1 |
> On 13 Jul 2022, at 19:50, Arthur Zamarin <arthurzam@g.o> wrote: |
2 |
> |
3 |
> On 13/07/2022 11.12, Ulrich Mueller wrote: |
4 |
>>>>>>> On Mon, 11 Jul 2022, Ulrich Mueller wrote: |
5 |
>> |
6 |
>> |
7 |
>> So, any opinions? Should we go for the longer transition time (and make |
8 |
>> overlay maintainers happy), or for a shorter time so that we can tidy up |
9 |
>> eclasses sooner? |
10 |
>> |
11 |
>> Ulrich |
12 |
> |
13 |
> My personal take on this question. Faster deprecation of EAPI ebuilds in |
14 |
> ::gentoo repo (as we can control it), but longer time until we remove it |
15 |
> from eclasses. Note that I don't mean here deprecation, only removal. |
16 |
> |
17 |
> I think that with current EAPI>=6 state, the "weight of supporting" |
18 |
> EAPI=6 isn't very heavy, so some extra time for overlays will be nice. I |
19 |
> do know that I don't help a lot in eclass maintenance, so if I wrong in |
20 |
> this statement, I won't complain of course. |
21 |
> |
22 |
> Maybe (?) this will also help during bumps of old systems (not that we |
23 |
> care too much, as we give them along timeframe for this and they can use |
24 |
> snapshots of repo, but why not as an extra bonus). |
25 |
> |
26 |
|
27 |
Yeah, I broadly agree with this. Making things predictable for |
28 |
downstreams is important. |
29 |
|
30 |
Best, |
31 |
sam |