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

Example 16 with ConstantFileSplitProvider

use of org.apache.hyracks.dataflow.std.file.ConstantFileSplitProvider 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 17 with ConstantFileSplitProvider

use of org.apache.hyracks.dataflow.std.file.ConstantFileSplitProvider 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 18 with ConstantFileSplitProvider

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

Example 19 with ConstantFileSplitProvider

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

Example 20 with ConstantFileSplitProvider

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

the class AbstractRTreeOperatorTest method setup.

@Before
public void setup() throws Exception {
    testHelper = createTestHelper();
    primaryFileName = testHelper.getPrimaryIndexName();
    primarySplitProvider = new ConstantFileSplitProvider(new FileSplit[] { new ManagedFileSplit(NC1_ID, primaryFileName) });
    primaryHelperFactory = new IndexDataflowHelperFactory(storageManager, primarySplitProvider);
    secondaryFileName = testHelper.getSecondaryIndexName();
    secondarySplitProvider = new ConstantFileSplitProvider(new FileSplit[] { new ManagedFileSplit(NC1_ID, secondaryFileName) });
    secondaryHelperFactory = new IndexDataflowHelperFactory(storageManager, secondarySplitProvider);
    // field, type and key declarations for primary index
    primaryTypeTraits[0] = UTF8StringPointable.TYPE_TRAITS;
    primaryTypeTraits[1] = UTF8StringPointable.TYPE_TRAITS;
    primaryTypeTraits[2] = UTF8StringPointable.TYPE_TRAITS;
    primaryTypeTraits[3] = UTF8StringPointable.TYPE_TRAITS;
    primaryTypeTraits[4] = UTF8StringPointable.TYPE_TRAITS;
    primaryTypeTraits[5] = UTF8StringPointable.TYPE_TRAITS;
    primaryTypeTraits[6] = UTF8StringPointable.TYPE_TRAITS;
    primaryTypeTraits[7] = UTF8StringPointable.TYPE_TRAITS;
    primaryTypeTraits[8] = UTF8StringPointable.TYPE_TRAITS;
    primaryTypeTraits[9] = UTF8StringPointable.TYPE_TRAITS;
    primaryComparatorFactories[0] = PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY);
    // field, type and key declarations for secondary indexes
    secondaryTypeTraits[0] = DoublePointable.TYPE_TRAITS;
    secondaryTypeTraits[1] = DoublePointable.TYPE_TRAITS;
    secondaryTypeTraits[2] = DoublePointable.TYPE_TRAITS;
    secondaryTypeTraits[3] = DoublePointable.TYPE_TRAITS;
    secondaryTypeTraits[4] = UTF8StringPointable.TYPE_TRAITS;
    secondaryComparatorFactories[0] = PointableBinaryComparatorFactory.of(DoublePointable.FACTORY);
    secondaryComparatorFactories[1] = PointableBinaryComparatorFactory.of(DoublePointable.FACTORY);
    secondaryComparatorFactories[2] = PointableBinaryComparatorFactory.of(DoublePointable.FACTORY);
    secondaryComparatorFactories[3] = PointableBinaryComparatorFactory.of(DoublePointable.FACTORY);
    // This only used for LSMRTree
    int[] rtreeFields = null;
    int[] filterFields = null;
    ITypeTraits[] filterTypes = null;
    IBinaryComparatorFactory[] filterCmpFactories = null;
    if (rTreeType == RTreeType.LSMRTREE || rTreeType == RTreeType.LSMRTREE_WITH_ANTIMATTER) {
        rtreeFields = new int[] { 0, 1, 2, 3, 4 };
        filterFields = new int[] { 4 };
        filterTypes = new ITypeTraits[] { UTF8StringPointable.TYPE_TRAITS };
        filterCmpFactories = new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) };
    }
    int[] btreeFields = null;
    if (rTreeType == RTreeType.LSMRTREE) {
        btreeKeyFieldCount = 1;
        btreeComparatorFactories = new IBinaryComparatorFactory[btreeKeyFieldCount];
        btreeComparatorFactories[0] = PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY);
        btreeFields = new int[] { 4 };
    } else {
        btreeComparatorFactories[0] = PointableBinaryComparatorFactory.of(DoublePointable.FACTORY);
        btreeComparatorFactories[1] = PointableBinaryComparatorFactory.of(DoublePointable.FACTORY);
        btreeComparatorFactories[2] = PointableBinaryComparatorFactory.of(DoublePointable.FACTORY);
        btreeComparatorFactories[3] = PointableBinaryComparatorFactory.of(DoublePointable.FACTORY);
        btreeComparatorFactories[4] = PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY);
    }
    IPrimitiveValueProviderFactory[] secondaryValueProviderFactories = RTreeUtils.createPrimitiveValueProviderFactories(secondaryComparatorFactories.length, DoublePointable.FACTORY);
    rtreeFactory = createSecondaryResourceFactory(secondaryValueProviderFactories, RTreePolicyType.RSTARTREE, btreeComparatorFactories, LSMRTreeUtils.proposeBestLinearizer(secondaryTypeTraits, secondaryComparatorFactories.length), btreeFields);
}