[jira] [Updated] (HIVE-21021) Scalar subquery with only aggregate in subquery (no group by) has unnecessary sq_count_check branch

["Vineet Garg (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HIVE-21021?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Vineet Garg updated HIVE-21021:
-------------------------------
    Status: Patch Available  (was: Open)

> Scalar subquery with only aggregate in subquery (no group by) has unnecessary sq_count_check branch
> ---------------------------------------------------------------------------------------------------
>
>                 Key: HIVE-21021
>                 URL: https://issues.apache.org/jira/browse/HIVE-21021
>             Project: Hive
>          Issue Type: Improvement
>    Affects Versions: 3.0.0
>            Reporter: Vineet Garg
>            Assignee: Vineet Garg
>            Priority: Major
>         Attachments: HIVE-21021.1.patch
>
>
> {code:sql}
> CREATE TABLE `store_sales`(
>   `ss_sold_date_sk` int,
>   `ss_quantity` int,
>   `ss_list_price` decimal(7,2));
> CREATE TABLE `date_dim`(
>   `d_date_sk` int,
>   `d_year` int);
> explain cbo with avg_sales as
>  (select avg(quantity*list_price) average_sales
>   from (select ss_quantity quantity
>              ,ss_list_price list_price
>        from store_sales
>            ,date_dim
>        where ss_sold_date_sk = d_date_sk
>          and d_year between 1999 and 2001 ) x)
> select * from store_sales where ss_list_price > (select average_sales from avg_sales);
> {code}
> {noformat}
> CBO PLAN:
> HiveProject(ss_sold_date_sk=[$0], ss_quantity=[$1], ss_list_price=[$2])
>   HiveJoin(condition=[true], joinType=[inner], algorithm=[none], cost=[{2.0 rows, 0.0 cpu, 0.0 io}])
>     HiveJoin(condition=[>($2, $3)], joinType=[inner], algorithm=[none], cost=[{2.0 rows, 0.0 cpu, 0.0 io}])
>       HiveProject(ss_sold_date_sk=[$0], ss_quantity=[$1], ss_list_price=[$2])
>         HiveTableScan(table=[[sub, store_sales]], table:alias=[store_sales])
>       HiveProject($f0=[/($0, $1)])
>         HiveAggregate(group=[{}], agg#0=[sum($0)], agg#1=[count($0)])
>           HiveProject($f0=[*(CAST($1):DECIMAL(10, 0), $2)])
>             HiveJoin(condition=[=($0, $3)], joinType=[inner], algorithm=[none], cost=[{2.0 rows, 0.0 cpu, 0.0 io}])
>               HiveProject(ss_sold_date_sk=[$0], ss_quantity=[$1], ss_list_price=[$2])
>                 HiveFilter(condition=[IS NOT NULL($0)])
>                   HiveTableScan(table=[[sub, store_sales]], table:alias=[store_sales])
>               HiveProject(d_date_sk=[$0])
>                 HiveFilter(condition=[AND(BETWEEN(false, $1, 1999, 2001), IS NOT NULL($0))])
>                   HiveTableScan(table=[[sub, date_dim]], table:alias=[date_dim])
>     HiveProject(cnt=[$0])
>       HiveFilter(condition=[<=(sq_count_check($0), 1)])
>         HiveProject(cnt=[$0])
>           HiveAggregate(group=[{}], cnt=[COUNT()])
>             HiveProject
>               HiveProject($f0=[$0])
>                 HiveAggregate(group=[{}], agg#0=[count($0)])
>                   HiveJoin(condition=[=($0, $3)], joinType=[inner], algorithm=[none], cost=[{2.0 rows, 0.0 cpu, 0.0 io}])
>                     HiveProject(ss_sold_date_sk=[$0], ss_quantity=[$1], ss_list_price=[$2])
>                       HiveFilter(condition=[IS NOT NULL($0)])
>                         HiveTableScan(table=[[sub, store_sales]], table:alias=[store_sales])
>                     HiveProject(d_date_sk=[$0])
>                       HiveFilter(condition=[AND(BETWEEN(false, $1, 1999, 2001), IS NOT NULL($0))])
>                         HiveTableScan(table=[[sub, date_dim]], table:alias=[date_dim])
> {noformat}



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