yt_dlp/extractor/olympics.py

   1 from .common import InfoExtractor
   2 from ..networking.exceptions import HTTPError
   3 from ..utils import (
   4     ExtractorError,
   5     int_or_none,
   6     parse_iso8601,
   7     parse_qs,
   8     try_get,
   9     update_url,
  10     url_or_none,
  11 )
  12 from ..utils.traversal import traverse_obj
  13
  14
  15 class OlympicsReplayIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.)?olympics\.com/[a-z]{2}/(?:paris-2024/)?(?:replay|videos?|original-series/episode)/(?P<id>[\w-]+)'
  17     _TESTS = [{
  18         'url': 'https://olympics.com/fr/video/men-s-109kg-group-a-weightlifting-tokyo-2020-replays',
  19         'info_dict': {
  20             'id': 'f6a0753c-8e6f-4b7d-a435-027054a4f8e9',
  21             'ext': 'mp4',
  22             'title': '+109kg (H) Groupe A - Haltérophilie | Replay de Tokyo 2020',
  23             'upload_date': '20210801',
  24             'timestamp': 1627797600,
  25             'description': 'md5:c66af4a5bc7429dbcc43d15845ff03b3',
  26             'thumbnail': 'https://img.olympics.com/images/image/private/t_1-1_1280/primary/nua4o7zwyaznoaejpbk2',
  27             'duration': 7017.0,
  28         },
  29     }, {
  30         'url': 'https://olympics.com/en/original-series/episode/b-boys-and-b-girls-take-the-spotlight-breaking-life-road-to-paris-2024',
  31         'info_dict': {
  32             'id': '32633650-c5ee-4280-8b94-fb6defb6a9b5',
  33             'ext': 'mp4',
  34             'title': 'B-girl Nicka - Breaking Life, Road to Paris 2024 | Episode 1',
  35             'upload_date': '20240517',
  36             'timestamp': 1715948200,
  37             'description': 'md5:f63d728a41270ec628f6ac33ce471bb1',
  38             'thumbnail': 'https://img.olympics.com/images/image/private/t_1-1_1280/primary/a3j96l7j6so3vyfijby1',
  39             'duration': 1321.0,
  40         },
  41     }, {
  42         'url': 'https://olympics.com/en/paris-2024/videos/men-s-preliminaries-gbr-esp-ned-rsa-hockey-olympic-games-paris-2024',
  43         'info_dict': {
  44             'id': '3d96db23-8eee-4b7c-8ef5-488a0361026c',
  45             'ext': 'mp4',
  46             'title': 'Men\'s Preliminaries GBR-ESP & NED-RSA | Hockey | Olympic Games Paris 2024',
  47             'upload_date': '20240727',
  48             'timestamp': 1722066600,
  49         },
  50         'skip': 'Geo-restricted to RU, BR, BT, NP, TM, BD, TL',
  51     }, {
  52         'url': 'https://olympics.com/en/paris-2024/videos/dnp-suni-lee-i-have-goals-and-i-have-expectations-for-myself-but-i-also-am-trying-to-give-myself-grace',
  53         'info_dict': {
  54             'id': 'a42f37ab-8a74-41d0-a7d9-af27b7b02a90',
  55             'ext': 'mp4',
  56             'title': 'md5:c7cfbc9918636a98e66400a812e4d407',
  57             'upload_date': '20240729',
  58             'timestamp': 1722288600,
  59         },
  60     }]
  61     _GEO_BYPASS = False
  62
  63     def _extract_from_nextjs_data(self, webpage, video_id):
  64         data = traverse_obj(self._search_nextjs_data(webpage, video_id, default={}), (
  65             'props', 'pageProps', 'page', 'items',
  66             lambda _, v: v['name'] == 'videoPlaylist', 'data', 'currentVideo', {dict}, any))
  67         if not data:
  68             return None
  69
  70         geo_countries = traverse_obj(data, ('countries', ..., {str}))
  71         if traverse_obj(data, ('geoRestrictedVideo', {bool})):
  72             self.raise_geo_restricted(countries=geo_countries)
  73
  74         is_live = traverse_obj(data, ('streamingStatus', {str})) == 'LIVE'
  75         m3u8_url = traverse_obj(data, ('videoUrl', {url_or_none})) or data['streamUrl']
