Recherche avancée

Sur d’autres sites (6118)

• Convert *.mov file to *.mp4 with media source support

  4 février 2017, par Yerzhan Torgayev

  My goal is to play converted *.mp4 file with HTML5 Media Source Extension (MSE).
  I made a small video with iMovie on my Mac and exported this video to *.mov file. Using ffmpeg I’ve converted *.mov file to fragmented *.mp4 file.
  I used code from Mozilla Media Source tutorial to make simple html5 video player.

  https://developer.mozilla.org/en-US/docs/Web/API/Media_Source_Extensions_API/Transcoding_assets_for_MSE

  But I couldn’t get my video played with this tutorial. Other *.mp4 videos play just fine. But I have troubles with converted *.mov videos.

  I’ve checked codec support for my video.
  On this page I’ve got my codec info :

  http://nickdesaulniers.github.io/mp4info/

  track #0 codec string: mp4v
  
  track #1 codec string: mp4a.40.2
  
  track #2 codec string: tmcd
  
  MediaSource.isTypeSupported('video/mp4; codecs="mp4v, mp4a.40.2, tmcd"'); // => false

  I found that tmcd codec is not supported by MSE :

  https://cconcolato.github.io/media-mime-support/

  I couldn’t find the way to get my video work with MSE. Can anyone help me with this issue ?

• dashenc : allow assigning all streams of a media type to an AdaptationSet

  29 janvier 2017, par Peter Große

  dashenc : allow assigning all streams of a media type to an AdaptationSet
  Using the characters "v" or "a" instead of stream index numbers for assigning
  
  streams in the adaption_set option, all streams matching that given type will
  
  be added to the AdaptationSet.
  Signed-off-by : Peter Große <pegro@friiks.de>
  
  Signed-off-by : Martin Storsjö <martin@martin.st>
  

  • [DBH] libavformat/dashenc.c

• Python - ffmpeg-like media seeking with requests

  8 janvier 2017, par Will

  Using ffmpeg, times in HTTP media resources, such as videos, can be seeked to very quickly with a command like ffmpeg -ss 00:01:00 -i https://example.com/video.mp4, presumably by downloading the video headers to look for the right byte offset. This means that it is very fast to run operations on a section of video, even if it is hours in.

  What I want to be able to do is to do the same thing but piping into ffmpeg with the requests module with something like (in partial pseudo-code) :

  stream = requests.get(url, stream=True)
  
  start_byte = get_byte_offset(stream, time=60)
  
  stream.seek_to(start_byte)
  
  
  
  process = subprocess.Popen(["ffmpeg", "-i" "pipe:" "out.mp4"], stdin=subprocess.PIPE)
  
  
  
  for chunk in stream.iter_content(chunk_size=64 * 2 ** 10):
  
      process.stdin.write(chunk)
  
  process.stdin.close()

  This would allow me to implement extra checks into the stream that I would normally need to parse ffmpeg for, such as if the stream returns a bad status code such as 404 or 403.

  How might I be able to implement this using the requests, and presumably other, modules ? I suppose the first part is using a tool to parse the video headers.