Store -1 in attdisbursion to signal 'no duplicates in column'.

Centralize att_disbursion readout logic.

3 files changed, 43 insertions, 102 deletions

diff --git a/src/backend/commands/vacuum.c b/src/backend/commands/vacuum.c
index 0418a8d3a38..b480b44f347 100644
--- a/src/backend/commands/vacuum.c
+++ b/src/backend/commands/vacuum.c
@@ -7,7 +7,7 @@
  *
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/commands/vacuum.c,v 1.117 1999/08/08 17:13:10 tgl Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/commands/vacuum.c,v 1.118 1999/08/09 03:16:47 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -2346,36 +2346,46 @@ vc_updstats(Oid relid, int num_pages, int num_tuples, bool hasindex, VRelStats *
 				}
 				else if (stats->null_cnt <= 1 && stats->best_cnt == 1)
 				{
-					/* looks like we have a unique-key attribute */
-					double		total = ((double) stats->nonnull_cnt) + ((double) stats->null_cnt);
-
-					selratio = 1.0 / total;
-				}
-				else if (VacAttrStatsLtGtValid(stats) && stats->min_cnt + stats->max_cnt == stats->nonnull_cnt)
-				{
-					/* exact result when there are just 1 or 2 values... */
-					double		min_cnt_d = stats->min_cnt,
-								max_cnt_d = stats->max_cnt,
-								null_cnt_d = stats->null_cnt;
-					double		total = ((double) stats->nonnull_cnt) + null_cnt_d;
-
-					selratio = (min_cnt_d * min_cnt_d + max_cnt_d * max_cnt_d + null_cnt_d * null_cnt_d) / (total * total);
+					/* looks like we have a unique-key attribute ---
+					 * flag this with special -1.0 flag value.
+					 *
+					 * The correct disbursion is 1.0/numberOfRows, but since
+					 * the relation row count can get updated without
+					 * recomputing disbursion, we want to store a "symbolic"
+					 * value and figure 1.0/numberOfRows on the fly.
+					 */
+					selratio = -1;
 				}
 				else
 				{
-					double		most = (double) (stats->best_cnt > stats->null_cnt ? stats->best_cnt : stats->null_cnt);
-					double		total = ((double) stats->nonnull_cnt) + ((double) stats->null_cnt);
+					if (VacAttrStatsLtGtValid(stats) &&
+						stats->min_cnt + stats->max_cnt == stats->nonnull_cnt)
+					{
+						/* exact result when there are just 1 or 2 values... */
+						double		min_cnt_d = stats->min_cnt,
+									max_cnt_d = stats->max_cnt,
+									null_cnt_d = stats->null_cnt;
+						double		total = ((double) stats->nonnull_cnt) + null_cnt_d;
 
-					/*
-					 * we assume count of other values are 20% of best
-					 * count in table
-					 */
-					selratio = (most * most + 0.20 * most * (total - most)) / (total * total);
+						selratio = (min_cnt_d * min_cnt_d + max_cnt_d * max_cnt_d + null_cnt_d * null_cnt_d) / (total * total);
+					}
+					else
+					{
+						double		most = (double) (stats->best_cnt > stats->null_cnt ? stats->best_cnt : stats->null_cnt);
+						double		total = ((double) stats->nonnull_cnt) + ((double) stats->null_cnt);
+
+						/*
+						 * we assume count of other values are 20% of best
+						 * count in table
+						 */
+						selratio = (most * most + 0.20 * most * (total - most)) / (total * total);
+					}
+					/* Make sure calculated values are in-range */
+					if (selratio < 0.0)
+						selratio = 0.0;
+					else if (selratio > 1.0)
+						selratio = 1.0;
 				}
-				if (selratio < 0.0)
-					selratio = 0.0;
-				else if (selratio > 1.0)
-					selratio = 1.0;
 				attp->attdisbursion = selratio;
 
 				/*
diff --git a/src/backend/optimizer/path/joinpath.c b/src/backend/optimizer/path/joinpath.c
index 57688deeb85..4a7018aa64a 100644
--- a/src/backend/optimizer/path/joinpath.c
+++ b/src/backend/optimizer/path/joinpath.c
@@ -7,7 +7,7 @@
  *
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/optimizer/path/joinpath.c,v 1.43 1999/08/06 04:00:15 tgl Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/optimizer/path/joinpath.c,v 1.44 1999/08/09 03:16:43 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -23,7 +23,7 @@
 #include "optimizer/pathnode.h"
 #include "optimizer/paths.h"
 #include "parser/parsetree.h"
-#include "utils/syscache.h"
+#include "utils/lsyscache.h"
 
 static Path *best_innerjoin(List *join_paths, List *outer_relid);
 static List *sort_inner_and_outer(RelOptInfo *joinrel, RelOptInfo *outerrel, RelOptInfo *innerrel,
@@ -586,7 +586,6 @@ hash_inner_and_outer(Query *root,
 
