]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/tudou.py
Merge tag 'upstream/2016.02.22'
[youtubedl] / youtube_dl / extractor / tudou.py
index c89de5ba4a46bb261987d8dbee5f55b3d05492da..f56b66d06f2f7323b60d2fc4788d26cb3274ef9e 100644 (file)
@@ -2,14 +2,18 @@
 
 from __future__ import unicode_literals
 
-import re
-import json
-
 from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    int_or_none,
+    float_or_none,
+    unescapeHTML,
+)
 
 
 class TudouIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?tudou\.com/(?:listplay|programs(?:/view)?|albumplay)/.*?/(?P<id>[^/?#]+?)(?:\.html)?/?(?:$|[?#])'
+    IE_NAME = 'tudou'
+    _VALID_URL = r'https?://(?:www\.)?tudou\.com/(?:(?:programs|wlplay)/view|(?:listplay|albumplay)/[\w-]{11})/(?P<id>[\w-]{11})'
     _TESTS = [{
         'url': 'http://www.tudou.com/listplay/zzdE77v6Mmo/2xN2duXMxmw.html',
         'md5': '140a49ed444bd22f93330985d8475fcb',
@@ -18,6 +22,11 @@ class TudouIE(InfoExtractor):
             'ext': 'f4v',
             'title': '卡马乔国足开大脚长传冲吊集锦',
             'thumbnail': 're:^https?://.*\.jpg$',
+            'timestamp': 1372113489000,
+            'description': '卡马乔卡家军,开大脚先进战术不完全集锦!',
+            'duration': 289.04,
+            'view_count': int,
+            'filesize': int,
         }
     }, {
         'url': 'http://www.tudou.com/programs/view/ajX3gyhL0pc/',
@@ -26,36 +35,40 @@ class TudouIE(InfoExtractor):
             'ext': 'f4v',
             'title': 'La Sylphide-Bolshoi-Ekaterina Krysanova & Vyacheslav Lopatin 2012',
             'thumbnail': 're:^https?://.*\.jpg$',
+            'timestamp': 1349207518000,
+            'description': 'md5:294612423894260f2dcd5c6c04fe248b',
+            'duration': 5478.33,
+            'view_count': int,
+            'filesize': int,
         }
     }]
 
-    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, 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):
         video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-
-        m = re.search(r'vcode:\s*[\'"](.+?)[\'"]', webpage)
-        if m and m.group(1):
-            return {
-                '_type': 'url',
-                'url': 'youku:' + m.group(1),
-                'ie_key': 'Youku'
-            }
+        item_data = self._download_json(
+            'http://www.tudou.com/tvp/getItemInfo.action?ic=%s' % video_id, video_id)
 
-        title = self._search_regex(
-            r",kw:\s*['\"](.+?)[\"']", webpage, 'title')
-        thumbnail_url = self._search_regex(
-            r",pic:\s*[\"'](.+?)[\"']", webpage, 'thumbnail URL', fatal=False)
+        youku_vcode = item_data.get('vcode')
+        if youku_vcode:
+            return self.url_result('youku:' + youku_vcode, ie='Youku')
 
-        segs_json = self._search_regex(r'segs: \'(.*)\'', webpage, 'segments')
-        segments = json.loads(segs_json)
+        title = unescapeHTML(item_data['kw'])
+        description = item_data.get('desc')
+        thumbnail_url = item_data.get('pic')
+        view_count = int_or_none(item_data.get('playTimes'))
+        timestamp = int_or_none(item_data.get('pt'))
+
+        segments = self._parse_json(item_data['itemSegs'], 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).
@@ -76,6 +89,14 @@ class TudouIE(InfoExtractor):
                 'ext': ext,
                 'title': title,
                 'thumbnail': thumbnail_url,
+                'description': description,
+                'view_count': view_count,
+                'timestamp': timestamp,
+                'duration': float_or_none(part.get('seconds'), 1000),
+                'filesize': int_or_none(part.get('size')),
+                'http_headers': {
+                    'Referer': self._PLAYER_URL,
+                },
             }
             result.append(part_info)
 
@@ -85,3 +106,47 @@ class TudouIE(InfoExtractor):
             'id': video_id,
             'title': title,
         }
+
+
+class TudouPlaylistIE(InfoExtractor):
+    IE_NAME = 'tudou:playlist'
+    _VALID_URL = r'https?://(?:www\.)?tudou\.com/listplay/(?P<id>[\w-]{11})\.html'
+    _TESTS = [{
+        'url': 'http://www.tudou.com/listplay/zzdE77v6Mmo.html',
+        'info_dict': {
+            'id': 'zzdE77v6Mmo',
+        },
+        'playlist_mincount': 209,
+    }]
+
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+        playlist_data = self._download_json(
+            'http://www.tudou.com/tvp/plist.action?lcode=%s' % playlist_id, playlist_id)
+        entries = [self.url_result(
+            'http://www.tudou.com/programs/view/%s' % item['icode'],
+            'Tudou', item['icode'],
+            item['kw']) for item in playlist_data['items']]
+        return self.playlist_result(entries, playlist_id)
+
+
+class TudouAlbumIE(InfoExtractor):
+    IE_NAME = 'tudou:album'
+    _VALID_URL = r'https?://(?:www\.)?tudou\.com/album(?:cover|play)/(?P<id>[\w-]{11})'
+    _TESTS = [{
+        'url': 'http://www.tudou.com/albumplay/v5qckFJvNJg.html',
+        'info_dict': {
+            'id': 'v5qckFJvNJg',
+        },
+        'playlist_mincount': 45,
+    }]
+
+    def _real_extract(self, url):
+        album_id = self._match_id(url)
+        album_data = self._download_json(
+            'http://www.tudou.com/tvp/alist.action?acode=%s' % album_id, album_id)
+        entries = [self.url_result(
+            'http://www.tudou.com/programs/view/%s' % item['icode'],
+            'Tudou', item['icode'],
+            item['kw']) for item in album_data['items']]
+        return self.playlist_result(entries, album_id)