Re: Type checking (was: The strengths of FORTH)

"Bruce McFarling" <agila61@[EMAIL PROTECTED]
> schrieb im Newsbeitrag 
news:6c807d21-1774-48f5-b10e-72801888c99c@[EMAIL PROTECTED]
> On Apr 9, 2:49 am, "Stephan Becher" <stephan.remove-this.bec...@[EMAIL PROTECTED]
> online.de> wrote:
>> Okay. Let me summarize how I interpret your (and Bernd's and Jonah's)
>> suggestions.
>
> Almost exactly.
>
> The only slight difference in what I envisioned was that the system
> would re-use the Forth-94 StrongForth scaffolding to export a word
> into StrongForth, and as you describe it, after switching into FORTH
> to load the Forth-94 library, you'd switch back and use a StrongForth
> tool to import the word.
>
> So, after its defined in Forth-94, export from Forth-94 or import into
> StrongForth?
>
> The only different that I can see is that importing with a SrongForth
> wordset that is itself operating under the strong static typing seems
> more secure, so, IOW:
>
> ... yes, what you said.

I think it's easier to import into StrongForth than to export from
Forth-94, 
because StrongForth already has all the words that are required to create 
stack diagrams. The syntax I have in mind is similar to StrongForth's 
version of ALIAS:

'ANS < IMPORT < ( SIGNED 1ST -- FLAG )
'ANS U< IMPORT < ( INTEGER 1ST -- FLAG )

Note than < is overloaded in StrongForth; INTEGER is an unsigned data
type.

Stephan