Java FacetsConfig类代码示例

import org.apache.lucene.facet.FacetsConfig; //导入依赖的package包/类
/**
 * Stores features from a specified feature file to the specified project's Lucene index
 * Sample query: featureId:rs44022* AND (variationType:del OR variationType:ins)
 *
 * @param featureFileId a FeatureFile, for which features to save
 * @param projectId a project, for which to write an index
 * @param entries a list of FeatureIndexEntry to write to index
 * @throws IOException
 */
public void writeLuceneIndexForProject(final Long featureFileId, final long projectId,
                                       final List<? extends FeatureIndexEntry> entries) throws IOException {
    try (
        StandardAnalyzer analyzer = new StandardAnalyzer();
        Directory index = fileManager.createIndexForProject(projectId);
        IndexWriter writer = new IndexWriter(index, new IndexWriterConfig(analyzer).setOpenMode(
                                                                IndexWriterConfig.OpenMode.CREATE_OR_APPEND))
    ) {
        FacetsConfig facetsConfig = new FacetsConfig();
        facetsConfig.setIndexFieldName(FeatureIndexFields.CHR_ID.getFieldName(),
                FeatureIndexFields.FACET_CHR_ID.getFieldName());

        for (FeatureIndexEntry entry : entries) {
            Document document = new Document();
            addCommonDocumentFields(document, entry, featureFileId);

            if (entry instanceof VcfIndexEntry) {
                addVcfDocumentFields(document, entry);
            }

            writer.addDocument(facetsConfig.build(document));
        }
    }
}
 

import org.apache.lucene.facet.FacetsConfig; //导入依赖的package包/类
/**
 * Process an intermediate form by carrying out, on the Lucene instance of
 * the shard, the deletes and the inserts (a ram index) in the form.
 * @param form  the intermediate form containing deletes and a ram index
 * @throws IOException
 */
public void process(IntermediateForm form, FacetsConfig facetsConfig) throws IOException {
  if (facetsConfig != null) {
    DirectoryTaxonomyWriter.OrdinalMap map = new DirectoryTaxonomyWriter.MemoryOrdinalMap();
    // merge the taxonomies
    taxoWriter.addTaxonomy(form.getTaxoDirectory(), map);
    int ordinalMap[] = map.getMap();
    DirectoryReader reader = DirectoryReader.open(form.getDirectory());
    try {
      List<AtomicReaderContext> leaves = reader.leaves();
      int numReaders = leaves.size();
      AtomicReader wrappedLeaves[] = new AtomicReader[numReaders];
      for (int i = 0; i < numReaders; i++) {
        wrappedLeaves[i] = new OrdinalMappingAtomicReader(leaves.get(i).reader(), ordinalMap, facetsConfig);
      }
      writer.addIndexes(new MultiReader(wrappedLeaves));
    } finally {
      reader.close();
    }
  } else {
    writer.addIndexes(new Directory[] { form.getDirectory() });
  }
  numForms++;
}
 

import org.apache.lucene.facet.FacetsConfig; //导入依赖的package包/类
/**
 * Stores features from a specified feature file to the specified project's Lucene index
 * Sample query: featureId:rs44022* AND (variationType:del OR variationType:ins)
 *
 * @param featureFileId a FeatureFile, for which features to save
 * @param projectId a project, for which to write an index
 * @param entries a list of FeatureIndexEntry to write to index
 * @throws IOException
 */
public void writeLuceneIndexForProject(final Long featureFileId, final long projectId,
        final List<? extends FeatureIndexEntry> entries) throws IOException {
    try (
            StandardAnalyzer analyzer = new StandardAnalyzer();
            Directory index = fileManager.createIndexForProject(projectId);
            IndexWriter writer = new IndexWriter(index, new IndexWriterConfig(analyzer).setOpenMode(
                    IndexWriterConfig.OpenMode.CREATE_OR_APPEND))
    ) {
        FacetsConfig facetsConfig = new FacetsConfig();
        facetsConfig.setIndexFieldName(FeatureIndexFields.CHR_ID.getFieldName(),
                FeatureIndexFields.FACET_CHR_ID.getFieldName());

        for (FeatureIndexEntry entry : entries) {
            Document document = new Document();
            addCommonDocumentFields(document, entry, featureFileId);

            if (entry instanceof VcfIndexEntry) {
                addVcfDocumentFields(document, entry);
            }

            writer.addDocument(facetsConfig.build(document));
        }
    }
}
 

import org.apache.lucene.facet.FacetsConfig; //导入依赖的package包/类
@Override
@Before
public void setUp() throws Exception {
  super.setUp();
  publishIndexDir = newDirectory();
  publishTaxoDir = newDirectory();
  handlerIndexDir = newMockDirectory();
  handlerTaxoDir = newMockDirectory();
  clientWorkDir = createTempDir("replicationClientTest");
  sourceDirFactory = new PerSessionDirectoryFactory(clientWorkDir);
  replicator = new LocalReplicator();
  callback = new IndexAndTaxonomyReadyCallback(handlerIndexDir, handlerTaxoDir);
  handler = new IndexAndTaxonomyReplicationHandler(handlerIndexDir, handlerTaxoDir, callback);
  client = new ReplicationClient(replicator, handler, sourceDirFactory);
  
  IndexWriterConfig conf = newIndexWriterConfig(null);
  conf.setIndexDeletionPolicy(new SnapshotDeletionPolicy(conf.getIndexDeletionPolicy()));
  publishIndexWriter = new IndexWriter(publishIndexDir, conf);
  publishTaxoWriter = new SnapshotDirectoryTaxonomyWriter(publishTaxoDir);
  config = new FacetsConfig();
  config.setHierarchical("A", true);
}
 

import org.apache.lucene.facet.FacetsConfig; //导入依赖的package包/类
/**
 * Merges the given taxonomy and index directories and commits the changes to
 * the given writers.
 */
public static void merge(Directory srcIndexDir, Directory srcTaxoDir, OrdinalMap map, IndexWriter destIndexWriter,
    DirectoryTaxonomyWriter destTaxoWriter, FacetsConfig srcConfig) throws IOException {
  
  // merge the taxonomies
  destTaxoWriter.addTaxonomy(srcTaxoDir, map);
  int ordinalMap[] = map.getMap();
  DirectoryReader reader = DirectoryReader.open(srcIndexDir);
  try {
    List<AtomicReaderContext> leaves = reader.leaves();
    int numReaders = leaves.size();
    AtomicReader wrappedLeaves[] = new AtomicReader[numReaders];
    for (int i = 0; i < numReaders; i++) {
      wrappedLeaves[i] = new OrdinalMappingAtomicReader(leaves.get(i).reader(), ordinalMap, srcConfig);
    }
    destIndexWriter.addIndexes(new MultiReader(wrappedLeaves));
    
    // commit changes to taxonomy and index respectively.
    destTaxoWriter.commit();
    destIndexWriter.commit();
  } finally {
    reader.close();
  }
}
 

import org.apache.lucene.facet.FacetsConfig; //导入依赖的package包/类
@Override
public List<FacetResult> getAllDims(int topN) throws IOException {
  int ord = children[TaxonomyReader.ROOT_ORDINAL];
  List<FacetResult> results = new ArrayList<>();
  while (ord != TaxonomyReader.INVALID_ORDINAL) {
    String dim = taxoReader.getPath(ord).components[0];
    FacetsConfig.DimConfig dimConfig = config.getDimConfig(dim);
    if (dimConfig.indexFieldName.equals(indexFieldName)) {
      FacetResult result = getTopChildren(topN, dim);
      if (result != null) {
        results.add(result);
      }
    }
    ord = siblings[ord];
  }

  // Sort by highest value, tie break by dim:
  Collections.sort(results, BY_VALUE_THEN_DIM);
  return results;
}
 

import org.apache.lucene.facet.FacetsConfig; //导入依赖的package包/类
public void testMixedTypesInSameIndexField() throws Exception {
  Directory dir = newDirectory();
  Directory taxoDir = newDirectory();
  
  TaxonomyWriter taxoWriter = new DirectoryTaxonomyWriter(taxoDir);
  FacetsConfig config = new FacetsConfig();
  RandomIndexWriter writer = new RandomIndexWriter(random(), dir);

