As you noted, it is possible, in principle, for a RTSP client to specify - to the server - that the server send the stream to a different destination. In practice, however, few servers implement this, because it would allow the client to launch a ‘denial of service’ attack by telling the server to stream to an unsuspecting client. (Our RTSP server implementation does support this, but in code that is #ifdef’d out by default. And our RTSP client implementation doesn’t support it at all.) So this approach is pretty much a non-starter.
So, the server will have to send the stream to the client machine. But what you can do is modify your RTSP client application so that it doesn’t actually receive the stream itself. Then, you can write your own, separate relay application - running on the client machine - that receives the stream, and relays it to whatever final destination you want. For an illustration of how to modify your RTSP client application so that it doesn't receive the stream, look at how our “openRTSP” client application implements the “-r” option. (See http://live555.com/openRTSP/#no-receive and “testProgs/playCommon.cpp”, line 188.) Ross Finlayson Live Networks, Inc. http://www.live555.com/ _______________________________________________ live-devel mailing list live-devel@lists.live555.com http://lists.live555.com/mailman/listinfo/live-devel
