sed命令

最新推荐文章于 2022-01-11 13:44:21 发布

原创最新推荐文章于 2022-01-11 13:44:21 发布 · 200 阅读

0 ·

CC 4.0 BY-SA版权

sed 专栏收录该内容

8 篇文章

订阅专栏

本文深入解析了SED命令的各种用法，包括基本语法、选项说明、常用命令如打印(p)、删除(d)、替换(s)等，并提供了丰富的实战案例，帮助读者掌握SED在文本处理中的强大功能。

sed命令的语法格式：

sed的命令格式： sed [option] 'sed command'filename

sed的脚本格式：sed [option] -f 'sed script'filename

sed命令的选项(option)：

-n ：只打印模式匹配的行

-e ：直接在命令行模式上进行sed动作编辑，此为默认选项

-f ：将sed的动作写在一个文件内，用–f filename 执行filename内的sed动作

-r ：支持扩展表达式

-i ：直接修改文件内容

测试文件sedtest.txt,内容为：

[root@localhost sed]# cat sedtest.txt 
my cat's name is betty
   it's a cat;
This is your dog
   its name is Lokia;
That is your cat
   it is a gitl
[root@localhost sed]#

打印：p命令：

打印sedtest.txt中包含"This"的行。

默认情况sed把所有行都打印到屏幕，如果某行匹配到模式，则把该行另外再打印一遍。例如：

[root@localhost sed]# sed  '/This/p' sedtest.txt 
my cat's name is betty
   it's a cat;
This is your dog
This is your dog
   its name is Lokia;
That is your cat
   it is a gitl
[root@localhost sed]#

和上面一样，只是去掉了sed的默认行为，只会打印匹配的行

[root@localhost sed]# sed -n '/This/p' sedtest.txt 
This is your dog
[root@localhost sed]#

打印第10行到20行，sed -n '10,20p' sedtest.txt

打印从'abc'行到'def'行，sed -n '/abc/,/def/p' sedtest.txt

间隔行：步进2行，sed -n '1~2p' sedtest.txt

删除：d命令:（不会修改源文件,-i可以修改原文件）

删除从第3行到最后一行的内容。

[root@localhost sed]# sed '3,$d' sedtest.txt 
my cat's name is betty
   it's a cat;
[root@localhost sed]#

删除第二行数据内容：
```
sed -i '2d' file
```
删除最后一行的内容

[root@localhost sed]# sed '$d' sedtest.txt 
my cat's name is betty
   it's a cat;
This is your dog
   its name is Lokia;
That is your cat
[root@localhost sed]#

删除包含“is”的行

[root@localhost sed]# sed '/is/d' sedtest.txt 
   it's a cat;
[root@localhost sed]#

替换：s命令（默认不会修改原文件，"-i"可以修改原文件）

把行内的所有"it"替换成"IT"，如果没有g,则只替换行内的第一个it

[root@localhost sed]# sed 's/it/IT/' sedtest.txt 
my cat's name is betty
   IT's a cat,it has a tail;
This is your dog
   ITs name is Lokia;
That is your cat
   IT is a gitl
[root@localhost sed]#

/g 表示全部替换

[root@localhost sed]# sed 's/it/IT/g' sedtest.txt 
my cat's name is betty
   IT's a cat,IT has a tail;
This is your dog
   ITs name is Lokia;
That is your cat
   IT is a gITl
[root@localhost sed]#

-n /p 只打印发生替换的那些行

[root@localhost sed]# sed  -n 's/it/It/gp' sedtest.txt 
   It's a cat,It has a tail;
   Its name is Lokia;
   It is a gItl
[root@localhost sed]#

sed ‘s/it/&123/’ sedtest.txt :在行内第一个出现的的it后面添加123（&表示匹配的内容）

[root@localhost sed]# sed  's/it/&123/' sedtest.txt 
my cat's name is betty
   it123's a cat,it has a tail;
This is your dog
   it123s name is Lokia;
That is your cat
   it123 is a gitl
[root@localhost sed]#

sed 's/it/&123/g' sedtest.txt：在所有的it后面添加123

[root@localhost sed]# sed  's/it/&123/g' sedtest.txt 
my cat's name is betty
   it123's a cat,it123 has a tail;
This is your dog
   it123s name is Lokia;
That is your cat
   it123 is a git123l
[root@localhost sed]#

sed -n 's/it/&123/gp' sedtest.txt ：在所有it后面添加123，打印受影响的行数

[root@localhost sed]# sed -n 's/it/&123/gp' sedtest.txt 
   it123's a cat,it123 has a tail;
   it123s name is Lokia;
   it123 is a git123l
[root@localhost sed]#

sed -i 's/^/{"index":{}}\n/g' test.txt: 在每一行前都添加一行{"index":{}}\n
sed 's#it#abc#g' sedtest.txt ：把所有的it替换成abc，跟在替换s后面的字符就是查找串和替换串之间的分割字符，本例中试#

[root@localhost sed]# sed 's#it#abc#g' sedtest.txt 
my cat's name is betty
   abc's a cat,abc has a tail;
This is your dog
   abcs name is Lokia;
That is your cat
   abc is a gabcl
[root@localhost sed]#

指定行的范围：逗号

sed -n '/dog/,/cat/p' sedtest.txt：打印模式dog到cat的行

[root@localhost sed]# sed -n '/dog/,/cat/p' sedtest.txt 
This is your dog
   its name is Lokia;
That is your cat
[root@localhost sed]#

sed -n '2,/Lokia/p' sedtest.txt ：打印从第2行到包含Lokia行之间的行

[root@localhost sed]# sed -n '2,/Lokia/p' sedtest.txt 
   it's a cat,it has a tail;
This is your dog
   its name is Lokia;
[root@localhost sed]#

sed '/dog/,/cat/s/is/IS/g' sedtest.txt ：修改从模式dog到模式cat之间的行，把is替换成IS

[root@localhost sed]# sed '/dog/,/cat/s/is/IS/g' sedtest.txt 
my cat's name is betty
   it's a cat,it has a tail;
ThIS IS your dog
   its name IS Lokia;
That IS your cat
   it is a gitl
[root@localhost sed]#

