goller 2004/04/16 02:17:05
Modified: src/test/org/apache/lucene/index TestMultiReader.java
TestIndexReader.java TestFilterIndexReader.java
src/java/org/apache/lucene/index MultiReader.java
FilterIndexReader.java IndexReader.java
SegmentReader.java
Log:
restructuring of IndexReader, SegmentReader,
and Multireader in order to improve directory locking
Revision Changes Path
1.3 +98 -98
jakarta-lucene/src/test/org/apache/lucene/index/TestMultiReader.java
Index: TestMultiReader.java
===================================================================
RCS file:
/home/cvs/jakarta-lucene/src/test/org/apache/lucene/index/TestMultiReader.java,v
retrieving revision 1.2
retrieving revision 1.3
diff -u -r1.2 -r1.3
--- TestMultiReader.java 29 Mar 2004 22:48:06 -0000 1.2
+++ TestMultiReader.java 16 Apr 2004 09:17:05 -0000 1.3
@@ -1,98 +1,98 @@
-package org.apache.lucene.index;
-
-/**
- * Copyright 2004 The Apache Software Foundation
- *
- * Licensed under the Apache License, Version 2.0 (the "License");
- * you may not use this file except in compliance with the License.
- * You may obtain a copy of the License at
- *
- * http://www.apache.org/licenses/LICENSE-2.0
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-import junit.framework.TestCase;
-import org.apache.lucene.document.Document;
-import org.apache.lucene.store.Directory;
-import org.apache.lucene.store.RAMDirectory;
-
-import java.io.IOException;
-
-public class TestMultiReader extends TestCase {
- private Directory dir = new RAMDirectory();
- private Document doc1 = new Document();
- private Document doc2 = new Document();
- private SegmentReader reader1;
- private SegmentReader reader2;
- private SegmentReader [] readers = new SegmentReader[2];
- private SegmentInfos sis = new SegmentInfos();
-
- public TestMultiReader(String s) {
- super(s);
- }
-
- protected void setUp() {
- DocHelper.setupDoc(doc1);
- DocHelper.setupDoc(doc2);
- DocHelper.writeDoc(dir, "seg-1", doc1);
- DocHelper.writeDoc(dir, "seg-2", doc2);
-
- try {
- sis.write(dir);
- reader1 = new SegmentReader(new SegmentInfo("seg-1", 1, dir));
- reader2 = new SegmentReader(new SegmentInfo("seg-2", 1, dir));
- readers[0] = reader1;
- readers[1] = reader2;
- } catch (IOException e) {
- e.printStackTrace();
- }
- }
-/*IndexWriter writer = new IndexWriter(dir, new WhitespaceAnalyzer(), true);
- writer.addDocument(doc1);
- writer.addDocument(doc2);
- writer.close();*/
- protected void tearDown() {
-
- }
-
- public void test() {
- assertTrue(dir != null);
- assertTrue(reader1 != null);
- assertTrue(reader2 != null);
- assertTrue(sis != null);
- }
-
- public void testDocument() {
- try {
- sis.read(dir);
- MultiReader reader = new MultiReader(dir, readers);
- assertTrue(reader != null);
- Document newDoc1 = reader.document(0);
- assertTrue(newDoc1 != null);
- assertTrue(DocHelper.numFields(newDoc1) == DocHelper.numFields(doc1) - 2);
- Document newDoc2 = reader.document(1);
- assertTrue(newDoc2 != null);
- assertTrue(DocHelper.numFields(newDoc2) == DocHelper.numFields(doc2) - 2);
- TermFreqVector vector = reader.getTermFreqVector(0,
DocHelper.TEXT_FIELD_2_KEY);
- assertTrue(vector != null);
- } catch (IOException e) {
- e.printStackTrace();
- assertTrue(false);
- }
- }
-
- public void testTermVectors() {
- try {
- MultiReader reader = new MultiReader(dir, readers);
- assertTrue(reader != null);
- } catch (IOException e) {
- e.printStackTrace();
- assertTrue(false);
- }
- }
-}
+package org.apache.lucene.index;
+
+/**
+ * Copyright 2004 The Apache Software Foundation
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import junit.framework.TestCase;
+import org.apache.lucene.document.Document;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.store.RAMDirectory;
+
+import java.io.IOException;
+
+public class TestMultiReader extends TestCase {
+ private Directory dir = new RAMDirectory();
+ private Document doc1 = new Document();
+ private Document doc2 = new Document();
+ private SegmentReader reader1;
+ private SegmentReader reader2;
+ private SegmentReader [] readers = new SegmentReader[2];
+ private SegmentInfos sis = new SegmentInfos();
+
+ public TestMultiReader(String s) {
+ super(s);
+ }
+
+ protected void setUp() {
+ DocHelper.setupDoc(doc1);
+ DocHelper.setupDoc(doc2);
+ DocHelper.writeDoc(dir, "seg-1", doc1);
+ DocHelper.writeDoc(dir, "seg-2", doc2);
+
+ try {
+ sis.write(dir);
+ reader1 = new SegmentReader(new SegmentInfo("seg-1", 1, dir));
+ reader2 = new SegmentReader(new SegmentInfo("seg-2", 1, dir));
+ readers[0] = reader1;
+ readers[1] = reader2;
+ } catch (IOException e) {
+ e.printStackTrace();
+ }
+ }
+/*IndexWriter writer = new IndexWriter(dir, new WhitespaceAnalyzer(), true);
+ writer.addDocument(doc1);
+ writer.addDocument(doc2);
+ writer.close();*/
+ protected void tearDown() {
+
+ }
+
+ public void test() {
