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/recurbate.py
coletdjnz 3d2623a898
[compat, networking] Deprecate old functions (#2861)
Authored by: coletdjnz, pukkandan
2023-07-15 16:18:35 +05:30

42 lines
1.9 KiB
Python
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

from .common import InfoExtractor
from ..networking.exceptions import HTTPError
from ..utils import ExtractorError, merge_dicts
class RecurbateIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?recurbate\.com/play\.php\?video=(?P<id>\d+)'
_TESTS = [{
'url': 'https://recurbate.com/play.php?video=39161415',
'md5': 'dd2b4ec57aa3e3572cb5cf0997fca99f',
'info_dict': {
'id': '39161415',
'ext': 'mp4',
'description': 'md5:db48d09e4d93fc715f47fd3d6b7edd51',
'title': 'Performer zsnicole33 show on 2022-10-25 20:23, Chaturbate Archive Recurbate',
'age_limit': 18,
},
'skip': 'Website require membership.',
}]
def _real_extract(self, url):
SUBSCRIPTION_MISSING_MESSAGE = 'This video is only available for registered users; Set your authenticated browser user agent via the --user-agent parameter.'
video_id = self._match_id(url)
try:
webpage = self._download_webpage(url, video_id)
except ExtractorError as e:
if isinstance(e.cause, HTTPError) and e.cause.status == 403:
self.raise_login_required(msg=SUBSCRIPTION_MISSING_MESSAGE, method='cookies')
raise
token = self._html_search_regex(r'data-token="([^"]+)"', webpage, 'token')
video_url = f'https://recurbate.com/api/get.php?video={video_id}&token={token}'
video_webpage = self._download_webpage(video_url, video_id)
if video_webpage == 'shall_subscribe':
self.raise_login_required(msg=SUBSCRIPTION_MISSING_MESSAGE, method='cookies')
entries = self._parse_html5_media_entries(video_url, video_webpage, video_id)
return merge_dicts({
'id': video_id,
'title': self._html_extract_title(webpage, 'title'),
'description': self._og_search_description(webpage),
'age_limit': self._rta_search(webpage),
}, entries[0])