Search in sources :

Example 6 with FileScanOperatorDescriptor

use of org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor in project asterixdb by apache.

the class HeapSortMergeTest method optimizedSortMergeTest01.

@Test
public void optimizedSortMergeTest01() throws Exception {
    JobSpecification spec = new JobSpecification();
    FileSplit[] ordersSplits = new FileSplit[] { new ManagedFileSplit(NC1_ID, "data" + File.separator + "tpch0.001" + File.separator + "orders-part1.tbl"), new ManagedFileSplit(NC2_ID, "data" + File.separator + "tpch0.001" + File.separator + "orders-part2.tbl") };
    IFileSplitProvider ordersSplitProvider = new ConstantFileSplitProvider(ordersSplits);
    RecordDescriptor ordersDesc = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer() });
    FileScanOperatorDescriptor ordScanner = new FileScanOperatorDescriptor(spec, ordersSplitProvider, new DelimitedDataTupleParserFactory(new IValueParserFactory[] { UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE }, '|'), ordersDesc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, ordScanner, NC1_ID, NC2_ID);
    // larger than the total record numbers.
    int outputLimit = 5;
    TopKSorterOperatorDescriptor sorter = new TopKSorterOperatorDescriptor(spec, 4, outputLimit, new int[] { 1, 0 }, null, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY), PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, ordersDesc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, sorter, NC1_ID, NC2_ID);
    ResultSetId rsId = new ResultSetId(1);
    spec.addResultSetId(rsId);
    FileSplit fs = createFile(nc1);
    IFileSplitProvider outputSplitProvider = new ConstantFileSplitProvider(new FileSplit[] { fs });
    IOperatorDescriptor printer = new PlainFileWriterOperatorDescriptor(spec, outputSplitProvider, "|");
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, printer, NC1_ID);
    spec.connect(new OneToOneConnectorDescriptor(spec), ordScanner, 0, sorter, 0);
    spec.connect(new MToNPartitioningMergingConnectorDescriptor(spec, new FieldHashPartitionComputerFactory(new int[] { 1, 0 }, new IBinaryHashFunctionFactory[] { PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY), PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY) }), new int[] { 1, 0 }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY), PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, new UTF8StringNormalizedKeyComputerFactory()), sorter, 0, printer, 0);
    runTest(spec);
    System.out.println("Result write into :" + fs.getPath() + " in node: " + fs.getNodeName());
}
Also used : UTF8StringNormalizedKeyComputerFactory(org.apache.hyracks.dataflow.common.data.normalizers.UTF8StringNormalizedKeyComputerFactory) 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) MToNPartitioningMergingConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.MToNPartitioningMergingConnectorDescriptor) ConstantFileSplitProvider(org.apache.hyracks.dataflow.std.file.ConstantFileSplitProvider) IBinaryComparatorFactory(org.apache.hyracks.api.dataflow.value.IBinaryComparatorFactory) 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) TopKSorterOperatorDescriptor(org.apache.hyracks.dataflow.std.sort.TopKSorterOperatorDescriptor) ManagedFileSplit(org.apache.hyracks.api.io.ManagedFileSplit) ResultSetId(org.apache.hyracks.api.dataset.ResultSetId) IOperatorDescriptor(org.apache.hyracks.api.dataflow.IOperatorDescriptor) PlainFileWriterOperatorDescriptor(org.apache.hyracks.dataflow.std.file.PlainFileWriterOperatorDescriptor) FileScanOperatorDescriptor(org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor) JobSpecification(org.apache.hyracks.api.job.JobSpecification) Test(org.junit.Test)

Example 7 with FileScanOperatorDescriptor

use of org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor in project asterixdb by apache.

the class HeapSortMergeTest method createSortMergeJobSpec.

