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'https?://(?:www\.)?tudou\.com/(?:listplay|programs(?:/view)?|albumplay)/?.*/(?P<id>[^/?#]+?)(?:\.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/cJAHGih4yYg.html',
  32         'only_matching': True,
  33     }]
  34
  35     _PLAYER_URL = 'http://js.tudouui.com/bin/lingtong/PortalPlayer_177.swf'
  36
  37     def _url_for_id(self, id, quality=None):
  38         info_url = "http://v2.tudou.com/f?id=" + str(id)
  39         if quality:
  40             info_url += '&hd' + quality
  41         webpage = self._download_webpage(info_url, id, "Opening the info webpage")
  42         final_url = self._html_search_regex('>(.+?)</f>', webpage, 'video url')
  43         return final_url
  44
  45     def _real_extract(self, url):
  46         video_id = self._match_id(url)
  47         webpage = self._download_webpage(url, video_id)
  48
  49         m = re.search(r'vcode:\s*[\'"](.+?)[\'"]', webpage)
  50         if m and m.group(1):
  51             return {
  52                 '_type': 'url',
  53                 'url': 'youku:' + m.group(1),
  54                 'ie_key': 'Youku'
  55             }
  56
  57         title = self._search_regex(
  58             r",kw:\s*['\"](.+?)[\"']", webpage, 'title')
  59         thumbnail_url = self._search_regex(
  60             r",pic:\s*[\"'](.+?)[\"']", webpage, 'thumbnail URL', fatal=False)
  61
  62         player_url = self._search_regex(
  63             r"playerUrl\s*:\s*['\"](.+?\.swf)[\"']",
  64             webpage, 'player URL', default=self._PLAYER_URL)
  65
  66         segs_json = self._search_regex(r'segs: \'(.*)\'', webpage, 'segments')
  67         segments = json.loads(segs_json)
  68         # It looks like the keys are the arguments that have to be passed as
  69         # the hd field in the request url, we pick the higher
  70         # Also, filter non-number qualities (see issue #3643).
  71         quality = sorted(filter(lambda k: k.isdigit(), segments.keys()),
  72                          key=lambda k: int(k))[-1]
  73         parts = segments[quality]
  74         result = []
  75         len_parts = len(parts)
  76         if len_parts > 1:
  77             self.to_screen('%s: found %s parts' % (video_id, len_parts))
  78         for part in parts:
  79             part_id = part['k']
  80             final_url = self._url_for_id(part_id, quality)
  81             ext = (final_url.split('?')[0]).split('.')[-1]
  82             part_info = {
  83                 'id': '%s' % part_id,
  84                 'url': final_url,
  85                 'ext': ext,
  86                 'title': title,
  87                 'thumbnail': thumbnail_url,
  88                 'http_headers': {
  89                     'Referer': player_url,
  90                 },
  91             }
  92             result.append(part_info)
  93
  94         return {
  95             '_type': 'multi_video',
  96             'entries': result,
  97             'id': video_id,
  98             'title': title,
  99         }