分类: LINUX
2008-04-08 14:22:42
引用: |
• 抽取域。 • 匹配正则表达式。 • 比较域。 • 增加、附加、替换。 • 基本的s e d命令和一行脚本。 |
代码: |
sed [选项] s e d命令输入文件。 |
代码: |
sed [选项] -f sed脚本文件输入文件 |
代码: |
s e d脚本文件[选项] 输入文件 |
引用: |
s e d选项如下: n 不打印;s e d不写编辑行到标准输出,缺省为打印所有行(编辑和未编辑)。p命令可以用来打印编辑行。 c 下一命令是编辑命令。使用多项编辑时加入此选项。如果只用到一条s e d命令,此选项无用,但指定它也没有关系。 f 如果正在调用s e d脚本文件,使用此选项。此选项通知s e d一个脚本文件支持所有的s e d命令,例如:sed -f myscript.sed input_file,这里m y s c r i p t . s e d即为支持s e d命令的文件。 |
代码: |
$sed 'some-sed-commands' input-file > myoutfile |
引用: |
1) 使用行号,可以是一个简单数字,或是一个行号范围。 2 ) 使用正则表达式 |
代码: |
x x为一行号,如1 x , y 表示行号范围从x到y,如2,5表示从第2行到第5行 / p a t t e r n / 查询包含模式的行。例如/ d i s k /或/[a-z]/ / p a t t e r n / p a t t e r n / 查询包含两个模式的行。例如/ d i s k / d i s k s / p a t t e r n / , x 在给定行号上查询包含模式的行。如/ r i b b o n / , 3 x , / p a t t e r n / 通过行号和模式查询匹配行。3 . / v d u / x , y ! 查询不包含指定行号x和y的行。1 , 2 ! |
代码: |
sed编辑命令 p 打印匹配行 = 显示文件行号 a \ 在定位行号后附加新文本信息 i \ 在定位行号后插入新文本信息 d 删除定位行 c \ 用新文本替换定位文本 s 使用替换模式替换相应模式 r 从另一个文件中读文本 w 写文本到一个文件 q 第一个模式匹配完成后推出或立即推出 l 显示与八进制A S C I I代码等价的控制字符 { } 在定位行执行的命令组 n 从另一个文件中读文本下一行,并附加在下一行 g 将模式2粘贴到/pattern n/ y 传送字符 n 延续到下一输入行;允许跨行的模式匹配语句 |
sed和正则表达式
s e d识别任何基本正则表达式和模式及其行匹配规则。记住规则之一是:如果要定位一特殊字符,必须使用( \)屏蔽其特殊含义
s e d例子中使用下述文本文件q u o t e . t x t。
代码: |
[sam@Linux_chenwy sam]$ cat quote.txt The honeysuckle band played all night long for only $90. It was an evening of splendid music and company. Too bad the disco floor fell through at 23:00. The local nurse Miss P.Neave was in attendance. |
1 使用p(rint)显示行
只打印第二行,用-n
代码: |
[sam@Linux_chenwy sam]$ sed -n '2p' quote.txt It was an evening of splendid music and company. |
2 打印范围
可以指定行的范围,现打印1到3行,用逗号分隔行号。
代码: |
[sam@Linux_chenwy sam]$ sed -n '1,3p' quote.txt The honeysuckle band played all night long for only $90. It was an evening of splendid music and company. Too bad the disco floor fell through at 23:00. |
3 打印模式
假定要匹配单词N e a v e,并打印此行,方法如下。使用模式/ p a t t e r n /格式,这里为/ N e a v e /。
代码: |
[sam@Linux_chenwy sam]$ sed -n '/The/p' quote.txt The honeysuckle band played all night long for only $90. The local nurse Miss P.Neave was in attendance. |
4 使用模式和行号进行查询
可以将行号和模式结合使用。假定要改动文件q u o t e . t x t最后一行中的单词t h e,使用s e d查询t h e,返回两行:
代码: |
[sam@Linux_chenwy sam]$ sed -n '/The/p' quote.txt The honeysuckle band played all night long for only $90. The local nurse Miss P.Neave was in attendance. |
使用模式与行号的混合方式可以剔除第一行,格式为l i n e _ n u m b e r, / p a t t e r n /。逗号用来分隔行号与模式开始部分。为达到预期结果,使用4 , / t h e /。意即只在第四行查询模式t h e,命令如下:
代码: |
[sam@Linux_chenwy sam]$ sed -n '4,/The/p' quote.txt The local nurse Miss P.Neave was in attendance. |
上面有错,其实是把第四行后的都打出来了
这个模式应该哪果指定行找不到符合条件的,就从下一行开始查找,直到找到为止,并把,找到行之前的全部打打印出来。
如果指定行本身就符合条伯,把本行及后面的行的全部打印出来
5 匹配元字符
匹配元字符$前,必须使用反斜线\屏蔽其特殊含义。模式为/\$/ p。
代码: |
[sam@Linux_chenwy sam]$ sed -n '/\$/p' quote.txt The honeysuckle band played all night long for only $90. |
6 显示整个文件
要打印整个文件,只需将行范围设为第一行到最后一行1 , $。$意为最后一行。
代码: |
[sam@Linux_chenwy sam]$ sed -n '1,$p' quote.txt The honeysuckle band played all night long for only $90. It was an evening of splendid music and company. Too bad the disco floor fell through at 23:00. The local nurse Miss P.Neave was in attendance. |
7 任意字符
匹配任意字母,后跟任意字母的0次或多次重复,并以i n g结尾,模式为/ . * i n g /。可以使用这个模式查询以i n g结尾的任意单词。
代码: |
[sam@Linux_chenwy sam]$ sed -n '/.*ing/p' quote.txt It was an evening of splendid music and company. |
8 首行
要打印文件第一行,使用行号:
代码: |
[sam@Linux_chenwy sam]$ sed -n '1p' quote.txt The honeysuckle band played all night long for only $90. |
9 最后一行
要打印最后一行,使用$。$是代表最后一行的元字符。
代码: |
[sam@Linux_chenwy sam]$ sed -n '$p' quote.txt The local nurse Miss P.Neave was in attendance. |
10 打印行号
要打印行号,使用等号=。打印模式匹配的行号,使用格式/ p a t t e r n / =。
代码: |
[sam@Linux_chenwy sam]$ sed -e '/music/=' quote.txt The honeysuckle band played all night long for only $90. 2 It was an evening of splendid music and company. Too bad the disco floor fell through at 23:00. The local nurse Miss P.Neave was in attendance. |
整个文件都打印出来,并且匹配行打印了行号。如果只关心实际行号,使用- e选项。
代码: |
[sam@Linux_chenwy sam]$ sed -n '/music/=' quote.txt 2 |
如果只打印行号及匹配行,必须使用两个s e d命令,并使用e选项。第一个命令打印模式匹配行,第二个使用=选项打印行号,格式为sed -n -e /pattern/p -e /pattern/=
代码: |
[sam@Linux_chenwy sam]$ sed -n -e '/music/p' -e '/music/=' quote.txt It was an evening of splendid music and company. 2 |
P.S:新版本的sed(GNU sed version 4.0.5)带有-i选项,允许直接编辑修改文件内容而不需要重定向到临时文件
例如:
/home/lee#cat file
old
abc
/home/lee#sed -i 's/old/new/' file
/home/lee#cat file
new
abc
11 附加文本
要
附加文本,使用符号a \,可以将指定文本一行或多行附加到指定行。如果不指定文本放置位置, s e
d缺省放在每一行后面。附加文本时不能指定范围,只允许一个地址模式。文本附加操作时,结果输出在标准输出上。注意它不能被编辑,因为s e
d执行时,首先将文件的一行文本拷贝至缓冲区,在这里s e d编辑命令执行所有操作(不是在初始文件上),因为文本直接输出到标准输出,s e
d并无拷贝。
要想在附加操作后编辑文本,必须保存文件,然后运行另一个s e d命令编辑它。这时文件的内容又被移至缓冲区。
附加操作格式如下:
代码: [address]a\
text\
text\
......
text
地
址指定一个模式或行号,定位新文本附加位置。a\ 通知s e d对a
\后的文本进行实际附加操作。观察格式,注意每一行后面有一斜划线,这个斜划线代表换行。s e
d执行到这儿,将创建一新行,然后插入下一文本行。最后一行不加斜划线, s e d假定这是附加命令结尾。
当附加或插入文本或键入几个s e d命令时,可以利用辅助的s h e l l提示符以输入多行命令。当附加或插入文本或键入几个s e d命令时,可以利用辅助的s h e l l提示符以输入多行命令。
创建sed脚本文件
创建脚本文件a p p e n d . s e d:
第一行是s e d命令解释行。脚本在这一行查找s e d以运行命令,这里定位在/ b i n。
第二行以/ c o m p a n y /开始,这是附加操作起始位置。a \通知s e d这是一个附加操作,首先应插入一个新行。
第三行是附加操作要加入到拷贝的实际文本。
输出显示附加结果。如果要保存输出,重定向到一个文件。
代码: [sam@chenwy sam]$ cat append.sed
#!/bin/sed -f
/company/ a\
Then suddenly it happed.
