00:00:00 --- log: started forth/03.12.27 00:28:16 * chrisrw is away: zZzZ 00:58:30 --- join: qFox (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth 01:25:19 : doe 3 begin dup . bl 1 - 0= until drop ; 01:25:24 why is this an endless loop? 01:29:18 oh the bl.. 01:29:46 oooh oops, i thought.. wrong :) 01:31:10 the tos is always 31? 01:31:17 32, ya 01:31:23 i thought bl would automaticly print :) 01:32:14 like cr does 01:33:52 --- quit: onetom (Read error: 60 (Operation timed out)) 02:28:38 ok, in this code: http://pastebot.nd.edu:80/135 02:28:52 the most inner loop somehow returns 0 02:32:15 i mean the counter of hte inner loop, its appearantly always 0.. 02:32:23 any reason for this? :) 02:45:41 hm ok, its the outer loop appearantly, but why.. 02:46:51 ah i figured it out :) 02:47:03 stack starts at 0, not at 1 :p 02:47:14 i mean, if you pick or whatever 03:00:34 --- join: networm_ (~networm@L0663P28.dipool.highway.telekom.at) joined #forth 03:08:57 --- quit: networm (Read error: 110 (Connection timed out)) 03:17:47 if i define a variable, will it's initial "size" be equal to cells bytes? 05:43:41 --- join: crc (~crc@ACAFFEC5.ipt.aol.com) joined #forth 07:33:29 --- quit: crc ("I was using TinyIRC! Visit http://www.tinyirc.net/ for more information.") 07:57:20 * warpzero is away: Down, down, down! 08:04:19 --- join: wossname (wossname@HSE-QuebecCity-ppp81282.qc.sympatico.ca) joined #forth 08:04:23 how can i check either a word is already defined? 08:04:48 without causing the code to stop running... 08:04:54 --- quit: wossname (Client Quit) 08:06:07 --- join: wossname (wossname@HSE-QuebecCity-ppp81282.qc.sympatico.ca) joined #forth 08:28:12 --- join: ree- (~jwm@ns.fasthost.net) joined #forth 08:28:23 --- nick: ree- -> reentry 08:29:33 --- nick: reentry -> ree_ 08:32:38 --- join: ree__ (~jwm@ns.fasthost.net) joined #forth 08:37:35 --- join: kc5tja (~kc5tja@66-91-231-74.san.rr.com) joined #forth 08:37:35 --- mode: ChanServ set +o kc5tja 08:41:28 kc5tja> how can i check either a word is already defined? 08:41:34 without stopping the code 08:42:16 Is the word's name known at design-time, or does it change as the program is running? 08:42:49 well atm i keep reloading a file, so i wont loose data if i create an endless loop 08:43:09 but except for the first time, he'll of course always warn me that the word or variable or whatever has already been defined 08:43:27 so i wanna check either it exists, if so forget or free it, and else just continue 08:44:24 From file: D:\download\apps\forth\wf.6.09.01\qfox.f word: TOP isn't unique 08:44:24 From file: D:\download\apps\forth\wf.6.09.01\qfox.f word: DECK isn't unique 08:44:25 etc 08:44:31 OK, so it's known at design-time then. 08:44:38 yes i guess so.. 08:44:45 I would use ' to determine if a word exists. 08:44:53 Oh wait... 08:44:57 but that will only continue if it does 08:45:02 that'll error out if the word isn't defined. 08:45:04 if so put an address to the stack 08:45:04 yes 08:45:08 Does the word "FIND" exist on your Forth? 08:45:18 yes 08:46:06 ah 08:46:10 c" doe" find 08:46:12 tnx 08:46:33 C" SomeWord" FIND IF DROP ( word exists ) ELSE 2DROP ( nope, doesn't exist ) THEN 08:46:42 ye :) 08:47:00 :) 08:47:27 btw, is if true on any non-zero value? 08:47:47 the returning of -1 if something is true, is still a bit unlogical to me... 08:48:03 illogical... eh doesnt make sense :) 08:49:47 --- quit: ree_ (Read error: 110 (Connection timed out)) 08:50:49 if "if" is compilation only, how can i check either the returned value of find is true or false? 09:00:42 [IF] 09:00:54 Yes, any non-zero value is considered true by IF or [IF] 09:01:05 And it's very logical. 