discoverer, uridecodebin, encodebin and multiple audio streams

[Christian Fredrik Kalager Schaller]

Hi,
I been working on supporting input files with multiple audio streams in
Transmageddon. The test file I am using got two mp3 audio streams which
in a technical level seems to be quite identical. What is not clear to
me is how to interact with them in a useful fashion. Can I assume that
the behaviour of discoverer and uridecodebin is deterministic? ie. the
first stream found by discoverer will be the first stream found by
uridecodebin?

Regardless it would be nice if we could come up with a system where
discoverer could create some kind of unique identifier for each stream
it finds. That way when I am connecting uridecodebin to encodebin or
other elements I can look for the pad with that unique identifier and
connect that, would probably simplify writing applications to handle
this case a lot. If discoverer and uridecodebin where also able to
automatically attach any language information to each stream, based on
tag data etc. that would also be great.

Christian

_______________________________________________
gstreamer-devel mailing list
[hidden email]
http://lists.freedesktop.org/mailman/listinfo/gstreamer-devel

[Tim-Philipp Müller-2]

On Sun, 2011-07-17 at 11:06 +0100, Christian Fredrik Kalager Schaller
wrote:

> Can I assume that the behaviour of discoverer and uridecodebin is
> deterministic? ie. the first stream found by discoverer will be the
> first stream found by uridecodebin?

Well, it may sometimes work, but you can't really assume that.

> Regardless it would be nice if we could come up with a system where
> discoverer could create some kind of unique identifier for each stream
> it finds. That way when I am connecting uridecodebin to encodebin or
> other elements I can look for the pad with that unique identifier and
> connect that, would probably simplify writing applications to handle
> this case a lot.

Yes, such stream IDs would be good. The demuxers/parsers/whatever should
create these. It's not really clear what should go into such a stream
though (Should it represent the topology? Or just be something unique to
he demuxer/file?)

This would best be pursued further in bugzilla, I see you just filed
https://bugzilla.gnome.org/show_bug.cgi?id=654830


> If discoverer and uridecodebin where also able to
> automatically attach any language information to each stream, based on
> tag data etc. that would also be great.

Not sure what you mean here. Attach what language information based on
tag data? (Besides the language tags that exist in the tag lists)

 Cheers
  -Tim


_______________________________________________
gstreamer-devel mailing list
[hidden email]
http://lists.freedesktop.org/mailman/listinfo/gstreamer-devel