org.apache.flink.streaming.connectors.kinesis.serialization.KinesisDeserializationSchemaWrapper Java Examples

@Test(expected = RuntimeException.class)
public void testIfNoShardsAreFoundShouldThrowException() throws Exception {
	List<String> fakeStreams = new LinkedList<>();
	fakeStreams.add("fakeStream1");
	fakeStreams.add("fakeStream2");

	HashMap<String, String> subscribedStreamsToLastSeenShardIdsUnderTest =
		KinesisDataFetcher.createInitialSubscribedStreamsToLastDiscoveredShardsState(fakeStreams);

	TestableKinesisDataFetcher<String> fetcher =
		new TestableKinesisDataFetcher<>(
			fakeStreams,
			new TestSourceContext<>(),
			TestUtils.getStandardProperties(),
			new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
			10,
			2,
			new AtomicReference<>(),
			new LinkedList<>(),
			subscribedStreamsToLastSeenShardIdsUnderTest,
			FakeKinesisBehavioursFactory.noShardsFoundForRequestedStreamsBehaviour());

	fetcher.runFetcher(); // this should throw RuntimeException
}
 

@Test(expected = RuntimeException.class)
public void testIfNoShardsAreFoundShouldThrowException() throws Exception {
	List<String> fakeStreams = new LinkedList<>();
	fakeStreams.add("fakeStream1");
	fakeStreams.add("fakeStream2");

	HashMap<String, String> subscribedStreamsToLastSeenShardIdsUnderTest =
		KinesisDataFetcher.createInitialSubscribedStreamsToLastDiscoveredShardsState(fakeStreams);

	TestableKinesisDataFetcher<String> fetcher =
		new TestableKinesisDataFetcher<>(
			fakeStreams,
			new TestSourceContext<>(),
			TestUtils.getStandardProperties(),
			new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
			10,
			2,
			new AtomicReference<>(),
			new LinkedList<>(),
			subscribedStreamsToLastSeenShardIdsUnderTest,
			FakeKinesisBehavioursFactory.noShardsFoundForRequestedStreamsBehaviour());

	fetcher.runFetcher(); // this should throw RuntimeException
}
 

@Test(expected = RuntimeException.class)
public void testIfNoShardsAreFoundShouldThrowException() throws Exception {
	List<String> fakeStreams = new LinkedList<>();
	fakeStreams.add("fakeStream1");
	fakeStreams.add("fakeStream2");

	HashMap<String, String> subscribedStreamsToLastSeenShardIdsUnderTest =
		KinesisDataFetcher.createInitialSubscribedStreamsToLastDiscoveredShardsState(fakeStreams);

	TestableKinesisDataFetcher<String> fetcher =
		new TestableKinesisDataFetcher<>(
			fakeStreams,
			new TestSourceContext<>(),
			TestUtils.getStandardProperties(),
			new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
			10,
			2,
			new AtomicReference<>(),
			new LinkedList<>(),
			subscribedStreamsToLastSeenShardIdsUnderTest,
			FakeKinesisBehavioursFactory.noShardsFoundForRequestedStreamsBehaviour());

	fetcher.runFetcher(); // this should throw RuntimeException
}
 

@Test
public void testMetricsReporting() {
	StreamShardHandle fakeToBeConsumedShard = getMockStreamShard("fakeStream", 0);

	LinkedList<KinesisStreamShardState> subscribedShardsStateUnderTest = new LinkedList<>();
	subscribedShardsStateUnderTest.add(
		new KinesisStreamShardState(
			KinesisDataFetcher.convertToStreamShardMetadata(fakeToBeConsumedShard),
			fakeToBeConsumedShard,
			new SequenceNumber("fakeStartingState")));

	TestSourceContext<String> sourceContext = new TestSourceContext<>();

	TestableKinesisDataFetcher<String> fetcher =
		new TestableKinesisDataFetcher<>(
			Collections.singletonList("fakeStream"),
			sourceContext,
			new Properties(),
			new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
			10,
			2,
			new AtomicReference<>(),
			subscribedShardsStateUnderTest,
			KinesisDataFetcher.createInitialSubscribedStreamsToLastDiscoveredShardsState(Collections.singletonList("fakeStream")),
			Mockito.mock(KinesisProxyInterface.class));

	ShardMetricsReporter shardMetricsReporter = new ShardMetricsReporter();
	long millisBehindLatest = 500L;
	new ShardConsumer<>(
		fetcher,
		0,
		subscribedShardsStateUnderTest.get(0).getStreamShardHandle(),
		subscribedShardsStateUnderTest.get(0).getLastProcessedSequenceNum(),
		FakeKinesisBehavioursFactory.totalNumOfRecordsAfterNumOfGetRecordsCalls(1000, 9, millisBehindLatest),
		shardMetricsReporter).run();

	// the millisBehindLatest metric should have been reported
	assertEquals(millisBehindLatest, shardMetricsReporter.getMillisBehindLatest());
}
 

@Test
public void testCorrectNumOfCollectedRecordsAndUpdatedState() {
	StreamShardHandle fakeToBeConsumedShard = getMockStreamShard("fakeStream", 0);

	LinkedList<KinesisStreamShardState> subscribedShardsStateUnderTest = new LinkedList<>();
	subscribedShardsStateUnderTest.add(
		new KinesisStreamShardState(KinesisDataFetcher.convertToStreamShardMetadata(fakeToBeConsumedShard),
			fakeToBeConsumedShard, new SequenceNumber("fakeStartingState")));

	TestSourceContext<String> sourceContext = new TestSourceContext<>();

	TestableKinesisDataFetcher<String> fetcher =
		new TestableKinesisDataFetcher<>(
			Collections.singletonList("fakeStream"),
			sourceContext,
			new Properties(),
			new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
			10,
			2,
			new AtomicReference<>(),
			subscribedShardsStateUnderTest,
			KinesisDataFetcher.createInitialSubscribedStreamsToLastDiscoveredShardsState(Collections.singletonList("fakeStream")),
			Mockito.mock(KinesisProxyInterface.class));

	int shardIndex = fetcher.registerNewSubscribedShardState(subscribedShardsStateUnderTest.get(0));
	new ShardConsumer<>(
		fetcher,
		shardIndex,
		subscribedShardsStateUnderTest.get(0).getStreamShardHandle(),
		subscribedShardsStateUnderTest.get(0).getLastProcessedSequenceNum(),
		FakeKinesisBehavioursFactory.totalNumOfRecordsAfterNumOfGetRecordsCalls(1000, 9, 500L),
		new ShardMetricsReporter()).run();

	assertEquals(1000, sourceContext.getCollectedOutputs().size());
	assertEquals(
		SentinelSequenceNumber.SENTINEL_SHARD_ENDING_SEQUENCE_NUM.get(),
		subscribedShardsStateUnderTest.get(0).getLastProcessedSequenceNum());
}
 

@Test
public void testMetricsReporting() {
	StreamShardHandle fakeToBeConsumedShard = getMockStreamShard("fakeStream", 0);

	LinkedList<KinesisStreamShardState> subscribedShardsStateUnderTest = new LinkedList<>();
	subscribedShardsStateUnderTest.add(
		new KinesisStreamShardState(
			KinesisDataFetcher.convertToStreamShardMetadata(fakeToBeConsumedShard),
			fakeToBeConsumedShard,
			new SequenceNumber("fakeStartingState")));

	TestSourceContext<String> sourceContext = new TestSourceContext<>();

	TestableKinesisDataFetcher<String> fetcher =
		new TestableKinesisDataFetcher<>(
			Collections.singletonList("fakeStream"),
			sourceContext,
			new Properties(),
			new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
			10,
			2,
			new AtomicReference<>(),
			subscribedShardsStateUnderTest,
			KinesisDataFetcher.createInitialSubscribedStreamsToLastDiscoveredShardsState(Collections.singletonList("fakeStream")),
			Mockito.mock(KinesisProxyInterface.class));

	ShardMetricsReporter shardMetricsReporter = new ShardMetricsReporter();
	long millisBehindLatest = 500L;
	new ShardConsumer<>(
		fetcher,
		0,
		subscribedShardsStateUnderTest.get(0).getStreamShardHandle(),
		subscribedShardsStateUnderTest.get(0).getLastProcessedSequenceNum(),
		FakeKinesisBehavioursFactory.totalNumOfRecordsAfterNumOfGetRecordsCalls(1000, 9, millisBehindLatest),
		shardMetricsReporter).run();

	// the millisBehindLatest metric should have been reported
	assertEquals(millisBehindLatest, shardMetricsReporter.getMillisBehindLatest());
}
 

@Test
public void testCorrectNumOfCollectedRecordsAndUpdatedState() {
	StreamShardHandle fakeToBeConsumedShard = getMockStreamShard("fakeStream", 0);

	LinkedList<KinesisStreamShardState> subscribedShardsStateUnderTest = new LinkedList<>();
	subscribedShardsStateUnderTest.add(
		new KinesisStreamShardState(KinesisDataFetcher.convertToStreamShardMetadata(fakeToBeConsumedShard),
			fakeToBeConsumedShard, new SequenceNumber("fakeStartingState")));

	TestSourceContext<String> sourceContext = new TestSourceContext<>();

	TestableKinesisDataFetcher<String> fetcher =
		new TestableKinesisDataFetcher<>(
			Collections.singletonList("fakeStream"),
			sourceContext,
			new Properties(),
			new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
			10,
			2,
			new AtomicReference<>(),
			subscribedShardsStateUnderTest,
			KinesisDataFetcher.createInitialSubscribedStreamsToLastDiscoveredShardsState(Collections.singletonList("fakeStream")),
			Mockito.mock(KinesisProxyInterface.class));

	int shardIndex = fetcher.registerNewSubscribedShardState(subscribedShardsStateUnderTest.get(0));
	new ShardConsumer<>(
		fetcher,
		shardIndex,
		subscribedShardsStateUnderTest.get(0).getStreamShardHandle(),
		subscribedShardsStateUnderTest.get(0).getLastProcessedSequenceNum(),
		FakeKinesisBehavioursFactory.totalNumOfRecordsAfterNumOfGetRecordsCalls(1000, 9, 500L),
		new ShardMetricsReporter()).run();

	assertEquals(1000, sourceContext.getCollectedOutputs().size());
	assertEquals(
		SentinelSequenceNumber.SENTINEL_SHARD_ENDING_SEQUENCE_NUM.get(),
		subscribedShardsStateUnderTest.get(0).getLastProcessedSequenceNum());
}
 

