On Wed, Feb 4, 2015 at 7:22 PM, Chuck Hipschman <ckhipschman@gmail.com> wrote:

7 / 8.0 roundTo 0.01 = 0.88 Aha! Thanks!

But:

100 * (1.05 raisedTo: 10) roundTo: 0.01 162.89000000000001
"162.88946267774418" "unrounded"
100 * (1.05 raisedTo: 15) roundTo: 0.01 207.89000000000001
"207.8928179411367" "unrounded"
Bug, or me again :-)

If you debug into #roundTo: you'll see its implementation is simple arithmetic, subject to the vagaries of float resolution.

Float(Number)>>roundTo:

^(self / quantum) rounded * quantum

Now are you wanting something nicely formatted for display, or are you rounding for some mathematic purpose? Perhaps instead you are wanting...

100 * (1.05 raisedTo: 10) printShowingDecimalPlaces: 2 " --> 162.89 "

0.105 printShowingDecimalPlaces: 2 " --> 0.10 "

0.115 printShowingDecimalPlaces: 2 " --> 0.12 "

btw, I only found these today, looking in the printing protocol of Number and Float.

cheers -ben