| blob | 86ca7f6a78a9b69294eb97bbbc1138376f40198b |
1 Submitting Patches
2 ==================
4 == Guidelines
6 Here are some guidelines for contributing back to this
7 project. There is also a link:MyFirstContribution.html[step-by-step tutorial]
8 available which covers many of these same guidelines.
10 [[patch-flow]]
11 === A typical life cycle of a patch series
13 To help us understand the reason behind various guidelines given later
14 in the document, first let's understand how the life cycle of a
15 typical patch series for this project goes.
17 . You come up with an itch. You code it up. You do not need any
18 pre-authorization from the project to do so.
19 +
20 Your patches will be reviewed by other contributors on the mailing
21 list, and the reviews will be done to assess the merit of various
22 things, like the general idea behind your patch (including "is it
23 solving a problem worth solving in the first place?"), the reason
24 behind the design of the solution, and the actual implementation.
25 The guidelines given here are there to help your patches by making
26 them easier to understand by the reviewers.
28 . You send the patches to the list and cc people who may need to know
29 about the change. Your goal is *not* necessarily to convince others
30 that what you are building is good. Your goal is to get help in
31 coming up with a solution for the "itch" that is better than what
32 you can build alone.
33 +
34 The people who may need to know are the ones who worked on the code
35 you are touching. These people happen to be the ones who are
36 most likely to be knowledgeable enough to help you, but
37 they have no obligation to help you (i.e. you ask them for help,
38 you don't demand). +git log -p {litdd} _$area_you_are_modifying_+ would
39 help you find out who they are.
41 . You get comments and suggestions for improvements. You may even get
42 them in an "on top of your change" patch form. You are expected to
43 respond to them with "Reply-All" on the mailing list, while taking
44 them into account while preparing an updated set of patches.
46 . Polish, refine, and re-send your patches to the list and to the people
47 who spent their time to improve your patch. Go back to step (2).
49 . While the above iterations improve your patches, the maintainer may
50 pick the patches up from the list and queue them to the `seen`
51 branch, in order to make it easier for people to play with it
52 without having to pick up and apply the patches to their trees
53 themselves. Being in `seen` has no other meaning. Specifically, it
54 does not mean the patch was "accepted" in any way.
56 . When the discussion reaches a consensus that the latest iteration of
57 the patches are in good enough shape, the maintainer includes the
58 topic in the "What's cooking" report that are sent out a few times a
59 week to the mailing list, marked as "Will merge to 'next'." This
60 decision is primarily made by the maintainer with help from those
61 who participated in the review discussion.
63 . After the patches are merged to the 'next' branch, the discussion
64 can still continue to further improve them by adding more patches on
65 top, but by the time a topic gets merged to 'next', it is expected
66 that everybody agrees that the scope and the basic direction of the
67 topic are appropriate, so such an incremental updates are limited to
68 small corrections and polishing. After a topic cooks for some time
69 (like 7 calendar days) in 'next' without needing further tweaks on
70 top, it gets merged to the 'master' branch and wait to become part
71 of the next major release.
73 In the following sections, many techniques and conventions are listed
74 to help your patches get reviewed effectively in such a life cycle.
77 [[choose-starting-point]]
78 === Choose a starting point.
80 As a preliminary step, you must first choose a starting point for your
81 work. Typically this means choosing a branch, although technically
82 speaking it is actually a particular commit (typically the HEAD, or tip,
83 of the branch).
85 There are several important branches to be aware of. Namely, there are
86 four integration branches as discussed in linkgit:gitworkflows[7]:
88 * maint
89 * master
90 * next
91 * seen
93 The branches lower on the list are typically descendants of the ones
94 that come before it. For example, `maint` is an "older" branch than
95 `master` because `master` usually has patches (commits) on top of
96 `maint`.
98 There are also "topic" branches, which contain work from other
99 contributors. Topic branches are created by the Git maintainer (in
100 their fork) to organize the current set of incoming contributions on
101 the mailing list, and are itemized in the regular "What's cooking in
102 git.git" announcements. To find the tip of a topic branch, run `git log
103 --first-parent master..seen` and look for the merge commit. The second
104 parent of this commit is the tip of the topic branch.
