You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@lucene.apache.org by mi...@apache.org on 2013/01/21 02:10:15 UTC

svn commit: r1436045 - in /lucene/dev/branches/lucene4547/lucene: core/src/java/org/apache/lucene/index/ core/src/java/org/apache/lucene/search/ core/src/test/org/apache/lucene/index/ facet/src/java/org/apache/lucene/facet/index/

Author: mikemccand
Date: Mon Jan 21 01:10:15 2013
New Revision: 1436045

URL: http://svn.apache.org/viewvc?rev=1436045&view=rev
Log:
add failing test

Added:
    lucene/dev/branches/lucene4547/lucene/core/src/test/org/apache/lucene/index/TestDocValuesWithThreads.java   (with props)
Modified:
    lucene/dev/branches/lucene4547/lucene/core/src/java/org/apache/lucene/index/SortedDocValues.java
    lucene/dev/branches/lucene4547/lucene/core/src/java/org/apache/lucene/search/FieldCacheImpl.java
    lucene/dev/branches/lucene4547/lucene/facet/src/java/org/apache/lucene/facet/index/FacetsPayloadMigrationReader.java

Modified: lucene/dev/branches/lucene4547/lucene/core/src/java/org/apache/lucene/index/SortedDocValues.java
URL: http://svn.apache.org/viewvc/lucene/dev/branches/lucene4547/lucene/core/src/java/org/apache/lucene/index/SortedDocValues.java?rev=1436045&r1=1436044&r2=1436045&view=diff
==============================================================================
--- lucene/dev/branches/lucene4547/lucene/core/src/java/org/apache/lucene/index/SortedDocValues.java (original)
+++ lucene/dev/branches/lucene4547/lucene/core/src/java/org/apache/lucene/index/SortedDocValues.java Mon Jan 21 01:10:15 2013
@@ -41,6 +41,14 @@ public abstract class SortedDocValues ex
     }
   }
 
+  // nocommit make this final, and impl seekExact(term) to
+  // fwd to lookupTerm
+
+  // nocommit should we nuke this?  the iterator can be
+  // efficiently built "on top" since ord is part of the
+  // API?  why must it be impl'd here...?
+  // SortedDocValuesTermsEnum.
+
   public TermsEnum getTermsEnum() {
     // nocommit who tests this base impl ...
     // Default impl just uses the existing API; subclasses
@@ -161,6 +169,8 @@ public abstract class SortedDocValues ex
    *  @param key Key to look up
    *  @param spare Spare BytesRef
    **/
+  // nocommit make this protected so codecs can impl better
+  // version ...
   public int lookupTerm(BytesRef key, BytesRef spare) {
 
     int low = 0;

Modified: lucene/dev/branches/lucene4547/lucene/core/src/java/org/apache/lucene/search/FieldCacheImpl.java
URL: http://svn.apache.org/viewvc/lucene/dev/branches/lucene4547/lucene/core/src/java/org/apache/lucene/search/FieldCacheImpl.java?rev=1436045&r1=1436044&r2=1436045&view=diff
==============================================================================
--- lucene/dev/branches/lucene4547/lucene/core/src/java/org/apache/lucene/search/FieldCacheImpl.java (original)
+++ lucene/dev/branches/lucene4547/lucene/core/src/java/org/apache/lucene/search/FieldCacheImpl.java Mon Jan 21 01:10:15 2013
@@ -1133,6 +1133,9 @@ class FieldCacheImpl implements FieldCac
       final int maxDoc = reader.maxDoc();
       SortedDocValues valuesIn = reader.getSortedDocValues(key.field);
       if (valuesIn != null) {
+        // nocommit we need thread DV test that would
+        // uncover this bug!!
+        // nocommit we should not cache in this case?
         return valuesIn;
       } else {
 

