00:00:00 --- log: started forth/02.09.19 00:24:29 --- join: cyberlok (CyberLOK@cyberlok.no-ip.com) joined #forth 00:27:13 --- quit: I440r ("Reality Strikes Again!") 00:48:50 robert: heh, well that's a tiny little lie, i've been around before 00:48:55 while thefox was here 00:49:37 hello cyberlok, i'd greet you and ask you questions about if you are new to forth, etc, but i have to go to bed now, so good night :) 01:15:32 night man 01:15:43 sorry busy compiling 02:44:30 --- join: Serg_penguin (~Z@nat-ch1.nat.comex.ru) joined #forth 04:05:00 --- join: Soap` (~flop@202-0-42-22.cable.paradise.net.nz) joined #forth 04:07:09 --- join: cleverdra (julianf@0-1pool36-249.nas2.florence1.sc.us.da.qwest.net) joined #forth 04:24:04 --- quit: Serg_penguin (Read error: 110 (Connection timed out)) 04:33:15 --- quit: Soap` (Read error: 104 (Connection reset by peer)) 04:36:51 --- join: Soap` (~flop@202-0-42-22.cable.paradise.net.nz) joined #forth 04:43:02 --- quit: Soap` (Read error: 104 (Connection reset by peer)) 04:45:48 --- join: Soap` (~flop@202-0-42-22.cable.paradise.net.nz) joined #forth 04:51:23 --- join: sif (~sifforth@ip68-14-11-38.ri.ri.cox.net) joined #forth 04:51:23 Type sif: (or /msg sif to play in private) 05:35:09 Hello. 07:14:29 # 07:14:29 Find a Forth or APL Guru 07:14:30 In those worlds, the terser your code and the more bizarre the way it works, the more you are revered. 07:14:33 Hah! 07:14:47 That's one of the top hints for code obfuscation. 07:19:40 --- join: ASau (ASau@158.250.48.197) joined #forth 07:29:46 Hello. 07:31:48 --- join: Serg_penguin (~Z@nat-ch1.nat.comex.ru) joined #forth 07:31:53 hi 07:33:36 whois Robert 07:34:16 Hello :) 07:34:20 Robertis here ;-) 07:35:05 working hard - 25 rigs till tomorrow 18h 07:35:15 "rigs"? 07:35:21 What's that? 07:36:12 rig = box that does work 07:36:20 ham slang 08:16:30 --- quit: Serg_penguin ("time to go home...") 08:17:43 Hello! 08:17:57 How fast Serg run! 08:18:07 I can't even see him! 08:32:09 Hehe. 08:32:16 Russians run fast >:) 08:35:48 Once he said he's like sub-marine: 08:36:06 look, what is going there upwards 08:36:12 and get away. 08:36:15 :) 08:45:45 Hrm. I feel a need to write a gravity simulator in Forth. 08:45:57 * Robert thinks of which Forth to use. 08:46:16 It's not as funny without graphics, but I don't know any good Linux Forth with graphics. 08:46:29 People say BigForth isn't that good. 08:46:39 I also need floating-point support. 08:46:57 Hrm, ASCII graphics in GForth would be boring ;) 08:47:01 gforth can link to graphics libraries, and has floating-point support. 08:47:49 Yeah, I know it has floating-point support, but how do I get it to do graphics? 08:48:20 You dynamically load a graphics library. 08:49:14 Do you know any examples of that? 08:50:18 No, but I've posted a 'hello server' to clf which uses the same mechanism to use libc socket functions. 08:50:38 search for cleverjulian@hotmail.com; it might be the most recent post, actually. 08:50:44 OK, thanks. 08:50:55 Robert: Take a look at Zen's float point. 08:51:08 It is very simple and adaptive. 08:51:53 Zen? 08:52:36 DCC? 08:53:47 --- quit: TreyB () 08:53:56 I found it :) http://groups.google.com/groups?q=cleverjulian%40hotmail.com+group:comp.lang.forth.*&hl=en&lr=&ie=UTF-8&scoring=d&selm=76ea4fd3.0207150958.8fda95b%40posting.google.com&rnum=6 08:54:06 ASau: What _is_ Zen? 08:54:56 "Less is more" floating point implementation by Martin Tracy 08:55:15 Put in public domain in 1984 Forml Proceedings 08:56:21 Hmm.. sounds interesting, please DCC. 08:58:51 Er... how do I unpack .arc files? 