Further tweaking of parsetree & plantree representation of SubLinks.

Simplify SubLink by storing just a List of operator OIDs, instead of
a list of incomplete OpExprs --- that was a bizarre and bulky choice,
with no redeeming social value since we have to build new OpExprs
anyway when forming the plan tree.

src/backend/catalog/dependency.c

@@ -8,7 +8,7 @@
  * Portions Copyright (c) 1994, Regents of the University of California
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/catalog/dependency.c,v 1.18 2002/12/12 15:49:21 tgl Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/catalog/dependency.c,v 1.19 2003/01/10 21:08:07 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -838,6 +838,18 @@ find_expr_references_walker(Node *node,
 						   &context->addrs);
 		/* fall through to examine arguments */
 	}
+	if (IsA(node, SubLink))
+	{
+		SubLink	   *sublink = (SubLink *) node;
+		List	   *opid;
+
+		foreach(opid, sublink->operOids)
+		{
+			add_object_address(OCLASS_OPERATOR, (Oid) lfirsti(opid), 0,
+							   &context->addrs);
+		}
+		/* fall through to examine arguments */
+	}
 	if (is_subplan(node))
 	{
 		/* Extra work needed here if we ever need this case */

src/backend/executor/execQual.c

@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/executor/execQual.c,v 1.121 2002/12/15 16:17:45 tgl Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/executor/execQual.c,v 1.122 2003/01/10 21:08:07 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -2023,8 +2023,8 @@ ExecInitExpr(Expr *node, PlanState *parent)
 				sstate->sub_estate = NULL;
 				sstate->planstate = NULL;
 
-				sstate->oper = (List *)
+				sstate->exprs = (List *)
-					ExecInitExpr((Expr *) subplan->oper, parent);
+					ExecInitExpr((Expr *) subplan->exprs, parent);
 				sstate->args = (List *)
 					ExecInitExpr((Expr *) subplan->args, parent);
 
@@ -2156,7 +2156,7 @@ ExecInitExprInitPlan(SubPlan *node, PlanState *parent)
 	sstate->sub_estate = NULL;
 	sstate->planstate = NULL;
 
-	sstate->oper = (List *) ExecInitExpr((Expr *) node->oper, parent);
+	sstate->exprs = (List *) ExecInitExpr((Expr *) node->exprs, parent);
 	sstate->args = (List *) ExecInitExpr((Expr *) node->args, parent);
 
 	sstate->xprstate.expr = (Expr *) node;

src/backend/executor/nodeSubplan.c

@@ -7,7 +7,7 @@
  * Portions Copyright (c) 1994, Regents of the University of California
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/executor/nodeSubplan.c,v 1.41 2003/01/09 20:50:50 tgl Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/executor/nodeSubplan.c,v 1.42 2003/01/10 21:08:08 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -110,6 +110,7 @@ ExecSubPlan(SubPlanState *node,
 		Datum		rowresult = BoolGetDatum(!useOr);
 		bool		rownull = false;
 		int			col = 1;
+		List	   *plst;
 
 		if (subLinkType == EXISTS_SUBLINK)
 		{
@@ -155,45 +156,19 @@ ExecSubPlan(SubPlanState *node,
 		 * For ALL, ANY, and MULTIEXPR sublinks, iterate over combining
 		 * operators for columns of tuple.
 		 */
-		foreach(lst, node->oper)
+		plst = subplan->paramIds;
+		foreach(lst, node->exprs)
 		{
 			ExprState  *exprstate = (ExprState *) lfirst(lst);
-			OpExpr	   *expr = (OpExpr *) exprstate->expr;
+			int			paramid = lfirsti(plst);
-			Param	   *prm = lsecond(expr->args);
 			ParamExecData *prmdata;
 			Datum		expresult;
 			bool		expnull;
 
 			/*
-			 * The righthand side of the expression should be either a
+			 * Load up the Param representing this column of the sub-select.
-			 * Param or a function call or RelabelType node taking a Param
-			 * as arg (these nodes represent run-time type coercions
-			 * inserted by the parser to get to the input type needed by
-			 * the operator). Find the Param node and insert the actual
-			 * righthand-side value into the param's econtext slot.
-			 *
-			 * XXX possible improvement: could make a list of the ParamIDs
-			 * at startup time, instead of repeating this check at each row.
 			 */
-			if (!IsA(prm, Param))
+			prmdata = &(econtext->ecxt_param_exec_vals[paramid]);
-			{
-				switch (nodeTag(prm))
-				{
-					case T_FuncExpr:
-						prm = lfirst(((FuncExpr *) prm)->args);
-						break;
-					case T_RelabelType:
-						prm = (Param *) (((RelabelType *) prm)->arg);
-						break;
-					default:
-						/* will fail below */
-						break;
-				}
-				if (!IsA(prm, Param))
-					elog(ERROR, "ExecSubPlan: failed to find placeholder for subplan result");
-			}
-			Assert(prm->paramkind == PARAM_EXEC);
-			prmdata = &(econtext->ecxt_param_exec_vals[prm->paramid]);
 			Assert(prmdata->execPlan == NULL);
 			prmdata->value = heap_getattr(tup, col, tdesc,
 										  &(prmdata->isnull));
@@ -236,6 +211,8 @@ ExecSubPlan(SubPlanState *node,
 					break;		/* needn't look at any more columns */
 				}
 			}
+
+			plst = lnext(plst);
 			col++;
 		}
 
@@ -312,6 +289,8 @@ ExecInitSubPlan(SubPlanState *node, EState *estate)
 	 */
 	node->needShutdown = false;
 	node->curTuple = NULL;
+	node->hashtable = NULL;
+	node->hashnulls = NULL;
 
 	/*
 	 * create an EState for the subplan

src/backend/nodes/copyfuncs.c

@@ -15,7 +15,7 @@
  * Portions Copyright (c) 1994, Regents of the University of California
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/nodes/copyfuncs.c,v 1.234 2003/01/09 20:50:50 tgl Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/nodes/copyfuncs.c,v 1.235 2003/01/10 21:08:10 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -809,10 +809,10 @@ _copySubLink(SubLink *from)
 	SubLink    *newnode = makeNode(SubLink);
 
 	COPY_SCALAR_FIELD(subLinkType);
-	COPY_SCALAR_FIELD(operIsEquals);
 	COPY_SCALAR_FIELD(useOr);
 	COPY_NODE_FIELD(lefthand);
-	COPY_NODE_FIELD(oper);
+	COPY_NODE_FIELD(operName);
+	COPY_INTLIST_FIELD(operOids);
 	COPY_NODE_FIELD(subselect);
 
 	return newnode;
@@ -828,10 +828,13 @@ _copySubPlan(SubPlan *from)
 
