1 |
On Wed, 25 Jul 2012 18:38:05 +0200 |
2 |
Ulrich Mueller <ulm@g.o> wrote: |
3 |
|
4 |
> >>>>> On Wed, 25 Jul 2012, Mike Frysinger wrote: |
5 |
> |
6 |
> >> Our current policy [1] requires that ebuilds must assign the seven |
7 |
> >> variables DESCRIPTION, HOMEPAGE, SRC_URI, LICENSE, SLOT, KEYWORDS, |
8 |
> >> and IUSE, even if their value is empty. |
9 |
> >> |
10 |
> >> Could we drop this requirement? Repoman already enforces that |
11 |
> >> DESCRIPTION, HOMEPAGE, LICENSE, SLOT, and KEYWORDS are non-empty |
12 |
> >> (with some exceptions for virtuals). I don't see why we need to |
13 |
> >> distinguish the "empty value" and "not assigned" cases. |
14 |
> |
15 |
> > i think we should clarify and say that when an eclass provides |
16 |
> > these, the ebuild need not. completely missing DESCRIPTION/HOMEPAGE |
17 |
> > should be a warning (and maybe KEYWORDS), and LICENSE should be an |
18 |
> > error. there are plenty of examples of SRC_URI not being set and |
19 |
> > that's fine (live ebuilds, ebuilds that only install out of |
20 |
> > $FILESDIR, virtuals, etc...). |
21 |
> |
22 |
> I think we have to distinguish between PMS and tree policy here. |
23 |
> The package manager should be able to handle any empty or missing |
24 |
> variables (except for DESCRIPTION and SLOT). Otherwise we'd have to |
25 |
> complicate the spec with additional case distinctions, e.g. for |
26 |
> virtuals. |
27 |
|
28 |
PMS should be able to handle empty DESCRIPTION (i.e. for hand-written |
29 |
dirty test ebuilds). And an empty SLOT has its meaning. |
30 |
|
31 |
-- |
32 |
Best regards, |
33 |
Michał Górny |