MDEV-22534 Decorrelate IN subquery

Transform

in (select inner_col' from inner_table where inner_col = outer_col)

to

, outer_col in (select inner_col', inner_col from inner_table)

Achieved by implementing Item_in_subselect::exists2in_processor(),
accompanied with comprehensive test coverage. Factored out common code
between the two transformations.

Caveat:

- Cannot recognise bad item mismatch in equalities that causes
  materialization to not materialize down the road

mysql-test/main/delete_use_source_engines.result

@@ -133,11 +133,13 @@ create table tmp as select * from t1 where c3 in (select distinct a.c1 from t1 a
 explain select * from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain delete from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 delete from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 affected rows: 8
 select * from t1;
@@ -175,15 +177,18 @@ create table tmp as select * from t1 where c1 in (select a.c2 from t1 a where a.
 explain select * from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using where
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using where
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 analyze delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	r_rows	filtered	r_filtered	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	32.00	100.00	12.50	Using where
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	24.50	3.12	1.02	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	0.25	100.00	100.00	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	32.00	100.00	100.00	
 select * from t1;
 c1	c2	c3
 1	2	2
@@ -227,8 +232,9 @@ order by c3 desc limit 1;
 explain select * from t1 where c1 in (select a.c2 from t1 a where a.c3 = t1.c3)
 order by c3 desc limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using temporary; Using filesort
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using filesort
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain delete from t1 where c1 in (select a.c2 from t1 a where a.c3 = t1.c3)
 order by c3 desc limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
@@ -731,15 +737,18 @@ create table tmp as select * from t1 where c1 in (select a.c2 from t1 a where a.
 explain select * from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using where
-1	PRIMARY	a	ref	t1_c2	t1_c2	5	test.t1.c1	5	Using index; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	index	t1_c2	t1_c2	10	NULL	32	Using index
 explain delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using where
-1	PRIMARY	a	ref	t1_c2	t1_c2	5	test.t1.c1	5	Using index; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	index	t1_c2	t1_c2	10	NULL	32	Using index
 analyze delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	r_rows	filtered	r_filtered	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	32.00	100.00	12.50	Using where
-1	PRIMARY	a	ref	t1_c2	t1_c2	5	test.t1.c1	5	0.25	18.75	100.00	Using index; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	0.25	100.00	100.00	
+2	MATERIALIZED	a	index	t1_c2	t1_c2	10	NULL	32	32.00	100.00	100.00	Using index
 select * from t1;
 c1	c2	c3
 1	2	2
@@ -783,8 +792,9 @@ order by c3 desc limit 1;
 explain select * from t1 where c1 in (select a.c2 from t1 a where a.c3 = t1.c3)
 order by c3 desc limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using temporary; Using filesort
-1	PRIMARY	a	ALL	t1_c2	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using filesort
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	t1_c2	NULL	NULL	NULL	32	
 explain delete from t1 where c1 in (select a.c2 from t1 a where a.c3 = t1.c3)
 order by c3 desc limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
@@ -1245,12 +1255,14 @@ drop table tmp;
 create table tmp as select * from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 explain select * from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; Start temporary
-1	PRIMARY	t1	eq_ref	PRIMARY	PRIMARY	4	test.a.c1	1	Using where; End temporary
+1	PRIMARY	t1	ALL	PRIMARY	NULL	NULL	NULL	32	
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain delete from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; Start temporary
-1	PRIMARY	t1	eq_ref	PRIMARY	PRIMARY	4	test.a.c1	1	Using where; End temporary
+1	PRIMARY	t1	ALL	PRIMARY	NULL	NULL	NULL	32	
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 delete from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 affected rows: 8
 select * from t1;
@@ -1287,16 +1299,19 @@ drop table tmp;
 create table tmp as select * from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 explain select * from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; Start temporary
-1	PRIMARY	t1	eq_ref	PRIMARY	PRIMARY	4	test.a.c2	1	Using where; End temporary
+1	PRIMARY	t1	ALL	PRIMARY	NULL	NULL	NULL	32	Using where
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; Start temporary
-1	PRIMARY	t1	eq_ref	PRIMARY	PRIMARY	4	test.a.c2	1	Using where; End temporary
+1	PRIMARY	t1	ALL	PRIMARY	NULL	NULL	NULL	32	Using where
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 analyze delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	r_rows	filtered	r_filtered	Extra
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	1.00	100.00	100.00	Using where; Start temporary
-1	PRIMARY	t1	eq_ref	PRIMARY	PRIMARY	4	test.a.c2	1	1.00	3.12	100.00	Using where; End temporary
+1	PRIMARY	t1	ALL	PRIMARY	NULL	NULL	NULL	32	1.00	100.00	100.00	Using where
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	1.00	100.00	100.00	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	32.00	100.00	100.00	
 select * from t1;
 c1	c2	c3
 1	2	2
@@ -1896,11 +1911,13 @@ create table tmp as select * from t1 where c3 in (select distinct a.c1 from t1 a
 explain select * from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain delete from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 delete from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 affected rows: 8
 select * from t1;
@@ -1938,15 +1955,18 @@ create table tmp as select * from t1 where c1 in (select a.c2 from t1 a where a.
 explain select * from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using where
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using where
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 analyze delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	r_rows	filtered	r_filtered	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	25.00	100.00	12.00	Using where
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	22.67	3.12	1.47	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	0.33	100.00	100.00	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	32.00	100.00	100.00	
 select * from t1;
 c1	c2	c3
 1	2	2
@@ -1990,8 +2010,9 @@ order by c3 desc limit 1;
 explain select * from t1 where c1 in (select a.c2 from t1 a where a.c3 = t1.c3)
 order by c3 desc limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using temporary; Using filesort
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using filesort
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain delete from t1 where c1 in (select a.c2 from t1 a where a.c3 = t1.c3)
 order by c3 desc limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
@@ -2716,8 +2737,8 @@ id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	a	ref	t1_c2	t1_c2	5	test.t1.c1	5	Using index; FirstMatch(t1)
 analyze delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	r_rows	filtered	r_filtered	Extra
-1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	27.00	100.00	14.81	Using where
-1	PRIMARY	a	ref	t1_c2	t1_c2	5	test.t1.c1	5	0.25	18.75	100.00	Using index; FirstMatch(t1)
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	17.00	100.00	11.76	Using where
+1	PRIMARY	a	ref	t1_c2	t1_c2	5	test.t1.c1	5	0.50	18.75	100.00	Using index; FirstMatch(t1)
 select * from t1;
 c1	c2	c3
 1	2	2
@@ -2761,8 +2782,9 @@ order by c3 desc limit 1;
 explain select * from t1 where c1 in (select a.c2 from t1 a where a.c3 = t1.c3)
 order by c3 desc limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using where; Using filesort
-1	PRIMARY	a	ref	t1_c2	t1_c2	5	test.t1.c1	5	Using where; FirstMatch(t1)
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using filesort
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	t1_c2	NULL	NULL	NULL	32	
 explain delete from t1 where c1 in (select a.c2 from t1 a where a.c3 = t1.c3)
 order by c3 desc limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
@@ -3394,7 +3416,7 @@ id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 analyze delete from t1 where c1 = 1 and exists (select 'X' from t1 a where a.c1 = t1.c2);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	r_rows	filtered	r_filtered	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	32.00	9.38	9.38	Using where
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	12.00	3.12	5.56	Using where; FirstMatch(t1)
+1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	12.67	3.12	5.26	Using where; FirstMatch(t1)
 select * from t1;
 c1	c2	c3
 1	3	3
@@ -3435,12 +3457,14 @@ drop table tmp;
 create table tmp as select * from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 explain select * from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; Start temporary
-1	PRIMARY	t1	eq_ref	PRIMARY	PRIMARY	4	test.a.c1	1	Using where; End temporary
+1	PRIMARY	t1	ALL	PRIMARY	NULL	NULL	NULL	32	
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain delete from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; Start temporary
-1	PRIMARY	t1	eq_ref	PRIMARY	PRIMARY	4	test.a.c1	1	Using where; End temporary
+1	PRIMARY	t1	ALL	PRIMARY	NULL	NULL	NULL	32	
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 delete from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 affected rows: 8
 select * from t1;
@@ -3477,16 +3501,19 @@ drop table tmp;
 create table tmp as select * from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 explain select * from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; Start temporary
-1	PRIMARY	t1	eq_ref	PRIMARY	PRIMARY	4	test.a.c2	1	Using where; End temporary
+1	PRIMARY	t1	ALL	PRIMARY	NULL	NULL	NULL	32	Using where
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; Start temporary
-1	PRIMARY	t1	eq_ref	PRIMARY	PRIMARY	4	test.a.c2	1	Using where; End temporary
+1	PRIMARY	t1	ALL	PRIMARY	NULL	NULL	NULL	32	Using where
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 analyze delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	r_rows	filtered	r_filtered	Extra
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	4.00	100.00	100.00	Using where; Start temporary
-1	PRIMARY	t1	eq_ref	PRIMARY	PRIMARY	4	test.a.c2	1	1.00	3.12	25.00	Using where; End temporary
+1	PRIMARY	t1	ALL	PRIMARY	NULL	NULL	NULL	32	16.00	100.00	18.75	Using where
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	0.33	100.00	100.00	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	32.00	100.00	100.00	
 select * from t1;
 c1	c2	c3
 1	2	2
@@ -4301,11 +4328,13 @@ create table tmp as select * from t1 where c3 in (select distinct a.c1 from t1 a
 explain select * from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain delete from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 delete from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 affected rows: 8
 select * from t1;
@@ -4343,15 +4372,18 @@ create table tmp as select * from t1 where c1 in (select a.c2 from t1 a where a.
 explain select * from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using where
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using where
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 analyze delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	r_rows	filtered	r_filtered	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	25.00	100.00	12.00	Using where
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	22.67	3.12	1.47	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	0.33	100.00	100.00	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	32.00	100.00	100.00	
 select * from t1;
 c1	c2	c3
 1	2	2
@@ -4395,8 +4427,9 @@ order by c3 desc limit 1;
 explain select * from t1 where c1 in (select a.c2 from t1 a where a.c3 = t1.c3)
 order by c3 desc limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using temporary; Using filesort
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using filesort
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain delete from t1 where c1 in (select a.c2 from t1 a where a.c3 = t1.c3)
 order by c3 desc limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
@@ -5121,8 +5154,8 @@ id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	a	ref	t1_c2	t1_c2	5	test.t1.c1	5	Using index; FirstMatch(t1)
 analyze delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	r_rows	filtered	r_filtered	Extra
-1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	29.00	100.00	13.79	Using where
-1	PRIMARY	a	ref	t1_c2	t1_c2	5	test.t1.c1	5	0.25	18.75	100.00	Using index; FirstMatch(t1)
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	17.00	100.00	11.76	Using where
+1	PRIMARY	a	ref	t1_c2	t1_c2	5	test.t1.c1	5	0.50	18.75	100.00	Using index; FirstMatch(t1)
 select * from t1;
 c1	c2	c3
 1	2	2
@@ -5166,8 +5199,9 @@ order by c3 desc limit 1;
 explain select * from t1 where c1 in (select a.c2 from t1 a where a.c3 = t1.c3)
 order by c3 desc limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using temporary; Using filesort
-1	PRIMARY	a	ALL	t1_c2	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using filesort
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	t1_c2	NULL	NULL	NULL	32	
 explain delete from t1 where c1 in (select a.c2 from t1 a where a.c3 = t1.c3)
 order by c3 desc limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
@@ -5799,7 +5833,7 @@ id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 analyze delete from t1 where c1 = 1 and exists (select 'X' from t1 a where a.c1 = t1.c2);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	r_rows	filtered	r_filtered	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	32.00	9.38	9.38	Using where
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	13.33	3.12	5.00	Using where; FirstMatch(t1)
+1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	13.67	3.12	4.88	Using where; FirstMatch(t1)
 select * from t1;
 c1	c2	c3
 1	3	3
@@ -5840,12 +5874,14 @@ drop table tmp;
 create table tmp as select * from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 explain select * from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; Start temporary
-1	PRIMARY	t1	eq_ref	PRIMARY	PRIMARY	4	test.a.c1	1	Using where; End temporary
+1	PRIMARY	t1	ALL	PRIMARY	NULL	NULL	NULL	32	
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain delete from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; Start temporary
-1	PRIMARY	t1	eq_ref	PRIMARY	PRIMARY	4	test.a.c1	1	Using where; End temporary
+1	PRIMARY	t1	ALL	PRIMARY	NULL	NULL	NULL	32	
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 delete from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 affected rows: 8
 select * from t1;
@@ -5882,16 +5918,19 @@ drop table tmp;
 create table tmp as select * from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 explain select * from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; Start temporary
-1	PRIMARY	t1	eq_ref	PRIMARY	PRIMARY	4	test.a.c2	1	Using where; End temporary
+1	PRIMARY	t1	ALL	PRIMARY	NULL	NULL	NULL	32	Using where
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; Start temporary
-1	PRIMARY	t1	eq_ref	PRIMARY	PRIMARY	4	test.a.c2	1	Using where; End temporary
+1	PRIMARY	t1	ALL	PRIMARY	NULL	NULL	NULL	32	Using where
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 analyze delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	r_rows	filtered	r_filtered	Extra
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	4.00	100.00	100.00	Using where; Start temporary
-1	PRIMARY	t1	eq_ref	PRIMARY	PRIMARY	4	test.a.c2	1	1.00	3.12	25.00	Using where; End temporary
+1	PRIMARY	t1	ALL	PRIMARY	NULL	NULL	NULL	32	26.00	100.00	11.54	Using where
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	0.33	100.00	100.00	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	32.00	100.00	100.00	
 select * from t1;
 c1	c2	c3
 1	2	2
@@ -6459,7 +6498,7 @@ and c1 = 2
 and exists (select 'X' from v1 a where a.c1 = v1.c1);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	r_rows	filtered	r_filtered	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	#	32.00	3.91	3.12	Using where
-3	DEPENDENT SUBQUERY	t1	ALL	NULL	NULL	NULL	NULL	#	3.00	25.00	33.33	Using where
+3	DEPENDENT SUBQUERY	t1	ALL	NULL	NULL	NULL	NULL	#	2.00	25.00	50.00	Using where
 2	DEPENDENT SUBQUERY	a	eq_ref	PRIMARY	PRIMARY	4	test.t1.c3	#	1.00	100.00	100.00	
 select * from t1;
 c1	c2	c3
@@ -6791,11 +6830,13 @@ create table tmp as select * from t1 where c3 in (select distinct a.c1 from t1 a
 explain select * from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain delete from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 delete from t1 where c3 in (select distinct a.c1 from t1 a where t1.c2=a.c2);
 affected rows: 8
 select * from t1;
@@ -6833,11 +6874,13 @@ create table tmp as select * from t1 where c1 in (select a.c2 from t1 a where a.
 explain select * from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using where
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using where
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 affected rows: 1
 select * from t1;
@@ -6883,8 +6926,9 @@ order by c3 desc limit 1;
 explain select * from t1 where c1 in (select a.c2 from t1 a where a.c3 = t1.c3)
 order by c3 desc limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using temporary; Using filesort
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using filesort
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain delete from t1 where c1 in (select a.c2 from t1 a where a.c3 = t1.c3)
 order by c3 desc limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
@@ -7588,11 +7632,13 @@ create table tmp as select * from t1 where c1 in (select a.c2 from t1 a where a.
 explain select * from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using where
-1	PRIMARY	a	ALL	t1_c2	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	t1_c2	NULL	NULL	NULL	32	
 explain delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using where
-1	PRIMARY	a	ALL	t1_c2	NULL	NULL	NULL	32	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	t1_c2	NULL	NULL	NULL	32	
 delete from t1 where c1 in (select a.c2 from t1 a where a.c2 = t1.c3) limit 1;
 affected rows: 1
 select * from t1;
@@ -7638,8 +7684,9 @@ order by c3 desc limit 1;
 explain select * from t1 where c1 in (select a.c2 from t1 a where a.c3 = t1.c3)
 order by c3 desc limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using temporary; Using filesort
-1	PRIMARY	a	ALL	t1_c2	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	Using filesort
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	t1_c2	NULL	NULL	NULL	32	
 explain delete from t1 where c1 in (select a.c2 from t1 a where a.c3 = t1.c3)
 order by c3 desc limit 1;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra

mysql-test/main/explain_json.result

@@ -906,7 +906,7 @@ EXPLAIN
           "buffer_type": "flat",
           "buffer_size": "141",
           "join_type": "BNL",
-          "attached_condition": "t1.b = t2.b and t1.a = t2.a"
+          "attached_condition": "t1.a = t2.a and t1.b = t2.b"
         }
       }
     ]
@@ -921,7 +921,8 @@ explain
 select * from t2 where t2.a in ( select a from t1 where t1.b=t2.b);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t2	ALL	NULL	NULL	NULL	NULL	10	
-1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	10	Using where; Start temporary; End temporary; Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	t1	ALL	NULL	NULL	NULL	NULL	10	
 explain format=json
 select * from t2 where t2.a in ( select a from t1 where t1.b=t2.b);
 EXPLAIN
@@ -929,6 +930,7 @@ EXPLAIN
   "query_block": {
     "select_id": 1,
     "cost": "COST_REPLACED",
+    "const_condition": "1",
     "nested_loop": [
       {
         "table": {
@@ -941,25 +943,36 @@ EXPLAIN
         }
       },
       {
-        "duplicates_removal": [
+        "table": {
+          "table_name": "<subquery2>",
+          "access_type": "eq_ref",
+          "possible_keys": ["distinct_key"],
+          "key": "distinct_key",
+          "key_length": "8",
+          "used_key_parts": ["a", "b"],
+          "ref": ["func", "func"],
+          "rows": 1,
+          "filtered": 100,
+          "materialized": {
+            "unique": 1,
+            "query_block": {
+              "select_id": 2,
+              "nested_loop": [
                 {
-            "block-nl-join": {
                   "table": {
                     "table_name": "t1",
                     "access_type": "ALL",
-                "loops": 10,
+                    "loops": 1,
                     "rows": 10,
                     "cost": "COST_REPLACED",
-                "filtered": 10
-              },
-              "buffer_type": "flat",
-              "buffer_size": "206",
-              "join_type": "BNL",
-              "attached_condition": "t1.b = t2.b and t1.a = t2.a"
+                    "filtered": 100
                   }
                 }
               ]
             }
+          }
+        }
+      }
     ]
   }
 }

mysql-test/main/innodb_ext_key.result

@@ -384,15 +384,16 @@ EXPLAIN
 SELECT a FROM t1 AS t, t2 as t2_out 
 WHERE t2_out.c = t.a AND t.b IN (SELECT b FROM t1, t2 WHERE b = t.b);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
-1	PRIMARY	t1	index	b	b	7	NULL	10	Using index; Start temporary
-1	PRIMARY	t	ref	a,b	b	3	test.t1.b	2	Using index
-1	PRIMARY	t2	index	NULL	PRIMARY	4	NULL	11	Using index; End temporary; Using join buffer (flat, BNL join)
+1	PRIMARY	t	index	a,b	b	7	NULL	10	Using index
 1	PRIMARY	t2_out	eq_ref	PRIMARY	PRIMARY	4	test.t.a	1	Using index
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	t1	index	b	b	7	NULL	10	Using index
+2	MATERIALIZED	t2	index	NULL	PRIMARY	4	NULL	11	Using index; Using join buffer (flat, BNL join)
 SELECT a FROM t1 AS t, t2 as t2_out 
 WHERE t2_out.c = t.a AND t.b IN (SELECT b FROM t1, t2 WHERE b = t.b);
 a
 24
-Last_query_cost	0.120558
+Last_query_cost	0.083937
 DROP TABLE t1,t2;
 #
 # LP Bug #923236: hash join + extended_keys = on 

mysql-test/main/rowid_filter_innodb.result

@@ -3435,11 +3435,11 @@ WHERE t1.c1 NOT IN (SELECT t2.c1 FROM t2, t1 AS a1
 WHERE t2.i1 = t1.pk AND t2.i1 BETWEEN 3 AND 5);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	filtered	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	60	100.00	Using where
-2	DEPENDENT SUBQUERY	t2	ref|filter	c1,i1	c1|i1	3|5	func	38 (25%)	25.00	Using where; Full scan on NULL key; Using rowid filter
-2	DEPENDENT SUBQUERY	a1	ALL	NULL	NULL	NULL	NULL	60	100.00	Using join buffer (flat, BNL join)
+2	MATERIALIZED	t2	ALL	c1,i1	NULL	NULL	NULL	192	25.00	Using where
+2	MATERIALIZED	a1	ALL	NULL	NULL	NULL	NULL	60	100.00	Using join buffer (flat, BNL join)
 Warnings:
 Note	1276	Field or reference 'test.t1.pk' of SELECT #2 was resolved in SELECT #1
-Note	1003	/* select#1 */ select `test`.`t1`.`pk` AS `pk`,`test`.`t1`.`c1` AS `c1` from `test`.`t1` where !<expr_cache><`test`.`t1`.`c1`,`test`.`t1`.`pk`>(<in_optimizer>(`test`.`t1`.`c1`,<exists>(/* select#2 */ select `test`.`t2`.`c1` from `test`.`t2` join `test`.`t1` `a1` where `test`.`t2`.`i1` = `test`.`t1`.`pk` and `test`.`t2`.`i1` between 3 and 5 and trigcond(<cache>(`test`.`t1`.`c1`) = `test`.`t2`.`c1`))))
+Note	1003	/* select#1 */ select `test`.`t1`.`pk` AS `pk`,`test`.`t1`.`c1` AS `c1` from `test`.`t1` where !<expr_cache><`test`.`t1`.`c1`,`test`.`t1`.`pk`>(<in_optimizer>((`test`.`t1`.`c1`,`test`.`t1`.`pk`),(`test`.`t1`.`c1`,`test`.`t1`.`pk`) in ( <materialize> (/* select#2 */ select `test`.`t2`.`c1`,`test`.`t2`.`i1` from `test`.`t2` join `test`.`t1` `a1` where `test`.`t2`.`i1` is not null and `test`.`t2`.`i1` between 3 and 5 ), <primary_index_lookup>(`test`.`t1`.`c1` in <temporary table> on distinct_key where `test`.`t1`.`c1` = `<subquery2>`.`c1` and `test`.`t1`.`pk` = `<subquery2>`.`i1`))))
 SELECT * FROM t1
 WHERE t1.c1 NOT IN (SELECT t2.c1 FROM t2, t1 AS a1
 WHERE t2.i1 = t1.pk AND t2.i1 BETWEEN 3 AND 5);

mysql-test/main/subselect4.result

@@ -225,7 +225,7 @@ EXPLAIN
 SELECT * FROM t1 WHERE NULL NOT IN (SELECT t2a.i FROM t2a WHERE t2a.pk = t1.pk);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	4	Using where
-2	DEPENDENT SUBQUERY	t2a	unique_subquery	PRIMARY	PRIMARY	8	const,test.t1.pk	1	Using index; Using where; Full scan on NULL key
+2	DEPENDENT SUBQUERY	t2a	unique_subquery	PRIMARY	PRIMARY	8	const,func	1	Using index; Using where; Full scan on NULL key
 SELECT * FROM t1 WHERE NULL NOT IN (SELECT t2a.i FROM t2a WHERE t2a.pk = t1.pk);
 pk	i
 SELECT * FROM t1 WHERE 1+NULL NOT IN (SELECT t2a.i FROM t2a WHERE t2a.pk = t1.pk);
@@ -265,7 +265,7 @@ EXPLAIN
 SELECT * FROM t1 WHERE NULL NOT IN (SELECT t2c.i FROM t2c WHERE t2c.pk = t1.pk);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	4	Using where
-2	DEPENDENT SUBQUERY	t2c	index_subquery	it2c	it2c	8	const,test.t1.pk	1	Using index; Using where; Full scan on NULL key
+2	DEPENDENT SUBQUERY	t2c	index_subquery	it2c	it2c	8	const,func	1	Using index; Using where; Full scan on NULL key
 SELECT * FROM t1 WHERE NULL NOT IN (SELECT t2c.i FROM t2c WHERE t2c.pk = t1.pk);
 pk	i
 SELECT * FROM t1 WHERE NULL     IN (SELECT t2c.i FROM t2c WHERE t2c.pk = t1.pk) IS UNKNOWN;
@@ -303,7 +303,7 @@ EXPLAIN
 SELECT * FROM t1 WHERE (NULL, 1) NOT IN (SELECT t2a.i, t2a.pk FROM t2a WHERE t2a.pk = t1.pk);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	4	Using where
-2	DEPENDENT SUBQUERY	t2a	unique_subquery	PRIMARY	PRIMARY	8	const,test.t1.pk	1	Using index; Using where; Full scan on NULL key
+2	DEPENDENT SUBQUERY	t2a	const	PRIMARY	PRIMARY	8	const,const	1	Using where; Using index; Full scan on NULL key
 SELECT * FROM t1 WHERE (NULL, 1) NOT IN (SELECT t2a.i, t2a.pk FROM t2a WHERE t2a.pk = t1.pk);
 pk	i
 0	10
@@ -335,7 +335,7 @@ EXPLAIN
 SELECT * FROM t1 WHERE (NULL, 1) NOT IN (SELECT t2c.i, t2c.pk FROM t2c WHERE t2c.pk = t1.pk);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	4	Using where
-2	DEPENDENT SUBQUERY	t2c	index_subquery	it2c	it2c	8	const,test.t1.pk	1	Using index; Using where; Full scan on NULL key
+2	DEPENDENT SUBQUERY	t2c	index_subquery	it2c	it2c	8	const,const	0	Using index; Using where; Full scan on NULL key
 SELECT * FROM t1 WHERE (NULL, 1) NOT IN (SELECT t2c.i, t2c.pk FROM t2c WHERE t2c.pk = t1.pk);
 pk	i
 0	10

