match_spider/zbytb/jsl/jsl.py

# -*- coding: utf-8 -*-
"""
Created on 2024-09-29 
---------
@summary:  加速乐
---------
@author: Dzr
"""
import json
import pathlib
import re

import execjs
import requests

_cookies = {}


def check_script(text):
    ret = re.findall('cookie=(.*?);location', text)
    assert len(ret) > 0


def ck1(url, headers, timeout=10, proxies=None):
    global _cookies
    response = requests.get(url, headers=headers, timeout=timeout, proxies=proxies)
    _cookies.update(response.cookies)
    text = response.content.decode()
    check_script(text)
    script = re.findall('cookie=(.*?);location', text)[0]
    _, val = next(iter(execjs.eval(script).split(';')), '').split('=')
    # print(val)  # jsl_clearance_s
    _cookies['__jsl_clearance_s'] = val
    # _cookies[_] = val


def get_params(url, headers, timeout, proxies=None):
    global _cookies
    # 通过携带 jsluid_s 和 jsl_clearance_s 值的 cookie 获取第二层响应内容
    request_params = dict(cookies=_cookies, timeout=timeout, proxies=proxies)
    response = requests.get(url=url, headers=headers, **request_params)
    # 获取 go 字典参数
    go_params = re.findall(';go\((.*?)\)</script>', response.content.decode())[0]
    params = json.loads(go_params)
    return params


def ck2(url, headers, timeout=10, proxies=None):
    filelib = (pathlib.Path(__file__).parent.absolute() / 'jsl.js').absolute()
    script = filelib.open('r', encoding='utf-8').read()
    params = get_params(url, headers, timeout, proxies)
    # 传入字典
    third_cookie = execjs.compile(script).call('cookies', params)
    _cookies.update(third_cookie)


def get_jsl_cookies(url, headers, **kwargs):
    global _cookies
    ck1(url, headers, **kwargs)
    ck2(url, headers, **kwargs)
    return _cookies


get_cookies = get_jsl_cookies


if __name__ == '__main__':
    jsl_headers = {
        'Referer': 'https://zjcs.cqggzy.com/cq-zjcs-pub/invalidProject/view/500103MB0T460482408210380',
        'Upgrade-Insecure-Requests': '1',
        'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/129.0.0.0 Safari/537.36',
    }
    jsl_url = 'https://zjcs.cqggzy.com/cq-zjcs-pub/invalidProject/view/500103MB0T460482408210380'
    print(get_cookies(jsl_url, jsl_headers))