Search in sources :

Example 46 with RecordDescriptor

use of org.apache.hyracks.api.dataflow.value.RecordDescriptor in project asterixdb by apache.

the class AggregationTest method multiKeyMinMaxStringPreClusterGroupTest.

@Test
public void multiKeyMinMaxStringPreClusterGroupTest() throws Exception {
    JobSpecification spec = new JobSpecification();
    FileScanOperatorDescriptor csvScanner = new FileScanOperatorDescriptor(spec, splitProvider, tupleParserFactory, desc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, csvScanner, NC2_ID);
    RecordDescriptor outputRec = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), IntegerSerializerDeserializer.INSTANCE, new UTF8StringSerializerDeserializer() });
    int[] keyFields = new int[] { 8, 0 };
    PreclusteredGroupOperatorDescriptor grouper = new PreclusteredGroupOperatorDescriptor(spec, keyFields, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY), PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, new MultiFieldsAggregatorFactory(new IFieldAggregateDescriptorFactory[] { new IntSumFieldAggregatorFactory(1, true), new MinMaxStringFieldAggregatorFactory(15, true, false) }), outputRec);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, grouper, NC2_ID, NC1_ID);
    IConnectorDescriptor conn1 = new MToNPartitioningConnectorDescriptor(spec, new FieldHashPartitionComputerFactory(keyFields, new IBinaryHashFunctionFactory[] { PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY), PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY) }));
    spec.connect(conn1, csvScanner, 0, grouper, 0);
    AbstractSingleActivityOperatorDescriptor printer = getPrinter(spec, "multiKeyMinMaxStringPreClusterGroupTest");
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, printer, NC2_ID, NC1_ID);
    IConnectorDescriptor conn2 = new OneToOneConnectorDescriptor(spec);
    spec.connect(conn2, grouper, 0, printer, 0);
    spec.addRoot(printer);
    runTest(spec);
}
Also used : IConnectorDescriptor(org.apache.hyracks.api.dataflow.IConnectorDescriptor) MultiFieldsAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.MultiFieldsAggregatorFactory) AbstractSingleActivityOperatorDescriptor(org.apache.hyracks.dataflow.std.base.AbstractSingleActivityOperatorDescriptor) RecordDescriptor(org.apache.hyracks.api.dataflow.value.RecordDescriptor) MToNPartitioningConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.MToNPartitioningConnectorDescriptor) OneToOneConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.OneToOneConnectorDescriptor) UTF8StringSerializerDeserializer(org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer) FieldHashPartitionComputerFactory(org.apache.hyracks.dataflow.common.data.partition.FieldHashPartitionComputerFactory) IBinaryHashFunctionFactory(org.apache.hyracks.api.dataflow.value.IBinaryHashFunctionFactory) FileScanOperatorDescriptor(org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor) PreclusteredGroupOperatorDescriptor(org.apache.hyracks.dataflow.std.group.preclustered.PreclusteredGroupOperatorDescriptor) MinMaxStringFieldAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.MinMaxStringFieldAggregatorFactory) IntSumFieldAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.IntSumFieldAggregatorFactory) JobSpecification(org.apache.hyracks.api.job.JobSpecification) IFieldAggregateDescriptorFactory(org.apache.hyracks.dataflow.std.group.IFieldAggregateDescriptorFactory) Test(org.junit.Test)

Example 47 with RecordDescriptor

use of org.apache.hyracks.api.dataflow.value.RecordDescriptor in project asterixdb by apache.

the class AggregationTest method singleKeyAvgPreClusterGroupTest.

