<p>My main problem is understanding of code flow. <br></p><p>Just want to make
your explanations more clear.</p><p> Am I on the right way?
</p><p> </p><p>> - Use the client code to open the RTSP
stream</p><div>Usage for RTSPClient
here?</div><div>RTSPClient.createNew(...).... (connect to camera's sdp
url)<br></div><div>RTSPClient.describeURL(...)...
</div><div>MediaSession.createNew(sdpDesciption)</div><div>Subsession iteration
logic .... <br></div><p> </p><p> <br></p><div>
> Create each RTPSink</div><div> Creation of RTPSinks(for audio and
video, as I understand)</div><div> How can I create RTPSinks? From existing
MediaSubsession? <br></div><div> </div><div>> For each RTPSink that you
just created, call<br>> sink->startPlaying(rtpSource,
...);<br>
> Call "doEventLoop()" to enter the event loop.</div><div>What is the way to
create RTPSource. </div><div>I have a gap in my understanding of the code
between the RTSPClinet, MedisSession/Subsession,</div><div>and your suggesions
to use Sink/Source objects.</div><div>I didn't find any samples in a testXXX
code describing such flows.<br></div><div> </div><div>If I understand you
right, I should create two passive media subsession in server object, and during
creation of passive subsessions,</div><div>use FramedSource object, that will be
received from RTPSink/RTPSouce/smth part of code.</div><div>My problem is
understanding of this part of code.
</div><div> </div><div>Thanks,</div><div> Vadim</div>