[extractor/generic] Improve generic SMIL detection

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

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index cd133a10c38c6d7bf081d550160bf730b456515e..6900ed96f6bc11397731acb2929d8c52a7436c8d 100644 (file)
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -276,14 +276,6 @@ class GenericIE(InfoExtractor):
                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
             },
         },
-        # BBC iPlayer embeds
-        {
-            'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER',
-            'info_dict': {
-                'title': 'BBC - Blogs -  Adam Curtis - BUGGER',
-            },
-            'playlist_mincount': 18,
-        },
         # RUTV embed
         {
             'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
@@ -1118,11 +1110,13 @@ class GenericIE(InfoExtractor):
 
         self.report_extraction(video_id)
 
-        # Is it an RSS feed?
+        # Is it an RSS feed or a SMIL file?
         try:
             doc = parse_xml(webpage)
             if doc.tag == 'rss':
                 return self._extract_rss(url, video_id, doc)
+            elif re.match(r'^(?:{[^}]+})?smil$', doc.tag):
+                return self._parse_smil(doc, url, video_id)
         except compat_xml_parse_error:
             pass
 
@@ -1702,7 +1696,7 @@ class GenericIE(InfoExtractor):
                 if refresh_header:
                     found = re.search(REDIRECT_REGEX, refresh_header)
             if found:
-                new_url = compat_urlparse.urljoin(url, found.group(1))
+                new_url = compat_urlparse.urljoin(url, unescapeHTML(found.group(1)))
                 self.report_following_redirect(new_url)
                 return {
                     '_type': 'url',