org.apache.flink.api.java.io.DiscardingOutputFormat Java Examples

@Test
public void testUnaryFunctionWildcardForwardedAnnotation() {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();

	@SuppressWarnings("unchecked")
	DataSet<Tuple3<Long, String, Integer>> input = env.fromElements(new Tuple3<Long, String, Integer>(3L, "test", 42));
	input.map(new WildcardForwardedMapper<Tuple3<Long, String, Integer>>()).output(new DiscardingOutputFormat<Tuple3<Long, String, Integer>>());
	Plan plan = env.createProgramPlan();

	GenericDataSinkBase<?> sink = plan.getDataSinks().iterator().next();
	MapOperatorBase<?, ?, ?> mapper = (MapOperatorBase<?, ?, ?>) sink.getInput();

	SingleInputSemanticProperties semantics = mapper.getSemanticProperties();

	FieldSet fw1 = semantics.getForwardingTargetFields(0, 0);
	FieldSet fw2 = semantics.getForwardingTargetFields(0, 1);
	FieldSet fw3 = semantics.getForwardingTargetFields(0, 2);
	assertNotNull(fw1);
	assertNotNull(fw2);
	assertNotNull(fw3);
	assertTrue(fw1.contains(0));
	assertTrue(fw2.contains(1));
	assertTrue(fw3.contains(2));
}
 

@Test
public void testReduce() {
	// construct the plan
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	env.setParallelism(DEFAULT_PARALLELISM);
	DataSet<Long> set1 = env.generateSequence(0,1);

	set1.reduceGroup(new IdentityGroupReducer<Long>()).name("Reduce1")
			.output(new DiscardingOutputFormat<Long>()).name("Sink");

	Plan plan = env.createProgramPlan();

	try {
		OptimizedPlan oPlan = compileNoStats(plan);
		JobGraphGenerator jobGen = new JobGraphGenerator();
		jobGen.compileJobGraph(oPlan);
	} catch(CompilerException ce) {
		ce.printStackTrace();
		fail("The pact compiler is unable to compile this plan correctly");
	}
}
 

/**
 * Source -> Map -> Reduce -> Cross -> Reduce -> Cross -> Reduce ->
 * |--------------------------/                  /
 * |--------------------------------------------/
 * 
 * First cross has SameKeyFirst output contract
 */
@Test
public void testTicket158() {
	// construct the plan
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	env.setParallelism(DEFAULT_PARALLELISM);
	DataSet<Long> set1 = env.generateSequence(0,1);

	set1.map(new IdentityMapper<Long>()).name("Map1")
			.groupBy("*").reduceGroup(new IdentityGroupReducer<Long>()).name("Reduce1")
			.cross(set1).with(new IdentityCrosser<Long>()).withForwardedFieldsFirst("*").name("Cross1")
			.groupBy("*").reduceGroup(new IdentityGroupReducer<Long>()).name("Reduce2")
			.cross(set1).with(new IdentityCrosser<Long>()).name("Cross2")
			.groupBy("*").reduceGroup(new IdentityGroupReducer<Long>()).name("Reduce3")
			.output(new DiscardingOutputFormat<Long>()).name("Sink");

	Plan plan = env.createProgramPlan();
	OptimizedPlan oPlan = compileNoStats(plan);

	JobGraphGenerator jobGen = new JobGraphGenerator();
	jobGen.compileJobGraph(oPlan);
}
 

@Test
public void testBranchesOnlyInBCVariables1() {
	try{
		ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
		env.setParallelism(100);

		DataSet<Long> input = env.generateSequence(1, 10);
		DataSet<Long> bc_input = env.generateSequence(1, 10);
		
		input
			.map(new IdentityMapper<Long>()).withBroadcastSet(bc_input, "name1")
			.map(new IdentityMapper<Long>()).withBroadcastSet(bc_input, "name2")
			.output(new DiscardingOutputFormat<Long>());
		
		Plan plan = env.createProgramPlan();
		compileNoStats(plan);
	}
	catch(Exception e){
		e.printStackTrace();
		fail(e.getMessage());
	}
}
 

@Test
public void testAccumulator() {
	try {
		final int numElements = 100;

		ExecutionEnvironment env = ExecutionEnvironment.createCollectionsEnvironment();

		env.generateSequence(1, numElements)
			.map(new CountingMapper())
			.output(new DiscardingOutputFormat<Long>());

		JobExecutionResult result = env.execute();

		assertTrue(result.getNetRuntime() >= 0);

		assertEquals(numElements, (int) result.getAccumulatorResult(ACCUMULATOR_NAME));
	}
	catch (Exception e) {
		e.printStackTrace();
		fail(e.getMessage());
	}
}
 

/**
 * Test inDegrees() with an edge having a trgId that does not exist in the vertex DataSet.
 */
@Test
public void testInDegreesInvalidEdgeTrgId() throws Exception {

	final ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	env.setParallelism(PARALLELISM);
	env.getConfig().disableSysoutLogging();

	Graph<Long, Long, Long> graph = Graph.fromDataSet(TestGraphUtils.getLongLongVertexData(env),
			TestGraphUtils.getLongLongEdgeInvalidTrgData(env), env);

	try {
		graph.inDegrees().output(new DiscardingOutputFormat<>());
		env.execute();

		fail("graph.inDegrees() did not fail.");
	} catch (Exception e) {
		// We expect the job to fail with an exception
	}
}
 

@Test
public void testUnaryFunctionForwardedInLine2() {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();

	@SuppressWarnings("unchecked")
	DataSet<Tuple3<Long, Long, Long>> input = env.fromElements(new Tuple3<Long, Long, Long>(3L, 2L, 1L));
	input.map(new ReadSetMapper<Tuple3<Long, Long, Long>>()).withForwardedFields("0->1; 2")
			.output(new DiscardingOutputFormat<Tuple3<Long, Long, Long>>());
	Plan plan = env.createProgramPlan();

	GenericDataSinkBase<?> sink = plan.getDataSinks().iterator().next();
	MapOperatorBase<?, ?, ?> mapper = (MapOperatorBase<?, ?, ?>) sink.getInput();

	SingleInputSemanticProperties semantics = mapper.getSemanticProperties();

	FieldSet fw1 = semantics.getForwardingTargetFields(0, 0);
	FieldSet fw2 = semantics.getForwardingTargetFields(0, 2);
	assertNotNull(fw1);
	assertNotNull(fw2);
	assertTrue(fw1.contains(1));
	assertTrue(fw2.contains(2));
}
 

public static void main(String[] args) throws Exception {

		ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
		env.getConfig().disableSysoutLogging();

		DataSet<Integer> data = env.createInput(new CustomInputFormat());

		data
			.map(new MapFunction<Integer, Tuple2<Integer, Double>>() {
				@Override
				public Tuple2<Integer, Double> map(Integer value) {
					return new Tuple2<Integer, Double>(value, value * 0.5);
				}
			})
			.output(new DiscardingOutputFormat<Tuple2<Integer, Double>>());

		env.execute();
	}
 

@Test
public void testUnaryFunctionAllForwardedExceptAnnotation() {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();

	@SuppressWarnings("unchecked")
	DataSet<Tuple3<Long, Long, Long>> input = env.fromElements(new Tuple3<Long, Long, Long>(3L, 2L, 1L));
	input.map(new AllForwardedExceptMapper<Tuple3<Long, Long, Long>>()).output(new DiscardingOutputFormat<Tuple3<Long, Long, Long>>());
	Plan plan = env.createProgramPlan();

	GenericDataSinkBase<?> sink = plan.getDataSinks().iterator().next();
	MapOperatorBase<?, ?, ?> mapper = (MapOperatorBase<?, ?, ?>) sink.getInput();

	SingleInputSemanticProperties semantics = mapper.getSemanticProperties();

	FieldSet fw1 = semantics.getForwardingTargetFields(0, 0);
	FieldSet fw2 = semantics.getForwardingTargetFields(0, 2);
	assertNotNull(fw1);
	assertNotNull(fw2);
	assertTrue(fw1.contains(0));
	assertTrue(fw2.contains(2));
}
 

@Test
public void testInvalidTypeAccumulator() throws Exception {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();

	env.getConfig().disableSysoutLogging();

	// Test Exception forwarding with faulty Accumulator implementation
	env.generateSequence(0, 10000)
		.map(new IncompatibleAccumulatorTypesMapper())
		.map(new IncompatibleAccumulatorTypesMapper2())
		.output(new DiscardingOutputFormat<>());

	try {
		env.execute();
		fail("Should have failed.");
	} catch (JobExecutionException e) {
		assertTrue("Root cause should be:",
				e.getCause() instanceof Exception);
		assertTrue("Root cause should be:",
				e.getCause().getCause() instanceof UnsupportedOperationException);
	}
}
 

@Test
public void testDisjointFlows() {
	try {
		ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
		
		// generate two different flows
		env.generateSequence(1, 10)
				.output(new DiscardingOutputFormat<Long>());
		env.generateSequence(1, 10)
				.output(new DiscardingOutputFormat<Long>());
		
		Plan p = env.createProgramPlan();
		OptimizedPlan op = compileNoStats(p);
		
		new JobGraphGenerator().compileJobGraph(op);
	}
	catch (Exception e) {
		e.printStackTrace();
		fail(e.getMessage());
	}
}
 

@Test
public void testCancelSortMatchWhileDoingHeavySorting() throws Exception {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	HeavyCompareGeneratorInputFormat input = new HeavyCompareGeneratorInputFormat(100);
	DataSet<Tuple2<HeavyCompare, Integer>> input1 = env.createInput(input);
	DataSet<Tuple2<HeavyCompare, Integer>> input2 = env.createInput(input);

	input1.join(input2, JoinOperatorBase.JoinHint.REPARTITION_SORT_MERGE)
			.where(0)
			.equalTo(0)
			.with(new JoinFunction<Tuple2<HeavyCompare, Integer>, Tuple2<HeavyCompare, Integer>, Tuple2<HeavyCompare, Integer>>() {
				@Override
				public Tuple2<HeavyCompare, Integer> join(
					Tuple2<HeavyCompare, Integer> first,
					Tuple2<HeavyCompare, Integer> second) throws Exception {
					throw new Exception("Job should be canceled in sort-merge phase, never run here ...");
				}
			})
			.output(new DiscardingOutputFormat<Tuple2<HeavyCompare, Integer>>());

	runAndCancelJob(env.createProgramPlan(), 30 * 1000, 60 * 1000);
}
 

@Test
public void testIdentityIteration() {
	try {
		ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
		env.setParallelism(43);
		
		IterativeDataSet<Long> iteration = env.generateSequence(-4, 1000).iterate(100);
		iteration.closeWith(iteration).output(new DiscardingOutputFormat<Long>());
		
		Plan p = env.createProgramPlan();
		OptimizedPlan op = compileNoStats(p);
		
		new JobGraphGenerator().compileJobGraph(op);
	}
	catch (Exception e) {
		e.printStackTrace();
		fail(e.getMessage());
	}
}
 

/**
 * Test outDegrees() with an edge having a srcId that does not exist in the vertex DataSet.
 */
@Test
public void testOutDegreesInvalidEdgeSrcId() throws Exception {

	final ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	env.setParallelism(PARALLELISM);
	env.getConfig().disableSysoutLogging();

	Graph<Long, Long, Long> graph = Graph.fromDataSet(TestGraphUtils.getLongLongVertexData(env),
			TestGraphUtils.getLongLongEdgeInvalidSrcData(env), env);

	try {
		graph.outDegrees().output(new DiscardingOutputFormat<>());
		env.execute();

		fail("graph.outDegrees() did not fail.");
	} catch (Exception e) {
		// We expect the job to fail with an exception
	}
}
 

/**
 * Test outDegrees() with an edge having a srcId that does not exist in the vertex DataSet.
 */
@Test
public void testOutDegreesInvalidEdgeSrcId() throws Exception {

	final ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	env.setParallelism(PARALLELISM);
	env.getConfig().disableSysoutLogging();

	Graph<Long, Long, Long> graph = Graph.fromDataSet(TestGraphUtils.getLongLongVertexData(env),
			TestGraphUtils.getLongLongEdgeInvalidSrcData(env), env);

	try {
		graph.outDegrees().output(new DiscardingOutputFormat<>());
		env.execute();

		fail("graph.outDegrees() did not fail.");
	} catch (Exception e) {
		// We expect the job to fail with an exception
	}
}
 

