awelless commented on code in PR #11857:
URL: https://github.com/apache/nifi/pull/11857#discussion_r4412224840
##########
nifi-extension-bundles/nifi-kafka-bundle/nifi-kafka-processors/src/main/java/org/apache/nifi/kafka/processors/consumer/convert/AbstractRecordStreamKafkaMessageConverter.java:
##########
@@ -87,6 +87,12 @@ protected AbstractRecordStreamKafkaMessageConverter(
public void toFlowFiles(final ProcessSession session, final
Iterator<ByteRecord> consumerRecords) {
while (consumerRecords.hasNext()) {
final ByteRecord consumerRecord = consumerRecords.next();
+ if (consumerRecord.isTombstone()) {
+ session.adjustCounter("Records Received from " +
consumerRecord.getTopic(), 1, false);
Review Comment:
I think we should use a separate counter for tombstones.
If there are only tombstones in the topic, "Records Received..." counter
will be increasing, but no FlowFiles would be produced. This is confusing.
I recommend adding another counter: "Tombstones skipped for ...", which is
going to be incremented whenever a tombstone record **is skipped**. We may
still increment the "Record received" counter, but it should be evident we're
skipping tombstones.
##########
nifi-extension-bundles/nifi-kafka-bundle/nifi-kafka-service-shared/src/main/java/org/apache/nifi/kafka/service/consumer/Kafka3ConsumerService.java:
##########
@@ -375,11 +375,10 @@ public ByteRecord next() {
recordHeaders.add(recordHeader);
});
- // Support Kafka tombstones
- byte[] value = consumerRecord.value();
- if (value == null) {
- value = new byte[0];
- }
+ // A null Kafka value is a tombstone. Keep that distinct from a
non-null empty value.
+ final byte[] originalValue = consumerRecord.value();
+ final boolean isTombstone = originalValue == null;
+ final byte[] value = isTombstone ? new byte[0] : originalValue;
Review Comment:
I assume this logic is going to be revised as a part of
https://issues.apache.org/jira/browse/NIFI-16517, correct?
##########
nifi-extension-bundles/nifi-kafka-bundle/nifi-kafka-processors/src/main/resources/docs/org.apache.nifi.kafka.processors.ConsumeKafka/additionalDetails.md:
##########
@@ -23,6 +23,8 @@ This processor offers multiple output strategies (configured
via processor prope
- 'Use Wrapper' emits FlowFile records containing the Kafka record key, value,
and headers, as well as additional metadata from the Kafka record.
- 'Inject Metadata' emits FlowFile records containing the Kafka record value
into which a sub record field has been added to hold metadata, headers and key
information.
+When Processing Strategy is RECORD, a Kafka tombstone (null value) is not
written for any Output Strategy. It is skipped and offset is committed.
Review Comment:
Nit: I'd put tombstone handling under a separate header. Could you
additionally describe the behavior for other Processing Strategies here?
It isn't related for your change, but would be helpful, provided the
tombstone behavior differs from strategy to strategy.
> offset is committed
This is true only when "Commit Offsets" is true.
##########
nifi-extension-bundles/nifi-kafka-bundle/nifi-kafka-3-integration/src/test/java/org/apache/nifi/kafka/processors/ConsumeKafkaRecordTombstoneIT.java:
##########
@@ -0,0 +1,422 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.nifi.kafka.processors;
+
+import com.fasterxml.jackson.databind.node.ArrayNode;
+import com.fasterxml.jackson.databind.node.ObjectNode;
+import org.apache.avro.Schema;
+import org.apache.avro.file.DataFileWriter;
+import org.apache.avro.generic.GenericData;
+import org.apache.avro.generic.GenericDatumWriter;
+import org.apache.avro.generic.GenericRecord;
+import org.apache.nifi.avro.AvroReader;
+import org.apache.nifi.kafka.processors.consumer.ProcessingStrategy;
+import org.apache.nifi.kafka.processors.producer.wrapper.InjectMetadataRecord;
+import org.apache.nifi.kafka.processors.producer.wrapper.WrapperRecord;
+import org.apache.nifi.kafka.service.api.consumer.AutoOffsetReset;
+import org.apache.nifi.kafka.shared.attribute.KafkaFlowFileAttribute;
+import org.apache.nifi.kafka.shared.property.OutputStrategy;
+import org.apache.nifi.reporting.InitializationException;
+import org.apache.nifi.schema.access.SchemaAccessUtils;
+import org.apache.nifi.util.MockFlowFile;
+import org.apache.nifi.util.TestRunner;
+import org.apache.nifi.util.TestRunners;
+import org.junit.jupiter.api.BeforeEach;
+import org.junit.jupiter.api.Test;
+
+import java.io.ByteArrayOutputStream;
+import java.io.IOException;
+import java.nio.charset.StandardCharsets;
+import java.time.Duration;
+import java.util.List;
+import java.util.UUID;
+import java.util.concurrent.ExecutionException;
+
+import static org.junit.jupiter.api.Assertions.assertEquals;
+import static org.junit.jupiter.api.Assertions.assertInstanceOf;
+
+/**
+ * RECORD output for a batch that contains a non-null empty value, a null
tombstone, and a real value.
