[trilulilu] improve extraction
[youtube-dl] / youtube_dl / extractor / nba.py
1 from __future__ import unicode_literals
2
3 from .common import InfoExtractor
4 from ..utils import (
5     remove_end,
6     parse_duration,
7 )
8
9
10 class NBAIE(InfoExtractor):
11     _VALID_URL = r'https?://(?:watch\.|www\.)?nba\.com/(?:nba/)?video(?P<id>/[^?]*?)/?(?:/index\.html)?(?:\?.*)?$'
12     _TESTS = [{
13         'url': 'http://www.nba.com/video/games/nets/2012/12/04/0021200253-okc-bkn-recap.nba/index.html',
14         'md5': 'c0edcfc37607344e2ff8f13c378c88a4',
15         'info_dict': {
16             'id': '0021200253-okc-bkn-recap.nba',
17             'ext': 'mp4',
18             'title': 'Thunder vs. Nets',
19             'description': 'Kevin Durant scores 32 points and dishes out six assists as the Thunder beat the Nets in Brooklyn.',
20             'duration': 181,
21         },
22     }, {
23         'url': 'http://www.nba.com/video/games/hornets/2014/12/05/0021400276-nyk-cha-play5.nba/',
24         'only_matching': True,
25     }, {
26         'url': 'http://watch.nba.com/nba/video/channels/playoffs/2015/05/20/0041400301-cle-atl-recap.nba',
27         'info_dict': {
28             'id': '0041400301-cle-atl-recap.nba',
29             'ext': 'mp4',
30             'title': 'NBA GAME TIME | Video: Hawks vs. Cavaliers Game 1',
31             'description': 'md5:8094c3498d35a9bd6b1a8c396a071b4d',
32             'duration': 228,
33         },
34         'params': {
35             'skip_download': True,
36         }
37     }]
38
39     def _real_extract(self, url):
40         video_id = self._match_id(url)
41         webpage = self._download_webpage(url, video_id)
42
43         video_url = 'http://ht-mobile.cdn.turner.com/nba/big' + video_id + '_nba_1280x720.mp4'
44
45         shortened_video_id = video_id.rpartition('/')[2]
46         title = remove_end(
47             self._og_search_title(webpage, default=shortened_video_id), ' : NBA.com')
48
49         description = self._og_search_description(webpage)
50         duration_str = self._html_search_meta(
51             'duration', webpage, 'duration', default=None)
52         if not duration_str:
53             duration_str = self._html_search_regex(
54                 r'Duration:</b>\s*(\d+:\d+)', webpage, 'duration', fatal=False)
55         duration = parse_duration(duration_str)
56
57         return {
58             'id': shortened_video_id,
59             'url': video_url,
60             'title': title,
61             'description': description,
62             'duration': duration,
63         }