
52 | --score-debug:: |
53 | Include debugging information related to the movement of | |
54 | lines between files (see `-C`) and lines moved within a | |
55 | file (see `-M`). The first number listed is the score. | |
56 | This is the number of alphanumeric characters detected | |
b89510f0 | 57 | as having been moved between or within files. This must be above |
0b444cdb | 58 | a certain threshold for 'git blame' to consider those lines |
635f4a30 | 59 | of code to have been moved. |
8f2b72a9 | 60 | |
3240240f SB |
61 | -f:: |
62 | --show-name:: | |
b89510f0 DM |
63 | Show the filename in the original commit. By default |
64 | the filename is shown if there is any line that came from a | |
65 | file with a different name, due to rename detection. | |
b24642b2 | 66 | |
3240240f SB |
67 | -n:: |
68 | --show-number:: | |
b89510f0 | 69 | Show the line number in the original commit (Default: off). |
b24642b2 | 70 | |
093dc5be | 71 | -s:: |
b89510f0 | 72 | Suppress the author name and timestamp from the output. |
093dc5be | 73 | |
1b8cdce9 KB |
74 | -e:: |
75 | --show-email:: | |
76 | Show the author email instead of author name (Default: off). | |
77 | ||
b82871b3 | 78 | -w:: |
b89510f0 DM |
79 | Ignore whitespace when comparing the parent's version and |
80 | the child's to find where the lines came from. | |
b82871b3 | 81 | |
84393bfd NK |
82 | --abbrev=<n>:: |
83 | Instead of using the default 7+1 hexadecimal digits as the | |
84 | abbreviated object name, use <n>+1 digits. Note that 1 column | |
85 | is used for a caret to mark the boundary commit. | |
86 | ||
b82871b3 | 87 | |
b24642b2 JH |
88 | THE PORCELAIN FORMAT |
89 | -------------------- | |
90 | ||
91 | In this format, each line is output after a header; the | |
23bfbb81 | 92 | header at the minimum has the first line which has: |
b24642b2 JH |
93 | |
94 | - 40-byte SHA-1 of the commit the line is attributed to; | |
95 | - the line number of the line in the original file; | |
96 | - the line number of the line in the final file; | |
b89510f0 | 97 | - on a line that starts a group of lines from a different |
b24642b2 JH |
98 | commit than the previous one, the number of lines in this |
99 | group. On subsequent lines this field is absent. | |
100 | ||
101 | This header line is followed by the following information | |
102 | at least once for each commit: | |
103 | ||
b89510f0 | 104 | - the author name ("author"), email ("author-mail"), time |
b24642b2 JH |
105 | ("author-time"), and timezone ("author-tz"); similarly |
106 | for committer. | |
b89510f0 | 107 | - the filename in the commit that the line is attributed to. |
b24642b2 JH |
108 | - the first line of the commit log message ("summary"). |
109 | ||
110 | The contents of the actual line is output after the above | |
111 | header, prefixed by a TAB. This is to allow adding more | |
112 | header elements later. | |
113 | ||
ed747dd5 JK |
114 | The porcelain format generally suppresses commit information that has |
115 | already been seen. For example, two lines that are blamed to the same | |
116 | commit will both be shown, but the details for that commit will be shown | |
117 | only once. This is more efficient, but may require more state be kept by | |
118 | the reader. The `--line-porcelain` option can be used to output full | |
119 | commit information for each line, allowing simpler (but less efficient) | |
120 | usage like: | |
121 | ||
122 | # count the number of lines attributed to each author | |
123 | git blame --line-porcelain file | | |
124 | sed -n 's/^author //p' | | |
125 | sort | uniq -c | sort -rn | |
126 | ||
acca687f | 127 | |
23bfbb81 RS |
128 | SPECIFYING RANGES |
129 | ----------------- | |
acca687f | 130 | |
0b444cdb | 131 | Unlike 'git blame' and 'git annotate' in older versions of git, the extent |
b89510f0 | 132 | of the annotation can be limited to both line ranges and revision |
acca687f | 133 | ranges. When you are interested in finding the origin for |
b89510f0 | 134 | lines 40-60 for file `foo`, you can use the `-L` option like so |
42f62db9 JH |
135 | (they mean the same thing -- both ask for 21 lines starting at |
