Examples with IFileSplitProvider org.apache.hyracks.dataflow.std.file.IFileSplitProvider used on opensource projects

Example 31 with IFileSplitProvider

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

the class BTreeSecondaryIndexSearchOperatorTest method searchSecondaryIndexTest.

@Test
public void searchSecondaryIndexTest() throws Exception {
    JobSpecification spec = new JobSpecification();
    // build tuple containing search keys (only use the first key as search
    // key)
    ArrayTupleBuilder tb = new ArrayTupleBuilder(DataSetConstants.secondaryKeyFieldCount);
    DataOutput dos = tb.getDataOutput();
    tb.reset();
    // low key
    new UTF8StringSerializerDeserializer().serialize("1998-07-21", dos);
    tb.addFieldEndOffset();
    // high key
    new UTF8StringSerializerDeserializer().serialize("2000-10-18", dos);
    tb.addFieldEndOffset();
    ISerializerDeserializer[] keyRecDescSers = { new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer() };
    RecordDescriptor keyRecDesc = new RecordDescriptor(keyRecDescSers);
    ConstantTupleSourceOperatorDescriptor keyProviderOp = new ConstantTupleSourceOperatorDescriptor(spec, keyRecDesc, tb.getFieldEndOffsets(), tb.getByteArray(), tb.getSize());
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, keyProviderOp, NC1_ID);
    int[] secondaryLowKeyFields = { 0 };
    int[] secondaryHighKeyFields = { 1 };
    // search secondary index
    BTreeSearchOperatorDescriptor secondaryBtreeSearchOp = new BTreeSearchOperatorDescriptor(spec, DataSetConstants.secondaryRecDesc, secondaryLowKeyFields, secondaryHighKeyFields, true, true, secondaryHelperFactory, false, false, null, NoOpOperationCallbackFactory.INSTANCE, null, null, false);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, secondaryBtreeSearchOp, NC1_ID);
    // second field from the tuples
    int[] primaryLowKeyFields = { 1 };
    // coming from secondary index
    // second field from the tuples
    int[] primaryHighKeyFields = { 1 };
    // coming from secondary index
    // search primary index
    BTreeSearchOperatorDescriptor primaryBtreeSearchOp = new BTreeSearchOperatorDescriptor(spec, DataSetConstants.primaryRecDesc, primaryLowKeyFields, primaryHighKeyFields, true, true, primaryHelperFactory, false, false, null, NoOpOperationCallbackFactory.INSTANCE, null, null, false);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, primaryBtreeSearchOp, 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, secondaryBtreeSearchOp, 0);
    spec.connect(new OneToOneConnectorDescriptor(spec), secondaryBtreeSearchOp, 0, primaryBtreeSearchOp, 0);
    spec.connect(new OneToOneConnectorDescriptor(spec), primaryBtreeSearchOp, 0, printer, 0);
    spec.addRoot(printer);
    runTest(spec);
}

Example 32 with IFileSplitProvider

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

the class BTreeSecondaryIndexUpsertOperatorTest method searchUpdatedSecondaryIndexTest.

@Test
public void searchUpdatedSecondaryIndexTest() throws Exception {
    JobSpecification spec = new JobSpecification();
    // build tuple containing search keys (only use the first key as search
    // key)
    ArrayTupleBuilder tb = new ArrayTupleBuilder(DataSetConstants.secondaryKeyFieldCount);
    DataOutput dos = tb.getDataOutput();
    tb.reset();
    // low key
    new UTF8StringSerializerDeserializer().serialize("1998-07-21", dos);
    tb.addFieldEndOffset();
    // high key
    new UTF8StringSerializerDeserializer().serialize("2000-10-18", dos);
    tb.addFieldEndOffset();
    ISerializerDeserializer[] keyRecDescSers = { new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer() };
    RecordDescriptor keyRecDesc = new RecordDescriptor(keyRecDescSers);
    ConstantTupleSourceOperatorDescriptor keyProviderOp = new ConstantTupleSourceOperatorDescriptor(spec, keyRecDesc, tb.getFieldEndOffsets(), tb.getByteArray(), tb.getSize());
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, keyProviderOp, NC1_ID);
    int[] secondaryLowKeyFields = { 0 };
    int[] secondaryHighKeyFields = { 1 };
    // search secondary index
    BTreeSearchOperatorDescriptor secondaryBtreeSearchOp = new BTreeSearchOperatorDescriptor(spec, DataSetConstants.secondaryRecDesc, secondaryLowKeyFields, secondaryHighKeyFields, true, true, secondaryHelperFactory, false, false, null, NoOpOperationCallbackFactory.INSTANCE, null, null, false);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, secondaryBtreeSearchOp, NC1_ID);
    // second field from the tuples coming from secondary index
    int[] primaryLowKeyFields = { 1 };
    // second field from the tuples coming from secondary index
    int[] primaryHighKeyFields = { 1 };
    // search primary index
    BTreeSearchOperatorDescriptor primaryBtreeSearchOp = new BTreeSearchOperatorDescriptor(spec, DataSetConstants.primaryRecDesc, primaryLowKeyFields, primaryHighKeyFields, true, true, primaryHelperFactory, false, false, null, NoOpOperationCallbackFactory.INSTANCE, null, null, false);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, primaryBtreeSearchOp, 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, secondaryBtreeSearchOp, 0);
    spec.connect(new OneToOneConnectorDescriptor(spec), secondaryBtreeSearchOp, 0, primaryBtreeSearchOp, 0);
    spec.connect(new OneToOneConnectorDescriptor(spec), primaryBtreeSearchOp, 0, printer, 0);
    spec.addRoot(printer);
    runTest(spec);
}

Example 33 with IFileSplitProvider

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

the class LSMBTreePrimaryIndexSearchOperatorTest method shouldWriteFilterValueIfAppendFilterIsTrue.

