Running pidgin in debug mode produces no useful debug output either.
To reproduce:
Run Pidgin, start a voice call with yourself (e.g. use 2 XMPP accounts). Your "calling" window will open, but Pidgin won't report an incoming call. Thus voice/video calls are impossible to make.
This happens since Ubuntu Precise shippes the Farsight library, which formerly was called FarStream2. The renaming was accompanied by an API change.
The reason is that the patch to the pidgin source package, debian/patches/70_farstream_rename.patch, not fully reflects the API change.
I'll post a an initial patch shortly, to keep it separate from the bug report.
ProblemType: Bug
DistroRelease: Ubuntu 12.04
Package: libpurple0 1:2.10.2-1ubuntu1
ProcVersionSignature: Ubuntu 3.2.0-21.34-generic 3.2.13
Uname: Linux 3.2.0-21-generic x86_64
ApportVersion: 2.0-0ubuntu2
Architecture: amd64
Date: Mon Apr 2 22:01:33 2012
InstallationMedia: Ubuntu 12.04 LTS "Precise Pangolin" - Alpha amd64 (20120303)
ProcEnviron:
TERM=xterm
PATH=(custom, user)
LANG=de_DE.UTF-8
SHELL=/bin/bash
SourcePackage: pidgin
UpgradeStatus: No upgrade log present (probably fresh install)
This is an initial solution or workaround for the bug. The following steps/reasoning were involved:
- The OpenSUSE project has a patch reflecting some (but not all) of the API changes at:
https:/
(queried at Mo 2. Apr 20:29:30 UTC 2012)
It still needs an minor change to build successfully, which the Ubuntu package already has in its debian/
--- pidgin-
+++ pidgin-
@@ -42,1 +42,1 @@
-#include <gst/farsight/
+#include <farstream/
- However, the build libpurple still isn't able to do voice/video calls. This is because of the following API change (see http://
* Return NULL in "codecs" unless they are ready (and "codecs-ready"
is not needed anymore)
This basically means, we can port the FarSight2 code to Farstream by behaving as if the "codecs-ready" property of a Farstream session always were TRUE. However, if (when) we had a waiting loop for "codecs-ready" to become TRUE before, we should now wait for "codecs" to become non-NULL instead.
- As of now I implemented the former part and as a result voice calls do work for me again. Video calls also get set up properly, but no video is transmitted/
I feel having only voice calls working is better than none at all, therefore I believe the patch is useful already at this point. I acknowledge a proper patch for the transition will need more logical cleanup, but I don't have the necessary insight to do that right now. Besides, it's an issue upstream will need to care about soon (but probably not before April 26th).
The patch file is intended to be used as a replacement for debian/