Examples with HyperUniqueFinalizingPostAggregator io.druid.query.aggregation.hyperloglog.HyperUniqueFinalizingPostAggregator used on opensource projects

Example 1 with HyperUniqueFinalizingPostAggregator

use of io.druid.query.aggregation.hyperloglog.HyperUniqueFinalizingPostAggregator in project druid by druid-io.

the class GroupByQueryRunnerTest method testGroupByWithUniquesAndPostAggWithSameName.

@Test(expected = IllegalArgumentException.class)
public void testGroupByWithUniquesAndPostAggWithSameName() {
    GroupByQuery query = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.dataSource).setQuerySegmentSpec(QueryRunnerTestHelper.firstToThird).setAggregatorSpecs(Arrays.<AggregatorFactory>asList(QueryRunnerTestHelper.rowsCount, new HyperUniquesAggregatorFactory("quality_uniques", "quality_uniques"))).setPostAggregatorSpecs(Arrays.<PostAggregator>asList(new HyperUniqueFinalizingPostAggregator("quality_uniques", "quality_uniques"))).setGranularity(QueryRunnerTestHelper.allGran).build();
    List<Row> expectedResults = Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "rows", 26L, "quality_uniques", QueryRunnerTestHelper.UNIQUES_9));
    Iterable<Row> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "");
}

Example 2 with HyperUniqueFinalizingPostAggregator

use of io.druid.query.aggregation.hyperloglog.HyperUniqueFinalizingPostAggregator in project druid by druid-io.

the class GroupByQueryRunnerTest method testGroupByWithHavingOnHyperUnique.

@Test
public void testGroupByWithHavingOnHyperUnique() {
    GroupByQuery query = new GroupByQuery.Builder().setDataSource(QueryRunnerTestHelper.dataSource).setGranularity(QueryRunnerTestHelper.allGran).setDimensions(Arrays.<DimensionSpec>asList(new DefaultDimensionSpec(QueryRunnerTestHelper.marketDimension, QueryRunnerTestHelper.marketDimension))).setInterval(QueryRunnerTestHelper.fullOnInterval).setLimitSpec(new DefaultLimitSpec(Lists.newArrayList(new OrderByColumnSpec(QueryRunnerTestHelper.uniqueMetric, OrderByColumnSpec.Direction.DESCENDING)), 3)).setHavingSpec(new GreaterThanHavingSpec(QueryRunnerTestHelper.uniqueMetric, 8)).setAggregatorSpecs(Lists.<AggregatorFactory>newArrayList(QueryRunnerTestHelper.qualityUniques)).setPostAggregatorSpecs(Lists.<PostAggregator>newArrayList(new HyperUniqueFinalizingPostAggregator(QueryRunnerTestHelper.hyperUniqueFinalizingPostAggMetric, QueryRunnerTestHelper.uniqueMetric))).build();
    List<Row> expectedResults = Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("1970-01-01T00:00:00.000Z", "market", "spot", QueryRunnerTestHelper.uniqueMetric, QueryRunnerTestHelper.UNIQUES_9, QueryRunnerTestHelper.hyperUniqueFinalizingPostAggMetric, QueryRunnerTestHelper.UNIQUES_9));
    // havingSpec equalTo/greaterThan/lessThan do not work on complex aggregators, even if they could be finalized.
    // See also: https://github.com/druid-io/druid/issues/2507
    expectedException.expect(ParseException.class);
    expectedException.expectMessage("Unknown type[class io.druid.hll.HLLCV1]");
    Iterable<Row> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "order-limit");
}

Example 3 with HyperUniqueFinalizingPostAggregator

use of io.druid.query.aggregation.hyperloglog.HyperUniqueFinalizingPostAggregator in project druid by druid-io.

the class GroupByQueryRunnerTest method testGroupByWithOrderOnHyperUnique.

@Test
public void testGroupByWithOrderOnHyperUnique() {
    GroupByQuery query = new GroupByQuery.Builder().setDataSource(QueryRunnerTestHelper.dataSource).setGranularity(QueryRunnerTestHelper.allGran).setDimensions(Arrays.<DimensionSpec>asList(new DefaultDimensionSpec(QueryRunnerTestHelper.marketDimension, QueryRunnerTestHelper.marketDimension))).setInterval(QueryRunnerTestHelper.fullOnInterval).setLimitSpec(new DefaultLimitSpec(Lists.newArrayList(new OrderByColumnSpec(QueryRunnerTestHelper.uniqueMetric, OrderByColumnSpec.Direction.DESCENDING)), 3)).setAggregatorSpecs(Lists.<AggregatorFactory>newArrayList(QueryRunnerTestHelper.qualityUniques)).setPostAggregatorSpecs(Lists.<PostAggregator>newArrayList(new HyperUniqueFinalizingPostAggregator(QueryRunnerTestHelper.hyperUniqueFinalizingPostAggMetric, QueryRunnerTestHelper.uniqueMetric))).build();
    List<Row> expectedResults = Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("1970-01-01T00:00:00.000Z", "market", "spot", QueryRunnerTestHelper.uniqueMetric, QueryRunnerTestHelper.UNIQUES_9, QueryRunnerTestHelper.hyperUniqueFinalizingPostAggMetric, QueryRunnerTestHelper.UNIQUES_9), GroupByQueryRunnerTestHelper.createExpectedRow("1970-01-01T00:00:00.000Z", "market", "upfront", QueryRunnerTestHelper.uniqueMetric, QueryRunnerTestHelper.UNIQUES_2, QueryRunnerTestHelper.hyperUniqueFinalizingPostAggMetric, QueryRunnerTestHelper.UNIQUES_2), GroupByQueryRunnerTestHelper.createExpectedRow("1970-01-01T00:00:00.000Z", "market", "total_market", QueryRunnerTestHelper.uniqueMetric, QueryRunnerTestHelper.UNIQUES_2, QueryRunnerTestHelper.hyperUniqueFinalizingPostAggMetric, QueryRunnerTestHelper.UNIQUES_2));
    Iterable<Row> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "order-limit");
}

