Re: [Chicken-users] callbacks without having a C_toplevel()

[Daniel B. Faken]

On Thu, 25 Aug 2005, Daniel B. Faken wrote:
>   I'm switching from using my own version of CHICKEN_eval_string()
> & friends to the builtin one, which is now a callback.
>   When I was using my own version, I defined it in the normal top-level 
> scheme file, and my eval-fn had access to all the units I included there.
>   Now that I've switched, they don't (I believe they have access to the 
> same units as the 'eval' library, since that is where the "boilerplate" 
> embedding API now resides).
>   This is not a problem - I can just call
>  CHICKEN_eval_string("(declare (uses my-funny-units) ...)").

I take this back.. It seems that macros are not available
 -- not even require-extension! -- in CHICKEN_eval_string() & friends.

(require 'ID) works, but if I try (include "chicken-more-macros"),
'include' is not defined either.

I know this is a tricky issue (deciding what is available without
requiring include/require/require-for-syntax), but I assume 
this is not the desired behaviour?

So, maybe I'll just go back to my own 'eval', over which I have full 
control (BTW, I don't think the boilerplate stuff handles multiple 
s-exprs).


On another note: (require 'library) fails, even in csi:
  _______   _     __
 / ___/ /  (_)___/ /_____ ___
/ /__/ _ \/ / __/  '_/ -_) _ \
\___/_//_/_/\__/_/\_\\__/_//_/

Version 2, Build 106 - linux-unix-gnu-x86 - [ dload ]
(c)2000-2005 Felix L. Winkelmann
#;1> (require 'library)
Error: (require) can not load extension: library


cheers,
Daniel