shunping commented on code in PR #29546:
URL: https://github.com/apache/beam/pull/29546#discussion_r1416553565
##########
sdks/java/io/kafka/src/main/java/org/apache/beam/sdk/io/kafka/ReadFromKafkaDoFn.java:
##########
@@ -431,36 +449,52 @@ public ProcessContinuation processElement(
if (!tracker.tryClaim(rawRecord.offset())) {
return ProcessContinuation.stop();
}
- KafkaRecord<K, V> kafkaRecord =
- new KafkaRecord<>(
- rawRecord.topic(),
- rawRecord.partition(),
- rawRecord.offset(),
- ConsumerSpEL.getRecordTimestamp(rawRecord),
- ConsumerSpEL.getRecordTimestampType(rawRecord),
- ConsumerSpEL.hasHeaders() ? rawRecord.headers() : null,
- ConsumerSpEL.deserializeKey(keyDeserializerInstance,
rawRecord),
- ConsumerSpEL.deserializeValue(valueDeserializerInstance,
rawRecord));
- int recordSize =
- (rawRecord.key() == null ? 0 : rawRecord.key().length)
- + (rawRecord.value() == null ? 0 : rawRecord.value().length);
- avgRecordSize
- .getUnchecked(kafkaSourceDescriptor.getTopicPartition())
- .update(recordSize, rawRecord.offset() - expectedOffset);
- rawSizes.update(recordSize);
- expectedOffset = rawRecord.offset() + 1;
- Instant outputTimestamp;
- // The outputTimestamp and watermark will be computed by
timestampPolicy, where the
- // WatermarkEstimator should be a manual one.
- if (timestampPolicy != null) {
- TimestampPolicyContext context =
- updateWatermarkManually(timestampPolicy, watermarkEstimator,
tracker);
- outputTimestamp = timestampPolicy.getTimestampForRecord(context,
kafkaRecord);
- } else {
- Preconditions.checkStateNotNull(this.extractOutputTimestampFn);
- outputTimestamp = extractOutputTimestampFn.apply(kafkaRecord);
+ try {
+ KafkaRecord<K, V> kafkaRecord =
+ new KafkaRecord<>(
+ rawRecord.topic(),
+ rawRecord.partition(),
+ rawRecord.offset(),
+ ConsumerSpEL.getRecordTimestamp(rawRecord),
+ ConsumerSpEL.getRecordTimestampType(rawRecord),
+ ConsumerSpEL.hasHeaders() ? rawRecord.headers() : null,
+ ConsumerSpEL.deserializeKey(keyDeserializerInstance,
rawRecord),
+ ConsumerSpEL.deserializeValue(valueDeserializerInstance,
rawRecord));
+ int recordSize =
+ (rawRecord.key() == null ? 0 : rawRecord.key().length)
+ + (rawRecord.value() == null ? 0 :
rawRecord.value().length);
+ avgRecordSize
+ .getUnchecked(kafkaSourceDescriptor.getTopicPartition())
+ .update(recordSize, rawRecord.offset() - expectedOffset);
+ rawSizes.update(recordSize);
+ expectedOffset = rawRecord.offset() + 1;
+ Instant outputTimestamp;
+ // The outputTimestamp and watermark will be computed by
timestampPolicy, where the
+ // WatermarkEstimator should be a manual one.
+ if (timestampPolicy != null) {
+ TimestampPolicyContext context =
+ updateWatermarkManually(timestampPolicy, watermarkEstimator,
tracker);
+ outputTimestamp = timestampPolicy.getTimestampForRecord(context,
kafkaRecord);
+ } else {
+ Preconditions.checkStateNotNull(this.extractOutputTimestampFn);
+ outputTimestamp = extractOutputTimestampFn.apply(kafkaRecord);
+ }
+ receiver
+ .get(recordTag)
+ .outputWithTimestamp(KV.of(kafkaSourceDescriptor,
kafkaRecord), outputTimestamp);
+ } catch (SerializationException e) {
+ // This exception should only occur during the key and value
deserialization when
+ // creating the Kafka Record
+ badRecordRouter.route(
Review Comment:
QQ: do we have timestamp in DLQ?
--
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]
