Re: [Live-devel] Possible missing parameter in MPEG4ESVideoRTPSink::auxSDPLine()

2012-03-29 Thread Ross Finlayson
No, you don't need to subclass "RTSPServer" to do what you think you need to do. You already implement the "createNewRTPSink()" virtual function, in which you create a "RTPSink" (subclass) object, given a "Groupsock" object. If you need to do something with the "Groupsock" object (or its socke

[Live-devel] Possible missing parameter in MPEG4ESVideoRTPSink::auxSDPLine()

2012-03-29 Thread Shaheed Abdol
Thank you Ross, Your information was spot-on, and I have discovered that I do not need to subclass the MPEG4ESVideoRTPSink class since the fSource parameter gets set when the subsession starts playing. I have come across an interesting limitation though, it turns out that I cannot send out data d

Re: [Live-devel] Possible missing parameter in MPEG4ESVideoRTPSink::auxSDPLine()

2012-03-28 Thread Ross Finlayson
The problem with MPEG-4 (and H.264) video is that - to properly serve the stream - you need (in the stream's SDP description) certain 'configuration' parameters. Often, these configuration parameters can be obtained only by reading the stream itself. In this case, we have a 'chicken and egg' p

[Live-devel] Possible missing parameter in MPEG4ESVideoRTPSink::auxSDPLine()

2012-03-28 Thread Shaheed Abdol
Good afternoon, I have hit an interesting obstacle while streaming RTSP video data from a live source. I am using the openRTSP client to retreive data from a testOnDemandRTSPServer, which I have added a new OnDemandServerMediaSubsession type to, named OnDemandVideoInputSubsession. I am streaming