我在以下代码中遇到了问题，这些代码具有不同的变量类型和用于查找 Bessel 函数的指针

Question

This assignment is one of my labs and it wants us to find the value of the 1st order Bessel function according to the input-ed user value. So long, my mindset is: first I will ask the user for input, then I use the number's rounding, the number above and the number below. After that, I perform the interpolation according to the numbers, find the error, then present everything on screen. (We must use pointers and arrays else we will not get any marks. That makes sense, how are we gonna list out every case for every rounded number from 0 to 10......)

The problem is, I seem to not get the pointers right, and the program always say that we have long int and *double mixed, even though I explicitly told the program that this is a double.

Code:

#include <stdio.h>
#include <math.h>
#include <Stdlib.h>

int y;

 int readdata(double rho[y], double Jrho[y])
{

double a, b, c, x;

FILE* fp;
fp= fopen("Bessel1.dat", "r");

for (y=0; y<=10; y++)
    {
        fscanf(fp, "%lf %lf", &rho[y], &Jrho[y]);
    }

}

double Lagrange2nd(int y, double rho[y], double Jrho[y], double x)
{

double a, b, c;

double output;

a=(x-rho[(y+1)])*((x-rho[y])*Jrho[((y-1))]+(rho[((y-1))]- 

x)*Jrho[y])/(rho[((y-1))]-rho[y]);

b=(((rho[(y-1)]-x)*((x-rho[(y+1)])*Jrho[y]+(rho[y]- 

x)*Jrho[(y+1)]))/(rho[y]-rho[(y+1)]));

output = c = (a+b)/(rho[(y-1)]-rho[(y+1)]);

return output;
}

double errx(int y, double rho[y], double Jrho[y], double x)
{
double a, b, c;

double output;

output= (abs(c-(Jrho[(y-1)]*(x-rho[y])+Jrho[y]*(rho[(y-1)]-x)))+abs(c- 
(Jrho[y]*(x-rho[(y+1)])+Jrho[(y+1)]*(rho[y]-x))))/2;

return output;

}

int main()
{
double x, Jrho[y], rho[y];

printf("x=");
scanf("%lf", &x);
y=round(x);

Lagrange2nd(&rho[y], &x, &Jrho[y], y);

errx(&Jrho[y], &x, &rho[y], y);

printf("%d, %d", Lagrange2nd, errx);

return(0);
}

I know this is messy, but I did not know any programming until 2 months ago, yet we now have to do this......gee, I am not very good at this, am I?

Answer 1

What compiler do you use? with gcc you will get MANY warnings. In particular, you have few functions that declare to return void, but return an int.

Looks like you compiler does treat 'void' as 'int' - which will cause lot of precision issue. If you have turned of warning - consider re-enabling them, or use better compiler:-)

void errx(double* rho[y], double* Jrho[y], double* x)
{
...
double output;

output= (abs(c-(Jrho[(y-1)]*(x-rho[y])+Jrho[y]*(rho[(y-1)]-x)))+abs(c-(Jrho[y]*(x- 
rho[(y+1)])+Jrho[(y+1)]*(rho[y]-x))))/2;

return output;
}

Style note: consider combing declaration and assignment in one statement. It will reduce many possible errors, and will get you better grade!

double output = (abs(c-(Jrho[(y-1)]*(x-rho[y])+Jrho[y]*(rho[(y-1)]-x)))+abs(c-(Jrho[y]*(x-rho[(y+1)])+Jrho[(y+1)]*(rho[y]-x))))/2;