You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@calcite.apache.org by "Julian Hyde (Jira)" <ji...@apache.org> on 2020/06/26 17:14:00 UTC
[jira] [Updated] (CALCITE-4090) When generating SQL for DB2, a
complex SELECT above a sub-query generates a bad table alias
[ https://issues.apache.org/jira/browse/CALCITE-4090?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Julian Hyde updated CALCITE-4090:
---------------------------------
Affects Version/s: 1.23.0
Summary: When generating SQL for DB2, a complex SELECT above a sub-query generates a bad table alias (was: DB2 aliasing breaks with a complex select above a subselect)
> When generating SQL for DB2, a complex SELECT above a sub-query generates a bad table alias
> -------------------------------------------------------------------------------------------
>
> Key: CALCITE-4090
> URL: https://issues.apache.org/jira/browse/CALCITE-4090
> Project: Calcite
> Issue Type: Bug
> Affects Versions: 1.23.0
> Reporter: Steven Talbot
> Assignee: Julian Hyde
> Priority: Major
> Fix For: 1.24.0
>
>
> Test in RelToSqlConverterTest:
> {code:java}
> @Test void testDb2DialectSubselect() {
> String query = "select count(foo), \"units_per_case\" "
> + "from (select \"units_per_case\", \"cases_per_pallet\", \"product_id\", 1 as foo from \"product\") where \"cases_per_pallet\" > 100 "
> + "group by \"product_id\", \"units_per_case\" "
> + "order by \"units_per_case\" desc";
> final String expected = "SELECT COUNT(*), t.units_per_case\n" +
> "FROM (SELECT product.units_per_case, product.cases_per_pallet, product.product_id, 1 AS " +
> "FOO\n" +
> "FROM foodmart.product AS product) AS t\n" +
> "WHERE t.cases_per_pallet > 100\n" +
> "GROUP BY t.product_id, t.units_per_case\n" +
> "ORDER BY t.units_per_case DESC";
> sql(query).withDb2().ok(expected);
> }
> {code}
> The test fails with the "t." alias qualifier in the group by/order by/main select actually being "t0.".
> From stepping through the code in the debugger, I believe this is a general problem with the way aliases are calculated in situations like this by SqlImplementor, but other dialects with hasImplicitTableAlias() do not force qualified contexts and therefore do not hit this.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)