为什么想学 shell？

shell 命令还是蛮重要的，因为 windows、mac、安卓、ios 等系统都拥有同一个祖先，Unix。因此，学会 shell 命令可以通吃服务端、客户端的测试。

最近在学习 Linux 三剑客，grep 、awk和sed。整理了一下学习笔记，方便以后查看。

linux 三剑客与 SQL 的类比

grep  相当于  select table like '%%'  
awk   相当于  select field from table
sed   相当于  update tabele set field = new  where field = old

grep

grep 扩展正则`{}`的用法

{}表示范围限定

[indeyo@VM_0_14_centos indeyo]$ echo '23456' | grep -oE '[2-5]' # 默认只匹配 1个
2
3
4
5
[indeyo@VM_0_14_centos indeyo]$ echo '23456' | grep -oE '[2-5]{1}'
2
3
4
5
[indeyo@VM_0_14_centos indeyo]$ echo '23456' | grep -oE '[2-5]{1,2}'
23
45
[indeyo@VM_0_14_centos indeyo]$ echo '23456' | grep -oE '[2-5]{1,3}' # 最大匹配 3个，剩下不够的就匹配 1个
234
5
[indeyo@VM_0_14_centos indeyo]$ echo '23456' | grep -oE '[2-5]{2,3}' # 最大匹配 3个，剩下的不足以匹配 2个则匹配完成
234

grep`[]`的用法

[indeyo@VM_0_14_centos ~]$ echo '
> 22
> 23
> 24' | grep -E '(23|24)'
23
24
[indeyo@VM_0_14_centos ~]$ echo '
22
23
24' | grep '2[24]'
22
24
[indeyo@VM_0_14_centos ~]$ echo '
22
23
24' | grep '[24]'
22
23
24

awk

awk 原理

awk 'BEGIN{ print "start" } pattern{ commands } END{ print "end" }' file

awk 执行顺序

加载输入流之前，运行 BEGIN 代码块
执行 pattern 代码块
语句运行完成后，执行 END 代码块

awk pattern 语法

[86413302@shell.testing-studio.com indeyo]$cat 4.txt
6656
123456
[86413302@shell.testing-studio.com indeyo]$awk '/123/' 4.txt  # 正则用法
123456
[86413302@shell.testing-studio.com indeyo]$cat 8.txt 
hello world
I gona eat korean roast meat .
Thinking of this , I feel excitd.


2390840928905825789

jskfjlsjfldkfjkdkfdjkddddddddddddddddd
fdsf 


[86413302@shell.testing-studio.com indeyo]$awk '/this/,/89/' 8.txt  # 取一个区间
Thinking of this , I feel excitd.


2390840928905825789
[86413302@shell.testing-studio.com indeyo]$awk '$2~/of/' 8.txt  # 某个字段等于某个值
Thinking of this , I feel excitd.
[86413302@shell.testing-studio.com indeyo]$awk 'NR==2' 8.txt  # 取第二行
I gona eat korean roast meat .
[86413302@shell.testing-studio.com indeyo]$awk 'NR>1' 8.txt  # 去掉第一行
I gona eat korean roast meat .
Thinking of this , I feel excitd.


2390840928905825789

jskfjlsjfldkfjkdkfdjkddddddddddddddddd
fdsf

awk 有正则表达的作用，一定程度可以代替 grep

[indeyo@VM_0_14_centos indeyo]$ awk '/111/' 1.sh
gogog1111111111abczzzzzzz
11111111sdfsdf222

NR 和 END 语句块

[86413302@shell.testing-studio.com indeyo]$echo $PATH | awk 'BEGIN{RS=":"}{print NR,$0}'
1 /usr/local/bin
2 /usr/bin
3 /usr/local/sbin
4 /usr/sbin
5 /home/86413302/.local/bin
6 /home/86413302/bin
[86413302@shell.testing-studio.com indeyo]$echo $PATH | awk 'BEGIN{RS=":"}END{print NR}'
6

NF 一般用来取倒数的字段

[indeyo@VM_0_14_centos indeyo]$ echo '1 2    3' | awk '{print $3}'
3
[indeyo@VM_0_14_centos indeyo]$ echo '1 2    3' | awk '{print $NF}' # 相当最后一个字段
3
[indeyo@VM_0_14_centos indeyo]$ echo '1 2    3' | awk '{print $(NF-1)}'  相当于倒数第二个字段
2