Example 4 with HyperUniqueFinalizingPostAggregator

use of io.druid.query.aggregation.hyperloglog.HyperUniqueFinalizingPostAggregator in project druid by druid-io.

the class TopNQueryRunnerTest method testTopNOverHyperUniqueFinalizingPostAggregator.

@Test
public void testTopNOverHyperUniqueFinalizingPostAggregator() {
    TopNQuery query = new TopNQueryBuilder().dataSource(QueryRunnerTestHelper.dataSource).granularity(QueryRunnerTestHelper.allGran).dimension(QueryRunnerTestHelper.marketDimension).metric(QueryRunnerTestHelper.hyperUniqueFinalizingPostAggMetric).threshold(3).intervals(QueryRunnerTestHelper.fullOnInterval).aggregators(Arrays.<AggregatorFactory>asList(QueryRunnerTestHelper.qualityUniques)).postAggregators(Arrays.<PostAggregator>asList(new HyperUniqueFinalizingPostAggregator(QueryRunnerTestHelper.hyperUniqueFinalizingPostAggMetric, QueryRunnerTestHelper.uniqueMetric))).build();
    List<Result<TopNResultValue>> expectedResults = Arrays.asList(new Result<>(new DateTime("2011-01-12T00:00:00.000Z"), new TopNResultValue(Arrays.<Map<String, Object>>asList(ImmutableMap.<String, Object>builder().put("market", "spot").put(QueryRunnerTestHelper.uniqueMetric, QueryRunnerTestHelper.UNIQUES_9).put(QueryRunnerTestHelper.hyperUniqueFinalizingPostAggMetric, QueryRunnerTestHelper.UNIQUES_9).build(), ImmutableMap.<String, Object>builder().put("market", "total_market").put(QueryRunnerTestHelper.uniqueMetric, QueryRunnerTestHelper.UNIQUES_2).put(QueryRunnerTestHelper.hyperUniqueFinalizingPostAggMetric, QueryRunnerTestHelper.UNIQUES_2).build(), ImmutableMap.<String, Object>builder().put("market", "upfront").put(QueryRunnerTestHelper.uniqueMetric, QueryRunnerTestHelper.UNIQUES_2).put(QueryRunnerTestHelper.hyperUniqueFinalizingPostAggMetric, QueryRunnerTestHelper.UNIQUES_2).build()))));
    assertExpectedResults(expectedResults, query);
}

Example 5 with HyperUniqueFinalizingPostAggregator

use of io.druid.query.aggregation.hyperloglog.HyperUniqueFinalizingPostAggregator in project druid by druid-io.

the class ApproxCountDistinctSqlAggregator method toDruidAggregation.

@Override
public Aggregation toDruidAggregation(final String name, final RowSignature rowSignature, final DruidOperatorTable operatorTable, final PlannerContext plannerContext, final List<Aggregation> existingAggregations, final Project project, final AggregateCall aggregateCall, final DimFilter filter) {
    final RexNode rexNode = Expressions.fromFieldAccess(rowSignature, project, Iterables.getOnlyElement(aggregateCall.getArgList()));
    final RowExtraction rex = Expressions.toRowExtraction(operatorTable, plannerContext, rowSignature.getRowOrder(), rexNode);
    if (rex == null) {
        return null;
    }
    final AggregatorFactory aggregatorFactory;
    if (rowSignature.getColumnType(rex.getColumn()) == ValueType.COMPLEX) {
        aggregatorFactory = new HyperUniquesAggregatorFactory(name, rex.getColumn());
    } else {
        final SqlTypeName sqlTypeName = rexNode.getType().getSqlTypeName();
        final ValueType outputType = Calcites.getValueTypeForSqlTypeName(sqlTypeName);
        if (outputType == null) {
            throw new ISE("Cannot translate sqlTypeName[%s] to Druid type for field[%s]", sqlTypeName, name);
        }
        final DimensionSpec dimensionSpec = rex.toDimensionSpec(rowSignature, null, ValueType.STRING);
        if (dimensionSpec == null) {
            return null;
        }
        aggregatorFactory = new CardinalityAggregatorFactory(name, ImmutableList.of(dimensionSpec), false);
    }
    return Aggregation.createFinalizable(ImmutableList.<AggregatorFactory>of(aggregatorFactory), null, new PostAggregatorFactory() {

        @Override
        public PostAggregator factorize(String outputName) {
            return new HyperUniqueFinalizingPostAggregator(outputName, name);
        }
    }).filter(filter);
}