[UstreamIE] [generic] Added support for Ustream embed URLs (Fixes #2694)

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 6a2ce0d6dba51019b270b23a21b345c4a5e104e5..6d3e4fb102a6d0435276d3faee5ea27a88285739 100644 (file)
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -184,6 +184,15 @@ class GenericIE(InfoExtractor):
                 'description': 'md5:ddb2a40ecd6b6a147e400e535874947b',
             }
         },
+        # Embeded Ustream video
+        {
+            'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
+            'md5': '27b99cdb639c9b12a79bca876a073417',
+            'info_dict': {
+                'uploader': 'AU SPA: The NSA and Privacy',
+                'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
+            }
+        },
         # nowvideo embed hidden behind percent encoding
         {
             'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
@@ -556,6 +565,12 @@ class GenericIE(InfoExtractor):
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'TED')
 
+        # Look for embedded Ustream videos
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'), 'Ustream')
+
         # Look for embedded arte.tv player
         mobj = re.search(
             r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',

diff --git a/youtube_dl/extractor/ustream.py b/youtube_dl/extractor/ustream.py
index 7fa2b9e159ed1a60c056140f05f51851663830e9..d1d2af19b8870353e66cbf227727e54e5bafb0e7 100644 (file)
--- a/youtube_dl/extractor/ustream.py
+++ b/youtube_dl/extractor/ustream.py
@@ -11,9 +11,9 @@ from ..utils import (
 
 
 class UstreamIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.ustream\.tv/recorded/(?P<videoID>\d+)'
+    _VALID_URL = r'https?://www\.ustream\.tv/(?P<type>recorded|embed)/(?P<videoID>\d+)'
     IE_NAME = 'ustream'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.ustream.tv/recorded/20274954',
         'file': '20274954.flv',
         'md5': '088f151799e8f572f84eb62f17d73e5c',
@@ -21,10 +21,27 @@ class UstreamIE(InfoExtractor):
             "uploader": "Young Americans for Liberty",
             "title": "Young Americans for Liberty February 7, 2012 2:28 AM",
         },
+    },
+    {
+        'url': 'http://www.ustream.tv/embed/17357891',
+        'file': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman-45734260.flv',
+        'md5': '27b99cdb639c9b12a79bca876a073417',
+        'info_dict': {
+            "uploader": "AU SPA: The NSA and Privacy",
+            "title": "NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman"
+    },
     }
+    ]
 
     def _real_extract(self, url):
         m = re.match(self._VALID_URL, url)
+        if m.group('type') == 'embed':
+            video_id = m.group('videoID')
+            webpage = self._download_webpage(url, video_id)
+            desktop_video_id = self._html_search_regex(r'ContentVideoIds=\["([^"]*?)"\]', webpage, 'desktop_video_id')
+            desktop_url = 'http://www.ustream.tv/recorded/' + desktop_video_id
+            return self.url_result(desktop_url, 'Ustream')
+
         video_id = m.group('videoID')
 
         video_url = 'http://tcdn.ustream.tv/video/%s' % video_id