yt_dlp/extractor/olympics.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     int_or_none,
   7     try_get
   8 )
   9
  10
  11 class OlympicsReplayIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?:www\.)?olympics\.com(?:/tokyo-2020)?/[a-z]{2}/(?:replay|video)/(?P<id>[^/#&?]+)'
  13     _TESTS = [{
  14         'url': 'https://olympics.com/fr/video/men-s-109kg-group-a-weightlifting-tokyo-2020-replays',
  15         'info_dict': {
  16             'id': 'f6a0753c-8e6f-4b7d-a435-027054a4f8e9',
  17             'ext': 'mp4',
  18             'title': '+109kg (H) Groupe A - Haltérophilie | Replay de Tokyo 2020',
  19             'upload_date': '20210801',
  20             'timestamp': 1627783200,
  21             'description': 'md5:c66af4a5bc7429dbcc43d15845ff03b3',
  22         },
  23         'params': {
  24             'format': 'bestvideo',
  25             'skip_download': True,
  26         },
  27     }, {
  28         'url': 'https://olympics.com/tokyo-2020/en/replay/bd242924-4b22-49a5-a846-f1d4c809250d/mens-bronze-medal-match-hun-esp',
  29         'only_matching': True,
  30     }]
  31
  32     def _real_extract(self, url):
  33         id = self._match_id(url)
  34
  35         webpage = self._download_webpage(url, id)
  36         title = self._html_search_meta(('title', 'og:title', 'twitter:title'), webpage)
  37         uuid = self._html_search_meta('episode_uid', webpage)
  38         m3u8_url = self._html_search_meta('video_url', webpage)
  39         json_ld = self._search_json_ld(webpage, uuid)
  40         thumbnails_list = json_ld.get('image')
  41         if not thumbnails_list:
  42             thumbnails_list = self._html_search_regex(
  43                 r'["\']image["\']:\s*["\']([^"\']+)["\']', webpage, 'images', default='')
  44             thumbnails_list = thumbnails_list.replace('[', '').replace(']', '').split(',')
  45             thumbnails_list = [thumbnail.strip() for thumbnail in thumbnails_list]
  46         thumbnails = []
  47         for thumbnail in thumbnails_list:
  48             width_a, height_a, width = self._search_regex(
  49                 r'/images/image/private/t_(?P<width_a>\d+)-(?P<height_a>\d+)_(?P<width>\d+)/primary/[\W\w\d]+',
  50                 thumbnail, 'thumb', group=(1, 2, 3), default=(None, None, None))
  51             width_a, height_a, width = int_or_none(width_a), int_or_none(height_a), int_or_none(width)
  52             thumbnails.append({
  53                 'url': thumbnail,
  54                 'width': width,
  55                 'height': int_or_none(try_get(width, lambda x: x * height_a / width_a))
  56             })
  57         m3u8_url = self._download_json(
  58             f'https://olympics.com/tokenGenerator?url={m3u8_url}', uuid, note='Downloading m3u8 url')
  59         formats, subtitles = self._extract_m3u8_formats_and_subtitles(m3u8_url, uuid, m3u8_id='hls')
  60         self._sort_formats(formats)
  61
  62         return {
  63             'id': uuid,
  64             'title': title,
  65             'timestamp': json_ld.get('timestamp'),
  66             'description': json_ld.get('description'),
  67             'thumbnails': thumbnails,
  68             'duration': json_ld.get('duration'),
  69             'formats': formats,
  70             'subtitles': subtitles,
  71         }