]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/pornhub.py
Prepare to upload
[youtubedl] / youtube_dl / extractor / pornhub.py
index 1dcc8df00b4c2eb2da3b2719a35d874eb290c4ed..3428458afa987fb3eb8c3be061742cefce6e2734 100644 (file)
@@ -186,7 +186,7 @@ class PornHubIE(InfoExtractor):
             title, thumbnail, duration = [None] * 3
 
         video_uploader = self._html_search_regex(
-            r'(?s)From:&nbsp;.+?<(?:a href="/users/|a href="/channels/|span class="username)[^>]+>(.+?)<',
+            r'(?s)From:&nbsp;.+?<(?:a\b[^>]+\bhref=["\']/(?:user|channel)s/|span\b[^>]+\bclass=["\']username)[^>]+>(.+?)<',
             webpage, 'uploader', fatal=False)
 
         view_count = self._extract_count(
@@ -227,13 +227,20 @@ class PornHubIE(InfoExtractor):
 
 class PornHubPlaylistBaseIE(InfoExtractor):
     def _extract_entries(self, webpage):
+        # Only process container div with main playlist content skipping
+        # drop-down menu that uses similar pattern for videos (see
+        # https://github.com/rg3/youtube-dl/issues/11594).
+        container = self._search_regex(
+            r'(?s)(<div[^>]+class=["\']container.+)', webpage,
+            'container', default=webpage)
+
         return [
             self.url_result(
                 'http://www.pornhub.com/%s' % video_url,
                 PornHubIE.ie_key(), video_title=title)
             for video_url, title in orderedSet(re.findall(
                 r'href="/?(view_video\.php\?.*\bviewkey=[\da-z]+[^"]*)"[^>]*\s+title="([^"]+)"',
-                webpage))
+                container))
         ]
 
     def _real_extract(self, url):
@@ -241,22 +248,18 @@ class PornHubPlaylistBaseIE(InfoExtractor):
 
         webpage = self._download_webpage(url, playlist_id)
 
-        # Only process container div with main playlist content skipping
-        # drop-down menu that uses similar pattern for videos (see
-        # https://github.com/rg3/youtube-dl/issues/11594).
-        container = self._search_regex(
-            r'(?s)(<div[^>]+class=["\']container.+)', webpage,
-            'container', default=webpage)
-
-        entries = self._extract_entries(container)
+        entries = self._extract_entries(webpage)
 
         playlist = self._parse_json(
             self._search_regex(
-                r'playlistObject\s*=\s*({.+?});', webpage, 'playlist'),
-            playlist_id)
+                r'(?:playlistObject|PLAYLIST_VIEW)\s*=\s*({.+?});', webpage,
+                'playlist', default='{}'),
+            playlist_id, fatal=False)
+        title = playlist.get('title') or self._search_regex(
+            r'>Videos\s+in\s+(.+?)\s+[Pp]laylist<', webpage, 'title', fatal=False)
 
         return self.playlist_result(
-            entries, playlist_id, playlist.get('title'), playlist.get('description'))
+            entries, playlist_id, title, playlist.get('description'))
 
 
 class PornHubPlaylistIE(PornHubPlaylistBaseIE):
@@ -296,6 +299,7 @@ class PornHubUserVideosIE(PornHubPlaylistBaseIE):
             except ExtractorError as e:
                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
                     break
+                raise
             page_entries = self._extract_entries(webpage)
             if not page_entries:
                 break