@Test
public void testCorrectNumOfCollectedRecordsAndUpdatedStateWithUnexpectedExpiredIterator() {
	StreamShardHandle fakeToBeConsumedShard = getMockStreamShard("fakeStream", 0);

	LinkedList<KinesisStreamShardState> subscribedShardsStateUnderTest = new LinkedList<>();
	subscribedShardsStateUnderTest.add(
		new KinesisStreamShardState(KinesisDataFetcher.convertToStreamShardMetadata(fakeToBeConsumedShard),
			fakeToBeConsumedShard, new SequenceNumber("fakeStartingState")));

	TestSourceContext<String> sourceContext = new TestSourceContext<>();

	KinesisDeserializationSchemaWrapper<String> deserializationSchema = new KinesisDeserializationSchemaWrapper<>(
		new SimpleStringSchema());
	TestableKinesisDataFetcher<String> fetcher =
		new TestableKinesisDataFetcher<>(
			Collections.singletonList("fakeStream"),
			sourceContext,
			new Properties(),
			deserializationSchema,
			10,
			2,
			new AtomicReference<>(),
			subscribedShardsStateUnderTest,
			KinesisDataFetcher.createInitialSubscribedStreamsToLastDiscoveredShardsState(Collections.singletonList("fakeStream")),
			Mockito.mock(KinesisProxyInterface.class));

	int shardIndex = fetcher.registerNewSubscribedShardState(subscribedShardsStateUnderTest.get(0));
	new ShardConsumer<>(
		fetcher,
		shardIndex,
		subscribedShardsStateUnderTest.get(0).getStreamShardHandle(),
		subscribedShardsStateUnderTest.get(0).getLastProcessedSequenceNum(),
		// Get a total of 1000 records with 9 getRecords() calls,
		// and the 7th getRecords() call will encounter an unexpected expired shard iterator
		FakeKinesisBehavioursFactory.totalNumOfRecordsAfterNumOfGetRecordsCallsWithUnexpectedExpiredIterator(
			1000, 9, 7, 500L),
		new ShardMetricsReporter(),
		deserializationSchema)
		.run();

	assertEquals(1000, sourceContext.getCollectedOutputs().size());
	assertEquals(
		SentinelSequenceNumber.SENTINEL_SHARD_ENDING_SEQUENCE_NUM.get(),
		subscribedShardsStateUnderTest.get(0).getLastProcessedSequenceNum());
}
 

@Test
public void testShardToSubtaskMappingWithCustomHashFunction() throws Exception {

	int totalCountOfSubtasks = 10;
	int shardCount = 3;

	for (int i = 0; i < 2; i++) {

		final int hash = i;
		final KinesisShardAssigner allShardsSingleSubtaskFn = (shard, subtasks) -> hash;
		Map<String, Integer> streamToShardCount = new HashMap<>();
		List<String> fakeStreams = new LinkedList<>();
		fakeStreams.add("fakeStream");
		streamToShardCount.put("fakeStream", shardCount);

		for (int j = 0; j < totalCountOfSubtasks; j++) {

			int subtaskIndex = j;
			// subscribe with default hashing
			final TestableKinesisDataFetcher fetcher =
				new TestableKinesisDataFetcher(
					fakeStreams,
					new TestSourceContext<>(),
					new Properties(),
					new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
					totalCountOfSubtasks,
					subtaskIndex,
					new AtomicReference<>(),
					new LinkedList<>(),
					KinesisDataFetcher.createInitialSubscribedStreamsToLastDiscoveredShardsState(fakeStreams),
					FakeKinesisBehavioursFactory.nonReshardedStreamsBehaviour(streamToShardCount));
			Whitebox.setInternalState(fetcher, "shardAssigner", allShardsSingleSubtaskFn); // override hashing
			List<StreamShardHandle> shards = fetcher.discoverNewShardsToSubscribe();
			fetcher.shutdownFetcher();

			String msg = String.format("for hash=%d, subtask=%d", hash, subtaskIndex);
			if (j == i) {
				assertEquals(msg, shardCount, shards.size());
			} else {
				assertEquals(msg, 0, shards.size());
			}
		}

	}

}
 

@Test
public void testCorrectNumOfCollectedRecordsAndUpdatedState() {
	StreamShardHandle fakeToBeConsumedShard = getMockStreamShard("fakeStream", 0);

	LinkedList<KinesisStreamShardState> subscribedShardsStateUnderTest = new LinkedList<>();
	subscribedShardsStateUnderTest.add(
		new KinesisStreamShardState(KinesisDataFetcher.convertToStreamShardMetadata(fakeToBeConsumedShard),
			fakeToBeConsumedShard, new SequenceNumber("fakeStartingState")));

	TestSourceContext<String> sourceContext = new TestSourceContext<>();

	KinesisDeserializationSchemaWrapper<String> deserializationSchema = new KinesisDeserializationSchemaWrapper<>(
		new SimpleStringSchema());
	TestableKinesisDataFetcher<String> fetcher =
		new TestableKinesisDataFetcher<>(
			Collections.singletonList("fakeStream"),
			sourceContext,
			new Properties(),
			deserializationSchema,
			10,
			2,
			new AtomicReference<>(),
			subscribedShardsStateUnderTest,
			KinesisDataFetcher.createInitialSubscribedStreamsToLastDiscoveredShardsState(Collections.singletonList("fakeStream")),
			Mockito.mock(KinesisProxyInterface.class));

	int shardIndex = fetcher.registerNewSubscribedShardState(subscribedShardsStateUnderTest.get(0));
	new ShardConsumer<>(
		fetcher,
		shardIndex,
		subscribedShardsStateUnderTest.get(0).getStreamShardHandle(),
		subscribedShardsStateUnderTest.get(0).getLastProcessedSequenceNum(),
		FakeKinesisBehavioursFactory.totalNumOfRecordsAfterNumOfGetRecordsCalls(1000, 9, 500L),
		new ShardMetricsReporter(),
		deserializationSchema)
		.run();

	assertEquals(1000, sourceContext.getCollectedOutputs().size());
	assertEquals(
		SentinelSequenceNumber.SENTINEL_SHARD_ENDING_SEQUENCE_NUM.get(),
		subscribedShardsStateUnderTest.get(0).getLastProcessedSequenceNum());
}
 

@Test
public void testMetricsReporting() {
	StreamShardHandle fakeToBeConsumedShard = getMockStreamShard("fakeStream", 0);

	LinkedList<KinesisStreamShardState> subscribedShardsStateUnderTest = new LinkedList<>();
	subscribedShardsStateUnderTest.add(
		new KinesisStreamShardState(
			KinesisDataFetcher.convertToStreamShardMetadata(fakeToBeConsumedShard),
			fakeToBeConsumedShard,
			new SequenceNumber("fakeStartingState")));

	TestSourceContext<String> sourceContext = new TestSourceContext<>();

	KinesisDeserializationSchemaWrapper<String> deserializationSchema = new KinesisDeserializationSchemaWrapper<>(
		new SimpleStringSchema());
	TestableKinesisDataFetcher<String> fetcher =
		new TestableKinesisDataFetcher<>(
			Collections.singletonList("fakeStream"),
			sourceContext,
			new Properties(),
			deserializationSchema,
			10,
			2,
			new AtomicReference<>(),
			subscribedShardsStateUnderTest,
			KinesisDataFetcher.createInitialSubscribedStreamsToLastDiscoveredShardsState(Collections.singletonList("fakeStream")),
			Mockito.mock(KinesisProxyInterface.class));

	ShardMetricsReporter shardMetricsReporter = new ShardMetricsReporter();
	long millisBehindLatest = 500L;
	new ShardConsumer<>(
		fetcher,
		0,
		subscribedShardsStateUnderTest.get(0).getStreamShardHandle(),
		subscribedShardsStateUnderTest.get(0).getLastProcessedSequenceNum(),
		FakeKinesisBehavioursFactory.totalNumOfRecordsAfterNumOfGetRecordsCalls(1000, 9, millisBehindLatest),
		shardMetricsReporter,
		deserializationSchema)
		.run();

	// the millisBehindLatest metric should have been reported
	assertEquals(millisBehindLatest, shardMetricsReporter.getMillisBehindLatest());
}
 

@SuppressWarnings("unchecked")
private void writeSnapshot(String path, HashMap<StreamShardMetadata, SequenceNumber> state) throws Exception {
	final List<StreamShardHandle> initialDiscoveryShards = new ArrayList<>(state.size());
	for (StreamShardMetadata shardMetadata : state.keySet()) {
		Shard shard = new Shard();
		shard.setShardId(shardMetadata.getShardId());

		SequenceNumberRange sequenceNumberRange = new SequenceNumberRange();
		sequenceNumberRange.withStartingSequenceNumber("1");
		shard.setSequenceNumberRange(sequenceNumberRange);

		initialDiscoveryShards.add(new StreamShardHandle(shardMetadata.getStreamName(), shard));
	}

	final TestFetcher<String> fetcher = new TestFetcher<>(
		Collections.singletonList(TEST_STREAM_NAME),
		new TestSourceContext<>(),
		new TestRuntimeContext(true, 1, 0),
		TestUtils.getStandardProperties(),
		new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
		state,
		initialDiscoveryShards);

	final DummyFlinkKinesisConsumer<String> consumer = new DummyFlinkKinesisConsumer<>(
		fetcher, new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()));

	StreamSource<String, DummyFlinkKinesisConsumer<String>> consumerOperator = new StreamSource<>(consumer);

	final AbstractStreamOperatorTestHarness<String> testHarness =
			new AbstractStreamOperatorTestHarness<>(consumerOperator, 1, 1, 0);

	testHarness.setTimeCharacteristic(TimeCharacteristic.ProcessingTime);

	testHarness.setup();
	testHarness.open();

	final AtomicReference<Throwable> error = new AtomicReference<>();

	// run the source asynchronously
	Thread runner = new Thread() {
		@Override
		public void run() {
			try {
				consumer.run(new TestSourceContext<>());
			} catch (Throwable t) {
				t.printStackTrace();
				error.set(t);
			}
		}
	};
	runner.start();

	fetcher.waitUntilRun();

	final OperatorSubtaskState snapshot;
	synchronized (testHarness.getCheckpointLock()) {
		snapshot = testHarness.snapshot(0L, 0L);
	}

	OperatorSnapshotUtil.writeStateHandle(snapshot, path);

	consumerOperator.close();
	runner.join();
}
 

@Test
public void testRestoreWithReshardedStream() throws Exception {
	final List<StreamShardHandle> initialDiscoveryShards = new ArrayList<>(TEST_STATE.size());
	for (StreamShardMetadata shardMetadata : TEST_STATE.keySet()) {
		// setup the closed shard
		Shard closedShard = new Shard();
		closedShard.setShardId(shardMetadata.getShardId());

		SequenceNumberRange closedSequenceNumberRange = new SequenceNumberRange();
		closedSequenceNumberRange.withStartingSequenceNumber("1");
		closedSequenceNumberRange.withEndingSequenceNumber("1087654321"); // this represents a closed shard
		closedShard.setSequenceNumberRange(closedSequenceNumberRange);

		initialDiscoveryShards.add(new StreamShardHandle(shardMetadata.getStreamName(), closedShard));

		// setup the new shards
		Shard newSplitShard1 = new Shard();
		newSplitShard1.setShardId(KinesisShardIdGenerator.generateFromShardOrder(1));

		SequenceNumberRange newSequenceNumberRange1 = new SequenceNumberRange();
		newSequenceNumberRange1.withStartingSequenceNumber("1087654322");
		newSplitShard1.setSequenceNumberRange(newSequenceNumberRange1);

		newSplitShard1.setParentShardId(TEST_SHARD_ID);

		Shard newSplitShard2 = new Shard();
		newSplitShard2.setShardId(KinesisShardIdGenerator.generateFromShardOrder(2));

		SequenceNumberRange newSequenceNumberRange2 = new SequenceNumberRange();
		newSequenceNumberRange2.withStartingSequenceNumber("2087654322");
		newSplitShard2.setSequenceNumberRange(newSequenceNumberRange2);

		newSplitShard2.setParentShardId(TEST_SHARD_ID);

		initialDiscoveryShards.add(new StreamShardHandle(shardMetadata.getStreamName(), newSplitShard1));
		initialDiscoveryShards.add(new StreamShardHandle(shardMetadata.getStreamName(), newSplitShard2));
	}

	final TestFetcher<String> fetcher = new TestFetcher<>(
		Collections.singletonList(TEST_STREAM_NAME),
		new TestSourceContext<>(),
		new TestRuntimeContext(true, 1, 0),
		TestUtils.getStandardProperties(),
		new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
		null,
		initialDiscoveryShards);

	final DummyFlinkKinesisConsumer<String> consumerFunction = new DummyFlinkKinesisConsumer<>(
		fetcher, new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()));

	StreamSource<String, DummyFlinkKinesisConsumer<String>> consumerOperator =
		new StreamSource<>(consumerFunction);

	final AbstractStreamOperatorTestHarness<String> testHarness =
		new AbstractStreamOperatorTestHarness<>(consumerOperator, 1, 1, 0);

	testHarness.setup();
	testHarness.initializeState(
		OperatorSnapshotUtil.getResourceFilename(
			"kinesis-consumer-migration-test-flink" + testMigrateVersion + "-snapshot"));
	testHarness.open();

	consumerFunction.run(new TestSourceContext<>());

	// assert that state is correctly restored
	assertNotEquals(null, consumerFunction.getRestoredState());
	assertEquals(1, consumerFunction.getRestoredState().size());
	assertEquals(TEST_STATE, removeEquivalenceWrappers(consumerFunction.getRestoredState()));

	// assert that the fetcher is registered with all shards, including new shards
	assertEquals(3, fetcher.getSubscribedShardsState().size());

	KinesisStreamShardState restoredClosedShardState = fetcher.getSubscribedShardsState().get(0);
	assertEquals(TEST_STREAM_NAME, restoredClosedShardState.getStreamShardHandle().getStreamName());
	assertEquals(TEST_SHARD_ID, restoredClosedShardState.getStreamShardHandle().getShard().getShardId());
	assertTrue(restoredClosedShardState.getStreamShardHandle().isClosed());
	assertEquals(TEST_SEQUENCE_NUMBER, restoredClosedShardState.getLastProcessedSequenceNum());

	KinesisStreamShardState restoredNewSplitShard1 = fetcher.getSubscribedShardsState().get(1);
	assertEquals(TEST_STREAM_NAME, restoredNewSplitShard1.getStreamShardHandle().getStreamName());
	assertEquals(KinesisShardIdGenerator.generateFromShardOrder(1), restoredNewSplitShard1.getStreamShardHandle().getShard().getShardId());
	assertFalse(restoredNewSplitShard1.getStreamShardHandle().isClosed());
	// new shards should be consumed from the beginning
	assertEquals(SentinelSequenceNumber.SENTINEL_EARLIEST_SEQUENCE_NUM.get(), restoredNewSplitShard1.getLastProcessedSequenceNum());

	KinesisStreamShardState restoredNewSplitShard2 = fetcher.getSubscribedShardsState().get(2);
	assertEquals(TEST_STREAM_NAME, restoredNewSplitShard2.getStreamShardHandle().getStreamName());
	assertEquals(KinesisShardIdGenerator.generateFromShardOrder(2), restoredNewSplitShard2.getStreamShardHandle().getShard().getShardId());
	assertFalse(restoredNewSplitShard2.getStreamShardHandle().isClosed());
	// new shards should be consumed from the beginning
	assertEquals(SentinelSequenceNumber.SENTINEL_EARLIEST_SEQUENCE_NUM.get(), restoredNewSplitShard2.getLastProcessedSequenceNum());

	consumerOperator.close();
	consumerOperator.cancel();
}
 

@Test
public void testRestore() throws Exception {
	final List<StreamShardHandle> initialDiscoveryShards = new ArrayList<>(TEST_STATE.size());
	for (StreamShardMetadata shardMetadata : TEST_STATE.keySet()) {
		Shard shard = new Shard();
		shard.setShardId(shardMetadata.getShardId());

		SequenceNumberRange sequenceNumberRange = new SequenceNumberRange();
		sequenceNumberRange.withStartingSequenceNumber("1");
		shard.setSequenceNumberRange(sequenceNumberRange);

		initialDiscoveryShards.add(new StreamShardHandle(shardMetadata.getStreamName(), shard));
	}

	final TestFetcher<String> fetcher = new TestFetcher<>(
		Collections.singletonList(TEST_STREAM_NAME),
		new TestSourceContext<>(),
		new TestRuntimeContext(true, 1, 0),
		TestUtils.getStandardProperties(),
		new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
		null,
		initialDiscoveryShards);

	final DummyFlinkKinesisConsumer<String> consumerFunction = new DummyFlinkKinesisConsumer<>(
		fetcher, new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()));

	StreamSource<String, DummyFlinkKinesisConsumer<String>> consumerOperator =
		new StreamSource<>(consumerFunction);

	final AbstractStreamOperatorTestHarness<String> testHarness =
		new AbstractStreamOperatorTestHarness<>(consumerOperator, 1, 1, 0);

	testHarness.setup();
	testHarness.initializeState(
		OperatorSnapshotUtil.getResourceFilename(
			"kinesis-consumer-migration-test-flink" + testMigrateVersion + "-snapshot"));
	testHarness.open();

	consumerFunction.run(new TestSourceContext<>());

	// assert that state is correctly restored
	assertNotEquals(null, consumerFunction.getRestoredState());
	assertEquals(1, consumerFunction.getRestoredState().size());
	assertEquals(TEST_STATE, removeEquivalenceWrappers(consumerFunction.getRestoredState()));
	assertEquals(1, fetcher.getSubscribedShardsState().size());
	assertEquals(TEST_SEQUENCE_NUMBER, fetcher.getSubscribedShardsState().get(0).getLastProcessedSequenceNum());

	KinesisStreamShardState restoredShardState = fetcher.getSubscribedShardsState().get(0);
	assertEquals(TEST_STREAM_NAME, restoredShardState.getStreamShardHandle().getStreamName());
	assertEquals(TEST_SHARD_ID, restoredShardState.getStreamShardHandle().getShard().getShardId());
	assertFalse(restoredShardState.getStreamShardHandle().isClosed());
	assertEquals(TEST_SEQUENCE_NUMBER, restoredShardState.getLastProcessedSequenceNum());

	consumerOperator.close();
	consumerOperator.cancel();
}
 

@Test
public void testRestoreWithEmptyState() throws Exception {
	final List<StreamShardHandle> initialDiscoveryShards = new ArrayList<>(TEST_STATE.size());
	for (StreamShardMetadata shardMetadata : TEST_STATE.keySet()) {
		Shard shard = new Shard();
		shard.setShardId(shardMetadata.getShardId());

		SequenceNumberRange sequenceNumberRange = new SequenceNumberRange();
		sequenceNumberRange.withStartingSequenceNumber("1");
		shard.setSequenceNumberRange(sequenceNumberRange);

		initialDiscoveryShards.add(new StreamShardHandle(shardMetadata.getStreamName(), shard));
	}

	final TestFetcher<String> fetcher = new TestFetcher<>(
		Collections.singletonList(TEST_STREAM_NAME),
		new TestSourceContext<>(),
		new TestRuntimeContext(true, 1, 0),
		TestUtils.getStandardProperties(),
		new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
		null,
		initialDiscoveryShards);

	final DummyFlinkKinesisConsumer<String> consumerFunction = new DummyFlinkKinesisConsumer<>(
		fetcher, new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()));

	StreamSource<String, DummyFlinkKinesisConsumer<String>> consumerOperator = new StreamSource<>(consumerFunction);

	final AbstractStreamOperatorTestHarness<String> testHarness =
		new AbstractStreamOperatorTestHarness<>(consumerOperator, 1, 1, 0);

	testHarness.setup();
	testHarness.initializeState(
		OperatorSnapshotUtil.getResourceFilename(
			"kinesis-consumer-migration-test-flink" + testMigrateVersion + "-empty-snapshot"));
	testHarness.open();

	consumerFunction.run(new TestSourceContext<>());

	// assert that no state was restored
	assertTrue(consumerFunction.getRestoredState().isEmpty());

	// although the restore state is empty, the fetcher should still have been registered the initial discovered shard;
	// furthermore, the discovered shard should be considered a newly created shard while the job wasn't running,
	// and therefore should be consumed from the earliest sequence number
	KinesisStreamShardState restoredShardState = fetcher.getSubscribedShardsState().get(0);
	assertEquals(TEST_STREAM_NAME, restoredShardState.getStreamShardHandle().getStreamName());
	assertEquals(TEST_SHARD_ID, restoredShardState.getStreamShardHandle().getShard().getShardId());
	assertFalse(restoredShardState.getStreamShardHandle().isClosed());
	assertEquals(SentinelSequenceNumber.SENTINEL_EARLIEST_SEQUENCE_NUM.get(), restoredShardState.getLastProcessedSequenceNum());

	consumerOperator.close();
	consumerOperator.cancel();
}
 

@Test
public void testPeriodicWatermark() {
	final MutableLong clock = new MutableLong();
	final MutableBoolean isTemporaryIdle = new MutableBoolean();
	final List<Watermark> watermarks = new ArrayList<>();

	String fakeStream1 = "fakeStream1";
	StreamShardHandle shardHandle =
		new StreamShardHandle(
			fakeStream1,
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(0)));

	TestSourceContext<String> sourceContext =
		new TestSourceContext<String>() {
			@Override
			public void emitWatermark(Watermark mark) {
				watermarks.add(mark);
			}

			@Override
			public void markAsTemporarilyIdle() {
				isTemporaryIdle.setTrue();
			}
		};

	HashMap<String, String> subscribedStreamsToLastSeenShardIdsUnderTest = new HashMap<>();

	final KinesisDataFetcher<String> fetcher =
		new TestableKinesisDataFetcher<String>(
			Collections.singletonList(fakeStream1),
			sourceContext,
			new java.util.Properties(),
			new KinesisDeserializationSchemaWrapper<>(new org.apache.flink.streaming.util.serialization.SimpleStringSchema()),
			1,
			1,
			new AtomicReference<>(),
			new LinkedList<>(),
			subscribedStreamsToLastSeenShardIdsUnderTest,
			FakeKinesisBehavioursFactory.nonReshardedStreamsBehaviour(new HashMap<>())) {

			@Override
			protected long getCurrentTimeMillis() {
				return clock.getValue();
			}
		};
	Whitebox.setInternalState(fetcher, "periodicWatermarkAssigner", watermarkAssigner);

	SequenceNumber seq = new SequenceNumber("fakeSequenceNumber");
	// register shards to subsequently emit records
	int shardIndex =
		fetcher.registerNewSubscribedShardState(
			new KinesisStreamShardState(
				KinesisDataFetcher.convertToStreamShardMetadata(shardHandle), shardHandle, seq));

	StreamRecord<String> record1 =
		new StreamRecord<>(String.valueOf(Long.MIN_VALUE), Long.MIN_VALUE);
	fetcher.emitRecordAndUpdateState(record1.getValue(), record1.getTimestamp(), shardIndex, seq);
	Assert.assertEquals(record1, sourceContext.getCollectedOutputs().poll());

	fetcher.emitWatermark();
	Assert.assertTrue("potential watermark equals previous watermark", watermarks.isEmpty());

	StreamRecord<String> record2 = new StreamRecord<>(String.valueOf(1), 1);
	fetcher.emitRecordAndUpdateState(record2.getValue(), record2.getTimestamp(), shardIndex, seq);
	Assert.assertEquals(record2, sourceContext.getCollectedOutputs().poll());

	fetcher.emitWatermark();
	Assert.assertFalse("watermark advanced", watermarks.isEmpty());
	Assert.assertEquals(new Watermark(record2.getTimestamp()), watermarks.remove(0));
	Assert.assertFalse("not idle", isTemporaryIdle.booleanValue());

	// test idle timeout
	long idleTimeout = 10;
	// advance clock idleTimeout
	clock.add(idleTimeout + 1);
	fetcher.emitWatermark();
	Assert.assertFalse("not idle", isTemporaryIdle.booleanValue());
	Assert.assertTrue("not idle, no new watermark", watermarks.isEmpty());

	// activate idle timeout
	Whitebox.setInternalState(fetcher, "shardIdleIntervalMillis", idleTimeout);
	fetcher.emitWatermark();
	Assert.assertTrue("idle", isTemporaryIdle.booleanValue());
	Assert.assertTrue("idle, no watermark", watermarks.isEmpty());
}
 

@SuppressWarnings("unchecked")
private void writeSnapshot(String path, HashMap<StreamShardMetadata, SequenceNumber> state) throws Exception {
	final List<StreamShardHandle> initialDiscoveryShards = new ArrayList<>(state.size());
	for (StreamShardMetadata shardMetadata : state.keySet()) {
		Shard shard = new Shard();
		shard.setShardId(shardMetadata.getShardId());

		SequenceNumberRange sequenceNumberRange = new SequenceNumberRange();
		sequenceNumberRange.withStartingSequenceNumber("1");
		shard.setSequenceNumberRange(sequenceNumberRange);

		initialDiscoveryShards.add(new StreamShardHandle(shardMetadata.getStreamName(), shard));
	}

	final TestFetcher<String> fetcher = new TestFetcher<>(
		Collections.singletonList(TEST_STREAM_NAME),
		new TestSourceContext<>(),
		new TestRuntimeContext(true, 1, 0),
		TestUtils.getStandardProperties(),
		new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
		state,
		initialDiscoveryShards);

	final DummyFlinkKinesisConsumer<String> consumer = new DummyFlinkKinesisConsumer<>(
		fetcher, new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()));

	StreamSource<String, DummyFlinkKinesisConsumer<String>> consumerOperator = new StreamSource<>(consumer);

	final AbstractStreamOperatorTestHarness<String> testHarness =
			new AbstractStreamOperatorTestHarness<>(consumerOperator, 1, 1, 0);

	testHarness.setTimeCharacteristic(TimeCharacteristic.ProcessingTime);

	testHarness.setup();
	testHarness.open();

	final AtomicReference<Throwable> error = new AtomicReference<>();

	// run the source asynchronously
	Thread runner = new Thread() {
		@Override
		public void run() {
			try {
				consumer.run(new TestSourceContext<>());
			} catch (Throwable t) {
				t.printStackTrace();
				error.set(t);
			}
		}
	};
	runner.start();

	fetcher.waitUntilRun();

	final OperatorSubtaskState snapshot;
	synchronized (testHarness.getCheckpointLock()) {
		snapshot = testHarness.snapshot(0L, 0L);
	}

	OperatorSnapshotUtil.writeStateHandle(snapshot, path);

	consumerOperator.close();
	runner.join();
}
 

@Test
public void testCorrectNumOfCollectedRecordsAndUpdatedStateWithAdaptiveReads() {
	Properties consumerProperties = new Properties();
	consumerProperties.put("flink.shard.adaptivereads", "true");

	StreamShardHandle fakeToBeConsumedShard = getMockStreamShard("fakeStream", 0);

	LinkedList<KinesisStreamShardState> subscribedShardsStateUnderTest = new LinkedList<>();
	subscribedShardsStateUnderTest.add(
		new KinesisStreamShardState(KinesisDataFetcher.convertToStreamShardMetadata(fakeToBeConsumedShard),
			fakeToBeConsumedShard, new SequenceNumber("fakeStartingState")));

	TestSourceContext<String> sourceContext = new TestSourceContext<>();

	KinesisDeserializationSchemaWrapper<String> deserializationSchema = new KinesisDeserializationSchemaWrapper<>(
		new SimpleStringSchema());
	TestableKinesisDataFetcher<String> fetcher =
		new TestableKinesisDataFetcher<>(
			Collections.singletonList("fakeStream"),
			sourceContext,
			consumerProperties,
			deserializationSchema,
			10,
			2,
			new AtomicReference<>(),
			subscribedShardsStateUnderTest,
			KinesisDataFetcher.createInitialSubscribedStreamsToLastDiscoveredShardsState(Collections.singletonList("fakeStream")),
			Mockito.mock(KinesisProxyInterface.class));

	int shardIndex = fetcher.registerNewSubscribedShardState(subscribedShardsStateUnderTest.get(0));
	new ShardConsumer<>(
		fetcher,
		shardIndex,
		subscribedShardsStateUnderTest.get(0).getStreamShardHandle(),
		subscribedShardsStateUnderTest.get(0).getLastProcessedSequenceNum(),
		// Initial number of records to fetch --> 10
		FakeKinesisBehavioursFactory.initialNumOfRecordsAfterNumOfGetRecordsCallsWithAdaptiveReads(10, 2, 500L),
		new ShardMetricsReporter(),
		deserializationSchema)
		.run();

	// Avg record size for first batch --> 10 * 10 Kb/10 = 10 Kb
	// Number of records fetched in second batch --> 2 Mb/10Kb * 5 = 40
	// Total number of records = 10 + 40 = 50
	assertEquals(50, sourceContext.getCollectedOutputs().size());
	assertEquals(
		SentinelSequenceNumber.SENTINEL_SHARD_ENDING_SEQUENCE_NUM.get(),
		subscribedShardsStateUnderTest.get(0).getLastProcessedSequenceNum());
}
 

@Test
public void testSkipCorruptedRecord() throws Exception {
	final String stream = "fakeStream";
	final int numShards = 3;

	final LinkedList<KinesisStreamShardState> testShardStates = new LinkedList<>();
	final TestSourceContext<String> sourceContext = new TestSourceContext<>();

	final TestableKinesisDataFetcher<String> fetcher = new TestableKinesisDataFetcher<>(
		Collections.singletonList(stream),
		sourceContext,
		TestUtils.getStandardProperties(),
		new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
		1,
		0,
		new AtomicReference<>(),
		testShardStates,
		new HashMap<>(),
		FakeKinesisBehavioursFactory.nonReshardedStreamsBehaviour(Collections.singletonMap(stream, numShards)));

	// FlinkKinesisConsumer is responsible for setting up the fetcher before it can be run;
	// run the consumer until it reaches the point where the fetcher starts to run
	final DummyFlinkKinesisConsumer<String> consumer = new DummyFlinkKinesisConsumer<>(TestUtils.getStandardProperties(), fetcher, 1, 0);

	CheckedThread consumerThread = new CheckedThread() {
		@Override
		public void go() throws Exception {
			consumer.run(new TestSourceContext<>());
		}
	};
	consumerThread.start();

	fetcher.waitUntilRun();
	consumer.cancel();
	consumerThread.sync();

	assertEquals(numShards, testShardStates.size());

	for (int i = 0; i < numShards; i++) {
		fetcher.emitRecordAndUpdateState("record-" + i, 10L, i, new SequenceNumber("seq-num-1"));
		assertEquals(new SequenceNumber("seq-num-1"), testShardStates.get(i).getLastProcessedSequenceNum());
		assertEquals(new StreamRecord<>("record-" + i, 10L), sourceContext.removeLatestOutput());
	}

	// emitting a null (i.e., a corrupt record) should not produce any output, but still have the shard state updated
	fetcher.emitRecordAndUpdateState(null, 10L, 1, new SequenceNumber("seq-num-2"));
		assertEquals(new SequenceNumber("seq-num-2"), testShardStates.get(1).getLastProcessedSequenceNum());
	assertEquals(null, sourceContext.removeLatestOutput()); // no output should have been collected
}
 

@Test
public void testStreamToLastSeenShardStateIsCorrectlySetWhenNotRestoringFromFailure() throws Exception {
	List<String> fakeStreams = new LinkedList<>();
	fakeStreams.add("fakeStream1");
	fakeStreams.add("fakeStream2");
	fakeStreams.add("fakeStream3");
	fakeStreams.add("fakeStream4");

	HashMap<String, String> subscribedStreamsToLastSeenShardIdsUnderTest =
		KinesisDataFetcher.createInitialSubscribedStreamsToLastDiscoveredShardsState(fakeStreams);

	Map<String, Integer> streamToShardCount = new HashMap<>();
	Random rand = new Random();
	for (String fakeStream : fakeStreams) {
		streamToShardCount.put(fakeStream, rand.nextInt(5) + 1);
	}

	final TestableKinesisDataFetcher<String> fetcher =
		new TestableKinesisDataFetcher<>(
			fakeStreams,
			new TestSourceContext<>(),
			TestUtils.getStandardProperties(),
			new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
			10,
			2,
			new AtomicReference<>(),
			new LinkedList<>(),
			subscribedStreamsToLastSeenShardIdsUnderTest,
			FakeKinesisBehavioursFactory.nonReshardedStreamsBehaviour(streamToShardCount));

	final DummyFlinkKinesisConsumer<String> consumer = new DummyFlinkKinesisConsumer<>(
			TestUtils.getStandardProperties(), fetcher, 1, 0);

	CheckedThread consumerThread = new CheckedThread() {
		@Override
		public void go() throws Exception {
			consumer.run(new TestSourceContext<>());
		}
	};
	consumerThread.start();

	fetcher.waitUntilRun();
	consumer.cancel();
	consumerThread.sync();

	// assert that the streams tracked in the state are identical to the subscribed streams
	Set<String> streamsInState = subscribedStreamsToLastSeenShardIdsUnderTest.keySet();
	assertEquals(fakeStreams.size(), streamsInState.size());
	assertTrue(streamsInState.containsAll(fakeStreams));

	// assert that the last seen shards in state is correctly set
	for (Map.Entry<String, String> streamToLastSeenShard : subscribedStreamsToLastSeenShardIdsUnderTest.entrySet()) {
		assertEquals(
			KinesisShardIdGenerator.generateFromShardOrder(streamToShardCount.get(streamToLastSeenShard.getKey()) - 1),
			streamToLastSeenShard.getValue());
	}
}
 

@Test
public void testStreamToLastSeenShardStateIsCorrectlySetWhenNoNewShardsSinceRestoredCheckpoint() throws Exception {
	List<String> fakeStreams = new LinkedList<>();
	fakeStreams.add("fakeStream1");
	fakeStreams.add("fakeStream2");

	Map<StreamShardHandle, String> restoredStateUnderTest = new HashMap<>();

	// fakeStream1 has 3 shards before restore
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream1",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(0))),
		UUID.randomUUID().toString());
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream1",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(1))),
		UUID.randomUUID().toString());
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream1",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(2))),
		UUID.randomUUID().toString());

	// fakeStream2 has 2 shards before restore
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream2",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(0))),
		UUID.randomUUID().toString());
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream2",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(1))),
		UUID.randomUUID().toString());

	Map<String, Integer> streamToShardCount = new HashMap<>();
	streamToShardCount.put("fakeStream1", 3); // fakeStream1 will still have 3 shards after restore
	streamToShardCount.put("fakeStream2", 2); // fakeStream2 will still have 2 shards after restore

	HashMap<String, String> subscribedStreamsToLastSeenShardIdsUnderTest =
		KinesisDataFetcher.createInitialSubscribedStreamsToLastDiscoveredShardsState(fakeStreams);

	final TestableKinesisDataFetcher<String> fetcher =
		new TestableKinesisDataFetcher<>(
			fakeStreams,
			new TestSourceContext<>(),
			TestUtils.getStandardProperties(),
			new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
			10,
			2,
			new AtomicReference<>(),
			new LinkedList<>(),
			subscribedStreamsToLastSeenShardIdsUnderTest,
			FakeKinesisBehavioursFactory.nonReshardedStreamsBehaviour(streamToShardCount));

	for (Map.Entry<StreamShardHandle, String> restoredState : restoredStateUnderTest.entrySet()) {
		fetcher.advanceLastDiscoveredShardOfStream(restoredState.getKey().getStreamName(), restoredState.getKey().getShard().getShardId());
		fetcher.registerNewSubscribedShardState(
			new KinesisStreamShardState(KinesisDataFetcher.convertToStreamShardMetadata(restoredState.getKey()),
				restoredState.getKey(), new SequenceNumber(restoredState.getValue())));
	}

	CheckedThread runFetcherThread = new CheckedThread() {
		@Override
		public void go() throws Exception {
			fetcher.runFetcher();
		}
	};
	runFetcherThread.start();

	fetcher.waitUntilInitialDiscovery();
	fetcher.shutdownFetcher();
	runFetcherThread.sync();

	// assert that the streams tracked in the state are identical to the subscribed streams
	Set<String> streamsInState = subscribedStreamsToLastSeenShardIdsUnderTest.keySet();
	assertEquals(fakeStreams.size(), streamsInState.size());
	assertTrue(streamsInState.containsAll(fakeStreams));

	// assert that the last seen shards in state is correctly set
	for (Map.Entry<String, String> streamToLastSeenShard : subscribedStreamsToLastSeenShardIdsUnderTest.entrySet()) {
		assertEquals(
			KinesisShardIdGenerator.generateFromShardOrder(streamToShardCount.get(streamToLastSeenShard.getKey()) - 1),
			streamToLastSeenShard.getValue());
	}
}
 

@Test
public void testStreamToLastSeenShardStateIsCorrectlySetWhenNewShardsFoundSinceRestoredCheckpoint() throws Exception {
	List<String> fakeStreams = new LinkedList<>();
	fakeStreams.add("fakeStream1");
	fakeStreams.add("fakeStream2");

	Map<StreamShardHandle, String> restoredStateUnderTest = new HashMap<>();

	// fakeStream1 has 3 shards before restore
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream1",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(0))),
		UUID.randomUUID().toString());
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream1",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(1))),
		UUID.randomUUID().toString());
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream1",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(2))),
		UUID.randomUUID().toString());

	// fakeStream2 has 2 shards before restore
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream2",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(0))),
		UUID.randomUUID().toString());
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream2",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(1))),
		UUID.randomUUID().toString());

	Map<String, Integer> streamToShardCount = new HashMap<>();
	streamToShardCount.put("fakeStream1", 3 + 1); // fakeStream1 had 3 shards before & 1 new shard after restore
	streamToShardCount.put("fakeStream2", 2 + 3); // fakeStream2 had 2 shards before & 3 new shard after restore

	HashMap<String, String> subscribedStreamsToLastSeenShardIdsUnderTest =
		KinesisDataFetcher.createInitialSubscribedStreamsToLastDiscoveredShardsState(fakeStreams);