public static JobSpecification createSortMergeJobSpec() throws Exception {
    JobSpecification spec = new JobSpecification();
    FileSplit[] ordersSplits = new FileSplit[] { new ManagedFileSplit(NC1_ID, "data" + File.separator + "tpch0.001" + File.separator + "orders-part1.tbl"), new ManagedFileSplit(NC2_ID, "data" + File.separator + "tpch0.001" + File.separator + "orders-part2.tbl") };
    IFileSplitProvider ordersSplitProvider = new ConstantFileSplitProvider(ordersSplits);
    RecordDescriptor ordersDesc = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer() });
    FileScanOperatorDescriptor ordScanner = new FileScanOperatorDescriptor(spec, ordersSplitProvider, new DelimitedDataTupleParserFactory(new IValueParserFactory[] { UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE }, '|'), ordersDesc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, ordScanner, NC1_ID, NC2_ID);
    int outputLimit = 20;
    TopKSorterOperatorDescriptor sorter = new TopKSorterOperatorDescriptor(spec, 4, outputLimit, new int[] { 1, 0 }, null, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY), PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, ordersDesc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, sorter, NC1_ID, NC2_ID);
    LimitOperatorDescriptor filter = new LimitOperatorDescriptor(spec, ordersDesc, outputLimit);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, filter, NC1_ID);
    ResultSetId rsId = new ResultSetId(1);
    spec.addResultSetId(rsId);
    IOperatorDescriptor printer = new ResultWriterOperatorDescriptor(spec, rsId, false, false, ResultSerializerFactoryProvider.INSTANCE.getResultSerializerFactoryProvider());
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, printer, NC1_ID);
    spec.connect(new OneToOneConnectorDescriptor(spec), ordScanner, 0, sorter, 0);
    spec.connect(new MToNPartitioningMergingConnectorDescriptor(spec, new FieldHashPartitionComputerFactory(new int[] { 1, 0 }, new IBinaryHashFunctionFactory[] { PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY), PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY) }), new int[] { 1, 0 }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY), PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, new UTF8StringNormalizedKeyComputerFactory()), sorter, 0, filter, 0);
    spec.connect(new OneToOneConnectorDescriptor(spec), filter, 0, printer, 0);
    return spec;
}
Also used : UTF8StringNormalizedKeyComputerFactory(org.apache.hyracks.dataflow.common.data.normalizers.UTF8StringNormalizedKeyComputerFactory) 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) MToNPartitioningMergingConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.MToNPartitioningMergingConnectorDescriptor) ConstantFileSplitProvider(org.apache.hyracks.dataflow.std.file.ConstantFileSplitProvider) IBinaryComparatorFactory(org.apache.hyracks.api.dataflow.value.IBinaryComparatorFactory) 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) ResultWriterOperatorDescriptor(org.apache.hyracks.dataflow.std.result.ResultWriterOperatorDescriptor) TopKSorterOperatorDescriptor(org.apache.hyracks.dataflow.std.sort.TopKSorterOperatorDescriptor) LimitOperatorDescriptor(org.apache.hyracks.dataflow.std.misc.LimitOperatorDescriptor) ManagedFileSplit(org.apache.hyracks.api.io.ManagedFileSplit) ResultSetId(org.apache.hyracks.api.dataset.ResultSetId) IOperatorDescriptor(org.apache.hyracks.api.dataflow.IOperatorDescriptor) FileScanOperatorDescriptor(org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor) JobSpecification(org.apache.hyracks.api.job.JobSpecification)

Example 8 with FileScanOperatorDescriptor

use of org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor in project asterixdb by apache.

the class AggregationTest method multiKeyAvgExtGroupTest.

