Examples with TokenizerFactory org.apache.lucene.analysis.util.TokenizerFactory used on opensource projects

Example 11 with TokenizerFactory

use of org.apache.lucene.analysis.util.TokenizerFactory in project lucene-solr by apache.

the class FieldType method getAnalyzerProperties.

/** 
   * Returns a description of the given analyzer, by either reporting the Analyzer class
   * name (and optionally luceneMatchVersion) if it's not a TokenizerChain, or if it is,
   * querying each analysis factory for its name and args.
   */
protected static SimpleOrderedMap<Object> getAnalyzerProperties(Analyzer analyzer) {
    SimpleOrderedMap<Object> analyzerProps = new SimpleOrderedMap<>();
    if (analyzer instanceof TokenizerChain) {
        Map<String, String> factoryArgs;
        TokenizerChain tokenizerChain = (TokenizerChain) analyzer;
        CharFilterFactory[] charFilterFactories = tokenizerChain.getCharFilterFactories();
        if (0 < charFilterFactories.length) {
            List<SimpleOrderedMap<Object>> charFilterProps = new ArrayList<>();
            for (CharFilterFactory charFilterFactory : charFilterFactories) {
                SimpleOrderedMap<Object> props = new SimpleOrderedMap<>();
                props.add(CLASS_NAME, charFilterFactory.getClassArg());
                factoryArgs = charFilterFactory.getOriginalArgs();
                if (null != factoryArgs) {
                    for (String key : factoryArgs.keySet()) {
                        if (!CLASS_NAME.equals(key)) {
                            if (LUCENE_MATCH_VERSION_PARAM.equals(key)) {
                                if (charFilterFactory.isExplicitLuceneMatchVersion()) {
                                    props.add(key, factoryArgs.get(key));
                                }
                            } else {
                                props.add(key, factoryArgs.get(key));
                            }
                        }
                    }
                }
                charFilterProps.add(props);
            }
            analyzerProps.add(CHAR_FILTERS, charFilterProps);
        }
        SimpleOrderedMap<Object> tokenizerProps = new SimpleOrderedMap<>();
        TokenizerFactory tokenizerFactory = tokenizerChain.getTokenizerFactory();
        tokenizerProps.add(CLASS_NAME, tokenizerFactory.getClassArg());
        factoryArgs = tokenizerFactory.getOriginalArgs();
        if (null != factoryArgs) {
            for (String key : factoryArgs.keySet()) {
                if (!CLASS_NAME.equals(key)) {
                    if (LUCENE_MATCH_VERSION_PARAM.equals(key)) {
                        if (tokenizerFactory.isExplicitLuceneMatchVersion()) {
                            tokenizerProps.add(key, factoryArgs.get(key));
                        }
                    } else {
                        tokenizerProps.add(key, factoryArgs.get(key));
                    }
                }
            }
        }
        analyzerProps.add(TOKENIZER, tokenizerProps);
        TokenFilterFactory[] filterFactories = tokenizerChain.getTokenFilterFactories();
        if (0 < filterFactories.length) {
            List<SimpleOrderedMap<Object>> filterProps = new ArrayList<>();
            for (TokenFilterFactory filterFactory : filterFactories) {
                SimpleOrderedMap<Object> props = new SimpleOrderedMap<>();
                props.add(CLASS_NAME, filterFactory.getClassArg());
                factoryArgs = filterFactory.getOriginalArgs();
                if (null != factoryArgs) {
                    for (String key : factoryArgs.keySet()) {
                        if (!CLASS_NAME.equals(key)) {
                            if (LUCENE_MATCH_VERSION_PARAM.equals(key)) {
                                if (filterFactory.isExplicitLuceneMatchVersion()) {
                                    props.add(key, factoryArgs.get(key));
                                }
                            } else {
                                props.add(key, factoryArgs.get(key));
                            }
                        }
                    }
                }
                filterProps.add(props);
            }
            analyzerProps.add(FILTERS, filterProps);
        }
    } else {
        // analyzer is not instanceof TokenizerChain
        analyzerProps.add(CLASS_NAME, analyzer.getClass().getName());
        if (analyzer.getVersion() != Version.LATEST) {
            analyzerProps.add(LUCENE_MATCH_VERSION_PARAM, analyzer.getVersion().toString());
        }
    }
    return analyzerProps;
}

Example 12 with TokenizerFactory

use of org.apache.lucene.analysis.util.TokenizerFactory in project lucene-solr by apache.

the class FieldTypePluginLoader method readAnalyzer.

