X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Ftudou.py;h=5f7ac4b35b6c4576e5fb0998d56eb21ec522b58a;hb=dcdfeb33d2666ca07d2fb73ebf9284a77e714f69;hp=c9d80a7efd97f2a3670560ca6d3c0a8645671247;hpb=141ba36996f77a420df69903a59792f6f93ae314;p=youtube-dl

diff --git a/youtube_dl/extractor/tudou.py b/youtube_dl/extractor/tudou.py
index c9d80a7ef..5f7ac4b35 100644
--- a/youtube_dl/extractor/tudou.py
+++ b/youtube_dl/extractor/tudou.py
@@ -3,6 +3,7 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..compat import compat_str
 
 
 class TudouIE(InfoExtractor):
@@ -31,12 +32,12 @@ class TudouIE(InfoExtractor):
 
     _PLAYER_URL = 'http://js.tudouui.com/bin/lingtong/PortalPlayer_177.swf'
 
-    def _url_for_id(self, id, quality=None):
-        info_url = "http://v2.tudou.com/f?id=" + str(id)
+    def _url_for_id(self, video_id, quality=None):
+        info_url = 'http://v2.tudou.com/f?id=' + compat_str(video_id)
         if quality:
             info_url += '&hd' + quality
-        webpage = self._download_webpage(info_url, id, "Opening the info webpage")
-        final_url = self._html_search_regex('>(.+?)</f>', webpage, 'video url')
+        xml_data = self._download_xml(info_url, video_id, "Opening the info XML page")
+        final_url = xml_data.text
         return final_url
 
     def _real_extract(self, url):
@@ -44,21 +45,21 @@ class TudouIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         youku_vcode = self._search_regex(
-            r'vcode:\s*[\'"](.+?)[\'"]', webpage, 'youku vcode', default=None)
+            r'vcode\s*:\s*[\'"]([^\'"]*)[\'"]', webpage, 'youku vcode', default=None)
         if youku_vcode:
             return self.url_result('youku:' + youku_vcode, ie='Youku')
 
         title = self._search_regex(
-            r",kw:\s*['\"](.+?)[\"']", webpage, 'title')
+            r',kw\s*:\s*[\'"]([^\'"]+)[\'"]', webpage, 'title')
         thumbnail_url = self._search_regex(
-            r",pic:\s*[\"'](.+?)[\"']", webpage, 'thumbnail URL', fatal=False)
+            r',pic\s*:\s*[\'"]([^\'"]+)[\'"]', webpage, 'thumbnail URL', fatal=False)
 
         player_url = self._search_regex(
-            r"playerUrl\s*:\s*['\"](.+?\.swf)[\"']",
+            r'playerUrl\s*:\s*[\'"]([^\'"]+\.swf)[\'"]',
             webpage, 'player URL', default=self._PLAYER_URL)
 
         segments = self._parse_json(self._search_regex(
-            r'segs: \'(.*)\'', webpage, 'segments'), video_id)
+            r'segs: \'([^\']+)\'', webpage, 'segments'), video_id)
         # It looks like the keys are the arguments that have to be passed as
         # the hd field in the request url, we pick the higher
         # Also, filter non-number qualities (see issue #3643).