在python

时间:2018-10-01 14:31:06

标签: python pandas

我有一个高度嵌套的json文件,该文件已拉平并得到如下输出,其中一行包含一列并附加了数值。有什么办法可以将其删除并按输出显示为行
输入文件

102_ip_addr, 102_ip_family, 102_ip_mask_addr, 102_email,    102_failed_attempts,103_ip_addr, 103_ip_family, 103_ip_mask_addr, 103_email,    103_failed_attempts,
3705824725, 2,  4294967295, abc@xyz.com,    0,3705824825,   4,  4294967625, sdf@xyz.com,    0

输出:

ip_addr, ip_family, ip_mask_addr, email, failed_attempts
3705824725, 2,  4294967295, abc@xyz.com,    0
3705824825, 4,  4294967625, sdf@xyz.com,    0

1 个答案:

答案 0 :(得分:2)

如果每个新行的宽度都固定为5,则可以使用reshape

pd.DataFrame(df.values.reshape(-1,5),columns=['addr','family','mask_addr','email','attempts'])
Out[580]: 
         addr family   mask_addr         email attempts
0  3705824725      2  4294967295   abc@xyz.com        0
1  3705824825      4  4294967625   sdf@xyz.com        0

更新

df.columns=df.columns.str.split('_',1).str[1]

df.melt().assign(newrow=lambda x : x.groupby(x['variable']).cumcount() ).pivot('newrow','variable','value')
Out[596]: 
variable         email failed_attempts     ip_addr ip_family ip_mask_addr
newrow                                                                   
0          abc@xyz.com               0  3705824725         2   4294967295
1          sdf@xyz.com               0  3705824825         4   4294967625