[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / ninenow.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 from .common import InfoExtractor
5 from ..compat import compat_str
6 from ..utils import (
7     ExtractorError,
8     int_or_none,
9     float_or_none,
10     smuggle_url,
11 )
12
13
14 class NineNowIE(InfoExtractor):
15     IE_NAME = '9now.com.au'
16     _VALID_URL = r'https?://(?:www\.)?9now\.com\.au/(?:[^/]+/){2}(?P<id>[^/?#]+)'
17     _GEO_COUNTRIES = ['AU']
18     _TESTS = [{
19         # clip
20         'url': 'https://www.9now.com.au/afl-footy-show/2016/clip-ciql02091000g0hp5oktrnytc',
21         'md5': '17cf47d63ec9323e562c9957a968b565',
22         'info_dict': {
23             'id': '16801',
24             'ext': 'mp4',
25             'title': 'St. Kilda\'s Joey Montagna on the potential for a player\'s strike',
26             'description': 'Is a boycott of the NAB Cup "on the table"?',
27             'uploader_id': '4460760524001',
28             'upload_date': '20160713',
29             'timestamp': 1468421266,
30         },
31         'skip': 'Only available in Australia',
32     }, {
33         # episode
34         'url': 'https://www.9now.com.au/afl-footy-show/2016/episode-19',
35         'only_matching': True,
36     }, {
37         # DRM protected
38         'url': 'https://www.9now.com.au/andrew-marrs-history-of-the-world/season-1/episode-1',
39         'only_matching': True,
40     }]
41     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/4460760524001/default_default/index.html?videoId=%s'
42
43     def _real_extract(self, url):
44         display_id = self._match_id(url)
45         webpage = self._download_webpage(url, display_id)
46         page_data = self._parse_json(self._search_regex(
47             r'window\.__data\s*=\s*({.*?});', webpage,
48             'page data', default='{}'), display_id, fatal=False)
49         if not page_data:
50             page_data = self._parse_json(self._parse_json(self._search_regex(
51                 r'window\.__data\s*=\s*JSON\.parse\s*\(\s*(".+?")\s*\)\s*;',
52                 webpage, 'page data'), display_id), display_id)
53
54         for kind in ('episode', 'clip'):
55             current_key = page_data.get(kind, {}).get(
56                 'current%sKey' % kind.capitalize())
57             if not current_key:
58                 continue
59             cache = page_data.get(kind, {}).get('%sCache' % kind, {})
60             if not cache:
61                 continue
62             common_data = (cache.get(current_key) or list(cache.values())[0])[kind]
63             break
64         else:
65             raise ExtractorError('Unable to find video data')
66
67         video_data = common_data['video']
68
69         if video_data.get('drm'):
70             raise ExtractorError('This video is DRM protected.', expected=True)
71
72         brightcove_id = video_data.get('brightcoveId') or 'ref:' + video_data['referenceId']
73         video_id = compat_str(video_data.get('id') or brightcove_id)
74         title = common_data['name']
75
76         thumbnails = [{
77             'id': thumbnail_id,
78             'url': thumbnail_url,
79             'width': int_or_none(thumbnail_id[1:])
80         } for thumbnail_id, thumbnail_url in common_data.get('image', {}).get('sizes', {}).items()]
81
82         return {
83             '_type': 'url_transparent',
84             'url': smuggle_url(
85                 self.BRIGHTCOVE_URL_TEMPLATE % brightcove_id,
86                 {'geo_countries': self._GEO_COUNTRIES}),
87             'id': video_id,
88             'title': title,
89             'description': common_data.get('description'),
90             'duration': float_or_none(video_data.get('duration'), 1000),
91             'thumbnails': thumbnails,
92             'ie_key': 'BrightcoveNew',
93         }