1GIT-FORMAT-PATCH(1) Git Manual GIT-FORMAT-PATCH(1)
2
6 git-format-patch - Prepare patches for e-mail submission
7
9 git format-patch [-k] [(-o|--output-directory) <dir> | --stdout]
10 [--no-thread | --thread[=<style>]]
11 [(--attach|--inline)[=<boundary>] | --no-attach]
12 [-s | --signoff]
13 [--signature=<signature> | --no-signature]
14 [--signature-file=<file>]
15 [-n | --numbered | -N | --no-numbered]
16 [--start-number <n>] [--numbered-files]
17 [--in-reply-to=<message id>] [--suffix=.<sfx>]
18 [--ignore-if-in-upstream] [--always]
19 [--cover-from-description=<mode>]
20 [--rfc] [--subject-prefix=<subject prefix>]
21 [(--reroll-count|-v) <n>]
22 [--to=<email>] [--cc=<email>]
23 [--[no-]cover-letter] [--quiet]
24 [--[no-]encode-email-headers]
25 [--no-notes | --notes[=<ref>]]
26 [--interdiff=<previous>]
27 [--range-diff=<previous> [--creation-factor=<percent>]]
28 [--filename-max-length=<n>]
29 [--progress]
30 [<common diff options>]
31 [ <since> | <revision range> ]
32
34 Prepare each non-merge commit with its "patch" in one "message" per
35 commit, formatted to resemble a UNIX mailbox. The output of this
36 command is convenient for e-mail submission or for use with git am.
37 A "message" generated by the command consists of three parts:
39 • A brief metadata header that begins with From <commit> with a fixed
41 Mon Sep 17 00:00:00 2001 datestamp to help programs like "file(1)"
42 to recognize that the file is an output from this command, fields
43 that record the author identity, the author date, and the title of
44 the change (taken from the first paragraph of the commit log
45 message).
46 • The second and subsequent paragraphs of the commit log message.
48 • The "patch", which is the "diff -p --stat" output (see git-diff(1))
50 between the commit and its parent.
51 The log message and the patch are separated by a line with a three-dash
53 line.
54 There are two ways to specify which commits to operate on.
56 1. A single commit, <since>, specifies that the commits leading to the
58 tip of the current branch that are not in the history that leads to
59 the <since> to be output.
60 2. Generic <revision range> expression (see "SPECIFYING REVISIONS"
62 section in gitrevisions(7)) means the commits in the specified
63 range.
64 The first rule takes precedence in the case of a single <commit>. To
66 apply the second rule, i.e., format everything since the beginning of
67 history up until <commit>, use the --root option: git format-patch
68 --root <commit>. If you want to format only <commit> itself, you can do
69 this with git format-patch -1 <commit>.
70 By default, each output file is numbered sequentially from 1, and uses
72 the first line of the commit message (massaged for pathname safety) as
73 the filename. With the --numbered-files option, the output file names
74 will only be numbers, without the first line of the commit appended.
75 The names of the output files are printed to standard output, unless
76 the --stdout option is specified.
77 If -o is specified, output files are created in <dir>. Otherwise they
79 are created in the current working directory. The default path can be
80 set with the format.outputDirectory configuration option. The -o option
81 takes precedence over format.outputDirectory. To store patches in the
82 current working directory even when format.outputDirectory points
83 elsewhere, use -o .. All directory components will be created.
84 By default, the subject of a single patch is "[PATCH] " followed by the
86 concatenation of lines from the commit message up to the first blank
87 line (see the DISCUSSION section of git-commit(1)).
88 When multiple patches are output, the subject prefix will instead be
90 "[PATCH n/m] ". To force 1/1 to be added for a single patch, use -n. To
91 omit patch numbers from the subject, use -N.
92 If given --thread, git-format-patch will generate In-Reply-To and
94 References headers to make the second and subsequent patch mails appear
95 as replies to the first mail; this also generates a Message-ID header
96 to reference.
97
99 -p, --no-stat
100 Generate plain patches without any diffstats.
101 -U<n>, --unified=<n>
103 Generate diffs with <n> lines of context instead of the usual
104 three.
105 --output=<file>
107 Output to a specific file instead of stdout.
108 --output-indicator-new=<char>, --output-indicator-old=<char>,
110 --output-indicator-context=<char>
111 Specify the character used to indicate new, old or context lines in
112 the generated patch. Normally they are +, - and ' ' respectively.
113 --indent-heuristic
115 Enable the heuristic that shifts diff hunk boundaries to make
116 patches easier to read. This is the default.
117 --no-indent-heuristic
119 Disable the indent heuristic.
120 --minimal
122 Spend extra time to make sure the smallest possible diff is
123 produced.
124 --patience
126 Generate a diff using the "patience diff" algorithm.
127 --histogram
129 Generate a diff using the "histogram diff" algorithm.
130 --anchored=<text>
132 Generate a diff using the "anchored diff" algorithm.
133 This option may be specified more than once.
135 If a line exists in both the source and destination, exists only
137 once, and starts with this text, this algorithm attempts to prevent
138 it from appearing as a deletion or addition in the output. It uses
139 the "patience diff" algorithm internally.
