Re: mpz_cmp_d and NaNs in = and <

[Marius Vollmer]

Kevin Ryde <address@hidden> writes:

> I applied this change.  I left in the tests on complex numbers in `=';
> someone smarter than me will have to advise whether it's true in fact
> that they're not reached (on account of the imaginary part always
> being non-zero).

I think that a lot of places expect it to be true that all complex
number objects in Guile have a non-zero imaginary part.  But your
conservative approach is OK, nevertheless, and should probably remain
that way.  For example, we can't currently represent negative
imaginary zero.  we might want to change that, maybe.

-- 
GPG: D5D4E405 - 2F9B BCCC 8527 692A 04E3  331E FAF8 226A D5D4 E405