@Test
public void singleKeyAvgPreClusterGroupTest() throws Exception {
    JobSpecification spec = new JobSpecification();
    FileScanOperatorDescriptor csvScanner = new FileScanOperatorDescriptor(spec, splitProvider, tupleParserFactory, desc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, csvScanner, NC2_ID);
    RecordDescriptor outputRec = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer(), IntegerSerializerDeserializer.INSTANCE, IntegerSerializerDeserializer.INSTANCE, FloatSerializerDeserializer.INSTANCE });
    int[] keyFields = new int[] { 0 };
    PreclusteredGroupOperatorDescriptor grouper = new PreclusteredGroupOperatorDescriptor(spec, keyFields, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, new MultiFieldsAggregatorFactory(new IFieldAggregateDescriptorFactory[] { new IntSumFieldAggregatorFactory(1, true), new CountFieldAggregatorFactory(true), new AvgFieldGroupAggregatorFactory(1, true) }), outputRec);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, grouper, NC2_ID, NC1_ID);
    IConnectorDescriptor conn1 = new MToNPartitioningConnectorDescriptor(spec, new FieldHashPartitionComputerFactory(keyFields, new IBinaryHashFunctionFactory[] { PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY) }));
    spec.connect(conn1, csvScanner, 0, grouper, 0);
    AbstractSingleActivityOperatorDescriptor printer = getPrinter(spec, "singleKeyAvgInmemGroupTest");
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, printer, NC2_ID, NC1_ID);
    IConnectorDescriptor conn2 = new OneToOneConnectorDescriptor(spec);
    spec.connect(conn2, grouper, 0, printer, 0);
    spec.addRoot(printer);
    runTest(spec);
}
Also used : IConnectorDescriptor(org.apache.hyracks.api.dataflow.IConnectorDescriptor) MultiFieldsAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.MultiFieldsAggregatorFactory) AbstractSingleActivityOperatorDescriptor(org.apache.hyracks.dataflow.std.base.AbstractSingleActivityOperatorDescriptor) RecordDescriptor(org.apache.hyracks.api.dataflow.value.RecordDescriptor) MToNPartitioningConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.MToNPartitioningConnectorDescriptor) CountFieldAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.CountFieldAggregatorFactory) OneToOneConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.OneToOneConnectorDescriptor) UTF8StringSerializerDeserializer(org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer) AvgFieldGroupAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.AvgFieldGroupAggregatorFactory) FieldHashPartitionComputerFactory(org.apache.hyracks.dataflow.common.data.partition.FieldHashPartitionComputerFactory) IBinaryHashFunctionFactory(org.apache.hyracks.api.dataflow.value.IBinaryHashFunctionFactory) FileScanOperatorDescriptor(org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor) PreclusteredGroupOperatorDescriptor(org.apache.hyracks.dataflow.std.group.preclustered.PreclusteredGroupOperatorDescriptor) IntSumFieldAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.IntSumFieldAggregatorFactory) JobSpecification(org.apache.hyracks.api.job.JobSpecification) IFieldAggregateDescriptorFactory(org.apache.hyracks.dataflow.std.group.IFieldAggregateDescriptorFactory) Test(org.junit.Test)

Example 48 with RecordDescriptor

use of org.apache.hyracks.api.dataflow.value.RecordDescriptor in project asterixdb by apache.

the class AggregationTest method singleKeySumPreClusterGroupTest.

