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     compat_urllib_request,
  10     ExtractorError,
  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<videoid>\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         mobj = re.match(self._VALID_URL, url)
  65         video_id = mobj.group('videoid')
  66
  67         api_url = 'http://api.digitalaccess.ru/api/json/'
  68
  69         data = {'method': 'da.content.get',
  70                 'params': [video_id, {'site': 's183',
  71                                       'referrer': 'http://www.ivi.ru/watch/%s' % video_id,
  72                                       'contentid': video_id
  73                                       }
  74                            ]
  75                 }
  76
  77         request = compat_urllib_request.Request(api_url, json.dumps(data))
  78
  79         video_json_page = self._download_webpage(request, video_id, 'Downloading video JSON')
  80         video_json = json.loads(video_json_page)
  81
  82         if 'error' in video_json:
  83             error = video_json['error']
  84             if error['origin'] == 'NoRedisValidData':
  85                 raise ExtractorError('Video %s does not exist' % video_id, expected=True)
  86             raise ExtractorError('Unable to download video %s: %s' % (video_id, error['message']), expected=True)
  87
  88         result = video_json['result']
  89
  90         formats = [{
  91             'url': x['url'],
  92             'format_id': x['content_format'],
  93             'preference': self._known_formats.index(x['content_format']),
  94         } for x in result['files'] if x['content_format'] in self._known_formats]
  95
  96         self._sort_formats(formats)
  97
  98         if not formats:
  99             raise ExtractorError('No media links available for %s' % video_id)
 100
 101         duration = result['duration']
 102         compilation = result['compilation']
 103         title = result['title']
 104
 105         title = '%s - %s' % (compilation, title) if compilation is not None else title
 106
 107         previews = result['preview']
 108         previews.sort(key=lambda fmt: self._known_thumbnails.index(fmt['content_format']))
 109         thumbnail = previews[-1]['url'] if len(previews) > 0 else None
 110
 111         video_page = self._download_webpage(url, video_id, 'Downloading video page')
 112         description = self._extract_description(video_page)
 113         comment_count = self._extract_comment_count(video_page)
 114
 115         return {
 116             'id': video_id,
 117             'title': title,
 118             'thumbnail': thumbnail,
 119             'description': description,
 120             'duration': duration,
 121             'comment_count': comment_count,
 122             'formats': formats,
 123         }
 124
 125
 126 class IviCompilationIE(InfoExtractor):
 127     IE_DESC = 'ivi.ru compilations'
 128     IE_NAME = 'ivi:compilation'
 129     _VALID_URL = r'https?://(?:www\.)?ivi\.ru/watch/(?!\d+)(?P<compilationid>[a-z\d_-]+)(?:/season(?P<seasonid>\d+))?$'
 130
 131     def _extract_entries(self, html, compilation_id):
 132         return [self.url_result('http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), 'Ivi')
 133                 for serie in re.findall(r'<strong><a href="/watch/%s/(\d+)">(?:[^<]+)</a></strong>' % compilation_id, html)]
 134
 135     def _real_extract(self, url):
 136         mobj = re.match(self._VALID_URL, url)
 137         compilation_id = mobj.group('compilationid')
 138         season_id = mobj.group('seasonid')
 139
 140         if season_id is not None: # Season link
 141             season_page = self._download_webpage(url, compilation_id, 'Downloading season %s web page' % season_id)
 142             playlist_id = '%s/season%s' % (compilation_id, season_id)
 143             playlist_title = self._html_search_meta('title', season_page, 'title')
 144             entries = self._extract_entries(season_page, compilation_id)
 145         else: # Compilation link
 146             compilation_page = self._download_webpage(url, compilation_id, 'Downloading compilation web page')
 147             playlist_id = compilation_id
 148             playlist_title = self._html_search_meta('title', compilation_page, 'title')
 149             seasons = re.findall(r'<a href="/watch/%s/season(\d+)">[^<]+</a>' % compilation_id, compilation_page)
 150             if len(seasons) == 0: # No seasons in this compilation
 151                 entries = self._extract_entries(compilation_page, compilation_id)
 152             else:
 153                 entries = []
 154                 for season_id in seasons:
 155                     season_page = self._download_webpage(
 156                         'http://www.ivi.ru/watch/%s/season%s' % (compilation_id, season_id),
 157                         compilation_id, 'Downloading season %s web page' % season_id)
 158                     entries.extend(self._extract_entries(season_page, compilation_id))
 159
 160         return self.playlist_result(entries, playlist_id, playlist_title)