]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vrv.py
c814a8a4a3a04d83cf3c478331cb86bf31f0b592
[youtubedl] / youtube_dl / extractor / vrv.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import base64
5 import json
6 import hashlib
7 import hmac
8 import random
9 import string
10 import time
11
12 from .common import InfoExtractor
13 from ..compat import (
14 compat_HTTPError,
15 compat_urllib_parse_urlencode,
16 compat_urllib_parse,
17 )
18 from ..utils import (
19 ExtractorError,
20 float_or_none,
21 int_or_none,
22 )
23
24
25 class VRVBaseIE(InfoExtractor):
26 _API_DOMAIN = None
27 _API_PARAMS = {}
28 _CMS_SIGNING = {}
29 _TOKEN = None
30 _TOKEN_SECRET = ''
31
32 def _call_api(self, path, video_id, note, data=None):
33 # https://tools.ietf.org/html/rfc5849#section-3
34 base_url = self._API_DOMAIN + '/core/' + path
35 query = [
36 ('oauth_consumer_key', self._API_PARAMS['oAuthKey']),
37 ('oauth_nonce', ''.join([random.choice(string.ascii_letters) for _ in range(32)])),
38 ('oauth_signature_method', 'HMAC-SHA1'),
39 ('oauth_timestamp', int(time.time())),
40 ]
41 if self._TOKEN:
42 query.append(('oauth_token', self._TOKEN))
43 encoded_query = compat_urllib_parse_urlencode(query)
44 headers = self.geo_verification_headers()
45 if data:
46 data = json.dumps(data).encode()
47 headers['Content-Type'] = 'application/json'
48 base_string = '&'.join([
49 'POST' if data else 'GET',
50 compat_urllib_parse.quote(base_url, ''),
51 compat_urllib_parse.quote(encoded_query, '')])
52 oauth_signature = base64.b64encode(hmac.new(
53 (self._API_PARAMS['oAuthSecret'] + '&' + self._TOKEN_SECRET).encode('ascii'),
54 base_string.encode(), hashlib.sha1).digest()).decode()
55 encoded_query += '&oauth_signature=' + compat_urllib_parse.quote(oauth_signature, '')
56 try:
57 return self._download_json(
58 '?'.join([base_url, encoded_query]), video_id,
59 note='Downloading %s JSON metadata' % note, headers=headers, data=data)
60 except ExtractorError as e:
61 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
62 raise ExtractorError(json.loads(e.cause.read().decode())['message'], expected=True)
63 raise
64
65 def _call_cms(self, path, video_id, note):
66 if not self._CMS_SIGNING:
67 self._CMS_SIGNING = self._call_api('index', video_id, 'CMS Signing')['cms_signing']
68 return self._download_json(
69 self._API_DOMAIN + path, video_id, query=self._CMS_SIGNING,
70 note='Downloading %s JSON metadata' % note, headers=self.geo_verification_headers())
71
72 def _get_cms_resource(self, resource_key, video_id):
73 return self._call_api(
74 'cms_resource', video_id, 'resource path', data={
75 'resource_key': resource_key,
76 })['__links__']['cms_resource']['href']
77
78 def _real_initialize(self):
79 webpage = self._download_webpage(
80 'https://vrv.co/', None, headers=self.geo_verification_headers())
81 self._API_PARAMS = self._parse_json(self._search_regex(
82 [
83 r'window\.__APP_CONFIG__\s*=\s*({.+?})(?:</script>|;)',
84 r'window\.__APP_CONFIG__\s*=\s*({.+})'
85 ], webpage, 'app config'), None)['cxApiParams']
86 self._API_DOMAIN = self._API_PARAMS.get('apiDomain', 'https://api.vrv.co')
87
88
89 class VRVIE(VRVBaseIE):
90 IE_NAME = 'vrv'
91 _VALID_URL = r'https?://(?:www\.)?vrv\.co/watch/(?P<id>[A-Z0-9]+)'
92 _TESTS = [{
93 'url': 'https://vrv.co/watch/GR9PNZ396/Hidden-America-with-Jonah-Ray:BOSTON-WHERE-THE-PAST-IS-THE-PRESENT',
94 'info_dict': {
95 'id': 'GR9PNZ396',
96 'ext': 'mp4',
97 'title': 'BOSTON: WHERE THE PAST IS THE PRESENT',
98 'description': 'md5:4ec8844ac262ca2df9e67c0983c6b83f',
99 'uploader_id': 'seeso',
100 },
101 'params': {
102 # m3u8 download
103 'skip_download': True,
104 },
105 }, {
106 # movie listing
107 'url': 'https://vrv.co/watch/G6NQXZ1J6/Lily-CAT',
108 'info_dict': {
109 'id': 'G6NQXZ1J6',
110 'title': 'Lily C.A.T',
111 'description': 'md5:988b031e7809a6aeb60968be4af7db07',
112 },
113 'playlist_count': 2,
114 }]
115 _NETRC_MACHINE = 'vrv'
116
117 def _real_initialize(self):
118 super(VRVIE, self)._real_initialize()
119
120 email, password = self._get_login_info()
121 if email is None:
122 return
123
124 token_credentials = self._call_api(
125 'authenticate/by:credentials', None, 'Token Credentials', data={
126 'email': email,
127 'password': password,
128 })
129 self._TOKEN = token_credentials['oauth_token']
130 self._TOKEN_SECRET = token_credentials['oauth_token_secret']
131
132 def _extract_vrv_formats(self, url, video_id, stream_format, audio_lang, hardsub_lang):
133 if not url or stream_format not in ('hls', 'dash', 'adaptive_hls'):
134 return []
135 stream_id_list = []
136 if audio_lang:
137 stream_id_list.append('audio-%s' % audio_lang)
