Re: Creating and using arrays - again

In article
<53ab1684-c1ce-4305-9716-6dacceebe900@[EMAIL PROTECTED]
>,
Bruce McFarling  <agila61@[EMAIL PROTECTED]
> wrote:
>
>"Make this and it does that" is only real constraining if you really
>want something that can do multiple things, and then "Make this and it
>exposes its hooks for its actor words to work with" can be built with
>"Make this and it does that", and if that's not enough, then you want
>an Object Oriented programming system, where the individual pieces can
>then be built in terms of one of those two foundation patterns.
>
>And in terms of semantics:
>
>CREATE ... { ; | ;CODE ... END-CODE | DOES> ... ; }
>
>... ramifies to quite a lot of additional capacity for very few
>words ... given that CREATE is normally a factor of ``:'' that brings
>the semantics into scope and ;CODE is often a factor of CODE, so that
>it comes for free once you have ``CODE'' ... ``END-CODE''.

;CODE is similar to DOES> in that it stores some pointer into the
header of the latest defined word. This is trivial, but has very
little to do with code .. end-code.

Of course the assembler, that is needed for the dots at
``;code'' ... ``end-code'' is a non-trivial amount of work.

>
>But thank goodness that its not common for the C-based Forth's to
>compile C-code in a CODE ... END-CODE segment. ... or maybe not? Would
>that be a way to seduce C-programmers away from the dark side?

That would be hard to imagine, using a c-compiler interactively
within a Forth session. Also c-code needs a lot of environment,
assumes definitions, declaration, symbolic constants whatnot.
Just divining a way of accompli****ng this is hard.
Then the user manual would be complicated enough to shy away
both Forth and c-programmers.

Groetjes Albert


--
-- 
Albert van der Horst, UTRECHT,THE NETHERLANDS
Economic growth -- like all pyramid schemes -- ultimately falters.
albert@[EMAIL PROTECTED]
 &=n http://home.hccnet.nl/a.w.m.van.der.horst