Re: Confirmation of bad query plan generated by 7.4

"Jim C. Nasby" <jnasby@pervasive.com> writes:
> On Tue, Jun 13, 2006 at 06:04:42PM -0400, Tom Lane wrote:
>> It'd depend on the context, possibly, but it's easy to show that the
>> current planner does fold "now() - interval_constant" when making
>> estimates.  Simple example:

> Turns out the difference is between feeding a date vs a timestamp into the
> query... I would have thought that since date is a date that the WHERE clause
> would be casted to a date if it was a timestamptz, but I guess not...

Hmm ... worksforme.  Could you provide a complete test case?

            regards, tom lane