::显示文本前10行

sed 11,$d a.txt

::显示文本最后一行

sed -n $p a.txt

::显示文本最后一行

sed $!d a.txt

sed s/；/\n/g a.txt

sed -n "20,25p" a.txt

sed -n "p;n"  a.txt

sed -n "n;p"  a.txt

sed -e "s/爱你/【爱你】/g" -e "s/恨你/（恨你）/g" a.txt

sed "s// /g" a.txt

sed "s/^ *//;s/ *$//" a.txt

sed "s/^/echo/" a.txt

sed "s/。$//" a.txt

sed "s/^*//;s/*$//;s/ //g" file

 sed "s/。\t*$//"

删正行

sed  "/。$/d"

sed -e "s/\(..........\).*/\1/" if.txt

:: 12，如果某行的行尾是以"。"结尾，就删除"。"

sed -e "s/。$//" a.txt | more

:: 13，如果某行多于10个字符，只显示每一行的前10个字符

sed -n "/........../{s/\(..........\).*/\1/p}" a.txt | more

1

sed -n -e "1,10p;$p" file

::1	用-n禁止输出，这样打印1到10和末行



2

sed -e "s/;/\n/g;s/,/\n/g;" file

::2	简单替换



3

sed -e "20,25!d" file

::3	同你第一个，20到25不清空



4

sed -e "s/.*//g;N;" file

::清空ps并读入下一行,偶行输出

sed -e "1!{s/.*//g;N}" file

::同上，只是跳过第一行，奇行输出



(想下面时出了好点的解决方法：

sed -e "p;N;d" file

::直接打印一行，多读一行，删，奇行输出

sed -e "1d;p;N;d" file

::偶行输出，每一行直接删掉

)





5、

简单替换，不重复了



6、

仍是简单替换，不说了，只是注意转义



7、

这个有点难度，目前还不知道怎么在正则里面表达汉字，同求解



8、

sed -e "N;s/\n//" file

::N读取下一行，替换掉换行符



9、

sed -e "1{x;d};${x;p;x;b};/^5/{x;G;s/\n//;x;b};x" file

::暂时用5代表“我”，用中文的总是不正常，原因未明，期待答案



10

sed -e "s/^\s*//;s/\s*$//;" file

::\s代表空字符，一般是空格和TAB



11、12略

sed -n "/^.\{10\}/p" a.txt

sed '$!N;s/\n/ /' a.txt

Originally posted by 26933062 at 2008-5-31 11:45 PM:
显示纯英文行
sed -n "/^*$/p" a.txt

C:\test>type a.txt
abc
a啊
啊
啊a
我
ccc

C:\test>sed -n "/^*$/p" a.txt

C:\test>

Originally posted by zhengruijian at 2008-5-31 11:34 PM:
太好了.
我来说一个.
前提是对一个文件夹下的N个TXT文件进行处理,对每一个TXT里的每一行前面都加入000

sed "s/^/000/" *.txt

sed -n "1!G;h;$p" a.txt

sed "n;n;n;n;G;" a.txt

Originally posted by HAT at 2008-6-1 02:04 AM:

这个正则表达式写的很正确，在Unix下可以得到正确结果，但是在cmd下调用sed.exe却得不到任何结果。

Originally posted by sylovanas at 2008-6-2 09:34 PM:


C:\1>type 1.1.txt
abc
a啊
啊
啊a
我

C:\1>sed -n "/^.*$/p" 1.1.txt
abc
a啊
ccc

可以用

sed -n "/^*$/p" a.txt

Originally posted by sylovanas at 2008-6-2 09:34 PM:


C:\1>type 1.1.txt
abc
a啊
啊
啊a
我

C:\1>sed -n "/^.*$/p" 1.1.txt
abc
a啊
ccc

可以用

Originally posted by plp626 at 2008-6-3 07:11 AM:

多了一个点：

sed -n "/^*$/p" a.txt

Originally posted by HAT at 2008-6-3 09:33 AM:

不加那个点也不对，见26楼描述。

sed -e "/^*$/!d" file

sed -r -e "/^+$/!d" file

F:\tptst>np file



F:\tptst>type file

adfadfadfasdfasdfadf

4trqefddafKKLFDK:D

JJDJEXDSLDD





da13431431

o9590415143

ABCDE

BDEFG

BBDkkDDlda



;adfazxx13!#$#

F:\tptst>sed -e "/^*$/!d" file

adfadfadfasdfasdfadf

JJDJEXDSLDD

ABCDE

BDEFG

BBDkkDDlda



F:\tptst>sed -r -e "/^+$/!d" file

adfadfadfasdfasdfadf

JJDJEXDSLDD

ABCDE

BDEFG

BBDkkDDlda

Originally posted by pooronce at 2008-6-3 10:09 AM:


不加那个点的话，会匹配到空行，正确的应该是重复一次，这样：

sed -e "/^*$/!d" file

abc
a啊
啊
啊a
我

Originally posted by HAT at 2008-6-3 11:45 AM:

