youtube-dl

commit 2a7b4da9b2ee11e88976e0e93796fd8460aa053d
parent 069d098f846ca53073ec646f335f77dac4439844
Author: Jaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com>
Date:   Tue, 27 Aug 2013 10:25:38 +0200

[hark] get the song info in JSON and extract more information.

Diffstat:
M
youtube_dl/extractor/hark.py
 | 
22
++++++++++++----------

1 file changed, 12 insertions(+), 10 deletions(-)
diff --git a/youtube_dl/extractor/hark.py b/youtube_dl/extractor/hark.py
@@ -1,6 +1,7 @@
 # -*- coding: utf-8 -*-
 
 import re
+import json
 
 from .common import InfoExtractor
 from ..utils import determine_ext
@@ -12,24 +13,25 @@ class HarkIE(InfoExtractor):
         u'file': u'mmbzyhkgny.mp3',
         u'md5': u'6783a58491b47b92c7c1af5a77d4cbee',
         u'info_dict': {
-            u"title": u"Obama: 'Beyond The Afghan Theater, We Only Target Al Qaeda' On May 23, 2013 ",
+            u'title': u"Obama: 'Beyond The Afghan Theater, We Only Target Al Qaeda' on May 23, 2013",
+            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.',
+            u'duration': 11,
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group(1)
-        embed_url = "http://www.hark.com/clips/%s/homepage_embed" %(video_id)
-        webpage = self._download_webpage(embed_url, video_id)
-
-        final_url = self._search_regex(r'src="(.+?).mp3"',
-                                webpage, 'video url')+'.mp3'
-        title = self._html_search_regex(r'<title>(.+?)</title>',
-                                webpage, 'video title').replace(' Sound Clip and Quote - Hark','').replace(
-                                'Sound Clip , Quote, MP3, and Ringtone - Hark','')
+        json_url = "http://www.hark.com/clips/%s.json" %(video_id)
+        info_json = self._download_webpage(json_url, video_id)
+        info = json.loads(info_json)
+        final_url = info['url']
 
         return {'id': video_id,
                 'url' : final_url,
-                'title': title,
+                'title': info['name'],
                 'ext': determine_ext(final_url),
+                'description': info['description'],
+                'thumbnail': info['image_original'],
+                'duration': info['duration'],
                 }