LUCENE-2114: TestFilteredSearch tests on multi-segment index; fix Filter javadocs to call out that reader is per-segment

git-svn-id: https://svn.apache.org/repos/asf/lucene/java/trunk@887347 13f79535-47bb-0310-9956-ffa450edef68

CHANGES.txt

@@ -69,6 +69,10 @@ Test Cases
 * LUCENE-2065: Use Java 5 generics throughout our unit tests.  (Kay
   Kay via Mike McCandless)
 
+* LUCENE-2114: Change TestFilteredSearch to test on multi-segment
+  index as well; improve javadocs of Filter to call out that the
+  provided reader is per-segment (Simon Willnauer via Mike McCandless)
+
 ======================= Release 3.0.0 2009-11-25 =======================
 
 Changes in backwards compatibility policy

src/java/org/apache/lucene/search/Filter.java

@@ -22,12 +22,28 @@ import java.io.IOException;
 import org.apache.lucene.index.IndexReader;
 import org.apache.lucene.util.DocIdBitSet;
 
-/** Abstract base class providing a mechanism to use a subset of an index
+/** 
+ *  Abstract base class providing a mechanism to use a subset of an index
  *  for restriction or permission of index search results.
  *  <p>
  */
 public abstract class Filter implements java.io.Serializable {
+  
   /**
+   * Creates a {@link DocIdSet} that provides the documents which should be
+   * permitted or prohibited in search results. <b>NOTE:</b> null can be
+   * returned if no documents will be accepted by this Filter.
+   * <p>
+   * Note: This method might be called more than once during a search if the
+   * index has more than one segment. In such a case the {@link DocIdSet}
+   * must be relative to the document base of the given reader. Yet, the
+   * segment readers are passed in increasing document base order.
+   * 
+   * @param reader a {@link IndexReader} instance opened on the index currently
+   *         searched on. Note, it is likely that the provided reader does not
+   *         represent the whole underlying index i.e. if the index has more than
+   *         one segment the given reader only represents a single segment.
+   *          
    * @return a DocIdSet that provides the documents which should be permitted or
    *         prohibited in search results. <b>NOTE:</b> null can be returned if
    *         no documents will be accepted by this Filter.

src/test/org/apache/lucene/search/TestFilteredSearch.java

@@ -24,9 +24,12 @@ import org.apache.lucene.util.LuceneTestCase;
 import org.apache.lucene.analysis.WhitespaceAnalyzer;
 import org.apache.lucene.document.Document;
 import org.apache.lucene.document.Field;
+import org.apache.lucene.index.CorruptIndexException;
 import org.apache.lucene.index.IndexReader;
 import org.apache.lucene.index.IndexWriter;
 import org.apache.lucene.index.Term;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.store.LockObtainFailedException;
 import org.apache.lucene.store.RAMDirectory;
 import org.apache.lucene.util.OpenBitSet;
 
@@ -42,19 +45,32 @@ public class TestFilteredSearch extends LuceneTestCase {
 
   private static final String FIELD = "category";
   
-  public void testFilteredSearch() {
+  public void testFilteredSearch() throws CorruptIndexException, LockObtainFailedException, IOException {
+    boolean enforceSingleSegment = true;
     RAMDirectory directory = new RAMDirectory();
     int[] filterBits = {1, 36};
-    Filter filter = new SimpleDocIdSetFilter(filterBits);
+    SimpleDocIdSetFilter filter = new SimpleDocIdSetFilter(filterBits);
-    
-
-    try {
     IndexWriter writer = new IndexWriter(directory, new WhitespaceAnalyzer(), true, IndexWriter.MaxFieldLength.LIMITED);
+    searchFiltered(writer, directory, filter, enforceSingleSegment);
+    // run the test on more than one segment
+    enforceSingleSegment = false;
+    // reset - it is stateful
+    filter.reset();
+    writer = new IndexWriter(directory, new WhitespaceAnalyzer(), true, IndexWriter.MaxFieldLength.LIMITED);
+    // we index 60 docs - this will create 6 segments
+    writer.setMaxBufferedDocs(10);
+    searchFiltered(writer, directory, filter, enforceSingleSegment);
+  }
+
+  public void searchFiltered(IndexWriter writer, Directory directory, Filter filter, boolean optimize) {
+    try {
       for (int i = 0; i < 60; i++) {//Simple docs
         Document doc = new Document();
         doc.add(new Field(FIELD, Integer.toString(i), Field.Store.YES, Field.Index.NOT_ANALYZED));
         writer.addDocument(doc);
       }
+      if(optimize)
+        writer.optimize();
       writer.close();
 
       BooleanQuery booleanQuery = new BooleanQuery();
@@ -72,21 +88,30 @@ public class TestFilteredSearch extends LuceneTestCase {
     
   }
  
-
   public static final class SimpleDocIdSetFilter extends Filter {
-    private OpenBitSet bits;
+    private int docBase;
-
+    private final int[] docs;
+    private int index;
     public SimpleDocIdSetFilter(int[] docs) {
-      bits = new OpenBitSet();
+      this.docs = docs;
-      for(int i = 0; i < docs.length; i++){
-    	  bits.set(docs[i]);
     }
-      
-    }
-
     @Override
     public DocIdSet getDocIdSet(IndexReader reader) {
-      return bits;
+      final OpenBitSet set = new OpenBitSet();
+      final int limit = docBase+reader.maxDoc();
+      for (;index < docs.length; index++) {
+        final int docId = docs[index];
+        if(docId > limit)
+          break;
+        set.set(docId-docBase);
+      }
+      docBase = limit;
+      return set.isEmpty()?null:set;
+    }
+    
+    public void reset(){
+      index = 0;
+      docBase = 0;
     }
   }