 	COPY_SCALAR_FIELD(subLinkType);
 	COPY_SCALAR_FIELD(useOr);
-	COPY_NODE_FIELD(oper);
+	COPY_NODE_FIELD(exprs);
+	COPY_INTLIST_FIELD(paramIds);
 	COPY_NODE_FIELD(plan);
 	COPY_SCALAR_FIELD(plan_id);
 	COPY_NODE_FIELD(rtable);
+	COPY_SCALAR_FIELD(useHashTable);
+	COPY_SCALAR_FIELD(unknownEqFalse);
 	COPY_INTLIST_FIELD(setParam);
 	COPY_INTLIST_FIELD(parParam);
 	COPY_NODE_FIELD(args);

src/backend/nodes/equalfuncs.c

@@ -18,7 +18,7 @@
  * Portions Copyright (c) 1994, Regents of the University of California
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/nodes/equalfuncs.c,v 1.178 2003/01/09 20:50:50 tgl Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/nodes/equalfuncs.c,v 1.179 2003/01/10 21:08:10 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -287,10 +287,10 @@ static bool
 _equalSubLink(SubLink *a, SubLink *b)
 {
 	COMPARE_SCALAR_FIELD(subLinkType);
-	COMPARE_SCALAR_FIELD(operIsEquals);
 	COMPARE_SCALAR_FIELD(useOr);
 	COMPARE_NODE_FIELD(lefthand);
-	COMPARE_NODE_FIELD(oper);
+	COMPARE_NODE_FIELD(operName);
+	COMPARE_INTLIST_FIELD(operOids);
 	COMPARE_NODE_FIELD(subselect);
 
 	return true;
@@ -301,10 +301,13 @@ _equalSubPlan(SubPlan *a, SubPlan *b)
 {
 	COMPARE_SCALAR_FIELD(subLinkType);
 	COMPARE_SCALAR_FIELD(useOr);
-	COMPARE_NODE_FIELD(oper);
+	COMPARE_NODE_FIELD(exprs);
+	COMPARE_INTLIST_FIELD(paramIds);
 	/* should compare plans, but have to settle for comparing plan IDs */
 	COMPARE_SCALAR_FIELD(plan_id);
 	COMPARE_NODE_FIELD(rtable);
+	COMPARE_SCALAR_FIELD(useHashTable);
+	COMPARE_SCALAR_FIELD(unknownEqFalse);
 	COMPARE_INTLIST_FIELD(setParam);
 	COMPARE_INTLIST_FIELD(parParam);
 	COMPARE_NODE_FIELD(args);

src/backend/nodes/outfuncs.c

@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/nodes/outfuncs.c,v 1.191 2003/01/09 20:50:50 tgl Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/nodes/outfuncs.c,v 1.192 2003/01/10 21:08:11 tgl Exp $
  *
  * NOTES
  *	  Every node type that can appear in stored rules' parsetrees *must*
@@ -658,10 +658,10 @@ _outSubLink(StringInfo str, SubLink *node)
 	WRITE_NODE_TYPE("SUBLINK");
 
 	WRITE_ENUM_FIELD(subLinkType, SubLinkType);
-	WRITE_BOOL_FIELD(operIsEquals);
 	WRITE_BOOL_FIELD(useOr);
 	WRITE_NODE_FIELD(lefthand);
-	WRITE_NODE_FIELD(oper);
+	WRITE_NODE_FIELD(operName);
+	WRITE_INTLIST_FIELD(operOids);
 	WRITE_NODE_FIELD(subselect);
 }
 
@@ -672,10 +672,13 @@ _outSubPlan(StringInfo str, SubPlan *node)
 
 	WRITE_ENUM_FIELD(subLinkType, SubLinkType);
 	WRITE_BOOL_FIELD(useOr);
-	WRITE_NODE_FIELD(oper);
+	WRITE_NODE_FIELD(exprs);
+	WRITE_INTLIST_FIELD(paramIds);
 	WRITE_NODE_FIELD(plan);
 	WRITE_INT_FIELD(plan_id);
 	WRITE_NODE_FIELD(rtable);
+	WRITE_BOOL_FIELD(useHashTable);
+	WRITE_BOOL_FIELD(unknownEqFalse);
 	WRITE_INTLIST_FIELD(setParam);
 	WRITE_INTLIST_FIELD(parParam);
 	WRITE_NODE_FIELD(args);

src/backend/nodes/readfuncs.c

@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/nodes/readfuncs.c,v 1.145 2003/01/09 20:50:51 tgl Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/nodes/readfuncs.c,v 1.146 2003/01/10 21:08:11 tgl Exp $
  *
  * NOTES
  *	  Path and Plan nodes do not have any readfuncs support, because we
@@ -531,10 +531,10 @@ _readSubLink(void)
 	READ_LOCALS(SubLink);
 
 	READ_ENUM_FIELD(subLinkType, SubLinkType);
-	READ_BOOL_FIELD(operIsEquals);
 	READ_BOOL_FIELD(useOr);
 	READ_NODE_FIELD(lefthand);
-	READ_NODE_FIELD(oper);
+	READ_NODE_FIELD(operName);
+	READ_INTLIST_FIELD(operOids);
 	READ_NODE_FIELD(subselect);
 
 	READ_DONE();

src/backend/optimizer/plan/setrefs.c

@@ -9,7 +9,7 @@
  *
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/optimizer/plan/setrefs.c,v 1.86 2002/12/14 00:17:55 tgl Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/optimizer/plan/setrefs.c,v 1.87 2003/01/10 21:08:11 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -42,6 +42,7 @@ typedef struct
 
 static void fix_expr_references(Plan *plan, Node *node);
 static bool fix_expr_references_walker(Node *node, void *context);
+static void mark_qual_expressions(List *quals);
 static void set_join_references(Join *join, List *rtable);
 static void set_uppernode_references(Plan *plan, Index subvarno);
 static Node *join_references_mutator(Node *node,
@@ -88,10 +89,12 @@ set_plan_references(Plan *plan, List *rtable)
 		case T_SeqScan:
 			fix_expr_references(plan, (Node *) plan->targetlist);
 			fix_expr_references(plan, (Node *) plan->qual);
+			mark_qual_expressions(plan->qual);
 			break;
 		case T_IndexScan:
 			fix_expr_references(plan, (Node *) plan->targetlist);
 			fix_expr_references(plan, (Node *) plan->qual);
+			mark_qual_expressions(plan->qual);
 			fix_expr_references(plan,
 								(Node *) ((IndexScan *) plan)->indxqual);
 			fix_expr_references(plan,
@@ -100,6 +103,7 @@ set_plan_references(Plan *plan, List *rtable)
 		case T_TidScan:
 			fix_expr_references(plan, (Node *) plan->targetlist);
 			fix_expr_references(plan, (Node *) plan->qual);
+			mark_qual_expressions(plan->qual);
 			fix_expr_references(plan,
 								(Node *) ((TidScan *) plan)->tideval);
 			break;
@@ -114,6 +118,7 @@ set_plan_references(Plan *plan, List *rtable)
 				 */
 				fix_expr_references(plan, (Node *) plan->targetlist);
 				fix_expr_references(plan, (Node *) plan->qual);
+				mark_qual_expressions(plan->qual);
 
 				/* Recurse into subplan too */
 				rte = rt_fetch(((SubqueryScan *) plan)->scan.scanrelid,
@@ -129,6 +134,7 @@ set_plan_references(Plan *plan, List *rtable)
 
 				fix_expr_references(plan, (Node *) plan->targetlist);
 				fix_expr_references(plan, (Node *) plan->qual);
+				mark_qual_expressions(plan->qual);
 				rte = rt_fetch(((FunctionScan *) plan)->scan.scanrelid,
 							   rtable);
 				Assert(rte->rtekind == RTE_FUNCTION);
@@ -139,13 +145,17 @@ set_plan_references(Plan *plan, List *rtable)
 			set_join_references((Join *) plan, rtable);
 			fix_expr_references(plan, (Node *) plan->targetlist);
 			fix_expr_references(plan, (Node *) plan->qual);
+			mark_qual_expressions(plan->qual);
 			fix_expr_references(plan, (Node *) ((Join *) plan)->joinqual);
+			mark_qual_expressions(((Join *) plan)->joinqual);
 			break;
 		case T_MergeJoin:
 			set_join_references((Join *) plan, rtable);
 			fix_expr_references(plan, (Node *) plan->targetlist);
 			fix_expr_references(plan, (Node *) plan->qual);
+			mark_qual_expressions(plan->qual);
 			fix_expr_references(plan, (Node *) ((Join *) plan)->joinqual);
+			mark_qual_expressions(((Join *) plan)->joinqual);
 			fix_expr_references(plan,
 							(Node *) ((MergeJoin *) plan)->mergeclauses);
 			break;
@@ -153,7 +163,9 @@ set_plan_references(Plan *plan, List *rtable)
 			set_join_references((Join *) plan, rtable);
 			fix_expr_references(plan, (Node *) plan->targetlist);
 			fix_expr_references(plan, (Node *) plan->qual);
+			mark_qual_expressions(plan->qual);
 			fix_expr_references(plan, (Node *) ((Join *) plan)->joinqual);
+			mark_qual_expressions(((Join *) plan)->joinqual);
 			fix_expr_references(plan,
 							  (Node *) ((HashJoin *) plan)->hashclauses);
 			break;
@@ -180,6 +192,7 @@ set_plan_references(Plan *plan, List *rtable)
 			set_uppernode_references(plan, (Index) 0);
 			fix_expr_references(plan, (Node *) plan->targetlist);
 			fix_expr_references(plan, (Node *) plan->qual);
+			mark_qual_expressions(plan->qual);
 			break;
 		case T_Result:
 
