如何刪除一個文本中相同行中的其中一行？

free_2006

一個文本中相同的行有幾行，只想刪除其中一行如何處理？？
例如：文本test.txt中有如下幾行：
aaaaaaa
bbbbbbb
aaaaaaa
ddddddd
fffffffffffff
aaaaaaa
ccccccc


想刪除最后一個aaaaaaa所在行，該如何處理呢？多謝各位了�。�！

Herowinter

回復(fù) 1# free_2006
方法好像不是很好。

1. awk 'NR==FNR{a[$0]++;b[$0]++;next} {if((a[$0]==1)||(--b[$0]))print}' i i
   
2. aaaaaaa
   
3. bbbbbbb
   
4. aaaaaaa
   
5. ddddddd
   
6. fffffffffffff
   
7. ccccccc

復(fù)制代碼

jason680

回復(fù) 1# free_2006

1. using array in awk for file size is samller than memory
$ awk '{if($0~/aaaaaaa/)N=NR;a[++t]=$0}END{for(n=1;n<=t;n++)if(n!=N)print a[n]}' FILE
aaaaaaa
bbbbbbb
aaaaaaa
ddddddd
fffffffffffff
ccccccc

2. read file twice for big size file
$ awk 'NR==FNR{if($0~/aaaaaaa/)N=NR;next}{if(FNR!=N)print}' FILE FILE
aaaaaaa
bbbbbbb
aaaaaaa
ddddddd
fffffffffffff
ccccccc

   

tdounnyy

有N個重復(fù)行，保留N-1行，只刪除第N行？

free_2006

回復(fù) 3# jason680

awk命令如何在文件中直接生效呢？
   

jason680

回復(fù) 5# free_2006

Answer 1: awk
solution: make the new file and rename it

or you want to using this one, but I'm not sure it's safety.   
$ awk '{if($0~/aaaaaaa/)N=NR;a[++t]=$0}END{for(n=1;n<=t;n++)if(N!=n)print a[n] > "FILE"}' FILE


Answer 2: sed and perl can repleace FILE directly
but I don't know how to do

yestreenstars

回復(fù) 6# jason680

How about this?

for sed:

1. sed -r ':1;N;$!b1;s/(.*)\naaaaaaa/\1/'

復(fù)制代碼

but it is not good for big size file.

jason680

回復(fù) 7# yestreenstars

thank you!!!

Note: It's hard to me to read this sed code
   

Herowinter

回復(fù) 7# yestreenstars
Could you write a sed to delete the 1st row of aaaaaaa?
Thanks.

   

Herowinter

回復(fù) 8# jason680
Read all lines in file into pattern space, then
replace (.*\n)aaaaaaa\n  with  \1.