| blob | 925ff96e2cc901fda38e95fbbe010943a6d9ea65 |
1 '\" t
2 .\" Title: git-diff-index
3 .\" Author: [FIXME: author] [see http://www.docbook.org/tdg5/en/html/author]
4 .\" Generator: DocBook XSL Stylesheets vsnapshot <http://docbook.sf.net/>
5 .\" Date: 2024-02-27
6 .\" Manual: Git Manual
7 .\" Source: Git 2.44.0.53.g0f9d4d28b7
8 .\" Language: English
9 .\"
10 .TH "GIT\-DIFF\-INDEX" "1" "2024\-02\-27" "Git 2\&.44\&.0\&.53\&.g0f9d4d2" "Git Manual"
11 .\" -----------------------------------------------------------------
12 .\" * Define some portability stuff
13 .\" -----------------------------------------------------------------
14 .\" ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
15 .\" http://bugs.debian.org/507673
16 .\" http://lists.gnu.org/archive/html/groff/2009-02/msg00013.html
17 .\" ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
18 .ie \n(.g .ds Aq \(aq
19 .el .ds Aq '
20 .\" -----------------------------------------------------------------
21 .\" * set default formatting
22 .\" -----------------------------------------------------------------
23 .\" disable hyphenation
24 .nh
25 .\" disable justification (adjust text to left margin only)
26 .ad l
27 .\" -----------------------------------------------------------------
28 .\" * MAIN CONTENT STARTS HERE *
29 .\" -----------------------------------------------------------------
30 .SH "NAME"
31 git-diff-index \- Compare a tree to the working tree or index
32 .SH "SYNOPSIS"
33 .sp
34 .nf
35 \fIgit diff\-index\fR [\-m] [\-\-cached] [\-\-merge\-base] [<common\-diff\-options>] <tree\-ish> [<path>\&...]
36 .fi
37 .sp
38 .SH "DESCRIPTION"
39 .sp
40 Compare the content and mode of the blobs found in a tree object with the corresponding tracked files in the working tree, or with the corresponding paths in the index\&. When <path> arguments are present, compare only paths matching those patterns\&. Otherwise all tracked files are compared\&.
41 .SH "OPTIONS"
42 .PP
43 \-p, \-u, \-\-patch
44 .RS 4
45 Generate patch (see
46 the section called \(lqGENERATING PATCH TEXT WITH \-P\(rq)\&.
47 .RE
48 .PP
49 \-s, \-\-no\-patch
50 .RS 4
51 Suppress all output from the diff machinery\&. Useful for commands like
52 \fBgit show\fR
53 that show the patch by default to squelch their output, or to cancel the effect of options like
54 \fB\-\-patch\fR,
55 \fB\-\-stat\fR
56 earlier on the command line in an alias\&.
57 .RE
58 .PP
59 \-U<n>, \-\-unified=<n>
60 .RS 4
61 Generate diffs with <n> lines of context instead of the usual three\&. Implies
62 \fB\-\-patch\fR\&.
63 .RE
64 .PP
65 \-\-output=<file>
66 .RS 4
67 Output to a specific file instead of stdout\&.
68 .RE
69 .PP
70 \-\-output\-indicator\-new=<char>, \-\-output\-indicator\-old=<char>, \-\-output\-indicator\-context=<char>
71 .RS 4
72 Specify the character used to indicate new, old or context lines in the generated patch\&. Normally they are
73 \fI+\fR,
74 \fI\-\fR
75 and \*(Aq \*(Aq respectively\&.
76 .RE
77 .PP
78 \-\-raw
79 .RS 4
80 Generate the diff in raw format\&. This is the default\&.
81 .RE
82 .PP
83 \-\-patch\-with\-raw
84 .RS 4
85 Synonym for
86 \fB\-p \-\-raw\fR\&.
87 .RE
88 .PP
89 \-\-indent\-heuristic
90 .RS 4
91 Enable the heuristic that shifts diff hunk boundaries to make patches easier to read\&. This is the default\&.
92 .RE
93 .PP
94 \-\-no\-indent\-heuristic
95 .RS 4
96 Disable the indent heuristic\&.
97 .RE
98 .PP
99 \-\-minimal
100 .RS 4
101 Spend extra time to make sure the smallest possible diff is produced\&.
102 .RE
103 .PP
104 \-\-patience
105 .RS 4
106 Generate a diff using the "patience diff" algorithm\&.
107 .RE
108 .PP
109 \-\-histogram
110 .RS 4
111 Generate a diff using the "histogram diff" algorithm\&.
112 .RE
113 .PP
114 \-\-anchored=<text>
115 .RS 4
116 Generate a diff using the "anchored diff" algorithm\&.
117 .sp
118 This option may be specified more than once\&.
119 .sp
120 If a line exists in both the source and destination, exists only once, and starts with this text, this algorithm attempts to prevent it from appearing as a deletion or addition in the output\&. It uses the "patience diff" algorithm internally\&.
121 .RE
122 .PP
123 \-\-diff\-algorithm={patience|minimal|histogram|myers}
124 .RS 4
125 Choose a diff algorithm\&. The variants are as follows:
126 .PP
127 \fBdefault\fR, \fBmyers\fR
128 .RS 4
129 The basic greedy diff algorithm\&. Currently, this is the default\&.
130 .RE
131 .PP
132 \fBminimal\fR
133 .RS 4
134 Spend extra time to make sure the smallest possible diff is produced\&.
135 .RE
136 .PP
137 \fBpatience\fR
138 .RS 4
139 Use "patience diff" algorithm when generating patches\&.
140 .RE
141 .PP
142 \fBhistogram\fR
143 .RS 4
144 This algorithm extends the patience algorithm to "support low\-occurrence common elements"\&.
145 .RE
146 .sp
147 For instance, if you configured the
148 \fBdiff\&.algorithm\fR
149 variable to a non\-default value and want to use the default one, then you have to use
150 \fB\-\-diff\-algorithm=default\fR
151 option\&.
152 .RE
153 .PP
154 \-\-stat[=<width>[,<name\-width>[,<count>]]]
155 .RS 4
156 Generate a diffstat\&. By default, as much space as necessary will be used for the filename part, and the rest for the graph part\&. Maximum width defaults to terminal width, or 80 columns if not connected to a terminal, and can be overridden by
157 \fB<width>\fR\&. The width of the filename part can be limited by giving another width
158 \fB<name\-width>\fR
159 after a comma or by setting
160 \fBdiff\&.statNameWidth=<width>\fR\&. The width of the graph part can be limited by using
161 \fB\-\-stat\-graph\-width=<width>\fR
162 or by setting
163 \fBdiff\&.statGraphWidth=<width>\fR\&. Using
164 \fB\-\-stat\fR
165 or
166 \fB\-\-stat\-graph\-width\fR
167 affects all commands generating a stat graph, while setting
168 \fBdiff\&.statNameWidth\fR
169 or
170 \fBdiff\&.statGraphWidth\fR
171 does not affect
172 \fBgit format\-patch\fR\&. By giving a third parameter
173 \fB<count>\fR, you can limit the output to the first
174 \fB<count>\fR
175 lines, followed by
176 \fB\&.\&.\&.\fR
177 if there are more\&.
178 .sp
179 These parameters can also be set individually with
180 \fB\-\-stat\-width=<width>\fR,
181 \fB\-\-stat\-name\-width=<name\-width>\fR
182 and
183 \fB\-\-stat\-count=<count>\fR\&.
184 .RE
185 .PP
186 \-\-compact\-summary
187 .RS 4
188 Output a condensed summary of extended header information such as file creations or deletions ("new" or "gone", optionally "+l" if it\(cqs a symlink) and mode changes ("+x" or "\-x" for adding or removing executable bit respectively) in diffstat\&. The information is put between the filename part and the graph part\&. Implies
189 \fB\-\-stat\fR\&.
190 .RE
191 .PP
192 \-\-numstat
193 .RS 4
194 Similar to
195 \fB\-\-stat\fR, but shows number of added and deleted lines in decimal notation and pathname without abbreviation, to make it more machine friendly\&. For binary files, outputs two
196 \fB\-\fR
197 instead of saying
198 \fB0 0\fR\&.
199 .RE
200 .PP
201 \-\-shortstat
202 .RS 4
203 Output only the last line of the
204 \fB\-\-stat\fR
205 format containing total number of modified files, as well as number of added and deleted lines\&.
206 .RE
207 .PP
208 \-X[<param1,param2,\&...>], \-\-dirstat[=<param1,param2,\&...>]
209 .RS 4
210 Output the distribution of relative amount of changes for each sub\-directory\&. The behavior of
211 \fB\-\-dirstat\fR
212 can be customized by passing it a comma separated list of parameters\&. The defaults are controlled by the
213 \fBdiff\&.dirstat\fR
214 configuration variable (see
215 \fBgit-config\fR(1))\&. The following parameters are available:
216 .PP
217 \fBchanges\fR
218 .RS 4
219 Compute the dirstat numbers by counting the lines that have been removed from the source, or added to the destination\&. This ignores the amount of pure code movements within a file\&. In other words, rearranging lines in a file is not counted as much as other changes\&. This is the default behavior when no parameter is given\&.
220 .RE
221 .PP
222 \fBlines\fR
223 .RS 4
224 Compute the dirstat numbers by doing the regular line\-based diff analysis, and summing the removed/added line counts\&. (For binary files, count 64\-byte chunks instead, since binary files have no natural concept of lines)\&. This is a more expensive
225 \fB\-\-dirstat\fR
226 behavior than the
227 \fBchanges\fR
228 behavior, but it does count rearranged lines within a file as much as other changes\&. The resulting output is consistent with what you get from the other
229 \fB\-\-*stat\fR
230 options\&.
