[ask awk] remove certain row in a column

grail · 06-08-2010, 07:47 AM

Well if only to remove the last field, irrelevant of field number, then the first answer slightly changed works well:

Code:

awk 'NR < 5{$NF = ""}1' input_file

syg00 · 06-08-2010, 07:48 AM

Thanks colucix, saves me finding your previous explanation.
"1" is fine for me - I am not going to start using pi to 14 decimals ....

rkski · 06-08-2010, 08:18 AM

Code:

Thanks for your clue. I used awk '{for(i=NR;i<=4;i++)$2="";print $0}' input.dat > output dat and I got what I needed.

in the for statement you are calculating $2="" 5 times, then 4 times,... until NR >= 5 and then for statement is just tested&fails..which is a waste.

sysg00 soln is much better

2nd point: about the 1 trick. I wasn't sure about it till it was explained here. Basically easier to write 1 (or any true) than {print}.

David the H. · 06-08-2010, 09:36 AM

Yes, grail was more succinct, but colucix fleshed it out. Thanks to both of you for the explanations. I'd already discovered that it didn't have to be 1 exactly, so I suspected it was something like that, but it's nice to have it confirmed.

So it's basically just a shortcut for this:

Code:

awk 'NR<5 {$2=""} {print}' input.dat

And like syg00, I don't think I'll be playing around with it much either.