Examples with DirectoryTaxonomyReader org.apache.lucene.facet.taxonomy.directory.DirectoryTaxonomyReader used on opensource projects

Example 11 with DirectoryTaxonomyReader

use of org.apache.lucene.facet.taxonomy.directory.DirectoryTaxonomyReader in project lucene-solr by apache.

the class TestDrillSideways method testMultipleRequestsPerDim.

public void testMultipleRequestsPerDim() throws Exception {
    Directory dir = newDirectory();
    Directory taxoDir = newDirectory();
    RandomIndexWriter writer = new RandomIndexWriter(random(), dir);
    // Writes facet ords to a separate directory from the
    // main index:
    DirectoryTaxonomyWriter taxoWriter = new DirectoryTaxonomyWriter(taxoDir, IndexWriterConfig.OpenMode.CREATE);
    FacetsConfig config = new FacetsConfig();
    config.setHierarchical("dim", true);
    Document doc = new Document();
    doc.add(new FacetField("dim", "a", "x"));
    writer.addDocument(config.build(taxoWriter, doc));
    doc = new Document();
    doc.add(new FacetField("dim", "a", "y"));
    writer.addDocument(config.build(taxoWriter, doc));
    doc = new Document();
    doc.add(new FacetField("dim", "a", "z"));
    writer.addDocument(config.build(taxoWriter, doc));
    doc = new Document();
    doc.add(new FacetField("dim", "b"));
    writer.addDocument(config.build(taxoWriter, doc));
    doc = new Document();
    doc.add(new FacetField("dim", "c"));
    writer.addDocument(config.build(taxoWriter, doc));
    doc = new Document();
    doc.add(new FacetField("dim", "d"));
    writer.addDocument(config.build(taxoWriter, doc));
    // NRT open
    IndexSearcher searcher = newSearcher(writer.getReader());
    //System.out.println("searcher=" + searcher);
    // NRT open
    TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoWriter);
    DrillDownQuery ddq = new DrillDownQuery(config);
    ddq.add("dim", "a");
    DrillSidewaysResult r = getNewDrillSideways(searcher, config, taxoReader).search(null, ddq, 10);
    assertEquals(3, r.hits.totalHits);
    assertEquals("dim=dim path=[] value=6 childCount=4\n  a (3)\n  b (1)\n  c (1)\n  d (1)\n", r.facets.getTopChildren(10, "dim").toString());
    assertEquals("dim=dim path=[a] value=3 childCount=3\n  x (1)\n  y (1)\n  z (1)\n", r.facets.getTopChildren(10, "dim", "a").toString());
    writer.close();
    IOUtils.close(searcher.getIndexReader(), taxoReader, taxoWriter, dir, taxoDir);
}

Example 12 with DirectoryTaxonomyReader

use of org.apache.lucene.facet.taxonomy.directory.DirectoryTaxonomyReader in project lucene-solr by apache.

the class TestOrdinalMappingLeafReader method verifyResults.

private void verifyResults(Directory indexDir, Directory taxoDir) throws IOException {
    DirectoryReader indexReader = DirectoryReader.open(indexDir);
    DirectoryTaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);
    IndexSearcher searcher = newSearcher(indexReader);
    FacetsCollector collector = new FacetsCollector();
    FacetsCollector.search(searcher, new MatchAllDocsQuery(), 10, collector);
    // tag facets
    Facets tagFacets = new FastTaxonomyFacetCounts("$tags", taxoReader, facetConfig, collector);
    FacetResult result = tagFacets.getTopChildren(10, "tag");
    for (LabelAndValue lv : result.labelValues) {
        if (VERBOSE) {
            System.out.println(lv);
        }
        assertEquals(NUM_DOCS, lv.value.intValue());
    }
    // id facets
    Facets idFacets = new FastTaxonomyFacetCounts(taxoReader, facetConfig, collector);
    FacetResult idResult = idFacets.getTopChildren(10, "id");
    assertEquals(NUM_DOCS, idResult.childCount);
    // each "id" appears twice
    assertEquals(NUM_DOCS * 2, idResult.value);
    BinaryDocValues bdv = MultiDocValues.getBinaryValues(indexReader, "bdv");
    BinaryDocValues cbdv = MultiDocValues.getBinaryValues(indexReader, "cbdv");
    for (int i = 0; i < indexReader.maxDoc(); i++) {
        assertEquals(i, bdv.nextDoc());
        assertEquals(i, cbdv.nextDoc());
        assertEquals(Integer.parseInt(cbdv.binaryValue().utf8ToString()), Integer.parseInt(bdv.binaryValue().utf8ToString()) * 2);
    }
    IOUtils.close(indexReader, taxoReader);
}

Example 13 with DirectoryTaxonomyReader

use of org.apache.lucene.facet.taxonomy.directory.DirectoryTaxonomyReader in project lucene-solr by apache.

the class TestMultipleIndexFields method testSomeSameSomeDifferent.

