If a double is demoted to a float or a long double is demoted to a double or a float, and the initial value exceeds the maximum of the demoted type, then the value of the result undefined.
Non-Compliant Code Example
This non-compliant code illustrates possible undefined behavior associated with demoting floating point represented numbers.
long double ld; double d1; double d2; float f1; float f2; ... f1 = (float)d1; f2 = (float)ld; d2 = (double)ld;
In the assignments above, it is possible that the variable "d1" exceeds the maximum value that can be stored by a float or that the variable "ld" exceeds the maximum value that can be stored in either a float or a double. One cannot make any assumptions about the value stored if the value is too large to be represented.
Compliant Code Example
This compliant code properly checks to see whether the values to be stored are too large to be represented.
#include <float.h>
...
long double ld;
double d1;
double d2;
float f1;
float f2;
...
if(d1 > FLT_MAX) {
/* Handle error condition */
} else {
f1 = (float)d1;
}
if(ld > FLT_MAX) {
/* Handle error condition */
} else {
f2 = (float)ld;
}
if(ld > DBL_MAX) {
/* Handle error condition */
} else {
d2 = (double)ld;
}
(Extrapolate, 6.3.1.5.2)
(Cite C99)