IRC Log - 2025-05-22

01:03:22 ##forth <vms14> where do s" and c" store the string in memory?

01:03:30 ##forth <vms14> they do not seem to increment here

01:03:44 ##forth <vms14> I'm incrementing here with both

01:05:36 ##forth <vms14> they do not store it in the pad either

01:05:42 ##forth <vms14> at least in gforth

01:06:57 ##forth <vms14> it is wrong to increment here for s" and c"?

01:25:45 ##forth <forthBot> Environment for cleobuli_ inactive, freeing...

02:26:33 ##forth <xentrac> vms14: typically they store it in the colon definition

02:26:35 ##forth <xentrac> : hello s" goodbye" ; ' hello 48 dump

02:26:35 ##forth <xentrac> 7F7BF76C34A8: BC 7E 10 34 CA 55 00 00 - 00 00 00 00 00 00 00 00 .~.4.U..........

02:26:38 ##forth <xentrac> 7F7BF76C34B8: D9 80 10 34 CA 55 00 00 - D0 34 6C F7 7B 7F 00 00 ...4.U...4l.{...

02:26:41 ##forth <xentrac> 7F7BF76C34C8: 67 6F 6F 64 62 79 65 20 - BE 82 10 34 CA 55 00 00 goodbye ...4.U..

02:28:44 ##forth <xentrac> (in gforth)

02:38:28 ##forth <vms14> yeah but outside them they seem to use what they call "a transient buffer" which is not pad and will be reused the next time s" or c" are called

02:39:05 ##forth <vms14> my pad is actually from 0 to 300 since memory is just an array, here starts at 300

02:39:29 ##forth <vms14> and colon definitions do not reside in memory, they are just a perl subroutine xd

02:40:19 ##forth <vms14> I should think about making a dictionary in memory, but for now I will keep having memory only for the rest of stuff and the dictionary be a hash table

02:43:22 ##forth <vms14> that's the interpreter now https://paste.c-net.org/RemarkBreeds

02:44:13 ##forth <vms14> and the words I have for now https://termbin.com/y7fp

02:47:08 ##forth <vms14> it launches a repl if you load that with perl

02:52:25 ##forth <vms14> I should create DUMP

02:53:13 ##forth <xentrac> s" is called " in F83

02:53:29 ##forth <xentrac> in compilation state it stores the characters in the colon definition, same as GForth

02:54:43 ##forth <xentrac> in interpretation state it appends them to the dictionary anyway

02:54:54 ##forth <xentrac> just not as part of a dictionary entry necessarily

03:01:22 ##forth <vms14> then it increments HERE like I do

03:01:38 ##forth <vms14> cool

03:01:54 ##forth <vms14> I like it that way more because they can persist

03:02:25 ##forth <vms14> but maybe they shouldn't and that's why the space gets reused in gforth

03:02:40 ##forth <xentrac> well also it doesn't work

03:03:30 ##forth <vms14> what does not work

03:03:42 ##forth <xentrac> it doesn't push the address and length on the stack in interpret state

03:03:55 ##forth <vms14> in f83?

03:03:57 ##forth <xentrac> yeah

03:04:02 ##forth <vms14> oh

03:04:05 ##forth <xentrac> so there's no way to use the string you've thus stored

03:04:08 ##forth <xentrac> well

03:05:53 ##forth <xentrac> maybe if you saved the value of here from before storing the string you could find the counted string from it

03:07:39 ##forth <xentrac> like here " hello vms14" 2 + count type says "hello vms14 ok"

03:08:57 ##forth <vms14> I have ," for that

03:10:52 ##forth <vms14> " pushes the string directly on the stack or definition and the others work like in forth

03:11:18 ##forth <vms14> I have accept too xd

03:11:48 ##forth <vms14> I will try to see how much of starting forth code I can evaluate in this thing

03:11:55 ##forth <vms14> the F example works already

03:12:26 ##forth <vms14> https://paste.c-net.org/TypicalLarry

03:12:32 ##forth <vms14> although that was simple

03:15:28 ##forth <xentrac> this works in F83: here " yer mom" ' type , ' unnest , constant where : whoa where >r ; whoa printing "yer mom"

03:16:12 ##forth <xentrac> whoa is arranging to return into the code appended to the dictionary by " and the two calls to ,

03:16:22 ##forth <xentrac> whose location is at the constant where

14:23:18 ##forth <cleobuline> forthBot: LOAD "ini.fth"

14:23:18 ##forth <forthBot> File ini.fth loaded

14:23:34 ##forth <cleobuline> forthBot: MICRO 123456789123457 FPRIME? . MICRO SWAP - .

14:23:35 ##forth <forthBot> 1

14:23:35 ##forth <forthBot> 886487

14:23:50 ##forth <cleobuline> forthBot: MICRO 123456789123457 FPRIME2? . MICRO SWAP - .

14:23:50 ##forth <forthBot> 1

14:23:51 ##forth <forthBot> 604963

14:24:11 ##forth <cleobuline> less than 1 second

14:25:00 ##forth <cleobuline> forthBot: MICRO 123456789123457 PRIME? . MICRO SWAP - .

14:25:00 ##forth <forthBot> 1

14:25:01 ##forth <forthBot> 47

16:01:13 ##forth <cleobuline> bon

16:04:10 ##forth <vms14> I'm learning forth by implementing words

16:05:16 ##forth <vms14> but I should change the dictionary and have proper headers and body for words

16:16:03 ##forth <cleobuline> vms14: with assembler ?

16:16:08 ##forth <cleobuline> of c ?

16:16:17 ##forth <cleobuline> or c

16:19:37 ##forth <vms14> in perl xd

16:19:40 ##forth <vms14> https://paste.c-net.org/SpousesBerlin

16:20:18 ##forth <vms14> @memory[$addr .. $addr + $times] = ($element) x $times;

16:20:25 ##forth <vms14> that's the implementation for FILL

16:20:45 ##forth <vms14> 0 30 32 FILL

16:25:00 ##forth <forthBot> Environment for cleobuline inactive, freeing...

19:37:06 ##forth <cleobuline> forthBot: LOAD "ini.fth"

19:37:06 ##forth <forthBot> File ini.fth loaded

19:37:17 ##forth <cleobuline> forthBot: CREDIT

19:37:17 ##forth <forthBot> Brought to you by Cleobuline updated with hashtable https://github.com/cleobuline/forth-bot-gmp-irc-threaded-multi-users/tree/main Site https://labynet.fr

21:37:17 ##forth <forthBot> Environment for cleobuline inactive, freeing...

21:55:19 ##forth <vms14> I have created m" which is like s" but expects an address on the stack and puts the strirng there

21:56:01 ##forth <xentrac> huh, interesting!

21:56:19 ##forth <xentrac> so you can pad m"

21:56:27 ##forth <vms14> yes

21:56:41 ##forth <vms14> I should make a version with limit I guess

21:56:50 ##forth <vms14> but does not make sense

21:57:14 ##forth <vms14> at the end you are creating a literal string so you should know if it fits or not

21:57:32 ##forth <vms14> I assume it can be useful for data structures and alike

21:57:54 ##forth <vms14> I was starting to read Object-Oriented Forth: Implementation of Data Structures

21:57:57 ##forth <xentrac> I agree

21:58:23 ##forth <vms14> it shows a very simple way to create records with constants and does> + ;

21:58:56 ##forth <xentrac> I suppose that in compile mode it will be copying the characters of the string into the location it gets from the stack

21:59:03 ##forth <xentrac> from the colon definition

21:59:37 ##forth <vms14> 0 constant name 20 constant phone : record create 40 allot does> + ;

22:00:38 ##forth <vms14> record john-doe name john-doe m" John Doe"

22:00:47 ##forth <vms14> should work

22:00:52 ##forth <xentrac> you could imagine a simpler way to define the record structure

22:01:02 ##forth <vms14> yeah it will explain it later

22:01:08 ##forth <vms14> (the book)

22:01:22 ##forth <vms14> but it lead to the m" word which is cool

22:03:17 ##forth <vms14> should I name it !" ?

22:03:18 ##forth <xentrac> like : field create , does> @ + ; : record create 40 allot ; for example

22:03:26 ##forth <xentrac> !" sounds like a good name to me

22:03:27 ##forth <vms14> I used m" because of memory

22:03:33 ##forth <vms14> but store is !

22:03:36 ##forth <xentrac> yeah

22:03:51 ##forth <xentrac> should it return the address and length or not?

22:04:01 ##forth <vms14> does not return anything

22:04:02 ##forth <xentrac> I feel like it should at least return the length

22:04:26 ##forth <vms14> or put a counted string

22:04:33 ##forth <xentrac> counted strings were a mistake

22:04:41 ##forth <vms14> I was thinking on a variation for the counted string

22:04:44 ##forth <vms14> oh

22:05:14 ##forth <xentrac> but without returning the string length somewhere, if you want to have Forth type out the string, you have to count the bytes yourself and say name john-doe 8 type

22:05:52 ##forth <vdupras> In Dusk, I call it cmove" ( dstaddr -- a u ) because it does pretty much like cmove

22:06:13 ##forth <xentrac> is a dstaddr?

22:07:07 ##forth <vms14> why they were a mistake

22:07:33 ##forth <vms14> returning a number it's sad because it requires me to store it somewhere although you are right

22:07:39 ##forth <vms14> but a counted string solves both problems

22:07:49 ##forth <vms14> name john-doe count type

22:08:03 ##forth <xentrac> because if you have a regular string and you want to pass it to a word that wants a counted string, you have to allocate a buffer somewhere and copy the count and the string into it

22:08:15 ##forth <xentrac> also you have a bug if a string grows past 255 bytes

22:08:30 ##forth <xentrac> so I think it's a mistake to create words that want counted strings

22:08:37 ##forth <xentrac> unfortunately there is one of them in the Forth standard

22:08:47 ##forth <xentrac> find

22:09:26 ##forth <vms14> so the problem is wanting a counted string

22:09:28 ##forth <xentrac> the standard punched-card way to store variable-length strings in a fixed-width field is Forth's -trailing

22:09:54 ##forth <xentrac> you pad out the buffer with spaces and pass the buffer's address and size to -trailing

22:10:53 ##forth <xentrac> then you can type it

22:10:54 ##forth <xentrac> as a trivial example: : hi s" example " -trailing type ; hi example ok

22:10:56 ##forth <xentrac> oops

22:10:57 ##forth <vms14> like expect does

22:11:03 ##forth <xentrac> as a trivial example: : hi s" example " -trailing type ; hi example ok

22:11:23 ##forth <vms14> I was implementing expect today, filling the rest with spaces

22:11:47 ##forth <vms14> that's why in part I was wondering about expecting a limit for m"

22:11:51 ##forth <xentrac> I think accept obsoletes expect

22:11:53 ##forth <vms14> so it fills with spaces

22:12:01 ##forth <vms14> I have both accept and expect

22:13:58 ##forth <vms14> I guess m" should return a number or expect a limit and fill with spaces

22:14:20 ##forth <vms14> and I also want the counted string because it makes sense for john doe's name

22:15:03 ##forth <vms14> name john-doe m" John" name john-doe count type

22:15:22 ##forth <xentrac> what I did for string handling in Forth last millennium was to reimplement DJB's stralloc in Forth, on top of malloc/free

22:15:29 ##forth <xentrac> which I think are called allocate and free

22:15:29 ##forth <vms14> I guess I will use only the counted version for now and name it !"

22:15:39 ##forth <xentrac> yes

22:15:47 ##forth <vms14> because it's weird to not see ! when setting values

22:16:05 ##forth <xentrac> you will regret having used counted strings

22:16:08 ##forth <xentrac> ;)

22:16:09 ##forth <vms14> :/

22:16:29 ##forth <vms14> well right now I have to experiment with things so it's fine

22:16:40 ##forth <vms14> the next month all this code will likely not exist anymore

22:20:09 ##forth <cleobuline> hello at night here !

