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

Example 11 with DimFilter

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

the class GroupByQueryRunnerTest method testGroupByWithExtractionDimFilter.

// Extraction Filters testing
@Test
public void testGroupByWithExtractionDimFilter() {
    Map<String, String> extractionMap = new HashMap<>();
    extractionMap.put("automotive", "automotiveAndBusinessAndNewsAndMezzanine");
    extractionMap.put("business", "automotiveAndBusinessAndNewsAndMezzanine");
    extractionMap.put("mezzanine", "automotiveAndBusinessAndNewsAndMezzanine");
    extractionMap.put("news", "automotiveAndBusinessAndNewsAndMezzanine");
    MapLookupExtractor mapLookupExtractor = new MapLookupExtractor(extractionMap, false);
    LookupExtractionFn lookupExtractionFn = new LookupExtractionFn(mapLookupExtractor, false, null, true, false);
    List<DimFilter> dimFilters = Lists.newArrayList(new ExtractionDimFilter("quality", "automotiveAndBusinessAndNewsAndMezzanine", lookupExtractionFn, null), new SelectorDimFilter("quality", "entertainment", null), new SelectorDimFilter("quality", "health", null), new SelectorDimFilter("quality", "premium", null), new SelectorDimFilter("quality", "technology", null), new SelectorDimFilter("quality", "travel", null));
    GroupByQuery query = makeQueryBuilder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setQuerySegmentSpec(QueryRunnerTestHelper.FIRST_TO_THIRD).setDimensions(new DefaultDimensionSpec("quality", "alias")).setAggregatorSpecs(QueryRunnerTestHelper.ROWS_COUNT, new LongSumAggregatorFactory("idx", "index")).setGranularity(QueryRunnerTestHelper.DAY_GRAN).setDimFilter(new OrDimFilter(dimFilters)).build();
    List<ResultRow> expectedResults = Arrays.asList(makeRow(query, "2011-04-01", "alias", "automotive", "rows", 1L, "idx", 135L), makeRow(query, "2011-04-01", "alias", "business", "rows", 1L, "idx", 118L), makeRow(query, "2011-04-01", "alias", "entertainment", "rows", 1L, "idx", 158L), makeRow(query, "2011-04-01", "alias", "health", "rows", 1L, "idx", 120L), makeRow(query, "2011-04-01", "alias", "mezzanine", "rows", 3L, "idx", 2870L), makeRow(query, "2011-04-01", "alias", "news", "rows", 1L, "idx", 121L), makeRow(query, "2011-04-01", "alias", "premium", "rows", 3L, "idx", 2900L), makeRow(query, "2011-04-01", "alias", "technology", "rows", 1L, "idx", 78L), makeRow(query, "2011-04-01", "alias", "travel", "rows", 1L, "idx", 119L), makeRow(query, "2011-04-02", "alias", "automotive", "rows", 1L, "idx", 147L), makeRow(query, "2011-04-02", "alias", "business", "rows", 1L, "idx", 112L), makeRow(query, "2011-04-02", "alias", "entertainment", "rows", 1L, "idx", 166L), makeRow(query, "2011-04-02", "alias", "health", "rows", 1L, "idx", 113L), makeRow(query, "2011-04-02", "alias", "mezzanine", "rows", 3L, "idx", 2447L), makeRow(query, "2011-04-02", "alias", "news", "rows", 1L, "idx", 114L), makeRow(query, "2011-04-02", "alias", "premium", "rows", 3L, "idx", 2505L), makeRow(query, "2011-04-02", "alias", "technology", "rows", 1L, "idx", 97L), makeRow(query, "2011-04-02", "alias", "travel", "rows", 1L, "idx", 126L));
    Iterable<ResultRow> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "dim-extraction");
}

Example 12 with DimFilter

use of org.apache.druid.query.filter.DimFilter 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 13 with DimFilter

use of org.apache.druid.query.filter.DimFilter 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 14 with DimFilter

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

the class DruidJoinQueryRel method getDimFilter.

@Nullable
private static DimFilter getDimFilter(final PlannerContext plannerContext, final RowSignature rowSignature, @Nullable final Filter filter) {
    if (filter == null) {
        return null;
    }
    final RexNode condition = filter.getCondition();
    final DimFilter dimFilter = Expressions.toFilter(plannerContext, rowSignature, null, condition);
    if (dimFilter == null) {
        throw new CannotBuildQueryException(filter, condition);
    } else {
        return dimFilter;
    }
}

Example 15 with DimFilter

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

the class DruidQuery method getDimFilter.

@Nonnull
private static DimFilter getDimFilter(final PlannerContext plannerContext, final RowSignature rowSignature, @Nullable final VirtualColumnRegistry virtualColumnRegistry, final Filter filter) {
    final RexNode condition = filter.getCondition();
    final DimFilter dimFilter = Expressions.toFilter(plannerContext, rowSignature, virtualColumnRegistry, condition);
    if (dimFilter == null) {
        throw new CannotBuildQueryException(filter, condition);
    } else {
        return dimFilter;
    }
}