Wups, stuck ORDER BY on wrong query while trying to eliminate random

ordering of GROUP BY output due to hash-based aggregation.

src/test/regress/expected/rules.out

@@ -858,7 +858,7 @@ select * from rtest_view3;
 
 delete from rtest_view3;
 insert into rtest_view4 select * from rtest_vview4 where 3 > refcount;
-select * from rtest_view4;
+select * from rtest_view4 order by a, b;
  a |   b    | c 
 ---+--------+---
  2 | item 2 | 2
@@ -868,7 +868,7 @@ select * from rtest_view4;
 
 delete from rtest_view4;
 insert into rtest_view4 select * from rtest_vview5 where a > 2 and refcount = 0;
-select * from rtest_view4 order by a, b;
+select * from rtest_view4;
  a |   b    | c 
 ---+--------+---
  3 | item 3 | 0

src/test/regress/sql/rules.sql

@@ -467,11 +467,11 @@ select * from rtest_view3;
 delete from rtest_view3;
 
 insert into rtest_view4 select * from rtest_vview4 where 3 > refcount;
-select * from rtest_view4;
+select * from rtest_view4 order by a, b;
 delete from rtest_view4;
 
 insert into rtest_view4 select * from rtest_vview5 where a > 2 and refcount = 0;
-select * from rtest_view4 order by a, b;
+select * from rtest_view4;
 delete from rtest_view4;
 --
 -- Test for computations in views