@Test
public void singleKeySumPreClusterGroupTest() throws Exception {
    JobSpecification spec = new JobSpecification();
    FileScanOperatorDescriptor csvScanner = new FileScanOperatorDescriptor(spec, splitProvider, tupleParserFactory, desc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, csvScanner, NC2_ID);
    RecordDescriptor outputRec = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer(), IntegerSerializerDeserializer.INSTANCE, IntegerSerializerDeserializer.INSTANCE, FloatSerializerDeserializer.INSTANCE });
    int[] keyFields = new int[] { 0 };
    PreclusteredGroupOperatorDescriptor grouper = new PreclusteredGroupOperatorDescriptor(spec, keyFields, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, new MultiFieldsAggregatorFactory(new IFieldAggregateDescriptorFactory[] { new IntSumFieldAggregatorFactory(1, true), new IntSumFieldAggregatorFactory(3, true), new FloatSumFieldAggregatorFactory(5, true) }), outputRec);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, grouper, NC2_ID, NC1_ID);
    IConnectorDescriptor conn1 = new MToNPartitioningConnectorDescriptor(spec, new FieldHashPartitionComputerFactory(keyFields, new IBinaryHashFunctionFactory[] { PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY) }));
    spec.connect(conn1, csvScanner, 0, grouper, 0);
    AbstractSingleActivityOperatorDescriptor printer = getPrinter(spec, "singleKeySumInmemGroupTest");
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, printer, NC2_ID, NC1_ID);
    IConnectorDescriptor conn2 = new OneToOneConnectorDescriptor(spec);
    spec.connect(conn2, grouper, 0, printer, 0);
    spec.addRoot(printer);
    runTest(spec);
}
Also used : IConnectorDescriptor(org.apache.hyracks.api.dataflow.IConnectorDescriptor) MultiFieldsAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.MultiFieldsAggregatorFactory) AbstractSingleActivityOperatorDescriptor(org.apache.hyracks.dataflow.std.base.AbstractSingleActivityOperatorDescriptor) RecordDescriptor(org.apache.hyracks.api.dataflow.value.RecordDescriptor) MToNPartitioningConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.MToNPartitioningConnectorDescriptor) OneToOneConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.OneToOneConnectorDescriptor) UTF8StringSerializerDeserializer(org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer) FieldHashPartitionComputerFactory(org.apache.hyracks.dataflow.common.data.partition.FieldHashPartitionComputerFactory) IBinaryHashFunctionFactory(org.apache.hyracks.api.dataflow.value.IBinaryHashFunctionFactory) FloatSumFieldAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.FloatSumFieldAggregatorFactory) FileScanOperatorDescriptor(org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor) PreclusteredGroupOperatorDescriptor(org.apache.hyracks.dataflow.std.group.preclustered.PreclusteredGroupOperatorDescriptor) IntSumFieldAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.IntSumFieldAggregatorFactory) JobSpecification(org.apache.hyracks.api.job.JobSpecification) IFieldAggregateDescriptorFactory(org.apache.hyracks.dataflow.std.group.IFieldAggregateDescriptorFactory) Test(org.junit.Test)

Example 49 with RecordDescriptor

use of org.apache.hyracks.api.dataflow.value.RecordDescriptor in project asterixdb by apache.

the class SleepOperatorDescriptor method jobWithSleepOp.

private JobSpecification jobWithSleepOp() {
    JobSpecification spec = new JobSpecification();
    FileSplit[] ordersSplits = new FileSplit[] { new ManagedFileSplit(ASTERIX_IDS[0], "data" + File.separator + "tpch0.001" + File.separator + "orders-part1.tbl") };
    IFileSplitProvider ordersSplitsProvider = new ConstantFileSplitProvider(ordersSplits);
    RecordDescriptor recordDesc = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer() });
    // File scan operator.
    FileScanOperatorDescriptor scanOp = new FileScanOperatorDescriptor(spec, ordersSplitsProvider, new DelimitedDataTupleParserFactory(new IValueParserFactory[] { UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE }, '|'), recordDesc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, scanOp, ASTERIX_IDS[0]);
    // Sleep operator.
    SleepOperatorDescriptor sleepOp = new SleepOperatorDescriptor(spec);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, sleepOp, ASTERIX_IDS);
    // Sink operator.
    SinkOperatorDescriptor sinkOp = new SinkOperatorDescriptor(spec, 1);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, sinkOp, ASTERIX_IDS);
    // Hash-repartitioning connector.
    IConnectorDescriptor conn1 = new MToNPartitioningConnectorDescriptor(spec, new FieldHashPartitionComputerFactory(new int[] { 0 }, new IBinaryHashFunctionFactory[] { PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY) }));
    spec.connect(conn1, scanOp, 0, sleepOp, 0);
    // One-to-one connector.
    IConnectorDescriptor conn2 = new OneToOneConnectorDescriptor(spec);
    spec.connect(conn2, sleepOp, 0, sinkOp, 0);
    return spec;
}
Also used : IConnectorDescriptor(org.apache.hyracks.api.dataflow.IConnectorDescriptor) 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) SinkOperatorDescriptor(org.apache.hyracks.dataflow.std.misc.SinkOperatorDescriptor) MToNPartitioningConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.MToNPartitioningConnectorDescriptor) DelimitedDataTupleParserFactory(org.apache.hyracks.dataflow.std.file.DelimitedDataTupleParserFactory) OneToOneConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.OneToOneConnectorDescriptor) ManagedFileSplit(org.apache.hyracks.api.io.ManagedFileSplit) FileSplit(org.apache.hyracks.api.io.FileSplit) UTF8StringSerializerDeserializer(org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer) FieldHashPartitionComputerFactory(org.apache.hyracks.dataflow.common.data.partition.FieldHashPartitionComputerFactory) IBinaryHashFunctionFactory(org.apache.hyracks.api.dataflow.value.IBinaryHashFunctionFactory) ManagedFileSplit(org.apache.hyracks.api.io.ManagedFileSplit) FileScanOperatorDescriptor(org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor) JobSpecification(org.apache.hyracks.api.job.JobSpecification)

