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/albumplay/TenTw_JgiPM/PzsAs5usU9A.html',
  24         'info_dict': {
  25             'title': 'todo.mp4',
  26         },
  27         'add_ie': ['Youku'],
  28         'skip': 'Only works from China'
  29     }]
  30
  31     def _url_for_id(self, id, quality = None):
  32         info_url = "http://v2.tudou.com/f?id="+str(id)
  33         if quality:
  34             info_url += '&hd' + quality
  35         webpage = self._download_webpage(info_url, id, "Opening the info webpage")
  36         final_url = self._html_search_regex('>(.+?)</f>',webpage, 'video url')
  37         return final_url
  38
  39     def _real_extract(self, url):
  40         mobj = re.match(self._VALID_URL, url)
  41         video_id = mobj.group(2)
  42         webpage = self._download_webpage(url, video_id)
  43
  44         m = re.search(r'vcode:\s*[\'"](.+?)[\'"]', webpage)
  45         if m and m.group(1):
  46             return {
  47                 '_type': 'url',
  48                 'url': 'youku:' + m.group(1),
  49                 'ie_key': 'Youku'
  50             }
  51
  52         title = self._search_regex(
  53             r",kw:\s*['\"](.+?)[\"']", webpage, 'title')
  54         thumbnail_url = self._search_regex(
  55             r",pic:\s*[\"'](.+?)[\"']", webpage, 'thumbnail URL', fatal=False)
  56
  57         segs_json = self._search_regex(r'segs: \'(.*)\'', webpage, 'segments')
  58         segments = json.loads(segs_json)
  59         # It looks like the keys are the arguments that have to be passed as
  60         # the hd field in the request url, we pick the higher
  61         # Also, filter non-number qualities (see issue #3643).
  62         quality = sorted(filter(lambda k: k.isdigit(), segments.keys()),
  63                          key=lambda k: int(k))[-1]
  64         parts = segments[quality]
  65         result = []
  66         len_parts = len(parts)
  67         if len_parts > 1:
  68             self.to_screen(u'%s: found %s parts' % (video_id, len_parts))
  69         for part in parts:
  70             part_id = part['k']
  71             final_url = self._url_for_id(part_id, quality)
  72             ext = (final_url.split('?')[0]).split('.')[-1]
  73             part_info = {
  74                 'id': '%s' % part_id,
  75                 'url': final_url,
  76                 'ext': ext,
  77                 'title': title,
  78                 'thumbnail': thumbnail_url,
  79             }
  80             result.append(part_info)
  81
  82         return result