Examples with InDimFilter org.apache.druid.query.filter.InDimFilter used on opensource projects

Example 11 with InDimFilter

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

the class GroupByQueryRunnerTest method testGroupByDoubleColumnDescending.

@Test
public void testGroupByDoubleColumnDescending() {
    if (config.getDefaultStrategy().equals(GroupByStrategySelector.STRATEGY_V1)) {
        expectedException.expect(UnsupportedOperationException.class);
        expectedException.expectMessage("GroupBy v1 only supports dimensions with an outputType of STRING.");
    }
    GroupByQuery query = makeQueryBuilder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setQuerySegmentSpec(QueryRunnerTestHelper.FIRST_TO_THIRD).setDimensions(new DefaultDimensionSpec("qualityDouble", "alias", ColumnType.DOUBLE)).setDimFilter(new InDimFilter("quality", Arrays.asList("entertainment", "technology"), null)).setAggregatorSpecs(QueryRunnerTestHelper.ROWS_COUNT, new LongSumAggregatorFactory("idx", "index")).addOrderByColumn(new OrderByColumnSpec("alias", OrderByColumnSpec.Direction.DESCENDING, StringComparators.NUMERIC)).setGranularity(QueryRunnerTestHelper.ALL_GRAN).build();
    Assert.assertNotEquals(Functions.<Sequence<ResultRow>>identity(), query.getLimitSpec().build(query));
    List<ResultRow> expectedResults = Arrays.asList(makeRow(query, "2011-04-01", "alias", 17000.0d, "rows", 2L, "idx", 175L), makeRow(query, "2011-04-01", "alias", 12000.0d, "rows", 2L, "idx", 324L));
    Iterable<ResultRow> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "double");
}

Example 12 with InDimFilter

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

the class GroupByQueryRunnerTest method testGroupByDecorationOnNumerics.

@Test
public void testGroupByDecorationOnNumerics() {
    // Cannot vectorize due to filtered dimension spec.
    cannotVectorize();
    if (config.getDefaultStrategy().equals(GroupByStrategySelector.STRATEGY_V1)) {
        expectedException.expect(UnsupportedOperationException.class);
        expectedException.expectMessage("GroupBy v1 only supports dimensions with an outputType of STRING.");
    }
    RegexFilteredDimensionSpec regexSpec = new RegexFilteredDimensionSpec(new DefaultDimensionSpec("qualityLong", "ql", ColumnType.LONG), "1700");
    ListFilteredDimensionSpec listFilteredSpec = new ListFilteredDimensionSpec(new DefaultDimensionSpec("qualityFloat", "qf", ColumnType.FLOAT), Sets.newHashSet("17000.0"), true);
    GroupByQuery query = makeQueryBuilder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setQuerySegmentSpec(QueryRunnerTestHelper.FIRST_TO_THIRD).setDimensions(regexSpec, listFilteredSpec).setDimFilter(new InDimFilter("quality", Arrays.asList("entertainment", "technology"), null)).setAggregatorSpecs(new CountAggregatorFactory("count")).setGranularity(QueryRunnerTestHelper.ALL_GRAN).build();
    List<ResultRow> expectedResults;
    if (NullHandling.replaceWithDefault()) {
        expectedResults = Arrays.asList(makeRow(query, "2011-04-01", "ql", 0L, "qf", 0.0, "count", 2L), makeRow(query, "2011-04-01", "ql", 1700L, "qf", 17000.0, "count", 2L));
    } else {
        expectedResults = Arrays.asList(makeRow(query, "2011-04-01", "ql", null, "qf", null, "count", 2L), makeRow(query, "2011-04-01", "ql", 1700L, "qf", 17000.0, "count", 2L));
    }
    Iterable<ResultRow> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "numeric");
}

Example 13 with InDimFilter

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

the class CachingClusteredClientTest method testSingleDimensionPruning.

