refactor impl

Author: geooo109

sqlglot/optimizer/simplify.py

@@ -280,9 +280,9 @@ def _simplify_connectors(expression, left, right):
                 return exp.null()
             if always_true(left) and always_true(right):
                 return exp.true()
-            if is_true(left):
+            if always_true(left) and right.is_type(exp.DataType.Type.BOOLEAN):
                 return right
-            if is_true(right):
+            if always_true(right) and left.is_type(exp.DataType.Type.BOOLEAN):
                 return left
             return _simplify_comparison(expression, left, right)
         elif isinstance(expression, exp.Or):
@@ -294,9 +294,9 @@ def _simplify_connectors(expression, left, right):
                 or (always_false(left) and is_null(right))
             ):
                 return exp.null()
-            if is_false(left):
+            if is_false(left) and right.is_type(exp.DataType.Type.BOOLEAN):
                 return right
-            if is_false(right):
+            if is_false(right) and left.is_type(exp.DataType.Type.BOOLEAN):
                 return left
             return _simplify_comparison(expression, left, right, or_=True)
 
@@ -1130,7 +1130,7 @@ def remove_where_true(expression):
 
 
 def always_true(expression):
-    return is_true(expression) or (
+    return (isinstance(expression, exp.Boolean) and expression.this) or (
         isinstance(expression, exp.Literal) and expression.is_number and not is_zero(expression)
     )
 
@@ -1151,10 +1151,6 @@ def is_false(a: exp.Expression) -> bool:
     return type(a) is exp.Boolean and not a.this
 
 
-def is_true(a: exp.Expression):
-    return type(a) is exp.Boolean and a.this
-
-
 def is_null(a: exp.Expression) -> bool:
     return type(a) is exp.Null
 

tests/fixtures/optimizer/normalize.sql

@@ -2,13 +2,13 @@
 (A OR B) AND (B OR C) AND (E OR F);
 
 (A AND B) OR (B AND C AND D);
-(A OR C) AND (A OR D) AND B;
+(A OR C) AND (A OR D) AND B AND TRUE;
 
 (A OR B) AND (A OR C) AND (A OR D) AND (B OR C) AND (B OR D) AND B;
-(A OR C) AND (A OR D) AND B;
+(A OR C) AND (A OR D) AND B AND TRUE;
 
 (A AND E) OR (B AND C) OR (D AND (E OR F));
-(A OR B OR D) AND (A OR C OR D) AND (B OR D OR E) AND (B OR E OR F) AND (C OR D OR E) AND (C OR E OR F);
+(A OR B OR D) AND (A OR C OR D) AND (B OR D OR E) AND (B OR E OR F) AND (C OR D OR E) AND (C OR E OR F) AND TRUE;
 
 (A AND B AND C AND D AND E AND F AND G) OR (H AND I AND J AND K AND L AND M AND N) OR (O AND P AND Q);
 (A AND B AND C AND D AND E AND F AND G) OR (H AND I AND J AND K AND L AND M AND N) OR (O AND P AND Q);
@@ -32,7 +32,7 @@ A OR (B OR (C AND D));
 (A OR B OR C) AND (A OR B OR D);
 
 A OR ((((B OR C) AND (B OR D)) OR C) AND (((B OR C) AND (B OR D)) OR D));
-(A OR B OR C) AND (A OR B OR D);
+(A OR B OR C) AND (A OR B OR D) AND TRUE;
 
 (A AND B) OR (C AND D);
 (A OR C) AND (A OR D) AND (B OR C) AND (B OR D);

tests/fixtures/optimizer/optimizer.sql

@@ -534,7 +534,10 @@ SELECT
   ) AS "sort_order"
 FROM "unioned" AS "unioned"
 WHERE
-  "unioned"."source_system" = 'bamboohr' OR "unioned"."source_system" = 'workday'
+  (
+    "unioned"."source_system" = 'bamboohr' OR "unioned"."source_system" = 'workday'
+  )
+  AND TRUE
 QUALIFY
   ROW_NUMBER() OVER (
     PARTITION BY "unioned"."unique_filter_key"

tests/fixtures/optimizer/pushdown_predicates.sql

@@ -8,7 +8,7 @@ SELECT x.a FROM (SELECT * FROM x) AS x CROSS JOIN y WHERE y.a = 1 OR (x.a = 1 AN
 SELECT x.a FROM (SELECT * FROM x) AS x CROSS JOIN y WHERE (x.a = 1 AND x.b = 1) OR y.a = 1;
 
 SELECT x.a FROM (SELECT * FROM x) AS x JOIN y WHERE (x.a = y.a AND x.a = 1 AND x.b = 1) OR x.a = y.a;
-SELECT x.a FROM (SELECT * FROM x) AS x JOIN y ON x.a = y.a WHERE TRUE;
+SELECT x.a FROM (SELECT * FROM x) AS x JOIN y ON FALSE OR x.a = y.a WHERE TRUE;
 
 SELECT x.a FROM (SELECT * FROM x) AS x JOIN y WHERE (x.a = y.a AND x.a = 1 AND x.b = 1) OR x.a = y.b;
 SELECT x.a FROM (SELECT * FROM x) AS x JOIN y ON (x.a = 1 AND x.a = y.a AND x.b = 1) OR x.a = y.b WHERE (x.a = 1 AND x.a = y.a AND x.b = 1) OR x.a = y.b;

tests/fixtures/optimizer/simplify.sql

@@ -153,11 +153,17 @@ COALESCE(x, y) <> ALL (SELECT z FROM w);
 SELECT NOT (2 <> ALL (SELECT 2 UNION ALL SELECT 3));
 SELECT 2 = ANY(SELECT 2 UNION ALL SELECT 3);
 
