yt_dlp/extractor/techtalks.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     get_element_by_attribute,
   6     clean_html,
   7 )
   8
   9
  10 class TechTalksIE(InfoExtractor):
  11     _VALID_URL = r'https?://techtalks\.tv/talks/(?:[^/]+/)?(?P<id>\d+)'
  12
  13     _TESTS = [{
  14         'url': 'http://techtalks.tv/talks/learning-topic-models-going-beyond-svd/57758/',
  15         'info_dict': {
  16             'id': '57758',
  17             'title': 'Learning Topic Models --- Going beyond SVD',
  18         },
  19         'playlist': [
  20             {
  21                 'info_dict': {
  22                     'id': '57758',
  23                     'ext': 'flv',
  24                     'title': 'Learning Topic Models --- Going beyond SVD',
  25                 },
  26             },
  27             {
  28                 'info_dict': {
  29                     'id': '57758-slides',
  30                     'ext': 'flv',
  31                     'title': 'Learning Topic Models --- Going beyond SVD',
  32                 },
  33             },
  34         ],
  35         'params': {
  36             # rtmp download
  37             'skip_download': True,
  38         },
  39     }, {
  40         'url': 'http://techtalks.tv/talks/57758',
  41         'only_matching': True,
  42     }]
  43
  44     def _real_extract(self, url):
  45         mobj = self._match_valid_url(url)
  46         talk_id = mobj.group('id')
  47         webpage = self._download_webpage(url, talk_id)
  48         rtmp_url = self._search_regex(
  49             r'netConnectionUrl: \'(.*?)\'', webpage, 'rtmp url')
  50         play_path = self._search_regex(
  51             r'href=\'(.*?)\' [^>]*id="flowplayer_presenter"',
  52             webpage, 'presenter play path')
  53         title = clean_html(get_element_by_attribute('class', 'title', webpage))
  54         video_info = {
  55             'id': talk_id,
  56             'title': title,
  57             'url': rtmp_url,
  58             'play_path': play_path,
  59             'ext': 'flv',
  60         }
  61         m_slides = re.search(r'<a class="slides" href=\'(.*?)\'', webpage)
  62         if m_slides is None:
  63             return video_info
  64         else:
  65             return {
  66                 '_type': 'playlist',
  67                 'id': talk_id,
  68                 'title': title,
  69                 'entries': [
  70                     video_info,
  71                     # The slides video
  72                     {
  73                         'id': talk_id + '-slides',
  74                         'title': title,
  75                         'url': rtmp_url,
  76                         'play_path': m_slides.group(1),
  77                         'ext': 'flv',
  78                     },
  79                 ],
  80             }