org.apache.lucene.facet.taxonomy.FastTaxonomyFacetCounts Java Examples

/** User runs a query and counts facets. */
private List<FacetResult> facetsWithSearch() throws IOException {
  DirectoryReader indexReader = DirectoryReader.open(indexDir);
  IndexSearcher searcher = new IndexSearcher(indexReader);
  TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);

  FacetsCollector fc = new FacetsCollector();

  // MatchAllDocsQuery is for "browsing" (counts facets
  // for all non-deleted docs in the index); normally
  // you'd use a "normal" query:
  FacetsCollector.search(searcher, new MatchAllDocsQuery(), 10, fc);

  // Retrieve results
  List<FacetResult> results = new ArrayList<>();

  // Count both "Publish Date" and "Author" dimensions
  Facets facets = new FastTaxonomyFacetCounts(taxoReader, config, fc);
  results.add(facets.getTopChildren(10, "Author"));
  results.add(facets.getTopChildren(10, "Publish Date"));
  
  indexReader.close();
  taxoReader.close();
  
  return results;
}
 

/** User drills down on 'Publish Date/2010', and we
 *  return facets for 'Author' */
private FacetResult drillDown() throws IOException {
  DirectoryReader indexReader = DirectoryReader.open(indexDir);
  IndexSearcher searcher = new IndexSearcher(indexReader);
  TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);

  // Passing no baseQuery means we drill down on all
  // documents ("browse only"):
  DrillDownQuery q = new DrillDownQuery(config);

  // Now user drills down on Publish Date/2010:
  q.add("Publish Date", "2010");
  FacetsCollector fc = new FacetsCollector();
  FacetsCollector.search(searcher, q, 10, fc);

  // Retrieve results
  Facets facets = new FastTaxonomyFacetCounts(taxoReader, config, fc);
  FacetResult result = facets.getTopChildren(10, "Author");

  indexReader.close();
  taxoReader.close();
  
  return result;
}
 

/** User runs a query and counts facets. */
private List<FacetResult> facetsWithSearch() throws IOException {
    DirectoryReader indexReader = DirectoryReader.open(indexDir);
    IndexSearcher searcher = new IndexSearcher(indexReader);
    TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);

    FacetsCollector fc = new FacetsCollector();

    // MatchAllDocsQuery is for "browsing" (counts facets
    // for all non-deleted docs in the index); normally
    // you'd use a "normal" query:
    FacetsCollector.search(searcher, new MatchAllDocsQuery(), 10, fc);

    // Retrieve results
    List<FacetResult> results = new ArrayList<FacetResult>();

    // Count both "Publish Date" and "Author" dimensions
    Facets facets = new FastTaxonomyFacetCounts(taxoReader, config, fc);
    results.add(facets.getTopChildren(10, "Author"));
    results.add(facets.getTopChildren(10, "Publish Date"));

    indexReader.close();
    taxoReader.close();

    return results;
}
 

/** User drills down on 'Publish Date/2010', and we
 *  return facets for 'Author' */
private FacetResult drillDown() throws IOException {
    DirectoryReader indexReader = DirectoryReader.open(indexDir);
    IndexSearcher searcher = new IndexSearcher(indexReader);
    TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);

    // Passing no baseQuery means we drill down on all
    // documents ("browse only"):
    DrillDownQuery q = new DrillDownQuery(config);

    // Now user drills down on Publish Date/2010:
    q.add("Publish Date", "2010");
    FacetsCollector fc = new FacetsCollector();
    FacetsCollector.search(searcher, q, 10, fc);

    // Retrieve results
    Facets facets = new FastTaxonomyFacetCounts(taxoReader, config, fc);
    FacetResult result = facets.getTopChildren(10, "Author");

    indexReader.close();
    taxoReader.close();

    return result;
}
 

/** User runs a query and counts facets. */
private List<FacetResult> search() throws IOException {
  DirectoryReader indexReader = DirectoryReader.open(indexDir);
  IndexSearcher searcher = new IndexSearcher(indexReader);
  TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);

  FacetsCollector fc = new FacetsCollector();

  // MatchAllDocsQuery is for "browsing" (counts facets
  // for all non-deleted docs in the index); normally
  // you'd use a "normal" query:
  FacetsCollector.search(searcher, new MatchAllDocsQuery(), 10, fc);

  // Retrieve results
  List<FacetResult> results = new ArrayList<>();

