On Tue, 2006-06-06 at 10:27 -0400, Tom Lane wrote:
> Simon Riggs <simon@2ndquadrant.com> writes:
> > Why not implement an array of option parameters on pg_class, so both
> > heaps and indexes can be given additional parameters? That way you
> > wouldn't need a specific relfillfactor attribute. That would allow us to
> > keep CREATE TABLE free of additional keywords also.
>
> None of this should go anywhere near pg_class. IIRC the solutions we
> discussed involved adding some sort of array to pg_index.
Itagaki had suggested adding options to heaps also, so clearly we'd need
to add that to pg_class, rather than pg_index in that case.
PCTFREE would be useful for heaps as well as indexes, but there could be
other options also. Extending the thought for the general case, I see no
reason why we would want to permanently exclude heaps from having a more
flexible set of options when we aim to provide that for indexes.
-- Simon Riggs EnterpriseDB http://www.enterprisedb.com
