CODEWITHSUNDEEP
awk
user1059932
user1059932

Reputation: 19

How to do the sum in individual line one by one in linux?

How to do the sum in individual line in Linux?

I have one file :

Course Name: Math
Credit: 4
12345 1 4 5 1 1 1 1 1 5 10 1 2 2 20
34567 2 3 4 1 10 5 3 2 5 5 10 20 5
Course Name: English
Credit: 4
12345 1 4 5 1 1 1 1 1 5 10 1 20
34567 4 1 10 5 3 2 5 5 10 20 5

Its output will be come:

Course Name: Math
Credit: 4
12345     55
34567     75
Course Name: English
Credit: 4
12345     51
34567     70

I tried this code:

awk '{for (i=2; i<=NF; i++) {tot += $1}; print $1 "\t" tot; tot =0}' file > file2

The output is like this:

Course Name: 0
Credit: 4
12345     55
34567     75
Course Name: 0
Credit: 4
12345     51
34567     70

Actually I need to display a Course name too (Math and English). I am trying to fix it but I couldn't. Can you please help?

Upvotes: 1

Views: 170

Answers (3)

potong
potong

Reputation: 58401

This might work for you too!

sed '/:/{s/.*/echo "&"/;b};s/ /+/2g;s/\(\S*\) \(.*\)/echo "\1\t\$((\2))"/' file | sh

Upvotes: 0

glenn jackman
glenn jackman

Reputation: 246807

Just with the shell

while read line; do
  case $line in   
    Course*|Credit*) echo "$line" ;;
    *) set -- $line
       id=$1
       shift 1
       sum=$(IFS=+; echo "$*" | bc)
       printf "%s\t%d\n" $id $sum
       ;;
  esac
done  < filename

Upvotes: 0

William Pursell
William Pursell

Reputation: 212248

Try:

awk '/^[0-9]/{for (i=2; i<=NF; i++) {tot += $i}; print $1 "\t" tot; tot =0} !/^[0-9]/'

This will only sum lines that start with a digit, and simply print those that don't.

Upvotes: 2

Related Questions