Extract the base_url for the XML download from the JS snippet's 'server' variable.
authorElias Probst <mail@eliasprobst.eu>
Fri, 20 Jun 2014 23:00:48 +0000 (01:00 +0200)
committerElias Probst <mail@eliasprobst.eu>
Fri, 20 Jun 2014 23:00:48 +0000 (01:00 +0200)
youtube_dl/extractor/spiegel.py

index 94346daf6cc4ae1a09a05a8524ea3b3beccca672..87f1ee69462600f57c0bde46784307cb6e3e0963 100644 (file)
@@ -33,7 +33,13 @@ class SpiegelIE(InfoExtractor):
         video_title = self._html_search_regex(
             r'<div class="module-title">(.*?)</div>', webpage, 'title')
 
-        xml_url = 'http://video2.spiegel.de/flash/' + video_id + '.xml'
+        base_url = self._search_regex(
+            r'var\s+server\s+=\s+\"(http://video\d*\.spiegel\.de/flash/\d+/\d+/)\";',
+            webpage,
+            'base_url',
+        )
+
+        xml_url = base_url + video_id + '.xml'
         idoc = self._download_xml(
             xml_url, video_id,
             note='Downloading XML', errnote='Failed to download XML from "{0}"'.format(xml_url))