Example 50 with RecordDescriptor

use of org.apache.hyracks.api.dataflow.value.RecordDescriptor in project asterixdb by apache.

the class RTreeSecondaryIndexSearchOperatorTest method searchSecondaryIndexTest.

@Test
public void searchSecondaryIndexTest() throws Exception {
    JobSpecification spec = new JobSpecification();
    // build tuple
    ArrayTupleBuilder tb = new ArrayTupleBuilder(secondaryKeyFieldCount);
    DataOutput dos = tb.getDataOutput();
    tb.reset();
    DoubleSerializerDeserializer.INSTANCE.serialize(61.2894, dos);
    tb.addFieldEndOffset();
    DoubleSerializerDeserializer.INSTANCE.serialize(-149.624, dos);
    tb.addFieldEndOffset();
    DoubleSerializerDeserializer.INSTANCE.serialize(61.8894, dos);
    tb.addFieldEndOffset();
    DoubleSerializerDeserializer.INSTANCE.serialize(-149.024, dos);
    tb.addFieldEndOffset();
    ISerializerDeserializer[] keyRecDescSers = { DoubleSerializerDeserializer.INSTANCE, DoubleSerializerDeserializer.INSTANCE, DoubleSerializerDeserializer.INSTANCE, DoubleSerializerDeserializer.INSTANCE };
    RecordDescriptor keyRecDesc = new RecordDescriptor(keyRecDescSers);
    ConstantTupleSourceOperatorDescriptor keyProviderOp = new ConstantTupleSourceOperatorDescriptor(spec, keyRecDesc, tb.getFieldEndOffsets(), tb.getByteArray(), tb.getSize());
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, keyProviderOp, NC1_ID);
    int[] keyFields = { 0, 1, 2, 3 };
    RTreeSearchOperatorDescriptor secondarySearchOp = new RTreeSearchOperatorDescriptor(spec, secondaryRecDesc, keyFields, true, true, secondaryHelperFactory, false, false, null, NoOpOperationCallbackFactory.INSTANCE, null, null, false);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, secondarySearchOp, NC1_ID);
    // fifth field from the tuples coming from secondary index
    int[] primaryLowKeyFields = { 4 };
    // fifth field from the tuples coming from secondary index
    int[] primaryHighKeyFields = { 4 };
    // search primary index
    BTreeSearchOperatorDescriptor primarySearchOp = new BTreeSearchOperatorDescriptor(spec, primaryRecDesc, primaryLowKeyFields, primaryHighKeyFields, true, true, primaryHelperFactory, false, false, null, NoOpOperationCallbackFactory.INSTANCE, null, null, false);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, primarySearchOp, NC1_ID);
    IFileSplitProvider outSplits = new ConstantFileSplitProvider(new FileSplit[] { createFile(nc1) });
    IOperatorDescriptor printer = new PlainFileWriterOperatorDescriptor(spec, outSplits, ",");
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, printer, NC1_ID);
    spec.connect(new OneToOneConnectorDescriptor(spec), keyProviderOp, 0, secondarySearchOp, 0);
    spec.connect(new OneToOneConnectorDescriptor(spec), secondarySearchOp, 0, primarySearchOp, 0);
    spec.connect(new OneToOneConnectorDescriptor(spec), primarySearchOp, 0, printer, 0);
    spec.addRoot(printer);
    runTest(spec);
}
Also used : DataOutput(java.io.DataOutput) RecordDescriptor(org.apache.hyracks.api.dataflow.value.RecordDescriptor) IFileSplitProvider(org.apache.hyracks.dataflow.std.file.IFileSplitProvider) ConstantFileSplitProvider(org.apache.hyracks.dataflow.std.file.ConstantFileSplitProvider) BTreeSearchOperatorDescriptor(org.apache.hyracks.storage.am.btree.dataflow.BTreeSearchOperatorDescriptor) ArrayTupleBuilder(org.apache.hyracks.dataflow.common.comm.io.ArrayTupleBuilder) OneToOneConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.OneToOneConnectorDescriptor) ISerializerDeserializer(org.apache.hyracks.api.dataflow.value.ISerializerDeserializer) ConstantTupleSourceOperatorDescriptor(org.apache.hyracks.dataflow.std.misc.ConstantTupleSourceOperatorDescriptor) IOperatorDescriptor(org.apache.hyracks.api.dataflow.IOperatorDescriptor) PlainFileWriterOperatorDescriptor(org.apache.hyracks.dataflow.std.file.PlainFileWriterOperatorDescriptor) RTreeSearchOperatorDescriptor(org.apache.hyracks.storage.am.rtree.dataflow.RTreeSearchOperatorDescriptor) JobSpecification(org.apache.hyracks.api.job.JobSpecification) Test(org.junit.Test)

