Search in sources :

Example 26 with AlgebricksMetaOperatorDescriptor

use of org.apache.hyracks.algebricks.runtime.operators.meta.AlgebricksMetaOperatorDescriptor in project asterixdb by apache.

the class PushRuntimeTest method scanMicroSortGbySelectWrite.

@Test
public void scanMicroSortGbySelectWrite() throws Exception {
    JobSpecification spec = new JobSpecification(FRAME_SIZE);
    // the scanner
    FileSplit[] fileSplits = new FileSplit[1];
    fileSplits[0] = new ManagedFileSplit(AlgebricksHyracksIntegrationUtil.NC1_ID, "data" + File.separator + "tpch0.001" + File.separator + "customer.tbl");
    IFileSplitProvider splitProvider = new ConstantFileSplitProvider(fileSplits);
    RecordDescriptor scannerDesc = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE, new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), IntegerSerializerDeserializer.INSTANCE, new UTF8StringSerializerDeserializer(), FloatSerializerDeserializer.INSTANCE, new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer() });
    IValueParserFactory[] valueParsers = new IValueParserFactory[] { IntegerParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, IntegerParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, FloatParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE };
    FileScanOperatorDescriptor scanner = new FileScanOperatorDescriptor(spec, splitProvider, new DelimitedDataTupleParserFactory(valueParsers, '|'), scannerDesc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, scanner, new String[] { AlgebricksHyracksIntegrationUtil.NC1_ID });
    // the sort (by nation id)
    RecordDescriptor sortDesc = scannerDesc;
    InMemorySortRuntimeFactory sort = new InMemorySortRuntimeFactory(new int[] { 3 }, null, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(IntegerPointable.FACTORY) }, null);
    // the group-by
    NestedTupleSourceRuntimeFactory nts = new NestedTupleSourceRuntimeFactory();
    RecordDescriptor ntsDesc = sortDesc;
    AggregateRuntimeFactory agg = new AggregateRuntimeFactory(new IAggregateEvaluatorFactory[] { new TupleCountAggregateFunctionFactory() });
    RecordDescriptor aggDesc = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE });
    AlgebricksPipeline pipeline = new AlgebricksPipeline(new IPushRuntimeFactory[] { nts, agg }, new RecordDescriptor[] { ntsDesc, aggDesc });
    NestedPlansAccumulatingAggregatorFactory npaaf = new NestedPlansAccumulatingAggregatorFactory(new AlgebricksPipeline[] { pipeline }, new int[] { 3 }, new int[] {});
    RecordDescriptor gbyDesc = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE, IntegerSerializerDeserializer.INSTANCE });
    MicroPreClusteredGroupRuntimeFactory gby = new MicroPreClusteredGroupRuntimeFactory(new int[] { 3 }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(IntegerPointable.FACTORY) }, npaaf, sortDesc, gbyDesc, null);
    // the algebricks op.
    IScalarEvaluatorFactory cond = new IntegerEqualsEvalFactory(new IntegerConstantEvalFactory(3), // Canadian customers
    new TupleFieldEvaluatorFactory(0));
    StreamSelectRuntimeFactory select = new StreamSelectRuntimeFactory(cond, new int[] { 1 }, BinaryBooleanInspectorImpl.FACTORY, false, -1, null);
    RecordDescriptor selectDesc = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE });
    String filePath = PATH_ACTUAL + SEPARATOR + "scanSortGbySelectWrite.out";