138 if hardsub_lang:
139 stream_id_list.append('hardsub-%s' % hardsub_lang)
140 format_id = stream_format
141 if stream_id_list:
142 format_id += '-' + '-'.join(stream_id_list)
143 if 'hls' in stream_format:
144 adaptive_formats = self._extract_m3u8_formats(
145 url, video_id, 'mp4', m3u8_id=format_id,
146 note='Downloading %s information' % format_id,
147 fatal=False)
148 elif stream_format == 'dash':
149 adaptive_formats = self._extract_mpd_formats(
150 url, video_id, mpd_id=format_id,
151 note='Downloading %s information' % format_id,
152 fatal=False)
153 if audio_lang:
154 for f in adaptive_formats:
155 if f.get('acodec') != 'none':
156 f['language'] = audio_lang
157 return adaptive_formats
158
159 def _real_extract(self, url):
160 video_id = self._match_id(url)
161
162 object_data = self._call_cms(self._get_cms_resource(
163 'cms:/objects/' + video_id, video_id), video_id, 'object')['items'][0]
164 resource_path = object_data['__links__']['resource']['href']
165 video_data = self._call_cms(resource_path, video_id, 'video')
166 title = video_data['title']
167 description = video_data.get('description')
168
169 if video_data.get('__class__') == 'movie_listing':
170 items = self._call_cms(
171 video_data['__links__']['movie_listing/movies']['href'],
172 video_id, 'movie listing').get('items') or []
173 if len(items) != 1:
174 entries = []
175 for item in items:
176 item_id = item.get('id')
177 if not item_id:
178 continue
179 entries.append(self.url_result(
180 'https://vrv.co/watch/' + item_id,
181 self.ie_key(), item_id, item.get('title')))
182 return self.playlist_result(entries, video_id, title, description)
183 video_data = items[0]
184
185 streams_path = video_data['__links__'].get('streams', {}).get('href')
186 if not streams_path:
187 self.raise_login_required()
188 streams_json = self._call_cms(streams_path, video_id, 'streams')
189
190 audio_locale = streams_json.get('audio_locale')
191 formats = []
192 for stream_type, streams in streams_json.get('streams', {}).items():
193 if stream_type in ('adaptive_hls', 'adaptive_dash'):
194 for stream in streams.values():
195 formats.extend(self._extract_vrv_formats(
196 stream.get('url'), video_id, stream_type.split('_')[1],
197 audio_locale, stream.get('hardsub_locale')))
198 self._sort_formats(formats)
199
200 subtitles = {}
201 for k in ('captions', 'subtitles'):
202 for subtitle in streams_json.get(k, {}).values():
203 subtitle_url = subtitle.get('url')
204 if not subtitle_url:
205 continue
206 subtitles.setdefault(subtitle.get('locale', 'en-US'), []).append({
207 'url': subtitle_url,
208 'ext': subtitle.get('format', 'ass'),
209 })
210
211 thumbnails = []
212 for thumbnail in video_data.get('images', {}).get('thumbnails', []):
213 thumbnail_url = thumbnail.get('source')
214 if not thumbnail_url:
215 continue
216 thumbnails.append({
217 'url': thumbnail_url,
218 'width': int_or_none(thumbnail.get('width')),
219 'height': int_or_none(thumbnail.get('height')),
220 })
221
222 return {
223 'id': video_id,
224 'title': title,
225 'formats': formats,
226 'subtitles': subtitles,
227 'thumbnails': thumbnails,
228 'description': description,
229 'duration': float_or_none(video_data.get('duration_ms'), 1000),
230 'uploader_id': video_data.get('channel_id'),
231 'series': video_data.get('series_title'),
232 'season': video_data.get('season_title'),
233 'season_number': int_or_none(video_data.get('season_number')),
234 'season_id': video_data.get('season_id'),
235 'episode': title,
236 'episode_number': int_or_none(video_data.get('episode_number')),
237 'episode_id': video_data.get('production_episode_id'),
238 }
239
240
241 class VRVSeriesIE(VRVBaseIE):
242 IE_NAME = 'vrv:series'
243 _VALID_URL = r'https?://(?:www\.)?vrv\.co/series/(?P<id>[A-Z0-9]+)'
244 _TEST = {
245 'url': 'https://vrv.co/series/G68VXG3G6/The-Perfect-Insider',
246 'info_dict': {
247 'id': 'G68VXG3G6',
248 },
249 'playlist_mincount': 11,
250 }
251
252 def _real_extract(self, url):
253 series_id = self._match_id(url)
254
255 seasons_path = self._get_cms_resource(
256 'cms:/seasons?series_id=' + series_id, series_id)
257 seasons_data = self._call_cms(seasons_path, series_id, 'seasons')
258
259 entries = []
260 for season in seasons_data.get('items', []):
261 episodes_path = season['__links__']['season/episodes']['href']
262 episodes = self._call_cms(episodes_path, series_id, 'episodes')
263 for episode in episodes.get('items', []):
264 episode_id = episode['id']
265 entries.append(self.url_result(
266 'https://vrv.co/watch/' + episode_id,
267 'VRV', episode_id, episode.get('title')))
268
269 return self.playlist_result(entries, series_id)