0
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-12-18 05:19:59 +00:00
yt-dlp/yt_dlp/extractor/europeantour.py
sepro add96eb9f8
[cleanup] Add more ruff rules (#10149)
Authored by: seproDev

Reviewed-by: bashonly <88596187+bashonly@users.noreply.github.com>
Reviewed-by: Simon Sawicki <contact@grub4k.xyz>
2024-06-12 01:09:58 +02:00

34 lines
1.2 KiB
Python

import re
from .common import InfoExtractor
class EuropeanTourIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?europeantour\.com/dpworld-tour/news/video/(?P<id>[^/&?#$]+)'
_TESTS = [{
'url': 'https://www.europeantour.com/dpworld-tour/news/video/the-best-shots-of-the-2021-seasons/',
'info_dict': {
'id': '6287788195001',
'ext': 'mp4',
'title': 'The best shots of the 2021 seasons',
'duration': 2416.512,
'timestamp': 1640010141,
'uploader_id': '5136026580001',
'tags': ['prod-imported'],
'thumbnail': 'md5:fdac52bc826548860edf8145ee74e71a',
'upload_date': '20211220',
},
'params': {'skip_download': True},
}]
BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/default_default/index.html?videoId=%s'
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
vid, aid = re.search(r'(?s)brightcove-player\s?video-id="([^"]+)".*"ACCOUNT_ID":"([^"]+)"', webpage).groups()
if not aid:
aid = '5136026580001'
return self.url_result(
self.BRIGHTCOVE_URL_TEMPLATE % (aid, vid), 'BrightcoveNew')