231 .RE
232 .PP
233 \fBfiles\fR
234 .RS 4
235 Compute the dirstat numbers by counting the number of files changed\&. Each changed file counts equally in the dirstat analysis\&. This is the computationally cheapest
236 \fB\-\-dirstat\fR
237 behavior, since it does not have to look at the file contents at all\&.
238 .RE
239 .PP
240 \fBcumulative\fR
241 .RS 4
242 Count changes in a child directory for the parent directory as well\&. Note that when using
243 \fBcumulative\fR, the sum of the percentages reported may exceed 100%\&. The default (non\-cumulative) behavior can be specified with the
244 \fBnoncumulative\fR
245 parameter\&.
246 .RE
247 .PP
248 <limit>
249 .RS 4
250 An integer parameter specifies a cut\-off percent (3% by default)\&. Directories contributing less than this percentage of the changes are not shown in the output\&.
251 .RE
252 .sp
253 Example: The following will count changed files, while ignoring directories with less than 10% of the total amount of changed files, and accumulating child directory counts in the parent directories:
254 \fB\-\-dirstat=files,10,cumulative\fR\&.
255 .RE
256 .PP
257 \-\-cumulative
258 .RS 4
259 Synonym for \-\-dirstat=cumulative
260 .RE
261 .PP
262 \-\-dirstat\-by\-file[=<param1,param2>\&...]
263 .RS 4
264 Synonym for \-\-dirstat=files,<param1>,<param2>\&...
265 .RE
266 .PP
267 \-\-summary
268 .RS 4
269 Output a condensed summary of extended header information such as creations, renames and mode changes\&.
270 .RE
271 .PP
272 \-\-patch\-with\-stat
273 .RS 4
274 Synonym for
275 \fB\-p \-\-stat\fR\&.
276 .RE
277 .PP
278 \-z
279 .RS 4
280 When
281 \fB\-\-raw\fR,
282 \fB\-\-numstat\fR,
283 \fB\-\-name\-only\fR
284 or
285 \fB\-\-name\-status\fR
286 has been given, do not munge pathnames and use NULs as output field terminators\&.
287 .sp
288 Without this option, pathnames with "unusual" characters are quoted as explained for the configuration variable
289 \fBcore\&.quotePath\fR
290 (see
291 \fBgit-config\fR(1))\&.
292 .RE
293 .PP
294 \-\-name\-only
295 .RS 4
296 Show only names of changed files\&. The file names are often encoded in UTF\-8\&. For more information see the discussion about encoding in the
297 \fBgit-log\fR(1)
298 manual page\&.
299 .RE
300 .PP
301 \-\-name\-status
302 .RS 4
303 Show only names and status of changed files\&. See the description of the
304 \fB\-\-diff\-filter\fR
305 option on what the status letters mean\&. Just like
306 \fB\-\-name\-only\fR
307 the file names are often encoded in UTF\-8\&.
308 .RE
309 .PP
310 \-\-submodule[=<format>]
311 .RS 4
312 Specify how differences in submodules are shown\&. When specifying
313 \fB\-\-submodule=short\fR
314 the
315 \fIshort\fR
316 format is used\&. This format just shows the names of the commits at the beginning and end of the range\&. When
317 \fB\-\-submodule\fR
318 or
319 \fB\-\-submodule=log\fR
320 is specified, the
321 \fIlog\fR
322 format is used\&. This format lists the commits in the range like
323 \fBgit-submodule\fR(1)
324 \fBsummary\fR
325 does\&. When
326 \fB\-\-submodule=diff\fR
327 is specified, the
328 \fIdiff\fR
329 format is used\&. This format shows an inline diff of the changes in the submodule contents between the commit range\&. Defaults to
330 \fBdiff\&.submodule\fR
331 or the
332 \fIshort\fR
333 format if the config option is unset\&.
334 .RE
335 .PP
336 \-\-color[=<when>]
337 .RS 4
338 Show colored diff\&.
339 \fB\-\-color\fR
340 (i\&.e\&. without
341 \fI=<when>\fR) is the same as
342 \fB\-\-color=always\fR\&.
343 \fI<when>\fR
344 can be one of
345 \fBalways\fR,
346 \fBnever\fR, or
347 \fBauto\fR\&.
348 .RE
349 .PP
350 \-\-no\-color
351 .RS 4
352 Turn off colored diff\&. It is the same as
353 \fB\-\-color=never\fR\&.
354 .RE
355 .PP
356 \-\-color\-moved[=<mode>]
357 .RS 4
358 Moved lines of code are colored differently\&. The <mode> defaults to
359 \fIno\fR
360 if the option is not given and to
361 \fIzebra\fR
362 if the option with no mode is given\&. The mode must be one of:
363 .PP
364 no
365 .RS 4
366 Moved lines are not highlighted\&.
367 .RE
368 .PP
369 default
370 .RS 4
371 Is a synonym for
372 \fBzebra\fR\&. This may change to a more sensible mode in the future\&.
373 .RE
374 .PP
375 plain
376 .RS 4
377 Any line that is added in one location and was removed in another location will be colored with
378 \fIcolor\&.diff\&.newMoved\fR\&. Similarly
379 \fIcolor\&.diff\&.oldMoved\fR
380 will be used for removed lines that are added somewhere else in the diff\&. This mode picks up any moved line, but it is not very useful in a review to determine if a block of code was moved without permutation\&.
381 .RE
382 .PP
383 blocks
384 .RS 4
385 Blocks of moved text of at least 20 alphanumeric characters are detected greedily\&. The detected blocks are painted using either the
386 \fIcolor\&.diff\&.{old,new}Moved\fR
387 color\&. Adjacent blocks cannot be told apart\&.
388 .RE
389 .PP
390 zebra
391 .RS 4
392 Blocks of moved text are detected as in
393 \fIblocks\fR
394 mode\&. The blocks are painted using either the
395 \fIcolor\&.diff\&.{old,new}Moved\fR
396 color or
397 \fIcolor\&.diff\&.{old,new}MovedAlternative\fR\&. The change between the two colors indicates that a new block was detected\&.
398 .RE
399 .PP
400 dimmed\-zebra
401 .RS 4
402 Similar to
403 \fIzebra\fR, but additional dimming of uninteresting parts of moved code is performed\&. The bordering lines of two adjacent blocks are considered interesting, the rest is uninteresting\&.
404 \fBdimmed_zebra\fR
405 is a deprecated synonym\&.
406 .RE
407 .RE
408 .PP
409 \-\-no\-color\-moved
410 .RS 4
411 Turn off move detection\&. This can be used to override configuration settings\&. It is the same as
412 \fB\-\-color\-moved=no\fR\&.
413 .RE
414 .PP
415 \-\-color\-moved\-ws=<modes>
416 .RS 4
417 This configures how whitespace is ignored when performing the move detection for
418 \fB\-\-color\-moved\fR\&. These modes can be given as a comma separated list:
419 .PP
420 no
421 .RS 4
422 Do not ignore whitespace when performing move detection\&.
423 .RE
424 .PP
425 ignore\-space\-at\-eol
426 .RS 4
427 Ignore changes in whitespace at EOL\&.
428 .RE
429 .PP
430 ignore\-space\-change
431 .RS 4
432 Ignore changes in amount of whitespace\&. This ignores whitespace at line end, and considers all other sequences of one or more whitespace characters to be equivalent\&.
433 .RE
434 .PP
435 ignore\-all\-space
436 .RS 4
437 Ignore whitespace when comparing lines\&. This ignores differences even if one line has whitespace where the other line has none\&.
438 .RE
439 .PP
440 allow\-indentation\-change
441 .RS 4
442 Initially ignore any whitespace in the move detection, then group the moved code blocks only into a block if the change in whitespace is the same per line\&. This is incompatible with the other modes\&.
443 .RE
444 .RE
445 .PP
446 \-\-no\-color\-moved\-ws
447 .RS 4
448 Do not ignore whitespace when performing move detection\&. This can be used to override configuration settings\&. It is the same as
449 \fB\-\-color\-moved\-ws=no\fR\&.
450 .RE
451 .PP
452 \-\-word\-diff[=<mode>]
453 .RS 4
454 Show a word diff, using the <mode> to delimit changed words\&. By default, words are delimited by whitespace; see
455 \fB\-\-word\-diff\-regex\fR
456 below\&. The <mode> defaults to
457 \fIplain\fR, and must be one of:
458 .PP
459 color
460 .RS 4
461 Highlight changed words using only colors\&. Implies
462 \fB\-\-color\fR\&.
463 .RE
464 .PP
465 plain
466 .RS 4
467 Show words as
468 \fB[\-removed\-]\fR
469 and
470 \fB{+added+}\fR\&. Makes no attempts to escape the delimiters if they appear in the input, so the output may be ambiguous\&.
471 .RE
472 .PP
473 porcelain
474 .RS 4
475 Use a special line\-based format intended for script consumption\&. Added/removed/unchanged runs are printed in the usual unified diff format, starting with a
476 \fB+\fR/\fB\-\fR/` ` character at the beginning of the line and extending to the end of the line\&. Newlines in the input are represented by a tilde
477 \fB~\fR
478 on a line of its own\&.
479 .RE
480 .PP
481 none
482 .RS 4
483 Disable word diff again\&.
484 .RE
485 .sp
486 Note that despite the name of the first mode, color is used to highlight the changed parts in all modes if enabled\&.
487 .RE
488 .PP
489 \-\-word\-diff\-regex=<regex>
490 .RS 4
491 Use <regex> to decide what a word is, instead of considering runs of non\-whitespace to be a word\&. Also implies
492 \fB\-\-word\-diff\fR
493 unless it was already enabled\&.
494 .sp
495 Every non\-overlapping match of the <regex> is considered a word\&. Anything between these matches is considered whitespace and ignored(!) for the purposes of finding differences\&. You may want to append
496 \fB|[^[:space:]]\fR
497 to your regular expression to make sure that it matches all non\-whitespace characters\&. A match that contains a newline is silently truncated(!) at the newline\&.
