Thread: How can i see "divide error" ?

hello to all,
I am new to c programming and i have encountered some very frustrating problem.
so i was suppose to write a program that gives a divide error but output is not showing anything like it.
I am using orwell devc++ 5.4.2 and all the default built tools(gcc compiler etc.) with default settings.
so, what should i do to see this error?

Code:

#include<stdio.h>
#include<conio.h>

int main()
{
    int a,b,c;
    float x;

    a=250;
    b=85;
    c=25;
    x=a/(b-c);
    printf("x = %f\n",x);

    a=300;
    b=70;
    c=70;
    x=a/(b-c);
    printf("x = %f\n",x);

    getch();
    return 0;

}

thanks for reading it!

any suggestions are welcome!
Update-
output by orwell devc++

output by borlands turbo c++ ver3

As you can see borland's tc++ compiler does generate this error. But i still don't know why gcc doesn't do this.

Run fails at my pc, as I expected to do, because of the division by zero. What is your problem exactly? You want to track down this kind of division? If so, just use an if statement and check if denominator is zero. If yes, print out division error, else do the division.

Welcome to the forum.

if you want to make a float division you need to cast at least one of participating ints to float

Code:

x=a/(b - (float)c);

@std10093
thank for reply!
nah! Its just an exercise in 1st chapter of the book that i am reading and the solution code is suppose to generate this error, that's it.
by the way shouldn't be a compiler like gcc would do this.

What Vart said is true, but I did not mentioned it, because your code (it happened now, but maybe not next time you want to do floating division) works as integer division, because the number do not have any decimal points.

"-Wdiv-by-zero
Warn about compile-time integer division by zero. This is default. To inhibit the warning messages, use -Wno-div-by-zero. Floating point division by zero is not warned about, as it can be a legitimate way of obtaining infinities and NaNs."
Source: http://gcc.gnu.org/

Floating point divide errors don't generally "throw" an exception unless you change the default behavior.
Since you're just looking for a "divide error", by which I assume you mean a runtine "arithmetic exception" then you could use integers instead of floats.
Or do you have your heart set on doing it with floats?

EDIT: Oh ... you are using integers.

Firstly, the -Wdiv-by-zero warning's effect is only at compile time, not runtime. Secondly, it will only warn for divisions by zero that it can spot.

So this gives a "division by zero" warning from the compiler.

Code:

int main(void)
{
    int a = 1 / 0;
    return 0;
}

This doesn't give a compiler warning, but will cause an arithmetic exception at runtime (assuming no optimizations).

Code:

int main(void)
{
    int a, b = 1, c = 0;
    a = b / c;
    return 0;
}