org.ceryle.util.search
Class Indexer

java.lang.Object
  java.lang.Thread
      org.ceryle.util.search.Indexer

All Implemented Interfaces:

Runnable

Direct Known Subclasses:

FSIndexer, XNodeIndexer

public abstract class Indexer
extends Thread

An abstract class that extends Thread to build a search index. This acts as a base class to be extended for various index types. The list of document fields may also be extended as necessary.

Note that this class is not reentrant.

Since:

JDK1.4

Version:

$Id: Indexer.java,v 1.14 2007-06-20 04:41:59 altheim Exp $

Author:

Murray Altheim

Nested Class Summary

Nested classes/interfaces inherited from class java.lang.Thread

Thread.State, Thread.UncaughtExceptionHandler

Field Summary

static int	ANALYZER_SIMPLE An identifier for the SimpleAnalyzer used by the search engine.
static int	ANALYZER_STANDARD An identifier for the StandardAnalyzer used by the search engine.
static int	ANALYZER_TYPE The Analyzer type to be used by the search engine.
protected boolean	create When true, indexing operations will create a new index from scratch rather than modifying an existing index.
protected static org.apache.lucene.analysis.Analyzer	m_analyzer The Analyzer used by the search engine.
protected File	m_indexDirectory The directory used by the search engine for storing its index.
protected org.apache.lucene.index.IndexWriter	m_writer The IndexWriter used by the search engine.
protected MessageHandler	mh
protected MessageWriter	mw
protected ProgressBar	progress
static String	SEARCH_INDEX_PATH The identifier of the property containing the search index path.
protected Services	srvs
static int	THREAD_PRIORITY The default Thread priority (Thread.MAX_PRIORITY).

Fields inherited from class java.lang.Thread

MAX_PRIORITY, MIN_PRIORITY, NORM_PRIORITY

Constructor Summary

Indexer(String name, MessageWriter msgwriter)
Constructor with a name identifier (used by the thread group manager), the MessageWriter to receive indexing and search result messages, and a File reference for the index directory.

Method Summary

void	closeIndex() Close any open index writer.
void	deleteIndex() Delete the built index of this Indexer.
org.apache.lucene.analysis.Analyzer	getAnalyzer() Return the Analyzer to be used for indexing content.
File	getIndexDirectory() Returns the index directory.
org.apache.lucene.index.IndexReader	getIndexReader() Returns an IndexReader used to restore the index.
org.apache.lucene.index.IndexWriter	getIndexWriter() Return the IndexWriter used by this XNodeIndexer.
MessageHandler	getMessageHandler() Returns the MessageHandler, which receives general informational, log, warning and error messages.
MessageWriter	getMessageWriter() Returns the MessageWriter, which displays index- and search-related messages directly to the user.
ProgressBar	getProgressBar() Returns the optional ProgressBar, or null if it is not set.
static String[]	getStopWords() Return the stop words being used by the analyzer.
long	getVersion() Returns the index version number, -1 if it has not been set.
boolean	hasProgressBar() Returns true if the optional ProgressBar has been set.
void	index(boolean createIndex) Begins an indexing process.
boolean	isCreateIndex() Returns true if the index should be created/recreated when the index is built/rebuilt, rather than just modified.
boolean	isIndexed() Returns true if Indexer indicates an index completed state.
void	mergeIndex(Indexer indexer) Merge the index provided by indexer with this index.
abstract boolean	restoreIndex() Restore the existing index after opening a new session, returning true if successful.
abstract void	run()
void	setCreateIndex(boolean createIndex) Sets the property that when true indicates that when index are built, they should be built from scratch rather than just modified.
void	setIndexDirectory(File directory) Sets the directory used to store the index.
protected void	setIsIndexed(boolean indexed) Set the boolean value indicating an index completed/ready state.
void	setProgressBar(ProgressBar progressBar) Sets an optional ProgressBar to indicate process status.

Methods inherited from class java.lang.Thread

activeCount, checkAccess, countStackFrames, currentThread, destroy, dumpStack, enumerate, getAllStackTraces, getContextClassLoader, getDefaultUncaughtExceptionHandler, getId, getName, getPriority, getStackTrace, getState, getThreadGroup, getUncaughtExceptionHandler, holdsLock, interrupt, interrupted, isAlive, isDaemon, isInterrupted, join, join, join, resume, setContextClassLoader, setDaemon, setDefaultUncaughtExceptionHandler, setName, setPriority, setUncaughtExceptionHandler, sleep, sleep, start, stop, stop, suspend, toString, yield

Methods inherited from class java.lang.Object

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

Field Detail

SEARCH_INDEX_PATH

public static final String SEARCH_INDEX_PATH

The identifier of the property containing the search index path.

See Also:

Constant Field Values

ANALYZER_SIMPLE

public static final int ANALYZER_SIMPLE

An identifier for the SimpleAnalyzer used by the search engine.

