1 package org.apache.lucene.index;
4 * Licensed to the Apache Software Foundation (ASF) under one or more
5 * contributor license agreements. See the NOTICE file distributed with
6 * this work for additional information regarding copyright ownership.
7 * The ASF licenses this file to You under the Apache License, Version 2.0
8 * (the "License"); you may not use this file except in compliance with
9 * the License. You may obtain a copy of the License at
11 * http://www.apache.org/licenses/LICENSE-2.0
13 * Unless required by applicable law or agreed to in writing, software
14 * distributed under the License is distributed on an "AS IS" BASIS,
15 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
16 * See the License for the specific language governing permissions and
17 * limitations under the License.
20 import org.apache.lucene.util.LuceneTestCase;
21 import org.apache.lucene.store.Directory;
22 import org.apache.lucene.store.BufferedIndexInput;
23 import org.apache.lucene.analysis.MockAnalyzer;
24 import org.apache.lucene.document.Document;
25 import org.apache.lucene.document.Field;
26 import org.apache.lucene.document.Field.Index;
27 import org.apache.lucene.document.Field.Store;
28 import org.apache.lucene.index.IndexWriterConfig.OpenMode;
30 import java.io.IOException;
31 import java.util.Collection;
33 public class TestSegmentMerger extends LuceneTestCase {
34 //The variables for the new merged segment
35 private Directory mergedDir;
36 private String mergedSegment = "test";
37 //First segment to be merged
38 private Directory merge1Dir;
39 private Document doc1 = new Document();
40 private SegmentReader reader1 = null;
41 //Second Segment to be merged
42 private Directory merge2Dir;
43 private Document doc2 = new Document();
44 private SegmentReader reader2 = null;
47 public void setUp() throws Exception {
49 mergedDir = newDirectory();
50 merge1Dir = newDirectory();
51 merge2Dir = newDirectory();
52 DocHelper.setupDoc(doc1);
53 SegmentInfo info1 = DocHelper.writeDoc(random, merge1Dir, doc1);
54 DocHelper.setupDoc(doc2);
55 SegmentInfo info2 = DocHelper.writeDoc(random, merge2Dir, doc2);
56 reader1 = SegmentReader.get(true, info1, IndexReader.DEFAULT_TERMS_INDEX_DIVISOR);
57 reader2 = SegmentReader.get(true, info2, IndexReader.DEFAULT_TERMS_INDEX_DIVISOR);
61 public void tearDown() throws Exception {
71 assertTrue(mergedDir != null);
72 assertTrue(merge1Dir != null);
73 assertTrue(merge2Dir != null);
74 assertTrue(reader1 != null);
75 assertTrue(reader2 != null);
78 public void testMerge() throws IOException {
79 SegmentMerger merger = new SegmentMerger(mergedDir, IndexWriterConfig.DEFAULT_TERM_INDEX_INTERVAL, mergedSegment, null, null, new FieldInfos());
82 int docsMerged = merger.merge();
83 assertTrue(docsMerged == 2);
84 //Should be able to open a new SegmentReader against the new directory
85 SegmentReader mergedReader = SegmentReader.get(false, mergedDir,
86 new SegmentInfo(mergedSegment, docsMerged, mergedDir, false, true,
87 merger.fieldInfos().hasProx(),
88 merger.fieldInfos().hasVectors()),
89 BufferedIndexInput.BUFFER_SIZE, true, IndexReader.DEFAULT_TERMS_INDEX_DIVISOR);
91 assertTrue(mergedReader != null);
92 assertTrue(mergedReader.numDocs() == 2);
93 Document newDoc1 = mergedReader.document(0);
94 assertTrue(newDoc1 != null);
95 //There are 2 unstored fields on the document
96 assertTrue(DocHelper.numFields(newDoc1) == DocHelper.numFields(doc1) - DocHelper.unstored.size());
97 Document newDoc2 = mergedReader.document(1);
98 assertTrue(newDoc2 != null);
99 assertTrue(DocHelper.numFields(newDoc2) == DocHelper.numFields(doc2) - DocHelper.unstored.size());
101 TermDocs termDocs = mergedReader.termDocs(new Term(DocHelper.TEXT_FIELD_2_KEY, "field"));
102 assertTrue(termDocs != null);
103 assertTrue(termDocs.next() == true);
105 Collection<String> stored = mergedReader.getFieldNames(IndexReader.FieldOption.INDEXED_WITH_TERMVECTOR);
106 assertTrue(stored != null);
107 //System.out.println("stored size: " + stored.size());
108 assertTrue("We do not have 3 fields that were indexed with term vector",stored.size() == 3);
110 TermFreqVector vector = mergedReader.getTermFreqVector(0, DocHelper.TEXT_FIELD_2_KEY);
111 assertTrue(vector != null);
112 String [] terms = vector.getTerms();
113 assertTrue(terms != null);
114 //System.out.println("Terms size: " + terms.length);
115 assertTrue(terms.length == 3);
116 int [] freqs = vector.getTermFrequencies();
117 assertTrue(freqs != null);
118 //System.out.println("Freqs size: " + freqs.length);
119 assertTrue(vector instanceof TermPositionVector == true);
121 for (int i = 0; i < terms.length; i++) {
122 String term = terms[i];
124 //System.out.println("Term: " + term + " Freq: " + freq);
125 assertTrue(DocHelper.FIELD_2_TEXT.indexOf(term) != -1);
126 assertTrue(DocHelper.FIELD_2_FREQS[i] == freq);
129 TestSegmentReader.checkNorms(mergedReader);
130 mergedReader.close();
134 public void testInvalidFilesToCreateCompound() throws Exception {
135 Directory dir = newDirectory();
136 IndexWriterConfig iwc = newIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(random));
137 IndexWriter w = new IndexWriter(dir, iwc);
139 // Create an index w/ .del file
140 w.addDocument(new Document());
141 Document doc = new Document();
142 doc.add(new Field("c", "test", Store.NO, Index.ANALYZED));
145 w.deleteDocuments(new Term("c", "test"));
148 // Assert that SM fails if .del exists
149 SegmentMerger sm = new SegmentMerger(dir, 1, "a", null, null, null);
151 sm.createCompoundFile("b1", w.segmentInfos.info(0));
152 fail("should not have been able to create a .cfs with .del and .s* files");
153 } catch (AssertionError e) {
157 // Create an index w/ .s*
158 w = new IndexWriter(dir, new IndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(random)).setOpenMode(OpenMode.CREATE));
159 doc = new Document();
160 doc.add(new Field("c", "test", Store.NO, Index.ANALYZED));
163 IndexReader r = IndexReader.open(dir, false);
164 r.setNorm(0, "c", (byte) 1);
167 // Assert that SM fails if .s* exists
169 sm.createCompoundFile("b2", w.segmentInfos.info(0));
170 fail("should not have been able to create a .cfs with .del and .s* files");
171 } catch (AssertionError e) {