  Document doc = new Document();
  doc.add(new IntAssociationFacetField(14, "a", "x"));
  doc.add(new FloatAssociationFacetField(55.0f, "b", "y"));
  try {
    writer.addDocument(config.build(taxoWriter, doc));
    fail("did not hit expected exception");
  } catch (IllegalArgumentException exc) {
    // expected
  }
  IOUtils.close(writer, taxoWriter, dir, taxoDir);
}
 

import org.apache.lucene.facet.FacetsConfig; //导入依赖的package包/类
public void testNoHierarchy() throws Exception {
  Directory dir = newDirectory();
  Directory taxoDir = newDirectory();
  
  TaxonomyWriter taxoWriter = new DirectoryTaxonomyWriter(taxoDir);
  FacetsConfig config = new FacetsConfig();
  config.setHierarchical("a", true);
  RandomIndexWriter writer = new RandomIndexWriter(random(), dir);

  Document doc = new Document();
  doc.add(new IntAssociationFacetField(14, "a", "x"));
  try {
    writer.addDocument(config.build(taxoWriter, doc));
    fail("did not hit expected exception");
  } catch (IllegalArgumentException exc) {
    // expected
  }
  IOUtils.close(writer, taxoWriter, dir, taxoDir);
}
 

import org.apache.lucene.facet.FacetsConfig; //导入依赖的package包/类
public void testRequireDimCount() throws Exception {
  Directory dir = newDirectory();
  Directory taxoDir = newDirectory();
  
  TaxonomyWriter taxoWriter = new DirectoryTaxonomyWriter(taxoDir);
  FacetsConfig config = new FacetsConfig();
  config.setRequireDimCount("a", true);
  RandomIndexWriter writer = new RandomIndexWriter(random(), dir);

  Document doc = new Document();
  doc.add(new IntAssociationFacetField(14, "a", "x"));
  try {
    writer.addDocument(config.build(taxoWriter, doc));
    fail("did not hit expected exception");
  } catch (IllegalArgumentException exc) {
    // expected
  }
  IOUtils.close(writer, taxoWriter, dir, taxoDir);
}
 

import org.apache.lucene.facet.FacetsConfig; //导入依赖的package包/类
public void testReallyNoNormsForDrillDown() throws Exception {
  Directory dir = newDirectory();
  Directory taxoDir = newDirectory();
  IndexWriterConfig iwc = newIndexWriterConfig(new MockAnalyzer(random()));
  iwc.setSimilarity(new PerFieldSimilarityWrapper() {
      final Similarity sim = new DefaultSimilarity();

      @Override
      public Similarity get(String name) {
        assertEquals("field", name);
        return sim;
      }
    });
  TaxonomyWriter taxoWriter = new DirectoryTaxonomyWriter(taxoDir, IndexWriterConfig.OpenMode.CREATE);
  RandomIndexWriter writer = new RandomIndexWriter(random(), dir, iwc);
  FacetsConfig config = new FacetsConfig();

  Document doc = new Document();
  doc.add(newTextField("field", "text", Field.Store.NO));
  doc.add(new FacetField("a", "path"));
  writer.addDocument(config.build(taxoWriter, doc));
  IOUtils.close(writer, taxoWriter, dir, taxoDir);
}
 

import org.apache.lucene.facet.FacetsConfig; //导入依赖的package包/类
public void testDetectHierarchicalField() throws Exception {
  Directory dir = newDirectory();
  Directory taxoDir = newDirectory();
  TaxonomyWriter taxoWriter = new DirectoryTaxonomyWriter(taxoDir, IndexWriterConfig.OpenMode.CREATE);
  RandomIndexWriter writer = new RandomIndexWriter(random(), dir);
  FacetsConfig config = new FacetsConfig();

