[spankbang] Fix extraction (closes #23307, closes #23423, closes #23444)

This commit is contained in:
Sergey M․ 2019-12-31 22:16:39 +07:00
parent 3bed621750
commit 2b845c4086
No known key found for this signature in database
GPG key ID: 2C393E0F18A9236D

View file

@ -4,6 +4,7 @@
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
determine_ext,
ExtractorError, ExtractorError,
merge_dicts, merge_dicts,
orderedSet, orderedSet,
@ -75,6 +76,15 @@ def extract_format(format_id, format_url):
if not f_url: if not f_url:
return return
f = parse_resolution(format_id) f = parse_resolution(format_id)
ext = determine_ext(f_url)
if format_id.startswith('m3u8') or ext == 'm3u8':
formats.extend(self._extract_m3u8_formats(
f_url, video_id, 'mp4', entry_protocol='m3u8_native',
m3u8_id='hls', fatal=False))
elif format_id.startswith('mpd') or ext == 'mpd':
formats.extend(self._extract_mpd_formats(
f_url, video_id, mpd_id='dash', fatal=False))
elif ext == 'mp4' or f.get('width') or f.get('height'):
f.update({ f.update({
'url': f_url, 'url': f_url,
'format_id': format_id, 'format_id': format_id,
@ -93,28 +103,22 @@ def extract_format(format_id, format_url):
r'data-streamkey\s*=\s*(["\'])(?P<value>(?:(?!\1).)+)\1', r'data-streamkey\s*=\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
webpage, 'stream key', group='value') webpage, 'stream key', group='value')
sb_csrf_session = self._get_cookies(
'https://spankbang.com')['sb_csrf_session'].value
stream = self._download_json( stream = self._download_json(
'https://spankbang.com/api/videos/stream', video_id, 'https://spankbang.com/api/videos/stream', video_id,
'Downloading stream JSON', data=urlencode_postdata({ 'Downloading stream JSON', data=urlencode_postdata({
'id': stream_key, 'id': stream_key,
'data': 0, 'data': 0,
'sb_csrf_session': sb_csrf_session,
}), headers={ }), headers={
'Referer': url, 'Referer': url,
'X-CSRFToken': sb_csrf_session, 'X-Requested-With': 'XMLHttpRequest',
}) })
for format_id, format_url in stream.items(): for format_id, format_url in stream.items():
if format_id.startswith(STREAM_URL_PREFIX):
if format_url and isinstance(format_url, list): if format_url and isinstance(format_url, list):
format_url = format_url[0] format_url = format_url[0]
extract_format( extract_format(format_id, format_url)
format_id[len(STREAM_URL_PREFIX):], format_url)
self._sort_formats(formats) self._sort_formats(formats, field_preference=('preference', 'height', 'width', 'fps', 'tbr', 'format_id'))
info = self._search_json_ld(webpage, video_id, default={}) info = self._search_json_ld(webpage, video_id, default={})