@Test
public void testSingleDimensionPruning() {
    DimFilter filter = new AndDimFilter(new OrDimFilter(new SelectorDimFilter("dim1", "a", null), new BoundDimFilter("dim1", "from", "to", false, false, false, null, StringComparators.LEXICOGRAPHIC)), new AndDimFilter(new InDimFilter("dim2", Arrays.asList("a", "c", "e", "g"), null), new BoundDimFilter("dim2", "aaa", "hi", false, false, false, null, StringComparators.LEXICOGRAPHIC), new BoundDimFilter("dim2", "e", "zzz", true, true, false, null, StringComparators.LEXICOGRAPHIC)));
    final Druids.TimeseriesQueryBuilder builder = Druids.newTimeseriesQueryBuilder().dataSource(DATA_SOURCE).filters(filter).granularity(GRANULARITY).intervals(SEG_SPEC).context(CONTEXT).intervals("2011-01-05/2011-01-10").aggregators(RENAMED_AGGS).postAggregators(RENAMED_POST_AGGS);
    TimeseriesQuery query = builder.randomQueryId().build();
    final Interval interval1 = Intervals.of("2011-01-06/2011-01-07");
    final Interval interval2 = Intervals.of("2011-01-07/2011-01-08");
    final Interval interval3 = Intervals.of("2011-01-08/2011-01-09");
    QueryRunner runner = new FinalizeResultsQueryRunner(getDefaultQueryRunner(), new TimeseriesQueryQueryToolChest());
    final DruidServer lastServer = servers[random.nextInt(servers.length)];
    ServerSelector selector1 = makeMockSingleDimensionSelector(lastServer, "dim1", null, "b", 0);
    ServerSelector selector2 = makeMockSingleDimensionSelector(lastServer, "dim1", "e", "f", 1);
    ServerSelector selector3 = makeMockSingleDimensionSelector(lastServer, "dim1", "hi", "zzz", 2);
    ServerSelector selector4 = makeMockSingleDimensionSelector(lastServer, "dim2", "a", "e", 0);
    ServerSelector selector5 = makeMockSingleDimensionSelector(lastServer, "dim2", null, null, 1);
    ServerSelector selector6 = makeMockSingleDimensionSelector(lastServer, "other", "b", null, 0);
    timeline.add(interval1, "v", new NumberedPartitionChunk<>(0, 3, selector1));
    timeline.add(interval1, "v", new NumberedPartitionChunk<>(1, 3, selector2));
    timeline.add(interval1, "v", new NumberedPartitionChunk<>(2, 3, selector3));
    timeline.add(interval2, "v", new NumberedPartitionChunk<>(0, 2, selector4));
    timeline.add(interval2, "v", new NumberedPartitionChunk<>(1, 2, selector5));
    timeline.add(interval3, "v", new NumberedPartitionChunk<>(0, 1, selector6));
    final Capture<QueryPlus> capture = Capture.newInstance();
    final Capture<ResponseContext> contextCap = Capture.newInstance();
    QueryRunner mockRunner = EasyMock.createNiceMock(QueryRunner.class);
    EasyMock.expect(mockRunner.run(EasyMock.capture(capture), EasyMock.capture(contextCap))).andReturn(Sequences.empty()).anyTimes();
    EasyMock.expect(serverView.getQueryRunner(lastServer)).andReturn(mockRunner).anyTimes();
    EasyMock.replay(serverView);
    EasyMock.replay(mockRunner);
    List<SegmentDescriptor> descriptors = new ArrayList<>();
    descriptors.add(new SegmentDescriptor(interval1, "v", 0));
    descriptors.add(new SegmentDescriptor(interval1, "v", 2));
    descriptors.add(new SegmentDescriptor(interval2, "v", 1));
    descriptors.add(new SegmentDescriptor(interval3, "v", 0));
    MultipleSpecificSegmentSpec expected = new MultipleSpecificSegmentSpec(descriptors);
    runner.run(QueryPlus.wrap(query)).toList();
    Assert.assertEquals(expected, ((TimeseriesQuery) capture.getValue().getQuery()).getQuerySegmentSpec());
}

Example 14 with InDimFilter

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

the class CachingClusteredClientTest method testNoSegmentPruningForHashPartitionedSegments.

