agora inbox for [email protected]  
help / color / mirror / Atom feed
From: jesperpedersen <[email protected]>
Subject: [PATCH 1/2] Unique key
Date: Fri, 2 Aug 2019 07:52:08 -0400

Design by David Rowley.

Author: Jesper Pedersen
---
 src/backend/nodes/outfuncs.c           |  14 +++
 src/backend/nodes/print.c              |  39 +++++++
 src/backend/optimizer/path/Makefile    |   2 +-
 src/backend/optimizer/path/allpaths.c  |   8 ++
 src/backend/optimizer/path/costsize.c  |   5 +
 src/backend/optimizer/path/indxpath.c  |  41 +++++++
 src/backend/optimizer/path/pathkeys.c  |  72 ++++++++++--
 src/backend/optimizer/path/uniquekey.c | 147 +++++++++++++++++++++++++
 src/backend/optimizer/plan/planner.c   |  18 ++-
 src/backend/optimizer/util/pathnode.c  |  12 ++
 src/backend/optimizer/util/tlist.c     |   1 -
 src/include/nodes/nodes.h              |   1 +
 src/include/nodes/pathnodes.h          |  18 +++
 src/include/nodes/print.h              |   2 +-
 src/include/optimizer/pathnode.h       |   1 +
 src/include/optimizer/paths.h          |  11 ++
 16 files changed, 377 insertions(+), 15 deletions(-)
 create mode 100644 src/backend/optimizer/path/uniquekey.c

diff --git a/src/backend/nodes/outfuncs.c b/src/backend/nodes/outfuncs.c
index e6ce8e2110..9a4f3e8e4b 100644
--- a/src/backend/nodes/outfuncs.c
+++ b/src/backend/nodes/outfuncs.c
@@ -1720,6 +1720,7 @@ _outPathInfo(StringInfo str, const Path *node)
 	WRITE_FLOAT_FIELD(startup_cost, "%.2f");
 	WRITE_FLOAT_FIELD(total_cost, "%.2f");
 	WRITE_NODE_FIELD(pathkeys);
+	WRITE_NODE_FIELD(uniquekeys);
 }
 
 /*
@@ -2201,6 +2202,7 @@ _outPlannerInfo(StringInfo str, const PlannerInfo *node)
 	WRITE_NODE_FIELD(eq_classes);
 	WRITE_BOOL_FIELD(ec_merging_done);
 	WRITE_NODE_FIELD(canon_pathkeys);
+	WRITE_NODE_FIELD(canon_uniquekeys);
 	WRITE_NODE_FIELD(left_join_clauses);
 	WRITE_NODE_FIELD(right_join_clauses);
 	WRITE_NODE_FIELD(full_join_clauses);
@@ -2210,6 +2212,7 @@ _outPlannerInfo(StringInfo str, const PlannerInfo *node)
 	WRITE_NODE_FIELD(placeholder_list);
 	WRITE_NODE_FIELD(fkey_list);
 	WRITE_NODE_FIELD(query_pathkeys);
+	WRITE_NODE_FIELD(query_uniquekeys);
 	WRITE_NODE_FIELD(group_pathkeys);
 	WRITE_NODE_FIELD(window_pathkeys);
 	WRITE_NODE_FIELD(distinct_pathkeys);
@@ -2397,6 +2400,14 @@ _outPathKey(StringInfo str, const PathKey *node)
 	WRITE_BOOL_FIELD(pk_nulls_first);
 }
 
+static void
+_outUniqueKey(StringInfo str, const UniqueKey *node)
+{
+	WRITE_NODE_TYPE("UNIQUEKEY");
+
+	WRITE_NODE_FIELD(eq_clause);
+}
+
 static void
 _outPathTarget(StringInfo str, const PathTarget *node)
 {
@@ -4073,6 +4084,9 @@ outNode(StringInfo str, const void *obj)
 			case T_PathKey:
 				_outPathKey(str, obj);
 				break;
+			case T_UniqueKey:
+				_outUniqueKey(str, obj);
+				break;
 			case T_PathTarget:
 				_outPathTarget(str, obj);
 				break;
diff --git a/src/backend/nodes/print.c b/src/backend/nodes/print.c
index 4ecde6b421..62c9d4ef7a 100644
--- a/src/backend/nodes/print.c
+++ b/src/backend/nodes/print.c
@@ -459,6 +459,45 @@ print_pathkeys(const List *pathkeys, const List *rtable)
 	printf(")\n");
 }
 
