Recherche avancée

Sur d’autres sites (5585)

• Question about coding design of interface and receivers

  14 décembre 2018, par the_real_one

  I am trying to create an interface for an FFmpeg conversion scheme where :

  • New() should return whatever implements it
  • (receiver *interface) Convert() is the only method that should be implemented and the underlying details hidden (source files from : URL, File, etc.)

  How would I accomplish the above problem and design the code in such a way that you have various New() methods returning an FFmpegConverer that enforces the implementation of the Convert() method but with the underlying details behind each New() being different ? Or is that just a bad design to begin with ?

  My code may be less relevant than my problem statement but this is what I have so far which does not pass any go sanity checks :

  // FfmpegConverter is the interface for setting the public facing local
  
  // video files
  
  type FfmpegConverter interface {
  
    Convert() error
  
  }
  
  
  
  // localVideo holds the file information for a downloaded video
  
  type localVideo struct {
  
    inputAudioFile * os.File
  
    inputVideoFile * os.File
  
    outputVideoFile * os.File
  
  }
  
  
  
  // New initializes a new FfmpegConverter
  
  func New(inputVideoFilePath string, inputAudioFilePath string, outputVideoFilePath string)(ffmpegConverter FfmpegConverter, err error) {
  
    ffmpegConverter = & localVideo {}
  
  
  
    // Set input video file
  
    ffmpegConverter.inputVideoFile, err = os.Open(inputVideoFilePath)
  
    if err != nil {
  
      return
  
    }
  
    if err = ffmpegConverter.inputVideoFile.Close();
  
    err != nil {
  
      return
  
    }
  
  
  
    // Set input audio file
  
    ffmpegConverter.inputAudioFile, err = os.Open(inputAudioFilePath)
  
    if err != nil {
  
      return
  
    }
  
    if err = ffmpegConverter.inputAudioFile.Close();
  
    err != nil {
  
      return
  
    }
  
  
  
    // Set output video file
  
    ffmpegConverter.outputVideoFile, err = os.Create(outputVideoFilePath)
  
    if err != nil {
  
      return
  
    }
  
    if err = ffmpegConverter.outputVideoFile.Close();
  
    err != nil {
  
      return
  
    }
  
  
  
    return
  
  }
  
  
  
  // Convert will convert the input video and input audio files to an combined
  
  // output video file
  
  func(l * localVideo) Convert()(err error) {
  
    // Do stuff with the internal struct
  
    // ...
  
  }

• ffmpeg strange errors variables/undefined constants & question fonts

  21 mars 2019, par Rick

  I was trying to achieve get an animated text overlay. I found this thread :
  ffmpeg moving text drawtext

  When I run this code however :

  ffmpeg -i VideoInput.mp4 \
  
         -vf "drawtext=enable='between(t,12,15)':fontfile=myfont.otf:text='Test test': \
  
             x='if(lt(t-t2+1\,0)\,x1+(x2-x1)*(t-t1)/(t2-t1-1)\,x)': \
  
             y='if(lt(t-t2+1\,0)\,y1+(y2-y1)*(t-t1)/(t2-t1-1)\,y)':fontsize=65" \
  
         -acodec copy outputVideo.mp4

  I get the error :

  [Parsed_drawtext_0 @ 000002d862595c00][Eval @ 000000e1d79fe300] Undefined constant or missing '(' in 't2-t1-1),x)' 
  
  [Parsed_drawtext_0 @ 000002d862595c00] Failed to configure input pad on Parsed_drawtext_0

  A second strange error, which may be the cause for the problem at the top. Is that I can’t run the arguments on multiple lines. So with the code below I have to remove the slashes and put it on 1 line. I use PHP so this does not work (white screen, no output)

  $arguments = "-i out1.mp4 \
  
         -vf \"drawtext=enable='between(t,12,15)':fontfile=myfont.otf:text='Test test': \
  
             x='if(lt(t-t2+1\,0)\,x1+(x2-x1)*(t-t1)/(t2-t1-1)\,x)': \
  
             y='if(lt(t-t2+1\,0)\,y1+(y2-y1)*(t-t1)/(t2-t1-1)\,y)':fontsize=65\" \
  
         -acodec copy outputVideo.mp4 2>&1";
  
  $output = shell_exec('D:/wamp64/www/createslideshow/ffmpeg-20190319-f8075b2-win64-static/bin/ffmpeg.exe '.$arguments);

  This works but gives the undefined constant error :

  $arguments = "-i out1.mp4 -vf \"drawtext=enable='between(t,12,15)':fontfile=myfont.otf:text='Test test': x='if(lt(t-t2+1\,0)\,x1+(x2-x1)*(t-t1)/(t2-t1-1)\,x)': y='if(lt(t-t2+1\,0)\,y1+(y2-y1)*(t-t1)/(t2-t1-1)\,y)':fontsize=65\" -acodec copy outputVideo.mp4 2>&1";
  
  $output = shell_exec('D:/wamp64/www/createslideshow/ffmpeg-20190319-f8075b2-win64-static/bin/ffmpeg.exe '.$arguments);

  Last thing is that I keep getting the font error :
  Fontconfig error : Cannot load default config file
  [Parsed_drawtext_0 @ 000001d6ad86f180] Using "C :/WINDOWS/fonts/mingliub.ttc"

  I have tried :

  arial.ttf
  
  /windows/fonts/arial.ttf
  
  \windows\fonts\arial.ttf
  
  \\\windows\\\fonts\\\arial.ttf
  
  c\\:\\\windows\\\fonts\\\arial.ttf

• Which Platform/Language to choose for Video Management System [closed]

  5 décembre 2019, par Vivek Buddhadev

  We have started on a new project, Which enables for the multiple IP cameras(Around 1000 Cameras) to record as well as available for live streaming.

  Like https://www.security.honeywell.com/uk/product-repository/maxpro-vms

  Initially, we have tried using ffmpeg commands that saves data in chunks :

  ffmpeg -i {rtsp_url} -fflags flush_packets -max_delay 2 -flags -global_header \
  
      -hls_time 2 -hls_list_size 3 -vcodec copy -y /var/www/html/Cam01.m3u8

  Then we have started http server and using node.js we are displaying the stream.

  But as we are adding more cameras into it, the performance is going down.

  Kindly suggest if there is any other way to achieve this functionality which help us to fulfill the above requirement.