Teach parser to transform "x IS [NOT] DISTINCT FROM NULL" to a NullTest.

Now that we've nailed down the principle that NullTest with !argisrow
is fully equivalent to SQL's IS [NOT] DISTINCT FROM NULL, let's teach
the parser about it.  This produces a slightly more compact parse tree
and is much more amenable to optimization than a DistinctExpr, since
the planner knows a good deal about NullTest and next to nothing about
DistinctExpr.

I'm not sure that there are all that many queries in the wild that could
be improved by this, but at least one source of such cases is the patch
just made to postgres_fdw to emit IS [NOT] DISTINCT FROM NULL when
IS [NOT] NULL isn't semantically correct.

No back-patch, since to the extent that this does affect planning results,
it might be considered undesirable plan destabilization.

diff --git a/src/backend/nodes/outfuncs.c b/src/backend/nodes/outfuncs.c
index e19662273794d00edca8ef77f56608f325bd987d..acaf4ea5ebcd694b266001ddc8526891b56bcddb 100644 (file)
--- a/src/backend/nodes/outfuncs.c
+++ b/src/backend/nodes/outfuncs.c
@@ -2903,6 +2903,10 @@ _outAExpr(StringInfo str, const A_Expr *node)
                        appendStringInfoString(str, " DISTINCT ");
                        WRITE_NODE_FIELD(name);
                        break;
+               case AEXPR_NOT_DISTINCT:
+                       appendStringInfoString(str, " NOT_DISTINCT ");
+                       WRITE_NODE_FIELD(name);
+                       break;
                case AEXPR_NULLIF:
                        appendStringInfoString(str, " NULLIF ");
                        WRITE_NODE_FIELD(name);

diff --git a/src/backend/parser/gram.y b/src/backend/parser/gram.y
index edf4516dacd55c0f4859eae6a00ecbb23c2c2d31..0cae44641f83db5158245f1bd53dd518679154cd 100644 (file)
--- a/src/backend/parser/gram.y
+++ b/src/backend/parser/gram.y
@@ -11839,9 +11839,7 @@ a_expr:         c_expr                                                                  { $$ = $1; }
                                }
                        | a_expr IS NOT DISTINCT FROM a_expr            %prec IS
                                {
-                                       $$ = makeNotExpr((Node *) makeSimpleA_Expr(AEXPR_DISTINCT,
-                                                                                                                          "=", $1, $6, @2),
-                                                                        @2);
+                                       $$ = (Node *) makeSimpleA_Expr(AEXPR_NOT_DISTINCT, "=", $1, $6, @2);
                                }
                        | a_expr IS OF '(' type_list ')'                        %prec IS
                                {
@@ -12025,9 +12023,7 @@ b_expr:         c_expr
                                }
                        | b_expr IS NOT DISTINCT FROM b_expr    %prec IS
                                {
-                                       $$ = makeNotExpr((Node *) makeSimpleA_Expr(AEXPR_DISTINCT,
-                                                                                                                          "=", $1, $6, @2),
-                                                                        @2);
+                                       $$ = (Node *) makeSimpleA_Expr(AEXPR_NOT_DISTINCT, "=", $1, $6, @2);
                                }
                        | b_expr IS OF '(' type_list ')'                %prec IS
                                {

diff --git a/src/backend/parser/parse_expr.c b/src/backend/parser/parse_expr.c
index 8b285165d5f4d8c37f0165cc56ff63fa72e69beb..cead21283d00a3f201515de2808993b468be49c7 100644 (file)
--- a/src/backend/parser/parse_expr.c
+++ b/src/backend/parser/parse_expr.c
@@ -124,6 +124,8 @@ static Node *make_row_distinct_op(ParseState *pstate, List *opname,
                                         RowExpr *lrow, RowExpr *rrow, int location);
 static Expr *make_distinct_op(ParseState *pstate, List *opname,
                                 Node *ltree, Node *rtree, int location);
+static Node *make_nulltest_from_distinct(ParseState *pstate,
+                                                       A_Expr *distincta, Node *arg);
 static int     operator_precedence_group(Node *node, const char **nodename);
 static void emit_precedence_warnings(ParseState *pstate,
                                                 int opgroup, const char *opname,
@@ -224,6 +226,7 @@ transformExprRecurse(ParseState *pstate, Node *expr)
                                                result = transformAExprOpAll(pstate, a);
                                                break;
                                        case AEXPR_DISTINCT:
+                                       case AEXPR_NOT_DISTINCT:
                                                result = transformAExprDistinct(pstate, a);
                                                break;
                                        case AEXPR_NULLIF:
@@ -991,12 +994,23 @@ transformAExprDistinct(ParseState *pstate, A_Expr *a)
 {
        Node       *lexpr = a->lexpr;
        Node       *rexpr = a->rexpr;
+       Node       *result;
 
        if (operator_precedence_warning)
                emit_precedence_warnings(pstate, PREC_GROUP_INFIX_IS, "IS",
                                                                 lexpr, rexpr,
                                                                 a->location);
 