NR 表示行序号

[indeyo@VM_0_14_centos indeyo]$ cat 2.txt
dddddddd
ttttttttttttttt
hello  world

hey

[indeyo@VM_0_14_centos indeyo]$ awk 'NR>1' 2.txt |cat  # 默认输出$0
ttttttttttttttt
hello  world

hey

[indeyo@VM_0_14_centos indeyo]$ awk 'NR>1{print $0}' 2.txt |cat
ttttttttttttttt
hello  world

hey

[indeyo@VM_0_14_centos indeyo]$ awk 'NR>2' 2.txt |cat
hello  world

hey

NR 和 NF 的不同

NR表示每一行的序号，NF表示这一行的字段数。
计算行数要用 wc -l，END{print NR}也可用来输出行数

[indeyo@VM_0_14_centos indeyo]$ cat 2.txt
dddddddd
ttttttttttttttt
hello  world

hey

[indeyo@VM_0_14_centos indeyo]$ awk '{print $NR}' 2.txt  # 每一行的行数不同，第一行$1有值，直接打印出来，第二行$2没有值，所以显示空
dddddddd





[indeyo@VM_0_14_centos indeyo]$ awk '{print NR}' 2.txt  # 把每一行行号打印出来
1
2
3
4
5
6
[indeyo@VM_0_14_centos indeyo]$ awk 'NR>1{print $0}' 2.txt
ttttttttttttttt
hello  world

hey

[indeyo@VM_0_14_centos indeyo]$ sort 2.txt | uniq -c
      2 
      1 dddddddd
      1 hello  world
      1 hey
      1 ttttttttttttttt
[indeyo@VM_0_14_centos indeyo]$ wc -l 2.txt
6 2.txt
[indeyo@VM_0_14_centos indeyo]$ awk '{print $NF}' 2.txt
dddddddd
ttttttttttttttt
world

hey

[indeyo@VM_0_14_centos indeyo]$ awk '{print NF}' 2.txt  # 计算每一行的字段数
1
1
2
0
1
0

awk 单行变多行

[indeyo@VM_0_14_centos ~]$ echo $PATH
/usr/local/bin:/usr/bin:/usr/local/sbin:/usr/sbin:/home/indeyo/.local/bin:/home/indeyo/bin
[indeyo@VM_0_14_centos ~]$ echo $PATH | awk 'BEGIN{RS=":"}{print $0}'
/usr/local/bin
/usr/bin
/usr/local/sbin
/usr/sbin
/home/indeyo/.local/bin
/home/indeyo/bin

awk -F 参数，分隔符

[indeyo@VM_0_14_centos indeyo]$ echo '1 2 3' | awk -F ' ' '{print $1}'
1
[indeyo@VM_0_14_centos indeyo]$ echo '1 2 3' | awk -F ' ' '{print $2}'
2
[indeyo@VM_0_14_centos indeyo]$ echo '1 2 3' | awk -F ' ' '{print $3}'
3
[indeyo@VM_0_14_centos indeyo]$ echo '1 2 3' | awk -F '2' '{print $1}'
1 
[indeyo@VM_0_14_centos indeyo]$ echo '1 2 3' | awk -F '2' '{print $2}'
 3
[indeyo@VM_0_14_centos indeyo]$ echo '1 2 3' | awk '{print $2}' # 不填分隔符，默认空格分隔
2
[indeyo@VM_0_14_centos indeyo]$ echo '1 2 3' | awk '{print $1}'
1
[indeyo@VM_0_14_centos indeyo]$ echo '1 2 3' | awk '{print $3}'
3
[indeyo@VM_0_14_centos indeyo]$ echo '1 2       3' | awk '{print $3}' # 无论多少空格都当做分隔符
3
[indeyo@VM_0_14_centos indeyo]$ echo '1 2       3' | awk '{print $0}' # $0表示匹配到的字符串本身
1 2       3

awk OFS 的用法

[indeyo@VM_0_14_centos indeyo]$ echo '1 2    3' | awk '{OFS="-"}{print $1,$2,$3}'
1-2-3
[indeyo@VM_0_14_centos indeyo]$ echo '1 2    3' | awk '{OFS="-";print $1,$2,$3}'
1-2-3

