awk的详细用法demo

首先创建一个文件叫做test
内容为（hehe前面有几个空格）：

    hehe  lijie1:one   lijie2:two  lijie3:three

1.查找

awk '/lijie/' ./test
grep lijie ./test

2.默认查找分隔符

[root@lijie hadoop]# cat ./test | awk '{print $1}'
hehe

3.自定义分隔符

[root@lijie hadoop]# cat ./test | grep lijie | awk -F[:\ ] '{print $1}'

[root@lijie hadoop]# cat ./test | grep lijie | awk -F[:\ ] '{print $2}'
hehe
[root@lijie hadoop]# cat ./test | grep lijie | awk -F[:\ ] '{print $3}'

[root@lijie hadoop]# cat ./test | grep lijie | awk -F[:\ ] '{print $4}'
lijie1
[root@lijie hadoop]# cat ./test | grep lijie | awk -F[:\ ] '{print $5}'

4.

[root@lijie hadoop]# cat ./test | grep lijie | awk -F: '{print $1}'
    hehe  lijie1

5.空格

[root@lijie hadoop]# cat ./test | grep lijie | awk -F[\ ] '{print $2}'
hehe

6.自定义打印分隔符

[root@lijie hadoop]# cat ./test | grep lijie | awk -F: '{print $1"---"$2}'
    hehe  lijie1---one   lijie2

7.$0 打印整行信息

[root@lijie hadoop]# cat ./test | grep lijie | awk -F: '{print $0}'
    hehe  lijie1:one   lijie2:two  lijie3:three

8.转义字符

[root@lijie hadoop]# cat ./test | awk -F[:] '{print $1"\t---\n"$2}'
    hehe  lijie1    ---
one   lijie2

9.预定义表头

[root@lijie hadoop]# cat ./test | awk -F[:] 'BEGIN{print "table1 table2 table3"} {print $1,$2,$3}'
table1 table2 table3
    hehe  lijie1 one   lijie2 two  lijie3

10.预定义的分隔符

[root@lijie hadoop]# cat ./test | awk -F[:\ ] '{print $5}'
one
[root@lijie hadoop]# cat ./test | awk 'BEGIN{FS="[:\\ ]"}{print $5}'
one

11.预定义变量

[root@lijie hadoop]# cat ./test | awk 'BEGIN{FS="[:\\ ]";name="my name is LIJIE"}{print name,$5}'
my name is LIJIE one

12.awk的if else

[root@lijie hadoop]# cat ./test | awk 'BEGIN{FS=":"}{if ($1~"lijie12$")print $1;else print$2}'
one   lijie2
[root@lijie hadoop]# cat ./test | awk 'BEGIN{FS=":"}{if ($1~"lijie1$")print $1;else print$2}'
    hehe  lijie1

--可以写成
[root@lijie hadoop]# cat ./test | awk -F: '{if($1~"lijie1$") print $1;else print$2}'
    hehe  lijie1
[root@lijie hadoop]# cat ./test | awk -F: '{if($1~"lijie12$") print $1;else print$2}'
one   lijie2

--利用变量匹配
[root@lijie hadoop]# cat ./test | awk -F: 'BEGIN{name="lijie1$"} {if($1~name) print$1;elseprint$2}'
    hehe  lijie1

--逻辑判断
[root@lijie hadoop]# cat ./test | awk -F: 'BEGIN{name1="lijie1$";name2="lijie2$"} {if($1~name && $2~name2) print$1;elseprint$2}'
    hehe  lijie1

13.NF的用法

--取总共有多少列
[root@lijie hadoop]# cat ./test | awk -F: '{print NF}'
4

--取最后一个列
[root@lijie hadoop]# cat ./test | awk -F: '{print $NF}'
three

14.NR的用法（在之前test文件中添加一行heihei lijie666666:）

[root@lijie hadoop]# cat ./test | awk -F: '{print NR,$1}'
1  hehe  lijie1
2  heihei lijie66666

15.自定义打印列分隔符（OFS）

[root@lijie hadoop]# cat ./test | awk -F: 'BEGIN{OFS="###"}{print $1,$2,$3}'
   hehe  lijie1###one   lijie2###two  lijie3

16.RS和ORS的使用（ORS行分隔符默认为 \n可以不写）

[root@lijie hadoop]# cat ./test | awk 'BEGIN{RS=":"}{print $1,$2,$3}'
  hehe lijie1 
one lijie2 
two lijie3 
three

17.awk的循环

[root@lijie hadoop]# cat ./test | awk 'BEGIN{FS=":"}{i=1; while(i<NF) {print $i" end";i++}}'
   hehe  lijie1 end
one   lijie2 end
two  lijie3 end

18.awk打印格式化（-表示左对齐，不加表示右对齐）

[root@lijie hadoop]# cat ./test | awk '{printf "%15-s %15s %-15s", $1,$2,$3}'
hehe                 lijie1:one lijie2:two

--如果printf后面有自定义的打印也需要格式化处理
[root@lijie hadoop]# cat ./test | awk '{printf "%15-s %15s %-15s %d", $1,$2,$3,1000}'
hehe                 lijie1:one lijie2:two      1000

注意：
NF 计算列数量
$NF 最后一列
NR 计算行号

数据截取定义符
RS 行分隔符
FS 列分隔符

数据打印定义符
OFS 列分隔符
ORS 行分割符