140 --diff-algorithm={patience|minimal|histogram|myers}
142 Choose a diff algorithm. The variants are as follows:
143 default, myers
145 The basic greedy diff algorithm. Currently, this is the
146 default.
147 minimal
149 Spend extra time to make sure the smallest possible diff is
150 produced.
151 patience
153 Use "patience diff" algorithm when generating patches.
154 histogram
156 This algorithm extends the patience algorithm to "support
157 low-occurrence common elements".
158 For instance, if you configured the diff.algorithm variable to a
160 non-default value and want to use the default one, then you have to
161 use --diff-algorithm=default option.
162 --stat[=<width>[,<name-width>[,<count>]]]
164 Generate a diffstat. By default, as much space as necessary will be
165 used for the filename part, and the rest for the graph part.
166 Maximum width defaults to terminal width, or 80 columns if not
167 connected to a terminal, and can be overridden by <width>. The
168 width of the filename part can be limited by giving another width
169 <name-width> after a comma or by setting
170 diff.statNameWidth=<width>. The width of the graph part can be
171 limited by using --stat-graph-width=<width> or by setting
172 diff.statGraphWidth=<width>. Using --stat or --stat-graph-width
173 affects all commands generating a stat graph, while setting
174 diff.statNameWidth or diff.statGraphWidth does not affect git
175 format-patch. By giving a third parameter <count>, you can limit
176 the output to the first <count> lines, followed by ... if there
177 are more.
178 These parameters can also be set individually with
180 --stat-width=<width>, --stat-name-width=<name-width> and
181 --stat-count=<count>.
182 --compact-summary
184 Output a condensed summary of extended header information such as
185 file creations or deletions ("new" or "gone", optionally "+l" if
186 it’s a symlink) and mode changes ("+x" or "-x" for adding or
187 removing executable bit respectively) in diffstat. The information
188 is put between the filename part and the graph part. Implies
189 --stat.
190 --numstat
192 Similar to --stat, but shows number of added and deleted lines in
193 decimal notation and pathname without abbreviation, to make it more
194 machine friendly. For binary files, outputs two - instead of saying
195 0 0.
196 --shortstat
198 Output only the last line of the --stat format containing total
199 number of modified files, as well as number of added and deleted
200 lines.
201 -X[<param1,param2,...>], --dirstat[=<param1,param2,...>]
203 Output the distribution of relative amount of changes for each
204 sub-directory. The behavior of --dirstat can be customized by
205 passing it a comma separated list of parameters. The defaults are
206 controlled by the diff.dirstat configuration variable (see git-
207 config(1)). The following parameters are available:
208 changes
210 Compute the dirstat numbers by counting the lines that have
211 been removed from the source, or added to the destination. This
212 ignores the amount of pure code movements within a file. In
213 other words, rearranging lines in a file is not counted as much
214 as other changes. This is the default behavior when no
215 parameter is given.
216 lines
218 Compute the dirstat numbers by doing the regular line-based
219 diff analysis, and summing the removed/added line counts. (For
220 binary files, count 64-byte chunks instead, since binary files
221 have no natural concept of lines). This is a more expensive
222 --dirstat behavior than the changes behavior, but it does count
223 rearranged lines within a file as much as other changes. The
224 resulting output is consistent with what you get from the other
225 --*stat options.
226 files
228 Compute the dirstat numbers by counting the number of files
229 changed. Each changed file counts equally in the dirstat
230 analysis. This is the computationally cheapest --dirstat
231 behavior, since it does not have to look at the file contents
232 at all.
233 cumulative
235 Count changes in a child directory for the parent directory as
236 well. Note that when using cumulative, the sum of the
237 percentages reported may exceed 100%. The default
238 (non-cumulative) behavior can be specified with the
239 noncumulative parameter.
240 <limit>
242 An integer parameter specifies a cut-off percent (3% by
243 default). Directories contributing less than this percentage of
244 the changes are not shown in the output.
245 Example: The following will count changed files, while ignoring
247 directories with less than 10% of the total amount of changed
248 files, and accumulating child directory counts in the parent
249 directories: --dirstat=files,10,cumulative.
250 --cumulative
252 Synonym for --dirstat=cumulative
253 --dirstat-by-file[=<param1,param2>...]
255 Synonym for --dirstat=files,param1,param2...
256 --summary
258 Output a condensed summary of extended header information such as
259 creations, renames and mode changes.
260 --no-renames
262 Turn off rename detection, even when the configuration file gives
263 the default to do so.
264 --[no-]rename-empty
266 Whether to use empty blobs as rename source.
267 --full-index
269 Instead of the first handful of characters, show the full pre- and
270 post-image blob object names on the "index" line when generating
271 patch format output.
272 --binary
274 In addition to --full-index, output a binary diff that can be
275 applied with git-apply.
276 --abbrev[=<n>]
278 Instead of showing the full 40-byte hexadecimal object name in
279 diff-raw format output and diff-tree header lines, show the
280 shortest prefix that is at least <n> hexdigits long that uniquely
281 refers the object. In diff-patch output format, --full-index takes
282 higher precedence, i.e. if --full-index is specified, full blob
283 names will be shown regardless of --abbrev. Non default number of
284 digits can be specified with --abbrev=<n>.
