CODEWITHSUNDEEP
bashawk
Another.Chemist
Another.Chemist

Reputation: 2559

AWK|BASH, use double FS and ternary operator

Is it possible?

I was wondering how to do:

  1. Count fields differentiated by comma.
  2. Only the obtained first field of the previous step, count words differentiated by space.
  3. If there is more than 2 words, print NF, otherwise $0.

Input

cellular biol immunogenet, rosario
escuela estadist, medellin
medellin

Expected output

rosario
escuela estadist, medellin
medellin

Upvotes: 1

Views: 119

Answers (2)

bkmoney
bkmoney

Reputation: 1256

You can use this awk command

awk -F, '{if (split($1, a, " ") > 2) print $NF; else print}' file

This will output

 rosario
escuela estadist, medellin
medellin

If you want to get rid of the space before $NF, use this awk command

awk -F, '{if (split($1, a, " ") > 2) print substr($NF, 2); else print}' file

This will output

rosario
escuela estadist, medellin
medellin

Explanation

We use the -F, to set the field separator to a comma.

For every line, we split the first field into the array a, using a single space as the separator. We check if the return value is greater than 2 (the split function returns the number of elements), and then we print the substr($NF, 2), which removes the space at the beginning of $NF. Otherwise, we just print the whole line using print.

Upvotes: 2

geirha
geirha

Reputation: 6181

awk -F, 'split($1,a,/ /) > 2 {$0=$NF} 1' input.txt

split($1,a,/ /) splits $1 on the regex / / and stores them into the array a. It returns the resulting length of the array.

So if $1 is split into more than two words, then change the line $0 to $NF.

The lone 1 is a shorthand that makes sure it prints each line.

Upvotes: 1

Related Questions