498 .sp
499 For example,
500 \fB\-\-word\-diff\-regex=\&.\fR
501 will treat each character as a word and, correspondingly, show differences character by character\&.
502 .sp
503 The regex can also be set via a diff driver or configuration option, see
504 \fBgitattributes\fR(5)
505 or
506 \fBgit-config\fR(1)\&. Giving it explicitly overrides any diff driver or configuration setting\&. Diff drivers override configuration settings\&.
507 .RE
508 .PP
509 \-\-color\-words[=<regex>]
510 .RS 4
511 Equivalent to
512 \fB\-\-word\-diff=color\fR
513 plus (if a regex was specified)
514 \fB\-\-word\-diff\-regex=<regex>\fR\&.
515 .RE
516 .PP
517 \-\-no\-renames
518 .RS 4
519 Turn off rename detection, even when the configuration file gives the default to do so\&.
520 .RE
521 .PP
522 \-\-[no\-]rename\-empty
523 .RS 4
524 Whether to use empty blobs as rename source\&.
525 .RE
526 .PP
527 \-\-check
528 .RS 4
529 Warn if changes introduce conflict markers or whitespace errors\&. What are considered whitespace errors is controlled by
530 \fBcore\&.whitespace\fR
531 configuration\&. By default, trailing whitespaces (including lines that consist solely of whitespaces) and a space character that is immediately followed by a tab character inside the initial indent of the line are considered whitespace errors\&. Exits with non\-zero status if problems are found\&. Not compatible with \-\-exit\-code\&.
532 .RE
533 .PP
534 \-\-ws\-error\-highlight=<kind>
535 .RS 4
536 Highlight whitespace errors in the
537 \fBcontext\fR,
538 \fBold\fR
539 or
540 \fBnew\fR
541 lines of the diff\&. Multiple values are separated by comma,
542 \fBnone\fR
543 resets previous values,
544 \fBdefault\fR
545 reset the list to
546 \fBnew\fR
547 and
548 \fBall\fR
549 is a shorthand for
550 \fBold,new,context\fR\&. When this option is not given, and the configuration variable
551 \fBdiff\&.wsErrorHighlight\fR
552 is not set, only whitespace errors in
553 \fBnew\fR
554 lines are highlighted\&. The whitespace errors are colored with
555 \fBcolor\&.diff\&.whitespace\fR\&.
556 .RE
557 .PP
558 \-\-full\-index
559 .RS 4
560 Instead of the first handful of characters, show the full pre\- and post\-image blob object names on the "index" line when generating patch format output\&.
561 .RE
562 .PP
563 \-\-binary
564 .RS 4
565 In addition to
566 \fB\-\-full\-index\fR, output a binary diff that can be applied with
567 \fBgit\-apply\fR\&. Implies
568 \fB\-\-patch\fR\&.
569 .RE
570 .PP
571 \-\-abbrev[=<n>]
572 .RS 4
573 Instead of showing the full 40\-byte hexadecimal object name in diff\-raw format output and diff\-tree header lines, show the shortest prefix that is at least
574 \fI<n>\fR
575 hexdigits long that uniquely refers the object\&. In diff\-patch output format,
576 \fB\-\-full\-index\fR
577 takes higher precedence, i\&.e\&. if
578 \fB\-\-full\-index\fR
579 is specified, full blob names will be shown regardless of
580 \fB\-\-abbrev\fR\&. Non default number of digits can be specified with
581 \fB\-\-abbrev=<n>\fR\&.
582 .RE
583 .PP
584 \-B[<n>][/<m>], \-\-break\-rewrites[=[<n>][/<m>]]
585 .RS 4
586 Break complete rewrite changes into pairs of delete and create\&. This serves two purposes:
587 .sp
588 It affects the way a change that amounts to a total rewrite of a file not as a series of deletion and insertion mixed together with a very few lines that happen to match textually as the context, but as a single deletion of everything old followed by a single insertion of everything new, and the number
589 \fBm\fR
590 controls this aspect of the \-B option (defaults to 60%)\&.
591 \fB\-B/70%\fR
592 specifies that less than 30% of the original should remain in the result for Git to consider it a total rewrite (i\&.e\&. otherwise the resulting patch will be a series of deletion and insertion mixed together with context lines)\&.
593 .sp
594 When used with \-M, a totally\-rewritten file is also considered as the source of a rename (usually \-M only considers a file that disappeared as the source of a rename), and the number
595 \fBn\fR
596 controls this aspect of the \-B option (defaults to 50%)\&.
597 \fB\-B20%\fR
598 specifies that a change with addition and deletion compared to 20% or more of the file\(cqs size are eligible for being picked up as a possible source of a rename to another file\&.
599 .RE
600 .PP
601 \-M[<n>], \-\-find\-renames[=<n>]
602 .RS 4
603 Detect renames\&. If
604 \fBn\fR
605 is specified, it is a threshold on the similarity index (i\&.e\&. amount of addition/deletions compared to the file\(cqs size)\&. For example,
606 \fB\-M90%\fR
607 means Git should consider a delete/add pair to be a rename if more than 90% of the file hasn\(cqt changed\&. Without a
608 \fB%\fR
609 sign, the number is to be read as a fraction, with a decimal point before it\&. I\&.e\&.,
610 \fB\-M5\fR
611 becomes 0\&.5, and is thus the same as
612 \fB\-M50%\fR\&. Similarly,
613 \fB\-M05\fR
614 is the same as
615 \fB\-M5%\fR\&. To limit detection to exact renames, use
616 \fB\-M100%\fR\&. The default similarity index is 50%\&.
617 .RE
618 .PP
619 \-C[<n>], \-\-find\-copies[=<n>]
620 .RS 4
621 Detect copies as well as renames\&. See also
622 \fB\-\-find\-copies\-harder\fR\&. If
623 \fBn\fR
624 is specified, it has the same meaning as for
625 \fB\-M<n>\fR\&.
626 .RE
627 .PP
628 \-\-find\-copies\-harder
629 .RS 4
630 For performance reasons, by default,
631 \fB\-C\fR
632 option finds copies only if the original file of the copy was modified in the same changeset\&. This flag makes the command inspect unmodified files as candidates for the source of copy\&. This is a very expensive operation for large projects, so use it with caution\&. Giving more than one
633 \fB\-C\fR
634 option has the same effect\&.
635 .RE
636 .PP
637 \-D, \-\-irreversible\-delete
638 .RS 4
639 Omit the preimage for deletes, i\&.e\&. print only the header but not the diff between the preimage and
640 \fB/dev/null\fR\&. The resulting patch is not meant to be applied with
641 \fBpatch\fR
642 or
643 \fBgit apply\fR; this is solely for people who want to just concentrate on reviewing the text after the change\&. In addition, the output obviously lacks enough information to apply such a patch in reverse, even manually, hence the name of the option\&.
644 .sp
645 When used together with
646 \fB\-B\fR, omit also the preimage in the deletion part of a delete/create pair\&.
647 .RE
648 .PP
649 \-l<num>
650 .RS 4
651 The
652 \fB\-M\fR
653 and
654 \fB\-C\fR
655 options involve some preliminary steps that can detect subsets of renames/copies cheaply, followed by an exhaustive fallback portion that compares all remaining unpaired destinations to all relevant sources\&. (For renames, only remaining unpaired sources are relevant; for copies, all original sources are relevant\&.) For N sources and destinations, this exhaustive check is O(N^2)\&. This option prevents the exhaustive portion of rename/copy detection from running if the number of source/destination files involved exceeds the specified number\&. Defaults to diff\&.renameLimit\&. Note that a value of 0 is treated as unlimited\&.
656 .RE
657 .PP
658 \-\-diff\-filter=[(A|C|D|M|R|T|U|X|B)\&...[*]]
659 .RS 4
660 Select only files that are Added (\fBA\fR), Copied (\fBC\fR), Deleted (\fBD\fR), Modified (\fBM\fR), Renamed (\fBR\fR), have their type (i\&.e\&. regular file, symlink, submodule, \&...) changed (\fBT\fR), are Unmerged (\fBU\fR), are Unknown (\fBX\fR), or have had their pairing Broken (\fBB\fR)\&. Any combination of the filter characters (including none) can be used\&. When
661 \fB*\fR
662 (All\-or\-none) is added to the combination, all paths are selected if there is any file that matches other criteria in the comparison; if there is no file that matches other criteria, nothing is selected\&.
663 .sp
664 Also, these upper\-case letters can be downcased to exclude\&. E\&.g\&.
665 \fB\-\-diff\-filter=ad\fR
666 excludes added and deleted paths\&.
667 .sp
668 Note that not all diffs can feature all types\&. For instance, copied and renamed entries cannot appear if detection for those types is disabled\&.
669 .RE
670 .PP
671 \-S<string>
672 .RS 4
673 Look for differences that change the number of occurrences of the specified string (i\&.e\&. addition/deletion) in a file\&. Intended for the scripter\(cqs use\&.
674 .sp
675 It is useful when you\(cqre looking for an exact block of code (like a struct), and want to know the history of that block since it first came into being: use the feature iteratively to feed the interesting block in the preimage back into
676 \fB\-S\fR, and keep going until you get the very first version of the block\&.
677 .sp
678 Binary files are searched as well\&.
679 .RE
680 .PP
681 \-G<regex>
682 .RS 4
683 Look for differences whose patch text contains added/removed lines that match <regex>\&.
684 .sp
685 To illustrate the difference between
686 \fB\-S<regex> \-\-pickaxe\-regex\fR
687 and
688 \fB\-G<regex>\fR, consider a commit with the following diff in the same file:
689 .sp
690 .if n \{\
691 .RS 4
692 .\}
693 .nf
694 + return frotz(nitfol, two\->ptr, 1, 0);
695 \&.\&.\&.