136 | line 40): | |
acca687f JH |
137 | |
138 | git blame -L 40,60 foo | |
42f62db9 | 139 | git blame -L 40,+21 foo |
acca687f | 140 | |
b89510f0 | 141 | Also you can use a regular expression to specify the line range: |
18d5453e JH |
142 | |
143 | git blame -L '/^sub hello {/,/^}$/' foo | |
144 | ||
b89510f0 | 145 | which limits the annotation to the body of the `hello` subroutine. |
18d5453e | 146 | |
b89510f0 | 147 | When you are not interested in changes older than version |
acca687f | 148 | v2.6.18, or changes older than 3 weeks, you can use revision |
0b444cdb | 149 | range specifiers similar to 'git rev-list': |
acca687f JH |
150 | |
151 | git blame v2.6.18.. -- foo | |
152 | git blame --since=3.weeks -- foo | |
153 | ||
154 | When revision range specifiers are used to limit the annotation, | |
155 | lines that have not changed since the range boundary (either the | |
156 | commit v2.6.18 or the most recent commit that is more than 3 | |
157 | weeks old in the above example) are blamed for that range | |
158 | boundary commit. | |
159 | ||
b89510f0 | 160 | A particularly useful way is to see if an added file has lines |
acca687f JH |
161 | created by copy-and-paste from existing files. Sometimes this |
162 | indicates that the developer was being sloppy and did not | |
163 | refactor the code properly. You can first find the commit that | |
164 | introduced the file with: | |
165 | ||
166 | git log --diff-filter=A --pretty=short -- foo | |
167 | ||
168 | and then annotate the change between the commit and its | |
6cf378f0 | 169 | parents, using `commit^!` notation: |
acca687f JH |
170 | |
171 | git blame -C -C -f $commit^! -- foo | |
172 | ||
173 | ||
57e7a0a4 JH |
174 | INCREMENTAL OUTPUT |
175 | ------------------ | |
176 | ||
177 | When called with `--incremental` option, the command outputs the | |
178 | result as it is built. The output generally will talk about | |
179 | lines touched by more recent commits first (i.e. the lines will | |
180 | be annotated out of order) and is meant to be used by | |
181 | interactive viewers. | |
182 | ||
183 | The output format is similar to the Porcelain format, but it | |
184 | does not contain the actual lines from the file that is being | |
185 | annotated. | |
186 | ||
187 | . Each blame entry always starts with a line of: | |
188 | ||
189 | <40-byte hex sha1> <sourceline> <resultline> <num_lines> | |
190 | + | |
191 | Line numbers count from 1. | |
192 | ||
b89510f0 | 193 | . The first time that a commit shows up in the stream, it has various |
57e7a0a4 | 194 | other information about it printed out with a one-word tag at the |
b89510f0 DM |
195 | beginning of each line describing the extra commit information (author, |
196 | email, committer, dates, summary, etc.). | |
57e7a0a4 | 197 | |
b89510f0 | 198 | . Unlike the Porcelain format, the filename information is always |
57e7a0a4 JH |
199 | given and terminates the entry: |
200 | ||
201 | "filename" <whitespace-quoted-filename-goes-here> | |
202 | + | |
b89510f0 | 203 | and thus it is really quite easy to parse for some line- and word-oriented |
57e7a0a4 JH |
204 | parser (which should be quite natural for most scripting languages). |
205 | + | |
206 | [NOTE] | |
207 | For people who do parsing: to make it more robust, just ignore any | |
b89510f0 DM |
208 | lines between the first and last one ("<sha1>" and "filename" lines) |
209 | where you do not recognize the tag words (or care about that particular | |
57e7a0a4 JH |
210 | one) at the beginning of the "extended information" lines. That way, if |
211 | there is ever added information (like the commit encoding or extended | |
b89510f0 | 212 | commit commentary), a blame viewer will not care. |
57e7a0a4 JH |
213 | |
214 | ||
7d48e9e6 MSO |
215 | MAPPING AUTHORS |
216 | --------------- | |
217 | ||
218 | include::mailmap.txt[] | |
219 | ||
220 | ||
8f2b72a9 JF |
221 | SEE ALSO |
222 | -------- | |
5162e697 | 223 | linkgit:git-annotate[1] |
8f2b72a9 | 224 | |
8f2b72a9 JF |
225 | GIT |
226 | --- | |
9e1f0a85 | 227 | Part of the linkgit:git[1] suite |