+ assertTrue(dir != null);
+ assertTrue(reader1 != null);
+ assertTrue(reader2 != null);
+ assertTrue(sis != null);
+ }
+
+ public void testDocument() {
+ try {
+ sis.read(dir);
+ MultiReader reader = new MultiReader(dir, sis, false, readers);
+ assertTrue(reader != null);
+ Document newDoc1 = reader.document(0);
+ assertTrue(newDoc1 != null);
+ assertTrue(DocHelper.numFields(newDoc1) == DocHelper.numFields(doc1) - 2);
+ Document newDoc2 = reader.document(1);
+ assertTrue(newDoc2 != null);
+ assertTrue(DocHelper.numFields(newDoc2) == DocHelper.numFields(doc2) - 2);
+ TermFreqVector vector = reader.getTermFreqVector(0,
DocHelper.TEXT_FIELD_2_KEY);
+ assertTrue(vector != null);
+ } catch (IOException e) {
+ e.printStackTrace();
+ assertTrue(false);
+ }
+ }
+
+ public void testTermVectors() {
+ try {
+ MultiReader reader = new MultiReader(dir, sis, false, readers);
+ assertTrue(reader != null);
+ } catch (IOException e) {
+ e.printStackTrace();
+ assertTrue(false);
+ }
+ }
+}
1.7 +21 -5
jakarta-lucene/src/test/org/apache/lucene/index/TestIndexReader.java
Index: TestIndexReader.java
===================================================================
RCS file:
/home/cvs/jakarta-lucene/src/test/org/apache/lucene/index/TestIndexReader.java,v
retrieving revision 1.6
retrieving revision 1.7
diff -u -r1.6 -r1.7
--- TestIndexReader.java 29 Mar 2004 22:48:06 -0000 1.6
+++ TestIndexReader.java 16 Apr 2004 09:17:05 -0000 1.7
@@ -173,8 +173,15 @@
}
+ public void testDeleteReaderWriterConflictUnoptimized() throws IOException{
+ deleteReaderWriterConflict(false);
+ }
+
+ public void testDeleteReaderWriterConflictOptimized() throws IOException{
+ deleteReaderWriterConflict(true);
+ }
- public void testDeleteReaderWriterConflict() throws IOException
+ private void deleteReaderWriterConflict(boolean optimize) throws IOException
{
//Directory dir = new RAMDirectory();
Directory dir = getDirectory(true);
@@ -210,7 +217,8 @@
// searchers. Because of this, deletions made via a previously open
// reader, which would be applied to that reader's segment, are lost
// for subsequent searchers/readers
- writer.optimize();
+ if(optimize)
+ writer.optimize();
writer.close();
// The reader should not see the new data
@@ -288,8 +296,15 @@
dir = getDirectory(true);
}
-
- public void testDeleteReaderReaderConflict() throws IOException
+ public void testDeleteReaderReaderConflictUnoptimized() throws IOException{
+ deleteReaderReaderConflict(false);
+ }
+
+ public void testDeleteReaderReaderConflictOptimized() throws IOException{
+ deleteReaderReaderConflict(true);
+ }
+
+ private void deleteReaderReaderConflict(boolean optimize) throws IOException
{
Directory dir = getDirectory(true);
@@ -307,7 +322,8 @@
addDoc(writer, searchTerm2.text());
addDoc(writer, searchTerm3.text());
}
- writer.optimize();
+ if(optimize)
+ writer.optimize();
writer.close();
// OPEN TWO READERS
1.4 +2 -2
jakarta-lucene/src/test/org/apache/lucene/index/TestFilterIndexReader.java
Index: TestFilterIndexReader.java
===================================================================
RCS file:
/home/cvs/jakarta-lucene/src/test/org/apache/lucene/index/TestFilterIndexReader.java,v
retrieving revision 1.3
retrieving revision 1.4
diff -u -r1.3 -r1.4
--- TestFilterIndexReader.java 29 Mar 2004 22:48:06 -0000 1.3
+++ TestFilterIndexReader.java 16 Apr 2004 09:17:05 -0000 1.4
@@ -81,12 +81,12 @@
/** Filter terms with TestTermEnum. */
public TermEnum terms() throws IOException {
- return new TestTermEnum(in.terms());
+ return new TestTermEnum(baseReader.terms());
}
/** Filter positions with TestTermPositions. */
public TermPositions termPositions() throws IOException {
- return new TestTermPositions(in.termPositions());
+ return new TestTermPositions(baseReader.termPositions());
}
}
1.4 +63 -46 jakarta-lucene/src/java/org/apache/lucene/index/MultiReader.java
Index: MultiReader.java
===================================================================
RCS file:
/home/cvs/jakarta-lucene/src/java/org/apache/lucene/index/MultiReader.java,v
retrieving revision 1.3
retrieving revision 1.4
diff -u -r1.3 -r1.4
--- MultiReader.java 29 Mar 2004 22:48:02 -0000 1.3
+++ MultiReader.java 16 Apr 2004 09:17:05 -0000 1.4
@@ -31,32 +31,44 @@
* @version $Id$
*/
public class MultiReader extends IndexReader {
- private IndexReader[] readers;
+ private IndexReader[] subReaders;
private int[] starts; // 1st docno for each segment
private Hashtable normsCache = new Hashtable();
private int maxDoc = 0;
private int numDocs = -1;
private boolean hasDeletions = false;
- /** Construct reading the named set of readers. */
- public MultiReader(IndexReader[] readers) throws IOException {
- this(readers.length == 0 ? null : readers[0].directory(), readers);
+ /**
+ * <p>Construct a MultiReader aggregating the named set of (sub)readers.
