本文整理汇总了Java中org.carrot2.text.preprocessing.pipeline.BasicPreprocessingPipelineDescriptor类的典型用法代码示例。如果您正苦于以下问题：Java BasicPreprocessingPipelineDescriptor类的具体用法？Java BasicPreprocessingPipelineDescriptor怎么用？Java BasicPreprocessingPipelineDescriptor使用的例子？那么, 这里精选的类代码示例或许可以为您提供帮助。
BasicPreprocessingPipelineDescriptor类属于org.carrot2.text.preprocessing.pipeline包，在下文中一共展示了BasicPreprocessingPipelineDescriptor类的1个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Java代码示例。
示例1: init
import org.carrot2.text.preprocessing.pipeline.BasicPreprocessingPipelineDescriptor; //导入依赖的package包/类
@Override
@SuppressWarnings({ "unchecked", "rawtypes" })
public String init(NamedList config, final SolrCore core) {
this.core = core;
String result = super.init(config, core);
final SolrParams initParams = SolrParams.toSolrParams(config);
// Initialize Carrot2 controller. Pass initialization attributes, if any.
HashMap<String, Object> initAttributes = new HashMap<String, Object>();
extractCarrotAttributes(initParams, initAttributes);
// Customize the stemmer and tokenizer factories. The implementations we provide here
// are included in the code base of Solr, so that it's possible to refactor
// the Lucene APIs the factories rely on if needed.
// Additionally, we set a custom lexical resource factory for Carrot2 that
// will use both Carrot2 default stop words as well as stop words from
// the StopFilter defined on the field.
final AttributeBuilder attributeBuilder = BasicPreprocessingPipelineDescriptor.attributeBuilder(initAttributes);
attributeBuilder.lexicalDataFactory(SolrStopwordsCarrot2LexicalDataFactory.class);
if (!initAttributes.containsKey(BasicPreprocessingPipelineDescriptor.Keys.TOKENIZER_FACTORY)) {
attributeBuilder.tokenizerFactory(LuceneCarrot2TokenizerFactory.class);
}
if (!initAttributes.containsKey(BasicPreprocessingPipelineDescriptor.Keys.STEMMER_FACTORY)) {
attributeBuilder.stemmerFactory(LuceneCarrot2StemmerFactory.class);
}
// Pass the schema to SolrStopwordsCarrot2LexicalDataFactory.
initAttributes.put("solrIndexSchema", core.getSchema());
// Customize Carrot2's resource lookup to first look for resources
// using Solr's resource loader. If that fails, try loading from the classpath.
DefaultLexicalDataFactoryDescriptor.attributeBuilder(initAttributes).resourceLookup(
new ResourceLookup(
// Solr-specific resource loading.
new SolrResourceLocator(core, initParams),
// Using the class loader directly because this time we want to omit the prefix
new ClassLoaderLocator(core.getResourceLoader().getClassLoader())));
// Carrot2 uses current thread's context class loader to get
// certain classes (e.g. custom tokenizer/stemmer) at initialization time.
// To make sure classes from contrib JARs are available,
// we swap the context class loader for the time of clustering.
Thread ct = Thread.currentThread();
ClassLoader prev = ct.getContextClassLoader();
try {
ct.setContextClassLoader(core.getResourceLoader().getClassLoader());
this.controller.init(initAttributes);
} finally {
ct.setContextClassLoader(prev);
}
SchemaField uniqueField = core.getSchema().getUniqueKeyField();
if (uniqueField == null) {
throw new SolrException(SolrException.ErrorCode.SERVER_ERROR,
CarrotClusteringEngine.class.getSimpleName() + " requires the schema to have a uniqueKeyField");
}
this.idFieldName = uniqueField.getName();
// Make sure the requested Carrot2 clustering algorithm class is available
String carrotAlgorithmClassName = initParams.get(CarrotParams.ALGORITHM);
this.clusteringAlgorithmClass = core.getResourceLoader().findClass(carrotAlgorithmClassName, IClusteringAlgorithm.class);
return result;
}