+/*
+ * print_uniquekeys -
+ *	  unique_key an UniqueKey
+ */
+void
+print_uniquekeys(const List *uniquekeys, const List *rtable)
+{
+	ListCell   *l;
+
+	printf("(");
+	foreach(l, uniquekeys)
+	{
+		UniqueKey *unique_key = (UniqueKey *) lfirst(l);
+		EquivalenceClass *eclass = (EquivalenceClass *) unique_key->eq_clause;
+		ListCell   *k;
+		bool		first = true;
+
+		/* chase up */
+		while (eclass->ec_merged)
+			eclass = eclass->ec_merged;
+
+		printf("(");
+		foreach(k, eclass->ec_members)
+		{
+			EquivalenceMember *mem = (EquivalenceMember *) lfirst(k);
+
+			if (first)
+				first = false;
+			else
+				printf(", ");
+			print_expr((Node *) mem->em_expr, rtable);
+		}
+		printf(")");
+		if (lnext(uniquekeys, l))
+			printf(", ");
+	}
+	printf(")\n");
+}
+
 /*
  * print_tl
  *	  print targetlist in a more legible way.
diff --git a/src/backend/optimizer/path/Makefile b/src/backend/optimizer/path/Makefile
index 6864a62132..8249a6b6db 100644
--- a/src/backend/optimizer/path/Makefile
+++ b/src/backend/optimizer/path/Makefile
@@ -13,6 +13,6 @@ top_builddir = ../../../..
 include $(top_builddir)/src/Makefile.global
 
 OBJS = allpaths.o clausesel.o costsize.o equivclass.o indxpath.o \
-       joinpath.o joinrels.o pathkeys.o tidpath.o
+       joinpath.o joinrels.o pathkeys.o tidpath.o uniquekey.o
 
 include $(top_srcdir)/src/backend/common.mk
diff --git a/src/backend/optimizer/path/allpaths.c b/src/backend/optimizer/path/allpaths.c
index e9ee32b7f4..6f4c25f7dd 100644
--- a/src/backend/optimizer/path/allpaths.c
+++ b/src/backend/optimizer/path/allpaths.c
@@ -3957,6 +3957,14 @@ print_path(PlannerInfo *root, Path *path, int indent)
 		print_pathkeys(path->pathkeys, root->parse->rtable);
 	}
 
+	if (path->uniquekeys)
+	{
+		for (i = 0; i < indent; i++)
+			printf("\t");
+		printf("  uniquekeys: ");
+		print_uniquekeys(path->uniquekeys, root->parse->rtable);
+	}
+
 	if (join)
 	{
 		JoinPath   *jp = (JoinPath *) path;
diff --git a/src/backend/optimizer/path/costsize.c b/src/backend/optimizer/path/costsize.c
index 3a9a994733..2565dcf296 100644
--- a/src/backend/optimizer/path/costsize.c
+++ b/src/backend/optimizer/path/costsize.c
@@ -705,6 +705,11 @@ cost_index(IndexPath *path, PlannerInfo *root, double loop_count,
 		path->path.parallel_aware = true;
 	}
 
+	/* Consider cost based on unique key */
+	if (path->path.uniquekeys)
+	{
+	}
+
 	/*
 	 * Now interpolate based on estimated index order correlation to get total
 	 * disk I/O cost for main table accesses.
diff --git a/src/backend/optimizer/path/indxpath.c b/src/backend/optimizer/path/indxpath.c
index 5f339fdfde..4b90dd378a 100644
--- a/src/backend/optimizer/path/indxpath.c
+++ b/src/backend/optimizer/path/indxpath.c
@@ -189,6 +189,7 @@ static Expr *match_clause_to_ordering_op(IndexOptInfo *index,
 static bool ec_member_matches_indexcol(PlannerInfo *root, RelOptInfo *rel,
 									   EquivalenceClass *ec, EquivalenceMember *em,
 									   void *arg);
+static List *get_uniquekeys_for_index(PlannerInfo *root, List *pathkeys);
 
 
 /*
@@ -874,6 +875,7 @@ build_index_paths(PlannerInfo *root, RelOptInfo *rel,
 	List	   *orderbyclausecols;
 	List	   *index_pathkeys;
 	List	   *useful_pathkeys;
+	List	   *useful_uniquekeys;
 	bool		found_lower_saop_clause;
 	bool		pathkeys_possibly_useful;
 	bool		index_is_ordered;
@@ -1036,11 +1038,15 @@ build_index_paths(PlannerInfo *root, RelOptInfo *rel,
 	if (index_clauses != NIL || useful_pathkeys != NIL || useful_predicate ||
 		index_only_scan)
 	{
+		if (has_useful_uniquekeys(root))
+			useful_uniquekeys = get_uniquekeys_for_index(root, useful_pathkeys);
+
 		ipath = create_index_path(root, index,
 								  index_clauses,
 								  orderbyclauses,
 								  orderbyclausecols,
 								  useful_pathkeys,
+								  useful_uniquekeys,
 								  index_is_ordered ?
 								  ForwardScanDirection :
 								  NoMovementScanDirection,
@@ -1063,6 +1069,7 @@ build_index_paths(PlannerInfo *root, RelOptInfo *rel,
 									  orderbyclauses,
 									  orderbyclausecols,
 									  useful_pathkeys,
+									  useful_uniquekeys,
 									  index_is_ordered ?
 									  ForwardScanDirection :
 									  NoMovementScanDirection,
@@ -1093,11 +1100,15 @@ build_index_paths(PlannerInfo *root, RelOptInfo *rel,
 													index_pathkeys);
 		if (useful_pathkeys != NIL)
 		{
+			if (has_useful_uniquekeys(root))
+				useful_uniquekeys = get_uniquekeys_for_index(root, useful_pathkeys);
+
 			ipath = create_index_path(root, index,
 									  index_clauses,
 									  NIL,
 									  NIL,
 									  useful_pathkeys,
+									  useful_uniquekeys,
 									  BackwardScanDirection,
 									  index_only_scan,
 									  outer_relids,
@@ -1115,6 +1126,7 @@ build_index_paths(PlannerInfo *root, RelOptInfo *rel,
 										  NIL,
 										  NIL,
 										  useful_pathkeys,
+										  useful_uniquekeys,
 										  BackwardScanDirection,
 										  index_only_scan,
 										  outer_relids,
@@ -3369,6 +3381,35 @@ match_clause_to_ordering_op(IndexOptInfo *index,
 	return clause;
 }
 
+/*
+ * get_uniquekeys_for_index
+ */
+static List *
+get_uniquekeys_for_index(PlannerInfo *root, List *pathkeys)
+{
+	ListCell *lc;
+
+	if (pathkeys)
+	{
+		List *uniquekeys = NIL;
+		foreach(lc, pathkeys)
+		{
+			UniqueKey *unique_key;
+			PathKey *pk = (PathKey *) lfirst(lc);
+			EquivalenceClass *ec = (EquivalenceClass *) pk->pk_eclass;
+
+			unique_key = makeNode(UniqueKey);
+			unique_key->eq_clause = ec;
+			
+			lappend(uniquekeys, unique_key);
+		}
+
+		if (uniquekeys_contained_in(root->canon_uniquekeys, uniquekeys))
+			return uniquekeys;
+	}
+
+	return NIL;
+}
 
 /****************************************************************************
  *				----  ROUTINES TO DO PARTIAL INDEX PREDICATE TESTS	----
diff --git a/src/backend/optimizer/path/pathkeys.c b/src/backend/optimizer/path/pathkeys.c
index 2f4fea241a..0cba366c06 100644
--- a/src/backend/optimizer/path/pathkeys.c
+++ b/src/backend/optimizer/path/pathkeys.c
@@ -29,6 +29,7 @@
 #include "utils/lsyscache.h"
 
 
+static bool pathkey_is_unique(PathKey *new_pathkey, List *pathkeys);
 static bool pathkey_is_redundant(PathKey *new_pathkey, List *pathkeys);
 static bool matches_boolean_partition_clause(RestrictInfo *rinfo,
 											 RelOptInfo *partrel,
@@ -96,6 +97,30 @@ make_canonical_pathkey(PlannerInfo *root,
 	return pk;
 }
 
