Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/bloomberg.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7
   8
   9 class BloombergIE(InfoExtractor):
  10     _VALID_URL = r'https?://(?:www\.)?bloomberg\.com/(?:[^/]+/)*(?P<id>[^/?#]+)'
  11
  12     _TESTS = [{
  13         'url': 'http://www.bloomberg.com/news/videos/b/aaeae121-5949-481e-a1ce-4562db6f5df2',
  14         # The md5 checksum changes
  15         'info_dict': {
  16             'id': 'qurhIVlJSB6hzkVi229d8g',
  17             'ext': 'flv',
  18             'title': 'Shah\'s Presentation on Foreign-Exchange Strategies',
  19             'description': 'md5:a8ba0302912d03d246979735c17d2761',
  20         },
  21         'params': {
  22             'format': 'best[format_id^=hds]',
  23         },
  24     }, {
  25         # video ID in BPlayer(...)
  26         'url': 'http://www.bloomberg.com/features/2016-hello-world-new-zealand/',
  27         'info_dict': {
  28             'id': '938c7e72-3f25-4ddb-8b85-a9be731baa74',
  29             'ext': 'flv',
  30             'title': 'Meet the Real-Life Tech Wizards of Middle Earth',
  31             'description': 'Hello World, Episode 1: New Zealand’s freaky AI babies, robot exoskeletons, and a virtual you.',
  32         },
  33         'params': {
  34             'format': 'best[format_id^=hds]',
  35         },
  36     }, {
  37         # data-bmmrid=
  38         'url': 'https://www.bloomberg.com/politics/articles/2017-02-08/le-pen-aide-briefed-french-central-banker-on-plan-to-print-money',
  39         'only_matching': True,
  40     }, {
  41         'url': 'http://www.bloomberg.com/news/articles/2015-11-12/five-strange-things-that-have-been-happening-in-financial-markets',
  42         'only_matching': True,
  43     }, {
  44         'url': 'http://www.bloomberg.com/politics/videos/2015-11-25/karl-rove-on-jeb-bush-s-struggles-stopping-trump',
  45         'only_matching': True,
  46     }]
  47
  48     def _real_extract(self, url):
  49         name = self._match_id(url)
  50         webpage = self._download_webpage(url, name)
  51         video_id = self._search_regex(
  52             (r'["\']bmmrId["\']\s*:\s*(["\'])(?P<id>(?:(?!\1).)+)\1',
  53              r'videoId\s*:\s*(["\'])(?P<id>(?:(?!\1).)+)\1',
  54              r'data-bmmrid=(["\'])(?P<id>(?:(?!\1).)+)\1'),
  55             webpage, 'id', group='id', default=None)
  56         if not video_id:
  57             bplayer_data = self._parse_json(self._search_regex(
  58                 r'BPlayer\(null,\s*({[^;]+})\);', webpage, 'id'), name)
  59             video_id = bplayer_data['id']
  60         title = re.sub(': Video$', '', self._og_search_title(webpage))
  61
  62         embed_info = self._download_json(
  63             'http://www.bloomberg.com/api/embed?id=%s' % video_id, video_id)
  64         formats = []
  65         for stream in embed_info['streams']:
  66             stream_url = stream.get('url')
  67             if not stream_url:
  68                 continue
  69             if stream['muxing_format'] == 'TS':
  70                 formats.extend(self._extract_m3u8_formats(
  71                     stream_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
  72             else:
  73                 formats.extend(self._extract_f4m_formats(
  74                     stream_url, video_id, f4m_id='hds', fatal=False))
  75         self._sort_formats(formats)
  76
  77         return {
  78             'id': video_id,
  79             'title': title,
  80             'formats': formats,
  81             'description': self._og_search_description(webpage),
  82             'thumbnail': self._og_search_thumbnail(webpage),
  83         }