[
https://issues.apache.org/jira/browse/SPARK-10747?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15425547#comment-15425547
]
Xin Wu commented on SPARK-10747:
--------------------------------
[~hvanhovell] Yes. Since we have native parser now, we can do this within
SparkSQL. I can work on this. Thanks!
> add support for window specification to include how NULLS are ordered
> ---------------------------------------------------------------------
>
> Key: SPARK-10747
> URL: https://issues.apache.org/jira/browse/SPARK-10747
> Project: Spark
> Issue Type: Improvement
> Components: SQL
> Affects Versions: 1.5.0
> Reporter: N Campbell
>
> You cannot express how NULLS are to be sorted in the window order
> specification and have to use a compensating expression to simulate.
> Error: org.apache.spark.sql.AnalysisException: line 1:76 missing ) at 'nulls'
> near 'nulls'
> line 1:82 missing EOF at 'last' near 'nulls';
> SQLState: null
> Same limitation as Hive reported in Apache JIRA HIVE-9535 )
> This fails
> select rnum, c1, c2, c3, dense_rank() over(partition by c1 order by c3 desc
> nulls last) from tolap
> select rnum, c1, c2, c3, dense_rank() over(partition by c1 order by case when
> c3 is null then 1 else 0 end) from tolap
--
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
