Fix old bug with coercing the result of a COLLATE expression.

There are hacks in parse_coerce.c to push down a requested coercion
to below any CollateExpr that may appear.  However, we did that even
if the requested data type is non-collatable, leading to an invalid
expression tree in which CollateExpr is applied to a non-collatable
type.  The fix is just to drop the CollateExpr altogether, reasoning
that it's useless.

This bug is ten years old, dating to the original addition of
COLLATE support.  The lack of field complaints suggests that there
aren't a lot of user-visible consequences.  We noticed the problem
because it would trigger an assertion in DefineVirtualRelation if
the invalid structure appears as an output column of a view; however,
in a non-assert build, you don't see a crash just a (subtly incorrect)
complaint about applying collation to a non-collatable type.  I found
that by putting the incorrect structure further down in a view, I could
make a view definition that would fail dump/reload, per the added
regression test case.  But CollateExpr doesn't do anything at run-time,
so this likely doesn't lead to any really exciting consequences.

Per report from Yulin Pei.  Back-patch to all supported branches.

Discussion: https://postgr.es/m/HK0PR01MB22744393C474D503E16C8509F4709@HK0PR01MB2274.apcprd01.prod.exchangelabs.com

diff --git a/src/backend/parser/parse_coerce.c b/src/backend/parser/parse_coerce.c
index 903478d8ca73a7730e9f4a63e4cc2205c8a1c849..eae888e501159d78eefe7695aa0bcc5d3f995d1b 100644 (file)
--- a/src/backend/parser/parse_coerce.c
+++ b/src/backend/parser/parse_coerce.c
@@ -95,6 +95,7 @@ coerce_to_target_type(ParseState *pstate, Node *expr, Oid exprtype,
         * *must* know that to avoid possibly calling hide_coercion_node on
         * something that wasn't generated by coerce_type.  Note that if there are
         * multiple stacked CollateExprs, we just discard all but the topmost.
+        * Also, if the target type isn't collatable, we discard the CollateExpr.
         */
        origexpr = expr;
        while (expr && IsA(expr, CollateExpr))
@@ -114,7 +115,7 @@ coerce_to_target_type(ParseState *pstate, Node *expr, Oid exprtype,
                                                                ccontext, cformat, location,
                                                                (result != expr && !IsA(result, Const)));
 
-       if (expr != origexpr)
+       if (expr != origexpr && type_is_collatable(targettype))
        {
                /* Reinstall top CollateExpr */
                CollateExpr *coll = (CollateExpr *) origexpr;
@@ -382,20 +383,26 @@ coerce_type(ParseState *pstate, Node *node,
        {
                /*
                 * If we have a COLLATE clause, we have to push the coercion
-                * underneath the COLLATE.  This is really ugly, but there is little
-                * choice because the above hacks on Consts and Params wouldn't happen
+                * underneath the COLLATE; or discard the COLLATE if the target type
+                * isn't collatable.  This is really ugly, but there is little choice
+                * because the above hacks on Consts and Params wouldn't happen
                 * otherwise.  This kluge has consequences in coerce_to_target_type.
                 */
                CollateExpr *coll = (CollateExpr *) node;
-               CollateExpr *newcoll = makeNode(CollateExpr);
 
-               newcoll->arg = (Expr *)
-                       coerce_type(pstate, (Node *) coll->arg,
-                                               inputTypeId, targetTypeId, targetTypeMod,
-                                               ccontext, cformat, location);
-               newcoll->collOid = coll->collOid;
-               newcoll->location = coll->location;
-               return (Node *) newcoll;
+               result = coerce_type(pstate, (Node *) coll->arg,
+                                                        inputTypeId, targetTypeId, targetTypeMod,
+                                                        ccontext, cformat, location);
+               if (type_is_collatable(targetTypeId))
+               {
+                       CollateExpr *newcoll = makeNode(CollateExpr);
+
+                       newcoll->arg = (Expr *) result;
+                       newcoll->collOid = coll->collOid;
+                       newcoll->location = coll->location;
+                       result = (Node *) newcoll;
+               }
+               return result;
        }
        pathtype = find_coercion_pathway(targetTypeId, inputTypeId, ccontext,
                                                                         &funcId);

diff --git a/src/test/regress/expected/collate.out b/src/test/regress/expected/collate.out
index 0dee7d783a0c854216298d739a0e6d11f58e6692..6c7321bdf24a19299e5700427ab6946e8f054cc7 100644 (file)
--- a/src/test/regress/expected/collate.out
+++ b/src/test/regress/expected/collate.out
@@ -676,13 +676,26 @@ SELECT collation for ((SELECT b FROM collate_test1 LIMIT 1));
  "C"
 (1 row)
 
+-- old bug with not dropping COLLATE when coercing to non-collatable type
+CREATE VIEW collate_on_int AS
+SELECT c1+1 AS c1p FROM
+  (SELECT ('4' COLLATE "C")::INT AS c1) ss;
+\d+ collate_on_int
+                    View "collate_tests.collate_on_int"
+ Column |  Type   | Collation | Nullable | Default | Storage | Description 
+--------+---------+-----------+----------+---------+---------+-------------
+ c1p    | integer |           |          |         | plain   | 
+View definition:
+ SELECT ss.c1 + 1 AS c1p
+   FROM ( SELECT 4 AS c1) ss;
+
 --
 -- Clean up.  Many of these table names will be re-used if the user is
 -- trying to run any platform-specific collation tests later, so we
 -- must get rid of them.
 --
 DROP SCHEMA collate_tests CASCADE;
-NOTICE:  drop cascades to 17 other objects
+NOTICE:  drop cascades to 18 other objects
 DETAIL:  drop cascades to table collate_test1
 drop cascades to table collate_test_like
 drop cascades to table collate_test2
@@ -700,3 +713,4 @@ drop cascades to table collate_test21
 drop cascades to table collate_test22
 drop cascades to collation mycoll2
 drop cascades to table collate_test23
+drop cascades to view collate_on_int

diff --git a/src/test/regress/sql/collate.sql b/src/test/regress/sql/collate.sql
index 89de26a227e92d1200234e96cb729c99fedfc67e..7e2f81013c66d2de9d93dd59625c2f4ea2d9d345 100644 (file)
--- a/src/test/regress/sql/collate.sql
+++ b/src/test/regress/sql/collate.sql
@@ -258,6 +258,12 @@ SELECT collation for ('foo'::text);
 SELECT collation for ((SELECT a FROM collate_test1 LIMIT 1)); -- non-collatable type - error
 SELECT collation for ((SELECT b FROM collate_test1 LIMIT 1));
 
+-- old bug with not dropping COLLATE when coercing to non-collatable type
+CREATE VIEW collate_on_int AS
+SELECT c1+1 AS c1p FROM
+  (SELECT ('4' COLLATE "C")::INT AS c1) ss;
+\d+ collate_on_int
+
 
 --
 -- Clean up.  Many of these table names will be re-used if the user is