ORS 和 OFS

疑问：为什么不是 OFS？？？

可能是因为打印出来的是很多条记录，OFS 无法对多条记录起作用，只对一条记录。（自问自答）

[86413302@shell.testing-studio.com indeyo]$echo $PATH | awk 'BEGIN{RS=":"}{print $0}' | awk 'BEGIN{OFS=":"}{print $0}'
/usr/local/bin
/usr/bin
/usr/local/sbin
/usr/sbin
/home/86413302/.local/bin
/home/86413302/bin

[86413302@shell.testing-studio.com indeyo]$echo $PATH | awk 'BEGIN{RS=":"}{print $0}' | awk 'BEGIN{RS="\n";OFS=":"}{print $0}'
/usr/local/bin
/usr/bin
/usr/local/sbin
/usr/sbin
/home/86413302/.local/bin
/home/86413302/bin

[86413302@shell.testing-studio.com indeyo]$echo $PATH | awk 'BEGIN{RS=":"}{print $0}' | awk 'BEGIN{ORS=":"}{print $0}'
/usr/local/bin:/usr/bin:/usr/local/sbin:/usr/sbin:/home/86413302/.local/bin:/home/86413302/bin::[86413302@shell.testing-studio.com indeyo]$

awk 字段分割

[86413302@shell.testing-studio.com indeyo]$echo $PATH | awk 'BEGIN{RS=":"}{print $0}'  
/usr/local/bin
/usr/bin
/usr/local/sbin
/usr/sbin
/home/86413302/.local/bin
/home/86413302/bin

[86413302@shell.testing-studio.com indeyo]$echo $PATH | awk 'BEGIN{RS=":"}{print $0}'| awk -F / '{print $1,$2,$3,$4}'
 usr local bin
 usr bin 
 usr local sbin
 usr sbin 
 home 86413302 .local
 home 86413302 bin

[86413302@shell.testing-studio.com indeyo]$echo $PATH | awk 'BEGIN{RS=":"}{print $0}'| awk 'BEGIN{FS="/"}{print $1,$2,$3,$4}'
 usr local bin
 usr bin 
 usr local sbin
 usr sbin 
 home 86413302 .local
 home 86413302 bin

[86413302@shell.testing-studio.com indeyo]$echo $PATH | awk 'BEGIN{FS=":";OFS=" | "}{$1=$1;print $0}'
/usr/local/bin | /usr/bin | /usr/local/sbin | /usr/sbin | /home/86413302/.local/bin | /home/86413302/bin
[86413302@shell.testing-studio.com indeyo]$echo $PATH | awk 'BEGIN{FS=":";OFS=" | "}{$0=$0;print $0}'
/usr/local/bin:/usr/bin:/usr/local/sbin:/usr/sbin:/home/86413302/.local/bin:/home/86413302/bin
[86413302@shell.testing-studio.com indeyo]$echo $PATH | awk 'BEGIN{FS=":";OFS=" | "}{$2=$2;print $0}'
/usr/local/bin | /usr/bin | /usr/local/sbin | /usr/sbin | /home/86413302/.local/bin | /home/86413302/bin

$1=$1 是什么意思？和 $0=$0 有什么不同？

看了 chinaunix 帖子，发现 2011 那个时候的网络用语很有意思。大家喜欢用 “呵呵”，帅旋说那个时候用这个词可是很酷的。还有 “浮云”，啊，多有年代感的词啊哈哈。

而且奖品只是鼠标、文化衫和马克杯，但是却吸引了很多大神，想不通。最后一题是写一个五子棋的实现，真的有人写出来，250+ 行代码，好像只有一个人答了这题。可能真的很爱技术吧？

在帖子中看到一个大神的解释，非常易懂。

http://bbs.chinaunix.net/thread-2319120-1-1.html

命令	作用	使用场景
`$1 = $1`	重新计算`$0`	用了`OFS`，且需输出`$0`
`$0 = $0`	重新计算`$N`和`NF`	改变了`FS`，且需输出`$1`或`NF`时

