Re: [Chicken-users] update: complex egg

[Thomas Chust]

Hello,

TiM wrote:
> greetings programs.
>
> i have a few questions about the complex egg.
>
> first, wouldn't it be slightly more efficient to represent a complex
> number as a length 2 vector?

I don't really know how this is implemented in CHICKEN, but in many 
Schemes cons-cells are more time efficient than vectors or equally 
efficient. And of course, the acces procedures take less time to type ;)

>
> secondly, don't we need some sort of type identifier to distinguish a
> complex number from a pair, or a length 2 vector?
>
> i'm thinking of something like ('complex . (a . b)) or ('complex . 
#(a > b)).

It is not strictly necessary to have such a distinction, although it 
would be a bit cleaner code.
It would probably be possible to make the complex data type a tinyclos 
class for example or to use a custom record data type, as Felix 
suggested. The main reason I did it the way it is implemented, is that I 
wanted a high degree of portability across Scheme (and other Lisp 
dialects) implementations.

>
> i would dearly like to see chicken have a complete numeric system. is
> it possible that this complex egg could be used to give chicken native
> (perhaps optional?) support for complex numbers?

It would be possible, but for native support of complex numbers I would 
strongly suggest adding a new primitive data type to CHICKEN and using 
the C library routines for modification of complex numbers internally, 
because this would bring a huge performance gain compared to my native 
Scheme implementation.

> [...]

cu,
Thomas