]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/cbs.py
[iprima] Relax _VALID_URL (closes #18515)
[youtube-dl] / youtube_dl / extractor / cbs.py
1 from __future__ import unicode_literals
2
3 from .theplatform import ThePlatformFeedIE
4 from ..utils import (
5     ExtractorError,
6     int_or_none,
7     find_xpath_attr,
8     xpath_element,
9     xpath_text,
10     update_url_query,
11 )
12
13
14 class CBSBaseIE(ThePlatformFeedIE):
15     def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'):
16         closed_caption_e = find_xpath_attr(smil, self._xpath_ns('.//param', namespace), 'name', 'ClosedCaptionURL')
17         return {
18             'en': [{
19                 'ext': 'ttml',
20                 'url': closed_caption_e.attrib['value'],
21             }]
22         } if closed_caption_e is not None and closed_caption_e.attrib.get('value') else []
23
24
25 class CBSIE(CBSBaseIE):
26     _VALID_URL = r'(?:cbs:|https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/video|colbertlateshow\.com/(?:video|podcasts))/)(?P<id>[\w-]+)'
27
28     _TESTS = [{
29         'url': 'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
30         'info_dict': {
31             'id': '_u7W953k6la293J7EPTd9oHkSPs6Xn6_',
32             'ext': 'mp4',
33             'title': 'Connect Chat feat. Garth Brooks',
34             'description': 'Connect with country music singer Garth Brooks, as he chats with fans on Wednesday November 27, 2013. Be sure to tune in to Garth Brooks: Live from Las Vegas, Friday November 29, at 9/8c on CBS!',
35             'duration': 1495,
36             'timestamp': 1385585425,
37             'upload_date': '20131127',
38             'uploader': 'CBSI-NEW',
39         },
40         'params': {
41             # m3u8 download
42             'skip_download': True,
43         },
44         '_skip': 'Blocked outside the US',
45     }, {
46         'url': 'http://colbertlateshow.com/video/8GmB0oY0McANFvp2aEffk9jZZZ2YyXxy/the-colbeard/',
47         'only_matching': True,
48     }, {
49         'url': 'http://www.colbertlateshow.com/podcasts/dYSwjqPs_X1tvbV_P2FcPWRa_qT6akTC/in-the-bad-room-with-stephen/',
50         'only_matching': True,
51     }]
52
53     def _extract_video_info(self, content_id, site='cbs', mpx_acc=2198311517):
54         items_data = self._download_xml(
55             'http://can.cbs.com/thunder/player/videoPlayerService.php',
56             content_id, query={'partner': site, 'contentId': content_id})
57         video_data = xpath_element(items_data, './/item')
58         title = xpath_text(video_data, 'videoTitle', 'title', True)
59         tp_path = 'dJ5BDC/media/guid/%d/%s' % (mpx_acc, content_id)
60         tp_release_url = 'http://link.theplatform.com/s/' + tp_path
61
62         asset_types = []
63         subtitles = {}
64         formats = []
65         last_e = None
66         for item in items_data.findall('.//item'):
67             asset_type = xpath_text(item, 'assetType')
68             if not asset_type or asset_type in asset_types or asset_type in ('HLS_FPS', 'DASH_CENC'):
69                 continue
70             asset_types.append(asset_type)
71             query = {
72                 'mbr': 'true',
73                 'assetTypes': asset_type,
74             }
75             if asset_type.startswith('HLS') or asset_type in ('OnceURL', 'StreamPack'):
76                 query['formats'] = 'MPEG4,M3U'
77             elif asset_type in ('RTMP', 'WIFI', '3G'):
78                 query['formats'] = 'MPEG4,FLV'
79             try:
80                 tp_formats, tp_subtitles = self._extract_theplatform_smil(
81                     update_url_query(tp_release_url, query), content_id,
82                     'Downloading %s SMIL data' % asset_type)
83             except ExtractorError as e:
84                 last_e = e
85                 continue
86             formats.extend(tp_formats)
87             subtitles = self._merge_subtitles(subtitles, tp_subtitles)
88         if last_e and not formats:
89             raise last_e
90         self._sort_formats(formats)
91
92         info = self._extract_theplatform_metadata(tp_path, content_id)
93         info.update({
94             'id': content_id,
95             'title': title,
96             'series': xpath_text(video_data, 'seriesTitle'),
97             'season_number': int_or_none(xpath_text(video_data, 'seasonNumber')),
98             'episode_number': int_or_none(xpath_text(video_data, 'episodeNumber')),
99             'duration': int_or_none(xpath_text(video_data, 'videoLength'), 1000),
100             'thumbnail': xpath_text(video_data, 'previewImageURL'),
101             'formats': formats,
102             'subtitles': subtitles,
103         })
104         return info
105
106     def _real_extract(self, url):
107         content_id = self._match_id(url)
108         return self._extract_video_info(content_id)