[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#946241: ffmpeg: please make autopkgtests cross-test-friendly



On 2020-01-03 01:08:14, James Cowgill wrote:
> Hi,
> 
> On 06/12/2019 03:33, Steve Langasek wrote:
> > Package: ffmpeg
> > Version: 7:4.2.1-2
> > Severity: minor
> > Tags: patch
> > User: ubuntu-devel@lists.ubuntu.com
> > Usertags: origin-ubuntu focal ubuntu-patch
> > 
> > Dear maintainers,
> > 
> > In Ubuntu, we are in the process of moving the i386 architecture to a
> > compatibility-only layer on amd64, and therefore we are also moving our
> > autopkgtest infrastructure to test i386 binaries in a cross-environment.
> > 
> > This requires changes to some tests so that they are cross-aware and can do
> > the right thing.
> 
> [...]
> > 
> > --- ffmpeg-4.2.1/debian/tests/control	2019-11-01 18:17:31.000000000 -0700
> > +++ ffmpeg-4.2.1/debian/tests/control	2019-12-05 17:17:44.000000000 -0800
> > @@ -5,4 +5,4 @@
> >  Depends: ffmpeg
> >  
> >  Tests: encdec-extra
> > -Depends: ffmpeg, libavcodec-extra
> > +Depends: ffmpeg, libavcodec-extra58
> 
> Am I right in thinking this is necessary because libavcodec-extra is an
> "arch all multi-arch foreign" package?
> 
> I'm wondering if marking libavcodec-extra (and libavfilter-extra) like
> that is wrong because the behavior of these packages depends on what the
> native architecture is, but I'm not sure. It currently installs the
> native arch's libavcodec-extra library which doesn't seem right for a
> multi-arch foreign package. I don't know if anyone from the multimedia
> team knows why it's done this way (maybe just historical)?
> 
> Making the package "arch any multi-arch same" seems sensible to me, or
> just removing the packages altogether (we already have a Provides:
> libavcodec-extra on the real library packages so the existing
> dependencies should work still).

The package is intended for users that want the extra codecs. So
converting it to arch any would be fine, dropping it not.

Best

> 
> Thanks,
> James
> 




-- 
Sebastian Ramacher


Reply to: