com.orchestranetworks.addon.tese

Class SearchDistanceObject

java.lang.Object

com.orchestranetworks.addon.tese.SearchComparator

com.orchestranetworks.addon.tese.SearchDistanceObject

public abstract class SearchDistanceObject
extends SearchComparator

Abstract class for implementing distance algorithms between types. For String it is better to use SearchDistance.

Since:

1.4.2

Constructor Summary

Constructors

Constructor and Description
SearchDistanceObject(com.orchestranetworks.schema.SchemaNode schemaNode, Locale locale)

Method Summary

Modifier and Type	Method and Description
abstract float	getDistance(Object baseObject, Object otherObject) Returns the distance between two objects.
boolean	isNullSupported() Null value must be supported.
String	normalizeValue(String value) Normalizes the value before any comparison.

Methods inherited from class com.orchestranetworks.addon.tese.SearchComparator

cloneMatching, equals, getLocale, getParameter, getSchemaNode, hashCode, isDateSupported, isFullTableScanRequired, isNumericSupported, setParameter

Methods inherited from class java.lang.Object

getClass, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

SearchDistanceObject

public SearchDistanceObject(com.orchestranetworks.schema.SchemaNode schemaNode,
                            Locale locale)

Method Detail

normalizeValue

public final String normalizeValue(String value)

Description copied from class: SearchComparator

Normalizes the value before any comparison. Default behavior: value is changed into lower-case. Can be overridden in subclasses to change this default behavior.

Overrides:

normalizeValue in class SearchComparator

Throws:

IllegalStateException - in all cases, this method must not be called.

isNullSupported

public final boolean isNullSupported()

Null value must be supported.

Overrides:

isNullSupported in class SearchComparator

getDistance

public abstract float getDistance(Object baseObject,
                                  Object otherObject)

Returns the distance between two objects. 0 means completely different. The value returned for strict equality must be strictly superior to zero. This value will be the baseline to convert all other distance into a percentage. Considering baseObject to be compared with otherObject, the percentage of similarity will be calculated as follow.

baseline = getDistance(baseObject, baseObject) distance = getDistance(baseObject, otherObject) similarity = distance*baseObject/100