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

Example 66 with SelectorDimFilter

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

the class GroupByQueryRunnerTest method testGroupByTimeExtraction.

@Test
public void testGroupByTimeExtraction() {
    // Cannot vectorize due to extraction dimension spec.
    cannotVectorize();
    GroupByQuery query = makeQueryBuilder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setQuerySegmentSpec(QueryRunnerTestHelper.FULL_ON_INTERVAL_SPEC).setDimensions(new DefaultDimensionSpec("market", "market"), new ExtractionDimensionSpec(ColumnHolder.TIME_COLUMN_NAME, "dayOfWeek", new TimeFormatExtractionFn("EEEE", null, null, null, false))).setAggregatorSpecs(QueryRunnerTestHelper.ROWS_COUNT, QueryRunnerTestHelper.INDEX_DOUBLE_SUM).setPostAggregatorSpecs(Collections.singletonList(QueryRunnerTestHelper.ADD_ROWS_INDEX_CONSTANT)).setGranularity(QueryRunnerTestHelper.ALL_GRAN).setDimFilter(new OrDimFilter(Arrays.asList(new SelectorDimFilter("market", "spot", null), new SelectorDimFilter("market", "upfront", null)))).build();
    List<ResultRow> expectedResults = Arrays.asList(makeRow(query, "1970-01-01", "dayOfWeek", "Friday", "market", "spot", "index", 13219.574157714844, "rows", 117L, "addRowsIndexConstant", 13337.574157714844), makeRow(query, "1970-01-01", "dayOfWeek", "Monday", "market", "spot", "index", 13557.738830566406, "rows", 117L, "addRowsIndexConstant", 13675.738830566406), makeRow(query, "1970-01-01", "dayOfWeek", "Saturday", "market", "spot", "index", 13493.751281738281, "rows", 117L, "addRowsIndexConstant", 13611.751281738281), makeRow(query, "1970-01-01", "dayOfWeek", "Sunday", "market", "spot", "index", 13585.541015625, "rows", 117L, "addRowsIndexConstant", 13703.541015625), makeRow(query, "1970-01-01", "dayOfWeek", "Thursday", "market", "spot", "index", 14279.127197265625, "rows", 126L, "addRowsIndexConstant", 14406.127197265625), makeRow(query, "1970-01-01", "dayOfWeek", "Tuesday", "market", "spot", "index", 13199.471435546875, "rows", 117L, "addRowsIndexConstant", 13317.471435546875), makeRow(query, "1970-01-01", "dayOfWeek", "Wednesday", "market", "spot", "index", 14271.368591308594, "rows", 126L, "addRowsIndexConstant", 14398.368591308594), makeRow(query, "1970-01-01", "dayOfWeek", "Friday", "market", "upfront", "index", 27297.8623046875, "rows", 26L, "addRowsIndexConstant", 27324.8623046875), makeRow(query, "1970-01-01", "dayOfWeek", "Monday", "market", "upfront", "index", 27619.58447265625, "rows", 26L, "addRowsIndexConstant", 27646.58447265625), makeRow(query, "1970-01-01", "dayOfWeek", "Saturday", "market", "upfront", "index", 27820.83154296875, "rows", 26L, "addRowsIndexConstant", 27847.83154296875), makeRow(query, "1970-01-01", "dayOfWeek", "Sunday", "market", "upfront", "index", 24791.223876953125, "rows", 26L, "addRowsIndexConstant", 24818.223876953125), makeRow(query, "1970-01-01", "dayOfWeek", "Thursday", "market", "upfront", "index", 28562.748901367188, "rows", 28L, "addRowsIndexConstant", 28591.748901367188), makeRow(query, "1970-01-01", "dayOfWeek", "Tuesday", "market", "upfront", "index", 26968.280639648438, "rows", 26L, "addRowsIndexConstant", 26995.280639648438), makeRow(query, "1970-01-01", "dayOfWeek", "Wednesday", "market", "upfront", "index", 28985.5751953125, "rows", 28L, "addRowsIndexConstant", 29014.5751953125));
    Iterable<ResultRow> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "time-extraction");
}

Example 67 with SelectorDimFilter

use of org.apache.druid.query.filter.SelectorDimFilter 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 68 with SelectorDimFilter

use of org.apache.druid.query.filter.SelectorDimFilter 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 69 with SelectorDimFilter

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

the class ClientCompactionTaskTransformSpecTest method testSerde.

