Search in sources :

Example 26 with TopNQueryBuilder

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

the class TopNBenchmark method setupQueries.

private void setupQueries() {
    // queries for the basic schema
    Map<String, TopNQueryBuilder> basicQueries = new LinkedHashMap<>();
    GeneratorSchemaInfo basicSchema = GeneratorBasicSchemas.SCHEMA_MAP.get("basic");
    {
        // basic.A
        QuerySegmentSpec intervalSpec = new MultipleIntervalSegmentSpec(Collections.singletonList(basicSchema.getDataInterval()));
        List<AggregatorFactory> queryAggs = new ArrayList<>();
        queryAggs.add(new LongSumAggregatorFactory("sumLongSequential", "sumLongSequential"));
        queryAggs.add(new LongMaxAggregatorFactory("maxLongUniform", "maxLongUniform"));
        queryAggs.add(new DoubleSumAggregatorFactory("sumFloatNormal", "sumFloatNormal"));
        queryAggs.add(new DoubleMinAggregatorFactory("minFloatZipf", "minFloatZipf"));
        queryAggs.add(new HyperUniquesAggregatorFactory("hyperUniquesMet", "hyper"));
        TopNQueryBuilder queryBuilderA = new TopNQueryBuilder().dataSource("blah").granularity(Granularities.ALL).dimension("dimSequential").metric("sumFloatNormal").intervals(intervalSpec).aggregators(queryAggs);
        basicQueries.put("A", queryBuilderA);
    }
    {
        // basic.numericSort
        QuerySegmentSpec intervalSpec = new MultipleIntervalSegmentSpec(Collections.singletonList(basicSchema.getDataInterval()));
        List<AggregatorFactory> queryAggs = new ArrayList<>();
        queryAggs.add(new LongSumAggregatorFactory("sumLongSequential", "sumLongSequential"));
        TopNQueryBuilder queryBuilderA = new TopNQueryBuilder().dataSource("blah").granularity(Granularities.ALL).dimension("dimUniform").metric(new DimensionTopNMetricSpec(null, StringComparators.NUMERIC)).intervals(intervalSpec).aggregators(queryAggs);
        basicQueries.put("numericSort", queryBuilderA);
    }
    {
        // basic.alphanumericSort
        QuerySegmentSpec intervalSpec = new MultipleIntervalSegmentSpec(Collections.singletonList(basicSchema.getDataInterval()));
        List<AggregatorFactory> queryAggs = new ArrayList<>();
        queryAggs.add(new LongSumAggregatorFactory("sumLongSequential", "sumLongSequential"));
        TopNQueryBuilder queryBuilderA = new TopNQueryBuilder().dataSource("blah").granularity(Granularities.ALL).dimension("dimUniform").metric(new DimensionTopNMetricSpec(null, StringComparators.ALPHANUMERIC)).intervals(intervalSpec).aggregators(queryAggs);
        basicQueries.put("alphanumericSort", queryBuilderA);
    }
    SCHEMA_QUERY_MAP.put("basic", basicQueries);
}
Also used : TopNQueryBuilder(org.apache.druid.query.topn.TopNQueryBuilder) DoubleSumAggregatorFactory(org.apache.druid.query.aggregation.DoubleSumAggregatorFactory) GeneratorSchemaInfo(org.apache.druid.segment.generator.GeneratorSchemaInfo) LongSumAggregatorFactory(org.apache.druid.query.aggregation.LongSumAggregatorFactory) MultipleIntervalSegmentSpec(org.apache.druid.query.spec.MultipleIntervalSegmentSpec) DoubleMinAggregatorFactory(org.apache.druid.query.aggregation.DoubleMinAggregatorFactory) LinkedHashMap(java.util.LinkedHashMap) DimensionTopNMetricSpec(org.apache.druid.query.topn.DimensionTopNMetricSpec) HyperUniquesAggregatorFactory(org.apache.druid.query.aggregation.hyperloglog.HyperUniquesAggregatorFactory) QuerySegmentSpec(org.apache.druid.query.spec.QuerySegmentSpec) List(java.util.List) ArrayList(java.util.ArrayList) LongMaxAggregatorFactory(org.apache.druid.query.aggregation.LongMaxAggregatorFactory)

Example 27 with TopNQueryBuilder

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

the class TopNTypeInterfaceBenchmark method setup.

