Taylan Ulrich B. scripsit:

> > Note that the behavior of eqv? where either argument is NaN and the
> > other argument is inexact is deliberately left unspecified.
>
> Is this a typo?  Both the wording and the examples in draft 9 seem to
> indicate that for one NaN and one inexact, the result is #f; and for two
> NaNs it is unspecified.  I'll continue on that assumption.

Yes, of course that's what I should have said.

> I would move the two-NaNs situation to a side-note of some sort:
>
>      obj_1 and obj_2 are both inexact numbers such that they are
>      numerically unequal (in the sense of =) or they do not yield the
>      same results (in the sense of eqv?) when passed as arguments to any
>      other procedure that can be defined as a finite composition of
>      Scheme’s standard arithmetic procedures which does not result in a
>      NaN value.  As an exception, the behavior of eqv? is unspecified
>      when both obj_1 and obj_2 are a NaN value.

Adopted, with the following edits:

    for "they are" read "either they are"

    for "which" read ", provided that" (and likewise in the
    affirmative paragraph)

    for "are a NaN value" (which might suggest that they have to be
    the same NaN value) to "are NaN".

Thanks.

--
He played King Lear as though           John Cowan <cowan@ccil.org>
someone had played the ace.             http://www.ccil.org/~cowan
        --Eugene Field

--
You received this message because you are subscribed to the Google Groups "scheme-reports-wg1" group.
To unsubscribe from this group and stop receiving emails from it, send an email to scheme-reports-wg1+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.