Support TeamCoco URLs with video_id in the title
authorAGSPhoenix <lx45803@gmail.com>
Fri, 4 Apr 2014 17:42:34 +0000 (13:42 -0400)
committerAGSPhoenix <lx45803@gmail.com>
Fri, 4 Apr 2014 17:42:34 +0000 (13:42 -0400)
If the URL has the video_id in it, use that since the current method of
finding the id breaks on those pages.

Fixes 2698.

youtube_dl/extractor/teamcoco.py

index 9dcffead04d5466c14c6f2ff60995ecfb5435e6d..e0fc3e60fcdf593b58216c215b87f1ff87ce909a 100644 (file)
@@ -9,7 +9,7 @@ from ..utils import (
 
 
 class TeamcocoIE(InfoExtractor):
-    _VALID_URL = r'http://teamcoco\.com/video/(?P<url_title>.*)'
+    _VALID_URL = r'http://teamcoco\.com/video/(?P<video_id>\d*)?/?(?P<url_title>.*)'
     _TEST = {
         'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
         'file': '19705.mp4',
@@ -26,11 +26,13 @@ class TeamcocoIE(InfoExtractor):
             raise ExtractorError('Invalid URL: %s' % url)
         url_title = mobj.group('url_title')
         webpage = self._download_webpage(url, url_title)
-
-        video_id = self._html_search_regex(
-            r'<article class="video" data-id="(\d+?)"',
-            webpage, 'video id')
-
+        
+        video_id = mobj.group("video_id")
+        if video_id == '':
+            video_id = self._html_search_regex(
+                r'<article class="video" data-id="(\d+?)"',
+                webpage, 'video id')
+        
         self.report_extraction(video_id)
 
         data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id