Merge branch 'master' of https://github.com/linhua55/youtube-dl into linhua55-master

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

diff --git a/youtube_dl/extractor/tudou.py b/youtube_dl/extractor/tudou.py
index 3007b136f4f696eaff7eb332d00559e859841bb8..e800477e2c1014593df0763b9806c21ff73dd3f9 100644 (file)
--- a/youtube_dl/extractor/tudou.py
+++ b/youtube_dl/extractor/tudou.py
@@ -9,7 +9,7 @@ from .common import InfoExtractor
 
 
 class TudouIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?tudou\.com/(?:listplay|programs|albumplay)/(?:view|(.+?))/(?:([^/]+)|([^/]+))(?:\.html)?'
+    _VALID_URL = r'https?://(?:www\.)?tudou\.com/(?:listplay|programs(?:/view)?|albumplay)/?.*/(?P<id>[^/?#]+?)(?:\.html)?/?(?:$|[?#])'
     _TESTS = [{
         'url': 'http://www.tudou.com/listplay/zzdE77v6Mmo/2xN2duXMxmw.html',
         'md5': '140a49ed444bd22f93330985d8475fcb',
@@ -27,17 +27,12 @@ class TudouIE(InfoExtractor):
             'title': 'La Sylphide-Bolshoi-Ekaterina Krysanova & Vyacheslav Lopatin 2012',
             'thumbnail': 're:^https?://.*\.jpg$',
         }
-    }, {
-        'url': 'http://www.tudou.com/albumplay/TenTw_JgiPM/PzsAs5usU9A.html',
-        'info_dict': {
-            'title': 'todo.mp4',
-        },
-        'add_ie': ['Youku'],
-        'skip': 'Only works from China'
     }]
 
-    def _url_for_id(self, id, quality = None):
-        info_url = "http://v2.tudou.com/f?id=" +str(id)
+    _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)
         if quality:
             info_url += '&hd' + quality
         webpage = self._download_webpage(info_url, id, "Opening the info webpage")
@@ -45,8 +40,7 @@ class TudouIE(InfoExtractor):
         return final_url
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(2)
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
         m = re.search(r'vcode:\s*[\'"](.+?)[\'"]', webpage)
@@ -62,6 +56,10 @@ class TudouIE(InfoExtractor):
         thumbnail_url = self._search_regex(
             r",pic:\s*[\"'](.+?)[\"']", webpage, 'thumbnail URL', fatal=False)
 
+        player_url = self._search_regex(
+            r"playerUrl\s*:\s*['\"](.+?\.swf)[\"']",
+            webpage, 'player URL', default=self._PLAYER_URL)
+
         segs_json = self._search_regex(r'segs: \'(.*)\'', webpage, 'segments')
         segments = json.loads(segs_json)
         # It looks like the keys are the arguments that have to be passed as
@@ -73,7 +71,7 @@ class TudouIE(InfoExtractor):
         result = []
         len_parts = len(parts)
         if len_parts > 1:
-            self.to_screen(u'%s: found %s parts' % (video_id, len_parts))
+            self.to_screen('%s: found %s parts' % (video_id, len_parts))
         for part in parts:
             part_id = part['k']
             final_url = self._url_for_id(part_id, quality)
@@ -84,7 +82,15 @@ class TudouIE(InfoExtractor):
                 'ext': ext,
                 'title': title,
                 'thumbnail': thumbnail_url,
+                'http_headers': {
+                    'Referer': player_url,
+                },
             }
             result.append(part_info)
 
-        return result
+        return {
+            '_type': 'multi_video',
+            'entries': result,
+            'id': video_id,
+            'title': title,
+        }