robertwb commented on code in PR #29261:
URL: https://github.com/apache/beam/pull/29261#discussion_r1379453804
##########
sdks/java/io/kafka/src/main/java/org/apache/beam/sdk/io/kafka/KafkaWriteSchemaTransformProvider.java:
##########
@@ -167,8 +182,18 @@ public PCollectionRowTuple expand(PCollectionRowTuple
input) {
.withKeySerializer(ByteArraySerializer.class)
.withValueSerializer(ByteArraySerializer.class));
- return PCollectionRowTuple.of(
- "errors", outputTuple.get(ERROR_TAG).setRowSchema(ERROR_SCHEMA));
+ PCollection<Row> postWrite =
+ outputTuple
+ .get(OUTPUT_TAG)
+ .apply("post-write", ParDo.of(new NoOutputDoFn<>()))
Review Comment:
This isn't quite safe because we aren't guaranteed the write itself happened
yet, just the byte mapping. For now just emit either the empty or error tuple,
but drop a TODO to catch any errors KafkaIO itself throws and also follow its
output (once it's updated from PDone.)
--
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]
