You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hive.apache.org by "Gopal V (JIRA)" <ji...@apache.org> on 2018/10/16 05:46:00 UTC

[jira] [Updated] (HIVE-20749) Constraints: Null scan optimizer does not remove entire TableScan

     [ https://issues.apache.org/jira/browse/HIVE-20749?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Gopal V updated HIVE-20749:
---------------------------
    Affects Version/s: 4.0.0

> Constraints: Null scan optimizer does not remove entire TableScan
> -----------------------------------------------------------------
>
>                 Key: HIVE-20749
>                 URL: https://issues.apache.org/jira/browse/HIVE-20749
>             Project: Hive
>          Issue Type: Bug
>          Components: CBO, Logical Optimizer
>    Affects Versions: 4.0.0
>            Reporter: Gopal V
>            Priority: Major
>
> With PK constraints on store table, the following query fails to remove the entire join (and cut the query short).
> {code}
>  explain select count(1) from store_sales, store where ss_store_sk = s_store_sk and s_store_sk is null and ss_sold_date_sk between 2450816 and 2450816+365;
> {code}
> Calcite generates the following optimized query, but the NullScan optimizer does not kick in for the query.
> {code}
> SELECT COUNT(*) AS `$f0`
> FROM (SELECT *
> FROM (SELECT 0 AS `DUMMY`
> FROM `tpcds_bin_partitioned_orc_10000`.`store_sales`) AS `t`
> WHERE FALSE) AS `t0`,
> (SELECT *
> FROM (SELECT 0 AS `DUMMY`
> FROM `tpcds_bin_partitioned_orc_10000`.`store`) AS `t1`
> WHERE FALSE) AS `t2`
> {code}



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