modf, modff, modfl

Parameters

Return value

If no errors occur, returns the fractional part of x with the same sign as x. The integral part is put into the value pointed to by iptr.

The sum of the returned value and the value stored in *iptr gives arg (allowing for rounding).

Error handling

This function is not subject to any errors specified in math_errhandling.

If the implementation supports IEEE floating-point arithmetic (IEC 60559),

• If arg is ±0, ±0 is returned, and ±0 is stored in *iptr.
• If arg is ±∞, ±0 is returned, and ±∞ is stored in *iptr.
• If arg is NaN, NaN is returned, and NaN is stored in *iptr.
• The returned value is exact, the current rounding mode is ignored

Notes

This function behaves as if implemented as follows:

double modf(double value, double *iptr)
{
#pragma STDC FENV_ACCESS ON
    int save_round = fegetround();
    fesetround(FE_TOWARDZERO);
    *iptr = std::nearbyint(value);
    fesetround(save_round);
    return copysign(isinf(value) ? 0.0 : value - (*iptr), value);
}

Example

#include <stdio.h>
#include <math.h>
#include <float.h>
 
int main(void)
{
    double f = 123.45;
    printf("Given the number %.2f or %a in hex,\n", f, f);
 
    double f3;
    double f2 = modf(f, &f3);
    printf("modf() makes %.2f + %.2f\n", f3, f2);
 
    int i;
    f2 = frexp(f, &i);
    printf("frexp() makes %f * 2^%d\n", f2, i);
 
    i = ilogb(f);
    printf("logb()/ilogb() make %f * %d^%d\n", f/scalbn(1.0, i), FLT_RADIX, i);
 
    // special values
    f2 = modf(-0.0, &f3);
    printf("modf(-0) makes %.2f + %.2f\n", f3, f2);
    f2 = modf(-INFINITY, &f3);
    printf("modf(-Inf) makes %.2f + %.2f\n", f3, f2);
}

Given the number 123.45 or 0x1.edccccccccccdp+6 in hex,
modf() makes 123.00 + 0.45
frexp() makes 0.964453 * 2^7
logb()/ilogb() make 1.92891 * 2^6
modf(-0) makes -0.00 + -0.00
modf(-Inf) makes -INF + -0.00

References