yt_dlp/extractor/ivi.py

   1 import json
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     ExtractorError,
   7     int_or_none,
   8     qualities,
   9 )
  10
  11
  12 class IviIE(InfoExtractor):
  13     IE_DESC = 'ivi.ru'
  14     IE_NAME = 'ivi'
  15     _VALID_URL = r'https?://(?:www\.)?ivi\.(?:ru|tv)/(?:watch/(?:[^/]+/)?|video/player\?.*?videoId=)(?P<id>\d+)'
  16     _GEO_BYPASS = False
  17     _GEO_COUNTRIES = ['RU']
  18     _LIGHT_KEY = b'\xf1\x02\x32\xb7\xbc\x5c\x7a\xe8\xf7\x96\xc1\x33\x2b\x27\xa1\x8c'
  19     _LIGHT_URL = 'https://api.ivi.ru/light/'
  20
  21     _TESTS = [
  22         # Single movie
  23         {
  24             'url': 'http://www.ivi.ru/watch/53141',
  25             'md5': '6ff5be2254e796ed346251d117196cf4',
  26             'info_dict': {
  27                 'id': '53141',
  28                 'ext': 'mp4',
  29                 'title': 'Иван Васильевич меняет профессию',
  30                 'description': 'md5:b924063ea1677c8fe343d8a72ac2195f',
  31                 'duration': 5498,
  32                 'thumbnail': r're:^https?://.*\.jpg$',
  33             },
  34             'skip': 'Only works from Russia',
  35         },
  36         # Serial's series
  37         {
  38             'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/9549',
  39             'md5': '221f56b35e3ed815fde2df71032f4b3e',
  40             'info_dict': {
  41                 'id': '9549',
  42                 'ext': 'mp4',
  43                 'title': 'Двое из ларца - Дело Гольдберга (1 часть)',
  44                 'series': 'Двое из ларца',
  45                 'season': 'Сезон 1',
  46                 'season_number': 1,
  47                 'episode': 'Дело Гольдберга (1 часть)',
  48                 'episode_number': 1,
  49                 'duration': 2655,
  50                 'thumbnail': r're:^https?://.*\.jpg$',
  51             },
  52             'skip': 'Only works from Russia',
  53         },
  54         {
  55             # with MP4-HD720 format
  56             'url': 'http://www.ivi.ru/watch/146500',
  57             'md5': 'd63d35cdbfa1ea61a5eafec7cc523e1e',
  58             'info_dict': {
  59                 'id': '146500',
  60                 'ext': 'mp4',
  61                 'title': 'Кукла',
  62                 'description': 'md5:ffca9372399976a2d260a407cc74cce6',
  63                 'duration': 5599,
  64                 'thumbnail': r're:^https?://.*\.jpg$',
  65             },
  66             'skip': 'Only works from Russia',
  67         },
  68         {
  69             'url': 'https://www.ivi.tv/watch/33560/',
  70             'only_matching': True,
  71         },
  72     ]
  73
  74     # Sorted by quality
  75     _KNOWN_FORMATS = (
  76         'MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi',
  77         'MP4-SHQ', 'MP4-HD720', 'MP4-HD1080')
  78
  79     def _real_extract(self, url):
  80         video_id = self._match_id(url)
  81
  82         data = json.dumps({
  83             'method': 'da.content.get',
  84             'params': [
  85                 video_id, {
  86                     'site': 's%d',
  87                     'referrer': 'http://www.ivi.ru/watch/%s' % video_id,
  88                     'contentid': video_id
  89                 }
  90             ]
  91         })
  92
  93         for site in (353, 183):
  94             content_data = (data % site).encode()
  95             if site == 353:
  96                 try:
  97                     from Cryptodome.Cipher import Blowfish
  98                     from Cryptodome.Hash import CMAC
  99                     pycryptodome_found = True
 100                 except ImportError:
 101                     try:
 102                         from Crypto.Cipher import Blowfish
 103                         from Crypto.Hash import CMAC
 104                         pycryptodome_found = True
 105                     except ImportError:
 106                         pycryptodome_found = False
 107                         continue
 108
 109                 timestamp = (self._download_json(
 110                     self._LIGHT_URL, video_id,
 111                     'Downloading timestamp JSON', data=json.dumps({
 112                         'method': 'da.timestamp.get',
 113                         'params': []
 114                     }).encode(), fatal=False) or {}).get('result')
 115                 if not timestamp:
 116                     continue
 117
 118                 query = {
 119                     'ts': timestamp,
 120                     'sign': CMAC.new(self._LIGHT_KEY, timestamp.encode() + content_data, Blowfish).hexdigest(),
 121                 }
 122             else:
 123                 query = {}
 124
 125             video_json = self._download_json(
 126                 self._LIGHT_URL, video_id,
 127                 'Downloading video JSON', data=content_data, query=query)
 128
 129             error = video_json.get('error')
 130             if error:
 131                 origin = error.get('origin')
 132                 message = error.get('message') or error.get('user_message')
 133                 extractor_msg = 'Unable to download video %s'
 134                 if origin == 'NotAllowedForLocation':
 135                     self.raise_geo_restricted(message, self._GEO_COUNTRIES)
 136                 elif origin == 'NoRedisValidData':
 137                     extractor_msg = 'Video %s does not exist'
 138                 elif site == 353:
 139                     continue
 140                 elif not pycryptodome_found:
 141                     raise ExtractorError('pycryptodomex not found. Please install', expected=True)
 142                 elif message:
