[
https://issues.apache.org/jira/browse/HIVE-11880?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
WangMeng updated HIVE-11880:
----------------------------
Description:
For UNION ALL , when an union operator is constant(column 'a' such as '0L')
and it has incompatible type with the corresponding column A(INT type).
Query with filter condition on type incompatible column on this UNION ALL
will cause IndexOutOfBoundsException.
Such as TPC-H table "orders",in the fllowing query:
Type of 'o_custkey' is INT normally, while the type of corresponding column
constant "0" is BIGINT( `0L AS `o_custkey` ).
This query (with filter " incompatible column 'o_custkey' ") will fail with
java.lang.IndexOutOfBoundsException :
{code}
SELECT Count(1)
FROM (
SELECT `o_orderkey` ,
`o_custkey`
FROM `orders`
UNION ALL
SELECT `o_orderkey`,
0L AS `o_custkey`
FROM `rcfileorders`) `oo`
WHERE o_custkey<100000 limit 4 ;
{code}
When
{code}
set hive.ppd.remove.duplicatefilters=true
{code}
the fllowing query (with filter " incompatible column 'o_custkey' ") will
fail with java.lang.IndexOutOfBoundsException:
{code}
select count(1) from view_orders where o_custkey<100000
{code}
was:
For Hive UNION ALL , when an union column is constant(column a such as '0L')
and it has incompatible type with the corresponding column A(INT Type).
Query with filter condition on type incompatible column a on this UNION ALL
will cause IndexOutOfBoundsException.
Such as TPC-H table "orders", we create View by :
{code}
CREATE VIEW `view_orders` AS
SELECT `oo`.`o_orderkey` ,
`oo`.`o_custkey`
FROM (
SELECT `o_orderkey` , `0L AS `o_custkey`
FROM `rcfileorders`
UNION ALL
SELECT `o_orderkey` ,`o_custkey`
FROM `textfileorders`) `oo`.
{code}
In VIEW view_orders , type of 'o_custkey' is INT normally, while the type of
corresponding column constant "0" is BIGINT( `0L AS `o_custkey` ).
When
{code}
set hive.ppd.remove.duplicatefilters=true
{code}
the fllowing query (with filter " incompatible column 'o_custkey' ") will
fail with java.lang.IndexOutOfBoundsException:
{code}
select count(1) from view_orders where o_custkey<100000
{code}
> filter bug of UNION ALL when hive.ppd.remove.duplicatefilters=true and
> filter condition is type incompatible column
> ---------------------------------------------------------------------------------------------------------------------
>
> Key: HIVE-11880
> URL: https://issues.apache.org/jira/browse/HIVE-11880
> Project: Hive
> Issue Type: Bug
> Components: Logical Optimizer
> Affects Versions: 1.2.1
> Reporter: WangMeng
> Assignee: WangMeng
> Attachments: HIVE-11880.01.patch
>
>
> For UNION ALL , when an union operator is constant(column 'a' such as
> '0L') and it has incompatible type with the corresponding column A(INT
> type).
> Query with filter condition on type incompatible column on this UNION ALL
> will cause IndexOutOfBoundsException.
> Such as TPC-H table "orders",in the fllowing query:
> Type of 'o_custkey' is INT normally, while the type of corresponding
> column constant "0" is BIGINT( `0L AS `o_custkey` ).
> This query (with filter " incompatible column 'o_custkey' ") will fail
> with java.lang.IndexOutOfBoundsException :
> {code}
> SELECT Count(1)
> FROM (
> SELECT `o_orderkey` ,
> `o_custkey`
> FROM `orders`
> UNION ALL
> SELECT `o_orderkey`,
> 0L AS `o_custkey`
> FROM `rcfileorders`) `oo`
> WHERE o_custkey<100000 limit 4 ;
> {code}
> When
> {code}
> set hive.ppd.remove.duplicatefilters=true
> {code}
> the fllowing query (with filter " incompatible column 'o_custkey' ") will
> fail with java.lang.IndexOutOfBoundsException:
> {code}
> select count(1) from view_orders where o_custkey<100000
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
