00:00:00 --- log: started forth/02.03.14 01:01:35 --- join: onetom (tom@adsl52070.vnet.hu) joined #forth 01:03:25 --- quit: onetom (Client Quit) 01:04:09 --- join: onetom (tom@adsl52070.vnet.hu) joined #forth 01:04:14 testing 01:04:21 ah, it wrx now 01:31:35 --- nick: Soap- -> SoapZZzz 05:03:35 --- join: rob_ert (~robert@h173n2fls33o898.telia.com) joined #forth 05:17:12 --- join: Speuler (~l@195.30.184.52) joined #forth 05:17:21 'morning 05:19:54 'afternoon 05:20:03 hi rob 05:20:05 _ert 05:21:13 hey :) 05:33:52 brb 05:33:57 --- part: Speuler left #forth 05:47:55 --- join: Speuler_ (~l@195.30.184.51) joined #forth 06:07:09 anybody present? 06:07:41 hm 06:07:52 yeah 06:08:20 y ? 06:09:54 I440r: is always .... 06:10:08 hateing the ANS 06:10:19 personal freedom 06:10:29 what standard does he follow then? 06:10:55 think its called the i440r-standard 06:10:55 earlier ones? 06:11:02 f79? :) 06:11:07 not quite 06:11:09 like f83, f79, fig? 06:11:24 execute expects another address 06:11:31 with f79/fig 06:11:42 tick returns another address too 06:12:04 and where can i find docs about these other standards 06:12:12 i'd try the internet 06:12:21 imean, such a comprehensive docs, like the ans draft 06:12:36 is there any at all? 06:12:39 but, mind you, i've been looking for the f83 standard a while ago, to no avail 06:12:59 f79 wasn't really a standard. more an "industry standard" 06:13:01 aha... thats great... :< 06:13:15 like "ibm pc" is "standard" 06:13:30 fig was more a kind of "de-facto" standard 06:13:47 which forth vendors tried to improve 06:13:56 that's what became f79 06:14:18 as forth community wasn't pleased with this situation, f83 arose 06:14:57 f83 fixed a number of shortcomings in both fig and f79 06:15:14 and it broke legacy code 06:15:24 ANS tried not to do 06:22:24 aha, i did know about this last clause 06:22:56 so probably f83 could b considered "the best"? 06:23:38 coz, ans tries to b compatible w the heritage? 06:23:53 just like, say windows? 06:33:45 f83 is stricter than ans 06:33:55 does some assumptions about the system 06:33:58 ans doesn't 06:34:51 ans has a number of improvements too 06:34:56 but adds complexity 06:35:40 for example, "accept" i like better than "expect" 06:42:46 and f83 only has expect, right? 06:42:48 Is there any guru in here that can either explain what "create" "," and "does>" do, or point me to some page? 06:43:10 rob_ert: would non-guru do too ? 06:43:19 rob_ert: have u read that tut i gave u? 06:43:35 rob_ert: forth primer is not good enough, i think 06:43:36 onetom: uhmm.. where was it? 06:43:43 Speuler_: sure :P 06:43:50 create is simple 06:43:54 rob_ert: it doesnt talk about the internals @ all 06:43:56 create foo 06:44:08 rob_ert: sec, i'll look 4 it 06:44:10 makes a header 06:44:26 rob_ert: but u have gforth docs, dont u? 06:44:30 default action of "foo" is, returning address of foo 06:44:41 rob_ert: what gforth version do u have? 0.5.0? 06:44:41 like in: 06:44:59 onetom: yeah, think so 06:45:02 the latest ;) 06:45:04 : rob's_variable create 0 , ; 06:45:34 with does>, we can say: "use another run time behaviour" 06:45:54 which one, we specify after does> , using hi-level code 06:46:02 such as in: 06:46:20 : rob's_constant create , does> @ ; 06:46:31 5 rob's_constant five 06:47:01 rob's_constant is like a variable, but additionally fetching it's contents 06:47:08 rob_ert: http://dec.bournemouth.ac.uk/forth/forth.html 06:47:11 because we tell it so after does> 06:47:29 rob_ert: shame on me, i dont even read it through either :) 06:47:42 onetom: :) 06:47:43 rob_ert: tho it also talks about multitasking basics :))) 06:47:50 see the part after does> as a kind of subroutine, called by the created words 06:48:02 the @ in this case? 06:48:04 with the word's address on top of stack 06:48:07 yes 06:48:11 rob_ert: exactly what i was looking for in the last weeks 06:48:21 another example to try: 06:48:39 : emitter create , does> @ emit ; 06:48:46 '* emitter star 06:48:51 rob_ert: it also has very nice pictures that could help a lot 06:48:55 star ( * ) 06:48:59 What is "'*" ? 06:49:16 '* puts ascii of * ( 42 ) on stack 06:49:28 ( gforth does ) 06:49:35 ( not standard ) 06:49:36 Speuler_: not in gforth, iguess 06:49:50 Speuler_: oops, really? 06:50:03 same as char * 06:50:04 Speuler_: probably he should use char * 06:50:19 that's standard 06:50:20 and [char] * inside a definition 06:50:39 ( Speuler_ , u r faster than me ;) 06:50:52 my line was shorter :) 06:51:22 hehe 06:51:24 7 emitter beep 06:51:30 hm, there I have another problem 06:52:04 er. n/m 07:06:57 rob_ert: does that explain all you need to know about create does> ? 07:09:13 not quite... I still don't really get exactly what it does, and how to use it... 07:09:26 pinpoint the problem 07:09:36 it does: 07:09:49 change the code address of the word 07:10:23 to..? 07:10:23 (instead of pointing to runtime code of variable, it points to code which executes code after does> 07:10:42 correctly "changes run time semantics of created word" 07:11:17 what's this dictionary stack they talk about in fprimer? 07:11:31 it also compiles "exit" in place of does, so that the creating part doesn't run into the run-time part (code after does>) 07:11:47 .vocs 07:11:50 sorry 07:11:52 vocs 07:12:04 shows you available vocabularies 07:12:07 order 07:12:24 shows you the vocabularies on the dictionary stack 07:12:30 those are just searched 07:12:35 like PATH 07:13:05 with ALSO, PREVIOUS you add and drop vocs from that stack 07:13:28 ALSO is like a dup 07:13:36 hm 07:13:43 what is stored there? 07:13:56 address of start of vocabulary link 07:15:06 pseudo code: : vocabulary create newvocabulary does> addvocabulary ; 07:15:42 do: 07:15:45 see vocabulary 07:15:56 : Vocabulary 07:15:56 Create wordlist drop DOES> context ! ; ok 07:16:13 yeah... 07:16:25 did you know about see ? 07:16:32 yes.. 07:17:25 good example for create does> too 07:19:10 how to use it ... (2nd part of your question) : 07:19:21 need a new type of data structure ? 07:19:30 there you are 07:19:40 instead of: 07:19:47 : foo1 dosomething ; 07:19:50 : foo2 dosomething ; 07:19:53 : foo3 dosomething ; 07:19:57 you can do: 07:20:26 : newfoo create does> dosomething ; 07:20:30 newfoo foo1 07:20:32 newfoo foo2 07:20:33 newfoo foo3 07:20:48 instead of: 07:20:50 what did the "," do earlier? 07:21:06 here cell allot ! 07:21:27 : star [char] * emit ; 07:21:36 : space bl emit ; 07:22:01 : lf $0a emit ; 07:22:04 you can do: 07:22:28 : emitter create , does> @ emit ; 07:22:35 42 emitter star 07:22:45 bl emitter space 07:22:50 $0a emitter lf 07:23:07 need a character array ? 07:23:19 : carray create allot does> + ; 07:23:30 10 carray tenbytearray 07:23:41 44 4 tenbytearray c! 07:23:46 55 5 tenbytearray c! 07:23:58 4 tenbytearray c@ . ( 44 ) 07:24:15 need another one ? 07:24:25 100 carray hundredbytearray 07:24:53 I'm starting to get this... :-) 07:25:14 t'is easy. but unusual, that's where the problem understanding may come from 07:26:02 forth is full with unconventional stuff ... 07:26:35 know a little bit lisp ? 