@Test
public void testSomeSameSomeDifferent() throws Exception {
    Directory indexDir = newDirectory();
    Directory taxoDir = newDirectory();
    // create and open an index writer
    RandomIndexWriter iw = new RandomIndexWriter(random(), indexDir, newIndexWriterConfig(new MockAnalyzer(random(), MockTokenizer.WHITESPACE, false)));
    // create and open a taxonomy writer
    TaxonomyWriter tw = new DirectoryTaxonomyWriter(taxoDir, OpenMode.CREATE);
    FacetsConfig config = getConfig();
    config.setIndexFieldName("Band", "$music");
    config.setIndexFieldName("Composer", "$music");
    config.setIndexFieldName("Author", "$literature");
    seedIndex(tw, iw, config);
    IndexReader ir = iw.getReader();
    tw.commit();
    // prepare index reader and taxonomy.
    TaxonomyReader tr = new DirectoryTaxonomyReader(taxoDir);
    // prepare searcher to search against
    IndexSearcher searcher = newSearcher(ir);
    FacetsCollector sfc = performSearch(tr, ir, searcher);
    Map<String, Facets> facetsMap = new HashMap<>();
    Facets facets2 = getTaxonomyFacetCounts(tr, config, sfc, "$music");
    facetsMap.put("Band", facets2);
    facetsMap.put("Composer", facets2);
    facetsMap.put("Author", getTaxonomyFacetCounts(tr, config, sfc, "$literature"));
    Facets facets = new MultiFacets(facetsMap, getTaxonomyFacetCounts(tr, config, sfc));
    // Obtain facets results and hand-test them
    assertCorrectResults(facets);
    assertOrdinalsExist("$music", ir);
    assertOrdinalsExist("$literature", ir);
    iw.close();
    IOUtils.close(tr, ir, iw, tw, indexDir, taxoDir);
}

Example 14 with DirectoryTaxonomyReader

use of org.apache.lucene.facet.taxonomy.directory.DirectoryTaxonomyReader in project lucene-solr by apache.

the class TestTaxonomyFacetCounts method testWrongIndexFieldName.

public void testWrongIndexFieldName() throws Exception {
    Directory dir = newDirectory();
    Directory taxoDir = newDirectory();
    // Writes facet ords to a separate directory from the
    // main index:
    DirectoryTaxonomyWriter taxoWriter = new DirectoryTaxonomyWriter(taxoDir, IndexWriterConfig.OpenMode.CREATE);
    FacetsConfig config = new FacetsConfig();
    config.setIndexFieldName("a", "$facets2");
    RandomIndexWriter writer = new RandomIndexWriter(random(), dir);
    Document doc = new Document();
    doc.add(new FacetField("a", "foo1"));
    writer.addDocument(config.build(taxoWriter, doc));
    // NRT open
    IndexSearcher searcher = newSearcher(writer.getReader());
    // NRT open
    TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoWriter);
    FacetsCollector c = new FacetsCollector();
    searcher.search(new MatchAllDocsQuery(), c);
    // Uses default $facets field:
    Facets facets;
    if (random().nextBoolean()) {
        facets = new FastTaxonomyFacetCounts(taxoReader, config, c);
    } else {
        OrdinalsReader ordsReader = new DocValuesOrdinalsReader();
        if (random().nextBoolean()) {
            ordsReader = new CachedOrdinalsReader(ordsReader);
        }
        facets = new TaxonomyFacetCounts(ordsReader, taxoReader, config, c);
    }
    // Ask for top 10 labels for any dims that have counts:
    List<FacetResult> results = facets.getAllDims(10);
    assertTrue(results.isEmpty());
    expectThrows(IllegalArgumentException.class, () -> {
        facets.getSpecificValue("a");
    });
    expectThrows(IllegalArgumentException.class, () -> {
        facets.getTopChildren(10, "a");
    });
    writer.close();
    IOUtils.close(taxoWriter, searcher.getIndexReader(), taxoReader, taxoDir, dir);
}