+ * Directory locking for delete, undeleteAll, and setNorm operations is
+ * left to the subreaders. </p>
+ * <p>Note that all subreaders are closed if this Multireader is closed.</p>
+ * @param readers set of (sub)readers
+ * @throws IOException
+ */
+ public MultiReader(IndexReader[] subReaders) throws IOException {
+ super(subReaders.length == 0 ? null : subReaders[0].directory());
+ initialize(subReaders);
}
/** Construct reading the named set of readers. */
- public MultiReader(Directory directory, IndexReader[] readers)
+ MultiReader(Directory directory, SegmentInfos sis, boolean closeDirectory,
IndexReader[] subReaders)
throws IOException {
- super(directory);
- this.readers = readers;
- starts = new int[readers.length + 1]; // build starts array
- for (int i = 0; i < readers.length; i++) {
+ super(directory, sis, closeDirectory);
+ initialize(subReaders);
+ }
+
+ private void initialize(IndexReader[] subReaders) throws IOException{
+ this.subReaders = subReaders;
+ starts = new int[subReaders.length + 1]; // build starts array
+ for (int i = 0; i < subReaders.length; i++) {
starts[i] = maxDoc;
- maxDoc += readers[i].maxDoc(); // compute maxDocs
+ maxDoc += subReaders[i].maxDoc(); // compute maxDocs
- if (readers[i].hasDeletions())
+ if (subReaders[i].hasDeletions())
hasDeletions = true;
}
- starts[readers.length] = maxDoc;
+ starts[subReaders.length] = maxDoc;
}
@@ -69,20 +81,20 @@
public TermFreqVector[] getTermFreqVectors(int n)
throws IOException {
int i = readerIndex(n); // find segment num
- return readers[i].getTermFreqVectors(n - starts[i]); // dispatch to segment
+ return subReaders[i].getTermFreqVectors(n - starts[i]); // dispatch to segment
}
public TermFreqVector getTermFreqVector(int n, String field)
throws IOException {
int i = readerIndex(n); // find segment num
- return readers[i].getTermFreqVector(n - starts[i], field);
+ return subReaders[i].getTermFreqVector(n - starts[i], field);
}
public synchronized int numDocs() {
if (numDocs == -1) { // check cache
int n = 0; // cache miss--recompute
- for (int i = 0; i < readers.length; i++)
- n += readers[i].numDocs(); // sum from readers
+ for (int i = 0; i < subReaders.length; i++)
+ n += subReaders[i].numDocs(); // sum from readers
numDocs = n;
}
return numDocs;
@@ -94,32 +106,32 @@
public Document document(int n) throws IOException {
int i = readerIndex(n); // find segment num
- return readers[i].document(n - starts[i]); // dispatch to segment reader
+ return subReaders[i].document(n - starts[i]); // dispatch to segment reader
}
public boolean isDeleted(int n) {
int i = readerIndex(n); // find segment num
- return readers[i].isDeleted(n - starts[i]); // dispatch to segment reader
+ return subReaders[i].isDeleted(n - starts[i]); // dispatch to segment reader
}
public boolean hasDeletions() { return hasDeletions; }
- protected synchronized void doDelete(int n) throws IOException {
+ protected void doDelete(int n) throws IOException {
numDocs = -1; // invalidate cache
int i = readerIndex(n); // find segment num
- readers[i].doDelete(n - starts[i]); // dispatch to segment reader
+ subReaders[i].delete(n - starts[i]); // dispatch to segment reader
hasDeletions = true;
}
- public void undeleteAll() throws IOException {
- for (int i = 0; i < readers.length; i++)
- readers[i].undeleteAll();
+ protected void doUndeleteAll() throws IOException {
+ for (int i = 0; i < subReaders.length; i++)
+ subReaders[i].undeleteAll();
hasDeletions = false;
}
private int readerIndex(int n) { // find reader for doc n:
int lo = 0; // search starts array
- int hi = readers.length - 1; // for first element less
+ int hi = subReaders.length - 1; // for first element less
while (hi >= lo) {
int mid = (lo + hi) >> 1;
@@ -129,7 +141,7 @@
else if (n > midValue)
lo = mid + 1;
else { // found a match
- while (mid+1 < readers.length && starts[mid+1] == midValue) {
+ while (mid+1 < subReaders.length && starts[mid+1] == midValue) {
mid++; // scan to last match
}
return mid;
@@ -144,8 +156,8 @@
return bytes; // cache hit
bytes = new byte[maxDoc()];
- for (int i = 0; i < readers.length; i++)
- readers[i].norms(field, bytes, starts[i]);
