]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/cbs.py
Prepare to upload
[youtubedl] / 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 subtitles = {}
17 for k, ext in [('sMPTE-TTCCURL', 'tt'), ('ClosedCaptionURL', 'ttml'), ('webVTTCaptionURL', 'vtt')]:
18 cc_e = find_xpath_attr(smil, self._xpath_ns('.//param', namespace), 'name', k)
19 if cc_e is not None:
20 cc_url = cc_e.get('value')
21 if cc_url:
22 subtitles.setdefault(subtitles_lang, []).append({
23 'ext': ext,
24 'url': cc_url,
25 })
26 return subtitles
27
28
29 class CBSIE(CBSBaseIE):
30 _VALID_URL = r'(?:cbs:|https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/video|colbertlateshow\.com/(?:video|podcasts))/)(?P<id>[\w-]+)'
31
32 _TESTS = [{
33 'url': 'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
34 'info_dict': {
35 'id': '_u7W953k6la293J7EPTd9oHkSPs6Xn6_',
36 'ext': 'mp4',
37 'title': 'Connect Chat feat. Garth Brooks',
38 '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!',
39 'duration': 1495,
40 'timestamp': 1385585425,
41 'upload_date': '20131127',
42 'uploader': 'CBSI-NEW',
43 },
44 'params': {
45 # m3u8 download
46 'skip_download': True,
47 },
48 '_skip': 'Blocked outside the US',
49 }, {
50 'url': 'http://colbertlateshow.com/video/8GmB0oY0McANFvp2aEffk9jZZZ2YyXxy/the-colbeard/',
51 'only_matching': True,
52 }, {
53 'url': 'http://www.colbertlateshow.com/podcasts/dYSwjqPs_X1tvbV_P2FcPWRa_qT6akTC/in-the-bad-room-with-stephen/',
54 'only_matching': True,
55 }]
56
57 def _extract_video_info(self, content_id, site='cbs', mpx_acc=2198311517):
58 items_data = self._download_xml(
59 'http://can.cbs.com/thunder/player/videoPlayerService.php',
60 content_id, query={'partner': site, 'contentId': content_id})
61 video_data = xpath_element(items_data, './/item')
62 title = xpath_text(video_data, 'videoTitle', 'title', True)
63 tp_path = 'dJ5BDC/media/guid/%d/%s' % (mpx_acc, content_id)
64 tp_release_url = 'http://link.theplatform.com/s/' + tp_path
65
66 asset_types = []
67 subtitles = {}
68 formats = []
69 last_e = None
70 for item in items_data.findall('.//item'):
71 asset_type = xpath_text(item, 'assetType')
72 if not asset_type or asset_type in asset_types or 'HLS_FPS' in asset_type or 'DASH_CENC' in asset_type:
73 continue
74 asset_types.append(asset_type)
75 query = {
76 'mbr': 'true',
77 'assetTypes': asset_type,
78 }
79 if asset_type.startswith('HLS') or asset_type in ('OnceURL', 'StreamPack'):
80 query['formats'] = 'MPEG4,M3U'
81 elif asset_type in ('RTMP', 'WIFI', '3G'):
82 query['formats'] = 'MPEG4,FLV'
83 try:
84 tp_formats, tp_subtitles = self._extract_theplatform_smil(
85 update_url_query(tp_release_url, query), content_id,
86 'Downloading %s SMIL data' % asset_type)
87 except ExtractorError as e:
88 last_e = e
89 continue
90 formats.extend(tp_formats)
91 subtitles = self._merge_subtitles(subtitles, tp_subtitles)
92 if last_e and not formats:
93 raise last_e
94 self._sort_formats(formats)
95
96 info = self._extract_theplatform_metadata(tp_path, content_id)
97 info.update({
98 'id': content_id,
99 'title': title,
100 'series': xpath_text(video_data, 'seriesTitle'),
101 'season_number': int_or_none(xpath_text(video_data, 'seasonNumber')),
102 'episode_number': int_or_none(xpath_text(video_data, 'episodeNumber')),
103 'duration': int_or_none(xpath_text(video_data, 'videoLength'), 1000),
104 'thumbnail': xpath_text(video_data, 'previewImageURL'),
105 'formats': formats,
106 'subtitles': subtitles,
107 })
108 return info
109
110 def _real_extract(self, url):
111 content_id = self._match_id(url)
112 return self._extract_video_info(content_id)