Examples with LabelAndValue org.apache.lucene.facet.LabelAndValue used on opensource projects

Example 16 with LabelAndValue

use of org.apache.lucene.facet.LabelAndValue in project lucene-solr by apache.

the class ConcurrentSortedSetDocValuesFacetCounts method getDim.

private final FacetResult getDim(String dim, OrdRange ordRange, int topN) throws IOException {
    TopOrdAndIntQueue q = null;
    int bottomCount = 0;
    int dimCount = 0;
    int childCount = 0;
    TopOrdAndIntQueue.OrdAndValue reuse = null;
    //System.out.println("getDim : " + ordRange.start + " - " + ordRange.end);
    for (int ord = ordRange.start; ord <= ordRange.end; ord++) {
        //System.out.println("  ord=" + ord + " count=" + counts[ord]);
        if (counts.get(ord) > 0) {
            dimCount += counts.get(ord);
            childCount++;
            if (counts.get(ord) > bottomCount) {
                if (reuse == null) {
                    reuse = new TopOrdAndIntQueue.OrdAndValue();
                }
                reuse.ord = ord;
                reuse.value = counts.get(ord);
                if (q == null) {
                    // Lazy init, so we don't create this for the
                    // sparse case unnecessarily
                    q = new TopOrdAndIntQueue(topN);
                }
                reuse = q.insertWithOverflow(reuse);
                if (q.size() == topN) {
                    bottomCount = q.top().value;
                }
            }
        }
    }
    if (q == null) {
        return null;
    }
    LabelAndValue[] labelValues = new LabelAndValue[q.size()];
    for (int i = labelValues.length - 1; i >= 0; i--) {
        TopOrdAndIntQueue.OrdAndValue ordAndValue = q.pop();
        final BytesRef term = dv.lookupOrd(ordAndValue.ord);
        String[] parts = FacetsConfig.stringToPath(term.utf8ToString());
        labelValues[i] = new LabelAndValue(parts[1], ordAndValue.value);
    }
    return new FacetResult(dim, new String[0], dimCount, labelValues, childCount);
}

Example 17 with LabelAndValue

use of org.apache.lucene.facet.LabelAndValue in project lucene-solr by apache.

the class TestRangeFacetCounts method testRandomLongs.

