debian/NEWS: Update with notes about split audio/video downloads.

[youtubedl] / youtube_dl / extractor / ted.py

diff --git a/youtube_dl/extractor/ted.py b/youtube_dl/extractor/ted.py
index ccad1c7c2087b533712b62aedaed43abcfa0bf90..dfa1176a3e4e4eef333dcb829773c189bf9916ba 100644 (file)
--- a/youtube_dl/extractor/ted.py
+++ b/youtube_dl/extractor/ted.py
@@ -14,6 +14,15 @@ class TEDIE(InfoExtractor):
                    (/lang/(.*?))? # The url may contain the language
                    /(?P<name>\w+) # Here goes the name and then ".html"
                    '''
                    (/lang/(.*?))? # The url may contain the language
                    /(?P<name>\w+) # Here goes the name and then ".html"
                    '''

+    _TEST = {
+        u'url': u'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html',
+        u'file': u'102.mp4',
+        u'md5': u'2d76ee1576672e0bd8f187513267adf6',
+        u'info_dict': {
+            u"description": u"md5:c6fa72e6eedbd938c9caf6b2702f5922", 
+            u"title": u"Dan Dennett: The illusion of consciousness"
+        }
+    }

 
     @classmethod
     def suitable(cls, url):
 
     @classmethod
     def suitable(cls, url):


@@ -58,7 +67,7 @@ class TEDIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id, 'Downloading \"%s\" page' % video_name)
         self.report_extraction(video_name)
         # If the url includes the language we get the title translated
         webpage = self._download_webpage(url, video_id, 'Downloading \"%s\" page' % video_name)
         self.report_extraction(video_name)
         # If the url includes the language we get the title translated

-        title = self._html_search_regex(r'<span id="altHeadline" >(?P<title>.*)</span>',
+        title = self._html_search_regex(r'<span .*?id="altHeadline".+?>(?P<title>.*)</span>',

                                         webpage, 'title')
         json_data = self._search_regex(r'<script.*?>var talkDetails = ({.*?})</script>',
                                     webpage, 'json data')
                                         webpage, 'title')
         json_data = self._search_regex(r'<script.*?>var talkDetails = ({.*?})</script>',
                                     webpage, 'json data')


@@ -68,12 +77,20 @@ class TEDIE(InfoExtractor):
         
         thumbnail = self._search_regex(r'</span>[\s.]*</div>[\s.]*<img src="(.*?)"',
                                        webpage, 'thumbnail')
         
         thumbnail = self._search_regex(r'</span>[\s.]*</div>[\s.]*<img src="(.*?)"',
                                        webpage, 'thumbnail')

+        formats = [{
+            'ext': 'mp4',
+            'url': stream['file'],
+            'format': stream['id']
+            } for stream in info['htmlStreams']]

         info = {
         info = {

-                'id': info['id'],
-                'url': info['htmlStreams'][-1]['file'],
-                'ext': 'mp4',
-                'title': title,
-                'thumbnail': thumbnail,
-                'description': desc,
-                }
+            'id': info['id'],
+            'title': title,
+            'thumbnail': thumbnail,
+            'description': desc,
+            'formats': formats,
+        }
+
+        # TODO: Remove when #980 has been merged
+        info.update(info['formats'][-1])
+

         return info
         return info