Re: Python 2 removal in sid/bullseye: Progress and next steps

[Ondřej Surý <ondrej@sury.org>]

Sorry for top post, I am just on my phone. Just my two cents - the RoQA is a pretty standard thing when you deal with bigger transitions and there’s a lot of unmaintained packages.

Been there couple of times, and it was nowhere near py2 removal magnitude. So, I just want to send huge thanks towards the Python team and you have my support to make things as easier for you as you can. I think you drafted excellent and considerate plan.

O.

--

Ondřej Surý <ondrej@sury.org>

On 11 Nov 2019, at 22:15, Ondrej Novy <novy@ondrej.org> wrote:



Hi,

po 11. 11. 2019 v 16:27 odesílatel Norbert Preining <norbert@preining.info> napsal:

thanks for your work on the Python2 removal.

It's team work, I "only" sent that email :). It looks like others already replied, but

Could you please give a time line of how you are planning to proceed?

time line=now. 1.5 year is really short period for doing so much work in Debian. Imho we are already too late to make it, but let's try it. :)

I think requesting the removal of packages that you are **not**
maintaining is - to be polite - a bit unconventional (at least).
This remains at the discretion of the maintainer as far as I remember.

as other sad, RoQA. But maintainer can always stop this. Py2keep tag, fix package, even anyone else can NMU it. Removing is last and least preferred option. This is happening mostly for unmaintained/dead upstream packages, low popcons, MIA maintainers, QA packages, etc.

> All dependency fields in debian/control and debian/tests/control must
> also be updated to stop using the unversioned python

Are all you "must" statements "policy decisions"? Or your personal wish
list items?

Python policy update is underway.

Thank you.

--

Best regards
 Ondřej Nový