@@ -193,7 +206,9 @@ set_plan_references(Plan *plan, List *rtable)
 				set_uppernode_references(plan, (Index) OUTER);
 			fix_expr_references(plan, (Node *) plan->targetlist);
 			fix_expr_references(plan, (Node *) plan->qual);
+			mark_qual_expressions(plan->qual);
 			fix_expr_references(plan, ((Result *) plan)->resconstantqual);
+			mark_qual_expressions((List *) ((Result *) plan)->resconstantqual);
 			break;
 		case T_Append:
 
@@ -268,6 +283,28 @@ fix_expr_references_walker(Node *node, void *context)
 	return expression_tree_walker(node, fix_expr_references_walker, context);
 }
 
+/*
+ * mark_qual_expressions
+ *	  Do final cleanup on qualifier expressions (not targetlists!)
+ *
+ * SubPlans appearing at the top level of a qual expression are marked
+ * to indicate that they need not distinguish UNKNOWN (null) from FALSE
+ * results; this can save processing time in some cases.
+ */
+static void
+mark_qual_expressions(List *quals)
+{
+	List   *qual;
+
+	foreach(qual, quals)
+	{
+		Node   *node = lfirst(qual);
+
+		if (IsA(node, SubPlan))
+			((SubPlan *) node)->unknownEqFalse = true;
+	}
+}
+
 /*
  * set_join_references
  *	  Modifies the target list of a join node to reference its subplans,

src/backend/optimizer/plan/subselect.c

@@ -7,7 +7,7 @@
  * Portions Copyright (c) 1994, Regents of the University of California
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/optimizer/plan/subselect.c,v 1.62 2003/01/09 20:50:51 tgl Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/optimizer/plan/subselect.c,v 1.63 2003/01/10 21:08:11 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -15,6 +15,7 @@
 
 #include "catalog/pg_operator.h"
 #include "catalog/pg_type.h"
+#include "miscadmin.h"
 #include "nodes/makefuncs.h"
 #include "nodes/params.h"
 #include "optimizer/clauses.h"
@@ -25,6 +26,7 @@
 #include "parser/parsetree.h"
 #include "parser/parse_expr.h"
 #include "parser/parse_oper.h"
+#include "utils/lsyscache.h"
 #include "utils/syscache.h"
 
 
@@ -59,8 +61,9 @@ typedef struct finalize_primnode_results
 } finalize_primnode_results;
 
 
-static List *convert_sublink_opers(List *operlist, List *lefthand,
+static List *convert_sublink_opers(List *lefthand, List *operOids,
-								   List *targetlist, List **setParams);
+								   List *targetlist, List **paramIds);
+static bool subplan_is_hashable(SubLink *slink, SubPlan *node);
 static Node *replace_correlation_vars_mutator(Node *node, void *context);
 static Node *process_sublinks_mutator(Node *node, void *context);
 static bool finalize_primnode(Node *node, finalize_primnode_results *results);
@@ -222,11 +225,14 @@ make_subplan(SubLink *slink, List *lefthand)
 	node->rtable = subquery->rtable;
 
 	/*
-	 * Fill in other fields of the SubPlan node.
+	 * Initialize other fields of the SubPlan node.
 	 */
 	node->subLinkType = slink->subLinkType;
 	node->useOr = slink->useOr;
-	node->oper = NIL;
+	node->exprs = NIL;
+	node->paramIds = NIL;
+	node->useHashTable = false;
+	node->unknownEqFalse = false;
 	node->setParam = NIL;
 	node->parParam = NIL;
 	node->args = NIL;
@@ -267,6 +273,7 @@ make_subplan(SubLink *slink, List *lefthand)
 		TargetEntry *te = lfirst(plan->targetlist);
 		Param	   *prm;
 
+		Assert(!te->resdom->resjunk);
 		prm = generate_new_param(te->resdom->restype, te->resdom->restypmod);
 		node->setParam = lappendi(node->setParam, prm->paramid);
 		PlannerInitPlan = lappend(PlannerInitPlan, node);
@@ -274,19 +281,25 @@ make_subplan(SubLink *slink, List *lefthand)
 	}
 	else if (node->parParam == NIL && slink->subLinkType == MULTIEXPR_SUBLINK)
 	{
-		List   *oper;
+		List   *exprs;
 
-		/* Convert the oper list, but don't put it into the SubPlan node */
+		/* Convert the lefthand exprs and oper OIDs into executable exprs */
-		oper = convert_sublink_opers(slink->oper,
+		exprs = convert_sublink_opers(lefthand,
-									 lefthand,
+									  slink->operOids,
 									  plan->targetlist,
-									 &node->setParam);
+									  &node->paramIds);
+		node->setParam = nconc(node->setParam, listCopy(node->paramIds));
 		PlannerInitPlan = lappend(PlannerInitPlan, node);
-		if (length(oper) > 1)
+		/*
-			result = (Node *) (node->useOr ? make_orclause(oper) :
+		 * The executable expressions are returned to become part of the
-							   make_andclause(oper));
+		 * outer plan's expression tree; they are not kept in the initplan
+		 * node.
+		 */
+		if (length(exprs) > 1)
+			result = (Node *) (node->useOr ? make_orclause(exprs) :
+							   make_andclause(exprs));
 		else
-			result = (Node *) lfirst(oper);
+			result = (Node *) lfirst(exprs);
 	}
 	else
 	{
@@ -296,13 +309,20 @@ make_subplan(SubLink *slink, List *lefthand)
 		 * We can't convert subplans of ALL_SUBLINK or ANY_SUBLINK types
 		 * to initPlans, even when they are uncorrelated or undirect
 		 * correlated, because we need to scan the output of the subplan
-		 * for each outer tuple.  However, we have the option to tack a
+		 * for each outer tuple.  But if it's an IN (= ANY) test, we might
-		 * MATERIAL node onto the top of an uncorrelated/undirect
+		 * be able to use a hashtable to avoid comparing all the tuples.
-		 * correlated subplan, which lets us do the work of evaluating the
+		 */
-		 * subplan only once.  We do this if the subplan's top plan node
+		if (subplan_is_hashable(slink, node))
-		 * is anything more complicated than a plain sequential scan, and
+			node->useHashTable = true;
-		 * we do it even for seqscan if the qual appears selective enough
+		/*
-		 * to eliminate many tuples.
+		 * Otherwise, we have the option to tack a MATERIAL node onto the top
+		 * of the subplan, to reduce the cost of reading it repeatedly.  This
+		 * is pointless for a direct-correlated subplan, since we'd have to
+		 * recompute its results each time anyway.  For uncorrelated/undirect
+		 * correlated subplans, we add MATERIAL if the subplan's top plan node
+		 * is anything more complicated than a plain sequential scan, and we
+		 * do it even for seqscan if the qual appears selective enough to
+		 * eliminate many tuples.
 		 *
 		 * XXX It's pretty ugly to be inserting a MATERIAL node at this
 		 * point.  Since subquery_planner has already run SS_finalize_plan
@@ -310,7 +330,7 @@ make_subplan(SubLink *slink, List *lefthand)
 		 * the MATERIAL node.  Possibly this could be fixed by postponing
 		 * SS_finalize_plan processing until setrefs.c is run.
 		 */
-		if (node->parParam == NIL)
+		else if (node->parParam == NIL)
 		{
 			bool		use_material;
 
