com.jstatcom.util
Class UMath

java.lang.Object
  com.jstatcom.util.UMath

public final class UMath
extends java.lang.Object

This class is a collection of some mathematical functions. It can be used in addition to the Math class.

Author:

A. Benkwitz, Markus Kraetzig

Method Summary

static double[]	base10(double arg) Transforms arbitrary double values to a double in the interval (-10,-1] or [1,10) and a power of 10.
static int	closestMultiple(int value, int unit) Computes the closest multiple of unit for value.
static double	round(double arg, int n) Rounds the arg to a number with n decimal digits, for example 2.368=round(2.367849,3) If arg is NaN or +/-INF, then arg is returned.
static int	sign(int arg) Returns -1 if arg is smaller than zero and 1 if it is greater or equal to zero.

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Detail

base10

public static double[] base10(double arg)

Transforms arbitrary double values to a double in the interval (-10,-1] or [1,10) and a power of 10. It breaks number to a number of the form #.####... and a power of 10. { mant,pow } = base10(arg) with mant * 10^pow == arg returnValue[0]=mant, returnValue[1]=pow. If arg is NaN or +/-INF, then {arg, 0} is returned.

Parameters:

arg - the number to break

Returns:

two dimensional double array

• double[0] the mantisse
• double[1] the power of 10

closestMultiple

public static int closestMultiple(int value,
                                  int unit)

Computes the closest multiple of unit for value. If value is just in the middle between two multiples of unit, the higher multiple is choosen, for example 15 == closestMultiple(13, 5), 10 == closestMultiple(12, 5), 8 == closestMultiple(7, 4).

Parameters:

value - defines a closest multiple of unit

unit - defines the steps

Returns:

closest multiple of unit for value, returns value if (unit == 1), returns 0 if (unit == 0)

round

public static double round(double arg,
                           int n)

Rounds the arg to a number with n decimal digits, for example 2.368=round(2.367849,3) If arg is NaN or +/-INF, then arg is returned.

Parameters:

arg - the number to round

n - the number of decimal digits to round for

Returns:

the rounded number

sign

public static int sign(int arg)

Returns -1 if arg is smaller than zero and 1 if it is greater or equal to zero.

Parameters:

arg - the number to check the sign for

Returns:

-1 if (arg < 0), otherwise 1