Search in sources :

Example 1 with LongSumAggregatorFactory

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

the class TestDruidRecordWriter method testWrite.

// This test need this patch https://github.com/druid-io/druid/pull/3483
@Ignore
@Test
public void testWrite() throws IOException, SegmentLoadingException {
    final String dataSourceName = "testDataSource";
    final File segmentOutputDir = temporaryFolder.newFolder();
    final File workingDir = temporaryFolder.newFolder();
    Configuration config = new Configuration();
    final InputRowParser inputRowParser = new MapInputRowParser(new TimeAndDimsParseSpec(new TimestampSpec(DruidTable.DEFAULT_TIMESTAMP_COLUMN, "auto", null), new DimensionsSpec(ImmutableList.<DimensionSchema>of(new StringDimensionSchema("host")), null, null)));
    final Map<String, Object> parserMap = objectMapper.convertValue(inputRowParser, Map.class);
    DataSchema dataSchema = new DataSchema(dataSourceName, parserMap, new AggregatorFactory[] { new LongSumAggregatorFactory("visited_sum", "visited_sum"), new HyperUniquesAggregatorFactory("unique_hosts", "unique_hosts") }, new UniformGranularitySpec(Granularity.DAY, QueryGranularities.NONE, ImmutableList.of(INTERVAL_FULL)), objectMapper);
    RealtimeTuningConfig tuningConfig = RealtimeTuningConfig.makeDefaultTuningConfig(temporaryFolder.newFolder());
    LocalFileSystem localFileSystem = FileSystem.getLocal(config);
    DataSegmentPusher dataSegmentPusher = new LocalDataSegmentPusher(new LocalDataSegmentPusherConfig() {

        @Override
        public File getStorageDirectory() {
            return segmentOutputDir;
        }
    }, objectMapper);
    Path segmentDescriptroPath = new Path(workingDir.getAbsolutePath(), DruidStorageHandler.SEGMENTS_DESCRIPTOR_DIR_NAME);
    druidRecordWriter = new DruidRecordWriter(dataSchema, tuningConfig, dataSegmentPusher, 20, segmentDescriptroPath, localFileSystem);
    List<DruidWritable> druidWritables = Lists.transform(expectedRows, new Function<ImmutableMap<String, Object>, DruidWritable>() {

        @Nullable
        @Override
        public DruidWritable apply(@Nullable ImmutableMap<String, Object> input) {
            return new DruidWritable(ImmutableMap.<String, Object>builder().putAll(input).put(Constants.DRUID_TIMESTAMP_GRANULARITY_COL_NAME, Granularity.DAY.truncate(new DateTime((long) input.get(DruidTable.DEFAULT_TIMESTAMP_COLUMN))).getMillis()).build());
        }
    });
    for (DruidWritable druidWritable : druidWritables) {
        druidRecordWriter.write(druidWritable);
    }
    druidRecordWriter.close(false);
    List<DataSegment> dataSegmentList = DruidStorageHandlerUtils.getPublishedSegments(segmentDescriptroPath, config);
    Assert.assertEquals(1, dataSegmentList.size());
    File tmpUnzippedSegmentDir = temporaryFolder.newFolder();
    new LocalDataSegmentPuller().getSegmentFiles(dataSegmentList.get(0), tmpUnzippedSegmentDir);
    final QueryableIndex queryableIndex = DruidStorageHandlerUtils.INDEX_IO.loadIndex(tmpUnzippedSegmentDir);
    QueryableIndexStorageAdapter adapter = new QueryableIndexStorageAdapter(queryableIndex);
    Firehose firehose = new IngestSegmentFirehose(ImmutableList.of(new WindowedStorageAdapter(adapter, adapter.getInterval())), ImmutableList.of("host"), ImmutableList.of("visited_sum", "unique_hosts"), null, QueryGranularities.NONE);
    List<InputRow> rows = Lists.newArrayList();
    while (firehose.hasMore()) {
        rows.add(firehose.nextRow());
    }
    verifyRows(expectedRows, rows);
}
Also used : IngestSegmentFirehose(io.druid.segment.realtime.firehose.IngestSegmentFirehose) LocalDataSegmentPusher(io.druid.segment.loading.LocalDataSegmentPusher) DataSegmentPusher(io.druid.segment.loading.DataSegmentPusher) Configuration(org.apache.hadoop.conf.Configuration) MapInputRowParser(io.druid.data.input.impl.MapInputRowParser) LongSumAggregatorFactory(io.druid.query.aggregation.LongSumAggregatorFactory) DataSegment(io.druid.timeline.DataSegment) DateTime(org.joda.time.DateTime) TimeAndDimsParseSpec(io.druid.data.input.impl.TimeAndDimsParseSpec) UniformGranularitySpec(io.druid.segment.indexing.granularity.UniformGranularitySpec) LocalDataSegmentPuller(io.druid.segment.loading.LocalDataSegmentPuller) TimestampSpec(io.druid.data.input.impl.TimestampSpec) WindowedStorageAdapter(io.druid.segment.realtime.firehose.WindowedStorageAdapter) Path(org.apache.hadoop.fs.Path) Firehose(io.druid.data.input.Firehose) IngestSegmentFirehose(io.druid.segment.realtime.firehose.IngestSegmentFirehose) LocalDataSegmentPusherConfig(io.druid.segment.loading.LocalDataSegmentPusherConfig) QueryableIndexStorageAdapter(io.druid.segment.QueryableIndexStorageAdapter) RealtimeTuningConfig(io.druid.segment.indexing.RealtimeTuningConfig) LocalDataSegmentPusher(io.druid.segment.loading.LocalDataSegmentPusher) ImmutableMap(com.google.common.collect.ImmutableMap) StringDimensionSchema(io.druid.data.input.impl.StringDimensionSchema) DataSchema(io.druid.segment.indexing.DataSchema) DruidWritable(org.apache.hadoop.hive.druid.serde.DruidWritable) LocalFileSystem(org.apache.hadoop.fs.LocalFileSystem) QueryableIndex(io.druid.segment.QueryableIndex) HyperUniquesAggregatorFactory(io.druid.query.aggregation.hyperloglog.HyperUniquesAggregatorFactory) InputRow(io.druid.data.input.InputRow) DimensionsSpec(io.druid.data.input.impl.DimensionsSpec) MapInputRowParser(io.druid.data.input.impl.MapInputRowParser) InputRowParser(io.druid.data.input.impl.InputRowParser) File(java.io.File) DruidRecordWriter(org.apache.hadoop.hive.druid.io.DruidRecordWriter) Nullable(javax.annotation.Nullable) Ignore(org.junit.Ignore) Test(org.junit.Test)

