bug#69646: [PATCH] Have insert 'submit-emacs-patch' maintainers into X-Debbugs-Cc

[Philip Kaludercic <philipk@posteo.net>]

Andrea Corallo <acorallo@gnu.org> writes:

> Philip Kaludercic <philipk@posteo.net> writes:
>
>> Andrea Corallo <acorallo@gnu.org> writes:
>>
>>> Philip Kaludercic <philipk@posteo.net> writes:
>>>
>>>> Tags: patch
>>>>
>>>>
>>>> Here's a suggestion, to have the submit-emacs-patch command
>>>> automatically ping maintainers, in case they are not following the bug
>>>> tracker.  Basically it goes through all the files and extracts the
>>>> information from the Maintainer: header.
>>>>
>>>> One non-technical issue here is that the maintainer information might
>>>> not be up to date.  I've run a quick scan to see who is mentioned how
>>>> frequently.  It might be necessary to clean up this list or to check who
>>>> is still around:
>>>
>>> Hi Philip,
>>>
>>> IIUC with this patch I'd be in Cc of all bugs? If that's the case I'm
>>> not personally in favor of this, I score mails on bug-gnu-emacs also
>>> based on me being in Cc or not.
>>
>> Why should that be?  I have tested the new command out with bug#69647,
>> and you weren't added automatically.  In fact, a quick grep tells me you
>> aren't listed as the maintainer in any file, so you wouldn't be affected
>> by this.
>
> Maybe your patch should parse also admin/MAINTAINERS and collect names
> from there as well?

Imaginable, but that is much more difficult to parse.  If we want to do
this, when some more general function should be written to extract this
information.

>>> I think the bug is that maintainers should be subscribed to
>>> bug-gnu-emacs (otherwise I'm not sure how they can actually maintain).
>>
>> The issue is that for someone like me, who maintains a few little things
>> around Emacs, following all bugs and all discussions would be too much.
>>
>>> Isn't a long list of people in Cc technically just another mailing list?
>>
>> It shouldn't be a long list, just the people responsible for the
>> specific files being modified.  My idea is to have something comparable
>> to Linux's "get_maintainer.pl"[0], and a bit more streamlined (assuming
>> people even use M-x submit-emacs-patch).
>
> If we want to Cc maintainers who are impacted by a patch I'm in favor,
> if we want to Cc all maintainers then I'm not.  I haven't read your
> patch but from the initial description I thought was the first case,
> I've now the impression is the second.  Apologies if there was some
> miss-understanding on my side.

No, definitely the second case.  The data-flow is

 .patch file -> list of modified files -> list of maintainers of these files

and only those are added to X-Debbugs-CC.

> Thanks
>
>   Andrea

-- 
	Philip Kaludercic on peregrine