[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / dplay.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..compat import compat_HTTPError
8 from ..utils import (
9     determine_ext,
10     ExtractorError,
11     float_or_none,
12     int_or_none,
13     unified_timestamp,
14 )
15
16
17 class DPlayIE(InfoExtractor):
18     _VALID_URL = r'''(?x)https?://
19         (?P<domain>
20             (?:www\.)?(?P<host>dplay\.(?P<country>dk|fi|jp|se|no))|
21             (?P<subdomain_country>es|it)\.dplay\.com
22         )/[^/]+/(?P<id>[^/]+/[^/?#]+)'''
23
24     _TESTS = [{
25         # non geo restricted, via secure api, unsigned download hls URL
26         'url': 'https://www.dplay.se/videos/nugammalt-77-handelser-som-format-sverige/nugammalt-77-handelser-som-format-sverige-101',
27         'info_dict': {
28             'id': '13628',
29             'display_id': 'nugammalt-77-handelser-som-format-sverige/nugammalt-77-handelser-som-format-sverige-101',
30             'ext': 'mp4',
31             'title': 'Svensken lär sig njuta av livet',
32             'description': 'md5:d3819c9bccffd0fe458ca42451dd50d8',
33             'duration': 2649.856,
34             'timestamp': 1365453720,
35             'upload_date': '20130408',
36             'creator': 'Kanal 5',
37             'series': 'Nugammalt - 77 händelser som format Sverige',
38             'season_number': 1,
39             'episode_number': 1,
40         },
41         'params': {
42             'format': 'bestvideo',
43             'skip_download': True,
44         },
45     }, {
46         # geo restricted, via secure api, unsigned download hls URL
47         'url': 'http://www.dplay.dk/videoer/ted-bundy-mind-of-a-monster/ted-bundy-mind-of-a-monster',
48         'info_dict': {
49             'id': '104465',
50             'display_id': 'ted-bundy-mind-of-a-monster/ted-bundy-mind-of-a-monster',
51             'ext': 'mp4',
52             'title': 'Ted Bundy: Mind Of A Monster',
53             'description': 'md5:8b780f6f18de4dae631668b8a9637995',
54             'duration': 5290.027,
55             'timestamp': 1570694400,
56             'upload_date': '20191010',
57             'creator': 'ID - Investigation Discovery',
58             'series': 'Ted Bundy: Mind Of A Monster',
59             'season_number': 1,
60             'episode_number': 1,
61         },
62         'params': {
63             'format': 'bestvideo',
64             'skip_download': True,
65         },
66     }, {
67         # disco-api
68         'url': 'https://www.dplay.no/videoer/i-kongens-klr/sesong-1-episode-7',
69         'info_dict': {
70             'id': '40206',
71             'display_id': 'i-kongens-klr/sesong-1-episode-7',
72             'ext': 'mp4',
73             'title': 'Episode 7',
74             'description': 'md5:e3e1411b2b9aebeea36a6ec5d50c60cf',
75             'duration': 2611.16,
76             'timestamp': 1516726800,
77             'upload_date': '20180123',
78             'series': 'I kongens klær',
79             'season_number': 1,
80             'episode_number': 7,
81         },
82         'params': {
83             'format': 'bestvideo',
84             'skip_download': True,
85         },
86         'skip': 'Available for Premium users',
87     }, {
88         'url': 'http://it.dplay.com/nove/biografie-imbarazzanti/luigi-di-maio-la-psicosi-di-stanislawskij/',
89         'md5': '2b808ffb00fc47b884a172ca5d13053c',
90         'info_dict': {
91             'id': '6918',
92             'display_id': 'biografie-imbarazzanti/luigi-di-maio-la-psicosi-di-stanislawskij',
93             'ext': 'mp4',
94             'title': 'Luigi Di Maio: la psicosi di Stanislawskij',
95             'description': 'md5:3c7a4303aef85868f867a26f5cc14813',
96             'thumbnail': r're:^https?://.*\.jpe?g',
97             'upload_date': '20160524',
98             'timestamp': 1464076800,
99             'series': 'Biografie imbarazzanti',
100             'season_number': 1,
101             'episode': 'Episode 1',
102             'episode_number': 1,
103         },
104     }, {
105         'url': 'https://es.dplay.com/dmax/la-fiebre-del-oro/temporada-8-episodio-1/',
106         'info_dict': {
107             'id': '21652',
108             'display_id': 'la-fiebre-del-oro/temporada-8-episodio-1',
109             'ext': 'mp4',
110             'title': 'Episodio 1',
111             'description': 'md5:b9dcff2071086e003737485210675f69',
112             'thumbnail': r're:^https?://.*\.png',
113             'upload_date': '20180709',
114             'timestamp': 1531173540,
115             'series': 'La fiebre del oro',
116             'season_number': 8,
117             'episode': 'Episode 1',
118             'episode_number': 1,
119         },
120         'params': {
121             'skip_download': True,
122         },
123     }, {
124         'url': 'https://www.dplay.fi/videot/shifting-gears-with-aaron-kaufman/episode-16',
125         'only_matching': True,
126     }, {
127         'url': 'https://www.dplay.jp/video/gold-rush/24086',
128         'only_matching': True,
129     }]
130
131     def _get_disco_api_info(self, url, display_id, disco_host, realm, country):
132         geo_countries = [country.upper()]
133         self._initialize_geo_bypass({
134             'countries': geo_countries,
