用shell统计表格数据

今天有个人问了这样一个问题，图片是原题，在这个题的基础上写了一个实现方法

首先日志存到a.txt文本里，如下

Zhangsan|lisi1|0|
Zhangsan|lisi2|10|
Zhangsan|lisi3|6|
Zhangsan|lisi4|0|
Zhangsan|lisi5|0|
Lisi|list1|0|
Lisi|list2|5|

 

代码如下：

#!/bin/sh
awk 'BEGIN{FS="|";};{print $1"`"$2"`"$3}' a.txt|sort -t \` -k 1,1|awk 'BEGIN{FS="`";curline="";name="";renshu=0;no_cishu=0;cishu=0};
{
    if(curline==""){
        name=$1;
        renshu=1;
        no_cishu=1;
        chengong=0;
        cishu=$3;
        curline="name:"name"`renshu:"renshu"`no_cishu:"no_cishu"`cishu:"cishu"`chengong:"chengong;
    }else if(name==$1){
        renshu++;
        if($3==0){
            no_cishu++;
        }else{
            chengong++;
        }
        #print chengong
        cishu+=$3;
    }else{
        curline="name:"name"`renshu:"renshu"`no_cishu:"no_cishu"`cishu:"cishu"`chengong:"chengong;
        print curline;
        name=$1;
        cishu=$3;
        renshu=1;
        no_cishu=1;
        chengong=0;
    }

    };
        END{if(curline!=""){
        curline="name:"name"`renshu:"renshu"`no_cishu:"no_cishu"`cishu:"cishu"`chengong:"chengong;
        print curline;
        }

}'

 结果如下：　

name:Lisi`renshu:2`no_cishu:1`cishu:5`chengong:1
name:Zhangsan`renshu:5`no_cishu:3`cishu:16`chengong:2