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

   1 # encoding: 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     sanitized_Request,
  12 )
  13
  14
  15 class IviIE(InfoExtractor):
  16     IE_DESC = 'ivi.ru'
  17     IE_NAME = 'ivi'
  18     _VALID_URL = r'https?://(?:www\.)?ivi\.ru/(?:watch/(?:[^/]+/)?|video/player\?.*?videoId=)(?P<id>\d+)'
  19
  20     _TESTS = [
  21         # Single movie
  22         {
  23             'url': 'http://www.ivi.ru/watch/53141',
  24             'md5': '6ff5be2254e796ed346251d117196cf4',
  25             'info_dict': {
  26                 'id': '53141',
  27                 'ext': 'mp4',
  28                 'title': 'Иван Васильевич меняет профессию',
  29                 'description': 'md5:b924063ea1677c8fe343d8a72ac2195f',
  30                 'duration': 5498,
  31                 'thumbnail': 're:^https?://.*\.jpg$',
  32             },
  33             'skip': 'Only works from Russia',
  34         },
  35         # Serial's serie
  36         {
  37             'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/9549',
  38             'md5': '221f56b35e3ed815fde2df71032f4b3e',
  39             'info_dict': {
  40                 'id': '9549',
  41                 'ext': 'mp4',
  42                 'title': 'Двое из ларца - Дело Гольдберга (1 часть)',
  43                 'series': 'Двое из ларца',
  44                 'episode': 'Дело Гольдберга (1 часть)',
  45                 'episode_number': 1,
  46                 'duration': 2655,
  47                 'thumbnail': 're:^https?://.*\.jpg$',
  48             },
  49             'skip': 'Only works from Russia',
  50         }
  51     ]
  52
  53     # Sorted by quality
  54     _KNOWN_FORMATS = ['MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi', 'MP4-SHQ']
  55
  56     def _real_extract(self, url):
  57         video_id = self._match_id(url)
  58
  59         data = {
  60             'method': 'da.content.get',
  61             'params': [
  62                 video_id, {
  63                     'site': 's183',
  64                     'referrer': 'http://www.ivi.ru/watch/%s' % video_id,
  65                     'contentid': video_id
  66                 }
  67             ]
  68         }
  69
  70         request = sanitized_Request(
  71             'http://api.digitalaccess.ru/api/json/', json.dumps(data))
  72         video_json = self._download_json(
  73             request, video_id, 'Downloading video JSON')
  74
  75         if 'error' in video_json:
  76             error = video_json['error']
  77             if error['origin'] == 'NoRedisValidData':
  78                 raise ExtractorError('Video %s does not exist' % video_id, expected=True)
  79             raise ExtractorError(
  80                 'Unable to download video %s: %s' % (video_id, error['message']),
  81                 expected=True)
  82
  83         result = video_json['result']
  84
  85         formats = [{
  86             'url': x['url'],
  87             'format_id': x['content_format'],
  88             'preference': self._KNOWN_FORMATS.index(x['content_format']),
  89         } for x in result['files'] if x['content_format'] in self._KNOWN_FORMATS]
  90
  91         self._sort_formats(formats)
  92
  93         title = result['title']
  94
  95         duration = int_or_none(result.get('duration'))
  96         compilation = result.get('compilation')
  97         episode = title if compilation else None
  98
  99         title = '%s - %s' % (compilation, title) if compilation is not None else title
 100
 101         thumbnails = [{
 102             'url': preview['url'],
 103             'id': preview.get('content_format'),
 104         } for preview in result.get('preview', []) if preview.get('url')]
 105
 106         webpage = self._download_webpage(url, video_id)
 107
 108         episode_number = int_or_none(self._search_regex(
 109             r'<meta[^>]+itemprop="episode"[^>]*>\s*<meta[^>]+itemprop="episodeNumber"[^>]+content="(\d+)',
 110             webpage, 'episode number', default=None))
 111
 112         description = self._og_search_description(webpage, default=None) or self._html_search_meta(
 113             'description', webpage, 'description', default=None)
 114
 115         return {
 116             'id': video_id,
 117             'title': title,
 118             'series': compilation,
 119             'episode': episode,
 120             'episode_number': episode_number,
 121             'thumbnails': thumbnails,
 122             'description': description,
 123             'duration': duration,
 124             'formats': formats,
 125         }
 126
 127
 128 class IviCompilationIE(InfoExtractor):
 129     IE_DESC = 'ivi.ru compilations'
 130     IE_NAME = 'ivi:compilation'
 131     _VALID_URL = r'https?://(?:www\.)?ivi\.ru/watch/(?!\d+)(?P<compilationid>[a-z\d_-]+)(?:/season(?P<seasonid>\d+))?$'
 132     _TESTS = [{
 133         'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa',
 134         'info_dict': {
 135             'id': 'dvoe_iz_lartsa',
 136             'title': 'Двое из ларца (2006 - 2008)',
 137         },
 138         'playlist_mincount': 24,
 139     }, {
 140         'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/season1',
 141         'info_dict': {
 142             'id': 'dvoe_iz_lartsa/season1',
 143             'title': 'Двое из ларца (2006 - 2008) 1 сезон',
 144         },
 145         'playlist_mincount': 12,
 146     }]
 147
 148     def _extract_entries(self, html, compilation_id):
 149         return [
 150             self.url_result(
 151                 'http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), IviIE.ie_key())
 152             for serie in re.findall(
 153                 r'<a href="/watch/%s/(\d+)"[^>]+data-id="\1"' % compilation_id, html)]
 154
 155     def _real_extract(self, url):
 156         mobj = re.match(self._VALID_URL, url)
 157         compilation_id = mobj.group('compilationid')
 158         season_id = mobj.group('seasonid')
 159
 160         if season_id is not None:  # Season link
 161             season_page = self._download_webpage(
 162                 url, compilation_id, 'Downloading season %s web page' % season_id)
 163             playlist_id = '%s/season%s' % (compilation_id, season_id)
 164             playlist_title = self._html_search_meta('title', season_page, 'title')
 165             entries = self._extract_entries(season_page, compilation_id)
 166         else:  # Compilation link
 167             compilation_page = self._download_webpage(url, compilation_id, 'Downloading compilation web page')
 168             playlist_id = compilation_id
 169             playlist_title = self._html_search_meta('title', compilation_page, 'title')
 170             seasons = re.findall(
 171                 r'<a href="/watch/%s/season(\d+)' % compilation_id, compilation_page)
 172             if not seasons:  # No seasons in this compilation
 173                 entries = self._extract_entries(compilation_page, compilation_id)
 174             else:
 175                 entries = []
 176                 for season_id in seasons:
 177                     season_page = self._download_webpage(
 178                         'http://www.ivi.ru/watch/%s/season%s' % (compilation_id, season_id),
 179                         compilation_id, 'Downloading season %s web page' % season_id)
 180                     entries.extend(self._extract_entries(season_page, compilation_id))
 181
 182         return self.playlist_result(entries, playlist_id, playlist_title)