Switch codebase to use sanitized_Request instead of
[youtube-dl] / youtube_dl / extractor / viewster.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 from .common import InfoExtractor
5 from ..compat import (
6     compat_HTTPError,
7     compat_urllib_parse,
8     compat_urllib_parse_unquote,
9 )
10 from ..utils import (
11     determine_ext,
12     ExtractorError,
13     int_or_none,
14     parse_iso8601,
15     sanitized_Request,
16     HEADRequest,
17 )
18
19
20 class ViewsterIE(InfoExtractor):
21     _VALID_URL = r'https?://(?:www\.)?viewster\.com/(?:serie|movie)/(?P<id>\d+-\d+-\d+)'
22     _TESTS = [{
23         # movie, Type=Movie
24         'url': 'http://www.viewster.com/movie/1140-11855-000/the-listening-project/',
25         'md5': 'e642d1b27fcf3a4ffa79f194f5adde36',
26         'info_dict': {
27             'id': '1140-11855-000',
28             'ext': 'mp4',
29             'title': 'The listening Project',
30             'description': 'md5:bac720244afd1a8ea279864e67baa071',
31             'timestamp': 1214870400,
32             'upload_date': '20080701',
33             'duration': 4680,
34         },
35     }, {
36         # series episode, Type=Episode
37         'url': 'http://www.viewster.com/serie/1284-19427-001/the-world-and-a-wall/',
38         'md5': '9243079a8531809efe1b089db102c069',
39         'info_dict': {
40             'id': '1284-19427-001',
41             'ext': 'mp4',
42             'title': 'The World and a Wall',
43             'description': 'md5:24814cf74d3453fdf5bfef9716d073e3',
44             'timestamp': 1428192000,
45             'upload_date': '20150405',
46             'duration': 1500,
47         },
48     }, {
49         # serie, Type=Serie
50         'url': 'http://www.viewster.com/serie/1303-19426-000/',
51         'info_dict': {
52             'id': '1303-19426-000',
53             'title': 'Is It Wrong to Try to Pick up Girls in a Dungeon?',
54             'description': 'md5:eeda9bef25b0d524b3a29a97804c2f11',
55         },
56         'playlist_count': 13,
57     }, {
58         # unfinished serie, no Type
59         'url': 'http://www.viewster.com/serie/1284-19427-000/baby-steps-season-2/',
60         'info_dict': {
61             'id': '1284-19427-000',
62             'title': 'Baby Steps—Season 2',
63             'description': 'md5:e7097a8fc97151e25f085c9eb7a1cdb1',
64         },
65         'playlist_mincount': 16,
66     }, {
67         # geo restricted series
68         'url': 'https://www.viewster.com/serie/1280-18794-002/',
69         'only_matching': True,
70     }, {
71         # geo restricted video
72         'url': 'https://www.viewster.com/serie/1280-18794-002/what-is-extraterritoriality-lawo/',
73         'only_matching': True,
74     }]
75
76     _ACCEPT_HEADER = 'application/json, text/javascript, */*; q=0.01'
77
78     def _download_json(self, url, video_id, note='Downloading JSON metadata', fatal=True):
79         request = sanitized_Request(url)
80         request.add_header('Accept', self._ACCEPT_HEADER)
81         request.add_header('Auth-token', self._AUTH_TOKEN)
82         return super(ViewsterIE, self)._download_json(request, video_id, note, fatal=fatal)
83
84     def _real_extract(self, url):
85         video_id = self._match_id(url)
86         # Get 'api_token' cookie
87         self._request_webpage(HEADRequest('http://www.viewster.com/'), video_id)
88         cookies = self._get_cookies('http://www.viewster.com/')
89         self._AUTH_TOKEN = compat_urllib_parse_unquote(cookies['api_token'].value)
90
91         info = self._download_json(
92             'https://public-api.viewster.com/search/%s' % video_id,
93             video_id, 'Downloading entry JSON')
94
95         entry_id = info.get('Id') or info['id']
96
97         # unfinished serie has no Type
98         if info.get('Type') in ('Serie', None):
99             try:
100                 episodes = self._download_json(
101                     'https://public-api.viewster.com/series/%s/episodes' % entry_id,
102                     video_id, 'Downloading series JSON')
103             except ExtractorError as e:
104                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
105                     self.raise_geo_restricted()
106                 else:
107                     raise
108             entries = [
109                 self.url_result(
110                     'http://www.viewster.com/movie/%s' % episode['OriginId'], 'Viewster')
111                 for episode in episodes]
112             title = (info.get('Title') or info['Synopsis']['Title']).strip()
113             description = info.get('Synopsis', {}).get('Detailed')
114             return self.playlist_result(entries, video_id, title, description)
115
116         formats = []
117         for media_type in ('application/f4m+xml', 'application/x-mpegURL', 'video/mp4'):
118             media = self._download_json(
119                 'https://public-api.viewster.com/movies/%s/video?mediaType=%s'
120                 % (entry_id, compat_urllib_parse.quote(media_type)),
121                 video_id, 'Downloading %s JSON' % media_type, fatal=False)
122             if not media:
123                 continue
124             video_url = media.get('Uri')
125             if not video_url:
126                 continue
127             ext = determine_ext(video_url)
128             if ext == 'f4m':
129                 video_url += '&' if '?' in video_url else '?'
130                 video_url += 'hdcore=3.2.0&plugin=flowplayer-3.2.0.1'
131                 formats.extend(self._extract_f4m_formats(
132                     video_url, video_id, f4m_id='hds'))
133             elif ext == 'm3u8':
134                 m3u8_formats = self._extract_m3u8_formats(
135                     video_url, video_id, 'mp4', m3u8_id='hls',
136                     fatal=False)  # m3u8 sometimes fail
137                 if m3u8_formats:
138                     formats.extend(m3u8_formats)
139             else:
140                 format_id = media.get('Bitrate')
141                 f = {
142                     'url': video_url,
143                     'format_id': 'mp4-%s' % format_id,
144                     'height': int_or_none(media.get('Height')),
145                     'width': int_or_none(media.get('Width')),
146                     'preference': 1,
147                 }
148                 if format_id and not f['height']:
149                     f['height'] = int_or_none(self._search_regex(
150                         r'^(\d+)[pP]$', format_id, 'height', default=None))
151                 formats.append(f)
152
153         if not formats and not info.get('LanguageSets') and not info.get('VODSettings'):
154             self.raise_geo_restricted()
155
156         self._sort_formats(formats)
157
158         synopsis = info.get('Synopsis', {})
159         # Prefer title outside synopsis since it's less messy
160         title = (info.get('Title') or synopsis['Title']).strip()
161         description = synopsis.get('Detailed') or info.get('Synopsis', {}).get('Short')
162         duration = int_or_none(info.get('Duration'))
163         timestamp = parse_iso8601(info.get('ReleaseDate'))
164
165         return {
166             'id': video_id,
167             'title': title,
168             'description': description,
169             'timestamp': timestamp,
170             'duration': duration,
171             'formats': formats,
172         }