/**
 * Test groupReduceOnNeighbors() -NeighborsFunctionWithVertexValue-
 * with an edge having a trgId that does not exist in the vertex DataSet.
 */
@Test
public void testGroupReduceOnNeighborsWithVVInvalidEdgeTrgId() throws Exception {

	final ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	env.setParallelism(PARALLELISM);
	env.getConfig().disableSysoutLogging();

	Graph<Long, Long, Long> graph = Graph.fromDataSet(TestGraphUtils.getLongLongVertexData(env),
			TestGraphUtils.getLongLongEdgeInvalidTrgData(env), env);

	try {
		DataSet<Tuple2<Long, Long>> verticesWithSumOfOutNeighborValues =
				graph.groupReduceOnNeighbors(new SumAllNeighbors(), EdgeDirection.ALL);

		verticesWithSumOfOutNeighborValues.output(new DiscardingOutputFormat<>());
		env.execute();

		fail("Expected an exception.");
	} catch (Exception e) {
		// We expect the job to fail with an exception
	}
}
 

private void executeTaskWithGenerator(
		JoinFunction<Tuple2<Integer, Integer>, Tuple2<Integer, Integer>, Tuple2<Integer, Integer>> joiner,
		int keys, int vals, int msecsTillCanceling, int maxTimeTillCanceled) throws Exception {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	DataSet<Tuple2<Integer, Integer>> input1 = env.createInput(new UniformIntTupleGeneratorInputFormat(keys, vals));
	DataSet<Tuple2<Integer, Integer>> input2 = env.createInput(new UniformIntTupleGeneratorInputFormat(keys, vals));

	input1.join(input2, JoinOperatorBase.JoinHint.REPARTITION_SORT_MERGE)
			.where(0)
			.equalTo(0)
			.with(joiner)
			.output(new DiscardingOutputFormat<Tuple2<Integer, Integer>>());

	env.setParallelism(PARALLELISM);

	runAndCancelJob(env.createProgramPlan(), msecsTillCanceling, maxTimeTillCanceled);
}
 

/**
 * Test groupReduceOnNeighbors() -NeighborsFunctionWithVertexValue-
 * with an edge having a srcId that does not exist in the vertex DataSet.
 */
@Test
public void testGroupReduceOnNeighborsWithVVInvalidEdgeSrcId() throws Exception {

	final ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	env.setParallelism(PARALLELISM);
	env.getConfig().disableSysoutLogging();

	Graph<Long, Long, Long> graph = Graph.fromDataSet(TestGraphUtils.getLongLongVertexData(env),
			TestGraphUtils.getLongLongEdgeInvalidSrcData(env), env);

	try {
		DataSet<Tuple2<Long, Long>> verticesWithSumOfOutNeighborValues =
				graph.groupReduceOnNeighbors(new SumAllNeighbors(), EdgeDirection.ALL);

		verticesWithSumOfOutNeighborValues.output(new DiscardingOutputFormat<>());
		env.execute();

		fail("Expected an exception.");
	} catch (Exception e) {
		// We expect the job to fail with an exception
	}
}
 

@Test
public void testCustomPartitioningTupleAgg() {
	try {
		ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
		
		DataSet<Tuple2<Integer, Integer>> data = env.fromElements(new Tuple2<Integer, Integer>(0, 0))
				.rebalance().setParallelism(4);
		
		data.groupBy(0).withPartitioner(new TestPartitionerInt())
			.sum(1)
			.output(new DiscardingOutputFormat<Tuple2<Integer, Integer>>());
		
		Plan p = env.createProgramPlan();
		OptimizedPlan op = compileNoStats(p);
		
		SinkPlanNode sink = op.getDataSinks().iterator().next();
		SingleInputPlanNode reducer = (SingleInputPlanNode) sink.getInput().getSource();
		SingleInputPlanNode combiner = (SingleInputPlanNode) reducer.getInput().getSource();
		
		assertEquals(ShipStrategyType.FORWARD, sink.getInput().getShipStrategy());
		assertEquals(ShipStrategyType.PARTITION_CUSTOM, reducer.getInput().getShipStrategy());
		assertEquals(ShipStrategyType.FORWARD, combiner.getInput().getShipStrategy());
	}
	catch (Exception e) {
		e.printStackTrace();
		fail(e.getMessage());
	}
}
 