106 There is one guiding principle for choosing the right starting point: in
107 general, always base your work on the oldest integration branch that
108 your change is relevant to (see "Merge upwards" in
109 linkgit:gitworkflows[7]). What this principle means is that for the
110 vast majority of cases, the starting point for new work should be the
111 latest HEAD commit of `maint` or `master` based on the following cases:
113 * If you are fixing bugs in the released version, use `maint` as the
114 starting point (which may mean you have to fix things without using
115 new API features on the cutting edge that recently appeared in
116 `master` but were not available in the released version).
118 * Otherwise (such as if you are adding new features) use `master`.
121 NOTE: In exceptional cases, a bug that was introduced in an old
122 version may have to be fixed for users of releases that are much older
123 than the recent releases. `git describe --contains X` may describe
124 `X` as `v2.30.0-rc2-gXXXXXX` for the commit `X` that introduced the
125 bug, and the bug may be so high-impact that we may need to issue a new
126 maintenance release for Git 2.30.x series, when "Git 2.41.0" is the
127 current release. In such a case, you may want to use the tip of the
128 maintenance branch for the 2.30.x series, which may be available in the
129 `maint-2.30` branch in https://github.com/gitster/git[the maintainer's
130 "broken out" repo].
132 This also means that `next` or `seen` are inappropriate starting points
133 for your work, if you want your work to have a realistic chance of
134 graduating to `master`. They are simply not designed to be used as a
135 base for new work; they are only there to make sure that topics in
136 flight work well together. This is why both `next` and `seen` are
137 frequently re-integrated with incoming patches on the mailing list and
138 force-pushed to replace previous versions of themselves. A topic that is
139 literally built on top of `next` cannot be merged to `master` without
140 dragging in all the other topics in `next`, some of which may not be
141 ready.
143 For example, if you are making tree-wide changes, while somebody else is
144 also making their own tree-wide changes, your work may have severe
145 overlap with the other person's work. This situation may tempt you to
146 use `next` as your starting point (because it would have the other
147 person's work included in it), but doing so would mean you'll not only
148 depend on the other person's work, but all the other random things from
149 other contributors that are already integrated into `next`. And as soon
150 as `next` is updated with a new version, all of your work will need to
151 be rebased anyway in order for them to be cleanly applied by the
152 maintainer.
154 Under truly exceptional circumstances where you absolutely must depend
155 on a select few topic branches that are already in `next` but not in
156 `master`, you may want to create your own custom base-branch by forking
157 `master` and merging the required topic branches into it. You could then
158 work on top of this base-branch. But keep in mind that this base-branch
159 would only be known privately to you. So when you are ready to send
160 your patches to the list, be sure to communicate how you created it in
161 your cover letter. This critical piece of information would allow
162 others to recreate your base-branch on their end in order for them to
163 try out your work.
165 Finally, note that some parts of the system have dedicated maintainers
166 with their own separate source code repositories (see the section
167 "Subsystems" below).
169 [[separate-commits]]
170 === Make separate commits for logically separate changes.
172 Unless your patch is really trivial, you should not be sending
173 out a patch that was generated between your working tree and
174 your commit head. Instead, always make a commit with complete
175 commit message and generate a series of patches from your
176 repository. It is a good discipline.
178 Give an explanation for the change(s) that is detailed enough so
179 that people can judge if it is good thing to do, without reading
180 the actual patch text to determine how well the code does what
181 the explanation promises to do.
183 If your description starts to get too long, that's a sign that you
184 probably need to split up your commit to finer grained pieces.
185 That being said, patches which plainly describe the things that
186 help reviewers check the patch, and future maintainers understand
187 the code, are the most beautiful patches. Descriptions that summarize
188 the point in the subject well, and describe the motivation for the
189 change, the approach taken by the change, and if relevant how this
190 differs substantially from the prior version, are all good things
191 to have.
193 Make sure that you have tests for the bug you are fixing. See
194 `t/README` for guidance.
196 [[tests]]
197 When adding a new feature, make sure that you have new tests to show
198 the feature triggers the new behavior when it should, and to show the
199 feature does not trigger when it shouldn't. After any code change,
200 make sure that the entire test suite passes. When fixing a bug, make
201 sure you have new tests that break if somebody else breaks what you
202 fixed by accident to avoid regression. Also, try merging your work to
203 'next' and 'seen' and make sure the tests still pass; topics by others
204 that are still in flight may have unexpected interactions with what
205 you are trying to do in your topic.
