Optimizing GROUP BY and DISTINCT Clauses in Subqueries

A DISTINCT clause and a GROUP BY without a corresponding HAVING clause have no meaning in IN/ALL/ANY/SOME/EXISTS subqueries. The reason is that IN/ALL/ANY/SOME/EXISTS only check if an outer row satisfies some condition with respect to all or any row in the subquery result. Therefore is doesn't matter if the subquery has duplicate result rows or not - if some condition is true for some row of the subquery, this condition will be true for all duplicates of this row. Notice that GROUP BY without a corresponding HAVING clause is equivalent to a DISTINCT.

MariaDB 5.3 and later versions automatically remove DISTINCT and GROUP BY without HAVING if these clauses appear in an IN/ALL/ANY/SOME/EXISTS subquery. For instance:

select * from t1
where t1.a > ALL(select distinct b from t2 where t2.c > 100)

is transformed to:

select * from t1
where t1.a > ALL(select b from t2 where t2.c > 100)

Removing these unnecessary clauses allows the optimizer to find more efficient query plans because it doesn't need to take care of post-processing the subquery result to satisfy DISTINCT / GROUP BY.

Content reproduced on this site is the property of its respective owners, and this content is not reviewed in advance by MariaDB. The views, information and opinions expressed by this content do not necessarily represent those of MariaDB or any other party.

© 2019 MariaDB
Licensed under the Creative Commons Attribution 3.0 Unported License and the GNU Free Documentation License.