2 from __future__ import unicode_literals
7 from .common import InfoExtractor
14 class IviIE(InfoExtractor):
17 _VALID_URL = r'https?://(?:www\.)?ivi\.ru/(?:watch/(?:[^/]+/)?|video/player\?.*?videoId=)(?P<videoid>\d+)'
22 'url': 'http://www.ivi.ru/watch/53141',
23 'md5': '6ff5be2254e796ed346251d117196cf4',
27 'title': 'Иван Васильевич меняет профессию',
28 'description': 'md5:b924063ea1677c8fe343d8a72ac2195f',
30 'thumbnail': 'http://thumbs.ivi.ru/f20.vcp.digitalaccess.ru/contents/d/1/c3c885163a082c29bceeb7b5a267a6.jpg',
32 'skip': 'Only works from Russia',
36 'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/9549',
37 'md5': '221f56b35e3ed815fde2df71032f4b3e',
41 'title': 'Двое из ларца - Серия 1',
43 'thumbnail': 'http://thumbs.ivi.ru/f15.vcp.digitalaccess.ru/contents/8/4/0068dc0677041f3336b7c2baad8fc0.jpg',
45 'skip': 'Only works from Russia',
50 _known_formats = ['MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi', 'MP4-SHQ']
53 _known_thumbnails = ['Thumb-120x90', 'Thumb-160', 'Thumb-640x480']
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
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
63 def _real_extract(self, url):
64 mobj = re.match(self._VALID_URL, url)
65 video_id = mobj.group('videoid')
67 api_url = 'http://api.digitalaccess.ru/api/json/'
69 data = {'method': 'da.content.get',
70 'params': [video_id, {'site': 's183',
71 'referrer': 'http://www.ivi.ru/watch/%s' % video_id,
77 request = compat_urllib_request.Request(api_url, json.dumps(data))
79 video_json_page = self._download_webpage(request, video_id, 'Downloading video JSON')
80 video_json = json.loads(video_json_page)
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)
88 result = video_json['result']
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]
96 self._sort_formats(formats)
99 raise ExtractorError('No media links available for %s' % video_id)
101 duration = result['duration']
102 compilation = result['compilation']
103 title = result['title']
105 title = '%s - %s' % (compilation, title) if compilation is not None else title
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
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)
118 'thumbnail': thumbnail,
119 'description': description,
120 'duration': duration,
121 'comment_count': comment_count,
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+))?$'
131 'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa',
133 'id': 'dvoe_iz_lartsa',
134 'title': 'Двое из ларца (2006 - 2008)',
136 'playlist_mincount': 24,
138 'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/season1',
140 'id': 'dvoe_iz_lartsa/season1',
141 'title': 'Двое из ларца (2006 - 2008) 1 сезон',
143 'playlist_mincount': 12,
146 def _extract_entries(self, html, compilation_id):
147 return [self.url_result('http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), 'Ivi')
148 for serie in re.findall(r'<strong><a href="/watch/%s/(\d+)">(?:[^<]+)</a></strong>' % compilation_id, html)]
150 def _real_extract(self, url):
151 mobj = re.match(self._VALID_URL, url)
152 compilation_id = mobj.group('compilationid')
153 season_id = mobj.group('seasonid')
155 if season_id is not None: # Season link
156 season_page = self._download_webpage(url, compilation_id, 'Downloading season %s web page' % season_id)
157 playlist_id = '%s/season%s' % (compilation_id, season_id)
158 playlist_title = self._html_search_meta('title', season_page, 'title')
159 entries = self._extract_entries(season_page, compilation_id)
160 else: # Compilation link
161 compilation_page = self._download_webpage(url, compilation_id, 'Downloading compilation web page')
162 playlist_id = compilation_id
163 playlist_title = self._html_search_meta('title', compilation_page, 'title')
164 seasons = re.findall(r'<a href="/watch/%s/season(\d+)">[^<]+</a>' % compilation_id, compilation_page)
165 if len(seasons) == 0: # No seasons in this compilation
166 entries = self._extract_entries(compilation_page, compilation_id)
169 for season_id in seasons:
170 season_page = self._download_webpage(
171 'http://www.ivi.ru/watch/%s/season%s' % (compilation_id, season_id),
172 compilation_id, 'Downloading season %s web page' % season_id)
173 entries.extend(self._extract_entries(season_page, compilation_id))
175 return self.playlist_result(entries, playlist_id, playlist_title)