1 |
On Thu, 15 Aug 2013 10:04:47 +0200 |
2 |
Pacho Ramos <pacho@g.o> wrote: |
3 |
> Wouldn't be much easy to try to get sets support approved for the next |
4 |
> eapi? (eapi6 I think). Once we get the usual problems, we can complain |
5 |
> but, who knows, maybe (as it's already implemented in a PM) it doesn't |
6 |
> take so long to get approved (or maybe I am being too optimistic :( ) |
7 |
|
8 |
Of course. All you have to do is propose a sane format for them -- this |
9 |
has been the blocker the last few times this issue has come up. |
10 |
|
11 |
The big question the Council will probably want answered is whether or |
12 |
not sets are allowed in package.mask and the like. If the answer is no, |
13 |
you're removing a large part of their usefulness. If the answer is yes, |
14 |
how are you controlling backwards compatibility? |
15 |
|
16 |
> > (not well maintained: simple patches take months to get applied, and |
17 |
> > even then often need council interference to be applied. Does not |
18 |
> > reflect reality: Multiple cases like mandating bash 3.2 that we |
19 |
> > don't even have in tree anymore, so no compliance testing possible. |
20 |
> |
21 |
> Maybe a quick new eapi bump (5.1?) including this and other small |
22 |
> changes that are quick to implement could help :/ |
23 |
|
24 |
People seem to be opposed to "lots of EAPIs" or "too many new EAPIs". |
25 |
There's a fairly long delay between them because that's what developers |
26 |
have been asking for. |
27 |
|
28 |
> > Not |
29 |
> > documenting package.mask as a directory for EAPI0 even when that |
30 |
> > feature existed in portage before the initial release of PMS. Etc. |
31 |
> > etc.) |
32 |
> > |
33 |
> |
34 |
> I wasn't aware of this issue at all, does it have a bug report |
35 |
> tracking it? (for knowing its status, why it is being ignored or |
36 |
> bringing the problem to the council if needed) Please take care that |
37 |
> not all people are aware of the PMS related issues :) |
38 |
|
39 |
It's not an issue at all. PMS followed the Portage documentation at the |
40 |
time (and unless it's changed recently, what the Portage documentation |
41 |
still says). It's just that Portage reuses code in such a way that |
42 |
there are accidental undocumented "features" every now and again, and |
43 |
this is one of them that someone spotted and started using. Directories |
44 |
for package.mask were introduced as a user config feature, not a tree |
45 |
feature (read the commit message that added the feature to Portage). |
46 |
|
47 |
-- |
48 |
Ciaran McCreesh |