	// using a non-resharded streams kinesis behaviour to represent that Kinesis is not resharded AFTER the restore
	final TestableKinesisDataFetcher<String> fetcher =
		new TestableKinesisDataFetcher<>(
			fakeStreams,
			new TestSourceContext<>(),
			TestUtils.getStandardProperties(),
			new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
			10,
			2,
			new AtomicReference<>(),
			new LinkedList<>(),
			subscribedStreamsToLastSeenShardIdsUnderTest,
			FakeKinesisBehavioursFactory.nonReshardedStreamsBehaviour(streamToShardCount));

	for (Map.Entry<StreamShardHandle, String> restoredState : restoredStateUnderTest.entrySet()) {
		fetcher.advanceLastDiscoveredShardOfStream(restoredState.getKey().getStreamName(), restoredState.getKey().getShard().getShardId());
		fetcher.registerNewSubscribedShardState(
			new KinesisStreamShardState(KinesisDataFetcher.convertToStreamShardMetadata(restoredState.getKey()),
				restoredState.getKey(), new SequenceNumber(restoredState.getValue())));
	}

	CheckedThread runFetcherThread = new CheckedThread() {
		@Override
		public void go() throws Exception {
			fetcher.runFetcher();
		}
	};
	runFetcherThread.start();

	fetcher.waitUntilInitialDiscovery();
	fetcher.shutdownFetcher();
	runFetcherThread.sync();

	// assert that the streams tracked in the state are identical to the subscribed streams
	Set<String> streamsInState = subscribedStreamsToLastSeenShardIdsUnderTest.keySet();
	assertEquals(fakeStreams.size(), streamsInState.size());
	assertTrue(streamsInState.containsAll(fakeStreams));

	// assert that the last seen shards in state is correctly set
	for (Map.Entry<String, String> streamToLastSeenShard : subscribedStreamsToLastSeenShardIdsUnderTest.entrySet()) {
		assertEquals(
			KinesisShardIdGenerator.generateFromShardOrder(streamToShardCount.get(streamToLastSeenShard.getKey()) - 1),
			streamToLastSeenShard.getValue());
	}
}
 

@Test
public void testStreamToLastSeenShardStateIsCorrectlySetWhenNoNewShardsSinceRestoredCheckpointAndSomeStreamsDoNotExist() throws Exception {
	List<String> fakeStreams = new LinkedList<>();
	fakeStreams.add("fakeStream1");
	fakeStreams.add("fakeStream2");
	fakeStreams.add("fakeStream3"); // fakeStream3 will not have any shards
	fakeStreams.add("fakeStream4"); // fakeStream4 will not have any shards

	Map<StreamShardHandle, String> restoredStateUnderTest = new HashMap<>();

	// fakeStream1 has 3 shards before restore
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream1",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(0))),
		UUID.randomUUID().toString());
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream1",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(1))),
		UUID.randomUUID().toString());
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream1",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(2))),
		UUID.randomUUID().toString());

	// fakeStream2 has 2 shards before restore
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream2",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(0))),
		UUID.randomUUID().toString());
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream2",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(1))),
		UUID.randomUUID().toString());

	Map<String, Integer> streamToShardCount = new HashMap<>();
	streamToShardCount.put("fakeStream1", 3); // fakeStream1 has fixed 3 shards
	streamToShardCount.put("fakeStream2", 2); // fakeStream2 has fixed 2 shards
	streamToShardCount.put("fakeStream3", 0); // no shards can be found for fakeStream3
	streamToShardCount.put("fakeStream4", 0); // no shards can be found for fakeStream4

	HashMap<String, String> subscribedStreamsToLastSeenShardIdsUnderTest =
		KinesisDataFetcher.createInitialSubscribedStreamsToLastDiscoveredShardsState(fakeStreams);

	// using a non-resharded streams kinesis behaviour to represent that Kinesis is not resharded AFTER the restore
	final TestableKinesisDataFetcher<String> fetcher =
		new TestableKinesisDataFetcher<>(
			fakeStreams,
			new TestSourceContext<>(),
			TestUtils.getStandardProperties(),
			new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
			10,
			2,
			new AtomicReference<>(),
			new LinkedList<>(),
			subscribedStreamsToLastSeenShardIdsUnderTest,
			FakeKinesisBehavioursFactory.nonReshardedStreamsBehaviour(streamToShardCount));

	for (Map.Entry<StreamShardHandle, String> restoredState : restoredStateUnderTest.entrySet()) {
		fetcher.advanceLastDiscoveredShardOfStream(restoredState.getKey().getStreamName(), restoredState.getKey().getShard().getShardId());
		fetcher.registerNewSubscribedShardState(
			new KinesisStreamShardState(KinesisDataFetcher.convertToStreamShardMetadata(restoredState.getKey()),
				restoredState.getKey(), new SequenceNumber(restoredState.getValue())));
	}

	CheckedThread runFetcherThread = new CheckedThread() {
		@Override
		public void go() throws Exception {
			fetcher.runFetcher();
		}
	};
	runFetcherThread.start();

	fetcher.waitUntilInitialDiscovery();
	fetcher.shutdownFetcher();
	runFetcherThread.sync();

	// assert that the streams tracked in the state are identical to the subscribed streams
	Set<String> streamsInState = subscribedStreamsToLastSeenShardIdsUnderTest.keySet();
	assertEquals(fakeStreams.size(), streamsInState.size());
	assertTrue(streamsInState.containsAll(fakeStreams));

	// assert that the last seen shards in state is correctly set
	assertEquals(
		KinesisShardIdGenerator.generateFromShardOrder(2),
		subscribedStreamsToLastSeenShardIdsUnderTest.get("fakeStream1"));
	assertEquals(
		KinesisShardIdGenerator.generateFromShardOrder(1),
		subscribedStreamsToLastSeenShardIdsUnderTest.get("fakeStream2"));
	assertNull(subscribedStreamsToLastSeenShardIdsUnderTest.get("fakeStream3"));
	assertNull(subscribedStreamsToLastSeenShardIdsUnderTest.get("fakeStream4"));
}
 

@Test
public void testStreamToLastSeenShardStateIsCorrectlySetWhenNewShardsFoundSinceRestoredCheckpointAndSomeStreamsDoNotExist() throws Exception {
	List<String> fakeStreams = new LinkedList<>();
	fakeStreams.add("fakeStream1");
	fakeStreams.add("fakeStream2");
	fakeStreams.add("fakeStream3"); // fakeStream3 will not have any shards
	fakeStreams.add("fakeStream4"); // fakeStream4 will not have any shards

	Map<StreamShardHandle, String> restoredStateUnderTest = new HashMap<>();

	// fakeStream1 has 3 shards before restore
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream1",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(0))),
		UUID.randomUUID().toString());
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream1",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(1))),
		UUID.randomUUID().toString());
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream1",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(2))),
		UUID.randomUUID().toString());

	// fakeStream2 has 2 shards before restore
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream2",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(0))),
		UUID.randomUUID().toString());
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream2",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(1))),
		UUID.randomUUID().toString());

	Map<String, Integer> streamToShardCount = new HashMap<>();
	streamToShardCount.put("fakeStream1", 3 + 1); // fakeStream1 had 3 shards before & 1 new shard after restore
	streamToShardCount.put("fakeStream2", 2 + 3); // fakeStream2 had 2 shards before & 2 new shard after restore
	streamToShardCount.put("fakeStream3", 0); // no shards can be found for fakeStream3
	streamToShardCount.put("fakeStream4", 0); // no shards can be found for fakeStream4

	HashMap<String, String> subscribedStreamsToLastSeenShardIdsUnderTest =
		KinesisDataFetcher.createInitialSubscribedStreamsToLastDiscoveredShardsState(fakeStreams);

	// using a non-resharded streams kinesis behaviour to represent that Kinesis is not resharded AFTER the restore
	final TestableKinesisDataFetcher<String> fetcher =
		new TestableKinesisDataFetcher<>(
			fakeStreams,
			new TestSourceContext<>(),
			TestUtils.getStandardProperties(),
			new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
			10,
			2,
			new AtomicReference<>(),
			new LinkedList<>(),
			subscribedStreamsToLastSeenShardIdsUnderTest,
			FakeKinesisBehavioursFactory.nonReshardedStreamsBehaviour(streamToShardCount));

	for (Map.Entry<StreamShardHandle, String> restoredState : restoredStateUnderTest.entrySet()) {
		fetcher.advanceLastDiscoveredShardOfStream(restoredState.getKey().getStreamName(), restoredState.getKey().getShard().getShardId());
		fetcher.registerNewSubscribedShardState(
			new KinesisStreamShardState(KinesisDataFetcher.convertToStreamShardMetadata(restoredState.getKey()),
				restoredState.getKey(), new SequenceNumber(restoredState.getValue())));
	}

	CheckedThread runFetcherThread = new CheckedThread() {
		@Override
		public void go() throws Exception {
			fetcher.runFetcher();
		}
	};
	runFetcherThread.start();

	fetcher.waitUntilInitialDiscovery();
	fetcher.shutdownFetcher();
	runFetcherThread.sync();

	// assert that the streams tracked in the state are identical to the subscribed streams
	Set<String> streamsInState = subscribedStreamsToLastSeenShardIdsUnderTest.keySet();
	assertEquals(fakeStreams.size(), streamsInState.size());
	assertTrue(streamsInState.containsAll(fakeStreams));

	// assert that the last seen shards in state is correctly set
	assertEquals(
		KinesisShardIdGenerator.generateFromShardOrder(3),
		subscribedStreamsToLastSeenShardIdsUnderTest.get("fakeStream1"));
	assertEquals(
		KinesisShardIdGenerator.generateFromShardOrder(4),
		subscribedStreamsToLastSeenShardIdsUnderTest.get("fakeStream2"));
	assertNull(subscribedStreamsToLastSeenShardIdsUnderTest.get("fakeStream3"));
	assertNull(subscribedStreamsToLastSeenShardIdsUnderTest.get("fakeStream4"));
}
 

