[hark] Modernize

youtube_dl/extractor/hark.py

@@ -1,37 +1,33 @@
 # -*- coding: utf-8 -*-
-
+from __future__ import unicode_literals
-import re
-import json
 
 from .common import InfoExtractor
-from ..utils import determine_ext
+
 
 class HarkIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.hark\.com/clips/(.+?)-.+'
+    _VALID_URL = r'https?://www\.hark\.com/clips/(?P<id>.+?)-.+'
     _TEST = {
-        u'url': u'http://www.hark.com/clips/mmbzyhkgny-obama-beyond-the-afghan-theater-we-only-target-al-qaeda-on-may-23-2013',
+        'url': 'http://www.hark.com/clips/mmbzyhkgny-obama-beyond-the-afghan-theater-we-only-target-al-qaeda-on-may-23-2013',
-        u'file': u'mmbzyhkgny.mp3',
+        'md5': '6783a58491b47b92c7c1af5a77d4cbee',
-        u'md5': u'6783a58491b47b92c7c1af5a77d4cbee',
+        'info_dict': {
-        u'info_dict': {
+            'id': 'mmbzyhkgny',
-            u'title': u"Obama: 'Beyond The Afghan Theater, We Only Target Al Qaeda' on May 23, 2013",
+            'ext': 'mp3',
-            u'description': u'President Barack Obama addressed the nation live on May 23, 2013 in a speech aimed at addressing counter-terrorism policies including the use of drone strikes, detainees at Guantanamo Bay prison facility, and American citizens who are terrorists.',
+            'title': 'Obama: \'Beyond The Afghan Theater, We Only Target Al Qaeda\' on May 23, 2013',
-            u'duration': 11,
+            'description': 'President Barack Obama addressed the nation live on May 23, 2013 in a speech aimed at addressing counter-terrorism policies including the use of drone strikes, detainees at Guantanamo Bay prison facility, and American citizens who are terrorists.',
+            'duration': 11,
         }
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
-        video_id = mobj.group(1)
+        data = self._download_json(
-        json_url = "http://www.hark.com/clips/%s.json" %(video_id)
+            'http://www.hark.com/clips/%s.json' % video_id, video_id)
-        info_json = self._download_webpage(json_url, video_id)
-        info = json.loads(info_json)
-        final_url = info['url']
 
-        return {'id': video_id,
+        return {
-                'url' : final_url,
+            'id': video_id,
-                'title': info['name'],
+            'url': data['url'],
-                'ext': determine_ext(final_url),
+            'title': data['name'],
-                'description': info['description'],
+            'description': data.get('description'),
-                'thumbnail': info['image_original'],
+            'thumbnail': data.get('image_original'),
-                'duration': info['duration'],
+            'duration': data.get('duration'),
-                }
+        }