09:01:22 um.. 09:01:23 Error: create_deck_array is undefined 09:01:30 at this line 09:01:30 : create_deck_array 09:01:39 arent i defining it right there? :\ 09:01:52 I need to see your source code. I do not have enough context. 09:02:42 http://pastebot.nd.edu/138?ln=on&submit=Format+it%21 09:02:49 line 25 09:04:11 Yeah, that's correct Forth. 09:04:17 That sounds like a bug in Win32Forth to me. 09:04:21 hrm 09:04:22 :\ 09:04:29 Though I suspect something else is going on, because I'm almost positive it can handle that case. 09:05:55 It works in GForth. 09:05:59 its got something to do with whats above 09:06:08 if i remove all the stuff above it, it loads np 09:06:53 and now he works fine 09:06:54 wth 09:07:06 * qFox chuckles... 09:07:07 whatever 09:07:35 Hmm 09:07:38 No idea. 09:07:58 * warpzero is back (gone 01:10:39) 09:08:27 hm i dont think the forget trick works 09:09:27 and the problem comes back when i re-open winforth... the cause is the definition of top line 09:09:33 Are you using the [IF] and [THEN] words instead of IF and THEN? 09:09:40 yes, otherwise it wont work at all 09:09:44 Also, you might want to consider using MARKER instead of FORGET. 09:09:45 c" top" find [if] forget top [then] \ if top exists, forget it 09:09:50 but when i do that 09:09:56 Stack depth increased in file: D:\DOWNLOAD\APPS\FORTH\WF.6.09.01\QFOX.F at line: 09:09:56 19 Stack: [1] 1359136 09:10:06 and he still returns the top already defined warning 09:10:11 Yes, because you forgot the DROP and 2DROPs, like I posted in my example. 09:10:22 ack 09:10:37 FIND is the core of the Forth interpreter -- it finds a word, and either (a) returns its execution token on success (that's why you need DROP), or (b) returns the string address for conversion to a number (hence the 2DROP) 09:10:38 brb 09:11:01 Hey kc5tja. 09:13:48 back 09:13:49 re 09:13:59 Gotta go in 10 minutes -- have to get ready for work. 09:14:41 Okay, man, have fun. Or something. 09:14:54 still odd... untill the first time the file completely loaded, he will return the undefined error if : top ( n1 -- n1 ) dup ." -->" . cr \ prints --> TOS CR 09:14:59 is above it 09:15:15 oh 09:15:21 d'oh, guess what :) 09:15:50 how does one terminate a definition? ;) 09:16:06 how did it ever compile in gforth 09:17:36 qFox: I just took the definition of create_deck_array and its subsequent use. 09:17:39 I didn't compile the whole program. 09:17:42 oh right 09:18:58 would you happen to know why, the first time i run the code, the forget line give a stack underflow error on 2drop? 09:19:00 deck as in deque? 09:19:06 c" top" find [if] forget top drop ( found ) [else] 2drop ( not found) [then] 09:19:10 warpzero> cards 09:19:17 oh 09:19:30 set game; http://www.reed.edu/~mcphailb/applets/set/ 09:20:26 qFox: I probably made an error in recalling the stack effects for 'find'. Try using just DROP instead. I think it might return the count address instead of the string/length pair. 09:21:01 well the thing is 09:21:11 if the file loaded completely once 09:21:14 the error is gone 09:21:22 hm 09:21:26 actually that makes sense 09:21:53 but he still wont forget top... 09:21:58 Though, frankly, I would choose to use a marker word, since that would make things easier, I think: 09:21:58 From file: D:\download\apps\forth\wf.6.09.01\qfox.f word: TOP isn't unique 09:22:01 MARKER Top 09:22:12 oops 09:22:21 Scratch that last line. :0 09:22:22 :) even 09:22:22 dunno what marker is.. 09:22:41 C" Top" FIND [IF] Top [THEN] MARKER Top 09:23:01 MARKER defines a word which, when executed, FORGETs itself and all subsequent definitions. 09:23:27 ah, so a throw-away word.. 09:24:04 but do i mark it after the definition? 