data corruption reading lazy loaded fields: LUCENE-683

git-svn-id: https://svn.apache.org/repos/asf/lucene/java/trunk@464313 13f79535-47bb-0310-9956-ffa450edef68

CHANGES.txt

@@ -127,6 +127,8 @@ Bug fixes
     has no value.
     (Oliver Hutchison via Chris Hostetter)
 
+15. LUCENE-683: Fixed data corruption when reading lazy loaded fields.
+    (Yonik Seeley)
 
 Optimizations
 

src/java/org/apache/lucene/index/FieldsReader.java

@@ -39,12 +39,12 @@ import org.apache.lucene.store.IndexInput;
  * @version $Id$
  */
 final class FieldsReader {
-  private FieldInfos fieldInfos;
-  private IndexInput fieldsStream;
-  private IndexInput indexStream;
+  private final FieldInfos fieldInfos;
+  private final IndexInput fieldsStream;
+  private final IndexInput indexStream;
   private int size;
 
-  private static ThreadLocal fieldsStreamTL = new ThreadLocal();
+  private ThreadLocal fieldsStreamTL = new ThreadLocal();
 
   FieldsReader(Directory d, String segment, FieldInfos fn) throws IOException {
     fieldInfos = fn;

src/test/org/apache/lucene/index/TestLazyBug.java

@@ -0,0 +1,130 @@
+package org.apache.lucene.index;
+
+/**
+ * Copyright 2006 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.analysis.Analyzer;
+import org.apache.lucene.analysis.SimpleAnalyzer;
+import org.apache.lucene.document.*;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.store.RAMDirectory;
+
+import java.util.*;
+import java.lang.reflect.Array;
+
+
+/**
+ * Test demonstrating EOF bug on the last field of the last doc 
+ * if other docs have allready been accessed.
+ */
+public class TestLazyBug extends TestCase {
+
+  public static int BASE_SEED = 13;
+
+  public static int NUM_DOCS = 500;
+  public static int NUM_FIELDS = 100;
+
+  private static String[] data = new String[] {
+    "now",
+    "is the time",
+    "for all good men",
+    "to come to the aid",
+    "of their country!",
+    "this string contains big chars:{\u0111 \u0222 \u0333 \u1111 \u2222 \u3333}",
+    "this string is a bigger string, mary had a little lamb, little lamb, little lamb!"
+  };
+
+  private static Set dataset = new HashSet(Arrays.asList(data));
+  
+  private static String MAGIC_FIELD = "f"+(NUM_FIELDS/3);
+  
+  private static FieldSelector SELECTOR = new FieldSelector() {
+      public FieldSelectorResult accept(String f) {
+        if (f.equals(MAGIC_FIELD)) {
+          return FieldSelectorResult.LOAD;
+        }
+        return FieldSelectorResult.LAZY_LOAD;
+      }
+    };
+  
+  private static Directory makeIndex() throws RuntimeException { 
+    Directory dir = new RAMDirectory();
+    try {
+      Random r = new Random(BASE_SEED + 42) ; 
+      Analyzer analyzer = new SimpleAnalyzer();
+      IndexWriter writer = new IndexWriter(dir, analyzer, true);
+      
+      writer.setUseCompoundFile(false);
+      
+      for (int d = 1; d <= NUM_DOCS; d++) {
+        Document doc = new Document();
+        for (int f = 1; f <= NUM_FIELDS; f++ ) {
+          doc.add(new Field("f"+f, 
+                            data[f % data.length] 
+                            + '#' + data[r.nextInt(data.length)], 
+                            Field.Store.YES, 
+                            Field.Index.TOKENIZED));
+        }
+        writer.addDocument(doc);
+      }
+      writer.close();
+    } catch (Exception e) {
+      throw new RuntimeException(e);
+    }
+    return dir;
+  }
+  
+  public static void doTest(int[] docs) throws Exception {
+    Directory dir = makeIndex();
+    IndexReader reader = IndexReader.open(dir);
+    for (int i = 0; i < docs.length; i++) {
+      Document d = reader.document(docs[i], SELECTOR);
+      String trash = d.get(MAGIC_FIELD);
+      
+      List fields = d.getFields();
+      for (Iterator fi = fields.iterator(); fi.hasNext(); ) {
+        Fieldable f=null;
+        try {
+          f = (Fieldable) fi.next();
+          String fname = f.name();
+          String fval = f.stringValue();
+          assertNotNull(docs[i]+" FIELD: "+fname, fval);
+          String[] vals = fval.split("#");
+          if (!dataset.contains(vals[0]) || !dataset.contains(vals[1])) {        
+            fail("FIELD:"+fname+",VAL:"+fval);
+          }
+        } catch (Exception e) {
+          throw new Exception(docs[i]+" WTF: "+f.name(), e);
+        }
+      }
+    }
+    reader.close();
+  }
+
+  public void testLazyWorks() throws Exception {
+    doTest(new int[] { 399 });
+  }
+  
+  public void testLazyAlsoWorks() throws Exception {
+    doTest(new int[] { 399, 150 });
+  }
+
+  public void testLazyBroken() throws Exception {
+    doTest(new int[] { 150, 399 });
+  }
+
+}