07:27:06 nope 07:27:15 there are some similarities 07:27:35 for example, first element in a lisp list gives the operation 07:27:46 that's like the code field in a forth word 07:28:22 not exactly. somewhat ... 07:28:31 hmm 07:28:40 don't know anything about lisp :) 07:30:06 then there's precious little to compare forth with ... 07:31:14 don't try to compare it with c 07:31:25 gives you a headache trying it 07:31:57 actually, i tend to get headache from trying to program c 07:32:10 t'is so... clumsy 07:32:24 hm 07:32:30 I find C quite OK 07:33:04 but of course I see forth isn't like C 07:33:07 get some aspirin then 07:33:19 ;) 07:40:21 rob_ert: how about the doc i gave u? reading, reading? 07:40:29 rob_ert: coz u should..! 07:40:54 --- join: GnuVince_ (~vince@modemcable049.128-203-24.que.mc.videotron.ca) joined #forth 07:41:20 Hi 07:41:35 hi GnuVince 07:41:53 Hello Speuler_ 07:42:29 I'm not even a newbie Forth programmer yet, but I have a couple of questions about the language 07:42:41 go ahead 07:42:43 OK 07:42:58 Well first of all, is Forth a functionnal language or an imperative one? 07:43:16 hmmm 07:43:29 i'd call it an imperative language 07:44:24 OK. 07:44:55 Next, who uses Forth? I mean, what do people do with it? General purpose? Number crunching? 07:45:16 "imperative" as in "action oriented" 07:45:41 lot's of purposes. often used where other languages don't reach to. 07:45:51 t'is popular with embedded designs 07:46:03 i use it for working on prototypes 07:46:24 code prototyping is an example 07:46:26 robotics 07:46:33 hardware control in general 07:46:39 ok 07:46:51 --- join: Stepan (~stepan@Charybdis.suse.de) joined #forth 07:46:52 Are variables mutable? 07:47:47 as in "contents may change" ? 07:48:04 Yeah 07:48:12 a = 10; 07:48:16 wouldn't be a good variable otherwise 07:48:18 a = 10 + 10; 07:48:22 OK 07:48:27 Good thing :) 07:48:39 variable foo 07:48:42 10 foo ! 07:48:44 20 foo ! 07:48:51 Is the GNU implementation of Forth recommended? 07:48:52 foo @ . ( 20 ) 07:49:06 define "recommended" 07:49:11 recommended 4 what? 07:49:22 hi GnuVince_, anyway 07:49:44 variable a 07:49:46 10 a ! 07:49:49 10 a +! 07:49:54 a @ . ( 20 ) 07:50:19 or - 2 b more similar - 10 10 + a ! 07:50:45 i wanted to show the "mutable" aspect somewhat clearer 07:50:45 but cant imagine why 1 should ask such a question... 07:51:01 I mean, is Gforth a good implementation, or just a lame attempt to have a Forth implementation under the GPL by the GNU project? 07:51:01 Speuler_: could u explain GnuVince_ reasons? 07:51:42 onetom: y don't u ask GnuVince ? 07:52:10 gforth is quite good 07:52:16 ok 07:52:18 i agree 07:52:28 Good thing :) 07:52:29 --- join: futhin (~bloh@h24-64-174-2.cg.shawcable.net) joined #forth 07:52:29 Well 07:52:37 hihi 07:52:37 GnuVince_: what is this mutabitity mean? 07:52:48 howdy gnuvince :) 07:52:48 I think I'll give gforth a try :) 07:52:51 s/ mean/-thing/ 07:52:55 hi futhin 07:53:02 new to forth ? 07:53:11 GnuVince_: forth programmers tend to avoid variables 07:53:14 onetom well, in some functional languages, you cannot modify the value of a variable 07:53:23 10 10 + 07:53:29 ( result on stack ) 07:53:35 reverse polish notation? 07:53:40 yes 07:53:44 it's a stack-based language 07:53:52 heavily stack oriented 07:53:55 reverse polish notation is the _only_ way to manipulate the stack.. 07:54:13 GnuVince_: why r those things named variables? 