08:59:01 find pkxarc.com file 08:59:05 OS? 08:59:45 DCC unpacked? 09:01:37 Nope.. I'm using Linux. 09:01:41 Maybe apt got something. 09:01:49 Why can't you use normal file formats? :-) 09:06:04 It's quite normal for me :) 09:06:08 --- join: XeF4 (kindle@12-245-111-146.client.attbi.com) joined #forth 09:06:09 DCC unpacked? 09:06:25 --- join: TreyB (~trey@cpe-66-87-192-27.tx.sprintbbd.net) joined #forth 09:13:44 hello all 09:13:52 Robert: Another version. 09:15:29 I also have adapted for my Forth version. 09:16:19 NB! <# and #> may differ in semantics from ANTI/F83 09:16:42 Hello, XeF4, TreyB, cyberloc! 09:17:29 Hey, Robert! 09:17:59 Do you need ones? 09:20:36 Robert, you use 32 bit forth, so you need to change constant 6554 to your MAXINT/10 value. 09:24:39 Oops. 09:24:45 Sorry, I went to eat :) 09:29:32 Thanks. 09:29:37 I'll check if I can unpack it. 09:30:39 What's a .seq file? 09:31:43 --- quit: thin (Read error: 110 (Connection timed out)) 09:31:45 plain text 09:32:03 You have to touninx it 09:32:10 it is in CR-LF format 09:32:16 Oh. 09:32:20 Yuck :) 09:32:54 This is my adapted version. 09:33:22 I've deleted all aknowledgements as it is at home. 09:35:15 "D r . C a c t u s" from Belarus... haha! 09:35:43 That's what unrar said :) 09:36:33 He is RuFIG memner. 09:36:33 Hrm. 09:36:46 He is RuFIG member. 09:36:47 Hrm. 09:36:56 How to convert to UNIX newlines? 09:37:13 2 crlfs? 09:37:18 Filter from something that eats CRs (13 decimal) 09:37:31 0xD 09:37:35 How is that done? 09:37:36 ah, from crlfs 09:37:41 onetom: Yes. 09:37:47 0x0d 0x0a -> 0x0a 09:38:04 begin key dup 13 = 0= then emit again 09:38:13 joking ;) 09:38:36 its a bit more complex 09:38:57 Pff... 09:38:58 Robert: Write a small C program, as a solution. 09:39:22 Well, I've done that a zillion times... Getting tired of it. I'll check if I can find an old one. 09:39:26 sed -e 's/\r$//' 09:39:45 : unixkey key dup 0< if bye then ; : dos2unix' unixkey dup 10 <> if 13 emit then emit ; : dos2unix unixkey dup 13 = if drop dos2unix' exit then emit ; : filter begin dos2unix again ; 09:39:45 begin key dup 13 = 0= if emit else drop then 09:39:47 tr -d \\r 09:39:48 When I needed to convert CR-LF to CR, I wrote asm 09:40:06 again 09:40:17 That's a little more complicated so that it only translates \r\n to \n, instead of rudely stripping \r 09:40:57 tho, rude stripping usually sufficies in case of text files 09:41:03 True. 09:41:10 * @%) 09:41:30 Why this is such a big problem??? 09:41:47 It's not a big problem at all. 09:41:57 Hrm. 09:42:18 onetom: This is simple text file. 09:42:22 One off: : dos2unix begin unixkey dup 13 <> if emit then again ; 09:42:31 The .scr file contains an awful lot of spaces. 09:42:46 onetom: I don't understand why to convert a non-text file in such way. 09:42:59 Damnit, I'll use cleverdra's gforth library example instead. 09:43:11 unixkey as defined above; all it does is call BYE if input has ended (and return the value of KEY otherwise) -- useful for filters like this. 09:43:31 Robert: .scr is blocks of 1K bytes 09:43:52 Robert: but tr -d \\r is the simplest 09:43:54 Robert - .scr would probably be what ASau said. You can convert that to a more readable file rather simply: 09:44:29 cleverdra: It is not probably. It _IS_. 09:44:49 ASau - is probably. Watch me make a file that ends in .scr and which is not what you describe. 09:44:59 cleverdra: It _IS_, because I sent it to Robert. :) 09:45:05 Oh =) 09:45:41 : block-line 64 0 do unixkey emit loop ; : block2file begin block-line cr again ; 09:46:04 cleverdra: I've heard about UNIX fold util. 09:46:07 voila. That doesn't strip spaces, though. 