@Test
public void multiKeyAvgExtGroupTest() 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, IntegerSerializerDeserializer.INSTANCE, FloatSerializerDeserializer.INSTANCE });
    int[] keyFields = new int[] { 8, 0 };
    int frameLimits = 5;
    int tableSize = 8;
    long fileSize = frameLimits * spec.getFrameSize();
    ExternalGroupOperatorDescriptor grouper = new ExternalGroupOperatorDescriptor(spec, tableSize, fileSize, keyFields, frameLimits, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY), PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, new UTF8StringNormalizedKeyComputerFactory(), new MultiFieldsAggregatorFactory(new IFieldAggregateDescriptorFactory[] { new IntSumFieldAggregatorFactory(1, false), new CountFieldAggregatorFactory(false), new AvgFieldGroupAggregatorFactory(1, false) }), new MultiFieldsAggregatorFactory(new IFieldAggregateDescriptorFactory[] { new IntSumFieldAggregatorFactory(2, false), new IntSumFieldAggregatorFactory(3, false), new AvgFieldMergeAggregatorFactory(4, false) }), outputRec, outputRec, new HashSpillableTableFactory(new IBinaryHashFunctionFamily[] { UTF8StringBinaryHashFunctionFamily.INSTANCE, UTF8StringBinaryHashFunctionFamily.INSTANCE }));
    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, "multiKeyAvgExtGroupTest");
    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 : UTF8StringNormalizedKeyComputerFactory(org.apache.hyracks.dataflow.common.data.normalizers.UTF8StringNormalizedKeyComputerFactory) IConnectorDescriptor(org.apache.hyracks.api.dataflow.IConnectorDescriptor) MultiFieldsAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.MultiFieldsAggregatorFactory) AbstractSingleActivityOperatorDescriptor(org.apache.hyracks.dataflow.std.base.AbstractSingleActivityOperatorDescriptor) AvgFieldMergeAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.AvgFieldMergeAggregatorFactory) ExternalGroupOperatorDescriptor(org.apache.hyracks.dataflow.std.group.external.ExternalGroupOperatorDescriptor) RecordDescriptor(org.apache.hyracks.api.dataflow.value.RecordDescriptor) HashSpillableTableFactory(org.apache.hyracks.dataflow.std.group.HashSpillableTableFactory) 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) IntSumFieldAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.IntSumFieldAggregatorFactory) JobSpecification(org.apache.hyracks.api.job.JobSpecification) IFieldAggregateDescriptorFactory(org.apache.hyracks.dataflow.std.group.IFieldAggregateDescriptorFactory) IBinaryHashFunctionFamily(org.apache.hyracks.api.dataflow.value.IBinaryHashFunctionFamily) Test(org.junit.Test)

Example 9 with FileScanOperatorDescriptor

use of org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor in project asterixdb by apache.

the class CountOfCountsTest method countOfCountsMultiNC.

