CODEWITHSUNDEEP
bashshellawk
Fattaneh Talebi
Fattaneh Talebi

Reputation: 767

awk average in decimal

I want to write an script which gets the average of two column of a file where the first column is equal, I did that with awk. it works, but the in the output there was exponential format data
121323e+2
then I changed the print to printf and I added %f for the float

awk '
    NR>1{
        arr[$1]   += $2
        arr2[$1] += $3
        count[$1] += 1
    }
    END{
        for (a in arr) {
             printf "%4.3f",  a " " arr[a] / count[a] " " arr2[a] / count[a]
        }
    }
' t1.txt  > t2.txt

I have problem with this part:

printf "%4.3f",  a " " arr[a] / count[a] " " arr2[a] / count[a]

the output:

 0001204.0001125.0001118.0001053.0001046.0001039.000901.000822.000815.000808.000750

which I wanted

 0001 204.000 1125.000 
 1118.000 1053.000 1046.000 
 1039.000 901.000 822.000 
 815.000 808.000 750

Upvotes: 0

Views: 466

Answers (1)

Tom Fenech
Tom Fenech

Reputation: 74605

Assuming that the rest of the code does what you want, you should change the printf to something like this:

printf "%4.3f %4.3f %4.3f\n",  a, arr[a] / count[a], arr2[a] / count[a]

Here I've included one format specifier for each value to be inserted and added a newline \n to the end. Each value is then separated by commas.

Upvotes: 2

Related Questions