Examples of org.apache.lucene.facet.taxonomy.directory.DirectoryTaxonomyReader

Package org.apache.lucene.facet.taxonomy.directory

Examples of org.apache.lucene.facet.taxonomy.directory.DirectoryTaxonomyReader

org.apache.lucene.facet.taxonomy.directory.DirectoryTaxonomyReader
A {@link TaxonomyReader} which retrieves stored taxonomy information from a{@link Directory}.
Reading from the on-disk index on every method call is too slow, so this implementation employs caching: Some methods cache recent requests and their results, while other methods prefetch all the data into memory and then provide answers directly from in-memory tables. See the documentation of individual methods for comments on their performance. @lucene.experimental

    IndexSearcher searcher = newSearcher(writer.getReader());


    //System.out.println("searcher=" + searcher);


    // NRT open
    TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoWriter);


    DrillSideways ds = new DrillSideways(searcher, config, taxoReader);


    //  case: drill-down on a single field; in this
    // case the drill-sideways + drill-down counts ==

View Full Code Here

    IndexSearcher searcher = newSearcher(writer.getReader());


    //System.out.println("searcher=" + searcher);


    // NRT open
    TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoWriter);


    DrillDownQuery ddq = new DrillDownQuery(config);
    ddq.add("Author", "Lisa");
    DrillSidewaysResult r = new DrillSideways(searcher, config, taxoReader).search(null, ddq, 10);

View Full Code Here

    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 = new DrillSideways(searcher, config, taxoReader).search(null, ddq, 10);

View Full Code Here

    if (VERBOSE) {
      System.out.println("r.numDocs() = " + r.numDocs());
    }


    // NRT open
    TaxonomyReader tr = new DirectoryTaxonomyReader(tw);


    int numIters = atLeast(10);


    for(int iter=0;iter<numIters;iter++) {

View Full Code Here

    Directory dir = newDirectory();
    Directory taxoDir = newDirectory();
    RandomIndexWriter writer = new RandomIndexWriter(random(), dir);
    DirectoryTaxonomyWriter taxoWriter = new DirectoryTaxonomyWriter(taxoDir, IndexWriterConfig.OpenMode.CREATE);
    IndexSearcher searcher = newSearcher(writer.getReader());
    TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoWriter);


    // Count "Author"
    FacetsConfig config = new FacetsConfig();
    DrillSideways ds = new DrillSideways(searcher, config, taxoReader);
    DrillDownQuery ddq = new DrillDownQuery(config);

View Full Code Here

  public void testChildrenArraysGrowth() throws Exception {
    Directory indexDir = newDirectory();
    TaxonomyWriter tw = new DirectoryTaxonomyWriter(indexDir);
    tw.addCategory(new FacetLabel("hi", "there"));
    tw.commit();
    TaxonomyReader tr = new DirectoryTaxonomyReader(indexDir);
    ParallelTaxonomyArrays ca = tr.getParallelTaxonomyArrays();
    assertEquals(3, tr.getSize());
    assertEquals(3, ca.siblings().length);
    assertEquals(3, ca.children().length);
    assertTrue(Arrays.equals(new int[] { 1, 2, -1 }, ca.children()));
    assertTrue(Arrays.equals(new int[] { -1, -1, -1 }, ca.siblings()));
    tw.addCategory(new FacetLabel("hi", "ho"));
    tw.addCategory(new FacetLabel("hello"));
    tw.commit();
    // Before refresh, nothing changed..
    ParallelTaxonomyArrays newca = tr.getParallelTaxonomyArrays();
    assertSame(newca, ca); // we got exactly the same object
    assertEquals(3, tr.getSize());
    assertEquals(3, ca.siblings().length);
    assertEquals(3, ca.children().length);
    // After the refresh, things change:
    TaxonomyReader newtr = TaxonomyReader.openIfChanged(tr);
    assertNotNull(newtr);
    tr.close();
    tr = newtr;
    ca = tr.getParallelTaxonomyArrays();
    assertEquals(5, tr.getSize());
    assertEquals(5, ca.siblings().length);
    assertEquals(5, ca.children().length);
    assertTrue(Arrays.equals(new int[] { 4, 3, -1, -1, -1 }, ca.children()));
    assertTrue(Arrays.equals(new int[] { -1, -1, -1, 2, 1 }, ca.siblings()));
    tw.close();
    tr.close();
    indexDir.close();
  }