public void testRandomLongs() throws Exception {
    Directory dir = newDirectory();
    RandomIndexWriter w = new RandomIndexWriter(random(), dir);
    int numDocs = atLeast(1000);
    if (VERBOSE) {
        System.out.println("TEST: numDocs=" + numDocs);
    }
    long[] values = new long[numDocs];
    long minValue = Long.MAX_VALUE;
    long maxValue = Long.MIN_VALUE;
    for (int i = 0; i < numDocs; i++) {
        Document doc = new Document();
        long v = random().nextLong();
        values[i] = v;
        doc.add(new NumericDocValuesField("field", v));
        doc.add(new LongPoint("field", v));
        w.addDocument(doc);
        minValue = Math.min(minValue, v);
        maxValue = Math.max(maxValue, v);
    }
    IndexReader r = w.getReader();
    IndexSearcher s = newSearcher(r, false);
    FacetsConfig config = new FacetsConfig();
    int numIters = atLeast(10);
    for (int iter = 0; iter < numIters; iter++) {
        if (VERBOSE) {
            System.out.println("TEST: iter=" + iter);
        }
        int numRange = TestUtil.nextInt(random(), 1, 100);
        LongRange[] ranges = new LongRange[numRange];
        int[] expectedCounts = new int[numRange];
        long minAcceptedValue = Long.MAX_VALUE;
        long maxAcceptedValue = Long.MIN_VALUE;
        for (int rangeID = 0; rangeID < numRange; rangeID++) {
            long min;
            if (rangeID > 0 && random().nextInt(10) == 7) {
                // Use an existing boundary:
                LongRange prevRange = ranges[random().nextInt(rangeID)];
                if (random().nextBoolean()) {
                    min = prevRange.min;
                } else {
                    min = prevRange.max;
                }
            } else {
                min = random().nextLong();
            }
            long max;
            if (rangeID > 0 && random().nextInt(10) == 7) {
                // Use an existing boundary:
                LongRange prevRange = ranges[random().nextInt(rangeID)];
                if (random().nextBoolean()) {
                    max = prevRange.min;
                } else {
                    max = prevRange.max;
                }
            } else {
                max = random().nextLong();
            }
            if (min > max) {
                long x = min;
                min = max;
                max = x;
            }
            boolean minIncl;
            boolean maxIncl;
            // NOTE: max - min >= 0 is here to handle the common overflow case!
            if (max - min >= 0 && max - min < 2) {
                // If max == min or max == min+1, we always do inclusive, else we might pass an empty range and hit exc from LongRange's ctor:
                minIncl = true;
                maxIncl = true;
            } else {
                minIncl = random().nextBoolean();
                maxIncl = random().nextBoolean();
            }
            ranges[rangeID] = new LongRange("r" + rangeID, min, minIncl, max, maxIncl);
            if (VERBOSE) {
                System.out.println("  range " + rangeID + ": " + ranges[rangeID]);
            }
            // expected count:
            for (int i = 0; i < numDocs; i++) {
                boolean accept = true;
                if (minIncl) {
                    accept &= values[i] >= min;
                } else {
                    accept &= values[i] > min;
                }
                if (maxIncl) {
                    accept &= values[i] <= max;
                } else {
                    accept &= values[i] < max;
                }
                if (accept) {
                    expectedCounts[rangeID]++;
                    minAcceptedValue = Math.min(minAcceptedValue, values[i]);
                    maxAcceptedValue = Math.max(maxAcceptedValue, values[i]);
                }
            }
        }
        FacetsCollector sfc = new FacetsCollector();
        s.search(new MatchAllDocsQuery(), sfc);
        Query fastMatchQuery;
        if (random().nextBoolean()) {
            if (random().nextBoolean()) {
                fastMatchQuery = LongPoint.newRangeQuery("field", minValue, maxValue);
            } else {
                fastMatchQuery = LongPoint.newRangeQuery("field", minAcceptedValue, maxAcceptedValue);
            }
        } else {
            fastMatchQuery = null;
        }
        LongValuesSource vs = LongValuesSource.fromLongField("field");
        Facets facets = new LongRangeFacetCounts("field", vs, sfc, fastMatchQuery, ranges);
        FacetResult result = facets.getTopChildren(10, "field");
        assertEquals(numRange, result.labelValues.length);
        for (int rangeID = 0; rangeID < numRange; rangeID++) {
            if (VERBOSE) {
                System.out.println("  range " + rangeID + " expectedCount=" + expectedCounts[rangeID]);
            }
            LabelAndValue subNode = result.labelValues[rangeID];
            assertEquals("r" + rangeID, subNode.label);
            assertEquals(expectedCounts[rangeID], subNode.value.intValue());
            LongRange range = ranges[rangeID];
            // Test drill-down:
            DrillDownQuery ddq = new DrillDownQuery(config);
            if (random().nextBoolean()) {
                ddq.add("field", LongPoint.newRangeQuery("field", range.min, range.max));
            } else {
                ddq.add("field", range.getQuery(fastMatchQuery, vs));
            }
            assertEquals(expectedCounts[rangeID], s.search(ddq, 10).totalHits);
        }
    }
    w.close();
    IOUtils.close(r, dir);
}

Example 18 with LabelAndValue

use of org.apache.lucene.facet.LabelAndValue in project lucene-solr by apache.

the class TestRangeFacetCounts method testRandomDoubles.

