You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
ok thanks for the qFuzzyCompare, this is fixing the <float, float> curve case.
however I've raised another bug which seems much more tricky but certainly relative to double precision.
for a <float, int> curve from (-10., -10) point to (10., 10) point curve->valueAt(-9.) returns -8 !
the test is in CurveTest.cpp and fails as I'm expecting to get -9.
this on feature/2d_curve branch in case you have the time to look at it.
there is a strange bug on Curve when trying to test valueAt for the <float, float> curve case :
QVERIFY(curveB->valueAt(-60.) == -60.); //! \note it returns 60. !
QVERIFY(curveB->valueAt(-40.) == -40.); //! \note it returns something like ~39.999... ?!?
QVERIFY(curveB->valueAt(-20.) == -20.); //! \note it returns something like ~19.999... ?!?
all the result are supposed to be made by the same formula in Curve.cpp and CurveSegmentLinear.
maybe you have an idea of what happen here @jcelerier ?
The text was updated successfully, but these errors were encountered: