delete column

[ovince]

hi All,

I would like to delete 2 columns from 60 columns data. How to do that except to write

awk '{print $1, $2 .... $60}' file.dat > newFile.dat


leaving out columns that I dont want in a newFile.dat?


thanks
oliver

 

[feherke]

Hi

If you do not care about the separators which will remain in place :

awk '{$[green][i]40[/i][/green]=$[green][i]50[/i][/green]=""}1' file.dat > newFile.dat

Or a less nice code with nicer output, although all field separators will be set to [tt]OFS[/tt] :

awk '{for(i=1;i<=NF;i++)if(i!=[green][i]40[/i][/green]&&i!=[green][i]50[/i][/green])printf$i OFS;print""}' file.dat > newFile.dat

Or we just can modify abit the first code to replace multiple field separators with one :

awk '{$[green][i]40[/i][/green]=$[green][i]50[/i][/green]="";gsub(FS "+",FS)}1' file.dat > newFile.dat

In my examples I removed columns 40 and 50.

Feherke.

http://rootshell.be/~feherke/

 

[Ygor]

Unless you have to use awk, perhaps consider using cut, e.g...

cut -f-39,41-49,51- file.dat > newFile.dat

 

[ovince]

thanks


I have a csv file and I would like to keep like that. This try does not work if I want to remuve 1 column (i.e it leaves ',' in front of lines)

awk 'BEGIN{FS=",";OFS=","}{$1=$6="";gsub(FS "+",FS)}1' file.dat > newFile.dat



Feherke's suggestion is OK

 

[ovince]

sory...Ygor's suggestion with cut works and

awk 'BEGIN{FS=",";OFS=","}{$1=$6="";gsub(FS "+",FS)}1' file.dat > newFile.dat

leave ',' in front of each line

 

[feherke]

Hi

Right. I just wondered how could it be possible.

I completely ignored the situation of removing the first or the last field. For those the shell itself is enough.

Feherke.

http://rootshell.be/~feherke/