207 Pushing to a fork of https://github.com/git/git will use their CI
208 integration to test your changes on Linux, Mac and Windows. See the
209 <<GHCI,GitHub CI>> section for details.
211 Do not forget to update the documentation to describe the updated
212 behavior and make sure that the resulting documentation set formats
213 well (try the Documentation/doc-diff script).
215 We currently have a liberal mixture of US and UK English norms for
216 spelling and grammar, which is somewhat unfortunate. A huge patch that
217 touches the files all over the place only to correct the inconsistency
218 is not welcome, though. Potential clashes with other changes that can
219 result from such a patch are not worth it. We prefer to gradually
220 reconcile the inconsistencies in favor of US English, with small and
221 easily digestible patches, as a side effect of doing some other real
222 work in the vicinity (e.g. rewriting a paragraph for clarity, while
223 turning en_UK spelling to en_US). Obvious typographical fixes are much
224 more welcomed ("teh -> "the"), preferably submitted as independent
225 patches separate from other documentation changes.
227 [[whitespace-check]]
228 Oh, another thing. We are picky about whitespaces. Make sure your
229 changes do not trigger errors with the sample pre-commit hook shipped
230 in `templates/hooks--pre-commit`. To help ensure this does not happen,
231 run `git diff --check` on your changes before you commit.
233 [[describe-changes]]
234 === Describe your changes well.
236 The log message that explains your changes is just as important as the
237 changes themselves. Your code may be clearly written with in-code
238 comment to sufficiently explain how it works with the surrounding
239 code, but those who need to fix or enhance your code in the future
240 will need to know _why_ your code does what it does, for a few
241 reasons:
243 . Your code may be doing something differently from what you wanted it
244 to do. Writing down what you actually wanted to achieve will help
245 them fix your code and make it do what it should have been doing
246 (also, you often discover your own bugs yourself, while writing the
247 log message to summarize the thought behind it).
249 . Your code may be doing things that were only necessary for your
250 immediate needs (e.g. "do X to directories" without implementing or
251 even designing what is to be done on files). Writing down why you
252 excluded what the code does not do will help guide future developers.
253 Writing down "we do X to directories, because directories have
254 characteristic Y" would help them infer "oh, files also have the same
255 characteristic Y, so perhaps doing X to them would also make sense?".
256 Saying "we don't do the same X to files, because ..." will help them
257 decide if the reasoning is sound (in which case they do not waste
258 time extending your code to cover files), or reason differently (in
259 which case, they can explain why they extend your code to cover
260 files, too).
262 The goal of your log message is to convey the _why_ behind your change
263 to help future developers. The reviewers will also make sure that
264 your proposed log message will serve this purpose well.
266 The first line of the commit message should be a short description (50
267 characters is the soft limit, see DISCUSSION in linkgit:git-commit[1]),
268 and should skip the full stop. It is also conventional in most cases to
269 prefix the first line with "area: " where the area is a filename or
270 identifier for the general area of the code being modified, e.g.
272 * doc: clarify distinction between sign-off and pgp-signing
273 * githooks.txt: improve the intro section
275 If in doubt which identifier to use, run `git log --no-merges` on the
276 files you are modifying to see the current conventions.
278 [[summary-section]]
279 The title sentence after the "area:" prefix omits the full stop at the
280 end, and its first word is not capitalized (the omission
281 of capitalization applies only to the word after the "area:"
282 prefix of the title) unless there is a reason to
283 capitalize it other than because it is the first word in the sentence.
284 E.g. "doc: clarify...", not "doc: Clarify...", or "githooks.txt:
285 improve...", not "githooks.txt: Improve...". But "refs: HEAD is also
286 treated as a ref" is correct, as we spell `HEAD` in all caps even when
287 it appears in the middle of a sentence.
289 [[meaningful-message]]
290 The body should provide a meaningful commit message, which:
292 . explains the problem the change tries to solve, i.e. what is wrong
293 with the current code without the change.
295 . justifies the way the change solves the problem, i.e. why the
296 result with the change is better.
298 . alternate solutions considered but discarded, if any.
300 [[present-tense]]
301 The problem statement that describes the status quo is written in the
302 present tense. Write "The code does X when it is given input Y",
303 instead of "The code used to do Y when given input X". You do not
304 have to say "Currently"---the status quo in the problem statement is
305 about the code _without_ your change, by project convention.