+ for (int i = 0; i < subReaders.length; i++)
+ subReaders[i].norms(field, bytes, starts[i]);
normsCache.put(field, bytes); // update cache
return bytes;
}
@@ -156,43 +168,48 @@
if (bytes != null) // cache hit
System.arraycopy(bytes, 0, result, offset, maxDoc());
- for (int i = 0; i < readers.length; i++) // read from segments
- readers[i].norms(field, result, offset + starts[i]);
+ for (int i = 0; i < subReaders.length; i++) // read from segments
+ subReaders[i].norms(field, result, offset + starts[i]);
}
- public synchronized void setNorm(int n, String field, byte value)
+ protected void doSetNorm(int n, String field, byte value)
throws IOException {
normsCache.remove(field); // clear cache
int i = readerIndex(n); // find segment num
- readers[i].setNorm(n-starts[i], field, value); // dispatch
+ subReaders[i].setNorm(n-starts[i], field, value); // dispatch
}
public TermEnum terms() throws IOException {
- return new MultiTermEnum(readers, starts, null);
+ return new MultiTermEnum(subReaders, starts, null);
}
public TermEnum terms(Term term) throws IOException {
- return new MultiTermEnum(readers, starts, term);
+ return new MultiTermEnum(subReaders, starts, term);
}
public int docFreq(Term t) throws IOException {
int total = 0; // sum freqs in segments
- for (int i = 0; i < readers.length; i++)
- total += readers[i].docFreq(t);
+ for (int i = 0; i < subReaders.length; i++)
+ total += subReaders[i].docFreq(t);
return total;
}
public TermDocs termDocs() throws IOException {
- return new MultiTermDocs(readers, starts);
+ return new MultiTermDocs(subReaders, starts);
}
public TermPositions termPositions() throws IOException {
- return new MultiTermPositions(readers, starts);
+ return new MultiTermPositions(subReaders, starts);
+ }
+
+ protected void doCommit() throws IOException {
+ for (int i = 0; i < subReaders.length; i++)
+ subReaders[i].commit();
}
protected synchronized void doClose() throws IOException {
- for (int i = 0; i < readers.length; i++)
- readers[i].close();
+ for (int i = 0; i < subReaders.length; i++)
+ subReaders[i].close();
}
/**
@@ -201,8 +218,8 @@
public Collection getFieldNames() throws IOException {
// maintain a unique set of field names
Set fieldSet = new HashSet();
- for (int i = 0; i < readers.length; i++) {
- IndexReader reader = readers[i];
+ for (int i = 0; i < subReaders.length; i++) {
+ IndexReader reader = subReaders[i];
Collection names = reader.getFieldNames();
// iterate through the field names and add them to the set
for (Iterator iterator = names.iterator(); iterator.hasNext();) {
@@ -219,8 +236,8 @@
public Collection getFieldNames(boolean indexed) throws IOException {
// maintain a unique set of field names
Set fieldSet = new HashSet();
- for (int i = 0; i < readers.length; i++) {
- IndexReader reader = readers[i];
+ for (int i = 0; i < subReaders.length; i++) {
+ IndexReader reader = subReaders[i];
Collection names = reader.getFieldNames(indexed);
fieldSet.addAll(names);
}
@@ -230,8 +247,8 @@
public Collection getIndexedFieldNames(boolean storedTermVector) {
// maintain a unique set of field names
Set fieldSet = new HashSet();
- for (int i = 0; i < readers.length; i++) {
- IndexReader reader = readers[i];
+ for (int i = 0; i < subReaders.length; i++) {
+ IndexReader reader = subReaders[i];
Collection names = reader.getIndexedFieldNames(storedTermVector);
fieldSet.addAll(names);
}
1.9 +34 -27
jakarta-lucene/src/java/org/apache/lucene/index/FilterIndexReader.java
Index: FilterIndexReader.java
===================================================================
RCS file:
/home/cvs/jakarta-lucene/src/java/org/apache/lucene/index/FilterIndexReader.java,v
retrieving revision 1.8
retrieving revision 1.9
diff -u -r1.8 -r1.9
--- FilterIndexReader.java 29 Mar 2004 22:48:02 -0000 1.8
+++ FilterIndexReader.java 16 Apr 2004 09:17:05 -0000 1.9
@@ -73,61 +73,68 @@
public void close() throws IOException { in.close(); }
}
- protected IndexReader in;
+ protected IndexReader baseReader;
- public FilterIndexReader(IndexReader in) {
- super(in.directory());
- segmentInfos = in.segmentInfos;
- this.in = in;
+ /**
+ * <p>Construct a FilterIndexReader based on the specified base reader.
