CODEWITHSUNDEEP
awkcount
Cris
Cris

Reputation: 65

Count a number of records in a field using awk

I have a big file like this one 

C1,C2,C3
C1,C2
C5
C3,C5

I expected one output like this

C1,C2,C3   3
C1,C2      2
C5         1
C3,C5      2

I would like to make this using shell. Could you help me guys, please? Thankyou

Upvotes: 0

Views: 826

Answers (2)

James Brown
James Brown

Reputation: 37404

Another in awk:

$ awk '{
    m=(m<(n=length($0))?n:m)                  # get the max record length
    a[NR]=$0 }                                # hash to a
END {
    for(i=1;i<=NR;i++)                        # iterate and (below) output nicely
        printf "%s%"(m-length(a[i])+4)"s\n",a[i],gsub(/,/,"&",a[i])+1 }
' file
C1,C2,C3   3
C1,C2      2
C5         1
C3,C5      2

IF you want to change the distance between fields and the length, toy with that +4 in the printf.

Upvotes: 1

sjsam
sjsam

Reputation: 21965

Something like 

awk 'BEGIN{FS=","}{printf "%-20s\t%d\n",$0,NF;}' file

should give

C1,C2,C3                3
C1,C2                   2
C5                      1
C3,C5                   2

Note You need to adjust the width logically considering the maximum length of your lines

Upvotes: 2

Related Questions