285 -B[<n>][/<m>], --break-rewrites[=[<n>][/<m>]]
287 Break complete rewrite changes into pairs of delete and create.
288 This serves two purposes:
289 It affects the way a change that amounts to a total rewrite of a
291 file not as a series of deletion and insertion mixed together with
292 a very few lines that happen to match textually as the context, but
293 as a single deletion of everything old followed by a single
294 insertion of everything new, and the number m controls this aspect
295 of the -B option (defaults to 60%). -B/70% specifies that less
296 than 30% of the original should remain in the result for Git to
297 consider it a total rewrite (i.e. otherwise the resulting patch
298 will be a series of deletion and insertion mixed together with
299 context lines).
300 When used with -M, a totally-rewritten file is also considered as
302 the source of a rename (usually -M only considers a file that
303 disappeared as the source of a rename), and the number n controls
304 this aspect of the -B option (defaults to 50%). -B20% specifies
305 that a change with addition and deletion compared to 20% or more of
306 the file’s size are eligible for being picked up as a possible
307 source of a rename to another file.
308 -M[<n>], --find-renames[=<n>]
310 Detect renames. If n is specified, it is a threshold on the
311 similarity index (i.e. amount of addition/deletions compared to the
312 file’s size). For example, -M90% means Git should consider a
313 delete/add pair to be a rename if more than 90% of the file hasn’t
314 changed. Without a % sign, the number is to be read as a fraction,
315 with a decimal point before it. I.e., -M5 becomes 0.5, and is thus
316 the same as -M50%. Similarly, -M05 is the same as -M5%. To limit
317 detection to exact renames, use -M100%. The default similarity
318 index is 50%.
319 -C[<n>], --find-copies[=<n>]
321 Detect copies as well as renames. See also --find-copies-harder. If
322 n is specified, it has the same meaning as for -M<n>.
323 --find-copies-harder
325 For performance reasons, by default, -C option finds copies only if
326 the original file of the copy was modified in the same changeset.
327 This flag makes the command inspect unmodified files as candidates
328 for the source of copy. This is a very expensive operation for
329 large projects, so use it with caution. Giving more than one -C
330 option has the same effect.
331 -D, --irreversible-delete
333 Omit the preimage for deletes, i.e. print only the header but not
334 the diff between the preimage and /dev/null. The resulting patch is
335 not meant to be applied with patch or git apply; this is solely for
336 people who want to just concentrate on reviewing the text after the
337 change. In addition, the output obviously lacks enough information
338 to apply such a patch in reverse, even manually, hence the name of
339 the option.
340 When used together with -B, omit also the preimage in the deletion
342 part of a delete/create pair.
343 -l<num>
345 The -M and -C options involve some preliminary steps that can
346 detect subsets of renames/copies cheaply, followed by an exhaustive
347 fallback portion that compares all remaining unpaired destinations
348 to all relevant sources. (For renames, only remaining unpaired
349 sources are relevant; for copies, all original sources are
350 relevant.) For N sources and destinations, this exhaustive check is
351 O(N^2). This option prevents the exhaustive portion of rename/copy
352 detection from running if the number of source/destination files
353 involved exceeds the specified number. Defaults to
354 diff.renameLimit. Note that a value of 0 is treated as unlimited.
355 -O<orderfile>
357 Control the order in which files appear in the output. This
358 overrides the diff.orderFile configuration variable (see git-
359 config(1)). To cancel diff.orderFile, use -O/dev/null.
360 The output order is determined by the order of glob patterns in
362 <orderfile>. All files with pathnames that match the first pattern
363 are output first, all files with pathnames that match the second
364 pattern (but not the first) are output next, and so on. All files
365 with pathnames that do not match any pattern are output last, as if
366 there was an implicit match-all pattern at the end of the file. If
367 multiple pathnames have the same rank (they match the same pattern
368 but no earlier patterns), their output order relative to each other
369 is the normal order.
370 <orderfile> is parsed as follows:
372 • Blank lines are ignored, so they can be used as separators for
374 readability.
375 • Lines starting with a hash ("#") are ignored, so they can be
377 used for comments. Add a backslash ("\") to the beginning of
378 the pattern if it starts with a hash.
379 • Each other line contains a single pattern.
381 Patterns have the same syntax and semantics as patterns used for
383 fnmatch(3) without the FNM_PATHNAME flag, except a pathname also
384 matches a pattern if removing any number of the final pathname
385 components matches the pattern. For example, the pattern "foo*bar"
386 matches "fooasdfbar" and "foo/bar/baz/asdf" but not "foobarx".
387 --skip-to=<file>, --rotate-to=<file>
389 Discard the files before the named <file> from the output (i.e.
390 skip to), or move them to the end of the output (i.e. rotate to).
391 These options were invented primarily for the use of the git
392 difftool command, and may not be very useful otherwise.
393 --relative[=<path>], --no-relative
395 When run from a subdirectory of the project, it can be told to
396 exclude changes outside the directory and show pathnames relative
397 to it with this option. When you are not in a subdirectory (e.g. in
398 a bare repository), you can name which subdirectory to make the
399 output relative to by giving a <path> as an argument.
400 --no-relative can be used to countermand both diff.relative config
401 option and previous --relative.