@Test
public void testParallelism() throws Exception {
	int parallelism = 2;

	Graph<LongValue, NullValue, NullValue> graph = new HypercubeGraph(env, 4)
		.setParallelism(parallelism)
		.generate();

	graph.getVertices().output(new DiscardingOutputFormat<>());
	graph.getEdges().output(new DiscardingOutputFormat<>());

	TestUtils.verifyParallelism(env, parallelism);
}
 

@Test
public void testDistinctDestroysPartitioningOfNonDistinctFields() {
	try {
		ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
		env.setParallelism(4);
		
		@SuppressWarnings("unchecked")
		DataSet<Tuple2<Long, Long>> data = env.fromElements(new Tuple2<Long, Long>(0L, 0L), new Tuple2<Long, Long>(1L, 1L))
				.map(new IdentityMapper<Tuple2<Long,Long>>()).setParallelism(4);
		
		data.distinct(1)
			.groupBy(0)
			.sum(1)
			.output(new DiscardingOutputFormat<Tuple2<Long, Long>>());
		
		Plan p = env.createProgramPlan();
		OptimizedPlan op = compileNoStats(p);
		
		SinkPlanNode sink = op.getDataSinks().iterator().next();
		SingleInputPlanNode reducer = (SingleInputPlanNode) sink.getInput().getSource();
		SingleInputPlanNode combiner = (SingleInputPlanNode) reducer.getInput().getSource();
		SingleInputPlanNode distinctReducer = (SingleInputPlanNode) combiner.getInput().getSource();
		
		assertEquals(ShipStrategyType.FORWARD, sink.getInput().getShipStrategy());
		
		// reducer must repartition, because it works on a different field
		assertEquals(ShipStrategyType.PARTITION_HASH, reducer.getInput().getShipStrategy());

		assertEquals(ShipStrategyType.FORWARD, combiner.getInput().getShipStrategy());
		
		// distinct reducer is partitioned
		assertEquals(ShipStrategyType.PARTITION_HASH, distinctReducer.getInput().getShipStrategy());
	}
	catch (Exception e) {
		e.printStackTrace();
		fail(e.getMessage());
	}
}
 

@Test
public void testParallelism() throws Exception {
	int parallelism = 2;

	Graph<LongValue, NullValue, NullValue> graph = new EchoGraph(env, 10, 3)
		.setParallelism(parallelism)
		.generate();

	graph.getVertices().output(new DiscardingOutputFormat<>());
	graph.getEdges().output(new DiscardingOutputFormat<>());

	TestUtils.verifyParallelism(env, parallelism);
}
 

private static void runConnectedComponents(ExecutionEnvironment env) throws Exception {

		env.setParallelism(PARALLELISM);
		env.getConfig().disableSysoutLogging();

		// read vertex and edge data
		DataSet<Long> vertices = ConnectedComponentsData.getDefaultVertexDataSet(env)
				.rebalance();

		DataSet<Tuple2<Long, Long>> edges = ConnectedComponentsData.getDefaultEdgeDataSet(env)
				.rebalance()
				.flatMap(new ConnectedComponents.UndirectEdge());

		// assign the initial components (equal to the vertex id)
		DataSet<Tuple2<Long, Long>> verticesWithInitialId = vertices
				.map(new ConnectedComponents.DuplicateValue<Long>());

		// open a delta iteration
		DeltaIteration<Tuple2<Long, Long>, Tuple2<Long, Long>> iteration =
				verticesWithInitialId.iterateDelta(verticesWithInitialId, 100, 0);

		// apply the step logic: join with the edges, select the minimum neighbor,
		// update if the component of the candidate is smaller
		DataSet<Tuple2<Long, Long>> changes = iteration.getWorkset().join(edges)
				.where(0).equalTo(0)
				.with(new ConnectedComponents.NeighborWithComponentIDJoin())

				.groupBy(0).aggregate(Aggregations.MIN, 1)

				.join(iteration.getSolutionSet())
				.where(0).equalTo(0)
				.with(new ConnectedComponents.ComponentIdFilter());

		// close the delta iteration (delta and new workset are identical)
		DataSet<Tuple2<Long, Long>> result = iteration.closeWith(changes, changes);

		result.output(new DiscardingOutputFormat<Tuple2<Long, Long>>());

		env.execute();
	}
 

@Test
public void translateDistinctExpressionKey() {
	try {
		final int parallelism = 8;
		ExecutionEnvironment env = ExecutionEnvironment.createLocalEnvironment(parallelism);

		DataSet<CustomType> initialData = getSourcePojoDataSet(env);

		initialData.distinct("myInt").output(new DiscardingOutputFormat<CustomType>());

		Plan p = env.createProgramPlan();

		GenericDataSinkBase<?> sink = p.getDataSinks().iterator().next();

		// currently distinct is translated to a Reduce
		ReduceOperatorBase<?, ?> reducer = (ReduceOperatorBase<?, ?>) sink.getInput();

		// check types
		assertEquals(initialData.getType(), reducer.getOperatorInfo().getInputType());
		assertEquals(initialData.getType(), reducer.getOperatorInfo().getOutputType());

		// check keys
		assertArrayEquals(new int[] {0}, reducer.getKeyColumns(0));

		// parallelism was not configured on the operator
		assertTrue(reducer.getParallelism() == 1 || reducer.getParallelism() == -1);

		assertTrue(reducer.getInput() instanceof GenericDataSourceBase<?, ?>);
	}
	catch (Exception e) {
		System.err.println(e.getMessage());
		e.printStackTrace();
		fail("Test caused an error: " + e.getMessage());
	}
}
 

@Test
public void testFunctionSkipCodeAnalysisAnnotationPrecedence() {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	env.getConfig().setCodeAnalysisMode(CodeAnalysisMode.OPTIMIZE);

	@SuppressWarnings("unchecked")
	DataSet<Tuple3<Long, String, Integer>> input = env.fromElements(Tuple3.of(3L, "test", 42));
	input
			.map(new WildcardForwardedMapperWithSkipAnnotation<Tuple3<Long, String, Integer>>())
			.output(new DiscardingOutputFormat<Tuple3<Long, String, Integer>>());
	Plan plan = env.createProgramPlan();

	GenericDataSinkBase<?> sink = plan.getDataSinks().iterator().next();
	MapOperatorBase<?, ?, ?> mapper = (MapOperatorBase<?, ?, ?>) sink.getInput();

	SingleInputSemanticProperties semantics = mapper.getSemanticProperties();

	FieldSet fw1 = semantics.getForwardingTargetFields(0, 0);
	FieldSet fw2 = semantics.getForwardingTargetFields(0, 1);
	FieldSet fw3 = semantics.getForwardingTargetFields(0, 2);
	assertNotNull(fw1);
	assertNotNull(fw2);
	assertNotNull(fw3);
	assertFalse(fw1.contains(0));
	assertFalse(fw2.contains(1));
	assertFalse(fw3.contains(2));
}
 

@Test
public void testFunctionForwardedAnnotationPrecedence() {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	env.getConfig().setCodeAnalysisMode(CodeAnalysisMode.OPTIMIZE);

	@SuppressWarnings("unchecked")
	DataSet<Tuple3<Long, String, Integer>> input = env.fromElements(Tuple3.of(3L, "test", 42));
	input
			.map(new WildcardForwardedMapperWithForwardAnnotation<Tuple3<Long, String, Integer>>())
			.output(new DiscardingOutputFormat<Tuple3<Long, String, Integer>>());
	Plan plan = env.createProgramPlan();

	GenericDataSinkBase<?> sink = plan.getDataSinks().iterator().next();
	MapOperatorBase<?, ?, ?> mapper = (MapOperatorBase<?, ?, ?>) sink.getInput();

	SingleInputSemanticProperties semantics = mapper.getSemanticProperties();

	FieldSet fw1 = semantics.getForwardingTargetFields(0, 0);
	FieldSet fw2 = semantics.getForwardingTargetFields(0, 1);
	FieldSet fw3 = semantics.getForwardingTargetFields(0, 2);
	assertNotNull(fw1);
	assertNotNull(fw2);
	assertNotNull(fw3);
	assertTrue(fw1.contains(0));
	assertFalse(fw2.contains(1));
	assertFalse(fw3.contains(2));
}
 

@Override
protected void testProgram() throws Exception {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();

	if (!isCollectionExecution()) {
		Assert.assertTrue(env.getParallelism() > 1);
	}

	env.generateSequence(1, 1000)
		.partitionCustom(new AllZeroPartitioner(), new IdKeySelector<Long>())
		.map(new FailExceptInPartitionZeroMapper())
		.output(new DiscardingOutputFormat<Long>());

	env.execute();
}
 

@Test
public void testGetExecutionPlanOfRangePartition() {
	ExecutionEnvironment env = new LocalEnvironment();
	env.getConfig().disableSysoutLogging();

	DataSet<Integer> baseSet = env.fromElements(1, 2);

	DataSet<Tuple2<Integer, Integer>> result = baseSet
		.map(new MapFunction<Integer, Tuple2<Integer, Integer>>() {
			@Override
			public Tuple2<Integer, Integer> map(Integer value) throws Exception {
				return new Tuple2(value, value * 2);
			}
		})
		.partitionByRange(0)
		.aggregate(Aggregations.MAX, 1);
	result.output(new DiscardingOutputFormat<Tuple2<Integer, Integer>>());

	try {
		env.getExecutionPlan();
		env.execute();
	}
	catch (Exception e) {
		e.printStackTrace();
		fail("Cannot run both #getExecutionPlan and #execute.");
	}
}
 

@Test
public void testBulkIterationInClosure() {
	try {
		ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
		
		DataSet<Long> data1 = env.generateSequence(1, 100);
		DataSet<Long> data2 = env.generateSequence(1, 100);
		
		IterativeDataSet<Long> firstIteration = data1.iterate(100);
		
		DataSet<Long> firstResult = firstIteration.closeWith(firstIteration.map(new IdentityMapper<Long>()));
		
		
		IterativeDataSet<Long> mainIteration = data2.map(new IdentityMapper<Long>()).iterate(100);
		
		DataSet<Long> joined = mainIteration.join(firstResult)
				.where(new IdentityKeyExtractor<Long>()).equalTo(new IdentityKeyExtractor<Long>())
				.with(new DummyFlatJoinFunction<Long>());
		
		DataSet<Long> mainResult = mainIteration.closeWith(joined);
		
		mainResult.output(new DiscardingOutputFormat<Long>());
		
		Plan p = env.createProgramPlan();
		
		// optimizer should be able to translate this
		OptimizedPlan op = compileNoStats(p);
		
		// job graph generator should be able to translate this
		new JobGraphGenerator().compileJobGraph(op);
	}
	catch (Exception e) {
		e.printStackTrace();
		fail(e.getMessage());
	}
}
 

@Test
public void testGeneratingJobGraphWithUnconsumedResultPartition() {

	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();

	DataSet<Tuple2<Long, Long>> input = env.fromElements(new Tuple2<>(1L, 2L))
		.setParallelism(1);

	DataSet<Tuple2<Long, Long>> ds = input.map(new IdentityMapper<>())
		.setParallelism(3);

	AbstractID intermediateDataSetID = new AbstractID();

	// this output branch will be excluded.
	ds.output(BlockingShuffleOutputFormat.createOutputFormat(intermediateDataSetID))
		.setParallelism(1);

	// this is the normal output branch.
	ds.output(new DiscardingOutputFormat<>())
		.setParallelism(1);

	JobGraph jobGraph = compileJob(env);

	Assert.assertEquals(3, jobGraph.getVerticesSortedTopologicallyFromSources().size());

	JobVertex mapVertex = jobGraph.getVerticesSortedTopologicallyFromSources().get(1);
	Assert.assertThat(mapVertex, Matchers.instanceOf(JobVertex.class));

	// there are 2 output result with one of them is ResultPartitionType.BLOCKING_PERSISTENT
	Assert.assertEquals(2, mapVertex.getProducedDataSets().size());

	Assert.assertTrue(mapVertex.getProducedDataSets().stream()
		.anyMatch(dataSet -> dataSet.getId().equals(new IntermediateDataSetID(intermediateDataSetID)) &&
			dataSet.getResultType() == ResultPartitionType.BLOCKING_PERSISTENT));
}