| 1 |
W dniu czw, 29.03.2018 o godzinie 13∶24 +0200, użytkownik Ulrich Mueller |
| 2 |
napisał: |
| 3 |
> > > > > > On Thu, 29 Mar 2018, Michał Górny wrote: |
| 4 |
> > Next item: provided that EAPI 7 is approved, we'd have 4 'live' EAPIs |
| 5 |
> > in motion [1]. I'd like therefore request the Council to vote on: |
| 6 |
> > a. banning EAPI 4 for new ebuilds (and EAPI bumps of existing ebuilds). |
| 7 |
> > It has been deprecated on 2015-10-11. In the past, deprecated EAPIs were |
| 8 |
> > banned within 11/23 months from deprecation, so we're overdue. |
| 9 |
> |
| 10 |
> Fine with me. |
| 11 |
> |
| 12 |
> > 2. deprecating EAPI 5. In case of EAPIs 3-4 they were deprecated 4-5 |
| 13 |
> > years after being added. |
| 14 |
> |
| 15 |
> I think a better indicator is the time between support for EAPI n+1 in |
| 16 |
> stable Portage, and deprecation of EAPI n (see [1]). Using this, I get |
| 17 |
> 37 months for EAPI 2, 35 months for EAPI 3, and 34 months for EAPI 4 |
| 18 |
> deprecation. |
| 19 |
> |
| 20 |
> > EAPI 6 has been added on 2015-11-13, and even toolchain team already |
| 21 |
> > uses it, so there's really no reason to use EAPI 5 anymore. |
| 22 |
> |
| 23 |
> Stable Portage supports EAPI 6 since 2016-01-17, i.e. since 26 months. |
| 24 |
> So we would be somewhat on the early side. |
| 25 |
|
| 26 |
Not that it's less than the supported upgrade path. |
| 27 |
|
| 28 |
> What worries me more is that deprecation of EAPI 5 would apply to |
| 29 |
> profiles too. However, all profiles are still at EAPI 5 at this point, |
| 30 |
> and I don't see any value in upgrading them to EAPI 6. |
| 31 |
|
| 32 |
That's a fair argument. However: |
| 33 |
|
| 34 |
1. Does deprecation really mean anything in terms of profiles? Even |
| 35 |
in the context of EAPI bans we explicitly stated that it affects new |
| 36 |
packages and EAPI bumps. I think deprecating it for ebuilds is still |
| 37 |
meaningful even if profiles would stay EAPI 5. |
| 38 |
|
| 39 |
2. Do we want to keep profiles EAPI 5 indefinitely? If we consider it |
| 40 |
a goal to reduce the number of EAPIs in use, I think it would be |
| 41 |
reasonable to bump profiles to EAPI 6 proactively, even if it doesn't |
| 42 |
change anything. |
| 43 |
|
| 44 |
> |
| 45 |
> Ulrich |
| 46 |
> |
| 47 |
> > [1]:https://wiki.gentoo.org/wiki/Project:Package_Manager_Specification#Council_approval_and_use_in_Gentoo_repository |
| 48 |
|
| 49 |
-- |
| 50 |
Best regards, |
| 51 |
Michał Górny |
Archives