0
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-16 01:25:06 +00:00
yt-dlp/yt_dlp/extractor/theholetv.py
Dosychev Peter 4e7f375c94
[extractor/theholetv] Add extractor (#4325)
Authored by: dosy4ev
2022-07-11 04:48:12 +05:30

36 lines
1.5 KiB
Python

from .common import InfoExtractor
from ..utils import extract_attributes, remove_end
class TheHoleTvIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?the-hole\.tv/episodes/(?P<id>[\w-]+)'
_TESTS = [{
'url': 'https://the-hole.tv/episodes/gromkii-vopros-sergey-orlov',
'md5': 'fea6682f47786f3ae5a6cbd635ec4bf9',
'info_dict': {
'id': 'gromkii-vopros-sergey-orlov',
'ext': 'mp4',
'title': 'Сергей Орлов — Громкий вопрос',
'thumbnail': 'https://assets-cdn.the-hole.tv/images/t8gan4n6zn627e7wni11b2uemqts',
'description': 'md5:45741a9202331f995d9fb76996759379'
}
}]
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
player_attrs = extract_attributes(self._search_regex(
r'(<div[^>]*\bdata-controller="player"[^>]*>)', webpage, 'video player'))
formats, subtitles = self._extract_m3u8_formats_and_subtitles(
player_attrs['data-player-source-value'], video_id, 'mp4')
self._sort_formats(formats)
return {
'id': video_id,
'title': remove_end(self._html_extract_title(webpage), ' — The Hole'),
'description': self._og_search_description(webpage),
'thumbnail': player_attrs.get('data-player-poster-value'),
'formats': formats,
'subtitles': subtitles
}