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

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_HTTPError,
   8     compat_urlparse,
   9 )
  10 from ..utils import (
  11     ExtractorError,
  12     parse_duration,
  13 )
  14
  15
  16 class VideoLecturesNetIE(InfoExtractor):
  17     _VALID_URL = r'http://(?:www\.)?videolectures\.net/(?P<id>[^/#?]+)/*(?:[#?].*)?$'
  18     IE_NAME = 'videolectures.net'
  19
  20     _TESTS = [{
  21         'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/',
  22         'info_dict': {
  23             'id': 'promogram_igor_mekjavic_eng',
  24             'ext': 'mp4',
  25             'title': 'Automatics, robotics and biocybernetics',
  26             'description': 'md5:815fc1deb6b3a2bff99de2d5325be482',
  27             'upload_date': '20130627',
  28             'duration': 565,
  29             'thumbnail': 're:http://.*\.jpg',
  30         },
  31     }, {
  32         'url': 'http://videolectures.net/deeplearning2015_montreal/',
  33         'info_dict': {
  34             'id': 'deeplearning2015_montreal',
  35             'title': 'Deep Learning Summer School, Montreal 2015',
  36             'description': 'md5:90121a40cc6926df1bf04dcd8563ed3b',
  37         },
  38         'playlist_count': 30,
  39     }]
  40
  41     def _real_extract(self, url):
  42         video_id = self._match_id(url)
  43
  44         smil_url = 'http://videolectures.net/%s/video/1/smil.xml' % video_id
  45
  46         try:
  47             smil = self._download_smil(smil_url, video_id)
  48         except ExtractorError as e:
  49             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
  50                 # Probably a playlist
  51                 webpage = self._download_webpage(url, video_id)
  52                 entries = [
  53                     self.url_result(compat_urlparse.urljoin(url, video_url), 'VideoLecturesNet')
  54                     for _, video_url in re.findall(r'<a[^>]+href=(["\'])(.+?)\1[^>]+id=["\']lec=\d+', webpage)]
  55                 playlist_title = self._html_search_meta('title', webpage, 'title', fatal=True)
  56                 playlist_description = self._html_search_meta('description', webpage, 'description')
  57                 return self.playlist_result(entries, video_id, playlist_title, playlist_description)
  58
  59         info = self._parse_smil(smil, smil_url, video_id)
  60
  61         info['id'] = video_id
  62
  63         switch = smil.find('.//switch')
  64         if switch is not None:
  65             info['duration'] = parse_duration(switch.attrib.get('dur'))
  66
  67         return info