mysql-test/main/subselect_decorrelate_in.result

+#
+# MDEV-22534 Trivial correlation detection/removal for IN subqueries
+#
+# Basic example
+set optimizer_switch='materialization=on,firstmatch=off';
+set optimizer_trace=1;
+create table ten(a int primary key);
+insert into ten values (0),(1),(2),(3),(4),(5),(6),(7),(8),(9);
+create table t1 (a int, b int, c int);
+insert into t1 select a,a,a from ten;
+create table t2 select * from t1;
+explain select * from t1 where a in (select a from t2 where t1.b=t2.b);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	10	
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	t2	ALL	NULL	NULL	NULL	NULL	10	
+select * from t1 where a in (select a from t2 where t1.b=t2.b);
+a	b	c
+0	0	0
+1	1	1
+2	2	2
+3	3	3
+4	4	4
+5	5	5
+6	6	6
+7	7	7
+8	8	8
+9	9	9
+explain select * from t1 where (a, b) in (select a, b from t2);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	10	
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	t2	ALL	NULL	NULL	NULL	NULL	10	
+select * from t1 where (a, b) in (select a, b from t2);
+a	b	c
+0	0	0
+1	1	1
+2	2	2
+3	3	3
+4	4	4
+5	5	5
+6	6	6
+7	7	7
+8	8	8
+9	9	9
+# Multiple equalities
+explain select * from t1 where a in (select a from t2 where t1.b=t2.b and t1.c=t2.c);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	10	
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	12	func,func,func	1	
+2	MATERIALIZED	t2	ALL	NULL	NULL	NULL	NULL	10	
+select * from t1 where a in (select a from t2 where t1.b=t2.b and t1.c=t2.c);
+a	b	c
+0	0	0
+1	1	1
+2	2	2
+3	3	3
+4	4	4
+5	5	5
+6	6	6
+7	7	7
+8	8	8
+9	9	9
+explain select * from t1 where a in (select a from t2 where (t1.b, t1.c)=(t2.b, t2.c));
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	10	
+1	PRIMARY	t2	ALL	NULL	NULL	NULL	NULL	10	Using where; Start temporary; End temporary; Using join buffer (flat, BNL join)
+select * from t1 where a in (select a from t2 where (t1.b, t1.c)=(t2.b, t2.c));
+a	b	c
+0	0	0
+1	1	1
+2	2	2
+3	3	3
+4	4	4
+5	5	5
+6	6	6
+7	7	7
+8	8	8
+9	9	9
+explain select * from t1 where (a, b) in (select a, b from t2 where t1.c=t2.c);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	10	
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	12	func,func,func	1	
+2	MATERIALIZED	t2	ALL	NULL	NULL	NULL	NULL	10	
+select * from t1 where (a, b) in (select a, b from t2 where t1.c=t2.c);
+a	b	c
+0	0	0
+1	1	1
+2	2	2
+3	3	3
+4	4	4
+5	5	5
+6	6	6
+7	7	7
+8	8	8
+9	9	9
+explain select * from t1 where a in (select a from t2 where t1.b=t2.b and t2.c+t2.b < 5);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	10	
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	t2	ALL	NULL	NULL	NULL	NULL	10	Using where
+select * from t1 where a in (select a from t2 where t1.b=t2.b and t2.c+t2.b < 5);
+a	b	c
+0	0	0
+1	1	1
+2	2	2
+explain select * from t1 where a + b in (select a * c from t2 where t1.b=t2.b and t2.c+t2.b < 9);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	10	
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	12	func,func	1	Using where
+2	MATERIALIZED	t2	ALL	NULL	NULL	NULL	NULL	10	Using where
+select * from t1 where a + b in (select a * c from t2 where t1.b=t2.b and t2.c+t2.b < 9);
+a	b	c
+0	0	0
+2	2	2
+# Still correlated after the transformation / no materialization
+explain select * from t1 where a in (select a from t2 where t1.b=t2.b and t1.c < 5);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	10	Using where
+1	PRIMARY	t2	ALL	NULL	NULL	NULL	NULL	10	Using where; Start temporary; End temporary; Using join buffer (flat, BNL join)
+select * from t1 where a in (select a from t2 where t1.b=t2.b and t1.c < 5);
+a	b	c
+0	0	0
+1	1	1
+2	2	2
+3	3	3
+4	4	4
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+JS
+["decorrelation", "materialization", "semijoin"]
+explain select * from t1 where a in (select a from t2 where t1.a = t2.b and t2.c+t1.a < 9);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	10	
+1	PRIMARY	t2	ALL	NULL	NULL	NULL	NULL	10	Using where; Start temporary; End temporary; Using join buffer (flat, BNL join)
+select * from t1 where a in (select a from t2 where t1.a = t2.b and t2.c+t1.a < 9);
+a	b	c
+0	0	0
+1	1	1
+2	2	2
+3	3	3
+4	4	4
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS 
+from information_schema.OPTIMIZER_TRACE;
+JS
+["decorrelation", "materialization", "semijoin"]
+# No transformation because of applicable conditions in subquery
+explain select * from t1 where a in (select a from t2 where t2.c+t1.a < 9);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	10	
+1	PRIMARY	t2	ALL	NULL	NULL	NULL	NULL	10	Using where; Start temporary; End temporary; Using join buffer (flat, BNL join)
+select * from t1 where a in (select a from t2 where t2.c+t1.a < 9);
+a	b	c
+0	0	0
+1	1	1
+2	2	2
+3	3	3
+4	4	4
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS 
+from information_schema.OPTIMIZER_TRACE;
+JS
+["materialization", "semijoin"]
+drop table ten, t1, t2;
+# group by and having checks in the transformer; also not in
+create table t1 (a1 int not null, b1 char(8) not null);
+create table t2 (a2 int not null, b2 char(8) not null);
+insert into t1 values (0, '2 - 00');
+insert into t1 values (1, '2 - 01');
+insert into t1 values (2, '2 - 02');
+insert into t2 values (1, '2 - 01');
+insert into t2 values (1, '2 - 01');
+insert into t2 values (2, '2 - 02');
+insert into t2 values (2, '2 - 02');
+insert into t2 values (3, '2 - 03');
+explain select * from t1 where a1 in (select count(*) from t2 where b1 = b2 group by b2);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	3	Using where
+2	DEPENDENT SUBQUERY	t2	ALL	NULL	NULL	NULL	NULL	5	Using where
+select * from t1 where a1 in (select count(*) from t2 where b1 = b2 group by b2);
+a1	b1
+2	2 - 02
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS 
+from information_schema.OPTIMIZER_TRACE;
+JS
+NULL
+explain select * from t2 where a2 in (select a1 from t1 where b1 = b2 having a1 > 0);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t2	ALL	NULL	NULL	NULL	NULL	5	Using where
+2	DEPENDENT SUBQUERY	t1	ALL	NULL	NULL	NULL	NULL	3	Using where
+select * from t2 where a2 in (select a1 from t1 where b1 = b2 having a1 > 0);
+a2	b2
+1	2 - 01
+1	2 - 01
+2	2 - 02
+2	2 - 02
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+JS
+NULL
+explain select * from t1 where a1 not in (select a2 from t2 where b1 = b2);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	3	Using where
+2	DEPENDENT SUBQUERY	t2	ALL	NULL	NULL	NULL	NULL	5	Using where
+select * from t1 where a1 not in (select a2 from t2 where b1 = b2);
+a1	b1
+0	2 - 00
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+JS
+["decorrelation", "materialization"]
+drop table t1, t2;
+# check for aggregate function in the select items in the subquery; test adapted from main.delete_use_source_engines
+create table t1 (c1 integer, c2 integer, c3 integer);
+insert into t1(c1,c2,c3)
+values (1,1,1),(1,2,2),(1,3,3),
+(2,1,4),(2,2,5),(2,3,6),
+(2,4,7),(2,5,8);
+insert into t1 select c1+10,c2,c3+10 from t1;
+insert into t1 select c1+20,c2+1,c3+20 from t1;
+create view v1 as select * from t1 where c2=2;
+create table tmp as select * from t1 where t1.c2 in ( select max(a.c2) from v1 a where a.c1 = t1.c1);
+drop table t1, tmp;
+drop view v1;
+# check for subselect in left_expr; adapted from main.ps
+create table t1 (a int, b int, c int);
+create table t2 (x int, y int, z int);
+create table t3 as select * from t1;
+insert into t1 values (1,2,3),(4,5,6),(100,200,300),(400,500,600);
+insert into t2 values (1,2,3),(7,8,9),(100,200,300),(400,500,600);
+insert into t3 values (1,2,3),(11,12,13),(100,0,0),(400,500,600);
+explain select * from t1 where (select a from t3 where t3.c=t1.c) in (select x from t2 where t1.c= t2.z);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	4	
+1	PRIMARY	t2	ALL	NULL	NULL	NULL	NULL	4	Using where; Start temporary; End temporary; Using join buffer (flat, BNL join)
+2	DEPENDENT SUBQUERY	t3	ALL	NULL	NULL	NULL	NULL	4	Using where
+select * from t1 where (select a from t3 where t3.c=t1.c) in (select x from t2 where t1.c= t2.z);
+a	b	c
+1	2	3
+400	500	600
+explain select * from t1 where ((select a from t3 where t3.c=t1.c),b) in (select x, y from t2 where t1.c= t2.z);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	4	
+1	PRIMARY	t2	ALL	NULL	NULL	NULL	NULL	4	Using where; Start temporary; End temporary; Using join buffer (flat, BNL join)
+2	DEPENDENT SUBQUERY	t3	ALL	NULL	NULL	NULL	NULL	4	Using where
+select * from t1 where ((select a from t3 where t3.c=t1.c),b) in (select x, y from t2 where t1.c= t2.z);
+a	b	c
+1	2	3
+400	500	600
+explain select * from t1 where (select a,b from t3 where t3.c=t1.c) in (select x,y from t2 where t1.c= t2.z);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	4	
+1	PRIMARY	t2	ALL	NULL	NULL	NULL	NULL	4	Using where; Start temporary; End temporary; Using join buffer (flat, BNL join)
+2	DEPENDENT SUBQUERY	t3	ALL	NULL	NULL	NULL	NULL	4	Using where
+select * from t1 where (select a,b from t3 where t3.c=t1.c) in (select x,y from t2 where t1.c= t2.z);
+a	b	c
+1	2	3
+400	500	600
+drop table t1, t2, t3;
+# check that subselect is not a part of a union
+create table t1 (a1 char(8), b1 char(8));
+create table t2 (a2 char(8), b2 char(8));
+create table t3 (c1 char(8), c2 char(8));
+insert into t1 values ('1 - 00', '2 - 00');
+insert into t1 values ('1 - 01', '2 - 01');
+insert into t1 values ('1 - 02', '2 - 02');
+insert into t2 values ('1 - 01', '2 - 01');
+insert into t2 values ('1 - 01', '2 - 01');
+insert into t2 values ('1 - 02', '2 - 02');
+insert into t2 values ('1 - 02', '2 - 02');
+insert into t2 values ('1 - 03', '2 - 03');
+insert into t3 values ('1 - 01', '2 - 01');
+insert into t3 values ('1 - 01', '2 - 01');
+insert into t3 values ('1 - 03', '2 - 03');
+select * from t1 where a1 in
+((select a2 from t2 where b1 = b2) union
+(select c2 from t3 where a1 = c1));
+a1	b1
+1 - 01	2 - 01
+1 - 02	2 - 02
+drop table t1, t2, t3;
+# all / any
+create table t1 (a1 int, b1 int);
+create table t2 (a2 int, b2 int);
+insert into t1 values (0, 0);
+insert into t1 values (1, 1);
+insert into t1 values (3, 2);
+insert into t2 values (1, 0);
+insert into t2 values (1, 1);
+insert into t2 values (2, 2);
+insert into t2 values (3, 3);
+insert into t2 values (3, 3);
+explain select * from t1 where a1 = any (select a2 from t2 where b1 = b2);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	3	
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	t2	ALL	NULL	NULL	NULL	NULL	5	
+select * from t1 where a1 = any (select a2 from t2 where b1 = b2);
+a1	b1
+1	1
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+JS
+["decorrelation", "materialization", "semijoin"]
+explain select * from t1 where a1 <> all (select a2 from t2 where b1 = b2);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	3	Using where
+2	DEPENDENT SUBQUERY	t2	ALL	NULL	NULL	NULL	NULL	5	Using where
+select * from t1 where a1 <> all (select a2 from t2 where b1 = b2);
+a1	b1
+0	0
+3	2
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+JS
+["decorrelation", "materialization"]
+# no transformation applied when no synonyms applied to any / all
+select * from t1 where a1 >= all (select a2 from t2 where b1 = b2);
+a1	b1
+1	1
+3	2
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS 
+from information_schema.OPTIMIZER_TRACE;
+JS
+NULL
+select * from t1 where a1 <> any (select a2 from t2 where b1 = b2);
+a1	b1
+0	0
+3	2
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS 
+from information_schema.OPTIMIZER_TRACE;
+JS
+NULL
+drop table t1, t2;
+# fix NOT IN with null values
+create table t1 (a1 int, b1 int);
+create table t2 (a2 int, b2 int);
+insert into t1 values (0, null), (1, 1);
+insert into t2 values (0, 1), (1, null);
+# returns (0, null), (1, 1)
+select * from t1 where not a1 in (select a2 from t2 where b1 = b2);
+a1	b1
+0	NULL
+1	1
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+JS
+["decorrelation", "materialization"]
+# returns empty
+select * from t1 where not (a1, b1) in (select a2, b2 from t2);
+a1	b1
+# returns (0, null), (1, 1)
+select * from t1 where not (b1 is not null and (a1, b1) in (select a2, b2 from t2 where b2 is not null));
+a1	b1
+0	NULL
+1	1
+drop table t1, t2;
+# skip transformation when neither toplevel IN nor toplevel NOT IN; testcase adapated from main.subslect4
