174d63a1ee4238aac88c76daee6f3f1bead0d6b5
1 The output format from "git-diff-index", "git-diff-tree" and
2 "git-diff-files" are very similar.
4 These commands all compare two sets of things; what is
5 compared differs:
7 git-diff-index <tree-ish>::
8 compares the <tree-ish> and the files on the filesystem.
10 git-diff-index --cached <tree-ish>::
11 compares the <tree-ish> and the index.
13 git-diff-tree [-r] <tree-ish-1> <tree-ish-2> [<pattern>...]::
14 compares the trees named by the two arguments.
16 git-diff-files [<pattern>...]::
17 compares the index and the files on the filesystem.
20 An output line is formatted this way:
22 ------------------------------------------------
23 in-place edit :100644 100644 bcd1234... 0123456... M file0
24 copy-edit :100644 100644 abcd123... 1234567... C68 file1 file2
25 rename-edit :100644 100644 abcd123... 1234567... R86 file1 file3
26 create :000000 100644 0000000... 1234567... A file4
27 delete :100644 000000 1234567... 0000000... D file5
28 unmerged :000000 000000 0000000... 0000000... U file6
29 ------------------------------------------------
31 That is, from the left to the right:
33 . a colon.
34 . mode for "src"; 000000 if creation or unmerged.
35 . a space.
36 . mode for "dst"; 000000 if deletion or unmerged.
37 . a space.
38 . sha1 for "src"; 0\{40\} if creation or unmerged.
39 . a space.
40 . sha1 for "dst"; 0\{40\} if creation, unmerged or "look at work tree".
41 . a space.
42 . status, followed by optional "score" number.
43 . a tab or a NUL when '-z' option is used.
44 . path for "src"
45 . a tab or a NUL when '-z' option is used; only exists for C or R.
46 . path for "dst"; only exists for C or R.
47 . an LF or a NUL when '-z' option is used, to terminate the record.
49 <sha1> is shown as all 0's if a file is new on the filesystem
50 and it is out of sync with the index.
52 Example:
54 ------------------------------------------------
55 :100644 100644 5be4a4...... 000000...... M file.c
56 ------------------------------------------------
58 When `-z` option is not used, TAB, LF, and backslash characters
59 in pathnames are represented as `\t`, `\n`, and `\\`,
60 respectively.
63 Generating patches with -p
64 --------------------------
66 When "git-diff-index", "git-diff-tree", or "git-diff-files" are run
67 with a '-p' option, they do not produce the output described above;
68 instead they produce a patch file.
70 The patch generation can be customized at two levels.
72 1. When the environment variable 'GIT_EXTERNAL_DIFF' is not set,
73 these commands internally invoke "diff" like this:
75 diff -L a/<path> -L b/<path> -pu <old> <new>
76 +
77 For added files, `/dev/null` is used for <old>. For removed
78 files, `/dev/null` is used for <new>
79 +
80 The "diff" formatting options can be customized via the
81 environment variable 'GIT_DIFF_OPTS'. For example, if you
82 prefer context diff:
84 GIT_DIFF_OPTS=-c git-diff-index -p HEAD
87 2. When the environment variable 'GIT_EXTERNAL_DIFF' is set, the
88 program named by it is called, instead of the diff invocation
89 described above.
90 +
91 For a path that is added, removed, or modified,
92 'GIT_EXTERNAL_DIFF' is called with 7 parameters:
94 path old-file old-hex old-mode new-file new-hex new-mode
95 +
96 where:
98 <old|new>-file:: are files GIT_EXTERNAL_DIFF can use to read the
99 contents of <old|new>,
100 <old|new>-hex:: are the 40-hexdigit SHA1 hashes,
101 <old|new>-mode:: are the octal representation of the file modes.
103 +
104 The file parameters can point at the user's working file
105 (e.g. `new-file` in "git-diff-files"), `/dev/null` (e.g. `old-file`
106 when a new file is added), or a temporary file (e.g. `old-file` in the
107 index). 'GIT_EXTERNAL_DIFF' should not worry about unlinking the
108 temporary file --- it is removed when 'GIT_EXTERNAL_DIFF' exits.
110 For a path that is unmerged, 'GIT_EXTERNAL_DIFF' is called with 1
111 parameter, <path>.
114 git specific extension to diff format
115 -------------------------------------
117 What -p option produces is slightly different from the
118 traditional diff format.
120 1. It is preceded with a "git diff" header, that looks like
121 this:
123 diff --git a/file1 b/file2
124 +
125 The `a/` and `b/` filenames are the same unless rename/copy is
126 involved. Especially, even for a creation or a deletion,
127 `/dev/null` is _not_ used in place of `a/` or `b/` filenames.