View Full Code Here

    TaxonomyWriter twBase = new DirectoryTaxonomyWriter(indexDirBase);
    twBase.addCategory(new FacetLabel("a", "0"));
    final FacetLabel abPath = new FacetLabel("a", "b");
    twBase.addCategory(abPath);
    twBase.commit();
    TaxonomyReader trBase = new DirectoryTaxonomyReader(indexDirBase);


    final ParallelTaxonomyArrays ca1 = trBase.getParallelTaxonomyArrays();
    
    final int abOrd = trBase.getOrdinal(abPath);
    final int abYoungChildBase1 = ca1.children()[abOrd]; 
    
    final int numCategories = atLeast(800);
    for (int i = 0; i < numCategories; i++) {
      twBase.addCategory(new FacetLabel("a", "b", Integer.toString(i)));
    }
    twBase.close();
    
    TaxonomyReader newTaxoReader = TaxonomyReader.openIfChanged(trBase);
    assertNotNull(newTaxoReader);
    trBase.close();
    trBase = newTaxoReader;
    
    final ParallelTaxonomyArrays ca2 = trBase.getParallelTaxonomyArrays();
    final int abYoungChildBase2 = ca2.children()[abOrd];
    
    int numRetries = atLeast(50);
    for (int retry = 0; retry < numRetries; retry++) {
      assertConsistentYoungestChild(abPath, abOrd, abYoungChildBase1, abYoungChildBase2, retry, numCategories);
    }
    
    trBase.close();
    indexDirBase.close();
  }

View Full Code Here

    TaxonomyWriter tw = new DirectoryTaxonomyWriter(indexDir);
    tw.addCategory(new FacetLabel("a", "0"));
    tw.addCategory(abPath);
    tw.commit();
    
    final DirectoryTaxonomyReader tr = new DirectoryTaxonomyReader(indexDir);
    for (int i = 0; i < numCategories; i++) {
      final FacetLabel cp = new FacetLabel("a", "b", Integer.toString(i));
      tw.addCategory(cp);
      assertEquals("Ordinal of "+cp+" must be invalid until Taxonomy Reader was refreshed", TaxonomyReader.INVALID_ORDINAL, tr.getOrdinal(cp));
    }
    tw.close();
    
    final AtomicBoolean stop = new AtomicBoolean(false);
    final Throwable[] error = new Throwable[] { null };
    final int retrieval[] = { 0 }; 
    
    Thread thread = new Thread("Child Arrays Verifier") {
      @Override
      public void run() {
        setPriority(1 + getPriority());
        try {
          while (!stop.get()) {
            int lastOrd = tr.getParallelTaxonomyArrays().parents().length - 1;
            assertNotNull("path of last-ord " + lastOrd + " is not found!", tr.getPath(lastOrd));
            assertChildrenArrays(tr.getParallelTaxonomyArrays(), retry, retrieval[0]++);
            sleep(10); // don't starve refresh()'s CPU, which sleeps every 50 bytes for 1 ms
          }
        } catch (Throwable e) {
          error[0] = e;
          stop.set(true);
        }
      }


      private void assertChildrenArrays(ParallelTaxonomyArrays ca, int retry, int retrieval) {
        final int abYoungChild = ca.children()[abOrd];
        assertTrue(
            "Retry "+retry+": retrieval: "+retrieval+": wrong youngest child for category "+abPath+" (ord="+abOrd+
            ") - must be either "+abYoungChildBase1+" or "+abYoungChildBase2+" but was: "+abYoungChild,
            abYoungChildBase1==abYoungChild ||
            abYoungChildBase2==ca.children()[abOrd]);
      }
    };
    thread.start();
    
    indexDir.setSleepMillis(1); // some delay for refresh
    TaxonomyReader newTaxoReader = TaxonomyReader.openIfChanged(tr);
    if (newTaxoReader != null) {
      newTaxoReader.close();
    }
    
    stop.set(true);
    thread.join();
    assertNull("Unexpcted exception at retry "+retry+" retrieval "+retrieval[0]+": \n"+stackTraceStr(error[0]), error[0]);
    
    tr.close();
  }

View Full Code Here

  @Test
  public void testSeparateReaderAndWriter() throws Exception {
    Directory indexDir = newDirectory();
    TaxonomyWriter tw = new DirectoryTaxonomyWriter(indexDir);
    tw.commit();
    TaxonomyReader tr = new DirectoryTaxonomyReader(indexDir);


