[hibernate-commits] Hibernate SVN: r15547 - in search/trunk/src: java/org/hibernate/search/backend/impl/lucene and 7 other directories.
hibernate-commits at lists.jboss.org
hibernate-commits at lists.jboss.org
Tue Nov 11 07:57:47 EST 2008
Author: hardy.ferentschik
Date: 2008-11-11 07:57:47 -0500 (Tue, 11 Nov 2008)
New Revision: 15547
Modified:
search/trunk/src/java/org/hibernate/search/backend/Workspace.java
search/trunk/src/java/org/hibernate/search/backend/impl/lucene/PerDPQueueProcessor.java
search/trunk/src/java/org/hibernate/search/engine/DocumentBuilder.java
search/trunk/src/java/org/hibernate/search/store/DirectoryProviderHelper.java
search/trunk/src/java/org/hibernate/search/store/RAMDirectoryProvider.java
search/trunk/src/test/org/hibernate/search/test/analyzer/AbstractTestAnalyzer.java
search/trunk/src/test/org/hibernate/search/test/filter/BestDriversFilter.java
search/trunk/src/test/org/hibernate/search/test/filter/ExcludeAllFilter.java
search/trunk/src/test/org/hibernate/search/test/filter/InstanceBasedExcludeAllFilter.java
search/trunk/src/test/org/hibernate/search/test/jms/master/JMSMasterTest.java
search/trunk/src/test/org/hibernate/search/test/reader/performance/IndexFillRunnable.java
search/trunk/src/test/org/hibernate/search/test/reader/performance/ReaderPerformance.java
search/trunk/src/test/org/hibernate/search/test/util/AnalyzerUtils.java
Log:
HSEARCH-283
* Updated deprecated field constants
* changed IndexWriter constructor
Modified: search/trunk/src/java/org/hibernate/search/backend/Workspace.java
===================================================================
--- search/trunk/src/java/org/hibernate/search/backend/Workspace.java 2008-11-11 10:14:22 UTC (rev 15546)
+++ search/trunk/src/java/org/hibernate/search/backend/Workspace.java 2008-11-11 12:57:47 UTC (rev 15547)
@@ -25,11 +25,11 @@
/**
* Lucene workspace for a DirectoryProvider.<p/>
* <ul>
- * <li>Before using getIndexWriter or getIndexReader the lock must be acquired, and resources must be closed
- * before releasing the lock.</li>
+ * <li>Before using {@link #getIndexWriter} or {@link #getIndexReader} the lock must be acquired,
+ * and resources must be closed before releasing the lock.</li>
* <li>One cannot get an IndexWriter when an IndexReader has been acquired and not closed, and vice-versa.</li>
- * <li>The recommended approach is to execute all the modifications on the IndexReader, and after that on
- * the IndexWriter</li>
+ * <li>The recommended approach is to execute all the modifications on the <code>IndexReader</code>, and after that on
+ * the <code>IndexWriter</code></li>.
* </ul>
*
* @author Emmanuel Bernard
@@ -173,8 +173,8 @@
return writer;
try {
// don't care about the Analyzer as it will be selected during usage of IndexWriter.
- //FIXME use the non deprecated constructor => requires to call #Commit()
- writer = new IndexWriter( directoryProvider.getDirectory(), SIMPLE_ANALYZER, false ); // has been created at init time
+ IndexWriter.MaxFieldLength fieldLength = new IndexWriter.MaxFieldLength( IndexWriter.DEFAULT_MAX_FIELD_LENGTH );
+ writer = new IndexWriter( directoryProvider.getDirectory(), SIMPLE_ANALYZER, false, fieldLength ); // has been created at init time
indexingParams.applyToWriter( writer, batchmode );
log.trace( "IndexWriter opened" );
}
@@ -186,6 +186,30 @@
}
/**
+ * Commits changes to a previously opened index writer.
+ *
+ * @throws SearchException on IOException during Lucene close operation.
+ * @throws AssertionFailure if there is no IndexWriter to close, or if the lock is not owned.
+ */
+ public synchronized void commitIndexWriter() {
+ assertOwnLock();
+ if ( writer != null ) {
+ try {
+ writer.commit();
+ log.trace( "Index changes commited." );
+ }
+ catch ( IOException e ) {
+ throw new SearchException( "Exception while commiting index changes", e );
+ }
+ }
+ else {
+ throw new AssertionFailure( "No open IndexWriter to commit changes." );
+ }
+ }
+
+
+
+ /**
* Closes a previously opened IndexWriter.
