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     _WORKING = False
  12     _VALID_URL = r'https?://(?:www\.)?kankanews\.com/a/\d+\-\d+\-\d+/(?P<id>\d+)\.shtml'
  13     _TESTS = [{
  14         'url': 'https://www.kankanews.com/a/2022-11-08/00310276054.shtml?appid=1088227',
  15         'md5': '05e126513c74b1258d657452a6f4eef9',
  16         'info_dict': {
  17             'id': '4485057',
  18             'url': 'http://mediaplay.kksmg.com/2022/11/08/h264_450k_mp4_1a388ad771e0e4cc28b0da44d245054e_ncm.mp4',
  19             'ext': 'mp4',
  20             'title': '视频｜第23个中国记者节，我们在进博切蛋糕',
  21             'thumbnail': r're:^https?://.*\.jpg*',
  22         }
  23     }]
  24
  25     def _real_extract(self, url):
  26         display_id = self._match_id(url)
  27         webpage = self._download_webpage(url, display_id)
  28         video_id = self._search_regex(r'omsid\s*=\s*"(\d+)"', webpage, 'video id')
  29
  30         params = {
  31             'nonce': ''.join(random.choices(string.ascii_lowercase + string.digits, k=8)),
  32             'omsid': video_id,
  33             'platform': 'pc',
  34             'timestamp': int(time.time()),
  35             'version': '1.0',
  36         }
  37         params['sign'] = hashlib.md5((hashlib.md5((
  38             urllib.parse.urlencode(params) + '&28c8edde3d61a0411511d3b1866f0636'
  39         ).encode()).hexdigest()).encode()).hexdigest()
  40
  41         meta = self._download_json('https://api-app.kankanews.com/kankan/pc/getvideo',
  42                                    video_id, query=params)['result']['video']
  43
  44         return {
  45             'id': video_id,
  46             'url': meta['videourl'],
  47             'title': self._search_regex(r'g\.title\s*=\s*"([^"]+)"', webpage, 'title'),
  48             'thumbnail': meta.get('titlepic'),
  49         }