yt_dlp/extractor/hollywoodreporter.py

   1 import functools
   2 import re
   3
   4 from .common import InfoExtractor
   5 from .jwplatform import JWPlatformIE
   6 from ..utils import (
   7     ExtractorError,
   8     OnDemandPagedList,
   9     extract_attributes,
  10     get_element_by_class,
  11     get_element_html_by_class,
  12 )
  13
  14
  15 class HollywoodReporterIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.)?hollywoodreporter\.com/video/(?P<id>[\w-]+)'
  17     _TESTS = [{
  18         'url': 'https://www.hollywoodreporter.com/video/chris-pine-michelle-rodriguez-dungeons-dragons-cast-directors-on-what-it-took-to-make-film-sxsw-2023/',
  19         'info_dict': {
  20             'id': 'zH4jZaR5',
  21             'ext': 'mp4',
  22             'title': 'md5:a9a1c073770a32f178955997712c4bd9',
  23             'description': 'The cast and directors of \'Dungeons & Dragons: Honor Among Thieves\' talk about their new film.',
  24             'thumbnail': 'https://cdn.jwplayer.com/v2/media/zH4jZaR5/poster.jpg?width=720',
  25             'upload_date': '20230312',
  26             'timestamp': 1678586423,
  27             'duration': 242.0,
  28         },
  29         'params': {'skip_download': 'm3u8'},
  30     }]
  31
  32     def _real_extract(self, url):
  33         display_id = self._match_id(url)
  34         webpage = self._download_webpage(url, display_id)
  35
  36         data = extract_attributes(get_element_html_by_class('vlanding-video-card__link', webpage) or '')
  37         video_id = data['data-video-showcase-trigger']
  38         showcase_type = data['data-video-showcase-type']
  39
  40         if showcase_type == 'jwplayer':
  41             return self.url_result(f'jwplatform:{video_id}', JWPlatformIE)
  42         elif showcase_type == 'youtube':
  43             return self.url_result(video_id, 'Youtube')
  44         else:
  45             raise ExtractorError(f'Unsupported showcase type "{showcase_type}"')
  46
  47
  48 class HollywoodReporterPlaylistIE(InfoExtractor):
  49     _VALID_URL = r'https?://(?:www\.)?hollywoodreporter\.com/vcategory/(?P<slug>[\w-]+)-(?P<id>\d+)'
  50     _TESTS = [{
  51         'url': 'https://www.hollywoodreporter.com/vcategory/heat-vision-breakdown-57822/',
  52         'playlist_mincount': 109,
  53         'info_dict': {
  54             'id': '57822',
  55             'title': 'heat-vision-breakdown',
  56         },
  57     }]
  58
  59     def _fetch_page(self, slug, pl_id, page):
  60         page += 1
  61         webpage = self._download_webpage(
  62             f'https://www.hollywoodreporter.com/vcategory/{slug}-{pl_id}/page/{page}/',
  63             pl_id, note=f'Downloading playlist page {page}')
  64         section = get_element_by_class('video-playlist-river', webpage) or ''
  65
  66         for url in re.findall(r'<a[^>]+href="([^"]+)"[^>]+class="c-title__link', section):
  67             yield self.url_result(url, HollywoodReporterIE)
  68
  69     def _real_extract(self, url):
  70         slug, pl_id = self._match_valid_url(url).group('slug', 'id')
  71         return self.playlist_result(
  72             OnDemandPagedList(functools.partial(self._fetch_page, slug, pl_id), 15), pl_id, slug)