Recherche avancée

Sur d’autres sites (7763)

• is it necessary in Anaconda for Matplotlib to install ffmpeg for animation saving ?

  2 mai 2018, par Vasyl Kolomiets

  For animation saving I have read some propositions
  1) manual installing ffmpeg this way :

    conda install -c conda-forge ffmpeg

  2) manual installing ImageMagick and changing matplotlib configuration file "matplotlibrc" this way :

  #animation.ffmpeg_args: ''         # Additional arguments to pass to ffmpeg
  
  #animation.avconv_path: 'avconv'   # Path to avconv binary. Without full path
  
                                     # $PATH is searched
  
  #animation.avconv_args: ''         # Additional arguments to pass to avconv
  
  #animation.mencoder_path: 'mencoder'
  
                                     # Path to mencoder binary. Without full path
  
                                     # $PATH is searched
  
  #animation.mencoder_args: ''       # Additional arguments to pass to mencoder
  
  #animation.convert_path: 'convert' # Path to ImageMagick's convert binary.
  
  animation.convert_path: F:\Games\ImageMagick-7.0.7-Q16\magick.exe
  
                                     # On Windows use the full path since convert
  
                                     # is also the name of a system tool.

  where line with pass to installed ImageMagick was added manually

  animation.convert_path: F:\Games\ImageMagick-7.0.7-Q16\magick.exe

  Two questions :
  1 - what way is better for Anaconda ?
  2 - really Anaconda has no better way ????

• How do I get the last line of a popen() callback in every iteration ?

  29 mai 2017, par Mike

  I’m trying to create a progress bar with FFMPEG using php and AJAX. When a user uploads a video file I want to be able to display the current percent until completion. I have managed to get everything working but I have one issue.

  The data returns what I want, but it also returns all the data from the previous iterations... like it just keeps stacking everything on top rather than flushing out the data from previous iterations. I tried to work with tail thinking it would return only the last line, but it did not return anything.

  Here is the code I’m working with :

  encode.php

  $video_path = 'test.mp4';
  
  $cmd        = 'ffmpeg -i ' . $video_path .' -y -hide_banner output.mp4 2>&1';
  
  
  
  while (@ ob_end_flush());
  
  
  
  $proc = popen($cmd, 'r');
  
  
  
  while (!feof($proc))
  
  {
  
  
  
      $file = escapeshellarg(fread($proc, 4096));
  
      //$line = `tail -n 1 $file`; // <-tried this with no luck
  
      echo fread($file, 4096) . "\n";
  
      @ flush();
  
  }
  
  return 'complete';
  
  pclose($proc);

  The above code returns :

  // first iteration
  
  frame=   52 fps= 13 q=29.0 size=     279kB time=00:00:00.10 bitrate=22856.9kbits/s
  
  
  
  // second iteration
  
  frame=   52 fps= 13 q=29.0 size=     279kB time=00:00:00.10 bitrate=22856.9kbits/s    
  
  frame=   54 fps= 12 q=29.0 size=     329kB time=00:00:00.16 bitrate=16146.6kbits/s 
  
  
  
  // third iteration
  
  frame=   52 fps= 13 q=29.0 size=     279kB time=00:00:00.10 bitrate=22856.9kbits/s    
  
  frame=   54 fps= 12 q=29.0 size=     329kB time=00:00:00.16 bitrate=16146.6kbits/s    
  
  frame=   57 fps= 11 q=29.0 size=     464kB time=00:00:00.26 bitrate=14233.2kbits/s 

  As you can see the data stacks, I need only the new line of data, not the data stacking.

  ** EDIT ** this has been marked as a duplicate, rather than me explain how it’s different, I’d like to hear how this is the same thing ? I am not writing to a log file and, like most people, do not feel that’s a good solution.

• Trying to fetch all audio streams with FFmpeg Python

  27 juillet 2022, par Apollo

  I'm using ffmpeg-python to fetch streams from a video and write some parameters (codec_name, resolution, etc.) for each stream into csv.

  


  video = 'test.mkv'
probe = ffmpeg.probe(video)
video_stream = next((stream for stream in probe['streams'] if stream['codec_type'] == 'video'), None)
print(video_stream['codec_long_name'])
audio_stream = next((stream for stream in probe['streams'] if stream['codec_type'] == 'audio'), None)
...


  


  My problem is that it works well for a video stream, but not for multiple audio (or subtitles) streams. If the video has several audio streams it returns only one audio stream.

  


  I've tried another approach, but it returns some streams 2-3 times and I get duplicates. So if the video sample has 4 audio tracks, I end up with 9 audio streams instread of 4.

  


  audio_streams = []
for audio in (probe['streams']):
    if (audio['codec_type'] == 'audio'):
        audio_streams.append(audio)
        pprint(audio_streams)


  


  All other ideas I tried don't work, I'm new to programming and I'm stuck with it.
How can I get all audio streams from a file without duplicates ?