]> gitweb @ CieloNegro.org - youtube-dl.git/commitdiff
[pornhub] Fix extraction (closes #12007)
authorThomas Christlieb <thomaschristlieb@hotmail.com>
Wed, 8 Feb 2017 12:53:39 +0000 (13:53 +0100)
committerSergey M․ <dstftw@gmail.com>
Thu, 9 Feb 2017 17:56:12 +0000 (00:56 +0700)
youtube_dl/extractor/pornhub.py

index 017f6c55219ff3db0cf9bf745f74031882f12c54..5e930f45e0ac271a3c37e5bffa4a23f791a2ebf0 100644 (file)
@@ -156,11 +156,24 @@ class PornHubIE(InfoExtractor):
         comment_count = self._extract_count(
             r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment')
 
+        video_variables = {}
+        for video_variablename, quote, video_variable in re.findall(
+                r'(player_quality_[0-9]{3,4}p[0-9a-z]+?)=\s*(["\'])(.*?)\2;', webpage):
+            video_variables[video_variablename] = video_variable
+
+        encoded_video_urls = []
+        for encoded_video_url in re.findall(
+                r'player_quality_[0-9]{3,4}p\s*=(.*?);', webpage):
+            encoded_video_urls.append(encoded_video_url)
+
+        # Decode the URLs 
         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)))
+        for url in encoded_video_urls:
+            for varname, varval in video_variables.items():
+                url = url.replace(varname, varval)
+            url = url.replace('+', '')
+            url = url.replace(' ', '')
+            video_urls.append(url)
 
         if webpage.find('"encrypted":true') != -1:
             password = compat_urllib_parse_unquote_plus(