Re: Lispification of Forth

In article <f6f37213a3429712a0ec487175a74360@www.novabbs.com>,
mhx <mhx@iae.nl> wrote:

On Tue, 27 May 2025 20:23:02 +0000, minforth wrote:
>

It might be less complicated to use tokens or hashes
instead of memory addresses in high-level Forth code.
All dictionaries, data or VM code sequences could then
be localised exclusively via a central token table or
hash table. Replacing/modifying a word would be very
simple by patching the token/hash table.

>
Not when a word is vectored and/or using ` ' woord , ` and the like.

"less complicated" . My idea is not complicated, but adding
hash codes is definitively makes absolutely no sense,
It definitively makes it more complicated.

And especially not if the very purpose of the execercise is
to constantly transform each individual field of a dictionary entry,
to get that dynamic lisp feel, that you can change the editor
that you are using.

A typical step after lispification:
Enhance each dictionary entry with one field:the source field.
Decompile and store the result as a string in the heap, now let the
source field point to the string.
Now there is a 2DROP in the decompilation . I want to replace 2DROP by
PDROP. It is easy to do that. Replace everywhere in the source fields.
Replace the name of `2DROP "2DROP" with "PDROP".
Done. The Forth is transformed.

Groetjes Albert
--
Temu exploits Christians: (Disclaimer, only 10 apostles)
Last Supper Acrylic Suncatcher - 15Cm Round Stained Glass- Style Wall
Art For Home, Office And Garden Decor - Perfect For Windows, Bars,
And Gifts For Friends Family And Colleagues.