Search in sources :

Example 16 with RuntimeUDFContext

use of org.apache.flink.api.common.functions.util.RuntimeUDFContext in project flink by apache.

the class GroupReduceOperatorTest method testGroupReduceCollectionWithRuntimeContext.

@Test
public void testGroupReduceCollectionWithRuntimeContext() {
    try {
        final String taskName = "Test Task";
        final AtomicBoolean opened = new AtomicBoolean();
        final AtomicBoolean closed = new AtomicBoolean();
        final RichGroupReduceFunction<Tuple2<String, Integer>, Tuple2<String, Integer>> reducer = new RichGroupReduceFunction<Tuple2<String, Integer>, Tuple2<String, Integer>>() {

            @Override
            public void reduce(Iterable<Tuple2<String, Integer>> values, Collector<Tuple2<String, Integer>> out) throws Exception {
                Iterator<Tuple2<String, Integer>> input = values.iterator();
                Tuple2<String, Integer> result = input.next();
                int sum = result.f1;
                while (input.hasNext()) {
                    Tuple2<String, Integer> next = input.next();
                    sum += next.f1;
                }
                result.f1 = sum;
                out.collect(result);
            }

            @Override
            public void open(Configuration parameters) throws Exception {
                opened.set(true);
                RuntimeContext ctx = getRuntimeContext();
                assertEquals(0, ctx.getIndexOfThisSubtask());
                assertEquals(1, ctx.getNumberOfParallelSubtasks());
                assertEquals(taskName, ctx.getTaskName());
            }

            @Override
            public void close() throws Exception {
                closed.set(true);
            }
        };
        GroupReduceOperatorBase<Tuple2<String, Integer>, Tuple2<String, Integer>, GroupReduceFunction<Tuple2<String, Integer>, Tuple2<String, Integer>>> op = new GroupReduceOperatorBase<Tuple2<String, Integer>, Tuple2<String, Integer>, GroupReduceFunction<Tuple2<String, Integer>, Tuple2<String, Integer>>>(reducer, new UnaryOperatorInformation<Tuple2<String, Integer>, Tuple2<String, Integer>>(TypeInfoParser.<Tuple2<String, Integer>>parse("Tuple2<String, Integer>"), TypeInfoParser.<Tuple2<String, Integer>>parse("Tuple2<String, Integer>")), new int[] { 0 }, "TestReducer");
        List<Tuple2<String, Integer>> input = new ArrayList<Tuple2<String, Integer>>(asList(new Tuple2<String, Integer>("foo", 1), new Tuple2<String, Integer>("foo", 3), new Tuple2<String, Integer>("bar", 2), new Tuple2<String, Integer>("bar", 4)));
        final TaskInfo taskInfo = new TaskInfo(taskName, 1, 0, 1, 0);
        ExecutionConfig executionConfig = new ExecutionConfig();
        executionConfig.disableObjectReuse();
        List<Tuple2<String, Integer>> resultMutableSafe = op.executeOnCollections(input, new RuntimeUDFContext(taskInfo, null, executionConfig, new HashMap<String, Future<Path>>(), new HashMap<String, Accumulator<?, ?>>(), new UnregisteredMetricsGroup()), executionConfig);
        executionConfig.enableObjectReuse();
        List<Tuple2<String, Integer>> resultRegular = op.executeOnCollections(input, new RuntimeUDFContext(taskInfo, null, executionConfig, new HashMap<String, Future<Path>>(), new HashMap<String, Accumulator<?, ?>>(), new UnregisteredMetricsGroup()), executionConfig);
        Set<Tuple2<String, Integer>> resultSetMutableSafe = new HashSet<Tuple2<String, Integer>>(resultMutableSafe);
        Set<Tuple2<String, Integer>> resultSetRegular = new HashSet<Tuple2<String, Integer>>(resultRegular);
        Set<Tuple2<String, Integer>> expectedResult = new HashSet<Tuple2<String, Integer>>(asList(new Tuple2<String, Integer>("foo", 4), new Tuple2<String, Integer>("bar", 6)));
        assertEquals(expectedResult, resultSetMutableSafe);
        assertEquals(expectedResult, resultSetRegular);
        assertTrue(opened.get());
        assertTrue(closed.get());
    } catch (Exception e) {
        e.printStackTrace();
        fail(e.getMessage());
    }
}
Also used : RichGroupReduceFunction(org.apache.flink.api.common.functions.RichGroupReduceFunction) UnregisteredMetricsGroup(org.apache.flink.metrics.groups.UnregisteredMetricsGroup) Configuration(org.apache.flink.configuration.Configuration) HashMap(java.util.HashMap) ArrayList(java.util.ArrayList) ExecutionConfig(org.apache.flink.api.common.ExecutionConfig) TaskInfo(org.apache.flink.api.common.TaskInfo) Collector(org.apache.flink.util.Collector) RuntimeUDFContext(org.apache.flink.api.common.functions.util.RuntimeUDFContext) HashSet(java.util.HashSet) Path(org.apache.flink.core.fs.Path) RichGroupReduceFunction(org.apache.flink.api.common.functions.RichGroupReduceFunction) GroupReduceFunction(org.apache.flink.api.common.functions.GroupReduceFunction) AtomicBoolean(java.util.concurrent.atomic.AtomicBoolean) Tuple2(org.apache.flink.api.java.tuple.Tuple2) RuntimeContext(org.apache.flink.api.common.functions.RuntimeContext) Test(org.junit.Test)

Example 17 with RuntimeUDFContext

use of org.apache.flink.api.common.functions.util.RuntimeUDFContext in project flink by apache.

the class InnerJoinOperatorBaseTest method testTupleBaseJoiner.

@Test
public void testTupleBaseJoiner() {
    final FlatJoinFunction<Tuple3<String, Double, Integer>, Tuple2<Integer, String>, Tuple2<Double, String>> joiner = new FlatJoinFunction<Tuple3<String, Double, Integer>, Tuple2<Integer, String>, Tuple2<Double, String>>() {

        @Override
        public void join(Tuple3<String, Double, Integer> first, Tuple2<Integer, String> second, Collector<Tuple2<Double, String>> out) {
            assertEquals(first.f0, second.f1);
            assertEquals(first.f2, second.f0);
            out.collect(new Tuple2<>(first.f1, second.f0.toString()));
        }
    };
    final TupleTypeInfo<Tuple3<String, Double, Integer>> leftTypeInfo = TupleTypeInfo.getBasicTupleTypeInfo(String.class, Double.class, Integer.class);
    final TupleTypeInfo<Tuple2<Integer, String>> rightTypeInfo = TupleTypeInfo.getBasicTupleTypeInfo(Integer.class, String.class);
    final TupleTypeInfo<Tuple2<Double, String>> outTypeInfo = TupleTypeInfo.getBasicTupleTypeInfo(Double.class, String.class);
    final int[] leftKeys = new int[] { 0, 2 };
    final int[] rightKeys = new int[] { 1, 0 };
    final String taskName = "Collection based tuple joiner";
    final BinaryOperatorInformation<Tuple3<String, Double, Integer>, Tuple2<Integer, String>, Tuple2<Double, String>> binaryOpInfo = new BinaryOperatorInformation<Tuple3<String, Double, Integer>, Tuple2<Integer, String>, Tuple2<Double, String>>(leftTypeInfo, rightTypeInfo, outTypeInfo);
    final InnerJoinOperatorBase<Tuple3<String, Double, Integer>, Tuple2<Integer, String>, Tuple2<Double, String>, FlatJoinFunction<Tuple3<String, Double, Integer>, Tuple2<Integer, String>, Tuple2<Double, String>>> base = new InnerJoinOperatorBase<Tuple3<String, Double, Integer>, Tuple2<Integer, String>, Tuple2<Double, String>, FlatJoinFunction<Tuple3<String, Double, Integer>, Tuple2<Integer, String>, Tuple2<Double, String>>>(joiner, binaryOpInfo, leftKeys, rightKeys, taskName);
    final List<Tuple3<String, Double, Integer>> inputData1 = new ArrayList<Tuple3<String, Double, Integer>>(Arrays.asList(new Tuple3<>("foo", 42.0, 1), new Tuple3<>("bar", 1.0, 2), new Tuple3<>("bar", 2.0, 3), new Tuple3<>("foobar", 3.0, 4), new Tuple3<>("bar", 3.0, 3)));
    final List<Tuple2<Integer, String>> inputData2 = new ArrayList<Tuple2<Integer, String>>(Arrays.asList(new Tuple2<>(3, "bar"), new Tuple2<>(4, "foobar"), new Tuple2<>(2, "foo")));
    final Set<Tuple2<Double, String>> expected = new HashSet<Tuple2<Double, String>>(Arrays.asList(new Tuple2<>(2.0, "3"), new Tuple2<>(3.0, "3"), new Tuple2<>(3.0, "4")));
    try {
        final TaskInfo taskInfo = new TaskInfo("op", 1, 0, 1, 0);
        ExecutionConfig executionConfig = new ExecutionConfig();
        executionConfig.disableObjectReuse();
        List<Tuple2<Double, String>> resultSafe = base.executeOnCollections(inputData1, inputData2, new RuntimeUDFContext(taskInfo, null, executionConfig, new HashMap<String, Future<Path>>(), new HashMap<String, Accumulator<?, ?>>(), new UnregisteredMetricsGroup()), executionConfig);
        executionConfig.enableObjectReuse();
        List<Tuple2<Double, String>> resultRegular = base.executeOnCollections(inputData1, inputData2, new RuntimeUDFContext(taskInfo, null, executionConfig, new HashMap<String, Future<Path>>(), new HashMap<String, Accumulator<?, ?>>(), new UnregisteredMetricsGroup()), executionConfig);
        assertEquals(expected, new HashSet<>(resultSafe));
        assertEquals(expected, new HashSet<>(resultRegular));
    } catch (Exception e) {
        e.printStackTrace();
        fail(e.getMessage());
    }
}
Also used : UnregisteredMetricsGroup(org.apache.flink.metrics.groups.UnregisteredMetricsGroup) HashMap(java.util.HashMap) ArrayList(java.util.ArrayList) FlatJoinFunction(org.apache.flink.api.common.functions.FlatJoinFunction) ExecutionConfig(org.apache.flink.api.common.ExecutionConfig) TaskInfo(org.apache.flink.api.common.TaskInfo) Collector(org.apache.flink.util.Collector) RuntimeUDFContext(org.apache.flink.api.common.functions.util.RuntimeUDFContext) BinaryOperatorInformation(org.apache.flink.api.common.operators.BinaryOperatorInformation) HashSet(java.util.HashSet) Path(org.apache.flink.core.fs.Path) Tuple2(org.apache.flink.api.java.tuple.Tuple2) Tuple3(org.apache.flink.api.java.tuple.Tuple3) Test(org.junit.Test)

Aggregations

TaskInfo (org.apache.flink.api.common.TaskInfo)17 RuntimeUDFContext (org.apache.flink.api.common.functions.util.RuntimeUDFContext)17 UnregisteredMetricsGroup (org.apache.flink.metrics.groups.UnregisteredMetricsGroup)17 HashMap (java.util.HashMap)15 ExecutionConfig (org.apache.flink.api.common.ExecutionConfig)13 Test (org.junit.Test)11 ArrayList (java.util.ArrayList)9 Path (org.apache.flink.core.fs.Path)7 Future (java.util.concurrent.Future)6 Accumulator (org.apache.flink.api.common.accumulators.Accumulator)6 AtomicBoolean (java.util.concurrent.atomic.AtomicBoolean)5 RuntimeContext (org.apache.flink.api.common.functions.RuntimeContext)5 Configuration (org.apache.flink.configuration.Configuration)5 HashSet (java.util.HashSet)4 List (java.util.List)4 Tuple2 (org.apache.flink.api.java.tuple.Tuple2)4 MetricGroup (org.apache.flink.metrics.MetricGroup)4 Collector (org.apache.flink.util.Collector)4 InvalidProgramException (org.apache.flink.api.common.InvalidProgramException)3 Map (java.util.Map)2