09:46:32 Are they so annoying? 09:46:37 ASau - fold(1)? I haven't heard of it; I'm still not as hip to shell-programming as I ought to be -- grew up on Perl =) 09:46:57 cleverdra: I don't know. 09:47:02 ASau - not at all annoying, except wasteful. The result is fine if you want to view the blocks. 09:47:03 cleverdra: I don't use UNIX. 09:47:36 ASau - what do you use? 09:47:52 dosonly.net :) 09:48:10 I'm tied to $$ $$$ 95 09:48:26 Anything that I can use is DOS. 09:48:50 Ah. 09:48:52 --- join: tathi (~josh@wsip68-15-54-54.ri.ri.cox.net) joined #forth 09:48:56 Hello. 09:49:01 Hello! 09:49:23 --- join: Herkamire (~jason@wsip68-15-54-54.ri.ri.cox.net) joined #forth 09:49:35 Hello to you too, Herkamire :) 09:49:35 Hello! 09:49:42 hi hi :) 09:52:38 Robert: You can use sed(1), I think. 09:52:45 Hmm... 09:52:50 Well, I'll mess with all that later. 09:52:53 Now I'll go out. 09:52:57 Thanks, and see you. 09:53:12 To convert \r\n -> \n and to strip spaces from converted .scr 09:54:32 perl -i.bak -pe 's/ *\r\n$/\n/' 09:57:16 I am to go. 09:57:29 Bye! 09:57:33 --- quit: ASau () 10:06:41 --- quit: cleverdra (Read error: 104 (Connection reset by peer)) 10:06:52 does anyone know why http://cln.no-ip.org/programs/httpd.asm will not work right in url submission engines? 10:06:59 what am I missing here? 10:11:38 --- join: cleverdra (julianf@0-1pool36-249.nas2.florence1.sc.us.da.qwest.net) joined #forth 10:12:08 cyberlok: the url and page (and http headers) look fine to me 10:13:48 demo site is 10:14:06 http://serverdisk.servehttp.com/ 10:14:22 which freshmeat and sourceforge freak out on 10:14:36 "encountered error... url seems to not be valid" 10:14:54 I also notice strange behavior with http benchmarking software 10:15:09 it literally does not exchange any info just kind of sits there 10:15:33 have tried siege httperf ptester ez_httpbench 10:15:44 I have no idea why though 10:49:36 --- quit: Robert ("I am the walrus. John Lennon wrote me in FORTH.") 10:54:42 --- join: Robert (~Robert@robost86.tsps1.freenet6.net) joined #forth 10:57:55 --- join: I440r (~mark4@1Cust199.tnt3.bloomington.in.da.uu.net) joined #forth 11:02:36 Hi, I440r. 11:04:03 hi 11:04:33 Why can't IsForth have floating point support? :) 11:05:47 I thought about making a gravity simulator in Forth (mostly to practice factoring ;), but I need floating point for that. 11:09:31 do it in integer :P 11:09:35 it will be faster 11:09:38 smaller 11:09:39 BETTER 11:10:32 Uhm. 11:10:35 Heh. 11:10:59 Sure, but think of things like the sun's mass (~10^30), the gravitation constant (~10^-11) etc. 11:11:11 lol 11:11:12 That's a factor of over 10^40 11:11:23 You'd need a 256 bit number :) 11:11:26 well if you INSIST on expressing numbers that way :P 11:11:55 there is NO floating point in nature 11:11:57 ANYWHERE 11:12:03 god uses integer arithmatic 11:12:28 God is a retarded guy working with chaos. 11:12:50 Well, I guess I could change the units a little bit... but still... you can't have too big differences in range/speed etc. 11:12:58 Floating-point is a lot more flexible. 11:14:19 Hrm.. I bet implementing a simulation of en einsteinan universe would be rather complicated. 11:15:36 heh 11:15:54 isforth may support FP some day 11:15:55 * Robert stays with uncle Newton. 11:16:00 Yay :) 11:16:01 it CANT do so without the assembler 11:16:12 Then code faster! :) 11:16:12 the assembler will support floating point instructions 11:16:22 :) 11:16:26 actually, im slowing down 11:16:32 Aww... 11:16:44 * Robert will have to do with GForth :) 11:17:05 trator!!!!!!! 