sed challenging... 1

[hokky]

Hi guys,

I just need to ask how to change the second column from dot delimiter using sed.

ie, data :

my home.1000.change

becomes

my home.90.change

Thanks guys

 

[feherke]

Hi

[gray]# just to contain the required counting to 2[/gray]

sed 's/[^.]\+\./90./2' /input/file

[gray]# or in this case there would be the same[/gray]

sed 's/\.[^.]\+/.90/' /input/file

[gray]# or[/gray]

sed 's/\.[^.]\+\./.90./' /input/file

Feherke.

http://rootshell.be/~feherke/

 

[PHV]

Another way with awk:
awk 'BEGIN{FS=OFS="."}{$2=90;print}' /path/to/input

Yet another (legacy) sed method:
sed 's!\([^.]*\)\.[^.]*\.!\1.90.!' /path/to/input

Hope This Helps, PH.
FAQ219-2884
FAQ181-2886

 

[hokky]

Thanks feherke,

1 more question :
what about if I want to delete the line that the second column contain 1000, Thanks.

 

[feherke]

Hi

Hmm... thinking again, it can be even shorter :

sed 's/[^.]\+/90/2' /input/file

Feherke.

http://rootshell.be/~feherke/

 

[feherke]

Hi

sed '/^[^.]\+\.1000\./d' /input/file

[gray]# or in the given example may be just[/gray]

sed '/\.1000\./d' /input/file

Feherke.

http://rootshell.be/~feherke/

 

[hokky]

Sorry feherke,

sometimes they contain in 7th column as well, I tried this doesn't work

sed '/^^^^^^[^.]\+\.1000\./d' /input/file

Cheers,

 

[feherke]

Hi

Let us try to learn from PHV :

sed '/^[red][^.][/red][^.][red]*[/red]\.1000\./d' /input/file

Or just modify PHV's more portable [tt]awk[/tt] code :

awk 'BEGIN{FS=OFS="."}[red]$2!=1000{[/red]print}' /path/to/input

Feherke.

http://rootshell.be/~feherke/