Search in sources :

Example 36 with CountAggregatorFactory

use of org.apache.druid.query.aggregation.CountAggregatorFactory in project druid by druid-io.

the class CalciteCorrelatedQueryTest method testCorrelatedSubqueryWithCorrelatedQueryFilter.

@Test
@Parameters(source = QueryContextForJoinProvider.class)
public void testCorrelatedSubqueryWithCorrelatedQueryFilter(Map<String, Object> queryContext) throws Exception {
    cannotVectorize();
    queryContext = withLeftDirectAccessEnabled(queryContext);
    testQuery("select country, ANY_VALUE(\n" + "        select max(\"users\") from (\n" + "            select floor(__time to day), count(user) \"users\" from visits f where f.country = visits.country and f.city = 'A' group by 1\n" + "        )\n" + "     ) as \"dailyVisits\"\n" + "from visits \n" + " where city = 'B'" + " group by 1", queryContext, ImmutableList.of(GroupByQuery.builder().setDataSource(join(new TableDataSource(CalciteTests.USERVISITDATASOURCE), new QueryDataSource(GroupByQuery.builder().setDataSource(GroupByQuery.builder().setDataSource(CalciteTests.USERVISITDATASOURCE).setQuerySegmentSpec(querySegmentSpec(Intervals.ETERNITY)).setVirtualColumns(new ExpressionVirtualColumn("v0", "timestamp_floor(\"__time\",'P1D',null,'UTC')", ColumnType.LONG, TestExprMacroTable.INSTANCE)).setDimensions(new DefaultDimensionSpec("v0", "d0", ColumnType.LONG), new DefaultDimensionSpec("country", "d1")).setAggregatorSpecs(new FilteredAggregatorFactory(new CountAggregatorFactory("a0"), not(selector("user", null, null)))).setDimFilter(and(selector("city", "A", null), not(selector("country", null, null)))).setContext(withTimestampResultContext(queryContext, "d0", Granularities.DAY)).setGranularity(new AllGranularity()).build()).setQuerySegmentSpec(querySegmentSpec(Intervals.ETERNITY)).setDimensions(new DefaultDimensionSpec("d1", "_d0")).setAggregatorSpecs(new LongMaxAggregatorFactory("_a0", "a0")).setGranularity(new AllGranularity()).setContext(queryContext).build()), "j0.", equalsCondition(makeColumnExpression("country"), makeColumnExpression("j0._d0")), JoinType.LEFT, selector("city", "B", null))).setQuerySegmentSpec(querySegmentSpec(Intervals.ETERNITY)).setDimensions(new DefaultDimensionSpec("country", "d0")).setAggregatorSpecs(new LongAnyAggregatorFactory("a0", "j0._a0")).setGranularity(new AllGranularity()).setContext(queryContext).build()), ImmutableList.of(new Object[] { "canada", 2L }));
}
Also used : FilteredAggregatorFactory(org.apache.druid.query.aggregation.FilteredAggregatorFactory) ExpressionVirtualColumn(org.apache.druid.segment.virtual.ExpressionVirtualColumn) TableDataSource(org.apache.druid.query.TableDataSource) QueryDataSource(org.apache.druid.query.QueryDataSource) CountAggregatorFactory(org.apache.druid.query.aggregation.CountAggregatorFactory) LongAnyAggregatorFactory(org.apache.druid.query.aggregation.any.LongAnyAggregatorFactory) AllGranularity(org.apache.druid.java.util.common.granularity.AllGranularity) LongMaxAggregatorFactory(org.apache.druid.query.aggregation.LongMaxAggregatorFactory) DefaultDimensionSpec(org.apache.druid.query.dimension.DefaultDimensionSpec) Parameters(junitparams.Parameters) Test(org.junit.Test)

Example 37 with CountAggregatorFactory

use of org.apache.druid.query.aggregation.CountAggregatorFactory in project druid by druid-io.

the class CalciteCorrelatedQueryTest method testCorrelatedSubquery.

