Re: Inaccurate results from numeric ln(), log(), exp() and pow()

BTW, something I find confusing and error-prone is that this patch keeps
on using the term "weight" to refer to numbers expressed in decimal digits
(ie, the approximate log10 of something). Basically everywhere in the
existing code, "weights" are measured in base-NBASE digits, while "scales"
are measured in decimal digits. I've not yet come across anyplace where
you got the units wrong, but it seems like a gotcha waiting to bite the
next hacker.

I thought for a bit about s/weight/scale/g in the patch, but that is not
le mot juste either, since weight is generally counting digits to the left
of the decimal point while scale is generally counting digits to the
right.

The best idea that has come to me is to use "dweight" to refer to a weight
measured in decimal digits. Anyone have a better thought?

regards, tom lane