00:00:00 --- log: started forth/06.06.28
00:37:27 --- quit: neceve (Remote closed the connection)
01:52:19 --- nick: segher_ -> segher
02:59:02 --- quit: JasonWoof ("off to bed")
02:59:54 --- quit: GoGinah (Remote closed the connection)
03:57:38 --- join: tathi (n=josh@pdpc/supporter/bronze/tathi) joined #forth
04:05:04 --- join: zpg (n=user@smaug.linux.pwf.cam.ac.uk) joined #forth
04:05:07 <zpg> afternoon
04:08:18 --- join: nighty (n=nighty@fr-reims-gw.origami-systems.com) joined #forth
04:10:45 <virl> hi
04:14:06 <zpg> hi virl 
04:40:57 --- quit: zpg ("ERC Version 5.0.4 $Revision: 1.726.2.19 $ (IRC client for Emacs)")
05:34:26 --- join: zpg (n=user@smaug.linux.pwf.cam.ac.uk) joined #forth
05:42:37 --- join: PoppaVic (n=pete@0-1pool66-66.nas22.chicago4.il.us.da.qwest.net) joined #forth
05:49:24 --- join: timlarson_ (n=timlarso@65.116.199.19) joined #forth
05:51:59 --- join: Ray_work (n=Raystm2@199.227.227.26) joined #forth
05:52:45 --- quit: Ray_work (Client Quit)
06:04:05 --- quit: nighty (Read error: 104 (Connection reset by peer))
06:04:49 --- join: nighty (n=nighty@fr-reims-gw.origami-systems.com) joined #forth
06:12:20 --- part: zpg left #forth
06:21:21 --- quit: PoppaVic ("Pulls the pin...")
06:31:34 --- join: silvousplait (n=trip_n_s@pool-151-196-179-211.balt.east.verizon.net) joined #forth
06:35:24 --- quit: silvousplait (Client Quit)
06:57:48 --- join: PoppaVic (n=pete@0-1pool73-111.nas24.chicago4.il.us.da.qwest.net) joined #forth
07:24:08 --- join: Queek (i=74081307@Orleans-ppp42653.sympatico.ca) joined #forth
07:43:51 --- part: Queek left #forth
08:14:39 --- join: Ray_work (n=Raystm2@199.227.227.26) joined #forth
08:41:45 <PoppaVic> You know..? I've been glaring for days, but I am beginning to think there are a variety of "depths" or "levels" or "contexts" or "states all thru most Forths.
08:42:28 <PoppaVic> state seems to be a member of a global struct, as does context and current
08:42:44 <PoppaVic> ..WHAT the latter two mean seems to be the issue.
08:43:48 <PoppaVic> are "threads" processes or states or vocstacks? are "words" no more than operators or operands?
09:03:07 --- join: rob1nhood (n=user@193.219.74.136) joined #forth
09:19:53 --- quit: nighty ("Disappears in a puff of smoke")
09:33:32 <OrngeTide> Quartus, well I'll just encourage him to use figforth until he can point to some legitimate reason not to use it on his 6502. 
09:34:02 <OrngeTide> (I fell asleep before you guys responded to me. thank goodness for logs)
09:39:22 --- quit: PoppaVic ("Pulls the pin...")
09:41:20 --- join: ttuttle (n=tom@unaffiliated/ttuttle) joined #forth
09:41:22 <ttuttle> Quartus: Hey.
09:44:47 --- part: rob1nhood left #forth
10:18:34 --- join: rabbitwhite (n=roger@136.160.196.114) joined #forth
10:20:47 --- join: JasonWoof (n=jason@pdpc/supporter/student/Herkamire) joined #forth
10:20:47 --- mode: ChanServ set +o JasonWoof
10:30:24 --- quit: rabbitwhite ()
10:35:23 --- quit: ttuttle ("leaving")
11:18:51 --- join: nighty (i=nighty@sushi.rural-networks.com) joined #forth
11:48:21 --- join: zpg (n=user@85-210-112-48.dsl.pipex.com) joined #forth
11:48:29 --- join: I440r (n=mark4@24-177-235-246.dhcp.gnvl.sc.charter.com) joined #forth
11:48:36 <zpg> evening
11:57:10 <I440r> hi
11:57:19 <I440r> zpg u new in here ?
11:59:01 <zpg> i am indeed, as new as i am to forth (which is most new)
11:59:09 <I440r> :)
11:59:16 <I440r> welcome :)
11:59:20 <zpg> many thanks
11:59:31 <I440r> what os do u use?
11:59:40 <I440r> what forths are u looking at ?
11:59:42 <zpg> OSX or BSD depending
11:59:51 <I440r> cool
11:59:54 <zpg> gforth installed
12:00:09 <I440r> are u an experienced coder but new to forth ?
12:00:17 <zpg> yeah, something like that.
12:00:24 <I440r> i dont like gforth but im biased
12:00:29 <zpg> not professionally so, but i've been hacking in various languages for quite a while.
12:00:42 <zpg> might you be a terp implementer/
12:00:43 <zpg> ?
12:00:43 <I440r> know asm ?
12:00:53 <I440r> isforth.clss.net
12:00:56 <I440r> my forth
12:01:03 <zpg> x86? yes, though my memory will likely disagree
12:01:11 <zpg> I440r: how did i guess you'd written your own :)
12:01:30 <I440r> everyone does :P
12:02:24 <zpg> "My hobbies include listening to and playing blues guitar," <-- I'm listening to Dylan's World Gone Wrong
12:02:32 <zpg> pretty strong record, not sure if you've heard it
12:02:46 <I440r> :)
12:03:12 <I440r> if u code x86 take a look at my kernel sources
12:03:20 <I440r> they might help
12:03:37 <zpg> so have you implemented isforth in asm?
12:03:49 <I440r> i just had my wisdom teeth out so im not 100% atm
12:03:50 <I440r> yes
12:03:56 <zpg> gas?
12:04:04 <zpg> (the assembler, not sideffect of denistry)
12:04:04 <I440r> if its not done in asm its not forth
12:04:07 <I440r> no
12:04:10 <I440r> nasm
12:04:13 <zpg> nice one
12:04:21 <zpg> tinkered with that a few years ago
12:04:35 <zpg> liked the portability (gas really threw me when i tried asm on linux)
12:04:52 <I440r> gas is crap on x86
12:04:58 <I440r> and arm
12:05:15 <zpg> typo sir: "This code too me about 3 days to write."
12:05:30 <I440r> aha :)
12:05:40 <zpg> hang on, did you rewrite in PPC assembly?
12:05:49 <I440r> i didnt
12:05:53 <I440r> tathi did
12:05:58 <zpg> i see
12:05:59 <I440r> but its not osx
12:06:05 <zpg> ppclinux?
12:06:14 <I440r> an osx port would be good
12:06:17 <I440r> ya
12:06:19 <zpg> k
12:06:23 <zpg> os it's all ELFed?
12:06:33 <I440r> i did the fbsd port tho
12:06:38 <I440r> on slavas box
12:06:43 <zpg> good stuff
12:06:47 <tathi> yeah, and syscalls are slightly different on OS X
12:06:50 <zpg> ah ok
12:06:54 <zpg> hey there tathi 
12:06:55 <tathi> but that part should be easy to fix
12:07:02 <tathi> hi zpg
12:07:06 <zpg> well, i'm still learning to walk in Forth at the moment.
12:07:11 <zpg> literally began ... yesterday
12:07:32 <zpg> been meaning to for a while; OFirmware is Forth, Raskin loved it, etc.
12:07:46 <tathi> yeah, open firmware is nice
12:07:56 * tathi can't believe they dropped it for the intel macs
12:07:57 <zpg> i'm looking at Brodie's Starting Forth at the moment
12:08:02 <zpg> really?
12:08:11 <zpg> didn't know that
12:11:52 <tathi> yeah, they switched to Intel's Extensible Firmware Interface for x86
12:14:02 <zpg> custom language?
12:14:26 <tathi> I don't know much about it, actually
12:22:44 <tathi> looks like Intel wanted something that was as backward-compatible as possible, so they wrote their own
12:23:07 <zpg> backward compatible with what?
12:23:19 <zpg> surely not bios
12:23:20 <tathi> the old x86 (PC/AT or whatever) system
12:23:28 <tathi> yeah, sounds like it.
12:24:16 <zpg> yikes
12:25:02 <tathi> yup.  Apple should totally have gone with Open Firmware.
