1
package org.apache.lucene.analysis.cz;
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 java.io.IOException;
21
import java.io.StringReader;
23
import org.apache.lucene.analysis.BaseTokenStreamTestCase;
24
import org.apache.lucene.analysis.CharArraySet;
25
import org.apache.lucene.analysis.KeywordMarkerFilter;
26
import org.apache.lucene.analysis.MockTokenizer;
29
* Test the Czech Stemmer.
31
* Note: its algorithmic, so some stems are nonsense
34
public class TestCzechStemmer extends BaseTokenStreamTestCase {
37
* Test showing how masculine noun forms conflate
39
public void testMasculineNouns() throws IOException {
40
CzechAnalyzer cz = new CzechAnalyzer(TEST_VERSION_CURRENT);
42
/* animate ending with a hard consonant */
43
assertAnalyzesTo(cz, "pán", new String[] { "pán" });
44
assertAnalyzesTo(cz, "páni", new String[] { "pán" });
45
assertAnalyzesTo(cz, "pánové", new String[] { "pán" });
46
assertAnalyzesTo(cz, "pána", new String[] { "pán" });
47
assertAnalyzesTo(cz, "pánů", new String[] { "pán" });
48
assertAnalyzesTo(cz, "pánovi", new String[] { "pán" });
49
assertAnalyzesTo(cz, "pánům", new String[] { "pán" });
50
assertAnalyzesTo(cz, "pány", new String[] { "pán" });
51
assertAnalyzesTo(cz, "páne", new String[] { "pán" });
52
assertAnalyzesTo(cz, "pánech", new String[] { "pán" });
53
assertAnalyzesTo(cz, "pánem", new String[] { "pán" });
55
/* inanimate ending with hard consonant */
56
assertAnalyzesTo(cz, "hrad", new String[] { "hrad" });
57
assertAnalyzesTo(cz, "hradu", new String[] { "hrad" });
58
assertAnalyzesTo(cz, "hrade", new String[] { "hrad" });
59
assertAnalyzesTo(cz, "hradem", new String[] { "hrad" });
60
assertAnalyzesTo(cz, "hrady", new String[] { "hrad" });
61
assertAnalyzesTo(cz, "hradech", new String[] { "hrad" });
62
assertAnalyzesTo(cz, "hradům", new String[] { "hrad" });
63
assertAnalyzesTo(cz, "hradů", new String[] { "hrad" });
65
/* animate ending with a soft consonant */
66
assertAnalyzesTo(cz, "muž", new String[] { "muh" });
67
assertAnalyzesTo(cz, "muži", new String[] { "muh" });
68
assertAnalyzesTo(cz, "muže", new String[] { "muh" });
69
assertAnalyzesTo(cz, "mužů", new String[] { "muh" });
70
assertAnalyzesTo(cz, "mužům", new String[] { "muh" });
71
assertAnalyzesTo(cz, "mužích", new String[] { "muh" });
72
assertAnalyzesTo(cz, "mužem", new String[] { "muh" });
74
/* inanimate ending with a soft consonant */
75
assertAnalyzesTo(cz, "stroj", new String[] { "stroj" });
76
assertAnalyzesTo(cz, "stroje", new String[] { "stroj" });
77
assertAnalyzesTo(cz, "strojů", new String[] { "stroj" });
78
assertAnalyzesTo(cz, "stroji", new String[] { "stroj" });
79
assertAnalyzesTo(cz, "strojům", new String[] { "stroj" });
80
assertAnalyzesTo(cz, "strojích", new String[] { "stroj" });
81
assertAnalyzesTo(cz, "strojem", new String[] { "stroj" });
84
assertAnalyzesTo(cz, "předseda", new String[] { "předsd" });
85
assertAnalyzesTo(cz, "předsedové", new String[] { "předsd" });
86
assertAnalyzesTo(cz, "předsedy", new String[] { "předsd" });
87
assertAnalyzesTo(cz, "předsedů", new String[] { "předsd" });
88
assertAnalyzesTo(cz, "předsedovi", new String[] { "předsd" });
89
assertAnalyzesTo(cz, "předsedům", new String[] { "předsd" });
90
assertAnalyzesTo(cz, "předsedu", new String[] { "předsd" });
91
assertAnalyzesTo(cz, "předsedo", new String[] { "předsd" });
92
assertAnalyzesTo(cz, "předsedech", new String[] { "předsd" });
93
assertAnalyzesTo(cz, "předsedou", new String[] { "předsd" });
96
assertAnalyzesTo(cz, "soudce", new String[] { "soudk" });
97
assertAnalyzesTo(cz, "soudci", new String[] { "soudk" });
98
assertAnalyzesTo(cz, "soudců", new String[] { "soudk" });
99
assertAnalyzesTo(cz, "soudcům", new String[] { "soudk" });
100
assertAnalyzesTo(cz, "soudcích", new String[] { "soudk" });
101
assertAnalyzesTo(cz, "soudcem", new String[] { "soudk" });
105
* Test showing how feminine noun forms conflate
107
public void testFeminineNouns() throws IOException {
108
CzechAnalyzer cz = new CzechAnalyzer(TEST_VERSION_CURRENT);
110
/* ending with hard consonant */
111
assertAnalyzesTo(cz, "kost", new String[] { "kost" });
112
assertAnalyzesTo(cz, "kosti", new String[] { "kost" });
113
assertAnalyzesTo(cz, "kostí", new String[] { "kost" });
114
assertAnalyzesTo(cz, "kostem", new String[] { "kost" });
115
assertAnalyzesTo(cz, "kostech", new String[] { "kost" });
116
assertAnalyzesTo(cz, "kostmi", new String[] { "kost" });
118
/* ending with a soft consonant */
119
// note: in this example sing nom. and sing acc. don't conflate w/ the rest
120
assertAnalyzesTo(cz, "píseň", new String[] { "písň" });
121
assertAnalyzesTo(cz, "písně", new String[] { "písn" });
122
assertAnalyzesTo(cz, "písni", new String[] { "písn" });
123
assertAnalyzesTo(cz, "písněmi", new String[] { "písn" });
124
assertAnalyzesTo(cz, "písních", new String[] { "písn" });
125
assertAnalyzesTo(cz, "písním", new String[] { "písn" });
128
assertAnalyzesTo(cz, "růže", new String[] { "růh" });
129
assertAnalyzesTo(cz, "růží", new String[] { "růh" });
130
assertAnalyzesTo(cz, "růžím", new String[] { "růh" });
131
assertAnalyzesTo(cz, "růžích", new String[] { "růh" });
132
assertAnalyzesTo(cz, "růžemi", new String[] { "růh" });
133
assertAnalyzesTo(cz, "růži", new String[] { "růh" });
136
assertAnalyzesTo(cz, "žena", new String[] { "žn" });
137
assertAnalyzesTo(cz, "ženy", new String[] { "žn" });
138
assertAnalyzesTo(cz, "žen", new String[] { "žn" });
139
assertAnalyzesTo(cz, "ženě", new String[] { "žn" });
140
assertAnalyzesTo(cz, "ženám", new String[] { "žn" });
141
assertAnalyzesTo(cz, "ženu", new String[] { "žn" });
142
assertAnalyzesTo(cz, "ženo", new String[] { "žn" });
143
assertAnalyzesTo(cz, "ženách", new String[] { "žn" });
144
assertAnalyzesTo(cz, "ženou", new String[] { "žn" });
145
assertAnalyzesTo(cz, "ženami", new String[] { "žn" });
149
* Test showing how neuter noun forms conflate
151
public void testNeuterNouns() throws IOException {
152
CzechAnalyzer cz = new CzechAnalyzer(TEST_VERSION_CURRENT);
155
assertAnalyzesTo(cz, "město", new String[] { "měst" });
156
assertAnalyzesTo(cz, "města", new String[] { "měst" });
157
assertAnalyzesTo(cz, "měst", new String[] { "měst" });
158
assertAnalyzesTo(cz, "městu", new String[] { "měst" });
159
assertAnalyzesTo(cz, "městům", new String[] { "měst" });
160
assertAnalyzesTo(cz, "městě", new String[] { "měst" });
161
assertAnalyzesTo(cz, "městech", new String[] { "měst" });
162
assertAnalyzesTo(cz, "městem", new String[] { "měst" });
163
assertAnalyzesTo(cz, "městy", new String[] { "měst" });
166
assertAnalyzesTo(cz, "moře", new String[] { "moř" });
167
assertAnalyzesTo(cz, "moří", new String[] { "moř" });
168
assertAnalyzesTo(cz, "mořím", new String[] { "moř" });
169
assertAnalyzesTo(cz, "moři", new String[] { "moř" });
170
assertAnalyzesTo(cz, "mořích", new String[] { "moř" });
171
assertAnalyzesTo(cz, "mořem", new String[] { "moř" });
174
assertAnalyzesTo(cz, "kuře", new String[] { "kuř" });
175
assertAnalyzesTo(cz, "kuřata", new String[] { "kuř" });
176
assertAnalyzesTo(cz, "kuřete", new String[] { "kuř" });
177
assertAnalyzesTo(cz, "kuřat", new String[] { "kuř" });
178
assertAnalyzesTo(cz, "kuřeti", new String[] { "kuř" });
179
assertAnalyzesTo(cz, "kuřatům", new String[] { "kuř" });
180
assertAnalyzesTo(cz, "kuřatech", new String[] { "kuř" });
181
assertAnalyzesTo(cz, "kuřetem", new String[] { "kuř" });
182
assertAnalyzesTo(cz, "kuřaty", new String[] { "kuř" });
185
assertAnalyzesTo(cz, "stavení", new String[] { "stavn" });
186
assertAnalyzesTo(cz, "stavením", new String[] { "stavn" });
187
assertAnalyzesTo(cz, "staveních", new String[] { "stavn" });
188
assertAnalyzesTo(cz, "staveními", new String[] { "stavn" });
192
* Test showing how adjectival forms conflate
194
public void testAdjectives() throws IOException {
195
CzechAnalyzer cz = new CzechAnalyzer(TEST_VERSION_CURRENT);
197
/* ending with ý/á/é */
198
assertAnalyzesTo(cz, "mladý", new String[] { "mlad" });
199
assertAnalyzesTo(cz, "mladí", new String[] { "mlad" });
200
assertAnalyzesTo(cz, "mladého", new String[] { "mlad" });
201
assertAnalyzesTo(cz, "mladých", new String[] { "mlad" });
202
assertAnalyzesTo(cz, "mladému", new String[] { "mlad" });
203
assertAnalyzesTo(cz, "mladým", new String[] { "mlad" });
204
assertAnalyzesTo(cz, "mladé", new String[] { "mlad" });
205
assertAnalyzesTo(cz, "mladém", new String[] { "mlad" });
206
assertAnalyzesTo(cz, "mladými", new String[] { "mlad" });
207
assertAnalyzesTo(cz, "mladá", new String[] { "mlad" });
208
assertAnalyzesTo(cz, "mladou", new String[] { "mlad" });
211
assertAnalyzesTo(cz, "jarní", new String[] { "jarn" });
212
assertAnalyzesTo(cz, "jarního", new String[] { "jarn" });
213
assertAnalyzesTo(cz, "jarních", new String[] { "jarn" });
214
assertAnalyzesTo(cz, "jarnímu", new String[] { "jarn" });
215
assertAnalyzesTo(cz, "jarním", new String[] { "jarn" });
216
assertAnalyzesTo(cz, "jarními", new String[] { "jarn" });
220
* Test some possessive suffixes
222
public void testPossessive() throws IOException {
223
CzechAnalyzer cz = new CzechAnalyzer(TEST_VERSION_CURRENT);
224
assertAnalyzesTo(cz, "Karlův", new String[] { "karl" });
225
assertAnalyzesTo(cz, "jazykový", new String[] { "jazyk" });
229
* Test some exceptional rules, implemented as rewrites.
231
public void testExceptions() throws IOException {
232
CzechAnalyzer cz = new CzechAnalyzer(TEST_VERSION_CURRENT);
234
/* rewrite of št -> sk */
235
assertAnalyzesTo(cz, "český", new String[] { "česk" });
236
assertAnalyzesTo(cz, "čeští", new String[] { "česk" });
238
/* rewrite of čt -> ck */
239
assertAnalyzesTo(cz, "anglický", new String[] { "anglick" });
240
assertAnalyzesTo(cz, "angličtí", new String[] { "anglick" });
242
/* rewrite of z -> h */
243
assertAnalyzesTo(cz, "kniha", new String[] { "knih" });
244
assertAnalyzesTo(cz, "knize", new String[] { "knih" });
246
/* rewrite of ž -> h */
247
assertAnalyzesTo(cz, "mazat", new String[] { "mah" });
248
assertAnalyzesTo(cz, "mažu", new String[] { "mah" });
250
/* rewrite of c -> k */
251
assertAnalyzesTo(cz, "kluk", new String[] { "kluk" });
252
assertAnalyzesTo(cz, "kluci", new String[] { "kluk" });
253
assertAnalyzesTo(cz, "klucích", new String[] { "kluk" });
255
/* rewrite of č -> k */
256
assertAnalyzesTo(cz, "hezký", new String[] { "hezk" });
257
assertAnalyzesTo(cz, "hezčí", new String[] { "hezk" });
259
/* rewrite of *ů* -> *o* */
260
assertAnalyzesTo(cz, "hůl", new String[] { "hol" });
261
assertAnalyzesTo(cz, "hole", new String[] { "hol" });
263
/* rewrite of e* -> * */
264
assertAnalyzesTo(cz, "deska", new String[] { "desk" });
265
assertAnalyzesTo(cz, "desek", new String[] { "desk" });
269
* Test that very short words are not stemmed.
271
public void testDontStem() throws IOException {
272
CzechAnalyzer cz = new CzechAnalyzer(TEST_VERSION_CURRENT);
273
assertAnalyzesTo(cz, "e", new String[] { "e" });
274
assertAnalyzesTo(cz, "zi", new String[] { "zi" });
277
public void testWithKeywordAttribute() throws IOException {
278
CharArraySet set = new CharArraySet(TEST_VERSION_CURRENT, 1, true);
280
CzechStemFilter filter = new CzechStemFilter(new KeywordMarkerFilter(
281
new MockTokenizer(new StringReader("hole desek"), MockTokenizer.WHITESPACE, false), set));
282
assertTokenStreamContents(filter, new String[] { "hole", "desk" });