402 -a, --text
404 Treat all files as text.
405 --ignore-cr-at-eol
407 Ignore carriage-return at the end of line when doing a comparison.
408 --ignore-space-at-eol
410 Ignore changes in whitespace at EOL.
411 -b, --ignore-space-change
413 Ignore changes in amount of whitespace. This ignores whitespace at
414 line end, and considers all other sequences of one or more
415 whitespace characters to be equivalent.
416 -w, --ignore-all-space
418 Ignore whitespace when comparing lines. This ignores differences
419 even if one line has whitespace where the other line has none.
420 --ignore-blank-lines
422 Ignore changes whose lines are all blank.
423 -I<regex>, --ignore-matching-lines=<regex>
425 Ignore changes whose all lines match <regex>. This option may be
426 specified more than once.
427 --inter-hunk-context=<lines>
429 Show the context between diff hunks, up to the specified number of
430 lines, thereby fusing hunks that are close to each other. Defaults
431 to diff.interHunkContext or 0 if the config option is unset.
432 -W, --function-context
434 Show whole function as context lines for each change. The function
435 names are determined in the same way as git diff works out patch
436 hunk headers (see Defining a custom hunk-header in
437 gitattributes(5)).
438 --ext-diff
440 Allow an external diff helper to be executed. If you set an
441 external diff driver with gitattributes(5), you need to use this
442 option with git-log(1) and friends.
443 --no-ext-diff
445 Disallow external diff drivers.
446 --textconv, --no-textconv
448 Allow (or disallow) external text conversion filters to be run when
449 comparing binary files. See gitattributes(5) for details. Because
450 textconv filters are typically a one-way conversion, the resulting
451 diff is suitable for human consumption, but cannot be applied. For
452 this reason, textconv filters are enabled by default only for git-
453 diff(1) and git-log(1), but not for git-format-patch(1) or diff
454 plumbing commands.
455 --ignore-submodules[=<when>]
457 Ignore changes to submodules in the diff generation. <when> can be
458 either "none", "untracked", "dirty" or "all", which is the default.
459 Using "none" will consider the submodule modified when it either
460 contains untracked or modified files or its HEAD differs from the
461 commit recorded in the superproject and can be used to override any
462 settings of the ignore option in git-config(1) or gitmodules(5).
463 When "untracked" is used submodules are not considered dirty when
464 they only contain untracked content (but they are still scanned for
465 modified content). Using "dirty" ignores all changes to the work
466 tree of submodules, only changes to the commits stored in the
467 superproject are shown (this was the behavior until 1.7.0). Using
468 "all" hides all changes to submodules.
469 --src-prefix=<prefix>
471 Show the given source prefix instead of "a/".
472 --dst-prefix=<prefix>
474 Show the given destination prefix instead of "b/".
475 --no-prefix
477 Do not show any source or destination prefix.
478 --default-prefix
480 Use the default source and destination prefixes ("a/" and "b/").
481 This is usually the default already, but may be used to override
482 config such as diff.noprefix.
483 --line-prefix=<prefix>
485 Prepend an additional prefix to every line of output.
486 --ita-invisible-in-index
488 By default entries added by "git add -N" appear as an existing
489 empty file in "git diff" and a new file in "git diff --cached".
490 This option makes the entry appear as a new file in "git diff" and
491 non-existent in "git diff --cached". This option could be reverted
492 with --ita-visible-in-index. Both options are experimental and
493 could be removed in future.
494 For more detailed explanation on these common options, see also
496 gitdiffcore(7).
497 -<n>
499 Prepare patches from the topmost <n> commits.
500 -o <dir>, --output-directory <dir>
502 Use <dir> to store the resulting files, instead of the current
503 working directory.
504 -n, --numbered
506 Name output in [PATCH n/m] format, even with a single patch.
507 -N, --no-numbered
509 Name output in [PATCH] format.
510 --start-number <n>
512 Start numbering the patches at <n> instead of 1.
513 --numbered-files
515 Output file names will be a simple number sequence without the
516 default first line of the commit appended.
517 -k, --keep-subject
519 Do not strip/add [PATCH] from the first line of the commit log
520 message.
521 -s, --signoff
523 Add a Signed-off-by trailer to the commit message, using the
524 committer identity of yourself. See the signoff option in git-
525 commit(1) for more information.
526 --stdout
528 Print all commits to the standard output in mbox format, instead of
529 creating a file for each one.
530 --attach[=<boundary>]
532 Create multipart/mixed attachment, the first part of which is the
533 commit message and the patch itself in the second part, with
534 Content-Disposition: attachment.
535 --no-attach
537 Disable the creation of an attachment, overriding the configuration
538 setting.
539 --inline[=<boundary>]
541 Create multipart/mixed attachment, the first part of which is the
542 commit message and the patch itself in the second part, with
543 Content-Disposition: inline.
544 --thread[=<style>], --no-thread
546 Controls addition of In-Reply-To and References headers to make the
547 second and subsequent mails appear as replies to the first. Also
548 controls generation of the Message-ID header to reference.
