On Wed, Nov 26, 2003 at 11:59:33AM -0500, Neil Conway wrote:
> > In some situations that doesn't really matter, as the same plan
> > would have gotten picked anyway.
>
> The hint is NOT "the chosen plan was non-optimal"; the hint is "the
> query planner did not produce an accurate row count estimate for this
> node in the query tree."
Maybe it could only be done for SeqScan and IndexScan nodes, which are
probably the most common source of bad estimates related to poor
statistics.
--
Alvaro Herrera (<alvherre[a]dcc.uchile.cl>)
"At least to kernel hackers, who really are human, despite occasional
rumors to the contrary" (LWN.net)