[Bug 642730] New features for AAC and other passthrough support

[GStreamer (bugzilla.gnome.org)]

https://bugzilla.gnome.org/show_bug.cgi?id=642730
  GStreamer | gst-plugins-base | 0.10.17

--- Comment #11 from Arun Raghavan <arun at accosted.net> 2011-04-14 08:34:34 UTC ---
(In reply to comment #9)
> Review of attachment 185929 [details]:

Thanks for the review!

> ::: gst-libs/gst/audio/gstbaseaudiosink.c
> @@ +1716,3 @@
> +done:
> +  if (out)
> +    gst_buffer_unref (out);
> 
> Does this unref here make sense if it is done only for the case when payloading
> was done? Shouldn't that buffer always be unreffed or never?

We need to keep this here because if we need to payload, we set buf = out and
thus can't actually unref that buffer until we're done working with it.

> @@ +1733,3 @@
> +payload_failed:
> +  {
> +    GST_DEBUG_OBJECT (sink, "failed to payload data for iec958 transmission");
> 
> This is not only for IEC958 I suppose? Just remove it from the debug output :)

The error itself is useful when debugging problems. I'll remove the IEC958 bit.

> ::: gst-libs/gst/audio/gstbaseaudiosink.h
> @@ +145,3 @@
> + *           payloading is required, returns a reffed copy of the original
> + *           buffer, else returns the payloaded buffer with all other metadata
> + *           copied.
> 
> Please add here that the subclass is supposed to take ownership of the passed
> buffer

The subclass doesn't actually take ownership of the buffer. Or am I
misunderstanding what you're saying?

-- 
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.