+ * A null tombstone is omitted from success and parse.failure. A JSON reader
emits only the real value.
+ * An Avro reader routes the non-null empty value to parse.failure and writes
the real value to success.
+ */
+class ConsumeKafkaRecordTombstoneIT extends AbstractConsumeKafkaIT {
+
+ private static final String NON_EMPTY_VALUE = "{\"id\":1,\"name\":\"A\"}";
+
+ private static final byte[] NON_EMPTY_BYTES =
NON_EMPTY_VALUE.getBytes(StandardCharsets.UTF_8);
+
+ private static final byte[] EMPTY_BYTES = new byte[0];
+
+ private static final String AVRO_SCHEMA = """
+ {
+ "type": "record",
+ "name": "message",
+ "fields": [
+ {"name": "id", "type": "int"},
+ {"name": "name", "type": "string"}
+ ]
+ }
+ """;
+
+ private static final String MESSAGE_KEY = "key";
+
+ private static final int FIRST_PARTITION = 0;
+
+ private static final Duration POLL_TIMEOUT = Duration.ofSeconds(30);
+
+ private TestRunner runner;
+
+ @BeforeEach
+ void setRunner() throws InitializationException {
+ runner = TestRunners.newTestRunner(ConsumeKafka.class);
+ addKafkaConnectionService(runner);
+ runner.setProperty(ConsumeKafka.CONNECTION_SERVICE,
CONNECTION_SERVICE_ID);
+ addRecordReaderService(runner);
+ addRecordWriterService(runner);
+ }
+
+ @Test
+ void testOutputStrategyUseValue() throws ExecutionException,
InterruptedException, IOException {
+ final String topic = configureRecord(OutputStrategy.USE_VALUE);
+ runner.run(1, false, true);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, EMPTY_BYTES);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, null);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, NON_EMPTY_BYTES);
+
+ runUntil(runner, testRunner ->
!testRunner.getFlowFilesForRelationship(ConsumeKafka.SUCCESS).isEmpty(),
POLL_TIMEOUT);
+ runner.run(1, true, false);
+
+ runner.assertTransferCount(ConsumeKafka.PARSE_FAILURE, 0);
+ final List<MockFlowFile> flowFiles =
runner.getFlowFilesForRelationship(ConsumeKafka.SUCCESS);
+ assertEquals(1, flowFiles.size());
+
+ final MockFlowFile flowFile = flowFiles.getFirst();
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_TOPIC,
topic);
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_OFFSET,
"2");
+
flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_MAX_OFFSET, "2");
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_COUNT,
"1");
+ flowFile.assertAttributeEquals("record.count", "1");
+
flowFile.assertAttributeNotExists(KafkaFlowFileAttribute.KAFKA_TOMBSTONE);
+
+ final ArrayNode records = assertInstanceOf(ArrayNode.class,
objectMapper.readTree(flowFile.getContent()));
+ assertEquals(1, records.size());
+ final ObjectNode record = assertInstanceOf(ObjectNode.class,
records.get(0));
+ assertEquals(1, record.get("id").asInt());
+ assertEquals("A", record.get("name").asText());
+ }
+
+ @Test
+ void testOutputStrategyUseWrapper() throws ExecutionException,
InterruptedException, IOException {
+ final String topic = configureRecord(OutputStrategy.USE_WRAPPER);
+ runner.run(1, false, true);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, EMPTY_BYTES);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, null);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, NON_EMPTY_BYTES);
+
+ runUntil(runner, testRunner ->
!testRunner.getFlowFilesForRelationship(ConsumeKafka.SUCCESS).isEmpty(),
POLL_TIMEOUT);
+ runner.run(1, true, false);
+
+ runner.assertTransferCount(ConsumeKafka.PARSE_FAILURE, 0);
+ final List<MockFlowFile> flowFiles =
runner.getFlowFilesForRelationship(ConsumeKafka.SUCCESS);
+ assertEquals(1, flowFiles.size());
+
+ final MockFlowFile flowFile = flowFiles.getFirst();
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_TOPIC,
topic);
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_OFFSET,
"2");
+
flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_MAX_OFFSET, "2");
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_COUNT,
"1");
+ flowFile.assertAttributeEquals("record.count", "1");
+
flowFile.assertAttributeNotExists(KafkaFlowFileAttribute.KAFKA_TOMBSTONE);
+
+ final ArrayNode records = assertInstanceOf(ArrayNode.class,
objectMapper.readTree(flowFile.getContent()));
+ assertEquals(1, records.size());
+ final ObjectNode record = assertInstanceOf(ObjectNode.class,
records.get(0));
+ final ObjectNode value = assertInstanceOf(ObjectNode.class,
record.get(WrapperRecord.VALUE));
+ assertEquals(1, value.get("id").asInt());
+ assertEquals("A", value.get("name").asText());
+ final ObjectNode metadata = assertInstanceOf(ObjectNode.class,
record.get(WrapperRecord.METADATA));
+ assertEquals(topic, metadata.get(WrapperRecord.TOPIC).asText());
+ assertEquals(2L, metadata.get(WrapperRecord.OFFSET).asLong());
+ }
+
+ @Test
+ void testOutputStrategyInjectMetadata() throws ExecutionException,
InterruptedException, IOException {
+ final String topic = configureRecord(OutputStrategy.INJECT_METADATA);
+ runner.run(1, false, true);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, EMPTY_BYTES);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, null);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, NON_EMPTY_BYTES);
+
+ runUntil(runner, testRunner ->
!testRunner.getFlowFilesForRelationship(ConsumeKafka.SUCCESS).isEmpty(),
POLL_TIMEOUT);
+ runner.run(1, true, false);
+
+ runner.assertTransferCount(ConsumeKafka.PARSE_FAILURE, 0);
+ final List<MockFlowFile> flowFiles =
runner.getFlowFilesForRelationship(ConsumeKafka.SUCCESS);
+ assertEquals(1, flowFiles.size());
+
+ final MockFlowFile flowFile = flowFiles.getFirst();
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_TOPIC,
topic);
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_OFFSET,
"2");
+
flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_MAX_OFFSET, "2");
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_COUNT,
"1");
+ flowFile.assertAttributeEquals("record.count", "1");
+
flowFile.assertAttributeNotExists(KafkaFlowFileAttribute.KAFKA_TOMBSTONE);
+
+ final ArrayNode records = assertInstanceOf(ArrayNode.class,
objectMapper.readTree(flowFile.getContent()));
+ assertEquals(1, records.size());
+ final ObjectNode record = assertInstanceOf(ObjectNode.class,
records.get(0));
+ assertEquals(1, record.get("id").asInt());
+ assertEquals("A", record.get("name").asText());
+ final ObjectNode metadata = assertInstanceOf(ObjectNode.class,
record.get(InjectMetadataRecord.METADATA));
+ assertEquals(topic, metadata.get(InjectMetadataRecord.TOPIC).asText());
+ assertEquals(2L, metadata.get(InjectMetadataRecord.OFFSET).asLong());
+ }
+
+ @Test
+ void testOutputStrategyInjectOffset() throws ExecutionException,
InterruptedException, IOException {
+ final String topic = configureRecord(OutputStrategy.INJECT_OFFSET);
+ runner.run(1, false, true);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, EMPTY_BYTES);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, null);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, NON_EMPTY_BYTES);
+
+ runUntil(runner, testRunner ->
!testRunner.getFlowFilesForRelationship(ConsumeKafka.SUCCESS).isEmpty(),
POLL_TIMEOUT);
+ runner.run(1, true, false);
+
+ runner.assertTransferCount(ConsumeKafka.PARSE_FAILURE, 0);
+ final List<MockFlowFile> flowFiles =
runner.getFlowFilesForRelationship(ConsumeKafka.SUCCESS);
+ assertEquals(1, flowFiles.size());
+
+ final MockFlowFile flowFile = flowFiles.getFirst();
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_TOPIC,
topic);
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_OFFSET,
"2");
+
flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_MAX_OFFSET, "2");
