Re: Creating and using arrays - again

On Mar 22, 10:29 am, Jonah Thomas <jethom...@[EMAIL PROTECTED]
> wrote:
> If you're willing, look at how you'd get nearly the same effect without
> DOES> . There are a few Forths that don't give you DOES> and someday you
> DOES> might use one and need to work around it. And it shows you just
> DOES> what DOES> gives you, when you look at the alternative.

Define "need to" as used in this sentence. (!)

More im****tant is when you want to do something else with the item
you've created, where a passive item that just places its access info
on the stack is easier to work with than ticking an active item to use
``>BODY''.

Cf.

ONLY-FORTH
WORDLIST CONSTANT Nicl-tools

versus

ONLY-FORTH
VOCABULARY Nicl-tools

I want to place Nicl-tools into the Nicl hierarchy as the "tools"
wordlist, its just:

ONLY-FORTH
Nicl SET-CURRENT
Nicl-tools CONSTANT tools

If its vocabulary, there might be a wordlist-id at

' voc-name >BODY

.... or there might not be, so to have a ****table library oversight
system, I would need to implement my own VOCABULARY inside Nicl-tools,
and Nicl-tools would have to be bootstrapped as a vocabulary of my
sort to avoid interfering with the local system VOCABULARY.

Now, DOES> is a great word, and, for example, its a convenience to be
able to make a vocabulary

: make-vocabulary ( wordlist-id -- ) CREATE ,
    DOES> >R GET-ORDER R> ROT DROP SET-ORDER ;

.... but, just like parsing words and non-parsing words, sometimes its
handy to build an "active" word on top of a "passive" word that has
the action, "this is who/where I am, do what you want about it".