Search in sources :

Example 1 with AggregatorFactory

use of io.druid.query.aggregation.AggregatorFactory in project hive by apache.

the class DruidGroupByQueryRecordReader method initExtractors.

private void initExtractors() throws IOException {
    extractors = new Extract[query.getAggregatorSpecs().size() + query.getPostAggregatorSpecs().size()];
    int counter = 0;
    for (int i = 0; i < query.getAggregatorSpecs().size(); i++, counter++) {
        AggregatorFactory af = query.getAggregatorSpecs().get(i);
        switch(af.getTypeName().toUpperCase()) {
            case DruidSerDeUtils.FLOAT_TYPE:
                extractors[counter] = Extract.FLOAT;
                break;
            case DruidSerDeUtils.LONG_TYPE:
                extractors[counter] = Extract.LONG;
                break;
            default:
                throw new IOException("Type not supported");
        }
    }
    for (int i = 0; i < query.getPostAggregatorSpecs().size(); i++, counter++) {
        extractors[counter] = Extract.FLOAT;
    }
}
Also used : IOException(java.io.IOException) AggregatorFactory(io.druid.query.aggregation.AggregatorFactory)

Example 2 with AggregatorFactory

use of io.druid.query.aggregation.AggregatorFactory in project hive by apache.

the class DruidSerDe method inferSchema.

/* GroupBy query */
private void inferSchema(GroupByQuery query, List<String> columnNames, List<PrimitiveTypeInfo> columnTypes) {
    // Timestamp column
    columnNames.add(DruidTable.DEFAULT_TIMESTAMP_COLUMN);
    columnTypes.add(TypeInfoFactory.timestampTypeInfo);
    // Dimension columns
    for (DimensionSpec ds : query.getDimensions()) {
        columnNames.add(ds.getOutputName());
        columnTypes.add(TypeInfoFactory.stringTypeInfo);
    }
    // Aggregator columns
    for (AggregatorFactory af : query.getAggregatorSpecs()) {
        columnNames.add(af.getName());
        columnTypes.add(DruidSerDeUtils.convertDruidToHiveType(af.getTypeName()));
    }
    // different types for post-aggregation functions
    for (PostAggregator pa : query.getPostAggregatorSpecs()) {
        columnNames.add(pa.getName());
        columnTypes.add(TypeInfoFactory.floatTypeInfo);
    }
}
Also used : DimensionSpec(io.druid.query.dimension.DimensionSpec) PostAggregator(io.druid.query.aggregation.PostAggregator) AggregatorFactory(io.druid.query.aggregation.AggregatorFactory)

Example 3 with AggregatorFactory

use of io.druid.query.aggregation.AggregatorFactory in project druid by druid-io.

the class DruidParquetReadSupport method getPartialReadSchema.

private MessageType getPartialReadSchema(InitContext context) {
    MessageType fullSchema = context.getFileSchema();
    String name = fullSchema.getName();
    HadoopDruidIndexerConfig config = HadoopDruidIndexerConfig.fromConfiguration(context.getConfiguration());
    String tsField = config.getParser().getParseSpec().getTimestampSpec().getTimestampColumn();
    List<DimensionSchema> dimensionSchema = config.getParser().getParseSpec().getDimensionsSpec().getDimensions();
    Set<String> dimensions = Sets.newHashSet();
    for (DimensionSchema dim : dimensionSchema) {
        dimensions.add(dim.getName());
    }
    Set<String> metricsFields = Sets.newHashSet();
    for (AggregatorFactory agg : config.getSchema().getDataSchema().getAggregators()) {
        metricsFields.addAll(agg.requiredFields());
    }
    List<Type> partialFields = Lists.newArrayList();
    for (Type type : fullSchema.getFields()) {
        if (tsField.equals(type.getName()) || metricsFields.contains(type.getName()) || dimensions.size() > 0 && dimensions.contains(type.getName()) || dimensions.size() == 0) {
            partialFields.add(type);
        }
    }
    return new MessageType(name, partialFields);
}
Also used : MessageType(org.apache.parquet.schema.MessageType) Type(org.apache.parquet.schema.Type) HadoopDruidIndexerConfig(io.druid.indexer.HadoopDruidIndexerConfig) AggregatorFactory(io.druid.query.aggregation.AggregatorFactory) DimensionSchema(io.druid.data.input.impl.DimensionSchema) MessageType(org.apache.parquet.schema.MessageType)

