youtube_dl/extractor/tudou.py

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