On 17 Oct 1999 22:20:24 CDT, the world broke into rejoicing as
Rob Browning <[EMAIL PROTECTED]>  said:
> [EMAIL PROTECTED] writes:
> 
> > (The Even Cooler Idea would be to use continued fractions, but
> > everyone is *certainly* allowed to trash that idea as ludicrously
> > impractical... That only works if we rewrite the whole thing in CL
> > :-)).
> 
> Or how about gmp?  We'd have arbitrary precision math that's
> apparently quite fast given what it does :>
> 
> I haven't meesed with it other than to look at it for stylistic hints
> for some guile work I'm doing, but from that investigation, it looked
> like it was plenty sophisticated.  Might be hitting a gnat with a
> cannon though...

The suggestion wasn't intended to be terribly serious.  Straightforward
fractions would be a not-too-bad idea, particularly if a parameter
gets tossed in to handle the consideration that currencied amounts
may need some shifting of decimal places.

The situation where the size of the value (e.g. - how "multiple" the
"multiple precision" is) results in the structure in which amounts
are stored varying in size depending on the value.

Essentially, this makes the size of the basic currency value dynamic,
when it would really be preferable, in a non-dynamic language like C,
to keep the representation static.

If we were writing the whole system in Scheme, it might be reasonable
to have a funky numeric type that can vary in its representation; when
the engine is in C, that won't fly nearly so well.
--
This program posts news to billions of machines throughout the galaxy.  Your
message will cost the net enough to bankrupt your entire planet.  As a result
your species will be sold into slavery.  Be sure you know what you are doing.
Are you absolutely sure you want to do this? [yn] y
[EMAIL PROTECTED] <http://www.hex.net/~cbbrowne/lsf.html>

--
Gnucash Developer's List 
To unsubscribe send empty email to: [EMAIL PROTECTED]

Reply via email to