Add a concept of "placeholder" variables to the planner. These are variables

that represent some expression that we desire to compute below the top level
of the plan, and then let that value "bubble up" as though it were a plain
Var (ie, a column value).

The immediate application is to allow sub-selects to be flattened even when
they are below an outer join and have non-nullable output expressions.
Formerly we couldn't flatten because such an expression wouldn't properly
go to NULL when evaluated above the outer join.  Now, we wrap it in a
PlaceHolderVar and arrange for the actual evaluation to occur below the outer
join.  When the resulting Var bubbles up through the join, it will be set to
NULL if necessary, yielding the correct results.  This fixes a planner
limitation that's existed since 7.1.

In future we might want to use this mechanism to re-introduce some form of
Hellerstein's "expensive functions" optimization, ie place the evaluation of
an expensive function at the most suitable point in the plan tree.

1 files changed, 24 insertions, 10 deletions

diff --git a/src/backend/optimizer/plan/createplan.c b/src/backend/optimizer/plan/createplan.c
index f812c939794..7bddf33e5c8 100644
--- a/src/backend/optimizer/plan/createplan.c
+++ b/src/backend/optimizer/plan/createplan.c
@@ -10,7 +10,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/optimizer/plan/createplan.c,v 1.250 2008/10/07 19:27:04 tgl Exp $
+ *	  $PostgreSQL: pgsql/src/backend/optimizer/plan/createplan.c,v 1.251 2008/10/21 20:42:53 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -37,7 +37,7 @@
 
 static Plan *create_scan_plan(PlannerInfo *root, Path *best_path);
 static List *build_relation_tlist(RelOptInfo *rel);
-static bool use_physical_tlist(RelOptInfo *rel);
+static bool use_physical_tlist(PlannerInfo *root, RelOptInfo *rel);
 static void disuse_physical_tlist(Plan *plan, Path *path);
 static Plan *create_gating_plan(PlannerInfo *root, Plan *plan, List *quals);
 static Plan *create_join_plan(PlannerInfo *root, JoinPath *best_path);
@@ -212,7 +212,7 @@ create_scan_plan(PlannerInfo *root, Path *best_path)
 	 * planner.c may replace the tlist we generate here, forcing projection to
 	 * occur.)
 	 */
-	if (use_physical_tlist(rel))
+	if (use_physical_tlist(root, rel))
 	{
 		tlist = build_physical_tlist(root, rel);
 		/* if fail because of dropped cols, use regular method */
@@ -325,9 +325,9 @@ build_relation_tlist(RelOptInfo *rel)
 	foreach(v, rel->reltargetlist)
 	{
 		/* Do we really need to copy here?	Not sure */
-		Var		   *var = (Var *) copyObject(lfirst(v));
+		Node   *node = (Node *) copyObject(lfirst(v));
 
-		tlist = lappend(tlist, makeTargetEntry((Expr *) var,
+		tlist = lappend(tlist, makeTargetEntry((Expr *) node,
 											   resno,
 											   NULL,
 											   false));
@@ -342,9 +342,10 @@ build_relation_tlist(RelOptInfo *rel)
  *		rather than only those Vars actually referenced.
  */
 static bool
-use_physical_tlist(RelOptInfo *rel)
+use_physical_tlist(PlannerInfo *root, RelOptInfo *rel)
 {
 	int			i;
+	ListCell   *lc;
 
 	/*
 	 * We can do this for real relation scans, subquery scans, function scans,
@@ -365,9 +366,9 @@ use_physical_tlist(RelOptInfo *rel)
 		return false;
 
 	/*
-	 * Can't do it if any system columns or whole-row Vars are requested,
-	 * either.	(This could possibly be fixed but would take some fragile
-	 * assumptions in setrefs.c, I think.)
+	 * Can't do it if any system columns or whole-row Vars are requested.
+	 * (This could possibly be fixed but would take some fragile assumptions
+	 * in setrefs.c, I think.)
 	 */
 	for (i = rel->min_attr; i <= 0; i++)
 	{
@@ -375,6 +376,19 @@ use_physical_tlist(RelOptInfo *rel)
 			return false;
 	}
 
+	/*
+	 * Can't do it if the rel is required to emit any placeholder expressions,
+	 * either.
+	 */
+	foreach(lc, root->placeholder_list)
+	{
+		PlaceHolderInfo *phinfo = (PlaceHolderInfo *) lfirst(lc);
+
+		if (bms_nonempty_difference(phinfo->ph_needed, rel->relids) &&
+			bms_is_subset(phinfo->ph_eval_at, rel->relids))
+			return false;
+	}
+
 	return true;
 }
 
@@ -2925,7 +2939,7 @@ make_sort_from_pathkeys(PlannerInfo *root, Plan *lefttree, List *pathkeys,
 					if (em->em_is_const || em->em_is_child)
 						continue;
 					sortexpr = em->em_expr;
-					exprvars = pull_var_clause((Node *) sortexpr, false);
+					exprvars = pull_var_clause((Node *) sortexpr, true);
 					foreach(k, exprvars)
 					{
 						if (!tlist_member_ignore_relabel(lfirst(k), tlist))