09:24:13 MARKER is the defining word. 09:24:26 Look at the line I posted above. 09:24:31 aye.. 09:24:59 If Top is defined, execute it (thus causing it and all subsequent definitions to be forgotten). Then, afterwards, (re-)define Top as a marker. 09:26:30 Oh well. 09:26:36 I have to jump in the shower and otherwise get ready for work. :( 09:26:45 yea go. tnx for the help 09:27:45 n/p 09:27:47 --- quit: kc5tja ("THX QSO ES 73 DE KC5TJA/6 CL ES QRT AR SK") 09:32:33 i guess i'll just live with the error for now.. 09:32:36 warning.. 10:05:10 Moo. 10:13:41 can you roll in the reverse direction? 10:13:54 it crashes with negative arguments.. 10:19:22 --- quit: warpzero ("Wave 29: 23000 points!") 10:19:58 roll? 10:20:05 * chrisrw has never used either pick or roll 10:21:14 never? 10:21:29 hmmm is it considered sloppy/bad? 10:21:38 like jumps for instance 10:22:53 --- join: schihei (~schihei@pD9E5CD6D.dip.t-dialin.net) joined #forth 10:22:54 anyways i'm off 10:40:05 --- join: onetom (~tom@cab.bio.u-szeged.hu) joined #forth 10:42:22 --- quit: networm_ (Read error: 113 (No route to host)) 10:59:51 --- join: networm_ (~networm@L0656P09.dipool.highway.telekom.at) joined #forth 11:19:36 --- quit: schihei (Client Quit) 11:57:58 --- nick: ree__ -> ree_ 12:59:02 --- quit: wossname (Read error: 60 (Operation timed out)) 13:07:34 --- quit: ree_ (Read error: 60 (Operation timed out)) 13:08:36 --- join: ree_ (~jwm@ns.fasthost.net) joined #forth 13:20:41 hey all 14:12:40 --- join: TheBlueWizard (TheBlueWiz@pc91dn1d.ppp.FCC.NET) joined #forth 14:12:40 --- mode: ChanServ set +o TheBlueWizard 14:12:44 hiya all 14:25:28 --- join: ree (thin@bespin.org) joined #forth 14:25:29 --- part: ree left #forth 14:25:53 --- join: ree__ (~jwm@ns.fasthost.net) joined #forth 14:26:05 --- join: ree (thin@bespin.org) joined #forth 14:26:21 --- part: ree left #forth 14:35:29 --- quit: ree_ (Read error: 110 (Connection timed out)) 14:46:45 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 14:47:21 --- nick: ree__ -> ree- 14:54:18 --- quit: tathi ("leaving") 15:14:54 --- quit: networm_ (Remote closed the connection) 15:54:27 --- join: warpzero (~warpzero@dsl.142.mt.onewest.net) joined #forth 16:07:40 --- nick: ree- -> ree 16:31:22 --- join: Herkamire (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth 16:43:51 hi tbw hi ree 16:44:01 hey chris 16:44:25 hiya chrisrw 16:46:06 :) 16:46:14 whats up? 16:46:38 finally figured out how to properly implement the final pieces of the system 16:46:52 yay :) 16:47:47 yeah, I am excited 16:48:44 ree: what sort of system are you making? 16:48:52 we talked about it before I think 16:49:11 large networked general development system 16:49:15 and also a media network 16:49:24 oh right :) 16:50:56 I've started studying stack machines again and came across how to implement this one big feature I wanted to have in the core design 16:51:01 can you roll in the reverse direction? it crashes with negative arguments.. 16:51:09 roll? 16:51:13 dont use roll. 16:51:27 use dup, over, rot, drop, r>, and >r 16:51:37 dont use pick, either 16:51:42 what are you trying to do? 16:51:44 why's that? 16:51:51 roll is bad style 16:51:56 k 16:52:12 4 loops at once 16:52:19 and i need the counters 16:52:27 i mean, the counters value have a meaning.. 16:52:35 (aside from counting :) 16:53:19 may i ask why its considered bad? 16:53:40 because at first glance i wouldnt really know how else to accomplish it 16:55:15 i { j { k { l { save ijkl } } } } , where ijk and l range from 1 to 3 16:56:09 ... 