@Test
public void countOfCountsMultiNC() throws Exception {
    JobSpecification spec = new JobSpecification();
    FileSplit[] splits = new FileSplit[] { new ManagedFileSplit(NC2_ID, "data" + File.separator + "words.txt") };
    IFileSplitProvider splitProvider = new ConstantFileSplitProvider(splits);
    RecordDescriptor desc = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer() });
    FileScanOperatorDescriptor csvScanner = new FileScanOperatorDescriptor(spec, splitProvider, new DelimitedDataTupleParserFactory(new IValueParserFactory[] { UTF8StringParserFactory.INSTANCE }, ','), desc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, csvScanner, NC2_ID);
    InMemorySortOperatorDescriptor sorter = new InMemorySortOperatorDescriptor(spec, new int[] { 0 }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, desc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, sorter, NC1_ID, NC2_ID, NC1_ID, NC2_ID);
    RecordDescriptor desc2 = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer(), IntegerSerializerDeserializer.INSTANCE });
    PreclusteredGroupOperatorDescriptor group = new PreclusteredGroupOperatorDescriptor(spec, new int[] { 0 }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, new MultiFieldsAggregatorFactory(new IFieldAggregateDescriptorFactory[] { new CountFieldAggregatorFactory(true) }), desc2);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, group, NC1_ID, NC2_ID, NC1_ID, NC2_ID);
    InMemorySortOperatorDescriptor sorter2 = new InMemorySortOperatorDescriptor(spec, new int[] { 1 }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(IntegerPointable.FACTORY) }, desc2);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, sorter2, NC1_ID, NC2_ID);
    RecordDescriptor desc3 = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE, IntegerSerializerDeserializer.INSTANCE });
    PreclusteredGroupOperatorDescriptor group2 = new PreclusteredGroupOperatorDescriptor(spec, new int[] { 1 }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(IntegerPointable.FACTORY) }, new MultiFieldsAggregatorFactory(new IFieldAggregateDescriptorFactory[] { new CountFieldAggregatorFactory(true) }), desc3);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, group2, NC1_ID, NC2_ID);
    ResultSetId rsId = new ResultSetId(1);
    IOperatorDescriptor printer = new ResultWriterOperatorDescriptor(spec, rsId, true, false, ResultSerializerFactoryProvider.INSTANCE.getResultSerializerFactoryProvider());
    spec.addResultSetId(rsId);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, printer, NC1_ID);
    IConnectorDescriptor conn1 = new MToNPartitioningConnectorDescriptor(spec, new FieldHashPartitionComputerFactory(new int[] { 0 }, new IBinaryHashFunctionFactory[] { PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY) }));
    spec.connect(conn1, csvScanner, 0, sorter, 0);
    IConnectorDescriptor conn2 = new OneToOneConnectorDescriptor(spec);
    spec.connect(conn2, sorter, 0, group, 0);
    IConnectorDescriptor conn3 = new MToNPartitioningConnectorDescriptor(spec, new FieldHashPartitionComputerFactory(new int[] { 1 }, new IBinaryHashFunctionFactory[] { PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY) }));
    spec.connect(conn3, group, 0, sorter2, 0);
    IConnectorDescriptor conn4 = new OneToOneConnectorDescriptor(spec);
    spec.connect(conn4, sorter2, 0, group2, 0);
    IConnectorDescriptor conn5 = new MToNBroadcastConnectorDescriptor(spec);
    spec.connect(conn5, group2, 0, printer, 0);
    spec.addRoot(printer);
    runTest(spec);
}
Also used : IFileSplitProvider(org.apache.hyracks.dataflow.std.file.IFileSplitProvider) RecordDescriptor(org.apache.hyracks.api.dataflow.value.RecordDescriptor) CountFieldAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.CountFieldAggregatorFactory) 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) IBinaryHashFunctionFactory(org.apache.hyracks.api.dataflow.value.IBinaryHashFunctionFactory) ResultWriterOperatorDescriptor(org.apache.hyracks.dataflow.std.result.ResultWriterOperatorDescriptor) MToNBroadcastConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.MToNBroadcastConnectorDescriptor) ManagedFileSplit(org.apache.hyracks.api.io.ManagedFileSplit) ResultSetId(org.apache.hyracks.api.dataset.ResultSetId) FileScanOperatorDescriptor(org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor) JobSpecification(org.apache.hyracks.api.job.JobSpecification) IFieldAggregateDescriptorFactory(org.apache.hyracks.dataflow.std.group.IFieldAggregateDescriptorFactory) IConnectorDescriptor(org.apache.hyracks.api.dataflow.IConnectorDescriptor) MultiFieldsAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.MultiFieldsAggregatorFactory) IValueParserFactory(org.apache.hyracks.dataflow.common.data.parsers.IValueParserFactory) InMemorySortOperatorDescriptor(org.apache.hyracks.dataflow.std.sort.InMemorySortOperatorDescriptor) ConstantFileSplitProvider(org.apache.hyracks.dataflow.std.file.ConstantFileSplitProvider) MToNPartitioningConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.MToNPartitioningConnectorDescriptor) DelimitedDataTupleParserFactory(org.apache.hyracks.dataflow.std.file.DelimitedDataTupleParserFactory) FieldHashPartitionComputerFactory(org.apache.hyracks.dataflow.common.data.partition.FieldHashPartitionComputerFactory) IOperatorDescriptor(org.apache.hyracks.api.dataflow.IOperatorDescriptor) PreclusteredGroupOperatorDescriptor(org.apache.hyracks.dataflow.std.group.preclustered.PreclusteredGroupOperatorDescriptor) Test(org.junit.Test)

Example 10 with FileScanOperatorDescriptor

use of org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor in project asterixdb by apache.

the class CountOfCountsTest method countOfCountsExternalSortMultiNC.