+ * Directory locking for delete, undeleteAll, and setNorm operations is
+ * left to the base reader.</p>
+ * <p>Note that base reader is closed if this FilterIndexReader is closed.</p>
+ * @param in specified base reader.
+ */
+ public FilterIndexReader(IndexReader baseReader) {
+ super(baseReader.directory());
+ this.baseReader = baseReader;
}
public TermFreqVector[] getTermFreqVectors(int docNumber)
throws IOException {
- return in.getTermFreqVectors(docNumber);
+ return baseReader.getTermFreqVectors(docNumber);
}
public TermFreqVector getTermFreqVector(int docNumber, String field)
throws IOException {
- return in.getTermFreqVector(docNumber, field);
+ return baseReader.getTermFreqVector(docNumber, field);
}
- public int numDocs() { return in.numDocs(); }
- public int maxDoc() { return in.maxDoc(); }
+ public int numDocs() { return baseReader.numDocs(); }
+ public int maxDoc() { return baseReader.maxDoc(); }
- public Document document(int n) throws IOException { return in.document(n); }
+ public Document document(int n) throws IOException { return
baseReader.document(n); }
- public boolean isDeleted(int n) { return in.isDeleted(n); }
- public boolean hasDeletions() { return in.hasDeletions(); }
- public void undeleteAll() throws IOException { in.undeleteAll(); }
+ public boolean isDeleted(int n) { return baseReader.isDeleted(n); }
+ public boolean hasDeletions() { return baseReader.hasDeletions(); }
+ protected void doUndeleteAll() throws IOException { baseReader.undeleteAll(); }
- public byte[] norms(String f) throws IOException { return in.norms(f); }
+ public byte[] norms(String f) throws IOException { return baseReader.norms(f); }
public void norms(String f, byte[] bytes, int offset) throws IOException {
- in.norms(f, bytes, offset);
+ baseReader.norms(f, bytes, offset);
}
- public void setNorm(int d, String f, byte b) throws IOException {
- in.setNorm(d, f, b);
+ protected void doSetNorm(int d, String f, byte b) throws IOException {
+ baseReader.setNorm(d, f, b);
}
- public TermEnum terms() throws IOException { return in.terms(); }
- public TermEnum terms(Term t) throws IOException { return in.terms(t); }
+ public TermEnum terms() throws IOException { return baseReader.terms(); }
+ public TermEnum terms(Term t) throws IOException { return baseReader.terms(t); }
- public int docFreq(Term t) throws IOException { return in.docFreq(t); }
+ public int docFreq(Term t) throws IOException { return baseReader.docFreq(t); }
- public TermDocs termDocs() throws IOException { return in.termDocs(); }
+ public TermDocs termDocs() throws IOException { return baseReader.termDocs(); }
public TermPositions termPositions() throws IOException {
- return in.termPositions();
+ return baseReader.termPositions();
}
- protected void doDelete(int n) throws IOException { in.doDelete(n); }
- protected void doClose() throws IOException { in.doClose(); }
+ protected void doDelete(int n) throws IOException { baseReader.delete(n); }
+ protected void doCommit() throws IOException { baseReader.commit(); }
+ protected void doClose() throws IOException { baseReader.close(); }
public Collection getFieldNames() throws IOException {
- return in.getFieldNames();
+ return baseReader.getFieldNames();
}
public Collection getFieldNames(boolean indexed) throws IOException {
- return in.getFieldNames(indexed);
+ return baseReader.getFieldNames(indexed);
}
/**
@@ -137,6 +144,6 @@
* @return Collection of Strings indicating the names of the fields
*/
public Collection getIndexedFieldNames(boolean storedTermVector) {
- return in.getIndexedFieldNames(storedTermVector);
+ return baseReader.getIndexedFieldNames(storedTermVector);
}
}
1.30 +125 -27 jakarta-lucene/src/java/org/apache/lucene/index/IndexReader.java
Index: IndexReader.java
===================================================================
RCS file:
/home/cvs/jakarta-lucene/src/java/org/apache/lucene/index/IndexReader.java,v
retrieving revision 1.29
retrieving revision 1.30
diff -u -r1.29 -r1.30
--- IndexReader.java 29 Mar 2004 22:48:02 -0000 1.29
+++ IndexReader.java 16 Apr 2004 09:17:05 -0000 1.30
@@ -44,31 +44,69 @@
@version $Id$
*/
public abstract class IndexReader {
+
+ /**
+ * Constructor used if IndexReader is not owner of its directory.
