1 package org.apache.lucene.analysis;
3 import org.apache.lucene.analysis.standard.ClassicAnalyzer;
5 import org.apache.lucene.document.Document;
6 import org.apache.lucene.document.Field;
7 import org.apache.lucene.index.IndexReader;
8 import org.apache.lucene.index.IndexWriter;
9 import org.apache.lucene.index.IndexWriterConfig;
10 import org.apache.lucene.index.Term;
11 import org.apache.lucene.index.TermPositions;
12 import org.apache.lucene.store.RAMDirectory;
13 import org.apache.lucene.util.Version;
15 import java.io.IOException;
16 import java.util.Arrays;
20 * Copyright 2004 The Apache Software Foundation
22 * Licensed under the Apache License, Version 2.0 (the "License");
23 * you may not use this file except in compliance with the License.
24 * You may obtain a copy of the License at
26 * http://www.apache.org/licenses/LICENSE-2.0
28 * Unless required by applicable law or agreed to in writing, software
29 * distributed under the License is distributed on an "AS IS" BASIS,
30 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
31 * See the License for the specific language governing permissions and
32 * limitations under the License.
35 public class TestClassicAnalyzer extends BaseTokenStreamTestCase {
37 private Analyzer a = new ClassicAnalyzer(TEST_VERSION_CURRENT);
39 public void testMaxTermLength() throws Exception {
40 ClassicAnalyzer sa = new ClassicAnalyzer(TEST_VERSION_CURRENT);
41 sa.setMaxTokenLength(5);
42 assertAnalyzesTo(sa, "ab cd toolong xy z", new String[]{"ab", "cd", "xy", "z"});
45 public void testMaxTermLength2() throws Exception {
46 ClassicAnalyzer sa = new ClassicAnalyzer(TEST_VERSION_CURRENT);
47 assertAnalyzesTo(sa, "ab cd toolong xy z", new String[]{"ab", "cd", "toolong", "xy", "z"});
48 sa.setMaxTokenLength(5);
50 assertAnalyzesTo(sa, "ab cd toolong xy z", new String[]{"ab", "cd", "xy", "z"}, new int[]{1, 1, 2, 1});
53 public void testMaxTermLength3() throws Exception {
54 char[] chars = new char[255];
55 for(int i=0;i<255;i++)
57 String longTerm = new String(chars, 0, 255);
59 assertAnalyzesTo(a, "ab cd " + longTerm + " xy z", new String[]{"ab", "cd", longTerm, "xy", "z"});
60 assertAnalyzesTo(a, "ab cd " + longTerm + "a xy z", new String[]{"ab", "cd", "xy", "z"});
63 public void testAlphanumeric() throws Exception {
64 // alphanumeric tokens
65 assertAnalyzesTo(a, "B2B", new String[]{"b2b"});
66 assertAnalyzesTo(a, "2B", new String[]{"2b"});
69 public void testUnderscores() throws Exception {
70 // underscores are delimiters, but not in email addresses (below)
71 assertAnalyzesTo(a, "word_having_underscore", new String[]{"word", "having", "underscore"});
72 assertAnalyzesTo(a, "word_with_underscore_and_stopwords", new String[]{"word", "underscore", "stopwords"});
75 public void testDelimiters() throws Exception {
76 // other delimiters: "-", "/", ","
77 assertAnalyzesTo(a, "some-dashed-phrase", new String[]{"some", "dashed", "phrase"});
78 assertAnalyzesTo(a, "dogs,chase,cats", new String[]{"dogs", "chase", "cats"});
79 assertAnalyzesTo(a, "ac/dc", new String[]{"ac", "dc"});
82 public void testApostrophes() throws Exception {
83 // internal apostrophes: O'Reilly, you're, O'Reilly's
84 // possessives are actually removed by StardardFilter, not the tokenizer
85 assertAnalyzesTo(a, "O'Reilly", new String[]{"o'reilly"});
86 assertAnalyzesTo(a, "you're", new String[]{"you're"});
87 assertAnalyzesTo(a, "she's", new String[]{"she"});
88 assertAnalyzesTo(a, "Jim's", new String[]{"jim"});
89 assertAnalyzesTo(a, "don't", new String[]{"don't"});
90 assertAnalyzesTo(a, "O'Reilly's", new String[]{"o'reilly"});
93 public void testTSADash() throws Exception {
94 // t and s had been stopwords in Lucene <= 2.0, which made it impossible
95 // to correctly search for these terms:
96 assertAnalyzesTo(a, "s-class", new String[]{"s", "class"});
97 assertAnalyzesTo(a, "t-com", new String[]{"t", "com"});
98 // 'a' is still a stopword:
99 assertAnalyzesTo(a, "a-class", new String[]{"class"});
102 public void testCompanyNames() throws Exception {
104 assertAnalyzesTo(a, "AT&T", new String[]{"at&t"});
105 assertAnalyzesTo(a, "Excite@Home", new String[]{"excite@home"});
108 public void testLucene1140() throws Exception {
110 ClassicAnalyzer analyzer = new ClassicAnalyzer(TEST_VERSION_CURRENT);
111 assertAnalyzesTo(analyzer, "www.nutch.org.", new String[]{ "www.nutch.org" }, new String[] { "<HOST>" });
112 } catch (NullPointerException e) {
113 fail("Should not throw an NPE and it did");
118 public void testDomainNames() throws Exception {
119 // Current lucene should not show the bug
120 ClassicAnalyzer a2 = new ClassicAnalyzer(TEST_VERSION_CURRENT);
123 assertAnalyzesTo(a2, "www.nutch.org", new String[]{"www.nutch.org"});
124 //Notice the trailing . See https://issues.apache.org/jira/browse/LUCENE-1068.
