Examples with KStream org.apache.kafka.streams.kstream.KStream used on opensource projects

Example 11 with KStream

use of org.apache.kafka.streams.kstream.KStream in project kafka-streams-examples by confluentinc.

the class EventDeduplicationLambdaIntegrationTest method shouldRemoveDuplicatesFromTheInput.

@Test
public void shouldRemoveDuplicatesFromTheInput() throws Exception {
    // e.g. "4ff3cb44-abcb-46e3-8f9a-afb7cc74fbb8"
    String firstId = UUID.randomUUID().toString();
    String secondId = UUID.randomUUID().toString();
    String thirdId = UUID.randomUUID().toString();
    List<String> inputValues = Arrays.asList(firstId, secondId, firstId, firstId, secondId, thirdId, thirdId, firstId, secondId);
    List<String> expectedValues = Arrays.asList(firstId, secondId, thirdId);
    // 
    // Step 1: Configure and start the processor topology.
    // 
    StreamsBuilder builder = new StreamsBuilder();
    Properties streamsConfiguration = new Properties();
    streamsConfiguration.put(StreamsConfig.APPLICATION_ID_CONFIG, "deduplication-lambda-integration-test");
    streamsConfiguration.put(StreamsConfig.BOOTSTRAP_SERVERS_CONFIG, CLUSTER.bootstrapServers());
    streamsConfiguration.put(StreamsConfig.DEFAULT_KEY_SERDE_CLASS_CONFIG, Serdes.ByteArray().getClass().getName());
    streamsConfiguration.put(StreamsConfig.DEFAULT_VALUE_SERDE_CLASS_CONFIG, Serdes.String().getClass().getName());
    // The commit interval for flushing records to state stores and downstream must be lower than
    // this integration test's timeout (30 secs) to ensure we observe the expected processing results.
    streamsConfiguration.put(StreamsConfig.COMMIT_INTERVAL_MS_CONFIG, TimeUnit.SECONDS.toMillis(10));
    streamsConfiguration.put(ConsumerConfig.AUTO_OFFSET_RESET_CONFIG, "earliest");
    // Use a temporary directory for storing state, which will be automatically removed after the test.
    streamsConfiguration.put(StreamsConfig.STATE_DIR_CONFIG, TestUtils.tempDirectory().getAbsolutePath());
    // How long we "remember" an event.  During this time, any incoming duplicates of the event
    // will be, well, dropped, thereby de-duplicating the input data.
    // 
    // The actual value depends on your use case.  To reduce memory and disk usage, you could
    // decrease the size to purge old windows more frequently at the cost of potentially missing out
    // on de-duplicating late-arriving records.
    long maintainDurationPerEventInMs = TimeUnit.MINUTES.toMillis(10);
    // The number of segments has no impact on "correctness".
    // Using more segments implies larger overhead but allows for more fined grained record expiration
    // Note: the specified retention time is a _minimum_ time span and no strict upper time bound
    int numberOfSegments = 3;
    // retention period must be at least window size -- for this use case, we don't need a longer retention period
    // and thus just use the window size as retention time
    long retentionPeriod = maintainDurationPerEventInMs;
    StoreBuilder<WindowStore<String, Long>> dedupStoreBuilder = Stores.windowStoreBuilder(Stores.persistentWindowStore(storeName, retentionPeriod, numberOfSegments, maintainDurationPerEventInMs, false), Serdes.String(), Serdes.Long());
    builder.addStateStore(dedupStoreBuilder);
    KStream<byte[], String> input = builder.stream(inputTopic);
    KStream<byte[], String> deduplicated = input.transform(// function as needed.
    () -> new DeduplicationTransformer<>(maintainDurationPerEventInMs, (key, value) -> value), storeName);
    deduplicated.to(outputTopic);
    KafkaStreams streams = new KafkaStreams(builder.build(), streamsConfiguration);
    streams.start();
    // 
    // Step 2: Produce some input data to the input topic.
    // 
    Properties producerConfig = new Properties();
    producerConfig.put(ProducerConfig.BOOTSTRAP_SERVERS_CONFIG, CLUSTER.bootstrapServers());
    producerConfig.put(ProducerConfig.ACKS_CONFIG, "all");
    producerConfig.put(ProducerConfig.RETRIES_CONFIG, 0);
    producerConfig.put(ProducerConfig.KEY_SERIALIZER_CLASS_CONFIG, ByteArraySerializer.class);
    producerConfig.put(ProducerConfig.VALUE_SERIALIZER_CLASS_CONFIG, StringSerializer.class);
    IntegrationTestUtils.produceValuesSynchronously(inputTopic, inputValues, producerConfig);
    // 
    // Step 3: Verify the application's output data.
    // 
    Properties consumerConfig = new Properties();
    consumerConfig.put(ConsumerConfig.BOOTSTRAP_SERVERS_CONFIG, CLUSTER.bootstrapServers());
    consumerConfig.put(ConsumerConfig.GROUP_ID_CONFIG, "deduplication-integration-test-standard-consumer");
    consumerConfig.put(ConsumerConfig.AUTO_OFFSET_RESET_CONFIG, "earliest");
    consumerConfig.put(ConsumerConfig.KEY_DESERIALIZER_CLASS_CONFIG, ByteArrayDeserializer.class);
    consumerConfig.put(ConsumerConfig.VALUE_DESERIALIZER_CLASS_CONFIG, StringDeserializer.class);
    List<String> actualValues = IntegrationTestUtils.waitUntilMinValuesRecordsReceived(consumerConfig, outputTopic, expectedValues.size());
    streams.close();
    assertThat(actualValues).containsExactlyElementsOf(expectedValues);
}

Example 12 with KStream

use of org.apache.kafka.streams.kstream.KStream in project kafka-streams-examples by confluentinc.

the class HandlingCorruptedInputRecordsIntegrationTest method shouldIgnoreCorruptInputRecords.

@Test
public void shouldIgnoreCorruptInputRecords() throws Exception {
    List<Long> inputValues = Arrays.asList(1L, 2L, 3L);
    List<Long> expectedValues = inputValues.stream().map(x -> 2 * x).collect(Collectors.toList());
    // 
    // Step 1: Configure and start the processor topology.
    // 
    StreamsBuilder builder = new StreamsBuilder();
    Properties streamsConfiguration = new Properties();
    streamsConfiguration.put(StreamsConfig.APPLICATION_ID_CONFIG, "failure-handling-integration-test");
    streamsConfiguration.put(StreamsConfig.BOOTSTRAP_SERVERS_CONFIG, CLUSTER.bootstrapServers());
    streamsConfiguration.put(StreamsConfig.DEFAULT_KEY_SERDE_CLASS_CONFIG, Serdes.ByteArray().getClass().getName());
    streamsConfiguration.put(StreamsConfig.DEFAULT_VALUE_SERDE_CLASS_CONFIG, Serdes.ByteArray().getClass().getName());
    streamsConfiguration.put(ConsumerConfig.AUTO_OFFSET_RESET_CONFIG, "earliest");
    Serde<String> stringSerde = Serdes.String();
    Serde<Long> longSerde = Serdes.Long();
    KStream<byte[], byte[]> input = builder.stream(inputTopic);
    // Note how the returned stream is of type `KStream<String, Long>`.
    KStream<String, Long> doubled = input.flatMap((k, v) -> {
        try {
            // Attempt deserialization
            String key = stringSerde.deserializer().deserialize("input-topic", k);
            long value = longSerde.deserializer().deserialize("input-topic", v);
            // checking.
            return Collections.singletonList(KeyValue.pair(key, 2 * value));
        } catch (SerializationException e) {
            // Ignore/skip the corrupted record by catching the exception.
            // Optionally, we can log the fact that we did so:
            System.err.println("Could not deserialize record: " + e.getMessage());
        }
        return Collections.emptyList();
    });
    // Write the processing results (which was generated from valid records only) to Kafka.
    doubled.to(outputTopic, Produced.with(stringSerde, longSerde));
    KafkaStreams streams = new KafkaStreams(builder.build(), streamsConfiguration);
    streams.start();
    // 
    // Step 2: Produce some corrupt input data to the input topic.
    // 
    Properties producerConfigForCorruptRecords = new Properties();
    producerConfigForCorruptRecords.put(ProducerConfig.BOOTSTRAP_SERVERS_CONFIG, CLUSTER.bootstrapServers());
    producerConfigForCorruptRecords.put(ProducerConfig.ACKS_CONFIG, "all");
    producerConfigForCorruptRecords.put(ProducerConfig.RETRIES_CONFIG, 0);
    producerConfigForCorruptRecords.put(ProducerConfig.KEY_SERIALIZER_CLASS_CONFIG, ByteArraySerializer.class);
    producerConfigForCorruptRecords.put(ProducerConfig.VALUE_SERIALIZER_CLASS_CONFIG, StringSerializer.class);
    IntegrationTestUtils.produceValuesSynchronously(inputTopic, Collections.singletonList("corrupt"), producerConfigForCorruptRecords);
    // 
    // Step 3: Produce some (valid) input data to the input topic.
    // 
    Properties producerConfig = new Properties();
    producerConfig.put(ProducerConfig.BOOTSTRAP_SERVERS_CONFIG, CLUSTER.bootstrapServers());
    producerConfig.put(ProducerConfig.ACKS_CONFIG, "all");
    producerConfig.put(ProducerConfig.RETRIES_CONFIG, 0);
    producerConfig.put(ProducerConfig.KEY_SERIALIZER_CLASS_CONFIG, ByteArraySerializer.class);
    producerConfig.put(ProducerConfig.VALUE_SERIALIZER_CLASS_CONFIG, LongSerializer.class);
    IntegrationTestUtils.produceValuesSynchronously(inputTopic, inputValues, producerConfig);
    // 
    // Step 4: Verify the application's output data.
    // 
    Properties consumerConfig = new Properties();
    consumerConfig.put(ConsumerConfig.BOOTSTRAP_SERVERS_CONFIG, CLUSTER.bootstrapServers());
    consumerConfig.put(ConsumerConfig.GROUP_ID_CONFIG, "map-function-lambda-integration-test-standard-consumer");
    consumerConfig.put(ConsumerConfig.AUTO_OFFSET_RESET_CONFIG, "earliest");
    consumerConfig.put(ConsumerConfig.KEY_DESERIALIZER_CLASS_CONFIG, ByteArrayDeserializer.class);
    consumerConfig.put(ConsumerConfig.VALUE_DESERIALIZER_CLASS_CONFIG, LongDeserializer.class);
    List<Long> actualValues = IntegrationTestUtils.waitUntilMinValuesRecordsReceived(consumerConfig, outputTopic, expectedValues.size());
    streams.close();
    assertThat(actualValues).isEqualTo(expectedValues);
}

Example 13 with KStream

use of org.apache.kafka.streams.kstream.KStream in project kafka by apache.

the class KStreamMapTest method testMap.

@Test
public void testMap() {
    final StreamsBuilder builder = new StreamsBuilder();
    final String topicName = "topic";
    final int[] expectedKeys = new int[] { 0, 1, 2, 3 };
    final MockApiProcessorSupplier<String, Integer, Void, Void> supplier = new MockApiProcessorSupplier<>();
    final KStream<Integer, String> stream = builder.stream(topicName, Consumed.with(Serdes.Integer(), Serdes.String()));
    stream.map((key, value) -> KeyValue.pair(value, key)).process(supplier);
    try (final TopologyTestDriver driver = new TopologyTestDriver(builder.build(), props)) {
        for (final int expectedKey : expectedKeys) {
            final TestInputTopic<Integer, String> inputTopic = driver.createInputTopic(topicName, new IntegerSerializer(), new StringSerializer(), Instant.ofEpochMilli(0L), Duration.ZERO);
            inputTopic.pipeInput(expectedKey, "V" + expectedKey, 10L - expectedKey);
        }
    }
    final KeyValueTimestamp[] expected = new KeyValueTimestamp[] { new KeyValueTimestamp<>("V0", 0, 10), new KeyValueTimestamp<>("V1", 1, 9), new KeyValueTimestamp<>("V2", 2, 8), new KeyValueTimestamp<>("V3", 3, 7) };
    assertEquals(4, supplier.theCapturedProcessor().processed().size());
    for (int i = 0; i < expected.length; i++) {
        assertEquals(expected[i], supplier.theCapturedProcessor().processed().get(i));
    }
}

Example 14 with KStream

use of org.apache.kafka.streams.kstream.KStream in project kafka by apache.

the class KStreamMapValuesTest method testMapValuesWithKeys.