07:54:48 onetom: I have no idea. 07:54:56 GnuVince_: constants are supposed to be immutable, but contents can be (non-standard) modified too 07:55:01 And what does stack oriented mean? 07:55:32 your working data is on stack, operators work on stack too 07:55:45 10 20 07:55:52 ( puts two numbers on stack ) 07:55:53 + 07:56:09 ( takes two numbers from stack, adds them, puts back result ) 07:56:28 Speuler_: i was already about asking u about this "tend to avoid variables" philosphy 07:56:41 Speuler_: could u explain it a bit? 07:56:42 variables are evil 07:56:43 I do not exactly know what stack is... 07:57:10 : addmultiply + * ; 07:57:15 1 2 3 addmultiply 07:57:16 Speuler_: but u have to store the data somewhere u r manipulating 07:57:29 GnuVince_: visualize a number to be a sheet of paper. pile paper up 07:57:34 that's a stacl 07:57:34 result: 9 07:57:36 stack 07:57:39 GnuVince_: stack = LIFO 07:58:12 GnuVince_: vs. queue/row = FIFO 07:58:16 technically, a stack is a memory region, with a pointer pointing somewhere into it 07:58:30 where pointer points to, that's top of stack 07:58:41 OK 07:58:51 is Forth a safe language? 07:58:53 onetom: data you work on can stay on stack 07:59:14 GnuVince_: define "safe" 07:59:30 I mean, is there a risk of stack overflow, or segmentation faults (that ringed a bell since you said 'pointers') 07:59:40 yes 07:59:51 Speuler_: ez faszsag 08:00:01 Yes? 08:00:05 Yes it's safe? 08:00:09 or yes there's a risk? 08:00:19 no, the safety lies in the programmer 08:00:20 onetom: snap ik niet 08:00:28 it's not some lazy language for lazy coders :P 08:00:39 Speuler_: the data u r working w @ the very moment should b stored/handled on the stack 08:00:40 although plently forth coders are reasonably lazy and still chose forth :) 08:01:10 GnuVince_: "yes" like risk. compiler doesn't trap all programmer errors 08:01:28 ok 08:01:40 you can call it, the compiler leaves the programmer lots of freedom 08:01:41 Speuler_: but u cant store matrices on stack, can u... 08:01:59 Support for arbitrary large integers (I like those) 08:02:01 onetom: you can put a pointer to a matrix on stack 08:02:14 gnuvince: it is one of the most flexible language i have come across, you can do _anything_ in forth. the elegance and simplicity and flexibility of forth come at a price, it is not safe, there is no type-checking. but i think it was one of the best languages in the world :) 08:02:23 GnuVince_: not part of standard, but can be added 08:02:45 Speuler_: that is: the actual data is stored in a variable, finally, isnt it? 08:03:03 onetom: in memory. not strictly in a variable 08:03:06 OK 08:03:18 Well 08:03:21 I think I'm gonna give it a fair try 08:03:26 oh, hell 08:03:35 GnuVince_: what languages do you know ? 08:03:38 arent variables stored in memory? 08:03:44 and even if I don't use it after learning it, the learning experience will probably be very rewarding anyway 08:03:52 yes. but not everything stored in mem is a variable 08:04:09 Speuler_: Python, Ruby, RPG/400, O'Caml, VB (though I wish I didn't) 08:04:15 bit of C/C++ and Pascal 08:04:19 GnuVince_: it is said that forth is more than a programming language, but a programming philosophy too 08:04:36 guess that goes for other languages too 08:04:44 i tend 2 experience that people misunderstanding this "var r evil" thing 08:05:15 forth variables tend to be of global nature 08:05:27 sure 08:05:30 thats right 08:05:47 but everything which returns an address, to write data to, BEHAVES like a variable 08:05:49 give me an example of this evilness 08:06:03 variable temp 08:06:04 ah, i agree again 08:06:14 s/ah/aha 08:06:23 I would maybe agree or disagree, but I am completely lost :) 08:06:32 : xxx begin 1 temp +! again ; 08:06:44 The Forth community seems helpful though, so I guess I'll have fun 08:06:44 brb ... 