Add testcase

arenadata · Dec 27, 2024 · d0eeb6a · d0eeb6a
1 parent 72cc8b2
commit d0eeb6a
Show file tree

Hide file tree

Showing 4 changed files with 127 additions and 7 deletions.
diff --git a/src/backend/gpopt/translate/CTranslatorDXLToScalar.cpp b/src/backend/gpopt/translate/CTranslatorDXLToScalar.cpp
@@ -751,6 +751,8 @@ CTranslatorDXLToScalar::TranslateDXLScalarSubplanToScalar(
 	// Generate var mapping to handle test expression based on the required
 	// output context. Test expression may contain vars from external
 	// (relative to subplan) nodes, so add their context too.
+	// We create a copy of the external mapping because we don't want to
+	// add TestExpr params to the original.
 	CMappingColIdVarPlStmt test_expr_var_mapping = CMappingColIdVarPlStmt(
 		m_mp, m_base_table_context, outer_child_contexts,
 		&test_expr_output_ctxt, plstmt->GetDXLToPlStmtContext());
@@ -762,7 +764,6 @@ CTranslatorDXLToScalar::TranslateDXLScalarSubplanToScalar(
 		dxlop->GetDxlTestExpr(), slink, &test_expr_var_mapping);
 
 	const CDXLColRefArray *outer_refs = dxlop->GetDxlOuterColRefsArray();
-
 	const ULONG len = outer_refs->Size();
 
 	// Translate a copy of the translate context: the param mappings from the outer scope get copied in the constructor

diff --git a/src/test/regress/expected/subselect.out b/src/test/regress/expected/subselect.out
@@ -1161,9 +1161,9 @@ drop table tl3;
 drop table tl4;
 -- Check deep tree support in Test expression node when outer params are present with =ANY or IN queries
 CREATE TABLE test_tbl_t (text_t text, int_t int) DISTRIBUTED BY (int_t);
-INSERT INTO test_tbl_t VALUES ('0', 0), ('0', 1);
+INSERT INTO test_tbl_t VALUES ('0', 0);
 CREATE TABLE test_tbl_d (text_d text, int_d int) DISTRIBUTED BY (int_d);
-INSERT INTO test_tbl_d VALUES ('0', 0), ('0', 1);
+INSERT INTO test_tbl_d VALUES ('0', 0);
 CREATE TABLE test_tbl_p (text_p text, int_p int) DISTRIBUTED BY (int_p);
 INSERT INTO test_tbl_p VALUES ('0', 0), ('0', 1);
 EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM test_tbl_t WHERE
@@ -1202,6 +1202,53 @@ EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM test_tbl_t WHERE
  Settings: optimizer=off
 (30 rows)
 
+SELECT * FROM test_tbl_t WHERE
+  (int_t IN (SELECT int_d FROM test_tbl_d)) IN (SELECT int_p = 1 from test_tbl_p) ORDER BY int_t;
+ text_t | int_t 
+--------+-------
+ 0      |     0
+(1 row)
+
+INSERT INTO test_tbl_t VALUES ('0', 1);
+INSERT INTO test_tbl_d VALUES ('0', 1);
+INSERT INTO test_tbl_p VALUES ('0', 0);
+ANALYZE test_tbl_t;
+EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM test_tbl_t WHERE
+  (int_t IN (SELECT int_d FROM test_tbl_d)) IN (SELECT int_p = 1 from test_tbl_p) ORDER BY int_t;
+                                        QUERY PLAN                                         
+-------------------------------------------------------------------------------------------
+ Gather Motion 3:1  (slice4; segments: 3)
+   Output: test_tbl_t.text_t, test_tbl_t.int_t
+   Merge Key: test_tbl_t.int_t
+   ->  Sort
+         Output: test_tbl_t.text_t, test_tbl_t.int_t
+         Sort Key: test_tbl_t.int_t
+         ->  Hash Semi Join
+               Output: test_tbl_t.text_t, test_tbl_t.int_t
+               Hash Cond: (((hashed SubPlan 1)) = (test_tbl_p.int_p = 1))
+               ->  Redistribute Motion 3:3  (slice2; segments: 3)
+                     Output: test_tbl_t.text_t, test_tbl_t.int_t, ((hashed SubPlan 1))
+                     Hash Key: ((hashed SubPlan 1))
+                     ->  Seq Scan on public.test_tbl_t
+                           Output: test_tbl_t.text_t, test_tbl_t.int_t, (hashed SubPlan 1)
+                           SubPlan 1  (slice2; segments: 3)
+                             ->  Materialize
+                                   Output: test_tbl_d.int_d
+                                   ->  Broadcast Motion 3:3  (slice1; segments: 3)
+                                         Output: test_tbl_d.int_d
+                                         ->  Seq Scan on public.test_tbl_d
+                                               Output: test_tbl_d.int_d
+               ->  Hash
+                     Output: test_tbl_p.int_p
+                     ->  Redistribute Motion 3:3  (slice3; segments: 3)
+                           Output: test_tbl_p.int_p
+                           Hash Key: (test_tbl_p.int_p = 1)
+                           ->  Seq Scan on public.test_tbl_p
+                                 Output: test_tbl_p.int_p
+ Optimizer: Postgres query optimizer
+ Settings: optimizer=off
+(30 rows)
+
 SELECT * FROM test_tbl_t WHERE
   (int_t IN (SELECT int_d FROM test_tbl_d)) IN (SELECT int_p = 1 from test_tbl_p) ORDER BY int_t;
  text_t | int_t 