+/*
+ * pathkey_is_unique
+ *	   Part of pathkey_is_redundant, that is reponsible for the case, when the
+ *	   new pathkey's equivalence class is the same as that of any existing
+ *	   member of the pathkey list.
+ */
+static bool
+pathkey_is_unique(PathKey *new_pathkey, List *pathkeys)
+{
+	EquivalenceClass *new_ec = new_pathkey->pk_eclass;
+	ListCell   *lc;
+
+	/* If same EC already is already in the list, then not unique */
+	foreach(lc, pathkeys)
+	{
+		PathKey    *old_pathkey = (PathKey *) lfirst(lc);
+
+		if (new_ec == old_pathkey->pk_eclass)
+			return false;
+	}
+
+	return true;
+}
+
 /*
  * pathkey_is_redundant
  *	   Is a pathkey redundant with one already in the given list?
@@ -135,22 +160,12 @@ static bool
 pathkey_is_redundant(PathKey *new_pathkey, List *pathkeys)
 {
 	EquivalenceClass *new_ec = new_pathkey->pk_eclass;
-	ListCell   *lc;
 
 	/* Check for EC containing a constant --- unconditionally redundant */
 	if (EC_MUST_BE_REDUNDANT(new_ec))
 		return true;
 
-	/* If same EC already used in list, then redundant */
-	foreach(lc, pathkeys)
-	{
-		PathKey    *old_pathkey = (PathKey *) lfirst(lc);
-
-		if (new_ec == old_pathkey->pk_eclass)
-			return true;
-	}
-
-	return false;
+	return !pathkey_is_unique(new_pathkey, pathkeys);
 }
 
 /*
@@ -1098,6 +1113,41 @@ make_pathkeys_for_sortclauses(PlannerInfo *root,
 	return pathkeys;
 }
 
