Search in sources :

Example 11 with LinearShardSpec

use of io.druid.timeline.partition.LinearShardSpec in project druid by druid-io.

the class QuantileSqlAggregatorTest method setUp.

@Before
public void setUp() throws Exception {
    Calcites.setSystemProperties();
    // Note: this is needed in order to properly register the serde for Histogram.
    new ApproximateHistogramDruidModule().configure(null);
    final QueryableIndex index = IndexBuilder.create().tmpDir(temporaryFolder.newFolder()).indexMerger(TestHelper.getTestIndexMergerV9()).schema(new IncrementalIndexSchema.Builder().withMetrics(new AggregatorFactory[] { new CountAggregatorFactory("cnt"), new DoubleSumAggregatorFactory("m1", "m1"), new ApproximateHistogramAggregatorFactory("hist_m1", "m1", null, null, null, null) }).withRollup(false).build()).rows(CalciteTests.ROWS1).buildMMappedIndex();
    walker = new SpecificSegmentsQuerySegmentWalker(CalciteTests.queryRunnerFactoryConglomerate()).add(DataSegment.builder().dataSource(DATA_SOURCE).interval(index.getDataInterval()).version("1").shardSpec(new LinearShardSpec(0)).build(), index);
    final PlannerConfig plannerConfig = new PlannerConfig();
    final SchemaPlus rootSchema = Calcites.createRootSchema(CalciteTests.createMockSchema(walker, plannerConfig));
    final DruidOperatorTable operatorTable = new DruidOperatorTable(ImmutableSet.<SqlAggregator>of(new QuantileSqlAggregator()), ImmutableSet.<SqlExtractionOperator>of());
    plannerFactory = new PlannerFactory(rootSchema, walker, operatorTable, plannerConfig);
}
Also used : DoubleSumAggregatorFactory(io.druid.query.aggregation.DoubleSumAggregatorFactory) ApproximateHistogramDruidModule(io.druid.query.aggregation.histogram.ApproximateHistogramDruidModule) LinearShardSpec(io.druid.timeline.partition.LinearShardSpec) IndexBuilder(io.druid.segment.IndexBuilder) SchemaPlus(org.apache.calcite.schema.SchemaPlus) AggregatorFactory(io.druid.query.aggregation.AggregatorFactory) CountAggregatorFactory(io.druid.query.aggregation.CountAggregatorFactory) ApproximateHistogramFoldingAggregatorFactory(io.druid.query.aggregation.histogram.ApproximateHistogramFoldingAggregatorFactory) FilteredAggregatorFactory(io.druid.query.aggregation.FilteredAggregatorFactory) ApproximateHistogramAggregatorFactory(io.druid.query.aggregation.histogram.ApproximateHistogramAggregatorFactory) DoubleSumAggregatorFactory(io.druid.query.aggregation.DoubleSumAggregatorFactory) DruidOperatorTable(io.druid.sql.calcite.planner.DruidOperatorTable) CountAggregatorFactory(io.druid.query.aggregation.CountAggregatorFactory) SpecificSegmentsQuerySegmentWalker(io.druid.sql.calcite.util.SpecificSegmentsQuerySegmentWalker) QueryableIndex(io.druid.segment.QueryableIndex) PlannerConfig(io.druid.sql.calcite.planner.PlannerConfig) PlannerFactory(io.druid.sql.calcite.planner.PlannerFactory) ApproximateHistogramAggregatorFactory(io.druid.query.aggregation.histogram.ApproximateHistogramAggregatorFactory) Before(org.junit.Before)

Example 12 with LinearShardSpec

use of io.druid.timeline.partition.LinearShardSpec in project druid by druid-io.

the class DefaultOfflineAppenderatorFactoryTest method testBuild.

@Test
public void testBuild() throws IOException, SegmentNotWritableException {
    Injector injector = Initialization.makeInjectorWithModules(GuiceInjectors.makeStartupInjector(), ImmutableList.<Module>of(new Module() {

        @Override
        public void configure(Binder binder) {
            binder.bindConstant().annotatedWith(Names.named("serviceName")).to("druid/tool");
            binder.bindConstant().annotatedWith(Names.named("servicePort")).to(9999);
            binder.bind(DruidProcessingConfig.class).toInstance(new DruidProcessingConfig() {

                @Override
                public String getFormatString() {
                    return "processing-%s";
                }

                @Override
                public int intermediateComputeSizeBytes() {
                    return 100 * 1024 * 1024;
                }

                @Override
                public int getNumThreads() {
                    return 1;
                }

                @Override
                public int columnCacheSizeBytes() {
                    return 25 * 1024 * 1024;
                }
            });
            binder.bind(ColumnConfig.class).to(DruidProcessingConfig.class);
        }
    }));
    ObjectMapper objectMapper = injector.getInstance(ObjectMapper.class);
    AppenderatorFactory defaultOfflineAppenderatorFactory = objectMapper.reader(AppenderatorFactory.class).readValue("{\"type\":\"offline\"}");
    final Map<String, Object> parserMap = objectMapper.convertValue(new MapInputRowParser(new JSONParseSpec(new TimestampSpec("ts", "auto", null), new DimensionsSpec(null, null, null), null, null)), Map.class);
    DataSchema schema = new DataSchema("dataSourceName", parserMap, new AggregatorFactory[] { new CountAggregatorFactory("count"), new LongSumAggregatorFactory("met", "met") }, new UniformGranularitySpec(Granularities.MINUTE, Granularities.NONE, null), objectMapper);
    RealtimeTuningConfig tuningConfig = new RealtimeTuningConfig(75000, null, null, temporaryFolder.newFolder(), null, null, null, null, null, null, 0, 0, null, null);
    try (Appenderator appenderator = defaultOfflineAppenderatorFactory.build(schema, tuningConfig, new FireDepartmentMetrics())) {
        Assert.assertEquals("dataSourceName", appenderator.getDataSource());
        Assert.assertEquals(null, appenderator.startJob());
        SegmentIdentifier identifier = new SegmentIdentifier("dataSourceName", new Interval("2000/2001"), "A", new LinearShardSpec(0));
        Assert.assertEquals(0, ((AppenderatorImpl) appenderator).getRowsInMemory());
        appenderator.add(identifier, AppenderatorTest.IR("2000", "bar", 1), Suppliers.ofInstance(Committers.nil()));
        Assert.assertEquals(1, ((AppenderatorImpl) appenderator).getRowsInMemory());
        appenderator.add(identifier, AppenderatorTest.IR("2000", "baz", 1), Suppliers.ofInstance(Committers.nil()));
        Assert.assertEquals(2, ((AppenderatorImpl) appenderator).getRowsInMemory());
        appenderator.close();
        Assert.assertEquals(0, ((AppenderatorImpl) appenderator).getRowsInMemory());
    }
}
Also used : ColumnConfig(io.druid.segment.column.ColumnConfig) MapInputRowParser(io.druid.data.input.impl.MapInputRowParser) LinearShardSpec(io.druid.timeline.partition.LinearShardSpec) LongSumAggregatorFactory(io.druid.query.aggregation.LongSumAggregatorFactory) Binder(com.google.inject.Binder) UniformGranularitySpec(io.druid.segment.indexing.granularity.UniformGranularitySpec) Injector(com.google.inject.Injector) TimestampSpec(io.druid.data.input.impl.TimestampSpec) JSONParseSpec(io.druid.data.input.impl.JSONParseSpec) ObjectMapper(com.fasterxml.jackson.databind.ObjectMapper) RealtimeTuningConfig(io.druid.segment.indexing.RealtimeTuningConfig) DataSchema(io.druid.segment.indexing.DataSchema) FireDepartmentMetrics(io.druid.segment.realtime.FireDepartmentMetrics) CountAggregatorFactory(io.druid.query.aggregation.CountAggregatorFactory) DimensionsSpec(io.druid.data.input.impl.DimensionsSpec) Module(com.google.inject.Module) DruidProcessingConfig(io.druid.query.DruidProcessingConfig) Interval(org.joda.time.Interval) Test(org.junit.Test)

Example 13 with LinearShardSpec

use of io.druid.timeline.partition.LinearShardSpec in project druid by druid-io.

the class ActionBasedUsedSegmentCheckerTest method testBasic.

