Re: Patch to fix FK-related selectivity estimates with constants

I wrote:
> Over in the thread at [1] it's discussed how our code for making
> selectivity estimates using knowledge about FOREIGN KEY constraints
> is busted in the face of EquivalenceClasses including constants.
> ...
> Attached is a patch series that attacks it that way.

I'd failed to generate a test case I liked yesterday, but perhaps
the attached will do.  (While the new code is exercised in the
core regression tests already, it doesn't produce any visible
plan changes.)  I'm a little nervous about whether the plan
shape will be stable in the buildfarm, but it works for me on
both 64-bit and 32-bit machines, so probably it's OK.

			regards, tom lane