youtube-dl

commit e1f6e61e6ab9d1ab6c27bbbd855e991001db216c
parent 0932300e3a7d609f203bdfad4bcadaf1fafee67c
Author: Jaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com>
Date:   Thu, 18 Jul 2013 20:47:10 +0200

Add an extractor for 56.com (related #1039)

Diffstat:
M
youtube_dl/extractor/__init__.py
 | 
1
+
A
youtube_dl/extractor/c56.py
 | 
36
++++++++++++++++++++++++++++++++++++

2 files changed, 37 insertions(+), 0 deletions(-)
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
@@ -71,6 +71,7 @@ from .veoh import VeohIE
 from .vevo import VevoIE
 from .vimeo import VimeoIE
 from .vine import VineIE
+from .c56 import C56IE
 from .wat import WatIE
 from .wimp import WimpIE
 from .worldstarhiphop import WorldStarHipHopIE
diff --git a/youtube_dl/extractor/c56.py b/youtube_dl/extractor/c56.py
@@ -0,0 +1,36 @@
+# coding: utf-8
+
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import determine_ext
+
+class C56IE(InfoExtractor):
+    _VALID_URL = r'https?://((www|player)\.)?56\.com/(.+?/)?(v_|(play_album.+-))(?P<textid>.+?)\.(html|swf)'
+    IE_NAME = u'56.com'
+
+    _TEST ={
+        u'url': u'http://www.56.com/u39/v_OTM0NDA3MTY.html',
+        u'file': u'93440716.mp4',
+        u'md5': u'9dc07b5c8e978112a6441f9e75d2b59e',
+        u'info_dict': {
+            u'title': u'网事知多少 第32期：车怒',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
+        text_id = mobj.group('textid')
+        info_page = self._download_webpage('http://vxml.56.com/json/%s/' % text_id,
+                                           text_id, u'Downloading video info')
+        info = json.loads(info_page)['info']
+        best_format = sorted(info['rfiles'], key=lambda f: int(f['filesize']))[-1]
+        video_url = best_format['url']
+
+        return {'id': info['vid'],
+                'title': info['Subject'],
+                'url': video_url,
+                'ext': determine_ext(video_url),
+                'thumbnail': info.get('bimg') or info.get('img'),
+                }