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

Example 1 with UnionDataSource

use of org.apache.druid.query.UnionDataSource 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 }));
}

Example 2 with UnionDataSource

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

the class DataSourceAnalysisTest method testJoinTableUnionToLookup.

@Test
public void testJoinTableUnionToLookup() {
    final UnionDataSource unionDataSource = new UnionDataSource(ImmutableList.of(TABLE_FOO, TABLE_BAR));
    final JoinDataSource joinDataSource = join(unionDataSource, LOOKUP_LOOKYLOO, "1.", JoinType.INNER);
    final DataSourceAnalysis analysis = DataSourceAnalysis.forDataSource(joinDataSource);
    Assert.assertTrue(analysis.isConcreteBased());
    Assert.assertTrue(analysis.isConcreteTableBased());
    Assert.assertFalse(analysis.isGlobal());
    Assert.assertFalse(analysis.isQuery());
    Assert.assertEquals(joinDataSource, analysis.getDataSource());
    Assert.assertEquals(Optional.empty(), analysis.getBaseTableDataSource());
    Assert.assertEquals(Optional.empty(), analysis.getJoinBaseTableFilter());
    Assert.assertEquals(Optional.of(unionDataSource), analysis.getBaseUnionDataSource());
    Assert.assertEquals(unionDataSource, analysis.getBaseDataSource());
    Assert.assertEquals(Optional.empty(), analysis.getBaseQuery());
    Assert.assertEquals(Optional.empty(), analysis.getBaseQuerySegmentSpec());
    Assert.assertEquals(ImmutableList.of(new PreJoinableClause("1.", LOOKUP_LOOKYLOO, JoinType.INNER, joinClause("1."))), analysis.getPreJoinableClauses());
    Assert.assertTrue(analysis.isJoin());
}

Example 3 with UnionDataSource

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

the class DataSourceAnalysisTest method testUnion.

@Test
public void testUnion() {
    final UnionDataSource unionDataSource = new UnionDataSource(ImmutableList.of(TABLE_FOO, TABLE_BAR));
    final DataSourceAnalysis analysis = DataSourceAnalysis.forDataSource(unionDataSource);
    Assert.assertTrue(analysis.isConcreteBased());
    Assert.assertTrue(analysis.isConcreteTableBased());
    Assert.assertFalse(analysis.isGlobal());
    Assert.assertFalse(analysis.isQuery());
    Assert.assertEquals(unionDataSource, analysis.getDataSource());
    Assert.assertEquals(unionDataSource, analysis.getBaseDataSource());
    Assert.assertEquals(Optional.empty(), analysis.getBaseTableDataSource());
    Assert.assertEquals(Optional.of(unionDataSource), analysis.getBaseUnionDataSource());
    Assert.assertEquals(Optional.empty(), analysis.getBaseQuery());
    Assert.assertEquals(Optional.empty(), analysis.getBaseQuerySegmentSpec());
    Assert.assertEquals(Collections.emptyList(), analysis.getPreJoinableClauses());
    Assert.assertFalse(analysis.isJoin());
}

Example 4 with UnionDataSource

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

the class DruidUnionDataSourceRel method toDruidQuery.