Aggregations

RecordDescriptor (org.apache.hyracks.api.dataflow.value.RecordDescriptor)169 JobSpecification (org.apache.hyracks.api.job.JobSpecification)90 UTF8StringSerializerDeserializer (org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer)74 OneToOneConnectorDescriptor (org.apache.hyracks.dataflow.std.connectors.OneToOneConnectorDescriptor)72 Test (org.junit.Test)69 IFileSplitProvider (org.apache.hyracks.dataflow.std.file.IFileSplitProvider)64 FileScanOperatorDescriptor (org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor)55 ConstantFileSplitProvider (org.apache.hyracks.dataflow.std.file.ConstantFileSplitProvider)53 IOperatorDescriptor (org.apache.hyracks.api.dataflow.IOperatorDescriptor)52 DelimitedDataTupleParserFactory (org.apache.hyracks.dataflow.std.file.DelimitedDataTupleParserFactory)41 IConnectorDescriptor (org.apache.hyracks.api.dataflow.IConnectorDescriptor)40 FileSplit (org.apache.hyracks.api.io.FileSplit)38 ManagedFileSplit (org.apache.hyracks.api.io.ManagedFileSplit)37 ISerializerDeserializer (org.apache.hyracks.api.dataflow.value.ISerializerDeserializer)36 IValueParserFactory (org.apache.hyracks.dataflow.common.data.parsers.IValueParserFactory)35 FieldHashPartitionComputerFactory (org.apache.hyracks.dataflow.common.data.partition.FieldHashPartitionComputerFactory)35 ILogicalOperator (org.apache.hyracks.algebricks.core.algebra.base.ILogicalOperator)33 IBinaryHashFunctionFactory (org.apache.hyracks.api.dataflow.value.IBinaryHashFunctionFactory)31 MToNPartitioningConnectorDescriptor (org.apache.hyracks.dataflow.std.connectors.MToNPartitioningConnectorDescriptor)27 IBinaryComparatorFactory (org.apache.hyracks.api.dataflow.value.IBinaryComparatorFactory)25