@Test
public void testBasic() throws IOException {
    final TaskActionClient taskActionClient = EasyMock.createMock(TaskActionClient.class);
    EasyMock.expect(taskActionClient.submit(new SegmentListUsedAction("bar", null, ImmutableList.of(new Interval("2002/P1D"))))).andReturn(ImmutableList.of(DataSegment.builder().dataSource("bar").interval(new Interval("2002/P1D")).shardSpec(new LinearShardSpec(0)).version("b").build(), DataSegment.builder().dataSource("bar").interval(new Interval("2002/P1D")).shardSpec(new LinearShardSpec(1)).version("b").build()));
    EasyMock.expect(taskActionClient.submit(new SegmentListUsedAction("foo", null, ImmutableList.of(new Interval("2000/P1D"), new Interval("2001/P1D"))))).andReturn(ImmutableList.of(DataSegment.builder().dataSource("foo").interval(new Interval("2000/P1D")).shardSpec(new LinearShardSpec(0)).version("a").build(), DataSegment.builder().dataSource("foo").interval(new Interval("2000/P1D")).shardSpec(new LinearShardSpec(1)).version("a").build(), DataSegment.builder().dataSource("foo").interval(new Interval("2001/P1D")).shardSpec(new LinearShardSpec(1)).version("b").build(), DataSegment.builder().dataSource("foo").interval(new Interval("2002/P1D")).shardSpec(new LinearShardSpec(1)).version("b").build()));
    EasyMock.replay(taskActionClient);
    final UsedSegmentChecker checker = new ActionBasedUsedSegmentChecker(taskActionClient);
    final Set<DataSegment> segments = checker.findUsedSegments(ImmutableSet.of(new SegmentIdentifier("foo", new Interval("2000/P1D"), "a", new LinearShardSpec(1)), new SegmentIdentifier("foo", new Interval("2001/P1D"), "b", new LinearShardSpec(0)), new SegmentIdentifier("bar", new Interval("2002/P1D"), "b", new LinearShardSpec(0))));
    Assert.assertEquals(ImmutableSet.of(DataSegment.builder().dataSource("foo").interval(new Interval("2000/P1D")).shardSpec(new LinearShardSpec(1)).version("a").build(), DataSegment.builder().dataSource("bar").interval(new Interval("2002/P1D")).shardSpec(new LinearShardSpec(0)).version("b").build()), segments);
    EasyMock.verify(taskActionClient);
}
Also used : TaskActionClient(io.druid.indexing.common.actions.TaskActionClient) SegmentIdentifier(io.druid.segment.realtime.appenderator.SegmentIdentifier) LinearShardSpec(io.druid.timeline.partition.LinearShardSpec) UsedSegmentChecker(io.druid.segment.realtime.appenderator.UsedSegmentChecker) SegmentListUsedAction(io.druid.indexing.common.actions.SegmentListUsedAction) DataSegment(io.druid.timeline.DataSegment) Interval(org.joda.time.Interval) Test(org.junit.Test)

Example 14 with LinearShardSpec

use of io.druid.timeline.partition.LinearShardSpec in project druid by druid-io.

the class SegmentAllocateActionTest method testAddToExistingLinearShardSpecsSameGranularity.

@Test
public void testAddToExistingLinearShardSpecsSameGranularity() throws Exception {
    final Task task = new NoopTask(null, 0, 0, null, null, null);
    taskActionTestKit.getMetadataStorageCoordinator().announceHistoricalSegments(ImmutableSet.of(DataSegment.builder().dataSource(DATA_SOURCE).interval(Granularities.HOUR.bucket(PARTY_TIME)).version(PARTY_TIME.toString()).shardSpec(new LinearShardSpec(0)).build(), DataSegment.builder().dataSource(DATA_SOURCE).interval(Granularities.HOUR.bucket(PARTY_TIME)).version(PARTY_TIME.toString()).shardSpec(new LinearShardSpec(1)).build()));
    taskActionTestKit.getTaskLockbox().add(task);
    final SegmentIdentifier id1 = allocate(task, PARTY_TIME, Granularities.NONE, Granularities.HOUR, "s1", null);
    final SegmentIdentifier id2 = allocate(task, PARTY_TIME, Granularities.NONE, Granularities.HOUR, "s1", id1.getIdentifierAsString());
    assertSameIdentifier(id1, new SegmentIdentifier(DATA_SOURCE, Granularities.HOUR.bucket(PARTY_TIME), PARTY_TIME.toString(), new LinearShardSpec(2)));
    assertSameIdentifier(id2, new SegmentIdentifier(DATA_SOURCE, Granularities.HOUR.bucket(PARTY_TIME), PARTY_TIME.toString(), new LinearShardSpec(3)));
}
Also used : Task(io.druid.indexing.common.task.Task) NoopTask(io.druid.indexing.common.task.NoopTask) SegmentIdentifier(io.druid.segment.realtime.appenderator.SegmentIdentifier) LinearShardSpec(io.druid.timeline.partition.LinearShardSpec) NoopTask(io.druid.indexing.common.task.NoopTask) Test(org.junit.Test)

Example 15 with LinearShardSpec

use of io.druid.timeline.partition.LinearShardSpec in project druid by druid-io.

the class DruidSchemaTest method setUp.