@Setup
public void setup() throws IOException {
    log.info("SETUP CALLED AT " + System.currentTimeMillis());
    ComplexMetrics.registerSerde("hyperUnique", new HyperUniquesSerde());
    setupQueries();
    schemaInfo = GeneratorBasicSchemas.SCHEMA_MAP.get("basic");
    queryBuilder = SCHEMA_QUERY_MAP.get("basic").get("string");
    queryBuilder.threshold(threshold);
    stringQuery = queryBuilder.build();
    TopNQueryBuilder longBuilder = SCHEMA_QUERY_MAP.get("basic").get("long");
    longBuilder.threshold(threshold);
    longQuery = longBuilder.build();
    TopNQueryBuilder floatBuilder = SCHEMA_QUERY_MAP.get("basic").get("float");
    floatBuilder.threshold(threshold);
    floatQuery = floatBuilder.build();
    incIndexes = new ArrayList<>();
    for (int i = 0; i < numSegments; i++) {
        log.info("Generating rows for segment " + i);
        DataGenerator gen = new DataGenerator(schemaInfo.getColumnSchemas(), RNG_SEED + i, schemaInfo.getDataInterval(), rowsPerSegment);
        IncrementalIndex incIndex = makeIncIndex();
        for (int j = 0; j < rowsPerSegment; j++) {
            InputRow row = gen.nextRow();
            if (j % 10000 == 0) {
                log.info(j + " rows generated.");
            }
            incIndex.add(row);
        }
        incIndexes.add(incIndex);
    }
    File tmpFile = FileUtils.createTempDir();
    log.info("Using temp dir: " + tmpFile.getAbsolutePath());
    tmpFile.deleteOnExit();
    qIndexes = new ArrayList<>();
    for (int i = 0; i < numSegments; i++) {
        File indexFile = INDEX_MERGER_V9.persist(incIndexes.get(i), tmpFile, new IndexSpec(), null);
        QueryableIndex qIndex = INDEX_IO.loadIndex(indexFile);
        qIndexes.add(qIndex);
    }
    factory = new TopNQueryRunnerFactory(new StupidPool<>("TopNBenchmark-compute-bufferPool", new OffheapBufferGenerator("compute", 250000000), 0, Integer.MAX_VALUE), new TopNQueryQueryToolChest(new TopNQueryConfig()), QueryBenchmarkUtil.NOOP_QUERYWATCHER);
}
Also used : TopNQueryBuilder(org.apache.druid.query.topn.TopNQueryBuilder) IndexSpec(org.apache.druid.segment.IndexSpec) IncrementalIndex(org.apache.druid.segment.incremental.IncrementalIndex) OnheapIncrementalIndex(org.apache.druid.segment.incremental.OnheapIncrementalIndex) HyperUniquesSerde(org.apache.druid.query.aggregation.hyperloglog.HyperUniquesSerde) OffheapBufferGenerator(org.apache.druid.offheap.OffheapBufferGenerator) TopNQueryConfig(org.apache.druid.query.topn.TopNQueryConfig) QueryableIndex(org.apache.druid.segment.QueryableIndex) DataGenerator(org.apache.druid.segment.generator.DataGenerator) InputRow(org.apache.druid.data.input.InputRow) TopNQueryRunnerFactory(org.apache.druid.query.topn.TopNQueryRunnerFactory) StupidPool(org.apache.druid.collections.StupidPool) TopNQueryQueryToolChest(org.apache.druid.query.topn.TopNQueryQueryToolChest) File(java.io.File) Setup(org.openjdk.jmh.annotations.Setup)

Example 28 with TopNQueryBuilder

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

the class DistinctCountTopNQueryTest method testTopNWithDistinctCountAgg.

