1 package org.wikimedia.search.extra.analysis.turkish;
2
3 import static org.hamcrest.CoreMatchers.equalTo;
4
5 import java.io.IOException;
6
7 import org.apache.lucene.analysis.Analyzer;
8 import org.apache.lucene.analysis.TokenStream;
9 import org.apache.lucene.analysis.tokenattributes.CharTermAttribute;
10 import org.elasticsearch.Version;
11 import org.elasticsearch.cluster.metadata.IndexMetadata;
12 import org.elasticsearch.common.settings.Settings;
13 import org.elasticsearch.env.Environment;
14 import org.elasticsearch.index.IndexSettings;
15 import org.elasticsearch.index.analysis.IndexAnalyzers;
16 import org.elasticsearch.test.ESTestCase;
17 import org.elasticsearch.test.IndexSettingsModule;
18 import org.hamcrest.MatcherAssert;
19 import org.junit.Test;
20
21 public class BetterApostropheFilterESTest extends ESTestCase {
22 private IndexAnalyzers indexAnalyzers;
23
24 @Test
25 public void testPrebuilt() throws IOException {
26 assertAnalyzerAvailable("turkish_prebuilt", "prebuilt.json");
27 }
28
29 @Test
30 public void testRedefined() throws IOException {
31 assertAnalyzerAvailable("turkish_redefined", "redefined.json");
32 }
33
34 private void assertAnalyzerAvailable(String analyzerName, String analysisResource) throws IOException {
35 Settings indexSettings = settings(Version.CURRENT)
36 .loadFromStream(analysisResource, this.getClass().getResourceAsStream(analysisResource), false)
37 .put(IndexMetadata.SETTING_VERSION_CREATED, Version.CURRENT)
38 .build();
39 IndexSettings indexProps = IndexSettingsModule.newIndexSettings("test", indexSettings);
40 Settings settings = Settings.builder()
41 .put(Environment.PATH_HOME_SETTING.getKey(), createTempDir())
42 .build();
43 indexAnalyzers = createTestAnalysis(indexProps, settings, new ExtraAnalysisTurkishPlugin()).indexAnalyzers;
44 match(analyzerName,
45 "Vikipedi'nin ana sunucuları ABD'nin Florida Eyaleti'ndeki Tampa kentinde",
46 "Vikipedi ana sunucuları ABD Florida Eyaleti Tampa kentinde");
47 }
48
49 private void match(String analyzerName, String source, String target) throws IOException {
50 Analyzer analyzer = indexAnalyzers.get(analyzerName).analyzer();
51
52 TokenStream stream = analyzer.tokenStream("_all", source);
53 stream.reset();
54 CharTermAttribute termAtt = stream.addAttribute(CharTermAttribute.class);
55
56 StringBuilder sb = new StringBuilder();
57 while (stream.incrementToken()) {
58 sb.append(termAtt.toString()).append(" ");
59 }
60
61 MatcherAssert.assertThat(target, equalTo(sb.toString().trim()));
62 }
63 }