+/*
+ * make_pathkeys_for_uniquekeyclauses
+ *		Generate a pathkeys list to be used for uniquekey clauses
+ */
+List *
+make_pathkeys_for_uniquekeys(PlannerInfo *root,
+							 List *sortclauses,
+							 List *tlist)
+{
+	List	   *pathkeys = NIL;
+	ListCell   *l;
+
+	foreach(l, sortclauses)
+	{
+		SortGroupClause *sortcl = (SortGroupClause *) lfirst(l);
+		Expr	   *sortkey;
+		PathKey    *pathkey;
+
+		sortkey = (Expr *) get_sortgroupclause_expr(sortcl, tlist);
+		Assert(OidIsValid(sortcl->sortop));
+		pathkey = make_pathkey_from_sortop(root,
+										   sortkey,
+										   root->nullable_baserels,
+										   sortcl->sortop,
+										   sortcl->nulls_first,
+										   sortcl->tleSortGroupRef,
+										   true);
+
+		if (pathkey_is_unique(pathkey, pathkeys))
+			pathkeys = lappend(pathkeys, pathkey);
+	}
+
+	return pathkeys;
+}
+
 /****************************************************************************
  *		PATHKEYS AND MERGECLAUSES
  ****************************************************************************/
diff --git a/src/backend/optimizer/path/uniquekey.c b/src/backend/optimizer/path/uniquekey.c
new file mode 100644
index 0000000000..13d4ebb98c
--- /dev/null
+++ b/src/backend/optimizer/path/uniquekey.c
@@ -0,0 +1,147 @@
+/*-------------------------------------------------------------------------
+ *
+ * uniquekey.c
+ *	  Utilities for matching and building unique keys
+ *
+ * Portions Copyright (c) 1996-2019, PostgreSQL Global Development Group
+ * Portions Copyright (c) 1994, Regents of the University of California
+ *
+ * IDENTIFICATION
+ *	  src/backend/optimizer/path/uniquekey.c
+ *
+ *-------------------------------------------------------------------------
+ */
+#include "postgres.h"
+
+#include "optimizer/pathnode.h"
+#include "optimizer/paths.h"
+#include "nodes/pg_list.h"
+
+static UniqueKey *make_canonical_uniquekey(PlannerInfo *root, EquivalenceClass *eclass);
+
+/*
+ * Build a list of unique keys
+ */
+List*
+build_uniquekeys(PlannerInfo *root, List *sortclauses)
+{
+	List *result = NIL;
+	List *sortkeys;
+	ListCell *l;
+
+	sortkeys = make_pathkeys_for_uniquekeys(root,
+											sortclauses,
+											root->processed_tlist);
+
+	/* Create a uniquekey and add it to the list */
+	foreach(l, sortkeys)
+	{
+		PathKey    *pathkey = (PathKey *) lfirst(l);
+		EquivalenceClass *ec = pathkey->pk_eclass;
+		UniqueKey *unique_key = make_canonical_uniquekey(root, ec);
+
+		result = lappend(result, unique_key);
+	}
+
+	return result;
+}
+
+/*
+ * uniquekeys_contained_in
+ *	  Are the keys2 included in the keys1 superset
+ */
+bool
+uniquekeys_contained_in(List *keys1, List *keys2)
+{
+	ListCell   *key1,
+			   *key2;
+
+	/*
+	 * Fall out quickly if we are passed two identical lists.  This mostly
+	 * catches the case where both are NIL, but that's common enough to
+	 * warrant the test.
+	 */
+	if (keys1 == keys2)
+		return true;
+
+	foreach(key2, keys2)
+	{
+		bool found = false;
+		UniqueKey  *uniquekey2 = (UniqueKey *) lfirst(key2);
+
+		foreach(key1, keys1)
+		{
+			UniqueKey  *uniquekey1 = (UniqueKey *) lfirst(key1);
+
+			if (uniquekey1->eq_clause == uniquekey2->eq_clause)
+			{
+				found = true;
+				break;
+			}
+		}
+
+		if (!found)
+			return false;
+	}
+
+	return true;
+}
+
+/*
+ * has_useful_uniquekeys
+ *		Detect whether the planner could have any uniquekeys that are
+ *		useful.
+ */
+bool
+has_useful_uniquekeys(PlannerInfo *root)
+{
+	if (root->query_uniquekeys != NIL)
+		return true;	/* there are some */
+	return false;		/* definitely useless */
+}
+
+/*
+ * make_canonical_uniquekey
+ *	  Given the parameters for a UniqueKey, find any pre-existing matching
+ *	  uniquekey in the query's list of "canonical" uniquekeys.  Make a new
+ *	  entry if there's not one already.
+ *
+ * Note that this function must not be used until after we have completed
+ * merging EquivalenceClasses.  (We don't try to enforce that here; instead,
+ * equivclass.c will complain if a merge occurs after root->canon_uniquekeys
+ * has become nonempty.)
+ */
+static UniqueKey *
+make_canonical_uniquekey(PlannerInfo *root,
+						 EquivalenceClass *eclass)
+{
+	UniqueKey  *uk;
+	ListCell   *lc;
+	MemoryContext oldcontext;
+
+	/* The passed eclass might be non-canonical, so chase up to the top */
+	while (eclass->ec_merged)
+		eclass = eclass->ec_merged;
+
+	foreach(lc, root->canon_uniquekeys)
+	{
+		uk = (UniqueKey *) lfirst(lc);
+		if (eclass == uk->eq_clause)
+			return uk;
+	}
+
+	/*
+	 * Be sure canonical uniquekeys are allocated in the main planning context.
+	 * Not an issue in normal planning, but it is for GEQO.
+	 */
+	oldcontext = MemoryContextSwitchTo(root->planner_cxt);
+
+	uk = makeNode(UniqueKey);
+	uk->eq_clause = eclass;
+
+	root->canon_uniquekeys = lappend(root->canon_uniquekeys, uk);
+
+	MemoryContextSwitchTo(oldcontext);
+
+	return uk;
+}
diff --git a/src/backend/optimizer/plan/planner.c b/src/backend/optimizer/plan/planner.c
index 8f51f59f8a..5ee9ee6595 100644
--- a/src/backend/optimizer/plan/planner.c
+++ b/src/backend/optimizer/plan/planner.c
@@ -3656,15 +3656,31 @@ standard_qp_callback(PlannerInfo *root, void *extra)
 	 * much easier, since we know that the parser ensured that one is a
 	 * superset of the other.
 	 */
+	root->canon_uniquekeys = NIL;
+	root->query_uniquekeys = NIL;
+
 	if (root->group_pathkeys)
+	{
 		root->query_pathkeys = root->group_pathkeys;
+
+		if (!root->parse->hasAggs)
+			root->query_uniquekeys = build_uniquekeys(root, qp_extra->groupClause);
+	}
 	else if (root->window_pathkeys)
 		root->query_pathkeys = root->window_pathkeys;
 	else if (list_length(root->distinct_pathkeys) >
 			 list_length(root->sort_pathkeys))
+	{
 		root->query_pathkeys = root->distinct_pathkeys;
+		root->query_uniquekeys = build_uniquekeys(root, parse->distinctClause);
+	}
 	else if (root->sort_pathkeys)
+	{
 		root->query_pathkeys = root->sort_pathkeys;
+
+		if (root->distinct_pathkeys)
+			root->query_uniquekeys = build_uniquekeys(root, parse->distinctClause);
+	}
 	else
 		root->query_pathkeys = NIL;
 }
@@ -6222,7 +6238,7 @@ plan_cluster_use_sort(Oid tableOid, Oid indexOid)
 
