sirius.kernel.commons

Class Doubles

java.lang.Object

sirius.kernel.commons.Doubles

public class Doubles
extends Object

Helper class for working with double numbers.

Field Summary

Fields

Modifier and Type	Field and Description
static double	EPSILON Used as maximal epsilon (difference) when comparing numbers.

Method Summary

Modifier and Type	Method and Description
static boolean	areEqual(double a, double b) Determines if the given numbers are equal.
static double	frac(double val) Returns the fractional part of the given value.
static boolean	isZero(double val) Determines if the given number is zero (or very very close to).

Methods inherited from class java.lang.Object

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

Field Detail

EPSILON

public static final double EPSILON

Used as maximal epsilon (difference) when comparing numbers. Doubles are not compared using == due to numeric constraints when using floating point numbers.

See Also:

Constant Field Values

Method Detail

frac

public static double frac(double val)

Returns the fractional part of the given value.

The fractional part is "everything behind the decimal separator". So for 1.454 it will be 0.454. This method always returns an absolute value so the fractional part of -4.5656 will be 0.5656.

Parameters:

val - the number for return the fractional part of

Returns:

the fractional part [0..1) of the number

areEqual

public static boolean areEqual(double a,
                               double b)

Determines if the given numbers are equal.

The numbers are not compared for absolute exactness (as == would do) but a minimal delta (EPSILON) is permitted to make up for rounding errors which are in the nature of floating point numbers.

Parameters:

a - the first number to compare

b - the second number to compare

Returns:

true if the absolute difference of the given numbers is less than EPSILON, false otherwise

isZero

public static boolean isZero(double val)

Determines if the given number is zero (or very very close to).

Determines if the given number is zero or less than EPSILON away from it. This is used to make up for rounding errors which are in the nature of floating point numbers.

Parameters:

val - the number to check if it is 0.

Returns:

true if the absolute value of the given numbers is less than EPSILON, false otherwise