实践了一下，确实如此。

[indeyo@VM_0_14_centos indeyo]$ echo 'A B C D' | awk '{OFS="|";print $0;$1=$1;print $0}' # $1=$1 重新计算了 $0
A B C D
A|B|C|D
[indeyo@VM_0_14_centos indeyo]$ echo 'A B C D' | awk 'BEGIN{OFS="|"}{print $0;$1=$1;print $0}'  # BEGIN 对 OFS 不起作用
A B C D
A|B|C|D
[indeyo@VM_0_14_centos indeyo]$ echo 'A|B|C|D' | awk '{FS="|";print $1,NF;$0=$0;print $1,NF}'  # $0=$0 改变了 $1 和 NF
A|B|C|D 1
A 4
[indeyo@VM_0_14_centos indeyo]$ echo 'A|B|C|D' | awk 'BEGIN{FS="|"}{print $1,NF;$0=$0;print $1,NF}'  # 用 BEGIN 可以代替 $0=$0 的效果
A 4
A 4

awk 数据计算

[indeyo@VM_0_14_centos indeyo]$ echo '1,10
> 2,20
> 3,30' | awk 'BEGIN{a=0;FS=","}{a+=$2}END{print a/NR}' # 计算平均值
20
[indeyo@VM_0_14_centos ~]$ echo '
1
2
3
4' | awk '/^..*/{t+=$1;print NR,$1,t,t/NR}' # /^..*/ 过滤掉第一行空白行。这个有点累赘，可以用/^./，表示有任意字符就行，相当于去掉空白行
2 1 1 0.5
3 2 3 1
4 3 6 1.5
5 4 10 2
[86413302@shell.testing-studio.com indeyo]$echo ' # /^./表示有任意字符就行，相当于去掉空白行


1
2
3
4 ' | awk '/^./{print $0}'
1
2
3
4

sed

sed的主要功能是替换，类似于 sql 里面的 update 语句。

sed 的一些常见用法

[indeyo@VM_0_14_centos ~]$ echo '1
> 2
> 3
> 4' | sed -n 1p   # 打印
1
[indeyo@VM_0_14_centos ~]$ echo '1
2
3
4' | sed -n 2p
2
[indeyo@VM_0_14_centos ~]$ echo '1
2
3
4' | sed 's/2/xxxx/'  # 替换
1
xxxx
3
4
[indeyo@VM_0_14_centos ~]$ echo '1
2
3
4' | sed -e 's/2/xxxx/' -e 's&3&33333&'  # 多处替换
1
xxxx
33333
4
[indeyo@VM_0_14_centos ~]$ echo '1
2
3
4' | sed -e 's/2/xxxx/' -e 's#3#3334444#'   # 可以用除了单引号以外的符号作为分隔符
1
xxxx
3334444
4
[indeyo@VM_0_14_centos ~]$ echo 12341234 | sed 's/2/7/'
17341234
[indeyo@VM_0_14_centos ~]$ echo 12341234 | sed 's^2^7^g'
17341734
[indeyo@VM_0_14_centos ~]$ echo 12341234 | sed 's/2/7/g'  # g 表示全部替换
17341734

其他 linux 命令

/^$/ 表示空行
which 当前环境变量是否包含这个文件

[86413302@shell.testing-studio.com indeyo]$which 1.sh
/usr/bin/which: no 1.sh in (/usr/local/bin:/usr/bin:/usr/local/sbin:/usr/sbin:/home/86413302/.local/bin:/home/86413302/bin)
[86413302@shell.testing-studio.com indeyo]$which python
/usr/bin/python

echo 打印
a=1 赋值
$PATH, $PWD, $USER, $HOME 预定义变量
export path=$PATH:$PWD 添加环境变量
$变量名 引用变量
字符串中引用变量，需要用双引号

[86413302@shell.testing-studio.com indeyo]$a=1
[86413302@shell.testing-studio.com indeyo]$echo $a
1
[86413302@shell.testing-studio.com indeyo]$echo '$a'
$a
[86413302@shell.testing-studio.com indeyo]$echo "$a"
1

