Examples with TokenStream org.apache.lucene.analysis.TokenStream used on opensource projects

Example 11 with TokenStream

use of org.apache.lucene.analysis.TokenStream in project elasticsearch by elastic.

the class SimpleUkrainianAnalyzerTests method testAnalyzer.

private static void testAnalyzer(String source, String... expected_terms) throws IOException {
    TestAnalysis analysis = createTestAnalysis(new Index("test", "_na_"), Settings.EMPTY, new AnalysisUkrainianPlugin());
    Analyzer analyzer = analysis.indexAnalyzers.get("ukrainian").analyzer();
    TokenStream ts = analyzer.tokenStream("test", source);
    CharTermAttribute term1 = ts.addAttribute(CharTermAttribute.class);
    ts.reset();
    for (String expected : expected_terms) {
        assertThat(ts.incrementToken(), equalTo(true));
        assertThat(term1.toString(), equalTo(expected));
    }
    assertThat(ts.incrementToken(), equalTo(false));
}

Example 12 with TokenStream

use of org.apache.lucene.analysis.TokenStream in project elasticsearch by elastic.

the class NGramTokenizerFactoryTests method testBackwardsCompatibilityEdgeNgramTokenFilter.

public void testBackwardsCompatibilityEdgeNgramTokenFilter() throws Exception {
    int iters = scaledRandomIntBetween(20, 100);
    for (int i = 0; i < iters; i++) {
        final Index index = new Index("test", "_na_");
        final String name = "ngr";
        Version v = randomVersion(random());
        Builder builder = newAnalysisSettingsBuilder().put("min_gram", 2).put("max_gram", 3);
        boolean reverse = random().nextBoolean();
        if (reverse) {
            builder.put("side", "back");
        }
        Settings settings = builder.build();
        Settings indexSettings = newAnalysisSettingsBuilder().put(IndexMetaData.SETTING_VERSION_CREATED, v.id).build();
        Tokenizer tokenizer = new MockTokenizer();
        tokenizer.setReader(new StringReader("foo bar"));
        TokenStream edgeNGramTokenFilter = new EdgeNGramTokenFilterFactory(IndexSettingsModule.newIndexSettings(index, indexSettings), null, name, settings).create(tokenizer);
        if (reverse) {
            assertThat(edgeNGramTokenFilter, instanceOf(ReverseStringFilter.class));
        } else {
            assertThat(edgeNGramTokenFilter, instanceOf(EdgeNGramTokenFilter.class));
        }
    }
}

Example 13 with TokenStream

use of org.apache.lucene.analysis.TokenStream in project elasticsearch by elastic.

the class CompoundAnalysisTests method analyze.

private List<String> analyze(Settings settings, String analyzerName, String text) throws IOException {
    IndexSettings idxSettings = IndexSettingsModule.newIndexSettings("test", settings);
    AnalysisModule analysisModule = new AnalysisModule(new Environment(settings), singletonList(new AnalysisPlugin() {

        @Override
        public Map<String, AnalysisProvider<TokenFilterFactory>> getTokenFilters() {
            return singletonMap("myfilter", MyFilterTokenFilterFactory::new);
        }
    }));
    IndexAnalyzers indexAnalyzers = analysisModule.getAnalysisRegistry().build(idxSettings);
    Analyzer analyzer = indexAnalyzers.get(analyzerName).analyzer();
    AllEntries allEntries = new AllEntries();
    allEntries.addText("field1", text, 1.0f);
    TokenStream stream = AllTokenStream.allTokenStream("_all", text, 1.0f, analyzer);
    stream.reset();
    CharTermAttribute termAtt = stream.addAttribute(CharTermAttribute.class);
    List<String> terms = new ArrayList<>();
    while (stream.incrementToken()) {
        String tokText = termAtt.toString();
        terms.add(tokText);
    }
    return terms;
}

Example 14 with TokenStream

use of org.apache.lucene.analysis.TokenStream in project elasticsearch by elastic.

the class StemmerTokenFilterFactoryTests method testEnglishFilterFactory.

public void testEnglishFilterFactory() throws IOException {
    int iters = scaledRandomIntBetween(20, 100);
    for (int i = 0; i < iters; i++) {
        Version v = VersionUtils.randomVersion(random());
        Settings settings = Settings.builder().put("index.analysis.filter.my_english.type", "stemmer").put("index.analysis.filter.my_english.language", "english").put("index.analysis.analyzer.my_english.tokenizer", "whitespace").put("index.analysis.analyzer.my_english.filter", "my_english").put(SETTING_VERSION_CREATED, v).put(Environment.PATH_HOME_SETTING.getKey(), createTempDir().toString()).build();
        ESTestCase.TestAnalysis analysis = AnalysisTestsHelper.createTestAnalysisFromSettings(settings);
        TokenFilterFactory tokenFilter = analysis.tokenFilter.get("my_english");
        assertThat(tokenFilter, instanceOf(StemmerTokenFilterFactory.class));
        Tokenizer tokenizer = new WhitespaceTokenizer();
        tokenizer.setReader(new StringReader("foo bar"));
        TokenStream create = tokenFilter.create(tokenizer);
        IndexAnalyzers indexAnalyzers = analysis.indexAnalyzers;
        NamedAnalyzer analyzer = indexAnalyzers.get("my_english");
        assertThat(create, instanceOf(PorterStemFilter.class));
        assertAnalyzesTo(analyzer, "consolingly", new String[] { "consolingli" });
    }
}

Example 15 with TokenStream

use of org.apache.lucene.analysis.TokenStream in project elasticsearch by elastic.

the class DocumentFieldMapperTests method assertAnalyzes.

private void assertAnalyzes(Analyzer analyzer, String field, String output) throws IOException {
    try (TokenStream tok = analyzer.tokenStream(field, new StringReader(""))) {
        CharTermAttribute term = tok.addAttribute(CharTermAttribute.class);
        assertTrue(tok.incrementToken());
        assertEquals(output, term.toString());
    }
}