|
1 package org.apache.lucene.search;
2
3
18
19 import junit.framework.TestCase;
20 import org.apache.lucene.analysis.SimpleAnalyzer;
21 import org.apache.lucene.document.Document;
22 import org.apache.lucene.document.Field;
23 import org.apache.lucene.index.*;
24 import org.apache.lucene.store.Directory;
25 import org.apache.lucene.store.RAMDirectory;
26 import org.apache.lucene.util.English;
27
28 import java.io.IOException  ;
29 import java.util.HashMap ;
30 import java.util.Map ;
31
32 public class TestTermVectors extends TestCase {
33 private IndexSearcher searcher;
34 private RAMDirectory directory = new RAMDirectory();
35 public TestTermVectors(String s) {
36 super(s);
37 }
38
39 public void setUp() throws Exception {
40 IndexWriter writer
41 = new IndexWriter(directory, new SimpleAnalyzer(), true);
42 StringBuffer buffer = new StringBuffer ();
45 for (int i = 0; i < 1000; i++) {
46 Document doc = new Document();
47 doc.add(Field.Text("field", English.intToEnglish(i), true));
48 writer.addDocument(doc);
49 }
50 writer.close();
51 searcher = new IndexSearcher(directory);
52 }
53
54 protected void tearDown() {
55
56 }
57
58 public void test() {
59 assertTrue(searcher != null);
60 }
61
62 public void testTermVectors() {
63 Query query = new TermQuery(new Term("field", "seventy"));
64 try {
65 Hits hits = searcher.search(query);
66 assertEquals(100, hits.length());
67
68 for (int i = 0; i < hits.length(); i++)
69 {
70 TermFreqVector [] vector = searcher.reader.getTermFreqVectors(hits.id(i));
71 assertTrue(vector != null);
72 assertTrue(vector.length == 1);
73 }
75 TermFreqVector [] vector = searcher.reader.getTermFreqVectors(hits.id(50));
76 } catch (IOException e) {
79 assertTrue(false);
80 }
81 }
82
83 public void testTermPositionVectors() {
84 Query query = new TermQuery(new Term("field", "fifty"));
85 try {
86 Hits hits = searcher.search(query);
87 assertEquals(100, hits.length());
88
89 for (int i = 0; i < hits.length(); i++)
90 {
91 TermFreqVector [] vector = searcher.reader.getTermFreqVectors(hits.id(i));
92 assertTrue(vector != null);
93 assertTrue(vector.length == 1);
94 }
96 } catch (IOException e) {
97 assertTrue(false);
98 }
99 }
100
101 public void testKnownSetOfDocuments() {
102 String [] termArray = {"eating", "chocolate", "in", "a", "computer", "lab", "grows", "old", "colored",
103 "with", "an"};
104 String test1 = "eating chocolate in a computer lab"; String test2 = "computer in a computer lab"; String test3 = "a chocolate lab grows old"; String test4 = "eating chocolate with a chocolate lab in an old chocolate colored computer lab"; Map test4Map = new HashMap ();
109 test4Map.put("chocolate", new Integer (3));
110 test4Map.put("lab", new Integer (2));
111 test4Map.put("eating", new Integer (1));
112 test4Map.put("computer", new Integer (1));
113 test4Map.put("with", new Integer (1));
114 test4Map.put("a", new Integer (1));
115 test4Map.put("colored", new Integer (1));
116 test4Map.put("in", new Integer (1));
117 test4Map.put("an", new Integer (1));
118 test4Map.put("computer", new Integer (1));
119 test4Map.put("old", new Integer (1));
120
121 Document testDoc1 = new Document();
122 setupDoc(testDoc1, test1);
123 Document testDoc2 = new Document();
124 setupDoc(testDoc2, test2);
125 Document testDoc3 = new Document();
126 setupDoc(testDoc3, test3);
127 Document testDoc4 = new Document();
128 setupDoc(testDoc4, test4);
129
130 Directory dir = new RAMDirectory();
131
132 try {
133 IndexWriter writer = new IndexWriter(dir, new SimpleAnalyzer(), true);
134 assertTrue(writer != null);
135 writer.addDocument(testDoc1);
136 writer.addDocument(testDoc2);
137 writer.addDocument(testDoc3);
138 writer.addDocument(testDoc4);
139 writer.close();
140 IndexSearcher knownSearcher = new IndexSearcher(dir);
141 TermEnum termEnum = knownSearcher.reader.terms();
142 TermDocs termDocs = knownSearcher.reader.termDocs();
143
145 Similarity sim = knownSearcher.getSimilarity();
146 while (termEnum.next() == true)
147 {
148 Term term = termEnum.term();
149 termDocs.seek(term);
151 while (termDocs.next())
152 {
153 int docId = termDocs.doc();
154 int freq = termDocs.freq();
155 TermFreqVector vector = knownSearcher.reader.getTermFreqVector(docId, "field");
157 float tf = sim.tf(freq);
158 float idf = sim.idf(term, knownSearcher);
159 float lNorm = sim.lengthNorm("field", vector.getTerms().length);
162 assertTrue(vector != null);
165 String [] vTerms = vector.getTerms();
166 int [] freqs = vector.getTermFrequencies();
167 for (int i = 0; i < vTerms.length; i++)
168 {
169 if (term.text().equals(vTerms[i]) == true)
170 {
171 assertTrue(freqs[i] == freq);
172 }
173 }
174
175 }
176 }
178 Query query = new TermQuery(new Term("field", "chocolate"));
179 Hits hits = knownSearcher.search(query);
180 assertTrue(hits.length() == 3);
182 float score = hits.score(0);
183
189 assertTrue(testDoc3.toString().equals(hits.doc(0).toString()));
190 assertTrue(testDoc4.toString().equals(hits.doc(1).toString()));
191 assertTrue(testDoc1.toString().equals(hits.doc(2).toString()));
192 TermFreqVector vector = knownSearcher.reader.getTermFreqVector(hits.id(1), "field");
193 assertTrue(vector != null);
194 String [] terms = vector.getTerms();
196 int [] freqs = vector.getTermFrequencies();
197 assertTrue(terms != null && terms.length == 10);
198 for (int i = 0; i < terms.length; i++) {
199 String term = terms[i];
200 int freq = freqs[i];
202 assertTrue(test4.indexOf(term) != -1);
203 Integer freqInt = (Integer )test4Map.get(term);
204 assertTrue(freqInt != null);
205 assertTrue(freqInt.intValue() == freq);
206 }
207 knownSearcher.close();
208 } catch (IOException e) {
209 e.printStackTrace();
210 assertTrue(false);
211 }
212
213
214 }
215
216 private void setupDoc(Document doc, String text)
217 {
218 doc.add(Field.Text("field", text, true));
219 }
221
222
223 }
224 |
Popular Tags |
Java API By Example, From Geeks To Geeks.