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': 'https://www.bloomberg.com/news/videos/2021-09-14/apple-unveils-the-new-iphone-13-stock-doesn-t-move-much-video',
  11         'info_dict': {
  12             'id': 'V8cFcYMxTHaMcEiiYVr39A',
  13             'ext': 'flv',
  14             'title': 'Apple Unveils the New IPhone 13, Stock Doesn\'t Move Much',
  15         },
  16         'params': {
  17             'format': 'best[format_id^=hds]',
  18         },
  19     }, {
  20         # video ID in BPlayer(...)
  21         'url': 'http://www.bloomberg.com/features/2016-hello-world-new-zealand/',
  22         'info_dict': {
  23             'id': '938c7e72-3f25-4ddb-8b85-a9be731baa74',
  24             'ext': 'flv',
  25             'title': 'Meet the Real-Life Tech Wizards of Middle Earth',
  26             'description': 'Hello World, Episode 1: New Zealand’s freaky AI babies, robot exoskeletons, and a virtual you.',
  27         },
  28         'params': {
  29             'format': 'best[format_id^=hds]',
  30         },
  31     }, {
  32         # data-bmmrid=
  33         'url': 'https://www.bloomberg.com/politics/articles/2017-02-08/le-pen-aide-briefed-french-central-banker-on-plan-to-print-money',
  34         'only_matching': True,
  35     }, {
  36         'url': 'http://www.bloomberg.com/news/articles/2015-11-12/five-strange-things-that-have-been-happening-in-financial-markets',
  37         'only_matching': True,
  38     }, {
  39         'url': 'http://www.bloomberg.com/politics/videos/2015-11-25/karl-rove-on-jeb-bush-s-struggles-stopping-trump',
  40         'only_matching': True,
  41     }]
  42
  43     def _real_extract(self, url):
  44         name = self._match_id(url)
  45         webpage = self._download_webpage(url, name)
  46         video_id = self._search_regex(
  47             (r'["\']bmmrId["\']\s*:\s*(["\'])(?P<id>(?:(?!\1).)+)\1',
  48              r'videoId\s*:\s*(["\'])(?P<id>(?:(?!\1).)+)\1',
  49              r'data-bmmrid=(["\'])(?P<id>(?:(?!\1).)+)\1'),
  50             webpage, 'id', group='id', default=None)
  51         if not video_id:
  52             bplayer_data = self._parse_json(self._search_regex(
  53                 r'BPlayer\(null,\s*({[^;]+})\);', webpage, 'id'), name)
  54             video_id = bplayer_data['id']
  55         title = re.sub(': Video$', '', self._og_search_title(webpage))
  56
  57         embed_info = self._download_json(
  58             f'http://www.bloomberg.com/multimedia/api/embed?id={video_id}', video_id)
  59         formats = []
  60         for stream in embed_info['streams']:
  61             stream_url = stream.get('url')
  62             if not stream_url:
  63                 continue
  64             if stream['muxing_format'] == 'TS':
  65                 formats.extend(self._extract_m3u8_formats(
  66                     stream_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
  67             else:
  68                 formats.extend(self._extract_f4m_formats(
  69                     stream_url, video_id, f4m_id='hds', fatal=False))
  70
  71         return {
  72             'id': video_id,
  73             'title': title,
  74             'formats': formats,
  75             'description': self._og_search_description(webpage),
  76             'thumbnail': self._og_search_thumbnail(webpage),
  77         }