yt_dlp/extractor/ivi.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 ..utils import (
   9     ExtractorError,
  10     int_or_none,
  11     qualities,
  12 )
  13
  14
  15 class IviIE(InfoExtractor):
  16     IE_DESC = 'ivi.ru'
  17     IE_NAME = 'ivi'
  18     _VALID_URL = r'https?://(?:www\.)?ivi\.(?:ru|tv)/(?:watch/(?:[^/]+/)?|video/player\?.*?videoId=)(?P<id>\d+)'
  19     _GEO_BYPASS = False
  20     _GEO_COUNTRIES = ['RU']
  21     _LIGHT_KEY = b'\xf1\x02\x32\xb7\xbc\x5c\x7a\xe8\xf7\x96\xc1\x33\x2b\x27\xa1\x8c'
  22     _LIGHT_URL = 'https://api.ivi.ru/light/'
  23
  24     _TESTS = [
  25         # Single movie
  26         {
  27             'url': 'http://www.ivi.ru/watch/53141',
  28             'md5': '6ff5be2254e796ed346251d117196cf4',
  29             'info_dict': {
  30                 'id': '53141',
  31                 'ext': 'mp4',
  32                 'title': 'Иван Васильевич меняет профессию',
  33                 'description': 'md5:b924063ea1677c8fe343d8a72ac2195f',
  34                 'duration': 5498,
  35                 'thumbnail': r're:^https?://.*\.jpg$',
  36             },
  37             'skip': 'Only works from Russia',
  38         },
  39         # Serial's series
  40         {
  41             'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/9549',
  42             'md5': '221f56b35e3ed815fde2df71032f4b3e',
  43             'info_dict': {
  44                 'id': '9549',
  45                 'ext': 'mp4',
  46                 'title': 'Двое из ларца - Дело Гольдберга (1 часть)',
  47                 'series': 'Двое из ларца',
  48                 'season': 'Сезон 1',
  49                 'season_number': 1,
  50                 'episode': 'Дело Гольдберга (1 часть)',
  51                 'episode_number': 1,
  52                 'duration': 2655,
  53                 'thumbnail': r're:^https?://.*\.jpg$',
  54             },
  55             'skip': 'Only works from Russia',
  56         },
  57         {
  58             # with MP4-HD720 format
  59             'url': 'http://www.ivi.ru/watch/146500',
  60             'md5': 'd63d35cdbfa1ea61a5eafec7cc523e1e',
  61             'info_dict': {
  62                 'id': '146500',
  63                 'ext': 'mp4',
  64                 'title': 'Кукла',
  65                 'description': 'md5:ffca9372399976a2d260a407cc74cce6',
  66                 'duration': 5599,
  67                 'thumbnail': r're:^https?://.*\.jpg$',
  68             },
  69             'skip': 'Only works from Russia',
  70         },
  71         {
  72             'url': 'https://www.ivi.tv/watch/33560/',
  73             'only_matching': True,
  74         },
  75     ]
  76
  77     # Sorted by quality
  78     _KNOWN_FORMATS = (
  79         'MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi',
  80         'MP4-SHQ', 'MP4-HD720', 'MP4-HD1080')
  81
  82     def _real_extract(self, url):
  83         video_id = self._match_id(url)
  84
  85         data = json.dumps({
  86             'method': 'da.content.get',
  87             'params': [
  88                 video_id, {
  89                     'site': 's%d',
  90                     'referrer': 'http://www.ivi.ru/watch/%s' % video_id,
  91                     'contentid': video_id
  92                 }
  93             ]
  94         })
  95
  96         for site in (353, 183):
  97             content_data = (data % site).encode()
  98             if site == 353:
  99                 try:
 100                     from Cryptodome.Cipher import Blowfish
 101                     from Cryptodome.Hash import CMAC
 102                     pycryptodome_found = True
 103                 except ImportError:
 104                     try:
 105                         from Crypto.Cipher import Blowfish
 106                         from Crypto.Hash import CMAC
 107                         pycryptodome_found = True
 108                     except ImportError:
 109                         pycryptodome_found = False
 110                         continue
 111
 112                 timestamp = (self._download_json(
 113                     self._LIGHT_URL, video_id,
 114                     'Downloading timestamp JSON', data=json.dumps({
 115                         'method': 'da.timestamp.get',
 116                         'params': []
 117                     }).encode(), fatal=False) or {}).get('result')
 118                 if not timestamp:
 119                     continue
 120
 121                 query = {
 122                     'ts': timestamp,
 123                     'sign': CMAC.new(self._LIGHT_KEY, timestamp.encode() + content_data, Blowfish).hexdigest(),
 124                 }
 125             else:
 126                 query = {}
 127
 128             video_json = self._download_json(
 129                 self._LIGHT_URL, video_id,
 130                 'Downloading video JSON', data=content_data, query=query)
 131
 132             error = video_json.get('error')
 133             if error:
 134                 origin = error.get('origin')
 135                 message = error.get('message') or error.get('user_message')
 136                 extractor_msg = 'Unable to download video %s'
 137                 if origin == 'NotAllowedForLocation':
 138                     self.raise_geo_restricted(message, self._GEO_COUNTRIES)
 139                 elif origin == 'NoRedisValidData':
 140                     extractor_msg = 'Video %s does not exist'
 141                 elif site == 353:
 142                     continue
 143                 elif not pycryptodome_found:
 144                     raise ExtractorError('pycryptodomex not found. Please install', expected=True)
 145                 elif message:
