Examples with UTF8StringSerializerDeserializer org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer used on opensource projects

Example 26 with UTF8StringSerializerDeserializer

use of org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer 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 27 with UTF8StringSerializerDeserializer

use of org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer in project asterixdb by apache.

the class AggregationTest method singleKeyAvgExtGroupTest.

@Test
public void singleKeyAvgExtGroupTest() 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 };
    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) }, new UTF8StringNormalizedKeyComputerFactory(), new MultiFieldsAggregatorFactory(new IFieldAggregateDescriptorFactory[] { new IntSumFieldAggregatorFactory(1, false), new CountFieldAggregatorFactory(false), new AvgFieldGroupAggregatorFactory(1, false) }), new MultiFieldsAggregatorFactory(new IFieldAggregateDescriptorFactory[] { new IntSumFieldAggregatorFactory(1, false), new IntSumFieldAggregatorFactory(2, false), new AvgFieldMergeAggregatorFactory(3, false) }), outputRec, outputRec, new HashSpillableTableFactory(new IBinaryHashFunctionFamily[] { UTF8StringBinaryHashFunctionFamily.INSTANCE }));
    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, "singleKeyAvgExtGroupTest");
    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);
}

Example 28 with UTF8StringSerializerDeserializer

use of org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer in project asterixdb by apache.

the class OrderedIndexExamplesTest method deleteExample.

/**
     * Deletion Example. Create a BTree with one variable-length key field and
     * one variable-length value field. Fill B-tree with random values using
     * insertions, then delete entries one-by-one. Repeat procedure a few times
     * on same BTree.
     */
@Test
public void deleteExample() throws Exception {
    if (LOGGER.isLoggable(Level.INFO)) {
        LOGGER.info("Deletion Example");
    }
    // Declare fields.
    int fieldCount = 2;
    ITypeTraits[] typeTraits = new ITypeTraits[fieldCount];
    typeTraits[0] = UTF8StringPointable.TYPE_TRAITS;
    typeTraits[1] = UTF8StringPointable.TYPE_TRAITS;
    // Declare field serdes.
    ISerializerDeserializer[] fieldSerdes = { new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer() };
    // Declare keys.
    int keyFieldCount = 1;
    IBinaryComparatorFactory[] cmpFactories = new IBinaryComparatorFactory[keyFieldCount];
    cmpFactories[0] = PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY);
    // This is only used for the LSM-BTree.
    int[] bloomFilterKeyFields = new int[keyFieldCount];
    bloomFilterKeyFields[0] = 0;
    ITreeIndex treeIndex = createTreeIndex(typeTraits, cmpFactories, bloomFilterKeyFields, null, null, null, null);
    treeIndex.create();
    treeIndex.activate();
    ArrayTupleBuilder tb = new ArrayTupleBuilder(fieldCount);
    ArrayTupleReference tuple = new ArrayTupleReference();
    IIndexAccessor indexAccessor = treeIndex.createAccessor(TestOperationCallback.INSTANCE, TestOperationCallback.INSTANCE);
    // Max string length to be generated.
    int runs = 3;
    for (int run = 0; run < runs; run++) {
        if (LOGGER.isLoggable(Level.INFO)) {
            LOGGER.info("Deletion example run: " + (run + 1) + "/" + runs);
            LOGGER.info("Inserting into tree...");
        }
        int maxLength = 10;
        int ins = 10000;
        String[] f0s = new String[ins];
        String[] f1s = new String[ins];
        int insDone = 0;
        int[] insDoneCmp = new int[ins];
        for (int i = 0; i < ins; i++) {
            String f0 = randomString(Math.abs(rnd.nextInt()) % maxLength + 1, rnd);
            String f1 = randomString(Math.abs(rnd.nextInt()) % maxLength + 1, rnd);
            TupleUtils.createTuple(tb, tuple, fieldSerdes, f0, f1);
            f0s[i] = f0;
            f1s[i] = f1;
            if (LOGGER.isLoggable(Level.INFO)) {
                if (i % 1000 == 0) {
                    LOGGER.info("Inserting " + i);
                }
            }
            try {
                indexAccessor.insert(tuple);
                insDone++;
            } catch (HyracksDataException e) {
                if (e.getErrorCode() != ErrorCode.DUPLICATE_KEY) {
                    throw e;
                }
            }
            insDoneCmp[i] = insDone;
        }
        if (LOGGER.isLoggable(Level.INFO)) {
            LOGGER.info("Deleting from tree...");
        }
        int delDone = 0;
        for (int i = 0; i < ins; i++) {
            TupleUtils.createTuple(tb, tuple, fieldSerdes, f0s[i], f1s[i]);
            if (LOGGER.isLoggable(Level.INFO)) {
                if (i % 1000 == 0) {
                    LOGGER.info("Deleting " + i);
                }
            }
            try {
                indexAccessor.delete(tuple);
                delDone++;
            } catch (HyracksDataException e) {
                if (e.getErrorCode() != ErrorCode.UPDATE_OR_DELETE_NON_EXISTENT_KEY) {
                    throw e;
                }
            }
            if (insDoneCmp[i] != delDone) {
                if (LOGGER.isLoggable(Level.INFO)) {
                    LOGGER.info("INCONSISTENT STATE, ERROR IN DELETION EXAMPLE.");
                    LOGGER.info("INSDONECMP: " + insDoneCmp[i] + " " + delDone);
                }
                break;
            }
        }
        if (insDone != delDone) {
            if (LOGGER.isLoggable(Level.INFO)) {
                LOGGER.info("ERROR! INSDONE: " + insDone + " DELDONE: " + delDone);
            }
            break;
        }
    }
    treeIndex.validate();
    treeIndex.deactivate();
    treeIndex.destroy();
}

