com.goldencode.p2j.schema

Class NameNode.AbbreviationComparator

java.lang.Object

com.goldencode.p2j.schema.NameNode.ExactMatchComparator

com.goldencode.p2j.schema.NameNode.AbbreviationComparator

All Implemented Interfaces:

java.io.Serializable, java.util.Comparator<NameNode>

Enclosing class:

NameNode

private static class NameNode.AbbreviationComparator
extends NameNode.ExactMatchComparator

Comparator which will report two NameNode objects as equivalent if the second node's name represents an abbreviation for the first node's name. This is used when looking up a name which may be abbreviated, when an exact match lookup has already failed.

Note: this comparator imposes orderings that are inconsistent with equals.

Constructor Summary

Constructors

Modifier	Constructor and Description
private	AbbreviationComparator()

Method Summary

Modifier and Type	Method and Description
int	compare(NameNode n1, NameNode n2) Resolves a comparison of two NameNode objects.

Methods inherited from class java.lang.Object

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

Methods inherited from interface java.util.Comparator

comparing, comparing, comparingDouble, comparingInt, comparingLong, equals, naturalOrder, nullsFirst, nullsLast, reversed, reverseOrder, thenComparing, thenComparing, thenComparing, thenComparingDouble, thenComparingInt, thenComparingLong

Constructor Detail

AbbreviationComparator

private AbbreviationComparator()

Method Detail

compare

public int compare(NameNode n1,
                   NameNode n2)

Resolves a comparison of two NameNode objects. The criteria for this comparison to determine equivalence is either an exact lexical match of the nodes' names, or that the name of node n1 lexically begins with the name of node n2.

Important Node: this implementation supports the lookup of an abbreviated name within a sorted list of unabbreviated names. It relies upon the binary search implementation provided by java.util.Collections, in that it assumes the search key will always be n2, and n1 always represents a node from the list being searched. This assumption is fundamental to the correct behavior of the Namespace.find(java.lang.String, boolean, java.util.Set<java.lang.String>) method.

Specified by:

compare in interface java.util.Comparator<NameNode>

Overrides:

compare in class NameNode.ExactMatchComparator

Parameters:

n1 - First NameNode for comparison.

n2 - Second NameNode for comparison.

Returns:

0 if nodes' names are considered equal according to the rules described above; -1 if o1 is considered lexically less than o2; 1 if o1 is considered lexically greater than o2.