` 可引用命令

[86413302@shell.testing-studio.com indeyo]$echo `ll`
total 124 -rw-rw-r-- 1 86413302 86413302 104 Mar 4 14:37 1 -rwxrwxr-x 1 86413302 86413302 18 Dec 7 22:30 1.sh -rw-rw-r-- 1 86413302 86413302 478 Dec 12 16:40 20191212.txt -rw-rw-r-- 1 86413302 86413302 104 Dec 13 12:30 20191213.txt -rw-rw-r-- 1 86413302 86413302 259 Dec 10 12:41 2.txt lrwxrwxrwx 1 86413302 86413302 5 Dec 8 11:01 3.txt -> 2.txt -rw-rw-r-- 1 86413302 86413302 12 Dec 8 11:35 4.txt -rw-rw-r-- 1 86413302 86413302 147 Dec 13 12:36 8.txt -rw-rw-r-- 1 86413302 86413302 97646 Dec 8 16:23 bing_shell.txt
[86413302@shell.testing-studio.com indeyo]$ll
total 124
-rw-rw-r-- 1 86413302 86413302   104 Mar  4 14:37 1
-rwxrwxr-x 1 86413302 86413302    18 Dec  7 22:30 1.sh
-rw-rw-r-- 1 86413302 86413302   478 Dec 12 16:40 20191212.txt
-rw-rw-r-- 1 86413302 86413302   104 Dec 13 12:30 20191213.txt
-rw-rw-r-- 1 86413302 86413302   259 Dec 10 12:41 2.txt
lrwxrwxrwx 1 86413302 86413302     5 Dec  8 11:01 3.txt -> 2.txt
-rw-rw-r-- 1 86413302 86413302    12 Dec  8 11:35 4.txt
-rw-rw-r-- 1 86413302 86413302   147 Dec 13 12:36 8.txt
-rw-rw-r-- 1 86413302 86413302 97646 Dec  8 16:23 bing_shell.txt

a=(1 2 3 4 5) 数组赋值
${a[下标]}引用数组
${a[*]} , ${a[#]} 输出所有数组元素
${#a[*]} 计算元素个数

[86413302@shell.testing-studio.com indeyo]$b=(1 2 3)
[86413302@shell.testing-studio.com indeyo]$echo b
b
[86413302@shell.testing-studio.com indeyo]$echo $b
1
[86413302@shell.testing-studio.com indeyo]$echo ${b[1]}
2
[86413302@shell.testing-studio.com indeyo]$echo ${b[0]}
1
[86413302@shell.testing-studio.com indeyo]$echo ${b[2]}
3
[86413302@shell.testing-studio.com indeyo]$echo ${b[*]}
1 2 3
[86413302@shell.testing-studio.com indeyo]$echo ${#b[*]}
3

$?表示上一条语句运行结果，0 表示真，非 0 表示假
vim 文件名编辑文件，输入 i，insert 表示输入状态，ESC一下， :q表示退出，:wq保存并退出

[indeyo@VM_0_14_centos indeyo]$ ll # 查看当前文件列表
total 112
-rw-rw-r-- 1 indeyo indeyo   167 Dec 16 07:21 1.sh
-rw-rw-r-- 1 indeyo indeyo    44 Dec 27 08:28 2.txt
-rw-rw-r-- 1 indeyo indeyo 98808 Dec 11 22:59 shell_bing.txt
[indeyo@VM_0_14_centos indeyo]$ cat 2.txt # 打开一个文件
dddddddd
ttttttttttttttt
hello  world

hey

[indeyo@VM_0_14_centos indeyo]$ vim 2.txt # 编辑文件
[indeyo@VM_0_14_centos indeyo]$ less shell_bing.txt # 页数很多的时候用less查看文件，可翻页

掐头去尾

${string#匹配字符串} 掐头

${string##*匹配字符串} 贪婪匹配掐头

${string%匹配字符串} 去尾

${string%%匹配字符串*} 贪婪匹配去尾

[86413302@izuf60jasqavbxb9efockpz indeyo]$ s="hello testerhome"
[86413302@izuf60jasqavbxb9efockpz indeyo]$ echo ${s#h}
ello testerhome
[86413302@izuf60jasqavbxb9efockpz indeyo]$ echo ${s##*h}
ome
[86413302@izuf60jasqavbxb9efockpz indeyo]$ echo ${s%terhome}
hello tes
[86413302@izuf60jasqavbxb9efockpz indeyo]$ echo ${s%%t*}
hello