696 \- hit = frotz(nitfol, mf2\&.ptr, 1, 0);
697 .fi
698 .if n \{\
699 .RE
700 .\}
701 .sp
702 While
703 \fBgit log \-G"frotz\e(nitfol"\fR
704 will show this commit,
705 \fBgit log \-S"frotz\e(nitfol" \-\-pickaxe\-regex\fR
706 will not (because the number of occurrences of that string did not change)\&.
707 .sp
708 Unless
709 \fB\-\-text\fR
710 is supplied patches of binary files without a textconv filter will be ignored\&.
711 .sp
712 See the
713 \fIpickaxe\fR
714 entry in
715 \fBgitdiffcore\fR(7)
716 for more information\&.
717 .RE
718 .PP
719 \-\-find\-object=<object\-id>
720 .RS 4
721 Look for differences that change the number of occurrences of the specified object\&. Similar to
722 \fB\-S\fR, just the argument is different in that it doesn\(cqt search for a specific string but for a specific object id\&.
723 .sp
724 The object can be a blob or a submodule commit\&. It implies the
725 \fB\-t\fR
726 option in
727 \fBgit\-log\fR
728 to also find trees\&.
729 .RE
730 .PP
731 \-\-pickaxe\-all
732 .RS 4
733 When
734 \fB\-S\fR
735 or
736 \fB\-G\fR
737 finds a change, show all the changes in that changeset, not just the files that contain the change in <string>\&.
738 .RE
739 .PP
740 \-\-pickaxe\-regex
741 .RS 4
742 Treat the <string> given to
743 \fB\-S\fR
744 as an extended POSIX regular expression to match\&.
745 .RE
746 .PP
747 \-O<orderfile>
748 .RS 4
749 Control the order in which files appear in the output\&. This overrides the
750 \fBdiff\&.orderFile\fR
751 configuration variable (see
752 \fBgit-config\fR(1))\&. To cancel
753 \fBdiff\&.orderFile\fR, use
754 \fB\-O/dev/null\fR\&.
755 .sp
756 The output order is determined by the order of glob patterns in <orderfile>\&. All files with pathnames that match the first pattern are output first, all files with pathnames that match the second pattern (but not the first) are output next, and so on\&. All files with pathnames that do not match any pattern are output last, as if there was an implicit match\-all pattern at the end of the file\&. If multiple pathnames have the same rank (they match the same pattern but no earlier patterns), their output order relative to each other is the normal order\&.
757 .sp
758 <orderfile> is parsed as follows:
759 .sp
760 .RS 4
761 .ie n \{\
762 \h'-04'\(bu\h'+03'\c
763 .\}
764 .el \{\
765 .sp -1
766 .IP \(bu 2.3
767 .\}
768 Blank lines are ignored, so they can be used as separators for readability\&.
769 .RE
770 .sp
771 .RS 4
772 .ie n \{\
773 \h'-04'\(bu\h'+03'\c
774 .\}
775 .el \{\
776 .sp -1
777 .IP \(bu 2.3
778 .\}
779 Lines starting with a hash ("\fB#\fR") are ignored, so they can be used for comments\&. Add a backslash ("\fB\e\fR") to the beginning of the pattern if it starts with a hash\&.
780 .RE
781 .sp
782 .RS 4
783 .ie n \{\
784 \h'-04'\(bu\h'+03'\c
785 .\}
786 .el \{\
787 .sp -1
788 .IP \(bu 2.3
789 .\}
790 Each other line contains a single pattern\&.
791 .RE
792 .sp
793 Patterns have the same syntax and semantics as patterns used for fnmatch(3) without the FNM_PATHNAME flag, except a pathname also matches a pattern if removing any number of the final pathname components matches the pattern\&. For example, the pattern "\fBfoo*bar\fR" matches "\fBfooasdfbar\fR" and "\fBfoo/bar/baz/asdf\fR" but not "\fBfoobarx\fR"\&.
794 .RE
795 .PP
796 \-\-skip\-to=<file>, \-\-rotate\-to=<file>
797 .RS 4
798 Discard the files before the named <file> from the output (i\&.e\&.
799 \fIskip to\fR), or move them to the end of the output (i\&.e\&.
800 \fIrotate to\fR)\&. These options were invented primarily for the use of the
801 \fBgit difftool\fR
802 command, and may not be very useful otherwise\&.
803 .RE
804 .PP
805 \-R
806 .RS 4
807 Swap two inputs; that is, show differences from index or on\-disk file to tree contents\&.
808 .RE
809 .PP
810 \-\-relative[=<path>], \-\-no\-relative
811 .RS 4
812 When run from a subdirectory of the project, it can be told to exclude changes outside the directory and show pathnames relative to it with this option\&. When you are not in a subdirectory (e\&.g\&. in a bare repository), you can name which subdirectory to make the output relative to by giving a <path> as an argument\&.
813 \fB\-\-no\-relative\fR
814 can be used to countermand both
815 \fBdiff\&.relative\fR
816 config option and previous
817 \fB\-\-relative\fR\&.
818 .RE
819 .PP
820 \-a, \-\-text
821 .RS 4
822 Treat all files as text\&.
823 .RE
824 .PP
825 \-\-ignore\-cr\-at\-eol
826 .RS 4
827 Ignore carriage\-return at the end of line when doing a comparison\&.
828 .RE
829 .PP
830 \-\-ignore\-space\-at\-eol
831 .RS 4
832 Ignore changes in whitespace at EOL\&.
833 .RE
834 .PP
835 \-b, \-\-ignore\-space\-change
836 .RS 4
837 Ignore changes in amount of whitespace\&. This ignores whitespace at line end, and considers all other sequences of one or more whitespace characters to be equivalent\&.
838 .RE
839 .PP
840 \-w, \-\-ignore\-all\-space
841 .RS 4
842 Ignore whitespace when comparing lines\&. This ignores differences even if one line has whitespace where the other line has none\&.
843 .RE
844 .PP
845 \-\-ignore\-blank\-lines
846 .RS 4
847 Ignore changes whose lines are all blank\&.
848 .RE
849 .PP
850 \-I<regex>, \-\-ignore\-matching\-lines=<regex>
851 .RS 4
852 Ignore changes whose all lines match <regex>\&. This option may be specified more than once\&.
853 .RE
854 .PP
855 \-\-inter\-hunk\-context=<lines>
856 .RS 4
857 Show the context between diff hunks, up to the specified number of lines, thereby fusing hunks that are close to each other\&. Defaults to
858 \fBdiff\&.interHunkContext\fR
859 or 0 if the config option is unset\&.
860 .RE
861 .PP
862 \-W, \-\-function\-context
863 .RS 4
864 Show whole function as context lines for each change\&. The function names are determined in the same way as
865 \fBgit diff\fR
866 works out patch hunk headers (see
867 \fIDefining a custom hunk\-header\fR
868 in
869 \fBgitattributes\fR(5))\&.
870 .RE
871 .PP
872 \-\-exit\-code
873 .RS 4
874 Make the program exit with codes similar to diff(1)\&. That is, it exits with 1 if there were differences and 0 means no differences\&.
875 .RE
876 .PP
877 \-\-quiet
878 .RS 4
879 Disable all output of the program\&. Implies
880 \fB\-\-exit\-code\fR\&.
881 .RE
882 .PP
883 \-\-ext\-diff
884 .RS 4
885 Allow an external diff helper to be executed\&. If you set an external diff driver with
886 \fBgitattributes\fR(5), you need to use this option with
887 \fBgit-log\fR(1)
888 and friends\&.
889 .RE
890 .PP
891 \-\-no\-ext\-diff
892 .RS 4
893 Disallow external diff drivers\&.
894 .RE
895 .PP
896 \-\-textconv, \-\-no\-textconv
897 .RS 4
898 Allow (or disallow) external text conversion filters to be run when comparing binary files\&. See
899 \fBgitattributes\fR(5)
900 for details\&. Because textconv filters are typically a one\-way conversion, the resulting diff is suitable for human consumption, but cannot be applied\&. For this reason, textconv filters are enabled by default only for
901 \fBgit-diff\fR(1)
902 and
903 \fBgit-log\fR(1), but not for
904 \fBgit-format-patch\fR(1)
905 or diff plumbing commands\&.
906 .RE
907 .PP
908 \-\-ignore\-submodules[=<when>]
909 .RS 4
910 Ignore changes to submodules in the diff generation\&. <when> can be either "none", "untracked", "dirty" or "all", which is the default\&. Using "none" will consider the submodule modified when it either contains untracked or modified files or its HEAD differs from the commit recorded in the superproject and can be used to override any settings of the
911 \fIignore\fR
912 option in
913 \fBgit-config\fR(1)
914 or
915 \fBgitmodules\fR(5)\&. When "untracked" is used submodules are not considered dirty when they only contain untracked content (but they are still scanned for modified content)\&. Using "dirty" ignores all changes to the work tree of submodules, only changes to the commits stored in the superproject are shown (this was the behavior until 1\&.7\&.0)\&. Using "all" hides all changes to submodules\&.
916 .RE
917 .PP
918 \-\-src\-prefix=<prefix>
919 .RS 4
920 Show the given source prefix instead of "a/"\&.
921 .RE
922 .PP
923 \-\-dst\-prefix=<prefix>
924 .RS 4
925 Show the given destination prefix instead of "b/"\&.
926 .RE
927 .PP
928 \-\-no\-prefix
929 .RS 4
930 Do not show any source or destination prefix\&.
931 .RE
932 .PP
933 \-\-default\-prefix
934 .RS 4
935 Use the default source and destination prefixes ("a/" and "b/")\&. This is usually the default already, but may be used to override config such as
936 \fBdiff\&.noprefix\fR\&.
