Tom Lane wrote:
> Bruce Momjian <pgman@candle.pha.pa.us> writes:
> > This has been saved for the 8.2 release:
> > http://momjian.postgresql.org/cgi-bin/pgpatches_hold
>
> Uh, why do we need this at all? "NOT (tid = tid)" covers the
> functionality already.
tid should be a fully functional type, at least for = and !=.
> I disagree strongly with renumbering existing hand-assigned OIDs for
> this. There's too much risk of breakage and no benefit.
Agreed.
> Also, you forgot to add the negator cross-links between the operators.
OK.
--
Bruce Momjian | http://candle.pha.pa.us
pgman@candle.pha.pa.us | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ Christ can be your backup. | Newtown Square, Pennsylvania 19073