[jira] [Commented] (SPARK-52969) DataSourceV2 ORCScan should implements SupportsRuntimeV2Filtering to support DPP

["Shao Feng Shi (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SPARK-52969?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18013824#comment-18013824 ] 

Shao Feng Shi commented on SPARK-52969:
---------------------------------------

+1, we need this feature, looking forward to this!

> DataSourceV2 ORCScan should implements SupportsRuntimeV2Filtering to support DPP
> --------------------------------------------------------------------------------
>
>                 Key: SPARK-52969
>                 URL: https://issues.apache.org/jira/browse/SPARK-52969
>             Project: Spark
>          Issue Type: New Feature
>          Components: SQL
>    Affects Versions: 3.4.4, 3.5.5, 4.0.0
>            Reporter: TIsNotT
>            Priority: Major
>              Labels: pull-request-available
>
> When I was reading the ORC table based on data source v2, I found that the execution plan could not generate DPP plan. When checking the reason, it was found that ORCscan did not implement SupportsRuntimeV2Filtering. Do we have a plan to do this, or do we have other ways to fulfill this requirement.
> The relevant code is as follows：
> case (resExp, r @ DataSourceV2ScanRelation(_, {color:red}scan: SupportsRuntimeV2Filtering{color}, _, _, _)) =>
>         val filterAttrs = V2ExpressionUtils.resolveRefs[Attribute](scan.filterAttributes, r)
>         if (resExp.references.subsetOf(AttributeSet(filterAttrs))) {
>           Some(r)
>         } else {
>           None
>         }



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

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org