Anurag Mantripragada created SPARK-59945:
--------------------------------------------
Summary: Reject a bucket count outside the INT range in partition
and ordering transforms
Key: SPARK-59945
URL: https://issues.apache.org/jira/browse/SPARK-59945
Project: Spark
Issue Type: Bug
Components: SQL
Affects Versions: 5.0.0
Reporter: Anurag Mantripragada
In AstBuilder.visitTransform, a bucket count is converted with .toInt, so
bucket(4294967300, id) is silently truncated to bucket(4, id), while a DECIMAL
or TINYINT count is rejected. This affects PARTITIONED BY and, since
SPARK-34586 (https://github.com/apache/spark/pull/58153), ORDERED BY in
CREATE/REPLACE TABLE.
Reject a count outside the INT range with the existing invalid-bucket-number
error instead of truncating it.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]