  // Count both "Publish Date" and "Author" dimensions
  Facets author = new FastTaxonomyFacetCounts("author", taxoReader, config, fc);
  results.add(author.getTopChildren(10, "Author"));

  Facets pubDate = new FastTaxonomyFacetCounts("pubdate", taxoReader, config, fc);
  results.add(pubDate.getTopChildren(10, "Publish Date"));

  indexReader.close();
  taxoReader.close();

  return results;
}
 

/** User runs a query and counts facets only without collecting the matching documents.*/
private List<FacetResult> facetsOnly() throws IOException {
  DirectoryReader indexReader = DirectoryReader.open(indexDir);
  IndexSearcher searcher = new IndexSearcher(indexReader);
  TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);

  FacetsCollector fc = new FacetsCollector();

  // MatchAllDocsQuery is for "browsing" (counts facets
  // for all non-deleted docs in the index); normally
  // you'd use a "normal" query:
  searcher.search(new MatchAllDocsQuery(), fc);

  // Retrieve results
  List<FacetResult> results = new ArrayList<>();

  // Count both "Publish Date" and "Author" dimensions
  Facets facets = new FastTaxonomyFacetCounts(taxoReader, config, fc);
 
  results.add(facets.getTopChildren(10, "Author"));
  results.add(facets.getTopChildren(10, "Publish Date"));
  
  indexReader.close();
  taxoReader.close();
  
  return results;
}
 

@Override
public Boolean call() throws Exception {
  if (indexReader == null) {
    indexReader = DirectoryReader.open(indexDir);
    lastIndexGeneration = indexReader.getIndexCommit().getGeneration();
    taxoReader = new DirectoryTaxonomyReader(taxoDir);
  } else {
    // verify search index
    DirectoryReader newReader = DirectoryReader.openIfChanged(indexReader);
    assertNotNull("should not have reached here if no changes were made to the index", newReader);
    long newGeneration = newReader.getIndexCommit().getGeneration();
    assertTrue("expected newer generation; current=" + lastIndexGeneration + " new=" + newGeneration, newGeneration > lastIndexGeneration);
    indexReader.close();
    indexReader = newReader;
    lastIndexGeneration = newGeneration;
    TestUtil.checkIndex(indexDir);
    
    // verify taxonomy index
    DirectoryTaxonomyReader newTaxoReader = TaxonomyReader.openIfChanged(taxoReader);
    if (newTaxoReader != null) {
      taxoReader.close();
      taxoReader = newTaxoReader;
    }
    TestUtil.checkIndex(taxoDir);
    
    // verify faceted search
    int id = Integer.parseInt(indexReader.getIndexCommit().getUserData().get(VERSION_ID), 16);
    IndexSearcher searcher = new IndexSearcher(indexReader);
    FacetsCollector fc = new FacetsCollector();
    searcher.search(new MatchAllDocsQuery(), fc);
    Facets facets = new FastTaxonomyFacetCounts(taxoReader, config, fc);
    assertEquals(1, facets.getSpecificValue("A", Integer.toString(id, 16)).intValue());
    
    DrillDownQuery drillDown = new DrillDownQuery(config);
    drillDown.add("A", Integer.toString(id, 16));
    TopDocs docs = searcher.search(drillDown, 10);
    assertEquals(1, docs.totalHits.value);
  }
  return null;
}
 

/** User runs a query and counts facets only without collecting the matching documents.*/
private List<FacetResult> facetsOnly() throws IOException {
    DirectoryReader indexReader = DirectoryReader.open(indexDir);
    IndexSearcher searcher = new IndexSearcher(indexReader);
    TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);

    FacetsCollector fc = new FacetsCollector();

    // MatchAllDocsQuery is for "browsing" (counts facets
    // for all non-deleted docs in the index); normally
    // you'd use a "normal" query:
    searcher.search(new MatchAllDocsQuery(), null /*Filter */, fc);

    // Retrieve results
    List<FacetResult> results = new ArrayList<FacetResult>();

    // Count both "Publish Date" and "Author" dimensions
    Facets facets = new FastTaxonomyFacetCounts(taxoReader, config, fc);

    results.add(facets.getTopChildren(10, "Author"));
    results.add(facets.getTopChildren(10, "Publish Date"));

    indexReader.close();
    taxoReader.close();

    return results;
}
 

private void fetchFacet() {
  if (queryContext.facet) {
    FacetsCollector facetsCollector = new FacetsCollector(true);

    try {

      String[] pathFacet = null;
      if (queryContext.isDrillDown()) {
        DrillDownQuery drillDownQuery = new DrillDownQuery(queryContext.getFacetConfig(), query);
        String[] path = queryContext.getDrillDownQuery().split(":");
        pathFacet = path[1].split("/");
        drillDownQuery.add(path[0], pathFacet);
        FacetsCollector.search(queryContext.searcher, drillDownQuery, PAGE_SIZE, facetsCollector);
      } else {
        FacetsCollector.search(queryContext.searcher, query, PAGE_SIZE, facetsCollector);
      }

