Bug#296211: libgtk2.0-0 is broken in sid; doesn't export necessary symbols

Szucs Gabor Szucs Gabor <sniper@sniper.sch.bme.hu>, 296211@bugs.debian.org
Mon, 21 Feb 2005 10:33:58 +0100


--nextPart2148905.fgMVaJd4l1
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

On Monday 21 February 2005 09.25, you wrote:
> severity 296211 important
> tags 296211 - experimental
> thanks
>
>         Hi,
>
> On lun, f=E9v 21, 2005, Szucs Gabor wrote:
> > Package: libgtk2.0-0
> > Version: 2.4.14-2
> > Severity: critical
> > Tags: experimental
>
>  This is the second bug you report with severity higher than important
>  and with the tag "experimental".  Why are you using that tag?
>
Thought that the tag is independent from the severity of the bug. I'm sorry.

> > Please excuse me if it's a false report, but for me, it caused a
> > serious problem. Some apps (including firefox) failed to start,
> > dropping error messages about unresolved symbols (g_assert_warning and
> > g_option_context_new, as I remember; libglib related stuff).
> > Reverting to libgtk2.0-0 in the testing branch solved the issue. (I'm
> > using unstable) However, the cross-dependencies are still there in the
> > testing distribution.
>
>  Please provide the exact error message, firefox is running on my box
>  which is tracking unstable.
>
Install time:

Setting up libgtk2.0-bin (2.6.2-3) ...
Updating the IM modules list for GTK+-2.4.0...Cannot load   =20
module /usr/lib/gtk-2.0/2.4.0/immodules/im-xim.so: /usr/lib/gtk-2.0/2.4.0/i=
mmodules/im-xim.so:=20
undefined symbol: g_assert_warning

Runtime:
Will be difficult to cite the message exactly; it's the standard message ab=
out=20
unresolved symbols in a library. If you're sure you need that, I'll install=
=20
the broken packages again and post the message. The previous one was still =
in=20
the console cache.
I'm using firefox1.0* if that helps.

>  You seem to have kept libgtk2.0-common in the version from unstable,
>  that is probably the cause of the error with symbols. =20

It isn't, because of the reason I've mentioned below. It's completely=20
unrelated to this issue.

>  Please upgrade=20
>  completely to unstable or be sure you're tracking sarge.
>
I'm using a complete unstable system (in fact, I _were_ using). The very on=
ly=20
packages from sarge are libgtk2.0-0 and libgtk2.0-bin. I did this (ie. that=
 I=20
downgraded (downgraded due to the problems with libgtk2.0* in sid) only 2=20
packages) to minimalize the incompatibility between the two branches and th=
e=20
impact on my system.

Bye,

Sniper

>    Regards,

--nextPart2148905.fgMVaJd4l1
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)

iD8DBQBCGasKv1tUvJtrnTURAgs+AJ4vfWD5O1Ix0mAfaK1cYJKNrbBXbwCfShYq
8fesCkTQTM9DztGvgx8HSXI=
=LT0X
-----END PGP SIGNATURE-----

--nextPart2148905.fgMVaJd4l1--