Re: Allowing NOT IN to use ANTI joins

On Mon, Jun 09, 2014 at 12:36:30AM +1200, David Rowley wrote:
> Currently pull_up_sublinks_qual_recurse only changes the plan for NOT
> EXISTS queries and leaves NOT IN alone. The reason for this is because the
> values returned by a subquery in the IN clause could have NULLs.

Awesome. I've had a brief look at the patch and other than a line of
extraneous whitespace it looks sane.

Since it is only testing on NOT IN queries I don't think there are any
issues with it slowing down simple queries.

I also note you can't prove "id+1" not null. At first I thought you
might be able to prove this not null if the operator/function was
strict, but then I realised that strict only means "null if input is
null" not "output is only null if inputs are null". Pity.

Nice work.
--
Martijn van Oosterhout <kleptog(at)svana(dot)org> http://svana.org/kleptog/
> He who writes carelessly confesses thereby at the very outset that he does
> not attach much importance to his own thoughts.
-- Arthur Schopenhauer