搜狗微信搜索文章列表接口实战：Cookie风控加固+内容结构化清洗+全自动分页（Python合规生产版）

import requests
import time
import re
from urllib.parse import quote

class SogouWechatSearch:
    def __init__(self):
        self.base_url = "https://weixin.sogou.com/weixin"
        self.session = requests.Session()
        # 初始化获取风控Cookie，规避首次请求拦截
        self.session.get("https://weixin.sogou.com", timeout=15)
        # 生产级请求头+Cookie指纹加固
        self.headers = {
            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36",
            "Referer": "https://weixin.sogou.com/",
            "Accept-Language": "zh-CN,zh;q=0.9"
        }

    def clean_html(self, text):
        """清洗文章标题、描述中的HTML标签和特殊符号"""
        if not text:
            return ""
        text = re.sub(r"<.*?>", "", text)
        text = re.sub(r"[\r\n\t]", "", text)
        return text.strip()

    def single_page_search(self, keyword, page=1):
        """单页关键词文章搜索，处理风控与数据清洗"""
        if len(keyword) < 1 or len(keyword) > 30:
            return {"code": -1, "msg": "关键词长度1-30字符", "article_list": []}
        encoded_keyword = quote(keyword, encoding="utf-8")
        params = {
            "type": 2,
            "query": encoded_keyword,
            "page": page,
            "ie": "utf8",
            "_sug_": "n",
            "_sug_type_": ""
        }
        try:
            resp = self.session.get(self.base_url, headers=self.headers, params=params, timeout=15)
            # 检测验证码拦截
            if "验证" in resp.text or "captcha" in resp.text.lower():
                time.sleep(10)
                return self.single_page_search(keyword, page)
            # 提取文章核心数据
            article_pattern = re.compile(r'<div class="txt-box".*?<h3><a.*?>(.*?)</a></h3>.*?<a.*?>(.*?)</a>.*?<span>(.*?)</span>.*?<a href="(.*?)"', re.S)
            results = article_pattern.findall(resp.text)
            article_list = []
            for title, author, pub_time, raw_url in results:
                # 拼接完整文章链接
                full_url = f"https://weixin.sogou.com{raw_url}" if raw_url.startswith("/") else raw_url
                article_list.append({
                    "title": self.clean_html(title),
                    "author": self.clean_html(author),
                    "pub_time": self.clean_html(pub_time),
                    "article_url": full_url
                })
            time.sleep(3)
            return {"code": 200, "article_list": article_list, "has_next": len(article_list) > 0}
        except Exception as e:
            return {"code": -2, "msg": f"网络请求异常：{str(e)}", "article_list": []}

    def get_all_articles(self, keyword):
        """全自动分页拉取全部文章"""
        all_articles = []
        page = 1
        while True:
            res = self.single_page_search(keyword, page)
            if res["code"] != 200 or not res["article_list"]:
                break
            all_articles.extend(res["article_list"])
            if not res["has_next"]:
                break
            page += 1
        return {"keyword": keyword, "total": len(all_articles), "data": all_articles}

# 调用示例
if __name__ == "__main__":
    client = SogouWechatSearch()
    # 搜索Python相关微信文章
    result = client.get_all_articles("Python开发")
    print(result)