12:25:24 <tathi> oh well
12:29:27 <zpg> in other news, SWAP and OVER are confusing the hell out of me :) haven't exercised RPN-braincells in quite a while
12:29:53 <tathi> hehe
12:30:29 <zpg> one of those "i think i've got it, wait that makes no sense, ah the answers says ... hrmph?!" moments
12:30:41 <JasonWoof> 1 2 3 4   swap   1 2 4 3
12:30:50 <JasonWoof> 1 2 3 4  over  1 2 3 4 3
12:31:06 <JasonWoof> over adds a new element to the stack that is a copy of the one just under the top
12:31:13 <zpg> indeed, in itself a fairly straightforward concept
12:31:19 <zpg> it's the equation conversion that's confusing
12:31:30 <JasonWoof> swap swaps the top two
12:31:32 <zpg> i.e., how to correctly use these words
12:31:39 <JasonWoof> over reaches over the top one and graps a copy of the one past it
12:31:42 <zpg> as well as seeing the patterns <-- this is lacking
12:31:48 * zpg nods
12:31:48 <JasonWoof> ahh, usage
12:31:51 <zpg> :) indeed
12:31:53 <JasonWoof> avoid using swap if possible
12:32:29 <tathi> now, see, that's one of Jeff and Chuck's things that I totally disagree with.
12:32:45 <tathi> there are quite a lot of situations where it's completely impossible to avoid using swap
12:33:47 <zpg> heh
12:33:54 <zpg> so Chuck Moore advises against swap too?
12:34:16 <tathi> well, one of his little crazy-minimalist stack chips doesn't have swap, just over.
12:34:20 <zpg> (incidentally, what do you guys use Forth for? I know I440r works in the Embedded realm)
12:34:26 <virl> heh don't kill the swap dragon!
12:34:33 <tathi> it simplifies the hardware design slightly because you never have to write to the second stack element, I guess.
12:35:25 <virl> was that the seaforth chip?
12:35:52 <tathi> mostly just for little stuff
12:36:13 <tathi> I still don't have a forth that has good enough string support to easily do a lot of the scripting stuff that I do
12:36:47 <tathi> virl: I don't think the actual specs on the Intellasys chips are out yet; I was thinking of his earlier designs.
12:36:57 <tathi> I'm pretty sure c18 didn't, and possibly f21 also
12:47:54 * zpg gets off the phone and returns to SWAP 
12:51:04 <Quartus> Avoid SWAP?  I hadn't heard that one.
12:51:24 <Quartus> I'm the author of Quartus Forth, zpg, for the Palm OS.
12:54:21 <tathi> Quartus: well, actually I've only heard it from Jeff.  So don't take it too seriously :)
12:57:59 <Quartus> Ah. Disregarding now. :)
12:58:32 <zpg> oh neat, hadn't even though of putting Forth on my palm
12:58:47 <Quartus> The evaluation version is free, if you're interested; http://www.quartus.net
12:59:49 <zpg> Jeff [Fox]?
12:59:56 * zpg looks at quartus
13:00:28 <Quartus> I assume son.
13:00:30 <Quartus> -n
13:01:28 <tathi> yes
13:01:34 <zpg> Quartus looks neat, though I think I'd find programming with the stylus rather frustrating
13:01:51 <I440r> luckilly Quartus did :)
13:02:04 <Quartus> Did?
13:02:23 <Quartus> zpg, it's not so bad, I wrote many apps with the stylus alone, but the external keyboard does come in handy.
13:02:32 <I440r> thought of putting forth on a palm :)
13:02:37 <Quartus> Ah :)
13:03:17 <zpg> good stuff
13:03:27 <zpg> yeah i guess it's a matter of practise too, though external kbd would be handy
13:04:37 <Quartus> I recommend it for heavy use.
13:05:50 * zpg nods
13:06:01 <zpg> so do your use forth predominantly on the palm?
13:06:16 <Quartus> It only runs on Palm OS, so yes. :)
13:07:16 <Quartus> Heard RIM and Palm may merge.
13:07:29 <Quartus> That'd be neat.
13:08:54 <zpg> rim?
13:10:36 <zpg> http://home.iae.nl/users/mhx/sf2/sf2.html, Quizzie 2-c, # 6 was a kick in the teeth :E )
13:11:27 <Quartus> Research In Motion, Canadian company that makes the Blackberry wireless email device.
13:11:58 <zpg> ah, Blackberry+Palm; that sounds pretty neat indeed.
13:12:29 <Quartus> Did you figure out 2-c?
13:13:29 <zpg> 1-5, yes
13:13:37 <zpg> 6 led to crosseyes
13:16:54 <Quartus> Bit trickier.
13:17:00 <zpg> indeed
13:17:06 <Quartus> Solved?
13:19:08 <zpg> well i looked at the answer when i thought i had a solution, it was wrong
13:19:23 <zpg> i'll come back fresh later on and try again (when solution firmly forgotten - nearing that stage now)
13:19:56 <zpg> any of you guys use forth as a main language outside the embedded context?
13:20:46 <zpg> personally, i'm enjoying the different form of expression, RPN postfix and cleanliness (similar reasons to my enjoyment of lisp; an emacser in fact reminded me to check out Forth properly)
13:21:03 <zpg> obviously lisp = prefix
13:22:33 <tathi> heh.  I thought their answer was wrong for a second there
13:23:26 <zpg> nothing like that, we were just discussing the elegance of certain languages, and a chap mentioned factor and forth in the same breath as lisp, which really generated interest. i'd never looked into forth properly, it was little beyond a name and vague association with RPN
13:24:31 <tathi> did you check out factor?
13:25:20 <zpg> decided to begin with Forth before even looking at Factor.
13:25:48 <tathi> ah.
13:28:00 <tathi> I'm not sure they have much in common any more, beyond the RPN syntax
13:28:25 <Quartus> It's a primary language for me, zpg.
13:28:51 <Quartus> I use it for desktop apps, web scripting, etc. Not exclusively, of course, but if I have a choice.
13:28:57 <zpg> interesting.
13:29:22 <zpg> how's library support, ipc, etc?
13:29:37 <Quartus> Depends on the task at hand, and the Forth in question.
13:30:12 <zpg> oh sure; but i take it you use a particular set of tools do achieve the scripting and app writing....
13:30:47 <Quartus> I have yet to run into anything I wanted to do that I couldn't.  I've used SwiftForth and Win32Forth under Windows.
13:31:19 <I440r> im using isforth on this contract :)
13:31:24 <zpg> hehe
13:31:37 --- quit: timlarson_ ("Leaving")
13:32:08 <zpg> anyone tinkering with OSX + Forth (other than firmware)?
13:32:23 <zpg> though i guess X11/Unix would work fine too, which is handy.
13:33:35 <Quartus> Not that you're necessarily leaning this way, but I recommend learning the language first, how to think in Forth, before worrying about how to take over the world with it.  The techniques you'll learn will make you a better programmer in any language.
13:34:24 <zpg> i agree and take the advice on board.
13:34:30 <Quartus> :)
13:34:51 <zpg> as i say, at one level i'm simply engaged with the language-as-a-language, not a means but an end.
13:35:06 <zpg> just intrigued to scout out how people in #forth are using the language in 2006.
13:35:08 <Quartus> I mention it only because I've seen quite a few newcomers who look at some distant target, don't see the specific library linkage they think they'll need, and so decide not to learn it.
13:35:19 * zpg nods
13:35:30 <zpg> then opt for C(++) instead
13:35:43 <Quartus> Well, perhaps they were just looking for an excuse not to have to make the effort.
13:36:01 <zpg> this is possible
13:36:05 <zpg> "i hear forth can do this ..."
13:36:36 <Quartus> I've fielded any number of "can Forth interface with <xxx>"?
13:36:43 <zpg> it's also important, i think, to give a new language a chance, especially where there's paradigm shift (e.g., C->Smalltalk, Smalltalk -> Lisp)
13:37:21 <Quartus> Learning different languages is helpful, particularly when they're not all Algol-68 variants.
13:37:25 <zpg> didn't mean to come across as tedious; as i say, just inquisitive about usage today
13:37:35 <Quartus> No, you're not coming across that way.
13:37:39 <zpg> :)
13:37:59 <zpg> i'm particularly fond of smalltalk and lisp which might indicate where i'm coming from re: forth
13:38:27 <Quartus> The Lisp will give you a head-start.