保存它,增加可执行权限,运行
代码: [sam@chenwy sam]chmod u+x append.sed
[sam@chenwy sam]$ ./append.sed quote.txt
The honeysuckle band played all night long for only $90.
It was an evening of splendid music and company.
Then suddenly it happed.
Too bad the disco floor fell through at 23:00.
The local nurse Miss P.Neave was in attendance.
或直接用命令行:
代码: [sam@chenwy sam]$ sed "/company/a\Then suddenly it happened." quote.txt
The honeysuckle band played all night long for only $90.
It was an evening of splendid music and company.
Then suddenly it happened.
Too bad the disco floor fell through at 23:00.
The local nurse Miss P.Neave was in attendance.
[sam@chenwy sam]$ sed "/company/i\utter confusion followed." quote.txt
The honeysuckle band played all night long for only $90.
utter confusion followed.
It was an evening of splendid music and company.
Too bad the disco floor fell through at 23:00.
The local nurse Miss P.Neave was in attendance.
插入文本:
插入命令类似于附加命令,只是在指定行前面插入。和附加命令一样,它也只接受一个地址。
如在a t t e n d a n c e结尾的行前插入文本utter confusion followed。
代码: [sam@chenwy sam]$ sed "/company/i\Utter confusion followed." quote.txt
也可以指定行:
代码: [sam@chenwy sam]$ cat insert.sed
#!/bin/sed -f
4 i\
Utter confusion followed.
执行结果
代码: [sam@chenwy sam]$ chmod u+x insert.sed
[sam@chenwy sam]$ ./insert.sed quote.txt
The honeysuckle band played all night long for only $90.
It was an evening of splendid music and company.
Too bad the disco floor fell through at 23:00.
Utter confusion followed.
The local nurse Miss P.Neave was in attendance.
修改文本
修改命令将在匹配模式空间的指定行用新文本加以替代,格式如下:
将
第一行The honeysuckle band played all night long for only $90替换为The office
Di b b l e band played well。首先要匹配第一行的任何部分,可使用模式‘ / H o n e y s u c k l
e /’。s e d脚本文件为c h a n g e . s e d。内容如下:
代码: [sam@chenwy sam]$ cat change.sed
#!/bin/sed -f
3 c\
The office Dibble band played well.
代码: [sam@chenwy sam]$ chmod u+x change.sed
[sam@chenwy sam]$ ./change.sed quote.txt
The honeysuckle band played all night long for only $90.
It was an evening of splendid music and company.
The office Dibble band played well.
The local nurse Miss P.Neave was in attendance.
或命令行:
代码: [sam@chenwy sam]$ sed "/honeysuck/c\The Office Dibble band played well." quote.txt
The Office Dibble band played well.
It was an evening of splendid music and company.
Too bad the disco floor fell through at 23:00.
The local nurse Miss P.Neave was in attendance.
可以对同一个脚本中的相同文件进行修改、附加、插入三种动作匹配和混合操作。
代码: [sam@chenwy sam]$ cat mix.sed
#!/bin/sed -f
1 c\
The Dibble band were grooving.
/evening/ i\
They played some great tunes.
3 a\
Where was the nurse to help?
代码: [sam@chenwy sam]$ chmod u+x mix.sed
[sam@chenwy sam]$ ./mix.sed quote.txt
The Dibble band were grooving.
They played some great tunes.
It was an evening of splendid music and company.
Too bad the disco floor fell through at 23:00.
Where was the nurse to help?
The local nurse Miss P.Neave was in attendance.
删除文本
s e d删除文本格式:
代码: [ a d d r e s s [,a d d r e s s ] ] d
删除第一行;1 d意为删除第一行。
代码: [sam@chenwy sam]$ sed '1d' quote.txt
It was an evening of splendid music and company.
Too bad the disco floor fell through at 23:00.
The local nurse Miss P.Neave was in attendance.
删除第一到第三行:
代码: [sam@chenwy sam]$ sed '1,3d' quote.txt
The local nurse Miss P.Neave was in attendance.
删除最后一行:
代码: [sam@chenwy sam]$ sed '$d' quote.txt
The honeysuckle band played all night long for only $90.
