[
https://issues.apache.org/jira/browse/CALCITE-4133?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jiatao Tao updated CALCITE-4133:
--------------------------------
Description:
Consider case like this:
{code:java}
String sql =
"select \n"
+ " gender, \n"
+ " count(*) \n"
+ "from \n"
+ "(\n"
+ " (select gender, empno from EMPS) \n"
+ " union \n"
+ " (select gender, empno from EMPS)\n"
+ ") group by gender";
{code}
And the plan is:
{code:java}
EnumerableAggregate(group=[{0}], EXPR$1=[COUNT()])
EnumerableUnion(all=[false])
EnumerableProject(GENDER=[$3])
EnumerableTableScan(table=[[SALES, EMPS]])
EnumerableProject(GENDER=[$3])
EnumerableTableScan(table=[[SALES, EMPS]])
{code}
The col empno has been trimmed, this changes the semantics cuz union will
distinct values.
> Shouldn't trim fields when it's under Union
> -------------------------------------------
>
> Key: CALCITE-4133
> URL: https://issues.apache.org/jira/browse/CALCITE-4133
> Project: Calcite
> Issue Type: Bug
> Reporter: Jiatao Tao
> Assignee: Jiatao Tao
> Priority: Major
>
> Consider case like this:
>
> {code:java}
> String sql =
> "select \n"
> + " gender, \n"
> + " count(*) \n"
> + "from \n"
> + "(\n"
> + " (select gender, empno from EMPS) \n"
> + " union \n"
> + " (select gender, empno from EMPS)\n"
> + ") group by gender";
> {code}
> And the plan is:
>
> {code:java}
> EnumerableAggregate(group=[{0}], EXPR$1=[COUNT()])
> EnumerableUnion(all=[false])
> EnumerableProject(GENDER=[$3])
> EnumerableTableScan(table=[[SALES, EMPS]])
> EnumerableProject(GENDER=[$3])
> EnumerableTableScan(table=[[SALES, EMPS]])
> {code}
>
> The col empno has been trimmed, this changes the semantics cuz union will
> distinct values.
>
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
