extracted all of the soundcloud information including description (not tested), need...

[youtube-dl.git] / youtube-dl

diff --git a/youtube-dl b/youtube-dl
index 30a02e5..4861601 100755 (executable)
--- a/youtube-dl
+++ b/youtube-dl
@@ -3481,6 +3481,95 @@ class XVideosIE(InfoExtractor):
                        self._downloader.trouble(u'\nERROR: unable to download ' + video_id)
 
 
+class SoundcloudIE(InformationExtractor):
+       """Information extractor for soundcloud.com
+       To access the media, the uid of the song and a stream token
+       must be extracted from the page source and the script must make
+       a request to media.soundcloud.com/crossdomain.xml. Then
+       the media can be grabbed by requesting from an url composed
+       of the stream token and uid
+     """
+
+       _VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/([\w\d-]+)'
+       IE_NAME = u'soundcloud'
+
+    def __init__(self, downloader=None):
+        InfoExtractor.__init__(self, downloader)
+
+       def report_webpage(self, video_id):
+               """Report information extraction."""
+               self._downloader.to_screen(u'[%s] %s: Downloading webpage' % (self.IE_NAME, video_id))
+
+       def report_extraction(self, video_id):
+               """Report information extraction."""
+               self._downloader.to_screen(u'[%s] %s: Extracting information' % (self.IE_NAME, video_id))
+
+    def _real_initialize(self):
+        return
+
+       def _real_extract(self, url):
+               htmlParser = HTMLParser.HTMLParser()
+
+               mobj = re.match(self._VALID_URL, url)
+               if mobj is None:
+                       self._downloader.trouble(u'ERROR: invalid URL: %s' % url)
+                       return
+
+        # extract uploader (which is in the url)
+        uploader = mobj.group(3).decode('utf-8')
+        # extract simple title (uploader + slug of song title)
+               slug_title =  mobj.group(4).decode('utf-8')
+               simple_title = uploader + '-' + slug_title
+
+               self.report_webpage('%s/%s' % (uploader, slug_title))
+
+               request = urllib2.Request('http://soundcloud.com/%s/%s' % (uploader, slug_title))
+               try:
+                       webpage = urllib2.urlopen(request).read()
+               except (urllib2.URLError, httplib.HTTPException, socket.error), err:
+                       self._downloader.trouble(u'ERROR: unable to download video webpage: %s' % str(err))
+                       return
+
+               self.report_extraction('%s/%s' % (uploader, slug_title))
+
+        # extract uid and access token
+        mobj = re.search('"uid":"([\w\d]+?)".*?stream_token=([\w\d]+)', page)   
+        if mobj:
+            video_id = match.group(1)
+            stream_token = match.group(2)
+
+        # construct media url (with uid/token) to request song
+        mediaURL = "http://media.soundcloud.com/stream/%s?stream_token=%s"
+        mediaURL = mediaURL % (video_id, stream_token)
+
+        # description
+        description = u'No description available'
+        mobj = re.search('track-description-value"><p>(.*?)</p>', page)
+        if mobj:
+            description = mobj.group(1)
+        
+        # upload date
+        mobj = re.search("pretty-date'>on ([\w]+ [\d]+, [\d]+ \d+:\d+)</abbr></h2>", page)
+        if mobj:
+            try:
+                   upload_date = datetime.datetime.strptime(match.group(1), '%B %d, %Y %H:%M').strftime('%Y%m%d')
+            except:
+                pass
+
+        try:
+            self._download.process_info({
+                               'id':           video_id,
+                               'url':          video_url,
+                               'uploader':     uploader,
+                               'upload_date':  upload_date,
+                               'title':        video_title,
+                               'stitle':       simple_title,
+                               'ext':          u'mp3',
+                               'format':       u'NA',
+                               'player_url':   None,
+                'description': description
+            })
+
 class PostProcessor(object):
        """Post Processor class.