307 [[imperative-mood]]
308 Describe your changes in imperative mood, e.g. "make xyzzy do frotz"
309 instead of "[This patch] makes xyzzy do frotz" or "[I] changed xyzzy
310 to do frotz", as if you are giving orders to the codebase to change
311 its behavior. Try to make sure your explanation can be understood
312 without external resources. Instead of giving a URL to a mailing list
313 archive, summarize the relevant points of the discussion.
315 [[commit-reference]]
317 There are a few reasons why you may want to refer to another commit in
318 the "more stable" part of the history (i.e. on branches like `maint`,
319 `master`, and `next`):
321 . A commit that introduced the root cause of a bug you are fixing.
323 . A commit that introduced a feature that you are enhancing.
325 . A commit that conflicts with your work when you made a trial merge
326 of your work into `next` and `seen` for testing.
328 When you reference a commit on a more stable branch (like `master`,
329 `maint` and `next`), use the format "abbreviated hash (subject,
330 date)", like this:
332 ....
333 Commit f86a374 (pack-bitmap.c: fix a memleak, 2015-03-30)
334 noticed that ...
335 ....
337 The "Copy commit reference" command of gitk can be used to obtain this
338 format (with the subject enclosed in a pair of double-quotes), or this
339 invocation of `git show`:
341 ....
342 git show -s --pretty=reference <commit>
343 ....
345 or, on an older version of Git without support for --pretty=reference:
347 ....
348 git show -s --date=short --pretty='format:%h (%s, %ad)' <commit>
349 ....
351 [[sign-off]]
352 === Certify your work by adding your `Signed-off-by` trailer
354 To improve tracking of who did what, we ask you to certify that you
355 wrote the patch or have the right to pass it on under the same license
356 as ours, by "signing off" your patch. Without sign-off, we cannot
357 accept your patches.
359 If (and only if) you certify the below D-C-O:
361 [[dco]]
362 .Developer's Certificate of Origin 1.1
363 ____
364 By making a contribution to this project, I certify that:
366 a. The contribution was created in whole or in part by me and I
367 have the right to submit it under the open source license
368 indicated in the file; or
370 b. The contribution is based upon previous work that, to the best
371 of my knowledge, is covered under an appropriate open source
372 license and I have the right under that license to submit that
373 work with modifications, whether created in whole or in part
374 by me, under the same open source license (unless I am
375 permitted to submit under a different license), as indicated
376 in the file; or
378 c. The contribution was provided directly to me by some other
379 person who certified (a), (b) or (c) and I have not modified
380 it.
382 d. I understand and agree that this project and the contribution
383 are public and that a record of the contribution (including all
384 personal information I submit with it, including my sign-off) is
385 maintained indefinitely and may be redistributed consistent with
386 this project or the open source license(s) involved.
387 ____
389 you add a "Signed-off-by" trailer to your commit, that looks like
390 this:
392 ....
393 Signed-off-by: Random J Developer <random@developer.example.org>
394 ....
396 This line can be added by Git if you run the git-commit command with
397 the -s option.
399 Notice that you can place your own `Signed-off-by` trailer when
400 forwarding somebody else's patch with the above rules for
401 D-C-O. Indeed you are encouraged to do so. Do not forget to
402 place an in-body "From: " line at the beginning to properly attribute
403 the change to its true author (see (2) above).
405 This procedure originally came from the Linux kernel project, so our
406 rule is quite similar to theirs, but what exactly it means to sign-off
407 your patch differs from project to project, so it may be different
408 from that of the project you are accustomed to.
410 [[real-name]]
411 Please use a known identity in the `Signed-off-by` trailer, since we cannot
412 accept anonymous contributions. It is common, but not required, to use some form
413 of your real name. We realize that some contributors are not comfortable doing
414 so or prefer to contribute under a pseudonym or preferred name and we can accept
415 your patch either way, as long as the name and email you use are distinctive,
416 identifying, and not misleading.
418 The goal of this policy is to allow us to have sufficient information to contact
419 you if questions arise about your contribution.
421 [[commit-trailers]]
422 If you like, you can put extra trailers at the end:
424 . `Reported-by:` is used to credit someone who found the bug that
425 the patch attempts to fix.
426 . `Acked-by:` says that the person who is more familiar with the area
427 the patch attempts to modify liked the patch.