+CREATE TABLE t1 ( a INT, b INT );
+INSERT INTO t1 VALUES ( 1, NULL ), ( 2, NULL );
+CREATE TABLE t2 ( c INT, d INT );
+INSERT INTO t2 VALUES ( NULL, 3 ), ( NULL, 4 );
+SELECT * FROM t1 WHERE a NOT IN ( SELECT c FROM t2 where b = d ) IS NULL;
+a	b
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+JS
+NULL
+SELECT * FROM t1 WHERE a IN ( SELECT c FROM t2 where b = d ) IS NULL;
+a	b
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS 
+from information_schema.OPTIMIZER_TRACE;
+JS
+NULL
+drop table t1, t2;
+# 2nd ps execution / why we need to backup and restore the query arena; test also in main.subselect_sj2_mat
+CREATE TABLE t1 ( c varchar(1));
+INSERT INTO t1 VALUES ('r');
+CREATE TABLE t2 ( a integer, b varchar(1), c varchar(1));
+INSERT INTO t2 VALUES (1,'r','r');
+CREATE OR REPLACE VIEW v1 AS SELECT * FROM t1;
+PREPARE st2 FROM 'SELECT * FROM t2 WHERE a IN (SELECT a FROM v1 WHERE v1.c = t2.c)';
+EXECUTE st2;
+a	b	c
+1	r	r
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+JS
+["decorrelation", "materialization", "semijoin"]
+EXECUTE st2;
+a	b	c
+1	r	r
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+JS
+NULL
+drop table t1, t2;
+drop view v1;
+# 2nd ps execution / why we skip the transformation if we are in a ps execution and eligible equalities intersects with the free list; test also in main.subselect4
+CREATE TABLE t1 (a INT);
+INSERT INTO t1 VALUES (4),(6);
+CREATE TABLE t2 (b INT);
+INSERT INTO t2 VALUES (1),(8);
+PREPARE st5 FROM "
+SELECT * FROM t2
+HAVING 0 IN (
+  SELECT a FROM t1
+  WHERE a IN (
+    SELECT a FROM t1
+    WHERE b = a
+  )
+)
+";
+EXECUTE st5;
+b
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+JS
+["materialization", "semijoin"]
+EXECUTE st5;
+b
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+JS
+NULL
+drop table t1, t2;
+# test case where the in subselect is wrapped in an Item_in_optimizer / the necessity to update the Item_in_optimizer; test also in main.subselect4
+CREATE TABLE t1 (pk INT NOT NULL, i INT NOT NULL);
+INSERT INTO t1 VALUES (0,10), (1,20), (2,30), (3,40);
+CREATE TABLE t2a (pk INT NOT NULL, i INT NOT NULL, PRIMARY KEY(i,pk));
+INSERT INTO t2a VALUES (0,0), (1,1), (2,2), (3,3);
+EXPLAIN SELECT * FROM t1 WHERE (NULL, 1) NOT IN (SELECT t2a.i, t2a.pk FROM t2a WHERE t2a.pk = t1.pk);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	4	Using where
+2	DEPENDENT SUBQUERY	t2a	const	PRIMARY	PRIMARY	8	const,const	1	Using where; Using index; Full scan on NULL key
+SELECT * FROM t1 WHERE (NULL, 1) NOT IN (SELECT t2a.i, t2a.pk FROM t2a WHERE t2a.pk = t1.pk);
+pk	i
+0	10
+2	30
+3	40
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') AS JS
+from information_schema.OPTIMIZER_TRACE;
+JS
+["decorrelation", "materialization"]
+drop table t1, t2a;
+# misc cases
+CREATE TABLE t2 (pk INT PRIMARY KEY, b INT);
+INSERT INTO t2 VALUES (1,1),(2,7);
+SELECT * FROM t2 AS alias1, t2 AS alias2
+WHERE alias1.pk in ( SELECT 1 ) AND (alias2.pk = alias1.b )
+ORDER BY alias1.b;
+pk	b	pk	b
+1	1	1	1
+drop table t2;
+create table t1 (a1 char(8), b1 char(8));
+create table t2 (a2 char(8), b2 char(8));
+insert into t1 values ('1 - 00', '2 - 00');
+insert into t1 values ('1 - 01', '2 - 01');
+insert into t1 values ('1 - 02', '2 - 02');
+insert into t2 values ('1 - 01', '2 - 01');
+insert into t2 values ('1 - 01', '2 - 01');
+insert into t2 values ('1 - 02', '2 - 02');
+insert into t2 values ('1 - 03', '2 - 02');
+insert into t2 values ('1 - 03', '2 - 03');
+select * from t1 where a1 in (select a2 from t2 where b1 = b2 order by b2 desc limit 1, 2);
+ERROR 42000: This version of MariaDB doesn't yet support 'LIMIT & IN/ALL/ANY/SOME subquery'
+drop table t1, t2;
+# outer_exp has subquery
+create table t1 (a1 int, b1 int);
+create table t2 (a2 int, b2 int);
+insert into t1 values (0, 0);
+insert into t1 values (1, 1);
+insert into t1 values (2, 2);
+insert into t2 values (1, 0);
+insert into t2 values (1, 1);
+insert into t2 values (2, 2);
+insert into t2 values (3, 2);
+insert into t2 values (3, 3);
+select * from t1 where a1 in
+(select a2 from t2 where
+a1 + (select b2 from t2) = b2);
+ERROR 21000: Subquery returns more than 1 row
+explain select * from t1 where a1 in
+(select a2 from t2 where
+(select a1 + b2 from t2 order by b2 limit 1) = a2);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	3	
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	Using where
+2	MATERIALIZED	t2	ALL	NULL	NULL	NULL	NULL	5	
+3	DEPENDENT SUBQUERY	t2	ALL	NULL	NULL	NULL	NULL	5	Using filesort
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+JS
+["decorrelation", "materialization", "semijoin"]
+select * from t1 where a1 in
+(select a2 from t2 where
+(select a1 + b2 from t2 order by b2 limit 1) = a2);
+a1	b1
+1	1
+2	2
+drop table t1, t2;
+# outer_exp has fulltext function
+create table t1 (a1 int, b1 int);
+create table t2 (a2 int, b2 text, fulltext(b2));
+insert into t1 values (0, 0);
+insert into t1 values (1, 1);
+insert into t1 values (2, 0);
+insert into t2 values (1, 'MariaDB overview');
+insert into t2 values (1, 'https://mariadb.com/');
+insert into t2 values (2, 'Foreign keys');
+insert into t2 values (3, 'Indexes');
+insert into t2 values (3, 'Triggers');
+explain select * from t1 where a1 in
+(select a2 from t2 where
+(select b1 + count(*) from t2 where
+match(b2) against('mariadb')) = a2);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	3	
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	Using where
+2	MATERIALIZED	t2	ALL	NULL	NULL	NULL	NULL	5	
+3	DEPENDENT SUBQUERY	t2	fulltext	b2	b2	0		1	Using where
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+JS
+["decorrelation", "materialization", "semijoin"]
+select * from t1 where a1 in
+(select a2 from t2 where
+(select b1 + count(*) from t2 where
+match(b2) against('mariadb')) = a2);
+a1	b1
+2	0
+drop table t1, t2;
+# type mismatch leading to no materialization after transformation
+create table t1 (a1 int, b1 int);
+create table t2 (a2 int);
+insert into t1 values (0, 0);
+insert into t1 values (1, 1);
+insert into t1 values (2, 2);
+insert into t2 values (1);
+insert into t2 values (1);
+insert into t2 values (2);
+insert into t2 values (3);
+insert into t2 values (3);
+alter table t2 add b2 varchar(32);
+explain select * from t1 where a1 in (select 1 from t2 where b1 = b2);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	3	Using where
+1	PRIMARY	t2	ALL	NULL	NULL	NULL	NULL	5	Using where; Start temporary; End temporary; Using join buffer (flat, BNL join)
+select * from t1 where a1 in (select 1 from t2 where b1 = b2);
+a1	b1
+select 
+json_detailed(json_extract(trace, '$**.join_optimization.steps[*].transformation')) as JS 
+from information_schema.OPTIMIZER_TRACE;
+JS
+[
+    {
+        "select_id": 2,
+        "from": "IN (SELECT)",
+        "to": "decorrelation",
+        "is_correlated": false
+    },
+    {
+        "select_id": 2,
+        "from": "IN (SELECT)",
+        "to": "materialization",
+        "possible": false,
+        "cause": "types mismatch"
+    },
+    {
+        "select_id": 2,
+        "from": "IN (SELECT)",
+        "to": "semijoin",
+        "converted_to_semi_join": true
+    }
+]
+drop table t1, t2;
+# non-semijoin materialization
+create table ten(a int);
+insert into ten values (0),(1),(2),(3),(4),(5),(6),(7),(8),(9);
+create table one_h(a int);
+insert into one_h select A.a + B.a* 10 from ten A, ten B;
+create table ten_k(a int primary key);
+insert into ten_k select A.a + B.a* 100 from one_h A, one_h B;
+create table t1 (a int, b int, c int );
+insert into t1 select a, a, a from one_h;
+create table t2 (a int, b int, c int );
+insert into t2  select a,a,a from ten_k;
+explain select * from t1 where 1 in (select 1 from t2 where t2.a=t1.a and t2.b=t1.b) or t1.c<333;
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	100	Using where
+2	MATERIALIZED	t2	ALL	NULL	NULL	NULL	NULL	10000	
+select 
+json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+JS
+["decorrelation", "materialization"]
+drop table ten, one_h, ten_k, t1, t2;
+#
+#
+#
+create table t1 ( a int, b int, c int);
+insert into t1 select seq, seq, seq from seq_1_to_10;
+create table t2 ( x int, y int, z int);
+insert into t2 select seq, seq, seq from seq_1_to_10;
+insert into t1 values
+( NULL ,   12 ,   12 ),
+(   12 , NULL ,   12 ),
+( NULL ,   13 ,   13 ),
+(   13 , NULL ,   13 ),
+(   21 ,   21 ,   21 ),
+(   22 ,   22 ,   22 ),
+(   12 ,   12 ,   12 ),
+(   22 , NULL ,   22 ),
+( NULL ,   21 ,   21 ),
+(   23 , NULL ,   23 ),
+( NULL ,   23 ,   23 );
+create table one(w int);
+insert into one values (1);
+alter table t2 add key(x);
+alter table t2 add key(y);
+set @tmp=@@optimizer_switch, optimizer_switch='materialization=off';
+# This will not show full-scan-on-null-key:
+explain format=json 
+select * from t1 
+where t1.a not in (select t2.x from one,t2 ignore index(x) where t1.b=t2.y and t2.z=33);
+EXPLAIN
+{
+  "query_block": {
+    "select_id": 1,
+    "cost": "COST_REPLACED",
+    "nested_loop": [
+      {
+        "table": {
+          "table_name": "t1",
+          "access_type": "ALL",
+          "loops": 1,
+          "rows": 21,
+          "cost": "COST_REPLACED",
+          "filtered": 100,
+          "attached_condition": "!(t1.b is not null and <in_optimizer>((t1.a,t1.b),<exists>(subquery#2)))"
+        }
+      }
+    ],
+    "subqueries": [
+      {
+        "query_block": {
+          "select_id": 2,
+          "cost": "COST_REPLACED",
+          "having_condition": "trigcond(t2.x is null)",
+          "nested_loop": [
+            {
+              "table": {
+                "table_name": "one",
+                "access_type": "system",
+                "rows": 1,
+                "filtered": 100
+              }
+            },
+            {
+              "table": {
+                "table_name": "t2",
+                "access_type": "ref",
+                "possible_keys": ["y"],
+                "key": "y",
+                "key_length": "5",
+                "used_key_parts": ["y"],
+                "ref": ["func"],
+                "loops": 1,
+                "rows": 1,
+                "cost": "COST_REPLACED",
+                "filtered": 100,
+                "index_condition": "t2.y is not null",
+                "attached_condition": "t2.z = 33 and trigcond(trigcond(<cache>(t1.a) = t2.x or t2.x is null))"
+              }
+            }
+          ]
+        }
+      }
+    ]
+  }
+}
+# This will show full-scan-on-null-key:
+explain format=json 
+select * from t1 
+where t1.a not in (select t2.x from one,t2 ignore index(y) where t1.b=t2.y and t2.z=33);
+EXPLAIN
+{
+  "query_block": {
+    "select_id": 1,
+    "cost": "COST_REPLACED",
+    "nested_loop": [
+      {
+        "table": {
+          "table_name": "t1",
+          "access_type": "ALL",
+          "loops": 1,
+          "rows": 21,
+          "cost": "COST_REPLACED",
+          "filtered": 100,
+          "attached_condition": "!(t1.b is not null and <in_optimizer>((t1.a,t1.b),<exists>(subquery#2)))"
+        }
+      }
+    ],
+    "subqueries": [
+      {
+        "query_block": {
+          "select_id": 2,
+          "cost": "COST_REPLACED",
+          "having_condition": "trigcond(t2.x is null)",
+          "nested_loop": [
+            {
+              "table": {
+                "table_name": "one",
+                "access_type": "system",
+                "rows": 1,
+                "filtered": 100
+              }
+            },
+            {
+              "full-scan-on-null_key": {
+                "table": {
+                  "table_name": "t2",
+                  "access_type": "ref_or_null",
+                  "possible_keys": ["x"],
+                  "key": "x",
+                  "key_length": "5",
+                  "used_key_parts": ["x"],
+                  "ref": ["func"],
+                  "loops": 1,
+                  "rows": 4,
+                  "cost": "COST_REPLACED",
+                  "filtered": 100,
+                  "attached_condition": "t2.z = 33 and t2.y is not null and trigcond(trigcond(<cache>(t1.a) = t2.x or t2.x is null)) and <cache>(t1.b) = t2.y"
+                }
+              }
+            }
+          ]
+        }
+      }
+    ]
+  }
+}
+drop table one,t1,t2;