-x AND TRUE;
-x;
+SELECT t_bool.b AND TRUE FROM t_bool;
+SELECT t_bool.b FROM t_bool;
 
-TRUE AND x;
-x;
+SELECT TRUE AND t_bool.b FROM t_bool;
+SELECT t_bool.b FROM t_bool;
+
+SELECT t_bool.b OR FALSE FROM t_bool;
+SELECT t_bool.b FROM t_bool;
+
+SELECT FALSE OR t_bool.b FROM t_bool;
+SELECT t_bool.b FROM t_bool;
 
 --------------------------------------
 -- Absorption
@@ -166,49 +172,49 @@ x;
 (A OR B) AND (C OR NOT A);
 
 A AND (A OR B);
-A;
+A AND TRUE;
 
 A AND D AND E AND (B OR A);
-A AND D AND E;
+A AND D AND E AND TRUE;
 
 D AND A AND E AND (B OR A);
-A AND D AND E;
+A AND D AND E AND TRUE;
 
 (A OR B) AND A;
-A;
+A AND TRUE;
 
 C AND D AND (A OR B) AND E AND F AND A;
-A AND C AND D AND E AND F;
+A AND C AND D AND E AND F AND TRUE;
 
 A OR (A AND B);
-A;
+A OR FALSE;
 
 (A AND B) OR A;
-A;
+A OR FALSE;
 
 A AND (NOT A OR B);
-A AND B;
+A AND (B OR FALSE);
 
 (NOT A OR B) AND A;
-A AND B;
+A AND (B OR FALSE);
 
 A OR (NOT A AND B);
-A OR B;
+A OR (B AND TRUE);
 
 (A OR C) AND ((A OR C) OR B);
-A OR C;
+(A OR C) AND TRUE;
 
 (A OR C) AND (A OR B OR C);
-A OR C;
+(A OR C) AND TRUE;
 
 A AND (B AND C) AND (D AND E);
 A AND B AND C AND D AND E;
 
 A AND (A OR B) AND (A OR B OR C);
-A;
+A AND TRUE;
 
 (A OR B) AND (A OR C) AND (A OR B OR C);
-(A OR B) AND (A OR C);
+(A OR B) AND (A OR C) AND TRUE;
 
 --------------------------------------
 -- Elimination
@@ -863,41 +869,41 @@ COALESCE(x);
 x;
 
 COALESCE(x, 1) = 2;
-NOT x IS NULL AND x = 2;
+FALSE OR (NOT x IS NULL AND x = 2);
 
 # dialect: redshift
 COALESCE(x, 1) = 2;
 COALESCE(x, 1) = 2;
 
 2 = COALESCE(x, 1);
-NOT x IS NULL AND x = 2;
+FALSE OR (NOT x IS NULL AND x = 2);
 
 COALESCE(x, 1, 1) = 1 + 1;
-NOT x IS NULL AND x = 2;
+FALSE OR (NOT x IS NULL AND x = 2);
 
 COALESCE(x, 1, 2) = 2;
-NOT x IS NULL AND x = 2;
+FALSE OR (NOT x IS NULL AND x = 2);
 
 COALESCE(x, 3) <= 2;
-NOT x IS NULL AND x <= 2;
+FALSE OR (NOT x IS NULL AND x <= 2);
 
 COALESCE(x, 1) <> 2;
-x <> 2 OR x IS NULL;
+(NOT x IS NULL AND x <> 2) OR (TRUE AND x IS NULL);
 
 COALESCE(x, 1) <= 2;
-x <= 2 OR x IS NULL;
+(NOT x IS NULL AND x <= 2) OR (TRUE AND x IS NULL);
 
 COALESCE(x, 1) = 1;
-x = 1 OR x IS NULL;
+(NOT x IS NULL AND x = 1) OR (TRUE AND x IS NULL);
 
 COALESCE(x, 1) IS NULL;
 FALSE;
 
 COALESCE(ROW() OVER (), 1) = 1;
-ROW() OVER () = 1 OR ROW() OVER () IS NULL;
+(NOT ROW() OVER () IS NULL AND ROW() OVER () = 1) OR (TRUE AND ROW() OVER () IS NULL);
 
 a AND b AND COALESCE(ROW() OVER (), 1) = 1;
-(ROW() OVER () = 1 OR ROW() OVER () IS NULL) AND a AND b;
+((NOT ROW() OVER () IS NULL AND ROW() OVER () = 1) OR (TRUE AND ROW() OVER () IS NULL)) AND a AND b;
 
 COALESCE(1, 2);
 1;

tests/fixtures/optimizer/tpc-ds/tpc-ds.sql

@@ -5774,8 +5774,11 @@ JOIN "item" AS "item"
 LEFT JOIN "_u_0" AS "_u_0"
   ON "_u_0"."i_item_id" = "item"."i_item_id"
 WHERE
-  NOT "_u_0"."i_item_id" IS NULL
-  OR SUBSTRING("customer_address"."ca_zip", 1, 5) IN ('85669', '86197', '88274', '83405', '86475', '85392', '85460', '80348', '81792')
+  (
+    NOT "_u_0"."i_item_id" IS NULL
+    OR SUBSTRING("customer_address"."ca_zip", 1, 5) IN ('85669', '86197', '88274', '83405', '86475', '85392', '85460', '80348', '81792')
+  )
+  AND TRUE
 GROUP BY
   "customer_address"."ca_zip",
   "customer_address"."ca_state"