On Sat, Dec 8, 2012 at 3:35 PM, Mark H Weaver <[email protected]> wrote:

>
> Most importantly, 'eqv?' is now completely unspecified for non-real
> inexact numbers.  The R6RS used the operational equivalence language for
> _all_ inexact numbers, but here you have restricted those clauses to the
> inexact reals.
>

Sorry, the editors are very, very tired and missed this.
I will restore the definition to refer to all inexact numbers.


> This wording also does not address the many known problems with the R6RS
> definition, such as the NaN problem that you discovered


I misread the definition - I thought it had included my
NaN fix but it didn't, I'll restore that as well.

-- 
Alex
_______________________________________________
Scheme-reports mailing list
[email protected]
http://lists.scheme-reports.org/cgi-bin/mailman/listinfo/scheme-reports

Reply via email to