 	/* Estimate the cost of index scan */
 	indexScanPath = create_index_path(root, indexInfo,
-									  NIL, NIL, NIL, NIL,
+									  NIL, NIL, NIL, NIL, NIL,
 									  ForwardScanDirection, false,
 									  NULL, 1.0, false);
 
diff --git a/src/backend/optimizer/util/pathnode.c b/src/backend/optimizer/util/pathnode.c
index 0ac73984d2..ac0b937895 100644
--- a/src/backend/optimizer/util/pathnode.c
+++ b/src/backend/optimizer/util/pathnode.c
@@ -941,6 +941,7 @@ create_seqscan_path(PlannerInfo *root, RelOptInfo *rel,
 	pathnode->parallel_safe = rel->consider_parallel;
 	pathnode->parallel_workers = parallel_workers;
 	pathnode->pathkeys = NIL;	/* seqscan has unordered result */
+	pathnode->uniquekeys = NIL;
 
 	cost_seqscan(pathnode, root, rel, pathnode->param_info);
 
@@ -965,6 +966,7 @@ create_samplescan_path(PlannerInfo *root, RelOptInfo *rel, Relids required_outer
 	pathnode->parallel_safe = rel->consider_parallel;
 	pathnode->parallel_workers = 0;
 	pathnode->pathkeys = NIL;	/* samplescan has unordered result */
+	pathnode->uniquekeys = NIL;
 
 	cost_samplescan(pathnode, root, rel, pathnode->param_info);
 
@@ -1001,6 +1003,7 @@ create_index_path(PlannerInfo *root,
 				  List *indexorderbys,
 				  List *indexorderbycols,
 				  List *pathkeys,
+				  List *uniquekeys,
 				  ScanDirection indexscandir,
 				  bool indexonly,
 				  Relids required_outer,
@@ -1019,6 +1022,7 @@ create_index_path(PlannerInfo *root,
 	pathnode->path.parallel_safe = rel->consider_parallel;
 	pathnode->path.parallel_workers = 0;
 	pathnode->path.pathkeys = pathkeys;
+	pathnode->path.uniquekeys = uniquekeys;
 
 	pathnode->indexinfo = index;
 	pathnode->indexclauses = indexclauses;
@@ -1062,6 +1066,7 @@ create_bitmap_heap_path(PlannerInfo *root,
 	pathnode->path.parallel_safe = rel->consider_parallel;
 	pathnode->path.parallel_workers = parallel_degree;
 	pathnode->path.pathkeys = NIL;	/* always unordered */
+	pathnode->path.uniquekeys = NIL;
 
 	pathnode->bitmapqual = bitmapqual;
 
@@ -1923,6 +1928,7 @@ create_functionscan_path(PlannerInfo *root, RelOptInfo *rel,
 	pathnode->parallel_safe = rel->consider_parallel;
 	pathnode->parallel_workers = 0;
 	pathnode->pathkeys = pathkeys;
+	pathnode->uniquekeys = NIL;
 
 	cost_functionscan(pathnode, root, rel, pathnode->param_info);
 
@@ -1949,6 +1955,7 @@ create_tablefuncscan_path(PlannerInfo *root, RelOptInfo *rel,
 	pathnode->parallel_safe = rel->consider_parallel;
 	pathnode->parallel_workers = 0;
 	pathnode->pathkeys = NIL;	/* result is always unordered */
+	pathnode->uniquekeys = NIL;
 
 	cost_tablefuncscan(pathnode, root, rel, pathnode->param_info);
 
@@ -1975,6 +1982,7 @@ create_valuesscan_path(PlannerInfo *root, RelOptInfo *rel,
 	pathnode->parallel_safe = rel->consider_parallel;
 	pathnode->parallel_workers = 0;
 	pathnode->pathkeys = NIL;	/* result is always unordered */
+	pathnode->uniquekeys = NIL;
 
 	cost_valuesscan(pathnode, root, rel, pathnode->param_info);
 
