[
https://issues.apache.org/jira/browse/CALCITE-2327?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16594915#comment-16594915
]
Zoltan Haindrich commented on CALCITE-2327:
-------------------------------------------
I disagree with "off the track" - but even if I was: why didn't you
communicated at least that?
1) the over optimization have caused invalid results (that optimization have a
counterpart on the "unknownAsFalse=true" branch which was not touched)
I think that not having a bad optimization is better than having it...
2) if you think about the materializationview related test; I think in those
cases the original idea was to use withUnknownAsFalse=true...that's why I've
mentioned [~jcamachorodriguez] for the first place... (as I say in my *first*
comment in this jira...)
....and you have just pushed another commit...cool...at least it seems okay
now...
> In 3 valued logic mode (b and not b) may not be simplified to false
> -------------------------------------------------------------------
>
> Key: CALCITE-2327
> URL: https://issues.apache.org/jira/browse/CALCITE-2327
> Project: Calcite
> Issue Type: Bug
> Reporter: Zoltan Haindrich
> Assignee: Zoltan Haindrich
> Priority: Major
> Fix For: 1.18.0
>
>
> currently its simplified to false; but that is not correct when a is unknown
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
