Documentation: filter-branch: show --ignore-unmatch in main index-filter example
[git] / Documentation / SubmittingPatches
CommitLineData
56333bac
JS
1Checklist (and a short version for the impatient):
2
a7af09d2
JA
3 Commits:
4
56333bac
JS
5 - make commits of logical units
6 - check for unnecessary whitespace with "git diff --check"
7 before committing
8 - do not check in commented out code or unneeded files
9 - provide a meaningful commit message
10 - the first line of the commit message should be a short
11 description and should skip the full stop
12 - if you want your work included in git.git, add a
8e7425da 13 "Signed-off-by: Your Name <you@example.com>" line to the
56333bac
JS
14 commit message (or just use the option "-s" when
15 committing) to confirm that you agree to the Developer's
16 Certificate of Origin
d3017e93
JS
17 - make sure that you have tests for the bug you are fixing
18 - make sure that the test suite passes after your commit
a7af09d2
JA
19
20 Patch:
21
56333bac 22 - use "git format-patch -M" to create the patch
a7af09d2 23 - do not PGP sign your patch
56333bac
JS
24 - do not attach your patch, but read in the mail
25 body, unless you cannot teach your mailer to
26 leave the formatting of the patch alone.
27 - be careful doing cut & paste into your mailer, not to
28 corrupt whitespaces.
29 - provide additional information (which is unsuitable for
30 the commit message) between the "---" and the diffstat
15320175
AR
31 - if you change, add, or remove a command line option or
32 make some other user interface change, the associated
33 documentation should be updated as well.
d3017e93
JS
34 - if your name is not writable in ASCII, make sure that
35 you send off a message in the correct encoding.
13d4e6f7 36 - send the patch to the list (git@vger.kernel.org) and the
0b059940
JH
37 maintainer (gitster@pobox.com) if (and only if) the patch
38 is ready for inclusion. If you use git-send-email(1),
39 please test it first by sending email to yourself.
56333bac
JS
40
41Long version:
42
31408251
JH
43I started reading over the SubmittingPatches document for Linux
44kernel, primarily because I wanted to have a document similar to
45it for the core GIT to make sure people understand what they are
46doing when they write "Signed-off-by" line.
47
48But the patch submission requirements are a lot more relaxed
45d2b286
JH
49here on the technical/contents front, because the core GIT is
50thousand times smaller ;-). So here is only the relevant bits.
31408251
JH
51
52
53(1) Make separate commits for logically separate changes.
54
55Unless your patch is really trivial, you should not be sending
56out a patch that was generated between your working tree and
57your commit head. Instead, always make a commit with complete
58commit message and generate a series of patches from your
59repository. It is a good discipline.
60
61Describe the technical detail of the change(s).
62
45d2b286 63If your description starts to get too long, that's a sign that you
31408251
JH
64probably need to split up your commit to finer grained pieces.
65
45d2b286
JH
66Oh, another thing. I am picky about whitespaces. Make sure your
67changes do not trigger errors with the sample pre-commit hook shipped
16507fcf
BL
68in templates/hooks--pre-commit. To help ensure this does not happen,
69run git diff --check on your changes before you commit.
31408251 70
31408251 71
243bfd33
JS
72(1a) Try to be nice to older C compilers
73
8b1d88e8 74We try to support a wide range of C compilers to compile
243bfd33
JS
75git with. That means that you should not use C99 initializers, even
76if a lot of compilers grok it.
77
78Also, variables have to be declared at the beginning of the block
79(you can check this with gcc, using the -Wdeclaration-after-statement
80option).
81
82Another thing: NULL pointers shall be written as NULL, not as 0.
83
84
45d2b286
JH
85(2) Generate your patch using git tools out of your commits.
86
87git based diff tools (git, Cogito, and StGIT included) generate
88unidiff which is the preferred format.
89
31408251
JH
90You do not have to be afraid to use -M option to "git diff" or
91"git format-patch", if your patch involves file renames. The
92receiving end can handle them just fine.
93
94Please make sure your patch does not include any extra files
95which do not belong in a patch submission. Make sure to review
96your patch after generating it, to ensure accuracy. Before
97sending out, please make sure it cleanly applies to the "master"
45d2b286
JH
98branch head. If you are preparing a work based on "next" branch,
99that is fine, but please mark it as such.
