_ Git - youtube-dl/blob - youtube_dl/extractor/ivi.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import json
   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         try:
  84             from Crypto.Cipher import Blowfish
  85             from Crypto.Hash import CMAC
  86         except ImportError:
  87             raise ExtractorError('pycrypto not found. Please install it.', expected=True)
  88
  89         video_id = self._match_id(url)
  90
  91         timestamp = self._download_json(
  92             self._LIGHT_URL, video_id,
  93             'Downloading timestamp JSON', data=json.dumps({
  94                 'method': 'da.timestamp.get',
  95                 'params': []
  96             }).encode())['result']
  97
  98         data = json.dumps({
  99             'method': 'da.content.get',
 100             'params': [
 101                 video_id, {
 102                     'site': 's353',
 103                     'referrer': 'http://www.ivi.ru/watch/%s' % video_id,
 104                     'contentid': video_id
 105                 }
 106             ]
 107         }).encode()
 108
 109         video_json = self._download_json(
 110             self._LIGHT_URL, video_id,
 111             'Downloading video JSON', data=data, query={
 112                 'ts': timestamp,
 113                 'sign': CMAC.new(self._LIGHT_KEY, timestamp.encode() + data, Blowfish).hexdigest(),
 114             })
 115
 116         error = video_json.get('error')
 117         if error:
 118             origin = error['origin']
 119             if origin == 'NotAllowedForLocation':
 120                 self.raise_geo_restricted(
 121                     msg=error['message'], countries=self._GEO_COUNTRIES)
 122             elif origin == 'NoRedisValidData':
 123                 raise ExtractorError('Video %s does not exist' % video_id, expected=True)
 124             raise ExtractorError(
 125                 'Unable to download video %s: %s' % (video_id, error['message']),
 126                 expected=True)
 127
 128         result = video_json['result']
 129         title = result['title']
 130
 131         quality = qualities(self._KNOWN_FORMATS)
 132
 133         formats = []
 134         for f in result.get('files', []):
 135             f_url = f.get('url')
 136             content_format = f.get('content_format')
 137             if not f_url or '-MDRM-' in content_format or '-FPS-' in content_format:
 138                 continue
 139             formats.append({
 140                 'url': f_url,
 141                 'format_id': content_format,
 142                 'quality': quality(content_format),
 143                 'filesize': int_or_none(f.get('size_in_bytes')),
 144             })
 145         self._sort_formats(formats)
 146
 147         compilation = result.get('compilation')
 148         episode = title if compilation else None
 149
 150         title = '%s - %s' % (compilation, title) if compilation is not None else title
 151
 152         thumbnails = [{
 153             'url': preview['url'],
 154             'id': preview.get('content_format'),
 155         } for preview in result.get('preview', []) if preview.get('url')]
 156
 157         webpage = self._download_webpage(url, video_id)
 158
 159         season = self._search_regex(
 160             r'<li[^>]+class="season active"[^>]*><a[^>]+>([^<]+)',
 161             webpage, 'season', default=None)
 162         season_number = int_or_none(self._search_regex(
 163             r'<li[^>]+class="season active"[^>]*><a[^>]+data-season(?:-index)?="(\d+)"',
 164             webpage, 'season number', default=None))
 165
 166         episode_number = int_or_none(self._search_regex(
 167             r'[^>]+itemprop="episode"[^>]*>\s*<meta[^>]+itemprop="episodeNumber"[^>]+content="(\d+)',
 168             webpage, 'episode number', default=None))
 169
 170         description = self._og_search_description(webpage, default=None) or self._html_search_meta(
 171             'description', webpage, 'description', default=None)
 172
 173         return {
 174             'id': video_id,
 175             'title': title,
 176             'series': compilation,
 177             'season': season,
 178             'season_number': season_number,
 179             'episode': episode,
 180             'episode_number': episode_number,
 181             'thumbnails': thumbnails,
 182             'description': description,
 183             'duration': int_or_none(result.get('duration')),
 184             'formats': formats,
 185         }
 186
 187
 188 class IviCompilationIE(InfoExtractor):
 189     IE_DESC = 'ivi.ru compilations'
 190     IE_NAME = 'ivi:compilation'
 191     _VALID_URL = r'https?://(?:www\.)?ivi\.ru/watch/(?!\d+)(?P<compilationid>[a-z\d_-]+)(?:/season(?P<seasonid>\d+))?$'
 192     _TESTS = [{
 193         'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa',
 194         'info_dict': {
 195             'id': 'dvoe_iz_lartsa',
 196             'title': 'Двое из ларца (2006 - 2008)',
 197         },
 198         'playlist_mincount': 24,
 199     }, {
 200         'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/season1',
 201         'info_dict': {
 202             'id': 'dvoe_iz_lartsa/season1',
 203             'title': 'Двое из ларца (2006 - 2008) 1 сезон',
 204         },
 205         'playlist_mincount': 12,
 206     }]
 207
 208     def _extract_entries(self, html, compilation_id):
 209         return [
 210             self.url_result(
 211                 'http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), IviIE.ie_key())
 212             for serie in re.findall(
 213                 r'<a href="/watch/%s/(\d+)"[^>]+data-id="\1"' % compilation_id, html)]
 214
 215     def _real_extract(self, url):
 216         mobj = re.match(self._VALID_URL, url)
 217         compilation_id = mobj.group('compilationid')
 218         season_id = mobj.group('seasonid')
 219
 220         if season_id is not None:  # Season link
 221             season_page = self._download_webpage(
 222                 url, compilation_id, 'Downloading season %s web page' % season_id)
 223             playlist_id = '%s/season%s' % (compilation_id, season_id)
 224             playlist_title = self._html_search_meta('title', season_page, 'title')
 225             entries = self._extract_entries(season_page, compilation_id)
 226         else:  # Compilation link
 227             compilation_page = self._download_webpage(url, compilation_id, 'Downloading compilation web page')
 228             playlist_id = compilation_id
 229             playlist_title = self._html_search_meta('title', compilation_page, 'title')
 230             seasons = re.findall(
 231                 r'<a href="/watch/%s/season(\d+)' % compilation_id, compilation_page)
 232             if not seasons:  # No seasons in this compilation
 233                 entries = self._extract_entries(compilation_page, compilation_id)
 234             else:
 235                 entries = []
 236                 for season_id in seasons:
 237                     season_page = self._download_webpage(
 238                         'http://www.ivi.ru/watch/%s/season%s' % (compilation_id, season_id),
 239                         compilation_id, 'Downloading season %s web page' % season_id)
 240                     entries.extend(self._extract_entries(season_page, compilation_id))
 241
 242         return self.playlist_result(entries, playlist_id, playlist_title)