[
https://issues.apache.org/jira/browse/CALCITE-4467?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17264950#comment-17264950
]
Jesus Camacho Rodriguez commented on CALCITE-4467:
--------------------------------------------------
{quote}
It will not be easy to remove all simplifications concerning equality.
{quote}
[~julianhyde], are there other simplifications that come to your mind? Tbh I
did not take an exhaustive look but I assumed Calcite would not do folding and
instead relies on specific executor implementation (which would then enforce
the correct implementation-specific semantics), and that probably limits the
number of cases on which we will hit this issue.
{quote}
We could introduce a configuration property; people who strictly follow the
standard can benefit from the simplifications and those who don't can still run
the simplifier.
{quote}
In this specific case, I hit the issue because the RelBuilder calls
{{simplify}} when a Project is created. That's already configurable but I did
not want to disable all simplifications when a Project is created.
I could introduce some configuration variable to control the flow but I am not
sure if it is worth it if this would be the only simplification that would be
disabled? I don't have a strong opinion, I'd like to hear your ideas.
> Incorrect simplification for 'NaN' value
> ----------------------------------------
>
> Key: CALCITE-4467
> URL: https://issues.apache.org/jira/browse/CALCITE-4467
> Project: Calcite
> Issue Type: Bug
> Components: core
> Reporter: Jesus Camacho Rodriguez
> Assignee: Jesus Camacho Rodriguez
> Priority: Major
> Labels: pull-request-available
> Time Spent: 10m
> Remaining Estimate: 0h
>
> {{RexSimplify}} simplifies {{x = x}} to {{null or x is not null}} (similarly
> <= and >=), and {{x != x}} to {{null and x is null}} (similarly < and >).
> https://github.com/apache/calcite/blob/master/core/src/main/java/org/apache/calcite/rex/RexSimplify.java#L363
> This may not be applicable in some cases. For instance, if the type of x is
> floating-point, x could be 'NaN'. While some RDBMS consider 'NaN' = 'NaN'
> (e.g., Postgres), some others consider 'NaN' != 'NaN' following the IEEE 754
> standard. For the latest, the rewriting above will result in incorrect
> results.
> I think we should simply ignore this simplification for floating-point type.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