@@ -365,11 +385,11 @@ make_subplan(SubLink *slink, List *lefthand)
 			}
 		}
 
-		/* Convert the SubLink's oper list into executable form */
+		/* Convert the lefthand exprs and oper OIDs into executable exprs */
-		node->oper = convert_sublink_opers(slink->oper,
+		node->exprs = convert_sublink_opers(lefthand,
-										   lefthand,
+											slink->operOids,
 											plan->targetlist,
-										   NULL);
+											&node->paramIds);
 
 		/*
 		 * Make node->args from parParam.
@@ -398,29 +418,26 @@ make_subplan(SubLink *slink, List *lefthand)
 }
 
 /*
- * convert_sublink_opers: convert a SubLink's oper list from the
+ * convert_sublink_opers: given a lefthand-expressions list and a list of
- * parser/rewriter format into the executor's format.
+ * operator OIDs, build a list of actually executable expressions.  The
+ * righthand sides of the expressions are Params representing the results
+ * of the sub-select.
  *
- * The oper list is initially a list of OpExpr nodes with NIL args.  We
+ * The paramids of the Params created are returned in the *paramIds list.
- * convert it to a list of actually executable expressions, in which the
- * specified operators are applied to corresponding elements of the
- * lefthand list and Params representing the results of the subplan.
- *
- * If setParams is not NULL, the paramids of the Params created are added
- * to the *setParams list.
  */
 static List *
-convert_sublink_opers(List *operlist, List *lefthand,
+convert_sublink_opers(List *lefthand, List *operOids,
-					  List *targetlist, List **setParams)
+					  List *targetlist, List **paramIds)
 {
-	List	   *newoper = NIL;
+	List	   *result = NIL;
-	List	   *leftlist = lefthand;
 	List	   *lst;
 
-	foreach(lst, operlist)
+	*paramIds = NIL;
+
+	foreach(lst, operOids)
 	{
-		OpExpr	   *oper = (OpExpr *) lfirst(lst);
+		Oid			opid = (Oid) lfirsti(lst);
-		Node	   *leftop = lfirst(leftlist);
+		Node	   *leftop = lfirst(lefthand);
 		TargetEntry *te = lfirst(targetlist);
 		Param	   *prm;
 		Operator	tup;
@@ -428,21 +445,21 @@ convert_sublink_opers(List *operlist, List *lefthand,
 		Node	   *left,
 				   *right;
 
+		Assert(!te->resdom->resjunk);
+
 		/* Make the Param node representing the subplan's result */
 		prm = generate_new_param(te->resdom->restype,
 								 te->resdom->restypmod);
 
-		/* Record its ID if needed */
+		/* Record its ID */
-		if (setParams)
+		*paramIds = lappendi(*paramIds, prm->paramid);
-			*setParams = lappendi(*setParams, prm->paramid);
 
-		/* Look up the operator to check its declared input types */
+		/* Look up the operator to get its declared input types */
-		Assert(IsA(oper, OpExpr));
 		tup = SearchSysCache(OPEROID,
-							 ObjectIdGetDatum(oper->opno),
+							 ObjectIdGetDatum(opid),
 							 0, 0, 0);
 		if (!HeapTupleIsValid(tup))
-			elog(ERROR, "cache lookup failed for operator %u", oper->opno);
+			elog(ERROR, "cache lookup failed for operator %u", opid);
 		opform = (Form_pg_operator) GETSTRUCT(tup);
 
 		/*
@@ -453,20 +470,86 @@ convert_sublink_opers(List *operlist, List *lefthand,
 		 */
 		left = make_operand(leftop, exprType(leftop), opform->oprleft);
 		right = make_operand((Node *) prm, prm->paramtype, opform->oprright);
-		newoper = lappend(newoper,
+		result = lappend(result,
-						  make_opclause(oper->opno,
+						 make_opclause(opid,
-										oper->opresulttype,
+									   opform->oprresult,
-										oper->opretset,
+									   false, /* set-result not allowed */
 									   (Expr *) left,
 									   (Expr *) right));
 
 		ReleaseSysCache(tup);
 
-		leftlist = lnext(leftlist);
+		lefthand = lnext(lefthand);
 		targetlist = lnext(targetlist);
 	}
 
-	return newoper;
+	return result;
+}
+
+/*
+ * subplan_is_hashable: decide whether we can implement a subplan by hashing
+ *
+ * Caution: the SubPlan node is not completely filled in yet.  We can rely
+ * on its plan and parParam fields, however.
+ */
+static bool
+subplan_is_hashable(SubLink *slink, SubPlan *node)
+{
+	double		subquery_size;
+	List	   *opids;
+
+	/*
+	 * The sublink type must be "= ANY" --- that is, an IN operator.
+	 * (We require the operator name to be unqualified, which may be
+	 * overly paranoid, or may not be.)  XXX since we also check that the
+	 * operators are hashable, the test on operator name may be redundant?
+	 */
+	if (slink->subLinkType != ANY_SUBLINK)
+		return false;
+	if (length(slink->operName) != 1 ||
+		strcmp(strVal(lfirst(slink->operName)), "=") != 0)
+		return false;
+	/*
+	 * The subplan must not have any direct correlation vars --- else we'd
+	 * have to recompute its output each time, so that the hashtable wouldn't
+	 * gain anything.
+	 */
+	if (node->parParam != NIL)
+		return false;
+	/*
+	 * The estimated size of the subquery result must fit in SortMem.
+	 * (XXX what about hashtable overhead?)
+	 */
+	subquery_size = node->plan->plan_rows *
+		(MAXALIGN(node->plan->plan_width) + MAXALIGN(sizeof(HeapTupleData)));
+	if (subquery_size > SortMem * 1024L)
+		return false;
+	/*
+	 * The combining operators must be hashable and strict.  (Without
+	 * strictness, behavior in the presence of nulls is too unpredictable.
+	 * We actually must assume even more than plain strictness, see
+	 * nodeSubplan.c for details.)
+	 */
+	foreach(opids, slink->operOids)
+	{
+		Oid			opid = (Oid) lfirsti(opids);
+		HeapTuple	tup;
+		Form_pg_operator optup;
+
+		tup = SearchSysCache(OPEROID,
+							 ObjectIdGetDatum(opid),
+							 0, 0, 0);
+		if (!HeapTupleIsValid(tup))
+			elog(ERROR, "cache lookup failed for operator %u", opid);
+		optup = (Form_pg_operator) GETSTRUCT(tup);
+		if (!optup->oprcanhash || !func_strict(optup->oprcode))
+		{
+			ReleaseSysCache(tup);
+			return false;
+		}
+		ReleaseSysCache(tup);
+	}
+	return true;
 }
 