428 . `Reviewed-by:`, unlike the other trailers, can only be offered by the
429 reviewers themselves when they are completely satisfied with the
430 patch after a detailed analysis.
431 . `Tested-by:` is used to indicate that the person applied the patch
432 and found it to have the desired effect.
433 . `Co-authored-by:` is used to indicate that people exchanged drafts
434 of a patch before submitting it.
435 . `Helped-by:` is used to credit someone who suggested ideas for
436 changes without providing the precise changes in patch form.
437 . `Mentored-by:` is used to credit someone with helping develop a
438 patch as part of a mentorship program (e.g., GSoC or Outreachy).
439 . `Suggested-by:` is used to credit someone with suggesting the idea
440 for a patch.
442 While you can also create your own trailer if the situation warrants it, we
443 encourage you to instead use one of the common trailers in this project
444 highlighted above.
446 Only capitalize the very first letter of the trailer, i.e. favor
447 "Signed-off-by" over "Signed-Off-By" and "Acked-by:" over "Acked-By".
449 [[git-tools]]
450 === Generate your patch using Git tools out of your commits.
452 Git based diff tools generate unidiff which is the preferred format.
454 You do not have to be afraid to use `-M` option to `git diff` or
455 `git format-patch`, if your patch involves file renames. The
456 receiving end can handle them just fine.
458 [[review-patch]]
459 Please make sure your patch does not add commented out debugging code,
460 or include any extra files which do not relate to what your patch
461 is trying to achieve. Make sure to review
462 your patch after generating it, to ensure accuracy. Before
463 sending out, please make sure it cleanly applies to the starting point you
464 have chosen in the "Choose a starting point" section.
466 NOTE: From the perspective of those reviewing your patch, the `master`
467 branch is the default expected starting point. So if you have chosen a
468 different starting point, please communicate this choice in your cover
469 letter.
472 [[send-patches]]
473 === Sending your patches.
475 ==== Choosing your reviewers
477 :security-ml: footnoteref:[security-ml,The Git Security mailing list: git-security@googlegroups.com]
479 NOTE: Patches that may be
480 security relevant should be submitted privately to the Git Security
481 mailing list{security-ml}, instead of the public mailing list.
483 :contrib-scripts: footnoteref:[contrib-scripts,Scripts under `contrib/` are +
484 not part of the core `git` binary and must be called directly. Clone the Git +
485 codebase and run `perl contrib/contacts/git-contacts`.]
487 Send your patch with "To:" set to the mailing list, with "cc:" listing
488 people who are involved in the area you are touching (the `git-contacts`
489 script in `contrib/contacts/`{contrib-scripts} can help to
490 identify them), to solicit comments and reviews. Also, when you made
491 trial merges of your topic to `next` and `seen`, you may have noticed
492 work by others conflicting with your changes. There is a good possibility
493 that these people may know the area you are touching well.
495 If you are using `send-email`, you can feed it the output of `git-contacts` like
496 this:
498 ....
499 git send-email --cc-cmd='perl contrib/contacts/git-contacts' feature/*.patch
500 ....
502 :current-maintainer: footnote:[The current maintainer: gitster@pobox.com]
503 :git-ml: footnote:[The mailing list: git@vger.kernel.org]
505 After the list reached a consensus that it is a good idea to apply the
506 patch, re-send it with "To:" set to the maintainer{current-maintainer}
507 and "cc:" the list{git-ml} for inclusion. This is especially relevant
508 when the maintainer did not heavily participate in the discussion and
509 instead left the review to trusted others.
511 Do not forget to add trailers such as `Acked-by:`, `Reviewed-by:` and
512 `Tested-by:` lines as necessary to credit people who helped your
513 patch, and "cc:" them when sending such a final version for inclusion.
515 ==== `format-patch` and `send-email`
517 Learn to use `format-patch` and `send-email` if possible. These commands
518 are optimized for the workflow of sending patches, avoiding many ways
519 your existing e-mail client (often optimized for "multipart/*" MIME
520 type e-mails) might render your patches unusable.
522 NOTE: Here we outline the procedure using `format-patch` and
523 `send-email`, but you can instead use GitGitGadget to send in your
524 patches (see link:MyFirstContribution.html[MyFirstContribution]).
