awk 常用选项及数组的用法和模拟生产环境数据统计

awk 常用选项总结

在 awk 中使用外部的环境变量 （-v）

awk -v num2="$num1" -v var1="$var" 'BEGIN{print num2,var1}'

-f 选项 文件中读取表达式

1.awk

BEGIN{
    str="I hava a tream"
    location=index(str,"ea")
    print location
}

awk -f 1.awk　

2.wak

BEGIN{
    str="Transaction 243 Start,Event ID:9002"
    count=sub(/[0-9]+/,"$",str)
    print str
}

-F 指定分隔符

awk -F ":" '{print $7}' passwd 

显示版本号

awk -V

awk中数组的用法及模拟生产环境数据统计

shell中的数组的用法：

• shell数组中的下标是从0开始的

    array=("Allen" "Mike" "Messi" "Jerry" "Hanmeimei" "Wang")
    打印元素：        echo ${array[2]}
    打印元素个数:        echo ${#array[@]}
    打印某个元素长度:    echo ${#array[3]}
        
    给元素赋值：    array[3]=ui;
    删除元素:    unset array[2];unset array # 删除数组
    分片访问:    echo ${array[@]:1:3}
    元素内容替换:    ${array[@]/e/E} 只替换第一个e；${array[@]//e/E} 替换所有的e

　　

数组的遍历：

for a in ${array[@]}
do
    echo $a
done

　　

awk中数组的用法：

• 在awk中，使用数组时，不仅可以使用1.2..n作为数组小标，也可以使用字符串作为数组下标

典型常用例子：

统计主机上所有的tcp链接状态数，按照每个tcp状态分类

netstat -an | grep tcp | awk '{arr[$6]++}END{for (i in arr) print i,arr[i]}'

计算横向数据综合，计算纵向数据总和

statics.awk

        BEGIN{
            printf "%-30s%-30s%-30s%-30s%-30s%-30s
","Name","Yuwen","Math","English","Physical","total"
        }
        {
            total=$2+$3+$4+$5
            yuwen_sum+=$2
            math_sum+=$3
            english_sum+=$4
            physical_sum+=$5
            printf "%-30s%-30d%-30d%-30d%-30d%-30d
",$1,$2,$3,$4,$5,total
        }
        END{
            printf "%-30s%-30d%-30d%-30d%-30d
","every_total",yuwen_sum,math_sum,english_sum,physical_sum
        }

　

awk -f statics.awk student.txt 

计算字符串的长度：

str="test string"
echo ${#str}

修改数组元素

array=("Allen" "Mike" "Messi" "Jerry" "Hanmeimei" "Wang")
array[1]="Jerry"
echo ${array[@]}

删除第3个元素

echo ${array[@]}
unset array[2]
echo ${array[@]}

在数组中删除下标为1的元素，即Mike被删除，再次删除下标为1的元素，发现数组不变，说明数组虽然删除了元素，下标还是不变保存在内存中

array=("Allen" "Mike" "Messi" "Jerry" "Hanmeimei" "Wang")
unset array[1]
echo ${array[*]}

分片访问，数组为1的开始遍历3个元素

array=("Allen" "Mike" "Messi" "Jerry" "Hanmeimei" "Wang")
echo ${array[@]:1:3}

1到最后

echo ${array[@]:1}

替换1个，替换所有

echo ${array[@]}
echo ${array[@]/e/E}
echo ${array[@]//e/E}

遍历数组

for a in ${array[@]};do echo $a;done

计算横向和、纵向和

awk -f stu.awk student.txt 

　