 /*

src/backend/optimizer/util/clauses.c

@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/optimizer/util/clauses.c,v 1.120 2002/12/15 16:17:50 tgl Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/optimizer/util/clauses.c,v 1.121 2003/01/10 21:08:13 tgl Exp $
  *
  * HISTORY
  *	  AUTHOR			DATE			MAJOR EVENT
@@ -721,7 +721,7 @@ check_subplans_for_ungrouped_vars_walker(Node *node,
  * mistakenly think that something like "WHERE random() < 0.5" can be treated
  * as a constant qualification.
  *
- * XXX we do not examine sublinks/subplans to see if they contain uses of
+ * XXX we do not examine sub-selects to see if they contain uses of
  * mutable functions.  It's not real clear if that is correct or not...
  */
 bool
@@ -759,6 +759,18 @@ contain_mutable_functions_walker(Node *node, void *context)
 			return true;
 		/* else fall through to check args */
 	}
+	if (IsA(node, SubLink))
+	{
+		SubLink	   *sublink = (SubLink *) node;
+		List	   *opid;
+
+		foreach(opid, sublink->operOids)
+		{
+			if (op_volatile((Oid) lfirsti(opid)) != PROVOLATILE_IMMUTABLE)
+				return true;
+		}
+		/* else fall through to check args */
+	}
 	return expression_tree_walker(node, contain_mutable_functions_walker,
 								  context);
 }
@@ -776,7 +788,7 @@ contain_mutable_functions_walker(Node *node, void *context)
  * volatile function) is found. This test prevents invalid conversions
  * of volatile expressions into indexscan quals.
  *
- * XXX we do not examine sublinks/subplans to see if they contain uses of
+ * XXX we do not examine sub-selects to see if they contain uses of
  * volatile functions.	It's not real clear if that is correct or not...
  */
 bool
@@ -814,6 +826,18 @@ contain_volatile_functions_walker(Node *node, void *context)
 			return true;
 		/* else fall through to check args */
 	}
+	if (IsA(node, SubLink))
+	{
+		SubLink	   *sublink = (SubLink *) node;
+		List	   *opid;
+
+		foreach(opid, sublink->operOids)
+		{
+			if (op_volatile((Oid) lfirsti(opid)) == PROVOLATILE_VOLATILE)
+				return true;
+		}
+		/* else fall through to check args */
+	}
 	return expression_tree_walker(node, contain_volatile_functions_walker,
 								  context);
 }
@@ -830,7 +854,7 @@ contain_volatile_functions_walker(Node *node, void *context)
  * Returns true if any nonstrict construct is found --- ie, anything that
  * could produce non-NULL output with a NULL input.
  *
- * XXX we do not examine sublinks/subplans to see if they contain uses of
+ * XXX we do not examine sub-selects to see if they contain uses of
  * nonstrict functions.	It's not real clear if that is correct or not...
  * for the current usage it does not matter, since inline_function()
  * rejects cases with sublinks.
@@ -887,6 +911,18 @@ contain_nonstrict_functions_walker(Node *node, void *context)
 		return true;
 	if (IsA(node, BooleanTest))
 		return true;
+	if (IsA(node, SubLink))
+	{
+		SubLink	   *sublink = (SubLink *) node;
+		List	   *opid;
+
+		foreach(opid, sublink->operOids)
+		{
+			if (!op_strict((Oid) lfirsti(opid)))
+				return true;
+		}
+		/* else fall through to check args */
+	}
 	return expression_tree_walker(node, contain_nonstrict_functions_walker,
 								  context);
 }
@@ -2130,8 +2166,8 @@ substitute_actual_parameters_mutator(Node *node,
  * walker on all the expression subtrees of the given Query node.
  *
  * expression_tree_walker will handle SubPlan nodes by recursing normally
- * into the "oper" and "args" lists (which are expressions belonging to the
+ * into the "exprs" and "args" lists (which are expressions belonging to
- * outer plan).  It will not touch the completed subplan, however.  Since
+ * the outer plan).  It will not touch the completed subplan, however.  Since
  * there is no link to the original Query, it is not possible to recurse into
  * subselects of an already-planned expression tree.  This is OK for current
  * uses, but may need to be revisited in future.
@@ -2224,11 +2260,6 @@ expression_tree_walker(Node *node,
 			{
 				SubLink    *sublink = (SubLink *) node;
 
-				/*
-				 * We only recurse into the lefthand list (the incomplete
-				 * OpExpr nodes in the oper list are deemed uninteresting,
-				 * perhaps even confusing).
-				 */
 				if (expression_tree_walker((Node *) sublink->lefthand,
 										   walker, context))
 					return true;
@@ -2243,8 +2274,8 @@ expression_tree_walker(Node *node,
 			{
 				SubPlan *subplan = (SubPlan *) node;
 
-				/* recurse into the oper list, but not into the Plan */
+				/* recurse into the exprs list, but not into the Plan */
-				if (expression_tree_walker((Node *) subplan->oper,
+				if (expression_tree_walker((Node *) subplan->exprs,
 										   walker, context))
 					return true;
 				/* also examine args list */
@@ -2451,7 +2482,7 @@ query_tree_walker(Query *query,
  * and qualifier clauses during the planning stage.
  *
  * expression_tree_mutator will handle a SubPlan node by recursing into
- * the "oper" and "args" lists (which belong to the outer plan), but it
+ * the "exprs" and "args" lists (which belong to the outer plan), but it
  * will simply copy the link to the inner plan, since that's typically what
  * expression tree mutators want.  A mutator that wants to modify the subplan
  * can force appropriate behavior by recognizing SubPlan expression nodes
@@ -2567,8 +2598,7 @@ expression_tree_mutator(Node *node,
 		case T_SubLink:
 			{
 				/*
-				 * We transform the lefthand side, but not the oper list nor
+				 * We transform the lefthand side, but not the subquery.
-				 * the subquery.
 				 */
 				SubLink    *sublink = (SubLink *) node;
 				SubLink    *newnode;
@@ -2584,10 +2614,10 @@ expression_tree_mutator(Node *node,
 				SubPlan	   *newnode;
 
 				FLATCOPY(newnode, subplan, SubPlan);
+				/* transform exprs list */
+				MUTATE(newnode->exprs, subplan->exprs, List *);
 				/* transform args list (params to be passed to subplan) */
 				MUTATE(newnode->args, subplan->args, List *);
-				/* transform oper list as well */
-				MUTATE(newnode->oper, subplan->oper, List *);
 				/* but not the sub-Plan itself, which is referenced as-is */
 				return (Node *) newnode;
 			}

src/backend/optimizer/util/var.c

@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/optimizer/util/var.c,v 1.42 2002/12/14 00:17:59 tgl Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/optimizer/util/var.c,v 1.43 2003/01/10 21:08:13 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -107,13 +107,13 @@ pull_varnos_walker(Node *node, pull_varnos_context *context)
 	{
 		/*
 		 * Already-planned subquery.  Examine the args list (parameters to
-		 * be passed to subquery), as well as the "oper" list which is
+		 * be passed to subquery), as well as the exprs list which is
 		 * executed by the outer query.  But short-circuit recursion into
 		 * the subquery itself, which would be a waste of effort.
 		 */
 		SubPlan *subplan = (SubPlan *) node;
 
-		if (pull_varnos_walker((Node *) subplan->oper,
+		if (pull_varnos_walker((Node *) subplan->exprs,
 							   context))
 			return true;
 		if (pull_varnos_walker((Node *) subplan->args,
@@ -190,13 +190,13 @@ contain_var_reference_walker(Node *node,
 	{
 		/*
 		 * Already-planned subquery.  Examine the args list (parameters to
-		 * be passed to subquery), as well as the "oper" list which is
+		 * be passed to subquery), as well as the exprs list which is
 		 * executed by the outer query.  But short-circuit recursion into
 		 * the subquery itself, which would be a waste of effort.
 		 */
 		SubPlan *subplan = (SubPlan *) node;
 