    assertEquals(1, tr.getSize()); // the empty taxonomy has size 1 (the root)
    tw.addCategory(new FacetLabel("Author"));
    assertEquals(1, tr.getSize()); // still root only...
    assertNull(TaxonomyReader.openIfChanged(tr)); // this is not enough, because tw.commit() hasn't been done yet
    assertEquals(1, tr.getSize()); // still root only...
    tw.commit();
    assertEquals(1, tr.getSize()); // still root only...
    TaxonomyReader newTaxoReader = TaxonomyReader.openIfChanged(tr);
    assertNotNull(newTaxoReader);
    tr.close();
    tr = newTaxoReader;
    
    int author = 1;
    try {
      assertEquals(TaxonomyReader.ROOT_ORDINAL, tr.getParallelTaxonomyArrays().parents()[author]);
      // ok
    } catch (ArrayIndexOutOfBoundsException e) {
      fail("After category addition, commit() and refresh(), getParent for "+author+" should NOT throw exception");
    }
    assertEquals(2, tr.getSize()); // finally, see there are two categories


    // now, add another category, and verify that after commit and refresh
    // the parent of this category is correct (this requires the reader
    // to correctly update its prefetched parent vector), and that the
    // old information also wasn't ruined:
    tw.addCategory(new FacetLabel("Author", "Richard Dawkins"));
    int dawkins = 2;
    tw.commit();
    newTaxoReader = TaxonomyReader.openIfChanged(tr);
    assertNotNull(newTaxoReader);
    tr.close();
    tr = newTaxoReader;
    int[] parents = tr.getParallelTaxonomyArrays().parents();
    assertEquals(author, parents[dawkins]);
    assertEquals(TaxonomyReader.ROOT_ORDINAL, parents[author]);
    assertEquals(TaxonomyReader.INVALID_ORDINAL, parents[TaxonomyReader.ROOT_ORDINAL]);
    assertEquals(3, tr.getSize()); 
    tw.close();
    tr.close();
    indexDir.close();
  }

View Full Code Here

  @Test
  public void testSeparateReaderAndWriter2() throws Exception {
    Directory indexDir = newDirectory();
    TaxonomyWriter tw = new DirectoryTaxonomyWriter(indexDir);
    tw.commit();
    TaxonomyReader tr = new DirectoryTaxonomyReader(indexDir);


    // Test getOrdinal():
    FacetLabel author = new FacetLabel("Author");


    assertEquals(1, tr.getSize()); // the empty taxonomy has size 1 (the root)
    assertEquals(TaxonomyReader.INVALID_ORDINAL, tr.getOrdinal(author));
    tw.addCategory(author);
    // before commit and refresh, no change:
    assertEquals(TaxonomyReader.INVALID_ORDINAL, tr.getOrdinal(author));
    assertEquals(1, tr.getSize()); // still root only...
    assertNull(TaxonomyReader.openIfChanged(tr)); // this is not enough, because tw.commit() hasn't been done yet
    assertEquals(TaxonomyReader.INVALID_ORDINAL, tr.getOrdinal(author));
    assertEquals(1, tr.getSize()); // still root only...
    tw.commit();
    // still not enough before refresh:
    assertEquals(TaxonomyReader.INVALID_ORDINAL, tr.getOrdinal(author));
    assertEquals(1, tr.getSize()); // still root only...
    TaxonomyReader newTaxoReader = TaxonomyReader.openIfChanged(tr);
    assertNotNull(newTaxoReader);
    tr.close();
    tr = newTaxoReader;
    assertEquals(1, tr.getOrdinal(author));
    assertEquals(2, tr.getSize());
    tw.close();
    tr.close();
    indexDir.close();
  }

View Full Code Here

0 1 2 3 4 5 6 7 8 9

TOP

Related Classes of org.apache.lucene.facet.taxonomy.directory.DirectoryTaxonomyReader

com.gentics.cr.lucene.facets.taxonomy.taxonomyaccessor.DefaultTaxonomyAccessor

org.apache.lucene.benchmark.byTask.tasks.OpenTaxonomyReaderTask

org.apache.lucene.demo.facet.AssociationsFacetsExample

org.apache.lucene.demo.facet.ExpressionAggregationFacetsExample

org.apache.lucene.demo.facet.MultiCategoryListsFacetsExample

org.apache.lucene.demo.facet.SimpleFacetsExample

org.apache.lucene.document.Document

org.apache.lucene.facet.associations.AssociationsFacetRequestTest

org.apache.lucene.facet.complements.TestTotalFacetCounts

org.apache.lucene.facet.complements.TestTotalFacetCountsCache

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.