Re: [rtcweb] Inspect a received SDP *before* using it

[Harald Alvestrand <[e-mail]>]

Since this is strictly an API issue, I believe it belongs on the 
[e-mail] list.

On 01/04/2013 11:24 AM, Iñaki Baz Castillo wrote:
> Hi, the WebRTC JavaScript API does not define a mechanism for
> parsing/inspecting a received SDP (a string), so there are two
> "solutions":
>
>
> 1) Manually parse de SDP at JS level, which is ugly taking into
> account that the browser includes a tested SDP parser.
>
> 2) Call to RTCPeerconnection.setRemoteDescription(RECEIVED_SDP), which
> means that the media stream begins.
>
>
> The point here is to know what the remote peer is offering to us
> (audio? audio+video?) before calling to getUserMedia(), so we (the web
> application) can tell the user about which kind of media session the
> remote peer is offering to him.
This relationship is of course highly application dependent.
It's also likely that you will want to call setRemoteDescription before 
calling getUserMedia, so that if setRemoteDescription fails, you won't 
bother to call getUserMedia at all. Again, app dependent.
>
>
> Do I miss something? As said above the specs don't seem to provide
> this funcionality at JS API.
They don't. This has been mostly deemed not strictly necessary for 
version 1, and experience will be gathered on what we really need to get 
out of the SDP by some real life experience.

>
>
> Thanks a lot.
>
>