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_COUNT,
"1");
+ flowFile.assertAttributeEquals("record.count", "1");
+
flowFile.assertAttributeNotExists(KafkaFlowFileAttribute.KAFKA_TOMBSTONE);
+
+ final ArrayNode records = assertInstanceOf(ArrayNode.class,
objectMapper.readTree(flowFile.getContent()));
+ assertEquals(1, records.size());
+ final ObjectNode record = assertInstanceOf(ObjectNode.class,
records.get(0));
+ assertEquals(1, record.get("id").asInt());
+ assertEquals("A", record.get("name").asText());
+ assertEquals(2L, record.get("kafkaOffset").asLong());
+ }
+
+ @Test
+ void testOutputStrategyUseValueWithAvroReader() throws ExecutionException,
InterruptedException, IOException, InitializationException {
+ final String topic = configureRecord(OutputStrategy.USE_VALUE);
+ useAvroReader();
+ runner.run(1, false, true);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, EMPTY_BYTES);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, null);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, avroRecord());
+
+ runUntil(runner, testRunner ->
!testRunner.getFlowFilesForRelationship(ConsumeKafka.SUCCESS).isEmpty()
+ &&
!testRunner.getFlowFilesForRelationship(ConsumeKafka.PARSE_FAILURE).isEmpty(),
POLL_TIMEOUT);
+ runner.run(1, true, false);
+
+ final List<MockFlowFile> failures =
runner.getFlowFilesForRelationship(ConsumeKafka.PARSE_FAILURE);
+ assertEquals(1, failures.size());
+
+ final MockFlowFile emptyFailure = failures.getFirst();
+ emptyFailure.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_TOPIC,
topic);
+
emptyFailure.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_OFFSET, "0");
+ emptyFailure.assertContentEquals("");
+
emptyFailure.assertAttributeNotExists(KafkaFlowFileAttribute.KAFKA_TOMBSTONE);
+
+ final List<MockFlowFile> flowFiles =
runner.getFlowFilesForRelationship(ConsumeKafka.SUCCESS);
+ assertEquals(1, flowFiles.size());
+
+ final MockFlowFile flowFile = flowFiles.getFirst();
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_TOPIC,
topic);
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_OFFSET,
"2");
+
flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_MAX_OFFSET, "2");
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_COUNT,
"1");
+ flowFile.assertAttributeEquals("record.count", "1");
+
flowFile.assertAttributeNotExists(KafkaFlowFileAttribute.KAFKA_TOMBSTONE);
+
+ final ArrayNode records = assertInstanceOf(ArrayNode.class,
objectMapper.readTree(flowFile.getContent()));
+ assertEquals(1, records.size());
+ final ObjectNode record = assertInstanceOf(ObjectNode.class,
records.get(0));
+ assertEquals(1, record.get("id").asInt());
+ assertEquals("A", record.get("name").asText());
+ }
+
+ @Test
+ void testOutputStrategyUseWrapperWithAvroReader() throws
ExecutionException, InterruptedException, IOException, InitializationException {
+ final String topic = configureRecord(OutputStrategy.USE_WRAPPER);
+ useAvroReader();
+ runner.run(1, false, true);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, EMPTY_BYTES);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, null);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, avroRecord());
+
+ runUntil(runner, testRunner ->
!testRunner.getFlowFilesForRelationship(ConsumeKafka.SUCCESS).isEmpty()
+ &&
!testRunner.getFlowFilesForRelationship(ConsumeKafka.PARSE_FAILURE).isEmpty(),
POLL_TIMEOUT);
+ runner.run(1, true, false);
+
+ final List<MockFlowFile> failures =
runner.getFlowFilesForRelationship(ConsumeKafka.PARSE_FAILURE);
+ assertEquals(1, failures.size());
+
+ final MockFlowFile emptyFailure = failures.getFirst();
+ emptyFailure.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_TOPIC,
topic);
+
emptyFailure.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_OFFSET, "0");