      Facets facets = new FastTaxonomyFacetCounts(queryContext.reader, queryContext.getFacetConfig(), facetsCollector);

      FacetResult facetResult = null;
      if (pathFacet != null) {
        facetResult = facets.getTopChildren(PAGE_SIZE, queryContext.getFacetField(), pathFacet);
      } else {
        facetResult = facets.getTopChildren(PAGE_SIZE, queryContext.getFacetField());
      }

      if (facetResult != null) {
        List<ODocument> documents = new ArrayList<ODocument>();
        // for (FacetResult facetResult : res) {

        ODocument doc = new ODocument();

        doc.field("childCount", facetResult.childCount);
        doc.field("value", facetResult.value);
        doc.field("dim", facetResult.dim);
        List<ODocument> labelsAndValue = new ArrayList<ODocument>();
        for (LabelAndValue labelValue : facetResult.labelValues) {
          ODocument doc1 = new ODocument();
          doc1.field("label", labelValue.label);
          doc1.field("value", labelValue.value);
          labelsAndValue.add(doc1);

        }
        doc.field("labelsValue", labelsAndValue);
        documents.add(doc);
        queryContext.context.setVariable("$facet", documents);
      }
      // }

    } catch (IOException e) {
      e.printStackTrace();
    }
  }
}
 

private void searchWithFacets(FacetRequest facetRequest, Query q, IndexSearcher indexSearcher, TopDocsCollector<?> collector,
		SegmentResponse.Builder segmentReponseBuilder) throws Exception {
	FacetsCollector facetsCollector = new FacetsCollector();
	indexSearcher.search(q, MultiCollector.wrap(collector, facetsCollector));

	Facets facets = new FastTaxonomyFacetCounts(taxoReader, facetsConfig, facetsCollector);

	for (CountRequest countRequest : facetRequest.getCountRequestList()) {

		String label = countRequest.getFacetField().getLabel();

		if (!indexConfig.existingFacet(label)) {
			throw new Exception(label + " is not defined as a facetable field");
		}

		if (countRequest.hasSegmentFacets()) {
			if (indexConfig.getNumberOfSegments() == 1) {
				log.info("Segment facets is ignored with segments of 1 for facet <" + label + "> on index <" + indexName + ">");
			}
			if (countRequest.getSegmentFacets() < countRequest.getMaxFacets()) {
				throw new IllegalArgumentException("Segment facets must be greater than or equal to max facets");
			}
		}

		int numOfFacets;
		if (indexConfig.getNumberOfSegments() > 1) {
			if (countRequest.getSegmentFacets() != 0) {
				numOfFacets = countRequest.getSegmentFacets();
			}
			else {
				numOfFacets = countRequest.getMaxFacets() * 8;
			}

		}
		else {
			numOfFacets = countRequest.getMaxFacets();
		}

		FacetResult facetResult = null;

		try {

			if (indexConfig.getNumberOfSegments() > 1) {
				if (countRequest.hasSegmentFacets() && countRequest.getSegmentFacets() == 0) {
					//TODO: this not ideal
					numOfFacets = taxoReader.getSize();
				}
			}

			facetResult = facets.getTopChildren(numOfFacets, label);
		}
		catch (UncheckedExecutionException e) {
			Throwable cause = e.getCause();
			if (cause.getMessage().contains(" was not indexed with SortedSetDocValues")) {
				//this is when no data has been indexing into a facet or facet does not exist
			}
			else {
				throw e;
			}
		}
		FacetGroup.Builder fg = FacetGroup.newBuilder();
		fg.setCountRequest(countRequest);

		if (facetResult != null) {

			for (LabelAndValue subResult : facetResult.labelValues) {
				FacetCount.Builder facetCountBuilder = FacetCount.newBuilder();
				facetCountBuilder.setCount(subResult.value.longValue());
				facetCountBuilder.setFacet(subResult.label);
				fg.addFacetCount(facetCountBuilder);
			}
		}
		segmentReponseBuilder.addFacetGroup(fg);
	}
}