Example 29 with UTF8StringSerializerDeserializer

use of org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer in project asterixdb by apache.

the class OrderedIndexExamplesTest method varLenKeyValueExample.

/**
     * Variable-Length Example. Create a BTree with one variable-length key
     * field and one variable-length value field. Fill BTree with random values
     * using insertions (not bulk load) Perform ordered scans and range search.
     */
@Test
public void varLenKeyValueExample() throws Exception {
    if (LOGGER.isLoggable(Level.INFO)) {
        LOGGER.info("Variable-Length Key,Value Example");
    }
    // Declare fields.
    int fieldCount = 2;
    ITypeTraits[] typeTraits = new ITypeTraits[fieldCount];
    typeTraits[0] = UTF8StringPointable.TYPE_TRAITS;
    typeTraits[1] = UTF8StringPointable.TYPE_TRAITS;
    // Declare field serdes.
    ISerializerDeserializer[] fieldSerdes = { new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer() };
    // Declare keys.
    int keyFieldCount = 1;
    IBinaryComparatorFactory[] cmpFactories = new IBinaryComparatorFactory[keyFieldCount];
    cmpFactories[0] = PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY);
    // This is only used for the LSM-BTree.
    int[] bloomFilterKeyFields = new int[keyFieldCount];
    bloomFilterKeyFields[0] = 0;
    ITreeIndex treeIndex = createTreeIndex(typeTraits, cmpFactories, bloomFilterKeyFields, null, null, null, null);
    treeIndex.create();
    treeIndex.activate();
    long start = System.currentTimeMillis();
    if (LOGGER.isLoggable(Level.INFO)) {
        LOGGER.info("Inserting into tree...");
    }
    ArrayTupleBuilder tb = new ArrayTupleBuilder(fieldCount);
    ArrayTupleReference tuple = new ArrayTupleReference();
    IIndexAccessor indexAccessor = treeIndex.createAccessor(TestOperationCallback.INSTANCE, TestOperationCallback.INSTANCE);
    // Max string length to be generated.
    int maxLength = 10;
    int numInserts = 10000;
    for (int i = 0; i < 10000; i++) {
        String f0 = randomString(Math.abs(rnd.nextInt()) % maxLength + 1, rnd);
        String f1 = randomString(Math.abs(rnd.nextInt()) % maxLength + 1, rnd);
        TupleUtils.createTuple(tb, tuple, fieldSerdes, f0, f1);
        if (LOGGER.isLoggable(Level.INFO)) {
            if (i % 1000 == 0) {
                LOGGER.info("Inserting[" + i + "] " + f0 + " " + f1);
            }
        }
        try {
            indexAccessor.insert(tuple);
        } catch (HyracksDataException e) {
            if (e.getErrorCode() != ErrorCode.DUPLICATE_KEY) {
                throw e;
            }
        }
    }
    long end = System.currentTimeMillis();
    if (LOGGER.isLoggable(Level.INFO)) {
        LOGGER.info(numInserts + " inserts in " + (end - start) + "ms");
    }
    orderedScan(indexAccessor, fieldSerdes);
    diskOrderScan(indexAccessor, fieldSerdes);
    // Build low key.
    ArrayTupleBuilder lowKeyTb = new ArrayTupleBuilder(1);
    ArrayTupleReference lowKey = new ArrayTupleReference();
    TupleUtils.createTuple(lowKeyTb, lowKey, fieldSerdes, "cbf");
    // Build high key.
    ArrayTupleBuilder highKeyTb = new ArrayTupleBuilder(1);
    ArrayTupleReference highKey = new ArrayTupleReference();
    TupleUtils.createTuple(highKeyTb, highKey, fieldSerdes, "cc7");
    rangeSearch(cmpFactories, indexAccessor, fieldSerdes, lowKey, highKey, null, null);
    treeIndex.validate();
    treeIndex.deactivate();
    treeIndex.destroy();
}

