I have FILE from which I can extract two numbers using grep. The numbers appear in the last column.

 $ grep number FILE 
 number1:    123
 number2:    456

I would like to assign the numbers to variables, e.g. $num1 and $num2, and do some arithmetic operations using the variables.

How can I do this using bash commands?

CodePudding user response：

Assumptions:

• we want to match on lines that start with the string number
• we will always find 2 matches for ^number from the input file
• not interested in storing values in an array

Sample data:

$ cat file.dat
number1: 123
not a number: abc
number: 456

We'll use awk to find the desired values and print all to a single line of output:

$ awk '/^number/ { printf "%s ",$2 }' file.dat
123 456

From here we can use read to load the variables:

$ read -r num1 num2 < <(awk '/^number/ { printf "%s ",$2 }' file.dat)

$ typeset -p num1 num2
declare -- num1="123"
declare -- num2="456"

$ echo ".${num1}.${num2}."
.123.456.

NOTE: periods added as visual delimiters

CodePudding user response：

Firstly, you need to extract the numbers from the file. Assuming that the file is always in the format stated, then you can use a while loop, combined with the the read command to read the numbers into a named variable, one row at a time.

You can then use the $(( )) operator to perform integer arithmetic to keep a running total of the incoming numbers.

For example:

#!/bin/bash

declare -i total=0           # -i declares an integer.

while read discard number; do # read returns false at EOF. discard is ignored.
  total=$((total number))     # Variables don't need '$' prefix in this case.
done < FILE                   # while loop passes STDIN to the 'read' command.

echo "Total is: ${total}"