  Document doc = new Document();
  doc.add(newTextField("field", "text", Field.Store.NO));
  doc.add(new FacetField("a", "path", "other"));
  try {
    config.build(taxoWriter, doc);
    fail("did not hit expected exception");
  } catch (IllegalArgumentException iae) {
    // expected
  }
  IOUtils.close(writer, taxoWriter, dir, taxoDir);
}
 

import org.apache.lucene.facet.FacetsConfig; //导入依赖的package包/类
public void testDetectMultiValuedField() throws Exception {
  Directory dir = newDirectory();
  Directory taxoDir = newDirectory();
  TaxonomyWriter taxoWriter = new DirectoryTaxonomyWriter(taxoDir, IndexWriterConfig.OpenMode.CREATE);
  RandomIndexWriter writer = new RandomIndexWriter(random(), dir);
  FacetsConfig config = new FacetsConfig();

  Document doc = new Document();
  doc.add(newTextField("field", "text", Field.Store.NO));
  doc.add(new FacetField("a", "path"));
  doc.add(new FacetField("a", "path2"));
  try {
    config.build(taxoWriter, doc);
    fail("did not hit expected exception");
  } catch (IllegalArgumentException iae) {
    // expected
  }
  IOUtils.close(writer, taxoWriter, dir, taxoDir);
}
 

import org.apache.lucene.facet.FacetsConfig; //导入依赖的package包/类
public void testChildCount() throws Exception {
  // LUCENE-4885: FacetResult.numValidDescendants was not set properly by FacetsAccumulator
  Directory indexDir = newDirectory();
  Directory taxoDir = newDirectory();
  
  DirectoryTaxonomyWriter taxoWriter = new DirectoryTaxonomyWriter(taxoDir);
  IndexWriter iw = new IndexWriter(indexDir, newIndexWriterConfig(new MockAnalyzer(random())));
  FacetsConfig config = new FacetsConfig();
  for (int i = 0; i < 10; i++) {
    Document doc = new Document();
    doc.add(new FacetField("a", Integer.toString(i)));
    iw.addDocument(config.build(taxoWriter, doc));
  }
  
  DirectoryReader r = DirectoryReader.open(iw, true);
  DirectoryTaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoWriter);
  
  FacetsCollector sfc = new FacetsCollector();
  newSearcher(r).search(new MatchAllDocsQuery(), sfc);
  Facets facets = getTaxonomyFacetCounts(taxoReader, config, sfc);
  
  assertEquals(10, facets.getTopChildren(2, "a").childCount);

  IOUtils.close(taxoWriter, iw, taxoReader, taxoDir, r, indexDir);
}
 

import org.apache.lucene.facet.FacetsConfig; //导入依赖的package包/类
private void indexTwoDocs(TaxonomyWriter taxoWriter, IndexWriter indexWriter, FacetsConfig config, boolean withContent) throws Exception {
  for (int i = 0; i < 2; i++) {
    Document doc = new Document();
    if (withContent) {
      doc.add(new StringField("f", "a", Field.Store.NO));
    }
    if (config != null) {
      doc.add(new FacetField("A", Integer.toString(i)));
      indexWriter.addDocument(config.build(taxoWriter, doc));
    } else {
      indexWriter.addDocument(doc);
    }
  }
  
  indexWriter.commit();
}
 

import org.apache.lucene.facet.FacetsConfig; //导入依赖的package包/类
private static void addFacets(Document doc, FacetsConfig config, boolean updateTermExpectedCounts) 
    throws IOException {
  List<FacetField> docCategories = randomCategories(random());
  for (FacetField ff : docCategories) {
    doc.add(ff);
    String cp = ff.dim + "/" + ff.path[0];
    allExpectedCounts.put(cp, allExpectedCounts.get(cp) + 1);
    if (updateTermExpectedCounts) {
      termExpectedCounts.put(cp, termExpectedCounts.get(cp) + 1);
    }
  }
  // add 1 to each NO_PARENTS dimension
  allExpectedCounts.put(CP_B, allExpectedCounts.get(CP_B) + 1);
  allExpectedCounts.put(CP_C, allExpectedCounts.get(CP_C) + 1);
  allExpectedCounts.put(CP_D, allExpectedCounts.get(CP_D) + 1);
  if (updateTermExpectedCounts) {
    termExpectedCounts.put(CP_B, termExpectedCounts.get(CP_B) + 1);
    termExpectedCounts.put(CP_C, termExpectedCounts.get(CP_C) + 1);
    termExpectedCounts.put(CP_D, termExpectedCounts.get(CP_D) + 1);
  }
}