Recherche avancée

Sur d’autres sites (9396)

• Is async.js needed to process multiple ffmpeg conversions at the same time ?

  15 février 2019, par jurelik

  I’m trying to convert youtube videos to mp3 via my Node.js server, using ’ytdl-core’ and ’fluent-ffmpeg’. Since the server is intended to process multiple requests at the same time, it got me thinking whether or not async.js is needed to convert videos in a time efficient manner.

  The interesting thing however, is that upon testing the handling of multiple requests with and without using async.js, the result seems to be the same both ways - the time it takes to convert 3 videos is the same.

  Here is the code I’m using without async.js :

  server.get('/download/:id', (req, res) => {
  
  
  
    const id = req.params.id;
  
    let stream = ytdl(`https://www.youtube.com/watch?v=${id}`);
  
  
  
    ffmpeg(stream)
  
      .audioCodec('libmp3lame')
  
      .audioBitrate(128)
  
      .toFormat('mp3')
  
      .save(`public/downloads/${id}.mp3`)
  
      .on('error', err => {
  
        console.log(err);
  
      })
  
      .on('end', () => {
  
        console.log('file downloaded');
  
        send(req, `public/downloads/${id}.mp3`).pipe(res);
  
      });
  
  });

  And this is the code using async.js :

  let queue = async.queue((task, callback) => {
  
    let stream = ytdl(`https://www.youtube.com/watch?v=${task.id}`);
  
  
  
    ffmpeg(stream)
  
    .audioCodec('libmp3lame')
  
    .audioBitrate(128)
  
    .toFormat('mp3')
  
    .save(`public/downloads/${task.id}.mp3`)
  
    .on('error', err => {
  
      console.log(err);
  
      callback(err)
  
    })
  
    .on('end', () => {
  
      send(task.req, `public/downloads/${task.id}.mp3`).pipe(task.res);
  
      callback('file sucessfully downloaded');
  
    });
  
  }, 5);
  
  
  
  queue.drain = function() {
  
    console.log('all items downloaded');
  
  }
  
  
  
  server.get('/download/:id', (req, res) => {
  
    queue.push({req: req, id: req.params.id, res: res}, err => {
  
      console.log(err);
  
    });
  
  });

  Does anyone have any ideas why both methods seem to finish conversion at roughly the same time ? I would imagine using async.js should finish converting the videos faster due to processing in parallel, but that isn’t the case.

  Any thoughts would be much appreciated !

• avcodec/cfhd : Move peak_table() and difference_coding() calls after the existing...

  13 août 2018, par Michael Niedermayer

  avcodec/cfhd : Move peak_table() and difference_coding() calls after the existing coefficient count check
  Fixes : out of array access
  
  Fixes : 9509/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_CFHD_fuzzer-5283250636324864
  
  Fixes : 9572/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_CFHD_fuzzer-4920757409808384
  
  Fixes : 9596/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_CFHD_fuzzer-5114917580439552
  
  Fixes : 9640/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_CFHD_fuzzer-6247840698335232
  
  Fixes : 9659/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_CFHD_fuzzer-6079554987753472
  Found-by : continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
  
  Signed-off-by : Michael Niedermayer <michael@niedermayer.cc>
  

  • [DH] libavcodec/cfhd.c

• Why doesn't extracting one second from a video at least result in a multiple of the frame time ?

  21 avril 2018, par joeycato

  Trying to get a better understanding on why I’m unable to extract a segment from a movie clip with exact timing, even if I account for clip’s frame rate.

  For example, if I run the following command on a 29.97 fps video clip in an attempt to extract 1 second of video :

  ffmpeg -i input.avi -ss 0 -t 1 -c:v huffyuv -an output.avi

  the duration results in 1.001 seconds.

  Now this would make more sense to me if the result was actually a multiple of 0.02997 seconds (29.97 fps / 1000 ms) but doing the math to check the multiples (0.02997, 0.05994, ..., 0.98901, 1.01898, 1.04895), I see that the closest multiple is actually be 1.01898 or 1.04895

  What am I missing here ?