937 .RE
938 .PP
939 \-\-line\-prefix=<prefix>
940 .RS 4
941 Prepend an additional prefix to every line of output\&.
942 .RE
943 .PP
944 \-\-ita\-invisible\-in\-index
945 .RS 4
946 By default entries added by "git add \-N" appear as an existing empty file in "git diff" and a new file in "git diff \-\-cached"\&. This option makes the entry appear as a new file in "git diff" and non\-existent in "git diff \-\-cached"\&. This option could be reverted with
947 \fB\-\-ita\-visible\-in\-index\fR\&. Both options are experimental and could be removed in future\&.
948 .RE
949 .sp
950 For more detailed explanation on these common options, see also \fBgitdiffcore\fR(7)\&.
951 .PP
952 <tree\-ish>
953 .RS 4
954 The id of a tree object to diff against\&.
955 .RE
956 .PP
957 \-\-cached
958 .RS 4
959 Do not consider the on\-disk file at all\&.
960 .RE
961 .PP
962 \-\-merge\-base
963 .RS 4
964 Instead of comparing <tree\-ish> directly, use the merge base between <tree\-ish> and HEAD instead\&. <tree\-ish> must be a commit\&.
965 .RE
966 .PP
967 \-m
968 .RS 4
969 By default, files recorded in the index but not checked out are reported as deleted\&. This flag makes
970 \fIgit diff\-index\fR
971 say that all non\-checked\-out files are up to date\&.
972 .RE
973 .SH "RAW OUTPUT FORMAT"
974 .sp
975 The raw output format from "git\-diff\-index", "git\-diff\-tree", "git\-diff\-files" and "git diff \-\-raw" are very similar\&.
976 .sp
977 These commands all compare two sets of things; what is compared differs:
978 .PP
979 git\-diff\-index <tree\-ish>
980 .RS 4
981 compares the <tree\-ish> and the files on the filesystem\&.
982 .RE
983 .PP
984 git\-diff\-index \-\-cached <tree\-ish>
985 .RS 4
986 compares the <tree\-ish> and the index\&.
987 .RE
988 .PP
989 git\-diff\-tree [\-r] <tree\-ish\-1> <tree\-ish\-2> [<pattern>\&...]
990 .RS 4
991 compares the trees named by the two arguments\&.
992 .RE
993 .PP
994 git\-diff\-files [<pattern>\&...]
995 .RS 4
996 compares the index and the files on the filesystem\&.
997 .RE
998 .sp
999 The "git\-diff\-tree" command begins its output by printing the hash of what is being compared\&. After that, all the commands print one output line per changed file\&.
1000 .sp
1001 An output line is formatted this way:
1002 .sp
1003 .if n \{\
1004 .RS 4
1005 .\}
1006 .nf
1007 in\-place edit :100644 100644 bcd1234 0123456 M file0
1008 copy\-edit :100644 100644 abcd123 1234567 C68 file1 file2
1009 rename\-edit :100644 100644 abcd123 1234567 R86 file1 file3
1010 create :000000 100644 0000000 1234567 A file4
1011 delete :100644 000000 1234567 0000000 D file5
1012 unmerged :000000 000000 0000000 0000000 U file6
1013 .fi
1014 .if n \{\
1015 .RE
1016 .\}
1017 .sp
1018 .sp
1019 That is, from the left to the right:
1020 .sp
1021 .RS 4
1022 .ie n \{\
1023 \h'-04' 1.\h'+01'\c
1024 .\}
1025 .el \{\
1026 .sp -1
1027 .IP " 1." 4.2
1028 .\}
1029 a colon\&.
1030 .RE
1031 .sp
1032 .RS 4
1033 .ie n \{\
1034 \h'-04' 2.\h'+01'\c
1035 .\}
1036 .el \{\
1037 .sp -1
1038 .IP " 2." 4.2
1039 .\}
1040 mode for "src"; 000000 if creation or unmerged\&.
1041 .RE
1042 .sp
1043 .RS 4
1044 .ie n \{\
1045 \h'-04' 3.\h'+01'\c
1046 .\}
1047 .el \{\
1048 .sp -1
1049 .IP " 3." 4.2
1050 .\}
1051 a space\&.
1052 .RE
1053 .sp
1054 .RS 4
1055 .ie n \{\
1056 \h'-04' 4.\h'+01'\c
1057 .\}
1058 .el \{\
1059 .sp -1
1060 .IP " 4." 4.2
1061 .\}
1062 mode for "dst"; 000000 if deletion or unmerged\&.
1063 .RE
1064 .sp
1065 .RS 4
1066 .ie n \{\
1067 \h'-04' 5.\h'+01'\c
1068 .\}
1069 .el \{\
1070 .sp -1
1071 .IP " 5." 4.2
1072 .\}
1073 a space\&.
1074 .RE
1075 .sp
1076 .RS 4
1077 .ie n \{\
1078 \h'-04' 6.\h'+01'\c
1079 .\}
1080 .el \{\
1081 .sp -1
1082 .IP " 6." 4.2
1083 .\}
1084 sha1 for "src"; 0{40} if creation or unmerged\&.
1085 .RE
1086 .sp
1087 .RS 4
1088 .ie n \{\
1089 \h'-04' 7.\h'+01'\c
1090 .\}
1091 .el \{\
1092 .sp -1
1093 .IP " 7." 4.2
1094 .\}
1095 a space\&.
1096 .RE
1097 .sp
1098 .RS 4
1099 .ie n \{\
1100 \h'-04' 8.\h'+01'\c
1101 .\}
1102 .el \{\
1103 .sp -1
1104 .IP " 8." 4.2
1105 .\}
1106 sha1 for "dst"; 0{40} if deletion, unmerged or "work tree out of sync with the index"\&.
1107 .RE
1108 .sp
1109 .RS 4
1110 .ie n \{\
1111 \h'-04' 9.\h'+01'\c
1112 .\}
1113 .el \{\
1114 .sp -1
1115 .IP " 9." 4.2
1116 .\}
1117 a space\&.
1118 .RE
1119 .sp
1120 .RS 4
1121 .ie n \{\
1122 \h'-04'10.\h'+01'\c
1123 .\}
1124 .el \{\
1125 .sp -1
1126 .IP "10." 4.2
1127 .\}
1128 status, followed by optional "score" number\&.
1129 .RE
1130 .sp
1131 .RS 4
1132 .ie n \{\
1133 \h'-04'11.\h'+01'\c
1134 .\}
1135 .el \{\
1136 .sp -1
1137 .IP "11." 4.2
1138 .\}
1139 a tab or a NUL when
1140 \fB\-z\fR
1141 option is used\&.
1142 .RE
1143 .sp
1144 .RS 4
1145 .ie n \{\
1146 \h'-04'12.\h'+01'\c
1147 .\}
1148 .el \{\
1149 .sp -1
1150 .IP "12." 4.2
1151 .\}
1152 path for "src"
1153 .RE
1154 .sp
1155 .RS 4
1156 .ie n \{\
1157 \h'-04'13.\h'+01'\c
1158 .\}
1159 .el \{\
1160 .sp -1
1161 .IP "13." 4.2
1162 .\}
1163 a tab or a NUL when
1164 \fB\-z\fR
1165 option is used; only exists for C or R\&.
1166 .RE
1167 .sp
1168 .RS 4
1169 .ie n \{\
1170 \h'-04'14.\h'+01'\c
1171 .\}
1172 .el \{\
1173 .sp -1
1174 .IP "14." 4.2
1175 .\}
1176 path for "dst"; only exists for C or R\&.
1177 .RE
1178 .sp
1179 .RS 4
1180 .ie n \{\
1181 \h'-04'15.\h'+01'\c
1182 .\}
1183 .el \{\
1184 .sp -1
1185 .IP "15." 4.2
1186 .\}
1187 an LF or a NUL when
1188 \fB\-z\fR
1189 option is used, to terminate the record\&.
1190 .RE
1191 .sp
1192 Possible status letters are:
1193 .sp
1194 .RS 4
1195 .ie n \{\
1196 \h'-04'\(bu\h'+03'\c
1197 .\}
1198 .el \{\
1199 .sp -1
1200 .IP \(bu 2.3
1201 .\}
1202 A: addition of a file
1203 .RE
1204 .sp
1205 .RS 4
1206 .ie n \{\
1207 \h'-04'\(bu\h'+03'\c
1208 .\}
1209 .el \{\
1210 .sp -1
1211 .IP \(bu 2.3
1212 .\}
1213 C: copy of a file into a new one
1214 .RE
1215 .sp
1216 .RS 4
1217 .ie n \{\
1218 \h'-04'\(bu\h'+03'\c
1219 .\}
1220 .el \{\
1221 .sp -1
1222 .IP \(bu 2.3
1223 .\}
1224 D: deletion of a file
1225 .RE
1226 .sp
1227 .RS 4
1228 .ie n \{\
1229 \h'-04'\(bu\h'+03'\c
1230 .\}
1231 .el \{\
1232 .sp -1
1233 .IP \(bu 2.3
1234 .\}
1235 M: modification of the contents or mode of a file
1236 .RE
1237 .sp
1238 .RS 4
1239 .ie n \{\
1240 \h'-04'\(bu\h'+03'\c
1241 .\}
1242 .el \{\
1243 .sp -1
1244 .IP \(bu 2.3
1245 .\}
1246 R: renaming of a file
1247 .RE
1248 .sp
1249 .RS 4
1250 .ie n \{\
1251 \h'-04'\(bu\h'+03'\c
1252 .\}
1253 .el \{\
1254 .sp -1
1255 .IP \(bu 2.3
1256 .\}
1257 T: change in the type of the file (regular file, symbolic link or submodule)
1258 .RE
1259 .sp
1260 .RS 4
1261 .ie n \{\
1262 \h'-04'\(bu\h'+03'\c
1263 .\}
1264 .el \{\
1265 .sp -1
1266 .IP \(bu 2.3
1267 .\}
1268 U: file is unmerged (you must complete the merge before it can be committed)
1269 .RE
1270 .sp
1271 .RS 4
1272 .ie n \{\
1273 \h'-04'\(bu\h'+03'\c
1274 .\}
1275 .el \{\
1276 .sp -1
1277 .IP \(bu 2.3
1278 .\}
1279 X: "unknown" change type (most probably a bug, please report it)
1280 .RE
1281 .sp
1282 Status letters C and R are always followed by a score (denoting the percentage of similarity between the source and target of the move or copy)\&. Status letter M may be followed by a score (denoting the percentage of dissimilarity) for file rewrites\&.
