1 from __future__ import unicode_literals
5 from ..compat import compat_urlparse
12 from .common import InfoExtractor
15 class VikiIE(InfoExtractor):
18 _VALID_URL = r'^https?://(?:www\.)?viki\.com/videos/(?P<id>[0-9]+v)'
20 'url': 'http://www.viki.com/videos/1023585v-heirs-episode-14',
24 'title': 'Heirs Episode 14',
26 'description': 'md5:c4b17b9626dd4b143dcc4d855ba3474e',
27 'upload_date': '20131121',
30 'skip': 'Blocked in the US',
33 def _real_extract(self, url):
34 video_id = self._match_id(url)
36 webpage = self._download_webpage(url, video_id)
37 title = self._og_search_title(webpage)
38 description = self._og_search_description(webpage)
39 thumbnail = self._og_search_thumbnail(webpage)
41 uploader_m = re.search(
42 r'<strong>Broadcast Network: </strong>\s*([^<]*)<', webpage)
43 if uploader_m is None:
46 uploader = uploader_m.group(1).strip()
48 rating_str = self._html_search_regex(
49 r'<strong>Rating: </strong>\s*([^<]*)<', webpage,
50 'rating information', default='').strip()
51 age_limit = US_RATINGS.get(rating_str)
53 info_url = 'http://www.viki.com/player5_fragment/%s?action=show&controller=videos' % video_id
54 info_webpage = self._download_webpage(
55 info_url, video_id, note='Downloading info page')
56 if re.match(r'\s*<div\s+class="video-error', info_webpage):
58 'Video %s is blocked from your location.' % video_id,
60 video_url = self._html_search_regex(
61 r'<source[^>]+src="([^"]+)"', info_webpage, 'video URL')
63 upload_date_str = self._html_search_regex(
64 r'"created_at":"([^"]+)"', info_webpage, 'upload date')
66 unified_strdate(upload_date_str)
67 if upload_date_str is not None
72 video_subtitles = self.extract_subtitles(video_id, info_webpage)
78 'description': description,
79 'thumbnail': thumbnail,
80 'age_limit': age_limit,
82 'subtitles': video_subtitles,
83 'upload_date': upload_date,
86 def _get_subtitles(self, video_id, info_webpage):
88 for sturl_html in re.findall(r'<track src="([^"]+)"', info_webpage):
89 sturl = unescapeHTML(sturl_html)
90 m = re.search(r'/(?P<lang>[a-z]+)\.vtt', sturl)
93 res[m.group('lang')] = [{
94 'url': compat_urlparse.urljoin('http://www.viki.com', sturl),