125 // the following should be recognized as HOST:
126 assertAnalyzesTo(a2, "www.nutch.org.", new String[]{ "www.nutch.org" }, new String[] { "<HOST>" });
128 // 2.3 should show the bug
129 a2 = new ClassicAnalyzer(org.apache.lucene.util.Version.LUCENE_23);
130 assertAnalyzesTo(a2, "www.nutch.org.", new String[]{ "wwwnutchorg" }, new String[] { "<ACRONYM>" });
132 // 2.4 should not show the bug
133 a2 = new ClassicAnalyzer(Version.LUCENE_24);
134 assertAnalyzesTo(a2, "www.nutch.org.", new String[]{ "www.nutch.org" }, new String[] { "<HOST>" });
137 public void testEMailAddresses() throws Exception {
138 // email addresses, possibly with underscores, periods, etc
139 assertAnalyzesTo(a, "test@example.com", new String[]{"test@example.com"});
140 assertAnalyzesTo(a, "first.lastname@example.com", new String[]{"first.lastname@example.com"});
141 assertAnalyzesTo(a, "first_lastname@example.com", new String[]{"first_lastname@example.com"});
144 public void testNumeric() throws Exception {
145 // floating point, serial, model numbers, ip addresses, etc.
146 // every other segment must have at least one digit
147 assertAnalyzesTo(a, "21.35", new String[]{"21.35"});
148 assertAnalyzesTo(a, "R2D2 C3PO", new String[]{"r2d2", "c3po"});
149 assertAnalyzesTo(a, "216.239.63.104", new String[]{"216.239.63.104"});
150 assertAnalyzesTo(a, "1-2-3", new String[]{"1-2-3"});
151 assertAnalyzesTo(a, "a1-b2-c3", new String[]{"a1-b2-c3"});
152 assertAnalyzesTo(a, "a1-b-c3", new String[]{"a1-b-c3"});
155 public void testTextWithNumbers() throws Exception {
157 assertAnalyzesTo(a, "David has 5000 bones", new String[]{"david", "has", "5000", "bones"});
160 public void testVariousText() throws Exception {
162 assertAnalyzesTo(a, "C embedded developers wanted", new String[]{"c", "embedded", "developers", "wanted"});
163 assertAnalyzesTo(a, "foo bar FOO BAR", new String[]{"foo", "bar", "foo", "bar"});
164 assertAnalyzesTo(a, "foo bar . FOO <> BAR", new String[]{"foo", "bar", "foo", "bar"});
165 assertAnalyzesTo(a, "\"QUOTED\" word", new String[]{"quoted", "word"});
168 public void testAcronyms() throws Exception {
169 // acronyms have their dots stripped
170 assertAnalyzesTo(a, "U.S.A.", new String[]{"usa"});
173 public void testCPlusPlusHash() throws Exception {
174 // It would be nice to change the grammar in StandardTokenizer.jj to make "C#" and "C++" end up as tokens.
175 assertAnalyzesTo(a, "C++", new String[]{"c"});
176 assertAnalyzesTo(a, "C#", new String[]{"c"});
179 public void testKorean() throws Exception {
181 assertAnalyzesTo(a, "안녕하세요 한글입니다", new String[]{"안녕하세요", "한글입니다"});
184 // Compliance with the "old" JavaCC-based analyzer, see:
185 // https://issues.apache.org/jira/browse/LUCENE-966#action_12516752
187 public void testComplianceFileName() throws Exception {
188 assertAnalyzesTo(a, "2004.jpg",
189 new String[]{"2004.jpg"},
190 new String[]{"<HOST>"});
193 public void testComplianceNumericIncorrect() throws Exception {
194 assertAnalyzesTo(a, "62.46",
195 new String[]{"62.46"},
196 new String[]{"<HOST>"});
199 public void testComplianceNumericLong() throws Exception {
200 assertAnalyzesTo(a, "978-0-94045043-1",
201 new String[]{"978-0-94045043-1"},
202 new String[]{"<NUM>"});
205 public void testComplianceNumericFile() throws Exception {
208 "78academyawards/rules/rule02.html",
209 new String[]{"78academyawards/rules/rule02.html"},
210 new String[]{"<NUM>"});
213 public void testComplianceNumericWithUnderscores() throws Exception {
216 "2006-03-11t082958z_01_ban130523_rtridst_0_ozabs",
217 new String[]{"2006-03-11t082958z_01_ban130523_rtridst_0_ozabs"},
218 new String[]{"<NUM>"});
221 public void testComplianceNumericWithDash() throws Exception {
222 assertAnalyzesTo(a, "mid-20th", new String[]{"mid-20th"},
223 new String[]{"<NUM>"});
226 public void testComplianceManyTokens() throws Exception {
229 "/money.cnn.com/magazines/fortune/fortune_archive/2007/03/19/8402357/index.htm "
230 + "safari-0-sheikh-zayed-grand-mosque.jpg",
231 new String[]{"money.cnn.com", "magazines", "fortune",
232 "fortune", "archive/2007/03/19/8402357", "index.htm",
233 "safari-0-sheikh", "zayed", "grand", "mosque.jpg"},
234 new String[]{"<HOST>", "<ALPHANUM>", "<ALPHANUM>",
235 "<ALPHANUM>", "<NUM>", "<HOST>", "<NUM>", "<ALPHANUM>",
236 "<ALPHANUM>", "<HOST>"});
239 public void testJava14BWCompatibility() throws Exception {
240 ClassicAnalyzer sa = new ClassicAnalyzer(Version.LUCENE_30);
241 assertAnalyzesTo(sa, "test\u02C6test", new String[] { "test", "test" });
245 * Make sure we skip wicked long terms.
247 public void testWickedLongTerm() throws IOException {
248 RAMDirectory dir = new RAMDirectory();
249 IndexWriter writer = new IndexWriter(dir, new IndexWriterConfig(
250 TEST_VERSION_CURRENT, new ClassicAnalyzer(TEST_VERSION_CURRENT)));
252 char[] chars = new char[IndexWriter.MAX_TERM_LENGTH];
253 Arrays.fill(chars, 'x');
254 Document doc = new Document();
255 final String bigTerm = new String(chars);
257 // This produces a too-long term:
258 String contents = "abc xyz x" + bigTerm + " another term";
259 doc.add(new Field("content", contents, Field.Store.NO, Field.Index.ANALYZED));
260 writer.addDocument(doc);
262 // Make sure we can add another normal document
263 doc = new Document();
264 doc.add(new Field("content", "abc bbb ccc", Field.Store.NO, Field.Index.ANALYZED));
265 writer.addDocument(doc);
268 IndexReader reader = IndexReader.open(dir, true);
270 // Make sure all terms < max size were indexed
271 assertEquals(2, reader.docFreq(new Term("content", "abc")));
272 assertEquals(1, reader.docFreq(new Term("content", "bbb")));
273 assertEquals(1, reader.docFreq(new Term("content", "term")));
274 assertEquals(1, reader.docFreq(new Term("content", "another")));
276 // Make sure position is still incremented when
277 // massive term is skipped:
278 TermPositions tps = reader.termPositions(new Term("content", "another"));
279 assertTrue(tps.next());
280 assertEquals(1, tps.freq());
281 assertEquals(3, tps.nextPosition());
283 // Make sure the doc that has the massive term is in
285 assertEquals("document with wicked long term should is not in the index!", 2, reader.numDocs());
289 // Make sure we can add a document with exactly the
290 // maximum length term, and search on that term:
291 doc = new Document();
292 doc.add(new Field("content", bigTerm, Field.Store.NO, Field.Index.ANALYZED));
293 ClassicAnalyzer sa = new ClassicAnalyzer(TEST_VERSION_CURRENT);
294 sa.setMaxTokenLength(100000);
295 writer = new IndexWriter(dir, new IndexWriterConfig(TEST_VERSION_CURRENT, sa));
296 writer.addDocument(doc);
298 reader = IndexReader.open(dir, true);
299 assertEquals(1, reader.docFreq(new Term("content", bigTerm)));
305 /** blast some random strings through the analyzer */
306 public void testRandomStrings() throws Exception {
307 checkRandomData(random, new ClassicAnalyzer(TEST_VERSION_CURRENT), 10000*RANDOM_MULTIPLIER);