Documentation/git-blame.txt

   1 git-blame(1)
   2 ============
   3
   4 NAME
   5 ----
   6 git-blame - Show what revision and author last modified each line of a file
   7
   8 SYNOPSIS
   9 --------
  10 [verse]
  11 'git blame' [-c] [-b] [-l] [--root] [-t] [-f] [-n] [-s] [-e] [-p] [-w] [--incremental]
  12             [-L <range>] [-S <revs-file>] [-M] [-C] [-C] [-C] [--since=<date>]
  13             [--ignore-rev <rev>] [--ignore-revs-file <file>]
  14             [--progress] [--abbrev=<n>] [<rev> | --contents <file> | --reverse <rev>..<rev>]
  15             [--] <file>
  16
  17 DESCRIPTION
  18 -----------
  19
  20 Annotates each line in the given file with information from the revision which
  21 last modified the line. Optionally, start annotating from the given revision.
  22
  23 When specified one or more times, `-L` restricts annotation to the requested
  24 lines.
  25
  26 The origin of lines is automatically followed across whole-file
  27 renames (currently there is no option to turn the rename-following
  28 off). To follow lines moved from one file to another, or to follow
  29 lines that were copied and pasted from another file, etc., see the
  30 `-C` and `-M` options.
  31
  32 The report does not tell you anything about lines which have been deleted or
  33 replaced; you need to use a tool such as 'git diff' or the "pickaxe"
  34 interface briefly mentioned in the following paragraph.
  35
  36 Apart from supporting file annotation, Git also supports searching the
  37 development history for when a code snippet occurred in a change. This makes it
  38 possible to track when a code snippet was added to a file, moved or copied
  39 between files, and eventually deleted or replaced. It works by searching for
  40 a text string in the diff. A small example of the pickaxe interface
  41 that searches for `blame_usage`:
  42
  43 -----------------------------------------------------------------------------
  44 $ git log --pretty=oneline -S'blame_usage'
  45 5040f17eba15504bad66b14a645bddd9b015ebb7 blame -S <ancestry-file>
  46 ea4c7f9bf69e781dd0cd88d2bccb2bf5cc15c9a7 git-blame: Make the output
  47 -----------------------------------------------------------------------------
  48
  49 OPTIONS
  50 -------
  51 include::blame-options.txt[]
  52
  53 -c::
  54         Use the same output mode as linkgit:git-annotate[1] (Default: off).
  55
  56 --score-debug::
  57         Include debugging information related to the movement of
  58         lines between files (see `-C`) and lines moved within a
  59         file (see `-M`).  The first number listed is the score.
  60         This is the number of alphanumeric characters detected
  61         as having been moved between or within files.  This must be above
  62         a certain threshold for 'git blame' to consider those lines
  63         of code to have been moved.
  64
  65 -f::
  66 --show-name::
  67         Show the filename in the original commit.  By default
  68         the filename is shown if there is any line that came from a
  69         file with a different name, due to rename detection.
  70
  71 -n::
  72 --show-number::
  73         Show the line number in the original commit (Default: off).
  74
  75 -s::
  76         Suppress the author name and timestamp from the output.
  77
  78 -e::
  79 --show-email::
  80         Show the author email instead of author name (Default: off).
  81         This can also be controlled via the `blame.showEmail` config
  82         option.
  83
  84 -w::
  85         Ignore whitespace when comparing the parent's version and
  86         the child's to find where the lines came from.
  87
  88 --abbrev=<n>::
  89         Instead of using the default 7+1 hexadecimal digits as the
  90         abbreviated object name, use <n>+1 digits. Note that 1 column
  91         is used for a caret to mark the boundary commit.
  92
  93
  94 THE PORCELAIN FORMAT
  95 --------------------
  96
  97 In this format, each line is output after a header; the
  98 header at the minimum has the first line which has:
  99
 100 - 40-byte SHA-1 of the commit the line is attributed to;
 101 - the line number of the line in the original file;
 102 - the line number of the line in the final file;
 103 - on a line that starts a group of lines from a different
 104   commit than the previous one, the number of lines in this
 105   group.  On subsequent lines this field is absent.
 106
 107 This header line is followed by the following information
 108 at least once for each commit:
 109
 110 - the author name ("author"), email ("author-mail"), time
 111   ("author-time"), and time zone ("author-tz"); similarly
 112   for committer.
 113 - the filename in the commit that the line is attributed to.