31408251
JH
100
101
102(3) Sending your patches.
103
45d2b286 104People on the git mailing list need to be able to read and
31408251
JH
105comment on the changes you are submitting. It is important for
106a developer to be able to "quote" your changes, using standard
107e-mail tools, so that they may comment on specific portions of
addf88e4 108your code. For this reason, all patches should be submitted
45d2b286
JH
109"inline". WARNING: Be wary of your MUAs word-wrap
110corrupting your patch. Do not cut-n-paste your patch; you can
111lose tabs that way if you are not careful.
31408251 112
45d2b286 113It is a common convention to prefix your subject line with
31408251 114[PATCH]. This lets people easily distinguish patches from other
4e891acf
JH
115e-mail discussions. Use of additional markers after PATCH and
116the closing bracket to mark the nature of the patch is also
117encouraged. E.g. [PATCH/RFC] is often used when the patch is
118not ready to be applied but it is for discussion, [PATCH v2],
119[PATCH v3] etc. are often seen when you are sending an update to
120what you have previously sent.
31408251
JH
121
122"git format-patch" command follows the best current practice to
123format the body of an e-mail message. At the beginning of the
124patch should come your commit message, ending with the
125Signed-off-by: lines, and a line that consists of three dashes,
126followed by the diffstat information and the patch itself. If
127you are forwarding a patch from somebody else, optionally, at
128the beginning of the e-mail message just before the commit
129message starts, you can put a "From: " line to name that person.
130
131You often want to add additional explanation about the patch,
132other than the commit message itself. Place such "cover letter"
133material between the three dash lines and the diffstat.
134
135Do not attach the patch as a MIME attachment, compressed or not.
e30b217b
JH
136Do not let your e-mail client send quoted-printable. Do not let
137your e-mail client send format=flowed which would destroy
138whitespaces in your patches. Many
31408251
JH
139popular e-mail applications will not always transmit a MIME
140attachment as plain text, making it impossible to comment on
141your code. A MIME attachment also takes a bit more time to
142process. This does not decrease the likelihood of your
143MIME-attached change being accepted, but it makes it more likely
144that it will be postponed.
145
146Exception: If your mailer is mangling patches then someone may ask
9847f7e0 147you to re-send them using MIME, that is OK.
31408251 148
9847f7e0
JH
149Do not PGP sign your patch, at least for now. Most likely, your
150maintainer or other people on the list would not have your PGP
151key and would not bother obtaining it anyway. Your patch is not
152judged by who you are; a good patch from an unknown origin has a
153far better chance of being accepted than a patch from a known,
154respected origin that is done poorly or does incorrect things.
155
156If you really really really really want to do a PGP signed
157patch, format it as "multipart/signed", not a text/plain message
158that starts with '-----BEGIN PGP SIGNED MESSAGE-----'. That is
159not a text/plain, it's something else.
160
161Note that your maintainer does not necessarily read everything
162on the git mailing list. If your patch is for discussion first,
163send it "To:" the mailing list, and optionally "cc:" him. If it
164is trivially correct or after the list reached a consensus, send
0b059940
JH
165it "To:" the maintainer and optionally "cc:" the list for
166inclusion.
31408251 167
04d24455
JH
168Also note that your maintainer does not actively involve himself in
169maintaining what are in contrib/ hierarchy. When you send fixes and
170enhancements to them, do not forget to "cc: " the person who primarily
171worked on that hierarchy in contrib/.
172
31408251 173
84ab7b6f 174(4) Sign your work
31408251
JH
175
176To improve tracking of who did what, we've borrowed the
177"sign-off" procedure from the Linux kernel project on patches
178that are being emailed around. Although core GIT is a lot
179smaller project it is a good discipline to follow it.