549 The optional <style> argument can be either shallow or deep.
551 shallow threading makes every mail a reply to the head of the
552 series, where the head is chosen from the cover letter, the
553 --in-reply-to, and the first patch mail, in this order. deep
554 threading makes every mail a reply to the previous one.
555 The default is --no-thread, unless the format.thread configuration
557 is set. --thread without an argument is equivalent to
558 --thread=shallow.
559 Beware that the default for git send-email is to thread emails
561 itself. If you want git format-patch to take care of threading, you
562 will want to ensure that threading is disabled for git send-email.
563 --in-reply-to=<message id>
565 Make the first mail (or all the mails with --no-thread) appear as a
566 reply to the given <message id>, which avoids breaking threads to
567 provide a new patch series.
568 --ignore-if-in-upstream
570 Do not include a patch that matches a commit in <until>..<since>.
571 This will examine all patches reachable from <since> but not from
572 <until> and compare them with the patches being generated, and any
573 patch that matches is ignored.
574 --always
576 Include patches for commits that do not introduce any change, which
577 are omitted by default.
578 --cover-from-description=<mode>
580 Controls which parts of the cover letter will be automatically
581 populated using the branch’s description.
582 If <mode> is message or default, the cover letter subject will be
584 populated with placeholder text. The body of the cover letter will
585 be populated with the branch’s description. This is the default
586 mode when no configuration nor command line option is specified.
587 If <mode> is subject, the first paragraph of the branch description
589 will populate the cover letter subject. The remainder of the
590 description will populate the body of the cover letter.
591 If <mode> is auto, if the first paragraph of the branch description
593 is greater than 100 bytes, then the mode will be message, otherwise
594 subject will be used.
595 If <mode> is none, both the cover letter subject and body will be
597 populated with placeholder text.
598 --description-file=<file>
600 Use the contents of <file> instead of the branch’s description for
601 generating the cover letter.
602 --subject-prefix=<subject prefix>
604 Instead of the standard [PATCH] prefix in the subject line, instead
605 use [<subject prefix>]. This can be used to name a patch series,
606 and can be combined with the --numbered option.
607 The configuration variable format.subjectPrefix may also be used to
609 configure a subject prefix to apply to a given repository for all
610 patches. This is often useful on mailing lists which receive
611 patches for several repositories and can be used to disambiguate
612 the patches (with a value of e.g. "PATCH my-project").
613 --filename-max-length=<n>
615 Instead of the standard 64 bytes, chomp the generated output
616 filenames at around <n> bytes (too short a value will be silently
617 raised to a reasonable length). Defaults to the value of the
618 format.filenameMaxLength configuration variable, or 64 if
619 unconfigured.
620 --rfc
622 Prepends "RFC" to the subject prefix (producing "RFC PATCH" by
623 default). RFC means "Request For Comments"; use this when sending
624 an experimental patch for discussion rather than application.
625 -v <n>, --reroll-count=<n>
627 Mark the series as the <n>-th iteration of the topic. The output
628 filenames have v<n> prepended to them, and the subject prefix
629 ("PATCH" by default, but configurable via the --subject-prefix
630 option) has ` v<n>` appended to it. E.g. --reroll-count=4 may
631 produce v4-0001-add-makefile.patch file that has "Subject: [PATCH
632 v4 1/20] Add makefile" in it. <n> does not have to be an integer
633 (e.g. "--reroll-count=4.4", or "--reroll-count=4rev2" are allowed),
634 but the downside of using such a reroll-count is that the
635 range-diff/interdiff with the previous version does not state
636 exactly which version the new iteration is compared against.
637 --to=<email>
639 Add a To: header to the email headers. This is in addition to any
640 configured headers, and may be used multiple times. The negated
641 form --no-to discards all To: headers added so far (from config or
642 command line).
643 --cc=<email>
645 Add a Cc: header to the email headers. This is in addition to any
646 configured headers, and may be used multiple times. The negated
647 form --no-cc discards all Cc: headers added so far (from config or
648 command line).
649 --from, --from=<ident>
651 Use ident in the From: header of each commit email. If the author
652 ident of the commit is not textually identical to the provided
653 ident, place a From: header in the body of the message with the
654 original author. If no ident is given, use the committer ident.
655 Note that this option is only useful if you are actually sending
657 the emails and want to identify yourself as the sender, but retain
658 the original author (and git am will correctly pick up the in-body
659 header). Note also that git send-email already handles this
660 transformation for you, and this option should not be used if you
661 are feeding the result to git send-email.
662 --[no-]force-in-body-from
664 With the e-mail sender specified via the --from option, by default,
665 an in-body "From:" to identify the real author of the commit is
666 added at the top of the commit log message if the sender is
667 different from the author. With this option, the in-body "From:" is
668 added even when the sender and the author have the same name and
669 address, which may help if the mailing list software mangles the
670 sender’s identity. Defaults to the value of the
671 format.forceInBodyFrom configuration variable.
672 --add-header=<header>
674 Add an arbitrary header to the email headers. This is in addition
675 to any configured headers, and may be used multiple times. For
676 example, --add-header="Organization: git-foo". The negated form
677 --no-add-header discards all (To:, Cc:, and custom) headers added
678 so far from config or command line.
679 --[no-]cover-letter
681 In addition to the patches, generate a cover letter file containing
682 the branch description, shortlog and the overall diffstat. You can
683 fill in a description in the file before sending it out.