Example 4 with AggregatorFactory

use of io.druid.query.aggregation.AggregatorFactory in project druid by druid-io.

the class IncrementalIndexStorageAdapterTest method testSingleValueTopN.

@Test
public void testSingleValueTopN() throws IOException {
    IncrementalIndex index = indexCreator.createIndex();
    DateTime t = DateTime.now();
    index.add(new MapBasedInputRow(t.minus(1).getMillis(), Lists.newArrayList("sally"), ImmutableMap.<String, Object>of("sally", "bo")));
    TopNQueryEngine engine = new TopNQueryEngine(new StupidPool<ByteBuffer>("TopNQueryEngine-bufferPool", new Supplier<ByteBuffer>() {

        @Override
        public ByteBuffer get() {
            return ByteBuffer.allocate(50000);
        }
    }));
    final Iterable<Result<TopNResultValue>> results = Sequences.toList(engine.query(new TopNQueryBuilder().dataSource("test").granularity(Granularities.ALL).intervals(Lists.newArrayList(new Interval(0, new DateTime().getMillis()))).dimension("sally").metric("cnt").threshold(10).aggregators(Lists.<AggregatorFactory>newArrayList(new LongSumAggregatorFactory("cnt", "cnt"))).build(), new IncrementalIndexStorageAdapter(index)), Lists.<Result<TopNResultValue>>newLinkedList());
    Assert.assertEquals(1, Iterables.size(results));
    Assert.assertEquals(1, results.iterator().next().getValue().getValue().size());
}
Also used : TopNQueryBuilder(io.druid.query.topn.TopNQueryBuilder) TopNResultValue(io.druid.query.topn.TopNResultValue) LongSumAggregatorFactory(io.druid.query.aggregation.LongSumAggregatorFactory) AggregatorFactory(io.druid.query.aggregation.AggregatorFactory) CountAggregatorFactory(io.druid.query.aggregation.CountAggregatorFactory) JavaScriptAggregatorFactory(io.druid.query.aggregation.JavaScriptAggregatorFactory) LongSumAggregatorFactory(io.druid.query.aggregation.LongSumAggregatorFactory) ByteBuffer(java.nio.ByteBuffer) DateTime(org.joda.time.DateTime) TopNQueryEngine(io.druid.query.topn.TopNQueryEngine) Result(io.druid.query.Result) Supplier(com.google.common.base.Supplier) MapBasedInputRow(io.druid.data.input.MapBasedInputRow) Interval(org.joda.time.Interval) Test(org.junit.Test)

Example 5 with AggregatorFactory

use of io.druid.query.aggregation.AggregatorFactory in project druid by druid-io.

the class SpatialFilterTest method testSpatialQueryWithOtherSpatialDim.

