use of org.apache.hyracks.algebricks.runtime.operators.std.SinkWriterRuntimeFactory in project asterixdb by apache.
the class PushRuntimeTest method etsUnnestRunningaggregateWrite.
@Test
public void etsUnnestRunningaggregateWrite() throws Exception {
JobSpecification spec = new JobSpecification(FRAME_SIZE);
EmptyTupleSourceRuntimeFactory ets = new EmptyTupleSourceRuntimeFactory();
RecordDescriptor etsDesc = new RecordDescriptor(new ISerializerDeserializer[] {});
IUnnestingEvaluatorFactory aggregFactory = new IntArrayUnnester(new int[] { 100, 200, 300 });
UnnestRuntimeFactory unnest = new UnnestRuntimeFactory(0, aggregFactory, new int[] { 0 }, false, null);
RecordDescriptor unnestDesc = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE });
RunningAggregateRuntimeFactory ragg = new RunningAggregateRuntimeFactory(new int[] { 1 }, new IRunningAggregateEvaluatorFactory[] { new TupleCountRunningAggregateFunctionFactory() }, new int[] { 0, 1 });
RecordDescriptor raggDesc = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE, IntegerSerializerDeserializer.INSTANCE });
String filePath = PATH_ACTUAL + SEPARATOR + "etsUnnestRunningaggregateWrite.out";
File outFile = new File(filePath);
SinkWriterRuntimeFactory writer = new SinkWriterRuntimeFactory(new int[] { 1 }, new IPrinterFactory[] { IntegerPrinterFactory.INSTANCE }, outFile, PrinterBasedWriterFactory.INSTANCE, raggDesc);
AlgebricksMetaOperatorDescriptor algebricksOp = new AlgebricksMetaOperatorDescriptor(spec, 0, 0, new IPushRuntimeFactory[] { ets, unnest, ragg, writer }, new RecordDescriptor[] { etsDesc, unnestDesc, raggDesc, null });
PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, algebricksOp, new String[] { AlgebricksHyracksIntegrationUtil.NC1_ID });
spec.addRoot(algebricksOp);
AlgebricksHyracksIntegrationUtil.runJob(spec);
StringBuilder buf = new StringBuilder();
readFileToString(outFile, buf);
Assert.assertEquals("123", buf.toString());
outFile.delete();
}
use of org.apache.hyracks.algebricks.runtime.operators.std.SinkWriterRuntimeFactory in project asterixdb by apache.
the class PushRuntimeTest method scanMicroSortWrite.
@Test
public void scanMicroSortWrite() throws Exception {
JobSpecification spec = new JobSpecification(FRAME_SIZE);
// the scanner
FileSplit[] fileSplits = new FileSplit[1];
fileSplits[0] = new ManagedFileSplit(AlgebricksHyracksIntegrationUtil.NC1_ID, "data" + File.separator + "tpch0.001" + File.separator + "nation.tbl");
IFileSplitProvider splitProvider = new ConstantFileSplitProvider(fileSplits);
RecordDescriptor scannerDesc = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE, new UTF8StringSerializerDeserializer(), IntegerSerializerDeserializer.INSTANCE, new UTF8StringSerializerDeserializer() });
IValueParserFactory[] valueParsers = new IValueParserFactory[] { IntegerParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, IntegerParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE };
FileScanOperatorDescriptor scanner = new FileScanOperatorDescriptor(spec, splitProvider, new DelimitedDataTupleParserFactory(valueParsers, '|'), scannerDesc);
PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, scanner, new String[] { AlgebricksHyracksIntegrationUtil.NC1_ID });
// the algebricks op.
