Re: New approach to ye olde cross-datatype indexing problem

Tom Lane <tgl@sss.pgh.pa.us> writes:

> The other idea I had was to somehow create associations of opclasses,
> so that you could say that "int2_ops, int4_ops, and int8_ops all sort
> values compatibly, so it's okay to go find the operator you need in
> int4_ops or int2_ops".  I don't have any concrete idea about how to
> represent that though.  Any thoughts about that, or other ideas
> altogether?

At the risk of saying something obvious and not understanding the scope of the
problem, you add a column pg_amop.superclass that is a unique integer except
for groups of operators like this where they can all share the same value.

That's assuming you only need a transitive relationship, which seems like what
you need.

-- 
greg