1 |
On Mon, Oct 1, 2018 at 8:27 AM Gilles Dartiguelongue <eva@g.o> wrote: |
2 |
> |
3 |
> Le lundi 01 octobre 2018 à 07:40 +0200, Rémi Cardona a écrit : |
4 |
> > Le 01/10/2018 à 00:50, Mike Gilbert a écrit : |
5 |
> > > update-desktop-database and update-mime-database are called from |
6 |
> > > ROOT in |
7 |
> > > pkg functions, so the related dependenices belong in RDEPEND. |
8 |
> > > |
9 |
> > > Signed-off-by: Mike Gilbert <floppym@g.o> |
10 |
> > |
11 |
> > As far as the eclass goes, this is correct. But AFAIR, this was |
12 |
> > needed |
13 |
> > because some packages look for those tools at build time. It was ages |
14 |
> > ago so maybe it no longer applies. |
15 |
> > |
16 |
> > Rémi |
17 |
> > |
18 |
> |
19 |
> A lot of autotools based packages do explicitly search for it. I don't |
20 |
> think they fail if it is missing though but a tinderbox run would be |
21 |
> welcome. |
22 |
|
23 |
That's... interesting. I think it would make more sense for these |
24 |
packages to state this dependency explicitly instead of relying on the |
25 |
dep from the eclass. |
26 |
|
27 |
> The reason why it is in DEPEND though is that none of these tools are |
28 |
> required at runtime. They are needed at postinst and postrm stages |
29 |
> which afaik makes them DEPEND on EAPI previous to EAPI 7 and BDEPEND in |
30 |
> EAPI 7 if I'm not mistaken. |
31 |
|
32 |
This is incorrect; programs executed in pkg_postinst/pkg_postrm should |
33 |
be present in RDEPEND. Consider a binpkg installation, which would |
34 |
ignore DEPEND, but would still execute pkg_postinst. |
35 |
|
36 |
Another option: since xdg-utils.eclass functions will politely skip |
37 |
the updates if the tools are missing, we could just drop the |
38 |
dependency from xdg.eclass entirely and wait for the tools to get |
39 |
installed through some indirect means. |