problem with g++

[putnam120]

I am using Visual C++ and when I compile the following code there are no errors or warnings. However, when I compile with g++ I get 3 errors and 1 warning.

Here is the part of the code that generated the errors.
----------------------------------------------------------------------------------------------------------------

int divsum(int x, vector<int> y) //returns the sum of the PROPER divisors
{
    int t;
    double sum=1;
    int L=y.size();
    for(t=0; t<L; t++) //go through each element of the list of generated primes
    {
        int e=1;
        double psum=1;
        if(x%y[t]==0)//test for divisibility
        {
            while(fmod(x,pow(y[t],e))==0) //sum of p^e where p^e divides y
            {
                psum=(int) psum+pow(y[t],e);
                e++;
            }
            sum=sum*psum; //formula for divisor sum
        }
    }
    int rsum; rsum= (int)sum;
    return rsum-x;
}

-------------------------------------------------------------------------------------------------------------
The error messages all read something similar to: "divisorsum.cpp:73: error: call of overloaded âpow(int&, int&)â is ambiguous"

What is the mistake, and how can I avoid making it in the future?

[Laharl]

pow() is an old C function, which for some strange reason was never written to take two integer parameters or return an integer. You could write your own version, but the valid signatures for <cmath> pow() are:

double pow(double base, double exponent)
long double pow(long double base, long double exponent)
float pow(float base, float exponent)

You can use a cast to change your int into a double - not a static_cast<double> or anything similar (those are for real emergencies) - just a simple cast and then you could use floor() to return it to integer form. Since your doubles will be whole numbers in this problem, there is no issue.

int a = 4, b = 2;
int c = floor(pow(double(4), double(2))); //c = 16
 

[Ganon11]

Ah, so our mystery pow() problem seems to be g++ specific? Or, at least, Visual C++ has provided an int pow(int, int) function in their <cmath> header file.

Anyway, your warning is probably, "WARNING: no newline at end of file." I think this is a ridiculously silly warning, but you can get rid of it by entering your source code file, pressing Ctrl+End to go to the end of your file, pressing Enter/Return once, and saving.