* @throws SearchException on IOException during Lucene close operation.
* @throws AssertionFailure if there is no IndexWriter to close, or if the lock is not owned.
Modified: search/trunk/src/java/org/hibernate/search/backend/impl/lucene/PerDPQueueProcessor.java
===================================================================
--- search/trunk/src/java/org/hibernate/search/backend/impl/lucene/PerDPQueueProcessor.java 2008-11-11 10:14:22 UTC (rev 15546)
+++ search/trunk/src/java/org/hibernate/search/backend/impl/lucene/PerDPQueueProcessor.java 2008-11-11 12:57:47 UTC (rev 15547)
@@ -103,6 +103,7 @@
for (LuceneWork lw : workOnWriter) {
lw.getWorkDelegate( worker ).performWork( lw, indexWriter );
}
+ workspace.commitIndexWriter();
//TODO next line is assuming the OptimizerStrategy will need an IndexWriter;
// would be nicer to have the strategy put an OptimizeWork on the queue,
// or just return "yes please" (true) to some method?
Modified: search/trunk/src/java/org/hibernate/search/engine/DocumentBuilder.java
===================================================================
--- search/trunk/src/java/org/hibernate/search/engine/DocumentBuilder.java 2008-11-11 10:14:22 UTC (rev 15546)
+++ search/trunk/src/java/org/hibernate/search/engine/DocumentBuilder.java 2008-11-11 12:57:47 UTC (rev 15547)
@@ -583,11 +583,11 @@
case NO:
return Field.Index.NO;
case NO_NORMS:
- return Field.Index.NO_NORMS;
+ return Field.Index.NOT_ANALYZED_NO_NORMS;
case TOKENIZED:
- return Field.Index.TOKENIZED;
+ return Field.Index.ANALYZED;
case UN_TOKENIZED:
- return Field.Index.UN_TOKENIZED;
+ return Field.Index.NOT_ANALYZED;
default:
throw new AssertionFailure( "Unexpected Index: " + index );
}
@@ -750,10 +750,10 @@
}
{
Field classField =
- new Field( CLASS_FIELDNAME, entityType.getName(), Field.Store.YES, Field.Index.UN_TOKENIZED, Field.TermVector.NO );
+ new Field( CLASS_FIELDNAME, entityType.getName(), Field.Store.YES, Field.Index.NOT_ANALYZED, Field.TermVector.NO );
doc.add( classField );
LuceneOptions luceneOptions = new LuceneOptionsImpl( Field.Store.YES,
- Field.Index.UN_TOKENIZED, Field.TermVector.NO, idBoost );
+ Field.Index.NOT_ANALYZED, Field.TermVector.NO, idBoost );
idBridge.set( idKeywordName, id, doc, luceneOptions );
}
buildDocumentFields( instance, doc, rootPropertiesMetadata );
Modified: search/trunk/src/java/org/hibernate/search/store/DirectoryProviderHelper.java
===================================================================
--- search/trunk/src/java/org/hibernate/search/store/DirectoryProviderHelper.java 2008-11-11 10:14:22 UTC (rev 15546)
+++ search/trunk/src/java/org/hibernate/search/store/DirectoryProviderHelper.java 2008-11-11 12:57:47 UTC (rev 15547)
@@ -77,7 +77,8 @@
FSDirectory fsDirectory = FSDirectory.getDirectory( indexDir );
if ( ! IndexReader.indexExists( fsDirectory ) ) {
log.debug( "Initialize index: '{}'", indexDir.getAbsolutePath() );
- IndexWriter iw = new IndexWriter( fsDirectory, new StandardAnalyzer(), true );
+ IndexWriter.MaxFieldLength fieldLength = new IndexWriter.MaxFieldLength( IndexWriter.DEFAULT_MAX_FIELD_LENGTH );
+ IndexWriter iw = new IndexWriter( fsDirectory, new StandardAnalyzer(), true, fieldLength );
iw.close();
}
return fsDirectory;
Modified: search/trunk/src/java/org/hibernate/search/store/RAMDirectoryProvider.java
===================================================================
--- search/trunk/src/java/org/hibernate/search/store/RAMDirectoryProvider.java 2008-11-11 10:14:22 UTC (rev 15546)
+++ search/trunk/src/java/org/hibernate/search/store/RAMDirectoryProvider.java 2008-11-11 12:57:47 UTC (rev 15547)
@@ -28,7 +28,8 @@
public void start() {
directory = new RAMDirectory();