+ * This is used for IndexReaders that are used within other IndexReaders that
take care or locking directories.
+ *
+ * @param directory Directory where IndexReader files reside.
+ */
protected IndexReader(Directory directory) {
this.directory = directory;
- stale = false;
segmentInfos = null;
+ directoryOwner = false;
+ closeDirectory = false;
+ stale = false;
+ hasChanges = false;
+ writeLock = null;
+ }
+
+ /**
+ * Constructor used if IndexReader is owner of its directory.
+ * If IndexReader is owner of its directory, it locks its directory in case of
write operations.
+ *
+ * @param directory Directory where IndexReader files reside.
+ * @param segmentInfos Used for write-l
+ * @param closeDirectory
+ */
+ protected IndexReader(Directory directory, SegmentInfos segmentInfos, boolean
closeDirectory) {
+ this.directory = directory;
+ this.segmentInfos = segmentInfos;
+ directoryOwner = true;
+ this.closeDirectory = closeDirectory;
+ stale = false;
+ hasChanges = false;
+ writeLock = null;
}
- private Directory directory;
+ final private Directory directory;
+
+ final private boolean directoryOwner;
+ final private SegmentInfos segmentInfos;
private Lock writeLock;
- SegmentInfos segmentInfos = null;
- private boolean stale = false;
+ private boolean stale;
+ private boolean hasChanges;
+
+ final private boolean closeDirectory;
/** Returns an IndexReader reading the index in an FSDirectory in the named
path. */
public static IndexReader open(String path) throws IOException {
- return open(FSDirectory.getDirectory(path, false));
+ return open(FSDirectory.getDirectory(path, false), true);
}
/** Returns an IndexReader reading the index in an FSDirectory in the named
path. */
public static IndexReader open(File path) throws IOException {
- return open(FSDirectory.getDirectory(path, false));
+ return open(FSDirectory.getDirectory(path, false), true);
}
-
+
/** Returns an IndexReader reading the index in the given Directory. */
public static IndexReader open(final Directory directory) throws IOException {
+ return open(directory, false);
+ }
+
+ private static IndexReader open(final Directory directory, final boolean
closeDirectory) throws IOException {
synchronized (directory) { // in- & inter-process sync
return (IndexReader)new Lock.With(
directory.makeLock(IndexWriter.COMMIT_LOCK_NAME),
@@ -77,12 +115,12 @@
SegmentInfos infos = new SegmentInfos();
infos.read(directory);
if (infos.size() == 1) { // index is optimized
- return new SegmentReader(infos, infos.info(0), true);
+ return new SegmentReader(infos, infos.info(0), closeDirectory);
} else {
IndexReader[] readers = new IndexReader[infos.size()];
for (int i = 0; i < infos.size(); i++)
- readers[i] = new SegmentReader(infos, infos.info(i),
i==infos.size()-1);
- return new MultiReader(directory, readers);
+ readers[i] = new SegmentReader(infos.info(i));
+ return new MultiReader(directory, infos, closeDirectory, readers);
}
}
}.run();
@@ -272,7 +310,16 @@
* @see #norms(String)
* @see Similarity#decodeNorm(byte)
*/
- public abstract void setNorm(int doc, String field, byte value)
+ public final synchronized void setNorm(int doc, String field, byte value)
+ throws IOException{
+ if(directoryOwner)
+ aquireWriteLock();
+ doSetNorm(doc, field, value);
+ hasChanges = true;
+ }
+
+ /** Implements setNorm in subclass.*/
+ protected abstract void doSetNorm(int doc, String field, byte value)
throws IOException;
/** Expert: Resets the normalization factor for the named field of the named
@@ -346,16 +393,15 @@
/** Returns an unpositioned [EMAIL PROTECTED] TermPositions} enumerator. */
public abstract TermPositions termPositions() throws IOException;
- /** Deletes the document numbered <code>docNum</code>. Once a document is
- deleted it will not appear in TermDocs or TermPostitions enumerations.
- Attempts to read its field with the [EMAIL PROTECTED] #document}
- method will result in an error. The presence of this document may still be
- reflected in the [EMAIL PROTECTED] #docFreq} statistic, though
- this will be corrected eventually as the index is further modified.
+ /**
+ * Trys to acquire the WriteLock on this directory.
+ * this method is only valid if this IndexReader is directory owner.
+ *
+ * @throws IOException If WriteLock cannot be acquired.