Example 15 with DirectoryTaxonomyReader

use of org.apache.lucene.facet.taxonomy.directory.DirectoryTaxonomyReader in project lucene-solr by apache.

the class TestTaxonomyFacetCounts method testBasic.

public void testBasic() throws Exception {
    Directory dir = newDirectory();
    Directory taxoDir = newDirectory();
    // Writes facet ords to a separate directory from the
    // main index:
    DirectoryTaxonomyWriter taxoWriter = new DirectoryTaxonomyWriter(taxoDir, IndexWriterConfig.OpenMode.CREATE);
    FacetsConfig config = new FacetsConfig();
    config.setHierarchical("Publish Date", true);
    RandomIndexWriter writer = new RandomIndexWriter(random(), dir);
    Document doc = new Document();
    doc.add(new FacetField("Author", "Bob"));
    doc.add(new FacetField("Publish Date", "2010", "10", "15"));
    writer.addDocument(config.build(taxoWriter, doc));
    doc = new Document();
    doc.add(new FacetField("Author", "Lisa"));
    doc.add(new FacetField("Publish Date", "2010", "10", "20"));
    writer.addDocument(config.build(taxoWriter, doc));
    doc = new Document();
    doc.add(new FacetField("Author", "Lisa"));
    doc.add(new FacetField("Publish Date", "2012", "1", "1"));
    writer.addDocument(config.build(taxoWriter, doc));
    doc = new Document();
    doc.add(new FacetField("Author", "Susan"));
    doc.add(new FacetField("Publish Date", "2012", "1", "7"));
    writer.addDocument(config.build(taxoWriter, doc));
    doc = new Document();
    doc.add(new FacetField("Author", "Frank"));
    doc.add(new FacetField("Publish Date", "1999", "5", "5"));
    writer.addDocument(config.build(taxoWriter, doc));
    // NRT open
    IndexSearcher searcher = newSearcher(writer.getReader());
    // NRT open
    TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoWriter);
    Facets facets = getAllFacets(FacetsConfig.DEFAULT_INDEX_FIELD_NAME, searcher, taxoReader, config);
    // Retrieve & verify results:
    assertEquals("dim=Publish Date path=[] value=5 childCount=3\n  2010 (2)\n  2012 (2)\n  1999 (1)\n", facets.getTopChildren(10, "Publish Date").toString());
    assertEquals("dim=Author path=[] value=5 childCount=4\n  Lisa (2)\n  Bob (1)\n  Susan (1)\n  Frank (1)\n", facets.getTopChildren(10, "Author").toString());
    // Now user drills down on Publish Date/2010:
    DrillDownQuery q2 = new DrillDownQuery(config);
    q2.add("Publish Date", "2010");
    FacetsCollector c = new FacetsCollector();
    searcher.search(q2, c);
    facets = new FastTaxonomyFacetCounts(taxoReader, config, c);
    assertEquals("dim=Author path=[] value=2 childCount=2\n  Bob (1)\n  Lisa (1)\n", facets.getTopChildren(10, "Author").toString());
    assertEquals(1, facets.getSpecificValue("Author", "Lisa"));
    assertNull(facets.getTopChildren(10, "Non exitent dim"));
    // Smoke test PrintTaxonomyStats:
    ByteArrayOutputStream bos = new ByteArrayOutputStream();
    PrintTaxonomyStats.printStats(taxoReader, new PrintStream(bos, false, IOUtils.UTF_8), true);
    String result = bos.toString(IOUtils.UTF_8);
    assertTrue(result.indexOf("/Author: 4 immediate children; 5 total categories") != -1);
    assertTrue(result.indexOf("/Publish Date: 3 immediate children; 12 total categories") != -1);
    // Make sure at least a few nodes of the tree came out:
    assertTrue(result.indexOf("  /1999") != -1);
    assertTrue(result.indexOf("  /2012") != -1);
    assertTrue(result.indexOf("      /20") != -1);
    writer.close();
    IOUtils.close(taxoWriter, searcher.getIndexReader(), taxoReader, taxoDir, dir);
}