try {
- IndexWriter iw = new IndexWriter( directory, new StandardAnalyzer(), true );
+ IndexWriter.MaxFieldLength fieldLength = new IndexWriter.MaxFieldLength( IndexWriter.DEFAULT_MAX_FIELD_LENGTH );
+ IndexWriter iw = new IndexWriter( directory, new StandardAnalyzer(), true, fieldLength );
iw.close();
}
catch (IOException e) {
Modified: search/trunk/src/test/org/hibernate/search/test/analyzer/AbstractTestAnalyzer.java
===================================================================
--- search/trunk/src/test/org/hibernate/search/test/analyzer/AbstractTestAnalyzer.java 2008-11-11 10:14:22 UTC (rev 15546)
+++ search/trunk/src/test/org/hibernate/search/test/analyzer/AbstractTestAnalyzer.java 2008-11-11 12:57:47 UTC (rev 15547)
@@ -23,12 +23,17 @@
private class InternalTokenStream extends TokenStream {
private int position;
- public Token next() throws IOException {
- if ( position >= getTokens().length) {
+ public Token next(final Token reusableToken) throws IOException {
+ assert reusableToken != null;
+ if ( position >= getTokens().length ) {
return null;
}
else {
- return new Token( getTokens()[position++], 0, 0 );
+ reusableToken.clear();
+ reusableToken.setTermBuffer( getTokens()[position++] );
+ reusableToken.setStartOffset( 0 );
+ reusableToken.setEndOffset( 0 );
+ return reusableToken;
}
}
}
Modified: search/trunk/src/test/org/hibernate/search/test/filter/BestDriversFilter.java
===================================================================
--- search/trunk/src/test/org/hibernate/search/test/filter/BestDriversFilter.java 2008-11-11 10:14:22 UTC (rev 15546)
+++ search/trunk/src/test/org/hibernate/search/test/filter/BestDriversFilter.java 2008-11-11 12:57:47 UTC (rev 15547)
@@ -1,22 +1,23 @@
//$Id$
package org.hibernate.search.test.filter;
-import java.util.BitSet;
import java.io.IOException;
-import org.apache.lucene.search.Filter;
import org.apache.lucene.index.IndexReader;
+import org.apache.lucene.index.Term;
import org.apache.lucene.index.TermDocs;
-import org.apache.lucene.index.Term;
+import org.apache.lucene.search.DocIdSet;
+import org.apache.lucene.search.Filter;
+import org.apache.lucene.util.OpenBitSet;
/**
* @author Emmanuel Bernard
*/
public class BestDriversFilter extends Filter {
- public BitSet bits(IndexReader reader) throws IOException {
- BitSet bitSet = new BitSet( reader.maxDoc() );
- TermDocs termDocs = reader.termDocs( new Term("score", "5") );
+ public DocIdSet getDocIdSet(IndexReader reader) throws IOException {
+ OpenBitSet bitSet = new OpenBitSet( reader.maxDoc() );
+ TermDocs termDocs = reader.termDocs( new Term( "score", "5" ) );
while ( termDocs.next() ) {
bitSet.set( termDocs.doc() );
}
Modified: search/trunk/src/test/org/hibernate/search/test/filter/ExcludeAllFilter.java
===================================================================
--- search/trunk/src/test/org/hibernate/search/test/filter/ExcludeAllFilter.java 2008-11-11 10:14:22 UTC (rev 15546)
+++ search/trunk/src/test/org/hibernate/search/test/filter/ExcludeAllFilter.java 2008-11-11 12:57:47 UTC (rev 15547)
@@ -7,6 +7,7 @@
import org.apache.lucene.search.DocIdSet;
import org.apache.lucene.search.Filter;
import org.apache.lucene.index.IndexReader;
+
import org.hibernate.search.filter.EmptyDocIdBitSet;
/**
@@ -14,13 +15,15 @@
*/
@SuppressWarnings("serial")
public class ExcludeAllFilter extends Filter {
-
- //ugly but useful for test purposes
+
+ // ugly but useful for test purposes
private static volatile boolean done = false;
@Override
public BitSet bits(IndexReader reader) throws IOException {
- if ( done ) throw new IllegalStateException( "Called twice" );
+ if ( done ) {
+ throw new IllegalStateException( "Called twice" );
