[
https://issues.apache.org/jira/browse/SPARK-14990?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Reynold Xin resolved SPARK-14990.
---------------------------------
Resolution: Fixed
Assignee: Reynold Xin
Fix Version/s: 2.0.0
> nvl, coalesce, array functions with parameter of type "array"
> -------------------------------------------------------------
>
> Key: SPARK-14990
> URL: https://issues.apache.org/jira/browse/SPARK-14990
> Project: Spark
> Issue Type: Bug
> Components: SQL
> Affects Versions: 1.6.1
> Reporter: Oleg Danilov
> Assignee: Reynold Xin
> Priority: Minor
> Fix For: 2.0.0
>
>
> Steps to reproduce:
> 1. create table tmp(col1 int, col2 array<int>)
> 2. insert values:
> {code}
> 1, [1]
> 2, [2]
> 3, NULL
> {code}
> 3. run query
> select col1, coalesce(col2, array(1,2,3)) from tmp;
> Expected result:
> {code}
> 1, [1]
> 2, [2]
> 3, [1,2,3]
> {code}
> Current result:
> {code}
> Error: org.apache.spark.sql.AnalysisException: cannot resolve
> 'coalesce(col2,array(1,2,3))' due to data type mismatch: input to function
> coalesce should all be the same type, but it's [array<int>, array<int>]; line
> 1 pos 38 (state=,code=0)
> {code}
> The fix seems to be pretty easy, will create a PR soon.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
