[
https://issues.apache.org/jira/browse/SPARK-55980?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Hyukjin Kwon resolved SPARK-55980.
----------------------------------
Fix Version/s: 4.2.0
Resolution: Fixed
Issue resolved by pull request 54779
[https://github.com/apache/spark/pull/54779]
> Always apply _cast_back_float in numeric arithmetic
> ---------------------------------------------------
>
> Key: SPARK-55980
> URL: https://issues.apache.org/jira/browse/SPARK-55980
> Project: Spark
> Issue Type: Bug
> Components: Pandas API on Spark
> Affects Versions: 4.1.1
> Reporter: Devin Petersohn
> Assignee: Devin Petersohn
> Priority: Major
> Labels: pull-request-available
> Fix For: 4.2.0
>
>
> `_cast_back_float()` preserves float32 dtype in arithmetic results, matching
> pandas behavior. Under ANSI mode, Spark widens float32 to float64 during
> arithmetic with non-float32 numeric operands, and this helper casts it back.
> Currently the helper is only called when ANSI mode is enabled, but it should
> always be called so behavior is the same with or without ANSI mode enabled.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
