]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/pbs.py
Merge branch 'ustream-embed-recorded2' of https://github.com/anovicecodemonkey/youtub...
[youtube-dl] / youtube_dl / extractor / pbs.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from ..utils import (
7     US_RATINGS,
8 )
9
10
11 class PBSIE(InfoExtractor):
12     _VALID_URL = r'''(?x)https?://
13         (?:
14             # Direct video URL
15             video\.pbs\.org/(?:viralplayer|video)/(?P<id>[0-9]+)/? |
16             # Article with embedded player
17            (?:www\.)?pbs\.org/(?:[^/]+/){2,5}(?P<presumptive_id>[^/]+)/?(?:$|[?\#]) |
18            # Player
19            video\.pbs\.org/(?:widget/)?partnerplayer/(?P<player_id>[^/]+)/
20         )
21     '''
22
23     _TEST = {
24         'url': 'http://www.pbs.org/tpt/constitution-usa-peter-sagal/watch/a-more-perfect-union/',
25         'md5': 'ce1888486f0908d555a8093cac9a7362',
26         'info_dict': {
27             'id': '2365006249',
28             'ext': 'mp4',
29             'title': 'A More Perfect Union',
30             'description': 'md5:ba0c207295339c8d6eced00b7c363c6a',
31             'duration': 3190,
32         },
33     }
34
35     def _real_extract(self, url):
36         mobj = re.match(self._VALID_URL, url)
37
38         presumptive_id = mobj.group('presumptive_id')
39         display_id = presumptive_id
40         if presumptive_id:
41             webpage = self._download_webpage(url, display_id)
42             url = self._search_regex(
43                 r'<iframe\s+id=["\']partnerPlayer["\'].*?\s+src=["\'](.*?)["\']>',
44                 webpage, 'player URL')
45             mobj = re.match(self._VALID_URL, url)
46
47         player_id = mobj.group('player_id')
48         if not display_id:
49             display_id = player_id
50         if player_id:
51             player_page = self._download_webpage(
52                 url, display_id, note='Downloading player page',
53                 errnote='Could not download player page')
54             video_id = self._search_regex(
55                 r'<div\s+id="video_([0-9]+)"', player_page, 'video ID')
56         else:
57             video_id = mobj.group('id')
58             display_id = video_id
59
60         info_url = 'http://video.pbs.org/videoInfo/%s?format=json' % video_id
61         info = self._download_json(info_url, display_id)
62
63         rating_str = info.get('rating')
64         if rating_str is not None:
65             rating_str = rating_str.rpartition('-')[2]
66         age_limit = US_RATINGS.get(rating_str)
67
68         return {
69             'id': video_id,
70             'title': info['title'],
71             'url': info['alternate_encoding']['url'],
72             'ext': 'mp4',
73             'description': info['program'].get('description'),
74             'thumbnail': info.get('image_url'),
75             'duration': info.get('duration'),
76             'age_limit': age_limit,
77         }