On Mon, 2008-08-11 at 15:33 -0400, Christopher Aillon wrote:
> Ville-Pekka Vainio wrote:
> > Braden McDaniel wrote:
> >> On Tue, 2008-07-29 at 11:54 -0400, Will Woods wrote:
> >>> In last week's QA meeting, Chris Aillon (aka caillon, our fearless
> >>> firefox/xulrunner maintainer) stopped by to tell us what happened with
> >>> xulrunner dep breakage, and how package maintainers can help reduce /
> >>> prevent it in the future. Here's a quick summary:
> >>>
> >>> There are two APIs provided by xulrunner - the stable API (gecko-devel),
> >>> and the unstable one (gecko-devel-unstable).
> >> Why does xulrunner-devel-unstable provide some of the same headers (at a
> >> different path) that xulrunner-devel does? I'm specifically noticing
> >> SpiderMonkey headers; though there might be others.
> >
> > It's been almost two weeks and this question hasn't been answered yet, any
> > comments anyone?
>
> Hadn't noticed it sorry. Unstable builds need to be able to use the
> stable headers, too. The duplication is unfortunate, but it's a
> byproduct of the way the pkgconfig files are done. They include either
> a stable or unstable subdirectory, and sorting out the symlinks would be
> a pain.
Having the unstable headers depend on the stable ones seems like the
right approach here.
Modifying the pkg-config metadata to include -I flags for both the
stable and unstable directories is easy enough; so a state other than
that strikes me as another symptom of the problem rather than the point
of failure itself.
Is the problem you allude to ("sorting out the symlinks") that upstream
is dumping these stable headers (or symlinks thereto) in the unstable
hierarchy? If so, that seems worthy of an upstream bug report.
--
Braden McDaniel e-mail: <braden@xxxxxxxxxxxxx>
<http://endoframe.com> Jabber: <braden@xxxxxxxxxx>
--
fedora-devel-list mailing list
fedora-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/fedora-devel-list
