Re: Reducing the overhead of NUMERIC data

Simon Riggs <simon@2ndquadrant.com> writes:
> On Thu, 2005-11-03 at 10:32 -0500, Tom Lane wrote:
>> I think we could make it go by cramming the sign and
>> the high-order dscale bit into the first NumericDigit --- the
>> digit itself can only be 0..9999 so there are a couple of bits
>> to spare.

> I've got a working version of the code using the above scheme,

Really?  After I woke up a bit more I realized there was only one bit
and change to spare, not two, so I don't see how it would work.

(Unless you want to force a minimum of two NumericDigits when the dscale
exceeds 255, and flag the scale in the second one --- which might be OK
but it's getting messier and messier...)
        regards, tom lane