net.java.sen.dictionary
Class Tokenizer

java.lang.Object
  net.java.sen.dictionary.Tokenizer

Direct Known Subclasses:

JapaneseTokenizer

public abstract class Tokenizer
extends Object

A String Tokenizer

The Tokenizer uses a Dictionary to assist the decomposition of strings into potential morphemes

Field Summary

protected Node	bosNode A Node representing a beginning-of-string
protected Dictionary	dictionary The Dictionary used to find possible morphemes
protected Node	eosNode A Node representing an end-of-string
protected CToken	unknownCToken A CToken representing an unknown morpheme
protected String	unknownPartOfSpeechDescription The part-of-speech code to use for unknown tokens

Constructor Summary

Tokenizer(Dictionary dictionary, String unknownPartOfSpeechDescription)
Constructs a new Tokenizer that uses the specified Dictionary to find possible morphemes within a given string

Method Summary

Node	getBOSNode() Creates a unique beginning-of-string Node.
Dictionary	getDictionary()
Node	getEOSNode() Creates a unique end-of-string Node.
Node	getUnknownNode(char[] surface, int start, int length, int span) Creates an "unknown morpheme" Node with the specified characteristics.
abstract Node	lookup(SentenceIterator iterator, char[] surface) Searches for possible morphemes from the given SentenceIterator.

Methods inherited from class java.lang.Object

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

Field Detail

dictionary

protected final Dictionary dictionary

The Dictionary used to find possible morphemes

unknownCToken

protected final CToken unknownCToken

A CToken representing an unknown morpheme

bosNode

protected final Node bosNode

A Node representing a beginning-of-string

eosNode

protected final Node eosNode

A Node representing an end-of-string

unknownPartOfSpeechDescription

protected final String unknownPartOfSpeechDescription

The part-of-speech code to use for unknown tokens

Constructor Detail

Tokenizer

public Tokenizer(Dictionary dictionary,
                 String unknownPartOfSpeechDescription)

Constructs a new Tokenizer that uses the specified Dictionary to find possible morphemes within a given string

Parameters:

dictionary - The Dictionary to search within

unknownPartOfSpeechDescription - The part-of-speech code to use for unknown tokens

Method Detail

getDictionary

public Dictionary getDictionary()

Returns:

Returns the dictionary used to find possible morphemes

getBOSNode

public Node getBOSNode()

Creates a unique beginning-of-string Node. The Node returned by this method is freshly cloned and not an alias of any other Node

Returns:

A beginning-of-string Node

getEOSNode

public Node getEOSNode()

Creates a unique end-of-string Node. The Node returned by this method is freshly cloned and not an alias of any other Node

Returns:

An end-of-string Node

getUnknownNode

public Node getUnknownNode(char[] surface,
                           int start,
                           int length,
                           int span)

Creates an "unknown morpheme" Node with the specified characteristics. The Node returned by this method is freshly cloned and not an alias of any other Node

Parameters:

surface - The underlying surface of which the Node is part

start - The index of the first character of the surface within the Node

length - The length of the Node

span - The span of the Node

Returns:

The new "unknown morpheme" Node

lookup

public abstract Node lookup(SentenceIterator iterator,
                            char[] surface)
                     throws IOException

Searches for possible morphemes from the given SentenceIterator. The Node that is returned links through Node.rnext to a list of matches which may be of varying lengths

Parameters:

iterator - The iterator to search from

surface - The underlying character surface

Returns:

The head of a chain of Nodes representing the possible morphemes beginning at the given index

Throws:

IOException