[pornhub] Fix extraction (closes #11997)
authorSergey M․ <dstftw@gmail.com>
Mon, 6 Feb 2017 18:52:59 +0000 (01:52 +0700)
committerSergey M․ <dstftw@gmail.com>
Mon, 6 Feb 2017 18:52:59 +0000 (01:52 +0700)
youtube_dl/extractor/pornhub.py

index 3eaf56973ec35072d8f0549c5850357ca94ed12b..017f6c55219ff3db0cf9bf745f74031882f12c54 100644 (file)
@@ -156,7 +156,12 @@ class PornHubIE(InfoExtractor):
         comment_count = self._extract_count(
             r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment')
 
-        video_urls = list(map(compat_urllib_parse_unquote, re.findall(r"player_quality_[0-9]{3}p\s*=\s*'([^']+)'", webpage)))
+        video_urls = []
+        for quote, video_url in re.findall(
+                r'player_quality_[0-9]{3,4}p\s*=\s*(["\'])(.+?)\1;', webpage):
+            video_urls.append(compat_urllib_parse_unquote(re.sub(
+                r'{0}\s*\+\s*{0}'.format(quote), '', video_url)))
+
         if webpage.find('"encrypted":true') != -1:
             password = compat_urllib_parse_unquote_plus(
                 self._search_regex(r'"video_title":"([^"]+)', webpage, 'password'))