_ 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     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(?:/(?P<compilationid>[^/]+))?/(?P<videoid>\d+)'
  18
  19     _TESTS = [
  20         # Single movie
  21         {
  22             'url': 'http://www.ivi.ru/watch/53141',
  23             'file': '53141.mp4',
  24             'md5': '6ff5be2254e796ed346251d117196cf4',
  25             'info_dict': {
  26                 'title': 'Иван Васильевич меняет профессию',
  27                 'description': 'md5:b924063ea1677c8fe343d8a72ac2195f',
  28                 'duration': 5498,
  29                 'thumbnail': 'http://thumbs.ivi.ru/f20.vcp.digitalaccess.ru/contents/d/1/c3c885163a082c29bceeb7b5a267a6.jpg',
  30             },
  31             'skip': 'Only works from Russia',
  32         },
  33         # Serial's serie
  34         {
  35             'url': 'http://www.ivi.ru/watch/dezhurnyi_angel/74791',
  36             'file': '74791.mp4',
  37             'md5': '3e6cc9a848c1d2ebcc6476444967baa9',
  38             'info_dict': {
  39                 'title': 'Дежурный ангел - 1 серия',
  40                 'duration': 2490,
  41                 'thumbnail': 'http://thumbs.ivi.ru/f7.vcp.digitalaccess.ru/contents/8/e/bc2f6c2b6e5d291152fdd32c059141.jpg',
  42             },
  43             'skip': 'Only works from Russia',
  44          }
  45     ]
  46
  47     # Sorted by quality
  48     _known_formats = ['MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi', 'MP4-SHQ']
  49
  50     # Sorted by size
  51     _known_thumbnails = ['Thumb-120x90', 'Thumb-160', 'Thumb-640x480']
  52
  53     def _extract_description(self, html):
  54         m = re.search(r'<meta name="description" content="(?P<description>[^"]+)"/>', html)
  55         return m.group('description') if m is not None else None
  56
  57     def _extract_comment_count(self, html):
  58         m = re.search('(?s)<a href="#" id="view-comments" class="action-button dim gradient">\s*Комментарии:\s*(?P<commentcount>\d+)\s*</a>', html)
  59         return int(m.group('commentcount')) if m is not None else 0
  60
  61     def _real_extract(self, url):
  62         mobj = re.match(self._VALID_URL, url)
  63         video_id = mobj.group('videoid')
  64
  65         api_url = 'http://api.digitalaccess.ru/api/json/'
  66
  67         data = {'method': 'da.content.get',
  68                 'params': [video_id, {'site': 's183',
  69                                       'referrer': 'http://www.ivi.ru/watch/%s' % video_id,
  70                                       'contentid': video_id
  71                                       }
  72                            ]
  73                 }
  74
  75         request = compat_urllib_request.Request(api_url, json.dumps(data))
  76
  77         video_json_page = self._download_webpage(request, video_id, 'Downloading video JSON')
  78         video_json = json.loads(video_json_page)
  79
  80         if 'error' in video_json:
  81             error = video_json['error']
  82             if error['origin'] == 'NoRedisValidData':
  83                 raise ExtractorError('Video %s does not exist' % video_id, expected=True)
  84             raise ExtractorError('Unable to download video %s: %s' % (video_id, error['message']), expected=True)
  85
  86         result = video_json['result']
  87
  88         formats = [{
  89             'url': x['url'],
  90             'format_id': x['content_format'],
  91             'preference': self._known_formats.index(x['content_format']),
  92         } for x in result['files'] if x['content_format'] in self._known_formats]
  93
  94         self._sort_formats(formats)
  95
  96         if not formats:
  97             raise ExtractorError('No media links available for %s' % video_id)
  98
  99         duration = result['duration']
 100         compilation = result['compilation']
 101         title = result['title']
 102
 103         title = '%s - %s' % (compilation, title) if compilation is not None else title
 104
 105         previews = result['preview']
 106         previews.sort(key=lambda fmt: self._known_thumbnails.index(fmt['content_format']))
 107         thumbnail = previews[-1]['url'] if len(previews) > 0 else None
 108
 109         video_page = self._download_webpage(url, video_id, 'Downloading video page')
 110         description = self._extract_description(video_page)
 111         comment_count = self._extract_comment_count(video_page)
 112
 113         return {
 114             'id': video_id,
 115             'title': title,
 116             'thumbnail': thumbnail,
 117             'description': description,
 118             'duration': duration,
 119             'comment_count': comment_count,
 120             'formats': formats,
 121         }
 122
 123
 124 class IviCompilationIE(InfoExtractor):
 125     IE_DESC = 'ivi.ru compilations'
 126     IE_NAME = 'ivi:compilation'
 127     _VALID_URL = r'^https?://(?:www\.)?ivi\.ru/watch/(?!\d+)(?P<compilationid>[a-z\d_-]+)(?:/season(?P<seasonid>\d+))?$'
 128
 129     def _extract_entries(self, html, compilation_id):
 130         return [self.url_result('http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), 'Ivi')
 131                 for serie in re.findall(r'<strong><a href="/watch/%s/(\d+)">(?:[^<]+)</a></strong>' % compilation_id, html)]
 132
 133     def _real_extract(self, url):
 134         mobj = re.match(self._VALID_URL, url)
 135         compilation_id = mobj.group('compilationid')
 136         season_id = mobj.group('seasonid')
 137
 138         if season_id is not None: # Season link
 139             season_page = self._download_webpage(url, compilation_id, 'Downloading season %s web page' % season_id)
 140             playlist_id = '%s/season%s' % (compilation_id, season_id)
 141             playlist_title = self._html_search_meta('title', season_page, 'title')
 142             entries = self._extract_entries(season_page, compilation_id)
 143         else: # Compilation link
 144             compilation_page = self._download_webpage(url, compilation_id, 'Downloading compilation web page')
 145             playlist_id = compilation_id
 146             playlist_title = self._html_search_meta('title', compilation_page, 'title')
 147             seasons = re.findall(r'<a href="/watch/%s/season(\d+)">[^<]+</a>' % compilation_id, compilation_page)
 148             if len(seasons) == 0: # No seasons in this compilation
 149                 entries = self._extract_entries(compilation_page, compilation_id)
 150             else:
 151                 entries = []
 152                 for season_id in seasons:
 153                     season_page = self._download_webpage(
 154                         'http://www.ivi.ru/watch/%s/season%s' % (compilation_id, season_id),
 155                         compilation_id, 'Downloading season %s web page' % season_id)
 156                     entries.extend(self._extract_entries(season_page, compilation_id))
 157
 158         return self.playlist_result(entries, playlist_id, playlist_title)