+ emptyFailure.assertContentEquals("");
+
emptyFailure.assertAttributeNotExists(KafkaFlowFileAttribute.KAFKA_TOMBSTONE);
+
+ final List<MockFlowFile> flowFiles =
runner.getFlowFilesForRelationship(ConsumeKafka.SUCCESS);
+ assertEquals(1, flowFiles.size());
+
+ final MockFlowFile flowFile = flowFiles.getFirst();
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_TOPIC,
topic);
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_OFFSET,
"2");
+
flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_MAX_OFFSET, "2");
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_COUNT,
"1");
+ flowFile.assertAttributeEquals("record.count", "1");
+
flowFile.assertAttributeNotExists(KafkaFlowFileAttribute.KAFKA_TOMBSTONE);
+
+ final ArrayNode records = assertInstanceOf(ArrayNode.class,
objectMapper.readTree(flowFile.getContent()));
+ assertEquals(1, records.size());
+ final ObjectNode record = assertInstanceOf(ObjectNode.class,
records.get(0));
+ final ObjectNode value = assertInstanceOf(ObjectNode.class,
record.get(WrapperRecord.VALUE));
+ assertEquals(1, value.get("id").asInt());
+ assertEquals("A", value.get("name").asText());
+ final ObjectNode metadata = assertInstanceOf(ObjectNode.class,
record.get(WrapperRecord.METADATA));
+ assertEquals(topic, metadata.get(WrapperRecord.TOPIC).asText());
+ assertEquals(2L, metadata.get(WrapperRecord.OFFSET).asLong());
+ }
+
+ @Test
+ void testOutputStrategyInjectMetadataWithAvroReader() throws
ExecutionException, InterruptedException, IOException, InitializationException {
+ final String topic = configureRecord(OutputStrategy.INJECT_METADATA);
+ useAvroReader();
+ runner.run(1, false, true);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, EMPTY_BYTES);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, null);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, avroRecord());
+
+ runUntil(runner, testRunner ->
!testRunner.getFlowFilesForRelationship(ConsumeKafka.SUCCESS).isEmpty()
+ &&
!testRunner.getFlowFilesForRelationship(ConsumeKafka.PARSE_FAILURE).isEmpty(),
POLL_TIMEOUT);
+ runner.run(1, true, false);
+
+ final List<MockFlowFile> failures =
runner.getFlowFilesForRelationship(ConsumeKafka.PARSE_FAILURE);
+ assertEquals(1, failures.size());
+
+ final MockFlowFile emptyFailure = failures.getFirst();
+ emptyFailure.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_TOPIC,
topic);
+
emptyFailure.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_OFFSET, "0");
+ emptyFailure.assertContentEquals("");
+
emptyFailure.assertAttributeNotExists(KafkaFlowFileAttribute.KAFKA_TOMBSTONE);
+
+ final List<MockFlowFile> flowFiles =
runner.getFlowFilesForRelationship(ConsumeKafka.SUCCESS);
+ assertEquals(1, flowFiles.size());
+
+ final MockFlowFile flowFile = flowFiles.getFirst();
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_TOPIC,
topic);
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_OFFSET,
"2");
+
flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_MAX_OFFSET, "2");
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_COUNT,
"1");
+ flowFile.assertAttributeEquals("record.count", "1");
+
flowFile.assertAttributeNotExists(KafkaFlowFileAttribute.KAFKA_TOMBSTONE);
+
+ final ArrayNode records = assertInstanceOf(ArrayNode.class,
objectMapper.readTree(flowFile.getContent()));
+ assertEquals(1, records.size());
+ final ObjectNode record = assertInstanceOf(ObjectNode.class,
records.get(0));
+ assertEquals(1, record.get("id").asInt());
+ assertEquals("A", record.get("name").asText());
+ final ObjectNode metadata = assertInstanceOf(ObjectNode.class,
record.get(InjectMetadataRecord.METADATA));
+ assertEquals(topic, metadata.get(InjectMetadataRecord.TOPIC).asText());