@Test
@Parameters(source = QueryContextForJoinProvider.class)
public void testCorrelatedSubquery(Map<String, Object> queryContext) throws Exception {
    cannotVectorize();
    queryContext = withLeftDirectAccessEnabled(queryContext);
    testQuery("select country, ANY_VALUE(\n" + "        select avg(\"users\") from (\n" + "            select floor(__time to day), count(distinct user) \"users\" from visits f where f.country = visits.country group by 1\n" + "        )\n" + "     ) as \"DAU\"\n" + "from visits \n" + "group by 1", queryContext, ImmutableList.of(GroupByQuery.builder().setDataSource(join(new TableDataSource(CalciteTests.USERVISITDATASOURCE), new QueryDataSource(GroupByQuery.builder().setDataSource(GroupByQuery.builder().setDataSource(CalciteTests.USERVISITDATASOURCE).setQuerySegmentSpec(querySegmentSpec(Intervals.ETERNITY)).setVirtualColumns(new ExpressionVirtualColumn("v0", "timestamp_floor(\"__time\",'P1D',null,'UTC')", ColumnType.LONG, TestExprMacroTable.INSTANCE)).setDimFilter(not(selector("country", null, null))).setDimensions(new DefaultDimensionSpec("v0", "d0", ColumnType.LONG), new DefaultDimensionSpec("country", "d1")).setAggregatorSpecs(new CardinalityAggregatorFactory("a0:a", null, Collections.singletonList(new DefaultDimensionSpec("user", "user")), false, true)).setPostAggregatorSpecs(Collections.singletonList(new HyperUniqueFinalizingPostAggregator("a0", "a0:a"))).setContext(withTimestampResultContext(queryContext, "d0", Granularities.DAY)).setGranularity(new AllGranularity()).build()).setQuerySegmentSpec(querySegmentSpec(Intervals.ETERNITY)).setDimensions(new DefaultDimensionSpec("d1", "_d0")).setAggregatorSpecs(new LongSumAggregatorFactory("_a0:sum", "a0"), useDefault ? new CountAggregatorFactory("_a0:count") : new FilteredAggregatorFactory(new CountAggregatorFactory("_a0:count"), not(selector("a0", null, null)))).setPostAggregatorSpecs(Collections.singletonList(new ArithmeticPostAggregator("_a0", "quotient", Arrays.asList(new FieldAccessPostAggregator(null, "_a0:sum"), new FieldAccessPostAggregator(null, "_a0:count"))))).setGranularity(new AllGranularity()).setContext(queryContext).build()), "j0.", equalsCondition(makeColumnExpression("country"), makeColumnExpression("j0._d0")), JoinType.LEFT)).setQuerySegmentSpec(querySegmentSpec(Intervals.ETERNITY)).setDimensions(new DefaultDimensionSpec("country", "d0")).setAggregatorSpecs(new LongAnyAggregatorFactory("a0", "j0._a0")).setGranularity(new AllGranularity()).setContext(queryContext).build()), ImmutableList.of(new Object[] { "India", 2L }, new Object[] { "USA", 1L }, new Object[] { "canada", 3L }));
}
Also used : FilteredAggregatorFactory(org.apache.druid.query.aggregation.FilteredAggregatorFactory) ArithmeticPostAggregator(org.apache.druid.query.aggregation.post.ArithmeticPostAggregator) FieldAccessPostAggregator(org.apache.druid.query.aggregation.post.FieldAccessPostAggregator) AllGranularity(org.apache.druid.java.util.common.granularity.AllGranularity) LongSumAggregatorFactory(org.apache.druid.query.aggregation.LongSumAggregatorFactory) HyperUniqueFinalizingPostAggregator(org.apache.druid.query.aggregation.hyperloglog.HyperUniqueFinalizingPostAggregator) DefaultDimensionSpec(org.apache.druid.query.dimension.DefaultDimensionSpec) ExpressionVirtualColumn(org.apache.druid.segment.virtual.ExpressionVirtualColumn) TableDataSource(org.apache.druid.query.TableDataSource) QueryDataSource(org.apache.druid.query.QueryDataSource) CountAggregatorFactory(org.apache.druid.query.aggregation.CountAggregatorFactory) LongAnyAggregatorFactory(org.apache.druid.query.aggregation.any.LongAnyAggregatorFactory) CardinalityAggregatorFactory(org.apache.druid.query.aggregation.cardinality.CardinalityAggregatorFactory) Parameters(junitparams.Parameters) Test(org.junit.Test)

Example 38 with CountAggregatorFactory

use of org.apache.druid.query.aggregation.CountAggregatorFactory in project druid by druid-io.

the class CalciteJoinQueryTest method testInnerJoinTwoLookupsToTableUsingNumericColumnInReverse.