@Before
public void setUp() throws Exception {
    Calcites.setSystemProperties();
    final File tmpDir = temporaryFolder.newFolder();
    final QueryableIndex index1 = IndexBuilder.create().tmpDir(new File(tmpDir, "1")).indexMerger(TestHelper.getTestIndexMergerV9()).schema(new IncrementalIndexSchema.Builder().withMetrics(new AggregatorFactory[] { new CountAggregatorFactory("cnt"), new DoubleSumAggregatorFactory("m1", "m1"), new HyperUniquesAggregatorFactory("unique_dim1", "dim1") }).withRollup(false).build()).rows(ROWS1).buildMMappedIndex();
    final QueryableIndex index2 = IndexBuilder.create().tmpDir(new File(tmpDir, "2")).indexMerger(TestHelper.getTestIndexMergerV9()).schema(new IncrementalIndexSchema.Builder().withMetrics(new AggregatorFactory[] { new LongSumAggregatorFactory("m1", "m1") }).withRollup(false).build()).rows(ROWS2).buildMMappedIndex();
    walker = new SpecificSegmentsQuerySegmentWalker(CalciteTests.queryRunnerFactoryConglomerate()).add(DataSegment.builder().dataSource(CalciteTests.DATASOURCE1).interval(new Interval("2000/P1Y")).version("1").shardSpec(new LinearShardSpec(0)).build(), index1).add(DataSegment.builder().dataSource(CalciteTests.DATASOURCE1).interval(new Interval("2001/P1Y")).version("1").shardSpec(new LinearShardSpec(0)).build(), index2).add(DataSegment.builder().dataSource(CalciteTests.DATASOURCE2).interval(index2.getDataInterval()).version("1").shardSpec(new LinearShardSpec(0)).build(), index2);
    schema = new DruidSchema(walker, new TestServerInventoryView(walker.getSegments()), PLANNER_CONFIG_DEFAULT);
    schema.start();
    schema.awaitInitialization();
}
Also used : DoubleSumAggregatorFactory(io.druid.query.aggregation.DoubleSumAggregatorFactory) TestServerInventoryView(io.druid.sql.calcite.util.TestServerInventoryView) LinearShardSpec(io.druid.timeline.partition.LinearShardSpec) IndexBuilder(io.druid.segment.IndexBuilder) LongSumAggregatorFactory(io.druid.query.aggregation.LongSumAggregatorFactory) AggregatorFactory(io.druid.query.aggregation.AggregatorFactory) CountAggregatorFactory(io.druid.query.aggregation.CountAggregatorFactory) HyperUniquesAggregatorFactory(io.druid.query.aggregation.hyperloglog.HyperUniquesAggregatorFactory) DoubleSumAggregatorFactory(io.druid.query.aggregation.DoubleSumAggregatorFactory) LongSumAggregatorFactory(io.druid.query.aggregation.LongSumAggregatorFactory) CountAggregatorFactory(io.druid.query.aggregation.CountAggregatorFactory) SpecificSegmentsQuerySegmentWalker(io.druid.sql.calcite.util.SpecificSegmentsQuerySegmentWalker) QueryableIndex(io.druid.segment.QueryableIndex) HyperUniquesAggregatorFactory(io.druid.query.aggregation.hyperloglog.HyperUniquesAggregatorFactory) File(java.io.File) Interval(org.joda.time.Interval) Before(org.junit.Before)

Aggregations

LinearShardSpec (io.druid.timeline.partition.LinearShardSpec)21 Interval (org.joda.time.Interval)17 DataSegment (io.druid.timeline.DataSegment)12 Test (org.junit.Test)12 DataSegmentPusher (io.druid.segment.loading.DataSegmentPusher)8 File (java.io.File)8 MetadataStorageTablesConfig (io.druid.metadata.MetadataStorageTablesConfig)7 HdfsDataSegmentPusher (io.druid.storage.hdfs.HdfsDataSegmentPusher)7 HdfsDataSegmentPusherConfig (io.druid.storage.hdfs.HdfsDataSegmentPusherConfig)7 LocalFileSystem (org.apache.hadoop.fs.LocalFileSystem)7 Path (org.apache.hadoop.fs.Path)7 CountAggregatorFactory (io.druid.query.aggregation.CountAggregatorFactory)5 SegmentIdentifier (io.druid.segment.realtime.appenderator.SegmentIdentifier)5 AggregatorFactory (io.druid.query.aggregation.AggregatorFactory)4 IOException (java.io.IOException)4 LongSumAggregatorFactory (io.druid.query.aggregation.LongSumAggregatorFactory)3 QueryableIndex (io.druid.segment.QueryableIndex)3 DateTime (org.joda.time.DateTime)3 ObjectMapper (com.fasterxml.jackson.databind.ObjectMapper)2 InputRow (io.druid.data.input.InputRow)2