Shell Programming and Scripting

Hey, I'm writing this bash script that will test print me many copies of the same program but with different combos of 4 variables being between 1 and 100. Here's the code:

#! /bin/bash

x=0

for ((a=1; a < 101; a++))
do
for ((b=1; b < 101; b++))
do
for ((c=1; c < 101; c++))
do
for ((d=1; d < 101; d++))
do

((x=$x+1))


<echoing my program copies here into different txt files>


done
done
done
done


I'm no math major, but I noticed that assuming this script took a second per combo/file creation (and I'm estimating it does), then this code would take about 100^100^100^100 seconds...a (very)^100 big number haha. This means it'd take days or more to process. So, I thought I'd cut down on it by making sure that both c and d are always equal or greater than both a and b (the program segfaults if a or b is bigger than c and d anyways, so it'd be a good idea to try filtering these out of the copies and cut down on process time at the same time). Anyone know how to do such a thing? And how much do you guys think it will cut down on time?

Traditionally you do boundary checking.

So if something is supposed to work on values from 1 to 100, you test something like

0,1,2,50,99,100,101

I don't understand what you mean...could you explain more please?

nevermind...I just added:

if [ $a -le $c ] && [ $a -le $d ] && [ $b -le $c ] && [ $b -le $d ]; then

below the loops

and fi before the loops end

seems to work

Say you have some code that adds to numbers, psuedo code below:

Code:

print a + b

and ask for somebody to test it, do they have to test every combination of numbers from minus infinity to plus infinity in order to confirm your program works?

No, you specifiy the resolution of your numbers representation and the maximum and minimum. Then the tester only needs to test a much smaller subset at either end of the range (the boundary conditions) and in the middle (the likely conditions).