Merge remote-tracking branch 'glisignoli/master'

diff --git a/test/tests.json b/test/tests.json
index 5c46af2c8896af2a0bc58ca8fd0f827f7ee2e0e5..a3c31ae51e0646cef3315258f9a82e7d80995273 100644 (file)
--- a/test/tests.json
+++ b/test/tests.json
@@ -286,5 +286,23 @@
       "title": "test chars: \"'/\\ä<>This is a test video for youtube-dl.For more information, contact phihag@phihag.de ."
     }
 
+  },
+  {
+    "name": "TED",
+    "url": "http://www.ted.com/talks/dan_dennett_on_our_consciousness.html",
+    "file": "102.mp4",
+    "md5": "7bc087e71d16f18f9b8ab9fa62a8a031",
+    "info_dict": {
+        "title": "Dan Dennett: The illusion of consciousness"
+    }
+  },
+  {
+    "name": "MySpass",
+    "url": "http://www.myspass.de/myspass/shows/tvshows/absolute-mehrheit/Absolute-Mehrheit-vom-17022013-Die-Highlights-Teil-2--/11741/",
+    "file": "11741.mp4",
+    "md5": "0b49f4844a068f8b33f4b7c88405862b",
+    "info_dict": {
+        "title": "Absolute Mehrheit vom 17.02.2013 - Die Highlights, Teil 2"
+    }
   }
 ]

diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py
index ac69f82fe95345dbb67e7441ed3f66de4627e613..fe9bd97d0070dfbbee5b0b6904d048ab1935a5b2 100755 (executable)
--- a/youtube_dl/InfoExtractors.py
+++ b/youtube_dl/InfoExtractors.py
@@ -3968,6 +3968,86 @@ class KeekIE(InfoExtractor):
         }
         return [info]
 
+class TEDIE(InfoExtractor):
+    _VALID_URL=r'http://www.ted.com/talks/(?P<videoName>\w+)'
+    def _real_extract(self, url):
+        m=re.match(self._VALID_URL, url)
+        videoName=m.group('videoName')
+        webpage=self._download_webpage(url, 0, 'Downloading \"%s\" page' % videoName)
+        #If the url includes the language we get the title translated
+        title_RE=r'<h1><span id="altHeadline" >(?P<title>[\s\w:/\.\?=\+-\\\']*)</span></h1>'
+        title=re.search(title_RE, webpage).group('title')
+        info_RE=r'''<script\ type="text/javascript">var\ talkDetails\ =(.*?)
+                        "id":(?P<videoID>[\d]+).*?
+                        "mediaSlug":"(?P<mediaSlug>[\w\d]+?)"'''
+        info_match=re.search(info_RE,webpage,re.VERBOSE)
+        video_id=info_match.group('videoID')
+        mediaSlug=info_match.group('mediaSlug')
+        video_url='http://download.ted.com/talks/%s.mp4' % mediaSlug
+        info = {
+                'id':video_id,
+                'url':video_url,
+                'ext': 'mp4',
+                'title': title
+        }
+        return [info]
+
+class MySpassIE(InfoExtractor):
+    _VALID_URL = r'http://www.myspass.de/.*'
+    
+    def _real_extract(self, url):
+        META_DATA_URL_TEMPLATE = 'http://www.myspass.de/myspass/includes/apps/video/getvideometadataxml.php?id=%s'
+
+        # video id is the last path element of the URL
+        # usually there is a trailing slash, so also try the second but last
+        url_path = compat_urllib_parse_urlparse(url).path
+        url_parent_path, video_id = os.path.split(url_path)
+        if not video_id:
+            _, video_id = os.path.split(url_parent_path)
+        
+        # get metadata
+        metadata_url = META_DATA_URL_TEMPLATE % video_id
+        metadata_text = self._download_webpage(metadata_url, video_id)
+        metadata = xml.etree.ElementTree.fromstring(metadata_text.encode('utf-8'))
+        
+        # extract values from metadata
+        url_flv_el = metadata.find('url_flv')
+        if url_flv_el is None:
+            self._downloader.trouble(u'ERROR: unable to extract download url')
+            return
+        video_url = url_flv_el.text
+        extension = os.path.splitext(video_url)[1][1:]
+        title_el = metadata.find('title')
+        if title_el is None:
+            self._downloader.trouble(u'ERROR: unable to extract title')
+            return
+        title = title_el.text
+        format_id_el = metadata.find('format_id')
+        if format_id_el is None:
+            format = ext
+        else:
+            format = format_id_el.text
+        description_el = metadata.find('description')
+        if description_el is not None:
+            description = description_el.text
+        else:
+            description = None
+        imagePreview_el = metadata.find('imagePreview')
+        if imagePreview_el is not None:
+            thumbnail = imagePreview_el.text
+        else:
+            thumbnail = None
+        info = {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'ext': extension,
+            'format': format,
+            'thumbnail': thumbnail,
+            'description': description
+        }
+        return [info]
+
 def gen_extractors():
     """ Return a list of an instance of every supported extractor.
     The order does matter; the first extractor matched is the one handling the URL.
@@ -4015,6 +4095,8 @@ def gen_extractors():
         RBMARadioIE(),
         EightTracksIE(),
         KeekIE(),
+        TEDIE(),
+        MySpassIE(),
         GenericIE()
     ]