[
https://issues.apache.org/jira/browse/SPARK-28018?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Dongjoon Hyun resolved SPARK-28018.
-----------------------------------
Resolution: Invalid
This is closed by the author due to the loss of precision. Please see the PR
comments for the detail.
> Allow upcasting decimal to double/float
> ---------------------------------------
>
> Key: SPARK-28018
> URL: https://issues.apache.org/jira/browse/SPARK-28018
> Project: Spark
> Issue Type: Improvement
> Components: SQL
> Affects Versions: 3.0.0
> Reporter: Gengliang Wang
> Priority: Major
>
> Currently, Spark only allows upcasting DecimalType to IntegralType or
> DecimalType.
> This PR proposes: if the target DecimalType is tighter than DoubleType or
> FloatType, we can upcast it as well.
> The upcasting matters because it blocks
> https://github.com/apache/spark/pull/24806. E.g, if there is a table "t" with
> only one column of double type,
> ```
> INSERT INTO TABLE t SELECT 10.0
> ```
> The data type of value 10.0 is DecimalType(3, 1). In the current code, the
> upcasting failed.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
