Re: reducing NUMERIC size for 9.1

Robert Haas <robertmhaas(at)gmail(dot)com> writes:
> On Fri, Jul 30, 2010 at 1:13 AM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>> Perhaps, but I think you're confused on at least one point.
>> numeric(2,1) has to be able to hold 2 decimal digits, not 2
>> NumericDigits (which'd actually be 8 decimal digits given
>> the current code).

> I get that. The point is: if one of those 2 decimal digits is before
> the decimal point and the other is after it, then two NumericDigits
> will be used.

Ah, I see. Maybe we should allow for one more NumericDigit in the
calculation for such cases. I guess you could look at the scale too
to detect if the case is possible, but not sure it's worth the trouble.

regards, tom lane