On Tue, 2008-08-05 at 13:07 -0400, Tom Lane wrote:
> We could probably fix this by complicating the logic in ExecUnique,
> but I wonder whether it wouldn't be better to just stop treating
> Unique nodes as backwards-scannable.
No problem there.
> The only reason for that
> node type to exist (as opposed to using Group nodes) is that it's
> simple and low-overhead. So complicating it to support a corner case
> that no one has noticed in many years might be counterproductive.
> Thoughts?
I've never seen anyone scan backwards like this at all in practical use.
I knew it was possible, but never seen it done.
It seems entirely probable nobody else has either. It's a PostgreSQL
extension, so people arriving from outside don't even know it exists,
plus its always had bugs so those in-the-know don't use it either:
http://archives.postgresql.org/pgsql-bugs/1998-06/msg00049.php
My perceptions may not match others...
--
Simon Riggs www.2ndQuadrant.com
PostgreSQL Training, Services and Support