    File outFile = new File(filePath);
    SinkWriterRuntimeFactory writer = new SinkWriterRuntimeFactory(new int[] { 0 }, new IPrinterFactory[] { IntegerPrinterFactory.INSTANCE }, outFile, PrinterBasedWriterFactory.INSTANCE, selectDesc);
    AlgebricksMetaOperatorDescriptor algebricksOp = new AlgebricksMetaOperatorDescriptor(spec, 1, 0, new IPushRuntimeFactory[] { sort, gby, select, writer }, new RecordDescriptor[] { sortDesc, gbyDesc, selectDesc, null });
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, algebricksOp, new String[] { AlgebricksHyracksIntegrationUtil.NC1_ID });
    spec.connect(new OneToOneConnectorDescriptor(spec), scanner, 0, algebricksOp, 0);
    spec.addRoot(algebricksOp);
    AlgebricksHyracksIntegrationUtil.runJob(spec);
    StringBuilder buf = new StringBuilder();
    readFileToString(outFile, buf);
    Assert.assertEquals("9", buf.toString());
    outFile.delete();
}
Also used : InMemorySortRuntimeFactory(org.apache.hyracks.algebricks.runtime.operators.sort.InMemorySortRuntimeFactory) AggregateRuntimeFactory(org.apache.hyracks.algebricks.runtime.operators.aggreg.AggregateRuntimeFactory) RunningAggregateRuntimeFactory(org.apache.hyracks.algebricks.runtime.operators.std.RunningAggregateRuntimeFactory) MicroPreClusteredGroupRuntimeFactory(org.apache.hyracks.algebricks.runtime.operators.group.MicroPreClusteredGroupRuntimeFactory) IFileSplitProvider(org.apache.hyracks.dataflow.std.file.IFileSplitProvider) RecordDescriptor(org.apache.hyracks.api.dataflow.value.RecordDescriptor) OneToOneConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.OneToOneConnectorDescriptor) FileSplit(org.apache.hyracks.api.io.FileSplit) ManagedFileSplit(org.apache.hyracks.api.io.ManagedFileSplit) UTF8StringSerializerDeserializer(org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer) TupleCountAggregateFunctionFactory(org.apache.hyracks.algebricks.runtime.aggregators.TupleCountAggregateFunctionFactory) NestedTupleSourceRuntimeFactory(org.apache.hyracks.algebricks.runtime.operators.std.NestedTupleSourceRuntimeFactory) ManagedFileSplit(org.apache.hyracks.api.io.ManagedFileSplit) FileScanOperatorDescriptor(org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor) JobSpecification(org.apache.hyracks.api.job.JobSpecification) AlgebricksPipeline(org.apache.hyracks.algebricks.runtime.base.AlgebricksPipeline) TupleFieldEvaluatorFactory(org.apache.hyracks.algebricks.runtime.evaluators.TupleFieldEvaluatorFactory) IValueParserFactory(org.apache.hyracks.dataflow.common.data.parsers.IValueParserFactory) ConstantFileSplitProvider(org.apache.hyracks.dataflow.std.file.ConstantFileSplitProvider) AlgebricksMetaOperatorDescriptor(org.apache.hyracks.algebricks.runtime.operators.meta.AlgebricksMetaOperatorDescriptor) DelimitedDataTupleParserFactory(org.apache.hyracks.dataflow.std.file.DelimitedDataTupleParserFactory) NestedPlansAccumulatingAggregatorFactory(org.apache.hyracks.algebricks.runtime.operators.aggreg.NestedPlansAccumulatingAggregatorFactory) IScalarEvaluatorFactory(org.apache.hyracks.algebricks.runtime.base.IScalarEvaluatorFactory) StreamSelectRuntimeFactory(org.apache.hyracks.algebricks.runtime.operators.std.StreamSelectRuntimeFactory) SinkWriterRuntimeFactory(org.apache.hyracks.algebricks.runtime.operators.std.SinkWriterRuntimeFactory) File(java.io.File) Test(org.junit.Test)