526 People on the Git mailing list need to be able to read and
527 comment on the changes you are submitting. It is important for
528 a developer to be able to "quote" your changes, using standard
529 e-mail tools, so that they may comment on specific portions of
530 your code. For this reason, each patch should be submitted
531 "inline" in a separate message.
533 All subsequent versions of a patch series and other related patches should be
534 grouped into their own e-mail thread to help readers find all parts of the
535 series. To that end, send them as replies to either an additional "cover
536 letter" message (see below), the first patch, or the respective preceding patch.
537 Here is a link:MyFirstContribution.html#v2-git-send-email[step-by-step guide] on
538 how to submit updated versions of a patch series.
540 If your log message (including your name on the
541 `Signed-off-by` trailer) is not writable in ASCII, make sure that
542 you send off a message in the correct encoding.
544 WARNING: Be wary of your MUAs word-wrap
545 corrupting your patch. Do not cut-n-paste your patch; you can
546 lose tabs that way if you are not careful.
548 It is a common convention to prefix your subject line with
549 [PATCH]. This lets people easily distinguish patches from other
550 e-mail discussions. Use of markers in addition to PATCH within
551 the brackets to describe the nature of the patch is also
552 encouraged. E.g. [RFC PATCH] (where RFC stands for "request for
553 comments") is often used to indicate a patch needs further
554 discussion before being accepted, [PATCH v2], [PATCH v3] etc.
555 are often seen when you are sending an update to what you have
556 previously sent.
558 The `git format-patch` command follows the best current practice to
559 format the body of an e-mail message. At the beginning of the
560 patch should come your commit message, ending with the
561 `Signed-off-by` trailers, and a line that consists of three dashes,
562 followed by the diffstat information and the patch itself. If
563 you are forwarding a patch from somebody else, optionally, at
564 the beginning of the e-mail message just before the commit
565 message starts, you can put a "From: " line to name that person.
566 To change the default "[PATCH]" in the subject to "[<text>]", use
567 `git format-patch --subject-prefix=<text>`. As a shortcut, you
568 can use `--rfc` instead of `--subject-prefix="RFC PATCH"`, or
569 `-v <n>` instead of `--subject-prefix="PATCH v<n>"`.
571 You often want to add additional explanation about the patch,
572 other than the commit message itself. Place such "cover letter"
573 material between the three-dash line and the diffstat. For
574 patches requiring multiple iterations of review and discussion,
575 an explanation of changes between each iteration can be kept in
576 Git-notes and inserted automatically following the three-dash
577 line via `git format-patch --notes`.
579 [[the-topic-summary]]
580 *This is EXPERIMENTAL*.
582 When sending a topic, you can propose a one-paragraph summary that
583 should appear in the "What's cooking" report when it is picked up to
584 explain the topic. If you choose to do so, please write a 2-5 line
585 paragraph that will fit well in our release notes (see many bulleted
586 entries in the Documentation/RelNotes/* files for examples), and make
587 it the first paragraph of the cover letter. For a single-patch
588 series, use the space between the three-dash line and the diffstat, as
589 described earlier.
591 [[attachment]]
592 Do not attach the patch as a MIME attachment, compressed or not.
593 Do not let your e-mail client send quoted-printable. Do not let
594 your e-mail client send format=flowed which would destroy
595 whitespaces in your patches. Many
596 popular e-mail applications will not always transmit a MIME
597 attachment as plain text, making it impossible to comment on
598 your code. A MIME attachment also takes a bit more time to
599 process. This does not decrease the likelihood of your
600 MIME-attached change being accepted, but it makes it more likely
601 that it will be postponed.
603 Exception: If your mailer is mangling patches then someone may ask
604 you to re-send them using MIME, that is OK.
606 [[pgp-signature]]
607 Do not PGP sign your patch. Most likely, your maintainer or other people on the
608 list would not have your PGP key and would not bother obtaining it anyway.
609 Your patch is not judged by who you are; a good patch from an unknown origin
610 has a far better chance of being accepted than a patch from a known, respected
611 origin that is done poorly or does incorrect things.
613 If you really really really really want to do a PGP signed
614 patch, format it as "multipart/signed", not a text/plain message
615 that starts with `-----BEGIN PGP SIGNED MESSAGE-----`. That is
616 not a text/plain, it's something else.