 114 - the first line of the commit log message ("summary").
 115
 116 The contents of the actual line is output after the above
 117 header, prefixed by a TAB. This is to allow adding more
 118 header elements later.
 119
 120 The porcelain format generally suppresses commit information that has
 121 already been seen. For example, two lines that are blamed to the same
 122 commit will both be shown, but the details for that commit will be shown
 123 only once. This is more efficient, but may require more state be kept by
 124 the reader. The `--line-porcelain` option can be used to output full
 125 commit information for each line, allowing simpler (but less efficient)
 126 usage like:
 127
 128         # count the number of lines attributed to each author
 129         git blame --line-porcelain file |
 130         sed -n 's/^author //p' |
 131         sort | uniq -c | sort -rn
 132
 133
 134 SPECIFYING RANGES
 135 -----------------
 136
 137 Unlike 'git blame' and 'git annotate' in older versions of git, the extent
 138 of the annotation can be limited to both line ranges and revision
 139 ranges. The `-L` option, which limits annotation to a range of lines, may be
 140 specified multiple times.
 141
 142 When you are interested in finding the origin for
 143 lines 40-60 for file `foo`, you can use the `-L` option like so
 144 (they mean the same thing -- both ask for 21 lines starting at
 145 line 40):
 146
 147         git blame -L 40,60 foo
 148         git blame -L 40,+21 foo
 149
 150 Also you can use a regular expression to specify the line range:
 151
 152         git blame -L '/^sub hello {/,/^}$/' foo
 153
 154 which limits the annotation to the body of the `hello` subroutine.
 155
 156 When you are not interested in changes older than version
 157 v2.6.18, or changes older than 3 weeks, you can use revision
 158 range specifiers  similar to 'git rev-list':
 159
 160         git blame v2.6.18.. -- foo
 161         git blame --since=3.weeks -- foo
 162
 163 When revision range specifiers are used to limit the annotation,
 164 lines that have not changed since the range boundary (either the
 165 commit v2.6.18 or the most recent commit that is more than 3
 166 weeks old in the above example) are blamed for that range
 167 boundary commit.
 168
 169 A particularly useful way is to see if an added file has lines
 170 created by copy-and-paste from existing files.  Sometimes this
 171 indicates that the developer was being sloppy and did not
 172 refactor the code properly.  You can first find the commit that
 173 introduced the file with:
 174
 175         git log --diff-filter=A --pretty=short -- foo
 176
 177 and then annotate the change between the commit and its
 178 parents, using `commit^!` notation:
 179
 180         git blame -C -C -f $commit^! -- foo
 181
 182
 183 INCREMENTAL OUTPUT
 184 ------------------
 185
 186 When called with `--incremental` option, the command outputs the
 187 result as it is built.  The output generally will talk about
 188 lines touched by more recent commits first (i.e. the lines will
 189 be annotated out of order) and is meant to be used by
 190 interactive viewers.
 191
 192 The output format is similar to the Porcelain format, but it
 193 does not contain the actual lines from the file that is being
 194 annotated.
 195
 196 . Each blame entry always starts with a line of:
 197
 198         <40-byte hex sha1> <sourceline> <resultline> <num_lines>
 199 +
 200 Line numbers count from 1.
 201
 202 . The first time that a commit shows up in the stream, it has various
 203   other information about it printed out with a one-word tag at the
 204   beginning of each line describing the extra commit information (author,
 205   email, committer, dates, summary, etc.).
 206
 207 . Unlike the Porcelain format, the filename information is always
 208   given and terminates the entry:
 209
 210         "filename" <whitespace-quoted-filename-goes-here>
 211 +
 212 and thus it is really quite easy to parse for some line- and word-oriented
 213 parser (which should be quite natural for most scripting languages).
 214 +
 215 [NOTE]
 216 For people who do parsing: to make it more robust, just ignore any
 217 lines between the first and last one ("<sha1>" and "filename" lines)
 218 where you do not recognize the tag words (or care about that particular
 219 one) at the beginning of the "extended information" lines. That way, if
 220 there is ever added information (like the commit encoding or extended
 221 commit commentary), a blame viewer will not care.
 222
 223
 224 MAPPING AUTHORS
 225 ---------------
 226
 227 include::mailmap.txt[]
 228
 229
 230 SEE ALSO
 231 --------
 232 linkgit:git-annotate[1]
 233
 234 GIT
 235 ---
 236 Part of the linkgit:git[1] suite