[freesound] Minor improvements

youtube_dl/extractor/__init__.py

@@ -20,7 +20,7 @@ from .eighttracks import EightTracksIE
 from .escapist import EscapistIE
 from .facebook import FacebookIE
 from .flickr import FlickrIE
-from .freesound import FreeSoundIE
+from .freesound import FreesoundIE
 from .funnyordie import FunnyOrDieIE
 from .gamespot import GameSpotIE
 from .gametrailers import GametrailersIE

youtube_dl/extractor/freesound.py

@@ -1,36 +1,36 @@
-# -*- coding: utf-8 -*-
 import re
 
 from .common import InfoExtractor
+from ..utils import determine_ext
 
-class FreeSoundIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?freesound\.org/people/([^/]+)/sounds/([^/]+)'
+class FreesoundIE(InfoExtractor):
+    _VALID_URL = r'(?:https?://)?(?:www\.)?freesound\.org/people/([^/]+)/sounds/(?P<id>[^/]+)'
     _TEST = {
         u'url': u'http://www.freesound.org/people/miklovan/sounds/194503/',
         u'file': u'194503.mp3',
         u'md5': u'12280ceb42c81f19a515c745eae07650',
         u'info_dict': {
-            u"title": u"gulls in the city.wav by miklovan",
-            u"uploader" : u"miklovan"
+            u"title": u"gulls in the city.wav",
+            u"uploader" : u"miklovan",
+            u'description': u'the sounds of seagulls in the city',
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        music_id = mobj.group(2)
+        music_id = mobj.group('id')
         webpage = self._download_webpage(url, music_id)
-        title = self._html_search_regex(r'<meta property="og:title" content="([^"]*)"',
-                                webpage, 'music title')
-        music_url = self._html_search_regex(r'<meta property="og:audio" content="([^"]*)"',
-                                webpage, 'music url')       
-        uploader = self._html_search_regex(r'<meta property="og:audio:artist" content="([^"]*)"',
-                                webpage, 'music uploader')                                                                        
-        ext = music_url.split('.')[-1]
+        title = self._html_search_regex(r'<div id="single_sample_header">.*?<a href="#">(.+?)</a>',
+                                webpage, 'music title', flags=re.DOTALL)
+        music_url = self._og_search_property('audio', webpage, 'music url')
+        description = self._html_search_regex(r'<div id="sound_description">(.*?)</div>',
+                                webpage, 'description', fatal=False, flags=re.DOTALL)
 
         return [{
             'id':       music_id,
             'title':    title,            
             'url':      music_url,
-            'uploader': uploader,
-            'ext':      ext,
+            'uploader': self._og_search_property('audio:artist', webpage, 'music uploader'),
+            'ext':      determine_ext(music_url),
+            'description': description,
         }]