Major Description
The second query should return two rows but returns just one.
ij> select * from t1;
I |V
----------------------
1 |row 1
2 |row 2
2 rows selected
ij> select * from t1, (select distinct 1 from t1) as sub(c);
I |V |C
----------------------------------
1 |row 1 |1
1 row selected
More specifically, this bug will show up when the sql layer does duplication elimination for distinct without a sorter (i.e. the underlying result set is already sorted). The result set will return correct data the first time, but after it is closed and re-opened it will not return any data rows.