int main() {
double inf = INFINITY;
double pi = acos(-1.0);
printf("[1]: %f %f\n", atan(inf) / pi, atan(-inf) / pi);
printf("[2]: %f %f\n", tan(inf) / pi, tan(-inf) / pi);
return 0;
}
outputs
[1]: 0.500000 -0.500000
[2]: -nan -nan
Is such behaviour defined by the standard? Is [2] undefined behaviour? unspecified?
I want to be sure at least [1] is a guaranteed result.
Both are well defined behaviour. Quoting from http://en.cppreference.com
tan
If the argument is ±0, it is returned unmodified.
If the argument is ±∞, NaN is returned and FE_INVALID is raised.
If the argument is NaN, NaN is returned.
atan
If the argument is ±0, it is returned unmodified.
If the argument is +∞, +π/2 is returned.
If the argument is -∞, -π/2 is returned.
If the argument is NaN, NaN is returned.