[nhk:vod] Improve extraction (Closes #10424)
[youtube-dl] / youtube_dl / extractor / nhk.py
1 from __future__ import unicode_literals
2
3 from .common import InfoExtractor
4
5
6 class NhkVodIE(InfoExtractor):
7     _VALID_URL = r'https?://www3\.nhk\.or\.jp/nhkworld/en/vod/(?P<id>.+?)\.html'
8     _TEST = {
9         # Videos available only for a limited period of time. Visit
10         # http://www3.nhk.or.jp/nhkworld/en/vod/ for working samples.
11         'url': 'http://www3.nhk.or.jp/nhkworld/en/vod/tokyofashion/20160815.html',
12         'info_dict': {
13             'id': 'A1bnNiNTE6nY3jLllS-BIISfcC_PpvF5',
14             'ext': 'flv',
15             'title': 'TOKYO FASHION EXPRESS - The Kimono as Global Fashion',
16             'description': 'md5:db338ee6ce8204f415b754782f819824',
17             'series': 'TOKYO FASHION EXPRESS',
18             'episode': 'The Kimono as Global Fashion',
19         },
20         'skip': 'Videos available only for a limited period of time',
21     }
22
23     def _real_extract(self, url):
24         video_id = self._match_id(url)
25
26         webpage = self._download_webpage(url, video_id)
27
28         embed_code = self._search_regex(
29             r'nw_vod_ooplayer\([^,]+,\s*(["\'])(?P<id>(?:(?!\1).)+)\1',
30             webpage, 'ooyala embed code', group='id')
31
32         title = self._search_regex(
33             r'<div[^>]+class=["\']episode-detail["\']>\s*<h\d+>([^<]+)',
34             webpage, 'title', default=None)
35         description = self._html_search_regex(
36             r'(?s)<p[^>]+class=["\']description["\'][^>]*>(.+?)</p>',
37             webpage, 'description', default=None)
38         series = self._search_regex(
39             r'<h2[^>]+class=["\']detail-top-player-title[^>]+><a[^>]+>([^<]+)',
40             webpage, 'series', default=None)
41
42         return {
43             '_type': 'url_transparent',
44             'ie_key': 'Ooyala',
45             'url': 'ooyala:%s' % embed_code,
46             'title': '%s - %s' % (series, title) if series and title else title,
47             'description': description,
48             'series': series,
49             'episode': title,
50         }