+ }
BitSet bitSet = new BitSet( reader.maxDoc() );
done = true;
return bitSet;
@@ -28,9 +31,10 @@
@Override
public DocIdSet getDocIdSet(IndexReader reader) throws IOException {
- if ( done ) throw new IllegalStateException( "Called twice" );
+ if ( done ) {
+ throw new IllegalStateException( "Called twice" );
+ }
done = true;
return EmptyDocIdBitSet.instance;
}
-
}
Modified: search/trunk/src/test/org/hibernate/search/test/filter/InstanceBasedExcludeAllFilter.java
===================================================================
--- search/trunk/src/test/org/hibernate/search/test/filter/InstanceBasedExcludeAllFilter.java 2008-11-11 10:14:22 UTC (rev 15546)
+++ search/trunk/src/test/org/hibernate/search/test/filter/InstanceBasedExcludeAllFilter.java 2008-11-11 12:57:47 UTC (rev 15547)
@@ -5,17 +5,32 @@
import org.apache.lucene.index.IndexReader;
import org.apache.lucene.search.Filter;
+import org.apache.lucene.search.DocIdSet;
+import org.hibernate.search.filter.EmptyDocIdBitSet;
+
/**
* @author Emmanuel Bernard
*/
public class InstanceBasedExcludeAllFilter extends Filter {
private volatile boolean done = false;
+ @Override
public BitSet bits(IndexReader reader) throws IOException {
- if (done) throw new IllegalStateException("Called twice");
+ if ( done ) {
+ throw new IllegalStateException( "Called twice" );
+ }
BitSet bitSet = new BitSet( reader.maxDoc() );
done = true;
return bitSet;
}
+
+ @Override
+ public DocIdSet getDocIdSet(IndexReader reader) throws IOException {
+ if ( done ) {
+ throw new IllegalStateException( "Called twice" );
+ }
+ done = true;
+ return EmptyDocIdBitSet.instance;
+ }
}
Modified: search/trunk/src/test/org/hibernate/search/test/jms/master/JMSMasterTest.java
===================================================================
--- search/trunk/src/test/org/hibernate/search/test/jms/master/JMSMasterTest.java 2008-11-11 10:14:22 UTC (rev 15546)
+++ search/trunk/src/test/org/hibernate/search/test/jms/master/JMSMasterTest.java 2008-11-11 12:57:47 UTC (rev 15547)
@@ -54,11 +54,11 @@
s.close();
//create the work queue to send
Document doc = new Document();
- Field field = new Field( DocumentBuilder.CLASS_FIELDNAME, ts.getClass().getName(), Field.Store.YES, Field.Index.UN_TOKENIZED );
+ Field field = new Field( DocumentBuilder.CLASS_FIELDNAME, ts.getClass().getName(), Field.Store.YES, Field.Index.NOT_ANALYZED );
doc.add( field );
- field = new Field("id", "1", Field.Store.YES, Field.Index.UN_TOKENIZED );
+ field = new Field("id", "1", Field.Store.YES, Field.Index.NOT_ANALYZED );
doc.add( field );
- field = new Field("logo", ts.getLogo(), Field.Store.NO, Field.Index.TOKENIZED );
+ field = new Field("logo", ts.getLogo(), Field.Store.NO, Field.Index.ANALYZED );
doc.add( field );
LuceneWork luceneWork = new AddLuceneWork(ts.getId(), String.valueOf( ts.getId() ), ts.getClass(), doc );
List<LuceneWork> queue = new ArrayList<LuceneWork>();
Modified: search/trunk/src/test/org/hibernate/search/test/reader/performance/IndexFillRunnable.java
===================================================================
--- search/trunk/src/test/org/hibernate/search/test/reader/performance/IndexFillRunnable.java 2008-11-11 10:14:22 UTC (rev 15546)
+++ search/trunk/src/test/org/hibernate/search/test/reader/performance/IndexFillRunnable.java 2008-11-11 12:57:47 UTC (rev 15547)
@@ -16,23 +16,29 @@
private volatile int jobSeed = 0;
private final IndexWriter iw;
-
+
public IndexFillRunnable(IndexWriter iw) {
super();
this.iw = iw;
}
public void run() {
- Field f1 = new Field("name", "Some One " + jobSeed++, Store.NO, Index.TOKENIZED );
- Field f2 = new Field("physicalDescription", " just more people sitting around and filling my index... ", Store.NO, Index.TOKENIZED );