@Test
public void testShardToSubtaskMappingWithCustomHashFunction() throws Exception {

	int totalCountOfSubtasks = 10;
	int shardCount = 3;

	for (int i = 0; i < 2; i++) {

		final int hash = i;
		final KinesisShardAssigner allShardsSingleSubtaskFn = (shard, subtasks) -> hash;
		Map<String, Integer> streamToShardCount = new HashMap<>();
		List<String> fakeStreams = new LinkedList<>();
		fakeStreams.add("fakeStream");
		streamToShardCount.put("fakeStream", shardCount);

		for (int j = 0; j < totalCountOfSubtasks; j++) {

			int subtaskIndex = j;
			// subscribe with default hashing
			final TestableKinesisDataFetcher fetcher =
				new TestableKinesisDataFetcher(
					fakeStreams,
					new TestSourceContext<>(),
					new Properties(),
					new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
					totalCountOfSubtasks,
					subtaskIndex,
					new AtomicReference<>(),
					new LinkedList<>(),
					KinesisDataFetcher.createInitialSubscribedStreamsToLastDiscoveredShardsState(fakeStreams),
					FakeKinesisBehavioursFactory.nonReshardedStreamsBehaviour(streamToShardCount));
			Whitebox.setInternalState(fetcher, "shardAssigner", allShardsSingleSubtaskFn); // override hashing
			List<StreamShardHandle> shards = fetcher.discoverNewShardsToSubscribe();
			fetcher.shutdownFetcher();

			String msg = String.format("for hash=%d, subtask=%d", hash, subtaskIndex);
			if (j == i) {
				assertEquals(msg, shardCount, shards.size());
			} else {
				assertEquals(msg, 0, shards.size());
			}
		}

	}

}
 

@Test
public void testPeriodicWatermark() {
	final MutableLong clock = new MutableLong();
	final MutableBoolean isTemporaryIdle = new MutableBoolean();
	final List<Watermark> watermarks = new ArrayList<>();

	String fakeStream1 = "fakeStream1";
	StreamShardHandle shardHandle =
		new StreamShardHandle(
			fakeStream1,
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(0)));

	TestSourceContext<String> sourceContext =
		new TestSourceContext<String>() {
			@Override
			public void emitWatermark(Watermark mark) {
				watermarks.add(mark);
			}

			@Override
			public void markAsTemporarilyIdle() {
				isTemporaryIdle.setTrue();
			}
		};

	HashMap<String, String> subscribedStreamsToLastSeenShardIdsUnderTest = new HashMap<>();

	final KinesisDataFetcher<String> fetcher =
		new TestableKinesisDataFetcher<String>(
			Collections.singletonList(fakeStream1),
			sourceContext,
			new java.util.Properties(),
			new KinesisDeserializationSchemaWrapper<>(new org.apache.flink.streaming.util.serialization.SimpleStringSchema()),
			1,
			1,
			new AtomicReference<>(),
			new LinkedList<>(),
			subscribedStreamsToLastSeenShardIdsUnderTest,
			FakeKinesisBehavioursFactory.nonReshardedStreamsBehaviour(new HashMap<>())) {

			@Override
			protected long getCurrentTimeMillis() {
				return clock.getValue();
			}
		};
	Whitebox.setInternalState(fetcher, "periodicWatermarkAssigner", watermarkAssigner);

	SequenceNumber seq = new SequenceNumber("fakeSequenceNumber");
	// register shards to subsequently emit records
	int shardIndex =
		fetcher.registerNewSubscribedShardState(
			new KinesisStreamShardState(
				KinesisDataFetcher.convertToStreamShardMetadata(shardHandle), shardHandle, seq));

	StreamRecord<String> record1 =
		new StreamRecord<>(String.valueOf(Long.MIN_VALUE), Long.MIN_VALUE);
	fetcher.emitRecordAndUpdateState(record1.getValue(), record1.getTimestamp(), shardIndex, seq);
	Assert.assertEquals(record1, sourceContext.getCollectedOutputs().poll());

	fetcher.emitWatermark();
	Assert.assertTrue("potential watermark equals previous watermark", watermarks.isEmpty());

	StreamRecord<String> record2 = new StreamRecord<>(String.valueOf(1), 1);
	fetcher.emitRecordAndUpdateState(record2.getValue(), record2.getTimestamp(), shardIndex, seq);
	Assert.assertEquals(record2, sourceContext.getCollectedOutputs().poll());

	fetcher.emitWatermark();
	Assert.assertFalse("watermark advanced", watermarks.isEmpty());
	Assert.assertEquals(new Watermark(record2.getTimestamp()), watermarks.remove(0));
	Assert.assertFalse("not idle", isTemporaryIdle.booleanValue());

	// test idle timeout
	long idleTimeout = 10;
	// advance clock idleTimeout
	clock.add(idleTimeout + 1);
	fetcher.emitWatermark();
	Assert.assertFalse("not idle", isTemporaryIdle.booleanValue());
	Assert.assertTrue("not idle, no new watermark", watermarks.isEmpty());

	// activate idle timeout
	Whitebox.setInternalState(fetcher, "shardIdleIntervalMillis", idleTimeout);
	fetcher.emitWatermark();
	Assert.assertTrue("idle", isTemporaryIdle.booleanValue());
	Assert.assertTrue("idle, no watermark", watermarks.isEmpty());
}
 

@Test
public void testRestoreWithEmptyState() throws Exception {
	final List<StreamShardHandle> initialDiscoveryShards = new ArrayList<>(TEST_STATE.size());
	for (StreamShardMetadata shardMetadata : TEST_STATE.keySet()) {
		Shard shard = new Shard();
		shard.setShardId(shardMetadata.getShardId());

		SequenceNumberRange sequenceNumberRange = new SequenceNumberRange();
		sequenceNumberRange.withStartingSequenceNumber("1");
		shard.setSequenceNumberRange(sequenceNumberRange);

		initialDiscoveryShards.add(new StreamShardHandle(shardMetadata.getStreamName(), shard));
	}

	final TestFetcher<String> fetcher = new TestFetcher<>(
		Collections.singletonList(TEST_STREAM_NAME),
		new TestSourceContext<>(),
		new TestRuntimeContext(true, 1, 0),
		TestUtils.getStandardProperties(),
		new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
		null,
		initialDiscoveryShards);

	final DummyFlinkKinesisConsumer<String> consumerFunction = new DummyFlinkKinesisConsumer<>(
		fetcher, new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()));

	StreamSource<String, DummyFlinkKinesisConsumer<String>> consumerOperator = new StreamSource<>(consumerFunction);

	final AbstractStreamOperatorTestHarness<String> testHarness =
		new AbstractStreamOperatorTestHarness<>(consumerOperator, 1, 1, 0);

	testHarness.setup();
	testHarness.initializeState(
		OperatorSnapshotUtil.getResourceFilename(
			"kinesis-consumer-migration-test-flink" + testMigrateVersion + "-empty-snapshot"));
	testHarness.open();

	consumerFunction.run(new TestSourceContext<>());

	// assert that no state was restored
	assertTrue(consumerFunction.getRestoredState().isEmpty());

	// although the restore state is empty, the fetcher should still have been registered the initial discovered shard;
	// furthermore, the discovered shard should be considered a newly created shard while the job wasn't running,
	// and therefore should be consumed from the earliest sequence number
	KinesisStreamShardState restoredShardState = fetcher.getSubscribedShardsState().get(0);
	assertEquals(TEST_STREAM_NAME, restoredShardState.getStreamShardHandle().getStreamName());
	assertEquals(TEST_SHARD_ID, restoredShardState.getStreamShardHandle().getShard().getShardId());
	assertFalse(restoredShardState.getStreamShardHandle().isClosed());
	assertEquals(SentinelSequenceNumber.SENTINEL_EARLIEST_SEQUENCE_NUM.get(), restoredShardState.getLastProcessedSequenceNum());

	consumerOperator.close();
	consumerOperator.cancel();
}
 

@Test
public void testRestore() throws Exception {
	final List<StreamShardHandle> initialDiscoveryShards = new ArrayList<>(TEST_STATE.size());
	for (StreamShardMetadata shardMetadata : TEST_STATE.keySet()) {
		Shard shard = new Shard();
		shard.setShardId(shardMetadata.getShardId());

		SequenceNumberRange sequenceNumberRange = new SequenceNumberRange();
		sequenceNumberRange.withStartingSequenceNumber("1");
		shard.setSequenceNumberRange(sequenceNumberRange);

		initialDiscoveryShards.add(new StreamShardHandle(shardMetadata.getStreamName(), shard));
	}

	final TestFetcher<String> fetcher = new TestFetcher<>(
		Collections.singletonList(TEST_STREAM_NAME),
		new TestSourceContext<>(),
		new TestRuntimeContext(true, 1, 0),
		TestUtils.getStandardProperties(),
		new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
		null,
		initialDiscoveryShards);

	final DummyFlinkKinesisConsumer<String> consumerFunction = new DummyFlinkKinesisConsumer<>(
		fetcher, new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()));

	StreamSource<String, DummyFlinkKinesisConsumer<String>> consumerOperator =
		new StreamSource<>(consumerFunction);

	final AbstractStreamOperatorTestHarness<String> testHarness =
		new AbstractStreamOperatorTestHarness<>(consumerOperator, 1, 1, 0);

	testHarness.setup();
	testHarness.initializeState(
		OperatorSnapshotUtil.getResourceFilename(
			"kinesis-consumer-migration-test-flink" + testMigrateVersion + "-snapshot"));
	testHarness.open();

	consumerFunction.run(new TestSourceContext<>());

	// assert that state is correctly restored
	assertNotEquals(null, consumerFunction.getRestoredState());
	assertEquals(1, consumerFunction.getRestoredState().size());
	assertEquals(TEST_STATE, removeEquivalenceWrappers(consumerFunction.getRestoredState()));
	assertEquals(1, fetcher.getSubscribedShardsState().size());
	assertEquals(TEST_SEQUENCE_NUMBER, fetcher.getSubscribedShardsState().get(0).getLastProcessedSequenceNum());

