Merge branch 'crooksandliars' of https://github.com/fstirlitz/youtube-dl into fstirli...

[youtube-dl] / youtube_dl / extractor / ultimedia.py

diff --git a/youtube_dl/extractor/ultimedia.py b/youtube_dl/extractor/ultimedia.py
index 0c1b08d7dc5d54e1754802dfa3750bda12c4d068..96c809eaf7155290210e0f8b18d3a2c7c948ba97 100644 (file)
--- a/youtube_dl/extractor/ultimedia.py
+++ b/youtube_dl/extractor/ultimedia.py
@@ -1,6 +1,8 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
@@ -40,7 +42,6 @@ class UltimediaIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-
         webpage = self._download_webpage(url, video_id)
 
         deliver_url = self._search_regex(
@@ -60,12 +61,18 @@ class UltimediaIE(InfoExtractor):
             video_id)
 
         quality = qualities(['flash', 'html5'])
-
-        formats = [{
-            'url': mode['config']['file'],
-            'format_id': mode.get('type'),
-            'quality': quality(mode.get('type')),
-        } for mode in player['modes']]
+        formats = []
+        for mode in player['modes']:
+            video_url = mode.get('config', {}).get('file')
+            if not video_url:
+                continue
+            if re.match(r'https?://www\.youtube\.com/.+?', video_url):
+                return self.url_result(video_url, 'Youtube')
+            formats.append({
+                'url': video_url,
+                'format_id': mode.get('type'),
+                'quality': quality(mode.get('type')),
+            })
         self._sort_formats(formats)
 
         thumbnail = player.get('image')
@@ -73,8 +80,8 @@ class UltimediaIE(InfoExtractor):
         title = clean_html((
             self._html_search_regex(
                 r'(?s)<div\s+id="catArticle">.+?</div>(.+?)</h1>',
-                webpage, 'title', default=None)
-            or self._search_regex(
+                webpage, 'title', default=None) or
+            self._search_regex(
                 r"var\s+nameVideo\s*=\s*'([^']+)'",
                 deliver_page, 'title')))