_ 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 compat_urlparse
   7 from ..utils import parse_duration
   8
   9
  10 class VideoLecturesNetIE(InfoExtractor):
  11     _VALID_URL = r'http://(?:www\.)?videolectures\.net/(?P<id>[^/#?]+)/*(?:[#?].*)?$'
  12     IE_NAME = 'videolectures.net'
  13
  14     _TEST = {
  15         'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/',
  16         'info_dict': {
  17             'id': 'promogram_igor_mekjavic_eng',
  18             'ext': 'mp4',
  19             'title': 'Automatics, robotics and biocybernetics',
  20             'description': 'md5:815fc1deb6b3a2bff99de2d5325be482',
  21             'upload_date': '20130627',
  22             'duration': 565,
  23             'thumbnail': 're:http://.*\.jpg',
  24         },
  25     }
  26
  27     def _real_extract(self, url):
  28         video_id = self._match_id(url)
  29
  30         smil_url = 'http://videolectures.net/%s/video/1/smil.xml' % video_id
  31         smil = self._download_smil(smil_url, video_id, fatal=False)
  32
  33         # Probably a playlist
  34         if smil is False:
  35             webpage = self._download_webpage(url, video_id)
  36             entries = [
  37                 self.url_result(compat_urlparse.urljoin(url, video_url), 'VideoLecturesNet')
  38                 for _, video_url in re.findall(r'<a[^>]+href=(["\'])(.+?)\1[^>]+id=["\']lec=\d+', webpage)]
  39             playlist_title = self._html_search_meta('title', webpage, 'title', fatal=True)
  40             playlist_description = self._html_search_meta('description', webpage, 'description')
  41             return self.playlist_result(entries, video_id, playlist_title, playlist_description)
  42
  43         info = self._parse_smil(smil, smil_url, video_id)
  44
  45         info['id'] = video_id
  46
  47         switch = smil.find('.//switch')
  48         if switch is not None:
  49             info['duration'] = parse_duration(switch.attrib.get('dur'))
  50
  51         return info