[GitHub] [beam] dpcollins-google commented on a diff in pull request #22612: Reimplement Pub/Sub Lite's I/O using UnboundedSource.

[GitBox <gi...@apache.org>]

dpcollins-google commented on code in PR #22612:
URL: https://github.com/apache/beam/pull/22612#discussion_r939774567


##########
sdks/java/io/google-cloud-platform/src/main/java/org/apache/beam/sdk/io/gcp/pubsublite/internal/SubscribeTransform.java:
##########
@@ -134,19 +160,28 @@ private TopicPath getTopicPath() {
     }
   }
 
-  @Override
-  public PCollection<SequencedMessage> expand(PBegin input) {
-    PCollection<SubscriptionPartition> subscriptionPartitions;
-    subscriptionPartitions =
+  @SuppressWarnings("unused")
+  private PCollection<SequencedMessage> expandSdf(PBegin input) {

Review Comment:
   I'm not planning on deleting the SDF based implementation, but this seems like a lot of restructuring effort for ambiguous value. All implementations should support UnboundedSource at present IIUC (either directly or via wrapping for runnerv2)
   
   > potentially more performant
   
   The amount of work actually performed by either the SDF or UnboundedSource implementation is so miniscule (pulling out of an in-memory buffer) that I think this is probably moot. I'd note that it does not appear to perform better in testing.
   
   



-- 
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: github-unsubscribe@beam.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org