_ Git - youtube-dl/blob - youtube_dl/extractor/cloudflarestream.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import base64
   5 import re
   6
   7 from .common import InfoExtractor
   8
   9
  10 class CloudflareStreamIE(InfoExtractor):
  11     _DOMAIN_RE = r'(?:cloudflarestream\.com|(?:videodelivery|bytehighway)\.net)'
  12     _VALID_URL = r'''(?x)
  13                     https?://
  14                         (?:
  15                             (?:watch\.)?%s/|
  16                             embed\.%s/embed/[^/]+\.js\?.*?\bvideo=
  17                         )
  18                         (?P<id>[\da-f]{32}|[\w-]+\.[\w-]+\.[\w-]+)
  19                     ''' % (_DOMAIN_RE, _DOMAIN_RE)
  20     _TESTS = [{
  21         'url': 'https://embed.cloudflarestream.com/embed/we4g.fla9.latest.js?video=31c9291ab41fac05471db4e73aa11717',
  22         'info_dict': {
  23             'id': '31c9291ab41fac05471db4e73aa11717',
  24             'ext': 'mp4',
  25             'title': '31c9291ab41fac05471db4e73aa11717',
  26         },
  27         'params': {
  28             'skip_download': True,
  29         },
  30     }, {
  31         'url': 'https://watch.cloudflarestream.com/9df17203414fd1db3e3ed74abbe936c1',
  32         'only_matching': True,
  33     }, {
  34         'url': 'https://cloudflarestream.com/31c9291ab41fac05471db4e73aa11717/manifest/video.mpd',
  35         'only_matching': True,
  36     }, {
  37         'url': 'https://embed.videodelivery.net/embed/r4xu.fla9.latest.js?video=81d80727f3022488598f68d323c1ad5e',
  38         'only_matching': True,
  39     }]
  40
  41     @staticmethod
  42     def _extract_urls(webpage):
  43         return [
  44             mobj.group('url')
  45             for mobj in re.finditer(
  46                 r'<script[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//embed\.(?:cloudflarestream\.com|videodelivery\.net)/embed/[^/]+\.js\?.*?\bvideo=[\da-f]+?.*?)\1',
  47                 webpage)]
  48
  49     def _real_extract(self, url):
  50         video_id = self._match_id(url)
  51         domain = 'bytehighway.net' if 'bytehighway.net/' in url else 'videodelivery.net'
  52         base_url = 'https://%s/%s/' % (domain, video_id)
  53         if '.' in video_id:
  54             video_id = self._parse_json(base64.urlsafe_b64decode(
  55                 video_id.split('.')[1]), video_id)['sub']
  56         manifest_base_url = base_url + 'manifest/video.'
  57
  58         formats = self._extract_m3u8_formats(
  59             manifest_base_url + 'm3u8', video_id, 'mp4',
  60             'm3u8_native', m3u8_id='hls', fatal=False)
  61         formats.extend(self._extract_mpd_formats(
  62             manifest_base_url + 'mpd', video_id, mpd_id='dash', fatal=False))
  63         self._sort_formats(formats)
  64
  65         return {
  66             'id': video_id,
  67             'title': video_id,
  68             'thumbnail': base_url + 'thumbnails/thumbnail.jpg',
  69             'formats': formats,
  70         }