180
181The sign-off is a simple line at the end of the explanation for
182the patch, which certifies that you wrote it or otherwise have
183the right to pass it on as a open-source patch. The rules are
184pretty simple: if you can certify the below:
185
186 Developer's Certificate of Origin 1.1
187
188 By making a contribution to this project, I certify that:
189
190 (a) The contribution was created in whole or in part by me and I
191 have the right to submit it under the open source license
192 indicated in the file; or
193
194 (b) The contribution is based upon previous work that, to the best
195 of my knowledge, is covered under an appropriate open source
196 license and I have the right under that license to submit that
197 work with modifications, whether created in whole or in part
198 by me, under the same open source license (unless I am
199 permitted to submit under a different license), as indicated
200 in the file; or
201
202 (c) The contribution was provided directly to me by some other
203 person who certified (a), (b) or (c) and I have not modified
204 it.
205
206 (d) I understand and agree that this project and the contribution
207 are public and that a record of the contribution (including all
208 personal information I submit with it, including my sign-off) is
209 maintained indefinitely and may be redistributed consistent with
210 this project or the open source license(s) involved.
211
212then you just add a line saying
213
214 Signed-off-by: Random J Developer <random@developer.example.org>
215
69945602
PC
216This line can be automatically added by git if you run the git-commit
217command with the -s option.
218
c11c3b56
JH
219Notice that you can place your own Signed-off-by: line when
220forwarding somebody else's patch with the above rules for
221D-C-O. Indeed you are encouraged to do so. Do not forget to
222place an in-body "From: " line at the beginning to properly attribute
223the change to its true author (see (2) above).
224
67275247
MV
225Also notice that a real name is used in the Signed-off-by: line. Please
226don't hide your real name.
227
c11c3b56
JH
228Some people also put extra tags at the end.
229
230"Acked-by:" says that the patch was reviewed by the person who
231is more familiar with the issues and the area the patch attempts
232to modify. "Tested-by:" says the patch was tested by the person
233and found to have the desired effect.
9740d289 234
a941fb4a
JH
235------------------------------------------------
236An ideal patch flow
237
238Here is an ideal patch flow for this project the current maintainer
239suggests to the contributors:
240
241 (0) You come up with an itch. You code it up.
242
243 (1) Send it to the list and cc people who may need to know about
244 the change.
245
246 The people who may need to know are the ones whose code you
247 are butchering. These people happen to be the ones who are
248 most likely to be knowledgeable enough to help you, but
249 they have no obligation to help you (i.e. you ask for help,
250 don't demand). "git log -p -- $area_you_are_modifying" would
251 help you find out who they are.
252
253 (2) You get comments and suggestions for improvements. You may
254 even get them in a "on top of your change" patch form.
255
256 (3) Polish, refine, and re-send to the list and the people who
257 spend their time to improve your patch. Go back to step (2).
258
259 (4) The list forms consensus that the last round of your patch is
260 good. Send it to the list and cc the maintainer.
261
262 (5) A topic branch is created with the patch and is merged to 'next',
263 and cooked further and eventually graduates to 'master'.
264
265In any time between the (2)-(3) cycle, the maintainer may pick it up
266from the list and queue it to 'pu', in order to make it easier for
267people play with it without having to pick up and apply the patch to
268their trees themselves.
269
9740d289
JH
270------------------------------------------------
271MUA specific hints
272
273Some of patches I receive or pick up from the list share common
274patterns of breakage. Please make sure your MUA is set up
275properly not to corrupt whitespaces. Here are two common ones
276I have seen:
277
278* Empty context lines that do not have _any_ whitespace.
279
280* Non empty context lines that have one extra whitespace at the
281 beginning.
282
9847f7e0
JH
283One test you could do yourself if your MUA is set up correctly is:
284
285* Send the patch to yourself, exactly the way you would, except
286 To: and Cc: lines, which would not contain the list and
287 maintainer address.
288
289* Save that patch to a file in UNIX mailbox format. Call it say
290 a.patch.
291
292* Try to apply to the tip of the "master" branch from the
293 git.git public repository:
294
295 $ git fetch http://kernel.org/pub/scm/git/git.git master:test-apply
296 $ git checkout test-apply
297 $ git reset --hard
59c8e2cb 298 $ git am a.patch
9847f7e0
JH
299
300If it does not apply correctly, there can be various reasons.
301
302* Your patch itself does not apply cleanly. That is _bad_ but
303 does not have much to do with your MUA. Please rebase the
304 patch appropriately.