@@ -2000,6 +2008,7 @@ create_ctescan_path(PlannerInfo *root, RelOptInfo *rel, Relids required_outer)
 	pathnode->parallel_safe = rel->consider_parallel;
 	pathnode->parallel_workers = 0;
 	pathnode->pathkeys = NIL;	/* XXX for now, result is always unordered */
+	pathnode->uniquekeys = NIL;
 
 	cost_ctescan(pathnode, root, rel, pathnode->param_info);
 
@@ -2026,6 +2035,7 @@ create_namedtuplestorescan_path(PlannerInfo *root, RelOptInfo *rel,
 	pathnode->parallel_safe = rel->consider_parallel;
 	pathnode->parallel_workers = 0;
 	pathnode->pathkeys = NIL;	/* result is always unordered */
+	pathnode->uniquekeys = NIL;
 
 	cost_namedtuplestorescan(pathnode, root, rel, pathnode->param_info);
 
@@ -2052,6 +2062,7 @@ create_resultscan_path(PlannerInfo *root, RelOptInfo *rel,
 	pathnode->parallel_safe = rel->consider_parallel;
 	pathnode->parallel_workers = 0;
 	pathnode->pathkeys = NIL;	/* result is always unordered */
+	pathnode->uniquekeys = NIL;
 
 	cost_resultscan(pathnode, root, rel, pathnode->param_info);
 
@@ -2078,6 +2089,7 @@ create_worktablescan_path(PlannerInfo *root, RelOptInfo *rel,
 	pathnode->parallel_safe = rel->consider_parallel;
 	pathnode->parallel_workers = 0;
 	pathnode->pathkeys = NIL;	/* result is always unordered */
+	pathnode->uniquekeys = NIL;
 
 	/* Cost is the same as for a regular CTE scan */
 	cost_ctescan(pathnode, root, rel, pathnode->param_info);
diff --git a/src/backend/optimizer/util/tlist.c b/src/backend/optimizer/util/tlist.c
index 7ccb10e4e1..618032e82c 100644
--- a/src/backend/optimizer/util/tlist.c
+++ b/src/backend/optimizer/util/tlist.c
@@ -427,7 +427,6 @@ get_sortgrouplist_exprs(List *sgClauses, List *targetList)
 	return result;
 }
 
-
 /*****************************************************************************
  *		Functions to extract data from a list of SortGroupClauses
  *
diff --git a/src/include/nodes/nodes.h b/src/include/nodes/nodes.h
index 4e2fb39105..a9b67c64f8 100644
--- a/src/include/nodes/nodes.h
+++ b/src/include/nodes/nodes.h
@@ -261,6 +261,7 @@ typedef enum NodeTag
 	T_EquivalenceMember,
 	T_PathKey,
 	T_PathTarget,
+	T_UniqueKey,
 	T_RestrictInfo,
 	T_IndexClause,
 	T_PlaceHolderVar,
diff --git a/src/include/nodes/pathnodes.h b/src/include/nodes/pathnodes.h
index e3c579ee44..c1d6f33fc0 100644
--- a/src/include/nodes/pathnodes.h
+++ b/src/include/nodes/pathnodes.h
@@ -269,6 +269,8 @@ struct PlannerInfo
 
 	List	   *canon_pathkeys; /* list of "canonical" PathKeys */
 
+	List	   *canon_uniquekeys; /* list of "canonical" UniqueKeys */
+
 	List	   *left_join_clauses;	/* list of RestrictInfos for mergejoinable
 									 * outer join clauses w/nonnullable var on
 									 * left */
@@ -297,6 +299,8 @@ struct PlannerInfo
 
 	List	   *query_pathkeys; /* desired pathkeys for query_planner() */
 
+	List	   *query_uniquekeys; /* */
+
 	List	   *group_pathkeys; /* groupClause pathkeys, if any */
 	List	   *window_pathkeys;	/* pathkeys of bottom window, if any */
 	List	   *distinct_pathkeys;	/* distinctClause pathkeys, if any */
@@ -1077,6 +1081,15 @@ typedef struct ParamPathInfo
 	List	   *ppi_clauses;	/* join clauses available from outer rels */
 } ParamPathInfo;
 
+/*
+ * UniqueKey
+ */
+typedef struct UniqueKey
+{
+	NodeTag		type;
+
+	EquivalenceClass *eq_clause;	/* equivalence class */
+} UniqueKey;
 
