]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/tudou.py
Initiate a new release.
[youtubedl] / youtube_dl / extractor / tudou.py
index 9ca860ab084f523c7dc5ed7b16b28ff0e91c3324..c89de5ba4a46bb261987d8dbee5f55b3d05492da 100644 (file)
@@ -1,32 +1,87 @@
+# coding: utf-8
+
+from __future__ import unicode_literals
+
 import re
+import json
 
 from .common import InfoExtractor
 
 
 class TudouIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?tudou\.com/(?:listplay|programs)/(?:view|(.+?))/(?:([^/]+)|([^/]+)\.html)'
+    _VALID_URL = r'https?://(?:www\.)?tudou\.com/(?:listplay|programs(?:/view)?|albumplay)/.*?/(?P<id>[^/?#]+?)(?:\.html)?/?(?:$|[?#])'
+    _TESTS = [{
+        'url': 'http://www.tudou.com/listplay/zzdE77v6Mmo/2xN2duXMxmw.html',
+        'md5': '140a49ed444bd22f93330985d8475fcb',
+        'info_dict': {
+            'id': '159448201',
+            'ext': 'f4v',
+            'title': '卡马乔国足开大脚长传冲吊集锦',
+            'thumbnail': 're:^https?://.*\.jpg$',
+        }
+    }, {
+        '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$',
+        }
+    }]
+
+    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 = self._match_id(url)
         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': '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
+        # 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)
+        if len_parts > 1:
+            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)
+            ext = (final_url.split('?')[0]).split('.')[-1]
+            part_info = {
+                'id': '%s' % part_id,
+                'url': final_url,
+                'ext': ext,
+                'title': title,
+                'thumbnail': thumbnail_url,
+            }
+            result.append(part_info)
+
+        return {
+            '_type': 'multi_video',
+            'entries': result,
+            'id': video_id,
+            'title': title,
+        }