305
59c8e2cb 306* Your MUA corrupted your patch; "am" would complain that
51ef1daa 307 the patch does not apply. Look at .git/rebase-apply/ subdirectory and
9847f7e0
JH
308 see what 'patch' file contains and check for the common
309 corruption patterns mentioned above.
310
311* While you are at it, check what are in 'info' and
312 'final-commit' files as well. If what is in 'final-commit' is
313 not exactly what you would want to see in the commit log
314 message, it is very likely that your maintainer would end up
315 hand editing the log message when he applies your patch.
316 Things like "Hi, this is my first patch.\n", if you really
317 want to put in the patch e-mail, should come after the
318 three-dash line that signals the end of the commit message.
319
9740d289
JH
320
321Pine
322----
323
324(Johannes Schindelin)
325
326I don't know how many people still use pine, but for those poor
327souls it may be good to mention that the quell-flowed-text is
328needed for recent versions.
329
330... the "no-strip-whitespace-before-send" option, too. AFAIK it
331was introduced in 4.60.
332
333(Linus Torvalds)
334
335And 4.58 needs at least this.
336
337---
338diff-tree 8326dd8350be64ac7fc805f6563a1d61ad10d32c (from e886a61f76edf5410573e92e38ce22974f9c40f1)
339Author: Linus Torvalds <torvalds@g5.osdl.org>
340Date: Mon Aug 15 17:23:51 2005 -0700
341
342 Fix pine whitespace-corruption bug
343
344 There's no excuse for unconditionally removing whitespace from
345 the pico buffers on close.
346
347diff --git a/pico/pico.c b/pico/pico.c
348--- a/pico/pico.c
349+++ b/pico/pico.c
350@@ -219,7 +219,9 @@ PICO *pm;
a6080a0a
JH
351 switch(pico_all_done){ /* prepare for/handle final events */
352 case COMP_EXIT : /* already confirmed */
353 packheader();
9740d289 354+#if 0
a6080a0a 355 stripwhitespace();
9740d289 356+#endif
a6080a0a
JH
357 c |= COMP_EXIT;
358 break;
359
9740d289 360
1eb446fa
JH
361(Daniel Barkalow)
362
363> A patch to SubmittingPatches, MUA specific help section for
364> users of Pine 4.63 would be very much appreciated.
365
366Ah, it looks like a recent version changed the default behavior to do the
367right thing, and inverted the sense of the configuration option. (Either
368that or Gentoo did it.) So you need to set the
369"no-strip-whitespace-before-send" option, unless the option you have is
370"strip-whitespace-before-send", in which case you should avoid checking
371it.
372
9740d289
JH
373
374Thunderbird
375-----------
376
377(A Large Angry SCM)
378
1a526d48
JW
379By default, Thunderbird will both wrap emails as well as flag them as
380being 'format=flowed', both of which will make the resulting email unusable
381by git.
382
9740d289 383Here are some hints on how to successfully submit patches inline using
cf6de18a 384Thunderbird.
9740d289 385
1a526d48
JW
386There are two different approaches. One approach is to configure
387Thunderbird to not mangle patches. The second approach is to use
388an external editor to keep Thunderbird from mangling the patches.
389
390Approach #1 (configuration):
391
392This recipe is current as of Thunderbird 2.0.0.19. Three steps:
393 1. Configure your mail server composition as plain text
394 Edit...Account Settings...Composition & Addressing,
395 uncheck 'Compose Messages in HTML'.
396 2. Configure your general composition window to not wrap
397 Edit..Preferences..Composition, wrap plain text messages at 0
398 3. Disable the use of format=flowed
399 Edit..Preferences..Advanced..Config Editor. Search for:
400 mailnews.send_plaintext_flowed
401 toggle it to make sure it is set to 'false'.
402
403After that is done, you should be able to compose email as you
404otherwise would (cut + paste, git-format-patch | git-imap-send, etc),
405and the patches should not be mangled.
406
407Approach #2 (external editor):
408
9740d289
JH
409This recipe appears to work with the current [*1*] Thunderbird from Suse.
