00:00:00 --- log: started forth/04.06.24 00:15:20 --- quit: jDoctor ("Lost terminal") 00:25:10 --- quit: Herkamire ("bed") 00:28:16 core linux, compiling kernel 00:28:17 ay 00:28:18 yay 00:28:37 --- join: lalalim (~lalalim@p508ABC17.dip.t-dialin.net) joined #forth 00:28:43 which in itself is somewhat redundant, since as soon as I do, I'm "rebooting" and installing 2.6.7 00:28:44 --- quit: lalalim_ (Read error: 60 (Operation timed out)) 00:28:49 ;) 00:35:48 --- join: htp123 (~tehsux@S0106000d6151238b.gv.shawcable.net) joined #forth 00:56:05 ack, this is pissing me off 00:56:10 VMWare should _NOT_ crash 01:43:37 --- join: zardon (~zardon@S0106000d6151238b.gv.shawcable.net) joined #forth 01:47:24 does anyone know how to make gdb disassemble memory that has changed since load time? 01:50:02 --- quit: I4_work (Read error: 110 (Connection timed out)) 02:40:17 --- join: Topaz (~top@spc1-horn1-6-0-cust217.cosh.broadband.ntl.com) joined #forth 03:17:26 --- quit: htp123 ("*") 03:31:18 --- quit: Klaw (Read error: 104 (Connection reset by peer)) 04:09:56 --- join: qFox (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth 04:51:03 --- join: I4_work (~mark4@168-215-246-243.gen.twtelecom.net) joined #forth 04:52:46 --- nick: I4_work -> I44wrk 04:52:57 hrm cant decide :P 04:53:03 --- nick: I44wrk -> I4_wrk 05:02:23 --- quit: skylan (orwell.freenode.net irc.freenode.net) 05:02:23 --- quit: zardon (orwell.freenode.net irc.freenode.net) 05:02:23 --- quit: kuvos (orwell.freenode.net irc.freenode.net) 05:02:23 --- quit: cmeme (orwell.freenode.net irc.freenode.net) 05:02:25 --- quit: ianp (orwell.freenode.net irc.freenode.net) 05:02:25 --- quit: ChanServ (orwell.freenode.net irc.freenode.net) 05:02:25 --- quit: lalalim (orwell.freenode.net irc.freenode.net) 05:02:25 --- quit: fridge (orwell.freenode.net irc.freenode.net) 05:02:26 --- quit: SDO (orwell.freenode.net irc.freenode.net) 05:02:26 --- quit: Topaz (orwell.freenode.net irc.freenode.net) 05:02:27 --- quit: Robert (orwell.freenode.net irc.freenode.net) 05:03:59 bubye 05:04:01 :p 05:05:07 --- join: ChanServ (ChanServ@services.) joined #forth 05:05:07 --- join: Topaz (~top@spc1-horn1-6-0-cust217.cosh.broadband.ntl.com) joined #forth 05:05:07 --- join: zardon (~zardon@S0106000d6151238b.gv.shawcable.net) joined #forth 05:05:07 --- join: lalalim (~lalalim@p508ABC17.dip.t-dialin.net) joined #forth 05:05:07 --- join: ianp (ian@inpuj.net) joined #forth 05:05:07 --- join: fridge (~fridge@dsl-203-33-162-85.NSW.netspace.net.au) joined #forth 05:05:07 --- join: SDO (~SDO@68.170.20.201) joined #forth 05:05:07 --- join: skylan (~sjh@vickesh01-4831.tbaytel.net) joined #forth 05:05:07 --- join: cmeme (~cmeme@216.184.11.30.unused.swcp.com) joined #forth 05:05:07 --- join: kuvos (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth 05:05:07 --- join: Robert (~snofs@c-bf5a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth 05:05:07 --- mode: irc.freenode.net set +o ChanServ 05:05:26 --- quit: skylan (orwell.freenode.net irc.freenode.net) 05:05:26 --- quit: zardon (orwell.freenode.net irc.freenode.net) 05:05:26 --- quit: kuvos (orwell.freenode.net irc.freenode.net) 05:05:27 --- quit: cmeme (orwell.freenode.net irc.freenode.net) 05:05:27 --- quit: ianp (orwell.freenode.net irc.freenode.net) 05:05:28 --- quit: ChanServ (orwell.freenode.net irc.freenode.net) 05:05:28 --- quit: fridge (orwell.freenode.net irc.freenode.net) 05:05:28 --- quit: lalalim (orwell.freenode.net irc.freenode.net) 05:05:29 --- quit: SDO (orwell.freenode.net irc.freenode.net) 05:05:29 --- quit: Topaz (orwell.freenode.net irc.freenode.net) 05:05:30 --- quit: Robert (orwell.freenode.net irc.freenode.net) 05:05:52 --- join: ChanServ (ChanServ@services.) joined #forth 05:05:52 --- join: Topaz (~top@spc1-horn1-6-0-cust217.cosh.broadband.ntl.com) joined #forth 05:05:52 --- join: zardon (~zardon@S0106000d6151238b.gv.shawcable.net) joined #forth 05:05:52 --- join: lalalim (~lalalim@p508ABC17.dip.t-dialin.net) joined #forth 05:05:52 --- join: ianp (ian@inpuj.net) joined #forth 05:05:52 --- join: fridge (~fridge@dsl-203-33-162-85.NSW.netspace.net.au) joined #forth 05:05:52 --- join: SDO (~SDO@68.170.20.201) joined #forth 05:05:52 --- join: skylan (~sjh@vickesh01-4831.tbaytel.net) joined #forth 05:05:52 --- join: cmeme (~cmeme@216.184.11.30.unused.swcp.com) joined #forth 05:05:52 --- join: kuvos (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth 05:05:52 --- join: Robert (~snofs@c-bf5a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth 05:05:52 --- mode: irc.freenode.net set +o ChanServ 07:15:16 --- join: onetom (~tom@novtan.bio.u-szeged.hu) joined #forth 07:15:50 hi 07:17:13 save6 200003 16 flux13 6 80003 move16 disk13 floppy!3 ; 07:17:18 tadaa 07:17:59 i can paste color text from within my editor through screen into irssi 07:18:21 :) 07:18:34 if anyone interested, just ask me about the solution 07:18:51 (what is not perfect yet...) 07:19:25 i need help developing a color text experimenting enviroment 07:19:49 what do you need it for? 07:20:42 exploring the strengths & weaknesses of colored text 07:21:06 hehe 07:21:39 eg we can use it for a more conscise xml representation 07:23:04 and of course we can use it for colorforth like programming languages 07:24:03 yeah 07:31:23 i imagine that many configuration files - even in a linux environment - can b replaced w a colored text what can b "run" instead of just "parsing" 07:32:29 eg 1 can generate virtualhosts for apache by definig : vhost-type1 ... dir ... log ... whatever ; 07:32:56 and then vhost-typeN asd.domain.com 07:33:49 instead of specifying 07:33:50 07:33:50 ServerName aron.dunacafe.hu 07:33:50 DocumentRoot /var/www/aron.dunacafe.hu 07:33:50 ErrorLog /var/log/apache2/dunacafe.hu/error.log 07:33:52 CustomLog /var/log/apache2/dunacafe.hu/access.log combined 07:33:55 07:34:09 such a bunch of lines tenth of times 07:35:26 most configuration files r hard to maintain after a while, coz u need variables cycles etc.. 07:37:41 true 07:57:39 wheres herk ? 08:18:01 --- quit: madwork (Read error: 104 (Connection reset by peer)) 08:20:32 --- join: madwork (~madgarden@derby.metrics.com) joined #forth 09:45:28 hi 09:47:01 Hi 09:55:18 hi 09:58:16 4 CED3 is16 a11 fullscreen6 editor8 and5 it14 rox! 09:58:30 8) 09:58:31 Nice colors. ;) 09:58:54 its pasted here from within ced 10:01:06 the main code is just these few line 10:01:07 s 10:01:08 : addspace >r 3 ( ^C) c!+ r> irccolors swap iexec 16 ( ^V) c!+ 16 c!+ 20 c!+ ; 10:01:08 : enter param ln> nip inp do i c@ dup space? if addspace else 10:01:11 c!+ then loop ascii ' c!+ 0 swap c! cmd system ; 10:01:14 --- join: Herkamire (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth 10:01:25 eh.. herk missed it ;) 10:02:18 hi onetom 10:03:47 hello herk 10:04:46 im just hacking tile & my color editor these days & im just very happy about it, coz it works reasonably 10:05:20 what's tile again? 10:05:26 and it is also able to interoperate w my irc client through gnu screen 10:05:43 so we could easyly talk about colored code now 10:05:58 nifty 10:06:19 ftp.gnu.org/gnu/non-gnu/tile-forth 10:06:45 4 CED3 is16 a11 fullscreen6 editor8 and5 it14 rox! 10:07:29 the basic editor functionalities r almost perfectly finished 10:07:42 bound checking for DEL is missing 10:07:48 backspace is missing 10:08:02 you are writing CED? 10:08:16 and intelligent SPACE & ENTER meaning hasnt been definied yet 10:08:21 Color EDitor 10:09:20 what are you doing with space and enter? 10:09:42 in herkforth ENTER executes the word on the spot, and space inserts it into the editor. 10:09:56 dont know 10:10:13 anyway its just an editor yet, not a workbench 10:11:02 @the moment enter pastes the "current line" into the irc window of my screen 10:12:25 tile looks very cool 10:12:35 its clean at least... 10:12:59 but its very dumb/minimal to some extent :( 10:13:23 yeah 10:13:31 my editor is pretty bare bones in many ways 10:13:39 i had to implement a bunch of words before i could do anything useful w it in linux.. 10:13:44 but has some mind bogglingly useful bits too 10:14:05 namely pressing SHIFT-F to go to the definition of the word under the cursor 10:15:27 there was no raw keyboard input, no system word, no double wordset, no "c,", no file wordset, no >= <= u> ... 10:15:55 hmm.. & how do u look for its definition? 10:16:04 raw keyboard... you mean getting keyups? 10:16:19 i mean key downs at first ;) 10:16:27 there is no KEY? 10:16:55 generally I don't look, unless it's moved 10:17:02 yes, there is, but the terminal is left in line buffering canonical mode.. 10:17:09 HashSet is a Set; so 10:17:10 my source words are an index into the dictionary and color bits 10:17:14 er ww 10:17:33 oh, that's a pain. I hate that terminal shit 10:18:03 Herkamire: ah, so ur editor is the chuck type huffman compressed halfcompiling editor? 10:18:04 so I just grad the field in the dictionary that says where the source is, make sure it's still there, and display. 10:18:22 (if it's not there then I search for a red word that points to that dict entry) 10:18:30 no, 10:18:33 no huffman shit 10:18:41 no text stored in the source tokens at all 10:18:46 terminal shit, huffman shit.. :) 10:18:47 the text is in the dictionary 10:19:24 source tokens are a 12 bit index into the dictionary, and 4 bits to indicate color 10:20:20 so 0x0031 would be the fourth word in the dictionary in green (1). 10:20:47 ooh :)) gotit. funny 10:21:00 how do you do colors in irc? 10:21:04 (irssi) 10:21:28 bbiab 10:27:31 back 10:27:52 I think the idea for storing source this way came from aha 10:28:08 and colorforth. I sorta merged the ideas 10:28:54 soon I will store each defenition seperately in a database, so I don't have to have that silly search thing. 10:29:11 and the stupid code I have that tries to guess where the begining and end of the definitions are 10:29:31 (the begining of a normal word is trivial to find, but for a constant, it is not) 10:54:07 Herkamire: google irc colors, 1st hit http://www.ircbeginner.com/ircinfo/colors.html 10:57:55 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 10:58:58 Herkamire: 10:58:58 #include irccolors.fs create irccolors 10:58:59 ] n red green white cyan purple yellow brown 10:58:59 n n darkgray n n n n n 10:58:59 n red green white cyan purple yellow brown 10:59:01 n n lightgray n n n n n lightgreen [ 10:59:03 : addspace >r 3 ( ^C) c!+ r> irccolors swap iexec 16 ( ^V) c!+ 16 c!+ 20 c!+ ; 10:59:06 : enter param ln> nip inp do i c@ dup space? if addspace else 10:59:09 c!+ then loop ascii ' c!+ 0 swap c! cmd system ; 11:08:44 #include? 11:13:12 qFox: learn C 11:13:17 onetom: thanks 11:13:29 oh come on, that was uncalled for :) 11:13:34 i was referring to #include in forth 11:15:23 well, that's inconsistant 11:15:34 ? 11:15:52 it says ^O puts it back to normal colors, then later it says it's ^0 11:16:44 Linux Terminal? 11:17:37 tathi: is there such a thing as ^0 on the linux terminal? 11:17:54 the doc says it works on mIRC and PIRCH 11:18:04 I thought irssi might be the same 11:18:06 apparently not 11:18:19 No. 11:18:28 I'd assume it's a typo. 11:19:59 At least, there's no standard character code for it... 11:20:24 usually ctrl-whatever means subtract $40 from the char code. 11:22:41 that would make it negative... 11:22:48 well, here goes 11:23:08 2blue 11:23:10 yeah. Normally you just have A-Z, and [ \ ] ^ _ 11:23:12 wow 11:23:53 in irssi you hit ^C to change colors. that's nerve wracking 11:24:30 bold 11:24:40 test 11:24:44 Herkamire: ? 11:24:46 Herkamire: how? 11:25:03 arke: showing blue here. 11:25:42 arke: http://wiki.irssi.org/cgi-bin/twiki/view/Irssi/Colors 11:25:57 ^C then a number 11:26:05 it shows funky on the input line, but shows correctly in the channel 11:26:14 arke: that might just be how irssi does bold though. 11:27:10 bold is ^B 11:27:17 1hi 11:27:20 2hi 11:27:24 2hi 11:27:27 3hi 11:27:30 4hi 11:27:32 ;) 11:31:27 4red 11:34:15 --- quit: qFox ("this.is.not a.real.netsplit") 11:34:25 --- join: qFox (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth 11:50:50 --- quit: tathi ("leaving") 12:17:22 --- join: wossname (wossname@HSE-MTL-ppp73479.qc.sympatico.ca) joined #forth 12:43:28 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 12:49:32 --- join: proteusguy (~proteusgu@69.79.24.31) joined #forth 12:53:37 Herkamire: sorry not being totaly clear about this color stuff, but i thought that code snippet will b enlighting enough 12:54:12 but of course it isnt. i didnt even tell the number base :) 12:55:33 onetom: I just glanced over it and it didn't make sense 12:57:20 tho thats the code what pastes the transformed colored source into the irc window ;) 12:58:19 100 $allot constant cmd 12:58:19 " screen -p 0 -X stuff '" dup cmd $cat $free cmd dup $length + constant param 12:58:19 : c!+ over c! 1+ ; : $end + 0 swap c! ; 12:58:27 i missed these lines 13:20:18 hi wossname 13:36:32 --- quit: tathi ("leaving") 13:38:30 --- quit: madwork (Read error: 110 (Connection timed out)) 13:44:37 --- quit: wossname (orwell.freenode.net irc.freenode.net) 13:44:37 --- quit: ianp (orwell.freenode.net irc.freenode.net) 13:44:37 --- quit: ChanServ (orwell.freenode.net irc.freenode.net) 13:44:38 --- quit: fridge (orwell.freenode.net irc.freenode.net) 13:44:38 --- quit: lalalim (orwell.freenode.net irc.freenode.net) 13:44:38 --- quit: SDO (orwell.freenode.net irc.freenode.net) 13:44:38 --- quit: proteusguy (orwell.freenode.net irc.freenode.net) 13:44:39 --- quit: Topaz (orwell.freenode.net irc.freenode.net) 13:47:42 --- join: ChanServ (ChanServ@services.) joined #forth 13:47:42 --- join: lalalim (~lalalim@p508ABC17.dip.t-dialin.net) joined #forth 13:47:42 --- join: SDO (~SDO@68.170.20.201) joined #forth 13:47:42 --- join: fridge (~fridge@dsl-203-33-162-85.NSW.netspace.net.au) joined #forth 13:47:42 --- mode: irc.freenode.net set +o ChanServ 13:48:16 --- join: wossname (wossname@HSE-MTL-ppp73479.qc.sympatico.ca) joined #forth 13:48:16 --- join: ianp (ian@inpuj.net) joined #forth 13:48:51 --- join: proteusguy (~proteusgu@69.79.24.31) joined #forth 13:48:51 --- join: Topaz (~top@spc1-horn1-6-0-cust217.cosh.broadband.ntl.com) joined #forth 13:58:01 --- join: Herkamir1 (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth 13:58:08 --- quit: Herkamir1 (Client Quit) 13:58:16 oops 14:07:50 --- quit: proteusguy (No route to host) 14:08:04 --- join: _proteus (~proteusgu@69.79.24.31) joined #forth 14:11:30 7w 19 14:12:22 mur: isn't it 6w20 though? I believe that was 0the attack for the Silver F0rost Blade.... 14:12:51 typo 14:13:14 ;) 14:13:51 3 14:15:08 hmm 14:15:36 i wish i had something in my head 14:16:12 don't we all, mur? 14:20:43 --- quit: wossname (";lA") 14:24:15 why does pygmyforth use PUSH and POP instead of >R and R>? 14:24:49 machine forth instructions? 14:24:51 i think its closer to the machine 14:25:05 since the return stack would be the hw stack? 14:25:09 you can always make aliases for them :) 14:34:12 thats why i dont use pygmy 14:34:40 that and it uses blocks (ick) 14:34:56 other than that pygmy is some awesome code! 14:36:49 you don' like blocks? 14:37:00 no 14:37:04 not for source files 14:37:43 you have to have your sources all stretched out horizontally leaving no room for ANY comments unless you fragment your sources by interleaving source and comments 14:37:49 code code code code code code codde 14:37:50 code code code code code code codde 14:37:55 comment comment comment 14:37:57 code code code code code code codde 14:37:58 comment comment comment 14:37:59 comment comment comment 14:38:03 looks like SHIT 14:38:57 --- join: warpzero (~warpzero@mi015.dn178.umontana.edu) joined #forth 14:39:22 I4_wrk: how about 14:39:24 code 14:39:25 comment 14:39:27 code 14:39:28 comment 14:39:29 bad 14:39:31 code 14:39:33 comment 14:39:36 ? 14:39:38 :) 14:39:39 it fragments the code up 14:39:50 it should. 14:39:51 use shadow blocks 14:39:55 leave just CODE to the left and comments to the right 14:40:00 hey guys 14:40:01 have pg up/pg down browse through source blocks 14:40:02 shaddow blocks are a horrible idea 14:40:05 each little thing should use its own little block, with its corresponding shadow block. 14:40:23 you have to keep switching from code to comments. my way you get both code and comments on the same page at the same time and they dont interfear with each other 14:40:32 you can chose to ignore either column 14:40:56 theres a reason newspapers publish articles in tall thin columns, its easier to scan down and speed read them 14:41:40 slava commented on my source files the other day... which looks better. blocks with shaddows or my sources ? 14:42:17 isforth has block files btw (but ive not fully tested the code) 14:42:33 im not sure about loading a block etc, probably doesnt work 14:43:56 I always figured you should display the shadow block to the right of the source block 14:44:37 there's certainly plenty of room on my screen 14:44:44 :) 14:44:54 I almost forgot about that. 14:45:00 blocks are just so "last century" 14:45:03 quite a while ago I planned out my screen usage 14:45:40 split most of the screen in half, so there are two big areas side by side taking up the top 3/4 of the screen 14:45:54 then the bottom can be used for text output/history 14:46:02 and a row somewhere for status information 14:46:35 aaarrrrg. mysql won't let me set a default for a TEXT field 14:46:52 it's right in the manual that you can't 14:47:09 I quit ;) 14:47:29 lol 14:47:40 is it just mysql or is it a SQL thin? 14:47:59 make it a binary field and just make sure you never put anything but TEXT in it :) 14:50:20 TEXT (just a case insensitive BLOB) is a field for large amounts of data 14:50:27 i think there must be some better way to organize source. there's a few nice things about blocks but they're too simple 14:50:49 I'm looking for better ways to organize source too 14:51:10 I'll get coding on my next idea as soon as I can make up my mind on how the db should work 14:53:43 here is the main() function of the new factor interpreter coded in C: 14:53:44 init_arena(); 14:53:44 load_image("factor.image"); 14:53:44 init_environment(); 14:53:45 run(); 14:53:46 source files NEED to be in flat files so that they can be editied with ANY editor on any system 14:54:14 no they don't 14:55:04 there's plenty of systems that get along fine without doing it that way 14:55:28 but you cant share the code 14:55:35 yes I can 14:55:35 they work on "that" system only 14:55:37 I do so regularly 14:55:50 not unless you convert it to a flat file 14:55:58 and why add an extra step just to share your sources 14:56:13 smalltalk 14:56:21 so I can avoid 8 extra steps to edit/parse etc 14:57:28 not using text makes my system much simpler 14:57:46 it's easy enough to output a text version of sources 14:57:55 in many cases I just copy from the terminal 14:58:23 colorforth has a couple utilities to convert blocks of source to html 14:58:50 it is not my intention to write large hunks of code that people can use in other forths as is. 14:59:30 to do that my forth would have to be too conventional 14:59:38 and would lose most of it's value 14:59:57 I want people to learn from the way it works, and perhaps find an idea that they'd like to use 15:00:24 the code won't work on other forths, so it doesn't matter if it's in a format you can paste it there. 15:02:56 it is important to be that it is in a format where people can easily browse and learn about it 15:03:08 which flat text files aren't all that good at imo 15:03:44 I've started writing some code to translate herkforth code to html that will display in color with everything being a link to it's definition. 15:07:29 Herkamire: cool 15:07:31 Herkamire: i have the same thing but by decomping definitions in the currently running image 15:15:39 excellent. image loading/saving seems to work for simple cases at least. 15:17:11 I compile to native ppc code, so decompiling is tricky 15:17:27 ah 15:17:35 (and pointless, since I have the source) 15:18:12 --- join: doublec (~doublec@202.36.204.41) joined #forth 15:20:56 hi doublec 15:21:05 hi slava 15:21:18 slava, I got some initial support for callbacks now. 15:21:58 "Press me" [ run-something ] quot-href 15:22:02 the C interpreter now successfully loads an image. 15:22:15 That will put an anchor link that when pressed runs the code then returns back to display the calling page. 15:22:27 cool 15:22:29 call, the c interpreter has made a lot of progress then! 15:22:39 I struck a problem with 'workspaces' in 0.60.1 though 15:22:45 what is it? 15:23:13 --- join: crc (crc@0-1pool176-28.nas6.philadelphia1.pa.us.da.qwest.net) joined #forth 15:23:16 I'm sending it in an email now...it's easy to explain that way 15:25:11 slava, just sent 15:25:30 basically starting an interpreter pointing to an existing filestore database gives errors. 15:27:04 i'll take a look. 15:27:10 btw the C interpreter will have a much simpler persistence system. 15:27:26 it loads/saves images, which are basically memory dumps with relocation done on load 15:28:26 cool 15:28:58 An advantage of the C interpreter is being able to run it on smaller devices (pda and phones come to mind). 15:33:07 doublec: i'm going to write a cross-compiler now 15:33:55 slava, to compile to C code? 15:44:15 --- quit: SDO (Read error: 110 (Connection timed out)) 15:48:33 doublec: no 15:48:37 doublec: to produce images 15:49:49 --- part: crc left #forth 15:52:21 slava, oh I see. So you'll be able to cross compile to ARM on an intel machine? Nice. 15:59:55 bye 15:59:56 work 16:04:28 --- quit: warpzero ("Tried to warn you about Chino and Daddy Gee, but I can't seem to get to you through the U.S. Mail.") 16:13:35 doublec: images depend on byte order and word size 16:13:43 although i'll lift the first restriction 16:15:44 doublec: grab the latest jedit.org/factor/CFactor.zip 16:16:45 slava, will do thanks. 16:19:04 slava, I get errors running build.sh when trying to compile some of the files. 16:19:16 shuffle.c:6: error: conflicting types for `primitive_drop' 16:19:17 stack.h:4: error: previous declaration of `primitive_drop' 16:19:17 shuffle.c: In function `primitive_drop': 16:19:17 shuffle.c:7: error: too many arguments to function `dpop' 16:19:17 shuffle.c: At top level: 16:19:19 shuffle.c:11: error: conflicting types for `primitive_dup' 16:19:21 stack.h:2: error: previous declaration of `primitive_dup' 16:19:23 shuffle.c: In function `primitive_dup': 16:19:25 shu 16:19:37 I'm using gcc 3.3, do I need 3.4? 16:19:38 what gcc release? 16:20:00 it compiles here with 3.4 and 2.96 16:20:17 i'll do some investigating... 16:20:31 oh my fault 16:20:34 zip file includes obsolete sources 16:20:38 ok 16:20:39 shuffle.c is not part if the system anymore 16:21:13 ok get it again 16:22:01 ok that worked, thanks. 16:22:10 its very crude 16:22:13 try this: 16:22:22 ./f a -- running it with any cli arg will write factor.iamge 16:22:33 ./f -- running with no arg loads factor.image and executes the bootstrap quotation 16:22:39 in this case, the bootstrap quotation is [ 30 fib ] 16:22:59 cool, runs fine. 16:23:07 now try 30 fib in java factor 16:23:20 --- join: madgarden (~madgarden@Kitchener-HSE-ppp3576399.sympatico.ca) joined #forth 16:23:33 hi madgarden 16:23:42 hi ya, slava. 16:23:46 quite a bit faster in C isn't it 16:24:02 doublec: look at run.c 16:24:34 madgarden: where in canada are you? 16:24:44 madgarden: youre making a C forth no? 16:25:22 I'm in Kitchener, ON. And yea, I've got a C Forth-like system. 16:25:40 http://sourceforge.net/projects/forthy 16:26:19 slava, so env.boot should contain the boot quotation that gets run on load of the image? 16:26:29 doublec: yup -- see image.c, load_image() 16:27:56 doublec: note that right now 'env' is a global variable but at some stage it will become a macro or a parameter that's passed around 16:28:30 If you want multiple VM instances, that's a good idea. 16:28:52 or threading 16:28:58 doublec: language change! 