On 22/11/13 08:25, Josh Allmann wrote:
> Instead of assuming a single payload type per RTSPStream,
> store a collection of payload types, along with the necessary
> data (indexes, transports, dynamic handlers, etc).

Nice =)

> Upon reading the first RTP-UDP packet, set the transport_priv
> pointer for the RTSPStream to the transport for the corresponding
> payload (via pick_transport_priv), and the rest of the processing
> proceeds as usual.

Probably it is the best approach, just... what happens when you get an
alternate?

> Only tested with RTP-UDP. Most likely breaks RTSP, TCP, RDT,
> leaks and/or corrupts data, etc. Clearly, a work-in-progress.

Write down in wiki.libav.org/Blueprint/RTSP what you are trying to
achieve so more people (beside me and Martin) can form an opinion =)

The overall problem is that for each m= (think about logical stream) you
might get more than an a= (think about stream as AVStream).

SDP can use this to have an Offer/Answer mediation or just have variants
available that you would hop from/to containing the same information.

We should have some sketchy thing to bind together streams are you
interested on sorting out that part as well?

lu
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to