13:39:05 <zpg> and #6 will give me a headache :)
13:39:13 <Quartus> Heh.
14:02:27 * zpg has installed isforth on the linux box
14:02:33 <zpg> something to play with over ssh from work tomorrow :)
14:17:32 <zpg> ok this postfix stuff is beginning to look very neat 
14:17:49 <zpg> http://home.iae.nl/users/mhx/sf4/sf4.html, definition of /check
14:32:44 <Quartus> What peaks your interest about /check ?
14:34:15 <tathi> other than that the stack comment is wrong :)
14:34:35 <zpg> hmm?
14:35:07 <zpg> Quartus: definition of /check in the link provided; the ability to offer either an operator or a drop word is pretty striking
14:35:13 <tathi> should be ( numerator denominator -- quotient )
14:36:55 <zpg> good point
14:37:09 <zpg> got a question regarding ?dup
14:39:05 <tathi> ask away
14:39:39 <zpg> the equivalence of ": /CHECK  DUP IF  /  ELSE DROP  THEN ;" and ": /CHECK  ?DUP IF  /  THEN ;" is mentioned. but doesn't IF check the top of the stack for truth? so in the case of ?dup, which doesn't duplicate false literals: where does IF look to recognise FALSE?
14:39:49 <zpg> the return stack? (<--- wild guess)
14:39:57 <tathi> nothing so complicated
14:40:20 <tathi> if it's 0, then ?DUP leaves it alone
14:40:25 <tathi> and then IF eats it
14:40:25 <Quartus> And then IF eats it.
14:40:31 <zpg> ahhhhhhhhh
14:40:31 <tathi> so you don't need a DROP.
14:40:32 <Quartus> Heh.
14:40:36 <zpg> genius
14:40:49 <Quartus> Personally I prefer not using ?DUP, the first method is clearer on later examination.
14:41:01 <zpg> i i think managed to forget the original purpose of the DUP in the course of trying to work that one out.
14:41:02 <tathi> likewise
14:41:17 <Quartus> There's also an inherent conditional in ?DUP, so it's not an optimization in that regard.
14:41:30 <zpg> thanks for the help
14:41:34 <Quartus> np
14:42:05 <zpg> btw - is this channel usually quite active?
14:42:28 <tathi> no
14:42:54 <tathi> not unless someone (e.g. you) comes along with questions :)
14:43:42 <zpg> heheh
14:44:11 <zpg> as i say, the help is appreciated
14:44:32 <Quartus> Or PoppaVic issues a monologue like hte one he did this morning.
14:55:30 --- quit: Snoopy42 ()
14:55:37 --- quit: I440r ("Leaving")
14:56:44 --- join: Amanita_Virosa (n=jenni@adsl-65-68-75-51.dsl.hstntx.swbell.net) joined #forth
15:04:47 <zpg> time to get some kip, cheers for now guys
15:04:49 --- quit: zpg ("night")
15:06:15 <crc> good evening
15:06:34 <JasonWoof> my 0; word solves the above problem...
15:06:42 <JasonWoof> : /check 0; / ;
15:07:03 <JasonWoof> 0; drops and exits if tos is zero. and does nothing otherwise
15:07:24 <JasonWoof> hi crc :)
15:09:31 <segher> two-level exit?  that's not standard forth ;-)
15:12:46 --- join: Snoopy42 (i=snoopy_1@dslb-084-058-152-145.pools.arcor-ip.net) joined #forth
15:13:16 <JasonWoof> two level?
15:13:17 <crc> segher: is that a problem?
15:13:44 <segher> exiting not only from the current definition, but from the caller, too
15:15:24 <JasonWoof> you could implement it like that
15:16:14 <JasonWoof> or : 0; postpone dup postpone 0if postpone drop postpone exit postpone then ; immediate
15:16:32 <JasonWoof> or the two level exit: : 0; dup if exit then drop rdrop ;
15:17:01 <segher> ah yes, you can always make it a macro, sure
15:29:36 --- quit: Amanita_Virosa ("wemp.")