mysql-test/main/subselect_decorrelate_in.test

+--echo #
+--echo # MDEV-22534 Trivial correlation detection/removal for IN subqueries
+--echo #
+
+# The test uses optimizer trace:
+--source include/not_embedded.inc
+--source include/have_sequence.inc
+
+--echo # Basic example
+set optimizer_switch='materialization=on,firstmatch=off';
+set optimizer_trace=1;
+create table ten(a int primary key);
+insert into ten values (0),(1),(2),(3),(4),(5),(6),(7),(8),(9);
+create table t1 (a int, b int, c int);
+insert into t1 select a,a,a from ten;
+create table t2 select * from t1;
+
+let $query=
+select * from t1 where a in (select a from t2 where t1.b=t2.b);
+eval explain $query;
+eval $query;
+
+let $query=
+select * from t1 where (a, b) in (select a, b from t2);
+eval explain $query;
+eval $query;
+
+--echo # Multiple equalities
+let $query=
+select * from t1 where a in (select a from t2 where t1.b=t2.b and t1.c=t2.c);
+eval explain $query;
+eval $query;
+
+# Row item equality; requires extending `check_equality_for_exist2in()` to allow expr=expr to decorrelate (MDEV-31229)
+let $query=
+select * from t1 where a in (select a from t2 where (t1.b, t1.c)=(t2.b, t2.c));
+eval explain $query;
+eval $query;
+
+# Row in left_expr
+let $query=
+select * from t1 where (a, b) in (select a, b from t2 where t1.c=t2.c);
+eval explain $query;
+eval $query;
+
+let $query=
+select * from t1 where a in (select a from t2 where t1.b=t2.b and t2.c+t2.b < 5);
+eval explain $query;
+eval $query;
+
+let $query=
+select * from t1 where a + b in (select a * c from t2 where t1.b=t2.b and t2.c+t2.b < 9);
+eval explain $query;
+eval $query;
+
+--echo # Still correlated after the transformation / no materialization
+let $query=
+select * from t1 where a in (select a from t2 where t1.b=t2.b and t1.c < 5);
+eval explain $query;
+eval $query;
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+
+let $query=
+select * from t1 where a in (select a from t2 where t1.a = t2.b and t2.c+t1.a < 9);
+eval explain $query;
+eval $query;
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS 
+from information_schema.OPTIMIZER_TRACE;
+
+--echo # No transformation because of applicable conditions in subquery
+let $query=
+select * from t1 where a in (select a from t2 where t2.c+t1.a < 9);
+eval explain $query;
+eval $query;
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS 
+from information_schema.OPTIMIZER_TRACE;
+
+drop table ten, t1, t2;
+
+--echo # group by and having checks in the transformer; also not in
+
+create table t1 (a1 int not null, b1 char(8) not null);
+create table t2 (a2 int not null, b2 char(8) not null);
+
+insert into t1 values (0, '2 - 00');
+insert into t1 values (1, '2 - 01');
+insert into t1 values (2, '2 - 02');
+
+insert into t2 values (1, '2 - 01');
+insert into t2 values (1, '2 - 01');
+insert into t2 values (2, '2 - 02');
+insert into t2 values (2, '2 - 02');
+insert into t2 values (3, '2 - 03');
+
+# skip on group by
+let $query=
+select * from t1 where a1 in (select count(*) from t2 where b1 = b2 group by b2);
+eval explain $query;
+eval $query;
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS 
+from information_schema.OPTIMIZER_TRACE;
+
+# skip having
+let $query=
+select * from t2 where a2 in (select a1 from t1 where b1 = b2 having a1 > 0);
+eval explain $query;
+eval $query;
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+
+# not in
+let $query=
+select * from t1 where a1 not in (select a2 from t2 where b1 = b2);
+eval explain $query;
+eval $query;
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+
+drop table t1, t2;
+
+--echo # check for aggregate function in the select items in the subquery; test adapted from main.delete_use_source_engines
+create table t1 (c1 integer, c2 integer, c3 integer);
+insert into t1(c1,c2,c3)
+values (1,1,1),(1,2,2),(1,3,3),
+       (2,1,4),(2,2,5),(2,3,6),
+       (2,4,7),(2,5,8);
+insert into t1 select c1+10,c2,c3+10 from t1;
+insert into t1 select c1+20,c2+1,c3+20 from t1;
+create view v1 as select * from t1 where c2=2;
+let $c=
+t1.c2 in ( select max(a.c2) from v1 a where a.c1 = t1.c1);
+eval create table tmp as select * from t1 where $c;
+drop table t1, tmp;
+drop view v1;
+
+--echo # check for subselect in left_expr; adapted from main.ps
+
+create table t1 (a int, b int, c int);
+create table t2 (x int, y int, z int);
+create table t3 as select * from t1;
+insert into t1 values (1,2,3),(4,5,6),(100,200,300),(400,500,600);
+insert into t2 values (1,2,3),(7,8,9),(100,200,300),(400,500,600);
+insert into t3 values (1,2,3),(11,12,13),(100,0,0),(400,500,600);
+
+# scalar subselect - ok
+let $query=
+select * from t1 where (select a from t3 where t3.c=t1.c) in (select x from t2 where t1.c= t2.z);
+eval explain $query;
+eval $query;
+
+# a row containing a subselect - ok
+let $query=
+select * from t1 where ((select a from t3 where t3.c=t1.c),b) in (select x, y from t2 where t1.c= t2.z);
+eval explain $query;
+eval $query;
+
+# a single non-scalar subselect - skip transformation
+let $query=
+select * from t1 where (select a,b from t3 where t3.c=t1.c) in (select x,y from t2 where t1.c= t2.z);
+eval explain $query;
+eval $query;
+
+drop table t1, t2, t3;
+
+--echo # check that subselect is not a part of a union
+
+# union
+create table t1 (a1 char(8), b1 char(8));
+create table t2 (a2 char(8), b2 char(8));
+create table t3 (c1 char(8), c2 char(8));
+
+insert into t1 values ('1 - 00', '2 - 00');
+insert into t1 values ('1 - 01', '2 - 01');
+insert into t1 values ('1 - 02', '2 - 02');
+
+insert into t2 values ('1 - 01', '2 - 01');
+insert into t2 values ('1 - 01', '2 - 01');
+insert into t2 values ('1 - 02', '2 - 02');
+insert into t2 values ('1 - 02', '2 - 02');
+insert into t2 values ('1 - 03', '2 - 03');
+
+insert into t3 values ('1 - 01', '2 - 01');
+insert into t3 values ('1 - 01', '2 - 01');
+insert into t3 values ('1 - 03', '2 - 03');
+
+
+select * from t1 where a1 in
+  ((select a2 from t2 where b1 = b2) union
+   (select c2 from t3 where a1 = c1));
+
+drop table t1, t2, t3;
+
+--echo # all / any
+create table t1 (a1 int, b1 int);
+create table t2 (a2 int, b2 int);
+
+insert into t1 values (0, 0);
+insert into t1 values (1, 1);
+insert into t1 values (3, 2);
+
+insert into t2 values (1, 0);
+insert into t2 values (1, 1);
+insert into t2 values (2, 2);
+insert into t2 values (3, 3);
+insert into t2 values (3, 3);
+
+# transformation applied because `= any` is synonym to `in`
+let $query=
+select * from t1 where a1 = any (select a2 from t2 where b1 = b2);
+eval explain $query;
+eval $query;
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+
+# transformation applied because `<> all` is synonym to `not in`
+let $query=
+select * from t1 where a1 <> all (select a2 from t2 where b1 = b2);
+eval explain $query;
+eval $query;
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+
+--echo # no transformation applied when no synonyms applied to any / all
+select * from t1 where a1 >= all (select a2 from t2 where b1 = b2);
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS 
+from information_schema.OPTIMIZER_TRACE;
+
+select * from t1 where a1 <> any (select a2 from t2 where b1 = b2);
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS 
+from information_schema.OPTIMIZER_TRACE;
+
+drop table t1, t2;
+
+--echo # fix NOT IN with null values
+create table t1 (a1 int, b1 int);
+create table t2 (a2 int, b2 int);
+
+insert into t1 values (0, null), (1, 1);
+insert into t2 values (0, 1), (1, null);
+
+--echo # returns (0, null), (1, 1)
+select * from t1 where not a1 in (select a2 from t2 where b1 = b2);
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+
+--echo # returns empty
+select * from t1 where not (a1, b1) in (select a2, b2 from t2);
+--echo # returns (0, null), (1, 1)
+select * from t1 where not (b1 is not null and (a1, b1) in (select a2, b2 from t2 where b2 is not null));
+
+drop table t1, t2;
+
+--echo # skip transformation when neither toplevel IN nor toplevel NOT IN; testcase adapated from main.subslect4
+CREATE TABLE t1 ( a INT, b INT );
+INSERT INTO t1 VALUES ( 1, NULL ), ( 2, NULL );
+CREATE TABLE t2 ( c INT, d INT );
+INSERT INTO t2 VALUES ( NULL, 3 ), ( NULL, 4 );
+# not top level, upper_not is not top level
+SELECT * FROM t1 WHERE a NOT IN ( SELECT c FROM t2 where b = d ) IS NULL;
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+
+# not top level, upper_not is null
+SELECT * FROM t1 WHERE a IN ( SELECT c FROM t2 where b = d ) IS NULL;
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS 
+from information_schema.OPTIMIZER_TRACE;
+
+drop table t1, t2;
+
+--echo # 2nd ps execution / why we need to backup and restore the query arena; test also in main.subselect_sj2_mat
+
+CREATE TABLE t1 ( c varchar(1));
+INSERT INTO t1 VALUES ('r');
+CREATE TABLE t2 ( a integer, b varchar(1), c varchar(1));
+INSERT INTO t2 VALUES (1,'r','r');
+CREATE OR REPLACE VIEW v1 AS SELECT * FROM t1;
+PREPARE st2 FROM 'SELECT * FROM t2 WHERE a IN (SELECT a FROM v1 WHERE v1.c = t2.c)';
+EXECUTE st2;
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+EXECUTE st2;
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+
+drop table t1, t2;
+drop view v1;
+
+--echo # 2nd ps execution / why we skip the transformation if we are in a ps execution and eligible equalities intersects with the free list; test also in main.subselect4
+
+CREATE TABLE t1 (a INT);
+INSERT INTO t1 VALUES (4),(6);
+CREATE TABLE t2 (b INT);
+INSERT INTO t2 VALUES (1),(8);
+PREPARE st5 FROM "
+SELECT * FROM t2
+HAVING 0 IN (
+  SELECT a FROM t1
+  WHERE a IN (
+    SELECT a FROM t1
+    WHERE b = a
+  )
+)
+";
+EXECUTE st5;
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+EXECUTE st5;
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+
+drop table t1, t2;
+
+--echo # test case where the in subselect is wrapped in an Item_in_optimizer / the necessity to update the Item_in_optimizer; test also in main.subselect4
+
+CREATE TABLE t1 (pk INT NOT NULL, i INT NOT NULL);
+INSERT INTO t1 VALUES (0,10), (1,20), (2,30), (3,40);
+
+CREATE TABLE t2a (pk INT NOT NULL, i INT NOT NULL, PRIMARY KEY(i,pk));
+INSERT INTO t2a VALUES (0,0), (1,1), (2,2), (3,3);
+
+let $query=
+SELECT * FROM t1 WHERE (NULL, 1) NOT IN (SELECT t2a.i, t2a.pk FROM t2a WHERE t2a.pk = t1.pk);
+eval EXPLAIN $query;
+eval $query;
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') AS JS
+from information_schema.OPTIMIZER_TRACE;
+
+drop table t1, t2a;
+
+--echo # misc cases
+# would-be empty leaf_tables if not for prior optimization, adapted from similar case for exists2in in main.subselect4
+CREATE TABLE t2 (pk INT PRIMARY KEY, b INT);
+INSERT INTO t2 VALUES (1,1),(2,7);
+SELECT * FROM t2 AS alias1, t2 AS alias2
+WHERE alias1.pk in ( SELECT 1 ) AND (alias2.pk = alias1.b )
+ORDER BY alias1.b;
+drop table t2;
+
+# limit is not supported in IN subquery, so we don't have to worry about it
+create table t1 (a1 char(8), b1 char(8));
+create table t2 (a2 char(8), b2 char(8));
+
+insert into t1 values ('1 - 00', '2 - 00');
+insert into t1 values ('1 - 01', '2 - 01');
+insert into t1 values ('1 - 02', '2 - 02');
+
+insert into t2 values ('1 - 01', '2 - 01');
+insert into t2 values ('1 - 01', '2 - 01');
+insert into t2 values ('1 - 02', '2 - 02');
+insert into t2 values ('1 - 03', '2 - 02');
+insert into t2 values ('1 - 03', '2 - 03');
+
+--error ER_NOT_SUPPORTED_YET
+select * from t1 where a1 in (select a2 from t2 where b1 = b2 order by b2 desc limit 1, 2);
+
+drop table t1, t2;
+
+--echo # outer_exp has subquery
+create table t1 (a1 int, b1 int);
+create table t2 (a2 int, b2 int);
+
+insert into t1 values (0, 0);
+insert into t1 values (1, 1);
+insert into t1 values (2, 2);
+
+insert into t2 values (1, 0);
+insert into t2 values (1, 1);
+insert into t2 values (2, 2);
+insert into t2 values (3, 2);
+insert into t2 values (3, 3);
+
+--error ER_SUBQUERY_NO_1_ROW
+select * from t1 where a1 in
+       (select a2 from t2 where
+               a1 + (select b2 from t2) = b2);
+
+let $query=
+select * from t1 where a1 in
+       (select a2 from t2 where
+               (select a1 + b2 from t2 order by b2 limit 1) = a2);
+
+eval explain $query;
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+eval $query;
+
+drop table t1, t2;
+
+--echo # outer_exp has fulltext function
+create table t1 (a1 int, b1 int);
+create table t2 (a2 int, b2 text, fulltext(b2));
+
+insert into t1 values (0, 0);
+insert into t1 values (1, 1);
+insert into t1 values (2, 0);
+
+insert into t2 values (1, 'MariaDB overview');
+insert into t2 values (1, 'https://mariadb.com/');
+insert into t2 values (2, 'Foreign keys');
+insert into t2 values (3, 'Indexes');
+insert into t2 values (3, 'Triggers');
+
+let $query=
+select * from t1 where a1 in
+       (select a2 from t2 where
+               (select b1 + count(*) from t2 where
+                       match(b2) against('mariadb')) = a2);
+# with ps-protocol the count(0) which is an Item_aggregate_ref is in
+# the runtime arena causing skipping the decorrelation
+--disable_ps_protocol
+eval explain $query;
+--enable_ps_protocol
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+eval $query;
+
+drop table t1, t2;
+
+--echo # type mismatch leading to no materialization after transformation
+create table t1 (a1 int, b1 int);
+create table t2 (a2 int);
+
+insert into t1 values (0, 0);
+insert into t1 values (1, 1);
+insert into t1 values (2, 2);
+
+insert into t2 values (1);
+insert into t2 values (1);
+insert into t2 values (2);
+insert into t2 values (3);
+insert into t2 values (3);
+
+alter table t2 add b2 varchar(32);
+let $query=
+select * from t1 where a1 in (select 1 from t2 where b1 = b2);
+eval explain $query;
+eval $query;
+select 
+  json_detailed(json_extract(trace, '$**.join_optimization.steps[*].transformation')) as JS 
+from information_schema.OPTIMIZER_TRACE;
+
+drop table t1, t2;
+
+--echo # non-semijoin materialization
+create table ten(a int);
+insert into ten values (0),(1),(2),(3),(4),(5),(6),(7),(8),(9);
+
+create table one_h(a int);
+insert into one_h select A.a + B.a* 10 from ten A, ten B;
+
+create table ten_k(a int primary key);
+insert into ten_k select A.a + B.a* 100 from one_h A, one_h B;
+
+create table t1 (a int, b int, c int );
+
+insert into t1 select a, a, a from one_h;
+
+create table t2 (a int, b int, c int );
+insert into t2  select a,a,a from ten_k;
+
+explain select * from t1 where 1 in (select 1 from t2 where t2.a=t1.a and t2.b=t1.b) or t1.c<333;
+select 
+  json_extract(trace, '$**.join_optimization.steps[*].transformation.to') as JS
+from information_schema.OPTIMIZER_TRACE;
+
+drop table ten, one_h, ten_k, t1, t2;
+
+--echo #
+--echo #
+--echo #
+create table t1 ( a int, b int, c int);
+insert into t1 select seq, seq, seq from seq_1_to_10;
+create table t2 ( x int, y int, z int);
+insert into t2 select seq, seq, seq from seq_1_to_10;
+
+insert into t1 values
+( NULL ,   12 ,   12 ),
+(   12 , NULL ,   12 ),
+( NULL ,   13 ,   13 ),
+(   13 , NULL ,   13 ),
+(   21 ,   21 ,   21 ),
+(   22 ,   22 ,   22 ),
+
+(   12 ,   12 ,   12 ),
+(   22 , NULL ,   22 ),
+( NULL ,   21 ,   21 ),
+(   23 , NULL ,   23 ),
+( NULL ,   23 ,   23 );
+
+create table one(w int);
+insert into one values (1);
+
+alter table t2 add key(x);
+alter table t2 add key(y);
+set @tmp=@@optimizer_switch, optimizer_switch='materialization=off';
+
+--echo # This will not show full-scan-on-null-key:
+--source include/explain-no-costs.inc
+explain format=json 
+select * from t1 
+where t1.a not in (select t2.x from one,t2 ignore index(x) where t1.b=t2.y and t2.z=33);
+
+--echo # This will show full-scan-on-null-key:
+--source include/explain-no-costs.inc
+explain format=json 
+select * from t1 
+where t1.a not in (select t2.x from one,t2 ignore index(y) where t1.b=t2.y and t2.z=33);
+
+drop table one,t1,t2;

mysql-test/main/subselect_extra_no_semijoin.result

@@ -454,7 +454,7 @@ WHERE t2.c = v1.c AND t2.c = v1.b AND v1.b = t3.c);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t3	ALL	NULL	NULL	NULL	NULL	2	Using where
 2	DEPENDENT SUBQUERY	t2	system	NULL	NULL	NULL	NULL	1	
-2	DEPENDENT SUBQUERY	<derived3>	ref	key1	key1	8	const,const	0	Using where
+2	DEPENDENT SUBQUERY	<derived3>	ref	key0	key0	8	const,const	0	Using where
 3	DERIVED	t1	ALL	NULL	NULL	NULL	NULL	3	Using where
 SELECT * FROM t3 
 WHERE t3.b IN (SELECT v1.b FROM  v1, t2 
@@ -476,7 +476,7 @@ EXPLAIN
 SELECT * FROM t1 WHERE t1.b IN (SELECT v2.a FROM v2 WHERE v2.b = t1.a);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	2	Using where
-2	DEPENDENT SUBQUERY	<derived3>	ref	key0	key0	5	test.t1.a	1	Using where
+2	DEPENDENT SUBQUERY	<derived3>	index_subquery	key0	key0	10	func,func	1	Using where
 3	DERIVED	t2	ALL	NULL	NULL	NULL	NULL	2	
 SELECT * FROM t1 WHERE t1.b IN (SELECT v2.a FROM v2 WHERE v2.b = t1.a);
 a	b

