[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Vectors as functions




On Monday, August 18, 2003, at 09:34 AM, Joe Marshall wrote:

>
> Funny you should mention that.....
>
> I wrote the Rebol 1.0 interpreter.  (but left the company right 
> afterwards)
>
> Rebol is *extremely* difficult to compile in any meaningful sense.
> Since the parse tree can change dynamically at runtime, there is
> a combinatoric explosion of code paths.  You can constrain
> things a bit by allowing the user to declare procedure arity, though.
>

I think it is cool. You get lisp like power without all the parens. 
Problem is you can't parse it statically.

  It is interesting that the discussion here has veered onto these 
topics because I've been toying with implementing a language that 
interprets like Rebol, executes like Icon and has first class scopes. 
I'm not interested in speed or optimization at all, just power of 
expression.

The Rebol-like interpretation does cause some trouble for Icon-like 
control flow, because if an expression fails you might not know how the 
rest of the program would have been parsed had it not failed. But there 
seems to be a simple way around this by using brackets.

-- 
--- james mccartney


-- 
--- james mccartney   james@audiosynth.com   <http://www.audiosynth.com>
SuperCollider - a real time audio synthesis programming language for 
MacOS X.