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 ..utils import (
   9     ExtractorError,
  10     sanitized_Request,
  11 )
  12
  13
  14 class IviIE(InfoExtractor):
  15     IE_DESC = 'ivi.ru'
  16     IE_NAME = 'ivi'
  17     _VALID_URL = r'https?://(?:www\.)?ivi\.ru/(?:watch/(?:[^/]+/)?|video/player\?.*?videoId=)(?P<id>\d+)'
  18
  19     _TESTS = [
  20         # Single movie
  21         {
  22             'url': 'http://www.ivi.ru/watch/53141',
  23             'md5': '6ff5be2254e796ed346251d117196cf4',
  24             'info_dict': {
  25                 'id': '53141',
  26                 'ext': 'mp4',
  27                 'title': 'Иван Васильевич меняет профессию',
  28                 'description': 'md5:b924063ea1677c8fe343d8a72ac2195f',
  29                 'duration': 5498,
  30                 'thumbnail': 'http://thumbs.ivi.ru/f20.vcp.digitalaccess.ru/contents/d/1/c3c885163a082c29bceeb7b5a267a6.jpg',
  31             },
  32             'skip': 'Only works from Russia',
  33         },
  34         # Serial's serie
  35         {
  36             'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/9549',
  37             'md5': '221f56b35e3ed815fde2df71032f4b3e',
  38             'info_dict': {
  39                 'id': '9549',
  40                 'ext': 'mp4',
  41                 'title': 'Двое из ларца - Серия 1',
  42                 'duration': 2655,
  43                 'thumbnail': 'http://thumbs.ivi.ru/f15.vcp.digitalaccess.ru/contents/8/4/0068dc0677041f3336b7c2baad8fc0.jpg',
  44             },
  45             'skip': 'Only works from Russia',
  46         }
  47     ]
  48
  49     # Sorted by quality
  50     _known_formats = ['MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi', 'MP4-SHQ']
  51
  52     # Sorted by size
  53     _known_thumbnails = ['Thumb-120x90', 'Thumb-160', 'Thumb-640x480']
  54
  55     def _extract_description(self, html):
  56         m = re.search(r'<meta name="description" content="(?P<description>[^"]+)"/>', html)
  57         return m.group('description') if m is not None else None
  58
  59     def _extract_comment_count(self, html):
  60         m = re.search('(?s)<a href="#" id="view-comments" class="action-button dim gradient">\s*Комментарии:\s*(?P<commentcount>\d+)\s*</a>', html)
  61         return int(m.group('commentcount')) if m is not None else 0
  62
  63     def _real_extract(self, url):
  64         video_id = self._match_id(url)
  65
  66         api_url = 'http://api.digitalaccess.ru/api/json/'
  67
  68         data = {
  69             'method': 'da.content.get',
  70             'params': [
  71                 video_id, {
  72                     'site': 's183',
  73                     'referrer': 'http://www.ivi.ru/watch/%s' % video_id,
  74                     'contentid': video_id
  75                 }
  76             ]
  77         }
  78
  79         request = sanitized_Request(api_url, json.dumps(data))
  80
  81         video_json_page = self._download_webpage(
  82             request, video_id, 'Downloading video JSON')
  83         video_json = json.loads(video_json_page)
  84
  85         if 'error' in video_json:
  86             error = video_json['error']
  87             if error['origin'] == 'NoRedisValidData':
  88                 raise ExtractorError('Video %s does not exist' % video_id, expected=True)
  89             raise ExtractorError(
  90                 'Unable to download video %s: %s' % (video_id, error['message']),
  91                 expected=True)
  92
  93         result = video_json['result']
  94
  95         formats = [{
  96             'url': x['url'],
  97             'format_id': x['content_format'],
  98             'preference': self._known_formats.index(x['content_format']),
  99         } for x in result['files'] if x['content_format'] in self._known_formats]
 100
 101         self._sort_formats(formats)
 102
 103         if not formats:
 104             raise ExtractorError('No media links available for %s' % video_id)
 105
 106         duration = result['duration']
 107         compilation = result['compilation']
 108         title = result['title']
 109
 110         title = '%s - %s' % (compilation, title) if compilation is not None else title
 111
 112         previews = result['preview']
 113         previews.sort(key=lambda fmt: self._known_thumbnails.index(fmt['content_format']))
 114         thumbnail = previews[-1]['url'] if len(previews) > 0 else None
 115
 116         video_page = self._download_webpage(url, video_id, 'Downloading video page')
 117         description = self._extract_description(video_page)
 118         comment_count = self._extract_comment_count(video_page)
 119
 120         return {
 121             'id': video_id,
 122             'title': title,
 123             'thumbnail': thumbnail,
 124             'description': description,
 125             'duration': duration,
 126             'comment_count': comment_count,
 127             'formats': formats,
 128         }
 129
 130
 131 class IviCompilationIE(InfoExtractor):
 132     IE_DESC = 'ivi.ru compilations'
 133     IE_NAME = 'ivi:compilation'
 134     _VALID_URL = r'https?://(?:www\.)?ivi\.ru/watch/(?!\d+)(?P<compilationid>[a-z\d_-]+)(?:/season(?P<seasonid>\d+))?$'
 135     _TESTS = [{
 136         'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa',
 137         'info_dict': {
 138             'id': 'dvoe_iz_lartsa',
 139             'title': 'Двое из ларца (2006 - 2008)',
 140         },
 141         'playlist_mincount': 24,
 142     }, {
 143         'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/season1',
 144         'info_dict': {
 145             'id': 'dvoe_iz_lartsa/season1',
 146             'title': 'Двое из ларца (2006 - 2008) 1 сезон',
 147         },
 148         'playlist_mincount': 12,
 149     }]
 150
 151     def _extract_entries(self, html, compilation_id):
 152         return [self.url_result('http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), 'Ivi')
 153                 for serie in re.findall(r'<strong><a href="/watch/%s/(\d+)">(?:[^<]+)</a></strong>' % 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(url, compilation_id, 'Downloading season %s web page' % season_id)
 162             playlist_id = '%s/season%s' % (compilation_id, season_id)
 163             playlist_title = self._html_search_meta('title', season_page, 'title')
 164             entries = self._extract_entries(season_page, compilation_id)
 165         else:  # Compilation link
 166             compilation_page = self._download_webpage(url, compilation_id, 'Downloading compilation web page')
 167             playlist_id = compilation_id
 168             playlist_title = self._html_search_meta('title', compilation_page, 'title')
 169             seasons = re.findall(r'<a href="/watch/%s/season(\d+)">[^<]+</a>' % compilation_id, compilation_page)
 170             if len(seasons) == 0:  # No seasons in this compilation
 171                 entries = self._extract_entries(compilation_page, compilation_id)
 172             else:
 173                 entries = []
 174                 for season_id in seasons:
 175                     season_page = self._download_webpage(
 176                         'http://www.ivi.ru/watch/%s/season%s' % (compilation_id, season_id),
 177                         compilation_id, 'Downloading season %s web page' % season_id)
 178                     entries.extend(self._extract_entries(season_page, compilation_id))
 179
 180         return self.playlist_result(entries, playlist_id, playlist_title)