postgres .. unique index on two fields

gistfile1.sql

test=# create table j (id1 int, id2 int);
CREATE TABLE
test=*# create unique index idx_j on j(greatest(id1, id2), least(id1, id2));
CREATE INDEX
test=*# insert into j values (1,2);
INSERT 0 1
test=*# insert into j values (2,1);
ERROR:  duplicate key violates unique constraint "idx_j"