[9now] Improve video data extraction (Closes #10561)
[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     int_or_none,
8     float_or_none,
9     ExtractorError,
10 )
11
12
13 class NineNowIE(InfoExtractor):
14     IE_NAME = '9now.com.au'
15     _VALID_URL = r'https?://(?:www\.)?9now\.com\.au/(?:[^/]+/){2}(?P<id>[^/?#]+)'
16     _TESTS = [{
17         # clip
18         'url': 'https://www.9now.com.au/afl-footy-show/2016/clip-ciql02091000g0hp5oktrnytc',
19         'md5': '17cf47d63ec9323e562c9957a968b565',
20         'info_dict': {
21             'id': '16801',
22             'ext': 'mp4',
23             'title': 'St. Kilda\'s Joey Montagna on the potential for a player\'s strike',
24             'description': 'Is a boycott of the NAB Cup "on the table"?',
25             'uploader_id': '4460760524001',
26             'upload_date': '20160713',
27             'timestamp': 1468421266,
28         },
29         'skip': 'Only available in Australia',
30     }, {
31         # episode
32         'url': 'https://www.9now.com.au/afl-footy-show/2016/episode-19',
33         'only_matching': True,
34     }, {
35         # DRM protected
36         'url': 'https://www.9now.com.au/andrew-marrs-history-of-the-world/season-1/episode-1',
37         'only_matching': True,
38     }]
39     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/4460760524001/default_default/index.html?videoId=%s'
40
41     def _real_extract(self, url):
42         display_id = self._match_id(url)
43         webpage = self._download_webpage(url, display_id)
44         page_data = self._parse_json(self._search_regex(
45             r'window\.__data\s*=\s*({.*?});', webpage,
46             'page data'), display_id)
47
48         for kind in ('episode', 'clip'):
49             current_key = page_data.get(kind, {}).get(
50                 'current%sKey' % kind.capitalize())
51             if not current_key:
52                 continue
53             cache = page_data.get(kind, {}).get('%sCache' % kind, {})
54             if not cache:
55                 continue
56             common_data = (cache.get(current_key) or list(cache.values())[0])[kind]
57             break
58         else:
59             raise ExtractorError('Unable to find video data')
60
61         video_data = common_data['video']
62
63         if video_data.get('drm'):
64             raise ExtractorError('This video is DRM protected.', expected=True)
65
66         brightcove_id = video_data.get('brightcoveId') or 'ref:' + video_data['referenceId']
67         video_id = compat_str(video_data.get('id') or brightcove_id)
68         title = common_data['name']
69
70         thumbnails = [{
71             'id': thumbnail_id,
72             'url': thumbnail_url,
73             'width': int_or_none(thumbnail_id[1:])
74         } for thumbnail_id, thumbnail_url in common_data.get('image', {}).get('sizes', {}).items()]
75
76         return {
77             '_type': 'url_transparent',
78             'url': self.BRIGHTCOVE_URL_TEMPLATE % brightcove_id,
79             'id': video_id,
80             'title': title,
81             'description': common_data.get('description'),
82             'duration': float_or_none(video_data.get('duration'), 1000),
83             'thumbnails': thumbnails,
84             'ie_key': 'BrightcoveNew',
85         }