_ Git - youtube-dl/blob - youtube_dl/extractor/viki.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from ..compat import (
   6     compat_urlparse,
   7     compat_urllib_request,
   8 )
   9 from ..utils import (
  10     ExtractorError,
  11     unescapeHTML,
  12     unified_strdate,
  13     US_RATINGS,
  14     clean_html,
  15     determine_ext,
  16     mimetype2ext,
  17 )
  18 from .common import InfoExtractor
  19
  20
  21 class VikiIE(InfoExtractor):
  22     IE_NAME = 'viki'
  23
  24     # iPad2
  25     _USER_AGENT = 'Mozilla/5.0(iPad; U; CPU OS 4_3 like Mac OS X; en-us) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8F191 Safari/6533.18.5'
  26
  27     _VALID_URL = r'^https?://(?:www\.)?viki\.com/videos/(?P<id>[0-9]+v)'
  28     _TESTS = [{
  29         'url': 'http://www.viki.com/videos/1023585v-heirs-episode-14',
  30         'info_dict': {
  31             'id': '1023585v',
  32             'ext': 'mp4',
  33             'title': 'Heirs Episode 14',
  34             'uploader': 'SBS',
  35             'description': 'md5:c4b17b9626dd4b143dcc4d855ba3474e',
  36             'upload_date': '20131121',
  37             'age_limit': 13,
  38         },
  39         'skip': 'Blocked in the US',
  40     }, {
  41         'url': 'http://www.viki.com/videos/1067139v-the-avengers-age-of-ultron-press-conference',
  42         'md5': 'ca6493e6f0a6ec07da9aa8d6304b4b2c',
  43         'info_dict': {
  44             'id': '1067139v',
  45             'ext': 'mp4',
  46             'description': 'md5:d70b2f9428f5488321bfe1db10d612ea',
  47             'upload_date': '20150430',
  48             'title': '\'The Avengers: Age of Ultron\' Press Conference',
  49         }
  50     }, {
  51         'url': 'http://www.viki.com/videos/1048879v-ankhon-dekhi',
  52         'info_dict': {
  53             'id': '1048879v',
  54             'ext': 'mp4',
  55             'upload_date': '20140820',
  56             'description': 'md5:54ff56d51bdfc7a30441ec967394e91c',
  57             'title': 'Ankhon Dekhi',
  58         },
  59         'params': {
  60             # requires ffmpeg
  61             'skip_download': True,
  62         }
  63     }]
  64
  65     def _real_extract(self, url):
  66         video_id = self._match_id(url)
  67
  68         webpage = self._download_webpage(url, video_id)
  69         title = self._og_search_title(webpage)
  70         description = self._og_search_description(webpage)
  71         thumbnail = self._og_search_thumbnail(webpage)
  72
  73         uploader_m = re.search(
  74             r'<strong>Broadcast Network: </strong>\s*([^<]*)<', webpage)
  75         if uploader_m is None:
  76             uploader = None
  77         else:
  78             uploader = uploader_m.group(1).strip()
  79
  80         rating_str = self._html_search_regex(
  81             r'<strong>Rating: </strong>\s*([^<]*)<', webpage,
  82             'rating information', default='').strip()
  83         age_limit = US_RATINGS.get(rating_str)
  84
  85         req = compat_urllib_request.Request(
  86             'http://www.viki.com/player5_fragment/%s?action=show&controller=videos' % video_id)
  87         req.add_header('User-Agent', self._USER_AGENT)
  88         info_webpage = self._download_webpage(
  89             req, video_id, note='Downloading info page')
  90         err_msg = self._html_search_regex(r'<div[^>]+class="video-error[^>]+>(.+)</div>', info_webpage, 'error message', default=None)
  91         if err_msg:
  92             err_msg = clean_html(err_msg)
  93             if 'not available in your region' in err_msg:
  94                 raise ExtractorError(
  95                     'Video %s is blocked from your location.' % video_id,
  96                     expected=True)
  97             else:
  98                 raise ExtractorError('Viki said: ' + err_msg)
  99         mobj = re.search(
 100             r'<source[^>]+type="(?P<mime_type>[^"]+)"[^>]+src="(?P<url>[^"]+)"', info_webpage)
 101         if not mobj:
 102             raise ExtractorError('Unable to find video URL')
 103         video_url = unescapeHTML(mobj.group('url'))
 104         video_ext = mimetype2ext(mobj.group('mime_type'))
 105
 106         if determine_ext(video_url) == 'm3u8':
 107             formats = self._extract_m3u8_formats(
 108                 video_url, video_id, ext=video_ext)
 109         else:
 110             formats = [{
 111                 'url': video_url,
 112                 'ext': video_ext,
 113             }]
 114
 115         upload_date_str = self._html_search_regex(
 116             r'"created_at":"([^"]+)"', info_webpage, 'upload date')
 117         upload_date = (
 118             unified_strdate(upload_date_str)
 119             if upload_date_str is not None
 120             else None
 121         )
 122
 123         # subtitles
 124         video_subtitles = self.extract_subtitles(video_id, info_webpage)
 125
 126         return {
 127             'id': video_id,
 128             'title': title,
 129             'formats': formats,
 130             'description': description,
 131             'thumbnail': thumbnail,
 132             'age_limit': age_limit,
 133             'uploader': uploader,
 134             'subtitles': video_subtitles,
 135             'upload_date': upload_date,
 136         }
 137
 138     def _get_subtitles(self, video_id, info_webpage):
 139         res = {}
 140         for sturl_html in re.findall(r'<track src="([^"]+)"', info_webpage):
 141             sturl = unescapeHTML(sturl_html)
 142             m = re.search(r'/(?P<lang>[a-z]+)\.vtt', sturl)
 143             if not m:
 144                 continue
 145             res[m.group('lang')] = [{
 146                 'url': compat_urlparse.urljoin('http://www.viki.com', sturl),
 147                 'ext': 'vtt',
 148             }]
 149         return res