Examples with SinkFunction org.apache.flink.streaming.api.functions.sink.SinkFunction used on opensource projects

Example 1 with SinkFunction

use of org.apache.flink.streaming.api.functions.sink.SinkFunction in project flink by apache.

the class KafkaProducerTestBase method runCustomPartitioningTest.

/**
	 * 
	 * <pre>
	 *             +------> (sink) --+--> [KAFKA-1] --> (source) -> (map) --+
	 *            /                  |                                       \
	 *           /                   |                                        \
	 * (source) ----------> (sink) --+--> [KAFKA-2] --> (source) -> (map) -----+-> (sink)
	 *           \                   |                                        /
	 *            \                  |                                       /
	 *             +------> (sink) --+--> [KAFKA-3] --> (source) -> (map) --+
	 * </pre>
	 * 
	 * The mapper validates that the values come consistently from the correct Kafka partition.
	 * 
	 * The final sink validates that there are no duplicates and that all partitions are present.
	 */
public void runCustomPartitioningTest() {
    try {
        LOG.info("Starting KafkaProducerITCase.testCustomPartitioning()");
        final String topic = "customPartitioningTestTopic";
        final int parallelism = 3;
        createTestTopic(topic, parallelism, 1);
        TypeInformation<Tuple2<Long, String>> longStringInfo = TypeInfoParser.parse("Tuple2<Long, String>");
        StreamExecutionEnvironment env = StreamExecutionEnvironment.createRemoteEnvironment("localhost", flinkPort);
        env.setRestartStrategy(RestartStrategies.noRestart());
        env.getConfig().disableSysoutLogging();
        TypeInformationSerializationSchema<Tuple2<Long, String>> serSchema = new TypeInformationSerializationSchema<>(longStringInfo, env.getConfig());
        TypeInformationSerializationSchema<Tuple2<Long, String>> deserSchema = new TypeInformationSerializationSchema<>(longStringInfo, env.getConfig());
        // ------ producing topology ---------
        // source has DOP 1 to make sure it generates no duplicates
        DataStream<Tuple2<Long, String>> stream = env.addSource(new SourceFunction<Tuple2<Long, String>>() {

            private boolean running = true;

            @Override
            public void run(SourceContext<Tuple2<Long, String>> ctx) throws Exception {
                long cnt = 0;
                while (running) {
                    ctx.collect(new Tuple2<Long, String>(cnt, "kafka-" + cnt));
                    cnt++;
                }
            }

            @Override
            public void cancel() {
                running = false;
            }
        }).setParallelism(1);
        Properties props = new Properties();
        props.putAll(FlinkKafkaProducerBase.getPropertiesFromBrokerList(brokerConnectionStrings));
        props.putAll(secureProps);
        // sink partitions into 
        kafkaServer.produceIntoKafka(stream, topic, new KeyedSerializationSchemaWrapper<>(serSchema), props, new CustomPartitioner(parallelism)).setParallelism(parallelism);
        // ------ consuming topology ---------
        Properties consumerProps = new Properties();
        consumerProps.putAll(standardProps);
        consumerProps.putAll(secureProps);
        FlinkKafkaConsumerBase<Tuple2<Long, String>> source = kafkaServer.getConsumer(topic, deserSchema, consumerProps);
        env.addSource(source).setParallelism(parallelism).map(new RichMapFunction<Tuple2<Long, String>, Integer>() {

            private int ourPartition = -1;

            @Override
            public Integer map(Tuple2<Long, String> value) {
                int partition = value.f0.intValue() % parallelism;
                if (ourPartition != -1) {
                    assertEquals("inconsistent partitioning", ourPartition, partition);
                } else {
                    ourPartition = partition;
                }
                return partition;
            }
        }).setParallelism(parallelism).addSink(new SinkFunction<Integer>() {

            private int[] valuesPerPartition = new int[parallelism];

            @Override
            public void invoke(Integer value) throws Exception {
                valuesPerPartition[value]++;
                boolean missing = false;
                for (int i : valuesPerPartition) {
                    if (i < 100) {
                        missing = true;
                        break;
                    }
                }
                if (!missing) {
                    throw new SuccessException();
                }
            }
        }).setParallelism(1);
        tryExecute(env, "custom partitioning test");
        deleteTestTopic(topic);
        LOG.info("Finished KafkaProducerITCase.testCustomPartitioning()");
    } catch (Exception e) {
        e.printStackTrace();
        fail(e.getMessage());
    }
}

Example 2 with SinkFunction

use of org.apache.flink.streaming.api.functions.sink.SinkFunction in project flink by apache.

the class BufferTimeoutITCase method testDisablingBufferTimeout.

