org.jaitools.numeric

Class CompareOp

java.lang.Object

org.jaitools.numeric.CompareOp

public class CompareOp
extends Object

Provides static methods to compare floating point values, taking into account an absolute or proportional tolerance. There are methods for both float and double values. The acompare and aequal methods use absolute tolerance while the pcompare and pequal methods use proportional tolerance.

For the proportional tolerance methods, a corresponding absolute tolerance is calculated as:

     atol = |ptol| * MAX(|x1|,|x2|)
 

Note: this class does not give any special consideration to the Float and Double constants NEGATIVE_INFINITY, POSITIVE_INFINITY and NaN over that provided by Java itself.

Since:

1.1

Version:

$Id$

Author:

Michael Bedward

Field Summary

Fields

Modifier and Type	Field and Description
static double	DTOL Default tolerance for double comparisons: 1.0e-8
static float	FTOL Default tolerance for float comparisons: 1.0e-4

Constructor Summary

Constructors

Constructor and Description
CompareOp()

Method Summary

Modifier and Type	Method and Description
static int	acompare(double x1, double x2) Compares two double values using the default tolerance.
static int	acompare(double x1, double x2, double tol) Compares two double values using the specified tolerance.
static int	acompare(float x1, float x2) Compares two float values using the default tolerance.
static int	acompare(float x1, float x2, float tol) Compares two float values using the specified tolerance.
static boolean	aequal(double x1, double x2) Tests if two double values are equal within the default tolerance.
static boolean	aequal(double x1, double x2, double tol) Tests if two double values are equal within the specified tolerance.
static boolean	aequal(float x1, float x2) Tests if two float values are equal within the default tolerance.
static boolean	aequal(float x1, float x2, float tol) Tests if two float values are equal within the specified tolerance.
static boolean	isZero(double x) Tests if the given double value is within the default tolerance of zero.
static boolean	isZero(double x, double tol) Tests if the given double value is within the specified tolerance of zero.
static boolean	isZero(float x) Tests if the given float value is within the default tolerance of zero.
static boolean	isZero(float x, float tol) Tests if the given float value is within the specified tolerance of zero.
static int	pcompare(double x1, double x2, double propTol) Compares two double values using the specified proportional tolerance.
static int	pcompare(float x1, float x2, float propTol) Compares two float values using the specified proportional tolerance.
static boolean	pequal(double x1, double x2, double propTol) Tests if two double values are equal within the specified proportional tolerance.
static boolean	pequal(float x1, float x2, float propTol) Tests if two float values are equal within the specified proportional tolerance.

Methods inherited from class java.lang.Object

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

Field Detail

DTOL

public static final double DTOL

Default tolerance for double comparisons: 1.0e-8

See Also:

Constant Field Values

FTOL

public static final float FTOL

Default tolerance for float comparisons: 1.0e-4

See Also:

Constant Field Values

Constructor Detail

CompareOp

public CompareOp()

Method Detail

isZero

public static boolean isZero(double x)

Tests if the given double value is within the default tolerance of zero.

Parameters:

x - the value

Returns:

true if zero; false otherwise

isZero

public static boolean isZero(float x)

Tests if the given float value is within the default tolerance of zero.

Parameters:

x - the value

Returns:

true if zero; false otherwise

isZero

public static boolean isZero(double x,
                             double tol)

Tests if the given double value is within the specified tolerance of zero. Note that performance reasons, tol is assumed to be positive, ie. this is not checked.

Parameters:

x - the value

tol - the tolerance

Returns:

true if zero; false otherwise

isZero

public static boolean isZero(float x,
                             float tol)

Tests if the given float value is within the specified tolerance of zero. Note that performance reasons, tol is assumed to be positive, ie. this is not checked.

Parameters:

x - the value

tol - the tolerance

Returns:

true if zero; false otherwise

acompare

public static int acompare(double x1,
                           double x2)

Compares two double values using the default tolerance.

Parameters:

x1 - first value

x2 - second value

Returns:

a value less than 0 if x1 is less than x2; 0 if x1 is equal to x2; a value greater than 0 if x1 is greater than x2

acompare

public static int acompare(float x1,
                           float x2)

Compares two float values using the default tolerance.