模拟生产环境数据脚本

db.log.20190608

2019-06-08 10:31:40 15459 Batches: user Jerry insert 5504 records into datebase:product table:detail, insert 5253 records successfully,failed 251 records
2019-06-08 10:31:40 15460 Batches: user Tracy insert 25114 records into datebase:product table:detail, insert 13340 records successfully,failed 11774 records
2019-06-08 10:31:40 15461 Batches: user Hanmeimei insert 13840 records into datebase:product table:detail, insert 5108 records successfully,failed 8732 records
2019-06-08 10:31:40 15462 Batches: user Lilei insert 32691 records into datebase:product table:detail, insert 5780 records successfully,failed 26911 records
2019-06-08 10:31:40 15463 Batches: user Allen insert 25902 records into datebase:product table:detail, insert 14027 records successfully,failed 11875 records

1 统计每个人分别插入了多少条record进数据库

exam1.awk

    BEGIN{
        printf "%-20s%-20s
","User","Total records"
    }

    {
        USER[$6]+=$8
    }

    END{
        for(u in USER)
            printf "%-20s%-20d
",u,USER[u]
    }

awk -f exam1.awk  db.log.20190608

2 统计每个人分别插入成功了多少record，失败了多少record

exam2.awk

    BEGIN{
        printf "%-30s%-30s%-30s
","User","Success records","Failed records"
    }

    {
        SUCCESS[$6]+=$14
        FAILED[$6]+=$17
    }

    END{
        for(u in SUCCESS)
            printf "%-30s%-30d%-30d
",u,SUCCESS[u],FAILED[u]
    }

awk -f exam2.awk db.log.20190608　

3 将例子1和例子2结合起来，一起输出，输出每个人分别插入多少条数据，多少成功，多少失败，并且要格式化输出，加上标题

exam3.awk

    BEGIN{
        printf "%-30s%-30s%-30s%-30s
","Name","total records","success records","failed records"
    }

    {
        TOTAL_RECORDS[$6]+=$8
        SUCCESS[$6]+=$14
        FAILED[$6]+=$17
    }

    END{
        for(u in TOTAL_RECORDS)
            printf "%-30s%-30d%-30d%-30d
",u,TOTAL_RECORDS[u],SUCCESS[u],FAILED[u]
    }

　　

awk -f exam3.awk db.log.20190608 

4 在例子3的基础上，加上结尾，统计全部插入记录数，成功记录数，失败记录数

exam4_b.awk

    BEGIN{
        printf "%-30s%-30s%-30s%-30s
","Name","total records","success records","failed records"
    }

    {
        TOTAL_RECORDS[$6]+=$8
        SUCCESS[$6]+=$14
        FAILED[$6]+=$17
    }

    END{
        for(u in TOTAL_RECORDS)
        {
            # 在统计出的结果数组中进行累加
            records_sum+=TOTAL_RECORDS[u]
            success_sum+=SUCCESS[u]
            failed_sum+=FAILED[u]
            printf "%-30s%-30d%-30d%-30d
",u,TOTAL_RECORDS[u],SUCCESS[u],FAILED[u]
        }

        printf "%-30s%-30d%-30d%-30d
","",records_sum,success_sum,failed_sum
    }

　　

awk -f exam4_b.awk db.log.20190608

方法2：

exam4.awk　　

    BEGIN{
        printf "%-30s%-30s%-30s%-30s
","Name","total records","success records","failed records"
    }

    {
        RECORDS[$6]+=$8
        SUCCESS[$6]+=$14
        FAILED[$6]+=$17
        
        # 在原始数据中进行汇总计算
        records_sum+=$8
        success_sum+=$14
        failed_sum+=$17    
    }

    END{
        for(u in RECORDS)
            printf "%-30s%-30d%-30d%-30d
",u,RECORDS[u],SUCCESS[u],FAILED[u]

        printf "%-30s%-30d%-30d%-30d
","total",records_sum,success_sum,failed_sum
    }

　　

5 查找丢失数据的现象，也就是成功+失败的记录数不等于一共插入的记录数，找出这些数据并显示行号和对应行的日志信息

awk '{if($8!=$14+$17) print NR,$0}' db.log.20190608

写入文件的方式

exam5.awk

    BEGIN{
    }
    {
        if($8!=$14+$17) 
        print NR,$0
    }

awk -f exam5.awk db.log.20190608