+       /*
+        * If either input is an undecorated NULL literal, transform to a NullTest
+        * on the other input. That's simpler to process than a full DistinctExpr,
+        * and it avoids needing to require that the datatype have an = operator.
+        */
+       if (exprIsNullConstant(rexpr))
+               return make_nulltest_from_distinct(pstate, a, lexpr);
+       if (exprIsNullConstant(lexpr))
+               return make_nulltest_from_distinct(pstate, a, rexpr);
+
        lexpr = transformExprRecurse(pstate, lexpr);
        rexpr = transformExprRecurse(pstate, rexpr);
 
@@ -1004,20 +1018,31 @@ transformAExprDistinct(ParseState *pstate, A_Expr *a)
                rexpr && IsA(rexpr, RowExpr))
        {
                /* ROW() op ROW() is handled specially */
-               return make_row_distinct_op(pstate, a->name,
-                                                                       (RowExpr *) lexpr,
-                                                                       (RowExpr *) rexpr,
-                                                                       a->location);
+               result = make_row_distinct_op(pstate, a->name,
+                                                                         (RowExpr *) lexpr,
+                                                                         (RowExpr *) rexpr,
+                                                                         a->location);
        }
        else
        {
                /* Ordinary scalar operator */
-               return (Node *) make_distinct_op(pstate,
-                                                                                a->name,
-                                                                                lexpr,
-                                                                                rexpr,
-                                                                                a->location);
+               result = (Node *) make_distinct_op(pstate,
+                                                                                  a->name,
+                                                                                  lexpr,
+                                                                                  rexpr,
+                                                                                  a->location);
        }
+
+       /*
+        * If it's NOT DISTINCT, we first build a DistinctExpr and then stick a
+        * NOT on top.
+        */
+       if (a->kind == AEXPR_NOT_DISTINCT)
+               result = (Node *) makeBoolExpr(NOT_EXPR,
+                                                                          list_make1(result),
+                                                                          a->location);
+
+       return result;
 }
 
 static Node *
@@ -2869,6 +2894,28 @@ make_distinct_op(ParseState *pstate, List *opname, Node *ltree, Node *rtree,
        return result;
 }
 
+/*
+ * Produce a NullTest node from an IS [NOT] DISTINCT FROM NULL construct
+ *
+ * "arg" is the untransformed other argument
+ */
+static Node *
+make_nulltest_from_distinct(ParseState *pstate, A_Expr *distincta, Node *arg)
+{
+       NullTest   *nt = makeNode(NullTest);
+
+       nt->arg = (Expr *) transformExprRecurse(pstate, arg);
+       /* the argument can be any type, so don't coerce it */
+       if (distincta->kind == AEXPR_NOT_DISTINCT)
+               nt->nulltesttype = IS_NULL;
+       else
+               nt->nulltesttype = IS_NOT_NULL;
+       /* argisrow = false is correct whether or not arg is composite */
+       nt->argisrow = false;
+       nt->location = distincta->location;
+       return (Node *) nt;
+}
+
 /*
  * Identify node's group for operator precedence warnings
  *
@@ -2971,7 +3018,8 @@ operator_precedence_group(Node *node, const char **nodename)
                        *nodename = strVal(llast(aexpr->name));
                        group = PREC_GROUP_POSTFIX_OP;
                }
-               else if (aexpr->kind == AEXPR_DISTINCT)
+               else if (aexpr->kind == AEXPR_DISTINCT ||
+                                aexpr->kind == AEXPR_NOT_DISTINCT)
                {
                        *nodename = "IS";
                        group = PREC_GROUP_INFIX_IS;

diff --git a/src/include/nodes/parsenodes.h b/src/include/nodes/parsenodes.h
index 3773dd9c2ffd004915cd51d88064719da2b83e52..1481fff57de9169b72b9704e8e106e0f8aeb3680 100644 (file)
--- a/src/include/nodes/parsenodes.h
+++ b/src/include/nodes/parsenodes.h
@@ -237,6 +237,7 @@ typedef enum A_Expr_Kind
        AEXPR_OP_ANY,                           /* scalar op ANY (array) */
        AEXPR_OP_ALL,                           /* scalar op ALL (array) */
        AEXPR_DISTINCT,                         /* IS DISTINCT FROM - name must be "=" */
+       AEXPR_NOT_DISTINCT,                     /* IS NOT DISTINCT FROM - name must be "=" */
        AEXPR_NULLIF,                           /* NULLIF - name must be "=" */
        AEXPR_OF,                                       /* IS [NOT] OF - name must be "=" or "<>" */
        AEXPR_IN,                                       /* [NOT] IN - name must be "=" or "<>" */