	KinesisStreamShardState restoredShardState = fetcher.getSubscribedShardsState().get(0);
	assertEquals(TEST_STREAM_NAME, restoredShardState.getStreamShardHandle().getStreamName());
	assertEquals(TEST_SHARD_ID, restoredShardState.getStreamShardHandle().getShard().getShardId());
	assertFalse(restoredShardState.getStreamShardHandle().isClosed());
	assertEquals(TEST_SEQUENCE_NUMBER, restoredShardState.getLastProcessedSequenceNum());

	consumerOperator.close();
	consumerOperator.cancel();
}
 

@Test
public void testRestoreWithReshardedStream() throws Exception {
	final List<StreamShardHandle> initialDiscoveryShards = new ArrayList<>(TEST_STATE.size());
	for (StreamShardMetadata shardMetadata : TEST_STATE.keySet()) {
		// setup the closed shard
		Shard closedShard = new Shard();
		closedShard.setShardId(shardMetadata.getShardId());

		SequenceNumberRange closedSequenceNumberRange = new SequenceNumberRange();
		closedSequenceNumberRange.withStartingSequenceNumber("1");
		closedSequenceNumberRange.withEndingSequenceNumber("1087654321"); // this represents a closed shard
		closedShard.setSequenceNumberRange(closedSequenceNumberRange);

		initialDiscoveryShards.add(new StreamShardHandle(shardMetadata.getStreamName(), closedShard));

		// setup the new shards
		Shard newSplitShard1 = new Shard();
		newSplitShard1.setShardId(KinesisShardIdGenerator.generateFromShardOrder(1));

		SequenceNumberRange newSequenceNumberRange1 = new SequenceNumberRange();
		newSequenceNumberRange1.withStartingSequenceNumber("1087654322");
		newSplitShard1.setSequenceNumberRange(newSequenceNumberRange1);

		newSplitShard1.setParentShardId(TEST_SHARD_ID);

		Shard newSplitShard2 = new Shard();
		newSplitShard2.setShardId(KinesisShardIdGenerator.generateFromShardOrder(2));

		SequenceNumberRange newSequenceNumberRange2 = new SequenceNumberRange();
		newSequenceNumberRange2.withStartingSequenceNumber("2087654322");
		newSplitShard2.setSequenceNumberRange(newSequenceNumberRange2);

		newSplitShard2.setParentShardId(TEST_SHARD_ID);

		initialDiscoveryShards.add(new StreamShardHandle(shardMetadata.getStreamName(), newSplitShard1));
		initialDiscoveryShards.add(new StreamShardHandle(shardMetadata.getStreamName(), newSplitShard2));
	}

	final TestFetcher<String> fetcher = new TestFetcher<>(
		Collections.singletonList(TEST_STREAM_NAME),
		new TestSourceContext<>(),
		new TestRuntimeContext(true, 1, 0),
		TestUtils.getStandardProperties(),
		new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
		null,
		initialDiscoveryShards);

	final DummyFlinkKinesisConsumer<String> consumerFunction = new DummyFlinkKinesisConsumer<>(
		fetcher, new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()));

	StreamSource<String, DummyFlinkKinesisConsumer<String>> consumerOperator =
		new StreamSource<>(consumerFunction);

	final AbstractStreamOperatorTestHarness<String> testHarness =
		new AbstractStreamOperatorTestHarness<>(consumerOperator, 1, 1, 0);

	testHarness.setup();
	testHarness.initializeState(
		OperatorSnapshotUtil.getResourceFilename(
			"kinesis-consumer-migration-test-flink" + testMigrateVersion + "-snapshot"));
	testHarness.open();

	consumerFunction.run(new TestSourceContext<>());

	// assert that state is correctly restored
	assertNotEquals(null, consumerFunction.getRestoredState());
	assertEquals(1, consumerFunction.getRestoredState().size());
	assertEquals(TEST_STATE, removeEquivalenceWrappers(consumerFunction.getRestoredState()));

	// assert that the fetcher is registered with all shards, including new shards
	assertEquals(3, fetcher.getSubscribedShardsState().size());

	KinesisStreamShardState restoredClosedShardState = fetcher.getSubscribedShardsState().get(0);
	assertEquals(TEST_STREAM_NAME, restoredClosedShardState.getStreamShardHandle().getStreamName());
	assertEquals(TEST_SHARD_ID, restoredClosedShardState.getStreamShardHandle().getShard().getShardId());
	assertTrue(restoredClosedShardState.getStreamShardHandle().isClosed());
	assertEquals(TEST_SEQUENCE_NUMBER, restoredClosedShardState.getLastProcessedSequenceNum());

	KinesisStreamShardState restoredNewSplitShard1 = fetcher.getSubscribedShardsState().get(1);
	assertEquals(TEST_STREAM_NAME, restoredNewSplitShard1.getStreamShardHandle().getStreamName());
	assertEquals(KinesisShardIdGenerator.generateFromShardOrder(1), restoredNewSplitShard1.getStreamShardHandle().getShard().getShardId());
	assertFalse(restoredNewSplitShard1.getStreamShardHandle().isClosed());
	// new shards should be consumed from the beginning
	assertEquals(SentinelSequenceNumber.SENTINEL_EARLIEST_SEQUENCE_NUM.get(), restoredNewSplitShard1.getLastProcessedSequenceNum());

	KinesisStreamShardState restoredNewSplitShard2 = fetcher.getSubscribedShardsState().get(2);
	assertEquals(TEST_STREAM_NAME, restoredNewSplitShard2.getStreamShardHandle().getStreamName());
	assertEquals(KinesisShardIdGenerator.generateFromShardOrder(2), restoredNewSplitShard2.getStreamShardHandle().getShard().getShardId());
	assertFalse(restoredNewSplitShard2.getStreamShardHandle().isClosed());
	// new shards should be consumed from the beginning
	assertEquals(SentinelSequenceNumber.SENTINEL_EARLIEST_SEQUENCE_NUM.get(), restoredNewSplitShard2.getLastProcessedSequenceNum());

	consumerOperator.close();
	consumerOperator.cancel();
}
 

@Test
public void testStreamToLastSeenShardStateIsCorrectlySetWhenNoNewShardsSinceRestoredCheckpointAndSomeStreamsDoNotExist() throws Exception {
	List<String> fakeStreams = new LinkedList<>();
	fakeStreams.add("fakeStream1");
	fakeStreams.add("fakeStream2");
	fakeStreams.add("fakeStream3"); // fakeStream3 will not have any shards
	fakeStreams.add("fakeStream4"); // fakeStream4 will not have any shards

	Map<StreamShardHandle, String> restoredStateUnderTest = new HashMap<>();

	// fakeStream1 has 3 shards before restore
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream1",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(0))),
		UUID.randomUUID().toString());
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream1",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(1))),
		UUID.randomUUID().toString());
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream1",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(2))),
		UUID.randomUUID().toString());

	// fakeStream2 has 2 shards before restore
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream2",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(0))),
		UUID.randomUUID().toString());
	restoredStateUnderTest.put(
		new StreamShardHandle(
			"fakeStream2",
			new Shard().withShardId(KinesisShardIdGenerator.generateFromShardOrder(1))),
		UUID.randomUUID().toString());

	Map<String, Integer> streamToShardCount = new HashMap<>();
	streamToShardCount.put("fakeStream1", 3); // fakeStream1 has fixed 3 shards
	streamToShardCount.put("fakeStream2", 2); // fakeStream2 has fixed 2 shards
	streamToShardCount.put("fakeStream3", 0); // no shards can be found for fakeStream3
	streamToShardCount.put("fakeStream4", 0); // no shards can be found for fakeStream4

	HashMap<String, String> subscribedStreamsToLastSeenShardIdsUnderTest =
		KinesisDataFetcher.createInitialSubscribedStreamsToLastDiscoveredShardsState(fakeStreams);

	// using a non-resharded streams kinesis behaviour to represent that Kinesis is not resharded AFTER the restore
	final TestableKinesisDataFetcher<String> fetcher =
		new TestableKinesisDataFetcher<>(
			fakeStreams,
			new TestSourceContext<>(),
			TestUtils.getStandardProperties(),
			new KinesisDeserializationSchemaWrapper<>(new SimpleStringSchema()),
			10,
			2,
			new AtomicReference<>(),
			new LinkedList<>(),
			subscribedStreamsToLastSeenShardIdsUnderTest,
			FakeKinesisBehavioursFactory.nonReshardedStreamsBehaviour(streamToShardCount));

	for (Map.Entry<StreamShardHandle, String> restoredState : restoredStateUnderTest.entrySet()) {
		fetcher.advanceLastDiscoveredShardOfStream(restoredState.getKey().getStreamName(), restoredState.getKey().getShard().getShardId());
		fetcher.registerNewSubscribedShardState(
			new KinesisStreamShardState(KinesisDataFetcher.convertToStreamShardMetadata(restoredState.getKey()),
				restoredState.getKey(), new SequenceNumber(restoredState.getValue())));
	}

	CheckedThread runFetcherThread = new CheckedThread() {
		@Override
		public void go() throws Exception {
			fetcher.runFetcher();
		}
	};
	runFetcherThread.start();

	fetcher.waitUntilInitialDiscovery();
	fetcher.shutdownFetcher();
	runFetcherThread.sync();

	// assert that the streams tracked in the state are identical to the subscribed streams
	Set<String> streamsInState = subscribedStreamsToLastSeenShardIdsUnderTest.keySet();
	assertEquals(fakeStreams.size(), streamsInState.size());
	assertTrue(streamsInState.containsAll(fakeStreams));

	// assert that the last seen shards in state is correctly set
	assertEquals(
		KinesisShardIdGenerator.generateFromShardOrder(2),
		subscribedStreamsToLastSeenShardIdsUnderTest.get("fakeStream1"));
	assertEquals(
		KinesisShardIdGenerator.generateFromShardOrder(1),
		subscribedStreamsToLastSeenShardIdsUnderTest.get("fakeStream2"));
	assertNull(subscribedStreamsToLastSeenShardIdsUnderTest.get("fakeStream3"));
	assertNull(subscribedStreamsToLastSeenShardIdsUnderTest.get("fakeStream4"));
}