[pornhub] Improve title extraction (closes #24184)
authorSergey M․ <dstftw@gmail.com>
Mon, 2 Mar 2020 23:23:39 +0000 (06:23 +0700)
committerSergey M․ <dstftw@gmail.com>
Mon, 2 Mar 2020 23:23:39 +0000 (06:23 +0700)
youtube_dl/extractor/pornhub.py

index b3251ccd9b2300188f7efce561c3c5d1fbff702e..b8f65af7cbc15c566c01422ee6229f197af3015b 100644 (file)
@@ -189,10 +189,10 @@ class PornHubIE(PornHubBaseIE):
         # http://www.pornhub.com/view_video.php?viewkey=1331683002), not relying
         # on that anymore.
         title = self._html_search_meta(
-            'twitter:title', webpage, default=None) or self._search_regex(
-            (r'<h1[^>]+class=["\']title["\'][^>]*>(?P<title>[^<]+)',
-             r'<div[^>]+data-video-title=(["\'])(?P<title>.+?)\1',
-             r'shareTitle\s*=\s*(["\'])(?P<title>.+?)\1'),
+            'twitter:title', webpage, default=None) or self._html_search_regex(
+            (r'(?s)<h1[^>]+class=["\']title["\'][^>]*>(?P<title>.+?)</h1>',
+             r'<div[^>]+data-video-title=(["\'])(?P<title>(?:(?!\1).)+)\1',
+             r'shareTitle["\']\s*[=:]\s*(["\'])(?P<title>(?:(?!\1).)+)\1'),
             webpage, 'title', group='title')
 
         video_urls = []