161e47624b383dfe76ea1e2ea6ec73395a97db53
[ytdl] / youtube_dl / extractor / tudou.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5 import re
6 import json
7
8 from .common import InfoExtractor
9
10
11 class TudouIE(InfoExtractor):
12     _VALID_URL = r'(?:http://)?(?:www\.)?tudou\.com/(?:listplay|programs|albumplay)/(?:view|(.+?))/(?:([^/]+)|([^/]+))(?:\.html)?'
13     _TESTS = [{
14         'url': 'http://www.tudou.com/listplay/zzdE77v6Mmo/2xN2duXMxmw.html',
15         'md5': '140a49ed444bd22f93330985d8475fcb',
16         'info_dict': {
17             'id': '159448201',
18             'ext': 'f4v',
19             'title': '卡马乔国足开大脚长传冲吊集锦',
20             'thumbnail': 're:^https?://.*\.jpg$',
21         }
22     }, {
23         'url': 'http://www.tudou.com/programs/view/ajX3gyhL0pc/',
24         'info_dict': {
25             'id': '117049447',
26             'ext': 'f4v',
27             'title': 'La Sylphide-Bolshoi-Ekaterina Krysanova & Vyacheslav Lopatin 2012',
28             'thumbnail': 're:^https?://.*\.jpg$',
29         }
30     }, {
31         'url': 'http://www.tudou.com/albumplay/TenTw_JgiPM/PzsAs5usU9A.html',
32         'info_dict': {
33             'title': 'todo.mp4',
34         },
35         'add_ie': ['Youku'],
36         'skip': 'Only works from China'
37     }]
38
39     def _url_for_id(self, id, quality=None):
40         info_url = "http://v2.tudou.com/f?id=" + str(id)
41         if quality:
42             info_url += '&hd' + quality
43         webpage = self._download_webpage(info_url, id, "Opening the info webpage")
44         final_url = self._html_search_regex('>(.+?)</f>', webpage, 'video url')
45         return final_url
46
47     def _real_extract(self, url):
48         mobj = re.match(self._VALID_URL, url)
49         video_id = mobj.group(2)
50         webpage = self._download_webpage(url, video_id)
51
52         m = re.search(r'vcode:\s*[\'"](.+?)[\'"]', webpage)
53         if m and m.group(1):
54             return {
55                 '_type': 'url',
56                 'url': 'youku:' + m.group(1),
57                 'ie_key': 'Youku'
58             }
59
60         title = self._search_regex(
61             r",kw:\s*['\"](.+?)[\"']", webpage, 'title')
62         thumbnail_url = self._search_regex(
63             r",pic:\s*[\"'](.+?)[\"']", webpage, 'thumbnail URL', fatal=False)
64
65         segs_json = self._search_regex(r'segs: \'(.*)\'', webpage, 'segments')
66         segments = json.loads(segs_json)
67         # It looks like the keys are the arguments that have to be passed as
68         # the hd field in the request url, we pick the higher
69         # Also, filter non-number qualities (see issue #3643).
70         quality = sorted(filter(lambda k: k.isdigit(), segments.keys()),
71                          key=lambda k: int(k))[-1]
72         parts = segments[quality]
73         result = []
74         len_parts = len(parts)
75         if len_parts > 1:
76             self.to_screen('%s: found %s parts' % (video_id, len_parts))
77         for part in parts:
78             part_id = part['k']
79             final_url = self._url_for_id(part_id, quality)
80             ext = (final_url.split('?')[0]).split('.')[-1]
81             part_info = {
82                 'id': '%s' % part_id,
83                 'url': final_url,
84                 'ext': ext,
85                 'title': title,
86                 'thumbnail': thumbnail_url,
87             }
88             result.append(part_info)
89
90         return result