22:20:20 ##forth <vms14> hi cleobuline

22:20:32 ##forth <xentrac> somehow that makes me sad

22:20:39 ##forth <cleobuline> L'heure locale pour Paris, Île-de-France, France métropolitaine, France est : 2025-05-22 22:20:24 (Fuseau horaire : Europe/Paris)

22:20:55 ##forth <xentrac> c'est un fuseau!

22:21:06 ##forth <vms14> I think a counted string makes sense in this case because it embeds the number, but I will remember to avoid expecting them and using count always

22:21:07 ##forth <xentrac> in Spanish it's an "huso"

22:21:37 ##forth <vms14> xentrac sad that I will throw everything? it means I will rewrite it changing what I have learned experimenting now

22:21:39 ##forth <xentrac> which sounds exactly like "uso", as in "usage"

22:21:53 ##forth <xentrac> yes, probably I am unreasonably attached to things

22:22:02 ##forth <xentrac> thank God my wife makes me throw things out

22:22:07 ##forth <vms14> I cannot have attachment to a prototype

22:22:25 ##forth <vms14> when I learn what I'm doing and have a proper design then ok

22:22:41 ##forth <vms14> but now the code and the design suck and really has to change

22:23:13 ##forth <vms14> it is my playground where I learn, has been like that for a while

22:23:40 ##forth <cleobuline> forthBot: STRING TOTO

22:24:07 ##forth <cleobuline> forthBot: "S This is a string "S TOTO !

22:24:07 ##forth <forthBot> Unknown word: This

22:24:07 ##forth <forthBot> Unknown word: is

22:24:08 ##forth <forthBot> Unknown word: a

22:24:08 ##forth <forthBot> Unknown word: string

22:24:08 ##forth <forthBot> Error: STORE: Stack underflow for string value

22:24:20 ##forth <cleobuline> forthBot: S" his is a string "S TOTO !

22:24:39 ##forth <cleobuline> forthBot: TOTO @ PRINT CR

22:24:40 ##forth <forthBot> his is a string

22:24:40 ##forth <xentrac> cleobuline: you really need to fix that bug where it can generate multiple lines of error messages for a single line of input

22:25:37 ##forth <cleobuline> it's not a bug it just evaluate a string

22:26:07 ##forth <cleobuline> i can't predict that the input is buggy

22:26:19 ##forth <cleobuline> forthBot: prout

22:26:20 ##forth <forthBot> Unknown word: prout

22:29:19 ##forth <xentrac> it is a bug

22:29:24 ##forth <xentrac> forthBot: a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a

22:29:24 ##forth <forthBot> Unknown word: a

22:29:25 ##forth <forthBot> Unknown word: a

22:29:26 ##forth <forthBot> Unknown word: a

22:29:27 ##forth <forthBot> Unknown word: a

22:29:28 ##forth <forthBot> Unknown word: a

22:29:29 ##forth <forthBot> Unknown word: a

22:29:30 ##forth <xentrac> a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a

22:29:30 ##forth <forthBot> Unknown word: a

22:29:36 ##forth <xentrac> a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a

22:29:39 ##forth <xentrac> see?

22:29:41 ##forth <forthBot> Unknown word: a

22:29:42 ##forth <forthBot> Unknown word: a

22:29:43 ##forth <forthBot> Unknown word: a

22:29:44 ##forth <forthBot> Unknown word: a

22:29:45 ##forth <forthBot> Unknown word: a

22:29:46 ##forth <forthBot> Unknown word: a

22:29:47 ##forth <forthBot> Unknown word: a

22:29:48 ##forth <forthBot> Unknown word: a

22:29:49 ##forth <forthBot> Unknown word: a

22:29:50 ##forth <forthBot> Unknown word: a

22:30:55 ##forth <forthBot> Unknown word: a

22:32:16 ##forth <cleobuline> il will try

IRC Log - 2025-05-22 - ##forth