diff --git a/src/test/regress/expected/subselect_optimizer.out b/src/test/regress/expected/subselect_optimizer.out
@@ -1237,11 +1237,73 @@ drop table tl3;
 drop table tl4;
 -- Check deep tree support in Test expression node when outer params are present with =ANY or IN queries
 CREATE TABLE test_tbl_t (text_t text, int_t int) DISTRIBUTED BY (int_t);
-INSERT INTO test_tbl_t VALUES ('0', 0), ('0', 1);
+INSERT INTO test_tbl_t VALUES ('0', 0);
 CREATE TABLE test_tbl_d (text_d text, int_d int) DISTRIBUTED BY (int_d);
-INSERT INTO test_tbl_d VALUES ('0', 0), ('0', 1);
+INSERT INTO test_tbl_d VALUES ('0', 0);
 CREATE TABLE test_tbl_p (text_p text, int_p int) DISTRIBUTED BY (int_p);
 INSERT INTO test_tbl_p VALUES ('0', 0), ('0', 1);
+EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM test_tbl_t WHERE
+  (int_t IN (SELECT int_d FROM test_tbl_d)) IN (SELECT int_p = 1 from test_tbl_p) ORDER BY int_t;
+                                                                                                          QUERY PLAN                                                                                                           
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Result
+   Output: test_tbl_t.text_t, test_tbl_t.int_t
+   Filter: (SubPlan 1)
+   ->  Sort
+         Output: test_tbl_t.text_t, test_tbl_t.int_t
+         Sort Key: test_tbl_t.int_t
+         ->  Gather Motion 3:1  (slice1; segments: 3)
+               Output: test_tbl_t.text_t, test_tbl_t.int_t
+               ->  Seq Scan on public.test_tbl_t
+                     Output: test_tbl_t.text_t, test_tbl_t.int_t
+   SubPlan 1  (slice0)
+     ->  Result
+           Output: (count("outer".ColRef_0028)), (sum((CASE WHEN ((test_tbl_t.int_t = test_tbl_d.int_d) IS NULL) THEN 1 ELSE 0 END))), ((test_tbl_p.int_p = 1))
+           ->  GroupAggregate
+                 Output: count("outer".ColRef_0028), sum((CASE WHEN ((test_tbl_t.int_t = test_tbl_d.int_d) IS NULL) THEN 1 ELSE 0 END)), ((test_tbl_p.int_p = 1)), test_tbl_p.int_p, test_tbl_p.ctid, test_tbl_p.gp_segment_id
+                 Group Key: test_tbl_p.int_p, test_tbl_p.ctid, test_tbl_p.gp_segment_id, ((test_tbl_p.int_p = 1))
+                 ->  Sort
+                       Output: ((test_tbl_p.int_p = 1)), (CASE WHEN ((test_tbl_t.int_t = test_tbl_d.int_d) IS NULL) THEN 1 ELSE 0 END), test_tbl_p.int_p, test_tbl_p.ctid, test_tbl_p.gp_segment_id, "outer".ColRef_0028
+                       Sort Key: test_tbl_p.int_p, test_tbl_p.ctid, test_tbl_p.gp_segment_id, ((test_tbl_p.int_p = 1))
+                       ->  Result
+                             Output: (test_tbl_p.int_p = 1), CASE WHEN ((test_tbl_t.int_t = test_tbl_d.int_d) IS NULL) THEN 1 ELSE 0 END, test_tbl_p.int_p, test_tbl_p.ctid, test_tbl_p.gp_segment_id, "outer".ColRef_0028
+                             ->  Nested Loop Left Join
+                                   Output: test_tbl_p.int_p, test_tbl_p.ctid, test_tbl_p.gp_segment_id, test_tbl_d.int_d, "outer".ColRef_0028
+                                   Join Filter: true
+                                   ->  Materialize
+                                         Output: test_tbl_p.int_p, test_tbl_p.ctid, test_tbl_p.gp_segment_id
+                                         ->  Gather Motion 3:1  (slice3; segments: 3)
+                                               Output: test_tbl_p.int_p, test_tbl_p.ctid, test_tbl_p.gp_segment_id
+                                               ->  Seq Scan on public.test_tbl_p
+                                                     Output: test_tbl_p.int_p, test_tbl_p.ctid, test_tbl_p.gp_segment_id
+                                   ->  Materialize
+                                         Output: "outer".ColRef_0028, test_tbl_d.int_d
+                                         ->  Result
+                                               Output: true, test_tbl_d.int_d
+                                               ->  Result
+                                                     Output: test_tbl_d.int_d
+                                                     Filter: ((test_tbl_t.int_t = test_tbl_d.int_d) IS NOT FALSE)
+                                                     ->  Materialize
+                                                           Output: test_tbl_d.int_d
+                                                           ->  Gather Motion 3:1  (slice2; segments: 3)
+                                                                 Output: test_tbl_d.int_d
+                                                                 ->  Seq Scan on public.test_tbl_d
+                                                                       Output: test_tbl_d.int_d
+ Optimizer: Pivotal Optimizer (GPORCA)
+ Settings: optimizer=on
+(45 rows)
+
+SELECT * FROM test_tbl_t WHERE
+  (int_t IN (SELECT int_d FROM test_tbl_d)) IN (SELECT int_p = 1 from test_tbl_p) ORDER BY int_t;
+ text_t | int_t 
+--------+-------
+ 0      |     0
+(1 row)
+
+INSERT INTO test_tbl_t VALUES ('0', 1);
+INSERT INTO test_tbl_d VALUES ('0', 1);
+INSERT INTO test_tbl_p VALUES ('0', 0);
+ANALYZE test_tbl_t;
 EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM test_tbl_t WHERE
   (int_t IN (SELECT int_d FROM test_tbl_d)) IN (SELECT int_p = 1 from test_tbl_p) ORDER BY int_t;
                                        QUERY PLAN                                       

