Merge tag 'upstream/2014.10.30'

[youtubedl] / youtube_dl / extractor / tudou.py

diff --git a/youtube_dl/extractor/tudou.py b/youtube_dl/extractor/tudou.py
index 1405b73f76ad5166d45d9a9eb9687c49fa8a0bde..dcd823d0838dca23b27298cbf05ad47cc4261637 100644 (file)
--- a/youtube_dl/extractor/tudou.py
+++ b/youtube_dl/extractor/tudou.py
@@ -1,5 +1,7 @@
 # coding: utf-8
 
+from __future__ import unicode_literals
+
 import re
 import json
 
@@ -7,15 +9,32 @@ from .common import InfoExtractor
 
 
 class TudouIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?tudou\.com/(?:listplay|programs)/(?:view|(.+?))/(?:([^/]+)|([^/]+))(?:\.html)?'
-    _TEST = {
-        u'url': u'http://www.tudou.com/listplay/zzdE77v6Mmo/2xN2duXMxmw.html',
-        u'file': u'159448201.f4v',
-        u'md5': u'140a49ed444bd22f93330985d8475fcb',
-        u'info_dict': {
-            u"title": u"卡马乔国足开大脚长传冲吊集锦"
+    _VALID_URL = r'(?:http://)?(?:www\.)?tudou\.com/(?:listplay|programs|albumplay)/(?:view|(.+?))/(?:([^/]+)|([^/]+))(?:\.html)?'
+    _TESTS = [{
+        'url': 'http://www.tudou.com/listplay/zzdE77v6Mmo/2xN2duXMxmw.html',
+        'md5': '140a49ed444bd22f93330985d8475fcb',
+        'info_dict': {
+            'id': '159448201',
+            'ext': 'f4v',
+            'title': '卡马乔国足开大脚长传冲吊集锦',
+            'thumbnail': 're:^https?://.*\.jpg$',
+        }
+    }, {
+        'url': 'http://www.tudou.com/programs/view/ajX3gyhL0pc/',
+        'info_dict': {
+            'id': '117049447',
+            'ext': 'f4v',
+            'title': 'La Sylphide-Bolshoi-Ekaterina Krysanova & Vyacheslav Lopatin 2012',
+            'thumbnail': 're:^https?://.*\.jpg$',
         }
-    }
+    }, {
+        '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)
@@ -29,20 +48,27 @@ class TudouIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group(2)
         webpage = self._download_webpage(url, video_id)
-        title = re.search(",kw:\"(.+)\"",webpage)
-        if title is None:
-            title = re.search(",kw: \'(.+)\'",webpage)
-        title = title.group(1)
-        thumbnail_url = re.search(",pic: \'(.+?)\'",webpage)
-        if thumbnail_url is None:
-            thumbnail_url = re.search(",pic:\"(.+?)\"",webpage)
-        thumbnail_url = thumbnail_url.group(1)
+
+        m = re.search(r'vcode:\s*[\'"](.+?)[\'"]', webpage)
+        if m and m.group(1):
+            return {
+                '_type': 'url',
+                'url': 'youku:' + m.group(1),
+                'ie_key': 'Youku'
+            }
+
+        title = self._search_regex(
+            r",kw:\s*['\"](.+?)[\"']", webpage, 'title')
+        thumbnail_url = self._search_regex(
+            r",pic:\s*[\"'](.+?)[\"']", webpage, 'thumbnail URL', fatal=False)
 
         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
         # the hd field in the request url, we pick the higher
-        quality = sorted(segments.keys())[-1]
+        # Also, filter non-number qualities (see issue #3643).
+        quality = sorted(filter(lambda k: k.isdigit(), segments.keys()),
+                         key=lambda k: int(k))[-1]
         parts = segments[quality]
         result = []
         len_parts = len(parts)
@@ -52,12 +78,13 @@ class TudouIE(InfoExtractor):
             part_id = part['k']
             final_url = self._url_for_id(part_id, quality)
             ext = (final_url.split('?')[0]).split('.')[-1]
-            part_info = {'id': part_id,
-                          'url': final_url,
-                          'ext': ext,
-                          'title': title,
-                          'thumbnail': thumbnail_url,
-                          }
+            part_info = {
+                'id': '%s' % part_id,
+                'url': final_url,
+                'ext': ext,
+                'title': title,
+                'thumbnail': thumbnail_url,
+            }
             result.append(part_info)
 
         return result