damccorm commented on code in PR #40611:
URL: https://github.com/apache/beam/pull/40611#discussion_r4493521680
##########
runners/flink/src/main/java/org/apache/beam/runners/flink/FlinkPipelineOptions.java:
##########
@@ -367,6 +367,16 @@ public Long create(PipelineOptions options) {
void setFasterCopy(Boolean fasterCopy);
+ @Description(
+ "Batch only. Use a dedicated operator for the final step of GroupByKey
and Combine in the "
+ + "global window with the default trigger. It relies on Flink
sorting the input by key "
+ + "and does not use keyed state or timers. Has no effect in
streaming mode. "
+ + "Enabled by default; set to false to disable.")
+ @Default.Boolean(true)
+ Boolean getSortedBatchCombine();
Review Comment:
Nit (matching `getLazyBatchGroupingInputs()` in #40609): since
`SortedBatchCombineOperator` is only wired in
`FlinkStreamingAggregationsTranslators`, on Flink 1.x this also requires
`--useDataStreamForBatch` in batch mode:
```suggestion
@Description(
"Batch only. Use a dedicated operator for the final step of GroupByKey
and Combine in the "
+ "global window with the default trigger. It relies on Flink
sorting the input by key "
+ "and does not use keyed state or timers. Has no effect in
streaming mode, nor in "
+ "batch mode unless --useDataStreamForBatch is set. "
+ "Enabled by default; set to false to disable.")
@Default.Boolean(true)
Boolean getSortedBatchCombine();
```
--
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]
