[Bug 762884] rtspsrc: allow to control headers via properties

[GStreamer (GNOME Bugzilla)]

https://bugzilla.gnome.org/show_bug.cgi?id=762884

--- Comment #35 from Matt Staples <staples255 at gmail.com> ---
(In reply to Sebastian Dröge (slomo) from comment #34)
> (In reply to Matt Staples from comment #33)
> > Review of attachment 344659 [details] [review] [review]:
> > 
> > ::: gst/rtsp/gstrtspsrc.c
> > @@ +431,3 @@
> > +
> > +  myboolean = g_value_get_boolean (handler_return);
> > +
> > 
> > This was just a cut/paste from select_stream_accum, following the same
> > pattern.  I'm not sure the debug message is needed here at all.  Ok to just
> > remove it?
> 
> Sure
> 
> > @@ +5379,3 @@
> > +  if (!allow_send) {
> > +    GST_DEBUG_OBJECT (src, "skipping message, disabled by signal");
> > +    return GST_RTSP_OK;
> > 
> > I'm not sure what you mean.  Should the signal handler take something other
> > than the allow_send boolean?  like, (send, skip or error)?
> 
> Maybe let the signal return a GstRTSPResult

The original idea of the boolean return value was to allow the signal handler
to indicate that the given RTSP message (e.g., PAUSE) should be skipped
entirely (i.e., not sent to the RTSP server) without otherwise causing an
error.  Returning something other than GST_RTSP_OK from here would cause the
whole protocol to end in error as if there had a problem communicating with the
RTSP server, and that wasn't the intended behavior.  I'm not sure why an
application would want to force the protocol to error out here, but it seems
like there are other ways it could achieve that.

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.