11:17:36 How does the corresponding ANS words to var and !> work? 11:17:44 Isn't that "value" and "to"? 11:18:30 yes. ans words are GAY 11:18:45 Why? 11:19:01 I don't mean to be rude, but I think "to" loos better. 11:19:04 looks 11:20:39 Hrm... how does ANS structures work? 11:20:46 s/does/do/, I think... 11:21:10 err 11:21:14 no idea 11:21:22 :) 11:21:26 Hrm. 11:21:34 I think I'll do this whole shit in C instead. 11:21:38 I don't think ANS has 'structures'. What do you really mean? 11:21:40 Just to make you happy :) 11:22:00 Uhm... something like IsForth's structures. 11:22:11 isforths structures are not BY me heh 11:22:22 i got the idea off mrreach who also didnt invent them :) 11:22:28 Well, of course. 11:22:39 I mean, ANS has to have something similar to them, or? 11:22:48 There are many structure libraries. Some rather trivial; some more involved. Anton has written a struct.fs that is rather involved. 11:23:00 Robert - no, ANS doesn't at all. Structures are higher level than Forth. 11:23:47 All you're dealing with is memory; structures tend to simply allocate memory and then define tiny words that add a constant to a pointer. 11:24:00 Yes. 11:24:09 Well, I guess I could do it that way. 11:24:36 That's how you will it. The only question is what syntax you'll use, and how nice and readable you want the structure definition. 11:25:36 GForth says its struct is defined as : struct does> 2@ ; 11:25:52 ugh 11:26:02 might as well just do ' 2@ alias struct 11:26:15 erm no 11:26:15 Yes; that one word's definition will not make its use clear. 11:26:26 struct is a 2value 11:26:35 a double sized var 11:26:41 weird code heh 11:27:14 0 constant struct : end-struct constant ; : foo dup create , does> @ + ; : %int 4 + foo ; : %char 1+ foo ; 11:27:35 ugh 11:27:46 struct %int >cursor:x %int >cursor:y end-struct cursor 11:27:56 Of course, you could define it at 'struct' instead. 11:28:06 Also, you might want multiples. 2 %INT , etc. 11:28:25 Robert - google for Anton struct.fs; there's a document online which explains the system that's in gforth. 11:28:40 sorry had to run to the bank 11:29:51 cyberlok just robbed his bank! 11:30:46 actually, I can't find it. Nevermind; it's in the gforth info document. 11:31:18 I440r - ugh? 11:32:35 heh 11:33:57 --- join: Kitanin (~clark@SCF61185.ab.hsia.telus.net) joined #forth 11:35:04 I440r hehe 11:35:12 hey I440r you good with asm? 11:35:37 need to find out why my httpd.asm does not work with http bench software and engines like sourceforge and freshmeat 11:36:00 it seems as though (observed in http bench software... all of the ones I can find) 11:36:10 the server sends no data to the program 11:36:33 like the bench program is requesting something or sending something to the server which it does not know how to respond to 11:42:55 --- join: I440r_ (~mark4@1Cust132.tnt1.bloomington.in.da.uu.net) joined #forth 11:43:19 grrr 11:43:25 fscking PPP crapped out on me 11:43:28 somehow 11:46:50 dial up? 11:49:15 use of PPP tends to imply dial-up. 11:50:49 re discussion of Erlang and concurrency, see http://www.erlang.se/productinfo/erlang_tech.shtml -- which gives a good brief overview of the language, including its concurrency model. 11:54:26 brb hang on 11:55:51 --- quit: I440r (No route to host) 12:02:41 back 12:02:43 sorta 12:02:45 :) 12:03:56 hehe 12:06:35 * I440r_ never has bee "all" there :) 12:06:39 been 12:06:51 btw the last thing you asked was "am i good with asm" did you see my reply ? 12:07:04 thats just about when my PPP decided to go south on me 12:09:48 nope 12:09:55 I think I have it solved though 12:10:01 need to check it though 12:10:28 hehe 12:10:30 cool 12:10:43 how are you doing dns queries btw - using libs ? 