yt_dlp/extractor/olympics.py

   1 from .common import InfoExtractor
   2 from ..utils import int_or_none, try_get
   3
   4
   5 class OlympicsReplayIE(InfoExtractor):
   6     _VALID_URL = r'https?://(?:www\.)?olympics\.com(?:/tokyo-2020)?/[a-z]{2}/(?:replay|video)/(?P<id>[^/#&?]+)'
   7     _TESTS = [{
   8         'url': 'https://olympics.com/fr/video/men-s-109kg-group-a-weightlifting-tokyo-2020-replays',
   9         'info_dict': {
  10             'id': 'f6a0753c-8e6f-4b7d-a435-027054a4f8e9',
  11             'ext': 'mp4',
  12             'title': '+109kg (H) Groupe A - Haltérophilie | Replay de Tokyo 2020',
  13             'upload_date': '20210801',
  14             'timestamp': 1627783200,
  15             'description': 'md5:c66af4a5bc7429dbcc43d15845ff03b3',
  16             'uploader': 'International Olympic Committee',
  17         },
  18         'params': {
  19             'skip_download': True,
  20         },
  21     }, {
  22         'url': 'https://olympics.com/tokyo-2020/en/replay/bd242924-4b22-49a5-a846-f1d4c809250d/mens-bronze-medal-match-hun-esp',
  23         'only_matching': True,
  24     }]
  25
  26     def _real_extract(self, url):
  27         id = self._match_id(url)
  28
  29         webpage = self._download_webpage(url, id)
  30         title = self._html_search_meta(('title', 'og:title', 'twitter:title'), webpage)
  31         uuid = self._html_search_meta('episode_uid', webpage)
  32         m3u8_url = self._html_search_meta('video_url', webpage)
  33         json_ld = self._search_json_ld(webpage, uuid)
  34         thumbnails_list = json_ld.get('image')
  35         if not thumbnails_list:
  36             thumbnails_list = self._html_search_regex(
  37                 r'["\']image["\']:\s*["\']([^"\']+)["\']', webpage, 'images', default='')
  38             thumbnails_list = thumbnails_list.replace('[', '').replace(']', '').split(',')
  39             thumbnails_list = [thumbnail.strip() for thumbnail in thumbnails_list]
  40         thumbnails = []
  41         for thumbnail in thumbnails_list:
  42             width_a, height_a, width = self._search_regex(
  43                 r'/images/image/private/t_(?P<width_a>\d+)-(?P<height_a>\d+)_(?P<width>\d+)/primary/[\W\w\d]+',
  44                 thumbnail, 'thumb', group=(1, 2, 3), default=(None, None, None))
  45             width_a, height_a, width = int_or_none(width_a), int_or_none(height_a), int_or_none(width)
  46             thumbnails.append({
  47                 'url': thumbnail,
  48                 'width': width,
  49                 'height': int_or_none(try_get(width, lambda x: x * height_a / width_a))
  50             })
  51         m3u8_url = self._download_json(
  52             f'https://olympics.com/tokenGenerator?url={m3u8_url}', uuid, note='Downloading m3u8 url')
  53         formats, subtitles = self._extract_m3u8_formats_and_subtitles(m3u8_url, uuid, 'mp4', m3u8_id='hls')
  54
  55         return {
  56             'id': uuid,
  57             'title': title,
  58             'thumbnails': thumbnails,
  59             'formats': formats,
  60             'subtitles': subtitles,
  61             **json_ld
  62         }