*/
- public final synchronized void delete(int docNum) throws IOException {
+ private void aquireWriteLock() throws IOException {
if (stale)
- throw new IOException("IndexReader out of date and no longer valid for
deletion");
+ throw new IOException("IndexReader out of date and no longer valid for
delete, undelete, or setNorm operations");
if (writeLock == null) {
Lock writeLock = directory.makeLock(IndexWriter.WRITE_LOCK_NAME);
@@ -365,14 +411,27 @@
// we have to check whether index has changed since this reader was opened.
// if so, this reader is no longer valid for deletion
- if (segmentInfos != null && SegmentInfos.readCurrentVersion(directory) >
segmentInfos.getVersion()) {
+ if (SegmentInfos.readCurrentVersion(directory) > segmentInfos.getVersion()) {
stale = true;
this.writeLock.release();
this.writeLock = null;
- throw new IOException("IndexReader out of date and no longer valid for
deletion");
+ throw new IOException("IndexReader out of date and no longer valid for
delete, undelete, or setNorm operations");
}
}
+ }
+
+ /** Deletes the document numbered <code>docNum</code>. Once a document is
+ deleted it will not appear in TermDocs or TermPostitions enumerations.
+ Attempts to read its field with the [EMAIL PROTECTED] #document}
+ method will result in an error. The presence of this document may still be
+ reflected in the [EMAIL PROTECTED] #docFreq} statistic, though
+ this will be corrected eventually as the index is further modified.
+ */
+ public final synchronized void delete(int docNum) throws IOException {
+ if(directoryOwner)
+ aquireWriteLock();
doDelete(docNum);
+ hasChanges = true;
}
/** Implements deletion of the document numbered <code>docNum</code>.
@@ -402,19 +461,58 @@
}
/** Undeletes all documents currently marked as deleted in this index.*/
- public abstract void undeleteAll() throws IOException;
+ public final synchronized void undeleteAll() throws IOException{
+ if(directoryOwner)
+ aquireWriteLock();
+ doUndeleteAll();
+ hasChanges = true;
+ }
+
+ /** Implements actual undeleteAll() in subclass. */
+ protected abstract void doUndeleteAll() throws IOException;
/**
+ * Commit changes resulting from delete, undeleteAll, or setNorm operations
+ *
+ * @throws IOException
+ */
+ protected final synchronized void commit() throws IOException{
+ if(hasChanges){
+ if(directoryOwner){
+ synchronized (directory) { // in- & inter-process sync
+ new Lock.With(directory.makeLock(IndexWriter.COMMIT_LOCK_NAME),
+ IndexWriter.COMMIT_LOCK_TIMEOUT) {
+ public Object doBody() throws IOException {
+ doCommit();
+ segmentInfos.write(directory);
+ return null;
+ }
+ }.run();
+ }
+ if (writeLock != null) {
+ writeLock.release(); // release write lock
+ writeLock = null;
+ }
+ }
+ else
+ doCommit();
+ }
+ hasChanges = false;
+ }
+
+ /** Implements commit. */
+ protected abstract void doCommit() throws IOException;
+
+ /**
* Closes files associated with this index.
* Also saves any new deletions to disk.
* No other methods should be called after this has been called.
*/
public final synchronized void close() throws IOException {
+ commit();
doClose();
- if (writeLock != null) {
- writeLock.release(); // release write lock
- writeLock = null;
- }
+ if(closeDirectory)
+ directory.close();
}
/** Implements close. */
1.21 +45 -63
jakarta-lucene/src/java/org/apache/lucene/index/SegmentReader.java
Index: SegmentReader.java
===================================================================
RCS file:
/home/cvs/jakarta-lucene/src/java/org/apache/lucene/index/SegmentReader.java,v
retrieving revision 1.20
retrieving revision 1.21
diff -u -r1.20 -r1.21
--- SegmentReader.java 29 Mar 2004 22:48:02 -0000 1.20
+++ SegmentReader.java 16 Apr 2004 09:17:05 -0000 1.21
@@ -37,7 +37,6 @@
* @version $Id$
*/
final class SegmentReader extends IndexReader {
- private boolean closeDirectory = false;
private String segment;
FieldInfos fieldInfos;
@@ -49,6 +48,7 @@
BitVector deletedDocs = null;
private boolean deletedDocsDirty = false;
private boolean normsDirty = false;
+ private boolean undeleteAll = false;
InputStream freqStream;
InputStream proxStream;
@@ -57,13 +57,18 @@
CompoundFileReader cfsReader;
private class Norm {
- public Norm(InputStream in) { this.in = in; }
+ public Norm(InputStream in, int number)
+ {
+ this.in = in;
+ this.number = number;
+ }
private InputStream in;
private byte[] bytes;
private boolean dirty;
+ private int number;
- private void reWrite(String name) throws IOException {
