yashmayya opened a new pull request, #17750:
URL: https://github.com/apache/pinot/pull/17750
- Currently, if we run a query like `SELECT DISTINCT_COUNT(xyz) FROM
myEmptyTable`, it returns an error like: `Cannot invoke "java.util.Set.size()"
because "intermediateResult" is null `.
- The bug is in how the multi-stage query engine handles empty tables during
the `FINAL` aggregation stage. Two code paths interact to cause the NPE:
- `processMerge()` skips null intermediate results. When a table is empty,
the leaf stage produces null intermediate results. The merge stage skips all of
them, so the merge result holders remain null.
- `extractFinalResult()` doesn't handle `null` in various different
aggregation functions.
- This patch fixes all the aggregation functions that don't currently handle
null intermediate results appropriately. While we could also patch the
multi-stage aggregation executor to first call `extractAggregationResult()`
before calling `extractFinalResult()` in these situations, it's more robust to
make sure that the aggregation functions' `extractFinalResult()` method
implementations also handle `null` values appropriately.
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