+ assertEquals(2L, metadata.get(InjectMetadataRecord.OFFSET).asLong());
+ }
+
+ @Test
+ void testOutputStrategyInjectOffsetWithAvroReader() throws
ExecutionException, InterruptedException, IOException, InitializationException {
+ final String topic = configureRecord(OutputStrategy.INJECT_OFFSET);
+ useAvroReader();
+ runner.run(1, false, true);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, EMPTY_BYTES);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, null);
+ produceBytes(topic, FIRST_PARTITION, MESSAGE_KEY, avroRecord());
+
+ runUntil(runner, testRunner ->
!testRunner.getFlowFilesForRelationship(ConsumeKafka.SUCCESS).isEmpty()
+ &&
!testRunner.getFlowFilesForRelationship(ConsumeKafka.PARSE_FAILURE).isEmpty(),
POLL_TIMEOUT);
+ runner.run(1, true, false);
+
+ final List<MockFlowFile> failures =
runner.getFlowFilesForRelationship(ConsumeKafka.PARSE_FAILURE);
+ assertEquals(1, failures.size());
+
+ final MockFlowFile emptyFailure = failures.getFirst();
+ emptyFailure.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_TOPIC,
topic);
+
emptyFailure.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_OFFSET, "0");
+ emptyFailure.assertContentEquals("");
+
emptyFailure.assertAttributeNotExists(KafkaFlowFileAttribute.KAFKA_TOMBSTONE);
+
+ final List<MockFlowFile> flowFiles =
runner.getFlowFilesForRelationship(ConsumeKafka.SUCCESS);
+ assertEquals(1, flowFiles.size());
+
+ final MockFlowFile flowFile = flowFiles.getFirst();
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_TOPIC,
topic);
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_OFFSET,
"2");
+
flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_MAX_OFFSET, "2");
+ flowFile.assertAttributeEquals(KafkaFlowFileAttribute.KAFKA_COUNT,
"1");
+ flowFile.assertAttributeEquals("record.count", "1");
+
flowFile.assertAttributeNotExists(KafkaFlowFileAttribute.KAFKA_TOMBSTONE);
+
+ final ArrayNode records = assertInstanceOf(ArrayNode.class,
objectMapper.readTree(flowFile.getContent()));
+ assertEquals(1, records.size());
+ final ObjectNode record = assertInstanceOf(ObjectNode.class,
records.get(0));
+ assertEquals(1, record.get("id").asInt());
+ assertEquals("A", record.get("name").asText());
+ assertEquals(2L, record.get("kafkaOffset").asLong());
+ }
+
+ private String configureRecord(final OutputStrategy outputStrategy) {
Review Comment:
Nit: it isn't clear what this method is doing. Let's remove the common setup
logic into `@BeforeEach`. And here set processing strategy only.
##########
nifi-extension-bundles/nifi-kafka-bundle/nifi-kafka-processors/src/test/java/org/apache/nifi/kafka/processors/ConsumeKafkaTest.java:
##########
@@ -365,6 +372,80 @@ public void
testOnTriggerDemarcatorRecordsOriginalKafkaBytes() throws Initializa
runner.assertTransferCount(ConsumeKafka.SUCCESS, 1);
}
+ @Test
+ public void testVerifySkipsTombstoneAndParsesFollowingRecord() throws
Exception {
Review Comment:
Nit: could you please move these tests right below the other `testVerify`
tests? Optionally, you can use `@Nested` class for all verification tests.
##########
nifi-extension-bundles/nifi-kafka-bundle/nifi-kafka-service-api/src/main/java/org/apache/nifi/kafka/service/api/record/ByteRecord.java:
##########
@@ -23,7 +23,8 @@
import java.util.Optional;
/**
- * Byte Record translation of Kafka Record with byte arrays for key and value
properties
+ * Byte Record translation of Kafka Record with byte arrays for key and value
properties.
+ * Kafka tombstone is represented by {@code tombstone == true} and an empty
{@code value} array.
Review Comment:
Nit: if `isTombstone()` returns `true`, we shouldn't even care what's put
inside the `value` field.
I'd omit this line from the doc, the presence of `isTombstone()` method
makes it evident how to check for tombstones.