@Test
public void testTopNWithDistinctCountAgg() throws Exception {
    TopNQueryEngine engine = new TopNQueryEngine(pool);
    IncrementalIndex index = new OnheapIncrementalIndex.Builder().setIndexSchema(new IncrementalIndexSchema.Builder().withQueryGranularity(Granularities.SECOND).withMetrics(new CountAggregatorFactory("cnt")).build()).setMaxRowCount(1000).build();
    String visitor_id = "visitor_id";
    String client_type = "client_type";
    DateTime time = DateTimes.of("2016-03-04T00:00:00.000Z");
    long timestamp = time.getMillis();
    index.add(new MapBasedInputRow(timestamp, Lists.newArrayList(visitor_id, client_type), ImmutableMap.of(visitor_id, "0", client_type, "iphone")));
    index.add(new MapBasedInputRow(timestamp, Lists.newArrayList(visitor_id, client_type), ImmutableMap.of(visitor_id, "1", client_type, "iphone")));
    index.add(new MapBasedInputRow(timestamp, Lists.newArrayList(visitor_id, client_type), ImmutableMap.of(visitor_id, "2", client_type, "android")));
    TopNQuery query = new TopNQueryBuilder().dataSource(QueryRunnerTestHelper.DATA_SOURCE).granularity(QueryRunnerTestHelper.ALL_GRAN).intervals(QueryRunnerTestHelper.FULL_ON_INTERVAL_SPEC).dimension(client_type).metric("UV").threshold(10).aggregators(QueryRunnerTestHelper.ROWS_COUNT, new DistinctCountAggregatorFactory("UV", visitor_id, null)).build();
    final Iterable<Result<TopNResultValue>> results = engine.query(query, new IncrementalIndexStorageAdapter(index), null).toList();
    List<Result<TopNResultValue>> expectedResults = Collections.singletonList(new Result<>(time, new TopNResultValue(Arrays.<Map<String, Object>>asList(ImmutableMap.of(client_type, "iphone", "UV", 2L, "rows", 2L), ImmutableMap.of(client_type, "android", "UV", 1L, "rows", 1L)))));
    TestHelper.assertExpectedResults(expectedResults, results);
}
Also used : TopNQueryBuilder(org.apache.druid.query.topn.TopNQueryBuilder) TopNResultValue(org.apache.druid.query.topn.TopNResultValue) IncrementalIndex(org.apache.druid.segment.incremental.IncrementalIndex) OnheapIncrementalIndex(org.apache.druid.segment.incremental.OnheapIncrementalIndex) TopNQueryBuilder(org.apache.druid.query.topn.TopNQueryBuilder) OnheapIncrementalIndex(org.apache.druid.segment.incremental.OnheapIncrementalIndex) DateTime(org.joda.time.DateTime) TopNQueryEngine(org.apache.druid.query.topn.TopNQueryEngine) Result(org.apache.druid.query.Result) CountAggregatorFactory(org.apache.druid.query.aggregation.CountAggregatorFactory) IncrementalIndexStorageAdapter(org.apache.druid.segment.incremental.IncrementalIndexStorageAdapter) TopNQuery(org.apache.druid.query.topn.TopNQuery) MapBasedInputRow(org.apache.druid.data.input.MapBasedInputRow) InitializedNullHandlingTest(org.apache.druid.testing.InitializedNullHandlingTest) Test(org.junit.Test)

Example 29 with TopNQueryBuilder

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

the class MultiValuedDimensionTest method testTopNExpression.

@Test
public void testTopNExpression() {
    TopNQuery query = new TopNQueryBuilder().dataSource("xx").granularity(Granularities.ALL).dimension(new DefaultDimensionSpec("texpr", "texpr")).virtualColumns(new ExpressionVirtualColumn("texpr", "map(x -> concat(x, 'foo'), tags)", ColumnType.STRING, TestExprMacroTable.INSTANCE)).metric("count").intervals(QueryRunnerTestHelper.FULL_ON_INTERVAL_SPEC).aggregators(new CountAggregatorFactory("count")).threshold(15).build();
    try (CloseableStupidPool<ByteBuffer> pool = TestQueryRunners.createDefaultNonBlockingPool()) {
        QueryRunnerFactory factory = new TopNQueryRunnerFactory(pool, new TopNQueryQueryToolChest(new TopNQueryConfig()), QueryRunnerTestHelper.NOOP_QUERYWATCHER);
        QueryRunner<Result<TopNResultValue>> runner = QueryRunnerTestHelper.makeQueryRunner(factory, new QueryableIndexSegment(queryableIndex, SegmentId.dummy("sid1")), null);
        Sequence<Result<TopNResultValue>> result = runner.run(QueryPlus.wrap(query));
        List<Map<String, Object>> expected = ImmutableList.<Map<String, Object>>builder().add(ImmutableMap.of("texpr", "t3foo", "count", 2L)).add(ImmutableMap.of("texpr", "t5foo", "count", 2L)).add(new HashMap<String, Object>() {