1283 .sp
1284 The sha1 for "dst" is shown as all 0\(cqs if a file on the filesystem is out of sync with the index\&.
1285 .sp
1286 Example:
1287 .sp
1288 .if n \{\
1289 .RS 4
1290 .\}
1291 .nf
1292 :100644 100644 5be4a4a 0000000 M file\&.c
1293 .fi
1294 .if n \{\
1295 .RE
1296 .\}
1297 .sp
1298 .sp
1299 Without the \fB\-z\fR option, pathnames with "unusual" characters are quoted as explained for the configuration variable \fBcore\&.quotePath\fR (see \fBgit-config\fR(1))\&. Using \fB\-z\fR the filename is output verbatim and the line is terminated by a NUL byte\&.
1300 .SH "DIFF FORMAT FOR MERGES"
1301 .sp
1302 "git\-diff\-tree", "git\-diff\-files" and "git\-diff \-\-raw" can take \fB\-c\fR or \fB\-\-cc\fR option to generate diff output also for merge commits\&. The output differs from the format described above in the following way:
1303 .sp
1304 .RS 4
1305 .ie n \{\
1306 \h'-04' 1.\h'+01'\c
1307 .\}
1308 .el \{\
1309 .sp -1
1310 .IP " 1." 4.2
1311 .\}
1312 there is a colon for each parent
1313 .RE
1314 .sp
1315 .RS 4
1316 .ie n \{\
1317 \h'-04' 2.\h'+01'\c
1318 .\}
1319 .el \{\
1320 .sp -1
1321 .IP " 2." 4.2
1322 .\}
1323 there are more "src" modes and "src" sha1
1324 .RE
1325 .sp
1326 .RS 4
1327 .ie n \{\
1328 \h'-04' 3.\h'+01'\c
1329 .\}
1330 .el \{\
1331 .sp -1
1332 .IP " 3." 4.2
1333 .\}
1334 status is concatenated status characters for each parent
1335 .RE
1336 .sp
1337 .RS 4
1338 .ie n \{\
1339 \h'-04' 4.\h'+01'\c
1340 .\}
1341 .el \{\
1342 .sp -1
1343 .IP " 4." 4.2
1344 .\}
1345 no optional "score" number
1346 .RE
1347 .sp
1348 .RS 4
1349 .ie n \{\
1350 \h'-04' 5.\h'+01'\c
1351 .\}
1352 .el \{\
1353 .sp -1
1354 .IP " 5." 4.2
1355 .\}
1356 tab\-separated pathname(s) of the file
1357 .RE
1358 .sp
1359 For \fB\-c\fR and \fB\-\-cc\fR, only the destination or final path is shown even if the file was renamed on any side of history\&. With \fB\-\-combined\-all\-paths\fR, the name of the path in each parent is shown followed by the name of the path in the merge commit\&.
1360 .sp
1361 Examples for \fB\-c\fR and \fB\-\-cc\fR without \fB\-\-combined\-all\-paths\fR:
1362 .sp
1363 .if n \{\
1364 .RS 4
1365 .\}
1366 .nf
1367 ::100644 100644 100644 fabadb8 cc95eb0 4866510 MM desc\&.c
1368 ::100755 100755 100755 52b7a2d 6d1ac04 d2ac7d7 RM bar\&.sh
1369 ::100644 100644 100644 e07d6c5 9042e82 ee91881 RR phooey\&.c
1370 .fi
1371 .if n \{\
1372 .RE
1373 .\}
1374 .sp
1375 .sp
1376 Examples when \fB\-\-combined\-all\-paths\fR added to either \fB\-c\fR or \fB\-\-cc\fR:
1377 .sp
1378 .if n \{\
1379 .RS 4
1380 .\}
1381 .nf
1382 ::100644 100644 100644 fabadb8 cc95eb0 4866510 MM desc\&.c desc\&.c desc\&.c
1383 ::100755 100755 100755 52b7a2d 6d1ac04 d2ac7d7 RM foo\&.sh bar\&.sh bar\&.sh
1384 ::100644 100644 100644 e07d6c5 9042e82 ee91881 RR fooey\&.c fuey\&.c phooey\&.c
1385 .fi
1386 .if n \{\
1387 .RE
1388 .\}
1389 .sp
1390 .sp
1391 Note that \fIcombined diff\fR lists only files which were modified from all parents\&.
1392 .SH "GENERATING PATCH TEXT WITH \-P"
1393 .sp
1394 Running \fBgit-diff\fR(1), \fBgit-log\fR(1), \fBgit-show\fR(1), \fBgit-diff-index\fR(1), \fBgit-diff-tree\fR(1), or \fBgit-diff-files\fR(1) with the \fB\-p\fR option produces patch text\&. You can customize the creation of patch text via the \fBGIT_EXTERNAL_DIFF\fR and the \fBGIT_DIFF_OPTS\fR environment variables (see \fBgit\fR(1)), and the \fBdiff\fR attribute (see \fBgitattributes\fR(5))\&.
1395 .sp
1396 What the \-p option produces is slightly different from the traditional diff format:
1397 .sp
1398 .RS 4
1399 .ie n \{\
1400 \h'-04' 1.\h'+01'\c
1401 .\}
1402 .el \{\
1403 .sp -1
1404 .IP " 1." 4.2
1405 .\}
1406 It is preceded by a "git diff" header that looks like this:
1407 .sp
1408 .if n \{\
1409 .RS 4
1410 .\}
1411 .nf
1412 diff \-\-git a/file1 b/file2
1413 .fi
1414 .if n \{\
1415 .RE
1416 .\}
1417 .sp
1418 The
1419 \fBa/\fR
1420 and
1421 \fBb/\fR
1422 filenames are the same unless rename/copy is involved\&. Especially, even for a creation or a deletion,
1423 \fB/dev/null\fR
1424 is
1425 \fInot\fR
1426 used in place of the
1427 \fBa/\fR
1428 or
1429 \fBb/\fR
1430 filenames\&.
1431 .sp
1432 When a rename/copy is involved,
1433 \fBfile1\fR
1434 and
1435 \fBfile2\fR
1436 show the name of the source file of the rename/copy and the name of the file that the rename/copy produces, respectively\&.
1437 .RE
1438 .sp
1439 .RS 4
1440 .ie n \{\
1441 \h'-04' 2.\h'+01'\c
1442 .\}
1443 .el \{\
1444 .sp -1
1445 .IP " 2." 4.2
1446 .\}
1447 It is followed by one or more extended header lines:
1448 .sp
1449 .if n \{\
1450 .RS 4
1451 .\}
1452 .nf
1453 old mode <mode>
1454 new mode <mode>
1455 deleted file mode <mode>
1456 new file mode <mode>
1457 copy from <path>
1458 copy to <path>
1459 rename from <path>
1460 rename to <path>
1461 similarity index <number>
1462 dissimilarity index <number>
1463 index <hash>\&.\&.<hash> <mode>
1464 .fi
1465 .if n \{\
1466 .RE
1467 .\}
1468 .sp
1469 File modes are printed as 6\-digit octal numbers including the file type and file permission bits\&.
1470 .sp
1471 Path names in extended headers do not include the
1472 \fBa/\fR
1473 and
1474 \fBb/\fR
1475 prefixes\&.
1476 .sp
1477 The similarity index is the percentage of unchanged lines, and the dissimilarity index is the percentage of changed lines\&. It is a rounded down integer, followed by a percent sign\&. The similarity index value of 100% is thus reserved for two equal files, while 100% dissimilarity means that no line from the old file made it into the new one\&.
1478 .sp
1479 The index line includes the blob object names before and after the change\&. The <mode> is included if the file mode does not change; otherwise, separate lines indicate the old and the new mode\&.
1480 .RE
1481 .sp
1482 .RS 4
1483 .ie n \{\
1484 \h'-04' 3.\h'+01'\c
1485 .\}
1486 .el \{\
1487 .sp -1
1488 .IP " 3." 4.2
1489 .\}
1490 Pathnames with "unusual" characters are quoted as explained for the configuration variable
1491 \fBcore\&.quotePath\fR
1492 (see
1493 \fBgit-config\fR(1))\&.
1494 .RE
1495 .sp
1496 .RS 4
1497 .ie n \{\
1498 \h'-04' 4.\h'+01'\c
1499 .\}
1500 .el \{\
1501 .sp -1
1502 .IP " 4." 4.2
1503 .\}
1504 All the
1505 \fBfile1\fR
1506 files in the output refer to files before the commit, and all the
1507 \fBfile2\fR
1508 files refer to files after the commit\&. It is incorrect to apply each change to each file sequentially\&. For example, this patch will swap a and b:
1509 .sp
1510 .if n \{\
1511 .RS 4
1512 .\}
1513 .nf
1514 diff \-\-git a/a b/b
1515 rename from a
1516 rename to b
1517 diff \-\-git a/b b/a
1518 rename from b
1519 rename to a
1520 .fi
1521 .if n \{\
1522 .RE
1523 .\}
1524 .RE
1525 .sp
1526 .RS 4
1527 .ie n \{\
1528 \h'-04' 5.\h'+01'\c
1529 .\}
1530 .el \{\
1531 .sp -1
1532 .IP " 5." 4.2
1533 .\}
1534 Hunk headers mention the name of the function to which the hunk applies\&. See "Defining a custom hunk\-header" in
1535 \fBgitattributes\fR(5)
1536 for details of how to tailor this to specific languages\&.