public void testRandomDoubles() throws Exception {
    Directory dir = newDirectory();
    RandomIndexWriter w = new RandomIndexWriter(random(), dir);
    int numDocs = atLeast(1000);
    double[] values = new double[numDocs];
    double minValue = Double.POSITIVE_INFINITY;
    double maxValue = Double.NEGATIVE_INFINITY;
    for (int i = 0; i < numDocs; i++) {
        Document doc = new Document();
        double v = random().nextDouble();
        values[i] = v;
        doc.add(new DoubleDocValuesField("field", v));
        doc.add(new DoublePoint("field", v));
        w.addDocument(doc);
        minValue = Math.min(minValue, v);
        maxValue = Math.max(maxValue, v);
    }
    IndexReader r = w.getReader();
    IndexSearcher s = newSearcher(r, false);
    FacetsConfig config = new FacetsConfig();
    int numIters = atLeast(10);
    for (int iter = 0; iter < numIters; iter++) {
        if (VERBOSE) {
            System.out.println("TEST: iter=" + iter);
        }
        int numRange = TestUtil.nextInt(random(), 1, 5);
        DoubleRange[] ranges = new DoubleRange[numRange];
        int[] expectedCounts = new int[numRange];
        double minAcceptedValue = Double.POSITIVE_INFINITY;
        double maxAcceptedValue = Double.NEGATIVE_INFINITY;
        for (int rangeID = 0; rangeID < numRange; rangeID++) {
            double min;
            if (rangeID > 0 && random().nextInt(10) == 7) {
                // Use an existing boundary:
                DoubleRange prevRange = ranges[random().nextInt(rangeID)];
                if (random().nextBoolean()) {
                    min = prevRange.min;
                } else {
                    min = prevRange.max;
                }
            } else {
                min = random().nextDouble();
            }
            double max;
            if (rangeID > 0 && random().nextInt(10) == 7) {
                // Use an existing boundary:
                DoubleRange prevRange = ranges[random().nextInt(rangeID)];
                if (random().nextBoolean()) {
                    max = prevRange.min;
                } else {
                    max = prevRange.max;
                }
            } else {
                max = random().nextDouble();
            }
            if (min > max) {
                double x = min;
                min = max;
                max = x;
            }
            boolean minIncl;
            boolean maxIncl;
            long minAsLong = NumericUtils.doubleToSortableLong(min);
            long maxAsLong = NumericUtils.doubleToSortableLong(max);
            // NOTE: maxAsLong - minAsLong >= 0 is here to handle the common overflow case!
            if (maxAsLong - minAsLong >= 0 && maxAsLong - minAsLong < 2) {
                minIncl = true;
                maxIncl = true;
            } else {
                minIncl = random().nextBoolean();
                maxIncl = random().nextBoolean();
            }
            ranges[rangeID] = new DoubleRange("r" + rangeID, min, minIncl, max, maxIncl);
            // expected count:
            for (int i = 0; i < numDocs; i++) {
                boolean accept = true;
                if (minIncl) {
                    accept &= values[i] >= min;
                } else {
                    accept &= values[i] > min;
                }
                if (maxIncl) {
                    accept &= values[i] <= max;
                } else {
                    accept &= values[i] < max;
                }
                if (accept) {
                    expectedCounts[rangeID]++;
                    minAcceptedValue = Math.min(minAcceptedValue, values[i]);
                    maxAcceptedValue = Math.max(maxAcceptedValue, values[i]);
                }
            }
        }
        FacetsCollector sfc = new FacetsCollector();
        s.search(new MatchAllDocsQuery(), sfc);
        Query fastMatchFilter;
        if (random().nextBoolean()) {
            if (random().nextBoolean()) {
                fastMatchFilter = DoublePoint.newRangeQuery("field", minValue, maxValue);
            } else {
                fastMatchFilter = DoublePoint.newRangeQuery("field", minAcceptedValue, maxAcceptedValue);
            }
        } else {
            fastMatchFilter = null;
        }
        DoubleValuesSource vs = DoubleValuesSource.fromDoubleField("field");
        Facets facets = new DoubleRangeFacetCounts("field", vs, sfc, fastMatchFilter, ranges);
        FacetResult result = facets.getTopChildren(10, "field");
        assertEquals(numRange, result.labelValues.length);
        for (int rangeID = 0; rangeID < numRange; rangeID++) {
            if (VERBOSE) {
                System.out.println("  range " + rangeID + " expectedCount=" + expectedCounts[rangeID]);
            }
            LabelAndValue subNode = result.labelValues[rangeID];
            assertEquals("r" + rangeID, subNode.label);
            assertEquals(expectedCounts[rangeID], subNode.value.intValue());
            DoubleRange range = ranges[rangeID];
            // Test drill-down:
            DrillDownQuery ddq = new DrillDownQuery(config);
            if (random().nextBoolean()) {
                ddq.add("field", DoublePoint.newRangeQuery("field", range.min, range.max));
            } else {
                ddq.add("field", range.getQuery(fastMatchFilter, vs));
            }
            assertEquals(expectedCounts[rangeID], s.search(ddq, 10).totalHits);
        }
    }
    w.close();
    IOUtils.close(r, dir);
}

Example 19 with LabelAndValue

use of org.apache.lucene.facet.LabelAndValue in project lucene-solr by apache.

the class TestTaxonomyFacetCounts2 method testNoParents.

