yt_dlp/extractor/lecture2go.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     determine_ext,
   6     determine_protocol,
   7     parse_duration,
   8     int_or_none,
   9 )
  10
  11
  12 class Lecture2GoIE(InfoExtractor):
  13     _VALID_URL = r'https?://lecture2go\.uni-hamburg\.de/veranstaltungen/-/v/(?P<id>\d+)'
  14     _TEST = {
  15         'url': 'https://lecture2go.uni-hamburg.de/veranstaltungen/-/v/17473',
  16         'md5': 'ac02b570883020d208d405d5a3fd2f7f',
  17         'info_dict': {
  18             'id': '17473',
  19             'ext': 'mp4',
  20             'title': '2 - Endliche Automaten und reguläre Sprachen',
  21             'creator': 'Frank Heitmann',
  22             'duration': 5220,
  23         },
  24         'params': {
  25             # m3u8 download
  26             'skip_download': True,
  27         }
  28     }
  29
  30     def _real_extract(self, url):
  31         video_id = self._match_id(url)
  32         webpage = self._download_webpage(url, video_id)
  33
  34         title = self._html_search_regex(r'<em[^>]+class="title">(.+)</em>', webpage, 'title')
  35
  36         formats = []
  37         for url in set(re.findall(r'var\s+playerUri\d+\s*=\s*"([^"]+)"', webpage)):
  38             ext = determine_ext(url)
  39             protocol = determine_protocol({'url': url})
  40             if ext == 'f4m':
  41                 formats.extend(self._extract_f4m_formats(url, video_id, f4m_id='hds'))
  42             elif ext == 'm3u8':
  43                 formats.extend(self._extract_m3u8_formats(url, video_id, ext='mp4', m3u8_id='hls'))
  44             else:
  45                 if protocol == 'rtmp':
  46                     continue  # XXX: currently broken
  47                 formats.append({
  48                     'format_id': protocol,
  49                     'url': url,
  50                 })
  51
  52         creator = self._html_search_regex(
  53             r'<div[^>]+id="description">([^<]+)</div>', webpage, 'creator', fatal=False)
  54         duration = parse_duration(self._html_search_regex(
  55             r'Duration:\s*</em>\s*<em[^>]*>([^<]+)</em>', webpage, 'duration', fatal=False))
  56         view_count = int_or_none(self._html_search_regex(
  57             r'Views:\s*</em>\s*<em[^>]+>(\d+)</em>', webpage, 'view count', fatal=False))
  58
  59         return {
  60             'id': video_id,
  61             'title': title,
  62             'formats': formats,
  63             'creator': creator,
  64             'duration': duration,
  65             'view_count': view_count,
  66         }