[youtube] Add support for invidiou.sh (#20309)
[youtube-dl] / youtube_dl / extractor / mtv.py
index fc098cd13677c33b9813338bb3e50b3d85bb3103..7a3b57abd852f4ffb8e49bdce9d9e8029fca65c3 100644 (file)
@@ -50,8 +50,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
         thumb_node = itemdoc.find(search_path)
         if thumb_node is None:
             return None
-        else:
-            return thumb_node.attrib['url']
+        return thumb_node.get('url') or thumb_node.text or None
 
     def _extract_mobile_video_formats(self, mtvn_id):
         webpage_url = self._MOBILE_TEMPLATE % mtvn_id
@@ -116,10 +115,17 @@ class MTVServicesInfoExtractor(InfoExtractor):
             if transcript.get('kind') != 'captions':
                 continue
             lang = transcript.get('srclang')
-            subtitles[lang] = [{
-                'url': compat_str(typographic.get('src')),
-                'ext': typographic.get('format')
-            } for typographic in transcript.findall('./typographic')]
+            for typographic in transcript.findall('./typographic'):
+                sub_src = typographic.get('src')
+                if not sub_src:
+                    continue
+                ext = typographic.get('format')
+                if ext == 'cea-608':
+                    ext = 'scc'
+                subtitles.setdefault(lang, []).append({
+                    'url': compat_str(sub_src),
+                    'ext': ext
+                })
         return subtitles
 
     def _get_video_info(self, itemdoc, use_hls=True):
@@ -259,7 +265,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
 
         if mgid is None or ':' not in mgid:
             mgid = self._search_regex(
-                [r'data-mgid="(.*?)"', r'swfobject.embedSWF\(".*?(mgid:.*?)"'],
+                [r'data-mgid="(.*?)"', r'swfobject\.embedSWF\(".*?(mgid:.*?)"'],
                 webpage, 'mgid', default=None)
 
         if not mgid: