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     get_element_by_class,
  10     get_element_html_by_id,
  11     join_nonempty,
  12     parse_duration,
  13     unified_timestamp,
  14 )
  15 from ..utils.traversal import traverse_obj
  16
  17
  18 class LearningOnScreenIE(InfoExtractor):
  19     _VALID_URL = r'https?://learningonscreen\.ac\.uk/ondemand/index\.php/prog/(?P<id>\w+)'
  20     _TESTS = [{
  21         'url': 'https://learningonscreen.ac.uk/ondemand/index.php/prog/005D81B2?bcast=22757013',
  22         'info_dict': {
  23             'id': '005D81B2',
  24             'ext': 'mp4',
  25             'title': 'Planet Earth',
  26             'duration': 3600.0,
  27             'timestamp': 1164567600.0,
  28             'upload_date': '20061126',
  29             'thumbnail': 'https://stream.learningonscreen.ac.uk/trilt-cover-images/005D81B2-Planet-Earth-2006-11-26T190000Z-BBC4.jpg',
  30         },
  31     }]
  32
  33     def _real_initialize(self):
  34         if not self._get_cookies('https://learningonscreen.ac.uk/').get('PHPSESSID-BOB-LIVE'):
  35             self.raise_login_required(
  36                 'Use --cookies for authentication. See '
  37                 ' https://github.com/yt-dlp/yt-dlp/wiki/FAQ#how-do-i-pass-cookies-to-yt-dlp  '
  38                 'for how to manually pass cookies', method=None)
  39
  40     def _real_extract(self, url):
  41         video_id = self._match_id(url)
  42         webpage = self._download_webpage(url, video_id)
  43
  44         details = traverse_obj(webpage, (
  45             {functools.partial(get_element_html_by_id, 'programme-details')}, {
  46                 'title': ({functools.partial(re.search, r'<h2>([^<]+)</h2>')}, 1, {clean_html}),
  47                 'timestamp': (
  48                     {functools.partial(get_element_by_class, 'broadcast-date')},
  49                     {functools.partial(re.match, r'([^<]+)')}, 1, {unified_timestamp}),
  50                 'duration': (
  51                     {functools.partial(get_element_by_class, 'prog-running-time')},
  52                     {clean_html}, {parse_duration}),
  53             }))
  54
  55         title = details.pop('title', None) or traverse_obj(webpage, (
  56             {functools.partial(get_element_html_by_id, 'add-to-existing-playlist')},
  57             {extract_attributes}, 'data-record-title', {clean_html}))
  58
  59         entries = self._parse_html5_media_entries(
  60             'https://stream.learningonscreen.ac.uk', webpage, video_id, m3u8_id='hls', mpd_id='dash',
  61             _headers={'Origin': 'https://learningonscreen.ac.uk', 'Referer': 'https://learningonscreen.ac.uk/'})
  62         if not entries:
  63             raise ExtractorError('No video found')
  64
  65         if len(entries) > 1:
  66             duration = details.pop('duration', None)
  67             for idx, entry in enumerate(entries, start=1):
  68                 entry.update(details)
  69                 entry['id'] = join_nonempty(video_id, idx)
  70                 entry['title'] = join_nonempty(title, idx)
  71             return self.playlist_result(entries, video_id, title, duration=duration)
  72
  73         return {
  74             **entries[0],
  75             **details,
  76             'id': video_id,
  77             'title': title,
  78         }