yt_dlp/extractor/bloomberg.py

   1 import re
   2
   3 from .common import InfoExtractor
   4
   5
   6 class BloombergIE(InfoExtractor):
   7     _VALID_URL = r'https?://(?:www\.)?bloomberg\.com/(?:[^/]+/)*(?P<id>[^/?#]+)'
   8
   9     _TESTS = [{
  10         'url': 'http://www.bloomberg.com/news/videos/b/aaeae121-5949-481e-a1ce-4562db6f5df2',
  11         # The md5 checksum changes
  12         'info_dict': {
  13             'id': 'qurhIVlJSB6hzkVi229d8g',
  14             'ext': 'flv',
  15             'title': 'Shah\'s Presentation on Foreign-Exchange Strategies',
  16             'description': 'md5:a8ba0302912d03d246979735c17d2761',
  17         },
  18         'params': {
  19             'format': 'best[format_id^=hds]',
  20         },
  21     }, {
  22         # video ID in BPlayer(...)
  23         'url': 'http://www.bloomberg.com/features/2016-hello-world-new-zealand/',
  24         'info_dict': {
  25             'id': '938c7e72-3f25-4ddb-8b85-a9be731baa74',
  26             'ext': 'flv',
  27             'title': 'Meet the Real-Life Tech Wizards of Middle Earth',
  28             'description': 'Hello World, Episode 1: New Zealand’s freaky AI babies, robot exoskeletons, and a virtual you.',
  29         },
  30         'params': {
  31             'format': 'best[format_id^=hds]',
  32         },
  33     }, {
  34         # data-bmmrid=
  35         'url': 'https://www.bloomberg.com/politics/articles/2017-02-08/le-pen-aide-briefed-french-central-banker-on-plan-to-print-money',
  36         'only_matching': True,
  37     }, {
  38         'url': 'http://www.bloomberg.com/news/articles/2015-11-12/five-strange-things-that-have-been-happening-in-financial-markets',
  39         'only_matching': True,
  40     }, {
  41         'url': 'http://www.bloomberg.com/politics/videos/2015-11-25/karl-rove-on-jeb-bush-s-struggles-stopping-trump',
  42         'only_matching': True,
  43     }]
  44
  45     def _real_extract(self, url):
  46         name = self._match_id(url)
  47         webpage = self._download_webpage(url, name)
  48         video_id = self._search_regex(
  49             (r'["\']bmmrId["\']\s*:\s*(["\'])(?P<id>(?:(?!\1).)+)\1',
  50              r'videoId\s*:\s*(["\'])(?P<id>(?:(?!\1).)+)\1',
  51              r'data-bmmrid=(["\'])(?P<id>(?:(?!\1).)+)\1'),
  52             webpage, 'id', group='id', default=None)
  53         if not video_id:
  54             bplayer_data = self._parse_json(self._search_regex(
  55                 r'BPlayer\(null,\s*({[^;]+})\);', webpage, 'id'), name)
  56             video_id = bplayer_data['id']
  57         title = re.sub(': Video$', '', self._og_search_title(webpage))
  58
  59         embed_info = self._download_json(
  60             'http://www.bloomberg.com/api/embed?id=%s' % video_id, video_id)
  61         formats = []
  62         for stream in embed_info['streams']:
  63             stream_url = stream.get('url')
  64             if not stream_url:
  65                 continue
  66             if stream['muxing_format'] == 'TS':
  67                 formats.extend(self._extract_m3u8_formats(
  68                     stream_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
  69             else:
  70                 formats.extend(self._extract_f4m_formats(
  71                     stream_url, video_id, f4m_id='hds', fatal=False))
  72         self._sort_formats(formats)
  73
  74         return {
  75             'id': video_id,
  76             'title': title,
  77             'formats': formats,
  78             'description': self._og_search_description(webpage),
  79             'thumbnail': self._og_search_thumbnail(webpage),
  80         }