16:10:54 --- quit: nighty (Read error: 104 (Connection reset by peer))
16:28:35 --- quit: Cheery ("Leaving")
17:28:19 --- quit: Hobart (Remote closed the connection)
17:44:22 --- quit: tathi ("leaving")
18:56:34 --- quit: uiuiuiu (Remote closed the connection)
18:56:38 --- join: uiuiuiu (i=ian@dslb-084-056-234-126.pools.arcor-ip.net) joined #forth
19:33:43 * ayrnieu fails at life :-(
19:39:25 --- join: Hobart (i=jb@jb.org) joined #forth
19:45:05 <ayrnieu> more specifically, at debugging this.
19:55:37 <ayrnieu> if anyone would like to help, http://sagacity.merseine.nu:8080/misc/nopaste.tgz contains a test and test data and the application and the libraries it uses.  The problem I see is that the question-structure gets corrupted, somehow, so that the test provided will display the question, the first choice, the second choice, and then fail on the third choice.
19:56:19 <ayrnieu> and I've traced I think most parts of the system, and still can't see why this happens.
19:56:55 <ayrnieu> I'm going to do something else for a while, before coming back to run with answer D  :-/
20:00:24 <ayrnieu> (the stack-effects for A and ANS are wrong; they leave quiz% on the stack)
20:06:02 <segher> .q seems wrong...  it does dup before calling q$@ which uses pick...  looks bad
20:06:52 <ayrnieu> stackarry.fs's PICK , rather, which is ( u a -- x )
20:07:06 <segher> oh ah
20:07:17 <segher> redefining standard words, bad boy
20:07:32 <segher> next thing you'll tell me you redefined  0  too :-)
20:08:27 <ayrnieu> PICK and DROP and DEPTH , but just in the context of stackarray -- so, that file and the indented part at the beginning of quiz.fs
20:10:14 <segher> tricky
20:11:58 <segher> what does "allocate" do?
20:12:26 <segher> ( u -- a )   ?
20:13:10 <ayrnieu> ANS Forth for dynamic allocation.  Pretty much that: ( u -- a-addr wior ) , where wior is consumed by THROW
20:13:23 <segher> ah yes
20:13:54 <segher> i normally write in OF, which has alloc-mem instead, which throws by itself
20:14:14 <ayrnieu> stackarray's are doubly-indirect to account for RESIZE being able to return a new addressc
20:14:29 <segher> right
20:18:10 <segher> well i guess i don't see it
20:18:15 <segher> the problem, that is
20:18:31 <segher> but boy this looks worse than ioccc :-)
20:19:00 <ayrnieu> aw, it can't be that bad.  Except by definition, since it includes a very difficult to spot bug.
20:20:31 <segher> well it has a few too many not-so-obvious-to-me abstractions
20:20:41 <segher> "new"  should call align btw
20:21:03 <segher> though i doubt that is your bug, seems to work fine the way it's called
20:21:39 <segher> "swap DROP" means use the standard "drop"?
20:21:51 <ayrnieu> *nod*, I figure now that the problem is in Quiz::UI or that first indented portion of quiz.fs
20:22:08 <segher> "swap drop" is called "nip"
20:22:13 <ayrnieu> no, that's the drop defined in stackarray.fs
20:22:33 <segher> ah okay
20:22:40 <segher> why in capital then?
20:22:44 <ayrnieu> less obviously a warning, in retrospect, with the local PICK right in front :-/
20:24:32 <segher> nice test question heh
20:25:22 <segher> esp. the correct answer ;-)
20:37:12 --- join: segher_ (n=segher@dslb-084-056-132-071.pools.arcor-ip.net) joined #forth
20:37:45 <segher_> got disconnected
20:38:00 <segher_> [05:33] 	segher	resize ( a size -- a' ior )  and it resizes to "size" bytes?
20:38:00 <segher_> [05:35] 	segher	try   : plumb ( a -- )  dup @ dup @ 1+ cells resize throw swap ! ;    in that case
20:40:39 <segher_> ayrnieu: ?
20:44:54 --- quit: segher (Read error: 110 (Connection timed out))
20:52:15 <ayrnieu> (in that case, what?)
20:52:49 <segher_> does resize do what i think it does?
20:53:03 <segher_> if so, add that 1+ in plumb
20:53:15 <ayrnieu> yes, it does that
20:53:25 <ayrnieu> oh, didn't see your 'try'
20:53:33 <segher_> heh
20:53:43 <ayrnieu> why the 1+ ?