diff --git a/src/test/regress/sql/subselect.sql b/src/test/regress/sql/subselect.sql
@@ -617,14 +617,24 @@ drop table tl4;
 
 -- Check deep tree support in Test expression node when outer params are present with =ANY or IN queries
 CREATE TABLE test_tbl_t (text_t text, int_t int) DISTRIBUTED BY (int_t);
-INSERT INTO test_tbl_t VALUES ('0', 0), ('0', 1);
+INSERT INTO test_tbl_t VALUES ('0', 0);
 
 CREATE TABLE test_tbl_d (text_d text, int_d int) DISTRIBUTED BY (int_d);
-INSERT INTO test_tbl_d VALUES ('0', 0), ('0', 1);
+INSERT INTO test_tbl_d VALUES ('0', 0);
 
 CREATE TABLE test_tbl_p (text_p text, int_p int) DISTRIBUTED BY (int_p);
 INSERT INTO test_tbl_p VALUES ('0', 0), ('0', 1);
 
+EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM test_tbl_t WHERE
+  (int_t IN (SELECT int_d FROM test_tbl_d)) IN (SELECT int_p = 1 from test_tbl_p) ORDER BY int_t;
+SELECT * FROM test_tbl_t WHERE
+  (int_t IN (SELECT int_d FROM test_tbl_d)) IN (SELECT int_p = 1 from test_tbl_p) ORDER BY int_t;
+
+INSERT INTO test_tbl_t VALUES ('0', 1);
+INSERT INTO test_tbl_d VALUES ('0', 1);
+INSERT INTO test_tbl_p VALUES ('0', 0);
+ANALYZE test_tbl_t;
+
 EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM test_tbl_t WHERE
   (int_t IN (SELECT int_d FROM test_tbl_d)) IN (SELECT int_p = 1 from test_tbl_p) ORDER BY int_t;
 SELECT * FROM test_tbl_t WHERE