1537 .RE
1538 .SH "COMBINED DIFF FORMAT"
1539 .sp
1540 Any diff\-generating command can take the \fB\-c\fR or \fB\-\-cc\fR option to produce a \fIcombined diff\fR when showing a merge\&. This is the default format when showing merges with \fBgit-diff\fR(1) or \fBgit-show\fR(1)\&. Note also that you can give suitable \fB\-\-diff\-merges\fR option to any of these commands to force generation of diffs in a specific format\&.
1541 .sp
1542 A "combined diff" format looks like this:
1543 .sp
1544 .if n \{\
1545 .RS 4
1546 .\}
1547 .nf
1548 diff \-\-combined describe\&.c
1549 index fabadb8,cc95eb0\&.\&.4866510
1550 \-\-\- a/describe\&.c
1551 +++ b/describe\&.c
1552 @@@ \-98,20 \-98,12 +98,20 @@@
1553 return (a_date > b_date) ? \-1 : (a_date == b_date) ? 0 : 1;
1554 }
1556 \- static void describe(char *arg)
1557 \-static void describe(struct commit *cmit, int last_one)
1558 ++static void describe(char *arg, int last_one)
1559 {
1560 + unsigned char sha1[20];
1561 + struct commit *cmit;
1562 struct commit_list *list;
1563 static int initialized = 0;
1564 struct commit_name *n;
1566 + if (get_sha1(arg, sha1) < 0)
1567 + usage(describe_usage);
1568 + cmit = lookup_commit_reference(sha1);
1569 + if (!cmit)
1570 + usage(describe_usage);
1571 +
1572 if (!initialized) {
1573 initialized = 1;
1574 for_each_ref(get_name);
1575 .fi
1576 .if n \{\
1577 .RE
1578 .\}
1579 .sp
1581 .sp
1582 .RS 4
1583 .ie n \{\
1584 \h'-04' 1.\h'+01'\c
1585 .\}
1586 .el \{\
1587 .sp -1
1588 .IP " 1." 4.2
1589 .\}
1590 It is preceded by a "git diff" header, that looks like this (when the
1591 \fB\-c\fR
1592 option is used):
1593 .sp
1594 .if n \{\
1595 .RS 4
1596 .\}
1597 .nf
1598 diff \-\-combined file
1599 .fi
1600 .if n \{\
1601 .RE
1602 .\}
1603 .sp
1604 or like this (when the
1605 \fB\-\-cc\fR
1606 option is used):
1607 .sp
1608 .if n \{\
1609 .RS 4
1610 .\}
1611 .nf
1612 diff \-\-cc file
1613 .fi
1614 .if n \{\
1615 .RE
1616 .\}
1617 .RE
1618 .sp
1619 .RS 4
1620 .ie n \{\
1621 \h'-04' 2.\h'+01'\c
1622 .\}
1623 .el \{\
1624 .sp -1
1625 .IP " 2." 4.2
1626 .\}
1627 It is followed by one or more extended header lines (this example shows a merge with two parents):
1628 .sp
1629 .if n \{\
1630 .RS 4
1631 .\}
1632 .nf
1633 index <hash>,<hash>\&.\&.<hash>
1634 mode <mode>,<mode>\&.\&.<mode>
1635 new file mode <mode>
1636 deleted file mode <mode>,<mode>
1637 .fi
1638 .if n \{\
1639 .RE
1640 .\}
1641 .sp
1642 The
1643 \fBmode <mode>,<mode>\&.\&.<mode>\fR
1644 line appears only if at least one of the <mode> is different from the rest\&. Extended headers with information about detected content movement (renames and copying detection) are designed to work with the diff of two <tree\-ish> and are not used by combined diff format\&.
1645 .RE
1646 .sp
1647 .RS 4
1648 .ie n \{\
1649 \h'-04' 3.\h'+01'\c
1650 .\}
1651 .el \{\
1652 .sp -1
1653 .IP " 3." 4.2
1654 .\}
1655 It is followed by a two\-line from\-file/to\-file header:
1656 .sp
1657 .if n \{\
1658 .RS 4
1659 .\}
1660 .nf
1661 \-\-\- a/file
1662 +++ b/file
1663 .fi
1664 .if n \{\
1665 .RE
1666 .\}
1667 .sp
1668 Similar to the two\-line header for the traditional
1669 \fIunified\fR
1670 diff format,
1671 \fB/dev/null\fR
1672 is used to signal created or deleted files\&.
1673 .sp
1674 However, if the \-\-combined\-all\-paths option is provided, instead of a two\-line from\-file/to\-file, you get an N+1 line from\-file/to\-file header, where N is the number of parents in the merge commit:
1675 .sp
1676 .if n \{\
1677 .RS 4
1678 .\}
1679 .nf
1680 \-\-\- a/file
1681 \-\-\- a/file
1682 \-\-\- a/file
1683 +++ b/file
1684 .fi
1685 .if n \{\
1686 .RE
1687 .\}
1688 .sp
1689 This extended format can be useful if rename or copy detection is active, to allow you to see the original name of the file in different parents\&.
1690 .RE
1691 .sp
1692 .RS 4
1693 .ie n \{\
1694 \h'-04' 4.\h'+01'\c
1695 .\}
1696 .el \{\
1697 .sp -1
1698 .IP " 4." 4.2
1699 .\}
1700 Chunk header format is modified to prevent people from accidentally feeding it to
1701 \fBpatch \-p1\fR\&. Combined diff format was created for review of merge commit changes, and was not meant to be applied\&. The change is similar to the change in the extended
1702 \fIindex\fR
1703 header:
1704 .sp
1705 .if n \{\
1706 .RS 4
1707 .\}
1708 .nf
1709 @@@ <from\-file\-range> <from\-file\-range> <to\-file\-range> @@@
1710 .fi
1711 .if n \{\
1712 .RE
1713 .\}
1714 .sp
1715 There are (number of parents + 1)
1716 \fB@\fR
1717 characters in the chunk header for combined diff format\&.
1718 .RE
1719 .sp
1720 Unlike the traditional \fIunified\fR diff format, which shows two files A and B with a single column that has \fB\-\fR (minus \(em appears in A but removed in B), \fB+\fR (plus \(em missing in A but added to B), or \fB" "\fR (space \(em unchanged) prefix, this format compares two or more files file1, file2,\&... with one file X, and shows how X differs from each of fileN\&. One column for each of fileN is prepended to the output line to note how X\(cqs line is different from it\&.
1721 .sp
1722 A \fB\-\fR character in the column N means that the line appears in fileN but it does not appear in the result\&. A \fB+\fR character in the column N means that the line appears in the result, and fileN does not have that line (in other words, the line was added, from the point of view of that parent)\&.
1723 .sp
1724 In the above example output, the function signature was changed from both files (hence two \fB\-\fR removals from both file1 and file2, plus \fB++\fR to mean one line that was added does not appear in either file1 or file2)\&. Also, eight other lines are the same from file1 but do not appear in file2 (hence prefixed with \fB+\fR)\&.
1725 .sp
1726 When shown by \fBgit diff\-tree \-c\fR, it compares the parents of a merge commit with the merge result (i\&.e\&. file1\&.\&.fileN are the parents)\&. When shown by \fBgit diff\-files \-c\fR, it compares the two unresolved merge parents with the working tree file (i\&.e\&. file1 is stage 2 aka "our version", file2 is stage 3 aka "their version")\&.
1727 .SH "OTHER DIFF FORMATS"
1728 .sp
1729 The \fB\-\-summary\fR option describes newly added, deleted, renamed and copied files\&. The \fB\-\-stat\fR option adds diffstat(1) graph to the output\&. These options can be combined with other options, such as \fB\-p\fR, and are meant for human consumption\&.
1730 .sp
1731 When showing a change that involves a rename or a copy, \fB\-\-stat\fR output formats the pathnames compactly by combining common prefix and suffix of the pathnames\&. For example, a change that moves \fBarch/i386/Makefile\fR to \fBarch/x86/Makefile\fR while modifying 4 lines will be shown like this:
1732 .sp
1733 .if n \{\
1734 .RS 4
1735 .\}
1736 .nf
1737 arch/{i386 => x86}/Makefile | 4 +\-\-
1738 .fi
1739 .if n \{\
1740 .RE
1741 .\}
1742 .sp
1743 .sp
1744 The \fB\-\-numstat\fR option gives the diffstat(1) information but is designed for easier machine consumption\&. An entry in \fB\-\-numstat\fR output looks like this:
1745 .sp
1746 .if n \{\
1747 .RS 4
1748 .\}
1749 .nf
1750 1 2 README
1751 3 1 arch/{i386 => x86}/Makefile
1752 .fi
1753 .if n \{\
1754 .RE
1755 .\}
1756 .sp
1757 .sp
1758 That is, from left to right:
1759 .sp
1760 .RS 4
1761 .ie n \{\
1762 \h'-04' 1.\h'+01'\c
1763 .\}
1764 .el \{\
1765 .sp -1
1766 .IP " 1." 4.2
1767 .\}
1768 the number of added lines;
1769 .RE
1770 .sp
1771 .RS 4
1772 .ie n \{\
1773 \h'-04' 2.\h'+01'\c
1774 .\}
1775 .el \{\
1776 .sp -1
1777 .IP " 2." 4.2
1778 .\}
1779 a tab;
1780 .RE
1781 .sp
1782 .RS 4
1783 .ie n \{\
1784 \h'-04' 3.\h'+01'\c
1785 .\}
1786 .el \{\
1787 .sp -1
1788 .IP " 3." 4.2
1789 .\}
1790 the number of deleted lines;
1791 .RE
1792 .sp
1793 .RS 4
1794 .ie n \{\
1795 \h'-04' 4.\h'+01'\c
1796 .\}
1797 .el \{\
1798 .sp -1
1799 .IP " 4." 4.2
1800 .\}
1801 a tab;
1802 .RE
1803 .sp
1804 .RS 4
1805 .ie n \{\
1806 \h'-04' 5.\h'+01'\c
1807 .\}
1808 .el \{\
1809 .sp -1
1810 .IP " 5." 4.2
1811 .\}
1812 pathname (possibly with rename/copy information);
1813 .RE
1814 .sp
1815 .RS 4
1816 .ie n \{\
1817 \h'-04' 6.\h'+01'\c
1818 .\}
1819 .el \{\
1820 .sp -1
1821 .IP " 6." 4.2
1822 .\}
1823 a newline\&.