@Test
public void testNoParents() throws Exception {
    DirectoryReader indexReader = DirectoryReader.open(indexDir);
    TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);
    IndexSearcher searcher = newSearcher(indexReader);
    FacetsCollector sfc = new FacetsCollector();
    searcher.search(new MatchAllDocsQuery(), sfc);
    Facets facets = getTaxonomyFacetCounts(taxoReader, getConfig(), sfc);
    FacetResult result = facets.getTopChildren(NUM_CHILDREN_CP_C, CP_C);
    assertEquals(allExpectedCounts.get(CP_C), result.value);
    for (LabelAndValue labelValue : result.labelValues) {
        assertEquals(allExpectedCounts.get(CP_C + "/" + labelValue.label), labelValue.value);
    }
    result = facets.getTopChildren(NUM_CHILDREN_CP_D, CP_D);
    assertEquals(allExpectedCounts.get(CP_C), result.value);
    for (LabelAndValue labelValue : result.labelValues) {
        assertEquals(allExpectedCounts.get(CP_D + "/" + labelValue.label), labelValue.value);
    }
    IOUtils.close(indexReader, taxoReader);
}

Example 20 with LabelAndValue

use of org.apache.lucene.facet.LabelAndValue in project lucene-solr by apache.

the class TestTaxonomyFacetCounts method testSegmentsWithoutCategoriesOrResults.

public void testSegmentsWithoutCategoriesOrResults() throws Exception {
    // tests the accumulator when there are segments with no results
    Directory indexDir = newDirectory();
    Directory taxoDir = newDirectory();
    IndexWriterConfig iwc = newIndexWriterConfig(new MockAnalyzer(random()));
    // prevent merges
    iwc.setMergePolicy(NoMergePolicy.INSTANCE);
    IndexWriter indexWriter = new IndexWriter(indexDir, iwc);
    TaxonomyWriter taxoWriter = new DirectoryTaxonomyWriter(taxoDir);
    FacetsConfig config = new FacetsConfig();
    // 1st segment, no content, with categories
    indexTwoDocs(taxoWriter, indexWriter, config, false);
    // 2nd segment, with content, no categories
    indexTwoDocs(taxoWriter, indexWriter, null, true);
    // 3rd segment ok
    indexTwoDocs(taxoWriter, indexWriter, config, true);
    // 4th segment, no content, or categories
    indexTwoDocs(taxoWriter, indexWriter, null, false);
    // 5th segment, with content, no categories
    indexTwoDocs(taxoWriter, indexWriter, null, true);
    // 6th segment, with content, with categories
    indexTwoDocs(taxoWriter, indexWriter, config, true);
    // 7th segment, with content, no categories
    indexTwoDocs(taxoWriter, indexWriter, null, true);
    indexWriter.close();
    IOUtils.close(taxoWriter);
    DirectoryReader indexReader = DirectoryReader.open(indexDir);
    TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);
    IndexSearcher indexSearcher = newSearcher(indexReader);
    // search for "f:a", only segments 1 and 3 should match results
    Query q = new TermQuery(new Term("f", "a"));
    FacetsCollector sfc = new FacetsCollector();
    indexSearcher.search(q, sfc);
    Facets facets = getTaxonomyFacetCounts(taxoReader, config, sfc);
    FacetResult result = facets.getTopChildren(10, "A");
    assertEquals("wrong number of children", 2, result.labelValues.length);
    for (LabelAndValue labelValue : result.labelValues) {
        assertEquals("wrong weight for child " + labelValue.label, 2, labelValue.value.intValue());
    }
    IOUtils.close(indexReader, taxoReader, indexDir, taxoDir);
}