@Override
public DruidQuery toDruidQuery(final boolean finalizeAggregations) {
    final List<TableDataSource> dataSources = new ArrayList<>();
    RowSignature signature = null;
    for (final RelNode relNode : unionRel.getInputs()) {
        final DruidRel<?> druidRel = (DruidRel<?>) relNode;
        if (!DruidRels.isScanOrMapping(druidRel, false)) {
            getPlannerContext().setPlanningError("SQL requires union between inputs that are not simple table scans " + "and involve a filter or aliasing");
            throw new CannotBuildQueryException(druidRel);
        }
        final DruidQuery query = druidRel.toDruidQuery(false);
        final DataSource dataSource = query.getDataSource();
        if (!(dataSource instanceof TableDataSource)) {
            getPlannerContext().setPlanningError("SQL requires union with input of '%s' type that is not supported." + " Union operation is only supported between regular tables. ", dataSource.getClass().getSimpleName());
            throw new CannotBuildQueryException(druidRel);
        }
        if (signature == null) {
            signature = query.getOutputRowSignature();
        }
        if (signature.getColumnNames().equals(query.getOutputRowSignature().getColumnNames())) {
            dataSources.add((TableDataSource) dataSource);
        } else {
            getPlannerContext().setPlanningError("There is a mismatch between the output row signature of input tables and the row signature of union output.");
            throw new CannotBuildQueryException(druidRel);
        }
    }
    if (signature == null) {
        // No inputs.
        throw new CannotBuildQueryException(unionRel);
    }
    // creation time.
    if (!signature.getColumnNames().equals(unionColumnNames)) {
        throw new CannotBuildQueryException(unionRel);
    }
    return partialQuery.build(new UnionDataSource(dataSources), signature, getPlannerContext(), getCluster().getRexBuilder(), finalizeAggregations);
}

Example 5 with UnionDataSource

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

the class ClientQuerySegmentWalkerTest method testJoinOnGroupByOnUnionOfTables.

@Test
public void testJoinOnGroupByOnUnionOfTables() {
    final UnionDataSource unionDataSource = new UnionDataSource(ImmutableList.of(new TableDataSource(FOO), new TableDataSource(BAR)));
    final GroupByQuery subquery = GroupByQuery.builder().setDataSource(unionDataSource).setGranularity(Granularities.ALL).setInterval(Collections.singletonList(INTERVAL)).setDimensions(DefaultDimensionSpec.of("s")).setDimFilter(new SelectorDimFilter("s", "y", null)).build();
    final GroupByQuery query = (GroupByQuery) GroupByQuery.builder().setDataSource(JoinDataSource.create(unionDataSource, new QueryDataSource(subquery), "j.", "\"j.s\" == \"s\"", JoinType.INNER, null, ExprMacroTable.nil())).setGranularity(Granularities.ALL).setInterval(Intervals.ONLY_ETERNITY).setDimensions(DefaultDimensionSpec.of("s"), DefaultDimensionSpec.of("j.s")).setAggregatorSpecs(new CountAggregatorFactory("cnt")).build().withId(DUMMY_QUERY_ID);
    testQuery(query, ImmutableList.of(ExpectedQuery.cluster(subquery.withDataSource(subquery.getDataSource().getChildren().get(0)).withId(DUMMY_QUERY_ID).withSubQueryId("2.1.foo.1")), ExpectedQuery.cluster(subquery.withDataSource(subquery.getDataSource().getChildren().get(1)).withId(DUMMY_QUERY_ID).withSubQueryId("2.1.bar.2")), ExpectedQuery.cluster(query.withDataSource(query.getDataSource().withChildren(ImmutableList.of(unionDataSource.getChildren().get(0), InlineDataSource.fromIterable(ImmutableList.of(new Object[] { "y" }), RowSignature.builder().add("s", ColumnType.STRING).build())))).withSubQueryId("foo.1")), ExpectedQuery.cluster(query.withDataSource(query.getDataSource().withChildren(ImmutableList.of(unionDataSource.getChildren().get(1), InlineDataSource.fromIterable(ImmutableList.of(new Object[] { "y" }), RowSignature.builder().add("s", ColumnType.STRING).build())))).withSubQueryId("bar.2"))), ImmutableList.of(new Object[] { "y", "y", 1L }));
    // note: this should really be 1, but in the interim queries that are composed of multiple queries count each
    // invocation of either the cluster or local walker in ClientQuerySegmentWalker
    Assert.assertEquals(4, scheduler.getTotalRun().get());
    Assert.assertEquals(4, scheduler.getTotalPrioritizedAndLaned().get());
    Assert.assertEquals(4, scheduler.getTotalAcquired().get());
    Assert.assertEquals(4, scheduler.getTotalReleased().get());
}