[
https://issues.apache.org/jira/browse/HIVE-15986?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15959905#comment-15959905
]
Pengcheng Xiong commented on HIVE-15986:
----------------------------------------
[~vgarg], could u try this please?
{code}
precedenceNotDistinctOperator
:
KW_IS KW_NOT KW_DISTINCT KW_FROM
;
precedenceEqualOperator
:
EQUAL | EQUAL_NS | NOTEQUAL | KW_IS KW_DISTINCT KW_FROM -> EQUAL_NS
;
precedenceEqualExpression
:
(precedenceSimilarExpression -> precedenceSimilarExpression)
(equal=precedenceEqualOperator p=precedenceSimilarExpression
-> ^($equal {$precedenceEqualExpression.tree} $p)
|
dist=precedenceNotDistinctOperator p=precedenceSimilarExpression
-> ^(KW_NOT ^(EQUAL_NS {$precedenceEqualExpression.tree} $p))
)*
-> {$precedenceEqualExpression.tree}
;
{code}
> Support "is [not] distinct from"
> --------------------------------
>
> Key: HIVE-15986
> URL: https://issues.apache.org/jira/browse/HIVE-15986
> Project: Hive
> Issue Type: Sub-task
> Components: SQL
> Reporter: Carter Shanklin
> Assignee: Vineet Garg
> Attachments: HIVE-15986.1.patch, HIVE-15986.2.patch,
> HIVE-15986.3.patch, HIVE-15986.4.patch
>
>
> Support standard "is [not] distinct from" syntax. For example this gives a
> standard way to do a comparison to null safe join: select * from t1 join t2
> on t1.x is not distinct from t2.y. SQL standard reference Section 8.15
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