16:56:11 er 16:56:29 show me C-style what exactly you do to i, j, k, l 16:57:26 this seems kinda fishy, lol 16:59:00 for (i=0; i<3; ++i) { for (i=0; j<3; ++j) { for (j=0; j<3; ++j) { for (k=0; k<3; ++k) { thingie[(i-1)*27+(j-1)*9+(k-1)*3+l-1] = ijkl } } } } 16:59:23 ijkl would probably be slightly different but whatever. 17:00:09 !!! 17:00:12 wtf!? 17:00:15 wow 17:00:16 sue me :\ 17:00:19 thats horrible, lol 17:00:28 how would you do it 17:00:31 do you mean i + 1 or i - 1 on all of those? 17:00:46 http://www.reed.edu/~mcphailb/applets/set/ 17:00:55 this is to build a full deck of cards 17:01:06 3^4 cards 17:01:22 with the 4 properties in different order 17:01:51 i'm sorry but making 4 loops is simply teh easiest the way i see it.. 17:03:28 (ok easier would be just hardcoding it i guess) 17:04:57 well, no 17:05:10 its just that that loop is kinda .... wel .... bastardized, I should say 17:05:11 lol 17:05:19 well how would you handle it? 17:05:35 the deck has 52 cards, right? 17:05:38 81 17:05:42 4 x 13 cards 17:05:45 81? 17:05:51 4 properties, 3 types for each property 17:06:07 one card for every combination of properties 17:06:18 Color (Red, Green, or Purple) 17:06:18 Shape (Circle, Diamond, or Squiggle) 17:06:18 Number (One, Two, or Three) 17:06:18 Shading (Filled, Striped, or Empty) 17:06:37 oh SET :) 17:06:37 hrm, lemme rethink that one. 17:06:39 yes 17:06:53 Herkamire> i just pasted it :) 17:07:33 what's ijkl? an int with bitfields? 17:07:37 a structure? 17:07:42 just int counters 17:07:53 counts from 1 to 3 17:08:07 qFox: you might want to rethink the way you store this. 17:08:18 how exactly do you represent each card right now? 17:08:25 its kinda hard to tell from the above. 17:08:30 just 1111 1323 17:08:42 with each digit representing the option one to three 17:08:55 pretty straightforward 17:08:58 yeah, usually structures are used for this 17:09:11 yes but i'm just beginning so cut me some slack 17:09:13 2 bits * 4 = 1 byte 17:09:33 who took away your slack? 17:10:01 chrisrw> yes well the way i save isnt the issue here, is it? i thought the 4 counters were, and how to do it without pick or roll... 17:10:07 : new-card-prop dup 3 mod +! 3 / ; : new-deck deck-buf >a 81 0 do i new-card-prop new-card-prop new-card-prop new-card-prop loop ; 17:10:34 whats +! ? 17:11:44 +! stores to the address in the A register, then increments the A register 17:11:53 (that's where I'm storing your deck) 17:11:54 well, i gotta shower, so ill think about it there, and once im done, ill tell you what I came up with :) 17:12:02 k :) 17:12:23 hmm and this will create all the ways 1111 to 3333 ? 17:14:01 0000 to 2222 17:15:03 hmm 17:15:48 here, try this: : ncp dup 3 mod . 3 / ; : new-deck 81 0 do i ncp ncp ncp cr loop ; 17:15:59 that prints it instead of storing it 17:16:19 and deck_buf is declared how? 17:16:51 create deck_buf xxx allot 17:16:57 never mind that 17:17:13 see my new example, and alter it to store wherever and in whatever format you want. 17:17:56 it just loops from 0 to 81 and converts to base 3 17:20:28 he seems to not like it, but hold on 17:20:45 qFox: who? 17:20:49 winforth 17:20:54 mod? 17:20:56 i'm getting an exception 17:21:07 any message? 17:21:08 in the new-card-prop yes 17:21:31 fails at word NEW-CARD-PROP 17:21:34 winforth probably doesn't have A register. 17:21:41 i did change that to >r 17:21:41 try this one: : ncp dup 3 mod . 