postgres.git / summary / log / commit / refs

commit    7adbe186f7fd7e036d2cab7d8ed8ee99354ec219
Author:   Tom Lane <tgl@sss.pgh.pa.us>
Date:     Fri Oct 01 18:59:35 2021 +0000
First released in: 15.0, 10.19, 11.14, 12.9, 13.5, 14.1

    Avoid believing incomplete MCV-only stats in get_variable_range().
    
    get_variable_range() would incautiously believe that statistics
    containing only an MCV list are sufficient to derive a range estimate.
    That's okay for an enum-like column that contains only MCVs, but
    otherwise the estimate could be pretty bad.  Make it report that the
    range is indeterminate unless the MCVs plus nullfrac account for
    the whole table.
    
    I don't think this needs a dedicated test case, since a quick code
    coverage check verifies that the existing regression tests traverse
    all the alternatives.  There is room to doubt that a future-proof
    test case could be built anyway, given that the submitted example
    accidentally doesn't fail before v11.
    
    Per bug #17207 from Simon Perepelitsa.  Back-patch to v10.
    In principle this has been broken all along, but I'm hesitant to
    make such changes in 9.6, since if anyone is unhappy with 9.6.24's
    behavior there will be no second chance to fix it.
    
    Discussion: https://postgr.es/m/17207-5265aefa79e333b4@postgresql.org


 src/backend/utils/adt/selfuncs.c | 30 +++++++++++++++++++++++++-----
 1 file changed, 25 insertions(+), 5 deletions(-)

diff --git a/src/backend/utils/adt/selfuncs.c b/src/backend/utils/adt/selfuncs.c
index c2aeb4b947e..81602674fda 100644
--- a/src/backend/utils/adt/selfuncs.c
+++ b/src/backend/utils/adt/selfuncs.c
@@ -5494,15 +5494,35 @@ get_variable_range(PlannerInfo *root, VariableStatData *vardata,
 	/*
 	 * If we have most-common-values info, look for extreme MCVs.  This is
 	 * needed even if we also have a histogram, since the histogram excludes
-	 * the MCVs.
+	 * the MCVs.  However, if we *only* have MCVs and no histogram, we should
+	 * be pretty wary of deciding that that is a full representation of the
+	 * data.  Proceed only if the MCVs represent the whole table (to within
+	 * roundoff error).
 	 */
 	if (get_attstatsslot(&sslot, vardata->statsTuple,
 						 STATISTIC_KIND_MCV, InvalidOid,
-						 ATTSTATSSLOT_VALUES))
+						 have_data ? ATTSTATSSLOT_VALUES :
+						 (ATTSTATSSLOT_VALUES | ATTSTATSSLOT_NUMBERS)))
 	{
-		get_stats_slot_range(&sslot, opfuncoid, &opproc,
-							 collation, typLen, typByVal,
-							 &tmin, &tmax, &have_data);
+		bool		use_mcvs = have_data;
+
+		if (!have_data)
+		{
+			double		sumcommon = 0.0;
+			double		nullfrac;
+			int			i;
+
+			for (i = 0; i < sslot.nnumbers; i++)
+				sumcommon += sslot.numbers[i];
+			nullfrac = ((Form_pg_statistic) GETSTRUCT(vardata->statsTuple))->stanullfrac;
+			if (sumcommon + nullfrac > 0.99999)
+				use_mcvs = true;
+		}
+
+		if (use_mcvs)
+			get_stats_slot_range(&sslot, opfuncoid, &opproc,
+								 collation, typLen, typByVal,
+								 &tmin, &tmax, &have_data);
 		free_attstatsslot(&sslot);
 	}
 


[parent: 04ef2021e3ca]