yt_dlp/extractor/learningonscreen.py

   1 import functools
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     ExtractorError,
   7     clean_html,
   8     extract_attributes,
   9     join_nonempty,
  10     parse_duration,
  11     unified_timestamp,
  12 )
  13 from ..utils.traversal import find_element, traverse_obj
  14
  15
  16 class LearningOnScreenIE(InfoExtractor):
  17     _VALID_URL = r'https?://learningonscreen\.ac\.uk/ondemand/index\.php/prog/(?P<id>\w+)'
  18     _TESTS = [{
  19         'url': 'https://learningonscreen.ac.uk/ondemand/index.php/prog/005D81B2?bcast=22757013',
  20         'info_dict': {
  21             'id': '005D81B2',
  22             'ext': 'mp4',
  23             'title': 'Planet Earth',
  24             'duration': 3600.0,
  25             'timestamp': 1164567600.0,
  26             'upload_date': '20061126',
  27             'thumbnail': 'https://stream.learningonscreen.ac.uk/trilt-cover-images/005D81B2-Planet-Earth-2006-11-26T190000Z-BBC4.jpg',
  28         },
  29     }]
  30
  31     def _real_initialize(self):
  32         if not self._get_cookies('https://learningonscreen.ac.uk/').get('PHPSESSID-BOB-LIVE'):
  33             self.raise_login_required(method='session_cookies')
  34
  35     def _real_extract(self, url):
  36         video_id = self._match_id(url)
  37         webpage = self._download_webpage(url, video_id)
  38
  39         details = traverse_obj(webpage, (
  40             {find_element(id='programme-details', html=True)}, {
  41                 'title': ({find_element(tag='h2')}, {clean_html}),
  42                 'timestamp': (
  43                     {find_element(cls='broadcast-date')},
  44                     {functools.partial(re.match, r'([^<]+)')}, 1, {unified_timestamp}),
  45                 'duration': (
  46                     {find_element(cls='prog-running-time')}, {clean_html}, {parse_duration}),
  47             }))
  48
  49         title = details.pop('title', None) or traverse_obj(webpage, (
  50             {find_element(id='add-to-existing-playlist', html=True)},
  51             {extract_attributes}, 'data-record-title', {clean_html}))
  52
  53         entries = self._parse_html5_media_entries(
  54             'https://stream.learningonscreen.ac.uk', webpage, video_id, m3u8_id='hls', mpd_id='dash',
  55             _headers={'Origin': 'https://learningonscreen.ac.uk', 'Referer': 'https://learningonscreen.ac.uk/'})
  56         if not entries:
  57             raise ExtractorError('No video found')
  58
  59         if len(entries) > 1:
  60             duration = details.pop('duration', None)
  61             for idx, entry in enumerate(entries, start=1):
  62                 entry.update(details)
  63                 entry['id'] = join_nonempty(video_id, idx)
  64                 entry['title'] = join_nonempty(title, idx)
  65             return self.playlist_result(entries, video_id, title, duration=duration)
  66
  67         return {
  68             **entries[0],
  69             **details,
  70             'id': video_id,
  71             'title': title,
  72         }