[hrti:playlist] Relax _VALID_URL
[ytdl] / youtube_dl / extractor / hrti.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import json
5 import re
6
7 from .common import InfoExtractor
8 from ..compat import compat_HTTPError
9 from ..utils import (
10     clean_html,
11     ExtractorError,
12     int_or_none,
13     parse_age_limit,
14     sanitized_Request,
15     try_get,
16 )
17
18
19 class HRTiBaseIE(InfoExtractor):
20     """
21         Base Information Extractor for Croatian Radiotelevision
22         video on demand site https://hrti.hrt.hr
23         Reverse engineered from the JavaScript app in app.min.js
24     """
25     _NETRC_MACHINE = 'hrti'
26
27     _APP_LANGUAGE = 'hr'
28     _APP_VERSION = '1.1'
29     _APP_PUBLICATION_ID = 'all_in_one'
30     _API_URL = 'http://clientapi.hrt.hr/client_api.php/config/identify/format/json'
31
32     def _initialize_api(self):
33         init_data = {
34             'application_publication_id': self._APP_PUBLICATION_ID
35         }
36
37         uuid = self._download_json(
38             self._API_URL, None, note='Downloading uuid',
39             errnote='Unable to download uuid',
40             data=json.dumps(init_data).encode('utf-8'))['uuid']
41
42         app_data = {
43             'uuid': uuid,
44             'application_publication_id': self._APP_PUBLICATION_ID,
45             'application_version': self._APP_VERSION
46         }
47
48         req = sanitized_Request(self._API_URL, data=json.dumps(app_data).encode('utf-8'))
49         req.get_method = lambda: 'PUT'
50
51         resources = self._download_json(
52             req, None, note='Downloading session information',
53             errnote='Unable to download session information')
54
55         self._session_id = resources['session_id']
56
57         modules = resources['modules']
58
59         self._search_url = modules['vod_catalog']['resources']['search']['uri'].format(
60             language=self._APP_LANGUAGE,
61             application_id=self._APP_PUBLICATION_ID)
62
63         self._login_url = (modules['user']['resources']['login']['uri'] +
64                            '/format/json').format(session_id=self._session_id)
65
66         self._logout_url = modules['user']['resources']['logout']['uri']
67
68     def _login(self):
69         (username, password) = self._get_login_info()
70         # TODO: figure out authentication with cookies
71         if username is None or password is None:
72             self.raise_login_required()
73
74         auth_data = {
75             'username': username,
76             'password': password,
77         }
78
79         try:
80             auth_info = self._download_json(
81                 self._login_url, None, note='Logging in', errnote='Unable to log in',
82                 data=json.dumps(auth_data).encode('utf-8'))
83         except ExtractorError as e:
84             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 406:
85                 auth_info = self._parse_json(e.cause.read().encode('utf-8'), None)
86             else:
87                 raise
88
89         error_message = auth_info.get('error', {}).get('message')
90         if error_message:
91             raise ExtractorError(
92                 '%s said: %s' % (self.IE_NAME, error_message),
93                 expected=True)
94
95         self._token = auth_info['secure_streaming_token']
96
97     def _real_initialize(self):
98         self._initialize_api()
99         self._login()
100
101
102 class HRTiIE(HRTiBaseIE):
103     _VALID_URL = r'''(?x)
104                         (?:
105                             hrti:(?P<short_id>[0-9]+)|
106                             https?://
107                                 hrti\.hrt\.hr/(?:\#/)?video/show/(?P<id>[0-9]+)/(?P<display_id>[^/]+)?
108                         )
109                     '''
110     _TESTS = [{
111         'url': 'https://hrti.hrt.hr/#/video/show/2181385/republika-dokumentarna-serija-16-hd',
112         'info_dict': {
113             'id': '2181385',
114             'display_id': 'republika-dokumentarna-serija-16-hd',
115             'ext': 'mp4',
116             'title': 'REPUBLIKA, dokumentarna serija (1/6) (HD)',
117             'description': 'md5:48af85f620e8e0e1df4096270568544f',
118             'duration': 2922,
119             'view_count': int,
120             'average_rating': int,
121             'episode_number': int,
122             'season_number': int,
123             'age_limit': 12,
124         },
125         'skip': 'Requires account credentials',
126     }, {
127         'url': 'https://hrti.hrt.hr/#/video/show/2181385/',
128         'only_matching': True,
129     }, {
130         'url': 'hrti:2181385',
131         'only_matching': True,
132     }, {
133         'url': 'https://hrti.hrt.hr/video/show/3873068/cuvar-dvorca-dramska-serija-14',
134         'only_matching': True,
135     }]
136
137     def _real_extract(self, url):
138         mobj = re.match(self._VALID_URL, url)
139         video_id = mobj.group('short_id') or mobj.group('id')
140         display_id = mobj.group('display_id') or video_id
141
142         video = self._download_json(
143             '%s/video_id/%s/format/json' % (self._search_url, video_id),
144             display_id, 'Downloading video metadata JSON')['video'][0]
145
146         title_info = video['title']
147         title = title_info['title_long']
148
149         movie = video['video_assets']['movie'][0]
150         m3u8_url = movie['url'].format(TOKEN=self._token)
151         formats = self._extract_m3u8_formats(
152             m3u8_url, display_id, 'mp4', entry_protocol='m3u8_native',
153             m3u8_id='hls')
154         self._sort_formats(formats)
155
156         description = clean_html(title_info.get('summary_long'))
157         age_limit = parse_age_limit(video.get('parental_control', {}).get('rating'))
158         view_count = int_or_none(video.get('views'))
159         average_rating = int_or_none(video.get('user_rating'))
160         duration = int_or_none(movie.get('duration'))
161
162         return {
163             'id': video_id,
164             'display_id': display_id,
165             'title': title,
166             'description': description,
167             'duration': duration,
168             'view_count': view_count,
169             'average_rating': average_rating,
170             'age_limit': age_limit,
171             'formats': formats,
172         }
173
174
175 class HRTiPlaylistIE(HRTiBaseIE):
176     _VALID_URL = r'https?://hrti\.hrt\.hr/(?:#/)?video/list/category/(?P<id>[0-9]+)/(?P<display_id>[^/]+)?'
177     _TESTS = [{
178         'url': 'https://hrti.hrt.hr/#/video/list/category/212/ekumena',
179         'info_dict': {
180             'id': '212',
181             'title': 'ekumena',
182         },
183         'playlist_mincount': 8,
184         'skip': 'Requires account credentials',
185     }, {
186         'url': 'https://hrti.hrt.hr/#/video/list/category/212/',
187         'only_matching': True,
188     }, {
189         'url': 'https://hrti.hrt.hr/video/list/category/212/ekumena',
190         'only_matching': True,
191     }]
192
193     def _real_extract(self, url):
194         mobj = re.match(self._VALID_URL, url)
195         category_id = mobj.group('id')
196         display_id = mobj.group('display_id') or category_id
197
198         response = self._download_json(
199             '%s/category_id/%s/format/json' % (self._search_url, category_id),
200             display_id, 'Downloading video metadata JSON')
201
202         video_ids = try_get(
203             response, lambda x: x['video_listings'][0]['alternatives'][0]['list'],
204             list) or [video['id'] for video in response.get('videos', []) if video.get('id')]
205
206         entries = [self.url_result('hrti:%s' % video_id) for video_id in video_ids]
207
208         return self.playlist_result(entries, category_id, display_id)