Example 2 with LongSumAggregatorFactory

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

the class IncrementalIndexStorageAdapterTest method testFilterByNull.

@Test
public void testFilterByNull() throws Exception {
    IncrementalIndex index = indexCreator.createIndex();
    index.add(new MapBasedInputRow(new DateTime().minus(1).getMillis(), Lists.newArrayList("billy"), ImmutableMap.<String, Object>of("billy", "hi")));
    index.add(new MapBasedInputRow(new DateTime().minus(1).getMillis(), Lists.newArrayList("sally"), ImmutableMap.<String, Object>of("sally", "bo")));
    GroupByQueryEngine engine = makeGroupByQueryEngine();
    final Sequence<Row> rows = engine.process(GroupByQuery.builder().setDataSource("test").setGranularity(Granularities.ALL).setInterval(new Interval(0, new DateTime().getMillis())).addDimension("billy").addDimension("sally").addAggregator(new LongSumAggregatorFactory("cnt", "cnt")).setDimFilter(DimFilters.dimEquals("sally", (String) null)).build(), new IncrementalIndexStorageAdapter(index));
    final ArrayList<Row> results = Sequences.toList(rows, Lists.<Row>newArrayList());
    Assert.assertEquals(1, results.size());
    MapBasedRow row = (MapBasedRow) results.get(0);
    Assert.assertEquals(ImmutableMap.of("billy", "hi", "cnt", 1L), row.getEvent());
}
Also used : LongSumAggregatorFactory(io.druid.query.aggregation.LongSumAggregatorFactory) DateTime(org.joda.time.DateTime) MapBasedRow(io.druid.data.input.MapBasedRow) MapBasedInputRow(io.druid.data.input.MapBasedInputRow) MapBasedInputRow(io.druid.data.input.MapBasedInputRow) Row(io.druid.data.input.Row) MapBasedRow(io.druid.data.input.MapBasedRow) GroupByQueryEngine(io.druid.query.groupby.GroupByQueryEngine) Interval(org.joda.time.Interval) Test(org.junit.Test)

Example 3 with LongSumAggregatorFactory

use of io.druid.query.aggregation.LongSumAggregatorFactory 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 4 with LongSumAggregatorFactory

use of io.druid.query.aggregation.LongSumAggregatorFactory 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)

Example 5 with LongSumAggregatorFactory

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

the class RealtimeManagerTest method testQueryWithSegmentSpec.

