fapaul commented on a change in pull request #16769:
URL: https://github.com/apache/flink/pull/16769#discussion_r686855696
##########
File path:
flink-connectors/flink-connector-kafka/src/main/java/org/apache/flink/streaming/connectors/kafka/table/KafkaDynamicSink.java
##########
@@ -172,20 +185,43 @@ public SinkRuntimeProvider getSinkRuntimeProvider(Context
context) {
final SerializationSchema<RowData> valueSerialization =
createSerialization(context, valueEncodingFormat,
valueProjection, null);
- final FlinkKafkaProducer<RowData> kafkaProducer =
- createKafkaProducer(keySerialization, valueSerialization);
-
if (flushMode.isEnabled() && upsertMode) {
BufferedUpsertSinkFunction buffedSinkFunction =
Review comment:
I think @twalthr is right we cannot easily remove the
`BufferedUpsertSinkFunction` because it drives a valid use case.
https://issues.apache.org/jira/browse/FLINK-21191 I can imagine we provide a
more generalized way to apply this "buffer" functionality to all sinks
supporting upsert.
tldr; we probably have to keep the old stack until we find a solution how to
implement this for upsert sinks with FLIP-143.
--
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]
