处理动态分页：自动翻页与增量数据抓取策略-数据议事厅

爬虫代理

一、案例场景

Lily（挥舞着数据报表）：“用户反馈我们的股票舆情分析总是缺失最新跟帖！这些动态分页像狡猾的狐狸，每次抓取都漏掉关键数据！”

小王（调试着爬虫代码）：“传统分页参数已经失效了。看！（指向屏幕）这个「加载更多」按钮会变异——每次点击都会生成新的加密参数！”

动态分页化身黑衣刺客，手持带有时间戳的毒镖：「想要新数据？先破解我的身份令牌！」UserAgent检测如同城门守卫，将没有伪装的爬虫拒之门外。

import requests
from bs4 import BeautifulSoup
import time
import json

class GubaCrawler:
    def __init__(self):
        # 亿牛云代理配置（www.16yun.cn）
        self.proxy = {
            "http": "http://16YUN:16IP@yn-proxy.16yun.cn:3111", 
            "https": "http://16YUN:16IP@yn-proxy.16yun.cn:3111"
        }
        self.headers = {
            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36",
            "Cookie": "em_hq_fls=js; sid=6d5b20..."  # 需要定期更新的动态cookie
        }
        self.visited_ids = set()  # 增量抓取存储器

    def parse_page(self, url):
        try:
            # 爬虫代理IP与浏览器指纹双保险
            response = requests.get(url, proxies=self.proxy, headers=self.headers, timeout=10)
            soup = BeautifulSoup(response.text, 'html.parser')
            
            # 东方财富股吧帖子解析
            posts = []
            for item in soup.select('.articleh'):
                post_id = item.get('data-postid')  # 唯一标识符
                if post_id in self.visited_ids:
                    continue
                
                title = item.select_one('.l3 a').text.strip()
                time = item.select_one('.l5').text
                # 更多字段解析...
                posts.append({"id":post_id, "title":title, "time":time})
                self.visited_ids.add(post_id)
            
            return posts
        except Exception as e:
            print(f"抓取异常：{str(e)}")
            return []

    def auto_pagination(self):
        base_url = "https://guba.eastmoney.com/list,002291_{}.html"
        page = 1
        while True:
            current_url = base_url.format(page)
            print(f"智能翻页中：{current_url}")
            
            data = self.parse_page(current_url)
            if not data:  # 终止条件判断
                print("到达最后一页！")
                break
                
            # 数据存储逻辑
            with open('guba_data.json', 'a', encoding='utf-8') as f:
                json.dump(data, f, ensure_ascii=False)
            
            page += 1
            time.sleep(3)  # 控制频率

if __name__ == '__main__':
    crawler = GubaCrawler()
    crawler.auto_pagination()