3 / ; : new-deck 81 0 do i ncp ncp ncp cr loop ; 17:22:02 qFox: that would make it crash :) 17:22:09 whats a then? 17:22:19 i figured they would both be stacks, my bad :\ 17:22:44 now with 4 17:22:52 oh nm thats just another ncp 17:23:16 no, this one is complete standalone version 17:23:27 yea but it needed 4 :) 17:23:38 fixed that anyways, now i'll analyze :) 17:24:10 what does the i behind do do? 17:24:22 i grabs the loop index 17:24:29 oh its code, ic 17:25:52 leaves the stack filled btw 17:26:02 with 81 zero's 17:26:03 oops, looks like I got the loop indexes wrong too. it doesn't print out 0 0 0 0 17:26:15 no actually its quite right 17:26:21 except it needs a drop somewhere 17:26:34 i changed it slightly for output 17:26:35 : ncp dup 3 mod 49 + emit 3 / ; 17:26:35 : new-deck 81 0 do i ncp ncp ncp ncp cr loop ; 17:26:35 qFox: you're right. add "ncp drop" before "cr" 17:28:09 is this a "standard technique"? 17:28:22 qFox: what do you intend to use this lookup table with the cards in it for? 17:28:32 deal the cards from it 17:28:35 qFox: what? 17:29:06 what about shuffling 17:29:20 well this is a workaround (imo) for counters 17:29:24 back 17:29:28 did you figure it out yet? 17:29:29 I mean "is what a 'standard technique'?" 17:30:00 qFox: oh, don't know if it's standard, it just seemed the simplest way to me. 17:30:04 k 17:30:26 and i can shuffle later, but i need to build the deck in order to use it dont i 17:30:41 many forths have J defined, and there's probably some with K and L. you could define them if you wanted to. 17:30:49 you know the set game (i guess?), so you know the rules and how the cards are dealt.. 17:30:51 qFox: I don't think so. 17:31:09 then how do you make sure a card isnt already dealt? 17:31:15 if you're going random... 17:31:43 my code ("ncp ncp ncp ncp drop") converts an into to a card, so you don't need a lookup table to do so. you will need to keep track of what cards have been delt though. 17:31:46 dealt flags 17:32:05 chrisrw: don't be absurd 17:32:14 there's no need to get so complicated. 17:32:14 yes but i mean, you still have to create that table with cards... 17:32:47 qFox: yeah, that's probably the best way to shuffle, and make sure you don't deal the same card twice. 17:33:08 but I don't think you're going to want the deck in order like we've been generating it so far. 17:33:36 also you have to decide how you want to store the card 17:33:50 that can actually be changed any time.. 17:34:14 if i save it as a byte first, but in bit's later that shouldnt be much of a problem to alter 17:34:49 qFox: yeah, probably not. 17:35:01 either way it'll fit in a register 17:35:07 :) 17:35:18 ye 17:35:20 ok, so you're going to generate a deck, then scramble it right? 17:35:30 well is that neccesary? 17:35:34 what i was thinking 17:35:43 then you can use them in the order you have them in the array 17:35:49 pick cards random from 1 to size-of-array 17:36:01 whenever you pick a card, replace that value with the last value of the array 17:36:50 yes thats the alternative, but that does require two arrays (ok not much of a problem :) 17:37:00 one array is plenty 17:37:16 i mean double the size 17:37:30 heres one 17:37:31 or do you really want to shuffle them? 17:37:43 for each card, pick another random card, and exchange it. 17:37:43 :) 17:37:51 8bit entries in the array is enough, and you could do 32 bit to make it easier on yourself if you want. don't go any bigger than that 17:38:11 chrisrw: yeah, that's how I was thinking of shuffling. 17:38:13 at first i'll just go with 1 cells 17:39:00 i'm still a bit... surprised by : ncp dup 3 mod 49 + emit 3 / ; ... i dont think i would've ever gotten to this.. 17:39:16 which kinda puts me down tbh :) 17:39:44 or better, go through each card, and exchange it with a random other card. then you do this exactly 81 times, and you gaurantee that you never leave any strings of consecutive cards unscrambled. 17:41:01 qFox: you're cards are simple permutations. 