from __future__ import unicode_literals
+import re
+
from .common import InfoExtractor
from .ooyala import OoyalaIE
class TheSunIE(InfoExtractor):
- _VALID_URL = r'https://(?:www\.)?thesun\.co\.uk/\w+/(?P<id>\d+)/[\w-]'
+ _VALID_URL = r'https://(?:www\.)?thesun\.co\.uk/[^/]+/(?P<id>\d+)'
_TEST = {
'url': 'https://www.thesun.co.uk/tvandshowbiz/2261604/orlando-bloom-and-katy-perry-post-adorable-instagram-video-together-celebrating-thanksgiving-after-split-rumours/',
- 'md5': '5667123b24f25f43f4c4f381ef34c5c2',
'info_dict': {
- 'id': 'h4OXN0NzE6rv6ObkEifKcNA-gYUw4xFf',
- 'ext': 'mp4',
- 'title': 'Katy Perry and Orlando Bloom shut down split rumours with cute Thanksgiving video',
- 'description': 'Still going strong',
- 'duration': 31.28,
- }
+ 'id': '2261604',
+ 'title': 'md5:cba22f48bad9218b64d5bbe0e16afddf',
+ },
+ 'playlist_count': 2,
}
def _real_extract(self, url):
- video_id = self._match_id(url)
+ article_id = self._match_id(url)
+
+ webpage = self._download_webpage(url, article_id)
- webpage = self._download_webpage(url, video_id)
- ooyala_id = self._search_regex(r'id\s*=\s*"thesun-ooyala-player-([^"]+)"', webpage, 'ooyala id')
+ entries = []
+ for ooyala_id in re.findall(
+ r'<[^>]+\b(?:id\s*=\s*"thesun-ooyala-player-|data-content-id\s*=\s*")([^"]+)',
+ webpage):
+ entries.append(OoyalaIE._build_url_result(ooyala_id))
- return OoyalaIE._build_url_result(ooyala_id)
+ return self.playlist_result(
+ entries, article_id, self._og_search_title(webpage, fatal=False))