private void testNoSegmentPruningForHashPartitionedSegments(boolean enableSegmentPruning, @Nullable HashPartitionFunction partitionFunction, boolean useEmptyPartitionDimensions) {
    DimFilter filter = new AndDimFilter(new SelectorDimFilter("dim1", "a", null), new BoundDimFilter("dim2", "e", "zzz", true, true, false, null, StringComparators.LEXICOGRAPHIC), // Equivalent filter of dim3 below is InDimFilter("dim3", Arrays.asList("c"), null)
    new AndDimFilter(new InDimFilter("dim3", Arrays.asList("a", "c", "e", "g"), null), new BoundDimFilter("dim3", "aaa", "ddd", false, false, false, null, StringComparators.LEXICOGRAPHIC)));
    final Map<String, Object> context = new HashMap<>(CONTEXT);
    context.put(QueryContexts.SECONDARY_PARTITION_PRUNING_KEY, enableSegmentPruning);
    final Druids.TimeseriesQueryBuilder builder = Druids.newTimeseriesQueryBuilder().dataSource(DATA_SOURCE).filters(filter).granularity(GRANULARITY).intervals(SEG_SPEC).intervals("2011-01-05/2011-01-10").aggregators(RENAMED_AGGS).postAggregators(RENAMED_POST_AGGS).context(context).randomQueryId();
    TimeseriesQuery query = builder.build();
    QueryRunner runner = new FinalizeResultsQueryRunner(getDefaultQueryRunner(), new TimeseriesQueryQueryToolChest());
    final Interval interval1 = Intervals.of("2011-01-06/2011-01-07");
    final Interval interval2 = Intervals.of("2011-01-07/2011-01-08");
    final Interval interval3 = Intervals.of("2011-01-08/2011-01-09");
    final DruidServer lastServer = servers[random.nextInt(servers.length)];
    List<String> partitionDimensions = useEmptyPartitionDimensions ? ImmutableList.of() : ImmutableList.of("dim1");
    final int numPartitions1 = 6;
    for (int i = 0; i < numPartitions1; i++) {
        ServerSelector selector = makeMockHashBasedSelector(lastServer, partitionDimensions, partitionFunction, i, numPartitions1);
        timeline.add(interval1, "v", new NumberedPartitionChunk<>(i, numPartitions1, selector));
    }
    partitionDimensions = useEmptyPartitionDimensions ? ImmutableList.of() : ImmutableList.of("dim2");
    final int numPartitions2 = 3;
    for (int i = 0; i < numPartitions2; i++) {
        ServerSelector selector = makeMockHashBasedSelector(lastServer, partitionDimensions, partitionFunction, i, numPartitions2);
        timeline.add(interval2, "v", new NumberedPartitionChunk<>(i, numPartitions2, selector));
    }
    partitionDimensions = useEmptyPartitionDimensions ? ImmutableList.of() : ImmutableList.of("dim1", "dim3");
    final int numPartitions3 = 4;
    for (int i = 0; i < numPartitions3; i++) {
        ServerSelector selector = makeMockHashBasedSelector(lastServer, partitionDimensions, partitionFunction, i, numPartitions3);
        timeline.add(interval3, "v", new NumberedPartitionChunk<>(i, numPartitions3, selector));
    }
    final Capture<QueryPlus> capture = Capture.newInstance();
    final Capture<ResponseContext> contextCap = Capture.newInstance();
    QueryRunner mockRunner = EasyMock.createNiceMock(QueryRunner.class);
    EasyMock.expect(mockRunner.run(EasyMock.capture(capture), EasyMock.capture(contextCap))).andReturn(Sequences.empty()).anyTimes();
    EasyMock.expect(serverView.getQueryRunner(lastServer)).andReturn(mockRunner).anyTimes();
    EasyMock.replay(serverView);
    EasyMock.replay(mockRunner);
    // Expected to read all segments
    Set<SegmentDescriptor> expcetedDescriptors = new HashSet<>();
    IntStream.range(0, numPartitions1).forEach(i -> expcetedDescriptors.add(new SegmentDescriptor(interval1, "v", i)));
    IntStream.range(0, numPartitions2).forEach(i -> expcetedDescriptors.add(new SegmentDescriptor(interval2, "v", i)));
    IntStream.range(0, numPartitions3).forEach(i -> expcetedDescriptors.add(new SegmentDescriptor(interval3, "v", i)));
    runner.run(QueryPlus.wrap(query)).toList();
    QuerySegmentSpec querySegmentSpec = ((TimeseriesQuery) capture.getValue().getQuery()).getQuerySegmentSpec();
    Assert.assertSame(MultipleSpecificSegmentSpec.class, querySegmentSpec.getClass());
    final Set<SegmentDescriptor> actualDescriptors = new HashSet<>(((MultipleSpecificSegmentSpec) querySegmentSpec).getDescriptors());
    Assert.assertEquals(expcetedDescriptors, actualDescriptors);
}

Example 15 with InDimFilter

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

the class JoinFilterAnalyzerTest method test_filterPushDown_factToCountryRightWithFilterOnChannelAndJoinableUsingLookup.

@Test
public void test_filterPushDown_factToCountryRightWithFilterOnChannelAndJoinableUsingLookup() {
    List<JoinableClause> joinableClauses = ImmutableList.of(factToCountryNameUsingIsoCodeLookup(JoinType.RIGHT));
    Filter originalFilter = new AndFilter(ImmutableList.of(new SelectorFilter("channel", "#de.wikipedia"), new SelectorFilter(FACT_TO_COUNTRY_ON_ISO_CODE_PREFIX + "v", "Germany")));
    JoinFilterPreAnalysis joinFilterPreAnalysis = makeDefaultConfigPreAnalysis(originalFilter, joinableClauses, VirtualColumns.EMPTY);
    HashJoinSegmentStorageAdapter adapter = new HashJoinSegmentStorageAdapter(factSegment.asStorageAdapter(), joinableClauses, joinFilterPreAnalysis);
    JoinTestHelper.verifyCursors(adapter.makeCursors(originalFilter, Intervals.ETERNITY, VirtualColumns.EMPTY, Granularities.ALL, false, null), ImmutableList.of("page", "countryIsoCode", "countryNumber", FACT_TO_COUNTRY_ON_ISO_CODE_PREFIX + "k", FACT_TO_COUNTRY_ON_ISO_CODE_PREFIX + "v"), ImmutableList.of(new Object[] { "Diskussion:Sebastian Schulz", "DE", 3L, "DE", "Germany" }));
    JoinFilterSplit expectedFilterSplit = new JoinFilterSplit(new AndFilter(ImmutableList.of(new SelectorFilter("channel", "#de.wikipedia"), new InDimFilter("countryIsoCode", ImmutableSet.of("DE"), null, null).toFilter())), new SelectorFilter(FACT_TO_COUNTRY_ON_ISO_CODE_PREFIX + "v", "Germany"), ImmutableSet.of());
    JoinFilterSplit actualFilterSplit = JoinFilterAnalyzer.splitFilter(joinFilterPreAnalysis);
    Assert.assertEquals(expectedFilterSplit, actualFilterSplit);
}