            {
                put("texpr", NullHandling.sqlCompatible() ? "foo" : null);
                put("count", 1L);
            }
        }).add(ImmutableMap.of("texpr", "t1foo", "count", 1L)).add(ImmutableMap.of("texpr", "t2foo", "count", 1L)).add(ImmutableMap.of("texpr", "t4foo", "count", 1L)).add(ImmutableMap.of("texpr", "t6foo", "count", 1L)).add(ImmutableMap.of("texpr", "t7foo", "count", 1L)).build();
        List<Result<TopNResultValue>> expectedResults = Collections.singletonList(new Result<TopNResultValue>(DateTimes.of("2011-01-12T00:00:00.000Z"), new TopNResultValue(expected)));
        TestHelper.assertExpectedObjects(expectedResults, result.toList(), "filteredDim");
    }
}
Also used : TopNQueryBuilder(org.apache.druid.query.topn.TopNQueryBuilder) QueryableIndexSegment(org.apache.druid.segment.QueryableIndexSegment) TopNResultValue(org.apache.druid.query.topn.TopNResultValue) ByteBuffer(java.nio.ByteBuffer) DefaultDimensionSpec(org.apache.druid.query.dimension.DefaultDimensionSpec) ExpressionVirtualColumn(org.apache.druid.segment.virtual.ExpressionVirtualColumn) CountAggregatorFactory(org.apache.druid.query.aggregation.CountAggregatorFactory) TopNQueryRunnerFactory(org.apache.druid.query.topn.TopNQueryRunnerFactory) TopNQueryConfig(org.apache.druid.query.topn.TopNQueryConfig) TopNQuery(org.apache.druid.query.topn.TopNQuery) TopNQueryRunnerFactory(org.apache.druid.query.topn.TopNQueryRunnerFactory) TopNQueryQueryToolChest(org.apache.druid.query.topn.TopNQueryQueryToolChest) Map(java.util.Map) ImmutableMap(com.google.common.collect.ImmutableMap) HashMap(java.util.HashMap) GroupByQueryRunnerTest(org.apache.druid.query.groupby.GroupByQueryRunnerTest) InitializedNullHandlingTest(org.apache.druid.testing.InitializedNullHandlingTest) Test(org.junit.Test)

Example 30 with TopNQueryBuilder

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

the class MultiValuedDimensionTest method testTopNExpressionAutoTransform.

