It could have been a rounding problem, but it's not , the following digits are 001. The serie is tailored on purpose, to have a short fraction.
It is based on 12345679/999999999 -> (1/81) You can also try to evaluate 9801 sqrt in a recent squeak or pharo, then 998001 sqrt. Nicolas 2012/1/27 Chris Cunningham <[email protected]>: > On Fri, Jan 27, 2012 at 2:08 AM, Norbert Hartl <[email protected]> wrote: >>... >> I think I've found a bug. Mine ends with >> >> ... >> >> 990991992993994995996997999000s3000 >> >> You can see after 997 there is 999. :) >> > I suspect it should be ...997998999 and not ...997999000. > The 'issue' is that ScaledDecimal rounds the the number off that is is > printing. > You can find this in ScaledDecimal>>printOn: > > -Chris >
