CODEWITHSUNDEEP
bashunixawk
Rajat
Rajat

Reputation: 5803

Print multiple fields in AWK but split one of them based on a different delimiter

Is there an easy way to split a field, get the first element from the said split field AND also print other fields in same awk?

I want to modify below such that is works for $2 not just when it's MCU:123 but also when it's DISNEY:654 such that $2 returns MCU or DISNEY depending on the string. Obviously substr was a bad choice there.

 awk 'BEGIN{FS=OFS="\t"}{print $1,substr($2,1,3)}'

Sample of two fields tab seperated

1  mcu:1234   
2  disney:234

Expected result tab seperated

1  mcu    
2  disney

Upvotes: 0

Views: 718

Answers (3)

Ed Morton
Ed Morton

Reputation: 203189

$ cut -d: -f1 file
1       mcu
2       disney

or if the real file is more convoluted than the example in your question then maybe this will do what you really need:

$ awk -F'[\t:]' -v OFS='\t' '{print $1, $2}' file
1       mcu
2       disney

Upvotes: 2

Rajat
Rajat

Reputation: 5803

I was surprisingly closer than I gave my myself credit so throwing this in there as an alternative.

awk 'BEGIN{FS=OFS="\t"}{print $1,substr($2,1,index($2,":")-1)}'

Note that this fails if $2 isn't : delimited. For my use case, $2 would be : delimited by design and even if wasn't, I only need to perform this for values in $2 that do have : in them. To account for that, I could modify this to

awk 'BEGIN{FS=OFS="\t"}{if($2 ~ ":") print $1,substr($2,1,index($2,":")-1)}'

Other solutions are more scalable and possibly more performant.

Upvotes: 1

RavinderSingh13
RavinderSingh13

Reputation: 133428

1st solution: Could you please try following. Written and tested with shown samples.

awk 'BEGIN{FS=OFS="\t"} {sub(/:.*/,"",$2)} 1' Input_file

2nd solution: Using split function of awk.

awk 'BEGIN{FS=OFS="\t"} {split($2,arr,":");$2=arr[1]} 1' Input_file

Upvotes: 2

Related Questions