X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/feb5020b37d7d3ba4005a8bac6f4efece4ce4b8c..8524e382a76e91d49486a78ba2b961789d29fa1d:/youtube_dl/extractor/tudou.py?ds=inline

diff --git a/youtube_dl/extractor/tudou.py b/youtube_dl/extractor/tudou.py
index 4681a6f..7a3891b 100644
--- a/youtube_dl/extractor/tudou.py
+++ b/youtube_dl/extractor/tudou.py
@@ -1,40 +1,78 @@
+# coding: utf-8
+
 import re
+import json
 
 from .common import InfoExtractor
 
 
 class TudouIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?tudou\.com/(?:listplay|programs)/(?:view|(.+?))/(?:([^/]+)|([^/]+)\.html)'
-    _TEST = {
+    _VALID_URL = r'(?:http://)?(?:www\.)?tudou\.com/(?:listplay|programs|albumplay)/(?:view|(.+?))/(?:([^/]+)|([^/]+))(?:\.html)?'
+    _TESTS = [{
         u'url': u'http://www.tudou.com/listplay/zzdE77v6Mmo/2xN2duXMxmw.html',
-        u'file': u'159447792.f4v',
-        u'md5': u'ad7c358a01541e926a1e413612c6b10a',
+        u'file': u'159448201.f4v',
+        u'md5': u'140a49ed444bd22f93330985d8475fcb',
         u'info_dict': {
-            u"title": u"\u5361\u9a6c\u4e54\u56fd\u8db3\u5f00\u5927\u811a\u957f\u4f20\u51b2\u540a\u96c6\u9526"
+            u"title": u"卡马乔国足开大脚长传冲吊集锦"
         }
-    }
+    },
+    {
+        u'url': u'http://www.tudou.com/albumplay/TenTw_JgiPM/PzsAs5usU9A.html',
+        u'file': u'todo.mp4',
+        u'md5': u'todo.mp4',
+        u'info_dict': {
+            u'title': u'todo.mp4',
+        },
+        u'add_ie': [u'Youku'],
+        u'skip': u'Only works from China'
+    }]
+
+    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")
+        final_url = self._html_search_regex('>(.+?)</f>',webpage, 'video url')
+        return final_url
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(2).replace('.html','')
+        video_id = mobj.group(2)
         webpage = self._download_webpage(url, video_id)
-        video_id = re.search('"k":(.+?),',webpage).group(1)
-        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)
-        info_url = "http://v2.tudou.com/f?id="+str(video_id)
-        webpage = self._download_webpage(info_url, video_id, "Opening the info webpage")
-        final_url = re.search('\>(.+?)\<\/f\>',webpage).group(1)
-        ext = (final_url.split('?')[0]).split('.')[-1]
-        return [{
-            'id':        video_id,
-            'url':       final_url,
-            'ext':       ext,
-            'title':     title,
-            'thumbnail': thumbnail_url,
-        }]
+
+        m = re.search(r'vcode:\s*[\'"](.+?)[\'"]', webpage)
+        if m and m.group(1):
+            return {
+                '_type': 'url',
+                'url': u'youku:' + m.group(1),
+                'ie_key': 'Youku'
+            }
+
+        title = self._search_regex(
+            r",kw:\s*['\"](.+?)[\"']", webpage, u'title')
+        thumbnail_url = self._search_regex(
+            r",pic:\s*[\"'](.+?)[\"']", webpage, u'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]
+        parts = segments[quality]
+        result = []
+        len_parts = len(parts)
+        if len_parts > 1:
+            self.to_screen(u'%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)
+            ext = (final_url.split('?')[0]).split('.')[-1]
+            part_info = {'id': part_id,
+                          'url': final_url,
+                          'ext': ext,
+                          'title': title,
+                          'thumbnail': thumbnail_url,
+                          }
+            result.append(part_info)
+
+        return result