LUCENE-1578: Support for loading unoptimized readers to the constructor of InstantiatedIndex. (Karl Wettin)

git-svn-id: https://svn.apache.org/repos/asf/lucene/java/trunk@784481 13f79535-47bb-0310-9956-ffa450edef68

contrib/CHANGES.txt

@@ -62,8 +62,11 @@ New features
     (Xiaoping Gao via Mike McCandless)
 
 
-6. LUCENE-1676: Added DelimitedPayloadTokenFilter class for automatically adding payloads "in-stream" (Grant Ingersoll)    
+ 6. LUCENE-1676: Added DelimitedPayloadTokenFilter class for automatically adding payloads "in-stream" (Grant Ingersoll)    
-  
+ 
+ 7. LUCENE-1578: Support for loading unoptimized readers to the
+    constructor of InstantiatedIndex. (Karl Wettin)
+ 
 Optimizations
 
   1. LUCENE-1643: Re-use the collation key (RawCollationKey) for

contrib/instantiated/src/java/org/apache/lucene/store/instantiated/InstantiatedIndex.java

@@ -110,7 +110,8 @@ public class InstantiatedIndex
   public InstantiatedIndex(IndexReader sourceIndexReader, Set<String> fields) throws IOException {
 
     if (!sourceIndexReader.isOptimized()) {
-      throw new IOException("Source index is not optimized.");
+      System.out.println(("Source index is not optimized."));      
+      //throw new IOException("Source index is not optimized.");
     }
 
 
@@ -170,11 +171,14 @@ public class InstantiatedIndex
     }
 
 
-    documentsByNumber = new InstantiatedDocument[sourceIndexReader.numDocs()];
+    documentsByNumber = new InstantiatedDocument[sourceIndexReader.maxDoc()];
+
 
     // create documents
-    for (int i = 0; i < sourceIndexReader.numDocs(); i++) {
+    for (int i = 0; i < sourceIndexReader.maxDoc(); i++) {
-      if (!sourceIndexReader.isDeleted(i)) {
+      if (sourceIndexReader.isDeleted(i)) {
+        deletedDocuments.add(i);
+      } else {
         InstantiatedDocument document = new InstantiatedDocument();
         // copy stored fields from source reader
         Document sourceDocument = sourceIndexReader.document(i);
@@ -259,6 +263,9 @@ public class InstantiatedIndex
 
     // load offsets to term-document informations
     for (InstantiatedDocument document : getDocumentsByNumber()) {
+      if (document == null) {
+        continue; // deleted
+      }
       for (Field field : (List<Field>) document.getDocument().getFields()) {
         if (field.isTermVectorStored() && field.isStoreOffsetWithTermVector()) {
           TermPositionVector termPositionVector = (TermPositionVector) sourceIndexReader.getTermFreqVector(document.getDocumentNumber(), field.name());

contrib/instantiated/src/test/org/apache/lucene/store/instantiated/TestIndicesEquals.java

@@ -40,6 +40,10 @@ import org.apache.lucene.index.TermPositionVector;
 import org.apache.lucene.index.TermPositions;
 import org.apache.lucene.store.Directory;
 import org.apache.lucene.store.RAMDirectory;
+import org.apache.lucene.search.IndexSearcher;
+import org.apache.lucene.search.TermQuery;
+import org.apache.lucene.search.TopDocCollector;
+import org.apache.lucene.search.TopScoreDocCollector;
 
 /**
  * Asserts equality of content and behaviour of two index readers.

contrib/instantiated/src/test/org/apache/lucene/store/instantiated/TestRealTime.java

@@ -24,6 +24,12 @@ import org.apache.lucene.document.Document;
 import org.apache.lucene.document.Field;
 import org.apache.lucene.index.Term;
 
+/**
+ * Assert that the content of an index 
+ * is instantly available
+ * for all open searchers
+ * also after a commit.
+ */
 public class TestRealTime extends TestCase {
 
   public void test() throws Exception {

contrib/instantiated/src/test/org/apache/lucene/store/instantiated/TestUnoptimizedReaderOnConstructor.java

@@ -0,0 +1,53 @@
+package org.apache.lucene.store.instantiated;
+
+import junit.framework.TestCase;
+
+import java.io.IOException;
+
+import org.apache.lucene.index.IndexReader;
+import org.apache.lucene.index.IndexWriter;
+import org.apache.lucene.store.RAMDirectory;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.analysis.WhitespaceAnalyzer;
+import org.apache.lucene.document.Document;
+import org.apache.lucene.document.Field;
+
+/**
+ * @author kalle
+ * @since 2009-mar-30 13:15:49
+ */
+public class TestUnoptimizedReaderOnConstructor extends TestCase {
+
+  public void test() throws Exception {
+    Directory dir = new RAMDirectory();
+    IndexWriter iw = new IndexWriter(dir, new WhitespaceAnalyzer(), true, IndexWriter.MaxFieldLength.UNLIMITED);
+    addDocument(iw, "Hello, world!");
+    addDocument(iw, "All work and no play makes jack a dull boy");
+    iw.commit("a");
+    iw.close();
+
+    iw = new IndexWriter(dir, new WhitespaceAnalyzer(), false, IndexWriter.MaxFieldLength.UNLIMITED);
+    addDocument(iw, "Hello, tellus!");
+    addDocument(iw, "All work and no play makes danny a dull boy");
+    iw.commit("b");
+    iw.close();
+
+    iw = new IndexWriter(dir, new WhitespaceAnalyzer(), false, IndexWriter.MaxFieldLength.UNLIMITED);
+    addDocument(iw, "Hello, earth!");
+    addDocument(iw, "All work and no play makes wendy a dull girl");
+    iw.commit("c");
+    iw.close();
+
+    IndexReader unoptimizedReader = IndexReader.open(dir);
+    unoptimizedReader.deleteDocument(2);
+
+    InstantiatedIndex ii = new InstantiatedIndex(unoptimizedReader);
+
+  }
+
+  private void addDocument(IndexWriter iw, String text) throws IOException {
+    Document doc = new Document();
+    doc.add(new Field("field", text, Field.Store.NO, Field.Index.ANALYZED));
+    iw.addDocument(doc);
+  }
+}