684 --encode-email-headers, --no-encode-email-headers
686 Encode email headers that have non-ASCII characters with
687 "Q-encoding" (described in RFC 2047), instead of outputting the
688 headers verbatim. Defaults to the value of the
689 format.encodeEmailHeaders configuration variable.
690 --interdiff=<previous>
692 As a reviewer aid, insert an interdiff into the cover letter, or as
693 commentary of the lone patch of a 1-patch series, showing the
694 differences between the previous version of the patch series and
695 the series currently being formatted. previous is a single
696 revision naming the tip of the previous series which shares a
697 common base with the series being formatted (for example git
698 format-patch --cover-letter --interdiff=feature/v1 -3 feature/v2).
699 --range-diff=<previous>
701 As a reviewer aid, insert a range-diff (see git-range-diff(1)) into
702 the cover letter, or as commentary of the lone patch of a 1-patch
703 series, showing the differences between the previous version of the
704 patch series and the series currently being formatted. previous
705 can be a single revision naming the tip of the previous series if
706 it shares a common base with the series being formatted (for
707 example git format-patch --cover-letter --range-diff=feature/v1 -3
708 feature/v2), or a revision range if the two versions of the series
709 are disjoint (for example git format-patch --cover-letter
710 --range-diff=feature/v1~3..feature/v1 -3 feature/v2).
711 Note that diff options passed to the command affect how the primary
713 product of format-patch is generated, and they are not passed to
714 the underlying range-diff machinery used to generate the
715 cover-letter material (this may change in the future).
716 --creation-factor=<percent>
718 Used with --range-diff, tweak the heuristic which matches up
719 commits between the previous and current series of patches by
720 adjusting the creation/deletion cost fudge factor. See git-range-
721 diff(1)) for details.
722 --notes[=<ref>], --no-notes
724 Append the notes (see git-notes(1)) for the commit after the
725 three-dash line.
726 The expected use case of this is to write supporting explanation
728 for the commit that does not belong to the commit log message
729 proper, and include it with the patch submission. While one can
730 simply write these explanations after format-patch has run but
731 before sending, keeping them as Git notes allows them to be
732 maintained between versions of the patch series (but see the
733 discussion of the notes.rewrite configuration options in git-
734 notes(1) to use this workflow).
735 The default is --no-notes, unless the format.notes configuration is
737 set.
738 --[no-]signature=<signature>
740 Add a signature to each message produced. Per RFC 3676 the
741 signature is separated from the body by a line with '-- ' on it. If
742 the signature option is omitted the signature defaults to the Git
743 version number.
744 --signature-file=<file>
746 Works just like --signature except the signature is read from a
747 file.
748 --suffix=.<sfx>
750 Instead of using .patch as the suffix for generated filenames, use
751 specified suffix. A common alternative is --suffix=.txt. Leaving
752 this empty will remove the .patch suffix.
753 Note that the leading character does not have to be a dot; for
755 example, you can use --suffix=-patch to get
756 0001-description-of-my-change-patch.
757 -q, --quiet
759 Do not print the names of the generated files to standard output.
760 --no-binary
762 Do not output contents of changes in binary files, instead display
763 a notice that those files changed. Patches generated using this
764 option cannot be applied properly, but they are still useful for
765 code review.
766 --zero-commit
768 Output an all-zero hash in each patch’s From header instead of the
769 hash of the commit.
770 --[no-]base[=<commit>]
772 Record the base tree information to identify the state the patch
773 series applies to. See the BASE TREE INFORMATION section below for
774 details. If <commit> is "auto", a base commit is automatically
775 chosen. The --no-base option overrides a format.useAutoBase
776 configuration.
777 --root
779 Treat the revision argument as a <revision range>, even if it is
780 just a single commit (that would normally be treated as a <since>).
781 Note that root commits included in the specified range are always
782 formatted as creation patches, independently of this flag.
783 --progress
785 Show progress reports on stderr as patches are generated.
786
788 You can specify extra mail header lines to be added to each message,
789 defaults for the subject prefix and file suffix, number patches when
790 outputting more than one patch, add "To:" or "Cc:" headers, configure
791 attachments, change the patch output directory, and sign off patches
792 with configuration variables.
793 [format]
795 headers = "Organization: git-foo\n"
796 subjectPrefix = CHANGE
797 suffix = .txt
798 numbered = auto
799 to = <email>
800 cc = <email>
801 attach [ = mime-boundary-string ]
802 signOff = true
803 outputDirectory = <directory>
804 coverLetter = auto
805 coverFromDescription = auto
806
808 The patch produced by git format-patch is in UNIX mailbox format, with
809 a fixed "magic" time stamp to indicate that the file is output from
810 format-patch rather than a real mailbox, like so:
811 From 8f72bad1baf19a53459661343e21d6491c3908d3 Mon Sep 17 00:00:00 2001
813 From: Tony Luck <tony.luck@intel.com>
814 Date: Tue, 13 Jul 2010 11:42:54 -0700
815 Subject: [PATCH] =?UTF-8?q?[IA64]=20Put=20ia64=20config=20files=20on=20the=20?=
816 =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig=20diet?=
817 MIME-Version: 1.0
818 Content-Type: text/plain; charset=UTF-8
819 Content-Transfer-Encoding: 8bit
820 arch/arm config files were slimmed down using a python script
822 (See commit c2330e286f68f1c408b4aa6515ba49d57f05beae comment)
823 Do the same for ia64 so we can have sleek & trim looking
825 ...