  76         tokenized_url = self._tokenize_url(m3u8_url, data['jwtToken'], is_live, video_id)
  77
  78         try:
  79             formats, subtitles = self._extract_m3u8_formats_and_subtitles(
  80                 tokenized_url, video_id, 'mp4', m3u8_id='hls')
  81         except ExtractorError as e:
  82             if isinstance(e.cause, HTTPError) and 'georestricted' in e.cause.msg:
  83                 self.raise_geo_restricted(countries=geo_countries)
  84             raise
  85
  86         return {
  87             'formats': formats,
  88             'subtitles': subtitles,
  89             'is_live': is_live,
  90             **traverse_obj(data, {
  91                 'id': ('videoID', {str}),
  92                 'title': ('title', {str}),
  93                 'timestamp': ('contentDate', {parse_iso8601}),
  94             }),
  95         }
  96
  97     def _tokenize_url(self, url, token, is_live, video_id):
  98         return self._download_json(
  99             'https://metering.olympics.com/tokengenerator', video_id,
 100             'Downloading tokenized m3u8 url', query={
 101                 **parse_qs(url),
 102                 'url': update_url(url, query=None),
 103                 'service-id': 'live' if is_live else 'vod',
 104                 'user-auth': token,
 105             })['data']['url']
 106
 107     def _legacy_tokenize_url(self, url, video_id):
 108         return self._download_json(
 109             'https://olympics.com/tokenGenerator', video_id,
 110             'Downloading legacy tokenized m3u8 url', query={'url': url})
 111
 112     def _real_extract(self, url):
 113         video_id = self._match_id(url)
 114         webpage = self._download_webpage(url, video_id)
 115
 116         if info := self._extract_from_nextjs_data(webpage, video_id):
 117             return info
 118
 119         title = self._html_search_meta(('title', 'og:title', 'twitter:title'), webpage)
 120         video_uuid = self._html_search_meta('episode_uid', webpage)
 121         m3u8_url = self._html_search_meta('video_url', webpage)
 122         json_ld = self._search_json_ld(webpage, video_uuid)
 123         thumbnails_list = json_ld.get('image')
 124         if not thumbnails_list:
 125             thumbnails_list = self._html_search_regex(
 126                 r'["\']image["\']:\s*["\']([^"\']+)["\']', webpage, 'images', default='')
 127             thumbnails_list = thumbnails_list.replace('[', '').replace(']', '').split(',')
 128             thumbnails_list = [thumbnail.strip() for thumbnail in thumbnails_list]
 129         thumbnails = []
 130         for thumbnail in thumbnails_list:
 131             width_a, height_a, width = self._search_regex(
 132                 r'/images/image/private/t_(?P<width_a>\d+)-(?P<height_a>\d+)_(?P<width>\d+)/primary/[\W\w\d]+',
 133                 thumbnail, 'thumb', group=(1, 2, 3), default=(None, None, None))
 134             width_a, height_a, width = int_or_none(width_a), int_or_none(height_a), int_or_none(width)
 135             thumbnails.append({
 136                 'url': thumbnail,
 137                 'width': width,
 138                 'height': int_or_none(try_get(width, lambda x: x * height_a / width_a)),
 139             })
 140
 141         formats, subtitles = self._extract_m3u8_formats_and_subtitles(
 142             self._legacy_tokenize_url(m3u8_url, video_uuid), video_uuid, 'mp4', m3u8_id='hls')
 143
 144         return {
 145             'id': video_uuid,
 146             'title': title,
 147             'thumbnails': thumbnails,
 148             'formats': formats,
 149             'subtitles': subtitles,
 150             **json_ld,
 151         }