function getting wrong values

so i have this function in C to calculate a power, and i'm using visual c++ 2010

power.h

void power();  
float get_power(float a, int n);  

power.c

void power()
{
    float a, r;
    int n;
    printf("-POWER-\n");
    printf("The base: ");
    scanf("%f", &a);
    n = -1;
    while (n < 0)
    {
        printf("The power: ");
        scanf("%d", &n);
        if (n < 0)
        {
            printf("Power must be equal or larger than 0!\n");
        }
        else
        {
            r = get_power(a, n);
            printf("%.2f ^ %d = %.2f", a, n, r);
        }
    };
}

float get_power(float a, int n)
{
    if (n == 0)
    {
        return 1;
}
    return a * get_power(a, n-1);
}

not the best way to do it, i know, but that's not it

when i debug it the values are scanned correctly (that is, the values are correct until just before the function call) but then upon entering the function a becomes 0 and n becomes 1074790400, and you can guess what happens next...

the first function is being called from the main file, i included the full code because i really have no idea what could be going on, and i can't even think on how to google for it...

strangely, i wrote the function in a single file and it works fine, but it d开发者_如何学JAVAefinitely should work both ways

any idea why this is happening?

---

Do you have

#include "power.h"

at the top of power.c?

If not, the compiler doesn't know what the prototype of get_power() is at the point of the call, so it will resort to promoting the first argument to a double instead of passing it as a float. It'll also incorrectly assume that the result is an int instead of the float that's being returned.

If the compiler sees the prototype before the call, things will work better.

---

You can use pow from math.h. In other way you have error in multiplication in get_power, better write this on for.

---

Was there a specific reason why you wanted to use recursion?

What about something simple like (untested):

float get_power(float a, int n)
{
    float result = 1.0;
    for (int i = 0; i < n; i++)
    {
        result = result * a;
    }
    return result;
}

---

Just to add to the existing answer, I ran this on xlc to see if it was a visual studio only problem. I got this answer:

"power.c", line 25.7: 1506-343 (S) Redeclaration of get_power differs from previous declaration on line 19 of "power.c".
"power.c", line 25.7: 1506-050 (I) Return type "float" in redeclaration is not compatible with the previous return type "int".
"power.c", line 25.7: 1506-379 (I) Prototype for function get_power must contain only promoted types if prototype and nonprototype declarations are mixed.
"power.c", line 25.7: 1506-380 (I) Parameter 1 has type "float" which promotes to "double".

I also tried it in a few other compilers I had around and got variations on this theme. You can see this as a warning in VS if you turn up the warning level.

So in conclusion, it SHOULDN'T compile, and VS is the only one that does compile and link it.