You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@calcite.apache.org by "Julian Hyde (JIRA)" <ji...@apache.org> on 2019/02/27 17:40:00 UTC

[jira] [Commented] (CALCITE-2879) Improve filter condition in Join

    [ https://issues.apache.org/jira/browse/CALCITE-2879?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16779560#comment-16779560 ] 

Julian Hyde commented on CALCITE-2879:
--------------------------------------

-1

I think there's already a way to do this, using pulled up predicates. It's superior to your "structural" approach, because it also allows predicates that have bubbled up from the sources.

> Improve filter condition in Join
> --------------------------------
>
>                 Key: CALCITE-2879
>                 URL: https://issues.apache.org/jira/browse/CALCITE-2879
>             Project: Calcite
>          Issue Type: Improvement
>            Reporter: yuqi
>            Priority: Major
>              Labels: pull-request-available
>          Time Spent: 10m
>  Remaining Estimate: 0h
>
> Currently, filter condition in join have not been optimized very well, let's take a simple example
> {code:sql}
> select a.id, b.name from Ta a inner join Tb b on a.id = b.id and a.id < 5;
> {code}
> now, only a.id < 5 can push down, however, as for the sql above,  
> we can infer a.id < 5 and b.id < 5 and both of them can push down. 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)