It was an evening of splendid music and company.
Too bad the disco floor fell through at 23:00.
也可以使用正则表达式进行删除操作。下面的例子删除包含文本‘ N e a v e’的行。
代码: [sam@chenwy sam]$ sed '/Neave/d' quote.txt
The honeysuckle band played all night long for only $90.
It was an evening of splendid music and company.
Too bad the disco floor fell through at 23:00.
替换文本
替换命令用替换模式替换指定模式,格式为:
代码: [ a d d r e s s [,address]] s/ pattern-to-find /replacement-pattern/[g p w n]
s选项通知s e d这是一个替换操作,并查询p a t t e r n - t o - f i n d,成功后用r e p l a c e m e n t - p a t t e r n替换它。
替换选项如下:
引用: g 缺省情况下只替换第一次出现模式,使用g选项替换全局所有出现模式。
p 缺省s e d将所有被替换行写入标准输出,加p选项将使- n选项无效。- n选项不打印输出结果。
w 文件名使用此选项将输出定向到一个文件。
如替换n i g h t为N I G H T,首先查询模式n i g h t,然后用文本N I G H T替换它。
代码: [sam@chenwy sam]$ sed 's/night/NIGHT/' quote.txt
The honeysuckle band played all NIGHT long for only $90.
It was an evening of splendid music and company.
Too bad the disco floor fell through at 23:00.
The local nurse Miss P.Neave was in attendance.
要从$ 9 0 中删除$ 符号(记住这是一个特殊符号,必须用\ 屏蔽其特殊含义),在r e p l a c e m e n t - p a t t e r n部分不写任何东西,保留空白,但仍需要用斜线括起来。在s e d中也可以这样删除一个字符串。
代码: [sam@chenwy sam]$ sed 's/\$//' quote.txt
The honeysuckle band played all night long for only 90.
It was an evening of splendid music and company.
Too bad the disco floor fell through at 23:00.
The local nurse Miss P.Neave was in attendance.
要进行全局替换,即替换所有出现模式,只需在命令后加g选项。下面的例子将所有T h e替换成Wo w!。
代码: [sam@chenwy sam]$ sed 's/The/Wow!/g' quote.txt
Wow! honeysuckle band played all night long for only $90.
It was an evening of splendid music and company.
Too bad the disco floor fell through at 23:00.
Wow! local nurse Miss P.Neave was in attendance.
将替换结果写入一个文件用w选项,下面的例子将s p l e n d i d替换为S P L E N D I D的替换结果写入文件s e d . o u t:
代码: [sam@chenwy sam]$ sed 's/splendid/SPLENDID/w sed.out' quote.txt
The honeysuckle band played all night long for only $90.
It was an evening of SPLENDID music and company.
Too bad the disco floor fell through at 23:00.
The local nurse Miss P.Neave was in attendance.
注意要将文件名括在s e d的单引号里。文件结果如下:
代码: [sam@chenwy sam]$ cat sed.out
It was an evening of SPLENDID music and company.
使用替换修改字符串
如果要附加或修改一个字符串,可以使用(&)命令,&命令保存发现模式以便重新调用它,然后把它放在替换字符串里面。
先给出一个被替换模式,然后是一个准备附加在第一个模式后的另一个模式,并且后面带有&,这样修改模式将放在匹配模式之前。
例如, s e d语句s/nurse/"Hello"&/p 的结果如下
代码: [sam@chenwy sam]$ sed -n 's/nurse/"hello" &/p' quote.txt
The local "hello" nurse Miss P.Neave was in attendance.
原句是文本行The local nurse Miss P.Neave was in attendance。
记住模式中要使用空格,因为输出结果表明应加入空格。
还有一个例子:
代码: [sam@chenwy sam]$ sed -n 's/played/from Hockering &/p' quote.txt
The honeysuckle band from Hockering played all night long for only $90.
原句是The honeysuckle band played all night long for only $90。
将sed结果写入文件命令
像使用>文件重定向发送输出到一个文件一样,在s e d命令中也可以将结果输入文件。格式有点像使用替换命令:
代码: [ a d d r e s s [,address]]w filename
‘w’选项通知s e d将结果写入文件。f i l e n a m e是自解释文件名。
下面有两个例子。
代码: [sam@chenwy sam]$ sed '1,2 w filedt' quote.txt
The honeysuckle band played all night long for only $90.
It was an evening of splendid music and company.
Too bad the disco floor fell through at 23:00.
The local nurse Miss P.Neave was in attendance.