 /*
  * Type "Path" is used as-is for sequential-scan paths, as well as some other
@@ -1106,6 +1119,9 @@ typedef struct ParamPathInfo
  *
  * "pathkeys" is a List of PathKey nodes (see above), describing the sort
  * ordering of the path's output rows.
+ *
+ * "uniquekeys", if not NIL, is a list of UniqueKey nodes (see above),
+ * describing the XXX.
  */
 typedef struct Path
 {
@@ -1129,6 +1145,8 @@ typedef struct Path
 
 	List	   *pathkeys;		/* sort ordering of path's output */
 	/* pathkeys is a List of PathKey nodes; see above */
+
+	List	   *uniquekeys;	/* the unique keys, or NIL if none */
 } Path;
 
 /* Macro for extracting a path's parameterization relids; beware double eval */
diff --git a/src/include/nodes/print.h b/src/include/nodes/print.h
index cbff56a724..31e8f0686e 100644
--- a/src/include/nodes/print.h
+++ b/src/include/nodes/print.h
@@ -16,7 +16,6 @@
 
 #include "executor/tuptable.h"
 
-
 #define nodeDisplay(x)		pprint(x)
 
 extern void print(const void *obj);
@@ -28,6 +27,7 @@ extern char *pretty_format_node_dump(const char *dump);
 extern void print_rt(const List *rtable);
 extern void print_expr(const Node *expr, const List *rtable);
 extern void print_pathkeys(const List *pathkeys, const List *rtable);
+extern void print_uniquekeys(const List *uniquekeys, const List *rtable);
 extern void print_tl(const List *tlist, const List *rtable);
 extern void print_slot(TupleTableSlot *slot);
 
diff --git a/src/include/optimizer/pathnode.h b/src/include/optimizer/pathnode.h
index 182ffeef4b..374cac157b 100644
--- a/src/include/optimizer/pathnode.h
+++ b/src/include/optimizer/pathnode.h
@@ -44,6 +44,7 @@ extern IndexPath *create_index_path(PlannerInfo *root,
 									List *indexorderbys,
 									List *indexorderbycols,
 									List *pathkeys,
+									List *uniquekeys,
 									ScanDirection indexscandir,
 									bool indexonly,
 									Relids required_outer,
diff --git a/src/include/optimizer/paths.h b/src/include/optimizer/paths.h
index 7345137d1d..c7976d4a90 100644
--- a/src/include/optimizer/paths.h
+++ b/src/include/optimizer/paths.h
@@ -209,6 +209,9 @@ extern List *build_join_pathkeys(PlannerInfo *root,
 extern List *make_pathkeys_for_sortclauses(PlannerInfo *root,
 										   List *sortclauses,
 										   List *tlist);
+extern List *make_pathkeys_for_uniquekeys(PlannerInfo *root,
+										  List *sortclauses,
+										  List *tlist);
 extern void initialize_mergeclause_eclasses(PlannerInfo *root,
 											RestrictInfo *restrictinfo);
 extern void update_mergeclause_eclasses(PlannerInfo *root,
@@ -235,4 +238,12 @@ extern PathKey *make_canonical_pathkey(PlannerInfo *root,
 extern void add_paths_to_append_rel(PlannerInfo *root, RelOptInfo *rel,
 									List *live_childrels);
 
+/*
+ * uniquekey.c
+ *	  Utilities for matching and building unique keys
+ */
+extern List *build_uniquekeys(PlannerInfo *root, List *sortclauses);
+extern bool uniquekeys_contained_in(List *keys1, List *keys2);
+extern bool has_useful_uniquekeys(PlannerInfo *root);
+
 #endif							/* PATHS_H */
-- 
2.21.0


--------------A6AA3F98E061114DE0DF40DC
Content-Type: text/x-patch;
 name="v23_0002-Index-skip-scan.patch"
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment;
 filename="v23_0002-Index-skip-scan.patch"



view thread (11+ messages)  latest in thread

reply

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Reply to all the recipients using the --to and --cc options:
  reply via email

  To: [email protected]
  Cc: [email protected]
  Subject: Re: [PATCH 1/2] Unique key
  In-Reply-To: <no-message-id-204241@localhost>

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

This inbox is served by agora; see mirroring instructions
for how to clone and mirror all data and code used for this inbox