 146                     extractor_msg += ': ' + message
 147                 raise ExtractorError(extractor_msg % video_id, expected=True)
 148             else:
 149                 break
 150
 151         result = video_json['result']
 152         title = result['title']
 153
 154         quality = qualities(self._KNOWN_FORMATS)
 155
 156         formats = []
 157         for f in result.get('files', []):
 158             f_url = f.get('url')
 159             content_format = f.get('content_format')
 160             if not f_url:
 161                 continue
 162             if (not self.get_param('allow_unplayable_formats')
 163                     and ('-MDRM-' in content_format or '-FPS-' in content_format)):
 164                 continue
 165             formats.append({
 166                 'url': f_url,
 167                 'format_id': content_format,
 168                 'quality': quality(content_format),
 169                 'filesize': int_or_none(f.get('size_in_bytes')),
 170             })
 171         self._sort_formats(formats)
 172
 173         compilation = result.get('compilation')
 174         episode = title if compilation else None
 175
 176         title = '%s - %s' % (compilation, title) if compilation is not None else title
 177
 178         thumbnails = [{
 179             'url': preview['url'],
 180             'id': preview.get('content_format'),
 181         } for preview in result.get('preview', []) if preview.get('url')]
 182
 183         webpage = self._download_webpage(url, video_id)
 184
 185         season = self._search_regex(
 186             r'<li[^>]+class="season active"[^>]*><a[^>]+>([^<]+)',
 187             webpage, 'season', default=None)
 188         season_number = int_or_none(self._search_regex(
 189             r'<li[^>]+class="season active"[^>]*><a[^>]+data-season(?:-index)?="(\d+)"',
 190             webpage, 'season number', default=None))
 191
 192         episode_number = int_or_none(self._search_regex(
 193             r'[^>]+itemprop="episode"[^>]*>\s*<meta[^>]+itemprop="episodeNumber"[^>]+content="(\d+)',
 194             webpage, 'episode number', default=None))
 195
 196         description = self._og_search_description(webpage, default=None) or self._html_search_meta(
 197             'description', webpage, 'description', default=None)
 198
 199         return {
 200             'id': video_id,
 201             'title': title,
 202             'series': compilation,
 203             'season': season,
 204             'season_number': season_number,
 205             'episode': episode,
 206             'episode_number': episode_number,
 207             'thumbnails': thumbnails,
 208             'description': description,
 209             'duration': int_or_none(result.get('duration')),
 210             'formats': formats,
 211         }
 212
 213
 214 class IviCompilationIE(InfoExtractor):
 215     IE_DESC = 'ivi.ru compilations'
 216     IE_NAME = 'ivi:compilation'
 217     _VALID_URL = r'https?://(?:www\.)?ivi\.ru/watch/(?!\d+)(?P<compilationid>[a-z\d_-]+)(?:/season(?P<seasonid>\d+))?$'
 218     _TESTS = [{
 219         'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa',
 220         'info_dict': {
 221             'id': 'dvoe_iz_lartsa',
 222             'title': 'Двое из ларца (2006 - 2008)',
 223         },
 224         'playlist_mincount': 24,
 225     }, {
 226         'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/season1',
 227         'info_dict': {
 228             'id': 'dvoe_iz_lartsa/season1',
 229             'title': 'Двое из ларца (2006 - 2008) 1 сезон',
 230         },
 231         'playlist_mincount': 12,
 232     }]
 233
 234     def _extract_entries(self, html, compilation_id):
 235         return [
 236             self.url_result(
 237                 'http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), IviIE.ie_key())
 238             for serie in re.findall(
 239                 r'<a\b[^>]+\bhref=["\']/watch/%s/(\d+)["\']' % compilation_id, html)]
 240
 241     def _real_extract(self, url):
 242         mobj = self._match_valid_url(url)
 243         compilation_id = mobj.group('compilationid')
 244         season_id = mobj.group('seasonid')
 245
 246         if season_id is not None:  # Season link
 247             season_page = self._download_webpage(
 248                 url, compilation_id, 'Downloading season %s web page' % season_id)
 249             playlist_id = '%s/season%s' % (compilation_id, season_id)
 250             playlist_title = self._html_search_meta('title', season_page, 'title')
 251             entries = self._extract_entries(season_page, compilation_id)
 252         else:  # Compilation link
 253             compilation_page = self._download_webpage(url, compilation_id, 'Downloading compilation web page')
 254             playlist_id = compilation_id
 255             playlist_title = self._html_search_meta('title', compilation_page, 'title')
 256             seasons = re.findall(
 257                 r'<a href="/watch/%s/season(\d+)' % compilation_id, compilation_page)
 258             if not seasons:  # No seasons in this compilation
 259                 entries = self._extract_entries(compilation_page, compilation_id)
 260             else:
 261                 entries = []
 262                 for season_id in seasons:
 263                     season_page = self._download_webpage(
 264                         'http://www.ivi.ru/watch/%s/season%s' % (compilation_id, season_id),
 265                         compilation_id, 'Downloading season %s web page' % season_id)
 266                     entries.extend(self._extract_entries(season_page, compilation_id))
 267
 268         return self.playlist_result(entries, playlist_id, playlist_title)