12:12:14 ummmm 12:12:21 some of them 12:12:26 depends 12:12:45 but mainly the httpd and ftpd dont need dns query 12:12:48 isforth uses NONE. its the prime directive 12:12:48 so they dont have it 12:12:57 no. the daemons dont 12:13:03 the ones that use it are like ping and others 12:13:05 uclibc 12:13:08 busybox 12:13:21 siege --url="http://63.228.32.91/index.html" --benchmark --time=1M 12:13:25 gives me this 12:13:34 Transaction rate: 0.00 trans/sec 12:13:41 like what the hell 12:13:46 why not :) 12:13:50 yet 12:14:29 run some sort of snoop program and see if your box is TRYING to send stuff 12:14:31 or not 12:14:43 hmmm 12:15:38 bingo I think I have it fixed 12:15:59 heh 12:16:21 one sec though got to make sure 12:17:59 comparing 12:18:17 takes it over a minute to bench now so it gets averages and stuff 12:20:16 crap need to test same file or else it will not work out right 12:20:26 :) 12:22:26 --- join: thin (~thin@h24-66-221-120.cg.shawcable.net) joined #forth 12:23:56 * I440r_ hides 12:24:06 hehe 12:24:11 ok almost done 12:24:26 i was hiding from thin... SHHH - dont tell him you saw me :) 12:25:32 OK, I440r, I didn't see you, I440r. 12:25:47 I440r, why are you (I440r) hiding from thin? 12:26:40 http://serverdisk.servehttp.com/stats.txt 12:26:43 check it out 12:26:59 heh 12:27:05 looking at url... 12:27:42 also 12:27:47 another thing to keep in mind 12:28:01 the 586 was tested on that machine locally over 127.0.0.1 12:28:01 your faster than apache :) 12:28:06 yet 12:28:12 the remote 486 beat local 12:28:44 * cleverdra notes that many are faster than apache -- such as Roxen and Yaws. 12:29:09 apache is NOT a good web server 12:29:09 also, *10 concurrent users*? One sec... 12:29:10 cleverdra its cuz it is entirely in asm 12:29:34 yeah siege -b plus 10 concurrent users 12:29:37 what that does 12:29:49 entirely in asm does not imply fast, but OK. 12:29:53 http://www.sics.se/~joe/apachevsyaws.html 12:29:53 is siege hammers httpd as fast as it can on 10 channels 12:30:17 * thin pounces on i440r_ 12:30:24 lol 12:30:28 thin is slow today :P 12:30:34 Compare your asm httpd at 80,000 concurrent sessions vs Yaws, and then tell me how fast it is =) 12:30:42 cleverdra its 535 bytes 12:30:47 its straight to the point access 12:30:53 * thin purrs, he is a friendly cat ;) 12:31:08 tell me whos 386 will be doing 80,000 12:31:12 cyberlok - that's useful, I suppose, if you are a cracker. 12:31:30 900k diskimage man 12:31:34 hope to get it even smaller 12:31:41 and include more stuff 12:31:41 * cleverdra nods. 12:32:01 lets do ummm 300 concurrent users 12:32:59 hmmm lets start 200 12:33:16 Apache dies at about 4000 concurrent users. I suppose that that's a good upper-limit for your program. 12:33:36 need to find out how to press past the thread limit 12:33:37 (in this test, I mean) 12:33:45 cyberlok - 'thread limit'? 12:33:49 siege just told me I cant go past 256 12:33:58 In order to exceed that number, you will have 12:34:00 cyberlok - 256 whats? 12:34:06 to run siege as root user or have your system 12:34:13 administrator increase your nproc limit. 12:34:22 oh, a restriction on siege. 12:34:24 concurrent user similation 12:34:35 Write your own. 12:35:20 ummm dont feel like it 12:35:22 --- join: proteusguy (~proteusgu@24-197-147-197.charterga.net) joined #forth 12:35:24 * cleverdra shrugs. 