Examples with TokenizerFactory org.opensearch.index.analysis.TokenizerFactory used on opensource projects

Example 1 with TokenizerFactory

use of org.opensearch.index.analysis.TokenizerFactory in project OpenSearch by opensearch-project.

the class ScriptedConditionTokenFilterFactory method getChainAwareTokenFilterFactory.

@Override
public TokenFilterFactory getChainAwareTokenFilterFactory(TokenizerFactory tokenizer, List<CharFilterFactory> charFilters, List<TokenFilterFactory> previousTokenFilters, Function<String, TokenFilterFactory> allFilters) {
    List<TokenFilterFactory> filters = new ArrayList<>();
    List<TokenFilterFactory> existingChain = new ArrayList<>(previousTokenFilters);
    for (String filter : filterNames) {
        TokenFilterFactory tff = allFilters.apply(filter);
        if (tff == null) {
            throw new IllegalArgumentException("ScriptedConditionTokenFilter [" + name() + "] refers to undefined token filter [" + filter + "]");
        }
        tff = tff.getChainAwareTokenFilterFactory(tokenizer, charFilters, existingChain, allFilters);
        filters.add(tff);
        existingChain.add(tff);
    }
    return new TokenFilterFactory() {

        @Override
        public String name() {
            return ScriptedConditionTokenFilterFactory.this.name();
        }

        @Override
        public TokenStream create(TokenStream tokenStream) {
            Function<TokenStream, TokenStream> filter = in -> {
                for (TokenFilterFactory tff : filters) {
                    in = tff.create(in);
                }
                return in;
            };
            return new ScriptedConditionTokenFilter(tokenStream, filter, factory.newInstance());
        }
    };
}

Example 2 with TokenizerFactory

use of org.opensearch.index.analysis.TokenizerFactory in project OpenSearch by opensearch-project.

the class AnalysisModuleTests method testPluginPreConfiguredCharFilters.

/**
 * Tests that plugins can register pre-configured char filters that vary in behavior based on OpenSearch version, Lucene version,
 * and that do not vary based on version at all.
 */
public void testPluginPreConfiguredCharFilters() throws IOException {
    boolean noVersionSupportsMultiTerm = randomBoolean();
    boolean luceneVersionSupportsMultiTerm = randomBoolean();
    boolean opensearchVersionSupportsMultiTerm = randomBoolean();
    AnalysisRegistry registry = new AnalysisModule(TestEnvironment.newEnvironment(emptyNodeSettings), singletonList(new AnalysisPlugin() {

        @Override
        public List<PreConfiguredCharFilter> getPreConfiguredCharFilters() {
            return Arrays.asList(PreConfiguredCharFilter.singleton("no_version", noVersionSupportsMultiTerm, tokenStream -> new AppendCharFilter(tokenStream, "no_version")), PreConfiguredCharFilter.luceneVersion("lucene_version", luceneVersionSupportsMultiTerm, (tokenStream, luceneVersion) -> new AppendCharFilter(tokenStream, luceneVersion.toString())), PreConfiguredCharFilter.openSearchVersion("opensearch_version", opensearchVersionSupportsMultiTerm, (tokenStream, esVersion) -> new AppendCharFilter(tokenStream, esVersion.toString())));
        }

        @Override
        public Map<String, AnalysisProvider<TokenizerFactory>> getTokenizers() {
            // Need mock keyword tokenizer here, because alpha / beta versions are broken up by the dash.
            return singletonMap("keyword", (indexSettings, environment, name, settings) -> TokenizerFactory.newFactory(name, () -> new MockTokenizer(MockTokenizer.KEYWORD, false)));
        }
    })).getAnalysisRegistry();
    Version version = VersionUtils.randomVersion(random());
    IndexAnalyzers analyzers = getIndexAnalyzers(registry, Settings.builder().put("index.analysis.analyzer.no_version.tokenizer", "keyword").put("index.analysis.analyzer.no_version.char_filter", "no_version").put("index.analysis.analyzer.lucene_version.tokenizer", "keyword").put("index.analysis.analyzer.lucene_version.char_filter", "lucene_version").put("index.analysis.analyzer.opensearch_version.tokenizer", "keyword").put("index.analysis.analyzer.opensearch_version.char_filter", "opensearch_version").put(IndexMetadata.SETTING_VERSION_CREATED, version).build());
    assertTokenStreamContents(analyzers.get("no_version").tokenStream("", "test"), new String[] { "testno_version" });
    assertTokenStreamContents(analyzers.get("lucene_version").tokenStream("", "test"), new String[] { "test" + version.luceneVersion });
    assertTokenStreamContents(analyzers.get("opensearch_version").tokenStream("", "test"), new String[] { "test" + version });
    assertEquals("test" + (noVersionSupportsMultiTerm ? "no_version" : ""), analyzers.get("no_version").normalize("", "test").utf8ToString());
    assertEquals("test" + (luceneVersionSupportsMultiTerm ? version.luceneVersion.toString() : ""), analyzers.get("lucene_version").normalize("", "test").utf8ToString());
    assertEquals("test" + (opensearchVersionSupportsMultiTerm ? version.toString() : ""), analyzers.get("opensearch_version").normalize("", "test").utf8ToString());
}

