Package org.apache.commons.codec.language.bm

Examples of org.apache.commons.codec.language.bm.PhoneticEngine$PhonemeBuilder

73
74
75
76
77
78
79
80
81
82
83
      @Override
      protected TokenStreamComponents createComponents(String fieldName, Reader reader) {
        Tokenizer tokenizer = new MockTokenizer(reader, MockTokenizer.WHITESPACE, false);
        return new TokenStreamComponents(tokenizer,
            new BeiderMorseFilter(tokenizer,
                new PhoneticEngine(NameType.GENERIC, RuleType.EXACT, true), languages));
      }
    };
    assertAnalyzesTo(analyzer, "Angelo",
        new String[] { "andZelo", "angelo", "anxelo" },
        new int[] { 0, 0, 0, },
View Full Code Here
101
102
103
104
105
106
107
108
109
110
  public void testEmptyTerm() throws IOException {
    Analyzer a = new Analyzer() {
      @Override
      protected TokenStreamComponents createComponents(String fieldName, Reader reader) {
        Tokenizer tokenizer = new KeywordTokenizer(reader);
        return new TokenStreamComponents(tokenizer, new BeiderMorseFilter(tokenizer, new PhoneticEngine(NameType.GENERIC, RuleType.EXACT, true)));
      }
    };
    checkOneTerm(a, "", "");
  }
View Full Code Here
110
111
112
113
114
115
116
117
118
119
120
  }
 
  public void testCustomAttribute() throws IOException {
    TokenStream stream = new KeywordTokenizer(new StringReader("D'Angelo"));
    stream = new PatternKeywordMarkerFilter(stream, Pattern.compile(".*"));
    stream = new BeiderMorseFilter(stream, new PhoneticEngine(NameType.GENERIC, RuleType.EXACT, true));
    KeywordAttribute keyAtt = stream.addAttribute(KeywordAttribute.class);
    stream.reset();
    int i = 0;
    while(stream.incrementToken()) {
      assertTrue(keyAtt.isKeyword());
View Full Code Here
115
116
117
118
119
120
121
122
123
124
125
126
127
    public TokenStream create(TokenStream tokenStream) {
        if (encoder == null) {
            if (ruletype != null && nametype != null) {
                if (languageset != null) {
                    final LanguageSet languages = LanguageSet.from(new HashSet(Arrays.asList(languageset)));
                    return new BeiderMorseFilter(tokenStream, new PhoneticEngine(nametype, ruletype, true), languages);
                }
                return new BeiderMorseFilter(tokenStream, new PhoneticEngine(nametype, ruletype, true));
            }
            if (maxcodelength > 0) {
                return new DoubleMetaphoneFilter(tokenStream, maxcodelength, !replace);
            }
        } else {
View Full Code Here
69
70
71
72
73
74
75
76
77
78
79
      @Override
      protected TokenStreamComponents createComponents(String fieldName, Reader reader) {
        Tokenizer tokenizer = new MockTokenizer(reader, MockTokenizer.WHITESPACE, false);
        return new TokenStreamComponents(tokenizer,
            new BeiderMorseFilter(tokenizer,
                new PhoneticEngine(NameType.GENERIC, RuleType.EXACT, true), languages));
      }
    };
    assertAnalyzesTo(analyzer, "Angelo",
        new String[] { "andZelo", "angelo", "anxelo" },
        new int[] { 0, 0, 0, },
View Full Code Here
98
99
100
101
102
103
104
105
106
107
  public void testEmptyTerm() throws IOException {
    Analyzer a = new ReusableAnalyzerBase() {
      @Override
      protected TokenStreamComponents createComponents(String fieldName, Reader reader) {
        Tokenizer tokenizer = new KeywordTokenizer(reader);
        return new TokenStreamComponents(tokenizer, new BeiderMorseFilter(tokenizer, new PhoneticEngine(NameType.GENERIC, RuleType.EXACT, true)));
      }
    };
    checkOneTermReuse(a, "", "");
  }
View Full Code Here
TOP

Related Classes of org.apache.commons.codec.language.bm.PhoneticEngine$PhonemeBuilder

  • org.apache.commons.codec.language.bm.Languages.LanguageSet
  • org.apache.commons.codec.language.bm.Rule.Phoneme
  • org.apache.lucene.analysis.phonetic.BeiderMorseFilterFactory
  • org.apache.lucene.analysis.phonetic.TestBeiderMorseFilter
  • org.elasticsearch.index.analysis.PhoneticTokenFilterFactory
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.