Re: [gentoo-dev] [RFC] Deadlines for next Python implementations

[Aaron Bauman <bman@gentoo.org>]

[-- Attachment #1: Type: text/plain, Size: 1797 bytes --]

On Mon, Jun 01, 2020 at 11:07:39PM +0200, Michał Górny wrote:
> On Mon, 2020-06-01 at 15:49 -0500, Matthias Maier wrote:
> > On Mon, Jun  1, 2020, at 15:27 CDT, Michał Górny <mgorny@gentoo.org> wrote:
> > 
> > > 2020-08-01  Python 3.7 migration deadline
> > > 
> > >    After this date, we lastrite all remaining packages that haven't been
> > >    ported.  This gives people roughly two months, with a ping one month
> > >    from now.
> > > [...]
> > > 2020-12-01  Python 3.8 migration deadline
> > > 
> > >    We lastrite all the unmigrated packages.
> > 
> > Most of the time (guess >99%) this "porting" simply consists of
> > "keywording" with the new python target, i.e., a one-line change in the
> > ebuild.
> > 
> > What about we "auto keyword" all remaining packages that have a
> > python3_6 target but lack the python3_7 target instead? Meaning, just
> > add the python3_7 value to the corresponding PYTHON_*TARGET.
> > 
> > Given the fact how little difference there is between python3_6 and
> > python3_7 this seems to be the appropriate, gentler approach here.
> > 
> 
> Most of these packages are unmaintained, seriously outdated and they may
> actually be broken with py3.7 (because they're so seriously outdated). 
> I don't see that as solving a problem, it merely shoves it under
> the carpet and leaves us with the same shove-under-the-carpet attitude
> for the next few years.
> 
> I like to think of these migrations as opportunity to fix some broken
> ebuilds, update some packages and last rite all the things that aren't
> maintained.
> 

Couldn't agree more here. Unfortunately, it is normally the big projects
that have to deal with cleaning up the cruft.

> -- 
> Best regards,
> Michał Górny
> 

-- 
Cheers,
Aaron

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]