youtube_dl/extractor/lecture2go.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     determine_ext,
   9     determine_protocol,
  10     parse_duration,
  11     int_or_none,
  12 )
  13
  14
  15 class Lecture2GoIE(InfoExtractor):
  16     _VALID_URL = r'https?://lecture2go\.uni-hamburg\.de/veranstaltungen/-/v/(?P<id>\d+)'
  17     _TEST = {
  18         'url': 'https://lecture2go.uni-hamburg.de/veranstaltungen/-/v/17473',
  19         'md5': 'ac02b570883020d208d405d5a3fd2f7f',
  20         'info_dict': {
  21             'id': '17473',
  22             'ext': 'mp4',
  23             'title': '2 - Endliche Automaten und reguläre Sprachen',
  24             'creator': 'Frank Heitmann',
  25             'duration': 5220,
  26         },
  27         'params': {
  28             # m3u8 download
  29             'skip_download': True,
  30         }
  31     }
  32
  33     def _real_extract(self, url):
  34         video_id = self._match_id(url)
  35         webpage = self._download_webpage(url, video_id)
  36
  37         title = self._html_search_regex(r'<em[^>]+class="title">(.+)</em>', webpage, 'title')
  38
  39         formats = []
  40         for url in set(re.findall(r'var\s+playerUri\d+\s*=\s*"([^"]+)"', webpage)):
  41             ext = determine_ext(url)
  42             protocol = determine_protocol({'url': url})
  43             if ext == 'f4m':
  44                 formats.extend(self._extract_f4m_formats(url, video_id, f4m_id='hds'))
  45             elif ext == 'm3u8':
  46                 formats.extend(self._extract_m3u8_formats(url, video_id, ext='mp4', m3u8_id='hls'))
  47             else:
  48                 if protocol == 'rtmp':
  49                     continue  # XXX: currently broken
  50                 formats.append({
  51                     'format_id': protocol,
  52                     'url': url,
  53                 })
  54
  55         self._sort_formats(formats)
  56
  57         creator = self._html_search_regex(
  58             r'<div[^>]+id="description">([^<]+)</div>', webpage, 'creator', fatal=False)
  59         duration = parse_duration(self._html_search_regex(
  60             r'Duration:\s*</em>\s*<em[^>]*>([^<]+)</em>', webpage, 'duration', fatal=False))
  61         view_count = int_or_none(self._html_search_regex(
  62             r'Views:\s*</em>\s*<em[^>]+>(\d+)</em>', webpage, 'view count', fatal=False))
  63
  64         return {
  65             'id': video_id,
  66             'title': title,
  67             'formats': formats,
  68             'creator': creator,
  69             'duration': duration,
  70             'view_count': view_count,
  71         }