【原创源码】Python工具批量将long型存储的NAT表IP地址转换回ipv4存储

ljl764439816

需要pandas支持，处理CSV文件
转前后文件均为csv文件
转换前示例如下：
after_ip,begain_port,end_port,before_ip // 转换前IP，起始端口，结束端口，转换后IP
3074337170,40960,41215,2111136891
转换后示例如下：
after_ip,begain_port,end_port,before_ip

['183.62.169.146'],40960,41215,['125.213.100.123']
同时转换后的['及去没有想到办法去掉。

期望输出（可以使用UE进行二次处理，去除符号）：

183.62.169.146,40960,41215,125.213.100.123

[Python] 纯文本查看 复制代码

import sys
import csv
import string
import pandas as pd
import numpy as np

def decode_after_ip():                      #计算转换前IP为IPv4
    after_ip_list = []
    for ip in after_ip:
        floor_list = []
        yushu=ip
        for i in reversed(range(4)):
            res=divmod(yushu,256**i)
            floor_list.append(str(res[0]))
            yushu=res[1]
        decode_ip = ('.'.join(floor_list))
        result_list = decode_ip.split()
        after_ip_list.append(result_list)
    return after_ip_list

def decode_before_ip():                      #计算转换后IP为IPv4
    before_ip_list = []
    for ip in before_ip:
        floor_list = []
        yushu=ip
        for i in reversed(range(4)):
            res=divmod(yushu,256**i)
            floor_list.append(str(res[0]))
            yushu=res[1]
        decode_ip = ('.'.join(floor_list))
        result_list = decode_ip.split()
        before_ip_list.append(result_list)
    return before_ip_list

def decode_begin_port():                      #转换起始端口为list
    begin_port_list = []
    for port in begain_port:
        begin_port_list.append(str(port))
    return begin_port_list

def decode_end_port():                        #转换结束端口为list
    end_port_list = []
    for port in end_port:
        end_port_list.append(str(port))
    return end_port_list

csv_reader = pd.read_csv('explongip.csv', usecols=['after_ip', 'begain_port', 'end_port', 'before_ip'],nrows = 200,chunksize=100)  
#nrow读取文件行数 chunksize=每次读取行数，此行为调试用
#csv_reader = pd.read_csv('explongip.csv', usecols=['after_ip', 'begain_port', 'end_port', 'before_ip'],chunksize=1000000)  
#实际处理文件测试每次读取100W行数据
csv_file = pd.DataFrame()
count = 0                 #循环计数确认循环次数
for chunk in csv_reader: #分块读取
    csv_file = csv_file.append(chunk,ignore_index=True)
    # 从DataFrame取出数据
    after_ip = csv_file.loc[:, 'after_ip']
    begain_port = csv_file.loc[:, 'begain_port']
    end_port = csv_file.loc[:, 'end_port']
    before_ip = csv_file.loc[:, 'before_ip']
    # 将数据转换为字典并转换为DataFrame
    csv = {'after_ip': decode_after_ip(),'begain_port': begain_port, 'end_port': end_port,'before_ip': decode_before_ip()}
    exp_file = pd.DataFrame(csv)
    # 追加输出csv
    exp_file.to_csv("decode_ip.csv", mode='a', index=False)
    count=count + 1
    print("The loop count:",count)
    csv_file = pd.DataFrame()
    #每次循环后将csv_file置为空，否则处理大量数据时会出现Memoryerror
    print(csv_file)
    #输出检查是否成功置为空值

wyp360

涨姿势，学习了