"Jeroen Frijters" <[EMAIL PROTECTED]> writes: > Hi, > > Attached is a small patch to java.lang.Math to make round(float) and > round(double) do the right thing for NaN. > > Style wise, is it a good idea to use "a != a" to test for NaN, or should > Double.isNaN() be used?
I went ahead and converted all the a != a to Double.isNaN or Float.isNaN as appropriate. As mentioned previously a good JIT should optimize this... maybe even a good byte-code compiler. Brian -- Brian Jones <[EMAIL PROTECTED]> _______________________________________________ Classpath mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/classpath
