00:00:00 --- log: started forth/04.06.14 00:00:10 :) 00:00:31 I have to admit, your site is the largest Forth-related collection of links I've ever seen. 00:01:13 some day i'm going to raid a gob of other link directorys 00:01:15 Taygeta's site is large too, but it is too sparse and not well organized. 00:01:19 i also want to have a few more babies 00:01:38 I could cross organize stuff if I get the URLs 00:01:53 * kc5tja nods 00:02:01 Well, I think I have had a pretty productive weekend. 00:02:01 I don't think that they BYTE magazine scanned into chucks site has a link from his site 00:02:08 Or maybe it does 00:02:17 I did ok too 00:02:42 do you have another 5 min? 00:02:47 Despite the diminutive version number upgrade, the changes made were very, very sizable. 00:02:50 Sure. 00:02:55 about my site 00:03:07 what about the idea of puting the links into table format? 00:03:19 and having another collumn for description? 00:03:33 would it clutter it up or provide more information? 00:03:57 My only concern is that the links be easily organized and easily searchable. 00:04:05 You may want to consider adding a search engine to your page. 00:04:16 idea 11 00:04:23 like a google buddy? 00:04:31 This way, it'll be easier to cross-organize things as you can't always put things in one and only one category. 00:04:57 if its more convenient for my users, then probably i will consider it 00:04:59 Not sure how google buddies work. 00:05:20 i think Forth Inc has one 00:05:22 Well, we'll see. For now it seems quite well done. 00:05:30 If you want to put the links in a table, that's OK too, I guess. 00:05:47 or perhaps not to guild the lily 00:06:08 I think i might put the site in a .zip though so other may rip it and improve it as the see fit 00:07:05 --- join: htp123 (~tehsux@S010600055d233ab7.gv.shawcable.net) joined #forth 00:10:27 Anyway, I'm off to bed. 00:10:29 Night! 00:10:33 bye 00:10:37 --- quit: kc5tja ("THX QSO ES 73 DE KC5TJA/6 CL ES QRT AR SK") 00:17:21 me too 00:17:23 --- quit: LOOP-HOG () 00:29:00 --- quit: crc (Read error: 110 (Connection timed out)) 00:58:41 --- part: futhin left #forth 02:25:46 --- quit: htp123 (Read error: 110 (Connection timed out)) 02:43:45 --- join: hash (~hash@port-204-54-210.fastadsl.net.nz) joined #forth 02:47:25 --- join: Topaz (~top@exten-halls-131.soton.ac.uk) joined #forth 03:11:45 anyone here? 03:20:18 --- quit: hash () 04:11:45 --- join: qFox (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth 04:31:36 --- join: aum (~aum@port-204-54-210.fastadsl.net.nz) joined #forth 04:32:11 --- join: warpzero (~warpzero@dsl.142.mt.onewest.net) joined #forth 04:32:18 hi - anyone home? 04:33:13 hi 04:33:17 anyone home? 04:34:07 yo 04:34:22 warpzero: do you know forth? 04:34:30 nope 04:34:34 hehe 04:34:45 what brings you to #forth? 04:34:56 well i am trying to learn it eh 04:35:10 do you program in any OO languages? 04:38:01 warpzero liar ;p 04:38:17 yeah 04:38:24 i program in OO languages 04:38:27 qFox: am not! 04:38:28 i want to discuss something forth-related for a minute or three 04:38:43 you want what? 04:38:44 forth related? 04:38:45 here? 04:38:50 i'm not sure if thats ok... 04:38:58 * aum groans 04:39:03 lol 04:39:04 * qFox smiles 04:39:20 i'm adding the OO bits to a forth i'm writing 04:39:34 i have 2 options for calling methods: 04:40:01 1) objinst -> methodname [[ arg1 arg2 ... ]] invoke 04:40:19 2) [[ arg1 arg2 ... ]] objinst -> methodname invoke 04:40:29 sir... that looks nothing like forth 04:40:39 3) arg1 arg2 ... argn n objinst -> methodname invoke 04:40:43 how do you see the syntax in practice? 04:40:53 that's why i'm discussing it here 04:40:58 i'm writing a forth over python 04:41:14 so every item on the data stack is actually a python object 04:41:14 aha :) 04:41:29 and python has lots of methods that accept variable number of args 04:41:39 yeah ok, but from a forth point of view that doesnt matter... 04:41:56 wait, it was forth you wanted to do this oo stuff for right? 04:41:57 i guess i'm canvassing for opinions on which syntax looks/feels more natural to others 04:42:10 qFox: well, something forth-ish i'm cooking up 04:42:25 okay 04:42:33 i can't go full ANS, because ANS is not really applicable in a true OO environment 04:42:59 so, which of the three syntaxes looks most comfortable? 04:43:31 (i cant help ya there) 04:43:57 if you had to read someone else's code, which of the three would feel most understandable? 04:44:30 i guess 1 04:44:47 ok 04:45:11 but there are more ppl ;) 04:45:19 the '[[' and ']]' words, and the words in between, construct a python list (which gets pushed as a single entity to the data stack) 04:45:55 oh 04:46:00 so each cell on the stack can be any object - an int, a string, a float, a list, even a nested stack 04:46:27 but doesnt that kind of defy the whole forth way? 04:46:31 sure does 04:46:40 oki, just checking :) 04:46:52 the aim is to leverage the nice features of both python and forth 04:47:00 python has a sweet OO model 04:47:13 forth has great freedom and flexibility 04:48:43 the trouble i've had with pure forth is that to build truly high-level and general abstract layers takes shitloads of work in between 04:49:22 yep :) 04:50:24 like flying from London to Paris via Delhi, St Petersburg, the Moon 04:50:46 so i'm experimenting with building a forth on top of an existing oo platform 04:52:30 bbiab 05:03:20 Morning, Forthers. 05:03:59 --- quit: fridge (Remote closed the connection) 05:44:43 --- quit: warpzero (Read error: 110 (Connection timed out)) 06:34:12 --- quit: I440r (Read error: 104 (Connection reset by peer)) 07:05:58 --- join: poingie (~chatzilla@user-38ldu06.dialup.mindspring.com) joined #forth 07:26:08 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 07:29:17 Hi 07:31:40 --- join: I440r (~FooBlah@168-215-246-243.gen.twtelecom.net) joined #forth 07:34:03 Hey I440r :) 07:54:16 hi 07:59:08 --- join: ASau (~asau@158.250.48.204) joined #forth 07:59:16 Dobryjj vecher! 08:01:12 anyone knows about firewire plug ins? 08:01:21 seen one and know how many pins it is 08:05:59 Firewire is 4 wires, IIRC. 08:06:15 There seem to be a couple of different plug shapes though. 08:07:49 Oh. Hmm...maybe there are 4/6/8 wire versions... 08:08:49 Because a Google image search is turning up pictures of a 4-wire plug, but the sockets on the back of this box have 6 connectors. 08:09:27 http://www.keyelco.com/kec/standpro/specpage/spec149b.htm (8 and 6) 08:13:08 Ah. 6-pin ones provide power, 4-pin ones don't. 08:14:57 --- join: crc (crc@0-1pool176-23.nas6.philadelphia1.pa.us.da.qwest.net) joined #forth 08:15:10 Good evening, crc! 08:15:23 Good evening ASau 08:15:51 I've made quite a bit of progress on RetroForth 7 over the weekend :-) 08:18:02 I've written a working block editor, a generic libc-backed port for x86 OSes using ELF format, the start of an assembler, and the start of a target compiler. 08:20:29 crc. I've found a great thing. 08:20:50 I'm very impressed now. 08:21:17 How well do you know awk? 08:21:45 How good... 08:22:22 I don't use AWK, just SED 08:22:54 I'm aware. 08:23:07 But just take a look at gawk. 08:23:36 It has several features that make it very good tool for forther. 08:26:30 Such as? 08:27:05 Regexp as field separator. 08:27:24 Just try set it to [ \t\r\n\f]+ 08:27:50 Setting record separator is better. 08:46:47 --- quit: poingie (Read error: 110 (Connection timed out)) 08:58:58 --- quit: crc (Read error: 104 (Connection reset by peer)) 09:00:23 --- join: crc (crc@0-1pool176-23.nas6.philadelphia1.pa.us.da.qwest.net) joined #forth 09:01:08 --- join: neoBeta (bbbbbbbbbb@200-70-116-45.mrse.com.ar) joined #forth 09:01:14 --- quit: neoBeta (Client Quit) 09:02:20 Hm. 09:02:44 I just realized that a 3.5" floppy has 1440 kB of storage space, which happens to be the number of minutes in a day. 09:02:48 Magic. 09:05:57 God kvael, Robert! 09:06:09 Two 'l', but almost right. 09:06:16 Dobryjj vecher, ASau :) 09:07:31 BTW, who asked me of Alan Pratt's C-Forth? 09:07:52 I've built it one more time. 09:08:10 No idea. 09:08:21 It might be interesting from one point of view. 09:08:38 There's Forth sources (almost) in Forth. 09:10:34 It includes Forth compiler source written (almost) in Forth. 09:12:56 Nice. 09:13:04 I'm working on one of those now. 09:13:35 Planning to write the 5 or so words needed to start compiling in assembly language, the rest can be in Forth. 09:18:49 One evening, I felt so lonely on my Linux machine, that I coded maybe a quorter of Forth compiler. 09:20:23 I've stopped late at night at implementing division in high level words. :) 09:24:14 --- join: Herkamire (~stjohns@h000094d30ba2.ne.client2.attbi.com) joined #forth 09:25:09 Dobryjj vecher, Herkamire! 09:26:14 bodroe utro, ASau 09:26:43 Bodroe? 09:26:52 That's very good! 09:27:06 that's the female version of good or something? 09:27:13 * Herkamire took notes 09:31:01 --- join: neoBeta (bbbbbbbbbb@200-70-116-27.mrse.com.ar) joined #forth 09:31:21 Good evening, neoBeta! 09:31:59 Herkamire, my dictionary says that the closest translation for "bodryjj" is "cheerful." 09:33:37 Well, I think it's good if you have "bodroe utro." 09:39:45 --- quit: neoBeta (Read error: 54 (Connection reset by peer)) 09:55:29 --- quit: Herkamire (calvino.freenode.net irc.freenode.net) 09:55:29 --- quit: arke (calvino.freenode.net irc.freenode.net) 09:55:29 --- quit: ChanServ (calvino.freenode.net irc.freenode.net) 09:55:31 --- quit: madgarden (calvino.freenode.net irc.freenode.net) 09:55:31 --- quit: Topaz (calvino.freenode.net irc.freenode.net) 09:55:32 --- quit: ianp (calvino.freenode.net irc.freenode.net) 09:55:33 --- quit: Robert (calvino.freenode.net irc.freenode.net) 09:59:14 --- join: ChanServ (ChanServ@services.) joined #forth 09:59:14 --- join: Herkamire (~stjohns@h000094d30ba2.ne.client2.attbi.com) joined #forth 09:59:14 --- join: Topaz (~top@exten-halls-131.soton.ac.uk) joined #forth 09:59:14 --- join: arke (~Chris@wbar8.lax1-4-11-100-108.dsl-verizon.net) joined #forth 09:59:14 --- join: madgarden (~madgarden@Kitchener-HSE-ppp3576399.sympatico.ca) joined #forth 09:59:14 --- join: ianp (ian@inpuj.net) joined #forth 09:59:14 --- join: Robert (~snofs@c-bf5a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth 09:59:14 --- mode: irc.freenode.net set +o ChanServ 09:59:43 ASau: :) 10:08:22 --- part: crc left #forth 10:47:24 --- join: truecoder (ed@adsl-64-168-28-146.dsl.snfc21.pacbell.net) joined #forth 10:51:14 hi truecoder 10:51:41 Good evening, truecoder! 10:54:11 Hi. 11:19:08 Well, I like it. 11:19:24 One more company to join MBTI race. 11:20:47 I've just been pointed to y.a. collection of famous people classified. 11:24:29 --- quit: Topaz (Remote closed the connection) 11:25:28 --- join: slava (~slava@CPE00096ba44261-CM000e5cdfda14.cpe.net.cable.rogers.com) joined #forth 11:25:38 hey 11:25:52 Hi slava :) 11:26:22 Dobryjj vecher! 11:28:08 what's MBTI? 11:29:30 That's those well-known Meyers-Briggs typology. 11:30:00 This channel's disease of a month or two ago. 11:32:50 --- join: warpzero (~warpzero@dsl.142.mt.onewest.net) joined #forth 11:33:16 Dobryjj vecher, warpzero! 11:34:37 oh my 11:37:00 hello there 11:37:12 Hej, warpzero 11:37:32 --- nick: lalalim_ -> lalalim 11:39:15 * slava is trying to clean up some code he wrote several months ago and no longer understands :( 11:39:26 thats why we comment 11:40:19 Uhm, right.. 11:43:33 slava thats why god invented comments! 11:43:34 brb 11:43:35 pizza 11:43:48 Hehe 11:44:11 Reminds me, a guy in #hamradio came asking if he was allowed to speak with God on the radio. 11:44:25 Or let others do it 11:45:56 I440r. If code is not readable, maybe it needs to be thrown away? 11:49:10 i wish :/ 11:50:20 ok, code is a bit clearer now 11:54:55 i'm trying to convert my compiler into a 1-pass 11:58:02 A month ago I cleaned my code. 11:58:45 I think it is 4 times smaller now. 12:05:04 oh hi I440r 12:05:12 hi :) 12:05:47 ppc isforth plz 12:05:52 i don't have a linux x86 rite now 12:07:04 well... help port it :) 12:07:59 okay 12:08:02 you add with add 12:08:05 and you subtrackt with sub 12:08:09 does that help 12:08:25 yea... now keep going. only 47265467 other words to go :L) 12:09:52 holy shit 12:10:02 wrote some assembly in your life ever/ 12:11:12 hey did you know that we invented HLLs so we wouldn't have to write 47265467 words of assembly 12:11:33 warpzero. C-Forth? 12:11:42 what? 12:12:00 What about one of C-coded Forthes? 12:12:31 they suck 12:12:46 they don't have uber terminal windowing 12:13:36 Your suggestion? 12:14:43 bootstrap 12:14:44 ing 12:14:45 they arent IN asm 12:15:03 and HLL's are all abominations 12:15:09 specially c :P 12:15:13 forth is a hll 12:15:17 yes and no 12:15:30 my lanugague isn't a bomination 12:15:38 :) 12:15:43 warpzero. Have you tried to bootstrap Forth? 12:15:48 no 12:15:53 Well. 12:15:57 im too busy trying to bootstrap my language 12:15:58 I did. 12:16:32 But I've stopped. 12:16:44 why 12:16:58 I'm lazy. 12:17:12 I think of another, easier way to do it. 12:17:39 i'm like this tempted to use perl to bootstrap my language 12:18:11 I tried m4+gas. 12:18:28 didn't work? 12:18:36 Now I'm thinking of awk+gas. 12:18:41 anyone here done anything with rs-485 ? 12:18:49 is that like an ibm box 12:19:18 M4 works. 12:19:26 But it is hard enough. 12:19:53 im definitely not gunna try to bootstrap x86 first 12:20:14 I even succeded to write almost portable way. 12:20:34 eh ??? 12:22:54 Something like this: 12:23:27 I440r: have you done any rs-232 stuff? IIRC they're pretty similar... 12:23:30 code(type) times(1) list(dup,cat,emit,onep) repeat(1) list(semi) 12:23:37 not really :) 12:23:46 485 is multidrop 12:23:55 Correction: 12:24:00 code(type) times(1) list(dup,cat,emit,onep) repeat(1) list(drop,semi) 12:24:27 postfix notation is the coolest thing since 12:24:36 i don't even know 12:26:06 what is multidrop 12:26:11 I440r: define rs-485 12:26:15 Now I think that m4 is bad choice. 12:26:28 warp0 www.google.com 12:26:30 heheh 12:26:35 Hmm...I thought 485 was the one the mac used. 12:26:38 i hate you 12:27:04 warpzero: it's a serial port standard. 12:27:31 i got that 12:27:32 thanks 12:27:55 Oh. 422 probably was the mac one. 12:28:07 Well, I've to go. 12:28:09 Bye! 12:28:11 --- quit: ASau () 12:28:28 Yup. My bad. 13:29:27 --- join: lalalim_ (~lalalim@p508AABD5.dip.t-dialin.net) joined #forth 13:47:56 --- quit: lalalim (Read error: 110 (Connection timed out)) 13:51:00 --- join: wossname (wossname@Toronto-HSE-ppp3699743.sympatico.ca) joined #forth 13:57:16 --- quit: warpzero (Read error: 110 (Connection timed out)) 14:06:01 --- quit: slava (calvino.freenode.net irc.freenode.net) 14:06:01 --- quit: Herkamire (calvino.freenode.net irc.freenode.net) 14:06:01 --- quit: arke (calvino.freenode.net irc.freenode.net) 14:06:01 --- quit: ChanServ (calvino.freenode.net irc.freenode.net) 14:06:01 --- quit: lalalim_ (calvino.freenode.net irc.freenode.net) 14:06:02 --- quit: madgarden (calvino.freenode.net irc.freenode.net) 14:06:04 --- quit: ianp (calvino.freenode.net irc.freenode.net) 14:06:05 --- quit: Robert (calvino.freenode.net irc.freenode.net) 14:06:05 --- quit: truecoder (calvino.freenode.net irc.freenode.net) 14:06:05 --- quit: qFox (calvino.freenode.net irc.freenode.net) 14:06:05 --- quit: Klaw (calvino.freenode.net irc.freenode.net) 14:06:05 --- quit: Fractal (calvino.freenode.net irc.freenode.net) 14:06:06 --- quit: mur (calvino.freenode.net irc.freenode.net) 14:06:06 --- quit: cmeme (calvino.freenode.net irc.freenode.net) 14:06:07 --- quit: onetom (calvino.freenode.net irc.freenode.net) 14:06:07 --- quit: SDO (calvino.freenode.net irc.freenode.net) 14:06:07 --- quit: kuvos (calvino.freenode.net irc.freenode.net) 14:07:57 --- join: ChanServ (ChanServ@services.) joined #forth 14:07:57 --- join: lalalim_ (~lalalim@p508AABD5.dip.t-dialin.net) joined #forth 14:07:57 --- join: slava (~slava@CPE00096ba44261-CM000e5cdfda14.cpe.net.cable.rogers.com) joined #forth 14:07:57 --- join: truecoder (ed@adsl-64-168-28-146.dsl.snfc21.pacbell.net) joined #forth 14:07:57 --- join: Robert (~snofs@c-bf5a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth 14:07:57 --- join: ianp (ian@inpuj.net) joined #forth 14:07:57 --- join: madgarden (~madgarden@Kitchener-HSE-ppp3576399.sympatico.ca) joined #forth 14:07:57 --- join: arke (~Chris@wbar8.lax1-4-11-100-108.dsl-verizon.net) joined #forth 14:07:57 --- join: Herkamire (~stjohns@h000094d30ba2.ne.client2.attbi.com) joined #forth 14:07:57 --- join: qFox (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth 14:07:57 --- join: onetom (~tom@novtan.bio.u-szeged.hu) joined #forth 14:07:57 --- join: SDO (~SDO@co-trinidad1a-156.clspco.adelphia.net) joined #forth 14:07:57 --- join: kuvos (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth 14:07:57 --- join: cmeme (~cmeme@216.184.11.30.unused.swcp.com) joined #forth 14:07:57 --- join: Klaw (chuck@ip68-225-243-23.oc.oc.cox.net) joined #forth 14:07:57 --- join: Fractal (jah@selling.kernels.to.linus.torvalds.at.hcsw.org) joined #forth 14:07:57 --- join: mur (~mur@smtp.uiah.fi) joined #forth 14:07:57 --- mode: irc.freenode.net set +o ChanServ 14:08:41 I440r: have time for some isforth/x86 asm questions? (since warpzero brought up a PPC port, I'm browsing through the sources again...) 14:09:08 :) 14:09:10 ask away 14:09:59 in src/kernel/isforth.asm - chk_tty, you call the TCGETS syscall, then do 14:10:05 sub eax, 1 14:10:09 sbb eax, eax 14:10:12 ? 14:10:37 then store to a variable. I don't understand that bit. 14:10:52 ok. if eax is equal to zero eax becomes -1 and carry becomes 1 14:11:03 subtract eax from eax with borrow and eax becomes -1 14:11:13 i.e. a true - eax WAZ zero to begin with 14:11:40 so it's a 0= basically. 14:11:44 yup :) 14:12:08 thats just chekcing if stdin is a terminal (stdout too) 14:12:19 if you pipe data to isforth stdin is NOT a terminal, its a file 14:12:35 So the syscall fails. 14:12:39 Gotcha. 14:12:58 zyzcall failures return -1 14:13:02 syscalls even 14:13:04 right. 14:13:05 this may be a bug 14:13:12 im not looking at the sources lol 14:13:28 On PPC that gets returned in a condition register flag, so I wasn't sure. 14:13:43 I figured it would be quicker to ask you than find a good explanation of syscalls on x86 linux :) 14:14:19 that would be good 14:14:40 carry set means eax = errno 14:14:50 instead of "eax = negative means eax = ~errno" 14:15:44 yup. 14:17:57 Cool. Otherwise it's mostly making sense. 14:18:02 heh 14:18:20 be warned im thinking of dropping kernel 2.2 support 14:18:28 so i can do anonymous shared memory mappings 14:19:43 does isforth run on freebsd? 14:19:55 only in linux emu 14:20:04 damn 14:20:14 but porting would be a snap 14:22:04 --- nick: lalalim_ -> lalalim 14:23:25 just have to change how you do syscalls for BSD, right? 14:26:38 yes 14:26:46 actually you wont need to write any code 14:26:49 you can delete code 14:26:57 oh. there are some LINUX ONLY syscalls that i use 14:45:08 --- quit: aum (Read error: 104 (Connection reset by peer)) 15:00:43 --- join: warpzero (~warpzero@dsl.142.mt.onewest.net) joined #forth 15:14:05 --- join: topher (~chris@lsanca1-ar42-4-61-175-184.lsanca1.dsl-verizon.net) joined #forth 15:19:19 --- join: htp123 (~tehsux@S010600055d233ab7.gv.shawcable.net) joined #forth 15:27:39 --- quit: I440r ("Leaving") 15:38:57 --- join: Sonarman (~matt@adsl-64-160-164-149.dsl.snfc21.pacbell.net) joined #forth 15:40:58 --- nick: topher -> topher|away 15:47:09 --- quit: warpzero (Read error: 110 (Connection timed out)) 16:18:15 --- quit: SDO (Read error: 60 (Operation timed out)) 16:25:51 --- join: randolm (wossname@HSE-MTL-ppp66543.qc.sympatico.ca) joined #forth 16:28:57 --- join: I440r (~mark4@216-110-82-1.gen.twtelecom.net) joined #forth 16:29:06 --- quit: wossname (Read error: 60 (Operation timed out)) 16:34:03 --- quit: truecoder ("Leaving") 16:48:42 --- join: blockhead (default@dialin-181-tnt.nyc.bestweb.net) joined #forth 16:58:12 --- join: SDO (~SDO@co-trinidad1a-156.clspco.adelphia.net) joined #forth 17:11:22 --- join: aum (~aum@port-204-54-210.fastadsl.net.nz) joined #forth 17:11:38 --- nick: randolm -> wossname 17:16:37 --- quit: qFox ("this is mirc's last attempt of communication...") 17:25:54 --- quit: topher|away (Client Quit) 17:29:08 anyone home? 17:29:35 nope. 17:29:53 i'm creating some non-standard oo words for a forth i'm writing 17:30:16 i've got a syntax: objectname -> methodname 17:30:31 which puts on the stack a ref to the callable method 'methodname' 17:30:41 I didn't realize there was such a thing as standard oo words 17:30:49 but now i've gotta build a word for calling this method 17:30:51 options: 17:31:01 1) objectname -> methodname invoke 17:31:08 2) objectname ->() methodname 17:31:08 why put an XT instead of just colling it? 17:31:19 s/colling/calling/ 17:31:46 Herkamire: because i'm building atop python API, where calls are explicit - callable methods/funcs are themselves objects 17:31:52 objectname -> methodname () 17:32:11 madgarden: sounds good 17:32:18 Looks kinda familiar too. ;) 17:32:22 and for passing args, i'm doing: 17:32:38 arg1 arg2 arg3 3 >[] objectname -> methodname () 17:32:38 ??? 17:32:49 --- join: topher (~chris@lsanca1-ar42-4-61-175-184.lsanca1.dsl-verizon.net) joined #forth 17:32:57 aum: I thought you were doing this in forth? 17:32:57 problem is that python makes extensive use of variable-length arg lists 17:33:06 Herkamire: i'm implementing a forth over python 17:33:25 Can't you just create a seperate data stack in Python and use that for your Forth stack? 17:33:43 why don't you pass arguments on the stack like a normal person? 17:33:45 i've implemented a data stack in python 17:34:07 Herkamire: because i'm calling python methods, which very often take variable number of args 17:34:51 so? 17:35:06 do they need to be in a special format or something? 17:35:09 no 17:35:33 but for bridging from forth to python, i have to determine how many of the args on the stack should actually get passed 17:35:54 ok 17:36:06 i could do arg1 arg2 arg3 arg4 4 objectname ->() methodname 17:36:06 How about something like this instead... 17:36:19 how about: objectname -> methodname (4) 17:36:26 (( arg1 arg2 arg3 )) objectname -> methodname () 17:36:26 how about: objectname -> methodname (3) 17:36:51 (( puts a marker on the stack, and )) counts the arguments up to the marker 17:37:04 so then, a bunch of words '()', '(1)', '(2)', '(3)' which take that number of args from stack 17:37:15 madgarden: what if you want to pass the value the marker is? 17:37:26 aum: sure 17:37:43 aum: you're goal here is to be able to call any python method? 17:37:51 Herkamire: correct 17:38:16 so i can have '()', '(1)', '(2)'..., also (n) which takes n args 17:38:42 or you pass the number of args to () 17:38:44 Herkamire, that wouldn't be valid for Python functions, so why would it be needed? 17:38:46 * aum is implementing ANS words where he can, but not trying too hard to be 'standard' forth 17:38:50 ... 3 () 17:39:02 Herkamire: ok, makes sense 17:39:06 () means n args 17:39:12 (0) means 0 args 17:39:15 (1) means 1 arg 17:39:29 should suffice just to do up to (10) 17:39:32 ... should mean n args I'd think. 17:39:34 (maybe (1), (2) etc is a bad idea... not very flexible, and you have to define way to many words) 17:39:40 Or (...) 17:40:03 Is there ANY reason why the python method should care about the marker value? 17:40:05 hmm, gforth doesn't know '((' or '))' 17:40:24 aum, that was my invention just now. :) 17:40:32 But, it would make for a nice clean syntax. 17:40:37 madgarden: ok, np, easy to implement 17:41:10 i've implemented >[], and []>, which convert n items to a python list object, and vice versa 17:41:26 so 22 33 44 55 2 >[] -- 22 33 [44 55] 17:41:40 and 22 33 [44 55] []> -- 22 33 44 55 17:41:55 the [44 55] is a single item on the stack, being a python list object 17:45:45 aum: how are you turning objectname into a reference to the python object? 17:47:17 Herkamire: i've implemented 'import', which creates a constant whose name is the imported module, and whose value is the module object 17:47:22 eg 'import sys' 17:47:34 which creates a constant called 'sys', holding a ref to the 'sys' module object 17:47:50 then, i've got a word '->' which does attribute access 17:47:50 ahh 17:48:10 so 'sys -> path' pushes onto the stack a ref to the 'sys.path' object 17:49:07 and you [will] have some way to take a ref to a python intager object and turn it into a normal stack item? 17:49:24 Herkamire: everything on the stack is a python object 17:49:38 so a python integer object is _already_ a normal stack item 17:49:53 oh I see 17:50:29 so instead of 'c int' or 32-bit machine word being the ubiquitous stack datatype, i'm using python object 17:50:38 do you have a word that will give you a python object ref without creating a constant? 17:50:48 yes 17:50:48 -> 17:51:00 only 'import' creates a constant 17:51:19 --- quit: I440r (Read error: 60 (Operation timed out)) 17:51:22 i could have another version of 'import' which just pushes a ref to the module 17:51:42 and another version again, which takes the module name off the stack (instead of from read-ahead) 17:51:44 :) factor factor factor 17:52:01 rigth you are 17:52:10 you should be able to easily do all three of those with no duplicated code 17:52:11 i need a word '(import)' 17:52:28 (import) does all the grunt work 17:52:42 and a few sugar-coated variants which use it 17:53:18 : pyref word count $pyref ; 17:53:37 eh? 17:54:19 oh crap, I don't know how to make a constant from a name in ADDR U 17:55:11 : $pyref ( addr u -- python-object-reference ) 17:55:32 i'm not using 'addr u' for strings 17:55:38 i'm using python objects for strings 17:55:42 whatever 17:56:26 but i see what you mean - i should have a base word which takes the attribute name from the stack 17:56:47 right, and you can implement the others in forth on top of that 17:56:59 makes sense - good factoring 17:57:07 but i think i'll call it (->) 17:57:15 :) 17:57:36 the convention of naming lower-level words with enclosing parens - eg (if) 17:57:36 naming is sometimes the hard part :) 17:57:54 naming is way easy - devising names that make sense to others is a different issue 17:57:59 I don't think it's lower-level that get's perens. 17:58:09 in forth, one can easily disappear up one's back orifice 17:58:26 I think perens designates a word that is normally compiled for you (and you shouldn't ever have to type it in) 17:58:29 detractors call forth a 'write-only language' 17:59:03 people who don't know much about forth say all sorts of rediculous things about it 17:59:13 Herkamire: word! 17:59:20 people can write illegible code in any language 17:59:57 there's a great article on how to do this in C BTW :) 18:00:09 oh yes, obfuscated c hacking competitions 18:00:11