+ Field f1 = new Field( "name", "Some One " + jobSeed++, Store.NO, Index.ANALYZED );
+ Field f2 = new Field(
+ "physicalDescription",
+ " just more people sitting around and filling my index... ",
+ Store.NO,
+ Index.ANALYZED
+ );
Document d = new Document();
d.add( f1 );
d.add( f2 );
try {
iw.addDocument( d );
- } catch (IOException e) {
+ }
+ catch ( IOException e ) {
e.printStackTrace();
}
}
-
+
}
Modified: search/trunk/src/test/org/hibernate/search/test/reader/performance/ReaderPerformance.java
===================================================================
--- search/trunk/src/test/org/hibernate/search/test/reader/performance/ReaderPerformance.java 2008-11-11 10:14:22 UTC (rev 15546)
+++ search/trunk/src/test/org/hibernate/search/test/reader/performance/ReaderPerformance.java 2008-11-11 12:57:47 UTC (rev 15547)
@@ -55,7 +55,8 @@
private void buildBigIndex() throws InterruptedException, CorruptIndexException, LockObtainFailedException, IOException {
System.out.println( "Going to create fake index..." );
FSDirectory directory = FSDirectory.getDirectory(new File(getBaseIndexDir(), Detective.class.getCanonicalName()));
- IndexWriter iw = new IndexWriter( directory, new SimpleAnalyzer(), true );
+ IndexWriter.MaxFieldLength fieldLength = new IndexWriter.MaxFieldLength( IndexWriter.DEFAULT_MAX_FIELD_LENGTH );
+ IndexWriter iw = new IndexWriter( directory, new SimpleAnalyzer(), true, fieldLength );
IndexFillRunnable filler = new IndexFillRunnable( iw );
ThreadPoolExecutor executor = (ThreadPoolExecutor) Executors.newFixedThreadPool( WORKER_THREADS );
for (int batch=0; batch<=5000000; batch++){
@@ -63,6 +64,7 @@
}
executor.shutdown();
executor.awaitTermination( 600, TimeUnit.SECONDS );
+ iw.commit();
iw.optimize();
iw.close();
System.out.println( "Index created." );
Modified: search/trunk/src/test/org/hibernate/search/test/util/AnalyzerUtils.java
===================================================================
--- search/trunk/src/test/org/hibernate/search/test/util/AnalyzerUtils.java 2008-11-11 10:14:22 UTC (rev 15546)
+++ search/trunk/src/test/org/hibernate/search/test/util/AnalyzerUtils.java 2008-11-11 12:57:47 UTC (rev 15547)
@@ -26,23 +26,24 @@
public static Token[] tokensFromAnalysis(Analyzer analyzer, String field, String text) throws IOException {
TokenStream stream = analyzer.tokenStream( field, new StringReader( text ) );
List<Token> tokenList = new ArrayList<Token>();
+ Token reusableToken = new Token();
while ( true ) {
- Token token = stream.next();
+
+ Token token = stream.next( reusableToken );
if ( token == null ) {
break;
}
- tokenList.add( token );
+ tokenList.add( ( Token ) token.clone() );
}
- return ( Token[] ) tokenList.toArray( new Token[0] );
+ return tokenList.toArray( new Token[tokenList.size()] );
}
public static void displayTokens(Analyzer analyzer, String field, String text) throws IOException {
Token[] tokens = tokensFromAnalysis( analyzer, field, text );
- for ( int i = 0; i < tokens.length; i++ ) {
- Token token = tokens[i];
+ for ( Token token : tokens ) {
log.debug( "[" + getTermText( token ) + "] " );
}
}
@@ -52,9 +53,7 @@
int position = 0;
- for ( int i = 0; i < tokens.length; i++ ) {
- Token token = tokens[i];
-
+ for ( Token token : tokens ) {
int increment = token.getPositionIncrement();
if ( increment > 0 ) {
@@ -72,9 +71,7 @@
StringBuilder builder = new StringBuilder();
int position = 0;
- for ( int i = 0; i < tokens.length; i++ ) {
- Token token = tokens[i];
-
+ for ( Token token : tokens ) {
int increment = token.getPositionIncrement();
if ( increment > 0 ) {
More information about the hibernate-commits
mailing list