##########
nifi-extension-bundles/nifi-kafka-bundle/nifi-kafka-3-integration/src/test/java/org/apache/nifi/kafka/processors/ConsumeKafkaRecordTombstoneIT.java:
##########
@@ -0,0 +1,422 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.nifi.kafka.processors;
+
+import com.fasterxml.jackson.databind.node.ArrayNode;
+import com.fasterxml.jackson.databind.node.ObjectNode;
+import org.apache.avro.Schema;
+import org.apache.avro.file.DataFileWriter;
+import org.apache.avro.generic.GenericData;
+import org.apache.avro.generic.GenericDatumWriter;
+import org.apache.avro.generic.GenericRecord;
+import org.apache.nifi.avro.AvroReader;
+import org.apache.nifi.kafka.processors.consumer.ProcessingStrategy;
+import org.apache.nifi.kafka.processors.producer.wrapper.InjectMetadataRecord;
+import org.apache.nifi.kafka.processors.producer.wrapper.WrapperRecord;
+import org.apache.nifi.kafka.service.api.consumer.AutoOffsetReset;
+import org.apache.nifi.kafka.shared.attribute.KafkaFlowFileAttribute;
+import org.apache.nifi.kafka.shared.property.OutputStrategy;
+import org.apache.nifi.reporting.InitializationException;
+import org.apache.nifi.schema.access.SchemaAccessUtils;
+import org.apache.nifi.util.MockFlowFile;
+import org.apache.nifi.util.TestRunner;
+import org.apache.nifi.util.TestRunners;
+import org.junit.jupiter.api.BeforeEach;
+import org.junit.jupiter.api.Test;
+
+import java.io.ByteArrayOutputStream;
+import java.io.IOException;
+import java.nio.charset.StandardCharsets;
+import java.time.Duration;
+import java.util.List;
+import java.util.UUID;
+import java.util.concurrent.ExecutionException;
+
+import static org.junit.jupiter.api.Assertions.assertEquals;
+import static org.junit.jupiter.api.Assertions.assertInstanceOf;
+
+/**
+ * RECORD output for a batch that contains a non-null empty value, a null
tombstone, and a real value.
+ * A null tombstone is omitted from success and parse.failure. A JSON reader
emits only the real value.
+ * An Avro reader routes the non-null empty value to parse.failure and writes
the real value to success.
+ */
+class ConsumeKafkaRecordTombstoneIT extends AbstractConsumeKafkaIT {
+
+ private static final String NON_EMPTY_VALUE = "{\"id\":1,\"name\":\"A\"}";
+
+ private static final byte[] NON_EMPTY_BYTES =
NON_EMPTY_VALUE.getBytes(StandardCharsets.UTF_8);
+
+ private static final byte[] EMPTY_BYTES = new byte[0];
+
+ private static final String AVRO_SCHEMA = """
+ {
+ "type": "record",
+ "name": "message",
+ "fields": [
+ {"name": "id", "type": "int"},
+ {"name": "name", "type": "string"}
+ ]
+ }
+ """;
+
+ private static final String MESSAGE_KEY = "key";
+
+ private static final int FIRST_PARTITION = 0;
+
+ private static final Duration POLL_TIMEOUT = Duration.ofSeconds(30);
+
+ private TestRunner runner;
+
+ @BeforeEach
+ void setRunner() throws InitializationException {
+ runner = TestRunners.newTestRunner(ConsumeKafka.class);
+ addKafkaConnectionService(runner);
+ runner.setProperty(ConsumeKafka.CONNECTION_SERVICE,
CONNECTION_SERVICE_ID);
+ addRecordReaderService(runner);
+ addRecordWriterService(runner);
+ }
+
+ @Test
+ void testOutputStrategyUseValue() throws ExecutionException,
InterruptedException, IOException {
Review Comment:
Multiple setup and verification steps are repeated across all of the tests.
Let's move them into helper setup/assertion methods
##########
nifi-extension-bundles/nifi-kafka-bundle/nifi-kafka-service-shared/src/main/java/org/apache/nifi/kafka/service/consumer/RecordIterable.java:
##########
@@ -58,11 +58,10 @@ public ByteRecord next() {
recordHeaders.add(recordHeader);
});
- // Support Kafka tombstones
- byte[] value = consumerRecord.value();
- if (value == null) {
- value = new byte[0];
- }
+ // A null Kafka value is a tombstone. Keep that distinct from a
non-null empty value.
+ final byte[] originalValue = consumerRecord.value();
+ final boolean isTombstone = originalValue == null;
+ final byte[] value = isTombstone ? new byte[0] : originalValue;
Review Comment:
I assume this logic is going to be revised as a part of
https://issues.apache.org/jira/browse/NIFI-16517, correct?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
