Clarify use of "statistics objects" in the code

michaelpq · michaelpq · commit 070d2e19e408 · 2021-09-29T15:29:38.000+09:00
The code inconsistently used "statistic object" or "statistics" where the correct term, as discussed, is actually "statistics object". This improves the state of the code to be more consistent. While on it, fix an incorrect error message introduced in a4d75c8. This error should never happen, as the code states, but it would be misleading. Author: Justin Pryzby Reviewed-by: Álvaro Herrera, Michael Paquier Discussion: https://postgr.es/m/20210924215827.GS831@telsasoft.com Backpatch-through: 14
diff --git a/src/backend/commands/statscmds.c b/src/backend/commands/statscmds.c
@@ -560,11 +560,11 @@ CreateStatistics(CreateStatsStmt *stmt)
 	}
 
 	/*
-	 * If there are no dependencies on a column, give the statistics an auto
-	 * dependency on the whole table.  In most cases, this will be redundant,
-	 * but it might not be if the statistics expressions contain no Vars
-	 * (which might seem strange but possible). This is consistent with what
-	 * we do for indexes in index_create.
+	 * If there are no dependencies on a column, give the statistics object an
+	 * auto dependency on the whole table.  In most cases, this will be
+	 * redundant, but it might not be if the statistics expressions contain no
+	 * Vars (which might seem strange but possible). This is consistent with
+	 * what we do for indexes in index_create.
 	 *
 	 * XXX We intentionally don't consider the expressions before adding this
 	 * dependency, because recordDependencyOnSingleRelExpr may not create any
@@ -649,9 +649,9 @@ AlterStatistics(AlterStatsStmt *stmt)
 	stxoid = get_statistics_object_oid(stmt->defnames, stmt->missing_ok);
 
 	/*
-	 * If we got here and the OID is not valid, it means the statistics does
-	 * not exist, but the command specified IF EXISTS. So report this as a
-	 * simple NOTICE and we're done.
+	 * If we got here and the OID is not valid, it means the statistics object
+	 * does not exist, but the command specified IF EXISTS. So report this as
+	 * a simple NOTICE and we're done.
 	 */
 	if (!OidIsValid(stxoid))
 	{
@@ -768,7 +768,7 @@ RemoveStatisticsById(Oid statsOid)
 }
 
 /*
- * Select a nonconflicting name for a new statistics.
+ * Select a nonconflicting name for a new statistics object.
  *
  * name1, name2, and label are used the same way as for makeObjectName(),
  * except that the label can't be NULL; digits will be appended to the label
@@ -815,9 +815,9 @@ ChooseExtendedStatisticName(const char *name1, const char *name2,
 }
 
 /*
- * Generate "name2" for a new statistics given the list of column names for it
- * This will be passed to ChooseExtendedStatisticName along with the parent
- * table name and a suitable label.
+ * Generate "name2" for a new statistics object given the list of column
+ * names for it.  This will be passed to ChooseExtendedStatisticName along
+ * with the parent table name and a suitable label.
  *
  * We know that less than NAMEDATALEN characters will actually be used,
  * so we can truncate the result once we've generated that many.
@@ -869,8 +869,8 @@ ChooseExtendedStatisticNameAddition(List *exprs)
 }
 
 /*
- * StatisticsGetRelation: given a statistics's relation OID, get the OID of
- * the relation it is an statistics on.  Uses the system cache.
+ * StatisticsGetRelation: given a statistics object's OID, get the OID of
+ * the relation it is defined on.  Uses the system cache.
  */
 Oid
 StatisticsGetRelation(Oid statId, bool missing_ok)
diff --git a/src/backend/commands/tablecmds.c b/src/backend/commands/tablecmds.c
@@ -12585,13 +12585,13 @@ RememberStatisticsForRebuilding(Oid stxoid, AlteredTableInfo *tab)
 	/*
 	 * This de-duplication check is critical for two independent reasons: we
 	 * mustn't try to recreate the same statistics object twice, and if the
-	 * statistics depends on more than one column whose type is to be altered,
-	 * we must capture its definition string before applying any of the type
-	 * changes. ruleutils.c will get confused if we ask again later.
+	 * statistics object depends on more than one column whose type is to be
+	 * altered, we must capture its definition string before applying any of
+	 * the type changes. ruleutils.c will get confused if we ask again later.
 	 */
 	if (!list_member_oid(tab->changedStatisticsOids, stxoid))
 	{
-		/* OK, capture the index's existing definition string */
+		/* OK, capture the statistics object's existing definition string */
 		char	   *defstring = pg_get_statisticsobjdef_string(stxoid);
 
