Shell Programming and Scripting

nawk seems to handle it fine -

awk 'BEGIN{Fraction=1/3;printf("%0.2f\n",Fraction)}'
0.33
awk 'BEGIN{Fraction=2/3;printf("%0.2f\n",Fraction)}'
0.67

awk,

The function works as well when displaying a fraction. However I would like to return the percentage. If possible I would like to use only one funciton to add, subtract, percentage, etc.

calc()
{
awk 'BEGIN {print '"$*"'; exit}'
}

calc "1188 / 1454.07"

Would return: 0.817017
But I would like to get: 81.70

Not only 81.70 but 81.70%

awk 'BEGIN{Fraction=1188 / 1454.07;printf("%0.2f%%\n",Fraction*100)}'
81.70%

awk,

When I try to substitue a vaiable I recieve a sysntax error. Any Ideas?

USED=1188
TOTAL=1454.07

awk 'BEGIN{Fraction=${USED} / ${TOTAL};printf("%0.2f%%\n",Fraction*100)}'

Syntax Error The source line is 1.
The error context is
>>> BEGIN{Fraction=${ <<<
awk: 0602-502 The statement cannot be correctly parsed. The source line is 1.
Syntax Error The source line is 1.
awk: 0602-540 There is a missing } character.

2 solutions

export A=1188
export B=1454.07

awk 'BEGIN{Fraction=ENVIRON["A"] / ENVIRON["B"];
printf("%0.2f%%\n", Fraction)}'

awk 'BEGIN{Fraction='$A' / '$B';
printf("%0.2f%%\n", Fraction)}'
0.82%
0.82%

The first may have issues with some awks, but I prefer it myself.

The latter is more universal, but be sure and not put any spaces before or after the $A and $B. It does not have the have the variables exported.

To me the first is easier to debug, which is why I prefer it.

Neither have error coding - left to you - ie. non-numerics or dividing by zero error.

Works great thanks awk.

A ksh co-process is a very fast solution because a single process can handle all of your calculations. In contrast, the above solution need one one awk process per calculation.

Code:

#! /usr/bin/ksh

bc |&
print -p scale=5

function calc
{
        typeset result
        print -p "$@"
        read -p result
        echo "$result"
}



calc 1.5 / 3
A=1.5
B=3.0

calc $A / $B


$ ./testcalc
.50000
.50000
$