@Test
public void testSerde() throws IOException {
    NullHandling.initializeForTests();
    final ClientCompactionTaskTransformSpec expected = new ClientCompactionTaskTransformSpec(new SelectorDimFilter("dim1", "foo", null));
    final ObjectMapper mapper = new DefaultObjectMapper();
    final byte[] json = mapper.writeValueAsBytes(expected);
    final ClientCompactionTaskTransformSpec fromJson = (ClientCompactionTaskTransformSpec) mapper.readValue(json, ClientCompactionTaskTransformSpec.class);
    Assert.assertEquals(expected, fromJson);
}

Example 70 with SelectorDimFilter

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

the class AggregatorFactoryTest method testResultArraySignature.

@Test
public void testResultArraySignature() {
    final TimeseriesQuery query = Druids.newTimeseriesQueryBuilder().dataSource("dummy").intervals("2000/3000").granularity(Granularities.HOUR).aggregators(new CountAggregatorFactory("count"), new JavaScriptAggregatorFactory("js", ImmutableList.of("col"), "function(a,b) { return a + b; }", "function() { return 0; }", "function(a,b) { return a + b }", new JavaScriptConfig(true)), // long aggs
    new LongSumAggregatorFactory("longSum", "long-col"), new LongMinAggregatorFactory("longMin", "long-col"), new LongMaxAggregatorFactory("longMax", "long-col"), new LongFirstAggregatorFactory("longFirst", "long-col", null), new LongLastAggregatorFactory("longLast", "long-col", null), new LongAnyAggregatorFactory("longAny", "long-col"), // double aggs
    new DoubleSumAggregatorFactory("doubleSum", "double-col"), new DoubleMinAggregatorFactory("doubleMin", "double-col"), new DoubleMaxAggregatorFactory("doubleMax", "double-col"), new DoubleFirstAggregatorFactory("doubleFirst", "double-col", null), new DoubleLastAggregatorFactory("doubleLast", "double-col", null), new DoubleAnyAggregatorFactory("doubleAny", "double-col"), new DoubleMeanAggregatorFactory("doubleMean", "double-col"), // float aggs
    new FloatSumAggregatorFactory("floatSum", "float-col"), new FloatMinAggregatorFactory("floatMin", "float-col"), new FloatMaxAggregatorFactory("floatMax", "float-col"), new FloatFirstAggregatorFactory("floatFirst", "float-col", null), new FloatLastAggregatorFactory("floatLast", "float-col", null), new FloatAnyAggregatorFactory("floatAny", "float-col"), // string aggregators
    new StringFirstAggregatorFactory("stringFirst", "col", null, 1024), new StringLastAggregatorFactory("stringLast", "col", null, 1024), new StringAnyAggregatorFactory("stringAny", "col", 1024), // sketch aggs
    new CardinalityAggregatorFactory("cardinality", ImmutableList.of(DefaultDimensionSpec.of("some-col")), false), new HyperUniquesAggregatorFactory("hyperUnique", "hyperunique"), new HistogramAggregatorFactory("histogram", "histogram", ImmutableList.of(0.25f, 0.5f, 0.75f)), // delegate aggs
    new FilteredAggregatorFactory(new HyperUniquesAggregatorFactory("filtered", "hyperunique"), new SelectorDimFilter("col", "hello", null)), new SuppressedAggregatorFactory(new HyperUniquesAggregatorFactory("suppressed", "hyperunique"))).postAggregators(new FinalizingFieldAccessPostAggregator("count-finalize", "count"), new FinalizingFieldAccessPostAggregator("js-finalize", "js"), // long aggs
    new FinalizingFieldAccessPostAggregator("longSum-finalize", "longSum"), new FinalizingFieldAccessPostAggregator("longMin-finalize", "longMin"), new FinalizingFieldAccessPostAggregator("longMax-finalize", "longMax"), new FinalizingFieldAccessPostAggregator("longFirst-finalize", "longFirst"), new FinalizingFieldAccessPostAggregator("longLast-finalize", "longLast"), new FinalizingFieldAccessPostAggregator("longAny-finalize", "longAny"), // double
