[ruby-talk:11336] Re: numerical limits

[e-mail] (Raja S.) wrote:
>
>Is there a way in Ruby to determine the largest/smallest representable
>Fixnum and Float?  Since Ruby levarages off the underlying C system this
>will be architecture dependent.  I'm looking for something like 
>Fixnum::MAX,
>Fixnum::MIN and similarly for Floats.

You could add a method to do it for Fixnum based on
the existing size method.  See:

http://www.rubycentral.com/book/ref_c_fixnum.html

Floats don't seem to have any way to do it:

http://www.rubycentral.com/book/ref_c_float.html

Which is mildly ironic since the automatic conversion
to Bignum makes the limit irrelevant for Fixnum, but
what the limit is matters for Float.

>If it isn't currently builtin, it seems that as the language evolves, this
>may be one of the dotting the 'i's and crossing the 't's issue.

I am not seeing it at the moment, but is there any
equivalent to Perl's Configure module?  In Perl this
gives in one place all of the information that Perl
discovered while being configured before being built.
This is then available and is used (see
ExtUtils::MakeMaker) for writing makefiles for all
following modules.  The result allows policy decisions
made while installing Perl to be transparently used by
all CPAN modules.

If Ruby doesn't have this, it would be very nice.  I
have seen RAA applications that just assume that they
know where they should install on my system, making it
hard to test them out with a private user installation
first.  That is the kind of thing that Configure is
meant to detect and avoid.

A sample reason to use this would be to test how a bunch
of applications will work with a new development version
of Ruby *without* having to replace your installed main
version...

Cheers,
Ben
_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com