mysql-test/main/subselect_mat.result

@@ -387,10 +387,10 @@ id	select_type	table	type	possible_keys	key	key_len	ref	rows	filtered	Extra
 6	MATERIALIZED	t2i	index	it2i2	it2i3	18	NULL	5	100.00	Using where; Using index
 2	DEPENDENT SUBQUERY	t2	ALL	NULL	NULL	NULL	NULL	5	100.00	Using where
 4	MATERIALIZED	t3b	ALL	NULL	NULL	NULL	NULL	4	100.00	Using where
-3	DEPENDENT SUBQUERY	t3a	ALL	NULL	NULL	NULL	NULL	4	100.00	Using where
+3	MATERIALIZED	t3a	ALL	NULL	NULL	NULL	NULL	4	100.00	
 Warnings:
 Note	1276	Field or reference 'test.t1.a1' of SELECT #3 was resolved in SELECT #1
-Note	1003	/* select#1 */ select `test`.`t1`.`a1` AS `a1`,`test`.`t1`.`a2` AS `a2` from `test`.`t1` where <expr_cache><`test`.`t1`.`a1`,`test`.`t1`.`a2`>(<in_optimizer>((`test`.`t1`.`a1`,`test`.`t1`.`a2`),<exists>(/* select#2 */ select `test`.`t2`.`b1`,`test`.`t2`.`b2` from `test`.`t2` where (<expr_cache><`test`.`t2`.`b2`,`test`.`t1`.`a1`>(<in_optimizer>(`test`.`t2`.`b2`,<exists>(/* select#3 */ select `test`.`t3a`.`c2` from `test`.`t3` `t3a` where `test`.`t3a`.`c1` = `test`.`t1`.`a1` and <cache>(`test`.`t2`.`b2`) = `test`.`t3a`.`c2`))) or <expr_cache><`test`.`t2`.`b2`>(<in_optimizer>(`test`.`t2`.`b2`,`test`.`t2`.`b2` in ( <materialize> (/* select#4 */ select `test`.`t3b`.`c2` from `test`.`t3` `t3b` where `test`.`t3b`.`c2` like '%03' ), <primary_index_lookup>(`test`.`t2`.`b2` in <temporary table> on distinct_key where `test`.`t2`.`b2` = `<subquery4>`.`c2`))))) and <cache>(`test`.`t1`.`a1`) = `test`.`t2`.`b1` and <cache>(`test`.`t1`.`a2`) = `test`.`t2`.`b2`))) and <expr_cache><`test`.`t1`.`a1`,`test`.`t1`.`a2`>(<in_optimizer>((`test`.`t1`.`a1`,`test`.`t1`.`a2`),(`test`.`t1`.`a1`,`test`.`t1`.`a2`) in ( <materialize> (/* select#5 */ select `test`.`t3c`.`c1`,`test`.`t3c`.`c2` from `test`.`t3` `t3c` where <expr_cache><`test`.`t3c`.`c1`,`test`.`t3c`.`c2`>(<in_optimizer>((`test`.`t3c`.`c1`,`test`.`t3c`.`c2`),(`test`.`t3c`.`c1`,`test`.`t3c`.`c2`) in ( <materialize> (/* select#6 */ select `test`.`t2i`.`b1`,`test`.`t2i`.`b2` from `test`.`t2i` where `test`.`t2i`.`b2` > '0' ), <primary_index_lookup>(`test`.`t3c`.`c1` in <temporary table> on distinct_key where `test`.`t3c`.`c1` = `<subquery6>`.`b1` and `test`.`t3c`.`c2` = `<subquery6>`.`b2`)))) ), <primary_index_lookup>(`test`.`t1`.`a1` in <temporary table> on distinct_key where `test`.`t1`.`a1` = `<subquery5>`.`c1` and `test`.`t1`.`a2` = `<subquery5>`.`c2`))))
+Note	1003	/* select#1 */ select `test`.`t1`.`a1` AS `a1`,`test`.`t1`.`a2` AS `a2` from `test`.`t1` where <expr_cache><`test`.`t1`.`a1`,`test`.`t1`.`a2`>(<in_optimizer>((`test`.`t1`.`a1`,`test`.`t1`.`a2`),<exists>(/* select#2 */ select `test`.`t2`.`b1`,`test`.`t2`.`b2` from `test`.`t2` where (<expr_cache><`test`.`t2`.`b2`,`test`.`t1`.`a1`>(<in_optimizer>((`test`.`t2`.`b2`,`test`.`t1`.`a1`),(`test`.`t2`.`b2`,`test`.`t1`.`a1`) in ( <materialize> (/* select#3 */ select `test`.`t3a`.`c2`,`test`.`t3a`.`c1` from `test`.`t3` `t3a` where 1 ), <primary_index_lookup>(`test`.`t2`.`b2` in <temporary table> on distinct_key where `test`.`t2`.`b2` = `<subquery3>`.`c2` and `test`.`t1`.`a1` = `<subquery3>`.`c1`)))) or <expr_cache><`test`.`t2`.`b2`>(<in_optimizer>(`test`.`t2`.`b2`,`test`.`t2`.`b2` in ( <materialize> (/* select#4 */ select `test`.`t3b`.`c2` from `test`.`t3` `t3b` where `test`.`t3b`.`c2` like '%03' ), <primary_index_lookup>(`test`.`t2`.`b2` in <temporary table> on distinct_key where `test`.`t2`.`b2` = `<subquery4>`.`c2`))))) and <cache>(`test`.`t1`.`a1`) = `test`.`t2`.`b1` and <cache>(`test`.`t1`.`a2`) = `test`.`t2`.`b2`))) and <expr_cache><`test`.`t1`.`a1`,`test`.`t1`.`a2`>(<in_optimizer>((`test`.`t1`.`a1`,`test`.`t1`.`a2`),(`test`.`t1`.`a1`,`test`.`t1`.`a2`) in ( <materialize> (/* select#5 */ select `test`.`t3c`.`c1`,`test`.`t3c`.`c2` from `test`.`t3` `t3c` where <expr_cache><`test`.`t3c`.`c1`,`test`.`t3c`.`c2`>(<in_optimizer>((`test`.`t3c`.`c1`,`test`.`t3c`.`c2`),(`test`.`t3c`.`c1`,`test`.`t3c`.`c2`) in ( <materialize> (/* select#6 */ select `test`.`t2i`.`b1`,`test`.`t2i`.`b2` from `test`.`t2i` where `test`.`t2i`.`b2` > '0' ), <primary_index_lookup>(`test`.`t3c`.`c1` in <temporary table> on distinct_key where `test`.`t3c`.`c1` = `<subquery6>`.`b1` and `test`.`t3c`.`c2` = `<subquery6>`.`b2`)))) ), <primary_index_lookup>(`test`.`t1`.`a1` in <temporary table> on distinct_key where `test`.`t1`.`a1` = `<subquery5>`.`c1` and `test`.`t1`.`a2` = `<subquery5>`.`c2`))))
 select * from t1
 where (a1, a2) in (select b1, b2 from t2
 where b2 in (select c2 from t3 t3a where c1 = a1) or
@@ -520,11 +520,11 @@ id	select_type	table	type	possible_keys	key	key_len	ref	rows	filtered	Extra
 6	DEPENDENT SUBQUERY	t2i	index_subquery	it2i1,it2i2,it2i3	it2i3	18	func,func	1	100.00	Using index; Using where
 2	DEPENDENT SUBQUERY	t2	ALL	NULL	NULL	NULL	NULL	5	100.00	Using where
 4	MATERIALIZED	t3b	ALL	NULL	NULL	NULL	NULL	4	100.00	Using where
-3	DEPENDENT SUBQUERY	t3a	ALL	NULL	NULL	NULL	NULL	4	100.00	Using where
+3	MATERIALIZED	t3a	ALL	NULL	NULL	NULL	NULL	4	100.00	
 Warnings:
 Note	1276	Field or reference 'test.t1.a1' of SELECT #3 was resolved in SELECT #1
 Note	1276	Field or reference 'test.t1.a2' of SELECT #6 was resolved in SELECT #1
-Note	1003	/* select#1 */ select `test`.`t1`.`a1` AS `a1`,`test`.`t1`.`a2` AS `a2` from `test`.`t1` where <expr_cache><`test`.`t1`.`a1`,`test`.`t1`.`a2`>(<in_optimizer>((`test`.`t1`.`a1`,`test`.`t1`.`a2`),<exists>(/* select#2 */ select `test`.`t2`.`b1`,`test`.`t2`.`b2` from `test`.`t2` where (<expr_cache><`test`.`t2`.`b2`,`test`.`t1`.`a1`>(<in_optimizer>(`test`.`t2`.`b2`,<exists>(/* select#3 */ select `test`.`t3a`.`c2` from `test`.`t3` `t3a` where `test`.`t3a`.`c1` = `test`.`t1`.`a1` and <cache>(`test`.`t2`.`b2`) = `test`.`t3a`.`c2`))) or <expr_cache><`test`.`t2`.`b2`>(<in_optimizer>(`test`.`t2`.`b2`,`test`.`t2`.`b2` in ( <materialize> (/* select#4 */ select `test`.`t3b`.`c2` from `test`.`t3` `t3b` where `test`.`t3b`.`c2` like '%03' ), <primary_index_lookup>(`test`.`t2`.`b2` in <temporary table> on distinct_key where `test`.`t2`.`b2` = `<subquery4>`.`c2`))))) and <cache>(`test`.`t1`.`a1`) = `test`.`t2`.`b1` and <cache>(`test`.`t1`.`a2`) = `test`.`t2`.`b2`))) and <expr_cache><`test`.`t1`.`a1`,`test`.`t1`.`a2`>(<in_optimizer>((`test`.`t1`.`a1`,`test`.`t1`.`a2`),<exists>(/* select#5 */ select `test`.`t3c`.`c1`,`test`.`t3c`.`c2` from `test`.`t3` `t3c` where <expr_cache><`test`.`t3c`.`c1`,`test`.`t3c`.`c2`,`test`.`t1`.`a2`>(<in_optimizer>((`test`.`t3c`.`c1`,`test`.`t3c`.`c2`),<exists>(<index_lookup>(<cache>(`test`.`t3c`.`c1`) in t2i on it2i3 where (`test`.`t2i`.`b2` > '0' or `test`.`t2i`.`b2` = `test`.`t1`.`a2`) and <cache>(`test`.`t3c`.`c1`) = `test`.`t2i`.`b1` and <cache>(`test`.`t3c`.`c2`) = `test`.`t2i`.`b2`)))) and <cache>(`test`.`t1`.`a1`) = `test`.`t3c`.`c1` and <cache>(`test`.`t1`.`a2`) = `test`.`t3c`.`c2`)))
+Note	1003	/* select#1 */ select `test`.`t1`.`a1` AS `a1`,`test`.`t1`.`a2` AS `a2` from `test`.`t1` where <expr_cache><`test`.`t1`.`a1`,`test`.`t1`.`a2`>(<in_optimizer>((`test`.`t1`.`a1`,`test`.`t1`.`a2`),<exists>(/* select#2 */ select `test`.`t2`.`b1`,`test`.`t2`.`b2` from `test`.`t2` where (<expr_cache><`test`.`t2`.`b2`,`test`.`t1`.`a1`>(<in_optimizer>((`test`.`t2`.`b2`,`test`.`t1`.`a1`),(`test`.`t2`.`b2`,`test`.`t1`.`a1`) in ( <materialize> (/* select#3 */ select `test`.`t3a`.`c2`,`test`.`t3a`.`c1` from `test`.`t3` `t3a` where 1 ), <primary_index_lookup>(`test`.`t2`.`b2` in <temporary table> on distinct_key where `test`.`t2`.`b2` = `<subquery3>`.`c2` and `test`.`t1`.`a1` = `<subquery3>`.`c1`)))) or <expr_cache><`test`.`t2`.`b2`>(<in_optimizer>(`test`.`t2`.`b2`,`test`.`t2`.`b2` in ( <materialize> (/* select#4 */ select `test`.`t3b`.`c2` from `test`.`t3` `t3b` where `test`.`t3b`.`c2` like '%03' ), <primary_index_lookup>(`test`.`t2`.`b2` in <temporary table> on distinct_key where `test`.`t2`.`b2` = `<subquery4>`.`c2`))))) and <cache>(`test`.`t1`.`a1`) = `test`.`t2`.`b1` and <cache>(`test`.`t1`.`a2`) = `test`.`t2`.`b2`))) and <expr_cache><`test`.`t1`.`a1`,`test`.`t1`.`a2`>(<in_optimizer>((`test`.`t1`.`a1`,`test`.`t1`.`a2`),<exists>(/* select#5 */ select `test`.`t3c`.`c1`,`test`.`t3c`.`c2` from `test`.`t3` `t3c` where <expr_cache><`test`.`t3c`.`c1`,`test`.`t3c`.`c2`,`test`.`t1`.`a2`>(<in_optimizer>((`test`.`t3c`.`c1`,`test`.`t3c`.`c2`),<exists>(<index_lookup>(<cache>(`test`.`t3c`.`c1`) in t2i on it2i3 where (`test`.`t2i`.`b2` > '0' or `test`.`t2i`.`b2` = `test`.`t1`.`a2`) and <cache>(`test`.`t3c`.`c1`) = `test`.`t2i`.`b1` and <cache>(`test`.`t3c`.`c2`) = `test`.`t2i`.`b2`)))) and <cache>(`test`.`t1`.`a1`) = `test`.`t3c`.`c1` and <cache>(`test`.`t1`.`a2`) = `test`.`t3c`.`c2`)))
 explain extended
 select * from t1 where (a1, a2) in (select '1 - 01', '2 - 01');
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	filtered	Extra

mysql-test/main/subselect_mat_cost_bugs.result

@@ -446,8 +446,8 @@ SELECT i2 FROM t2 RIGHT JOIN t3 ON (c3 = c2) WHERE pk3 = i1
 );
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	system	NULL	NULL	NULL	NULL	1	
-2	DEPENDENT SUBQUERY	t3	const	PRIMARY	PRIMARY	4	const	1	
-2	DEPENDENT SUBQUERY	t2	index	i2	i2	11	NULL	2	Using where; Using index
+2	DEPENDENT SUBQUERY	t3	eq_ref	PRIMARY	PRIMARY	4	func	1	
+2	DEPENDENT SUBQUERY	t2	index	NULL	i2	11	NULL	2	Using where; Using index; Using join buffer (flat, BNL join)
 DROP TABLE t1,t2,t3;
 #
 # MDEV-7599: in-to-exists chosen after min/max optimization

mysql-test/main/subselect_sj2.result

@@ -466,7 +466,7 @@ id	select_type	table	type	possible_keys	key	key_len	ref	rows	filtered	Extra
 1	PRIMARY	t1	ref	a	a	5	test.t0.a	1	100.00	Using where; FirstMatch(t2)
 Warnings:
 Note	1276	Field or reference 'test.t0.a' of SELECT #2 was resolved in SELECT #1
-Note	1003	select `test`.`t0`.`a` AS `a` from `test`.`t2` semi join (`test`.`t1`) join `test`.`t0` where `test`.`t2`.`a` = `test`.`t0`.`a` and `test`.`t1`.`a` = `test`.`t0`.`a` and `test`.`t1`.`b` = `test`.`t2`.`b`
+Note	1003	select `test`.`t0`.`a` AS `a` from `test`.`t2` semi join (`test`.`t1`) join `test`.`t0` where `test`.`t1`.`b` = `test`.`t2`.`b` and `test`.`t2`.`a` = `test`.`t0`.`a` and `test`.`t1`.`a` = `test`.`t0`.`a`
 select * from t0
 where t0.a in ( select t1.a from t1,t2 where t2.a=t0.a and
 t1.b=t2.b);

mysql-test/main/subselect_sj2_jcl6.result

@@ -472,7 +472,7 @@ id	select_type	table	type	possible_keys	key	key_len	ref	rows	filtered	Extra
 1	PRIMARY	t1	ref	a	a	5	test.t0.a	1	100.00	Using where; FirstMatch(t2); Using join buffer (incremental, BKA join); Key-ordered Rowid-ordered scan
 Warnings:
 Note	1276	Field or reference 'test.t0.a' of SELECT #2 was resolved in SELECT #1
-Note	1003	select `test`.`t0`.`a` AS `a` from `test`.`t2` semi join (`test`.`t1`) join `test`.`t0` where `test`.`t2`.`a` = `test`.`t0`.`a` and `test`.`t1`.`a` = `test`.`t0`.`a` and `test`.`t1`.`b` = `test`.`t2`.`b`
+Note	1003	select `test`.`t0`.`a` AS `a` from `test`.`t2` semi join (`test`.`t1`) join `test`.`t0` where `test`.`t1`.`b` = `test`.`t2`.`b` and `test`.`t2`.`a` = `test`.`t0`.`a` and `test`.`t1`.`a` = `test`.`t0`.`a`
 select * from t0
 where t0.a in ( select t1.a from t1,t2 where t2.a=t0.a and
 t1.b=t2.b);

mysql-test/main/subselect_sj2_mat.result

@@ -468,7 +468,7 @@ id	select_type	table	type	possible_keys	key	key_len	ref	rows	filtered	Extra
 1	PRIMARY	t1	ref	a	a	5	test.t0.a	1	100.00	Using where; FirstMatch(t2)
 Warnings:
 Note	1276	Field or reference 'test.t0.a' of SELECT #2 was resolved in SELECT #1
-Note	1003	select `test`.`t0`.`a` AS `a` from `test`.`t2` semi join (`test`.`t1`) join `test`.`t0` where `test`.`t2`.`a` = `test`.`t0`.`a` and `test`.`t1`.`a` = `test`.`t0`.`a` and `test`.`t1`.`b` = `test`.`t2`.`b`
+Note	1003	select `test`.`t0`.`a` AS `a` from `test`.`t2` semi join (`test`.`t1`) join `test`.`t0` where `test`.`t1`.`b` = `test`.`t2`.`b` and `test`.`t2`.`a` = `test`.`t0`.`a` and `test`.`t1`.`a` = `test`.`t0`.`a`
 select * from t0
 where t0.a in ( select t1.a from t1,t2 where t2.a=t0.a and
 t1.b=t2.b);

