Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/ivi.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6 import sys
   7
   8 from .common import InfoExtractor
   9 from ..utils import (
  10     ExtractorError,
  11     int_or_none,
  12     qualities,
  13 )
  14
  15
  16 class IviIE(InfoExtractor):
  17     IE_DESC = 'ivi.ru'
  18     IE_NAME = 'ivi'
  19     _VALID_URL = r'https?://(?:www\.)?ivi\.(?:ru|tv)/(?:watch/(?:[^/]+/)?|video/player\?.*?videoId=)(?P<id>\d+)'
  20     _GEO_BYPASS = False
  21     _GEO_COUNTRIES = ['RU']
  22     _LIGHT_KEY = b'\xf1\x02\x32\xb7\xbc\x5c\x7a\xe8\xf7\x96\xc1\x33\x2b\x27\xa1\x8c'
  23     _LIGHT_URL = 'https://api.ivi.ru/light/'
  24
  25     _TESTS = [
  26         # Single movie
  27         {
  28             'url': 'http://www.ivi.ru/watch/53141',
  29             'md5': '6ff5be2254e796ed346251d117196cf4',
  30             'info_dict': {
  31                 'id': '53141',
  32                 'ext': 'mp4',
  33                 'title': 'Иван Васильевич меняет профессию',
  34                 'description': 'md5:b924063ea1677c8fe343d8a72ac2195f',
  35                 'duration': 5498,
  36                 'thumbnail': r're:^https?://.*\.jpg$',
  37             },
  38             'skip': 'Only works from Russia',
  39         },
  40         # Serial's series
  41         {
  42             'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/9549',
  43             'md5': '221f56b35e3ed815fde2df71032f4b3e',
  44             'info_dict': {
  45                 'id': '9549',
  46                 'ext': 'mp4',
  47                 'title': 'Двое из ларца - Дело Гольдберга (1 часть)',
  48                 'series': 'Двое из ларца',
  49                 'season': 'Сезон 1',
  50                 'season_number': 1,
  51                 'episode': 'Дело Гольдберга (1 часть)',
  52                 'episode_number': 1,
  53                 'duration': 2655,
  54                 'thumbnail': r're:^https?://.*\.jpg$',
  55             },
  56             'skip': 'Only works from Russia',
  57         },
  58         {
  59             # with MP4-HD720 format
  60             'url': 'http://www.ivi.ru/watch/146500',
  61             'md5': 'd63d35cdbfa1ea61a5eafec7cc523e1e',
  62             'info_dict': {
  63                 'id': '146500',
  64                 'ext': 'mp4',
  65                 'title': 'Кукла',
  66                 'description': 'md5:ffca9372399976a2d260a407cc74cce6',
  67                 'duration': 5599,
  68                 'thumbnail': r're:^https?://.*\.jpg$',
  69             },
  70             'skip': 'Only works from Russia',
  71         },
  72         {
  73             'url': 'https://www.ivi.tv/watch/33560/',
  74             'only_matching': True,
  75         },
  76     ]
  77
  78     # Sorted by quality
  79     _KNOWN_FORMATS = (
  80         'MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi',
  81         'MP4-SHQ', 'MP4-HD720', 'MP4-HD1080')
  82
  83     def _real_extract(self, url):
  84         video_id = self._match_id(url)
  85
  86         data = json.dumps({
  87             'method': 'da.content.get',
  88             'params': [
  89                 video_id, {
  90                     'site': 's%d',
  91                     'referrer': 'http://www.ivi.ru/watch/%s' % video_id,
  92                     'contentid': video_id
  93                 }
  94             ]
  95         })
  96
  97         bundled = hasattr(sys, 'frozen')
  98
  99         for site in (353, 183):
 100             content_data = (data % site).encode()
 101             if site == 353:
 102                 if bundled:
 103                     continue
 104                 try:
 105                     from Cryptodome.Cipher import Blowfish
 106                     from Cryptodome.Hash import CMAC
 107                     pycryptodomex_found = True
 108                 except ImportError:
 109                     pycryptodomex_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 bundled:
 144                     raise ExtractorError(
 145                         'This feature does not work from bundled exe. Run youtube-dl from sources.',
 146                         expected=True)
 147                 elif not pycryptodomex_found:
 148                     raise ExtractorError(
 149                         'pycryptodomex not found. Please install it.',
 150                         expected=True)
 151                 elif message:
 152                     extractor_msg += ': ' + message
 153                 raise ExtractorError(extractor_msg % video_id, expected=True)
 154             else:
 155                 break
 156
 157         result = video_json['result']
 158         title = result['title']
 159
 160         quality = qualities(self._KNOWN_FORMATS)
 161
 162         formats = []
 163         for f in result.get('files', []):
 164             f_url = f.get('url')
 165             content_format = f.get('content_format')
 166             if not f_url or '-MDRM-' in content_format or '-FPS-' in content_format:
 167                 continue
 168             formats.append({
 169                 'url': f_url,
 170                 'format_id': content_format,
 171                 'quality': quality(content_format),
 172                 'filesize': int_or_none(f.get('size_in_bytes')),
 173             })
 174         self._sort_formats(formats)
 175
 176         compilation = result.get('compilation')
 177         episode = title if compilation else None
 178
 179         title = '%s - %s' % (compilation, title) if compilation is not None else title
 180
 181         thumbnails = [{
 182             'url': preview['url'],
 183             'id': preview.get('content_format'),
 184         } for preview in result.get('preview', []) if preview.get('url')]
 185
 186         webpage = self._download_webpage(url, video_id)
 187
 188         season = self._search_regex(
 189             r'<li[^>]+class="season active"[^>]*><a[^>]+>([^<]+)',
 190             webpage, 'season', default=None)
 191         season_number = int_or_none(self._search_regex(
 192             r'<li[^>]+class="season active"[^>]*><a[^>]+data-season(?:-index)?="(\d+)"',
 193             webpage, 'season number', default=None))
 194
 195         episode_number = int_or_none(self._search_regex(
 196             r'[^>]+itemprop="episode"[^>]*>\s*<meta[^>]+itemprop="episodeNumber"[^>]+content="(\d+)',
 197             webpage, 'episode number', default=None))
 198
 199         description = self._og_search_description(webpage, default=None) or self._html_search_meta(
 200             'description', webpage, 'description', default=None)
 201
 202         return {
 203             'id': video_id,
 204             'title': title,
 205             'series': compilation,
 206             'season': season,
 207             'season_number': season_number,
 208             'episode': episode,
 209             'episode_number': episode_number,
 210             'thumbnails': thumbnails,
 211             'description': description,
 212             'duration': int_or_none(result.get('duration')),
 213             'formats': formats,
 214         }
 215
 216
 217 class IviCompilationIE(InfoExtractor):
 218     IE_DESC = 'ivi.ru compilations'
 219     IE_NAME = 'ivi:compilation'
 220     _VALID_URL = r'https?://(?:www\.)?ivi\.ru/watch/(?!\d+)(?P<compilationid>[a-z\d_-]+)(?:/season(?P<seasonid>\d+))?$'
 221     _TESTS = [{
 222         'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa',
 223         'info_dict': {
 224             'id': 'dvoe_iz_lartsa',
 225             'title': 'Двое из ларца (2006 - 2008)',
 226         },
 227         'playlist_mincount': 24,
 228     }, {
 229         'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/season1',
 230         'info_dict': {
 231             'id': 'dvoe_iz_lartsa/season1',
 232             'title': 'Двое из ларца (2006 - 2008) 1 сезон',
 233         },
 234         'playlist_mincount': 12,
 235     }]
 236
 237     def _extract_entries(self, html, compilation_id):
 238         return [
 239             self.url_result(
 240                 'http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), IviIE.ie_key())
 241             for serie in re.findall(
 242                 r'<a\b[^>]+\bhref=["\']/watch/%s/(\d+)["\']' % compilation_id, html)]
 243
 244     def _real_extract(self, url):
 245         mobj = re.match(self._VALID_URL, url)
 246         compilation_id = mobj.group('compilationid')
 247         season_id = mobj.group('seasonid')
 248
 249         if season_id is not None:  # Season link
 250             season_page = self._download_webpage(
 251                 url, compilation_id, 'Downloading season %s web page' % season_id)
 252             playlist_id = '%s/season%s' % (compilation_id, season_id)
 253             playlist_title = self._html_search_meta('title', season_page, 'title')
 254             entries = self._extract_entries(season_page, compilation_id)
 255         else:  # Compilation link
 256             compilation_page = self._download_webpage(url, compilation_id, 'Downloading compilation web page')
 257             playlist_id = compilation_id
 258             playlist_title = self._html_search_meta('title', compilation_page, 'title')
 259             seasons = re.findall(
 260                 r'<a href="/watch/%s/season(\d+)' % compilation_id, compilation_page)
 261             if not seasons:  # No seasons in this compilation
 262                 entries = self._extract_entries(compilation_page, compilation_id)
 263             else:
 264                 entries = []
 265                 for season_id in seasons:
 266                     season_page = self._download_webpage(
 267                         'http://www.ivi.ru/watch/%s/season%s' % (compilation_id, season_id),
 268                         compilation_id, 'Downloading season %s web page' % season_id)
 269                     entries.extend(self._extract_entries(season_page, compilation_id))
 270
 271         return self.playlist_result(entries, playlist_id, playlist_title)