12:37:28 doin apache now 12:41:04 http://serverdisk.servehttp.com/stats2.txt 12:41:18 I am wooping apache's ass 12:41:53 hehe 12:43:23 --- quit: thin () 12:43:47 hehe cant wait to try and get forth stuff in here 12:44:12 heh 12:44:26 * Robert <-- coding forth :D 12:44:29 u been reading up onm forth ? 12:46:20 hard to read on something that has little documentation 12:46:43 well it would be nice to make how you say 12:47:11 a kind of busyboxish thing but instead of ping shell and other stuff to put in ftpd httpd smtpd pop3d and others 12:47:18 hm? Lots of documentation on Forth. What kind are you looking for? 12:47:33 a starting point 12:47:37 level 1 novice 12:48:21 gforth comes with two tutorials; forth.com's shareware SwiftForth comes with two pdf manuals. 12:50:06 The gforth info document is also a decent, if somewhat incomplete, reference manual. 12:50:28 hmmm kk 12:50:37 what is the big thing with forth anyways? 12:50:52 all programming languages have a main goal 12:51:06 This is an illusion. 12:51:34 haha 12:51:58 well the way I440r_ talks about it it is sounding like a cross btwn asm and c 12:52:05 also, forth tends to be a DIY thing, so the "main goal" can differ widely depending on who you ask :) 12:52:10 Forth is a general purpose programming language that delivers an innately low level of abstraction, but to which many language features are nicely tame. 12:52:19 cyber argh its nothing like c hehe 12:52:27 brb i gotta walk the siberian attack poodle:) 12:53:00 I mean in concept not execution or syntax 12:53:27 cyberlok - that *is* the concept. If you mean something else, you should give me an example. 12:53:29 cuz you talk about libs and then also extreme tight control individually 12:53:45 I don't follow you... 12:53:53 c is kind of a pass this off to something else kind of method 12:54:03 pass the hot potato 12:54:06 I still don't follow you... 12:54:10 ummm 12:54:11 ok 12:54:16 take for example httpd.asm 12:54:33 in c httpd.c would have calls to libs to take care of this or that 12:54:40 then have some main functions within the c 12:54:47 or maybe not even that 12:55:05 while with httpd.asm everything is taken care of by 1 bin 12:55:09 Very well. My immediate attempts at understanding your previous statements would lead me to conclude that your httpd.asm is a 'pass the hot potato' program, delivering the real work to the OS through interrupts. 12:55:23 heh 12:55:25 haha good point 12:55:47 ok its MORE of a passer then lets say asm 12:56:04 which is actually perfectly fine 12:56:12 except when you are talking size 12:56:41 Forth can be either -- and so can ASM, so I don't see how this is a useful discriminator between languages. 12:57:15 ok ok it is more popular in c and higher level programming languages 12:57:56 So, might the term you're looking for be 'low level programming language'? 12:59:04 its the language of programmers I personally dont believe in programming levels, but I figured that would be the easiest way of getting an understanding 12:59:25 Anyway, I've written Forth programs to A) control a PC speaker through (very low level) PPI interface, B) deal with TCP/IP communication through system calls, C) deal with TCP/IP communication through library calls. 12:59:57 cleverdra yup exactly that is what I440r_ said it could do which makes it very nice for diskette distros 13:00:19 he also said he could get it all (server stuff) in like 20k 13:00:24 which I HAVE to see :) 13:03:45 --- quit: proteusguy (Read error: 110 (Connection timed out)) 13:04:22 --- join: proteusguy (~proteusgu@24-197-147-197.charterga.net) joined #forth 13:15:18 maybe 13:15:22 im not promising anything heh 13:17:29 hehe 13:32:57 --- quit: cleverdra (Connection timed out) 14:03:09