文件q u o t e . t x t输出到屏幕。模式范围即1,2行输出到文件f i l e d t。
代码: [sam@chenwy sam]$ cat filedt
The honeysuckle band played all night long for only $90.
It was an evening of splendid music and company.
下面例子中查询模式N e a v e,匹配结果行写入文件f i l e d h t。
代码: [sam@chenwy sam]$ sed '/Neave/ w dht' quote.txt
The honeysuckle band played all night long for only $90.
It was an evening of splendid music and company.
Too bad the disco floor fell through at 23:00.
The local nurse Miss P.Neave was in attendance.
代码: [sam@chenwy sam]$ cat dht
The local nurse Miss P.Neave was in attendance.
从文件中读文本
处理文件时, s e d允许从另一个文件中读文本,并将其文本附加在当前文件。此命令放在模式匹配行后,格式为:
代码: address r filename
这里r通知s e d将从另一个文件源中读文本。f i l e n a m e是其文件名。
现在创建一个小文件s e d e x . t x t,内容如下:
代码: [sam@chenwy sam]$ echo "Boom boom went the music" >sedex.txt
[sam@chenwy sam]$ cat sedex.txt
Boom boom went the music
将s e d e x . t x t内容附加到文件q u o t e . t x t的拷贝。在模式匹配行/ c o m p a n y /后放置附加文本。本例为第三行。注意所读的文件名需要用单引号括起来。
代码: [sam@chenwy sam]$ sed '/company./r sedex.txt' quote.txt
The honeysuckle band played all night long for only $90.
It was an evening of splendid music and company.
Boom boom went the music
Too bad the disco floor fell through at 23:00.
The local nurse Miss P.Neave was in attendance.
匹配后退出
有时需要在模式匹配首次出现后退出s e d,以便执行其他处理脚本。退出命令格式为:
代码: address q
下面的例子假定查询模式/ . a . * /,意为任意字符后跟字符a,再跟任意字符0次或任意多次。
查询首次出现模式,然后退出。需要将q放在s e d语句末尾。
代码: [sam@chenwy sam]$ sed '/.a.*/q' quote.txt
The honeysuckle band played all night long for only $90.
显示文件中的控制字符
1、$vi dos.txt
进入vi后,用ctrl+v 再用ctrl+M产生控制字符^M不知对不对
使用cat -v filename命令查看编辑好的文件
代码: [sam@chenwy sam]$ cat -v dos.txt
12332##DISO##45.12^M
00332##LPSO##23.14^M
01299##USPD##34.46^M
s e d格式为:
代码: [ a d d r e s s,[ a d d r e s s ] ] l
‘l’意为列表。一般情况下要列出整个文件,而不是模式匹配行,因此使用l要从第一到最后一行。模式范围1,$即为此意。
代码: [sam@chenwy sam]$ sed -n '1,$l' dos.txt
12332##DISO##45.12\r$
00332##LPSO##23.14\r$
01299##USPD##34.46\r$
处理控制字符
使用s e d实现的一个重要功能是在另一个系统中下载的文件中剔除控制字符。
下面是传送过来的文件( d o s . t x t)的部分脚本。必须去除所有可疑字符,以便于帐号所有者使用文件。
删除所有的#字符很容易,可以使用全局替换命令。这里用一个空格替换两个或
更多的#符号。
代码: [sam@chenwy sam]$ sed 's/##/ /g' dos.txt
12332 DISO 45.12
00332 LPSO 23.14
01299 USPD 34.46
。删除所有行首的0。使用^符号表示模式从行首开始, ^ 0 *表示行首任意个0。模式s / ^ 0 * / / g设置替换部分为空,即为删除模式,正是要求所在。
代码: [sam@chenwy sam]$ sed 's/##/ /g;s/^0*/ /g' dos.txt
12332 DISO 45.12
332 LPSO 23.14
1299 USPD 34.46
最后去除行尾^ M符号,为此需做全局替换。设置替换部分为空。模式为:
‘s / ^ m / / g’,注意‘^ M’,这是一个控制字符。
在命令行里也必须用^M控制字符耶!?
或 代码: [sam@chenwy sam]$ sed 's/##/ /g;s/^0*/ /g;s/^M/ /g' dos.txt
12332 DISO 45.12
332 LPSO 23.14
1299 USPD 34.46
代码: [sam@chenwy sam]$ cat dos.txt | sed 's/^0*/ /g' | sed 's/^M/ /g' | sed 's/##/ /g'