-		if (contain_var_reference_walker((Node *) subplan->oper,
+		if (contain_var_reference_walker((Node *) subplan->exprs,
 										 context))
 			return true;
 		if (contain_var_reference_walker((Node *) subplan->args,

src/backend/parser/gram.y

@@ -11,7 +11,7 @@
  *
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/parser/gram.y,v 2.393 2003/01/10 11:02:51 petere Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/parser/gram.y,v 2.394 2003/01/10 21:08:13 tgl Exp $
  *
  * HISTORY
  *	  AUTHOR			DATE			MAJOR EVENT
@@ -5430,10 +5430,9 @@ opt_interval:
 r_expr:  row IN_P select_with_parens
 				{
 					SubLink *n = makeNode(SubLink);
-					n->lefthand = $1;
-					n->oper = (List *) makeSimpleA_Expr(OP, "=", NULL, NULL);
 					n->subLinkType = ANY_SUBLINK;
-					/* operIsEquals and useOr will be set later */
+					n->lefthand = $1;
+					n->operName = makeList1(makeString("="));
 					n->subselect = $3;
 					$$ = (Node *)n;
 				}
@@ -5441,10 +5440,9 @@ r_expr:  row IN_P select_with_parens
 				{
 					/* Make an IN node */
 					SubLink *n = makeNode(SubLink);
-					n->lefthand = $1;
-					n->oper = (List *) makeSimpleA_Expr(OP, "=", NULL, NULL);
 					n->subLinkType = ANY_SUBLINK;
-					/* operIsEquals and useOr will be set later */
+					n->lefthand = $1;
+					n->operName = makeList1(makeString("="));
 					n->subselect = $4;
 					/* Stick a NOT on top */
 					$$ = (Node *) makeA_Expr(NOT, NIL, NULL, (Node *) n);
@@ -5453,10 +5451,9 @@ r_expr:  row IN_P select_with_parens
 			%prec Op
 				{
 					SubLink *n = makeNode(SubLink);
-					n->lefthand = $1;
-					n->oper = (List *) makeA_Expr(OP, $2, NULL, NULL);
 					n->subLinkType = $3;
-					/* operIsEquals and useOr will be set later */
+					n->lefthand = $1;
+					n->operName = $2;
 					n->subselect = $4;
 					$$ = (Node *)n;
 				}
@@ -5464,10 +5461,9 @@ r_expr:  row IN_P select_with_parens
 			%prec Op
 				{
 					SubLink *n = makeNode(SubLink);
-					n->lefthand = $1;
-					n->oper = (List *) makeA_Expr(OP, $2, NULL, NULL);
 					n->subLinkType = MULTIEXPR_SUBLINK;
-					/* operIsEquals and useOr will be set later */
+					n->lefthand = $1;
+					n->operName = $2;
 					n->subselect = $3;
 					$$ = (Node *)n;
 				}
@@ -5848,10 +5844,9 @@ a_expr:		c_expr									{ $$ = $1; }
 					if (IsA($3, SubLink))
 					{
 							SubLink *n = (SubLink *)$3;
-							n->lefthand = makeList1($1);
-							n->oper = (List *) makeSimpleA_Expr(OP, "=", NULL, NULL);
 							n->subLinkType = ANY_SUBLINK;
-							/* operIsEquals and useOr will be set later */
+							n->lefthand = makeList1($1);
+							n->operName = makeList1(makeString("="));
 							$$ = (Node *)n;
 					}
 					else
@@ -5877,10 +5872,9 @@ a_expr:		c_expr									{ $$ = $1; }
 					{
 						/* Make an IN node */
 						SubLink *n = (SubLink *)$4;
-						n->lefthand = makeList1($1);
-						n->oper = (List *) makeSimpleA_Expr(OP, "=", NULL, NULL);
 						n->subLinkType = ANY_SUBLINK;
-						/* operIsEquals and useOr will be set later */
+						n->lefthand = makeList1($1);
+						n->operName = makeList1(makeString("="));
 						/* Stick a NOT on top */
 						$$ = (Node *) makeA_Expr(NOT, NIL, NULL, (Node *) n);
 					}
@@ -5903,10 +5897,9 @@ a_expr:		c_expr									{ $$ = $1; }
 			| a_expr qual_all_Op sub_type select_with_parens %prec Op
 				{
 					SubLink *n = makeNode(SubLink);
-					n->lefthand = makeList1($1);
-					n->oper = (List *) makeA_Expr(OP, $2, NULL, NULL);
 					n->subLinkType = $3;
-					/* operIsEquals and useOr will be set later */
+					n->lefthand = makeList1($1);
+					n->operName = $2;
 					n->subselect = $4;
 					$$ = (Node *)n;
 				}
@@ -6447,18 +6440,18 @@ c_expr:		columnref								{ $$ = (Node *) $1; }
 			| select_with_parens			%prec UMINUS
 				{
 					SubLink *n = makeNode(SubLink);
-					n->lefthand = NIL;
-					n->oper = NIL;
 					n->subLinkType = EXPR_SUBLINK;
+					n->lefthand = NIL;
+					n->operName = NIL;
 					n->subselect = $1;
 					$$ = (Node *)n;
 				}
 			| EXISTS select_with_parens
 				{
 					SubLink *n = makeNode(SubLink);
-					n->lefthand = NIL;
-					n->oper = NIL;
 					n->subLinkType = EXISTS_SUBLINK;
+					n->lefthand = NIL;
+					n->operName = NIL;
 					n->subselect = $2;
 					$$ = (Node *)n;
 				}
@@ -6610,6 +6603,7 @@ in_expr:	select_with_parens
 				{
 					SubLink *n = makeNode(SubLink);
 					n->subselect = $1;
+					/* other fields will be filled later */
 					$$ = (Node *)n;
 				}
 			| '(' expr_list ')'						{ $$ = (Node *)$2; }

src/backend/parser/parse_expr.c

@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/parser/parse_expr.c,v 1.139 2003/01/09 20:50:52 tgl Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/parser/parse_expr.c,v 1.140 2003/01/10 21:08:15 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -366,8 +366,8 @@ transformExpr(ParseState *pstate, Node *expr)
 					 * These fields should be NIL already, but make sure.
 					 */
 					sublink->lefthand = NIL;
-					sublink->oper = NIL;
+					sublink->operName = NIL;
-					sublink->operIsEquals = FALSE;
+					sublink->operOids = NIL;
 					sublink->useOr = FALSE;
 				}
 				else if (sublink->subLinkType == EXPR_SUBLINK)
@@ -392,8 +392,8 @@ transformExpr(ParseState *pstate, Node *expr)
 					 * fields should be NIL already, but make sure.
 					 */
 					sublink->lefthand = NIL;
-					sublink->oper = NIL;
+					sublink->operName = NIL;
-					sublink->operIsEquals = FALSE;
+					sublink->operOids = NIL;
 					sublink->useOr = FALSE;
 				}
 				else
@@ -403,20 +403,14 @@ transformExpr(ParseState *pstate, Node *expr)
 					List	   *right_list = qtree->targetList;
 					int			row_length = length(left_list);
 					bool		needNot = false;
-					List	   *op;
+					List	   *op = sublink->operName;
-					char	   *opname;
+					char	   *opname = strVal(llast(op));
 					List	   *elist;
 
 					/* transform lefthand expressions */
 					foreach(elist, left_list)
 						lfirst(elist) = transformExpr(pstate, lfirst(elist));
 
-					/* get the combining-operator name */
-					Assert(IsA(sublink->oper, A_Expr));
-					op = ((A_Expr *) sublink->oper)->name;
-					opname = strVal(llast(op));
-					sublink->oper = NIL;
-
 					/*
 					 * If the expression is "<> ALL" (with unqualified opname)
 					 * then convert it to "NOT IN".  This is a hack to improve
@@ -428,15 +422,10 @@ transformExpr(ParseState *pstate, Node *expr)
 						sublink->subLinkType = ANY_SUBLINK;
 						opname = pstrdup("=");
 						op = makeList1(makeString(opname));
+						sublink->operName = op;
 						needNot = true;
 					}
 