See Also:

Constant Field Values

ANALYZER_STANDARD

public static final int ANALYZER_STANDARD

An identifier for the StandardAnalyzer used by the search engine.

See Also:

Constant Field Values

ANALYZER_TYPE

public static int ANALYZER_TYPE

The Analyzer type to be used by the search engine. The default The default value is ANALYZER_STANDARD.

m_writer

protected org.apache.lucene.index.IndexWriter m_writer

The IndexWriter used by the search engine.

m_analyzer

protected static org.apache.lucene.analysis.Analyzer m_analyzer

The Analyzer used by the search engine.

m_indexDirectory

protected File m_indexDirectory

The directory used by the search engine for storing its index.

THREAD_PRIORITY

public static int THREAD_PRIORITY

The default Thread priority (Thread.MAX_PRIORITY).

create

protected boolean create

When true, indexing operations will create a new index from scratch rather than modifying an existing index. Default is true.

progress

protected ProgressBar progress

srvs

protected Services srvs

mw

protected MessageWriter mw

mh

protected MessageHandler mh

Constructor Detail

Indexer

public Indexer(String name,
               MessageWriter msgwriter)

Constructor with a name identifier (used by the thread group manager), the MessageWriter to receive indexing and search result messages, and a File reference for the index directory.

Method Detail

index

public void index(boolean createIndex)

Begins an indexing process. If createIndex is true, any existing index will be overwritten. This causes the VM to call Thread's run() method, and should be called in lieu of calling run() directly.

run

public abstract void run()

Specified by:

run in interface Runnable

Overrides:

run in class Thread

getIndexReader

public org.apache.lucene.index.IndexReader getIndexReader()
                                                   throws IOException

Returns an IndexReader used to restore the index.

Throws:

IOException

setIsIndexed

protected final void setIsIndexed(boolean indexed)

Set the boolean value indicating an index completed/ready state. Setting this false sets the index version number to -1.

isIndexed

public final boolean isIndexed()

Returns true if Indexer indicates an index completed state. This should be set false upon starting an index operation, and true by the index Thread upon completion. Subclasses are responsible for setting this member variable.

getVersion

public final long getVersion()

Returns the index version number, -1 if it has not been set.

hasProgressBar

public final boolean hasProgressBar()

Returns true if the optional ProgressBar has been set.

setProgressBar

public final void setProgressBar(ProgressBar progressBar)

Sets an optional ProgressBar to indicate process status.

getProgressBar

public final ProgressBar getProgressBar()

Returns the optional ProgressBar, or null if it is not set.

getMessageWriter

public final MessageWriter getMessageWriter()

Returns the MessageWriter, which displays index- and search-related messages directly to the user.

getMessageHandler

public final MessageHandler getMessageHandler()

Returns the MessageHandler, which receives general informational, log, warning and error messages. This can also be accessed by referring directly to the member variable mh, as in mh.error("an error has occurred").

setIndexDirectory

public void setIndexDirectory(File directory)
                       throws IOException

Sets the directory used to store the index. This is an optional method, as the directory should probably be set via a property within the implementing class. Because the property may be obtained external to the implementing class, this method is provided. The method in this abstract class is empty so as not to force an implementation. Storage of the member variable must occur in the implementing class.

Throws:

IOException

getIndexDirectory

public File getIndexDirectory()
                       throws IOException

Returns the index directory. This method should be subclassed to return the index directory suitable for the index type.

Throws:

IOException

restoreIndex

public abstract boolean restoreIndex()

Restore the existing index after opening a new session, returning true if successful.

setCreateIndex

public final void setCreateIndex(boolean createIndex)

Sets the property that when true indicates that when index are built, they should be built from scratch rather than just modified.

isCreateIndex

public final boolean isCreateIndex()

Returns true if the index should be created/recreated when the index is built/rebuilt, rather than just modified.

getIndexWriter

public org.apache.lucene.index.IndexWriter getIndexWriter()

Return the IndexWriter used by this XNodeIndexer.

deleteIndex

public void deleteIndex()

Delete the built index of this Indexer.

mergeIndex

public void mergeIndex(Indexer indexer)

Merge the index provided by indexer with this index.

closeIndex

public void closeIndex()

Close any open index writer.

getAnalyzer

public org.apache.lucene.analysis.Analyzer getAnalyzer()

Return the Analyzer to be used for indexing content. This is a static, stateless object and can be reused. This method will default to returning a SimpleAnalyzer on unrecognized values.

getStopWords

public static String[] getStopWords()

Return the stop words being used by the analyzer. This is language-dependent, and for non-English languages must be set via the ResourceBundle (see MessageId.STOP_WORDS. For English or other recognized English-speaking Locales (ENGLISH, US, UK, and CANADA), this method returns null, since the constructor for the Analyzer using the default has no parameters.