关键不是空行的问题，而是能否匹配纯英文行的问题。
你的代码能正确匹配吗？

F:\tptst>type file

adfadfadfasdfasdfadf

4trqefddafKKLFDK:D

JJDJEXDSLDD



我你他

a你我

他abcd你我

人民的常b



da13431431

o9590415143

ABCDE

BDEFG

BBDkkDDlda



;adfazxx13!#$#

F:\tptst>sed -e "/^*$/!d" file

adfadfadfasdfasdfadf

JJDJEXDSLDD

ABCDE

BDEFG

BBDkkDDlda



F:\tptst>sed -r -e "/^+$/!d" file

adfadfadfasdfasdfadf

JJDJEXDSLDD

ABCDE

BDEFG

BBDkkDDlda

F:\tptst>sed --version

GNU sed version 4.1.5

Copyright (C) 2003 Free Software Foundation, Inc.

This is free software; see the source for copying conditions.  There is NO

warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE,

to the extent permitted by law.

C:\Test>sed --version
GNU sed version 3.02

Copyright (C) 1998 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE,
to the extent permitted by law.

Originally posted by HAT at 2008-6-1 01:34:
13，如果某行多于10个字符，只显示每一行的前10个字符

我知道这样可以查找出多余10个字符的行，但是不知道怎样只显示前10个字符。

sed -n &qu ... 

:查找出多余10个字符的行 ,但只显示后面的10个字符

sed -n "/^.\{10\}/s/\(^.\{10\}\)\(.\{10\}\).*/\2/p" a.txt



:查找出多余10个字符的行，但只显示前10个字符

sed -n "/^.\{10\}/s/\(^.\{10\}\).*/\1/p" a.txt

 sed "10q" a.txt 

 sed -n "1,10p" a.txt 

 sed "1,10!d" a.txt 

 sed "11,$d" a.txt 

 sed "$!d" a.txt 

 sed -n "$p" a.txt 

 sed "s/；/\n/g;s/，/\n/g" a.txt 

 sed "20,25!d" a.txt 

 sed -n "20,25p" a.txt 

 sed -n "p;n" a.txt 

 sed -n "n;p" a.txt 

 sed "s/爱你/【爱你】/g;s/恨你/【恨你】/g" a.txt 

 sed "s// /g" a.txt 

 sed -n "/^*$/p" a.txt 

 sed -n "/^*$/p" a.txt 

 sed -n "/\|/p" a.txt 

 sed "N;s/\n/ /" a.txt 

 sed ":a;$!N;s/\n我/我/;ta;P;D" a.txt 

 sed "s/^ *//;s/ *$//" a.txt 

 sed "s/^/echo/" a.txt 

 sed "s/。$//" a.txt 

 sed "s/\(^.\{10\}\).*/\1/" a.txt 

 sed ":a;//s/^.\{,19\}$/ &/;ta" a.txt 

 sed "1!G;h;$!d" a.txt 

 sed "/\n/!G;s/\(.\)\(.*\n\)/&\2\1/;//D;s/.//" a.txt 

 sed "n;n;n;n;G" a.txt 

 sed -n "G;s/\n/&&/;/^\(.*\n\).*\n\1/{P;d};s/\n//;h" a.txt 

@echo off

for /f "delims=" %%a in (a.txt) do (

	if /i defined %%a (echo %%a) else set %%a=a

)

 sed -e "s/^.\{,10\}$/&          /;s/\(^.\{11\}\) *$/\1/;s`\(^.....\)\(.....\)\(.*\)`\1/*\2*/\3`" a.txt 

@echo off

for /f "delims=" %%a in ('dir /a-d/b *.txt') do (

	sed "s/^/000/w tmp_.txt" "%%a"

	move tmp_.txt "%%a"

)

@echo off

for /f %%i in ('sed -n "$=" a.txt') do set number=%%i

:loop

set /a num+=1

set /a n=(%num%-1)*100+1

set /a m=100*%num%

sed "%n%,%m%w a_%num%.txt" a.txt

if %m% LSS %number% goto :loop

@echo off

for /f %%i in ('sed -n "$=" a.txt') do set num=%%i

:loop

set /a n+=1,count+=1

sed -n %n%p a.txt

set /a m=10+%n%

sed -n %m%p a.txt

if %count% EQU 10 set /a n+=10 & set count=0

if %n% LSS %num% goto :loop

pause

@echo off

for /f %%i in ('sed -n "$=" a.txt') do set num=%%i

for /f %%i in ('sed -n "$=" b.txt') do set num2=%%i

if %num% LSS %num2% set num=%num2%

:loop

set /a n+=1

sed -n %n%p b.txt>>ba_tmp.txt

sed -n %n%p a.txt>>ba_tmp.txt

if %n% LSS %num% goto :loop

move ba_tmp.txt b.txt

split -100 a.txt

Originally posted by ngd at 2008-6-19 01:59 PM:
9，如果当前行是以"我"字开头，怎么将它合并到上一行？

sed ":a;$!N;s/\n我/我/;ta;P;D" a.txt

aaa
bbb
我ccc
ddd