Recherche avancée

Sur d’autres sites (4431)

• Can't get FFMPEG buildpack working in Heroku Node.js server (using Fluent-FFMPEG)

  30 juillet 2021, par Rayhan Memon

  I'm using a library called 'fluent-ffmpeg' on my Nodejs server that makes it easier to use an audio/video editing tool called FFmpeg, which is downloaded locally on my computer.

  


  When running on my computer, I point fluent-ffmpeg to the local executable files of FFmpeg and FFprobe on my computer, like so :

  


  import ffmpeg from "fluent-ffmpeg";

// When running locally, set FFmpeg and FFprobe path to the local executable files
ffmpeg.setFfmpegPath("C:/Program Files/FFMPEG/ffmpeg.exe");
ffmpeg.setFfprobePath("C:/Program Files/FFMPEG/ffprobe.exe");


  


  When deploying to Heroku, I must use an FFmpeg 'buildpack'.

  


  I've tried two :

  


  

  1. https://github.com/jonathanong/heroku-buildpack-ffmpeg-latest

  


  2. https://github.com/HasibulKabir/heroku-buildpack-ffmpeg-ffprobe

  


  


  Neither have worked for me. Whenever I use an FFmpeg command, I get the following error in my heroku logs :

  


  2021-07-24T15:22:52.970990+00:00 app[web.1]:   code: 'ENOENT',
2021-07-24T15:22:52.970990+00:00 app[web.1]:   syscall: 'spawn C:/Program Files/FFMPEG/',
2021-07-24T15:22:52.970990+00:00 app[web.1]:   path: 'C:/Program Files/FFMPEG/',
2021-07-24T15:22:52.970991+00:00 app[web.1]:   spawnargs: [
2021-07-24T15:22:52.970991+00:00 app[web.1]:     '-show_streams',
2021-07-24T15:22:52.970991+00:00 app[web.1]:     '-show_format',
2021-07-24T15:22:52.970992+00:00 app[web.1]:     'temp/cf3b5f1ae270df824921364573a4366b'
2021-07-24T15:22:52.970992+00:00 app[web.1]:   ]
2021-07-24T15:22:52.970992+00:00 app[web.1]: }


  


  How can I go about using fluent-ffmpeg in my Nodejs server when deploying to Heroku ?

  


  Thank you for the help in advance !

  


• RNFFMpeg.executeAsync : Unable to create gif of a video with space(s) in it's source path

  7 juillet 2021, par STBox

  Unable to make gif of a video which has space(s) in it's source directory. All the other video which don't have space(s), their gif is making fine. I had a video whose folder name had space and it was not making gif for it. I copied the same video to a folder without space in it's name, and it made gif for it successfully. Even if I replace the the space with source path with anything like %20 etc, then it says

  


  


  No such file or directory

  


  


  Code snippet

  


  


  


  RNFFmpeg.executeAsync(
          `-i ${source} -ss ${start} -t ${gifDuration}  -vf "fps=5,scale=160:-1:flags=lanczos,split[s0][s1];[s0]palettegen[p];[s1][p]paletteuse" -loop 0 ${destination}`, (completedExecution) => {
            if (completedExecution.returnCode === 0) {
              resolve({ path: destination });
            } else {
              resolve({ error: "invalid Gif" });  // always this part executes for videos with space in their source path
            }
          }
        );

  


  


  



  Expected behavior

  


  Regardless a video has or has not a space in it's path, it should make gif correctly and

  


  


  completedExecution.returnCode

  


  


  should be 0

  


  Current behavior
Currently, for all the videos having space in their source path, completedExecution.returnCode is always 1

  


  Logs of the same video without space in folder name

  


  


  [Wed Jul 07 2021 12:23:56.377] LOG file :///storage/emulated/0/AirDroid/VID-20210620-WA0024.mp4 [Wed Jul

  


  


  


  07 2021 12:23:56.511] LOG [Parsed_palettegen_3 @ 0x7d62eb5ec0] 255(+1) colors generated out of 249823 colors ; ratio=0.001021 [Wed Jul 07 2021 12:23:56.672] LOG video:48kB audio:0kB subtitle:0kB other streams:0kB

  


  


  


  global headers:0kB muxing overhead : [Wed Jul 07 2021 12:23:56.673] LOG

  