Example 27 with AlgebricksMetaOperatorDescriptor

use of org.apache.hyracks.algebricks.runtime.operators.meta.AlgebricksMetaOperatorDescriptor in project asterixdb by apache.

the class PushRuntimeTest method scanAggregateWrite.

@Test
public void scanAggregateWrite() throws Exception {
    JobSpecification spec = new JobSpecification(FRAME_SIZE);
    // the scanner
    FileSplit[] fileSplits = new FileSplit[1];
    fileSplits[0] = new ManagedFileSplit(AlgebricksHyracksIntegrationUtil.NC1_ID, "data" + File.separator + "tpch0.001" + File.separator + "customer-part1.tbl");
    IFileSplitProvider splitProvider = new ConstantFileSplitProvider(fileSplits);
    RecordDescriptor scannerDesc = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE, new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), IntegerSerializerDeserializer.INSTANCE, new UTF8StringSerializerDeserializer(), FloatSerializerDeserializer.INSTANCE, new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer() });
    IValueParserFactory[] valueParsers = new IValueParserFactory[] { IntegerParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, IntegerParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, FloatParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE };
    FileScanOperatorDescriptor scanner = new FileScanOperatorDescriptor(spec, splitProvider, new DelimitedDataTupleParserFactory(valueParsers, '|'), scannerDesc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, scanner, new String[] { AlgebricksHyracksIntegrationUtil.NC1_ID });
    // the algebricks op.
    AggregateRuntimeFactory agg = new AggregateRuntimeFactory(new IAggregateEvaluatorFactory[] { new TupleCountAggregateFunctionFactory() });
    RecordDescriptor aggDesc = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE });
    String filePath = PATH_ACTUAL + SEPARATOR + "scanAggregateWrite.out";
    File outFile = new File(filePath);
    SinkWriterRuntimeFactory writer = new SinkWriterRuntimeFactory(new int[] { 0 }, new IPrinterFactory[] { IntegerPrinterFactory.INSTANCE }, outFile, PrinterBasedWriterFactory.INSTANCE, aggDesc);
    AlgebricksMetaOperatorDescriptor algebricksOp = new AlgebricksMetaOperatorDescriptor(spec, 1, 0, new IPushRuntimeFactory[] { agg, writer }, new RecordDescriptor[] { aggDesc, null });
    PartitionConstraintHelper.addPartitionCountConstraint(spec, algebricksOp, 1);
    spec.connect(new OneToOneConnectorDescriptor(spec), scanner, 0, algebricksOp, 0);
    spec.addRoot(algebricksOp);
    AlgebricksHyracksIntegrationUtil.runJob(spec);
    StringBuilder buf = new StringBuilder();
    readFileToString(outFile, buf);
    Assert.assertEquals("75", buf.toString());
    outFile.delete();
}
Also used : AggregateRuntimeFactory(org.apache.hyracks.algebricks.runtime.operators.aggreg.AggregateRuntimeFactory) RunningAggregateRuntimeFactory(org.apache.hyracks.algebricks.runtime.operators.std.RunningAggregateRuntimeFactory) IValueParserFactory(org.apache.hyracks.dataflow.common.data.parsers.IValueParserFactory) IFileSplitProvider(org.apache.hyracks.dataflow.std.file.IFileSplitProvider) RecordDescriptor(org.apache.hyracks.api.dataflow.value.RecordDescriptor) ConstantFileSplitProvider(org.apache.hyracks.dataflow.std.file.ConstantFileSplitProvider) AlgebricksMetaOperatorDescriptor(org.apache.hyracks.algebricks.runtime.operators.meta.AlgebricksMetaOperatorDescriptor) DelimitedDataTupleParserFactory(org.apache.hyracks.dataflow.std.file.DelimitedDataTupleParserFactory) OneToOneConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.OneToOneConnectorDescriptor) FileSplit(org.apache.hyracks.api.io.FileSplit) ManagedFileSplit(org.apache.hyracks.api.io.ManagedFileSplit) UTF8StringSerializerDeserializer(org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer) TupleCountAggregateFunctionFactory(org.apache.hyracks.algebricks.runtime.aggregators.TupleCountAggregateFunctionFactory) ManagedFileSplit(org.apache.hyracks.api.io.ManagedFileSplit) SinkWriterRuntimeFactory(org.apache.hyracks.algebricks.runtime.operators.std.SinkWriterRuntimeFactory) FileScanOperatorDescriptor(org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor) JobSpecification(org.apache.hyracks.api.job.JobSpecification) File(java.io.File) Test(org.junit.Test)

Aggregations

AlgebricksMetaOperatorDescriptor (org.apache.hyracks.algebricks.runtime.operators.meta.AlgebricksMetaOperatorDescriptor)27 JobSpecification (org.apache.hyracks.api.job.JobSpecification)19 RecordDescriptor (org.apache.hyracks.api.dataflow.value.RecordDescriptor)16 Test (org.junit.Test)13 File (java.io.File)12 SinkWriterRuntimeFactory (org.apache.hyracks.algebricks.runtime.operators.std.SinkWriterRuntimeFactory)12 OneToOneConnectorDescriptor (org.apache.hyracks.dataflow.std.connectors.OneToOneConnectorDescriptor)11 AlgebricksPartitionConstraint (org.apache.hyracks.algebricks.common.constraints.AlgebricksPartitionConstraint)9 AssignRuntimeFactory (org.apache.hyracks.algebricks.runtime.operators.std.AssignRuntimeFactory)8 EmptyTupleSourceRuntimeFactory (org.apache.hyracks.algebricks.runtime.operators.std.EmptyTupleSourceRuntimeFactory)8 IScalarEvaluatorFactory (org.apache.hyracks.algebricks.runtime.base.IScalarEvaluatorFactory)7 IValueParserFactory (org.apache.hyracks.dataflow.common.data.parsers.IValueParserFactory)7 DelimitedDataTupleParserFactory (org.apache.hyracks.dataflow.std.file.DelimitedDataTupleParserFactory)7 IFileSplitProvider (org.apache.hyracks.dataflow.std.file.IFileSplitProvider)7 FileSplit (org.apache.hyracks.api.io.FileSplit)6 ManagedFileSplit (org.apache.hyracks.api.io.ManagedFileSplit)6 ConstantFileSplitProvider (org.apache.hyracks.dataflow.std.file.ConstantFileSplitProvider)6 FileScanOperatorDescriptor (org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor)6 IOperatorDescriptor (org.apache.hyracks.api.dataflow.IOperatorDescriptor)5 UTF8StringSerializerDeserializer (org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer)5