Postpone aggregate checks until after collation is assigned.

Previously, parseCheckAggregates was run before
assign_query_collations, but this causes problems if any expression
has already had a collation assigned by some transform function (e.g.
transformCaseExpr) before parseCheckAggregates runs. The differing
collations would cause expressions not to be recognized as equal to
the ones in the GROUP BY clause, leading to spurious errors about
unaggregated column references.

The result was that CASE expr WHEN val ... would fail when "expr"
contained a GROUPING() expression or matched one of the group by
expressions, and where collatable types were involved; whereas the
supposedly identical CASE WHEN expr = val ... would succeed.

Backpatch all the way; this appears to have been wrong ever since
collations were introduced.

Per report from Guillaume Lelarge, analysis and patch by me.

Discussion: https://postgr.es/m/CAECtzeVSO_US8C2Khgfv54ZMUOBR4sWq+6_bLrETnWExHT=rFg@mail.gmail.com
Discussion: https://postgr.es/m/[email protected]

5 files changed, 75 insertions, 6 deletions

diff --git a/src/backend/parser/analyze.c b/src/backend/parser/analyze.c
index 5ff6964d51..8f96558b3d 100644
--- a/src/backend/parser/analyze.c
+++ b/src/backend/parser/analyze.c
@@ -451,11 +451,13 @@ transformDeleteStmt(ParseState *pstate, DeleteStmt *stmt)
 	qry->hasWindowFuncs = pstate->p_hasWindowFuncs;
 	qry->hasTargetSRFs = pstate->p_hasTargetSRFs;
 	qry->hasAggs = pstate->p_hasAggs;
-	if (pstate->p_hasAggs)
-		parseCheckAggregates(pstate, qry);
 
 	assign_query_collations(pstate, qry);
 
+	/* this must be done after collations, for reliable comparison of exprs */
+	if (pstate->p_hasAggs)
+		parseCheckAggregates(pstate, qry);
+
 	return qry;
 }
 
@@ -1319,8 +1321,6 @@ transformSelectStmt(ParseState *pstate, SelectStmt *stmt)
 	qry->hasWindowFuncs = pstate->p_hasWindowFuncs;
 	qry->hasTargetSRFs = pstate->p_hasTargetSRFs;
 	qry->hasAggs = pstate->p_hasAggs;
-	if (pstate->p_hasAggs || qry->groupClause || qry->groupingSets || qry->havingQual)
-		parseCheckAggregates(pstate, qry);
 
 	foreach(l, stmt->lockingClause)
 	{
@@ -1330,6 +1330,10 @@ transformSelectStmt(ParseState *pstate, SelectStmt *stmt)
 
 	assign_query_collations(pstate, qry);
 
+	/* this must be done after collations, for reliable comparison of exprs */
+	if (pstate->p_hasAggs || qry->groupClause || qry->groupingSets || qry->havingQual)
+		parseCheckAggregates(pstate, qry);
+
 	return qry;
 }
 
@@ -1791,8 +1795,6 @@ transformSetOperationStmt(ParseState *pstate, SelectStmt *stmt)
 	qry->hasWindowFuncs = pstate->p_hasWindowFuncs;
 	qry->hasTargetSRFs = pstate->p_hasTargetSRFs;
 	qry->hasAggs = pstate->p_hasAggs;
-	if (pstate->p_hasAggs || qry->groupClause || qry->groupingSets || qry->havingQual)
-		parseCheckAggregates(pstate, qry);
 
 	foreach(l, lockingClause)
 	{
@@ -1802,6 +1804,10 @@ transformSetOperationStmt(ParseState *pstate, SelectStmt *stmt)
 
 	assign_query_collations(pstate, qry);
 
+	/* this must be done after collations, for reliable comparison of exprs */
+	if (pstate->p_hasAggs || qry->groupClause || qry->groupingSets || qry->havingQual)
+		parseCheckAggregates(pstate, qry);
+
 	return qry;
 }
 
diff --git a/src/test/regress/expected/aggregates.out b/src/test/regress/expected/aggregates.out
index 20dacfe4b0..918db6cb67 100644
--- a/src/test/regress/expected/aggregates.out
+++ b/src/test/regress/expected/aggregates.out
@@ -2244,3 +2244,22 @@ SELECT min(x ORDER BY y) FROM (VALUES(1, 2)) AS d(x,y);
    1
 (1 row)
 
