Question

对Python很新但很兴奋，我需要你的建议。我想出了以下代码来比较基于nmap扫描的两个CSV文件：

import pandas as pd
from pandas import DataFrame
import os
file = raw_input('\nEnter the Old CSV file: ')
file1 = raw_input('\nEnter the New CSV file: ')
A=set(pd.read_csv(file, index_col=False, header=None)[0])
B=set(pd.read_csv(file1, index_col=False, header=None)[0])
final=list(A-B)
df = pd.DataFrame(final, columns=["host"])
df.to_csv('DIFF_'+file)

print "Completed!"

当我运行它时，我得到以下结果：，

host
0,82.214.228.71;dsl-radius-02.direcpceu.com;PTR;tcp;111;rpcbind;open;;;syn-ack;;3;
1,82.214.228.70;dsl-radius-01.direcpceu.com;PTR;tcp;111;rpcbind;open;;;syn-ack;;3;

我的问题是如何在列2,3等上添加标签/ enter code here名称，例如：hostanme，port，port name，state等。我试过了： df [＆＃39;主机名＆＃39;] =范围（1，len（df）+ 1）但是当我用Excel打开文件时，这会在第一列和主机上添加主机名

Answer 1

我认为您首先需要read_csv参数sep=','和names来定义列名：

file = raw_input('\nEnter the Old CSV file: ')
file1 = raw_input('\nEnter the New CSV file: ')

cols = ['hostname','port','portname', ...]
A= pd.read_csv(file, index_col=False, header=None, sep=';', names=cols)
B= pd.read_csv(file1, index_col=False, header=None, sep=';', names=cols)

如果需要比较所有列，请使用merge与boolean indexing进行比较：

df = pd.merge(A, B, how='outer', indicator=True)
df = df[df['_merge']=='left_only'].drop('_merge',axis=1)

df.to_csv('DIFF_'+file)

print "Completed!"

<强>示例：

import pandas as pd
from pandas.compat import StringIO

temp=u"""82.214.228.71;dsl-radius-02.direcpceu.com;PTR;tcp;111;rpcbind;open;;;syn-ack;;3;
82.214.228.70;dsl-radius-01.direcpceu.com;PTR;tcp;111;rpcbind;open;;;syn-ack;;3;
82.214.228.74;dsl-radius-02.direcpceu.com;PTR;tcp;111;rpcbind;open;;;syn-ack;;3;
82.214.228.75;dsl-radius-01.direcpceu.com;PTR;tcp;111;rpcbind;open;;;syn-ack;;3;"""
#after testing replace 'StringIO(temp)' to 'filename.csv'
cols = ['hostname','port','portname', 'a','b','c','d','e','f','g','h','i', 'j']
A = pd.read_csv(StringIO(temp), sep=";", names=cols)
print (A)
        hostname                         port portname    a    b        c  \
0  82.214.228.71  dsl-radius-02.direcpceu.com      PTR  tcp  111  rpcbind   
1  82.214.228.70  dsl-radius-01.direcpceu.com      PTR  tcp  111  rpcbind   
2  82.214.228.74  dsl-radius-02.direcpceu.com      PTR  tcp  111  rpcbind   
3  82.214.228.75  dsl-radius-01.direcpceu.com      PTR  tcp  111  rpcbind   

      d   e   f        g   h  i   j  
0  open NaN NaN  syn-ack NaN  3 NaN  
1  open NaN NaN  syn-ack NaN  3 NaN  
2  open NaN NaN  syn-ack NaN  3 NaN  
3  open NaN NaN  syn-ack NaN  3 NaN

temp=u"""82.214.228.75;dsl-radius-02.direcpceu.com;PTR;tcp;111;rpcbind;open;;;syn-ack;;3;
82.214.228.70;dsl-radius-01.direcpceu.com;PTR;tcp;111;rpcbind;open;;;syn-ack;;3;
82.214.228.77;dsl-radius-02.direcpceu.com;PTR;tcp;111;rpcbind;open;;;syn-ack;;3;
"""
#after testing replace 'StringIO(temp)' to 'filename.csv'
cols = ['hostname','port','portname', 'a','b','c','d','e','f','g','h','i', 'j']
B = pd.read_csv(StringIO(temp), sep=";", names=cols)
print (B)
        hostname                         port portname    a    b        c  \
0  82.214.228.75  dsl-radius-02.direcpceu.com      PTR  tcp  111  rpcbind   
1  82.214.228.70  dsl-radius-01.direcpceu.com      PTR  tcp  111  rpcbind   
2  82.214.228.77  dsl-radius-02.direcpceu.com      PTR  tcp  111  rpcbind   

