[
https://issues.apache.org/jira/browse/SPARK-38558?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17507300#comment-17507300
]
Apache Spark commented on SPARK-38558:
--------------------------------------
User 'cashmand' has created a pull request for this issue:
https://github.com/apache/spark/pull/35863
> Remove unnecessary casts between IntegerType and IntDecimal
> -----------------------------------------------------------
>
> Key: SPARK-38558
> URL: https://issues.apache.org/jira/browse/SPARK-38558
> Project: Spark
> Issue Type: Improvement
> Components: SQL
> Affects Versions: 3.2.1
> Reporter: David Cashman
> Priority: Major
>
> In {{{}NTile{}}}, the number of rows per bucket is computed as {{{}n /
> buckets{}}}, where {{n}} is the partition size, and {{buckets}} is the
> argument to {{NTile}} (number of buckets). The code currently casts the
> arguments to IntDecimal, then casts the result back to IntegerType. This is
> unnecessary, since it is equivalent to just doing integer division, i.e.
> {{{}n div buckets{}}}.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
