[aparat] Fix extraction

closes #22285
closes #22611
closes #23348
closes #24354
closes #24591
closes #24904
closes #25418
closes #26070
closes #26350
closes #26738
closes #27563
This commit is contained in:
Remita Amine 2020-12-28 18:19:30 +01:00
parent 782ea947b4
commit 64e419bd73

View File

@ -3,6 +3,7 @@ from __future__ import unicode_literals
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
get_element_by_id,
int_or_none, int_or_none,
merge_dicts, merge_dicts,
mimetype2ext, mimetype2ext,
@ -39,23 +40,15 @@ class AparatIE(InfoExtractor):
webpage = self._download_webpage(url, video_id, fatal=False) webpage = self._download_webpage(url, video_id, fatal=False)
if not webpage: if not webpage:
# Note: There is an easier-to-parse configuration at
# http://www.aparat.com/video/video/config/videohash/%video_id
# but the URL in there does not work
webpage = self._download_webpage( webpage = self._download_webpage(
'http://www.aparat.com/video/video/embed/vt/frame/showvideo/yes/videohash/' + video_id, 'http://www.aparat.com/video/video/embed/vt/frame/showvideo/yes/videohash/' + video_id,
video_id) video_id)
options = self._parse_json( options = self._parse_json(self._search_regex(
self._search_regex( r'options\s*=\s*({.+?})\s*;', webpage, 'options'), video_id)
r'options\s*=\s*JSON\.parse\(\s*(["\'])(?P<value>(?:(?!\1).)+)\1\s*\)',
webpage, 'options', group='value'),
video_id)
player = options['plugins']['sabaPlayerPlugin']
formats = [] formats = []
for sources in player['multiSRC']: for sources in (options.get('multiSRC') or []):
for item in sources: for item in sources:
if not isinstance(item, dict): if not isinstance(item, dict):
continue continue
@ -85,11 +78,12 @@ class AparatIE(InfoExtractor):
info = self._search_json_ld(webpage, video_id, default={}) info = self._search_json_ld(webpage, video_id, default={})
if not info.get('title'): if not info.get('title'):
info['title'] = player['title'] info['title'] = get_element_by_id('videoTitle', webpage) or \
self._html_search_meta(['og:title', 'twitter:title', 'DC.Title', 'title'], webpage, fatal=True)
return merge_dicts(info, { return merge_dicts(info, {
'id': video_id, 'id': video_id,
'thumbnail': url_or_none(options.get('poster')), 'thumbnail': url_or_none(options.get('poster')),
'duration': int_or_none(player.get('duration')), 'duration': int_or_none(options.get('duration')),
'formats': formats, 'formats': formats,
}) })