 		tab->changedStatisticsOids = lappend_oid(tab->changedStatisticsOids,
diff --git a/src/backend/parser/parse_utilcmd.c b/src/backend/parser/parse_utilcmd.c
@@ -2892,7 +2892,7 @@ transformIndexStmt(Oid relid, IndexStmt *stmt, const char *queryString)
 /*
  * transformStatsStmt - parse analysis for CREATE STATISTICS
  *
- * To avoid race conditions, it's important that this function rely only on
+ * To avoid race conditions, it's important that this function relies only on
  * the passed-in relid (and not on stmt->relation) to determine the target
  * relation.
  */
@@ -2948,7 +2948,7 @@ transformStatsStmt(Oid relid, CreateStatsStmt *stmt, const char *queryString)
 	if (list_length(pstate->p_rtable) != 1)
 		ereport(ERROR,
 				(errcode(ERRCODE_INVALID_COLUMN_REFERENCE),
-				 errmsg("statistics expressions can refer only to the table being indexed")));
+				 errmsg("statistics expressions can refer only to the table being referenced")));
 
 	free_parsestate(pstate);
 
@@ -3615,7 +3615,7 @@ transformAlterTableStmt(Oid relid, AlterTableStmt *stmt,
 		newcmds = lappend(newcmds, newcmd);
 	}
 
-	/* Append extended statistic objects */
+	/* Append extended statistics objects */
 	transformExtendedStatistics(&cxt);
 
 	/* Close rel */
diff --git a/src/backend/statistics/extended_stats.c b/src/backend/statistics/extended_stats.c
@@ -182,7 +182,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows,
 			continue;
 		}
 
-		/* compute statistics target for this statistics */
+		/* compute statistics target for this statistics object */
 		stattarget = statext_compute_stattarget(stat->stattarget,
 												bms_num_members(stat->columns),
 												stats);
@@ -195,7 +195,7 @@ BuildRelationExtStatistics(Relation onerel, double totalrows,
 		if (stattarget == 0)
 			continue;
 
-		/* evaluate expressions (if the statistics has any) */
+		/* evaluate expressions (if the statistics object has any) */
 		data = make_build_data(onerel, stat, numrows, rows, stats, stattarget);
 
 		/* compute statistic of each requested type */
@@ -257,9 +257,9 @@ BuildRelationExtStatistics(Relation onerel, double totalrows,
  * when analyzing only some of the columns, this will skip statistics objects
  * that would require additional columns.
  *
- * See statext_compute_stattarget for details about how we compute statistics
- * target for a statistics object (from the object target, attribute targets
- * and default statistics target).
+ * See statext_compute_stattarget for details about how we compute the
+ * statistics target for a statistics object (from the object target,
+ * attribute targets and default statistics target).
  */
 int
 ComputeExtStatisticsRows(Relation onerel,
@@ -329,8 +329,8 @@ ComputeExtStatisticsRows(Relation onerel,
  *
  * When computing target for extended statistics objects, we consider three
  * places where the target may be set - the statistics object itself,
- * attributes the statistics is defined on, and then the default statistics
- * target.
+ * attributes the statistics object is defined on, and then the default
+ * statistics target.
  *
  * First we look at what's set for the statistics object itself, using the
  * ALTER STATISTICS ... SET STATISTICS command. If we find a valid value
@@ -1789,8 +1789,8 @@ statext_mcv_clauselist_selectivity(PlannerInfo *root, List *clauses, int varReli
 
