| 1 |
Dnia 2014-08-07, o godz. 11:41:31 |
| 2 |
Ulrich Mueller <ulm@g.o> napisał(a): |
| 3 |
|
| 4 |
> >>>>> On Thu, 7 Aug 2014, Michał Górny wrote: |
| 5 |
> |
| 6 |
> > Since the tree is still full of too broad dependencies and other |
| 7 |
> > mistakes, I would like to add a repoman check and a matching QA |
| 8 |
> > policy regarding how slots should be used in dependencies. |
| 9 |
> |
| 10 |
> > The check would apply to EAPI 5 and newer ebuilds only. Considering |
| 11 |
> > the past uproar against having slotted dependencies against |
| 12 |
> > single-slot packages, it would apply only to dependencies that match |
| 13 |
> > more than one slot of a package. |
| 14 |
> |
| 15 |
> For what type of dependencies would that check apply? IIUC, it would |
| 16 |
> only make sense for packages that appear both in DEPEND and RDEPEND. |
| 17 |
|
| 18 |
The test is applied to RDEPEND only. I don't think it's a good idea to |
| 19 |
play with intersections of dependencies -- that would be hard to |
| 20 |
implement and therefore fragile. |
| 21 |
|
| 22 |
Furthermore, :* guarantees runtime switching support. Lack of any |
| 23 |
operator indicates semi-undefined behavior, e.g. paludis tries |
| 24 |
the safest route possible and requires all slots installed at build |
| 25 |
time. |
| 26 |
|
| 27 |
-- |
| 28 |
Best regards, |
| 29 |
Michał Górny |
Archives