Sur d’autres sites (7895)

• How to create thumbnail about torrent without downloading video files ? [on hold]

  1er septembre 2019, par tty78

  Most of time, torrent sites like RARBG/The Pirate Bay/KickAss would return many query results even through I have typed in many detailed keywords, which is not strange because there are many people upload it.

  So, when i want to watch/download a torrent movie, it’s better to check the video first to find out that :

  • is it a fake video
  • is the video watermarked
  • is there some AD in the video
  • is it 1980*1080 or 640*480 ??

  What i want to do is creating thumbnail of the videos without download them, so that i can save a lot of broadband, disk space and time.

  Now I can create thumbnail with a python tool - vcsi, but it depends on the video file stored in disk.
  In fact, vcsi get meta information with ffprobe and create thumbnail using ffmpeg.

  def probe_media(self, path):
  
      """Probe video file using ffprobe
  
      """
  
      ffprobe_command = [
  
          "ffprobe",
  
          "-v", "quiet",
  
          "-print_format", "json",
  
          "-show_format",
  
          "-show_streams",
  
          path
  
      ]
  
  
  
      try:
  
          output = subprocess.check_output(ffprobe_command)
  
          self.ffprobe_dict = json.loads(output.decode("utf-8"))
  
      except FileNotFoundError:
  
          error = "Could not find 'ffprobe' executable. Please make sure ffmpeg/ffprobe is installed and is in your PATH."
  
          error_exit(error)

  ffmpeg_command = [
  
          "ffmpeg",
  
          "-ss", skip_time,
  
          "-i", self.path,
  
          "-ss", skip_delay,
  
          "-vframes", "1",
  
          "-s", "%sx%s" % (width, height),
  
      ]

  WebTorrent may be a solution but it’s not good because I have a lot of torrent file to scan.
  I want to create thumbnail on a ubuntu VPS because it can run day and night, and i can post thumbnail under the torrent comments.
  Some key frame, resolution and duration are enough, maybe.
  

• How to run FFMPEG with —enable-libfontconfig on Amazon Lambda

  20 avril 2024, par Adrien Kaczmarek

  Problem

  


  I want to run FFmpeg on AWS Lambda (Amazon Linux 2) with the configuration --enable-libfontconfig enable.

  


  Situation

  


  I already have FFmpeg running on AWS Lambda without the configuration --enable-libfontconfig.

  


  Here is the step I took to run FFmpeg on AWS Lambda (see official guide) :

  


  

  • Connect to Amazon EC2 running on AL2 (environment used by Lambda for Python 3.11)

  


  • Download and package FFmpeg from John Van Sickle

  


  • Create a Lambda Layer with FFmpeg

  


  


  Unfortunately, the version built by John Van Sickle doesn't have the configuration --enable-libfontconfig enabled.

  


  Unsuccessful Trials

  


  I tried to rebuilt it from scratch following the installation guide but without success (and the guide doesn't install font related dependencies)

  


  I tried to install it with brew but the command brew install ffmpeg didn't succeed on AL2.

  


  I tried to install ffmpeg from ffmpeg-master-latest-linux64-gpl.tar.xz. Unfortunately, this build of ffmpeg doesn't run on AL2 :

  


  ffmpeg: /lib64/libm.so.6: version `GLIBC_2.27' not found (required by ffmpeg)
ffmpeg: /lib64/libpthread.so.0: version `GLIBC_2.28' not found (required by ffmpeg)
ffmpeg: /lib64/libc.so.6: version `GLIBC_2.27' not found (required by ffmpeg)
ffmpeg: /lib64/libc.so.6: version `GLIBC_2.28' not found (required by ffmpeg)


  


  Any help would be greatly appreciated,

  


  Please make sure your answer is up to date and tested. Too many answers out there are auto-generated, too generic, or simple redirect without context.

  


  Thank you

  


• Upload a LIVE HLS output from FFMPEG to Amazon S3 bucket

  22 janvier 2023, par Pit Digger

  I have an ffmpeg command that outputs live hls stream from a vod file to local disk. I would like to upload the HLS chunks and manifest file in real time to Amazon S3 Bucket. Is it possible to upload it directly from memory ?

  


  Here is the ffmpeg command

  


  ffmpeg -stream_loop -1 -re -i bunny_1080p_30fps.mp4 -c:v libx264 \
-x264opts keyint=30:min-keyint=30:scenecut=-1 -tune zerolatency -s 1280x720 \
-b:v 1400k -bufsize 1400k -hls_start_number_source epoch -f hls stream.m3u8


  


  Note that I have checked all questions here but none of them do that for a live stream and for multiple files output like hls or dash stream.

  


  One option that I have tried is using node-watch module and upload file on change notification, but that does introduce some buffering. But if there was any better option. https://github.com/yuanchuan/node-watch