yt_dlp/extractor/kankanews.py

   1 import time
   2 import random
   3 import string
   4 import hashlib
   5 import urllib.parse
   6
   7 from .common import InfoExtractor
   8
   9
  10 class KankaNewsIE(InfoExtractor):
  11     _VALID_URL = r'https?://(?:www\.)?kankanews\.com/a/\d+\-\d+\-\d+/(?P<id>\d+)\.shtml'
  12     _TESTS = [{
  13         'url': 'https://www.kankanews.com/a/2022-11-08/00310276054.shtml?appid=1088227',
  14         'md5': '05e126513c74b1258d657452a6f4eef9',
  15         'info_dict': {
  16             'id': '4485057',
  17             'url': 'http://mediaplay.kksmg.com/2022/11/08/h264_450k_mp4_1a388ad771e0e4cc28b0da44d245054e_ncm.mp4',
  18             'ext': 'mp4',
  19             'title': '视频｜第23个中国记者节，我们在进博切蛋糕',
  20             'thumbnail': r're:^https?://.*\.jpg*',
  21         }
  22     }]
  23
  24     def _real_extract(self, url):
  25         display_id = self._match_id(url)
  26         webpage = self._download_webpage(url, display_id)
  27         video_id = self._search_regex(r'omsid\s*=\s*"(\d+)"', webpage, 'video id')
  28
  29         params = {
  30             'nonce': ''.join(random.choices(string.ascii_lowercase + string.digits, k=8)),
  31             'omsid': video_id,
  32             'platform': 'pc',
  33             'timestamp': int(time.time()),
  34             'version': '1.0',
  35         }
  36         params['sign'] = hashlib.md5((hashlib.md5((
  37             urllib.parse.urlencode(params) + '&28c8edde3d61a0411511d3b1866f0636'
  38         ).encode()).hexdigest()).encode()).hexdigest()
  39
  40         meta = self._download_json('https://api-app.kankanews.com/kankan/pc/getvideo',
  41                                    video_id, query=params)['result']['video']
  42
  43         return {
  44             'id': video_id,
  45             'url': meta['videourl'],
  46             'title': self._search_regex(r'g\.title\s*=\s*"([^"]+)"', webpage, 'title'),
  47             'thumbnail': meta.get('titlepic'),
  48         }