InMemorySortRuntimeFactory sort = new InMemorySortRuntimeFactory(new int[] { 1 }, null, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, null);
RecordDescriptor sortDesc = scannerDesc;
String fileName = "scanMicroSortWrite.out";
String filePath = PATH_ACTUAL + SEPARATOR + fileName;
String resultFilePath = PATH_EXPECTED + SEPARATOR + fileName;
File outFile = new File(filePath);
SinkWriterRuntimeFactory writer = new SinkWriterRuntimeFactory(new int[] { 0, 1, 2, 3 }, new IPrinterFactory[] { IntegerPrinterFactory.INSTANCE, UTF8StringPrinterFactory.INSTANCE, IntegerPrinterFactory.INSTANCE, UTF8StringPrinterFactory.INSTANCE }, outFile, PrinterBasedWriterFactory.INSTANCE, sortDesc);
AlgebricksMetaOperatorDescriptor algebricksOp = new AlgebricksMetaOperatorDescriptor(spec, 1, 0, new IPushRuntimeFactory[] { sort, writer }, new RecordDescriptor[] { sortDesc, null });
PartitionConstraintHelper.addPartitionCountConstraint(spec, algebricksOp, 1);
spec.connect(new OneToOneConnectorDescriptor(spec), scanner, 0, algebricksOp, 0);
spec.addRoot(algebricksOp);
AlgebricksHyracksIntegrationUtil.runJob(spec);
compareFiles(filePath, resultFilePath);
outFile.delete();
}
use of org.apache.hyracks.algebricks.runtime.operators.std.SinkWriterRuntimeFactory in project asterixdb by apache.
the class PushRuntimeTest method scanMicroSortGbySelectWrite.
@Test
public void scanMicroSortGbySelectWrite() throws Exception {
JobSpecification spec = new JobSpecification(FRAME_SIZE);
// the scanner
FileSplit[] fileSplits = new FileSplit[1];
fileSplits[0] = new ManagedFileSplit(AlgebricksHyracksIntegrationUtil.NC1_ID, "data" + File.separator + "tpch0.001" + File.separator + "customer.tbl");
IFileSplitProvider splitProvider = new ConstantFileSplitProvider(fileSplits);
RecordDescriptor scannerDesc = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE, new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), IntegerSerializerDeserializer.INSTANCE, new UTF8StringSerializerDeserializer(), FloatSerializerDeserializer.INSTANCE, new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer() });
IValueParserFactory[] valueParsers = new IValueParserFactory[] { IntegerParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, IntegerParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, FloatParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE };
FileScanOperatorDescriptor scanner = new FileScanOperatorDescriptor(spec, splitProvider, new DelimitedDataTupleParserFactory(valueParsers, '|'), scannerDesc);
PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, scanner, new String[] { AlgebricksHyracksIntegrationUtil.NC1_ID });
// the sort (by nation id)
RecordDescriptor sortDesc = scannerDesc;
InMemorySortRuntimeFactory sort = new InMemorySortRuntimeFactory(new int[] { 3 }, null, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(IntegerPointable.FACTORY) }, null);
// the group-by
NestedTupleSourceRuntimeFactory nts = new NestedTupleSourceRuntimeFactory();
RecordDescriptor ntsDesc = sortDesc;
AggregateRuntimeFactory agg = new AggregateRuntimeFactory(new IAggregateEvaluatorFactory[] { new TupleCountAggregateFunctionFactory() });
RecordDescriptor aggDesc = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE });
AlgebricksPipeline pipeline = new AlgebricksPipeline(new IPushRuntimeFactory[] { nts, agg }, new RecordDescriptor[] { ntsDesc, aggDesc });
NestedPlansAccumulatingAggregatorFactory npaaf = new NestedPlansAccumulatingAggregatorFactory(new AlgebricksPipeline[] { pipeline }, new int[] { 3 }, new int[] {});
RecordDescriptor gbyDesc = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE, IntegerSerializerDeserializer.INSTANCE });
MicroPreClusteredGroupRuntimeFactory gby = new MicroPreClusteredGroupRuntimeFactory(new int[] { 3 }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(IntegerPointable.FACTORY) }, npaaf, sortDesc, gbyDesc, null);
// the algebricks op.
