rtcService.remoteVideoTrack in later SDK versions

S’abonner

Rick LUPPINO

• 15 avril 2021 06:57

Hi, 

I was previously using the remoteVideoStream method of RTCService to get/render the remote video stream of a video call. Since upgrading the SDK, I have replaced this with the remoteVideoTrack method but now the remote video track is just blank. The notification for the remote video track being added is called but there is no stream. This happens for both version 1.85.0 and 1.86.0 of the SDK.

Any ideas?

0

• 

  Rick LUPPINO

  • 31 mai 2022 04:38

  This is still happening in the latest LTS version of the SDK (2.4.0). I'm adding the remote video track to the view (RTCMTLVideoView) in the didAddRemoteVideoTrack event but the stream does not render. If I fall back to the previous method as described above, the remote video stream renders correctly

  0

  Actions pour les commentaires Permalien
• 

  Mahdi QWAREEQ

  • 1 juin 2022 12:09

  Hello Rick,

  Could you please share the code that you are using to render with remoteVideoStream, as it's expected to have the same result when use remoteVideoTrack.

  0

  Actions pour les commentaires Permalien
• 

  Rick LUPPINO

  • 1 juin 2022 23:40
  • Modifié

  This is the code that works:

  if let remoteVideoStreams = ServicesManager.sharedInstance()?.rtcService.remoteVideoStream(for: self.currentCall!) {
     if remoteVideoStreams.videoTracks.count > 0 {
        remoteVideoTrack = remoteVideoStreams.videoTracks[0]
     }
  }
  
  

  This is the code that presents a blank view/stream

  remoteVideoTrack = ServicesManager.sharedInstance()?.rtcService.remoteVideoTrack(for: self.currentCall!)

   

  0

  Actions pour les commentaires Permalien
• 

  Mahdi QWAREEQ

  • 2 juin 2022 05:34

  Thanks Rick, yes it's expected to get the same results for both codes.

  To compare the tracks values, could you please try to log the remoteVideoTrack in both cases, and also to log remoteVideoStreams.videoTracks to check the list of video tracks in the remoteVideoStreams.

  Also it will help if you can share the full logs when run the tests.

  0

  Actions pour les commentaires Permalien
• 

  Rick LUPPINO

  • 2 juin 2022 05:45

  It appears that remoteVideoTrack is indeed the same in both cases.

  VideoStream approach:

  

  VideoTrack approach:

  

   

  0

  Actions pour les commentaires Permalien
• 

  Mahdi QWAREEQ

  • 2 juin 2022 06:37

  Did you try both approaches in same test, since in the second approach I see that the remoteVideoTrack has different pointer value (0x0000000110dbec10)

  0

  Actions pour les commentaires Permalien
• 

  Rick LUPPINO

  • 2 juin 2022 06:56

  Sorry, I misunderstood. Here is the result when doing both approaches in the same test. remoteVideoTrack is from the remoteVideoStreamFor approach and remoteVideoTrack2 is from the remoteVideoTrackFor approach.

  

  0

  Actions pour les commentaires Permalien
• 

  Mahdi QWAREEQ

  • 2 juin 2022 07:28

  Could you please try to log the values using NSLog, and also if you could share the full logs from the Xcode console.

  0

  Actions pour les commentaires Permalien
• 

  Rick LUPPINO

  • 2 juin 2022 07:37

  The log is 100's of lines and will be too large to paste here. Is there a way to send it to you directly? via the Rainbow app maybe? Or is there a specific section you're looking for?

  NSLog here:

  2022-06-02 17:03:31.171534+0930 FitMed[42064:2955061] remoteVideoTrack: RTC_OBJC_TYPE(RTCMediaStreamTrack):

  video

  b7fe46a8-2952-4d99-aef9-a728d20ab629

  enabled

  Live

  2022-06-02 17:03:31.171984+0930 FitMed[42064:2955061] remoteVideoTrack2: RTC_OBJC_TYPE(RTCMediaStreamTrack):

  video

  b7fe46a8-2952-4d99-aef9-a728d20ab629

  enabled

  Live

   

  0

  Actions pour les commentaires Permalien
• 

  Mahdi QWAREEQ

  • 2 juin 2022 07:50

  It's okay, no need to share full logs for now, where from the tracks logs we see that both tracks look identical.

  When render the video track, can you try to remove the renderer first before adding the renderer to see if this could help to avoid the problem when use remoteVideoTrack.

  0

  Actions pour les commentaires Permalien
• 

  Rick LUPPINO

  • 2 juin 2022 08:08

  OK. I have tried that but it didn't solve the problem. I can continue to use the video stream approach for the time being.

  0

  Actions pour les commentaires Permalien

Vous devez vous connecter pour laisser un commentaire.