[MRG] DOC: Added rebasing and squashing section to contributing.rst #6176

devashishd12 · 2016-01-17T11:22:53Z

Added rebasing and squashing section to contributing.rst as agreed here. Used this comment to add to the section.

devashishd12 · 2016-01-19T19:07:16Z

@amueller @rvraghav93 could you please take a look at this? Thanks!

raghavrv · 2016-01-19T20:15:09Z

doc/developers/contributing.rst

@@ -207,6 +207,51 @@ and Cython optimizations.
   <http://astropy.readthedocs.org/en/latest/development/workflow/development_workflow.html>`_
   sections.

+Rebasing and squashing commits
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^


Is this the correct format for header?

yes it gives this kind of a header.

thanks for checking!

Thanks for checking. Maybe it would be better to have a heading like this?

raghavrv · 2016-01-19T20:21:59Z

I think we can structure it something like this -

Resolving merge conflicts...
------------------

Why they happen - 1 or 2 lines

What to do when that happens - two options

Why we choose rebase over merge

How to rebase properly

External link/man page to rebase

Squashing commits
-------------------
Why we prefer less, but meaningful (number of) commits - (To keep history clean and to help backtrack regressions to a single commit)

How to squash

How to force push and update the branch

Some external links on interactive rebase/squashing

devashishd12 · 2016-01-20T20:03:11Z

@rvraghav93 is this a bit better?

raghavrv · 2016-01-20T20:11:45Z

This looks much better. I will take a good look at it in some time

raghavrv · 2016-01-20T22:23:02Z

doc/developers/contributing.rst

+
+	2. Make a backup just in case::
+
+		$ git branch my-feature-branch-bak my-feature-branch


Do we need this? There is reflog in extreme cases no? It would be better if we have it simple I feel

Yes true....

devashishd12 · 2016-01-22T08:22:42Z

@rvraghav93 I was thinking of putting the interactive rebasing method before the cherry-pick one..... Also I just copied the gist of squashing the commits that's why I kept the link to the scipy tut intact. What do you think?

devashishd12 · 2016-01-30T19:16:02Z

ping @rvraghav93. Sorry for disturbing but should I proceed with the above changes? Could you please have a look at this if you have the time?

raghavrv · 2016-01-30T19:22:39Z

Apologies for the late response! I'll take a look at it now and let you know

raghavrv · 2016-01-30T19:58:56Z

doc/developers/contributing.rst

+
+		$ git checkout my-feature-branch
+
+	2. Rebase on master::


You should maybe add a line on updating master before?

raghavrv · 2016-01-30T20:10:03Z

This looks very good. Would help a lot of people new to git :)

Ping @amueller and @jnothman for reviews

devashishd12 · 2016-01-30T20:12:59Z

Thanks a lot for reviewing! I'll address the comments and make the changes asap 👍

devashishd12 · 2016-01-30T21:21:39Z

I've addressed the comments. Hope this is better now :)

raghavrv · 2016-01-30T21:23:28Z

See if you can use your documentation to squash the commits of this PR ;)

devashishd12 · 2016-01-30T21:28:51Z

Hahahaha! I'll do that now!

amueller · 2016-02-09T20:37:55Z

is this ready for review? If so, please rename.

amueller · 2016-02-09T20:39:15Z

I think we should talk first about squashing, then rebasing. That really lowers the chance of having rebase conflicts.

devashishd12 · 2016-02-10T15:15:51Z

@amueller oh sorry I'll rename it now. Yes this is ready for review.

devashishd12 · 2016-02-14T21:32:38Z

@amueller I have addressed the comments and changed the order of squash and rebase. Could you please check? Thanks!

nelson-liu · 2016-02-24T05:01:13Z

doc/developers/contributing.rst

+^^^^^^^^^^^^^^^^^
+
+To keep the history clean and to help backtrack regressions to a single commit we
+prefer less (but meaningful) number of commits.


I'd reword this to we prefer to have less commits that are more meaningful.

umm I'm not quite sure about this.... It still retains the same meaning but makes it unnecessarily verbose. What do you think?