@Test
public void countOfCountsExternalSortMultiNC() throws Exception {
    JobSpecification spec = new JobSpecification();
    FileSplit[] splits = new FileSplit[] { new ManagedFileSplit(NC2_ID, "data" + File.separator + "words.txt") };
    IFileSplitProvider splitProvider = new ConstantFileSplitProvider(splits);
    RecordDescriptor desc = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer() });
    FileScanOperatorDescriptor csvScanner = new FileScanOperatorDescriptor(spec, splitProvider, new DelimitedDataTupleParserFactory(new IValueParserFactory[] { UTF8StringParserFactory.INSTANCE }, ','), desc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, csvScanner, NC2_ID);
    ExternalSortOperatorDescriptor sorter = new ExternalSortOperatorDescriptor(spec, 3, new int[] { 0 }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, desc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, sorter, NC1_ID, NC2_ID, NC1_ID, NC2_ID);
    RecordDescriptor desc2 = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer(), IntegerSerializerDeserializer.INSTANCE });
    PreclusteredGroupOperatorDescriptor group = new PreclusteredGroupOperatorDescriptor(spec, new int[] { 0 }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, new MultiFieldsAggregatorFactory(new IFieldAggregateDescriptorFactory[] { new CountFieldAggregatorFactory(true) }), desc2);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, group, NC1_ID, NC2_ID, NC1_ID, NC2_ID);
    InMemorySortOperatorDescriptor sorter2 = new InMemorySortOperatorDescriptor(spec, new int[] { 1 }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(IntegerPointable.FACTORY) }, desc2);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, sorter2, NC1_ID, NC2_ID);
    RecordDescriptor desc3 = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE, IntegerSerializerDeserializer.INSTANCE });
    PreclusteredGroupOperatorDescriptor group2 = new PreclusteredGroupOperatorDescriptor(spec, new int[] { 1 }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(IntegerPointable.FACTORY) }, new MultiFieldsAggregatorFactory(new IFieldAggregateDescriptorFactory[] { new CountFieldAggregatorFactory(true) }), desc3);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, group2, NC1_ID, NC2_ID);
    ResultSetId rsId = new ResultSetId(1);
    IOperatorDescriptor printer = new ResultWriterOperatorDescriptor(spec, rsId, true, false, ResultSerializerFactoryProvider.INSTANCE.getResultSerializerFactoryProvider());
    spec.addResultSetId(rsId);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, printer, NC1_ID);
    IConnectorDescriptor conn1 = new MToNPartitioningConnectorDescriptor(spec, new FieldHashPartitionComputerFactory(new int[] { 0 }, new IBinaryHashFunctionFactory[] { PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY) }));
    spec.connect(conn1, csvScanner, 0, sorter, 0);
    IConnectorDescriptor conn2 = new OneToOneConnectorDescriptor(spec);
    spec.connect(conn2, sorter, 0, group, 0);
    IConnectorDescriptor conn3 = new MToNPartitioningConnectorDescriptor(spec, new FieldHashPartitionComputerFactory(new int[] { 1 }, new IBinaryHashFunctionFactory[] { PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY) }));
    spec.connect(conn3, group, 0, sorter2, 0);
    IConnectorDescriptor conn4 = new OneToOneConnectorDescriptor(spec);
    spec.connect(conn4, sorter2, 0, group2, 0);
    IConnectorDescriptor conn5 = new MToNBroadcastConnectorDescriptor(spec);
    spec.connect(conn5, group2, 0, printer, 0);
    spec.addRoot(printer);
    runTest(spec);
}
Also used : IFileSplitProvider(org.apache.hyracks.dataflow.std.file.IFileSplitProvider) RecordDescriptor(org.apache.hyracks.api.dataflow.value.RecordDescriptor) CountFieldAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.CountFieldAggregatorFactory) 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) IBinaryHashFunctionFactory(org.apache.hyracks.api.dataflow.value.IBinaryHashFunctionFactory) ResultWriterOperatorDescriptor(org.apache.hyracks.dataflow.std.result.ResultWriterOperatorDescriptor) MToNBroadcastConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.MToNBroadcastConnectorDescriptor) ManagedFileSplit(org.apache.hyracks.api.io.ManagedFileSplit) ResultSetId(org.apache.hyracks.api.dataset.ResultSetId) FileScanOperatorDescriptor(org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor) JobSpecification(org.apache.hyracks.api.job.JobSpecification) IFieldAggregateDescriptorFactory(org.apache.hyracks.dataflow.std.group.IFieldAggregateDescriptorFactory) IConnectorDescriptor(org.apache.hyracks.api.dataflow.IConnectorDescriptor) MultiFieldsAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.MultiFieldsAggregatorFactory) IValueParserFactory(org.apache.hyracks.dataflow.common.data.parsers.IValueParserFactory) InMemorySortOperatorDescriptor(org.apache.hyracks.dataflow.std.sort.InMemorySortOperatorDescriptor) ConstantFileSplitProvider(org.apache.hyracks.dataflow.std.file.ConstantFileSplitProvider) MToNPartitioningConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.MToNPartitioningConnectorDescriptor) DelimitedDataTupleParserFactory(org.apache.hyracks.dataflow.std.file.DelimitedDataTupleParserFactory) FieldHashPartitionComputerFactory(org.apache.hyracks.dataflow.common.data.partition.FieldHashPartitionComputerFactory) IOperatorDescriptor(org.apache.hyracks.api.dataflow.IOperatorDescriptor) ExternalSortOperatorDescriptor(org.apache.hyracks.dataflow.std.sort.ExternalSortOperatorDescriptor) PreclusteredGroupOperatorDescriptor(org.apache.hyracks.dataflow.std.group.preclustered.PreclusteredGroupOperatorDescriptor) Test(org.junit.Test)

