[python-committers] Should I merge a PR that I approved if it was written by a different core developer?

[Guido van Rossum]

On Wed, Sep 20, 2017 at 2:27 PM, Victor Stinner <victor.stinner at gmail.com>
wrote:

> Before the GitHub era, in the old "Mercurial era", the unwritten rule
> was to not merge a patch written by a developer who has the commit
> bit, to not "steal" his/her work. The old workflow (patches attached
> to the bug tracker) didn't allow to easily keep the author. You had to
> find the author email and full name and specify it manually.
>
> Moreover, there was a written rule of using the name of the developer
> who actually pushed the commit, so the commiters took the
> responsability of any regression (reminder the old era with no
> pre-commit CI? ;-)).
>
> In the new Git era, the author and committer *can* be two different
> people. Examples with "git log --pretty=full":
>
> commit 9abee722d448c1c00c7d4e11ce242ec7b13e5c49
> Author: Victor Stinner <victor.stinner at gmail.com>
> Commit: GitHub <noreply at github.com>
>
> commit 8f51bb436f8adfd139cad046b91cd462c7f27f6c (tag: v3.7.0a1)
> Author: Ned Deily <nad at python.org>
> Commit: Ned Deily <nad at python.org>
>
> commit 9b47af65375fab9318e88ccb061394a36c8c6c33
> Author: svelankar <17737361+svelankar at users.noreply.github.com>
> Commit: Raymond Hettinger <rhettinger at users.noreply.github.com>
>
> My question is: is someone opposed that a core developer clicks on the
> [Merge] button for a PR proposed by a different core developer?
>
> IMHO having a committer different than the author is valuable since
> the responsability is now shared by two developers instead of single
> one. It's similar to the "Signed-Off" tags used by the Linux kernel,
> but the list is limited to a single Signed-Off :-) Well, the committer
> is usually seen as the most reponsible, but now we can complain to the
> author as well *if needed* :-D
>

I think it's a good idea in many cases, but not required. E.g. you may be
OK with the diff but still ask the author to clean up some small nits, and
then they can merge their own diff. Or you may be OK with the diff but want
to wait for some other reviewer's OK.

The good news is that it's no longer wrong, since the author is preserved
regardless of who merges.

-- 
--Guido van Rossum (python.org/~guido)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-committers/attachments/20170920/c134f53b/attachment.html>