Example 3 with TokenizerFactory

use of org.opensearch.index.analysis.TokenizerFactory in project OpenSearch by opensearch-project.

the class SynonymsAnalysisTests method testTokenFiltersBypassSynonymAnalysis.

public void testTokenFiltersBypassSynonymAnalysis() throws IOException {
    Settings settings = Settings.builder().put(IndexMetadata.SETTING_VERSION_CREATED, Version.CURRENT).put("path.home", createTempDir().toString()).putList("word_list", "a").put("hyphenation_patterns_path", "foo").build();
    IndexSettings idxSettings = IndexSettingsModule.newIndexSettings("index", settings);
    String[] bypassingFactories = new String[] { "dictionary_decompounder" };
    CommonAnalysisPlugin plugin = new CommonAnalysisPlugin();
    for (String factory : bypassingFactories) {
        TokenFilterFactory tff = plugin.getTokenFilters().get(factory).get(idxSettings, null, factory, settings);
        TokenizerFactory tok = new KeywordTokenizerFactory(idxSettings, null, "keyword", settings);
        SynonymTokenFilterFactory stff = new SynonymTokenFilterFactory(idxSettings, null, "synonym", settings);
        Analyzer analyzer = stff.buildSynonymAnalyzer(tok, Collections.emptyList(), Collections.singletonList(tff), null);
        try (TokenStream ts = analyzer.tokenStream("field", "text")) {
            assertThat(ts, instanceOf(KeywordTokenizer.class));
        }
    }
}

Example 4 with TokenizerFactory

use of org.opensearch.index.analysis.TokenizerFactory in project OpenSearch by opensearch-project.

the class SynonymsAnalysisTests method testDisallowedTokenFilters.

public void testDisallowedTokenFilters() throws IOException {
    Settings settings = Settings.builder().put(IndexMetadata.SETTING_VERSION_CREATED, VersionUtils.randomVersionBetween(random(), LegacyESVersion.V_7_0_0, Version.CURRENT)).put("path.home", createTempDir().toString()).putList("common_words", "a", "b").put("output_unigrams", "true").build();
    IndexSettings idxSettings = IndexSettingsModule.newIndexSettings("index", settings);
    CommonAnalysisPlugin plugin = new CommonAnalysisPlugin();
    String[] disallowedFactories = new String[] { "multiplexer", "cjk_bigram", "common_grams", "ngram", "edge_ngram", "word_delimiter", "word_delimiter_graph", "fingerprint" };
    for (String factory : disallowedFactories) {
        TokenFilterFactory tff = plugin.getTokenFilters().get(factory).get(idxSettings, null, factory, settings);
        TokenizerFactory tok = new KeywordTokenizerFactory(idxSettings, null, "keyword", settings);
        SynonymTokenFilterFactory stff = new SynonymTokenFilterFactory(idxSettings, null, "synonym", settings);
        IllegalArgumentException e = expectThrows(IllegalArgumentException.class, "Expected IllegalArgumentException for factory " + factory, () -> stff.buildSynonymAnalyzer(tok, Collections.emptyList(), Collections.singletonList(tff), null));
        assertEquals(factory, "Token filter [" + factory + "] cannot be used to parse synonyms", e.getMessage());
    }
}

Example 5 with TokenizerFactory

use of org.opensearch.index.analysis.TokenizerFactory in project OpenSearch by opensearch-project.

the class TransportAnalyzeAction method detailAnalyze.