1824 .RE
1825 .sp
1826 When \fB\-z\fR output option is in effect, the output is formatted this way:
1827 .sp
1828 .if n \{\
1829 .RS 4
1830 .\}
1831 .nf
1832 1 2 README NUL
1833 3 1 NUL arch/i386/Makefile NUL arch/x86/Makefile NUL
1834 .fi
1835 .if n \{\
1836 .RE
1837 .\}
1838 .sp
1839 .sp
1840 That is:
1841 .sp
1842 .RS 4
1843 .ie n \{\
1844 \h'-04' 1.\h'+01'\c
1845 .\}
1846 .el \{\
1847 .sp -1
1848 .IP " 1." 4.2
1849 .\}
1850 the number of added lines;
1851 .RE
1852 .sp
1853 .RS 4
1854 .ie n \{\
1855 \h'-04' 2.\h'+01'\c
1856 .\}
1857 .el \{\
1858 .sp -1
1859 .IP " 2." 4.2
1860 .\}
1861 a tab;
1862 .RE
1863 .sp
1864 .RS 4
1865 .ie n \{\
1866 \h'-04' 3.\h'+01'\c
1867 .\}
1868 .el \{\
1869 .sp -1
1870 .IP " 3." 4.2
1871 .\}
1872 the number of deleted lines;
1873 .RE
1874 .sp
1875 .RS 4
1876 .ie n \{\
1877 \h'-04' 4.\h'+01'\c
1878 .\}
1879 .el \{\
1880 .sp -1
1881 .IP " 4." 4.2
1882 .\}
1883 a tab;
1884 .RE
1885 .sp
1886 .RS 4
1887 .ie n \{\
1888 \h'-04' 5.\h'+01'\c
1889 .\}
1890 .el \{\
1891 .sp -1
1892 .IP " 5." 4.2
1893 .\}
1894 a NUL (only exists if renamed/copied);
1895 .RE
1896 .sp
1897 .RS 4
1898 .ie n \{\
1899 \h'-04' 6.\h'+01'\c
1900 .\}
1901 .el \{\
1902 .sp -1
1903 .IP " 6." 4.2
1904 .\}
1905 pathname in preimage;
1906 .RE
1907 .sp
1908 .RS 4
1909 .ie n \{\
1910 \h'-04' 7.\h'+01'\c
1911 .\}
1912 .el \{\
1913 .sp -1
1914 .IP " 7." 4.2
1915 .\}
1916 a NUL (only exists if renamed/copied);
1917 .RE
1918 .sp
1919 .RS 4
1920 .ie n \{\
1921 \h'-04' 8.\h'+01'\c
1922 .\}
1923 .el \{\
1924 .sp -1
1925 .IP " 8." 4.2
1926 .\}
1927 pathname in postimage (only exists if renamed/copied);
1928 .RE
1929 .sp
1930 .RS 4
1931 .ie n \{\
1932 \h'-04' 9.\h'+01'\c
1933 .\}
1934 .el \{\
1935 .sp -1
1936 .IP " 9." 4.2
1937 .\}
1938 a NUL\&.
1939 .RE
1940 .sp
1941 The extra \fBNUL\fR before the preimage path in renamed case is to allow scripts that read the output to tell if the current record being read is a single\-path record or a rename/copy record without reading ahead\&. After reading added and deleted lines, reading up to \fBNUL\fR would yield the pathname, but if that is \fBNUL\fR, the record will show two paths\&.
1942 .SH "OPERATING MODES"
1943 .sp
1944 You can choose whether you want to trust the index file entirely (using the \fB\-\-cached\fR flag) or ask the diff logic to show any files that don\(cqt match the stat state as being "tentatively changed"\&. Both of these operations are very useful indeed\&.
1945 .SH "CACHED MODE"
1946 .sp
1947 If \fB\-\-cached\fR is specified, it allows you to ask:
1948 .sp
1949 .if n \{\
1950 .RS 4
1951 .\}
1952 .nf
1953 show me the differences between HEAD and the current index
1954 contents (the ones I\*(Aqd write using \*(Aqgit write\-tree\*(Aq)
1955 .fi
1956 .if n \{\
1957 .RE
1958 .\}
1959 .sp
1960 For example, let\(cqs say that you have worked on your working directory, updated some files in the index and are ready to commit\&. You want to see exactly \fBwhat\fR you are going to commit, without having to write a new tree object and compare it that way, and to do that, you just do
1961 .sp
1962 .if n \{\
1963 .RS 4
1964 .\}
1965 .nf
1966 git diff\-index \-\-cached HEAD
1967 .fi
1968 .if n \{\
1969 .RE
1970 .\}
1971 .sp
1972 Example: let\(cqs say I had renamed \fBcommit\&.c\fR to \fBgit\-commit\&.c\fR, and I had done an \fBupdate\-index\fR to make that effective in the index file\&. \fBgit diff\-files\fR wouldn\(cqt show anything at all, since the index file matches my working directory\&. But doing a \fIgit diff\-index\fR does:
1973 .sp
1974 .if n \{\
1975 .RS 4
1976 .\}
1977 .nf
1978 torvalds@ppc970:~/git> git diff\-index \-\-cached HEAD
1979 :100644 000000 4161aecc6700a2eb579e842af0b7f22b98443f74 0000000000000000000000000000000000000000 D commit\&.c
1980 :000000 100644 0000000000000000000000000000000000000000 4161aecc6700a2eb579e842af0b7f22b98443f74 A git\-commit\&.c
1981 .fi
1982 .if n \{\
1983 .RE
1984 .\}
1985 .sp
1986 You can see easily that the above is a rename\&.
1987 .sp
1988 In fact, \fBgit diff\-index \-\-cached\fR \fBshould\fR always be entirely equivalent to actually doing a \fIgit write\-tree\fR and comparing that\&. Except this one is much nicer for the case where you just want to check where you are\&.
1989 .sp
1990 So doing a \fBgit diff\-index \-\-cached\fR is basically very useful when you are asking yourself "what have I already marked for being committed, and what\(cqs the difference to a previous tree"\&.
1991 .SH "NON\-CACHED MODE"
1992 .sp
1993 The "non\-cached" mode takes a different approach, and is potentially the more useful of the two in that what it does can\(cqt be emulated with a \fIgit write\-tree\fR + \fIgit diff\-tree\fR\&. Thus that\(cqs the default mode\&. The non\-cached version asks the question:
1994 .sp
1995 .if n \{\
1996 .RS 4
1997 .\}
1998 .nf
1999 show me the differences between HEAD and the currently checked out
2000 tree \- index contents _and_ files that aren\*(Aqt up to date
2001 .fi
2002 .if n \{\
2003 .RE
2004 .\}
2005 .sp
2006 which is obviously a very useful question too, since that tells you what you \fBcould\fR commit\&. Again, the output matches the \fIgit diff\-tree \-r\fR output to a tee, but with a twist\&.
2007 .sp
2008 The twist is that if some file doesn\(cqt match the index, we don\(cqt have a backing store thing for it, and we use the magic "all\-zero" sha1 to show that\&. So let\(cqs say that you have edited \fBkernel/sched\&.c\fR, but have not actually done a \fIgit update\-index\fR on it yet \- there is no "object" associated with the new state, and you get:
2009 .sp
2010 .if n \{\
2011 .RS 4
2012 .\}
2013 .nf
2014 torvalds@ppc970:~/v2\&.6/linux> git diff\-index \-\-abbrev HEAD
2015 :100644 100644 7476bb5ba 000000000 M kernel/sched\&.c
2016 .fi
2017 .if n \{\
2018 .RE
2019 .\}
2020 .sp
2021 i\&.e\&., it shows that the tree has changed, and that \fBkernel/sched\&.c\fR is not up to date and may contain new stuff\&. The all\-zero sha1 means that to get the real diff, you need to look at the object in the working directory directly rather than do an object\-to\-object diff\&.
2022 .if n \{\
2023 .sp
2024 .\}
2025 .RS 4
2026 .it 1 an-trap
2027 .nr an-no-space-flag 1
2028 .nr an-break-flag 1
2029 .br
2030 .ps +1
2031 \fBNote\fR
2032 .ps -1
2033 .br
2034 .sp
2035 As with other commands of this type, \fIgit diff\-index\fR does not actually look at the contents of the file at all\&. So maybe \fBkernel/sched\&.c\fR hasn\(cqt actually changed, and it\(cqs just that you touched it\&. In either case, it\(cqs a note that you need to \fIgit update\-index\fR it to make the index be in sync\&.
2036 .sp .5v
2037 .RE
2038 .if n \{\
2039 .sp
2040 .\}
2041 .RS 4
2042 .it 1 an-trap
2043 .nr an-no-space-flag 1
2044 .nr an-break-flag 1
2045 .br
2046 .ps +1
2047 \fBNote\fR
2048 .ps -1
2049 .br
2050 .sp
2051 You can have a mixture of files show up as "has been updated" and "is still dirty in the working directory" together\&. You can always tell which file is in which state, since the "has been updated" ones show a valid sha1, and the "not in sync with the index" ones will always have the special all\-zero sha1\&.
2052 .sp .5v
2053 .RE
2054 .SH "GIT"
2055 .sp
2056 Part of the \fBgit\fR(1) suite