单个微博评论爬取（可二级评论）

import requests
from bs4 import BeautifulSoup
import pandas as pd
import os
import time
import random



def fetchUrl(pid, uid, max_id):  # 请求函数
    time.sleep(random.random() * 3)  # 设置睡眠时间

    url = "https://weibo.com/ajax/statuses/buildComments"

    headers = {
        'Cookie': '自己的cookie',
        ### 填写Cookie
        "user-agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36 Edg/124.0.0.0",
        'Accept': 'application/json, text/plain, */*',
        'Accept-Encoding': 'gzip, deflate, br, zstd',
        'Accept-Language': 'zh-CN,zh;q=0.9,en;q=0.8,en-GB;q=0.7,en-US;q=0.6'
    }

    params = {
        "is_reload": 1,  # 固定
        "id": pid,  # 该微博id
        "is_show_bulletin": 1,  # 固定
        "is_mix": 0,  # 一级评论是0 二级评论是1
        "max_id": max_id,  # 起始为0，从上一个请求中找
        "count": 20,
        "uid": uid,  # 发微博的用户id 固定
        "fetch_level": 0,  # 固定
        "locale": "zh - CN"
    }

    r = requests.get(url, headers=headers, params=params)
    return r.json()


def fetchLevel2(cid, uid, cmax_id):  # 请求函数 获取二级评论
    time.sleep(random.random() * 3)  # 设置睡眠时间
    curl = "https://weibo.com/ajax/statuses/buildComments"
    headers = {
        'Cookie': '自己的cookie',
        ### 填写Cookie
        "user-agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36 Edg/124.0.0.0",
        'Accept': 'application/json, text/plain, */*',
        'Accept-Encoding': 'gzip, deflate, br, zstd',
        'Accept-Language': 'zh-CN,zh;q=0.9,en;q=0.8,en-GB;q=0.7,en-US;q=0.6'
    }
    params = {
        "flow": 1,  # 固定
        "is_reload": 1,  # 固定
        "id": cid,  # 该评论id
        "is_show_bulletin": 2,  # 固定
        "is_mix": 1,  # 一级评论是0 二级评论是1
        "fetch_level": 1,  # 一级评论是0，二级评论是1
        "max_id": cmax_id,  # 起始为0，从上一个请求中找
        "count": 20,
        "uid": uid,  # 发微博的用户id 固定
    }

    c = requests.get(curl, headers=headers, params=params)
    return c.json()


i = 1
allData = []  # 负责暂时保存数据


def parseJson(jsonObj):  # 一级评论解析函数
    data = jsonObj["data"]
    max_id = jsonObj["max_id"]
    global i
    for item in data:
        #print(i)
        i = i + 1
        # 回复人
        OneRespondent = Respondent(rid)  # 一级评论回复人函数
        # 一级评论id   后续获取二级评论要用到
        comment_Id = item["id"]
        # 评论内容
        content = BeautifulSoup(item["text"], "html.parser").text
        # 评论时间
        created_at = item["created_at"]
        # 点赞数
        like_counts = item["like_counts"]
        # 评论者信息
        user = item["user"]
        userID = user["id"]
        userName = user["screen_name"]
        usergender = user["gender"]#性别
        if usergender=='m':
            usergender="男"
        elif usergender=="f":
            usergender="女"
        userfollowers_count = user["followers_count"]#粉丝数
        userfriends_count = user["friends_count"]#关注数
        userlocation = user["location"]
        userSource = item.setdefault("source")

        dataItem = ["1", created_at, userID, userName,
                    usergender,userfollowers_count,userfriends_count,
                    like_counts, content, userSource, userlocation,
                    OneRespondent]  # 一级评论回复人

       # ["评论级别", "发布时间", "用户id", "用户昵称",
       #   '用户性别', '用户粉丝数', '用户关注数',
       #   "评论点赞数","评论内容", "发消息时所在地", "居住地",
       #   "回复人"]

        #print(dataItem)
        allData.append(dataItem)