不带双引号则默认忽略空格

[86413302@izuf60jasqavbxb9efockpz indeyo]$ s="hello from testerhome"
[86413302@izuf60jasqavbxb9efockpz indeyo]$ echo ${s#hello}
from testerhome
[86413302@izuf60jasqavbxb9efockpz indeyo]$ echo "${s#hello}"
 from testerhome

字符串替换

[86413302@izuf60jasqavbxb9efockpz indeyo]$ echo ${s/testerhome/indeyo}
hello from indeyo
[86413302@izuf60jasqavbxb9efockpz indeyo]$ echo ${s/from/to}
hello to testerhome
[86413302@izuf60jasqavbxb9efockpz indeyo]$ echo ${s//h/xxx}
xxxello from testerxxxome

逻辑判断

[86413302@izuf60jasqavbxb9efockpz indeyo]$ [ 3 -eq 2 ]
[86413302@izuf60jasqavbxb9efockpz indeyo]$ echo $?
1
[86413302@izuf60jasqavbxb9efockpz indeyo]$ ((3>2));echo $?
0
[86413302@izuf60jasqavbxb9efockpz indeyo]$ ((3>20));echo $?
1
[86413302@izuf60jasqavbxb9efockpz indeyo]$ a="testerhome"
[86413302@izuf60jasqavbxb9efockpz indeyo]$ echo $a
testerhome
[86413302@izuf60jasqavbxb9efockpz indeyo]$ b="hello from testerhome"
[86413302@izuf60jasqavbxb9efockpz indeyo]$ echo $b
hello from testerhome
[86413302@izuf60jasqavbxb9efockpz indeyo]$ [ "$a" = "$b" ]
[86413302@izuf60jasqavbxb9efockpz indeyo]$ echo $b
hello from testerhome
[86413302@izuf60jasqavbxb9efockpz indeyo]$ [ "$a" = "$b" ]
[86413302@izuf60jasqavbxb9efockpz indeyo]$ echo $?
1
[86413302@izuf60jasqavbxb9efockpz indeyo]$ [ "$a" != "$b" ]
[86413302@izuf60jasqavbxb9efockpz indeyo]$ echo $?
0
[86413302@izuf60jasqavbxb9efockpz indeyo]$ [ -n "$a" ]  # 非空判断
[86413302@izuf60jasqavbxb9efockpz indeyo]$ echo $?
0
[86413302@izuf60jasqavbxb9efockpz indeyo]$ [ -z "$a" ]
[86413302@izuf60jasqavbxb9efockpz indeyo]$ echo $?
1
[86413302@izuf60jasqavbxb9efockpz indeyo]$ [[ "$b" == h* ]]
[86413302@izuf60jasqavbxb9efockpz indeyo]$ echo $?
0
[86413302@izuf60jasqavbxb9efockpz indeyo]$ [[ "$b" == m* ]]
[86413302@izuf60jasqavbxb9efockpz indeyo]$ echo $?
1
[86413302@izuf60jasqavbxb9efockpz indeyo]$ [ 2 -ge 1 ];echo $?
0
[86413302@izuf60jasqavbxb9efockpz indeyo]$ [ 2 -le 1 ];echo $?
1
[86413302@izuf60jasqavbxb9efockpz indeyo]$ [ 2 -ge 1 -a 3 -ge 4 ];echo $?
1
[86413302@izuf60jasqavbxb9efockpz indeyo]$ [ 2 -ge 1 -o 3 -ge 4 ];echo $?
0
[86413302@izuf60jasqavbxb9efockpz indeyo]$ [[ 2 -ge 1 && 3 -ge 4 ]];echo $?
1
[86413302@izuf60jasqavbxb9efockpz indeyo]$ [[ 2 -ge 1 || 3 -ge 4 ]];echo $?
0
[86413302@izuf60jasqavbxb9efockpz indeyo]$ [ ! 2 -ge 1 ];echo $?