@Test(timeout = 10_000L)
public void testQueryWithSegmentSpec() throws IOException, InterruptedException {
    List<Row> expectedResults = Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "automotive", "rows", 1L, "idx", 135L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "business", "rows", 1L, "idx", 118L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "entertainment", "rows", 1L, "idx", 158L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "health", "rows", 1L, "idx", 120L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "mezzanine", "rows", 3L, "idx", 2870L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "news", "rows", 1L, "idx", 121L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "premium", "rows", 3L, "idx", 2900L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "technology", "rows", 1L, "idx", 78L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "travel", "rows", 1L, "idx", 119L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "automotive", "rows", 1L, "idx", 147L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "business", "rows", 1L, "idx", 112L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "entertainment", "rows", 1L, "idx", 166L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "health", "rows", 1L, "idx", 113L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "mezzanine", "rows", 3L, "idx", 2447L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "news", "rows", 1L, "idx", 114L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "premium", "rows", 3L, "idx", 2505L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "technology", "rows", 1L, "idx", 97L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "travel", "rows", 1L, "idx", 126L));
    chiefStartedLatch.await();
    for (QueryRunner runner : QueryRunnerTestHelper.makeQueryRunners((GroupByQueryRunnerFactory) factory)) {
        GroupByQuery query = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.dataSource).setQuerySegmentSpec(QueryRunnerTestHelper.firstToThird).setDimensions(Lists.<DimensionSpec>newArrayList(new DefaultDimensionSpec("quality", "alias"))).setAggregatorSpecs(Arrays.asList(QueryRunnerTestHelper.rowsCount, new LongSumAggregatorFactory("idx", "index"))).setGranularity(QueryRunnerTestHelper.dayGran).build();
        plumber.setRunners(ImmutableMap.of(query.getIntervals().get(0), runner));
        plumber2.setRunners(ImmutableMap.of(query.getIntervals().get(0), runner));
        Iterable<Row> results = GroupByQueryRunnerTestHelper.runQuery(factory, realtimeManager3.getQueryRunnerForSegments(query, ImmutableList.<SegmentDescriptor>of(new SegmentDescriptor(new Interval("2011-04-01T00:00:00.000Z/2011-04-03T00:00:00.000Z"), "ver", 0))), query);
        TestHelper.assertExpectedObjects(expectedResults, results, "");
        results = GroupByQueryRunnerTestHelper.runQuery(factory, realtimeManager3.getQueryRunnerForSegments(query, ImmutableList.<SegmentDescriptor>of(new SegmentDescriptor(new Interval("2011-04-01T00:00:00.000Z/2011-04-03T00:00:00.000Z"), "ver", 1))), query);
        TestHelper.assertExpectedObjects(expectedResults, results, "");
    }
}
Also used : DimensionSpec(io.druid.query.dimension.DimensionSpec) DefaultDimensionSpec(io.druid.query.dimension.DefaultDimensionSpec) GroupByQuery(io.druid.query.groupby.GroupByQuery) SegmentDescriptor(io.druid.query.SegmentDescriptor) LongSumAggregatorFactory(io.druid.query.aggregation.LongSumAggregatorFactory) InputRow(io.druid.data.input.InputRow) Row(io.druid.data.input.Row) SpecificSegmentQueryRunner(io.druid.query.spec.SpecificSegmentQueryRunner) QueryRunner(io.druid.query.QueryRunner) DefaultDimensionSpec(io.druid.query.dimension.DefaultDimensionSpec) Interval(org.joda.time.Interval) GroupByQueryRunnerTest(io.druid.query.groupby.GroupByQueryRunnerTest) Test(org.junit.Test)

Aggregations

LongSumAggregatorFactory (io.druid.query.aggregation.LongSumAggregatorFactory)160 Test (org.junit.Test)137 DefaultDimensionSpec (io.druid.query.dimension.DefaultDimensionSpec)87 Row (io.druid.data.input.Row)80 DimensionSpec (io.druid.query.dimension.DimensionSpec)58 ExtractionDimensionSpec (io.druid.query.dimension.ExtractionDimensionSpec)54 ListFilteredDimensionSpec (io.druid.query.dimension.ListFilteredDimensionSpec)46 RegexFilteredDimensionSpec (io.druid.query.dimension.RegexFilteredDimensionSpec)46 Interval (org.joda.time.Interval)46 AggregatorFactory (io.druid.query.aggregation.AggregatorFactory)38 DateTime (org.joda.time.DateTime)37 CountAggregatorFactory (io.druid.query.aggregation.CountAggregatorFactory)36 Result (io.druid.query.Result)30 Period (org.joda.time.Period)26 DoubleSumAggregatorFactory (io.druid.query.aggregation.DoubleSumAggregatorFactory)25 PeriodGranularity (io.druid.java.util.common.granularity.PeriodGranularity)24 LookupExtractionFn (io.druid.query.lookup.LookupExtractionFn)24 QueryRunner (io.druid.query.QueryRunner)22 HashMap (java.util.HashMap)22 FinalizeResultsQueryRunner (io.druid.query.FinalizeResultsQueryRunner)21