function redeclaration gives an error

When I build the following I get an error as follows:

'float computedecimal' redeclared as different kind of symbol
Previous declaration of 'float computedecimal (int, int, int)'

Code:

#include <iostream>

using namespace std;

int degrees {};
int minutes {};
int seconds {};
float decimal{};

const int secondsinminute {60};
const int  minutesindegree {60};

float computedecimal (int, int, int);


int main()
{
    cout<<"Please insert degrees : ";
    cin>> degrees;

    cout<<endl<<endl<<"Please insert minutes : ";
    cin>>minutes;

    cout<<endl<<endl<<"Please insert seconds : ";
    cin>>seconds;

    decimal=computedecimal (degrees, minutes, seconds);

    cout<<endl<<endl<<degrees<<" degrees, "<<minutes<<" minutes, "<<seconds<<" seconds = "
    <<decimal<<"  degrees";

    cin.get ();
    cin.get ();

    return 0;
}


float computedecimal (degrees, minutes, seconds)
{
    int top=(minutes*secondsinminutes)+seconds;
    int bottom=secondsinminutes*minutesindegree;

    decimal=degrees+(top/bottom);

    return decimal;
}

I don't understand this as I am passing 3 ints to the function and expecting a float in return.

Obviously I am doing something wrong but can't figure it out myself.

Your function definition:

Code:

float computedecimal (degrees, minutes, seconds)

does not match your function prototype:

Code:

float computedecimal (int, int, int);

You should keep the parameter names in the prototype to avoid this kind of mistake.

Thanks laserlight.

I changed the prototype to

Code:

float computedecimal (degrees, minutes, seconds);

But now get 2 errors :
error : expression list treated as compund expression in initializer [f-permissive]
and
'computedecimal' cannot be used as a function

Did I understand you correctly when I changed the initial declaration?

No you didn't understand laserlight's meaning.

Leave the function declaration (aka prototype) "as is" and change the implementation to

Code:

    float computedecimal (int degrees, int minutes, int seconds);

You need to be clear in your mind whether the function receives inputs as arguments or as variables declared at file scope (when variables or arguments in different scopes have the same name, you need to be careful about which one is affected by any statement).

Thanks grumpy,
That now works as expected, except for one thing - decimal which I declared as a float only produces an integer. Do the variables it uses to compute it also have to be floats, currently they are integers because that is what they will always be but of course when you divide two integers you can end up with a floating point answer.
I seem to be having problems even with the first chapters of my book, maybne it is because I am a competent BASIC programmer but have never tried C++ before - any tips for getting over that?

You're the one who specified that computedecimal() accepts three ints, and specified that the variables top and bottom in that function are of type int.

You may think that dividing two integers yields a floating point answer, but computer hardware specifications and the C standard both disagree with you. Dividing two integers does an integer division, not a floating point division. So 2/3 will be computed as zero (rounded down), and a floating point value will never be computed in that process.

If you want the division to be done as floating point, then you need to force the issue. One approach is to define top and bottom to be of type float (or double) rather than int: a division that involves a floating point operand is done in floating point. The other way is to convert one of the values to be of type float. If you read your text a bit more closely, you will find that described in early chapters.

As to being a "competent BASIC programmer" learning C++, the rules of the two languages are quite different, so you can't expect your experience with BASIC will help you all that much with C++. You are better off assuming you know nothing about computer programming, and reading your C++ text without assuming you can skip sections because you know BASIC. Otherwise you will spend more time "unlearning" BASIC than you will learning anything useful about C++. (The same advice would apply, in reverse, to a C++ programmer learning BASIC - with the caveat that BASIC is a much simpler language than C++, so it is easier to learn from scratch).

Thanks grumpy, good advice which I shall heed.

I had assumed that if the result was declared as a float then the division would give that result, but now I see the error of my ways, I have actually learnt quite a lot today so I thank you again, I appreciate it must be frustrating having to explain these little intricacies to a beginner, especially one who knows a different language.

Originally Posted by grumpy

As to being a "competent BASIC programmer"

I like how you put that in quotes, very droll.

Actually, what I meant is that your prototype should be:

Code:

float computedecimal(int degrees, int minutes, int seconds);

Then the definition starts with:

Code:

float computedecimal(int degrees, int minutes, int seconds)
{