128 +
129 When rename/copy is involved, `file1` and `file2` show the
130 name of the source file of the rename/copy and the name of
131 the file that rename/copy produces, respectively.
133 2. It is followed by one or more extended header lines:
135 old mode <mode>
136 new mode <mode>
137 deleted file mode <mode>
138 new file mode <mode>
139 copy from <path>
140 copy to <path>
141 rename from <path>
142 rename to <path>
143 similarity index <number>
144 dissimilarity index <number>
145 index <hash>..<hash> <mode>
147 3. TAB, LF, and backslash characters in pathnames are
148 represented as `\t`, `\n`, and `\\`, respectively.
151 combined diff format
152 --------------------
154 git-diff-tree and git-diff-files can take '-c' or '--cc' option
155 to produce 'combined diff', which looks like this:
157 ------------
158 diff --combined describe.c
159 index fabadb8,cc95eb0..4866510
160 --- a/describe.c
161 +++ b/describe.c
162 @@@ -98,20 -98,12 +98,20 @@@
163 return (a_date > b_date) ? -1 : (a_date == b_date) ? 0 : 1;
164 }
166 - static void describe(char *arg)
167 -static void describe(struct commit *cmit, int last_one)
168 ++static void describe(char *arg, int last_one)
169 {
170 + unsigned char sha1[20];
171 + struct commit *cmit;
172 struct commit_list *list;
173 static int initialized = 0;
174 struct commit_name *n;
176 + if (get_sha1(arg, sha1) < 0)
177 + usage(describe_usage);
178 + cmit = lookup_commit_reference(sha1);
179 + if (!cmit)
180 + usage(describe_usage);
181 +
182 if (!initialized) {
183 initialized = 1;
184 for_each_ref(get_name);
185 ------------
187 1. It is preceded with a "git diff" header, that looks like
188 this (when '-c' option is used):
190 diff --combined file
191 +
192 or like this (when '--cc' option is used):
194 diff --c file
196 2. It is followed by one or more extended header lines
197 (this example shows a merge with two parents):
199 index <hash>,<hash>..<hash>
200 mode <mode>,<mode>..<mode>
201 new file mode <mode>
202 deleted file mode <mode>,<mode>
203 +
204 The `mode <mode>,<mode>..<mode>` line appears only if at least one of
205 the <mode> is diferent from the rest. Extended headers with
206 information about detected contents movement (renames and
207 copying detection) are designed to work with diff of two
208 <tree-ish> and are not used by combined diff format.
210 3. It is followed by two-line from-file/to-file header
212 --- a/file
213 +++ b/file
214 +
215 Contrary to two-line header for traditional 'unified' diff
216 format, and similar to filenames in ordinary "diff header",
217 /dev/null is not used for creation or deletion.
219 4. Chunk header format is modified to prevent people from
220 accidentally feeding it to `patch -p1`. Combined diff format
221 was created for review of merge commit changes, and was not
222 meant for apply. The change is similar to the change in the
223 extended 'index' header:
225 @@@ <from-file-range> <from-file-range> <to-file-range> @@@
226 +
227 There are (number of parents + 1) `@` characters in the chunk
228 header for combined diff format.
230 Unlike the traditional 'unified' diff format, which shows two
231 files A and B with a single column that has `-` (minus --
232 appears in A but removed in B), `+` (plus -- missing in A but
233 added to B), or `" "` (space -- unchanged) prefix, this format
234 compares two or more files file1, file2,... with one file X, and
235 shows how X differs from each of fileN. One column for each of
236 fileN is prepended to the output line to note how X's line is
237 different from it.
239 A `-` character in the column N means that the line appears in
240 fileN but it does not appear in the result. A `+` character
241 in the column N means that the line appears in the last file,
242 and fileN does not have that line (in other words, the line was
243 added, from the point of view of that parent).
245 In the above example output, the function signature was changed
246 from both files (hence two `-` removals from both file1 and
247 file2, plus `++` to mean one line that was added does not appear
248 in either file1 nor file2). Also two other lines are the same
249 from file1 but do not appear in file2 (hence prefixed with ` +`).
251 When shown by `git diff-tree -c`, it compares the parents of a
252 merge commit with the merge result (i.e. file1..fileN are the
253 parents). When shown by `git diff-files -c`, it compares the
254 two unresolved merge parents with the working tree file
255 (i.e. file1 is stage 2 aka "our version", file2 is stage 3 aka
256 "their version").