UNIX for Beginners Questions & Answers

Again sorry,

Not the Headers but the entire first 9 columns is there a way to make the code work but exempt the first 9 columns (i.e i want to keep the information after the : on the the the first 9 columns only) = Remove a selection of columns

Thanks

Yes. Look at the code and how it loops across fields 10 until end-of-line. Now imagine what were needed to make it loop across fields 1 to 9...

Code:

awk 'NR > 1 {for (i=1; i<=9; i++) sub (/:.*$/, _, $i)} 1' OFS="\t" file

right?

I am trying to understand the logic behind this bit. I would be grateful if you can break it down ?

Code:

(/:.*$/, _, $i)

$ for the columns ? and the columns are specified earlier in I?
, _, substitute ? usually in sed i leave it empty for substitution when i want to remove it.
I always find awk is complicated, any good place to read more about it as I have more use for it for the same data to find in-common and unique values between range of columns (patients versus control)?

Thanks again
You are a life saver. I have been experimenting with this for 3 days :/

You got it, congrats! Here the sub logics (please note that $ means different things in regexes and in awk field designators):

Code:

(/:.*$/,        # regex greedily matching ":" then ANY char (multiple) until EOL designated by $
 _,             # _ is an undefined thus empty variable, equivalent to ""
 $i)            # in the loop, $i results in fields $1 .. $9