1 |
On Wed, 29 Aug 2012 18:18:20 -0400 |
2 |
Mike Frysinger <vapier@g.o> wrote: |
3 |
|
4 |
> On Wed, Aug 29, 2012 at 6:14 PM, Michał Górny wrote: |
5 |
> > On Wed, 29 Aug 2012 18:05:19 -0400 Mike Frysinger wrote: |
6 |
> >> On Wed, Aug 29, 2012 at 6:02 PM, Michał Górny wrote: |
7 |
> >> > On Wed, 29 Aug 2012 17:50:16 -0400 Mike Frysinger wrote: |
8 |
> >> >> On Wed, Aug 29, 2012 at 5:42 PM, Michał Górny wrote: |
9 |
> >> >> > In other words, pkg-config is only used when no other criteria |
10 |
> >> >> > allows it to classify the particular .la file as suitable for |
11 |
> >> >> > removal or not. Sadly, it's rather, ehm, unfriendly to ebuild |
12 |
> >> >> > developers who obviously don't even read the relevant part. |
13 |
> >> >> > |
14 |
> >> >> > Do you have any ideas how we can improve that? |
15 |
> >> >> |
16 |
> >> >> before the func executes pkg-config, run `has virtual/pkgconfig |
17 |
> >> >> ${DEPEND}` and spit an eqawarn if it's not found |
18 |
> >> > |
19 |
> >> > Ciaran will shot at me for doing that. |
20 |
> >> |
21 |
> >> it isn't violating anything and can find real bugs. i don't see a |
22 |
> >> problem here. |
23 |
> > |
24 |
> > It is violating the Holy PMS. |
25 |
> |
26 |
> does it actually ? are DEPEND variables not allowed to be expanded in |
27 |
> pkg_* src_* funcs ? |
28 |
> |
29 |
> we could probably add a similar check to autotools.eclass: grep for |
30 |
> PKG_PROG_PKG_CONFIG and check ${DEPEND} |
31 |
> |
32 |
> >> >> > One thing that comes into my mind is finally making pkgconfig |
33 |
> >> >> > a required, implicit part of toolchain (or @system). Since we |
34 |
> >> >> > have pkgconf now, this is more feasible than before. |
35 |
> >> >> |
36 |
> >> >> i don't think making it part of the toolchain makes sense. i'd |
37 |
> >> >> rather not add it to @system simply to keep a few packages from |
38 |
> >> >> sometimes failing. |
39 |
> >> > |
40 |
> >> > I'd add it to @system because a lot of packages actually need to |
41 |
> >> > DEPEND on pkgconfig because they use libraries using .pc files. |
42 |
> >> > And the number is going to increase, hopefully. |
43 |
> >> |
44 |
> >> sure, but keeping things in @system doesn't make much sense: |
45 |
> >> - there's a penalty (as noted in old threads) |
46 |
> >> - it isn't actually required at runtime, so it's bloat on reduced |
47 |
> >> systems |
48 |
> > |
49 |
> > I think it's practically the same as compiler. |
50 |
> |
51 |
> that isn't a bad view point, but for the purposes of this discussion, |
52 |
> i don't think it's relevant :) |
53 |
|
54 |
Will it be a better view point if I opened a separate discussion about |
55 |
putting pkg-config in @system? It could get more attention probably. |
56 |
|
57 |
-- |
58 |
Best regards, |
59 |
Michał Górny |