The following bug has been logged online:

Bug reference: 2839
Logged by: Vladimir
Email address: dvs (AT) fon (DOT) kamchatka.ru
PostgreSQL version: 8.1.3
Operating system: FreeBSD
Description: after record with NULL field, query result is wrong
Details:

create table tst (a int[]);
insert into tst values ('{1}'); -- (1)
insert into tst values (null);
insert into tst values ('{2}'); -- (3)
select z.a from tst z where exists (select 1 from
(
select generate_series(1,1) as s,* from (select z.a as arr0) qq
) q
where (q.arr0[q.s] is null) );
a
-----

{2}
(2 rows)

-- What difference between (1) and (3) records?
-- really, my query more complex (auto generated) and table has many
columns, wrong rows return after row where tested field is null

---------------------------(end of broadcast)---------------------------
TIP 6: explain analyze is your friend

"Vladimir" <dvs (AT) fon (DOT) kamchatka.ru> writes:
Hmm, surprising this hasn't been found before, because the underlying
bug has been around for a long time :-(. Fix committed --- thanks for
the report!

regards, tom lane

---------------------------(end of broadcast)---------------------------
TIP 4: Have you searched our list archives?

http://archives.postgresql.org