dawidwys commented on a change in pull request #16589:
URL: https://github.com/apache/flink/pull/16589#discussion_r682523739
##########
File path:
flink-connectors/flink-connector-kafka/src/test/java/org/apache/flink/streaming/connectors/kafka/shuffle/KafkaShuffleITCase.java
##########
@@ -56,6 +56,7 @@
import static
org.apache.flink.streaming.connectors.kafka.shuffle.FlinkKafkaShuffle.PARTITION_NUMBER;
import static
org.apache.flink.streaming.connectors.kafka.shuffle.FlinkKafkaShuffle.PRODUCER_PARALLELISM;
import static org.apache.flink.test.util.TestUtils.tryExecute;
+import static org.junit.Assert.assertEquals;
Review comment:
Changes in comments are related to the fix. There is a slight change in
behaviour for `ProcessingTime` that we emit the `MAX_WATERMARK`. Previously it
was filtered by the `StreamSourceContext`.
I moved emitting the `MAX_WATERMARK` out from the `StreamSource` (operator)
to the `StreamSourceTask` (task). I did it because if we restore a fully
finished task the StreamSourceContext is not initialised when calling
`advanceToEndOfTime`. We skip life cycle methods of an operator in such a case.
I think this change should be fine though. It is also how we do it in new
sources.
--
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]