mysql-test/main/update_use_source.result

@@ -504,11 +504,13 @@ test.t1	analyze	status	OK
 explain select * from t1 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain update t1 set c3=c3+110 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 update t1 set c3=c3+110 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 affected rows: 32
 info: Rows matched: 32  Changed: 32  Warnings: 0
@@ -1777,11 +1779,13 @@ test.t1	analyze	status	OK
 explain select * from t1 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain update t1 set c3=c3+110 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 update t1 set c3=c3+110 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 affected rows: 32
 info: Rows matched: 32  Changed: 32  Warnings: 0
@@ -3051,11 +3055,13 @@ test.t1	analyze	status	OK
 explain select * from t1 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain update t1 set c3=c3+110 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 update t1 set c3=c3+110 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 affected rows: 32
 info: Rows matched: 32  Changed: 32  Warnings: 0
@@ -4324,11 +4330,13 @@ test.t1	analyze	status	OK
 explain select * from t1 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain update t1 set c3=c3+110 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 update t1 set c3=c3+110 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 affected rows: 32
 info: Rows matched: 32  Changed: 32  Warnings: 0
@@ -5598,11 +5606,13 @@ test.t1	analyze	status	OK
 explain select * from t1 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain update t1 set c3=c3+110 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 update t1 set c3=c3+110 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 affected rows: 32
 info: Rows matched: 32  Changed: 32  Warnings: 0
@@ -6871,11 +6881,13 @@ test.t1	analyze	status	OK
 explain select * from t1 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain update t1 set c3=c3+110 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 update t1 set c3=c3+110 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 affected rows: 32
 info: Rows matched: 32  Changed: 32  Warnings: 0
@@ -8145,11 +8157,13 @@ test.t1	analyze	note	The storage engine for the table doesn't support analyze
 explain select * from t1 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain update t1 set c3=c3+110 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 update t1 set c3=c3+110 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 affected rows: 32
 info: Rows matched: 32  Changed: 32  Warnings: 0
@@ -9410,11 +9424,13 @@ test.t1	analyze	note	The storage engine for the table doesn't support analyze
 explain select * from t1 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1); Using join buffer (flat, BNL join)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 explain update t1 set c3=c3+110 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	PRIMARY	t1	ALL	NULL	NULL	NULL	NULL	32	
-1	PRIMARY	a	ALL	NULL	NULL	NULL	NULL	32	Using where; FirstMatch(t1)
+1	PRIMARY	<subquery2>	eq_ref	distinct_key	distinct_key	8	func,func	1	
+2	MATERIALIZED	a	ALL	NULL	NULL	NULL	NULL	32	
 update t1 set c3=c3+110 where c2 in (select distinct a.c2 from t1 a where t1.c1=a.c1);
 affected rows: 32
 info: Rows matched: 32  Changed: 32  Warnings: 0

sql/item_subselect.cc

@@ -43,6 +43,7 @@
 #include "sql_cte.h"
 #include "sql_test.h"
 #include "opt_trace.h"
+#include "opt_subselect.h"
 
 double get_post_group_estimate(JOIN* join, double join_op_rows);
 
@@ -1630,7 +1631,8 @@ Item_in_subselect::Item_in_subselect(THD *thd, Item * left_exp,
   is_jtbm_const_tab(FALSE), is_flattenable_semijoin(FALSE),
   is_registered_semijoin(FALSE),
   upper_item(0),
-  converted_from_in_predicate(FALSE)
+  converted_from_in_predicate(FALSE),
+  not_nulls_after(-1)
 {
   DBUG_ENTER("Item_in_subselect::Item_in_subselect");
   DBUG_PRINT("info", ("in_strategy: %u", (uint)in_strategy));
@@ -2705,7 +2707,10 @@ Item_in_subselect::create_row_in_to_exists_cond(JOIN * join,
                                      ref_pointer_array[i],
                                      no_matter_name,
                                      list_ref));
-      if (!is_top_level_item() && select_lex->ref_pointer_array[i]->maybe_null())
+      bool do_null_handling= (not_nulls_after == -1 || 
+                              (int)i <= not_nulls_after);
+      if (do_null_handling && !is_top_level_item() && 
+          select_lex->ref_pointer_array[i]->maybe_null())
       {
         Item *having_col_item=
           new (thd->mem_root)
@@ -2737,7 +2742,7 @@ Item_in_subselect::create_row_in_to_exists_cond(JOIN * join,
         }
         *having_item= and_items(thd, *having_item, having_col_item);
       }
-      if (!is_top_level_item() && left_expr->element_index(i)->maybe_null() &&
+      if (do_null_handling && !is_top_level_item() && left_expr->element_index(i)->maybe_null() &&
           get_cond_guard(i))
       {
         if (!(item= new (thd->mem_root)
@@ -3025,14 +3030,6 @@ static bool check_equality_for_exist2in(Item_func *func,
   return FALSE;
 }
 
-typedef struct st_eq_field_outer
-{
-  Item **eq_ref;
-  Item_ident *local_field;
-  Item *outer_exp;
-} EQ_FIELD_OUTER;
-
-
 /**
   Check if 'conds' is a set of AND-ed outer_expr=inner_table.col equalities
 
@@ -3099,6 +3096,439 @@ static bool find_inner_outer_equalities(Item **conds,
   return TRUE;
 }
 
+
+/* Check whether item tree intersects with the free list */
+static bool intersects_free_list(Item *item, THD *thd)
+{
+  for (const Item *to_find= thd->free_list; to_find; to_find= to_find->next)
+    if (item->walk(&Item::find_item_processor, 1, (void *) to_find))
+      return true;
+  return false;
+}
+
+
+/*
+  Prepare exists2in / decorrelation transformation
+
+  Pick out the equalities, prevent 2nd ps execution segfault, check for
+  correlation, etc.
+
+  @param thd                       The connection
+  @param eqs                   OUT The decorrelate-able equalities
+  @param will_be_correlated    OUT Whether the resulting IN subquery will
+                                  still be correlated
+
+  @retval FALSE        ok
+  @retval TRUE         error
+*/
+
+bool Item_exists_subselect::exists2in_prepare(
+  THD *thd, Dynamic_array<EQ_FIELD_OUTER> &eqs, bool *will_be_correlated)
+{
+  SELECT_LEX *first_select= unit->first_select();
+  DBUG_ENTER("exists2in_prepare");
+  DBUG_ASSERT(substype() == EXISTS_SUBS || substype() == IN_SUBS);
+  /* Iterate over the conditions and find decorrelatable equalities */
+  if (find_inner_outer_equalities(&first_select->join->conds, eqs))
+    DBUG_RETURN(TRUE);
+  DBUG_ASSERT(eqs.elements() != 0);
+
+  /*
+    If we are in a ps/sp execution, check for and skip on
+    intersection with the temporary free list to avoid 2nd ps execution
+    segfault
+  */
+  if (!thd->stmt_arena->is_conventional())
+  {
+    for (uint i= 0; i < (uint) eqs.elements(); i++)
+    {
+      if (intersects_free_list(*eqs.at(i).eq_ref, thd))
+        DBUG_RETURN(TRUE);
+    }
+  }
+
+  /* Determine whether the result will be correlated */
+  {
+    List<Item> unused;
+    Item::Collect_deps_prm prm= {&unused,          // parameters
+      first_select->nest_level_base,         // nest_level_base
+      0,                                     // count
+      first_select->nest_level,              // nest_level
+      FALSE                                  // collect
+    };
+    walk(&Item::collect_outer_ref_processor, TRUE, &prm);
+    DBUG_ASSERT(prm.count > 0);
+    DBUG_ASSERT(prm.count >= (uint)eqs.elements());
+    *will_be_correlated= prm.count > (uint)eqs.elements();
+    if (substype() == EXISTS_SUBS && upper_not && *will_be_correlated)
+      DBUG_RETURN(TRUE);
+  }
+
+  /*
+    Check that there are enough slots in
+    `first_select->ref_pointer_array` for the local fields 
+  */
+  if ((uint)eqs.elements() > (first_select->item_list.elements +
+                              first_select->select_n_reserved))
+    DBUG_RETURN(TRUE);
+  DBUG_RETURN(FALSE);
+}
+
+
+/*
+  @brief
+  Move the items around for the exists2in / decorrelate-in transformation
+
+  For exists2in, replace the inner select items with local fields; for
+  decorrelate-in, add them to the inner select items. For both move
+  the outer expressions to the left expr in the IN subquery.
+
+  @detail
+  Example of conversion for EXISTS:
+    EXISTS (SELECT smth FROM t WHERE subq_where AND t.col=outer_col)
+
+    ->
+
+    left_expr_ref=t.col
+    EXISTS (SELECT t.col FROM t WHERE subq_where AND "1")
+
+  Example of conversion for IN:
+
+    outer_expr IN (SELECT inner_expr 
+                   FROM t 
+                   WHERE subq_where AND t.col=outer_col)
+    ->
+    left_expr_ref=t.col
+    outer_expr IN (SELECT inner_expr, t.col -- <-- Note this
+                   FROM t WHERE subq_where AND "1")
+
+  The "1" is in quotes just to emphasize it.
+  In case of NULL-aware execution, instead of "1" we get "t.col IS NOT NULL".
+
+  @param thd               The connection
+  @param eqs               The decorrelatable equalities
+  @param left_exp_ref  OUT The left expr for the resulting IN subquery
+
+  @retval FALSE        ok
+  @retval TRUE         error
+*/
+
+bool Item_exists_subselect::exists2in_create_or_update_in(
+  THD *thd, const Dynamic_array<EQ_FIELD_OUTER> &eqs, Item** left_exp_ref)
+{
+  SELECT_LEX *first_select= unit->first_select();
+  List_iterator<Item> it;
+  List<Item> outer;
+  uint offset= 0;
+  Item_in_subselect *in_subs= NULL;
+  DBUG_ENTER("Item_exists_subselect::exists2in_create_or_update_in");
+
+  DBUG_ASSERT(substype() == EXISTS_SUBS || substype() == IN_SUBS);
+
+  /* Construct the items for left_expr */
+  if (substype() == EXISTS_SUBS)
+  {
+    /* For EXISTS, remove redundant inner select items */
+    while (first_select->item_list.elements > (uint)eqs.elements())
+    {
+      first_select->item_list.pop();
+      first_select->join->all_fields.elements--;
+    }
+    it.init(first_select->item_list);
+  }
+  else
+  {
+    /*
+      For IN, copy existing left expr items, and point the iterator
+      at the end of the inner select item list so that nothing gets
+      overwritten
+    */
+    it.init(first_select->item_list);
+    DBUG_ASSERT(substype() == IN_SUBS);
+    in_subs= (Item_in_subselect*) this;
+    offset= first_select->item_list.elements;
+    in_subs->not_nulls_after= offset - 1;
+    if (in_subs->left_expr->type() == Item::ROW_ITEM)
+    {
+      for (uint i= 0; i < in_subs->left_expr->cols(); i++, it++)
+        outer.push_back(in_subs->left_expr->element_index(i));
+    }
+    else
+    {
+      outer.push_back(in_subs->left_expr);
+      it++;
+    }
+    DBUG_ASSERT(outer.elements == first_select->item_list.elements);
+
+    List_iterator<Item> it(outer);
+    Item *item;
+    while ((item= it++))
+    {
+      item->fix_after_pullout(unit->outer_select(), it.ref(), FALSE);
+      item->update_used_tables();
+    }
+  }
+
+  /* Move items to outer and select item list */
+  for (uint i= 0; i < (uint)eqs.elements(); i++)
+  {
+    Item *item= it++;
+    Item **eq_ref= eqs.at(i).eq_ref;
+    Item_ident *local_field= eqs.at(i).local_field;
+    Item *outer_exp= eqs.at(i).outer_exp;
+    /* Replace or add items to the inner select item list */
+    if (item)
+      it.replace(local_field);
+    else
+    {
+      first_select->item_list.push_back(local_field, thd->mem_root);
+      if (substype() == EXISTS_SUBS)
+        first_select->join->all_fields.elements++;
+    }
+    first_select->ref_pointer_array[offset + i]= (Item *) local_field;
+
+    /* Replace the equalities with 1 or local_field IS NOT NULL */
+    if (!upper_not || !local_field->maybe_null())
+      *eq_ref= new (thd->mem_root) Item_int(thd, 1);
+    else
+    {
+      *eq_ref= new (thd->mem_root)
+        Item_func_isnotnull(thd,
+                            new (thd->mem_root)
+                            Item_field(thd,
+                                       ((Item_field *)
+                                        (local_field->real_item()))->context,
+                                       ((Item_field *)
+                                        (local_field->real_item()))->field));
+      if ((*eq_ref)->fix_fields(thd, (Item **) eq_ref))
+        DBUG_RETURN(TRUE);
+    }
+
+    outer_exp->fix_after_pullout(unit->outer_select(), &outer_exp, FALSE);
+    outer_exp->update_used_tables();
+    /* Add the outer_exp to the left expr list */
+    outer.push_back(outer_exp, thd->mem_root);
+  }
+
+  /* Update the left expr */
+  if (outer.elements == 1)
+    *left_exp_ref= outer.head();
+  else
+    if (!(*left_exp_ref= new (thd->mem_root) Item_row(thd, outer)))
+      DBUG_RETURN(TRUE);
+  if ((*left_exp_ref)->fix_fields_if_needed(thd, left_exp_ref))
+    DBUG_RETURN(TRUE);
+  DBUG_RETURN(FALSE);
+}
+
+
+/*
+  @brief
+  AND the IN subquery with IS NOT NULLs if upper_not
+
+  Due to the three-value logic, we need to AND the IN subquery with
+  outer_expr IS NOT NULL so that the exists2in / decorrelate-in
+  transformation is correct, i.e. new IN subquery evals to the same
+  value as the as the old EXISTS/IN subquery.
+  
+  @detail
+  Example transformation, offset=2:
+
+  NOT ( (orig_expr1, orig_expr2, added_expr1, added_expr2) IN 
+        (SELECT ... FROM ...))
+  ->
+
+  NOT (added_expr1 IS NOT NULL AND
+       added_expr2 IS NOT NULL AND
+       (orig_expr1, ..(same as before).. ) IN (SELECT ... FROM ...)
+      )
+
+  @param offset        The offset in the left expr coming from the new item
+  @param left_exp_ref  The reference pointer to the left expr
+
+  @retval FALSE        ok
+  @retval TRUE         error
+*/
+
+bool Item_exists_subselect::exists2in_and_is_not_nulls(uint offset,
+                                                       Item **left_exp_ref)
+{
+  DBUG_ENTER("Item_exists_subselect::exists2in_and_is_not_nulls");
+  DBUG_ASSERT(substype() == EXISTS_SUBS || substype() == IN_SUBS);
+  Item *left_exp= *left_exp_ref;
+
+  if (upper_not)
+  {
+    /* optimizer == 0 implies `this` is an Item_in_subselect */
+    Item *exp= optimizer ? (Item *) optimizer : (Item *) this;
+    List<Item> *and_list= new (thd->mem_root) List<Item>;
+    if (!and_list)
+      DBUG_RETURN(TRUE);
+    if (left_exp->type() != ROW_ITEM)
+    {
+      if (left_exp->maybe_null())
+      {
+        and_list->
+          push_front(new (thd->mem_root)
+                     Item_func_isnotnull(thd,
+                                         new (thd->mem_root)
+                                         Item_direct_ref(thd,
+                                                         &unit->outer_select()->context,
+                                                         left_exp_ref,
+                                                         no_matter_name,
+                                                         exists_outer_expr_name)),
+                     thd->mem_root);
+      }
+    }
+    else
+    {
+      for (uint i= offset; i < left_exp->cols(); i++)
+      {
+        if (left_exp->element_index(i)->maybe_null())
+        {
+          and_list->
+            push_front(new (thd->mem_root)
+                       Item_func_isnotnull(thd,
+                                           new (thd->mem_root)
+                                           Item_direct_ref(thd,
+                                                           &unit->outer_select()->context,
+                                                           left_exp->addr(i),
+                                                           no_matter_name,
+                                                           exists_outer_expr_name)),
+                       thd->mem_root);
+        }
+      }
+    }
+    
+    if (and_list->elements > 1)
+    {
+      and_list->push_front(exp, thd->mem_root);
+      exp= new (thd->mem_root) Item_cond_and(thd, *and_list);
+    }
+    /*
+      If we do not single out this case, and merge it with the >1
+      case, we would get an infinite loop in resolving
+      Item_direct_view_ref::real_item() like in MDEV-3881 when
+      left_exp has scalar type
+    */
+    else if (and_list->elements == 1)
+      exp= new (thd->mem_root) Item_cond_and(thd, and_list->head(), exp);
+
+    upper_not->arguments()[0]= exp;
+    if (exp->fix_fields_if_needed(thd, upper_not->arguments()))
+      DBUG_RETURN(TRUE);
+  }
+  DBUG_RETURN(FALSE);
+}
+
+
+/*
+  De-correlate where conditions in an IN subquery
+
+  Similar to `Item_exists_subselect::exists2in_processor()`, it checks
+  for equalities in the form of outer_exp = inner_field in the where
+  condition, and move the outer_exp to left_expr, and inner_field to
+  the select item list in the IN subquery
+
+  @param opt_arg      a cast to a THD*, the connection
+
+  @retval FALSE       ok
+  @retval TRUE        error
+*/
+
+bool Item_in_subselect::exists2in_processor(void *opt_arg)
+{
+  THD *thd= (THD *) opt_arg;
+  SELECT_LEX *first_select= unit->first_select();
+  JOIN *join= first_select->join;
+  bool will_be_correlated;
+  Dynamic_array<EQ_FIELD_OUTER> eqs(PSI_INSTRUMENT_MEM, 5, 5);
+  Query_arena *arena= NULL, backup;
+  int res= FALSE;
+  const uint offset= first_select->item_list.elements;
+  DBUG_ENTER("Item_in_subselect::exists2in_processor");
+
+  /*
+    Skip the transformation if it is not needed or would not help
+
+    (1) optimizer flag exists_to_in is off
+    (2) skip if the subquery is not toplevel IN nor toplevel NOT IN
+    (3) skip if the subquery is a part of a union
+    (4) skip if the subquery has group by
+    (5) skip if the subquery has aggregation in the select
+    (6) skip if the subquery has having
+    (7) skip if the subquery does not have where conditions
+    (8) skip if the left expr is a single nonscalar subselect
+  */
+  if (!optimizer_flag(thd, OPTIMIZER_SWITCH_EXISTS_TO_IN) || /* (1) */
+      !(is_top_level_item() ||
+        (upper_not && upper_not->is_top_level_item())) ||    /* (2) */
+      first_select->is_part_of_union() ||                    /* (3) */
+      first_select->group_list.elements ||                   /* (4) */
+      first_select->with_sum_func ||                         /* (5) */
+      join->having ||                                        /* (6) */
+      !join->conds ||                                        /* (7) */
+      (left_expr->type() == Item::SUBSELECT_ITEM &&
+       !left_expr->type_handler()->is_scalar_type())) /* (8) */
+    DBUG_RETURN(FALSE);
+
+  if (exists2in_prepare(thd, eqs, &will_be_correlated))
+    DBUG_RETURN(FALSE);
+  
+  /* Switch to the permanent arena if we are in a ps/sp execution */
+  arena= thd->activate_stmt_arena_if_needed(&backup);
+
+  if (exists2in_create_or_update_in(thd, eqs, &left_expr))
+  {
+    res= TRUE;
+    goto out;
+  }
+  left_expr_orig= left_expr;
+  is_correlated= will_be_correlated;
+
+  /* Update any Item_in_optimizer wrapper if exists */
+  if (optimizer)
+  {
+    optimizer->reset_cache();
+    if (optimizer->fix_left(thd))
+    {
+      res= TRUE;
+      goto out;
+    }
+  }
+
+  if (exists2in_and_is_not_nulls(offset, &left_expr))
+  {
+    res= TRUE;
+    goto out;
+  }
+
+  {
+    OPT_TRACE_TRANSFORM(thd, trace_wrapper, trace_transform,
+                        get_select_lex()->select_number, "IN (SELECT)",
+                        "decorrelation");
+    trace_transform.add("is_correlated", (is_correlated ? true : false));
+  }
+
+  /*
+    If not having materialization as a strategy and it is not a
+    semijoin thus will not have a chance of adding materialization later
+    in `check_and_do_in_subquery_rewrites()`, try to add materialization
+    again
+  */
+  if (!test_strategy(SUBS_MATERIALIZATION) && !is_registered_semijoin)
+  {
+    if (is_materialization_applicable(thd, this, first_select))
+      add_strategy(SUBS_MATERIALIZATION);
+  }
+
+out:
+  /* Switch back to the runtime arena */
+  if (arena)
+    thd->restore_active_arena(arena, &backup);
+  DBUG_RETURN(res);
+}
+
 /**
   Converts EXISTS subquery to IN subquery if it is possible and has sense
 