@Test
@Parameters(source = QueryContextForJoinProvider.class)
public void testInnerJoinTwoLookupsToTableUsingNumericColumnInReverse(Map<String, Object> queryContext) throws Exception {
    // Like "testInnerJoinTwoLookupsToTableUsingNumericColumn", but the tables are specified backwards.
    cannotVectorize();
    testQuery("SELECT COUNT(*)\n" + "FROM lookup.lookyloo l1\n" + "INNER JOIN lookup.lookyloo l2 ON l1.k = l2.k\n" + "INNER JOIN foo on l2.k = foo.m1", queryContext, ImmutableList.of(Druids.newTimeseriesQueryBuilder().dataSource(join(join(new LookupDataSource("lookyloo"), new LookupDataSource("lookyloo"), "j0.", equalsCondition(makeColumnExpression("k"), makeColumnExpression("j0.k")), JoinType.INNER), new QueryDataSource(newScanQueryBuilder().dataSource(CalciteTests.DATASOURCE1).intervals(querySegmentSpec(Filtration.eternity())).columns("m1").context(QUERY_CONTEXT_DEFAULT).build()), "_j0.", equalsCondition(makeExpression(ColumnType.DOUBLE, "CAST(\"j0.k\", 'DOUBLE')"), DruidExpression.ofColumn(ColumnType.DOUBLE, "_j0.m1")), JoinType.INNER)).intervals(querySegmentSpec(Filtration.eternity())).granularity(Granularities.ALL).aggregators(new CountAggregatorFactory("a0")).context(QUERY_CONTEXT_DEFAULT).build()), ImmutableList.of(new Object[] { 1L }));
}
Also used : QueryDataSource(org.apache.druid.query.QueryDataSource) CountAggregatorFactory(org.apache.druid.query.aggregation.CountAggregatorFactory) LookupDataSource(org.apache.druid.query.LookupDataSource) Parameters(junitparams.Parameters) Test(org.junit.Test)

Example 39 with CountAggregatorFactory

use of org.apache.druid.query.aggregation.CountAggregatorFactory in project druid by druid-io.

the class CalciteJoinQueryTest method testJoinUnionTablesOnLookup.

@Test
@Parameters(source = QueryContextForJoinProvider.class)
public void testJoinUnionTablesOnLookup(Map<String, Object> queryContext) throws Exception {
    // Cannot vectorize JOIN operator.
    cannotVectorize();
    testQuery("SELECT lookyloo.v, COUNT(*)\n" + "FROM\n" + "  (SELECT dim2 FROM foo UNION ALL SELECT dim2 FROM numfoo) u\n" + "  LEFT JOIN lookup.lookyloo ON u.dim2 = lookyloo.k\n" + "WHERE lookyloo.v <> 'xa'\n" + "GROUP BY lookyloo.v", queryContext, ImmutableList.of(GroupByQuery.builder().setDataSource(join(new UnionDataSource(ImmutableList.of(new TableDataSource(CalciteTests.DATASOURCE1), new TableDataSource(CalciteTests.DATASOURCE3))), new LookupDataSource("lookyloo"), "j0.", equalsCondition(makeColumnExpression("dim2"), makeColumnExpression("j0.k")), JoinType.LEFT)).setInterval(querySegmentSpec(Filtration.eternity())).setDimFilter(not(selector("j0.v", "xa", null))).setGranularity(Granularities.ALL).setDimensions(dimensions(new DefaultDimensionSpec("j0.v", "d0"))).setAggregatorSpecs(aggregators(new CountAggregatorFactory("a0"))).setContext(queryContext).build()), ImmutableList.of(new Object[] { NULL_STRING, 6L }, new Object[] { "xabc", 2L }));
}
Also used : GlobalTableDataSource(org.apache.druid.query.GlobalTableDataSource) TableDataSource(org.apache.druid.query.TableDataSource) CountAggregatorFactory(org.apache.druid.query.aggregation.CountAggregatorFactory) LookupDataSource(org.apache.druid.query.LookupDataSource) UnionDataSource(org.apache.druid.query.UnionDataSource) DefaultDimensionSpec(org.apache.druid.query.dimension.DefaultDimensionSpec) Parameters(junitparams.Parameters) Test(org.junit.Test)

Example 40 with CountAggregatorFactory

use of org.apache.druid.query.aggregation.CountAggregatorFactory in project druid by druid-io.

the class CalciteArraysQueryTest method testArrayAggGroupByArrayContainsSubquery.

