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 357613 - XvStopVideo in xvimagesink
XvStopVideo in xvimagesink
Status: RESOLVED FIXED
Product: GStreamer
Classification: Platform
Component: gst-plugins-base
git master
Other Linux
: Normal normal
: 0.10.11
Assigned To: GStreamer Maintainers
GStreamer Maintainers
Depends on:
Blocks:
 
 
Reported: 2006-09-25 13:53 UTC by Stefan Sauer (gstreamer, gtkdoc dev)
Modified: 2006-11-15 10:05 UTC
See Also:
GNOME target: ---
GNOME version: ---



Description Stefan Sauer (gstreamer, gtkdoc dev) 2006-09-25 13:53:12 UTC
in xvimagesink.c::gst_xvimagesink_change_state() on 
GST_STATE_CHANGE_READY_TO_PAUSED in calls
gst_xvimagesink_xwindow_clear() which in turn() calls XvStopVideo().

I would expect the gst_xvimagesink_xwindow_clear() call when doing GST_STATE_CHANGE_PAUSED_TO_READY:

Or maybe it needs to go to gst_xvimagesink_xwindow_destroy() before 'if (xwindow->internal)'
Comment 1 Stefan Sauer (gstreamer, gtkdoc dev) 2006-09-29 11:39:44 UTC
on the other hand this call was introduced when trying to fix PLAYING -> READY -> PLAYING, so adding it to gst_xvimagesink_xwindow_destroy() would call it wehn going to NULL.
Comment 2 Stefan Sauer (gstreamer, gtkdoc dev) 2006-10-02 10:40:28 UTC
without the XvStopVideo() call, the last frame stays on the video window, when pausing, but also when the playback is finished (eof).

More funnily, when playback crashed and one starts the playback again, the old picture is shown for a brief moment.

So I suggest to add it when going to READY or NULL. Adding t to gst_xvimagesink_xwindow_destroy() works well here.
Comment 3 Stefan Sauer (gstreamer, gtkdoc dev) 2006-11-09 07:07:22 UTC
2006-11-09  Stefan Kost  <ensonic@users.sf.net>

	* sys/xvimage/xvimagesink.c: (gst_xvimagesink_change_state):
	  clear xv when going to NULL, remove // commented non-existant proto
Comment 4 Tim-Philipp Müller 2006-11-14 19:04:45 UTC
Has this actually been committed? I can't find a trace of this in either the ChangeLog or ViewCVS ...
Comment 5 Stefan Sauer (gstreamer, gtkdoc dev) 2006-11-15 10:05:51 UTC
recommitted, cvs was bugging me with
cvs commit: sticky tag `1.162' for file `sys/xvimage/xvimagesink.c' is not a branch

something broke the local CVS/Entries file.