Parameters:

x1 - first value

x2 - second value

Returns:

a value less than 0 if x1 is less than x2; 0 if x1 is equal to x2; a value greater than 0 if x1 is greater than x2

acompare

public static int acompare(double x1,
                           double x2,
                           double tol)

Compares two double values using the specified tolerance. Note that performance reasons, tol is assumed to be positive, ie. this is not checked.

Parameters:

x1 - first value

x2 - second value

tol - comparison tolerance

Returns:

a value less than 0 if x1 is less than x2; 0 if x1 is equal to x2; a value greater than 0 if x1 is greater than x2

acompare

public static int acompare(float x1,
                           float x2,
                           float tol)

Compares two float values using the specified tolerance. Note that performance reasons, tol is assumed to be positive, ie. this is not checked.

Parameters:

x1 - first value

x2 - second value

tol - comparison tolerance

Returns:

a value less than 0 if x1 is less than x2; 0 if x1 is equal to x2; a value greater than 0 if x1 is greater than x2

pcompare

public static int pcompare(double x1,
                           double x2,
                           double propTol)

Compares two double values using the specified proportional tolerance. This is equivalent to:

     double absoluteTol = Math.abs(propTol) * Math.max(Math.abs(x1), Math.abs(x2));
     int comp = acompare(x1, x2, absTol);
 

Parameters:

x1 - first value

x2 - second value

propTol - proportional tolerance between 0 and 1

Returns:

a value less than 0 if x1 is less than x2; 0 if x1 is equal to x2; a value greater than 0 if x1 is greater than x2

pcompare

public static int pcompare(float x1,
                           float x2,
                           float propTol)

Compares two float values using the specified proportional tolerance. This is equivalent to:

     float absoluteTol = Math.abs(propTol) * Math.max(Math.abs(x1), Math.abs(x2));
     int comp = acompare(x1, x2, absTol);
 

Parameters:

x1 - first value

x2 - second value

propTol - proportional tolerance between 0 and 1

Returns:

a value less than 0 if x1 is less than x2; 0 if x1 is equal to x2; a value greater than 0 if x1 is greater than x2

aequal

public static boolean aequal(double x1,
                             double x2)

Tests if two double values are equal within the default tolerance. This is equivalent to dzero(x1 - x2).

Parameters:

x1 - first value

x2 - second value

Returns:

true if equal; false otherwise

aequal

public static boolean aequal(float x1,
                             float x2)

Tests if two float values are equal within the default tolerance. This is equivalent to dzero(x1 - x2).

Parameters:

x1 - first value

x2 - second value

Returns:

true if equal; false otherwise

aequal

public static boolean aequal(double x1,
                             double x2,
                             double tol)

Tests if two double values are equal within the specified tolerance. This is equivalent to dzero(x1 - x2, tol). Note that performance reasons, tol is assumed to be positive, ie. this is not checked.

Parameters:

x1 - first value

x2 - second value

tol - comparison tolerance

Returns:

true if equal; false otherwise

aequal

public static boolean aequal(float x1,
                             float x2,
                             float tol)

Tests if two float values are equal within the specified tolerance. This is equivalent to dzero(x1 - x2, tol). Note that performance reasons, tol is assumed to be positive, ie. this is not checked.

Parameters:

x1 - first value

x2 - second value

tol - comparison tolerance

Returns:

true if equal; false otherwise

pequal

public static boolean pequal(double x1,
                             double x2,
                             double propTol)

Tests if two double values are equal within the specified proportional tolerance. This is equivalent to:

     double absoluteTol = Math.abs(propTol) * Math.max(Math.abs(x1), Math.abs(x2));
     boolean b = aequal(x1, x2, absTol);
 

Parameters:

x1 - first value

x2 - second value

propTol - proportional tolerance between 0 and 1

Returns:

true if equal; false otherwise

pequal

public static boolean pequal(float x1,
                             float x2,
                             float propTol)

Tests if two float values are equal within the specified proportional tolerance. This is equivalent to:

     float absoluteTol = Math.abs(propTol) * Math.max(Math.abs(x1), Math.abs(x2));
     boolean b = aequal(x1, x2, absTol);
 

Parameters:

x1 - first value

x2 - second value

propTol - proportional tolerance between 0 and 1

Returns:

true if equal; false otherwise