develooper Front page | perl.perl5.porters | Postings from February 2022

Re: Why do we still have AUTHORS?

Thread Previous | Thread Next
From:
demerphq
Date:
February 26, 2022 07:52
Subject:
Re: Why do we still have AUTHORS?
Message ID:
CANgJU+WBM3ADj3ANWA6JeinNiXiKmdbEyMnbWuUmmsYH88FPmw@mail.gmail.com
On Thu, 24 Feb 2022 at 00:11, Leon Timmermans <fawaka@gmail.com> wrote:

> We've been maintaining an AUTHORS file for more than twenty years, but in
> these days of git I'm kind of wondering why we still do this.
>
> We actually have tests that check if any new addition matches git, why
> don't we just rely on git instead? And remain a HISTORIC-AUTHORS file
> honoring the contributors who predate our transition to git.
>
> Right now it's a recurring source of test failures for new contributors,
> without providing any clear benefit IME. Why would we keep doing this?
>
>
Just for the record, I have decided to make creating a script that can
generate/update AUTHORs properly my "low stress boredom breaking" hacking
project for the next couple of days.  At the same time I am working on
fixing up the .mailmap to include the many many entries that are missing
from it.  If anybody else has picked this up let me know and either we can
join forces or I can drop it, although FWIW I have gotten pretty far
already.

If anyone knows of any pertinent facts that might impact this then I would
love to hear them.

A couple of questions to pose to the world:

* There are lots of entries in our commits that are ignored one way or
another, For instance names like "(Ilya Martynov)", which has no email, but
does have a commit where that  name is the author. This name is excluded
from Authors even though it need not be.

* Emails in the checkAUTHORS.pl are obfuscated to not include an ampersand.
This seems to be because it made coding the script easier and dates back to
NIcholas Clarks first commit. However it is conceivable that folks are also
exploiting the fact that this would hide their email from harvesters. If i
move those entries into .mailmap then they would no longer be obfuscated.
Anybody think this matters?

* .mailmap entries are often overspecified where they need not be. Eg,

Yves Orton <demerphq@gmail.com> <whatever@whatever.com>

is sufficient to fix the name and email from anything by
whatever@whatever.com. In some cases, such as where the email is perlbug or
rt or something like that it makes sense to specify it like:

Dude <whereever@wherever.com> Whatever <whatever@whatever.com>

but in many cases it simply does not. Anybody care about this?

* anybody NOT want something to be in .mailmap? Let me know, and let me
know why.

* some folks are apparently not supposed to have an email associated with
their name in AUTHORS. That is fine. But whatabout mailmap? Should i put an
entry in .mailmap that says "<no-email> or something like that? That would
fix most people's view of this data via git log.

* Any other comments or requests or whatnot please let me know.

Cheers,
Yves

-- 
perl -Mre=debug -e "/just|another|perl|hacker/"

Thread Previous | Thread Next


nntp.perl.org: Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at ask@perl.org | Group listing | About