Fixed download from Dailymotion.
authorknagano <knagano@sodan.org>
Sat, 7 May 2011 13:53:37 +0000 (22:53 +0900)
committerknagano <knagano@sodan.org>
Sat, 7 May 2011 13:53:37 +0000 (22:53 +0900)
Fetches FLV URL from "sdURL" in addVariable("sequence") JSON,
instead of addVariable("video") which doesnot exist today.
Supports new title, uploader nickname format.

youtube-dl

index 3ac27a8570eeebd5889efbe9111a324f751dc01d..377ceff588241ccdb37bb19f972648cc898d5873 100755 (executable)
@@ -1320,6 +1320,7 @@ class DailymotionIE(InfoExtractor):
 
                # Retrieve video webpage to extract further information
                request = urllib2.Request(url)
+               request.add_header('Cookie', 'family_filter=off')
                try:
                        self.report_download_webpage(video_id)
                        webpage = urllib2.urlopen(request).read()
@@ -1329,25 +1330,29 @@ class DailymotionIE(InfoExtractor):
 
                # Extract URL, uploader and title from webpage
                self.report_extraction(video_id)
-               mobj = re.search(r'(?i)addVariable\(\"video\"\s*,\s*\"([^\"]*)\"\)', webpage)
+               mobj = re.search(r'(?i)addVariable\(\"sequence\"\s*,\s*\"([^\"]+?)\"\)', webpage)
                if mobj is None:
                        self._downloader.trouble(u'ERROR: unable to extract media URL')
                        return
-               mediaURL = urllib.unquote(mobj.group(1))
+               sequence = urllib.unquote(mobj.group(1))
+               mobj = re.search(r',\"sdURL\"\:\"([^\"]+?)\",', sequence)
+               if mobj is None:
+                       self._downloader.trouble(u'ERROR: unable to extract media URL')
+                       return
+               mediaURL = urllib.unquote(mobj.group(1)).replace('\\', '')
 
                # if needed add http://www.dailymotion.com/ if relative URL
 
                video_url = mediaURL
 
-               # '<meta\s+name="title"\s+content="Dailymotion\s*[:\-]\s*(.*?)"\s*\/\s*>'
-               mobj = re.search(r'(?im)<title>Dailymotion\s*[\-:]\s*(.+?)</title>', webpage)
+               mobj = re.search(r'(?im)<title>Dailymotion\s*-\s*(.+)\s*-\s*[^<]+?</title>', webpage)
                if mobj is None:
                        self._downloader.trouble(u'ERROR: unable to extract title')
                        return
                video_title = mobj.group(1).decode('utf-8')
                video_title = sanitize_title(video_title)
 
-               mobj = re.search(r'(?im)<Attribute name="owner">(.+?)</Attribute>', webpage)
+               mobj = re.search(r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a></span>', webpage)
                if mobj is None:
                        self._downloader.trouble(u'ERROR: unable to extract uploader nickname')
                        return