//
// <analyzer><tokenizer class="...."/><tokenizer class="...." arg="....">
//
//
private Analyzer readAnalyzer(Node node) throws XPathExpressionException {
    final SolrResourceLoader loader = schema.getResourceLoader();
    if (node == null)
        return null;
    NamedNodeMap attrs = node.getAttributes();
    String analyzerName = DOMUtil.getAttr(attrs, "class");
    // check for all of these up front, so we can error if used in 
    // conjunction with an explicit analyzer class.
    NodeList charFilterNodes = (NodeList) xpath.evaluate("./charFilter", node, XPathConstants.NODESET);
    NodeList tokenizerNodes = (NodeList) xpath.evaluate("./tokenizer", node, XPathConstants.NODESET);
    NodeList tokenFilterNodes = (NodeList) xpath.evaluate("./filter", node, XPathConstants.NODESET);
    if (analyzerName != null) {
        // own custom nodes (ie: <description> or something like that)
        if (0 != charFilterNodes.getLength() || 0 != tokenizerNodes.getLength() || 0 != tokenFilterNodes.getLength()) {
            throw new SolrException(SolrException.ErrorCode.SERVER_ERROR, "Configuration Error: Analyzer class='" + analyzerName + "' can not be combined with nested analysis factories");
        }
        try {
            // No need to be core-aware as Analyzers are not in the core-aware list
            final Class<? extends Analyzer> clazz = loader.findClass(analyzerName, Analyzer.class);
            Analyzer analyzer = clazz.newInstance();
            final String matchVersionStr = DOMUtil.getAttr(attrs, LUCENE_MATCH_VERSION_PARAM);
            final Version luceneMatchVersion = (matchVersionStr == null) ? schema.getDefaultLuceneMatchVersion() : Config.parseLuceneVersionString(matchVersionStr);
            if (luceneMatchVersion == null) {
                throw new SolrException(SolrException.ErrorCode.SERVER_ERROR, "Configuration Error: Analyzer '" + clazz.getName() + "' needs a 'luceneMatchVersion' parameter");
            }
            analyzer.setVersion(luceneMatchVersion);
            return analyzer;
        } catch (Exception e) {
            log.error("Cannot load analyzer: " + analyzerName, e);
            throw new SolrException(SolrException.ErrorCode.SERVER_ERROR, "Cannot load analyzer: " + analyzerName, e);
        }
    }
    // Load the CharFilters
    final ArrayList<CharFilterFactory> charFilters = new ArrayList<>();
    AbstractPluginLoader<CharFilterFactory> charFilterLoader = new AbstractPluginLoader<CharFilterFactory>("[schema.xml] analyzer/charFilter", CharFilterFactory.class, false, false) {

        @Override
        protected CharFilterFactory create(SolrResourceLoader loader, String name, String className, Node node) throws Exception {
            final Map<String, String> params = DOMUtil.toMap(node.getAttributes());
            String configuredVersion = params.remove(LUCENE_MATCH_VERSION_PARAM);
            params.put(LUCENE_MATCH_VERSION_PARAM, parseConfiguredVersion(configuredVersion, CharFilterFactory.class.getSimpleName()).toString());
            CharFilterFactory factory = loader.newInstance(className, CharFilterFactory.class, getDefaultPackages(), new Class[] { Map.class }, new Object[] { params });
            factory.setExplicitLuceneMatchVersion(null != configuredVersion);
            return factory;
        }

        @Override
        protected void init(CharFilterFactory plugin, Node node) throws Exception {
            if (plugin != null) {
                charFilters.add(plugin);
            }
        }

        @Override
        protected CharFilterFactory register(String name, CharFilterFactory plugin) {
            // used for map registration
            return null;
        }
    };
    charFilterLoader.load(loader, charFilterNodes);
    // Load the Tokenizer
    // Although an analyzer only allows a single Tokenizer, we load a list to make sure
    // the configuration is ok
    final ArrayList<TokenizerFactory> tokenizers = new ArrayList<>(1);
    AbstractPluginLoader<TokenizerFactory> tokenizerLoader = new AbstractPluginLoader<TokenizerFactory>("[schema.xml] analyzer/tokenizer", TokenizerFactory.class, false, false) {

