微博爬虫2025最新版

背景

之前是可以通过直接调https://m.weibo.cn/api/container/getIndex?containerid={containerid}&page={page} 的方式可以直接调取对应用户的微博，但是最近爬取的时候发现这个已经停了，所以需要一种新的方式来进行爬取。

新方法

通过利用https://weibo.com/ajax/statuses/mymblog?uid=××× 方式去爬

1. 电脑浏览器登录 weibo.com → 点开目标用户主页 → F12 → Network 里任意刷新一下。
2. 找一条 https://weibo.com/ajax/statuses/mymblog?uid=××× 的请求，把 Request Headers 里的
   a. Cookie 整段复制下来（只需要 SUB= 那一小段也可以，但整段最稳）。
   b. x-xsrf-token
   c. sec-fetch-* 系列（共 3 行）
   d. x-requested-with
   整段复制下来备用（x-xsrf-token 每次登录会变，但 24 h 内有效）。
3. 代码里补字段：

cookie = '把你刚才复制的 Cookie 整段粘进来'
uid    = '目标用户的 uid（网址里那串数字）'
save   = f'{uid}_weibo.csv'

HEAD = {
    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36',
    'Cookie': cookie,
    'Referer': f'https://weibo.com/u/{uid}',
    'X-Requested-With': 'XMLHttpRequest',
    'X-Xsrf-Token': '刚才复制的值',          # ← 新增
    'Sec-Fetch-Site': 'same-origin',        # ← 新增
    'Sec-Fetch-Mode': 'cors',               # ← 新增
    'Sec-Fetch-Dest': 'empty'               # ← 新增
}

import requests
import pandas as pd
from datetime import datetime
import re
from zoneinfo import ZoneInfo
import random
import time
import logging

def fetch_one(page: int):
    url = f'https://weibo.com/ajax/statuses/mymblog'
    params = {'uid': uid, 'page': page, 'feature': 0}
    r = requests.get(url, params=params, headers=HEAD, timeout=10)
    if r.status_code != 200:
        logging.info('!!! 被拦 / cookie 失效，先停 30s'); time.sleep(30); return []
    return r.json()['data']['list']
def saveToExcel(reposts, filename):
    df = pd.DataFrame(reposts)
    logging.info('开始保存数据')
    df.to_excel(filename, index=False)

def main():
    reposts = []
    page = 1
    while page < 5:
        posts = fetch_one(page)
        if not posts: break
        for card in posts:
            cleaned_text = ''
            repost_time = datetime.strptime(card.get('created_at'), '%a %b %d %H:%M:%S %z %Y').replace(tzinfo=None)
            if start_date <= repost_time and repost_time <= end_date:
                repost_content = card.get('text', '')
                repost_url = f"https://weibo.com/{uid}/{card.get('mid')}"
                if card.get('retweeted_status') != {}:
                    retweeted_content = card.get('retweeted_status')['text_raw']
                    cleaned_text = re.sub(r'[^\u4e00-\u9fa5，。！？、；：“”‘’（）…—～\n]', '', retweeted_content)
                    logging.info({
                    '时间': datetime.fromisoformat(repost_time.isoformat()).strftime("%Y-%m-%d %H:%M:%S"),
                    '文本': cleaned_text if cleaned_text else repost_content,
                    'url': repost_url
                })
                reposts.append({
                    '时间': datetime.fromisoformat(repost_time.isoformat()).strftime("%Y-%m-%d %H:%M:%S"),
                    '文本': cleaned_text if cleaned_text else repost_content,
                    'url': repost_url
                })
        logging.info(f'第 {page} 页完成，已存 {len(posts)} 条'.format(page=page, posts=posts))
        page += 1
        time.sleep(1.5)          # 保守 1.5 s，基本不会被 403
    saveToExcel(reposts,save)
    logging.info('全部搞定 →')

if __name__ == '__main__':
    start_time = datetime.strptime('2025-10-01', '%Y-%m-%d')
    end_date = datetime.strptime('2025-11-01', '%Y-%m-%d')
    logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
    main()