_ 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     _TEST = {
  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
  33     def _real_extract(self, url):
  34         video_id = self._match_id(url)
  35
  36         smil_url = 'http://videolectures.net/%s/video/1/smil.xml' % video_id
  37
  38         try:
  39             smil = self._download_smil(smil_url, video_id)
  40         except ExtractorError as e:
  41             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
  42                 # Probably a playlist
  43                 webpage = self._download_webpage(url, video_id)
  44                 entries = [
  45                     self.url_result(compat_urlparse.urljoin(url, video_url), 'VideoLecturesNet')
  46                     for _, video_url in re.findall(r'<a[^>]+href=(["\'])(.+?)\1[^>]+id=["\']lec=\d+', webpage)]
  47                 playlist_title = self._html_search_meta('title', webpage, 'title', fatal=True)
  48                 playlist_description = self._html_search_meta('description', webpage, 'description')
  49                 return self.playlist_result(entries, video_id, playlist_title, playlist_description)
  50
  51         info = self._parse_smil(smil, smil_url, video_id)
  52
  53         info['id'] = video_id
  54
  55         switch = smil.find('.//switch')
  56         if switch is not None:
  57             info['duration'] = parse_duration(switch.attrib.get('dur'))
  58
  59         return info