]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/bloomberg.py
bd538be50bc4a650d000eaffb5e292a4e8e76cbe
[youtubedl] / 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 'params': {
21 'format': 'best[format_id^=hds]',
22 },
23 }, {
24 'url': 'http://www.bloomberg.com/news/articles/2015-11-12/five-strange-things-that-have-been-happening-in-financial-markets',
25 'only_matching': True,
26 }, {
27 'url': 'http://www.bloomberg.com/politics/videos/2015-11-25/karl-rove-on-jeb-bush-s-struggles-stopping-trump',
28 'only_matching': True,
29 }]
30
31 def _real_extract(self, url):
32 name = self._match_id(url)
33 webpage = self._download_webpage(url, name)
34 video_id = self._search_regex(
35 r'["\']bmmrId["\']\s*:\s*(["\'])(?P<url>.+?)\1',
36 webpage, 'id', group='url')
37 title = re.sub(': Video$', '', self._og_search_title(webpage))
38
39 embed_info = self._download_json(
40 'http://www.bloomberg.com/api/embed?id=%s' % video_id, video_id)
41 formats = []
42 for stream in embed_info['streams']:
43 stream_url = stream.get('url')
44 if not stream_url:
45 continue
46 if stream['muxing_format'] == 'TS':
47 formats.extend(self._extract_m3u8_formats(
48 stream_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
49 else:
50 formats.extend(self._extract_f4m_formats(
51 stream_url, video_id, f4m_id='hds', fatal=False))
52 self._sort_formats(formats)
53
54 return {
55 'id': video_id,
56 'title': title,
57 'formats': formats,
58 'description': self._og_search_description(webpage),
59 'thumbnail': self._og_search_thumbnail(webpage),
60 }