[jira] [Commented] (CALCITE-5663) [TestKit] RelOptFixture does not enforce the collation

["Marieke Gueye (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CALCITE-5663?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17714715#comment-17714715 ] 

Marieke Gueye commented on CALCITE-5663:
----------------------------------------

My initial thought was to add a new method in the relSupplier that returns the relRoot if available, not sure this is the right place to add it though, any thoughts ?

> [TestKit] RelOptFixture does not enforce the collation
> ------------------------------------------------------
>
>                 Key: CALCITE-5663
>                 URL: https://issues.apache.org/jira/browse/CALCITE-5663
>             Project: Calcite
>          Issue Type: Bug
>            Reporter: Marieke Gueye
>            Priority: Minor
>
> In RelOptFixture, we currently change the traitsets to enforce  EnumerableConvention.INSTANCE, however, in this instance we forget to port the collation in the traitset.
> {code:java}
> if (planner instanceof VolcanoPlanner)
> { r2 = planner.changeTraits(relBefore, relBefore.getTraitSet().replace(EnumerableConvention.INSTANCE)); }
> else
> { r2 = relBefore; }
>  
> {code}
> The problem goes even deeper, as of now, there is no way to get access to the collation as it lives in the relRoot, and we currently only access the relNode through  the relSupplier
>  
> The consequence of this can be pretty dire as it may mean that some rules might be incorrectly tested.
>  
>  



--
This message was sent by Atlassian Jira
(v8.20.10#820010)