+ private void reWrite() throws IOException {
// NOTE: norms are re-written in regular directory, not cfs
OutputStream out = directory().createFile(segment + ".tmp");
try {
@@ -71,7 +76,7 @@
} finally {
out.close();
}
- String fileName = segment + ".f" + fieldInfos.fieldNumber(name);
+ String fileName = segment + ".f" + number;
directory().renameFile(segment + ".tmp", fileName);
this.dirty = false;
}
@@ -81,14 +86,17 @@
SegmentReader(SegmentInfos sis, SegmentInfo si, boolean closeDir)
throws IOException {
- this(si);
- closeDirectory = closeDir;
- segmentInfos = sis;
+ super(si.dir, sis, closeDir);
+ initialize(si);
}
- SegmentReader(SegmentInfo si)
- throws IOException {
+ SegmentReader(SegmentInfo si) throws IOException {
super(si.dir);
+ initialize(si);
+ }
+
+ private void initialize(SegmentInfo si) throws IOException
+ {
segment = si.name;
// Use compound file directory for some files, if it exists
@@ -119,42 +127,29 @@
}
}
- protected final synchronized void doClose() throws IOException {
- if (deletedDocsDirty || normsDirty) {
- synchronized (directory()) { // in- & inter-process sync
- new Lock.With(directory().makeLock(IndexWriter.COMMIT_LOCK_NAME),
- IndexWriter.COMMIT_LOCK_TIMEOUT) {
- public Object doBody() throws IOException {
-
- if (deletedDocsDirty) { // re-write deleted
- deletedDocs.write(directory(), segment + ".tmp");
- directory().renameFile(segment + ".tmp", segment + ".del");
- }
-
- if (normsDirty) { // re-write norms
- Enumeration keys = norms.keys();
- Enumeration values = norms.elements();
- while (values.hasMoreElements()) {
- String field = (String) keys.nextElement();
- Norm norm = (Norm) values.nextElement();
- if (norm.dirty) {
- norm.reWrite(field);
- }
- }
- }
-
- if (segmentInfos != null)
- segmentInfos.write(directory());
- else
- directory().touchFile("segments");
- return null;
- }
- }.run();
+ protected final void doCommit() throws IOException {
+ if (deletedDocsDirty) { // re-write deleted
+ deletedDocs.write(directory(), segment + ".tmp");
+ directory().renameFile(segment + ".tmp", segment + ".del");
+ }
+ if(undeleteAll && directory().fileExists(segment + ".del")){
+ directory().deleteFile(segment + ".del");
+ }
+ if (normsDirty) { // re-write norms
+ Enumeration values = norms.elements();
+ while (values.hasMoreElements()) {
+ Norm norm = (Norm) values.nextElement();
+ if (norm.dirty) {
+ norm.reWrite();
+ }
}
- deletedDocsDirty = false;
- normsDirty = false;
}
-
+ deletedDocsDirty = false;
+ normsDirty = false;
+ undeleteAll = false;
+ }
+
+ protected final void doClose() throws IOException {
fieldsReader.close();
tis.close();
@@ -168,9 +163,6 @@
if (cfsReader != null)
cfsReader.close();
-
- if (closeDirectory)
- directory().close();
}
static final boolean hasDeletions(SegmentInfo si) throws IOException {
@@ -186,30 +178,20 @@
return si.dir.fileExists(si.name + ".cfs");
}
- protected final synchronized void doDelete(int docNum) throws IOException {
+ protected final void doDelete(int docNum) throws IOException {
if (deletedDocs == null)
deletedDocs = new BitVector(maxDoc());
deletedDocsDirty = true;
+ undeleteAll = false;
deletedDocs.set(docNum);
}
- public synchronized void undeleteAll() throws IOException {
- synchronized (directory()) { // in- & inter-process sync
- new Lock.With(directory().makeLock(IndexWriter.COMMIT_LOCK_NAME),
- IndexWriter.COMMIT_LOCK_TIMEOUT) {
- public Object doBody() throws IOException {
- if (directory().fileExists(segment + ".del")) {
- directory().deleteFile(segment + ".del");
- }
- return null;
- }
- };
+ protected final void doUndeleteAll() throws IOException {
deletedDocs = null;
deletedDocsDirty = false;
- }
+ undeleteAll = true;
}
-
final Vector files() throws IOException {
Vector files = new Vector(16);
final String ext[] = new String[]{
@@ -334,7 +316,7 @@
return norm.bytes;
}
- public synchronized void setNorm(int doc, String field, byte value)
+ protected final void doSetNorm(int doc, String field, byte value)
throws IOException {
Norm norm = (Norm) norms.get(field);
if (norm == null) // not an indexed field
@@ -374,7 +356,7 @@
String fileName = segment + ".f" + fi.number;
// look first for re-written file, then in compound format
Directory d = directory().fileExists(fileName) ? directory() : cfsDir;
- norms.put(fi.name, new Norm(d.openFile(fileName)));
+ norms.put(fi.name, new Norm(d.openFile(fileName), fi.number));
}
}
}
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
