Examples with QueryDataSource org.apache.druid.query.QueryDataSource used on opensource projects

Example 86 with QueryDataSource

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

the class CalciteJoinQueryTest method testTopNFilterJoinWithProjection.

@Test
@Parameters(source = QueryContextForJoinProvider.class)
public void testTopNFilterJoinWithProjection(Map<String, Object> queryContext) throws Exception {
    // Cannot vectorize JOIN operator.
    cannotVectorize();
    // Filters on top N values of some dimension by using an inner join. Also projects the outer dimension.
    testQuery("SELECT SUBSTRING(t1.dim1, 1, 10), SUM(t1.cnt)\n" + "FROM druid.foo t1\n" + "  INNER JOIN (\n" + "  SELECT\n" + "    SUM(cnt) AS sum_cnt,\n" + "    dim2\n" + "  FROM druid.foo\n" + "  GROUP BY dim2\n" + "  ORDER BY 1 DESC\n" + "  LIMIT 2\n" + ") t2 ON (t1.dim2 = t2.dim2)\n" + "GROUP BY SUBSTRING(t1.dim1, 1, 10)", queryContext, ImmutableList.of(GroupByQuery.builder().setDataSource(join(new TableDataSource(CalciteTests.DATASOURCE1), new QueryDataSource(new TopNQueryBuilder().dataSource(CalciteTests.DATASOURCE1).intervals(querySegmentSpec(Filtration.eternity())).granularity(Granularities.ALL).dimension(new DefaultDimensionSpec("dim2", "d0")).aggregators(new LongSumAggregatorFactory("a0", "cnt")).metric("a0").threshold(2).context(QUERY_CONTEXT_DEFAULT).build()), "j0.", equalsCondition(makeColumnExpression("dim2"), makeColumnExpression("j0.d0")), JoinType.INNER)).setInterval(querySegmentSpec(Filtration.eternity())).setGranularity(Granularities.ALL).setDimensions(dimensions(new ExtractionDimensionSpec("dim1", "d0", ColumnType.STRING, new SubstringDimExtractionFn(0, 10)))).setAggregatorSpecs(aggregators(new LongSumAggregatorFactory("a0", "cnt"))).setContext(queryContext).build()), ImmutableList.of(new Object[] { NULL_STRING, 1L }, new Object[] { "1", 1L }));
}

Example 87 with QueryDataSource

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

the class CalciteJoinQueryTest method testUsingSubqueryWithExtractionFns.

@Test
@Parameters(source = QueryContextForJoinProvider.class)
public void testUsingSubqueryWithExtractionFns(Map<String, Object> queryContext) throws Exception {
    // Cannot vectorize JOIN operator.
    cannotVectorize();
    testQuery("SELECT dim2, COUNT(*) FROM druid.foo " + "WHERE substring(dim2, 1, 1) IN (SELECT substring(dim1, 1, 1) FROM druid.foo WHERE dim1 <> '')" + "group by dim2", queryContext, ImmutableList.of(GroupByQuery.builder().setDataSource(join(new TableDataSource(CalciteTests.DATASOURCE1), new QueryDataSource(GroupByQuery.builder().setDataSource(CalciteTests.DATASOURCE1).setInterval(querySegmentSpec(Filtration.eternity())).setGranularity(Granularities.ALL).setDimFilter(not(selector("dim1", "", null))).setDimensions(dimensions(new ExtractionDimensionSpec("dim1", "d0", new SubstringDimExtractionFn(0, 1)))).setContext(QUERY_CONTEXT_DEFAULT).build()), "j0.", equalsCondition(makeExpression("substring(\"dim2\", 0, 1)"), makeColumnExpression("j0.d0")), JoinType.INNER)).setInterval(querySegmentSpec(Filtration.eternity())).setGranularity(Granularities.ALL).setDimensions(dimensions(new DefaultDimensionSpec("dim2", "d0"))).setAggregatorSpecs(aggregators(new CountAggregatorFactory("a0"))).setContext(queryContext).build()), ImmutableList.of(new Object[] { "a", 2L }, new Object[] { "abc", 1L }));
}

Example 88 with QueryDataSource

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

the class CalciteJoinQueryTest method testInnerJoinLookupTableTableChained.

