If/else help

[Riemann28]

Hey folks I just started learning C++ a week ago and have been stuck on a project for the past 2 days now. I am building a limited purpose calculator which finds the value of one of five operations. Visual studio doesn't underline any errors in my program but every time I try to run it I get an error message. I believe it has something to do with the if/else but Im not sure. This isnt a project from my book so it has been no help. If anyone has the time and patience I would greatly appreciate you taking a look at my program and telling me what I am doing wrong. Thanks.

#include <iostream>
using namespace std;

int main()
{
	int a;
	int b;
	
	int sum = a + b;
	int difference = a - b;
	int product = a * b;
	int quotient = a / b;
	int square = a * a;

	cout << "Enter a number " << endl;
	cin >> a;

	cout << "Enter another number " << endl;
	cin >> b;

	cout << "Enter an operation." << endl;
	cin >> sum, difference, product, quotient, square;
	
	if (sum)
	{
		cout << "The sum is " << endl;
		cout << sum << endl;
	}
	else if (difference)
	{
		cout << "The difference is " << endl;
		cout << difference << endl;
	}
	else if (product)
	{
		cout << "The product is " << endl;
		cout << product << endl;
	}
	else if (quotient)
	{
		cout << "The quotient is " << endl;
		cout << quotient << endl;
	}
	else 
	{	
		cout << "The square is " << endl;
		cout << square << endl;
	}
	
system("pause");
	return 0;
}

[AbstractionAnon]

What are the values of a and b when lines 9-13 are executed?
Hint: You can't do a calculation on two numbers before you've assigned values to them.

What is line 22 supposed to be doing? You're asking the user to enter 5 integers.

[jmadsen]

I ran your program and got a floating point exception. The floating point exception is due to

int quotient = a / b;

Since you don't initialize a or b to anything, this is initially 0 / 0 which is NaN (not a number). If you simply cast a or b to a floating point number in this equation:

int quotient = static_cast<double>(a) / b;

or

int quotient = a / static_cast<double>(b);

The floating point exception goes away. This is because if either a or b is a floating point, when the operation is done, it promotes the other to a floating point.

If you simply make b initialized to something other than 0, the floating point exception also goes away. If you divide a number by zero it is either NaN of +/- infinity, which is not an integer.

Keeping a and b both at zero and simply changing quotient to a double won't fix it (i.e. double quotient = a/b; because when a / b is initially calculated both a and b are int's so the solution is of type int and only after it is calculated is the solution (assigned to the variable quotient) converted to a floating point number. Look through type conversion in a C++ book, operations are best done with float or double because 5/2 = 2 with ints.


Also, you are correct that your if/else has problems and so does you cin and you also need to recompute sum/difference/etc. after you read a and b in. As for your if/else, if(sum) (and all the other else if's) convert the variable in the parentheses to a boolean (meaning it can only have the value of 0 or 1). For boolean's 0 = false, 1 = true. If another type is converted to a boolean, 0 = false and anything other than zero evaluates to true.

Keeping things as simple as possible, what you actually want to do is make a string called operation. Read in operation, and use that for your if/else:

#include <string>
...
double a = 0., b = 1.;
string operation = "+";
cout << "Enter a number" << endl;
cin >> a;
cout << " enter another number" << endl;
cin >> b;
cout << "enter an operation (+, -, *, /, sq)" << endl;
cin >> operation;

if(operation == "+")
  cout << "The sum is " << (a+b) << endl;
else if(operation == "-")
  cout << "The difference is " << (a-b) << endl;
...
else if(operation == "sq")
  cout << "The square is " << (a*a) << endl;
else
  cout << "Not a valid operation. Valid operations are +, -, *, /, sq" << endl;

By the way, system("pause") works on Windows but this doesn't work on most UNIX systems. You may not care about cross-platform compatibility right now but it's generally bad practice to do so.

[Riemann28]

Yes! thank you based anon! It works kind of!

the values of a and b are what ever you input so I moved them below 15-19 and I fixed line 22 as well. Now it runs but it only calculates the sum of everything even if I enter difference or product. Hopefully I can tinker around to fix that but at least it actually runs now!

thank you thank you thank you!

[jmadsen]

Actually, you can do calculations on two numbers before you've assigned values to them because they are assigned to space a memory regardless of whether you set what value to be placed in that memory location. The newer C++ standards (starting with C++11, I believe, maybe C++0x) guarantee that uninitialized variables are set to their equivalent of zero (0 for int/unsigned/float/double/etc.) or empty for char/string. Older standards will usually have a random value from what was previously stored in its place in memory previously.

[Riemann28]

jmadsen, I much appreciate you looking over my program. Embarrassingly I forgot that zero is an integer too...I noticed that when anon helped me fix it a minute ago, running the program performs every operation except for division so I am trying to fix that right now. I am certainly going to have to tinker with this program quite a bit more to make it do what I want.

And thanks for posting a snippet of how you would code it, I don't want to copy it but it really simplifies my program quite elegantly so I am definitely going to study it and maybe borrow some tricks. Also thanks for the heads up on Unix, I'm obviously a windows user although I plan to one day install a linux distro (maybe after I can create programs on my own!). Thanks you so much for the guidance Ill post a status update after I sort some things out.