410
411The following Thunderbird extensions are needed:
412 AboutConfig 0.5
413 http://aboutconfig.mozdev.org/
ff62b7f3
LS
414 External Editor 0.7.2
415 http://globs.org/articles.php?lng=en&pg=8
9740d289
JH
416
4171) Prepare the patch as a text file using your method of choice.
418
4192) Before opening a compose window, use Edit->Account Settings to
420uncheck the "Compose messages in HTML format" setting in the
421"Composition & Addressing" panel of the account to be used to send the
422patch. [*2*]
423
4243) In the main Thunderbird window, _before_ you open the compose window
425for the patch, use Tools->about:config to set the following to the
426indicated values:
427 mailnews.send_plaintext_flowed => false
cf6de18a 428 mailnews.wraplength => 0
9740d289
JH
429
4304) Open a compose window and click the external editor icon.
431
4325) In the external editor window, read in the patch file and exit the
433editor normally.
434
4356) Back in the compose window: Add whatever other text you wish to the
436message, complete the addressing and subject fields, and press send.
437
4387) Optionally, undo the about:config/account settings changes made in
439steps 2 & 3.
440
441
442[Footnotes]
443*1* Version 1.0 (20041207) from the MozillaThunderbird-1.0-5 rpm of Suse
4449.3 professional updates.
445
446*2* It may be possible to do this with about:config and the following
447settings but I haven't tried, yet.
448 mail.html_compose => false
449 mail.identity.default.compose_html => false
450 mail.identity.id?.compose_html => false
451
0c3d26d2
LS
452(Lukas Sandström)
453
454There is a script in contrib/thunderbird-patch-inline which can help
455you include patches with Thunderbird in an easy way. To use it, do the
456steps above and then use the script as the external editor.
e30b217b 457
e30b217b
JH
458Gnus
459----
460
461'|' in the *Summary* buffer can be used to pipe the current
462message to an external program, and this is a handy way to drive
463"git am". However, if the message is MIME encoded, what is
464piped into the program is the representation you see in your
465*Article* buffer after unwrapping MIME. This is often not what
466you would want for two reasons. It tends to screw up non ASCII
467characters (most notably in people's names), and also
468whitespaces (fatal in patches). Running 'C-u g' to display the
469message in raw form before using '|' to run the pipe can work
470this problem around.
471
451fd65a
M
472
473KMail
474-----
475
476This should help you to submit patches inline using KMail.
477
4781) Prepare the patch as a text file.
479
4802) Click on New Mail.
481
4823) Go under "Options" in the Composer window and be sure that
483"Word wrap" is not set.
484
4854) Use Message -> Insert file... and insert the patch.
486
4875) Back in the compose window: add whatever other text you wish to the
488message, complete the addressing and subject fields, and press send.
c2163c6a
TPW
489
490
491Gmail
492-----
493
50dffd4e
JT
494GMail does not appear to have any way to turn off line wrapping in the web
495interface, so this will mangle any emails that you send. You can however
496use any IMAP email client to connect to the google imap server, and forward
497the emails through that. Just make sure to disable line wrapping in that
498email client. Alternatively, use "git send-email" instead.
499
c2163c6a
TPW
500Submitting properly formatted patches via Gmail is simple now that
501IMAP support is available. First, edit your ~/.gitconfig to specify your
502account settings:
503
504[imap]
505 folder = "[Gmail]/Drafts"
506 host = imaps://imap.gmail.com
507 user = user@gmail.com
508 pass = p4ssw0rd
509 port = 993
510 sslverify = false
511
50dffd4e
JT
512You might need to instead use: folder = "[Google Mail]/Drafts" if you get an error
513that the "Folder doesn't exist".
514
c2163c6a
TPW
515Next, ensure that your Gmail settings are correct. In "Settings" the
516"Use Unicode (UTF-8) encoding for outgoing messages" should be checked.
517
518Once your commits are ready to send to the mailing list, run the following
519command to send the patch emails to your Gmail Drafts folder.
520
521 $ git format-patch -M --stdout origin/master | git imap-send
522
523Go to your Gmail account, open the Drafts folder, find the patch email, fill
524in the To: and CC: fields and send away!
50dffd4e 525