private static AnalyzeAction.DetailAnalyzeResponse detailAnalyze(AnalyzeAction.Request request, Analyzer analyzer, int maxTokenCount) {
    AnalyzeAction.DetailAnalyzeResponse detailResponse;
    final Set<String> includeAttributes = new HashSet<>();
    if (request.attributes() != null) {
        for (String attribute : request.attributes()) {
            includeAttributes.add(attribute.toLowerCase(Locale.ROOT));
        }
    }
    // maybe unwrap analyzer from NamedAnalyzer
    Analyzer potentialCustomAnalyzer = analyzer;
    if (analyzer instanceof NamedAnalyzer) {
        potentialCustomAnalyzer = ((NamedAnalyzer) analyzer).analyzer();
    }
    if (potentialCustomAnalyzer instanceof AnalyzerComponentsProvider) {
        AnalyzerComponentsProvider customAnalyzer = (AnalyzerComponentsProvider) potentialCustomAnalyzer;
        // note: this is not field-name dependent in our cases so we can leave out the argument
        int positionIncrementGap = potentialCustomAnalyzer.getPositionIncrementGap("");
        int offsetGap = potentialCustomAnalyzer.getOffsetGap("");
        AnalyzerComponents components = customAnalyzer.getComponents();
        // divide charfilter, tokenizer tokenfilters
        CharFilterFactory[] charFilterFactories = components.getCharFilters();
        TokenizerFactory tokenizerFactory = components.getTokenizerFactory();
        TokenFilterFactory[] tokenFilterFactories = components.getTokenFilters();
        String[][] charFiltersTexts = new String[charFilterFactories != null ? charFilterFactories.length : 0][request.text().length];
        TokenListCreator[] tokenFiltersTokenListCreator = new TokenListCreator[tokenFilterFactories != null ? tokenFilterFactories.length : 0];
        TokenListCreator tokenizerTokenListCreator = new TokenListCreator(maxTokenCount);
        for (int textIndex = 0; textIndex < request.text().length; textIndex++) {
            String charFilteredSource = request.text()[textIndex];
            Reader reader = new StringReader(charFilteredSource);
            if (charFilterFactories != null) {
                for (int charFilterIndex = 0; charFilterIndex < charFilterFactories.length; charFilterIndex++) {
                    reader = charFilterFactories[charFilterIndex].create(reader);
                    Reader readerForWriteOut = new StringReader(charFilteredSource);
                    readerForWriteOut = charFilterFactories[charFilterIndex].create(readerForWriteOut);
                    charFilteredSource = writeCharStream(readerForWriteOut);
                    charFiltersTexts[charFilterIndex][textIndex] = charFilteredSource;
                }
            }
            // analyzing only tokenizer
            Tokenizer tokenizer = tokenizerFactory.create();
            tokenizer.setReader(reader);
            tokenizerTokenListCreator.analyze(tokenizer, includeAttributes, positionIncrementGap, offsetGap);
            // analyzing each tokenfilter
            if (tokenFilterFactories != null) {
                for (int tokenFilterIndex = 0; tokenFilterIndex < tokenFilterFactories.length; tokenFilterIndex++) {
                    if (tokenFiltersTokenListCreator[tokenFilterIndex] == null) {
                        tokenFiltersTokenListCreator[tokenFilterIndex] = new TokenListCreator(maxTokenCount);
                    }
                    TokenStream stream = createStackedTokenStream(request.text()[textIndex], charFilterFactories, tokenizerFactory, tokenFilterFactories, tokenFilterIndex + 1);
                    tokenFiltersTokenListCreator[tokenFilterIndex].analyze(stream, includeAttributes, positionIncrementGap, offsetGap);
                }
            }
        }
        AnalyzeAction.CharFilteredText[] charFilteredLists = new AnalyzeAction.CharFilteredText[charFiltersTexts.length];
        if (charFilterFactories != null) {
            for (int charFilterIndex = 0; charFilterIndex < charFiltersTexts.length; charFilterIndex++) {
                charFilteredLists[charFilterIndex] = new AnalyzeAction.CharFilteredText(charFilterFactories[charFilterIndex].name(), charFiltersTexts[charFilterIndex]);
            }
        }
        AnalyzeAction.AnalyzeTokenList[] tokenFilterLists = new AnalyzeAction.AnalyzeTokenList[tokenFiltersTokenListCreator.length];
        if (tokenFilterFactories != null) {
            for (int tokenFilterIndex = 0; tokenFilterIndex < tokenFiltersTokenListCreator.length; tokenFilterIndex++) {
                tokenFilterLists[tokenFilterIndex] = new AnalyzeAction.AnalyzeTokenList(tokenFilterFactories[tokenFilterIndex].name(), tokenFiltersTokenListCreator[tokenFilterIndex].getArrayTokens());
            }
        }
        detailResponse = new AnalyzeAction.DetailAnalyzeResponse(charFilteredLists, new AnalyzeAction.AnalyzeTokenList(tokenizerFactory.name(), tokenizerTokenListCreator.getArrayTokens()), tokenFilterLists);
    } else {
        String name;
        if (analyzer instanceof NamedAnalyzer) {
            name = ((NamedAnalyzer) analyzer).name();
        } else {
            name = analyzer.getClass().getName();
        }
        TokenListCreator tokenListCreator = new TokenListCreator(maxTokenCount);
        for (String text : request.text()) {
            tokenListCreator.analyze(analyzer.tokenStream("", text), includeAttributes, analyzer.getPositionIncrementGap(""), analyzer.getOffsetGap(""));
        }
        detailResponse = new AnalyzeAction.DetailAnalyzeResponse(new AnalyzeAction.AnalyzeTokenList(name, tokenListCreator.getArrayTokens()));
    }
    return detailResponse;
}