[
https://issues.apache.org/jira/browse/HIVE-9534?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15144744#comment-15144744
]
Aihua Xu commented on HIVE-9534:
--------------------------------
Thanks Yongzhi. Yeah. The current implementation will do a sort on the
partitioning columns so each partition data in the reducer is sorted and we
only need to check previous row against the current row to find if it's
distinct or not.
> incorrect result set for query that projects a windowed aggregate
> -----------------------------------------------------------------
>
> Key: HIVE-9534
> URL: https://issues.apache.org/jira/browse/HIVE-9534
> Project: Hive
> Issue Type: Bug
> Components: PTF-Windowing
> Reporter: N Campbell
> Assignee: Aihua Xu
> Attachments: HIVE-9534.1.patch, HIVE-9534.2.patch, HIVE-9534.3.patch,
> HIVE-9534.4.patch
>
>
> Result set returned by Hive has one row instead of 5
> {code}
> select avg(distinct tsint.csint) over () from tsint
> create table if not exists TSINT (RNUM int , CSINT smallint)
> ROW FORMAT DELIMITED FIELDS TERMINATED BY '|' LINES TERMINATED BY '\n'
> STORED AS TEXTFILE;
> 0|\N
> 1|-1
> 2|0
> 3|1
> 4|10
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
