[
https://issues.apache.org/jira/browse/SPARK-29067?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16929066#comment-16929066
]
Mangesh Rananavare commented on SPARK-29067:
Thank you. Also to add this is the case with all math operations. I think it
would be better to expose these with specific parameter types instead of having
the parameter as Object.
Current : divide(Object obj)
Instead : divide(Column col), divide(Number num)
> divide function does not throw an error, if a number is not passed to it
>
>
> Key: SPARK-29067
> URL: https://issues.apache.org/jira/browse/SPARK-29067
> Project: Spark
> Issue Type: Bug
> Components: Java API
>Affects Versions: 2.4.3
> Environment: Apache Spark Java : 2.4.3
>Reporter: Mangesh Rananavare
>Priority: Major
>
> Ex.
> dataset.where(col("col1").divide("col2")).$greater(10)).show();
> If you see closely I forgot to wrap the divide parameter "col2" into col()
> function, so basically I pass a String. This should give me a
> NumberFormatException!, but instead the where clause resolves to null and I
> get a empty dataset as a result of above computation!!
--
This message was sent by Atlassian Jira
(v8.3.2#803003)
-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org