Live555 流程上的一些分析

看了好多关于live555 的文章，其他部分不说了，关于 读取文件 的服务类 ByteStreamFileSource

这个类是用来从视频读取帧数据的，其中实现了doReadFromFile(..)函数来对文件进行读取。实现了doGetNextFrame等函数。

关于ByteStreamFileSource类的建立，有2处。

一次是在处理DESCRIBE请求handleCmd_DESCRIBE(...)时候 sdpLines(...)临时建立

用来获取sdp信息，创建适用完成之后就被close掉了；

另外一次是在处理SETUP命令的时候，调用了 subsession->getStreamParameters(...);

如果你的subsession 是从OnDemandServerMediaSubsession继承或者是该类的后代，那么这个getStreamParameters(..)原本是

ServerMediaSubsession的纯虚函数，至于具体实现在子类完成，真的执行到子类的getStreamParameters(..)时候是在

OnDemandServerMediaSubsession类里面实现的，在这个类里面调用了函数：

createNewStreamSource(...) 这个函数一时在ServerMediaSubsession中定义了的虚函数，那么这个函数具体的执行就要看你

的继承的子类如何实现的了，比如你建立MPEG4对应的subsession， 那么就会执行这个类对应的函数：

FramedSource* MPEG4VideoFileServerMediaSubsession
::createNewStreamSource(unsigned /*clientSessionId*/, unsigned& estBitrate)

在来看看这个函数的实现：

FramedSource* MPEG4VideoFileServerMediaSubsession
::createNewStreamSource(unsigned /*clientSessionId*/, unsigned& estBitrate)
{
  estBitrate = 500; // kbps, estimate

  // Create the video source:
  ByteStreamFileSource* fileSource
    = ByteStreamFileSource::createNew(envir(), fFileName);
  if (fileSource == NULL) return NULL;
  fFileSize = fileSource->fileSize();

  // Create a framer for the Video Elementary Stream:
  return MPEG4VideoStreamFramer::createNew(envir(), fileSource);
}

创建了一个ByteStreamFileSource类的对象fileSource。

前面说的，这个createNewStreamSource是在getStreamParameters(..)中建立的，然后在看看getStreamParameters这个函数把。

void OnDemandServerMediaSubsession
::getStreamParameters(unsigned clientSessionId,
		      netAddressBits clientAddress,
		      Port const& clientRTPPort,
		      Port const& clientRTCPPort,
		      int tcpSocketNum,
		      unsigned char rtpChannelId,
		      unsigned char rtcpChannelId,
		      netAddressBits& destinationAddress,
		      u_int8_t& /*destinationTTL*/,
		      Boolean& isMulticast,
		      Port& serverRTPPort,
		      Port& serverRTCPPort,
		      void*& streamToken) {
  if (destinationAddress == 0) destinationAddress = clientAddress;
  struct in_addr destinationAddr; destinationAddr.s_addr = destinationAddress;
  isMulticast = False;

  if (fLastStreamToken != NULL && fReuseFirstSource) {
    // Special case: Rather than creating a new 'StreamState',
    // we reuse the one that we've already created:
    serverRTPPort = ((StreamState*)fLastStreamToken)->serverRTPPort();
    serverRTCPPort = ((StreamState*)fLastStreamToken)->serverRTCPPort();
    ++((StreamState*)fLastStreamToken)->referenceCount();
    streamToken = fLastStreamToken;
  } else {
    // Normal case: Create a new media source:
    unsigned streamBitrate;
    FramedSource* mediaSource
      = createNewStreamSource(clientSessionId, streamBitrate);

    // Create 'groupsock' and 'sink' objects for the destination,
    // using previously unused server port numbers:
    RTPSink* rtpSink;
    BasicUDPSink* udpSink;
    Groupsock* rtpGroupsock;
    Groupsock* rtcpGroupsock;
    portNumBits serverPortNum;
    if (clientRTCPPort.num() == 0) {
      // We're streaming raw UDP (not RTP). Create a single groupsock:
      NoReuse dummy(envir()); // ensures that we skip over ports that are already in use
      for (serverPortNum = fInitialPortNum; ; ++serverPortNum) {
	struct in_addr dummyAddr; dummyAddr.s_addr = 0;

	serverRTPPort = serverPortNum;
	rtpGroupsock = new Groupsock(envir(), dummyAddr, serverRTPPort, 255);
	if (rtpGroupsock->socketNum() >= 0) break; // success
      }

      rtcpGroupsock = NULL;
      rtpSink = NULL;
      udpSink = BasicUDPSink::createNew(envir(), rtpGroupsock);
    } else {
      // Normal case: We're streaming RTP (over UDP or TCP).  Create a pair of
      // groupsocks (RTP and RTCP), with adjacent port numbers (RTP port number even):
      NoReuse dummy(envir()); // ensures that we skip over ports that are already in use
      for (portNumBits serverPortNum = fInitialPortNum; ; serverPortNum += 2) {
	struct in_addr dummyAddr; dummyAddr.s_addr = 0;

	serverRTPPort = serverPortNum;
	rtpGroupsock = new Groupsock(envir(), dummyAddr, serverRTPPort, 255);
	if (rtpGroupsock->socketNum() < 0) {
	  delete rtpGroupsock;
	  continue; // try again
	}

	serverRTCPPort = serverPortNum+1;
	rtcpGroupsock = new Groupsock(envir(), dummyAddr, serverRTCPPort, 255);
	if (rtcpGroupsock->socketNum() < 0) {
	  delete rtpGroupsock;
	  delete rtcpGroupsock;
	  continue; // try again
	}

	break; // success
      }

      unsigned char rtpPayloadType = 96 + trackNumber()-1; // if dynamic
      rtpSink = createNewRTPSink(rtpGroupsock, rtpPayloadType, mediaSource);
      udpSink = NULL;
    }

    // Turn off the destinations for each groupsock.  They'll get set later
    // (unless TCP is used instead):
    if (rtpGroupsock != NULL) rtpGroupsock->removeAllDestinations();
    if (rtcpGroupsock != NULL) rtcpGroupsock->removeAllDestinations();

    if (rtpGroupsock != NULL) {
      // Try to use a big send buffer for RTP -  at least 0.1 second of
      // specified bandwidth and at least 50 KB
      unsigned rtpBufSize = streamBitrate * 25 / 2; // 1 kbps * 0.1 s = 12.5 bytes
      if (rtpBufSize < 50 * 1024) rtpBufSize = 50 * 1024;
      increaseSendBufferTo(envir(), rtpGroupsock->socketNum(), rtpBufSize);
    }

    // Set up the state of the stream.  The stream will get started later:
    streamToken = fLastStreamToken
      = new StreamState(*this, serverRTPPort, serverRTCPPort, rtpSink, udpSink,
			streamBitrate, mediaSource,
			rtpGroupsock, rtcpGroupsock);
  }

  // Record these destinations as being for this client session id:
  Destinations* destinations;
  if (tcpSocketNum < 0) { // UDP
    destinations = new Destinations(destinationAddr, clientRTPPort, clientRTCPPort);
  } else { // TCP
    destinations = new Destinations(tcpSocketNum, rtpChannelId, rtcpChannelId);
  }
  fDestinationsHashTable->Add((char const*)clientSessionId, destinations);
}

在getStreamParameters(...)这个函数中做了 source 和 sink 的关联，后面读取数据，和打包数据的对象都是在这里建立的。

PS: 以上为个人分析鄙见，不对大家指正，我基础也是渣渣级别。

附上一张分析live555 流程的调用过程图片，分辨率略低。