After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 425580 - Gstreamer backends should advertise when a filter is actually added
Gstreamer backends should advertise when a filter is actually added
Status: RESOLVED FIXED
Product: rhythmbox
Classification: Other
Component: Plugins (other)
HEAD
Other Linux
: Normal enhancement
: ---
Assigned To: RhythmBox Maintainers
RhythmBox Maintainers
Depends on:
Blocks:
 
 
Reported: 2007-04-02 18:47 UTC by Christophe Dehais
Modified: 2007-06-02 02:16 UTC
See Also:
GNOME target: ---
GNOME version: ---


Attachments
add and remove callbacks for tees and filters (41.02 KB, patch)
2007-05-31 13:22 UTC, Jonathan Matthew
none Details | Review
patch (12.66 KB, patch)
2007-05-31 13:33 UTC, James "Doc" Livingston
committed Details | Review

Description Christophe Dehais 2007-04-02 18:47:59 UTC
When adding an element to the pipeline, a plugin has no way to know when the element is actually inserted and correctly linked inside the pipeline.

As a side effect there is currently no standard place where one can safely retrieve the gst bus an element is emitting messages to.

This could be done by adding a signal such as "element-inserted" to the PLAYER_GST_FILTER interface.
Comment 1 Jonathan Matthew 2007-05-31 13:22:06 UTC
Created attachment 89102 [details] [review]
add and remove callbacks for tees and filters

I've only tested this with the visualisation plugin when using the crossfading backend, but it works OK there.
Comment 2 James "Doc" Livingston 2007-05-31 13:33:06 UTC
Created attachment 89103 [details] [review]
patch

This adds the above signal, and a matching pre-remove signal, to the RBPlayerGst* interfaces and makes the backends use it. I haven't done a huge amount of testing, but it seems to work fine.
Comment 3 Jonathan Matthew 2007-06-01 00:39:39 UTC
OK, your approach is much less hideous than mine.  For some reason I thought that using signals would be a bit awkward, but that turns out not to be the case.
Comment 4 Jonathan Matthew 2007-06-01 00:43:27 UTC
Also, I've attached a patch to bug 441927 that makes use of the tee signals.
Comment 5 James "Doc" Livingston 2007-06-02 02:15:44 UTC
Committed to svn trunk.
Comment 6 James "Doc" Livingston 2007-06-02 02:16:04 UTC
Committed to svn trunk.