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

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6
   7
   8 class BloombergIE(InfoExtractor):
   9     _VALID_URL = r'https?://(?:www\.)?bloomberg\.com/(?:[^/]+/)*(?P<id>[^/?#]+)'
  10
  11     _TESTS = [{
  12         'url': 'http://www.bloomberg.com/news/videos/b/aaeae121-5949-481e-a1ce-4562db6f5df2',
  13         # The md5 checksum changes
  14         'info_dict': {
  15             'id': 'qurhIVlJSB6hzkVi229d8g',
  16             'ext': 'flv',
  17             'title': 'Shah\'s Presentation on Foreign-Exchange Strategies',
  18             'description': 'md5:a8ba0302912d03d246979735c17d2761',
  19         },
  20     }, {
  21         'url': 'http://www.bloomberg.com/news/articles/2015-11-12/five-strange-things-that-have-been-happening-in-financial-markets',
  22         'only_matching': True,
  23     }, {
  24         'url': 'http://www.bloomberg.com/politics/videos/2015-11-25/karl-rove-on-jeb-bush-s-struggles-stopping-trump',
  25         'only_matching': True,
  26     }]
  27
  28     def _real_extract(self, url):
  29         name = self._match_id(url)
  30         webpage = self._download_webpage(url, name)
  31         video_id = self._search_regex(r'"bmmrId":"(.+?)"', webpage, 'id')
  32         title = re.sub(': Video$', '', self._og_search_title(webpage))
  33
  34         embed_info = self._download_json(
  35             'http://www.bloomberg.com/api/embed?id=%s' % video_id, video_id)
  36         formats = []
  37         for stream in embed_info['streams']:
  38             if stream['muxing_format'] == 'TS':
  39                 formats.extend(self._extract_m3u8_formats(stream['url'], video_id))
  40             else:
  41                 formats.extend(self._extract_f4m_formats(stream['url'], video_id))
  42         self._sort_formats(formats)
  43
  44         return {
  45             'id': video_id,
  46             'title': title,
  47             'formats': formats,
  48             'description': self._og_search_description(webpage),
  49             'thumbnail': self._og_search_thumbnail(webpage),
  50         }