Examples with FileSplit org.apache.hyracks.api.io.FileSplit used on opensource projects

Example 6 with FileSplit

use of org.apache.hyracks.api.io.FileSplit 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 7 with FileSplit

use of org.apache.hyracks.api.io.FileSplit 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 8 with FileSplit

use of org.apache.hyracks.api.io.FileSplit 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);
}

Example 9 with FileSplit

use of org.apache.hyracks.api.io.FileSplit in project asterixdb by apache.

the class PushRuntimeTest method scanSelectWrite.

@Test
public void scanSelectWrite() throws Exception {
    JobSpecification spec = new JobSpecification(FRAME_SIZE);
    // the scanner
    FileSplit[] intFileSplits = new FileSplit[1];
    intFileSplits[0] = new ManagedFileSplit(AlgebricksHyracksIntegrationUtil.NC1_ID, "data" + File.separator + "simple" + File.separator + "int-part1.tbl");
    IFileSplitProvider intSplitProvider = new ConstantFileSplitProvider(intFileSplits);
    RecordDescriptor intScannerDesc = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE });
    IValueParserFactory[] valueParsers = new IValueParserFactory[] { IntegerParserFactory.INSTANCE };
    FileScanOperatorDescriptor intScanner = new FileScanOperatorDescriptor(spec, intSplitProvider, new DelimitedDataTupleParserFactory(valueParsers, '|'), intScannerDesc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, intScanner, DEFAULT_NODES);
    // the algebricks op.
    IScalarEvaluatorFactory cond = new IntegerGreaterThanEvalFactory(new IntegerConstantEvalFactory(2), new TupleFieldEvaluatorFactory(0));
    StreamSelectRuntimeFactory select = new StreamSelectRuntimeFactory(cond, new int[] { 0 }, BinaryBooleanInspectorImpl.FACTORY, false, -1, null);
    RecordDescriptor selectDesc = intScannerDesc;
    String filePath = PATH_ACTUAL + SEPARATOR + "scanSelectWrite.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[] { select, writer }, new RecordDescriptor[] { selectDesc, null });
    PartitionConstraintHelper.addPartitionCountConstraint(spec, algebricksOp, 1);
    spec.connect(new OneToOneConnectorDescriptor(spec), intScanner, 0, algebricksOp, 0);
    spec.addRoot(algebricksOp);
    AlgebricksHyracksIntegrationUtil.runJob(spec);
    StringBuilder buf = new StringBuilder();
    readFileToString(outFile, buf);
    Assert.assertEquals("0", buf.toString());
    outFile.delete();
}

Example 10 with FileSplit

use of org.apache.hyracks.api.io.FileSplit in project asterixdb by apache.

the class PushRuntimeTest method scanSplitWrite.

@Test
public void scanSplitWrite() throws Exception {
    final int outputArity = 2;
    JobSpecification spec = new JobSpecification(FRAME_SIZE);
    String[] inputFileName = { "data" + File.separator + "simple" + File.separator + "int-string-part1.tbl", "data" + File.separator + "simple" + File.separator + "int-string-part1-split-0.tbl", "data" + File.separator + "simple" + File.separator + "int-string-part1-split-1.tbl" };
    File[] inputFiles = new File[inputFileName.length];
    for (int i = 0; i < inputFileName.length; i++) {
        inputFiles[i] = new File(inputFileName[i]);
    }
    File[] outputFile = new File[outputArity];
    FileSplit[] outputFileSplit = new FileSplit[outputArity];
    for (int i = 0; i < outputArity; i++) {
        outputFileSplit[i] = createFile(AlgebricksHyracksIntegrationUtil.nc1);
        outputFile[i] = outputFileSplit[i].getFile(AlgebricksHyracksIntegrationUtil.nc1.getIoManager());
    }
    FileSplit[] inputSplits = new FileSplit[] { new ManagedFileSplit(AlgebricksHyracksIntegrationUtil.NC1_ID, inputFileName[0]) };
    IFileSplitProvider intSplitProvider = new ConstantFileSplitProvider(inputSplits);
    RecordDescriptor scannerDesc = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE, new UTF8StringSerializerDeserializer() });
    IValueParserFactory[] valueParsers = new IValueParserFactory[] { IntegerParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE };
    FileScanOperatorDescriptor intScanner = new FileScanOperatorDescriptor(spec, intSplitProvider, new DelimitedDataTupleParserFactory(valueParsers, '|'), scannerDesc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, intScanner, DEFAULT_NODES);
    SplitOperatorDescriptor splitOp = new SplitOperatorDescriptor(spec, scannerDesc, outputArity, new TupleFieldEvaluatorFactory(0), BinaryIntegerInspectorImpl.FACTORY);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, splitOp, new String[] { AlgebricksHyracksIntegrationUtil.NC1_ID });
    IOperatorDescriptor[] outputOp = new IOperatorDescriptor[outputFile.length];
    for (int i = 0; i < outputArity; i++) {
        outputOp[i] = new LineFileWriteOperatorDescriptor(spec, new FileSplit[] { outputFileSplit[i] });
        PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, outputOp[i], new String[] { AlgebricksHyracksIntegrationUtil.NC1_ID });
    }
    spec.connect(new OneToOneConnectorDescriptor(spec), intScanner, 0, splitOp, 0);
    for (int i = 0; i < outputArity; i++) {
        spec.connect(new OneToOneConnectorDescriptor(spec), splitOp, i, outputOp[i], 0);
    }
    for (int i = 0; i < outputArity; i++) {
        spec.addRoot(outputOp[i]);
    }
    AlgebricksHyracksIntegrationUtil.runJob(spec);
    for (int i = 0; i < outputArity; i++) {
        compareFiles("data" + File.separator + "device0" + File.separator + inputFileName[i + 1], outputFile[i].getAbsolutePath());
    }
}