[
https://issues.apache.org/jira/browse/HIVE-21928?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16874248#comment-16874248
]
Zoltan Haindrich commented on HIVE-21928:
-----------------------------------------
Yes, I agree - it was only correct for the last term of ANDs; the situation
before HIVE-20260 was that it scaled everything down too much...I remember
wanting to get back to that clearAffectedColumns() call; I should have placed
there a FIXME - to alert me later...
I think the following continue block should be removed; even thru the rowcount
is not changed; the affectedcolumns might have, is there any reason I don't see
why we should do it?
{code}
if (evaluatedRowCount == newNumRows) {
continue;
}
{code}
> Fix for statistics annotation in nested AND expressions
> -------------------------------------------------------
>
> Key: HIVE-21928
> URL: https://issues.apache.org/jira/browse/HIVE-21928
> Project: Hive
> Issue Type: Bug
> Components: Physical Optimizer
> Reporter: Jesus Camacho Rodriguez
> Assignee: Jesus Camacho Rodriguez
> Priority: Critical
> Attachments: HIVE-21928.patch
>
>
> Discovered while working on HIVE-21867. Having predicates with nested AND
> expressions may result in different stats, even if predicates are basically
> similar (from stats estimation standpoint).
> For instance, stats for {{AND(x=5, true, true)}} are different from {{x=5}}.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