Example 30 with UTF8StringSerializerDeserializer

use of org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer in project asterixdb by apache.

the class DataflowTest method testHDFSReadWriteOperators.

/**
     * Test a job with only HDFS read and writes.
     *
     * @throws Exception
     */
public void testHDFSReadWriteOperators() throws Exception {
    FileInputFormat.setInputPaths(conf, HDFS_INPUT_PATH);
    FileOutputFormat.setOutputPath(conf, new Path(HDFS_OUTPUT_PATH));
    conf.setInputFormat(TextInputFormat.class);
    Scheduler scheduler = new Scheduler(HyracksUtils.CC_HOST, HyracksUtils.TEST_HYRACKS_CC_CLIENT_PORT);
    InputSplit[] splits = conf.getInputFormat().getSplits(conf, numberOfNC * 4);
    String[] readSchedule = scheduler.getLocationConstraints(splits);
    JobSpecification jobSpec = new JobSpecification();
    RecordDescriptor recordDesc = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer() });
    String[] locations = new String[] { HyracksUtils.NC1_ID, HyracksUtils.NC1_ID, HyracksUtils.NC2_ID, HyracksUtils.NC2_ID };
    HDFSReadOperatorDescriptor readOperator = new HDFSReadOperatorDescriptor(jobSpec, recordDesc, conf, splits, readSchedule, new TextKeyValueParserFactory());
    PartitionConstraintHelper.addAbsoluteLocationConstraint(jobSpec, readOperator, locations);
    ExternalSortOperatorDescriptor sortOperator = new ExternalSortOperatorDescriptor(jobSpec, 10, new int[] { 0 }, new IBinaryComparatorFactory[] { RawBinaryComparatorFactory.INSTANCE }, recordDesc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(jobSpec, sortOperator, locations);
    HDFSWriteOperatorDescriptor writeOperator = new HDFSWriteOperatorDescriptor(jobSpec, conf, new TextTupleWriterFactory());
    PartitionConstraintHelper.addAbsoluteLocationConstraint(jobSpec, writeOperator, HyracksUtils.NC1_ID);
    jobSpec.connect(new OneToOneConnectorDescriptor(jobSpec), readOperator, 0, sortOperator, 0);
    jobSpec.connect(new MToNPartitioningMergingConnectorDescriptor(jobSpec, new FieldHashPartitionComputerFactory(new int[] { 0 }, new IBinaryHashFunctionFactory[] { RawBinaryHashFunctionFactory.INSTANCE }), new int[] { 0 }, new IBinaryComparatorFactory[] { RawBinaryComparatorFactory.INSTANCE }, null), sortOperator, 0, writeOperator, 0);
    jobSpec.addRoot(writeOperator);
    IHyracksClientConnection client = new HyracksConnection(HyracksUtils.CC_HOST, HyracksUtils.TEST_HYRACKS_CC_CLIENT_PORT);
    JobId jobId = client.startJob(jobSpec);
    client.waitForCompletion(jobId);
    Assert.assertEquals(true, checkResults());
}