Github user srdo commented on a diff in the pull request:
https://github.com/apache/storm/pull/2465#discussion_r157346634
--- Diff:
external/storm-kafka-client/src/main/java/org/apache/storm/kafka/spout/KafkaSpoutMessageId.java
---
@@ -18,36 +18,88 @@
package org.apache.storm.kafka.spout;
+import java.io.IOException;
--- End diff --
I think maybe we should split this object. We don't need half of the fields
for metadata, and several fields aren't needed for emit/ack handling either.
Maybe have an object with thread name and topology id for metadata, and have
KafkaSpoutMessageId only contain the data we need for ack handling? That way we
can also avoid doing stuff like storing the topic partition and offset in
metadata, which IMO makes no sense.
---
