diff --git a/src/backend/optimizer/util/pathnode.c b/src/backend/optimizer/util/pathnode.c index 78aee5df08e6247e88c67f7a70c4a935bad20b8b..140d0004b51a6b26ebfab47fb39cd532df220ae3 100644 --- a/src/backend/optimizer/util/pathnode.c +++ b/src/backend/optimizer/util/pathnode.c @@ -1508,6 +1508,17 @@ query_is_distinct_for(Query *query, List *colnos, List *opids) Assert(list_length(colnos) == list_length(opids)); + /* + * A set-returning function in the query's targetlist can result in + * returning duplicate rows, if the SRF is evaluated after the + * de-duplication step; so we play it safe and say "no" if there are any + * SRFs. (We could be certain that it's okay if SRFs appear only in the + * specified columns, since those must be evaluated before de-duplication; + * but it doesn't presently seem worth the complication to check that.) + */ + if (expression_returns_set((Node *) query->targetList)) + return false; + /* * DISTINCT (including DISTINCT ON) guarantees uniqueness if all the * columns in the DISTINCT clause appear in colnos and operator semantics diff --git a/src/test/regress/expected/subselect.out b/src/test/regress/expected/subselect.out index cde168e3c801f3af8c7828fec4520ef9bcd2640c..543709ac3fcac6a6f0d74c655d50df42e4142924 100644 --- a/src/test/regress/expected/subselect.out +++ b/src/test/regress/expected/subselect.out @@ -742,3 +742,36 @@ select * from int4_tbl where 0 (1 row) +-- +-- Check for incorrect optimization when IN subquery contains a SRF +-- +explain (verbose, costs off) +select * from int4_tbl o where (f1, f1) in + (select f1, generate_series(1,2) / 10 g from int4_tbl i group by f1); + QUERY PLAN +---------------------------------------------------------------------- + Hash Join + Output: o.f1 + Hash Cond: (o.f1 = "ANY_subquery".f1) + -> Seq Scan on public.int4_tbl o + Output: o.f1 + -> Hash + Output: "ANY_subquery".f1, "ANY_subquery".g + -> HashAggregate + Output: "ANY_subquery".f1, "ANY_subquery".g + -> Subquery Scan on "ANY_subquery" + Output: "ANY_subquery".f1, "ANY_subquery".g + Filter: ("ANY_subquery".f1 = "ANY_subquery".g) + -> HashAggregate + Output: i.f1, (generate_series(1, 2) / 10) + -> Seq Scan on public.int4_tbl i + Output: i.f1 +(16 rows) + +select * from int4_tbl o where (f1, f1) in + (select f1, generate_series(1,2) / 10 g from int4_tbl i group by f1); + f1 +---- + 0 +(1 row) + diff --git a/src/test/regress/sql/subselect.sql b/src/test/regress/sql/subselect.sql index 326fd70e4a06bea8db43f24b878e1a79377c02b4..1975902726f99e02c2c4d579cd9869e529c13ff2 100644 --- a/src/test/regress/sql/subselect.sql +++ b/src/test/regress/sql/subselect.sql @@ -422,3 +422,12 @@ select * from int4_tbl where select * from int4_tbl where (case when f1 in (select unique1 from tenk1 a) then f1 else null end) in (select ten from tenk1 b); + +-- +-- Check for incorrect optimization when IN subquery contains a SRF +-- +explain (verbose, costs off) +select * from int4_tbl o where (f1, f1) in + (select f1, generate_series(1,2) / 10 g from int4_tbl i group by f1); +select * from int4_tbl o where (f1, f1) in + (select f1, generate_series(1,2) / 10 g from int4_tbl i group by f1);