826 Typically it will be placed in a MUA’s drafts folder, edited to add
828 timely commentary that should not go in the changelog after the three
829 dashes, and then sent as a message whose body, in our example, starts
830 with "arch/arm config files were...". On the receiving end, readers can
831 save interesting patches in a UNIX mailbox and apply them with git-
832 am(1).
833 When a patch is part of an ongoing discussion, the patch generated by
835 git format-patch can be tweaked to take advantage of the git am
836 --scissors feature. After your response to the discussion comes a line
837 that consists solely of "-- >8 --" (scissors and perforation), followed
838 by the patch with unnecessary header fields removed:
839 ...
841 > So we should do such-and-such.
842 Makes sense to me. How about this patch?
844 -- >8 --
846 Subject: [IA64] Put ia64 config files on the Uwe Kleine-König diet
847 arch/arm config files were slimmed down using a python script
849 ...
850 When sending a patch this way, most often you are sending your own
852 patch, so in addition to the "From $SHA1 $magic_timestamp" marker you
853 should omit From: and Date: lines from the patch file. The patch title
854 is likely to be different from the subject of the discussion the patch
855 is in response to, so it is likely that you would want to keep the
856 Subject: line, like the example above.
857 Checking for patch corruption
859 Many mailers if not set up properly will corrupt whitespace. Here are
860 two common types of corruption:
861 • Empty context lines that do not have any whitespace.
863 • Non-empty context lines that have one extra whitespace at the
865 beginning.
866 One way to test if your MUA is set up correctly is:
868 • Send the patch to yourself, exactly the way you would, except with
870 To: and Cc: lines that do not contain the list and maintainer
871 address.
872 • Save that patch to a file in UNIX mailbox format. Call it a.patch,
874 say.
875 • Apply it:
877 $ git fetch <project> master:test-apply
879 $ git switch test-apply
880 $ git restore --source=HEAD --staged --worktree :/
881 $ git am a.patch
882 If it does not apply correctly, there can be various reasons.
884 • The patch itself does not apply cleanly. That is bad but does not
886 have much to do with your MUA. You might want to rebase the patch
887 with git-rebase(1) before regenerating it in this case.
888 • The MUA corrupted your patch; "am" would complain that the patch
890 does not apply. Look in the .git/rebase-apply/ subdirectory and see
891 what patch file contains and check for the common corruption
892 patterns mentioned above.
893 • While at it, check the info and final-commit files as well. If what
895 is in final-commit is not exactly what you would want to see in the
896 commit log message, it is very likely that the receiver would end
897 up hand editing the log message when applying your patch. Things
898 like "Hi, this is my first patch.\n" in the patch e-mail should
899 come after the three-dash line that signals the end of the commit
900 message.
901
903 Here are some hints on how to successfully submit patches inline using
904 various mailers.
905 GMail
907 GMail does not have any way to turn off line wrapping in the web
908 interface, so it will mangle any emails that you send. You can however
909 use "git send-email" and send your patches through the GMail SMTP
910 server, or use any IMAP email client to connect to the google IMAP
911 server and forward the emails through that.
912 For hints on using git send-email to send your patches through the
914 GMail SMTP server, see the EXAMPLE section of git-send-email(1).
915 For hints on submission using the IMAP interface, see the EXAMPLE
917 section of git-imap-send(1).
918 Thunderbird
920 By default, Thunderbird will both wrap emails as well as flag them as
921 being format=flowed, both of which will make the resulting email
922 unusable by Git.
923 There are three different approaches: use an add-on to turn off line
925 wraps, configure Thunderbird to not mangle patches, or use an external
926 editor to keep Thunderbird from mangling the patches.
927 Approach #1 (add-on)
929 Install the Toggle Word Wrap add-on that is available from
930 https://addons.mozilla.org/thunderbird/addon/toggle-word-wrap/ It
931 adds a menu entry "Enable Word Wrap" in the composer’s "Options"
932 menu that you can tick off. Now you can compose the message as you
933 otherwise do (cut + paste, git format-patch | git imap-send, etc),
934 but you have to insert line breaks manually in any text that you
935 type.
936 Approach #2 (configuration)
938 Three steps:
939 1. Configure your mail server composition as plain text: Edit...
941 Account Settings...Composition & Addressing, uncheck "Compose
942 Messages in HTML".
943 2. Configure your general composition window to not wrap.
945 In Thunderbird 2: Edit..Preferences..Composition, wrap plain
947 text messages at 0
948 In Thunderbird 3: Edit..Preferences..Advanced..Config Editor.
950 Search for "mail.wrap_long_lines". Toggle it to make sure it is
951 set to false. Also, search for "mailnews.wraplength" and set
952 the value to 0.
953 3. Disable the use of format=flowed:
955 Edit..Preferences..Advanced..Config Editor. Search for
956 "mailnews.send_plaintext_flowed". Toggle it to make sure it is
957 set to false.