    new FinalizingFieldAccessPostAggregator("doubleSum-finalize", "doubleSum"), new FinalizingFieldAccessPostAggregator("doubleMin-finalize", "doubleMin"), new FinalizingFieldAccessPostAggregator("doubleMax-finalize", "doubleMax"), new FinalizingFieldAccessPostAggregator("doubleFirst-finalize", "doubleFirst"), new FinalizingFieldAccessPostAggregator("doubleLast-finalize", "doubleLast"), new FinalizingFieldAccessPostAggregator("doubleAny-finalize", "doubleAny"), new FinalizingFieldAccessPostAggregator("doubleMean-finalize", "doubleMean"), // finalized floats
    new FinalizingFieldAccessPostAggregator("floatSum-finalize", "floatSum"), new FinalizingFieldAccessPostAggregator("floatMin-finalize", "floatMin"), new FinalizingFieldAccessPostAggregator("floatMax-finalize", "floatMax"), new FinalizingFieldAccessPostAggregator("floatFirst-finalize", "floatFirst"), new FinalizingFieldAccessPostAggregator("floatLast-finalize", "floatLast"), new FinalizingFieldAccessPostAggregator("floatAny-finalize", "floatAny"), // finalized strings
    new FinalizingFieldAccessPostAggregator("stringFirst-finalize", "stringFirst"), new FinalizingFieldAccessPostAggregator("stringLast-finalize", "stringLast"), new FinalizingFieldAccessPostAggregator("stringAny-finalize", "stringAny"), // finalized sketch
    new FinalizingFieldAccessPostAggregator("cardinality-finalize", "cardinality"), new FinalizingFieldAccessPostAggregator("hyperUnique-finalize", "hyperUnique"), new FinalizingFieldAccessPostAggregator("histogram-finalize", "histogram"), // finalized delegate
    new FinalizingFieldAccessPostAggregator("filtered-finalize", "filtered"), new FinalizingFieldAccessPostAggregator("suppressed-finalize", "suppressed")).build();
    Assert.assertEquals(RowSignature.builder().addTimeColumn().add("count", ColumnType.LONG).add("js", ColumnType.FLOAT).add("longSum", ColumnType.LONG).add("longMin", ColumnType.LONG).add("longMax", ColumnType.LONG).add("longFirst", ColumnType.LONG).add("longLast", ColumnType.LONG).add("longAny", ColumnType.LONG).add("doubleSum", ColumnType.DOUBLE).add("doubleMin", ColumnType.DOUBLE).add("doubleMax", ColumnType.DOUBLE).add("doubleFirst", ColumnType.DOUBLE).add("doubleLast", ColumnType.DOUBLE).add("doubleAny", ColumnType.DOUBLE).add("doubleMean", null).add("floatSum", ColumnType.FLOAT).add("floatMin", ColumnType.FLOAT).add("floatMax", ColumnType.FLOAT).add("floatFirst", ColumnType.FLOAT).add("floatLast", ColumnType.FLOAT).add("floatAny", ColumnType.FLOAT).add("stringFirst", null).add("stringLast", null).add("stringAny", ColumnType.STRING).add("cardinality", null).add("hyperUnique", null).add("histogram", null).add("filtered", null).add("suppressed", null).add("count-finalize", ColumnType.LONG).add("js-finalize", ColumnType.FLOAT).add("longSum-finalize", ColumnType.LONG).add("longMin-finalize", ColumnType.LONG).add("longMax-finalize", ColumnType.LONG).add("longFirst-finalize", ColumnType.LONG).add("longLast-finalize", ColumnType.LONG).add("longAny-finalize", ColumnType.LONG).add("doubleSum-finalize", ColumnType.DOUBLE).add("doubleMin-finalize", ColumnType.DOUBLE).add("doubleMax-finalize", ColumnType.DOUBLE).add("doubleFirst-finalize", ColumnType.DOUBLE).add("doubleLast-finalize", ColumnType.DOUBLE).add("doubleAny-finalize", ColumnType.DOUBLE).add("doubleMean-finalize", ColumnType.DOUBLE).add("floatSum-finalize", ColumnType.FLOAT).add("floatMin-finalize", ColumnType.FLOAT).add("floatMax-finalize", ColumnType.FLOAT).add("floatFirst-finalize", ColumnType.FLOAT).add("floatLast-finalize", ColumnType.FLOAT).add("floatAny-finalize", ColumnType.FLOAT).add("stringFirst-finalize", ColumnType.STRING).add("stringLast-finalize", ColumnType.STRING).add("stringAny-finalize", ColumnType.STRING).add("cardinality-finalize", ColumnType.DOUBLE).add("hyperUnique-finalize", ColumnType.DOUBLE).add("histogram-finalize", HistogramAggregatorFactory.TYPE_VISUAL).add("filtered-finalize", ColumnType.DOUBLE).add("suppressed-finalize", ColumnType.DOUBLE).build(), new TimeseriesQueryQueryToolChest().resultArraySignature(query));
}