 /*
  * Estimate disbursion of the specified Var
- *	  Generate some kind of estimate, no matter what...
  *
  * We use a default of 0.1 if we can't figure out anything better.
  * This will typically discourage use of a hash rather strongly,
@@ -598,24 +597,11 @@ static Cost
 estimate_disbursion(Query *root, Var *var)
 {
 	Oid			relid;
-	HeapTuple	atp;
-	double		disbursion;
 
 	if (! IsA(var, Var))
 		return 0.1;
 
 	relid = getrelid(var->varno, root->rtable);
 
-	atp = SearchSysCacheTuple(ATTNUM,
-							  ObjectIdGetDatum(relid),
-							  Int16GetDatum(var->varattno),
-							  0, 0);
-	if (! HeapTupleIsValid(atp))
-		return 0.1;
-
-	disbursion = ((Form_pg_attribute) GETSTRUCT(atp))->attdisbursion;
-	if (disbursion > 0.0)
-		return disbursion;
-
-	return 0.1;
+	return (Cost) get_attdisbursion(relid, var->varattno, 0.1);
 }
diff --git a/src/backend/utils/adt/selfuncs.c b/src/backend/utils/adt/selfuncs.c
index a0e0c7ad7fc..298c7aeed16 100644
--- a/src/backend/utils/adt/selfuncs.c
+++ b/src/backend/utils/adt/selfuncs.c
@@ -10,7 +10,7 @@
  *
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/utils/adt/selfuncs.c,v 1.37 1999/08/02 02:05:41 tgl Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/utils/adt/selfuncs.c,v 1.38 1999/08/09 03:16:45 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -52,7 +52,6 @@ static bool getattstatistics(Oid relid, AttrNumber attnum,
 							 Datum *commonval,
 							 Datum *loval,
 							 Datum *hival);
-static double getattdisbursion(Oid relid, AttrNumber attnum);
 
 
 /*
@@ -172,7 +171,7 @@ eqsel(Oid opid,
 			/* No VACUUM ANALYZE stats available, so make a guess using
 			 * the disbursion stat (if we have that, which is unlikely...)
 			 */
-			selec = getattdisbursion(relid, attno);
+			selec = get_attdisbursion(relid, attno, 0.01);
 		}
 
 		*result = (float64data) selec;
@@ -374,8 +373,8 @@ eqjoinsel(Oid opid,
 		*result = 0.1;
 	else
 	{
-		num1 = getattdisbursion(relid1, attno1);
-		num2 = getattdisbursion(relid2, attno2);
+		num1 = get_attdisbursion(relid1, attno1, 0.01);
+		num2 = get_attdisbursion(relid2, attno2, 0.01);
 		max = (num1 > num2) ? num1 : num2;
 		if (max <= 0)
 			*result = 1.0;
@@ -675,60 +674,6 @@ getattstatistics(Oid relid, AttrNumber attnum, Oid typid, int32 typmod,
 	return true;
 }
 
-/*
- * getattdisbursion
- *	  Retrieve the disbursion statistic for an attribute,
- *	  or produce an estimate if no info is available.
- */
-static double
-getattdisbursion(Oid relid, AttrNumber attnum)
-{
-	HeapTuple	atp;
-	double		disbursion;
-	int32		ntuples;
-
-	atp = SearchSysCacheTuple(ATTNUM,
-							  ObjectIdGetDatum(relid),
-							  Int16GetDatum(attnum),
-							  0, 0);
-	if (!HeapTupleIsValid(atp))
-	{
-		/* this should not happen */
-		elog(ERROR, "getattdisbursion: no attribute tuple %u %d",
-			 relid, attnum);
-		return 0.1;
-	}
-
-	disbursion = ((Form_pg_attribute) GETSTRUCT(atp))->attdisbursion;
-	if (disbursion > 0.0)
-		return disbursion;
-
-	/* VACUUM ANALYZE has not stored a disbursion statistic for us.
-	 * Produce an estimate = 1/numtuples.  This may produce
-	 * unreasonably small estimates for large tables, so limit
-	 * the estimate to no less than 0.01.
-	 */
-	atp = SearchSysCacheTuple(RELOID,
-							  ObjectIdGetDatum(relid),
-							  0, 0, 0);
-	if (!HeapTupleIsValid(atp))
-	{
-		/* this should not happen */
-		elog(ERROR, "getattdisbursion: no relation tuple %u", relid);
-		return 0.1;
-	}
-
-	ntuples = ((Form_pg_class) GETSTRUCT(atp))->reltuples;
-
-	if (ntuples > 0)
-		disbursion = 1.0 / (double) ntuples;
-
-	if (disbursion < 0.01)
-		disbursion = 0.01;
-
-	return disbursion;
-}
-
 float64
 btreesel(Oid operatorObjectId,
 		 Oid indrelid,