yt_dlp/extractor/umg.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     int_or_none,
   4     parse_filesize,
   5     parse_iso8601,
   6 )
   7
   8
   9 class UMGDeIE(InfoExtractor):
  10     _WORKING = False
  11     IE_NAME = 'umg:de'
  12     IE_DESC = 'Universal Music Deutschland'
  13     _VALID_URL = r'https?://(?:www\.)?universal-music\.de/[^/]+/videos/[^/?#]+-(?P<id>\d+)'
  14     _TEST = {
  15         'url': 'https://www.universal-music.de/sido/videos/jedes-wort-ist-gold-wert-457803',
  16         'md5': 'ebd90f48c80dcc82f77251eb1902634f',
  17         'info_dict': {
  18             'id': '457803',
  19             'ext': 'mp4',
  20             'title': 'Jedes Wort ist Gold wert',
  21             'timestamp': 1513591800,
  22             'upload_date': '20171218',
  23         },
  24     }
  25
  26     def _real_extract(self, url):
  27         video_id = self._match_id(url)
  28         video_data = self._download_json(
  29             'https://graphql.universal-music.de/',
  30             video_id, query={
  31                 'query': '''{
  32   universalMusic(channel:16) {
  33     video(id:%s) {
  34       headline
  35       formats {
  36         formatId
  37         url
  38         type
  39         width
  40         height
  41         mimeType
  42         fileSize
  43       }
  44       duration
  45       createdDate
  46     }
  47   }
  48 }''' % video_id})['data']['universalMusic']['video']  # noqa: UP031
  49
  50         title = video_data['headline']
  51         hls_url_template = 'http://mediadelivery.universal-music-services.de/vod/mp4:autofill/storage/' + '/'.join(list(video_id)) + '/content/%s/file/playlist.m3u8'
  52
  53         thumbnails = []
  54         formats = []
  55
  56         def add_m3u8_format(format_id):
  57             formats.extend(self._extract_m3u8_formats(
  58                 hls_url_template % format_id, video_id, 'mp4',
  59                 'm3u8_native', m3u8_id='hls', fatal=False))
  60
  61         for f in video_data.get('formats', []):
  62             f_url = f.get('url')
  63             mime_type = f.get('mimeType')
  64             if not f_url or mime_type == 'application/mxf':
  65                 continue
  66             fmt = {
  67                 'url': f_url,
  68                 'width': int_or_none(f.get('width')),
  69                 'height': int_or_none(f.get('height')),
  70                 'filesize': parse_filesize(f.get('fileSize')),
  71             }
  72             f_type = f.get('type')
  73             if f_type == 'Image':
  74                 thumbnails.append(fmt)
  75             elif f_type == 'Video':
  76                 format_id = f.get('formatId')
  77                 if format_id:
  78                     fmt['format_id'] = format_id
  79                     if mime_type == 'video/mp4':
  80                         add_m3u8_format(format_id)
  81                 urlh = self._request_webpage(f_url, video_id, fatal=False)
  82                 if urlh:
  83                     first_byte = urlh.read(1)
  84                     if first_byte not in (b'F', b'\x00'):
  85                         continue
  86                     formats.append(fmt)
  87         if not formats:
  88             for format_id in (867, 836, 940):
  89                 add_m3u8_format(format_id)
  90
  91         return {
  92             'id': video_id,
  93             'title': title,
  94             'duration': int_or_none(video_data.get('duration')),
  95             'timestamp': parse_iso8601(video_data.get('createdDate'), ' '),
  96             'thumbnails': thumbnails,
  97             'formats': formats,
  98         }