@SuppressWarnings("unchecked")
@Override
public List<String> getClusterByCarrotVersion2(String query) {
// TODO Auto-generated method stub
List<String> strs = new ArrayList<String>();
final Controller controller = ControllerFactory.createPooling();
final Map<String, Object> luceneGlobalAttributes = new HashMap<String, Object>();
LuceneDocumentSourceDescriptor.attributeBuilder(luceneGlobalAttributes).directory(directory);
SimpleFieldMapperDescriptor.attributeBuilder(luceneGlobalAttributes)
.titleField(TITLE_FIELD)
.contentField(CONTENTS_FIELD)
.searchFields(Arrays.asList(new String[] {TITLE_FIELD, CONTENTS_FIELD}));
controller.init(
new HashMap<String, Object>(),
new ProcessingComponentConfiguration(
LuceneDocumentSource.class, "lucene", luceneGlobalAttributes));
final Map<String, Object> processingAttributes = Maps.newHashMap();
CommonAttributesDescriptor.attributeBuilder(processingAttributes).query(query);
ProcessingResult process =
controller.process(
processingAttributes, "lucene", LingoClusteringAlgorithm.class.getName());
for (Cluster c : process.getClusters()) {
strs.add(c.getLabel() + " >>>> " + c.getAllDocuments().size());
}
return strs;
}
public static void main(String[] args) throws IOException {
/*
* We will use the CachingController for this example. Running
* LuceneDocumentSource within the CachingController will let us open the index
* once per component initialization and not once per query, which would be the
* case with SimpleController. We will also use this opportunity to show how
* component-specific attribute values can be passed during CachingComponent
* initialization.
*/
/*
* Create a caching controller that will reuse processing component instances, but
* will not perform any caching of results produced by components. We will leave
* caching of documents from Lucene index to Lucene and the operating system
* caches.
*/
final Controller controller = ControllerFactory.createPooling();
/*
* Prepare a map with component-specific attributes. Here, this map will contain
* the index location and names of fields to be used to fetch document title and
* summary.
*/
final Map<String, Object> luceneGlobalAttributes = new HashMap<String, Object>();
String indexPath = "put your index path here or pass as the first argument";
if (args.length == 1) {
indexPath = args[0];
}
LuceneDocumentSourceDescriptor.attributeBuilder(luceneGlobalAttributes)
.directory(FSDirectory.open(new File(indexPath)));
/*
* Specify fields providing data inside your Lucene index.
*/
SimpleFieldMapperDescriptor.attributeBuilder(luceneGlobalAttributes)
.titleField("title")
.contentField("snippet")
.searchFields(Arrays.asList(new String[] {"titleField", "fullContent"}));
/*
* Initialize the controller passing the above attributes as component-specific
* for Lucene. The global attributes map will be empty. Note that we've provided
* an identifier for our specially-configured Lucene component, we'll need to use
* this identifier when performing processing.
*/
controller.init(
new HashMap<String, Object>(),
new ProcessingComponentConfiguration(
LuceneDocumentSource.class, "lucene", luceneGlobalAttributes));
/*
* Perform processing.
*/
String query = "mining";
final Map<String, Object> processingAttributes = Maps.newHashMap();
CommonAttributesDescriptor.attributeBuilder(processingAttributes).query(query);
/*
* We need to refer to the Lucene component by its identifier we set during
* initialization. As we've not assigned any identifier to the
* LingoClusteringAlgorithm we want to use, we can its fully qualified class name.
*/
ProcessingResult process =
controller.process(
processingAttributes, "lucene", LingoClusteringAlgorithm.class.getName());
ConsoleFormatter.displayResults(process);
}