IScalarEvaluatorFactory cond = new IntegerEqualsEvalFactory(new IntegerConstantEvalFactory(3), // Canadian customers
new TupleFieldEvaluatorFactory(0));
StreamSelectRuntimeFactory select = new StreamSelectRuntimeFactory(cond, new int[] { 1 }, BinaryBooleanInspectorImpl.FACTORY, false, -1, null);
RecordDescriptor selectDesc = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE });
String filePath = PATH_ACTUAL + SEPARATOR + "scanSortGbySelectWrite.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[] { sort, gby, select, writer }, new RecordDescriptor[] { sortDesc, gbyDesc, selectDesc, null });
PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, algebricksOp, new String[] { AlgebricksHyracksIntegrationUtil.NC1_ID });
spec.connect(new OneToOneConnectorDescriptor(spec), scanner, 0, algebricksOp, 0);
spec.addRoot(algebricksOp);
AlgebricksHyracksIntegrationUtil.runJob(spec);
StringBuilder buf = new StringBuilder();
readFileToString(outFile, buf);
Assert.assertEquals("9", buf.toString());
outFile.delete();
}
use of org.apache.hyracks.algebricks.runtime.operators.std.SinkWriterRuntimeFactory in project asterixdb by apache.
the class PushRuntimeTest method scanAggregateWrite.
@Test
public void scanAggregateWrite() throws Exception {
JobSpecification spec = new JobSpecification(FRAME_SIZE);
// the scanner
FileSplit[] fileSplits = new FileSplit[1];
fileSplits[0] = new ManagedFileSplit(AlgebricksHyracksIntegrationUtil.NC1_ID, "data" + File.separator + "tpch0.001" + File.separator + "customer-part1.tbl");
IFileSplitProvider splitProvider = new ConstantFileSplitProvider(fileSplits);
RecordDescriptor scannerDesc = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE, new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), IntegerSerializerDeserializer.INSTANCE, new UTF8StringSerializerDeserializer(), FloatSerializerDeserializer.INSTANCE, new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer() });
IValueParserFactory[] valueParsers = new IValueParserFactory[] { IntegerParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, IntegerParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, FloatParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE };
FileScanOperatorDescriptor scanner = new FileScanOperatorDescriptor(spec, splitProvider, new DelimitedDataTupleParserFactory(valueParsers, '|'), scannerDesc);
PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, scanner, new String[] { AlgebricksHyracksIntegrationUtil.NC1_ID });
// the algebricks op.
AggregateRuntimeFactory agg = new AggregateRuntimeFactory(new IAggregateEvaluatorFactory[] { new TupleCountAggregateFunctionFactory() });
RecordDescriptor aggDesc = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE });
String filePath = PATH_ACTUAL + SEPARATOR + "scanAggregateWrite.out";
File outFile = new File(filePath);
SinkWriterRuntimeFactory writer = new SinkWriterRuntimeFactory(new int[] { 0 }, new IPrinterFactory[] { IntegerPrinterFactory.INSTANCE }, outFile, PrinterBasedWriterFactory.INSTANCE, aggDesc);
AlgebricksMetaOperatorDescriptor algebricksOp = new AlgebricksMetaOperatorDescriptor(spec, 1, 0, new IPushRuntimeFactory[] { agg, writer }, new RecordDescriptor[] { aggDesc, null });
PartitionConstraintHelper.addPartitionCountConstraint(spec, algebricksOp, 1);
spec.connect(new OneToOneConnectorDescriptor(spec), scanner, 0, algebricksOp, 0);
spec.addRoot(algebricksOp);
AlgebricksHyracksIntegrationUtil.runJob(spec);
StringBuilder buf = new StringBuilder();
readFileToString(outFile, buf);
Assert.assertEquals("75", buf.toString());
outFile.delete();
}
Aggregations