Thread: Supposed to return float but doesn't.

Code:

#include <iostream>

using namespace std;

float dividethesenumbers(int x, int y){
    return x/y;
}

int main()
{
    int x;
    int y;
    cout << "2 numbers to be divided. \n First number: ";
    cin >> x;
    cout << "Second number: ";
    cin >> y;
    cout << dividethesenumbers(x,y) << endl;
    system("pause");
    return 0;
}

I'm still very wet behind the ears with C++, please bear with.

I was going over function prototypes, to further remember the syntax, and came accross something I can't quite figure out.

Using the above code, I compiled and used 22 for x and 7 for y. (Pi) Instead of getting 3.14......, I got 3.

dividethesenumbers is supposed to return a float, right? I know I used integers for x and y as input but I thought the function would return the type I told it to.

I know the fix is to replace the int x, int y with float x, float y (as well as in main()) but I don't understand why my function won't return the type I ask.

Made sense and answered my query, thanks.

Originally Posted by Xipher

The problem is your doing integer math, and therfor returning a integer automaticly converted to a float (after the math)

inside your function, try something like
return float(x)/float(y);
instead. Now that isn't really the recomended way to do it, but it works, the actual recomended way would probably be

return static_cast<float>(x)/static_cast<float>(y);

try them both, they should give the same results, but the later is recomended (don't remeber why off the top of my head though)

static_cast and c-style cast are basically the same. The benifits of the static_cast is that they stick out and clearly indicate what is being done. Where the c-style looks like a function call(Especially how you did it) which is why c-style cast are recommended to be formatted like:

return (float)x/(float)y;

Since I know float is an instrinsic type, i would know yours was a cast, but had that been a user-defined type, I wouldn't know by looking at it if it was a cast or a function call.

Another benefit I often see listed is that you can do a search of your code for static_cast where you can't on the c-style cast.

Originally Posted by quzah

Well the location of the parenthesis might clue you in.

Quzah.

How so? I was referring to the float(x) style and not my suggested (float) x;

The only thing that clues me in in the former is the fact that it is built-in type (float), but if it had been;

something(x);

Is that a function or a cast? Most would assume function, but it could be a cast if this was in the header:

typedef float something;

but (something)x is clearly a cast and not a function call