[Rte] Improve extractor

[youtube-dl] / youtube_dl / extractor / generic.py

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 877447381b3bff9568ea8c68e8faea76c4b3d5f5..9b64988943b16bc93eedce5c878ebdc088d95906 100644 (file)
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -386,8 +386,9 @@ class GenericIE(InfoExtractor):
             'info_dict': {
                 'id': 'uxjb0lwrcz',
                 'ext': 'mp4',
-                'title': 'Conversation about Hexagonal Rails Part 1',
+                'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
                 'duration': 1715.0,
+                'uploader': 'thoughtworks.wistia.com',
             },   
         },
     ]
@@ -486,7 +487,8 @@ class GenericIE(InfoExtractor):
                      'Set --default-search "ytsearch" (or run  youtube-dl "ytsearch:%s" ) to search YouTube'
                     ) % (url, url), expected=True)
             else:
-                assert ':' in default_search
+                if ':' not in default_search:
+                    default_search += ':'
                 return self.url_result(default_search + url)
 
         url, smuggled_data = unsmuggle_url(url)
@@ -661,12 +663,8 @@ class GenericIE(InfoExtractor):
                     playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
 
         # Look for embedded Wistia player
-        mobj = re.search(r'<meta content=(["\'])(?P<url>https?://fast\.wistia\.net/embed/iframe/(?P<id>[a-zA-Z0-9_]+)(["\']))', webpage)
-        if mobj:
-            return self.url_result(mobj.group('url'), 'Wistia')
-        
         match = re.search(
-            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
+            r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
         if match:
             embed_url = self._proto_relative_url(
                 unescapeHTML(match.group('url')))