 143                     extractor_msg += ': ' + message
 144                 raise ExtractorError(extractor_msg % video_id, expected=True)
 145             else:
 146                 break
 147
 148         result = video_json['result']
 149         title = result['title']
 150
 151         quality = qualities(self._KNOWN_FORMATS)
 152
 153         formats = []
 154         for f in result.get('files', []):
 155             f_url = f.get('url')
 156             content_format = f.get('content_format')
 157             if not f_url:
 158                 continue
 159             if (not self.get_param('allow_unplayable_formats')
 160                     and ('-MDRM-' in content_format or '-FPS-' in content_format)):
 161                 continue
 162             formats.append({
 163                 'url': f_url,
 164                 'format_id': content_format,
 165                 'quality': quality(content_format),
 166                 'filesize': int_or_none(f.get('size_in_bytes')),
 167             })
 168         self._sort_formats(formats)
 169
 170         compilation = result.get('compilation')
 171         episode = title if compilation else None
 172
 173         title = '%s - %s' % (compilation, title) if compilation is not None else title
 174
 175         thumbnails = [{
 176             'url': preview['url'],
 177             'id': preview.get('content_format'),
 178         } for preview in result.get('preview', []) if preview.get('url')]
 179
 180         webpage = self._download_webpage(url, video_id)
 181
 182         season = self._search_regex(
 183             r'<li[^>]+class="season active"[^>]*><a[^>]+>([^<]+)',
 184             webpage, 'season', default=None)
 185         season_number = int_or_none(self._search_regex(
 186             r'<li[^>]+class="season active"[^>]*><a[^>]+data-season(?:-index)?="(\d+)"',
 187             webpage, 'season number', default=None))
 188
 189         episode_number = int_or_none(self._search_regex(
 190             r'[^>]+itemprop="episode"[^>]*>\s*<meta[^>]+itemprop="episodeNumber"[^>]+content="(\d+)',
 191             webpage, 'episode number', default=None))
 192
 193         description = self._og_search_description(webpage, default=None) or self._html_search_meta(
 194             'description', webpage, 'description', default=None)
 195
 196         return {
 197             'id': video_id,
 198             'title': title,
 199             'series': compilation,
 200             'season': season,
 201             'season_number': season_number,
 202             'episode': episode,
 203             'episode_number': episode_number,
 204             'thumbnails': thumbnails,
 205             'description': description,
 206             'duration': int_or_none(result.get('duration')),
 207             'formats': formats,
 208         }
 209
 210
 211 class IviCompilationIE(InfoExtractor):
 212     IE_DESC = 'ivi.ru compilations'
 213     IE_NAME = 'ivi:compilation'
 214     _VALID_URL = r'https?://(?:www\.)?ivi\.ru/watch/(?!\d+)(?P<compilationid>[a-z\d_-]+)(?:/season(?P<seasonid>\d+))?$'
 215     _TESTS = [{
 216         'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa',
 217         'info_dict': {
 218             'id': 'dvoe_iz_lartsa',
 219             'title': 'Двое из ларца (2006 - 2008)',
 220         },
 221         'playlist_mincount': 24,
 222     }, {
 223         'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/season1',
 224         'info_dict': {
 225             'id': 'dvoe_iz_lartsa/season1',
 226             'title': 'Двое из ларца (2006 - 2008) 1 сезон',
 227         },
 228         'playlist_mincount': 12,
 229     }]
 230
 231     def _extract_entries(self, html, compilation_id):
 232         return [
 233             self.url_result(
 234                 'http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), IviIE.ie_key())
 235             for serie in re.findall(
 236                 r'<a\b[^>]+\bhref=["\']/watch/%s/(\d+)["\']' % compilation_id, html)]
 237
 238     def _real_extract(self, url):
 239         mobj = self._match_valid_url(url)
 240         compilation_id = mobj.group('compilationid')
 241         season_id = mobj.group('seasonid')
 242
 243         if season_id is not None:  # Season link
 244             season_page = self._download_webpage(
 245                 url, compilation_id, 'Downloading season %s web page' % season_id)
 246             playlist_id = '%s/season%s' % (compilation_id, season_id)
 247             playlist_title = self._html_search_meta('title', season_page, 'title')
 248             entries = self._extract_entries(season_page, compilation_id)
 249         else:  # Compilation link
 250             compilation_page = self._download_webpage(url, compilation_id, 'Downloading compilation web page')
 251             playlist_id = compilation_id
 252             playlist_title = self._html_search_meta('title', compilation_page, 'title')
 253             seasons = re.findall(
 254                 r'<a href="/watch/%s/season(\d+)' % compilation_id, compilation_page)
 255             if not seasons:  # No seasons in this compilation
 256                 entries = self._extract_entries(compilation_page, compilation_id)
 257             else:
 258                 entries = []
 259                 for season_id in seasons:
 260                     season_page = self._download_webpage(
 261                         'http://www.ivi.ru/watch/%s/season%s' % (compilation_id, season_id),
 262                         compilation_id, 'Downloading season %s web page' % season_id)
 263                     entries.extend(self._extract_entries(season_page, compilation_id))
 264
 265         return self.playlist_result(entries, playlist_id, playlist_title)