@Test
public void testMapValuesWithKeys() {
    final StreamsBuilder builder = new StreamsBuilder();
    final ValueMapperWithKey<Integer, CharSequence, Integer> mapper = (readOnlyKey, value) -> value.length() + readOnlyKey;
    final int[] expectedKeys = { 1, 10, 100, 1000 };
    final KStream<Integer, String> stream = builder.stream(topicName, Consumed.with(Serdes.Integer(), Serdes.String()));
    stream.mapValues(mapper).process(supplier);
    try (final TopologyTestDriver driver = new TopologyTestDriver(builder.build(), props)) {
        final TestInputTopic<Integer, String> inputTopic = driver.createInputTopic(topicName, new IntegerSerializer(), new StringSerializer());
        for (final int expectedKey : expectedKeys) {
            inputTopic.pipeInput(expectedKey, Integer.toString(expectedKey), expectedKey / 2L);
        }
    }
    final KeyValueTimestamp[] expected = { new KeyValueTimestamp<>(1, 2, 0), new KeyValueTimestamp<>(10, 12, 5), new KeyValueTimestamp<>(100, 103, 50), new KeyValueTimestamp<>(1000, 1004, 500) };
    assertArrayEquals(expected, supplier.theCapturedProcessor().processed().toArray());
}

Example 15 with KStream

use of org.apache.kafka.streams.kstream.KStream in project kafka by apache.

the class CogroupedKStreamImplTest method shouldInsertRepartitionsTopicForCogroupsUsedTwice.

@Test
public void shouldInsertRepartitionsTopicForCogroupsUsedTwice() {
    final StreamsBuilder builder = new StreamsBuilder();
    final Properties properties = new Properties();
    final KStream<String, String> stream1 = builder.stream("one", stringConsumed);
    final KGroupedStream<String, String> groupedOne = stream1.map((k, v) -> new KeyValue<>(v, k)).groupByKey(Grouped.as("foo"));
    final CogroupedKStream<String, String> one = groupedOne.cogroup(STRING_AGGREGATOR);
    one.aggregate(STRING_INITIALIZER);
    one.aggregate(STRING_INITIALIZER);
    final String topologyDescription = builder.build(properties).describe().toString();
    assertThat(topologyDescription, equalTo("Topologies:\n" + "   Sub-topology: 0\n" + "    Source: KSTREAM-SOURCE-0000000000 (topics: [one])\n" + "      --> KSTREAM-MAP-0000000001\n" + "    Processor: KSTREAM-MAP-0000000001 (stores: [])\n" + "      --> foo-repartition-filter\n" + "      <-- KSTREAM-SOURCE-0000000000\n" + "    Processor: foo-repartition-filter (stores: [])\n" + "      --> foo-repartition-sink\n" + "      <-- KSTREAM-MAP-0000000001\n" + "    Sink: foo-repartition-sink (topic: foo-repartition)\n" + "      <-- foo-repartition-filter\n\n" + "  Sub-topology: 1\n" + "    Source: foo-repartition-source (topics: [foo-repartition])\n" + "      --> COGROUPKSTREAM-AGGREGATE-0000000006, COGROUPKSTREAM-AGGREGATE-0000000012\n" + "    Processor: COGROUPKSTREAM-AGGREGATE-0000000006 (stores: [COGROUPKSTREAM-AGGREGATE-STATE-STORE-0000000002])\n" + "      --> COGROUPKSTREAM-MERGE-0000000007\n" + "      <-- foo-repartition-source\n" + "    Processor: COGROUPKSTREAM-AGGREGATE-0000000012 (stores: [COGROUPKSTREAM-AGGREGATE-STATE-STORE-0000000008])\n" + "      --> COGROUPKSTREAM-MERGE-0000000013\n" + "      <-- foo-repartition-source\n" + "    Processor: COGROUPKSTREAM-MERGE-0000000007 (stores: [])\n" + "      --> none\n" + "      <-- COGROUPKSTREAM-AGGREGATE-0000000006\n" + "    Processor: COGROUPKSTREAM-MERGE-0000000013 (stores: [])\n" + "      --> none\n" + "      <-- COGROUPKSTREAM-AGGREGATE-0000000012\n\n"));
}