Aggregations

FileScanOperatorDescriptor (org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor)60 JobSpecification (org.apache.hyracks.api.job.JobSpecification)59 RecordDescriptor (org.apache.hyracks.api.dataflow.value.RecordDescriptor)55 OneToOneConnectorDescriptor (org.apache.hyracks.dataflow.std.connectors.OneToOneConnectorDescriptor)54 UTF8StringSerializerDeserializer (org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer)51 Test (org.junit.Test)47 ConstantFileSplitProvider (org.apache.hyracks.dataflow.std.file.ConstantFileSplitProvider)46 DelimitedDataTupleParserFactory (org.apache.hyracks.dataflow.std.file.DelimitedDataTupleParserFactory)45 IFileSplitProvider (org.apache.hyracks.dataflow.std.file.IFileSplitProvider)44 ManagedFileSplit (org.apache.hyracks.api.io.ManagedFileSplit)41 IConnectorDescriptor (org.apache.hyracks.api.dataflow.IConnectorDescriptor)39 FileSplit (org.apache.hyracks.api.io.FileSplit)39 IValueParserFactory (org.apache.hyracks.dataflow.common.data.parsers.IValueParserFactory)35 IOperatorDescriptor (org.apache.hyracks.api.dataflow.IOperatorDescriptor)34 FieldHashPartitionComputerFactory (org.apache.hyracks.dataflow.common.data.partition.FieldHashPartitionComputerFactory)33 IBinaryHashFunctionFactory (org.apache.hyracks.api.dataflow.value.IBinaryHashFunctionFactory)27 MToNPartitioningConnectorDescriptor (org.apache.hyracks.dataflow.std.connectors.MToNPartitioningConnectorDescriptor)25 ResultSetId (org.apache.hyracks.api.dataset.ResultSetId)24 ResultWriterOperatorDescriptor (org.apache.hyracks.dataflow.std.result.ResultWriterOperatorDescriptor)23 MultiFieldsAggregatorFactory (org.apache.hyracks.dataflow.std.group.aggregators.MultiFieldsAggregatorFactory)20