dsProject/dsLightRag/Test/TestCrawl.py

# 详解（一）Python + Selenium 批量采集微信公众号，搭建自己的微信公众号每日AI简报，告别信息焦虑
# https://blog.csdn.net/k352733625/article/details/149222945

# 微信爬爬猫---公众号文章抓取代码分析
# https://blog.csdn.net/yajuanpi4899/article/details/121584268
import datetime
import logging
import random
import re

import requests

# 1、安装Firefox软件【最新】
# https://www.firefox.com.cn/download/#product-desktop-release

# 2、下载geckodriver驱动【最新】
# https://splinter-docs-zh-cn.readthedocs.io/zh/latest/drivers/firefox.html
# https://github.com/mozilla/geckodriver/releases

# 3、Python爬虫实战系列：微信公众号文章爬取的5种技术方案总结及代码示例！
# 方案5：微信公众号后台引用链接方式爬取
# https://blog.csdn.net/Python_trys/article/details/146506009

"""
# 查看selenium版本
pip show selenium
4.34.2

# 查看Chrome浏览器版本
chrome://version/
138.0.7204.101 (正式版本) （64 位）

# 下载驱动包
https://googlechromelabs.github.io/chrome-for-testing/
https://storage.googleapis.com/chrome-for-testing-public/138.0.7204.94/win64/chromedriver-win64.zip
"""
import time
from selenium import webdriver
from selenium.webdriver.chrome.options import Options
from selenium.webdriver.chrome.service import Service as ChromeService
import json

if __name__ == '__main__':
    # 定义一个空的字典，存放cookies内容
    cookies = {}
    # 设置headers
    header = {
        "HOST": "mp.weixin.qq.com",
        "User-Agent": "Mozilla/5.0 (Windows NT 6.1; WOW64; rv:53.0) Gecko/20100101 Firefox/53.0"
    }
    # 用webdriver启动谷歌浏览器
    logging.info("启动浏览器，打开微信公众号登录界面")
    options = Options()
    options.add_argument('-headless')  # 无头参数
    service = ChromeService(executable_path=r"C:\Windows\System32\chromedriver.exe")
    driver = webdriver.Chrome(service=service)
    # 打开微信公众号登录页面
    driver.get('https://mp.weixin.qq.com/')
    # 等待5秒钟
    time.sleep(2)
    # # 拿手机扫二维码！
    logging.info("请拿手机扫码二维码登录公众号")
    time.sleep(20)

    # 重新载入公众号登录页，登录之后会显示公众号后台首页，从这个返回内容中获取cookies信息
    driver.get('https://mp.weixin.qq.com/')
    # 获取cookies
    cookie_items = driver.get_cookies()
    # 获取到的cookies是列表形式，将cookies转成json形式并存入本地名为cookie的文本中
    for cookie_item in cookie_items:
        cookies[cookie_item['name']] = cookie_item['value']

    if "slave_sid" not in cookies:
        logging.info("登录公众号失败，获取cookie失败")
        exit()
    # cookies = json.dumps(post)  # 注释掉这一行

    # 方法3：使用requests库发送请求获取重定向URL
    url = 'https://mp.weixin.qq.com'
    response = requests.get(url=url, allow_redirects=False, cookies=cookies)
    if 'Location' in response.headers:
        redirect_url = response.headers.get("Location")
        print("重定向URL:", redirect_url)
        token_match = re.findall(r'token=(\d+)', redirect_url)
        if token_match:
            token = token_match[0]
            print("获取到的token:", token)
            logging.info("微信token:" + token)

    article_urls = []
    gzlist = [{"account_name": "长春教育八卦阵", "account_id": "jybg100"}]
    for item in gzlist:
        account_name = item["account_name"]
        account_id = item["account_id"]
        # 搜索微信公众号的接口地址
        search_url = 'https://mp.weixin.qq.com/cgi-bin/searchbiz?'
        # 搜索微信公众号接口需要传入的参数，有三个变量：微信公众号token、随机数random、搜索的微信公众号名字
        query_id = {
            'action': 'search_biz',
            'token': token,
            'lang': 'zh_CN',
            'f': 'json',
            'ajax': '1',
            'random': random.random(),
            'query': account_name,
            'begin': '0',
            'count': '5'
        }
        # 打开搜索微信公众号接口地址，需要传入相关参数信息如：cookies、params、headers
        search_response = requests.get(search_url, cookies=cookies, headers=header, params=query_id)
        # 取搜索结果中的第一个公众号
        lists = search_response.json().get('list')[0]
        # 获取这个公众号的fakeid，后面爬取公众号文章需要此字段
        fakeid = lists.get('fakeid')
        logging.info("fakeid:" + fakeid)
        # 微信公众号文章接口地址
        appmsg_url = 'https://mp.weixin.qq.com/cgi-bin/appmsg?'
        # 搜索文章需要传入几个参数：登录的公众号token、要爬取文章的公众号fakeid、随机数random
        query_id_data = {
            'token': token,
            'lang': 'zh_CN',
            'f': 'json',
            'ajax': '1',
            'random': random.random(),
            'action': 'list_ex',
            'begin': '0',  # 不同页，此参数变化，变化规则为每页加5
            'count': '5',
            'query': '',
            'fakeid': fakeid,
            'type': '9'
        }
        # 打开搜索的微信公众号文章列表页
        query_fakeid_response = requests.get(appmsg_url, cookies=cookies, headers=header, params=query_id_data)
        fakeid_list = query_fakeid_response.json().get('app_msg_list')

        for item in fakeid_list:
            # 采集item示例
            new_article = {
                'title': item.get('title'),
                'article_url': item.get('link'),
                'account_id': account_id,
                'account_name': account_name,
                'publish_time': datetime.datetime.fromtimestamp(int(item.get("update_time"))).strftime('%Y-%m-%d %H:%M:%S'),
                'collection_time': datetime.datetime.now().strftime('%Y-%m-%d %H:%M:%S')
            }
            logging.info("new_article:", new_article)
            article_urls.append(item.get('link'))
            time.sleep(1)

    for article_url in article_urls:
        print("正在爬取文章：" + article_url)