prefuse.data.search
Class KeywordSearchTupleSet

java.lang.Object
  prefuse.data.tuple.AbstractTupleSet
      prefuse.data.tuple.DefaultTupleSet
          prefuse.data.search.SearchTupleSet
              prefuse.data.search.KeywordSearchTupleSet

All Implemented Interfaces:

EventConstants, TupleSet

public class KeywordSearchTupleSet
extends SearchTupleSet

SearchTupleSet implementation that performs text searches on indexed Tuple data using the Lucene search engine. Lucene is an open source search engine supporting full text indexing and keyword search. Please refer to the Lucene web page for more information. Note that for this class to be used by prefuse applications, the Lucene classes must be included on the application classpath.

Version:

1.0

Author:

jeffrey heer

See Also:

SearchQueryBinding

Field Summary

protected int	m_id
protected LuceneSearcher	m_lucene
protected IntObjectHashMap	m_map
protected java.lang.String	m_query
protected boolean	m_storeTermVectors

Fields inherited from class prefuse.data.tuple.DefaultTupleSet

m_tuples

Fields inherited from interface prefuse.data.event.EventConstants

ALL_COLUMNS, DELETE, INSERT, UPDATE

Fields inherited from interface prefuse.data.tuple.TupleSet

EMPTY_ARRAY

Constructor Summary

KeywordSearchTupleSet()
Creates a new KeywordSearchFocusSet using an in-memory search index.

KeywordSearchTupleSet(LuceneSearcher searcher)
Creates a new TextSearchFocusSet with the given LuceneSearcher.

Method Summary

void	clear() Removes all search hits and clears out the index.
protected org.apache.lucene.document.Document	getDocument(int id, java.lang.String text) Create a Lucene Document instance with the given document ID and text.
LuceneSearcher	getLuceneSearcher() Get the LuceneSearcher instance used by this class.
protected Tuple	getMatchingTuple(org.apache.lucene.document.Document d) Return the Tuple matching the given Lucene Document, if any.
java.lang.String	getQuery() Returns the current search query, if any.
IntObjectHashMap	getTupleMap() Returns a copy of the mapping from Lucene document IDs to prefuse Tuple instances.
void	index(Tuple t, java.lang.String field) Index an individual Tuple field, so that it can be searched for.
boolean	isUnindexSupported() Returns false, as unindexing values is not currently supported.
void	search(java.lang.String query) Searches the indexed Tuple fields for matching keywords, using the Lucene search engine.
void	unindex(Tuple t, java.lang.String attrName) This method throws an exception, as unidexing is not supported.

Methods inherited from class prefuse.data.search.SearchTupleSet

addTuple, index, removeTuple

Methods inherited from class prefuse.data.tuple.DefaultTupleSet

addInternal, clearInternal, containsTuple, getTupleCount, removeInternal, setTuple, toArray, tuples

Methods inherited from class prefuse.data.tuple.AbstractTupleSet

addColumn, addColumn, addColumn, addColumn, addColumns, addPropertyChangeListener, addPropertyChangeListener, addTupleSetListener, fireTupleEvent, fireTupleEvent, fireTupleEvent, getClientProperty, isAddColumnSupported, putClientProperty, removePropertyChangeListener, removePropertyChangeListener, removeTupleSetListener, tuples, tuples

Methods inherited from class java.lang.Object

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

Field Detail

m_map

protected IntObjectHashMap m_map

m_query

protected java.lang.String m_query

m_lucene

protected LuceneSearcher m_lucene

m_storeTermVectors

protected boolean m_storeTermVectors

m_id

protected int m_id

Constructor Detail

KeywordSearchTupleSet

public KeywordSearchTupleSet()

Creates a new KeywordSearchFocusSet using an in-memory search index.

KeywordSearchTupleSet

public KeywordSearchTupleSet(LuceneSearcher searcher)

Creates a new TextSearchFocusSet with the given LuceneSearcher.

Parameters:

searcher - the LuceneSearcher to use.

Method Detail

getQuery

public java.lang.String getQuery()

Returns the current search query, if any.

Specified by:

getQuery in class SearchTupleSet

Returns:

the currently active search query

search

public void search(java.lang.String query)

Searches the indexed Tuple fields for matching keywords, using the Lucene search engine. Matching Tuples are available as the members of this TupleSet.

Specified by:

search in class SearchTupleSet

Parameters:

query - the query string to search for

getMatchingTuple

protected Tuple getMatchingTuple(org.apache.lucene.document.Document d)

Return the Tuple matching the given Lucene Document, if any.

Parameters:

d - the Document to lookup.

Returns:

the matching Tuple, or null if none.

index

public void index(Tuple t,
                  java.lang.String field)

Description copied from class: SearchTupleSet

Index an individual Tuple field, so that it can be searched for.

Specified by:

index in class SearchTupleSet

Parameters:

t - the Tuple

field - the data field to index

See Also:

SearchTupleSet.index(prefuse.data.Tuple, java.lang.String)

isUnindexSupported

public boolean isUnindexSupported()

Returns false, as unindexing values is not currently supported.

Specified by:

isUnindexSupported in class SearchTupleSet

Returns:

true if unindex is supported, false otherwise.

See Also:

SearchTupleSet.isUnindexSupported()

unindex

public void unindex(Tuple t,
                    java.lang.String attrName)

This method throws an exception, as unidexing is not supported.

Specified by:

unindex in class SearchTupleSet

Parameters:

t - the Tuple

attrName - the data field to unindex

Throws:

java.lang.UnsupportedOperationException

See Also:

SearchTupleSet.unindex(prefuse.data.Tuple, java.lang.String)

getDocument

protected org.apache.lucene.document.Document getDocument(int id,
                                                          java.lang.String text)

Create a Lucene Document instance with the given document ID and text.

Parameters:

id - the document ID

text - the text the Document should contain

Returns:

a new Lucene Document instance

getLuceneSearcher

public LuceneSearcher getLuceneSearcher()

Get the LuceneSearcher instance used by this class.

Returns:

returns the backing lucene searcher.

getTupleMap

public IntObjectHashMap getTupleMap()

Returns a copy of the mapping from Lucene document IDs to prefuse Tuple instances.

Returns:

a copy of the map from lucene doc IDs to prefuse Tuples.

clear

public void clear()

Removes all search hits and clears out the index.

Specified by:

clear in interface TupleSet

Overrides:

clear in class DefaultTupleSet

See Also:

TupleSet.clear()