08:07:10 GnuVince_: i can b sure of it ;) 08:07:27 onetom: variables are evil because if you are using variables, it is most likely true that your code isn't very well designed & coded 08:07:55 k, show me an understandable example 08:08:08 (i doubt u can avoid vars 08:08:09 ) 08:08:24 onetom: what do you mean? i don't use variables at all.. 08:08:40 cant belive :) 08:08:54 what else do u use insted, then 08:08:58 dont say: the stack! 08:09:10 thats the other thing i cant believe :) 08:09:25 in a program u have to store 08:09:29 counters 08:09:32 well, i'm doing what chuck moore recommends.. keep the words 1 to 2 lines 08:09:36 semaphores 08:09:43 status info 08:09:43 and each word is referentially complete 08:09:46 buffers 08:09:56 these r all variables arent they? 08:11:05 saying what uve just said a sec b4 ([05:08] onetom: variables are evil because ...) 08:11:22 can totally mislead a beginner, i guess 08:11:59 or just get back to the prev. example 08:12:05 how can u handle a matrix? 08:12:18 u have to store it in a varibale 08:12:40 certainly u dont use the word 08:12:55 use create 08:12:56 VARIABLE for defining that 08:13:01 sure 08:13:30 but wont this create construct just look like the definition of variable? 08:14:02 so finally it will also b a variable 08:14:05 when i say variables are evil, and i don't recommend using them. i meant VARIABLE 08:14:29 if you avoid creating VARIABLEs 08:14:37 and you use some create for arrays and stuff 08:14:39 then that's fine 08:14:41 but VARIABLE 08:14:45 has a limited usage 08:14:53 I must say that Forth's syntax looks wierd though :) 08:14:54 unless you use ALLOT 08:15:08 gnuvince: yup, takes a bit to get used to 08:15:13 but certainly i use ;) 08:15:24 futhin, ok so I'm normal 08:15:28 GnuVince_: just @ a 1st glance 08:15:31 cause I'm looking at some examples 08:15:45 and it's hum... terse 08:15:47 GnuVince_: infact, forth doesnt has a syntax... 08:15:51 http://www.bagley.org/~doug/shootout/lang/gforth/allsrc.shtml 08:17:46 onetom: VARIABLE means you store a number, and it's globally available.. if you avoid using variable as much as you possibly can, the code is going to be much better.. 08:18:49 getting better 08:19:36 but generally saying vars r evil... well stating such is more avoidable then vars ;) 08:19:54 coz if u say something like that 08:20:05 onetom: yes, I'm reading it... and btw, that was the first forth doc I found ;) 08:20:06 u deny the need 4 global data area 08:20:41 back ... 08:20:52 wb Speuler_ 08:21:01 onetom: naw, when i or speuler say variables, we are refering to the word 08:21:36 "global storage for intermediate results" is what i mean with "variables are evil" 08:22:16 if your code flow corresponds to data flow, 08:22:27 intermediate data doesn't need to be stored 08:22:44 but can be processed on the stack, as it comes 08:22:55 if code flow and data flow don't match, 08:22:55 yeah, much better way to say it 08:23:12 you may need variables, or review your logic 08:23:55 in general, forth programmers are quite good at matching code and data flow 08:24:09 because they want to use the stack for intermediate results 08:24:32 instead of moving a result to variables, only to move it back to stack for another operation 08:24:36 that sounds much more correct 08:25:19 variables do exist, and are used sometimes 08:25:29 but by far not as often as in other languages 08:25:42 base would be an example of a variable 08:25:58