20:54:02 <segher_> because you actually need storage for n+1 cells
20:54:13 <segher_> first cell itself contains the # items
20:54:21 <segher_> the n
20:55:07 <segher_> and "push" for example wants to store in cell number n
20:55:19 <segher_> so you better have space for n+1 cells :-)
21:00:34 <ayrnieu> thanks, I think some confusion comes simply from keeping that number in dynamic memory
21:01:09 <segher_> yeah, should have put that with the main pointer, in normal data space
21:01:32 <segher_> it's a single number, doesn't need resizing ;-)
21:02:23 <ayrnieu> : new ( -- a ) align here 0 dup , allocate throw , ;  <-- looks better already :-)
21:03:03 <segher_> align here 0 , 0 allocate throw ,
21:03:18 <segher_> no need for the dup heh
21:03:28 <segher_> pointer first is easier i guess though
21:03:43 <segher_> align here 0 allocate throw , 0 ,
21:04:03 <ayrnieu> mm, yeah
21:04:21 <segher_> you always need the pointer, you don't always need the #
21:04:52 <segher_> but yeah, looks simpler definitely
21:08:35 <segher_> so did it work with the 1+ :-)
21:08:54 <ayrnieu> no change with that :-/
21:10:05 <segher_> heh i can test it myself
21:10:09 <segher_> it's just gforth
21:10:20 <segher_> it does change, it prints all questions now
21:10:29 <segher_> crashes on printing the answer though
21:10:58 <segher_> it used to print gibberish for the questions
21:11:01 <ayrnieu> oh?  hm.
21:13:02 <segher_> change .q to read    ans@ ccount type
21:13:10 <segher_> and it runs all the way through
21:13:41 <segher_> what do i win?
21:14:11 <ayrnieu> the satisfaction of knowing that you beat your competitors! :D  Also, thanks.
21:14:23 <segher_> :-)
21:14:29 <segher_> my pleasure
21:15:06 <JasonWoof> huh? why are you allocating a memory block with zero size/
21:15:07 <JasonWoof> ?
21:15:32 <segher_> can you give a 0 pointer to "resize"?
21:15:35 <ayrnieu> to have something to resize later
21:15:48 <JasonWoof> oh
21:16:55 <ayrnieu> stackarray.fs isn't the goal, really; I didn't even have it in my first iteration.  I just noticed that I had two similar data structures with some header information and then an arbitrary number of cells ...
21:20:11 <segher_> sure, such an abstraction can reduce the number of bugs you have -- once you get the abstraction implemented correctly ;-)
21:22:14 <ayrnieu> yow, I just realized that I put . at the end of my gforth search path, so my test keeps inexplicably giving me the same answer... because I'm editing nopaste/ files and not ~/forth/lib/ files.
21:22:45 <segher_> heh
21:23:01 <ayrnieu> with that change, my new saner stackarray.fs also works
21:25:46 <ayrnieu> yay, now I can move on to do something useful.
21:26:07 <segher_> have fun!
21:27:27 <ayrnieu> (for the curious, I updated that nopaste.tgz the aforementioned rewrites)
21:33:42 <JasonWoof> ahh, there's what's getting at me
21:34:03 <JasonWoof> it seems funny to allocate that little struct with the two cells, one of which points to a memory block.
21:34:12 <JasonWoof> (I'm guessing the first might be a count)
21:34:31 <JasonWoof> why not store that first cell in the memory block, then your data is in one piece
21:34:56 <JasonWoof> you could do:
21:35:21 <JasonWoof> : new ( -- addr ) 0 4 allocate throw dup ! 4 + ;
21:35:28 <JasonWoof> oops
21:35:44 <JasonWoof> : new ( -- addr ) 4 allocate throw 0 over ! 4 + ;
21:35:55 <ayrnieu> (that's what I did, initially.  It just confused me.)
21:36:32 <ayrnieu> I need redirection, anyway, to account for resize.
21:37:09 <JasonWoof> indirection
21:37:18 <JasonWoof> ok
21:37:23 <ayrnieu> yes.
21:37:48 <JasonWoof> my approach to movable data is to keep an array of pointers
21:38:05 <JasonWoof> and reference the data by the array index
22:19:51 --- quit: Ray_work (Read error: 113 (No route to host))
23:59:59 --- log: ended forth/06.06.28