@Test
public void testSpatialQueryWithOtherSpatialDim() {
    TimeseriesQuery query = Druids.newTimeseriesQueryBuilder().dataSource("test").granularity(Granularities.ALL).intervals(Arrays.asList(new Interval("2013-01-01/2013-01-07"))).filters(new SpatialDimFilter("spatialIsRad", new RadiusBound(new float[] { 0.0f, 0.0f }, 5))).aggregators(Arrays.<AggregatorFactory>asList(new CountAggregatorFactory("rows"), new LongSumAggregatorFactory("val", "val"))).build();
    List<Result<TimeseriesResultValue>> expectedResults = Arrays.asList(new Result<TimeseriesResultValue>(new DateTime("2013-01-01T00:00:00.000Z"), new TimeseriesResultValue(ImmutableMap.<String, Object>builder().put("rows", 1L).put("val", 13L).build())));
    try {
        TimeseriesQueryRunnerFactory factory = new TimeseriesQueryRunnerFactory(new TimeseriesQueryQueryToolChest(QueryRunnerTestHelper.NoopIntervalChunkingQueryRunnerDecorator()), new TimeseriesQueryEngine(), QueryRunnerTestHelper.NOOP_QUERYWATCHER);
        QueryRunner runner = new FinalizeResultsQueryRunner(factory.createRunner(segment), factory.getToolchest());
        TestHelper.assertExpectedResults(expectedResults, runner.run(query, Maps.newHashMap()));
    } catch (Exception e) {
        throw Throwables.propagate(e);
    }
}
Also used : TimeseriesResultValue(io.druid.query.timeseries.TimeseriesResultValue) TimeseriesQuery(io.druid.query.timeseries.TimeseriesQuery) LongSumAggregatorFactory(io.druid.query.aggregation.LongSumAggregatorFactory) TimeseriesQueryQueryToolChest(io.druid.query.timeseries.TimeseriesQueryQueryToolChest) AggregatorFactory(io.druid.query.aggregation.AggregatorFactory) CountAggregatorFactory(io.druid.query.aggregation.CountAggregatorFactory) LongSumAggregatorFactory(io.druid.query.aggregation.LongSumAggregatorFactory) DateTime(org.joda.time.DateTime) FinalizeResultsQueryRunner(io.druid.query.FinalizeResultsQueryRunner) QueryRunner(io.druid.query.QueryRunner) IOException(java.io.IOException) Result(io.druid.query.Result) TimeseriesQueryEngine(io.druid.query.timeseries.TimeseriesQueryEngine) SpatialDimFilter(io.druid.query.filter.SpatialDimFilter) TimeseriesQueryRunnerFactory(io.druid.query.timeseries.TimeseriesQueryRunnerFactory) RadiusBound(io.druid.collections.spatial.search.RadiusBound) CountAggregatorFactory(io.druid.query.aggregation.CountAggregatorFactory) FinalizeResultsQueryRunner(io.druid.query.FinalizeResultsQueryRunner) Interval(org.joda.time.Interval) Test(org.junit.Test)

Aggregations

AggregatorFactory (io.druid.query.aggregation.AggregatorFactory)148 Test (org.junit.Test)86 CountAggregatorFactory (io.druid.query.aggregation.CountAggregatorFactory)82 LongSumAggregatorFactory (io.druid.query.aggregation.LongSumAggregatorFactory)64 Interval (org.joda.time.Interval)45 DoubleSumAggregatorFactory (io.druid.query.aggregation.DoubleSumAggregatorFactory)38 DateTime (org.joda.time.DateTime)37 FilteredAggregatorFactory (io.druid.query.aggregation.FilteredAggregatorFactory)32 Result (io.druid.query.Result)31 DoubleMaxAggregatorFactory (io.druid.query.aggregation.DoubleMaxAggregatorFactory)27 HyperUniquesAggregatorFactory (io.druid.query.aggregation.hyperloglog.HyperUniquesAggregatorFactory)25 Row (io.druid.data.input.Row)24 PostAggregator (io.druid.query.aggregation.PostAggregator)24 DefaultDimensionSpec (io.druid.query.dimension.DefaultDimensionSpec)22 CardinalityAggregatorFactory (io.druid.query.aggregation.cardinality.CardinalityAggregatorFactory)19 LongMaxAggregatorFactory (io.druid.query.aggregation.LongMaxAggregatorFactory)18 LongFirstAggregatorFactory (io.druid.query.aggregation.first.LongFirstAggregatorFactory)18 LongLastAggregatorFactory (io.druid.query.aggregation.last.LongLastAggregatorFactory)18 DimensionSpec (io.druid.query.dimension.DimensionSpec)18 TimeseriesQuery (io.druid.query.timeseries.TimeseriesQuery)17