        # 嵌套循环  开始获取二级评论
        cid = comment_Id

        cmax_id = 0
        while (True):
            chtml = fetchLevel2(cid, uid, cmax_id)  # 请求函数 获取二级评论
            cmax_id = parseLevel2(chtml)  # 解析二级评论 返回cmax_id
            # cmax_id 为 0 时，表示爬取结束
            if cmax_id == 0:
                break;

    return max_id


def parseLevel2(jsonObj):  # 二级评论解析函数
    cdata = jsonObj["data"]
    cmax_id = jsonObj["max_id"]
    for item in cdata:
        # 回复人                                       # 找到screen_name
        TwoRespondent = item["reply_comment"]["user"]["screen_name"]
        # 二级评论内容
        ctext = BeautifulSoup(item["text"], "html.parser").text
        # 评论时间
        ccreated_at = item["created_at"]
        # 点赞数
        clike_counts = item["like_counts"]
        # 评论者信息
        cuser = item["user"]
        cuserID = cuser["id"]
        cuserName = cuser["screen_name"]
        cuserSource = item.setdefault("source")
        cuserlocation = cuser["location"]
        cuserName = cuser["screen_name"]
        cusergender = cuser["gender"]  # 性别
        if cusergender == 'm':
            cusergender = "男"
        elif cusergender == "f":
            cusergender = "女"
        cuserfollowers_count = cuser["followers_count"]  # 粉丝数
        cuserfriends_count = cuser["friends_count"]  # 关注数

        cdataItem = ['2', ccreated_at, cuserID, cuserName,
                     cusergender, cuserfollowers_count, cuserfriends_count,
                     clike_counts, ctext, cuserSource, cuserlocation,
                     TwoRespondent]
        # ["评论级别", "发布时间", "用户id", "用户昵称",
        #   '用户性别', '用户粉丝数', '用户关注数',
        #   "评论点赞数","评论内容", "发消息时所在地", "居住地",
        #   "回复人"]
        #print(cdataItem)  # 二级评论回复人
        allData.append(cdataItem)
    return cmax_id


def Respondent(rid):
    time.sleep(random.random() * 3)  # 设置睡眠时间
    rurl = 'https://weibo.com/ajax/statuses/show'  # url地址

    headers = {
        'Cookie': '自己的cookie',
        "user-agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36 Edg/124.0.0.0",
        'Accept': 'application/json, text/plain, */*',
        'Accept-Encoding': 'gzip, deflate, br, zstd',
        'Accept-Language': 'zh-CN,zh;q=0.9,en;q=0.8,en-GB;q=0.7,en-US;q=0.6'
    }

    paramas = {
        'id': rid,  # 参数
        "locale": "zh-CN",
    }
    r = requests.get(rurl, headers=headers, params=paramas).json()

    user = r["user"]
    user_name = user['screen_name']  # 找到名字
    return user_name


def save_data(data, path, filename):
    global allData
    #print("开始存取")
    if not os.path.exists(path):
        os.makedirs(path)

    dataframe = pd.DataFrame(data)
    dataframe.to_csv(path + filename, encoding='utf_8_sig', mode='a', index=False, sep=',', header=False)
    allData = []  # 清空alldata


if __name__ == "__main__":

    pid = 4722701329501610  # 微博id，固定
    uid = 6593199887  # 发微博的用户id，固定
    max_id = 0
    rid = 'L9wz2DRNM'  # 参数
    path = "F:\\shiyan\\weibo"
    # 保存的路径
    filename = "微博(神女批观).csv"  # 保存的文件名

    csvHeader = [["评论级别", "发布时间", "用户id", "用户昵称",
                  '用户性别','用户粉丝数','用户关注数', "评论点赞数",
                  "评论内容", "发消息时所在地", "居住地",
                  "回复人"]]  # csv文件第一行
    save_data(csvHeader, path, filename)  # 将csvheader保存至csv文件

    while (True):
        html = fetchUrl(pid, uid, max_id)  # 发起请求
        max_id = parseJson(html)  # 解析数据 返回max_id
        # max_id 为 0 时，表示单次爬取结束 存储数据至csv文件   通常情况下i=20...40...60...时存储
        if max_id == 0:
            break
        save_data(allData, path, filename)

使用指南（main里面）

1.修改保存的文件路径

2.根据对应微博，查找对应pid及id（uid）