[
https://issues.apache.org/jira/browse/HIVE-741?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12901508#action_12901508
]
He Yongqiang commented on HIVE-741:
-----------------------------------
also about Ning's comments:
>>2) SMBMapJoinOperator.compareKey() is called for each row so it is critical
>>for performance. In your code the hasNullElement() could be called 4 times in
>>the worse case. If you cache the result it can be called only twice.
Agree. Not sure how much overhead is there, will try to estimate the overhead
over production running. That will be great if you can try to cache the null
check results, so that it can only happen one time for each key.
> NULL is not handled correctly in join
> -------------------------------------
>
> Key: HIVE-741
> URL: https://issues.apache.org/jira/browse/HIVE-741
> Project: Hadoop Hive
> Issue Type: Bug
> Reporter: Ning Zhang
> Assignee: Amareshwari Sriramadasu
> Attachments: patch-741-1.txt, patch-741-2.txt, patch-741-3.txt,
> patch-741-4.txt, patch-741-5.txt, patch-741.txt, smbjoin_nulls.q.txt
>
>
> With the following data in table input4_cb:
> Key Value
> ------ --------
> NULL 325
> 18 NULL
> The following query:
> {code}
> select * from input4_cb a join input4_cb b on a.key = b.value;
> {code}
> returns the following result:
> NULL 325 18 NULL
> The correct result should be empty set.
> When 'null' is replaced by '' it works.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