16:29:33 doublec: FactorArrayStack is now FactorArray -- stack-push/pop is now array-push/pop, and there is array-nth and array-length words 16:30:24 slava, ok I'll make a note of that. 16:30:38 it means you can use arrays now basically. 16:31:04 slava, globals are also bad one some platforms like Symbian where you can't have global or static variables. 16:31:36 (Symbian being a phone OS) 16:33:14 doublec: on second thought -- a few 'array' words exist for working with bare java arrays, so i'm naming the new concept 'vectors' 16:33:31 5 over vector-push 16:33:51 slava, ok. 16:34:28 slava, how do push/pop on vectors compare to cons/car on lists for efficiency? 16:34:57 (ie. do the vectors grow and copy when reaching a certain length) 16:35:37 a pop is just a array fetch + pointer decrement 16:35:45 a push is a possible array doubling, + pointer increment 16:36:08 the benefit here is vector-nth for random access, and constant-time vector-length 16:37:07 --- join: blockhead (default@dialin-646-tnt.nyc.bestweb.net) joined #forth 16:37:34 definitely 16:40:32 doublec: i'd like to use your cont-responder for the wiki-editor. 16:40:36 --- nick: _proteus -> proteusguy 16:40:40 doublec: to avoid the explicit post request handler, and edit? urls 16:41:27 slava, it should work fine for that. It really needs a means of expiring the saved continuations though or memory will run out eventually. 16:41:42 slava, which is my next task. 16:41:50 I'm uploading the newer version now. 16:42:59 www.double.co.nz/factor/cont.txt and www.double.co.nz/factor/cont-responder.factor 16:43:13 This has the support for anchors to quotations 16:43:20 also what if people forge the continuation ids 16:43:28 i know its almost impossible but you should check for ip address too 16:43:56 brb 16:44:01 Yeah, that's the plan. Ideally the continuation id would be a cryptographically secure random number too 16:45:03 --- quit: madgarden ("*frotz*") 16:45:56 --- join: madgarden (~madgarden@Kitchener-HSE-ppp3576399.sympatico.ca) joined #forth 16:50:58 I also want to allow attaching callbacks to form input elements and buttons. This allows you to run code when select lists are chosen, data is entered in text fields, etc. It makes creating re-usable widgets quite easy. 16:51:23 that sounds really cool 16:51:42 ideally you'd create forms from s-expressions... 16:51:49 yes. 16:52:11 although I'm torn between generating everything into a string of html that gets sent in one go, or streaming the html as needed. 16:52:32 doublec: ? 16:52:34 streaming means it renders in the browser immediately and you can do nifty tricks with 'publish/subscribe' or javascript rpc things. 16:52:52 yes I use at the moment. 16:53:04 But that builds a string up which gets sent in one go. 16:53:23 (which is what I want it to do of course) 16:53:31 but streaming directly to the socket has some advantages 16:53:49 : with-string-stream ( quot -- string ) 16:53:49 #! Call the quotation with standard output bound to a string output 16:53:49 #! stream. Return the string on exit. 16:53:49 [ 16:53:49 "stdio" put call "stdio" get stream>str 16:53:50 ] bind ; 16:53:54 is what I use 16:54:30 maybe its better to have with-stream ( stream quot -- ) 16:54:36 which rebinds stdio and runs the quotation 16:54:46 good idea 16:54:48 at the mo we have with-html-stream with almost identical code 16:55:26 do you think s-expression based building of html would be a good fit with factor? 16:55:41 or sticking with building things up via [ ... ] form , etc 17:01:07 by s-expression building i mean the [ ... ] form :) 17:01:35 cool. I was wondering why you would be advocating (... ) :) 17:01:39 new functions: logor lognot logand 17:01:45 bitwise operations 17:01:46 How would attributes be handled in that form? 17:01:58 never mind 17:02:02