多重编辑-e

sed -e '1,3d' -e 's/is/IS/g' sedtest.txt：删除1-3行，然后把其余行的is替换成IS

[root@localhost sed]# sed -e '1,3d' -e 's/is/IS/g' sedtest.txt 
   its name IS Lokia;
That IS your cat
   it IS a gitl
[root@localhost sed]#

读文件：r命令

sed '/it/r sedtest2.txt' sedtest.txt：在匹配'it'的行的下一行开始加入文件sedtest.txt的内容

vincent@vincent-duan:~/temp$ sed '/it/r sedtest2.txt' ./sedtest.txt 
my cat's name is betty
   it's a cat;
this is sedtest2.txt
This is your dog
   its name is Lokia;
this is sedtest2.txt
That is your cat
   it is a gitl
this is sedtest2.txt
vincent@vincent-duan:~/temp$

写文件：w命令

sed '/it/w sedtest2.txt' ./sedtest.txt ：在包含‘it’的行写入sedtest2.txt

vincent@vincent-duan:~/temp$ cat sedtest2.txt 
hello world
vincent@vincent-duan:~/temp$ sed '/it/w sedtest2.txt' ./sedtest.txt 
my cat's name is betty
   it's a cat;
This is your dog
   its name is Lokia;
That is your cat
   it is a gitl
vincent@vincent-duan:~/temp$ cat sedtest2.txt 
   it's a cat;
   its name is Lokia;
   it is a gitl
vincent@vincent-duan:~/temp$

追加：a命令

sed '/it/a\def' sedtest.txt ：在包含‘it’行后新起一行，写入def

vincent@vincent-duan:~/temp$ sed '/it/a\def' sedtest.txt 
my cat's name is betty
   it's a cat;
def
This is your dog
   its name is Lokia;
def
That is your cat
   it is a gitl
def
vincent@vincent-duan:~/temp$

在第5行后添加一行：sed '5a=====' sedtest.txt

插入：i命令

sed '/it/i\def' sedtest.txt ：在包含‘it’行前插入一行，写入def

vincent@vincent-duan:~/temp$ sed '/it/i\def' sedtest.txt 
my cat's name is betty
def
   it's a cat;
This is your dog
def
   its name is Lokia;
That is your cat
def
   it is a gitl
vincent@vincent-duan:~/temp$

修改：c命令

sed '/This/c\THIS' sedtest.txt ：将This替换成THIS，不会修改原文件（-i可以实现修改原文件）

vincent@vincent-duan:~/temp$ sed '/This/c\THIS' sedtest.txt 
my cat's name is betty
   it's a cat;
THIS
   its name is Lokia;
That is your cat
   it is a gitl
vincent@vincent-duan:~/temp$

读取下一行：n命令

sed '/Lokia/{n;s/That/THAT/g;}' sedtest.txt：读取包含Lokia的行的下一行，替换That为THAT

vincent@vincent-duan:~/temp$ sed '/Lokia/{n;s/That/THAT/g;}' sedtest.txt 
my cat's name is betty
   it's a cat;
This
   its name is Lokia;
THAT is your cat
   it is a gitl
vincent@vincent-duan:

转换：y命令

sed 'y/sl/SL/' sedtest.txt ：将sedtest.txt中的s替换成S,l替换成L

vincent@vincent-duan:~/temp$ sed 'y/sl/SL/' sedtest.txt 
my cat'S name iS betty
   it'S a cat;
ThiS iS your dog
   itS name iS Lokia;
That iS your cat
   it iS a gitL
vincent@vincent-duan:~/temp$

退出：q命令

sed '/it/{n;s/This/THIS/;q}' sedtest.txt：将sedtest.txt中包含it的下一行，将This替换成THIS,然后退出

vincent@vincent-duan:~/temp$ sed '/it/{n;s/This/THIS/g;q}' sedtest.txt 
my cat's name is betty
   it's a cat;
THIS is your dog
vincent@vincent-duan:~/temp$

暂存和取用：h命令（把模式行存储到暂存缓冲区）和g（取出暂存缓冲区的行并覆盖模式缓冲区）G（取出临时缓冲区的行）命令

h和g是复制行为(覆盖），H和G表示追加。

sed -e '/is/h' -e '$G' sedtest.txt：包含abc的行通过h命令保存到暂存缓冲区，在第二条命令汇中，sed读到最后一行$时，G命令从暂存缓冲区中读取一行，追加到模式缓冲区的后面。即所有包含abc的行的最后一行被复制到文件末尾。

vincent@vincent-duan:~/temp$ sed -e '/is/h' -e '$G' sedtest.txt 
my cat's name is betty
   it's a cat;
This is your dog
   its name is Lokia;
That is your cat
   it is a gitl
   it is a gitl
vincent@vincent-duan:~/temp$

{}

sed '{20,30d;s/false/true/g}' file.txt 先将20行到30行删除，然后将flase替换成true

()

 :替换某种(部分)字符串(\1,\2)

删除每行第一个tab

sed -i 's/^[\t]//' file

替换{"username":"vincent","age":"25"}删除"username":"vincent"

sed -i 's/"username":".*"//g' ./file.txt

删除每行第一个:数字+\t

sed 's/[0-9]*\t/' ./file.txt

删除第一行到最后一行数据：

sed -i "1,\$d" ./file.txt