1 |
El jue, 25-07-2013 a las 22:30 -0600, Ryan Hill escribió: |
2 |
> On Thu, 25 Jul 2013 09:26:48 -0700 |
3 |
> ""Paweł Hajdan, Jr."" <phajdan.jr@g.o> wrote: |
4 |
> |
5 |
> > About one month ago I've filed |
6 |
> > <https://bugs.gentoo.org/show_bug.cgi?id=474358> about modernizing |
7 |
> > toolchain.eclass by creating new toolchain-r1.eclass and migrating |
8 |
> > ebuilds using it to the new eclass. |
9 |
> > |
10 |
> > Please see attachments and review the code. |
11 |
> > |
12 |
> > One issue has already been raised, and it's prefix-related changes. I |
13 |
> > don't know what to change there, but I'm happy to test suggested changes. |
14 |
> > |
15 |
> > Then there is a question whether toolchain packages should use EAPI 5, |
16 |
> > and I think providing an upgrade path is a good concern. Given |
17 |
> > portage/python constraints though, it seems to me it would be fine. If |
18 |
> > you think it'd be better, I could use a lower EAPI just in case. |
19 |
> > |
20 |
> > All feedback is welcome. |
21 |
> |
22 |
> I meant to work on this last week but got distracted. I have a bunch of |
23 |
> build changes testing locally but need to make some cross compilers. |
24 |
> |
25 |
> I don't think we will be moving to 5 very soon. I have nothing against it but |
26 |
> Mike might be a harder sell. I want USE deps so I'm going to do 2 at least, |
27 |
> then get the prefix guys on board for 3. |
28 |
> |
29 |
> Like I said on the bug I don't think we want to do a new eclass (or if we did I |
30 |
> would make a toolchain-next for masked versions and backport stuff). |
31 |
> |
32 |
> |
33 |
|
34 |
Last time I talked with him, Mike was ok with eapi4 for base-system |
35 |
packages, but no idea if toolchain will have a special treatment. Better |
36 |
wait for him to reply here :/ |