[Bug 646566] Protect against Pad-Parent disappearing

[GStreamer (bugzilla.gnome.org)]

https://bugzilla.gnome.org/show_bug.cgi?id=646566
  GStreamer | gstreamer (core) | git

--- Comment #18 from Olivier Crete (Tester) <olivier.crete at ocrete.ca> 2011-04-04 17:41:19 UTC ---
(In reply to comment #17)

I meant, I don't know any case outside of the unit tests where the callback
from the pads should be called without holding the parent (or without having a
parent set). My proposal was to never call the functions from the pads unless
the parent is set and "hold" the parent until the callback returns.

I'm also getting convinced that we should have something at the GstObject
level, so we can "lock" the gstobject parent as returned by
gst_object_get_parent(), so that it always succeeds during the callbacks (so we
don't have to do those checks). Or maybe just not allow changing the parent,
only setting it once and then only allow to unset it once?

-- 
Configure bugmail: https://bugzilla.gnome.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug.
You are the assignee for the bug.