[
https://issues.apache.org/jira/browse/CALCITE-2586?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16625915#comment-16625915
]
Vladimir Sitnikov commented on CALCITE-2586:
--------------------------------------------
[~julianhyde], fun fact: the optimization in question is "transformation of
CASE ... END to AND/OR". As we discussed, those expressions are likely to
have different semantics with regard to the predicate evaluation order.
So the question is: should we cut the simplification from RexSimplify
altogether?
> RexSimplify: case with repeated branches results in assertion error
> -------------------------------------------------------------------
>
> Key: CALCITE-2586
> URL: https://issues.apache.org/jira/browse/CALCITE-2586
> Project: Calcite
> Issue Type: Bug
> Components: core
> Affects Versions: 1.17.0
> Reporter: Vladimir Sitnikov
> Assignee: Julian Hyde
> Priority: Major
> Labels: newbie
>
> {code:java} case_(
> eq(falseLiteral, falseLiteral), falseLiteral,
> eq(falseLiteral, falseLiteral), isNotNull(trueLiteral),
> isNull(trueLiteral)
> );
> {code}
> {noformat}
> java.lang.AssertionError: result mismatch: when applied to {},
> CASE(=(false, false), false, =(false, false), IS NOT NULL(true), IS
> NULL(true)) yielded false,
> and =(false, false) yielded true{noformat}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