+-- check collation-sensitive matching between grouping expressions
+select v||'a', case v||'a' when 'aa' then 1 else 0 end, count(*)
+  from unnest(array['a','b']) u(v)
+ group by v||'a' order by 1;
+ ?column? | case | count 
+----------+------+-------
+ aa       |    1 |     1
+ ba       |    0 |     1
+(2 rows)
+
+select v||'a', case when v||'a' = 'aa' then 1 else 0 end, count(*)
+  from unnest(array['a','b']) u(v)
+ group by v||'a' order by 1;
+ ?column? | case | count 
+----------+------+-------
+ aa       |    1 |     1
+ ba       |    0 |     1
+(2 rows)
+
diff --git a/src/test/regress/expected/groupingsets.out b/src/test/regress/expected/groupingsets.out
index c7deec2ff4..381ebce8a1 100644
--- a/src/test/regress/expected/groupingsets.out
+++ b/src/test/regress/expected/groupingsets.out
@@ -1540,4 +1540,29 @@ explain (costs off)
          ->  Seq Scan on tenk1
 (12 rows)
 
+-- check collation-sensitive matching between grouping expressions
+-- (similar to a check for aggregates, but there are additional code
+-- paths for GROUPING, so check again here)
+select v||'a', case grouping(v||'a') when 1 then 1 else 0 end, count(*)
+  from unnest(array[1,1], array['a','b']) u(i,v)
+ group by rollup(i, v||'a') order by 1,3;
+ ?column? | case | count 
+----------+------+-------
+ aa       |    0 |     1
+ ba       |    0 |     1
+          |    1 |     2
+          |    1 |     2
+(4 rows)
+
+select v||'a', case when grouping(v||'a') = 1 then 1 else 0 end, count(*)
+  from unnest(array[1,1], array['a','b']) u(i,v)
+ group by rollup(i, v||'a') order by 1,3;
+ ?column? | case | count 
+----------+------+-------
+ aa       |    0 |     1
+ ba       |    0 |     1
+          |    1 |     2
+          |    1 |     2
+(4 rows)
+
 -- end
diff --git a/src/test/regress/sql/aggregates.sql b/src/test/regress/sql/aggregates.sql
index 8192d457e9..a09fa47ae4 100644
--- a/src/test/regress/sql/aggregates.sql
+++ b/src/test/regress/sql/aggregates.sql
@@ -976,3 +976,11 @@ SELECT dense_rank(x) WITHIN GROUP (ORDER BY x) FROM (VALUES (1),(1),(2),(2),(3),
 -- [email protected]
 SELECT min(x ORDER BY y) FROM (VALUES(1, NULL)) AS d(x,y);
 SELECT min(x ORDER BY y) FROM (VALUES(1, 2)) AS d(x,y);
+
+-- check collation-sensitive matching between grouping expressions
+select v||'a', case v||'a' when 'aa' then 1 else 0 end, count(*)
+  from unnest(array['a','b']) u(v)
+ group by v||'a' order by 1;
+select v||'a', case when v||'a' = 'aa' then 1 else 0 end, count(*)
+  from unnest(array['a','b']) u(v)
+ group by v||'a' order by 1;
diff --git a/src/test/regress/sql/groupingsets.sql b/src/test/regress/sql/groupingsets.sql
index c32d23b8d7..5d6485913b 100644
--- a/src/test/regress/sql/groupingsets.sql
+++ b/src/test/regress/sql/groupingsets.sql
@@ -415,4 +415,15 @@ explain (costs off)
          count(*)
     from tenk1 group by grouping sets (unique1,twothousand,thousand,hundred,ten,four,two);
 
+-- check collation-sensitive matching between grouping expressions
+-- (similar to a check for aggregates, but there are additional code
+-- paths for GROUPING, so check again here)
+
+select v||'a', case grouping(v||'a') when 1 then 1 else 0 end, count(*)
+  from unnest(array[1,1], array['a','b']) u(i,v)
+ group by rollup(i, v||'a') order by 1,3;
+select v||'a', case when grouping(v||'a') = 1 then 1 else 0 end, count(*)
+  from unnest(array[1,1], array['a','b']) u(i,v)
+ group by rollup(i, v||'a') order by 1,3;
+
 -- end