618 === Handling Conflicts and Iterating Patches
620 When revising changes made to your patches, it's important to
621 acknowledge the possibility of conflicts with other ongoing topics. To
622 navigate these potential conflicts effectively, follow the recommended
623 steps outlined below:
625 . Build on a suitable base branch, see the <<choose-starting-point, section above>>,
626 and format-patch the series. If you are doing "rebase -i" in-place to
627 update from the previous round, this will reuse the previous base so
628 (2) and (3) may become trivial.
630 . Find the base of where the last round was queued
631 +
632 $ mine='kn/ref-transaction-symref'
633 $ git checkout "origin/seen^{/^Merge branch '$mine'}...master"
635 . Apply your format-patch result. There are two cases
636 .. Things apply cleanly and tests fine. Go to (4).
637 .. Things apply cleanly but does not build or test fails, or things do
638 not apply cleanly.
639 +
640 In the latter case, you have textual or semantic conflicts coming from
641 the difference between the old base and the base you used to build in
642 (1). Identify what caused the breakages (e.g., a topic or two may have
643 merged since the base used by (2) until the base used by (1)).
644 +
645 Check out the latest 'origin/master' (which may be newer than the base
646 used by (2)), "merge --no-ff" the topics you newly depend on in there,
647 and use the result of the merge(s) as the base, rebuild the series and
648 test again. Run format-patch from the last such merges to the tip of
649 your topic. If you did
650 +
651 $ git checkout origin/master
652 $ git merge --no-ff --into-name kn/ref-transaction-symref fo/obar
653 $ git merge --no-ff --into-name kn/ref-transaction-symref ba/zqux
654 ... rebuild the topic ...
655 +
656 Then you'd just format your topic above these "preparing the ground"
657 merges, e.g.
658 +
659 $ git format-patch "HEAD^{/^Merge branch 'ba/zqux'}"..HEAD
660 +
661 Do not forget to write in the cover letter you did this, including the
662 topics you have in your base on top of 'master'. Then go to (4).
664 . Make a trial merge of your topic into 'next' and 'seen', e.g.
665 +
666 $ git checkout --detach 'origin/seen'
667 $ git revert -m 1 <the merge of the previous iteration into seen>
668 $ git merge kn/ref-transaction-symref
669 +
670 The "revert" is needed if the previous iteration of your topic is
671 already in 'seen' (like in this case). You could choose to rebuild
672 master..origin/seen from scratch while excluding your previous
673 iteration, which may emulate what happens on the maintainers end more
674 closely.
675 +
676 This trial merge may conflict. It is primarily to see what conflicts
677 _other_ topics may have with your topic. In other words, you do not
678 have to depend on it to make your topic work on 'master'. It may
679 become the job of the other topic owners to resolve conflicts if your
680 topic goes to 'next' before theirs.
681 +
682 Make a note on what conflict you saw in the cover letter. You do not
683 necessarily have to resolve them, but it would be a good opportunity to
684 learn what others are doing in related areas.
685 +
686 $ git checkout --detach 'origin/next'
687 $ git merge kn/ref-transaction-symref
688 +
689 This is to see what conflicts your topic has with other topics that are
690 already cooking. This should not conflict if (3)-2 prepared a base on
691 top of updated master plus dependent topics taken from 'next'. Unless
692 the context is severe (one way to tell is try the same trial merge with
693 your old iteration, which may conflict in a similar way), expect that it
694 will be handled on maintainers end (if it gets unmanageable, I'll ask to
695 rebase when I receive your patches).
697 == Subsystems with dedicated maintainers
699 Some parts of the system have dedicated maintainers with their own
700 repositories.
702 - `git-gui/` comes from the git-gui project, maintained by Johannes Sixt:
704 https://github.com/j6t/git-gui
706 Contibutions should go via the git mailing list.
708 - `gitk-git/` comes from the gitk project, maintained by Johannes Sixt:
710 https://github.com/j6t/gitk
712 Contibutions should go via the git mailing list.
714 - `po/` comes from the localization coordinator, Jiang Xin:
716 https://github.com/git-l10n/git-po/
718 Patches to these parts should be based on their trees.
720 - The "Git documentation translations" project, led by Jean-Noël
721 Avila, translates our documentation pages. Their work products are
722 maintained separately from this project, not as part of our tree:
724 https://github.com/jnavila/git-manpages-l10n/
727 == GitHub CI[[GHCI]]
729 With an account at GitHub, you can use GitHub CI to test your changes
730 on Linux, Mac and Windows. See
731 https://github.com/git/git/actions/workflows/main.yml for examples of
732 recent CI runs.