958 After that is done, you should be able to compose email as you
960 otherwise would (cut + paste, git format-patch | git imap-send,
961 etc), and the patches will not be mangled.
962 Approach #3 (external editor)
964 The following Thunderbird extensions are needed: AboutConfig from
965 http://aboutconfig.mozdev.org/ and External Editor from
966 http://globs.org/articles.php?lng=en&pg=8
967 1. Prepare the patch as a text file using your method of choice.
969 2. Before opening a compose window, use Edit→Account Settings to
971 uncheck the "Compose messages in HTML format" setting in the
972 "Composition & Addressing" panel of the account to be used to
973 send the patch.
974 3. In the main Thunderbird window, before you open the compose
976 window for the patch, use Tools→about:config to set the
977 following to the indicated values:
978 mailnews.send_plaintext_flowed => false
980 mailnews.wraplength => 0
981 4. Open a compose window and click the external editor icon.
983 5. In the external editor window, read in the patch file and exit
985 the editor normally.
986 Side note: it may be possible to do step 2 with about:config and
988 the following settings but no one’s tried yet.
989 mail.html_compose => false
991 mail.identity.default.compose_html => false
992 mail.identity.id?.compose_html => false
993 There is a script in contrib/thunderbird-patch-inline which can
995 help you include patches with Thunderbird in an easy way. To use
996 it, do the steps above and then use the script as the external
997 editor.
998 KMail
1000 This should help you to submit patches inline using KMail.
1001 1. Prepare the patch as a text file.
1003 2. Click on New Mail.
1005 3. Go under "Options" in the Composer window and be sure that "Word
1007 wrap" is not set.
1008 4. Use Message → Insert file... and insert the patch.
1010 5. Back in the compose window: add whatever other text you wish to the
1012 message, complete the addressing and subject fields, and press
1013 send.
1014
1016 The base tree information block is used for maintainers or third party
1017 testers to know the exact state the patch series applies to. It
1018 consists of the base commit, which is a well-known commit that is part
1019 of the stable part of the project history everybody else works off of,
1020 and zero or more prerequisite patches, which are well-known patches in
1021 flight that is not yet part of the base commit that need to be applied
1022 on top of base commit in topological order before the patches can be
1023 applied.
1024 The base commit is shown as "base-commit: " followed by the 40-hex of
1026 the commit object name. A prerequisite patch is shown as
1027 "prerequisite-patch-id: " followed by the 40-hex patch id, which can be
1028 obtained by passing the patch through the git patch-id --stable
1029 command.
1030 Imagine that on top of the public commit P, you applied well-known
1032 patches X, Y and Z from somebody else, and then built your three-patch
1033 series A, B, C, the history would be like:
1034 ---P---X---Y---Z---A---B---C
1036 With git format-patch --base=P -3 C (or variants thereof, e.g. with
1038 --cover-letter or using Z..C instead of -3 C to specify the range), the
1039 base tree information block is shown at the end of the first message
1040 the command outputs (either the first patch, or the cover letter), like
1041 this:
1042 base-commit: P
1044 prerequisite-patch-id: X
1045 prerequisite-patch-id: Y
1046 prerequisite-patch-id: Z
1047 For non-linear topology, such as
1049 ---P---X---A---M---C
1051 \ /
1052 Y---Z---B
1053 You can also use git format-patch --base=P -3 C to generate patches for
1055 A, B and C, and the identifiers for P, X, Y, Z are appended at the end
1056 of the first message.
1057 If set --base=auto in cmdline, it will automatically compute the base
1059 commit as the merge base of tip commit of the remote-tracking branch
1060 and revision-range specified in cmdline. For a local branch, you need
1061 to make it to track a remote branch by git branch --set-upstream-to
1062 before using this option.
1063
1065 • Extract commits between revisions R1 and R2, and apply them on top
1066 of the current branch using git am to cherry-pick them:
1067 $ git format-patch -k --stdout R1..R2 | git am -3 -k
1069 • Extract all commits which are in the current branch but not in the
1071 origin branch:
1072 $ git format-patch origin
1074 For each commit a separate file is created in the current
1076 directory.
1077 • Extract all commits that lead to origin since the inception of the
1079 project:
1080 $ git format-patch --root origin
1082 • The same as the previous one:
1084 $ git format-patch -M -B origin
1086 Additionally, it detects and handles renames and complete rewrites
1088 intelligently to produce a renaming patch. A renaming patch reduces
1089 the amount of text output, and generally makes it easier to review.
1090 Note that non-Git "patch" programs won’t understand renaming
1091 patches, so use it only when you know the recipient uses Git to
1092 apply your patch.
1093 • Extract three topmost commits from the current branch and format
1095 them as e-mailable patches:
1096 $ git format-patch -3
1098
1100 Note that format-patch will omit merge commits from the output, even if
1101 they are part of the requested range. A simple "patch" does not include
1102 enough information for the receiving end to reproduce the same merge
1103 commit.
1104
1106 git-am(1), git-send-email(1)
1107
1109 Part of the git(1) suite
1110Git 2.43.0 11/20/2023 GIT-FORMAT-PATCH(1)