0
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-23 02:25:11 +00:00
yt-dlp/yt_dlp/extractor/bfi.py
sepro 9751a457cf
[cleanup] Remove dead extractors (#8604)
Closes #1609, Closes #3232, Closes #4763, Closes #6026, Closes #6322, Closes #7912
Authored by: seproDev
2023-11-26 03:09:59 +00:00

35 lines
1.3 KiB
Python

import re
from .common import InfoExtractor
from ..utils import extract_attributes
class BFIPlayerIE(InfoExtractor):
_WORKING = False
IE_NAME = 'bfi:player'
_VALID_URL = r'https?://player\.bfi\.org\.uk/[^/]+/film/watch-(?P<id>[\w-]+)-online'
_TEST = {
'url': 'https://player.bfi.org.uk/free/film/watch-computer-doctor-1974-online',
'md5': 'e8783ebd8e061ec4bc6e9501ed547de8',
'info_dict': {
'id': 'htNnhlZjE60C9VySkQEIBtU-cNV1Xx63',
'ext': 'mp4',
'title': 'Computer Doctor',
'description': 'md5:fb6c240d40c4dbe40428bdd62f78203b',
},
'skip': 'BFI Player films cannot be played outside of the UK',
}
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
entries = []
for player_el in re.findall(r'(?s)<[^>]+class="player"[^>]*>', webpage):
player_attr = extract_attributes(player_el)
ooyala_id = player_attr.get('data-video-id')
if not ooyala_id:
continue
entries.append(self.url_result(
'ooyala:' + ooyala_id, 'Ooyala',
ooyala_id, player_attr.get('data-label')))
return self.playlist_result(entries)