nextafter() and nexttoward() in C/C++

How would you solve below problems in C/C++?

• What is the smallest representable positive floating point number in C/C++?
• What is the largest representable negative floating point number in C/C++?
• Given a positive floating point number x, find the largest representable floating point value smaller than x?

nextafter(x, y) and nexttoward(x.y)
In C and C++, both nextafter(x, y)  and nexttoward(x.y)  are similar functions defined in math.h or cmath header files. They both return next representable value after x in the direction of y. nexttoward() has more precise second parameter y.

The following program demonstrates the concept :

// C program to demonstrate use of nextafter() and nexttoward()
#include <stdio.h>
#include <math.h>
int main ()
{
    // using nextafter
    printf ("Smallest positive floating point number : %e\n",
            nextafter(0.0, 1.0));
    printf ("Largest negative floating point number :%e\n",
            nextafter(0.0, -1.0));
    printf ("Largest positive floating point number smaller than 0.5 : %e\n",
            nextafter(0.5, 0.0));

    // using nexttoward
    printf ("Smallest positive floating point number : %e\n",
            nexttoward(0.0, 1.0));
    printf ("Largest negative floating point number : %e\n",
            nexttoward(0.0, -1.0));
    printf ("Largest positive floating point number smaller than 0.5 : %e\n",
            nexttoward(0.5, 0.0));
    return (0);
}

Output:

nextafter first value greater than zero: 4.940656e-324
nextafter first value less than zero: -4.940656e-324
nexttoward first value greater than zero: 4.940656e-324
nexttoward first valnextafter first value greater than zero: 4.940656e-324
nextafter first value less than zero: -4.940656e-324
nexttoward first value greater than zero: 4.940656e-324
nexttoward first value less than zero: -4.940656e-324 ue less than zero: -4.940656e-324 

This article is contributed by Aditya Chatterjee. If you like GeeksforGeeks and would like to contribute, you can also write an article and mail your article to contribute@geeksforgeeks.org. See your article appearing on the GeeksforGeeks main page and help other Geeks.

Please write comments if you find anything incorrect, or you want to share more information about the topic discussed above