@Test
@Parameters(source = QueryContextForJoinProvider.class)
public void testInnerJoinLookupTableTableChained(Map<String, Object> queryContext) throws Exception {
    // Cannot vectorize JOIN operator.
    cannotVectorize();
    testQuery("SELECT l.k, l.v, SUM(f.m1), SUM(nf.m1)\n" + "FROM lookup.lookyloo l\n" + "INNER JOIN druid.foo f on f.dim1 = l.k\n" + "INNER JOIN druid.numfoo nf on nf.dim1 = f.dim1\n" + "GROUP BY 1, 2 ORDER BY 2", queryContext, ImmutableList.of(GroupByQuery.builder().setDataSource(join(join(new LookupDataSource("lookyloo"), new QueryDataSource(newScanQueryBuilder().dataSource(CalciteTests.DATASOURCE1).intervals(querySegmentSpec(Filtration.eternity())).columns("dim1", "m1").context(QUERY_CONTEXT_DEFAULT).build()), "j0.", equalsCondition(makeColumnExpression("k"), makeColumnExpression("j0.dim1")), JoinType.INNER), new QueryDataSource(newScanQueryBuilder().dataSource(CalciteTests.DATASOURCE3).intervals(querySegmentSpec(Filtration.eternity())).columns("dim1", "m1").context(QUERY_CONTEXT_DEFAULT).build()), "_j0.", equalsCondition(makeColumnExpression("j0.dim1"), makeColumnExpression("_j0.dim1")), JoinType.INNER)).setInterval(querySegmentSpec(Filtration.eternity())).setGranularity(Granularities.ALL).setDimensions(dimensions(new DefaultDimensionSpec("k", "d0"), new DefaultDimensionSpec("v", "d1"))).setAggregatorSpecs(aggregators(new DoubleSumAggregatorFactory("a0", "j0.m1"), new DoubleSumAggregatorFactory("a1", "_j0.m1"))).setLimitSpec(new DefaultLimitSpec(ImmutableList.of(new OrderByColumnSpec("d1", OrderByColumnSpec.Direction.ASCENDING)), null)).setContext(queryContext).build()), ImmutableList.of(new Object[] { "abc", "xabc", 6d, 6d }));
}

Example 89 with QueryDataSource

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

the class CalciteCorrelatedQueryTest method testCorrelatedSubqueryWithLeftFilter_leftDirectAccessDisabled.

@Test
@Parameters(source = QueryContextForJoinProvider.class)
public void testCorrelatedSubqueryWithLeftFilter_leftDirectAccessDisabled(Map<String, Object> queryContext) throws Exception {
    cannotVectorize();
    testQuery("select country, ANY_VALUE(\n" + "        select max(\"users\") from (\n" + "            select floor(__time to day), count(*) \"users\" from visits f where f.country = visits.country group by 1\n" + "        )\n" + "     ) as \"dailyVisits\"\n" + "from visits \n" + " where city = 'B' and __time between '2021-01-01 01:00:00' AND '2021-01-02 23:59:59'" + " group by 1", queryContext, ImmutableList.of(GroupByQuery.builder().setDataSource(join(new QueryDataSource(newScanQueryBuilder().dataSource(CalciteTests.USERVISITDATASOURCE).intervals(querySegmentSpec(Intervals.of("2021-01-01T01:00:00.000Z/2021-01-02T23:59:59.001Z"))).filters(selector("city", "B", null)).columns("__time", "city", "country").build()), 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 CountAggregatorFactory("a0")).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)).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", 4L }));
}

Example 90 with QueryDataSource

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

the class CalciteCorrelatedQueryTest method testCorrelatedSubqueryWithLeftFilter.

@Test
@Parameters(source = QueryContextForJoinProvider.class)
public void testCorrelatedSubqueryWithLeftFilter(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(*) \"users\" from visits f where f.country = visits.country group by 1\n" + "        )\n" + "     ) as \"dailyVisits\"\n" + "from visits \n" + " where city = 'B' and __time between '2021-01-01 01:00:00' AND '2021-01-02 23:59:59'" + " 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 CountAggregatorFactory("a0")).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.of("2021-01-01T01:00:00.000Z/2021-01-02T23:59:59.001Z"))).setDimensions(new DefaultDimensionSpec("country", "d0")).setAggregatorSpecs(new LongAnyAggregatorFactory("a0", "j0._a0")).setGranularity(new AllGranularity()).setContext(queryContext).build()), ImmutableList.of(new Object[] { "canada", 4L }));
}