@Test
public void testTopNExpressionAutoTransform() {
    TopNQuery query = new TopNQueryBuilder().dataSource("xx").granularity(Granularities.ALL).dimension(new DefaultDimensionSpec("texpr", "texpr")).virtualColumns(new ExpressionVirtualColumn("texpr", "concat(tags, 'foo')", ColumnType.STRING, TestExprMacroTable.INSTANCE)).metric("count").intervals(QueryRunnerTestHelper.FULL_ON_INTERVAL_SPEC).aggregators(new CountAggregatorFactory("count")).threshold(15).build();
    try (CloseableStupidPool<ByteBuffer> pool = TestQueryRunners.createDefaultNonBlockingPool()) {
        QueryRunnerFactory factory = new TopNQueryRunnerFactory(pool, new TopNQueryQueryToolChest(new TopNQueryConfig()), QueryRunnerTestHelper.NOOP_QUERYWATCHER);
        QueryRunner<Result<TopNResultValue>> runner = QueryRunnerTestHelper.makeQueryRunner(factory, new QueryableIndexSegment(queryableIndex, SegmentId.dummy("sid1")), null);
        Sequence<Result<TopNResultValue>> result = runner.run(QueryPlus.wrap(query));
        List<Map<String, Object>> expected = ImmutableList.<Map<String, Object>>builder().add(ImmutableMap.of("texpr", "t3foo", "count", 2L)).add(ImmutableMap.of("texpr", "t5foo", "count", 2L)).add(ImmutableMap.of("texpr", "foo", "count", 1L)).add(ImmutableMap.of("texpr", "t1foo", "count", 1L)).add(ImmutableMap.of("texpr", "t2foo", "count", 1L)).add(ImmutableMap.of("texpr", "t4foo", "count", 1L)).add(ImmutableMap.of("texpr", "t6foo", "count", 1L)).add(ImmutableMap.of("texpr", "t7foo", "count", 1L)).build();
        List<Result<TopNResultValue>> expectedResults = Collections.singletonList(new Result<TopNResultValue>(DateTimes.of("2011-01-12T00:00:00.000Z"), new TopNResultValue(expected)));
        TestHelper.assertExpectedObjects(expectedResults, result.toList(), "filteredDim");
    }
}
Also used : TopNQueryBuilder(org.apache.druid.query.topn.TopNQueryBuilder) QueryableIndexSegment(org.apache.druid.segment.QueryableIndexSegment) TopNResultValue(org.apache.druid.query.topn.TopNResultValue) ByteBuffer(java.nio.ByteBuffer) DefaultDimensionSpec(org.apache.druid.query.dimension.DefaultDimensionSpec) ExpressionVirtualColumn(org.apache.druid.segment.virtual.ExpressionVirtualColumn) CountAggregatorFactory(org.apache.druid.query.aggregation.CountAggregatorFactory) TopNQueryRunnerFactory(org.apache.druid.query.topn.TopNQueryRunnerFactory) TopNQueryConfig(org.apache.druid.query.topn.TopNQueryConfig) TopNQuery(org.apache.druid.query.topn.TopNQuery) TopNQueryRunnerFactory(org.apache.druid.query.topn.TopNQueryRunnerFactory) TopNQueryQueryToolChest(org.apache.druid.query.topn.TopNQueryQueryToolChest) Map(java.util.Map) ImmutableMap(com.google.common.collect.ImmutableMap) HashMap(java.util.HashMap) GroupByQueryRunnerTest(org.apache.druid.query.groupby.GroupByQueryRunnerTest) InitializedNullHandlingTest(org.apache.druid.testing.InitializedNullHandlingTest) Test(org.junit.Test)

Aggregations

TopNQueryBuilder (org.apache.druid.query.topn.TopNQueryBuilder)55 Test (org.junit.Test)44 TopNQuery (org.apache.druid.query.topn.TopNQuery)25 DefaultDimensionSpec (org.apache.druid.query.dimension.DefaultDimensionSpec)22 LongSumAggregatorFactory (org.apache.druid.query.aggregation.LongSumAggregatorFactory)16 InvertedTopNMetricSpec (org.apache.druid.query.topn.InvertedTopNMetricSpec)14 CountAggregatorFactory (org.apache.druid.query.aggregation.CountAggregatorFactory)13 TopNQueryQueryToolChest (org.apache.druid.query.topn.TopNQueryQueryToolChest)13 Result (org.apache.druid.query.Result)12 TopNQueryConfig (org.apache.druid.query.topn.TopNQueryConfig)12 TopNResultValue (org.apache.druid.query.topn.TopNResultValue)11 DoubleMinAggregatorFactory (org.apache.druid.query.aggregation.DoubleMinAggregatorFactory)9 DimensionTopNMetricSpec (org.apache.druid.query.topn.DimensionTopNMetricSpec)9 NumericTopNMetricSpec (org.apache.druid.query.topn.NumericTopNMetricSpec)9 DoubleMaxAggregatorFactory (org.apache.druid.query.aggregation.DoubleMaxAggregatorFactory)7 InitializedNullHandlingTest (org.apache.druid.testing.InitializedNullHandlingTest)7 ByteBuffer (java.nio.ByteBuffer)5 QueryRunner (org.apache.druid.query.QueryRunner)5 MultipleIntervalSegmentSpec (org.apache.druid.query.spec.MultipleIntervalSegmentSpec)5 HashMap (java.util.HashMap)4