`cp` 移动文件

在文件目录下，执行cp 文件名目标目录

[86413302@shell.testing-studio.com tmp]$cp nginx.log /home/86413302
[86413302@shell.testing-studio.com tmp]$cd /home/86413302
[86413302@shell.testing-studio.com ~]$ll
total 556
-rw-rw-r-- 1 86413302 86413302      0 Dec 10 07:34 b
-rw-rw-r-- 1 86413302 86413302 135583 Dec 10 07:25 bing_shell.txt
drwxrwxr-x 2 86413302 86413302   4096 Dec 13 12:36 indeyo
-rw-r--r-- 1 86413302 86413302 424573 Feb  4 09:55 nginx.log

sort 和 uniq

统计每条记录的数量，先用sort聚合，再用uniq -c统计，要用sort 待计量数据| uniq -c

[indeyo@VM_0_14_centos indeyo]$ cat 1.sh
gogog1111111111abczzzzzzz
q
11111111sdfsdf222

q
q
q


Today is Wednessday.
I wake up at 6.20.
I prepare for the soap.
It must be delicious.


2019-12-11
2019-12-12

[indeyo@VM_0_14_centos indeyo]$ sort 1.sh | uniq -c
      5 
      1  
      1 11111111sdfsdf222
      1 2019-12-11
      1 2019-12-12
      1 gogog1111111111abczzzzzzz
      1 I prepare for the soap.
      1 It must be delicious.
      1 I wake up at 6.20.
      4 q
      1 Today is Wednessday.

问题清单

这个清单是学习过程中的灵魂发问。有些问题也许不能立马就有答案，后面熟悉以后，有些是自己解答的。有种穿越过去回答问题的感觉，哈哈。

BEGIN到底是啥用法？不明白

awk 加载输入流文件之前，运行的代码块，一般用于变量初始化、打印表格表头之类的。
```
[86413302@shell.testing-studio.com indeyo]$awk 'BEGIN{ print "begin"}{ print } END{ print "end"}' 4.txt
begin
6656
123456
end
```

field 和 record 到底是哪个层面区分的？什么情况下是record？

一行数据表示record记录，一行数据里面的字段是field。通过正则取数据，取出来的就是一个record。通过 awk 的$2取出第二个field。

[indeyo@VM_0_14_centos indeyo]$ cat 2.txt
dddddddd
ttttttttttttttt
hello  world

hey

[indeyo@VM_0_14_centos indeyo]$ grep hello 2.txt
hello  world
[indeyo@VM_0_14_centos indeyo]$ grep hello 2.txt | awk '{print $2}'
world
[indeyo@VM_0_14_centos indeyo]$ grep hello 2.txt | awk '{print $1}'
hello
[indeyo@VM_0_14_centos indeyo]$ grep hello 2.txt | awk '{print $0}'
hello  world

在(GNU grep) 3.1上面，grep 的贪婪匹配不能工作的，为啥？

用grep -V查看版本号。可能是操作系统版本不支持。

?贪婪匹配在 GNU2.20 版本上，该功能无效，mac2.50 上是可以的。

[indeyo@VM_0_14_centos indeyo]$ grep -V
grep (GNU grep) 2.20
Copyright (C) 2014 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>.
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.

Written by Mike Haertel and others, see <http://git.sv.gnu.org/cgit/grep.git/tree/AUTHORS>.

为什么[^"]*可以代表任意字符串？

这里的^表示非运算，非双引号的字符
grep 命令，如果不用-E，就不能用扩展正则表达式了吗？

dei！

学习心得

作为小白来说，入门可能不那么容易。尤其是那些晦涩难记的语法格式，如果只是练习一次，随着时间递增，能记忆的部分只会线性递减。

刚学没多久的语法，上手做题的时候硬是想不出来，特别受挫。

感谢春节延长多出来的假期，有大段大段完整的时间把语法捡起来。命令多敲几遍，慢慢开始理解，之前的疑难杂症也解开了。

所以，除了多练习，没有太多的捷径。

当然，如果要记得更牢，还是尽量运用到工作中。

再啰嗦一点，要学会用man 命令名查看说明文档。英文看不懂，Google 翻译安排一下。不过，求助翻译之前，最好自己先尝试翻译一下，毕竟不会看英文文档的测试人员不是好测开。

学习链接

推荐原因：awk 语句执行顺序讲得很清楚，有较全的语法整理，方面查找

https://man.linuxde.net/awk

↙↙↙阅读原文可查看相关链接，并与作者交流