Tom Lane wrote:
> Alvaro Herrera <alvherre@2ndquadrant.com> writes:
> > I can, of course, just leave these files well enough alone and just rely
> > on the numbers not changing (which surely they won't anyway) and
> > remaining consistent with numbers used in new opclasses (which surely
> > they will lest they be born inconsistent with existing ones).
>
> Yeah, we do have checks in opr_sanity which will complain if inconsistent
> strategy numbers are used for similarly-named operators. That's a pretty
> weak test though, and operator names aren't exactly the right thing to
> check anyway. I'd be good with pushing all of that stuff to a new central
> header.
So here's a patch for this.
--
Álvaro Herrera http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services