/**
 * The test verifies that it is possible to disable explicit buffer flushing. It checks that
 * OutputFlasher thread would not be started when the task is running. But this doesn't
 * guarantee that the unfinished buffers can not be flushed by another events.
 */
@Test
public void testDisablingBufferTimeout() throws Exception {
    final StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
    env.setParallelism(1);
    env.setBufferTimeout(-1);
    final SharedReference<ArrayList<Integer>> results = sharedObjects.add(new ArrayList<>());
    env.addSource(new SourceFunction<Integer>() {

        @Override
        public void run(SourceContext<Integer> ctx) throws Exception {
            ctx.collect(1);
            // just sleep forever
            Thread.sleep(Long.MAX_VALUE);
        }

        @Override
        public void cancel() {
        }
    }).slotSharingGroup("source").addSink(new SinkFunction<Integer>() {

        @Override
        public void invoke(Integer value, Context context) {
            results.get().add(value);
        }
    }).slotSharingGroup("sink");
    final JobClient jobClient = env.executeAsync();
    CommonTestUtils.waitForAllTaskRunning(MINI_CLUSTER_RESOURCE.getMiniCluster(), jobClient.getJobID(), false);
    assertTrue(RecordWriter.DEFAULT_OUTPUT_FLUSH_THREAD_NAME + " thread is unexpectedly running", Thread.getAllStackTraces().keySet().stream().noneMatch(thread -> thread.getName().startsWith(RecordWriter.DEFAULT_OUTPUT_FLUSH_THREAD_NAME)));
}

Example 3 with SinkFunction

use of org.apache.flink.streaming.api.functions.sink.SinkFunction in project flink by apache.

the class StreamOperatorChainingTest method testMultiChainingWithSplit.

/**
 * Verify that multi-chaining works with object reuse enabled.
 */
private void testMultiChainingWithSplit(StreamExecutionEnvironment env) throws Exception {
    // set parallelism to 2 to avoid chaining with source in case when available processors is
    // 1.
    env.setParallelism(2);
    // the actual elements will not be used
    DataStream<Integer> input = env.fromElements(1, 2, 3);
    sink1Results = new ArrayList<>();
    sink2Results = new ArrayList<>();
    sink3Results = new ArrayList<>();
    input = input.map(value -> value);
    OutputTag<Integer> oneOutput = new OutputTag<Integer>("one") {
    };
    OutputTag<Integer> otherOutput = new OutputTag<Integer>("other") {
    };
    SingleOutputStreamOperator<Object> split = input.process(new ProcessFunction<Integer, Object>() {

        private static final long serialVersionUID = 1L;

        @Override
        public void processElement(Integer value, Context ctx, Collector<Object> out) throws Exception {
            if (value.equals(1)) {
                ctx.output(oneOutput, value);
            } else {
                ctx.output(otherOutput, value);
            }
        }
    });
    split.getSideOutput(oneOutput).map(value -> "First 1: " + value).addSink(new SinkFunction<String>() {

        @Override
        public void invoke(String value, Context ctx) throws Exception {
            sink1Results.add(value);
        }
    });
    split.getSideOutput(oneOutput).map(value -> "First 2: " + value).addSink(new SinkFunction<String>() {

        @Override
        public void invoke(String value, Context ctx) throws Exception {
            sink2Results.add(value);
        }
    });
    split.getSideOutput(otherOutput).map(value -> "Second: " + value).addSink(new SinkFunction<String>() {

        @Override
        public void invoke(String value, Context ctx) throws Exception {
            sink3Results.add(value);
        }
    });
    // be build our own StreamTask and OperatorChain
    JobGraph jobGraph = env.getStreamGraph().getJobGraph();
    Assert.assertTrue(jobGraph.getVerticesSortedTopologicallyFromSources().size() == 2);
    JobVertex chainedVertex = jobGraph.getVerticesSortedTopologicallyFromSources().get(1);
    Configuration configuration = chainedVertex.getConfiguration();
    StreamConfig streamConfig = new StreamConfig(configuration);
    StreamMap<Integer, Integer> headOperator = streamConfig.getStreamOperator(Thread.currentThread().getContextClassLoader());
    try (MockEnvironment environment = createMockEnvironment(chainedVertex.getName())) {
        StreamTask<Integer, StreamMap<Integer, Integer>> mockTask = createMockTask(streamConfig, environment);
        OperatorChain<Integer, StreamMap<Integer, Integer>> operatorChain = createOperatorChain(streamConfig, environment, mockTask);
        headOperator.setup(mockTask, streamConfig, operatorChain.getMainOperatorOutput());
        operatorChain.initializeStateAndOpenOperators(null);
        headOperator.processElement(new StreamRecord<>(1));
        headOperator.processElement(new StreamRecord<>(2));
        headOperator.processElement(new StreamRecord<>(3));
        assertThat(sink1Results, contains("First 1: 1"));
        assertThat(sink2Results, contains("First 2: 1"));
        assertThat(sink3Results, contains("Second: 2", "Second: 3"));
    }
}

Example 4 with SinkFunction

use of org.apache.flink.streaming.api.functions.sink.SinkFunction in project flink by apache.

the class SavepointITCase method testTriggerSavepointAndResumeWithNoClaim.

@Test
@Ignore("Disabling this test because it regularly fails on AZP. See FLINK-25427.")
public void testTriggerSavepointAndResumeWithNoClaim() throws Exception {
    final int numTaskManagers = 2;
    final int numSlotsPerTaskManager = 2;
    final int parallelism = numTaskManagers * numSlotsPerTaskManager;
    final StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
    env.setStateBackend(new EmbeddedRocksDBStateBackend(true));
    env.getCheckpointConfig().enableExternalizedCheckpoints(CheckpointConfig.ExternalizedCheckpointCleanup.RETAIN_ON_CANCELLATION);
    env.getCheckpointConfig().setCheckpointStorage(folder.newFolder().toURI());
    env.setParallelism(parallelism);
    final SharedReference<CountDownLatch> counter = sharedObjects.add(new CountDownLatch(10_000));
    env.fromSequence(1, Long.MAX_VALUE).keyBy(i -> i % parallelism).process(new KeyedProcessFunction<Long, Long, Long>() {

        private ListState<Long> last;

        @Override
        public void open(Configuration parameters) {
            // we use list state here to create sst files of a significant size
            // if sst files do not reach certain thresholds they are not stored
            // in files, but as a byte stream in checkpoints metadata
            last = getRuntimeContext().getListState(new ListStateDescriptor<>("last", BasicTypeInfo.LONG_TYPE_INFO));
        }

        @Override
        public void processElement(Long value, KeyedProcessFunction<Long, Long, Long>.Context ctx, Collector<Long> out) throws Exception {
            last.add(value);
            out.collect(value);
        }
    }).addSink(new SinkFunction<Long>() {

        @Override
        public void invoke(Long value) {
            counter.consumeSync(CountDownLatch::countDown);
        }
    }).setParallelism(1);
    final JobGraph jobGraph = env.getStreamGraph().getJobGraph();
    MiniClusterWithClientResource cluster = new MiniClusterWithClientResource(new MiniClusterResourceConfiguration.Builder().setNumberTaskManagers(numTaskManagers).setNumberSlotsPerTaskManager(numSlotsPerTaskManager).build());
    cluster.before();
    try {
        final JobID jobID1 = new JobID();
        jobGraph.setJobID(jobID1);
        cluster.getClusterClient().submitJob(jobGraph).get();
        CommonTestUtils.waitForAllTaskRunning(cluster.getMiniCluster(), jobID1, false);
        // wait for some records to be processed before taking the checkpoint
        counter.get().await();
        final String firstCheckpoint = cluster.getMiniCluster().triggerCheckpoint(jobID1).get();
        cluster.getClusterClient().cancel(jobID1).get();
        jobGraph.setSavepointRestoreSettings(SavepointRestoreSettings.forPath(firstCheckpoint, false, RestoreMode.NO_CLAIM));
        final JobID jobID2 = new JobID();
        jobGraph.setJobID(jobID2);
        cluster.getClusterClient().submitJob(jobGraph).get();
        CommonTestUtils.waitForAllTaskRunning(cluster.getMiniCluster(), jobID2, false);
        String secondCheckpoint = cluster.getMiniCluster().triggerCheckpoint(jobID2).get();
        cluster.getClusterClient().cancel(jobID2).get();
        // delete the checkpoint we restored from
        FileUtils.deleteDirectory(Paths.get(new URI(firstCheckpoint)).getParent().toFile());
        // we should be able to restore from the second checkpoint even though it has been built
        // on top of the first checkpoint
        jobGraph.setSavepointRestoreSettings(SavepointRestoreSettings.forPath(secondCheckpoint, false, RestoreMode.NO_CLAIM));
        final JobID jobID3 = new JobID();
        jobGraph.setJobID(jobID3);
        cluster.getClusterClient().submitJob(jobGraph).get();
        CommonTestUtils.waitForAllTaskRunning(cluster.getMiniCluster(), jobID3, false);
    } finally {
        cluster.after();
    }
}

Example 5 with SinkFunction

use of org.apache.flink.streaming.api.functions.sink.SinkFunction in project flink by apache.

the class StreamOperatorChainingTest method testMultiChaining.

/**
 * Verify that multi-chaining works.
 */
private void testMultiChaining(StreamExecutionEnvironment env) throws Exception {
    // set parallelism to 2 to avoid chaining with source in case when available processors is
    // 1.
    env.setParallelism(2);
    // the actual elements will not be used
    DataStream<Integer> input = env.fromElements(1, 2, 3);
    sink1Results = new ArrayList<>();
    sink2Results = new ArrayList<>();
    input = input.map(value -> value);
    input.map(value -> "First: " + value).addSink(new SinkFunction<String>() {

        @Override
        public void invoke(String value, Context ctx) throws Exception {
            sink1Results.add(value);
        }
    });
    input.map(value -> "Second: " + value).addSink(new SinkFunction<String>() {

        @Override
        public void invoke(String value, Context ctx) throws Exception {
            sink2Results.add(value);
        }
    });
    // be build our own StreamTask and OperatorChain
    JobGraph jobGraph = env.getStreamGraph().getJobGraph();
    Assert.assertTrue(jobGraph.getVerticesSortedTopologicallyFromSources().size() == 2);
    JobVertex chainedVertex = jobGraph.getVerticesSortedTopologicallyFromSources().get(1);
    Configuration configuration = chainedVertex.getConfiguration();
    StreamConfig streamConfig = new StreamConfig(configuration);
    StreamMap<Integer, Integer> headOperator = streamConfig.getStreamOperator(Thread.currentThread().getContextClassLoader());
    try (MockEnvironment environment = createMockEnvironment(chainedVertex.getName())) {
        StreamTask<Integer, StreamMap<Integer, Integer>> mockTask = createMockTask(streamConfig, environment);
        OperatorChain<Integer, StreamMap<Integer, Integer>> operatorChain = createOperatorChain(streamConfig, environment, mockTask);
        headOperator.setup(mockTask, streamConfig, operatorChain.getMainOperatorOutput());
        operatorChain.initializeStateAndOpenOperators(null);
        headOperator.processElement(new StreamRecord<>(1));
        headOperator.processElement(new StreamRecord<>(2));
        headOperator.processElement(new StreamRecord<>(3));
        assertThat(sink1Results, contains("First: 1", "First: 2", "First: 3"));
        assertThat(sink2Results, contains("Second: 1", "Second: 2", "Second: 3"));
    }
}