2014-07-11 17:14 GMT+02:00 Camille Teruel <camille.ter...@gmail.com>:
> > On 11 juil. 2014, at 17:02, Natalia Tymchuk <natalia.tymc...@unikernel.net> > wrote: > > But why > 185/10 = 18.5 -> true? > > > Because 18.5 has an exact float representation, while 0.2 hasn't. > > Couldn't we use #asApproximateFraction instead of #asTrueFraction in > Float>>adaptToFraction:andCompare: ? > No, it would be a bad thing. For example, it would break the fact that = should be an equivalence relation (a=b && (b=c) ==> (a=c)). See for example http://bugs.squeak.org/view.php?id=3374 for a genesis of this feature. And note that lispers have preceded us by many years. > 0.2 asApproximateFraction ==> (1/5) > 0.2 asTrueFraction ==> (3602879701896397/18014398509481984) > > > There is also asMinimalDecimalFraction or something like that. > > On Jul 11, 2014, at 4:40 PM, Norbert Hartl <norb...@hartl.name> wrote: > > > Am 11.07.2014 um 16:33 schrieb Yuriy Tymchuk <yuriy.tymc...@me.com>: > > > On 11 Jul 2014, at 16:30, Esteban Lorenzano <esteba...@gmail.com> wrote: > > On 11 Jul 2014, at 16:22, kilon alios <kilon.al...@gmail.com> wrote: > > why not round it before comparing it ? looks like an easy enough problem > to fix . I would expect that to be equal. > > > because it wouldn’t be correct :) > > > But then why 4/2 = 2? > > > The same reason why > > 4/3 = 8/6 > > :) > > Norbert > > > > > > On Fri, Jul 11, 2014 at 5:18 PM, Yuriy Tymchuk <yuriy.tymc...@me.com> > wrote: > >> That’s why mathematics and programming are two different things… >> >> >> On 11 Jul 2014, at 16:08, Esteban A. Maringolo <emaring...@gmail.com> >> wrote: >> >> > Usually comparing against floats is not as deterministic as you would >> > expect. And in other dialects you use #equals: instead of #= >> > >> > But in Pharo there is no #equals: and instead there is a #closeTo:, >> > but #closeTo: has an arbitrary decimal precision, more than enough for >> > commong arithmetic. >> > >> > Regards! >> > >> > Esteban A. Maringolo >> > >> > >> > 2014-07-11 10:59 GMT-03:00 Goubier Thierry <thierry.goub...@cea.fr>: >> >> :) >> >> >> >> in Smalltalk, the division of two integers is a fraction, not a float. >> >> >> >> i.e. 1/5 is 1/5. >> >> >> >> Thierry >> >> >> >> Le 11/07/2014 15:53, Natalia Tymchuk a écrit : >> >>> >> >>> Hello. >> >>> I found interesting thing: >> >>> Why it is like this? >> >>> >> >>> Best regards, >> >>> Natalia >> >>> >> >> >> >> -- >> >> Thierry Goubier >> >> CEA list >> >> Laboratoire des Fondations des Systèmes Temps Réel Embarqués >> >> 91191 Gif sur Yvette Cedex >> >> France >> >> Phone/Fax: +33 (0) 1 69 08 32 92 / 83 95 >> >> >> > >> >> >> > > > > > >