Added: lucene/dev/branches/lucene4547/lucene/core/src/test/org/apache/lucene/index/TestDocValuesWithThreads.java
URL: http://svn.apache.org/viewvc/lucene/dev/branches/lucene4547/lucene/core/src/test/org/apache/lucene/index/TestDocValuesWithThreads.java?rev=1436045&view=auto
==============================================================================
--- lucene/dev/branches/lucene4547/lucene/core/src/test/org/apache/lucene/index/TestDocValuesWithThreads.java (added)
+++ lucene/dev/branches/lucene4547/lucene/core/src/test/org/apache/lucene/index/TestDocValuesWithThreads.java Mon Jan 21 01:10:15 2013
@@ -0,0 +1,112 @@
+package org.apache.lucene.index;
+
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You 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 java.util.ArrayList;
+import java.util.List;
+import java.util.Random;
+import java.util.concurrent.CountDownLatch;
+
+import org.apache.lucene.analysis.MockAnalyzer;
+import org.apache.lucene.document.Document;
+import org.apache.lucene.document.PackedLongDocValuesField;
+import org.apache.lucene.document.SortedBytesDocValuesField;
+import org.apache.lucene.document.StraightBytesDocValuesField;
+import org.apache.lucene.search.FieldCache;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.util.BytesRef;
+import org.apache.lucene.util.LuceneTestCase;
+import org.apache.lucene.util._TestUtil;
+
+public class TestDocValuesWithThreads extends LuceneTestCase {
+
+  public void test() throws Exception {
+    Directory dir = newDirectory();
+    IndexWriter w = new IndexWriter(dir, newIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(random())).setMergePolicy(newLogMergePolicy()));
+
+    // nocommit binary, sorted too
+    final List<Long> numbers = new ArrayList<Long>();
+    final List<BytesRef> binary = new ArrayList<BytesRef>();
+    final List<BytesRef> sorted = new ArrayList<BytesRef>();
+    final int numDocs = atLeast(100);
+    for(int i=0;i<numDocs;i++) {
+      Document d = new Document();
+      long number = random().nextLong();
+      d.add(new PackedLongDocValuesField("number", number));
+      BytesRef bytes = new BytesRef(_TestUtil.randomRealisticUnicodeString(random()));
+      d.add(new StraightBytesDocValuesField("bytes", bytes));
+      binary.add(bytes);
+      bytes = new BytesRef(_TestUtil.randomRealisticUnicodeString(random()));
+      d.add(new SortedBytesDocValuesField("sorted", bytes));
+      sorted.add(bytes);
+      w.addDocument(d);
+      numbers.add(number);
+    }
+
+    w.forceMerge(1);
+    final IndexReader r = w.getReader();
+    w.close();
+
+    assertEquals(1, r.leaves().size());
+    final AtomicReader ar = r.leaves().get(0).reader();
+
+    int numThreads = _TestUtil.nextInt(random(), 2, 5);
+    List<Thread> threads = new ArrayList<Thread>();
+    final CountDownLatch startingGun = new CountDownLatch(1);
+    for(int t=0;t<numThreads;t++) {
+      final Random threadRandom = new Random(random().nextLong());
+      Thread thread = new Thread() {
+          @Override
+          public void run() {
+            try {
+              //NumericDocValues ndv = ar.getNumericDocValues("number");
+              FieldCache.Longs ndv = FieldCache.DEFAULT.getLongs(ar, "number", false);
+              //BinaryDocValues bdv = ar.getBinaryDocValues("bytes");
+              BinaryDocValues bdv = FieldCache.DEFAULT.getTerms(ar, "bytes");
+              SortedDocValues sdv = FieldCache.DEFAULT.getTermsIndex(ar, "sorted");
+              startingGun.await();
+              int iters = atLeast(1000);
+              BytesRef scratch = new BytesRef();
+              for(int iter=0;iter<iters;iter++) {
+                int docID = threadRandom.nextInt(numDocs);
+                assertEquals(numbers.get(docID).longValue(), ndv.get(docID));
+                bdv.get(docID, scratch);
+                assertEquals(binary.get(docID), scratch);
+                sdv.get(docID, scratch);
+                assertEquals(sorted.get(docID), scratch);
+              }
+            } catch (Exception e) {
+              throw new RuntimeException(e);
+            }
+          }
+        };
+      thread.start();
+      threads.add(thread);
+    }
+
+    startingGun.countDown();
+
+    for(Thread thread : threads) {
+      thread.join();
+    }
+
+    r.close();
+    dir.close();
+  }
+
+}

Modified: lucene/dev/branches/lucene4547/lucene/facet/src/java/org/apache/lucene/facet/index/FacetsPayloadMigrationReader.java
URL: http://svn.apache.org/viewvc/lucene/dev/branches/lucene4547/lucene/facet/src/java/org/apache/lucene/facet/index/FacetsPayloadMigrationReader.java?rev=1436045&r1=1436044&r2=1436045&view=diff
==============================================================================
--- lucene/dev/branches/lucene4547/lucene/facet/src/java/org/apache/lucene/facet/index/FacetsPayloadMigrationReader.java (original)
+++ lucene/dev/branches/lucene4547/lucene/facet/src/java/org/apache/lucene/facet/index/FacetsPayloadMigrationReader.java Mon Jan 21 01:10:15 2013
@@ -27,10 +27,8 @@ import org.apache.lucene.facet.index.par
 import org.apache.lucene.facet.index.params.FacetIndexingParams;
 import org.apache.lucene.index.AtomicReader;
 import org.apache.lucene.index.AtomicReaderContext;
+import org.apache.lucene.index.BinaryDocValues;
 import org.apache.lucene.index.DirectoryReader;
-import org.apache.lucene.index.DocValues;
-import org.apache.lucene.index.DocValues.Source;
-import org.apache.lucene.index.DocValues.Type;
 import org.apache.lucene.index.DocsAndPositionsEnum;
 import org.apache.lucene.index.FieldInfo;
 import org.apache.lucene.index.FieldInfos;
@@ -45,9 +43,9 @@ import org.apache.lucene.util.BytesRef;
 
 /**
  * A {@link FilterAtomicReader} for migrating a facets index which encodes
- * category ordinals in a payload to {@link DocValues}. To migrate the index,
+ * category ordinals in a payload to {@link BinaryDocValues}. To migrate the index,
  * you should build a mapping from a field (String) to term ({@link Term}),
- * which denotes under which DocValues field to put the data encoded in the
+ * which denotes under which BinaryDocValues field to put the data encoded in the
  * matching term's payload. You can follow the code example below to migrate an
  * existing index:
  *