Raphaël G. Git Repositories - youtubedl/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 ..compat import (
   9     compat_urllib_request,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13 )
  14
  15
  16 class IviIE(InfoExtractor):
  17     IE_DESC = 'ivi.ru'
  18     IE_NAME = 'ivi'
  19     _VALID_URL = r'https?://(?:www\.)?ivi\.ru/(?:watch/(?:[^/]+/)?|video/player\?.*?videoId=)(?P<id>\d+)'
  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': 'http://thumbs.ivi.ru/f20.vcp.digitalaccess.ru/contents/d/1/c3c885163a082c29bceeb7b5a267a6.jpg',
  33             },
  34             'skip': 'Only works from Russia',
  35         },
  36         # Serial's serie
  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 - Серия 1',
  44                 'duration': 2655,
  45                 'thumbnail': 'http://thumbs.ivi.ru/f15.vcp.digitalaccess.ru/contents/8/4/0068dc0677041f3336b7c2baad8fc0.jpg',
  46             },
  47             'skip': 'Only works from Russia',
  48         }
  49     ]
  50
  51     # Sorted by quality
  52     _known_formats = ['MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi', 'MP4-SHQ']
  53
  54     # Sorted by size
  55     _known_thumbnails = ['Thumb-120x90', 'Thumb-160', 'Thumb-640x480']
  56
  57     def _extract_description(self, html):
  58         m = re.search(r'<meta name="description" content="(?P<description>[^"]+)"/>', html)
  59         return m.group('description') if m is not None else None
  60
  61     def _extract_comment_count(self, html):
  62         m = re.search('(?s)<a href="#" id="view-comments" class="action-button dim gradient">\s*Комментарии:\s*(?P<commentcount>\d+)\s*</a>', html)
  63         return int(m.group('commentcount')) if m is not None else 0
  64
  65     def _real_extract(self, url):
  66         video_id = self._match_id(url)
  67
  68         api_url = 'http://api.digitalaccess.ru/api/json/'
  69
  70         data = {
  71             'method': 'da.content.get',
  72             'params': [
  73                 video_id, {
  74                     'site': 's183',
  75                     'referrer': 'http://www.ivi.ru/watch/%s' % video_id,
  76                     'contentid': video_id
  77                 }
  78             ]
  79         }
  80
  81         request = compat_urllib_request.Request(api_url, json.dumps(data))
  82
  83         video_json_page = self._download_webpage(
  84             request, video_id, 'Downloading video JSON')
  85         video_json = json.loads(video_json_page)
  86
  87         if 'error' in video_json:
  88             error = video_json['error']
  89             if error['origin'] == 'NoRedisValidData':
  90                 raise ExtractorError('Video %s does not exist' % video_id, expected=True)
  91             raise ExtractorError(
  92                 'Unable to download video %s: %s' % (video_id, error['message']),
  93                 expected=True)
  94
  95         result = video_json['result']
  96
  97         formats = [{
  98             'url': x['url'],
  99             'format_id': x['content_format'],
 100             'preference': self._known_formats.index(x['content_format']),
 101         } for x in result['files'] if x['content_format'] in self._known_formats]
 102
 103         self._sort_formats(formats)
 104
 105         if not formats:
 106             raise ExtractorError('No media links available for %s' % video_id)
 107
 108         duration = result['duration']
 109         compilation = result['compilation']
 110         title = result['title']
 111
 112         title = '%s - %s' % (compilation, title) if compilation is not None else title
 113
 114         previews = result['preview']
 115         previews.sort(key=lambda fmt: self._known_thumbnails.index(fmt['content_format']))
 116         thumbnail = previews[-1]['url'] if len(previews) > 0 else None
 117
 118         video_page = self._download_webpage(url, video_id, 'Downloading video page')
 119         description = self._extract_description(video_page)
 120         comment_count = self._extract_comment_count(video_page)
 121
 122         return {
 123             'id': video_id,
 124             'title': title,
 125             'thumbnail': thumbnail,
 126             'description': description,
 127             'duration': duration,
 128             'comment_count': comment_count,
 129             'formats': formats,
 130         }
 131
 132
 133 class IviCompilationIE(InfoExtractor):
 134     IE_DESC = 'ivi.ru compilations'
 135     IE_NAME = 'ivi:compilation'
 136     _VALID_URL = r'https?://(?:www\.)?ivi\.ru/watch/(?!\d+)(?P<compilationid>[a-z\d_-]+)(?:/season(?P<seasonid>\d+))?$'
 137     _TESTS = [{
 138         'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa',
 139         'info_dict': {
 140             'id': 'dvoe_iz_lartsa',
 141             'title': 'Двое из ларца (2006 - 2008)',
 142         },
 143         'playlist_mincount': 24,
 144     }, {
 145         'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/season1',
 146         'info_dict': {
 147             'id': 'dvoe_iz_lartsa/season1',
 148             'title': 'Двое из ларца (2006 - 2008) 1 сезон',
 149         },
 150         'playlist_mincount': 12,
 151     }]
 152
 153     def _extract_entries(self, html, compilation_id):
 154         return [self.url_result('http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), 'Ivi')
 155                 for serie in re.findall(r'<strong><a href="/watch/%s/(\d+)">(?:[^<]+)</a></strong>' % compilation_id, html)]
 156
 157     def _real_extract(self, url):
 158         mobj = re.match(self._VALID_URL, url)
 159         compilation_id = mobj.group('compilationid')
 160         season_id = mobj.group('seasonid')
 161
 162         if season_id is not None:  # Season link
 163             season_page = self._download_webpage(url, compilation_id, 'Downloading season %s web page' % season_id)
 164             playlist_id = '%s/season%s' % (compilation_id, season_id)
 165             playlist_title = self._html_search_meta('title', season_page, 'title')
 166             entries = self._extract_entries(season_page, compilation_id)
 167         else:  # Compilation link
 168             compilation_page = self._download_webpage(url, compilation_id, 'Downloading compilation web page')
 169             playlist_id = compilation_id
 170             playlist_title = self._html_search_meta('title', compilation_page, 'title')
 171             seasons = re.findall(r'<a href="/watch/%s/season(\d+)">[^<]+</a>' % compilation_id, compilation_page)
 172             if len(seasons) == 0:  # 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)