734 Follow these steps for the initial setup:
736 . Fork https://github.com/git/git to your GitHub account.
737 You can find detailed instructions how to fork here:
738 https://help.github.com/articles/fork-a-repo/
740 After the initial setup, CI will run whenever you push new changes
741 to your fork of Git on GitHub. You can monitor the test state of all your
742 branches here: `https://github.com/<Your GitHub handle>/git/actions/workflows/main.yml`
744 If a branch does not pass all test cases then it will be marked with a
745 red +x+, instead of a green check. In that case, you can click on the
746 failing job and navigate to "ci/run-build-and-tests.sh" and/or
747 "ci/print-test-failures.sh". You can also download "Artifacts" which
748 are zip archives containing tarred (or zipped) archives with test data
749 relevant for debugging.
751 Then fix the problem and push your fix to your GitHub fork. This will
752 trigger a new CI build to ensure all tests pass.
754 [[mua]]
755 == MUA specific hints
757 Some of the patches I receive or pick up from the list share common
758 patterns of breakage. Please make sure your MUA is set up
759 properly not to corrupt whitespaces.
761 See the DISCUSSION section of linkgit:git-format-patch[1] for hints on
762 checking your patch by mailing it to yourself and applying with
763 linkgit:git-am[1].
765 While you are at it, check the resulting commit log message from
766 a trial run of applying the patch. If what is in the resulting
767 commit is not exactly what you would want to see, it is very
768 likely that your maintainer would end up hand editing the log
769 message when he applies your patch. Things like "Hi, this is my
770 first patch.\n", if you really want to put in the patch e-mail,
771 should come after the three-dash line that signals the end of the
772 commit message.
775 === Pine
777 (Johannes Schindelin)
779 ....
780 I don't know how many people still use pine, but for those poor
781 souls it may be good to mention that the quell-flowed-text is
782 needed for recent versions.
784 ... the "no-strip-whitespace-before-send" option, too. AFAIK it
785 was introduced in 4.60.
786 ....
788 (Linus Torvalds)
790 ....
791 And 4.58 needs at least this.
793 diff-tree 8326dd8350be64ac7fc805f6563a1d61ad10d32c (from e886a61f76edf5410573e92e38ce22974f9c40f1)
794 Author: Linus Torvalds <torvalds@g5.osdl.org>
795 Date: Mon Aug 15 17:23:51 2005 -0700
797 Fix pine whitespace-corruption bug
799 There's no excuse for unconditionally removing whitespace from
800 the pico buffers on close.
802 diff --git a/pico/pico.c b/pico/pico.c
803 --- a/pico/pico.c
804 +++ b/pico/pico.c
805 @@ -219,7 +219,9 @@ PICO *pm;
806 switch(pico_all_done){ /* prepare for/handle final events */
807 case COMP_EXIT : /* already confirmed */
808 packheader();
809 +#if 0
810 stripwhitespace();
811 +#endif
812 c |= COMP_EXIT;
813 break;
814 ....
816 (Daniel Barkalow)
818 ....
819 > A patch to SubmittingPatches, MUA specific help section for
820 > users of Pine 4.63 would be very much appreciated.
822 Ah, it looks like a recent version changed the default behavior to do the
823 right thing, and inverted the sense of the configuration option. (Either
824 that or Gentoo did it.) So you need to set the
825 "no-strip-whitespace-before-send" option, unless the option you have is
826 "strip-whitespace-before-send", in which case you should avoid checking
827 it.
828 ....
830 === Thunderbird, KMail, GMail
832 See the MUA-SPECIFIC HINTS section of linkgit:git-format-patch[1].
834 === Gnus
836 "|" in the `*Summary*` buffer can be used to pipe the current
837 message to an external program, and this is a handy way to drive
838 `git am`. However, if the message is MIME encoded, what is
839 piped into the program is the representation you see in your
840 `*Article*` buffer after unwrapping MIME. This is often not what
841 you would want for two reasons. It tends to screw up non-ASCII
842 characters (most notably in people's names), and also
843 whitespaces (fatal in patches). Running "C-u g" to display the
844 message in raw form before using "|" to run the pipe can work
845 this problem around.