 			/*
 			 * The clause was not estimated yet, and we've extracted either
-			 * attnums of expressions from it. Ignore it if it's not fully
-			 * covered by the chosen statistics.
+			 * attnums or expressions from it. Ignore it if it's not fully
+			 * covered by the chosen statistics object.
 			 *
 			 * We need to check both attributes and expressions, and reject if
 			 * either is not covered.
diff --git a/src/backend/utils/adt/selfuncs.c b/src/backend/utils/adt/selfuncs.c
@@ -3443,13 +3443,13 @@ estimate_num_groups(PlannerInfo *root, List *groupExprs, double input_rows,
 		 * expression, treat it as a single variable even if it's really more
 		 * complicated.
 		 *
-		 * XXX This has the consequence that if there's a statistics on the
-		 * expression, we don't split it into individual Vars. This affects
-		 * our selection of statistics in estimate_multivariate_ndistinct,
-		 * because it's probably better to use more accurate estimate for each
-		 * expression and treat them as independent, than to combine estimates
-		 * for the extracted variables when we don't know how that relates to
-		 * the expressions.
+		 * XXX This has the consequence that if there's a statistics object on
+		 * the expression, we don't split it into individual Vars. This
+		 * affects our selection of statistics in
+		 * estimate_multivariate_ndistinct, because it's probably better to
+		 * use more accurate estimate for each expression and treat them as
+		 * independent, than to combine estimates for the extracted variables
+		 * when we don't know how that relates to the expressions.
 		 */
 		examine_variable(root, groupexpr, 0, &vardata);
 		if (HeapTupleIsValid(vardata.statsTuple) || vardata.isunique)
@@ -3918,7 +3918,7 @@ estimate_multivariate_ndistinct(PlannerInfo *root, RelOptInfo *rel,
 	if (!rel->statlist)
 		return false;
 
-	/* look for the ndistinct statistics matching the most vars */
+	/* look for the ndistinct statistics object matching the most vars */
 	nmatches_vars = 0;			/* we require at least two matches */
 	nmatches_exprs = 0;
 	foreach(lc, rel->statlist)
@@ -3964,7 +3964,7 @@ estimate_multivariate_ndistinct(PlannerInfo *root, RelOptInfo *rel,
 				continue;
 			}
 
-			/* expression - see if it's in the statistics */
+			/* expression - see if it's in the statistics object */
 			foreach(lc3, info->exprs)
 			{
 				Node	   *expr = (Node *) lfirst(lc3);
@@ -4053,7 +4053,7 @@ estimate_multivariate_ndistinct(PlannerInfo *root, RelOptInfo *rel,
 				if (!AttrNumberIsForUserDefinedAttr(attnum))
 					continue;
 
-				/* Is the variable covered by the statistics? */
+				/* Is the variable covered by the statistics object? */
 				if (!bms_is_member(attnum, matched_info->keys))
 					continue;
 
@@ -4075,7 +4075,7 @@ estimate_multivariate_ndistinct(PlannerInfo *root, RelOptInfo *rel,
 			if (found)
 				continue;
 
-			/* expression - see if it's in the statistics */
+			/* expression - see if it's in the statistics object */
 			idx = 0;
 			foreach(lc3, matched_info->exprs)
 			{
@@ -5252,7 +5252,7 @@ examine_variable(PlannerInfo *root, Node *node, int varRelid,
 				{
 					HeapTuple	t = statext_expressions_load(info->statOid, pos);
 
-					/* Get index's table for permission check */
+					/* Get statistics object's table for permission check */
 					RangeTblEntry *rte;
 					Oid			userid;
 
@@ -5276,8 +5276,8 @@ examine_variable(PlannerInfo *root, Node *node, int varRelid,
 					/*
 					 * For simplicity, we insist on the whole table being
 					 * selectable, rather than trying to identify which
-					 * column(s) the statistics depends on.  Also require all
-					 * rows to be selectable --- there must be no
+					 * column(s) the statistics object depends on.  Also
+					 * require all rows to be selectable --- there must be no
 					 * securityQuals from security barrier views or RLS
 					 * policies.
 					 */