-					/* Set operIsEquals if op is unqualified "=" */
-					if (length(op) == 1 && strcmp(opname, "=") == 0)
-						sublink->operIsEquals = TRUE;
-					else
-						sublink->operIsEquals = FALSE;
-
 					/* Set useOr if op is "<>" (possibly qualified) */
 					if (strcmp(opname, "<>") == 0)
 						sublink->useOr = TRUE;
@@ -451,19 +440,21 @@ transformExpr(ParseState *pstate, Node *expr)
 							 opname);
 
 					/*
-					 * Scan subquery's targetlist to find values that will
+					 * To build the list of combining operator OIDs, we must
+					 * scan subquery's targetlist to find values that will
 					 * be matched against lefthand values.	We need to
 					 * ignore resjunk targets, so doing the outer
 					 * iteration over right_list is easier than doing it
 					 * over left_list.
 					 */
+					sublink->operOids = NIL;
+
 					while (right_list != NIL)
 					{
 						TargetEntry *tent = (TargetEntry *) lfirst(right_list);
 						Node	   *lexpr;
 						Operator	optup;
 						Form_pg_operator opform;
-						OpExpr	   *newop;
 
 						right_list = lnext(right_list);
 						if (tent->resdom->resjunk)
@@ -496,14 +487,8 @@ transformExpr(ParseState *pstate, Node *expr)
 								 " to be used with quantified predicate subquery",
 								 opname);
 
-						newop = makeNode(OpExpr);
+						sublink->operOids = lappendi(sublink->operOids,
-						newop->opno = oprid(optup);
+													 oprid(optup));
-						newop->opfuncid = InvalidOid;
-						newop->opresulttype = opform->oprresult;
-						newop->opretset = false;
-						newop->args = NIL; /* for now */
-
-						sublink->oper = lappend(sublink->oper, newop);
 
 						ReleaseSysCache(optup);
 					}

src/backend/utils/adt/ruleutils.c

@@ -3,7 +3,7 @@
  *				back to source text
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/utils/adt/ruleutils.c,v 1.131 2003/01/09 20:50:52 tgl Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/utils/adt/ruleutils.c,v 1.132 2003/01/10 21:08:15 tgl Exp $
  *
  *	  This software is copyrighted by Jan Wieck - Hamburg.
  *
@@ -167,6 +167,7 @@ static bool tleIsArrayAssign(TargetEntry *tle);
 static char *generate_relation_name(Oid relid);
 static char *generate_function_name(Oid funcid, int nargs, Oid *argtypes);
 static char *generate_operator_name(Oid operid, Oid arg1, Oid arg2);
+static void print_operator_name(StringInfo buf, List *opname);
 static char *get_relid_attribute_name(Oid relid, AttrNumber attnum);
 
 #define only_marker(rte)  ((rte)->inh ? "" : "ONLY ")
@@ -2111,6 +2112,9 @@ get_rule_expr(Node *node, deparse_context *context,
 				 * rule deparsing, only while EXPLAINing a query
 				 * plan. For now, just punt.
 				 */
+				if (((SubPlan *) node)->useHashTable)
+					appendStringInfo(buf, "(hashed subplan)");
+				else
 					appendStringInfo(buf, "(subplan)");
 			}
 			break;
@@ -2619,7 +2623,6 @@ get_sublink_expr(SubLink *sublink, deparse_context *context)
 	Query	   *query = (Query *) (sublink->subselect);
 	List	   *l;
 	char	   *sep;
-	OpExpr	   *oper;
 	bool		need_paren;
 
 	appendStringInfoChar(buf, '(');
@@ -2647,11 +2650,10 @@ get_sublink_expr(SubLink *sublink, deparse_context *context)
 	need_paren = true;
 
 	/*
-	 * XXX we assume here that we can get away without qualifying the
+	 * XXX we regurgitate the originally given operator name, with or without
-	 * operator name.  Since the name may imply multiple physical
+	 * schema qualification.  This is not necessarily 100% right but it's
-	 * operators it's rather difficult to do otherwise --- in fact, if the
+	 * the best we can do, since the operators actually used might not all
-	 * operators are in different namespaces any attempt to qualify would
+	 * be in the same schema.
-	 * surely fail.
 	 */
 	switch (sublink->subLinkType)
 	{
@@ -2660,26 +2662,27 @@ get_sublink_expr(SubLink *sublink, deparse_context *context)
 			break;
 
 		case ANY_SUBLINK:
-			if (sublink->operIsEquals)
+			if (length(sublink->operName) == 1 &&
+				strcmp(strVal(lfirst(sublink->operName)), "=") == 0)
 			{
-				/* Represent it as IN */
+				/* Represent = ANY as IN */
 				appendStringInfo(buf, "IN ");
 			}
 			else
 			{
-				oper = (OpExpr *) lfirst(sublink->oper);
+				print_operator_name(buf, sublink->operName);
-				appendStringInfo(buf, "%s ANY ", get_opname(oper->opno));
+				appendStringInfo(buf, " ANY ");
 			}
 			break;
 
 		case ALL_SUBLINK:
-			oper = (OpExpr *) lfirst(sublink->oper);
+			print_operator_name(buf, sublink->operName);
-			appendStringInfo(buf, "%s ALL ", get_opname(oper->opno));
+			appendStringInfo(buf, " ALL ");
 			break;
 
 		case MULTIEXPR_SUBLINK:
-			oper = (OpExpr *) lfirst(sublink->oper);
+			print_operator_name(buf, sublink->operName);
-			appendStringInfo(buf, "%s ", get_opname(oper->opno));
+			appendStringInfoChar(buf, ' ');
 			break;
 
 		case EXPR_SUBLINK:
@@ -3274,6 +3277,29 @@ generate_operator_name(Oid operid, Oid arg1, Oid arg2)
 	return buf.data;
 }
 
+/*
+ * Print out a possibly-qualified operator name
+ */
+static void
+print_operator_name(StringInfo buf, List *opname)
+{
+	int		nnames = length(opname);
+
+	if (nnames == 1)
+		appendStringInfo(buf, "%s", strVal(lfirst(opname)));
+	else
+	{
+		appendStringInfo(buf, "OPERATOR(");
+		while (nnames-- > 1)
+		{
+			appendStringInfo(buf, "%s.",
+							 quote_identifier(strVal(lfirst(opname))));
+			opname = lnext(opname);
+		}
+		appendStringInfo(buf, "%s)", strVal(lfirst(opname)));
+	}
+}
+
 /*
  * get_relid_attribute_name
  *		Get an attribute name by its relations Oid and its attnum

src/include/catalog/catversion.h

@@ -37,7 +37,7 @@
  * Portions Copyright (c) 1996-2002, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $Id: catversion.h,v 1.171 2003/01/09 20:50:53 tgl Exp $
+ * $Id: catversion.h,v 1.172 2003/01/10 21:08:15 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -53,6 +53,6 @@
  */
 
 /*							yyyymmddN */
-#define CATALOG_VERSION_NO	200301091
+#define CATALOG_VERSION_NO	200301101
 
