yt_dlp/extractor/cloudflarestream.py

   1 import base64
   2
   3 from .common import InfoExtractor
   4
   5
   6 class CloudflareStreamIE(InfoExtractor):
   7     _DOMAIN_RE = r'(?:cloudflarestream\.com|(?:videodelivery|bytehighway)\.net)'
   8     _EMBED_RE = r'embed\.%s/embed/[^/]+\.js\?.*?\bvideo=' % _DOMAIN_RE
   9     _ID_RE = r'[\da-f]{32}|[\w-]+\.[\w-]+\.[\w-]+'
  10     _VALID_URL = r'''(?x)
  11                     https?://
  12                         (?:
  13                             (?:watch\.)?%s/|
  14                             %s
  15                         )
  16                         (?P<id>%s)
  17                     ''' % (_DOMAIN_RE, _EMBED_RE, _ID_RE)
  18     _EMBED_REGEX = [fr'<script[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//{_EMBED_RE}(?:{_ID_RE}).*?)\1']
  19     _TESTS = [{
  20         'url': 'https://embed.cloudflarestream.com/embed/we4g.fla9.latest.js?video=31c9291ab41fac05471db4e73aa11717',
  21         'info_dict': {
  22             'id': '31c9291ab41fac05471db4e73aa11717',
  23             'ext': 'mp4',
  24             'title': '31c9291ab41fac05471db4e73aa11717',
  25         },
  26         'params': {
  27             'skip_download': True,
  28         },
  29     }, {
  30         'url': 'https://watch.cloudflarestream.com/9df17203414fd1db3e3ed74abbe936c1',
  31         'only_matching': True,
  32     }, {
  33         'url': 'https://cloudflarestream.com/31c9291ab41fac05471db4e73aa11717/manifest/video.mpd',
  34         'only_matching': True,
  35     }, {
  36         'url': 'https://embed.videodelivery.net/embed/r4xu.fla9.latest.js?video=81d80727f3022488598f68d323c1ad5e',
  37         'only_matching': True,
  38     }]
  39
  40     def _real_extract(self, url):
  41         video_id = self._match_id(url)
  42         domain = 'bytehighway.net' if 'bytehighway.net/' in url else 'videodelivery.net'
  43         base_url = 'https://%s/%s/' % (domain, video_id)
  44         if '.' in video_id:
  45             video_id = self._parse_json(base64.urlsafe_b64decode(
  46                 video_id.split('.')[1] + '==='), video_id)['sub']
  47         manifest_base_url = base_url + 'manifest/video.'
  48
  49         formats = self._extract_m3u8_formats(
  50             manifest_base_url + 'm3u8', video_id, 'mp4',
  51             'm3u8_native', m3u8_id='hls', fatal=False)
  52         formats.extend(self._extract_mpd_formats(
  53             manifest_base_url + 'mpd', video_id, mpd_id='dash', fatal=False))
  54
  55         return {
  56             'id': video_id,
  57             'title': video_id,
  58             'thumbnail': base_url + 'thumbnails/thumbnail.jpg',
  59             'formats': formats,
  60         }