@Test
public void testArrayAggGroupByArrayContainsSubquery() throws Exception {
    cannotVectorize();
    List<Object[]> expectedResults;
    if (useDefault) {
        expectedResults = ImmutableList.of(new Object[] { "", 3L }, new Object[] { "a", 1L }, new Object[] { "abc", 1L });
    } else {
        expectedResults = ImmutableList.of(new Object[] { null, 2L }, new Object[] { "", 1L }, new Object[] { "a", 2L }, new Object[] { "abc", 1L });
    }
    testQuery("SELECT dim2, COUNT(*) FROM foo WHERE ARRAY_CONTAINS((SELECT ARRAY_AGG(DISTINCT dim1) FROM foo WHERE dim1 is not null), dim1) GROUP BY 1", ImmutableList.of(GroupByQuery.builder().setDataSource(join(new TableDataSource(CalciteTests.DATASOURCE1), new QueryDataSource(Druids.newTimeseriesQueryBuilder().dataSource(CalciteTests.DATASOURCE1).intervals(querySegmentSpec(Filtration.eternity())).granularity(Granularities.ALL).filters(not(selector("dim1", null, null))).aggregators(aggregators(new ExpressionLambdaAggregatorFactory("a0", ImmutableSet.of("dim1"), "__acc", "ARRAY<STRING>[]", "ARRAY<STRING>[]", true, true, false, "array_set_add(\"__acc\", \"dim1\")", "array_set_add_all(\"__acc\", \"a0\")", null, null, ExpressionLambdaAggregatorFactory.DEFAULT_MAX_SIZE_BYTES, TestExprMacroTable.INSTANCE))).context(QUERY_CONTEXT_DEFAULT).build()), "j0.", "1", JoinType.LEFT, null)).setInterval(querySegmentSpec(Filtration.eternity())).setDimFilter(new ExpressionDimFilter("array_contains(\"j0.a0\",\"dim1\")", TestExprMacroTable.INSTANCE)).setDimensions(dimensions(new DefaultDimensionSpec("dim2", "d0"))).setAggregatorSpecs(aggregators(new CountAggregatorFactory("a0"))).setGranularity(Granularities.ALL).setLimitSpec(NoopLimitSpec.instance()).setContext(QUERY_CONTEXT_DEFAULT).build()), expectedResults);
}
Also used : ExpressionLambdaAggregatorFactory(org.apache.druid.query.aggregation.ExpressionLambdaAggregatorFactory) TableDataSource(org.apache.druid.query.TableDataSource) QueryDataSource(org.apache.druid.query.QueryDataSource) CountAggregatorFactory(org.apache.druid.query.aggregation.CountAggregatorFactory) ExpressionDimFilter(org.apache.druid.query.filter.ExpressionDimFilter) DefaultDimensionSpec(org.apache.druid.query.dimension.DefaultDimensionSpec) Test(org.junit.Test)

Aggregations

CountAggregatorFactory (org.apache.druid.query.aggregation.CountAggregatorFactory)300 Test (org.junit.Test)249 DefaultDimensionSpec (org.apache.druid.query.dimension.DefaultDimensionSpec)103 InitializedNullHandlingTest (org.apache.druid.testing.InitializedNullHandlingTest)81 LongSumAggregatorFactory (org.apache.druid.query.aggregation.LongSumAggregatorFactory)62 TimeseriesQuery (org.apache.druid.query.timeseries.TimeseriesQuery)50 TableDataSource (org.apache.druid.query.TableDataSource)44 QueryDataSource (org.apache.druid.query.QueryDataSource)41 TimeseriesQueryQueryToolChest (org.apache.druid.query.timeseries.TimeseriesQueryQueryToolChest)40 GroupByQuery (org.apache.druid.query.groupby.GroupByQuery)37 Result (org.apache.druid.query.Result)36 DoubleSumAggregatorFactory (org.apache.druid.query.aggregation.DoubleSumAggregatorFactory)35 FilteredAggregatorFactory (org.apache.druid.query.aggregation.FilteredAggregatorFactory)30 FieldAccessPostAggregator (org.apache.druid.query.aggregation.post.FieldAccessPostAggregator)30 AggregatorFactory (org.apache.druid.query.aggregation.AggregatorFactory)28 IncrementalIndexSegment (org.apache.druid.segment.IncrementalIndexSegment)27 DimensionsSpec (org.apache.druid.data.input.impl.DimensionsSpec)26 GroupByQueryRunnerTest (org.apache.druid.query.groupby.GroupByQueryRunnerTest)26 IncrementalIndex (org.apache.druid.segment.incremental.IncrementalIndex)26 Parameters (junitparams.Parameters)24