  


  


  0.041083% [Wed Jul 07 2021 12:23:56.722] LOG "path" : "/data/user/0/com.tbox.pantry/cache/1625642625930.gif"

  


  


  Logs of the same video with space in folder name

  


  


  [Wed Jul 07 2021 12:20:45.577] LOG

  


  


  


  file :///storage/emulated/0/FMWhatsApp/Media/FMWhatsApp

  


  


  


  Video/VID-20210620-WA0024.mp4 [Wed Jul 07 2021 12:20:45.998] LOG

  


  


  


  file :///storage/emulated/0/FMWhatsApp/Media/FMWhatsApp : No such file

  


  


  


  or directory [Wed Jul 07 2021 12:20:45.998] LOG Conversion failed !

  


  


  


  [Wed Jul 07 2021 12:20:45.999] LOG "error" : "invalid Gif"

  


  


  Environment

  


  "react" : "16.13.1",
"react-native" : "^0.63.3",
"react-native-ffmpeg" : "0.5.1",
Android 11

  


• Can't use OpenCV to decode video on GPU with cv2.CAP_FFMPEG

  19 juillet 2021, par Reza

  I'm trying to get all frames of list of videos(lots of .mp4 videos) using OpenCV. In order to speed up the process I was thinking may be I can use FFMPEG with Nvidia Video Codec SDK to use hardware-accelerated decoding(I have T4 cards on my machine). My code is something like this :

  


  import os
os.environ["OPENCV_FFMPEG_CAPTURE_OPTIONS"]="video_codec;h264_cuvid"
import cv2

ret = cv2.VideoCapture("input.mp4", cv2.CAP_FFMPEG)
while True:
    x, y = ret.read()
    if x:
        continue
    break


  


  but I get these errors when I run my script.

  


  [ERROR:0] global /tmp/pip-req-build-afu9cjzs/opencv/modules/videoio/src/cap_ffmpeg_impl.hpp (1086) open Could not find decoder 'h264_cuvid'
[ERROR:0] global /tmp/pip-req-build-afu9cjzs/opencv/modules/videoio/src/cap_ffmpeg_impl.hpp (1109) open VIDEOIO/FFMPEG: Failed to initialize VideoCapture


  


  My ffmpeg configs, ffmpeg :

  


  ffmpeg version N-102965-gf531a1a4e8 Copyright (c) 2000-2021 the FFmpeg developers
  built with gcc 9 (Ubuntu 9.3.0-17ubuntu1~20.04)
  configuration: --enable-nonfree --enable-cuda --enable-cuvid --enable-nvdec --enable-cuda-nvcc --enable-nvenc --enable-libnpp --enable-shared --extra-cflags=-I/usr/local/cuda/include --extra-ldflags=-L/usr/local/cuda/lib64
  libavutil      57.  0.100 / 57.  0.100
  libavcodec     59.  3.101 / 59.  3.101
  libavformat    59.  4.100 / 59.  4.100
  libavdevice    59.  0.100 / 59.  0.100
  libavfilter     8.  0.103 /  8.  0.103
  libswscale      6.  0.100 /  6.  0.100
  libswresample   4.  0.100 /  4.  0.100
Hyper fast Audio and Video encoder


  


  and ffmpeg -hwaccels :

  


  ffmpeg version N-102965-gf531a1a4e8 Copyright (c) 2000-2021 the FFmpeg developers
  built with gcc 9 (Ubuntu 9.3.0-17ubuntu1~20.04)
  configuration: --enable-nonfree --enable-cuda --enable-cuvid --enable-nvdec --enable-cuda-nvcc --enable-nvenc --enable-libnpp --enable-shared --extra-cflags=-I/usr/local/cuda/include --extra-ldflags=-L/usr/local/cuda/lib64
  libavutil      57.  0.100 / 57.  0.100
  libavcodec     59.  3.101 / 59.  3.101
  libavformat    59.  4.100 / 59.  4.100
  libavdevice    59.  0.100 / 59.  0.100
  libavfilter     8.  0.103 /  8.  0.103
  libswscale      6.  0.100 /  6.  0.100
  libswresample   4.  0.100 /  4.  0.100
Hardware acceleration methods:
cuda


  


  OpenCV version opencv-python==4.5.3.56