17:41:07 ye 17:41:09 s/you're/your 17:42:03 Herkamire: the only problem with that method of shuffling is that the last card can never be the last card generated 17:42:15 chrisrw: sure it can 17:42:21 no it cant. 17:42:26 er 17:42:28 oh 17:42:30 yes it can 17:42:31 gotta go...bye all 17:42:31 lol 17:42:33 nevermind 17:42:33 :) 17:42:37 bye tbw 17:42:57 1) it can be moved two or more times. 2) your randome cell to switch with could be the same one it's at now. 17:43:06 bye chrisrw 17:43:14 later TheBlueWizard 17:43:29 bye Herkamire 17:43:40 --- part: TheBlueWizard left #forth 17:43:59 Herkamire: yeah, i dont know what i was thinking lol 17:44:23 chrisrw: I understand. the same thought passed through my head :) 17:44:54 too bad I can't spell today (randome) 17:48:53 btw how do i immidiatly declare an array, for some reason i cant seem to use create without defining a word for it first... 17:48:54 create deck-buf 81 cells allot 17:49:09 kinda odd :\ 17:49:17 you want an array without a name? 17:49:34 is that possible? 17:49:35 what do you mean immediately? 17:49:42 qFox - eh, yes. Just allocate space for it. 17:49:50 well i had this 17:49:51 : create_deck_array ( -- ) 17:49:51 create 17:49:51 81 4 * cells dup allot \ creates array and leaves size on the stack 17:49:51 cr ." Created an array and allocated " . ." bytes." ; 17:49:51 create_deck_array deck \ create array that holds the deck 17:50:03 because it wouldnt do it without creating a word first 17:50:11 qFox - take a look at HERE first. 17:50:31 HERE 81 4 * cells dup allot ( a n ) 17:50:59 qFox: you want create_deck to create allocate a new array in memory every time you call it? 17:51:14 its a onetime thing really 17:51:21 then don't compile a function to do it. 17:51:23 just do it 17:51:30 thats what i said :\ 17:51:46 create deck-buff 81 4 * allot 17:51:47 or 17:51:48 well i mustve done something wrong yesterday, it seems to work now 17:52:06 here 324 allot 17:52:34 this part seems wrong at first glance: "81 4 * cells" 17:53:00 unless you're talking a cell for each attribute 17:53:07 i defined it as "deck[81][4]" 17:53:12 :) 17:53:17 as bytes or cells? 17:53:24 81x4 cells 17:53:32 why? 17:53:36 cell per attribute? 17:53:39 ye 17:53:44 don't do that 17:53:51 i knew you were gonna say that :p 17:54:20 anything a cell or smaller (your card could be) is much much easier to work with 17:54:21 : advertise ( n -- ) ." Created an array and allocated " . ." bytes." ; : create-deck-array ( -- a n ) here [ 81 4 * cells ]L dup allot cr dup advertise ; 17:57:52 hm, how would you combine the numbers to one number? right now the method i'm thinking is probably wrong... multiply the numbers with 1000 100 or 10 and add them to the total 17:58:08 i dont know if there's some kind of "stringcopy" word for this? 17:58:14 (and i know its not a string) 17:58:20 just store them to your array one byte at a time 17:58:30 or a half nibble at a time 17:58:33 using shifts 17:58:52 hm 1 byte for each number is one cells 17:59:00 for all of them i mean 17:59:09 eh, whatever :) ok 17:59:14 or one byte total 17:59:16 if you do it my way 17:59:32 0xF 0xA 2n, ( store 2 nibbles at HERE ) 17:59:53 yes but that would make it somewhat harder for me to retrieve it, its 3 am here... i'm not too great on my feet 18:00:19 OK, then don't worry about it. 18:00:22 :) 18:00:26 I lshift J or lshift K or lshift l or lshift 18:00:47 chris - oh, at least hide the bit-shifting under a word. 18:00:55 you generally have to AND masks and shift and stuff. I make words to do those. my forth in native PPC and there's and instruction that shifts and masks (rlwinm (Rotate Left With Immediate, aNd with Mask) 18:01:29