        @Override
        protected TokenizerFactory create(SolrResourceLoader loader, String name, String className, Node node) throws Exception {
            final Map<String, String> params = DOMUtil.toMap(node.getAttributes());
            String configuredVersion = params.remove(LUCENE_MATCH_VERSION_PARAM);
            params.put(LUCENE_MATCH_VERSION_PARAM, parseConfiguredVersion(configuredVersion, TokenizerFactory.class.getSimpleName()).toString());
            TokenizerFactory factory = loader.newInstance(className, TokenizerFactory.class, getDefaultPackages(), new Class[] { Map.class }, new Object[] { params });
            factory.setExplicitLuceneMatchVersion(null != configuredVersion);
            return factory;
        }

        @Override
        protected void init(TokenizerFactory plugin, Node node) throws Exception {
            if (!tokenizers.isEmpty()) {
                throw new SolrException(SolrException.ErrorCode.SERVER_ERROR, "The schema defines multiple tokenizers for: " + node);
            }
            tokenizers.add(plugin);
        }

        @Override
        protected TokenizerFactory register(String name, TokenizerFactory plugin) {
            // used for map registration
            return null;
        }
    };
    tokenizerLoader.load(loader, tokenizerNodes);
    // Make sure something was loaded
    if (tokenizers.isEmpty()) {
        throw new SolrException(SolrException.ErrorCode.SERVER_ERROR, "analyzer without class or tokenizer");
    }
    // Load the Filters
    final ArrayList<TokenFilterFactory> filters = new ArrayList<>();
    AbstractPluginLoader<TokenFilterFactory> filterLoader = new AbstractPluginLoader<TokenFilterFactory>("[schema.xml] analyzer/filter", TokenFilterFactory.class, false, false) {

        @Override
        protected TokenFilterFactory create(SolrResourceLoader loader, String name, String className, Node node) throws Exception {
            final Map<String, String> params = DOMUtil.toMap(node.getAttributes());
            String configuredVersion = params.remove(LUCENE_MATCH_VERSION_PARAM);
            params.put(LUCENE_MATCH_VERSION_PARAM, parseConfiguredVersion(configuredVersion, TokenFilterFactory.class.getSimpleName()).toString());
            TokenFilterFactory factory = loader.newInstance(className, TokenFilterFactory.class, getDefaultPackages(), new Class[] { Map.class }, new Object[] { params });
            factory.setExplicitLuceneMatchVersion(null != configuredVersion);
            return factory;
        }

        @Override
        protected void init(TokenFilterFactory plugin, Node node) throws Exception {
            if (plugin != null) {
                filters.add(plugin);
            }
        }

        @Override
        protected TokenFilterFactory register(String name, TokenFilterFactory plugin) throws Exception {
            // used for map registration
            return null;
        }
    };
    filterLoader.load(loader, tokenFilterNodes);
    return new TokenizerChain(charFilters.toArray(new CharFilterFactory[charFilters.size()]), tokenizers.get(0), filters.toArray(new TokenFilterFactory[filters.size()]));
}

Example 13 with TokenizerFactory

use of org.apache.lucene.analysis.util.TokenizerFactory in project stanbol by apache.

the class LuceneLabelTokenizer method activate.