135         })
136         disco_base = 'https://%s/' % disco_host
137         token = self._download_json(
138             disco_base + 'token', display_id, 'Downloading token',
139             query={
140                 'realm': realm,
141             })['data']['attributes']['token']
142         headers = {
143             'Referer': url,
144             'Authorization': 'Bearer ' + token,
145         }
146         video = self._download_json(
147             disco_base + 'content/videos/' + display_id, display_id,
148             headers=headers, query={
149                 'fields[channel]': 'name',
150                 'fields[image]': 'height,src,width',
151                 'fields[show]': 'name',
152                 'fields[tag]': 'name',
153                 'fields[video]': 'description,episodeNumber,name,publishStart,seasonNumber,videoDuration',
154                 'include': 'images,primaryChannel,show,tags'
155             })
156         video_id = video['data']['id']
157         info = video['data']['attributes']
158         title = info['name'].strip()
159         formats = []
160         try:
161             streaming = self._download_json(
162                 disco_base + 'playback/videoPlaybackInfo/' + video_id,
163                 display_id, headers=headers)['data']['attributes']['streaming']
164         except ExtractorError as e:
165             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
166                 info = self._parse_json(e.cause.read().decode('utf-8'), display_id)
167                 error = info['errors'][0]
168                 error_code = error.get('code')
169                 if error_code == 'access.denied.geoblocked':
170                     self.raise_geo_restricted(countries=geo_countries)
171                 elif error_code == 'access.denied.missingpackage':
172                     self.raise_login_required()
173                 raise ExtractorError(info['errors'][0]['detail'], expected=True)
174             raise
175         for format_id, format_dict in streaming.items():
176             if not isinstance(format_dict, dict):
177                 continue
178             format_url = format_dict.get('url')
179             if not format_url:
180                 continue
181             ext = determine_ext(format_url)
182             if format_id == 'dash' or ext == 'mpd':
183                 formats.extend(self._extract_mpd_formats(
184                     format_url, display_id, mpd_id='dash', fatal=False))
185             elif format_id == 'hls' or ext == 'm3u8':
186                 formats.extend(self._extract_m3u8_formats(
187                     format_url, display_id, 'mp4',
188                     entry_protocol='m3u8_native', m3u8_id='hls',
189                     fatal=False))
190             else:
191                 formats.append({
192                     'url': format_url,
193                     'format_id': format_id,
194                 })
195         self._sort_formats(formats)
196
197         creator = series = None
198         tags = []
199         thumbnails = []
200         included = video.get('included') or []
201         if isinstance(included, list):
202             for e in included:
203                 attributes = e.get('attributes')
204                 if not attributes:
205                     continue
206                 e_type = e.get('type')
207                 if e_type == 'channel':
208                     creator = attributes.get('name')
209                 elif e_type == 'image':
210                     src = attributes.get('src')
211                     if src:
212                         thumbnails.append({
213                             'url': src,
214                             'width': int_or_none(attributes.get('width')),
215                             'height': int_or_none(attributes.get('height')),
216                         })
217                 if e_type == 'show':
218                     series = attributes.get('name')
219                 elif e_type == 'tag':
220                     name = attributes.get('name')
221                     if name:
222                         tags.append(name)
223
224         return {
225             'id': video_id,
226             'display_id': display_id,
227             'title': title,
228             'description': info.get('description'),
229             'duration': float_or_none(info.get('videoDuration'), 1000),
230             'timestamp': unified_timestamp(info.get('publishStart')),
231             'series': series,
232             'season_number': int_or_none(info.get('seasonNumber')),
233             'episode_number': int_or_none(info.get('episodeNumber')),
234             'creator': creator,
235             'tags': tags,
236             'thumbnails': thumbnails,
237             'formats': formats,
238         }
239
240     def _real_extract(self, url):
241         mobj = re.match(self._VALID_URL, url)
242         display_id = mobj.group('id')
243         domain = mobj.group('domain').lstrip('www.')
244         country = mobj.group('country') or mobj.group('subdomain_country')
245         host = 'disco-api.' + domain if domain.startswith('dplay.') else 'eu2-prod.disco-api.com'
246         return self._get_disco_api_info(
247             url, display_id, host, 'dplay' + country, country)