com.ibm.tivoli.maximo.integrity.repair

Class PossibilityFilter

java.lang.Object

com.ibm.tivoli.maximo.integrity.repair.PossibilityFilter

public class PossibilityFilter
extends java.lang.Object

The possibility filter is used to get a good idea whether an object belongs in a set. This is useful when that set is too big to fit in memory. Once trained, the set will return no false negatives, but can return false positives. The chance of false positives depends on the size of the filter.

Author:

John Neville

Constructor Summary

Constructors

Constructor and Description
PossibilityFilter() Constructor.
PossibilityFilter(int size) Constructor.

Method Summary

Modifier and Type	Method and Description
boolean	isPossible(java.lang.Object testValue) Is the value given possibly a value the filter was trained for?
void	train(java.lang.Object validationValue) Train the filter to recognize this value.

Methods inherited from class java.lang.Object

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

Constructor Detail

PossibilityFilter

public PossibilityFilter()

Constructor. Creates filter with the default size of 1,048,576, a.k.a. one megabyte.

PossibilityFilter

public PossibilityFilter(int size)

Constructor.

Parameters:

size - Size of the filter. The larger the size, the fewer false positives.

Method Detail

train

public void train(java.lang.Object validationValue)

Train the filter to recognize this value.

Parameters:

validationValue - A valid value the filter should recognize. An exception will be thrown if this parameter is null.

See Also:

Values given here will return true when given to the isPossible method.

isPossible

public boolean isPossible(java.lang.Object testValue)

Is the value given possibly a value the filter was trained for?

Parameters:

testValue - The value to test whether the filter recognizes it. An exception will be thrown if this parameter is null.

Returns:

This method will always return true if the test value matches a value this filter was trained with. If value isn't one trained for, there is a chance this method will still return true. If this method returns false, there is no chance the testValue was trained for.