@Activate
protected void activate(ComponentContext ctx) throws ConfigurationException {
    //init the Solr ResourceLoader used for initialising the components
    resourceLoader = new StanbolResourceLoader(parentResourceLoader);
    //init the Solr CharFilterFactory (optional)
    Object value = ctx.getProperties().get(PROPERTY_CHAR_FILTER_FACTORY);
    if (value != null && !value.toString().isEmpty() && !DEFAULT_CLASS_NAME_CONFIG.equals(value)) {
        Entry<String, Map<String, String>> charFilterConfig = parseConfigLine(PROPERTY_CHAR_FILTER_FACTORY, value.toString());
        charFilterFactory = initAnalyzer(PROPERTY_CHAR_FILTER_FACTORY, charFilterConfig.getKey(), CharFilterFactory.class, charFilterConfig.getValue());
    } else {
        charFilterFactory = null;
    }
    //now initialise the TokenizerFactory (required)
    value = ctx.getProperties().get(PROPERTY_TOKENIZER_FACTORY);
    if (value == null || value.toString().isEmpty() || DEFAULT_CLASS_NAME_CONFIG.equals(value)) {
        throw new ConfigurationException(PROPERTY_TOKENIZER_FACTORY, "The class name of the Lucene Tokemizer MUST BE configured");
    }
    Entry<String, Map<String, String>> tokenizerConfig = parseConfigLine(PROPERTY_CHAR_FILTER_FACTORY, value.toString());
    tokenizerFactory = initAnalyzer(PROPERTY_TOKENIZER_FACTORY, tokenizerConfig.getKey(), TokenizerFactory.class, tokenizerConfig.getValue());
    //initialise the list of Token Filters
    Collection<String> values;
    value = ctx.getProperties().get(PROPERTY_TOKEN_FILTER_FACTORY);
    if (value == null) {
        values = Collections.emptyList();
    } else if (value instanceof Collection<?>) {
        values = new ArrayList<String>(((Collection<?>) value).size());
        for (Object v : (Collection<Object>) value) {
            values.add(v.toString());
        }
    } else if (value instanceof String[]) {
        values = Arrays.asList((String[]) value);
    } else if (value instanceof String) {
        values = Collections.singleton((String) value);
    } else {
        throw new ConfigurationException(PROPERTY_TOKEN_FILTER_FACTORY, "The type '" + value.getClass() + "' of the parsed value is not supported (supported are " + "Collections, String[] and String values)!");
    }
    for (String filterConfigLine : values) {
        if (filterConfigLine == null || filterConfigLine.isEmpty() || DEFAULT_CLASS_NAME_CONFIG.equals(filterConfigLine)) {
            //ignore null, empty and the default value
            continue;
        }
        Entry<String, Map<String, String>> filterConfig = parseConfigLine(PROPERTY_CHAR_FILTER_FACTORY, filterConfigLine);
        TokenFilterFactory tff = initAnalyzer(PROPERTY_TOKEN_FILTER_FACTORY, filterConfig.getKey(), TokenFilterFactory.class, filterConfig.getValue());
        filterFactories.add(tff);
    }
    //init the language configuration
    value = ctx.getProperties().get(LabelTokenizer.SUPPORTED_LANUAGES);
    if (value == null) {
        throw new ConfigurationException(LabelTokenizer.SUPPORTED_LANUAGES, "The language " + "configuration MUST BE present!");
    }
    langConf.setConfiguration(ctx.getProperties());
}

Example 14 with TokenizerFactory

use of org.apache.lucene.analysis.util.TokenizerFactory in project jackrabbit-oak by apache.

the class NodeStateAnalyzerFactory method loadTokenizer.

private TokenizerFactory loadTokenizer(NodeState state) {
    String clazz = checkNotNull(state.getString(LuceneIndexConstants.ANL_NAME));
    Map<String, String> args = convertNodeState(state);
    TokenizerFactory tf = TokenizerFactory.forName(clazz, args);
    init(tf, state);
    return tf;
}

Example 15 with TokenizerFactory

use of org.apache.lucene.analysis.util.TokenizerFactory in project lucene-solr by apache.

the class FieldAnalysisRequestHandlerTest method testCustomAttribute.

//See SOLR-8460
@Test
public void testCustomAttribute() throws Exception {
    FieldAnalysisRequest request = new FieldAnalysisRequest();
    request.addFieldType("skutype1");
    request.setFieldValue("hi, 3456-12 a Test");
    request.setShowMatch(false);
    FieldType fieldType = new TextField();
    Analyzer analyzer = new TokenizerChain(new TokenizerFactory(Collections.emptyMap()) {

        @Override
        public Tokenizer create(AttributeFactory factory) {
            return new CustomTokenizer(factory);
        }
    }, new TokenFilterFactory[] { new TokenFilterFactory(Collections.emptyMap()) {

        @Override
        public TokenStream create(TokenStream input) {
            return new CustomTokenFilter(input);
        }
    } });
    fieldType.setIndexAnalyzer(analyzer);
    NamedList<NamedList> result = handler.analyzeValues(request, fieldType, "fieldNameUnused");
    // just test that we see "900" in the flags attribute here
    List<NamedList> tokenInfoList = (List<NamedList>) result.findRecursive("index", CustomTokenFilter.class.getName());
    // '1' from CustomTokenFilter plus 900 from CustomFlagsAttributeImpl.
    assertEquals(901, tokenInfoList.get(0).get("org.apache.lucene.analysis.tokenattributes.FlagsAttribute#flags"));
}