@@ -3112,13 +3542,9 @@ bool Item_exists_subselect::exists2in_processor(void *opt_arg)
   THD *thd= (THD *)opt_arg;
   SELECT_LEX *first_select=unit->first_select(), *save_select;
   JOIN *join= first_select->join;
-  Item **eq_ref= NULL;
-  Item_ident *local_field= NULL;
-  Item *outer_exp= NULL;
   Item *left_exp= NULL; Item_in_subselect *in_subs;
   Query_arena *arena= NULL, backup;
   int res= FALSE;
-  List<Item> outer;
   Dynamic_array<EQ_FIELD_OUTER> eqs(PSI_INSTRUMENT_MEM, 5, 5);
   bool will_be_correlated;
   DBUG_ENTER("Item_exists_subselect::exists2in_processor");
@@ -3163,99 +3589,19 @@ bool Item_exists_subselect::exists2in_processor(void *opt_arg)
   DBUG_ASSERT(first_select->group_list.elements == 0 &&
               first_select->having == NULL);
 
-  if (find_inner_outer_equalities(&join->conds, eqs))
-    DBUG_RETURN(FALSE);
-
-  DBUG_ASSERT(eqs.elements() != 0);
-
   save_select= thd->lex->current_select;
   thd->lex->current_select= first_select;
 
-  /* check that the subquery has only dependencies we are going pull out */
-  {
-    List<Item> unused;
-    Collect_deps_prm prm= {&unused,          // parameters
-      unit->first_select()->nest_level_base, // nest_level_base
-      0,                                     // count
-      unit->first_select()->nest_level,      // nest_level
-      FALSE                                  // collect
-    };
-    walk(&Item::collect_outer_ref_processor, TRUE, &prm);
-    DBUG_ASSERT(prm.count > 0);
-    DBUG_ASSERT(prm.count >= (uint)eqs.elements());
-    will_be_correlated= prm.count > (uint)eqs.elements();
-    if (upper_not && will_be_correlated)
-      goto out;
-  }
-
-  if ((uint)eqs.elements() > (first_select->item_list.elements +
-                              first_select->select_n_reserved))
-    goto out;
+  if (exists2in_prepare(thd, eqs, &will_be_correlated))
+    DBUG_RETURN(FALSE);
 
   arena= thd->activate_stmt_arena_if_needed(&backup);
 
-  while (first_select->item_list.elements > (uint)eqs.elements())
-  {
-    first_select->item_list.pop();
-    first_select->join->all_fields.elements--;
-  }
-  {
-    List_iterator<Item> it(first_select->item_list);
-
-    for (uint i= 0; i < (uint)eqs.elements(); i++)
-    {
-      Item *item= it++;
-      eq_ref= eqs.at(i).eq_ref;
-      local_field= eqs.at(i).local_field;
-      outer_exp= eqs.at(i).outer_exp;
-      /* Add the field to the SELECT_LIST */
-      if (item)
-        it.replace(local_field);
-      else
-      {
-        first_select->item_list.push_back(local_field, thd->mem_root);
-        first_select->join->all_fields.elements++;
-      }
-      first_select->ref_pointer_array[i]= (Item *)local_field;
-
-      /* remove the parts from condition */
-      if (!upper_not || !local_field->maybe_null())
-        *eq_ref= new (thd->mem_root) Item_int(thd, 1);
-      else
-      {
-        *eq_ref= new (thd->mem_root)
-          Item_func_isnotnull(thd,
-                              new (thd->mem_root)
-                              Item_field(thd,
-                                         ((Item_field*)(local_field->
-                                                        real_item()))->context,
-                                         ((Item_field*)(local_field->
-                                                        real_item()))->field));
-        if((*eq_ref)->fix_fields(thd, (Item **)eq_ref))
+  if (exists2in_create_or_update_in(thd, eqs, &left_exp))
   {
     res= TRUE;
     goto out;
   }
-      }
-      outer_exp->fix_after_pullout(unit->outer_select(), &outer_exp, FALSE);
-      outer_exp->update_used_tables();
-      outer.push_back(outer_exp, thd->mem_root);
-    }
-  }
-
-  join->conds->update_used_tables();
-
-  /* make IN SUBQUERY and put outer_exp as left part */
-  if (eqs.elements() == 1)
-    left_exp= outer_exp;
-  else
-  {
-    if (!(left_exp= new (thd->mem_root) Item_row(thd, outer)))
-    {
-      res= TRUE;
-      goto out;
-    }
-  }
 
   /* make EXISTS->IN permanet (see Item_subselect::init()) */
   set_exists_transformed();
@@ -3360,64 +3706,11 @@ bool Item_exists_subselect::exists2in_processor(void *opt_arg)
 
   first_select->fix_prepare_information(thd, &join->conds, &join->having);
 
-  if (upper_not)
-  {
-    Item *exp;
-    if (eqs.elements() == 1)
-    {
-      exp= (optimizer->arguments()[0]->maybe_null() ?
-            (Item*) new (thd->mem_root)
-            Item_cond_and(thd,
-                          new (thd->mem_root)
-                          Item_func_isnotnull(thd,
-                                              new (thd->mem_root)
-                                              Item_direct_ref(thd,
-                                                              &unit->outer_select()->context,
-                                                              optimizer->arguments(),
-                                                              no_matter_name,
-                                                              exists_outer_expr_name)),
-                          optimizer) :
-            (Item *)optimizer);
-    }
-    else
-    {
-      List<Item> *and_list= new (thd->mem_root) List<Item>;
-      if (!and_list)
+  if (exists2in_and_is_not_nulls(0, optimizer->arguments()))
   {
     res= TRUE;
     goto out;
   }
-      for (size_t i= 0; i < eqs.elements(); i++)
-      {
-        if (optimizer->arguments()[0]->maybe_null())
-        {
-          and_list->
-            push_front(new (thd->mem_root)
-                       Item_func_isnotnull(thd,
-                                           new (thd->mem_root)
-                                           Item_direct_ref(thd,
-                                                           &unit->outer_select()->context,
-                                                           optimizer->arguments()[0]->addr((int)i),
-                                                           no_matter_name,
-                                                           exists_outer_expr_name)),
-                       thd->mem_root);
-        }
-      }
-      if (and_list->elements > 0)
-      {
-        and_list->push_front(optimizer, thd->mem_root);
-        exp= new (thd->mem_root) Item_cond_and(thd, *and_list);
-      }
-      else
-        exp= optimizer;
-    }
-    upper_not->arguments()[0]= exp;
-    if (exp->fix_fields_if_needed(thd, upper_not->arguments()))
-    {
-      res= TRUE;
-      goto out;
-    }
-  }
 
 out:
   thd->lex->current_select= save_select;

sql/item_subselect.h

@@ -365,6 +365,13 @@ class Item_maxmin_subselect :public Item_singlerow_subselect
   void no_rows_in_result() override;
 };
 
+typedef struct st_eq_field_outer
+{
+  Item **eq_ref;
+  Item_ident *local_field;
+  Item *outer_exp;
+} EQ_FIELD_OUTER;
+
 /* exists subselect */
 
 class Item_exists_subselect :public Item_subselect
@@ -375,7 +382,12 @@ class Item_exists_subselect :public Item_subselect
 
   void init_length_and_dec();
   bool select_prepare_to_be_in();
-
+  bool exists2in_prepare(THD *thd, Dynamic_array<EQ_FIELD_OUTER> &eqs,
+                         bool *will_be_correlated);
+  bool exists2in_create_or_update_in(THD *thd,
+                                     const Dynamic_array<EQ_FIELD_OUTER> &eqs,
+                                     Item** left_exp_ref);
+  bool exists2in_and_is_not_nulls(uint offset, Item **left_exp_ref);
 public:
   /*
     Used by subquery optimizations to keep track about in which clause this
@@ -617,13 +629,22 @@ class Item_in_subselect :public Item_exists_subselect
   */
   bool converted_from_in_predicate;
   
+  /*
+    -1 means "do nothing"
+    Other value means that we can assume that left_expr->element_index(i) 
+    cannot be NULL (because decorrelation code has added IS NOT NULL before
+    the subquery predicate).
+  */
+  int  not_nulls_after;
+
   Item_in_subselect(THD *thd_arg, Item * left_expr, st_select_lex *select_lex);
   Item_in_subselect(THD *thd_arg):
     Item_exists_subselect(thd_arg), left_expr_cache(0), first_execution(TRUE),
     in_strategy(SUBS_NOT_TRANSFORMED),
     pushed_cond_guards(NULL), func(NULL), do_not_convert_to_sj(FALSE),
     is_jtbm_merged(FALSE), is_jtbm_const_tab(FALSE), upper_item(0),
-    converted_from_in_predicate(FALSE) {}
+    converted_from_in_predicate(FALSE),
+    not_nulls_after(-1) {}
   void cleanup() override;
   subs_type substype() override { return IN_SUBS; }
   void reset() override
@@ -757,10 +778,7 @@ class Item_in_subselect :public Item_exists_subselect
            Item_subselect::walk(processor, walk_subquery, arg);
   }
 
-  bool exists2in_processor(void *opt_arg __attribute__((unused))) override
-  {
-    return 0;
-  };
+  bool exists2in_processor(void *opt_arg) override;
 
   bool pushdown_cond_for_in_subquery(THD *thd, Item *cond);
 
@@ -804,6 +822,7 @@ class Item_allany_subselect :public Item_in_subselect
   bool is_maxmin_applicable(JOIN *join);
   bool transform_into_max_min(JOIN *join);
   void no_rows_in_result();
+  bool exists2in_processor(void *arg) override { return FALSE; }
 };
 
 

sql/opt_subselect.h

@@ -431,3 +431,6 @@ void get_delayed_table_estimates(TABLE *table,
 
 enum_nested_loop_state join_tab_execution_startup(JOIN_TAB *tab);
 
+bool is_materialization_applicable(THD *thd, Item_in_subselect *in_subs,
+                                   st_select_lex *child_select);
+