      d   e   f        g   h  i   j  
0  open NaN NaN  syn-ack NaN  3 NaN  
1  open NaN NaN  syn-ack NaN  3 NaN  
2  open NaN NaN  syn-ack NaN  3 NaN

df1 = pd.merge(A, B, how='outer', indicator=True)

print (df1)

        hostname                         port portname    a    b        c  \
0  82.214.228.71  dsl-radius-02.direcpceu.com      PTR  tcp  111  rpcbind   
1  82.214.228.70  dsl-radius-01.direcpceu.com      PTR  tcp  111  rpcbind   
2  82.214.228.74  dsl-radius-02.direcpceu.com      PTR  tcp  111  rpcbind   
3  82.214.228.75  dsl-radius-01.direcpceu.com      PTR  tcp  111  rpcbind   
4  82.214.228.75  dsl-radius-02.direcpceu.com      PTR  tcp  111  rpcbind   
5  82.214.228.77  dsl-radius-02.direcpceu.com      PTR  tcp  111  rpcbind   

      d   e   f        g   h  i   j      _merge  
0  open NaN NaN  syn-ack NaN  3 NaN   left_only  
1  open NaN NaN  syn-ack NaN  3 NaN        both  
2  open NaN NaN  syn-ack NaN  3 NaN   left_only  
3  open NaN NaN  syn-ack NaN  3 NaN   left_only  
4  open NaN NaN  syn-ack NaN  3 NaN  right_only  
5  open NaN NaN  syn-ack NaN  3 NaN  right_only

#only values in A
df1 = df1[df1['_merge']=='left_only'].drop('_merge',axis=1)
print (df1)
        hostname                         port portname    a    b        c  \
0  82.214.228.71  dsl-radius-02.direcpceu.com      PTR  tcp  111  rpcbind   
2  82.214.228.74  dsl-radius-02.direcpceu.com      PTR  tcp  111  rpcbind   
3  82.214.228.75  dsl-radius-01.direcpceu.com      PTR  tcp  111  rpcbind   

      d   e   f        g   h  i   j  
0  open NaN NaN  syn-ack NaN  3 NaN  
2  open NaN NaN  syn-ack NaN  3 NaN  
3  open NaN NaN  syn-ack NaN  3 NaN

#only values in B
df1 = pd.merge(A, B, how='outer', indicator=True)
df11 = df1[df1['_merge']=='right_only'].drop('_merge',axis=1)
print (df11)
        hostname                         port portname    a    b        c  \
4  82.214.228.75  dsl-radius-02.direcpceu.com      PTR  tcp  111  rpcbind   
5  82.214.228.77  dsl-radius-02.direcpceu.com      PTR  tcp  111  rpcbind   

      d   e   f        g   h  i   j  
4  open NaN NaN  syn-ack NaN  3 NaN  
5  open NaN NaN  syn-ack NaN  3 NaN

#same values in both dataframes
df12 = df1[df1['_merge']=='both'].drop('_merge',axis=1)
print (df12)
        hostname                         port portname    a    b        c  \
1  82.214.228.70  dsl-radius-01.direcpceu.com      PTR  tcp  111  rpcbind   

      d   e   f        g   h  i   j  
1  open NaN NaN  syn-ack NaN  3 NaN

但如果需要仅比较第一列hostname使用isin作为掩码，~与boolean indexing进行反转：

df2 = A[~A['hostname'].isin(B['hostname'])]
print (df2)
        hostname                         port portname    a    b        c  \
0  82.214.228.71  dsl-radius-02.direcpceu.com      PTR  tcp  111  rpcbind   
2  82.214.228.74  dsl-radius-02.direcpceu.com      PTR  tcp  111  rpcbind   

      d   e   f        g   h  i   j  
0  open NaN NaN  syn-ack NaN  3 NaN  
2  open NaN NaN  syn-ack NaN  3 NaN

Answer 2

您可以在定义数据框的位置添加标签。例如，以下内容应该有效

df = pd.DataFrame(final, columns=["host"].append([x for x in range(1, len(df) + 1)] ))

Pandas：如何在csv文件的数据框上添加列名

2 个答案: