1 package org.wikimedia.search.extra;
2
3 import java.util.Arrays;
4 import java.util.Collection;
5 import java.util.Collections;
6 import java.util.HashMap;
7 import java.util.Map;
8 import java.util.regex.Pattern;
9
10 import org.apache.lucene.analysis.Analyzer;
11 import org.apache.lucene.analysis.CharArraySet;
12 import org.apache.lucene.analysis.LowerCaseFilter;
13 import org.apache.lucene.analysis.TokenStream;
14 import org.apache.lucene.analysis.core.WhitespaceTokenizer;
15 import org.apache.lucene.analysis.el.GreekLowerCaseFilter;
16 import org.apache.lucene.analysis.en.EnglishAnalyzer;
17 import org.apache.lucene.analysis.ga.IrishLowerCaseFilter;
18 import org.apache.lucene.analysis.miscellaneous.KeywordRepeatFilter;
19 import org.apache.lucene.analysis.ngram.NGramTokenizer;
20 import org.apache.lucene.analysis.pattern.PatternReplaceFilter;
21 import org.apache.lucene.analysis.tr.TurkishLowerCaseFilter;
22 import org.elasticsearch.index.analysis.AbstractIndexAnalyzerProvider;
23 import org.elasticsearch.index.analysis.Analysis;
24 import org.elasticsearch.index.analysis.AnalyzerProvider;
25 import org.elasticsearch.index.analysis.TokenFilterFactory;
26 import org.elasticsearch.index.analysis.TokenizerFactory;
27 import org.elasticsearch.indices.analysis.AnalysisModule;
28 import org.elasticsearch.plugins.AnalysisPlugin;
29 import org.elasticsearch.plugins.Plugin;
30 import org.elasticsearch.test.ESIntegTestCase;
31 import org.elasticsearch.test.ESIntegTestCase.ClusterScope;
32
33 @ClusterScope(scope = ESIntegTestCase.Scope.SUITE, transportClientRatio = 0.0)
34 public class AbstractPluginIntegrationTest extends ESIntegTestCase {
35 @Override
36 protected Collection<Class<? extends Plugin>> nodePlugins() {
37 return Collections.<Class<? extends Plugin>>unmodifiableList(Arrays.asList(ExtraCorePlugin.class, MockPlugin.class));
38 }
39
40 public static class MockPlugin extends Plugin implements AnalysisPlugin {
41 @Override
42 public Map<String, AnalysisModule.AnalysisProvider<TokenFilterFactory>> getTokenFilters() {
43 Map<String, AnalysisModule.AnalysisProvider<TokenFilterFactory>> map = new HashMap<>();
44 map.put("lowercase", (isettings, env, name, settings) -> new TokenFilterFactory() {
45 @Override
46 public String name() {
47 return name;
48 }
49
50 @Override
51 public TokenStream create(TokenStream tokenStream) {
52 String lang = settings.get("language");
53 switch (lang) {
54 case "greek":
55 return new GreekLowerCaseFilter(tokenStream);
56 case "irish":
57 return new IrishLowerCaseFilter(tokenStream);
58 case "turkish":
59 return new TurkishLowerCaseFilter(tokenStream);
60 default:
61 return new LowerCaseFilter(tokenStream);
62 }
63 }
64 });
65 map.put("pattern_replace", (isettings, env, name, settings) -> new TokenFilterFactory() {
66 @Override
67 public String name() {
68 return name;
69 }
70
71 @Override
72 public TokenStream create(TokenStream tokenStream) {
73 Pattern p = Pattern.compile(settings.get("pattern"));
74 String repl = settings.get("replacement");
75 return new PatternReplaceFilter(tokenStream, p, repl, true);
76 }
77 });
78 map.put("keyword_repeat", (isettings, env, name, settings) -> new TokenFilterFactory() {
79 @Override
80 public String name() {
81 return name;
82 }
83
84 @Override
85 public TokenStream create(TokenStream tokenStream) {
86 return new KeywordRepeatFilter(tokenStream);
87 }
88 });
89 return Collections.unmodifiableMap(map);
90 }
91
92 @Override
93 public Map<String, AnalysisModule.AnalysisProvider<TokenizerFactory>> getTokenizers() {
94 Map<String, AnalysisModule.AnalysisProvider<TokenizerFactory>> map = new HashMap<>();
95 map.put("whitespace", (isettings, env, name, settings) -> TokenizerFactory.newFactory("whitespace", WhitespaceTokenizer::new));
96 map.put("nGram", (isettings, env, name, settings) ->
97 TokenizerFactory.newFactory("nGram", () ->
98 new NGramTokenizer(settings.getAsInt("min_gram", 3), settings.getAsInt("max_gram", 3))));
99
100 return Collections.unmodifiableMap(map);
101 }
102
103 @Override
104 public Map<String, AnalysisModule.AnalysisProvider<AnalyzerProvider<? extends Analyzer>>> getAnalyzers() {
105 return Collections.singletonMap("english",
106 (isettings, env, name, settings) -> new AbstractIndexAnalyzerProvider<Analyzer>(isettings, name, settings) {
107 @Override
108 public Analyzer get() {
109 return new EnglishAnalyzer(
110 Analysis.parseStopWords(env, settings, EnglishAnalyzer.getDefaultStopSet()),
111 Analysis.parseStemExclusion(settings, CharArraySet.EMPTY_SET));
112 }
113 }
114 );
115 }
116 }
117 }