Sum and count in for loop

Question

I have five files each with a number in the first line.

I am attempting to write a bash script incorporating a for-loop that can calculate both the number (count) of the files and the sum of the numbers within in the files.

This is what I have attempted so far:

for file in $* 
do
        $[head -1 $file] 
        echo $(head -1) 
done

I am unsure how to incorporate a sum and count element as yet.

Answer 1

I guess what you are trying to do can be achieved by this:

#!/bin/bash
sum=0
count=0
for file in "$@"
do
        number=$(head -1 "$file")
        count=$((count + 1))
        sum=$((sum + number))
done
echo "The sum of the $count file(s) is: $sum"

You just add variables to store the sum and the count of the files you are working with. You then increase the count each time you run through the loop. Also we add the number at the beginning line of the file to our sum-variable.

Answer 2

You could also do this with awk, avoiding the need to for head for each file:

gawk '{count += 1; sum += $1; nextfile} 
      END {printf "count: %d\t sum: %d\n", count, sum}' *

The first rule increments the count and sum, and then jumps to the next file, thus ignoring all but the first line of each file. In the END, we print out the numbers.

nextfile is a GNU thing, it might not work in other versions of awk. Another alternative would be to explicitly work only on the first line:

awk 'FNR == 1 {count += 1; sum += $1} 
      END {printf "count: %d\t sum: %d\n", count, sum}' *

This would however read the files in full, even if it doesn't do anything with the rest of the lines.