I actually think it should be we prefer less (but more meaningful) commits.

nelson-liu · 2016-02-24T05:03:45Z

doc/developers/contributing.rst

+
+.. note::
+
+  Detailed guide on solving merge conflicts can be found on `git-scm.com


Detailed guide... -> A detailed guide...

devashishd12 · 2016-02-24T10:38:40Z

@nelson-liu I've addressed the comments. Good catches!

nelson-liu · 2016-04-23T05:42:41Z

@dsquareindia why did you close this? I think it's a great improvement; the part about rebasing is a common question e.g. #6446 (comment)

nelson-liu · 2016-04-25T18:05:56Z

@dsquareindia would you mind if i cherry-picked your commit and opened a new pr? I think this addition is quite necessary.

devashishd12 · 2016-04-25T19:45:33Z

Sorry for that; I'll open it again. @nelson-liu I'll be happy to continue work on this if needed :)

nelson-liu · 2016-04-25T20:59:16Z

doc/developers/contributing.rst

+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Another way to squash commits is through interactive rebasing which is one the most powerful
+features of ``rebase``. If you want to squash the latest 3 commits together (this number can


use $ git log --oneline to see past commits?

nelson-liu · 2016-04-25T21:02:10Z

could you also rebase on master? maybe use your own instructions ;)

devashishd12 · 2016-06-28T15:07:30Z

@nelson-liu sorry for keeping this PR idle. You can cherry pick this if you want and open a new pr 👍 . Won't be getting a chance to work on this anytime soon...

jnothman · 2016-06-28T23:41:48Z

I think this needs to be cut back to discuss rebasing etc while working on a PR. We no longer generally want contributors to do squashing, in preference for the "squash and merge" facility provided by github.

nelson-liu · 2016-07-15T19:34:57Z

@jnothman I agree and I've seen it happen a few times recently. maybe having this in the docs + perhaps adding it to the PR template would be good tasks for the scipy sprint?

amueller · 2016-10-13T18:26:11Z

solved by #7541 IMHO. Feel free to reopen / update. I think rewriting the same docs over and over again is probably not the point.

devashishd12 changed the title ~~DOC: Added rebasing and squashing section to contributing.rst~~ [MRG] DOC: Added rebasing and squashing section to contributing.rst Jan 17, 2016

raghavrv reviewed Jan 19, 2016
View reviewed changes

devashishd12 changed the title ~~[MRG] DOC: Added rebasing and squashing section to contributing.rst~~ [WIP] DOC: Added rebasing and squashing section to contributing.rst Jan 20, 2016

raghavrv reviewed Jan 20, 2016
View reviewed changes

raghavrv reviewed Jan 30, 2016
View reviewed changes

DOC: Added rebasing and squashing section to contributing.rst

5fe97f5

devashishd12 changed the title ~~[WIP] DOC: Added rebasing and squashing section to contributing.rst~~ [MRG] DOC: Added rebasing and squashing section to contributing.rst Feb 10, 2016

changed rebase and squash order

2bb47e9

nelson-liu reviewed Feb 24, 2016
View reviewed changes

Reworded sentence, fixed grammatical error

e4c7839

devashishd12 closed this Apr 23, 2016

devashishd12 reopened this Apr 25, 2016

nelson-liu reviewed Apr 25, 2016
View reviewed changes

nelson-liu mentioned this pull request Sep 14, 2016

PR tags used in sklearn #6211

Closed

amueller closed this Oct 13, 2016


		2. Make a backup just in case::

		$ git branch my-feature-branch-bak my-feature-branch


		.. note::

		Detailed guide on solving merge conflicts can be found on `git-scm.com

[MRG] DOC: Added rebasing and squashing section to contributing.rst #6176

[MRG] DOC: Added rebasing and squashing section to contributing.rst #6176

Conversation

devashishd12 commented Jan 17, 2016

devashishd12 commented Jan 19, 2016

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

raghavrv commented Jan 19, 2016

devashishd12 commented Jan 20, 2016

raghavrv commented Jan 20, 2016

Choose a reason for hiding this comment

Choose a reason for hiding this comment

devashishd12 commented Jan 22, 2016

devashishd12 commented Jan 30, 2016

raghavrv commented Jan 30, 2016

Choose a reason for hiding this comment

raghavrv commented Jan 30, 2016

devashishd12 commented Jan 30, 2016

devashishd12 commented Jan 30, 2016

raghavrv commented Jan 30, 2016

devashishd12 commented Jan 30, 2016

amueller commented Feb 9, 2016

amueller commented Feb 9, 2016

devashishd12 commented Feb 10, 2016

devashishd12 commented Feb 14, 2016

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

devashishd12 commented Feb 24, 2016

nelson-liu commented Apr 23, 2016

nelson-liu commented Apr 25, 2016

devashishd12 commented Apr 25, 2016

Choose a reason for hiding this comment

nelson-liu commented Apr 25, 2016

devashishd12 commented Jun 28, 2016

jnothman commented Jun 28, 2016

nelson-liu commented Jul 15, 2016

amueller commented Oct 13, 2016