¶前言

NGA论坛刚刚开放了用户IP显示功能，早就想查查泥潭精英充分的我连夜花费数个小时写了个IP爬虫出来，看看都是哪些人在泥潭大漩涡板块活跃

¶爬虫

¶包与headers

首先是配置headers：

import requests as req
from lxml import etree
import numpy as np
import time
import re


headers = {
    # 在浏览器中，network查看
    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/103.0.5060.114 Safari/537.36 Edg/103.0.1264.62',
    'Cookie': '',
    'Connection':'close',
}


# API文档参考 https://github.com/wolfcon/NGA-API-Documents

¶版面页

然后是从网事杂谈板块前几页的爬取到各个帖子的链接(API接口参数可查看文档)

F12查找到对应元素(不准确，需要自行修改)方便抓取链接。

urls = []  # 保存页面uid
limit = 5  # 版面页数，请勿设置过多

for i in range(1,limit+1):  # 获取近期网事杂谈板块回复前limit页中的帖子地址
    
    time.sleep(1) 
    
    mainPage = req.get('https://bbs.nga.cn/thread.php?fid=-7&order_by=lastpostdesc&page='+str(i),headers=headers,verify=False)
    doc = etree.HTML(mainPage.text)
    pages_url = doc.xpath('//td[@class="c1"]/a')  # 查找对应元素
    
    for pg in pages_url:
        r = re.search(r'[0-9]+',pg.attrib['href']).group()   # 帖子uid
        urls.append(r)
        print('no.'+str(i)+' : '+str(r))

之后对抓取到的主题贴进行去重

1
2
3

urls = set(urls) # 帖子去重,注意此处顺序被打乱
urls = list(urls)
print(len(urls))

¶主题页

之后获取到主题贴第一页(默认)的内容，找到对应结果计算帖子页数，并获取到每页的用户uid, 用户uid可去重可不去重。

uid = [] 
for item in  urls:  # 帖子中用户uid获取
    
    time.sleep(1)
    
    page_url = 'https://bbs.nga.cn/read.php?tid='+str(item)+'&lite=js'         # 获取当前帖子页数
    mainPage = req.get(page_url,headers=headers,verify=False)
    txt = str(mainPage.text).replace('window.script_muti_get_var_store=','')
    Rows = re.findall(r'"__ROWS"\:[0-9]+',txt)
    
    if Rows:        # nga一小部分帖子js只传一半
         pass
    else:
        continue
        
    pageNum = int(int(Rows[0].replace('"__ROWS":',''))/20 + 1)  # 当前帖子页数
    print(str(item)+" pages: "+str(pageNum))
    
    if pageNum>100:  #去除超过100页的帖子
        continue
    
    for i in range(1,pageNum+1):       # 用户uid获取
        u = page_url+'&page='+str(i)
        mainPage = req.get(u,headers=headers,verify=False)
        txt = str(mainPage.text).replace('window.script_muti_get_var_store=','')
        tmp = re.findall(r'"uid"\:[0-9]+',txt)
        flag = 0
        for t in tmp:
            if(flag==0):
                flag=1
                continue
            i = t.replace('"uid":','')
            uid.append(i)

¶用户IP获取

通过uid查到用户信息，并筛出ipLoc数据：

url = 'https://bbs.nga.cn/nuke.php?lite=js&__lib=ucp&__act=get&uid='  # 用户IP查询
ips = [] # 保存ip
nums = 0
for person in uid:
    time.sleep(0.1)
    person_page_url = url + person
    mainPage = req.get(person_page_url,headers=headers,verify=False)
    txt = str(mainPage.text).replace('window.script_muti_get_var_store=','')
    tmp = re.findall(r'"ipLoc"\:"[\u4e00-\u9fa5]+',txt) # 正则查找
    if tmp:  # nga有概率js只传一半
        pass
    else:
        continue
    tmp = tmp[0].replace('"ipLoc":"','')
    ips.append(tmp)
    print(nums,tmp)  # 输出当前位置，方便网络中断后继续运行
    nums = nums+1

with open('.\\area.txt', mode='a',encoding='utf-8') as f:  # 写入文件保存
    for i in ips:
        f.write(i+'\n')