nerdcruft/libcruft-util
1
0
Fork 0
Code Issues Releases Activity

Fix almost_equal for NaN and Inf

Browse Source
This commit is contained in:
Danny Robson 2012-05-03 15:58:18 +10:00
parent 68adb32472
commit 32e3e61fa7
1 changed files with 7 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
float.cpp
View File

@ -80,10 +80,14 @@ template <unsigned int E, unsigned int S>
bool
ieee_float<E, S>::almost_equal (floating_t a,
floating_t b) {
static const double epsilon = 0.001;
static const floating_t epsilon = 0.001;
const floating_t diff = fabs (a - b);
return fabs(a - b) <= epsilon * std::fabs (a) ||
fabs(a - b) <= epsilon * std::fabs (b);
// * Use an exact equality first so that infinities are not indirectly compared. This would generate NaNs in the diff.
// * Do not use gte or lte. This stops an infinite from making infinities on both sides of the inequality.
return exactly_equal (a, b) ||
diff < epsilon * std::fabs (a) ||
diff < epsilon * std::fabs (b);
}