From: Dennis Ranke <exoticorn@...>
Date: 2003-12-24T22:01:53+09:00
Subject: Re: Ruby 1.8.1 preview4

Hi NaHi,

> On your ruby, "-0.0" shoud be interpreted 0.0.  Gotoken hacked and
> told me that if I expect "-0.0" to be interpreted as negative zero
> in ruby,
> "echo 'main(){double a = 0; a = -a; puts(1/a < 0 ? "n" : "p");}' |
> gcc -xc - && ./a.out"
> must dump "n".  May be "p" on your system?

Yes, it's "p".

> I can use "-1.0 / (1.0 / 0.0)" instead of "-0.0" for soap4r test.
> How does this dump?
>
> $ ruby -e 'p(-1.0 / (1.0 / 0.0))'

-0.0

and '1.0 / (-1.0 / (1.0 / 0.0))' is

-Infinity

> "float" in XML Schema Datatypes (and SOAP) is a single float.
> Ruby's is double.  So I try to "narrow"? the given double float
> to single with
>
> d = given_double # e.g. 1.4e-45
> f = (/\A\0*\z/ =~ [d].pack("f")) ? 0.0 : d
> sprintf("%.10g", f) # use this as a value of "float"
>
> Maybe I misunderstand something...
>
> $ ruby -e 'p([1.4e-45].pack("f"))'
>
> ?

That correctly gives "1.4e-45", while [1.4e-45].pack('f') is 
"\000\000\000\000".

Now, since the double is indeed converted to a single-precision float, this 
is a known limitation of the Playstation 2 hardware which doesn't support 
denormalized floats. The smallest (positive, non-zero) normalized float is 
2**-126, and all results smaller than that just (silently) underflow to 
zero.

The smallest positive non-zero single-precision float is exactly 2**-149, 
so you could change your code to:
f = (d.abs < 2**-149) ? 0.0 : d

>> Yes, and "unknown host" with "ping localhost", so i added localhost to 
>> my hosts file, which got rid of those errors... ;)
>
> :)  Thanks.
>
> There shoudl be some architecture/settings where "localhost" is
> not defined.  But I don't know portable way to test server/client.

Well, it seems that I was the only one to stumble across that up to now 
(and it was easily solved on my side), so I don't really think it is a big 
issue right now.

Oh, it's the 24th... Merry Christmas, everyone :)

-- 
exoticorn/farbrausch