 #endif

src/include/nodes/execnodes.h

@@ -7,7 +7,7 @@
  * Portions Copyright (c) 1996-2002, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $Id: execnodes.h,v 1.88 2002/12/18 00:14:47 tgl Exp $
+ * $Id: execnodes.h,v 1.89 2003/01/10 21:08:15 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -445,15 +445,21 @@ typedef struct BoolExprState
  *		SubPlanState node
  * ----------------
  */
+/* this struct is private in nodeSubplan.c: */
+typedef struct SubPlanHashTableData *SubPlanHashTable;
+
 typedef struct SubPlanState
 {
 	ExprState	xprstate;
 	EState	   *sub_estate;		/* subselect plan has its own EState */
 	struct PlanState *planstate; /* subselect plan's state tree */
+	List	   *exprs;			/* states of combining expression(s) */
+	List	   *args;			/* states of argument expression(s) */
 	bool		needShutdown;	/* TRUE = need to shutdown subplan */
 	HeapTuple	curTuple;		/* copy of most recent tuple from subplan */
-	List	   *oper;			/* states for executable combining exprs */
+	/* these are used when hashing the subselect's output: */
-	List	   *args;			/* states of argument expression(s) */
+	SubPlanHashTable hashtable;	/* hash table for no-nulls subselect rows */
+	SubPlanHashTable hashnulls;	/* hash table for rows with null(s) */
 } SubPlanState;
 
 /* ----------------

src/include/nodes/primnodes.h

@@ -10,7 +10,7 @@
  * Portions Copyright (c) 1996-2002, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $Id: primnodes.h,v 1.76 2003/01/09 20:50:53 tgl Exp $
+ * $Id: primnodes.h,v 1.77 2003/01/10 21:08:15 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -381,25 +381,18 @@ typedef struct BoolExpr
  * it must be replaced in the expression tree by a SubPlan node during
  * planning.
  *
- * NOTE: in the raw output of gram.y, lefthand contains a list of (raw)
+ * NOTE: in the raw output of gram.y, lefthand contains a list of raw
- * expressions, and oper contains a single A_Expr (not a list!) containing
+ * expressions; useOr and operOids are not filled in yet.  Also, subselect
- * the string name of the operator, but no arguments.  Also, subselect is
+ * is a raw parsetree.  During parse analysis, the parser transforms the
- * a raw parsetree.  During parse analysis, the parser transforms the
  * lefthand expression list using normal expression transformation rules.
- * It replaces oper with a list of OpExpr nodes, one per lefthand expression.
+ * It fills operOids with the OIDs representing the specific operator(s)
- * These nodes represent the parser's resolution of exactly which operator
+ * to apply to each pair of lefthand and targetlist expressions.
- * to apply to each pair of lefthand and targetlist expressions.  However,
+ * And subselect is transformed to a Query.  This is the representation
- * we have not constructed complete Expr trees for these operations yet:
+ * seen in saved rules and in the rewriter.
- * the args fields of the OpExpr nodes are NIL.  And subselect is transformed
- * to a Query.  This is the representation seen in saved rules and in the
- * rewriter.
  *
- * In EXISTS and EXPR SubLinks, both lefthand and oper are unused and are
+ * In EXISTS and EXPR SubLinks, lefthand, operName, and operOids are unused
- * always NIL.	useOr is not significant either for these sublink types.
+ * and are always NIL.  useOr is not significant either for these sublink
- *
+ * types.
- * The operIsEquals field is TRUE when the combining operator was written as
- * "=" --- if the subLinkType is ANY_SUBLINK, this means the operation is
- * equivalent to "IN".  This case allows special optimizations to be used.
  * ----------------
  */
 typedef enum SubLinkType
@@ -412,13 +405,12 @@ typedef struct SubLink
 {
 	Expr		xpr;
 	SubLinkType subLinkType;	/* EXISTS, ALL, ANY, MULTIEXPR, EXPR */
-	bool		operIsEquals;	/* TRUE if combining operator is "=" */
 	bool		useOr;			/* TRUE to combine column results with
 								 * "OR" not "AND" */
 	List	   *lefthand;		/* list of outer-query expressions on the
 								 * left */
-	List	   *oper;			/* list of arg-less OpExpr nodes for
+	List	   *operName;		/* originally specified operator name */
-								 * combining operators */
+	List	   *operOids;		/* OIDs of actual combining operators */
 	Node	   *subselect;		/* subselect as Query* or parsetree */
 } SubLink;
 
@@ -427,15 +419,16 @@ typedef struct SubLink
  *
  * The planner replaces SubLink nodes in expression trees with SubPlan
  * nodes after it has finished planning the subquery.  SubPlan contains
- * a sub-plantree and rtable instead of a sub-Query.  Its "oper" field
+ * a sub-plantree and rtable instead of a sub-Query.
- * corresponds to the original SubLink's oper list, but has been expanded
+ *
- * into valid executable expressions representing the application of the
+ * In an ordinary subplan, "exprs" points to a list of executable expressions
- * combining operator(s) to the lefthand expressions and values from the
+ * (OpExpr trees) for the combining operators; their left-hand arguments are
- * inner targetlist.  The original lefthand expressions now appear as
+ * the original lefthand expressions, and their right-hand arguments are
- * left-hand arguments of the OpExpr nodes, while the inner targetlist items
+ * PARAM_EXEC Param nodes representing the outputs of the sub-select.
- * are represented by PARAM_EXEC Param nodes.  (Note: if the sub-select
+ * (NOTE: runtime coercion functions may be inserted as well.)  But if the
- * becomes an InitPlan rather than a SubPlan, the rebuilt oper list is
+ * sub-select becomes an initplan rather than a subplan, these executable
- * part of the outer plan tree and so is not stored in the oper field.)
+ * expressions are part of the outer plan's expression tree (and the SubPlan
+ * node itself is not).  In this case "exprs" is NIL to avoid duplication.
  *
  * The planner also derives lists of the values that need to be passed into
  * and out of the subplan.  Input values are represented as a list "args" of
@@ -444,7 +437,7 @@ typedef struct SubLink
  * The values are assigned to the global PARAM_EXEC params indexed by parParam
  * (the parParam and args lists must have the same length).  setParam is a
  * list of the PARAM_EXEC params that are computed by the sub-select, if it
- * is an initPlan.
+ * is an initplan.
  */
 typedef struct SubPlan
 {
@@ -453,8 +446,9 @@ typedef struct SubPlan
 	SubLinkType subLinkType;	/* EXISTS, ALL, ANY, MULTIEXPR, EXPR */
 	bool		useOr;			/* TRUE to combine column results with
 								 * "OR" not "AND" */
-	List	   *oper;			/* list of executable expressions for
+	/* The combining operators, transformed to executable expressions: */
-								 * combining operators (with arguments) */
+	List	   *exprs;			/* list of OpExpr expression trees */
+	List	   *paramIds;		/* IDs of Params embedded in the above */
 	/* The subselect, transformed to a Plan: */
 	struct Plan *plan;			/* subselect plan itself */
 	int			plan_id;		/* dummy thing because of we haven't equal
@@ -462,10 +456,16 @@ typedef struct SubPlan
 								 * could put *plan itself somewhere else
 								 * (TopPlan node ?)... */
 	List	   *rtable;			/* range table for subselect */
+	/* Information about execution strategy: */
+	bool		useHashTable;	/* TRUE to store subselect output in a hash
+								 * table (implies we are doing "IN") */
+	bool		unknownEqFalse;	/* TRUE if it's okay to return FALSE when
+								 * the spec result is UNKNOWN; this allows
+								 * much simpler handling of null values */
 	/* Information for passing params into and out of the subselect: */
 	/* setParam and parParam are lists of integers (param IDs) */
-	List	   *setParam;		/* non-correlated EXPR & EXISTS subqueries
+	List	   *setParam;		/* initplan subqueries have to set these
-								 * have to set some Params for paren Plan */
+								 * Params for parent plan */
 	List	   *parParam;		/* indices of input Params from parent plan */
 	List	   *args;			/* exprs to pass as parParam values */
 } SubPlan;