Contrib packages that probably should be forced into testing

[Steve Langasek]

--oC1+HKm2/end4ao3
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Wed, Jan 05, 2005 at 11:11:40AM +0100, Arnaud Vandyck wrote:
> Wed, 5 Jan 2005 01:20:37 -0800,=20
> Steve Langasek <vorlon@debian.org> wrote:=20

> > On Wed, Jan 05, 2005 at 09:48:02AM +0100, Arnaud Vandyck wrote:
> >> Wed, 5 Jan 2005 02:02:29 +0100,=20
> >> Jeroen van Wolffelaar <jeroen@wolffelaar.nl> wrote:=20
> >
> >> >      * charva (- to 1.0.1-3)
> >> >           + Maintainer: Debian Java Maintainers
> >> >           + Section: contrib/libs
> >> >           + 342 days old (needed 10 days)
> >> >           + libcharva1-java/i386 unsatisfiable Depends: libcharva1-j=
ni
> >> > NOTE: don't understand the above dependency issue...=20
> >> >           + Valid candidate

> >> charva has some part written in java (arch: all) and some in C. I think
> >> I built it with my powerpc.

> > The testing scripts only check the installability of arch: all packages=
 on
> > one of our architectures -- for obvious reasons, that architecture is i=
386.
> > libcharva1-java seems to be a rara avis, an arch: all package that's not
> > installable on i386.  But it could be: I would suggest getting someone =
to do
> > a binary upload of libcharva1-jni on i386.  In general, I think getting
> > contrib builds of JNI packages for both i386 and powerpc would be a good
> > idea.

> I don't know if free tools can do it easily.

Well, presumably this is the challenge with contrib packages, yes?  What I
meant to say here was that I think it's a good idea to get JNI packages
built for i386 and powerpc when possible; but a contrib JNI package that
fails to build on "the" arch for non-free Java seems of limited use, IMHO,
and I'm inclined to give such a package a pass for stable until it's more
generally useful.  (How's free Java in main coming again? :)

> Anyway, I've just try to build it on x86 and it FTBFS :(=20

Is this still the case?  Is this related to the bugs in kaffe on i386?

Cheers,
--=20
Steve Langasek
postmodern programmer

--oC1+HKm2/end4ao3
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: Digital signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQFB5QFhKN6ufymYLloRAt8dAJwN4Fsqh64hH4YQH39KHCnLD7mGOACg0Qym
sITIWEKCUMxIvqCj3aNhgzE=
=kqTy
-----END PGP SIGNATURE-----

--oC1+HKm2/end4ao3--