[jira] [Updated] (IGNITE-16334) Calcite. Need to investigate why *Join*Rule ordering can affect the final plan.

["Evgeny Stanilovsky (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/IGNITE-16334?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Evgeny Stanilovsky updated IGNITE-16334:
----------------------------------------
    Summary: Calcite. Need to investigate why *Join*Rule ordering can affect the final plan.  (was: Calcite. Need to investigate why *JoinRule ordering can affect the final plan.)

> Calcite. Need to investigate why *Join*Rule ordering can affect the final plan.
> -------------------------------------------------------------------------------
>
>                 Key: IGNITE-16334
>                 URL: https://issues.apache.org/jira/browse/IGNITE-16334
>             Project: Ignite
>          Issue Type: Bug
>          Components: sql
>            Reporter: Evgeny Stanilovsky
>            Priority: Major
>              Labels: calcite, calcite2-required, calcite3-required
>
> For now in optimization phase we have such join rules ordering, if MergeJoinConverterRule occurs at the end of this enumeration some tests will be failed.
> {noformat}
>                     MergeJoinConverterRule.INSTANCE,
>                     CorrelatedNestedLoopJoinRule.INSTANCE,
>                     CorrelateToNestedLoopRule.INSTANCE,
>                     NestedLoopJoinConverterRule.INSTANCE,
> {noformat}
> Seems [1] can point out the root of such a problem.
> [1] https://issues.apache.org/jira/browse/CALCITE-4542



--
This message was sent by Atlassian Jira
(v8.20.1#820001)