Recherche avancée

Sur d’autres sites (16462)

• KeyError : 1 on Python when the code points to a value that exists [closed]

  10 décembre 2020, par oo92

  I am using the following code to traverse a list JSON output. There are about 100 streamers I want to take screens from but the code fails after the first one where I call the ffmpeg. I put a code after the ffmpeg to see if Python is running it at all and it does :

  


  Traceback (most recent call last):&#xA;  File "/home/onur/Desktop/pythonProject/main.py", line 12, in <module>&#xA;    username = streams_now[i][&#x27;channel&#x27;][&#x27;display_name&#x27;]&#xA;KeyError: 1&#xA;auronplay it is &#xA;</module>

  


  This is my code :

  


  import streamlink, os
from twitch import TwitchClient

client = TwitchClient(client_id='frl3dqgn21bbpp6tajjvg5pdevczac')

streams_now = client.streams.get_live_streams(limit=100)

print(streams_now[2]['channel']['display_name'])

for i in range(0, 99):
  username = streams_now[i]['channel']['display_name']
  streams_now = streamlink.streams('http://twitch.tv/' + username)
  stream = streams_now["best"].url
  dir_path = os.getcwd() + '/' + username
  os.mkdir(dir_path)
  os.system('ffmpeg -i ' + stream + f' -ss 20 -vframes 10 -r 0.2 -f image2 {dir_path}/' + username + '_output_%09d.jpg')
  print(username + ' it is ')


  


  The line that throws the error is username = streams_now[i]['channel']['display_name']. However, this line cannot throw an error because I am testing it before the for loop :

  


  print(streams_now[2]['channel']['display_name'])

  


  and I get this : shroud

  


• Output a new mp3 file from a lengthy mp3 mix which previews the audio at 5 different points of the mix

  17 novembre 2020, par Paul Hinett

  I basically have a large collection of audio mix files in mp3 format, say approx 2 hours worth of music per mp3.

  


  I am trying to work out the best way using the command line to output a preview MP3 file, this mp3 should have 5-10 second previews of the full mix at 6 or so points during the mix.

  


  Example
00:00:10 - 00:00:20
00:30:00 - 00:30:10
00:50:00 - 00:50:10
...

  


  So with the small mp3 file, you can kind of get an idea of the whole mix in a very short preview.

  


  I have seen ffmpeg etc that can preview say 5 seconds up to 15 seconds of an audio, but not sure if this is capable of doing multiple points of the audio.

  


  Any ideas or help is appreciated.

  


• ffmpeg command line code translate to fluent-ffmpeg

  8 novembre 2020, par Berk Coşar

  I have trouble understanding fluent-ffmpeg.

  


  I'm trying to make a screen recorder with Fluent-ffmpeg but couldn't find the correct code. The output should consist of a screen, webcam and microphone.

  


  ffmpeg command line code works successfully.


  


  ffmpeg code :

  


  ffmpeg -ss 00:00:3 -thread_queue_size 4096 -f avfoundation -framerate 25 -i 1 -thread_queue_size 4096 -f avfoundation -framerate 25 -video_size 320x240 -i 0:0 -c:v libx264 -crf 18 -preset ultrafast -filter_complex 'overlay=main_w-overlay_w-10:main_h-overlay_h-10' -r 25 video.mp4


  


  my fluent-ffmpeg code :(ı know bad code) :

  


   command = ffmpeg("1")
        .videoCodec('libx264')
        .inputFPS(25)
        .size("1280x800")
        .inputFormat('avfoundation')  
         .input("0")
         .inputFormat('avfoundation')  
         .inputFPS(25)
          .size("320x240")
         .addOptions(["-filter_complex", "overlay=main_w-overlay_w-10:main_h-overlay_h-10 "])
        
     
        .addOption([  "-preset" , "ultrafast"  ])
       .save(`/Users/macbook/Pictures${new Date().toLocaleString().split('-').join('').split(':').join('')}.mp4`)


  


  i tried this but also this not worked :

  


       //.size("320x240")

   .complexFilter(['scale=320:240[rescaled]',
{
   filter: 'overlay', options: { x: 'main_w-overlay_w-10', y: 'main_h-overlay_h-10' },
   inputs: ['[0:v]','[1:v]'], outputs: '[a1][a2]'
 }
],'[a1][a2]')


  


  I would be very happy if you show me the right method. I will be happier if you also add a microphone.
thank you so much