@Test
public void shouldWriteFilterValueIfAppendFilterIsTrue() throws Exception {
    JobSpecification spec = new JobSpecification();
    // build tuple containing low and high search key
    // high key and low key
    ArrayTupleBuilder tb = new ArrayTupleBuilder(DataSetConstants.primaryKeyFieldCount * 2);
    DataOutput dos = tb.getDataOutput();
    tb.reset();
    // low key
    new UTF8StringSerializerDeserializer().serialize("100", dos);
    tb.addFieldEndOffset();
    // high key
    new UTF8StringSerializerDeserializer().serialize("200", dos);
    tb.addFieldEndOffset();
    ISerializerDeserializer[] keyRecDescSers = { new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer() };
    RecordDescriptor keyRecDesc = new RecordDescriptor(keyRecDescSers);
    ConstantTupleSourceOperatorDescriptor keyProviderOp = new ConstantTupleSourceOperatorDescriptor(spec, keyRecDesc, tb.getFieldEndOffsets(), tb.getByteArray(), tb.getSize());
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, keyProviderOp, NC1_ID);
    int[] lowKeyFields = { 0 };
    int[] highKeyFields = { 1 };
    BTreeSearchOperatorDescriptor primaryBtreeSearchOp = new BTreeSearchOperatorDescriptor(spec, DataSetConstants.primaryAndFilterRecDesc, lowKeyFields, highKeyFields, true, true, primaryHelperFactory, false, false, NoopMissingWriterFactory.INSTANCE, NoOpOperationCallbackFactory.INSTANCE, null, null, true);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, primaryBtreeSearchOp, 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, primaryBtreeSearchOp, 0);
    spec.connect(new OneToOneConnectorDescriptor(spec), primaryBtreeSearchOp, 0, printer, 0);
    spec.addRoot(printer);
    runTest(spec);
}

Example 34 with IFileSplitProvider

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

the class LSMBTreeSecondaryIndexSearchOperatorTest method shouldCarryFilterValueToPrimaryIndexSearch.

@Test
public void shouldCarryFilterValueToPrimaryIndexSearch() throws Exception {
    JobSpecification spec = new JobSpecification();
    // build tuple containing search keys (only use the first key as search
    // key)
    ArrayTupleBuilder tb = new ArrayTupleBuilder(DataSetConstants.secondaryKeyFieldCount);
    DataOutput dos = tb.getDataOutput();
    tb.reset();
    // low key
    new UTF8StringSerializerDeserializer().serialize("1998-07-21", dos);
    tb.addFieldEndOffset();
    // high key
    new UTF8StringSerializerDeserializer().serialize("2000-10-18", dos);
    tb.addFieldEndOffset();
    ISerializerDeserializer[] keyRecDescSers = { new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer() };
    RecordDescriptor keyRecDesc = new RecordDescriptor(keyRecDescSers);
    ConstantTupleSourceOperatorDescriptor keyProviderOp = new ConstantTupleSourceOperatorDescriptor(spec, keyRecDesc, tb.getFieldEndOffsets(), tb.getByteArray(), tb.getSize());
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, keyProviderOp, NC1_ID);
    int[] secondaryLowKeyFields = { 0 };
    int[] secondaryHighKeyFields = { 1 };
    // search secondary index
    BTreeSearchOperatorDescriptor secondaryBtreeSearchOp = new BTreeSearchOperatorDescriptor(spec, DataSetConstants.secondaryWithFilterRecDesc, secondaryLowKeyFields, secondaryHighKeyFields, true, true, secondaryHelperFactory, false, false, NoopMissingWriterFactory.INSTANCE, NoOpOperationCallbackFactory.INSTANCE, null, null, true);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, secondaryBtreeSearchOp, NC1_ID);
    // second field from the tuples
    int[] primaryLowKeyFields = { 1 };
    // second field from the tuples
    int[] primaryHighKeyFields = { 1 };
    int[] minFilterFields = { 2 };
    int[] maxFilterFields = { 3 };
    // search primary index
    BTreeSearchOperatorDescriptor primaryBtreeSearchOp = new BTreeSearchOperatorDescriptor(spec, DataSetConstants.primaryRecDesc, primaryLowKeyFields, primaryHighKeyFields, true, true, primaryHelperFactory, false, false, NoopMissingWriterFactory.INSTANCE, NoOpOperationCallbackFactory.INSTANCE, minFilterFields, maxFilterFields, false);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, primaryBtreeSearchOp, 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, secondaryBtreeSearchOp, 0);
    spec.connect(new OneToOneConnectorDescriptor(spec), secondaryBtreeSearchOp, 0, primaryBtreeSearchOp, 0);
    spec.connect(new OneToOneConnectorDescriptor(spec), primaryBtreeSearchOp, 0, printer, 0);
    spec.addRoot(printer);
    runTest(spec);
}

Example 35 with IFileSplitProvider

use of org.apache.hyracks.dataflow.std.file.IFileSplitProvider 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);
}