Need some help

I am learning C++, working through the books recommended by this website. I'm trying to code Exercise 6-3 of Practical C++ Programming. Here's the problem:

Given an amount less than $1.00, compute the number of quarters, dimes, nickels, and pennies needed.

Here's my code:

Code:

#include <iostream>

int main(){
	float money; // amount of money
	int quarters, dimes, nickels, pennies;  // Stores the amount of each coin needed.

	while (true){
		std::cout<< "Enter a non-negative amount of money " // Prompt user for money
			<< "less than $1.00, or -1 to stop: ";	    // amount
		std::cin >> money;

		if (money == -1) // Loop until user enters -1
			break;

		if (money < -1 || money >= 1.0) // Start loop over if money
			continue;		// amount is invalid.

		std::cout << "$" << money << " in change is ";

		quarters = (money / 0.25);  // Stores the amount of quarters needed, then subtracts
		money -= (quarters * 0.25); // that amount in cents from the money total.
		
		dimes = (money / 0.1);  // Stores the amount of dimes needed, then subtracts
		money -= (dimes * 0.1); // that amount in cents from the money total.

		nickels = (money / 0.05);  // Stores the amount of nickels needed, then subtracts
		money -= (nickels * 0.05); // that amount in cents from the money total.

		pennies = (money * 100); // Money is in cents, so it is multiplied by 100 to convert
			               	 // it to the amount of pennies needed.

		std::cout << quarters << " quarters, " << dimes << " dimes, "
			<< nickels << " nickels, " << " and " << pennies << " pennies.\n";
	}

	return 0;
}

For some reason, it's not calculating the right amount. For example, if I run the program and put in 0.35, when it gets to this part

Code:

dimes = (money / 0.1)

, I know money is equal to 0.1 at that point, but dimes is not being set equal to 1. I put in

Code:

std::cout << money;

just to make sure that money equals 0.1 at that point, which it does. I also tried

Code:

std::cout << (money / 0.1)

to see that money, when it is 0.1, divided by 0.1 does indeed equal 1, which it does. But it won't set dimes equal to 1. If I run the program at 0.36, it calculates everything right. What am I doing wrong?

It has to do with the fact that computers cant handle floating point ("real numbers") perfectly. To see this, do:

Code:

#include <iostream>

int main(void)
{
    float foo = 0.1;

    if ( foo == 0.1)
        std::cout << "you would think its equal" << std::endl;
    else
        std::cout << "but its not equal" << std::endl;
}

Since "money" can be considered a discrete quantity, you could just work with pennies, as "ints", and after you get the money as a float just multiply it by 100, so you only work with pennies (discrete quantities). Later just divide by 100. This will ensure correct comparisons, since your working with ints.

Ok, I tried working with pennies. I did this after getting money as a float from user:

Code:

pennies = (money * 100); // stores money as pennies

It stores the wrong amount sometimes. If I put in 0.35, it stores 34 in pennies. Some numbers work, such as 0.36 or 0.37 (0.38 stores as 37).

So if you do

Code:

float f = 0.35;
int i = 100*f;
cout << i;

it prints "34"? Or even by doing "cin >> f" instead of hardcoding it, you get 34?

EDIT: Compilers are probably a factor too, what are you using?

Yeah. Using exactly this code:

Code:

#include <iostream>

int main(){
	float f;
	int i;
	f = 0.35;
	i = f * 100;
	std::cout << i << std::endl;

	return 0;
}

I got this:
34
Press any key to continue

I'm using Microsoft Visual Studio .NET 2003.

Try forcing an int by casting i.e.:

Code:

i = (int)100*f;

I get 34 when its interpreted as a float, i.e.:

Code:

i = 100.0*f;

and forcing it to be an int I get 35.

I tried that, and it still didn't work. However, I was getting the warning "truncation from double to float" on this line

Code:

	f = 0.35;

I tried changing it to a double, and it works now. Given that the double f equals 0.35, (f * 100) stored into an integer gives me 35.

if you do "f = (float) 0.35;", and leave f as a float, do you later get integer 35?

No, it's still 34. I tried this though

Code:

#include <iostream>

int main(){
	float f;
	int i;
	f = 0.35;
	f *= 100;
	i = f;
	std::cout << i << std::endl;

	return 0;
}

This prints out 35.

Prefer double over float.
To create a float, you must do x.yf. Notice the "f" at the end. That will make the compiler treat the number as a float rather than a double.
Nevertheless, I would still urge you to use integer math.