From: the.beez.speaks@gmail.com
On 27-04-2025 08:21, dxf wrote:
> On 26/04/2025 9:07 pm, albert@spenarnc.xs4all.nl wrote:
>> In article ,
>> dxf wrote:
>>> On 26/04/2025 2:34 am, Hans Bezemer wrote:
>>>> ...
>>>> Yeah, I have helped to make a proposal for PLACE and +PLACE - which never
even made it to the voting stage.
>>>
>>> It's a nice symmetry. OTOH the remaining vendors use APPEND and why
should they change?
>>>
>>
>> $+! was even earlier. It predates the IBM PC XT.
>> (Osborne, CP/M)
>
> Even PLACE was new back then!
>
> String stacks often had $+ or equiv. Somehow I never took to them.
> Not enough applications that warranted the effort?
Let's face it - the string support was notoriously bad in Forth. People
openly complained about that.
So when I did 4tH I ported some of my string libraries in C to Forth,
https://sourceforge.net/p/forth-4th/code/HEAD/tree/trunk/4th.src
lib/replace.4th
Never used it. Ok, maybe once or twice. Because all I actually needed
was PLACE and +PLACE. You can compose a lot of things with those two.
Other frequently used tools: CLIP ( a n -- a n-1) and CHOP ( a n - a+1
n-1). Tip: see how often you use 1 /STRING - that'll give you an idea.
And that from a man who has done some text processing in his life..
>NUMBER? Never use it - too weird. My own NUMBER is great. SEARCH?
Never use it. Rather use my own INSTR routine that simply returns a
position or a pointer. Doesn't need COMPARE either. BTW, COMPARE is
rather useful..
So with a few minor additions there is nothing wrong with the text
processing capabilities of Forth, actually.
Just a shame the very basics are missing. The only thing holding Forth
back is the refusal to abstract strings. Wanna make it a counted string
with a cell count? Do it. Wanna keep on doing counted strings? Do it.
Wanna do ASCIIZ strings? No problem.
All you need is a word that initializes a string - like PLACE. A word
that appends to an existing string, +PLACE. And a word that converts a
string to a string/count, like COUNT. Add a word that converts a
character to a string: C>S:
32 string MyStr \ create a string that can hold 32 chars
s" This is my life" MyStr PLACE
char ! C>S MyStr +PLACE
Another handy word is >STRING that converts a "raw" string to a fully
qualified string ( a1 n1 -- a2) that can handle a COUNT (you have the
carnal knowledge - you take care of the "how"). E.g. for counted strings
it might be:
: >STRING swap over over dup char+ rot cmove c! ;
Yeah, you take care of the available string capacity. It's Forth, baby!
Now, how much wiggle room do you need additionally to do your very
special string stuff? Especially when we had this wonderful CHAR+ and
CHARS abstractions that you so carelessly discarded?
Sure, it might give rise to a whole explosion of string formats - but
does that matter if they're all compatible at the abstraction level?
I have introduced a light dynamic string format in 4tH a while ago. It
only contains the equivalents of PLACE, +PLACE and COUNT - and I'm
still astonished how versatile it is. C>S is still useful, but I haven't
needed >STRING yet.
Take this example from FFL est, which I had to change from:
end-of-line count tuck r@ count + swap cmove r@ c@ + r@ c!
To:
0 end-of-line count tuck r@ count + dup >r swap cmove chars r> + c!
All can be replaced (portably) by:
end-of-line r@ +place
That's it. Works for you, works for me. It's much easier.
Hans Bezemer
--- SoupGate-DOS v1.05
* Origin: you cannot sedate... all the things you hate (1:229/2)
|
