00:00:00 --- log: started forth/03.03.12 00:44:35 --- join: Speuler (~Speuler@mnch-d9ba4d42.pool.mediaWays.net) joined #forth 00:57:10 --- part: njd left #forth 02:47:16 --- join: Serg_Penguin (~Z@nat-ch1.nat.comex.ru) joined #forth 02:47:36 hi 02:55:16 HI 02:55:34 i set up modem at home ^) 03:01:39 :) 03:07:51 but it's 14400 and holds just for few minutes - bad line 03:11:27 thats not very good 03:11:32 you seen it :( 03:16:57 even good USR courier held just 10-15 min at 9600-19700 03:17:38 when it rains here the phones get really bad and people se a lot of dropouts 03:17:44 but the rest of the time its ok 03:19:58 here it shorts at spring, then underground cabling tunnels are full of water 03:27:28 'morning 03:32:00 hi 03:35:46 --- quit: Serg_Penguin () 03:40:24 Heh. 04:40:49 --- join: Serg_Penguin (~Z@nat-ch1.nat.comex.ru) joined #forth 05:03:05 --- quit: Serg_Penguin () 07:15:43 --- join: gilbertdeb (~gilbert@fl-nked-ubr2-c3a-29.dad.adelphia.net) joined #forth 07:16:06 . 07:16:21 2 2 + . 07:16:22 5 ok 07:17:40 :) 07:17:49 redefined math 07:18:20 That's F84. 07:20:02 or intel FPU 07:20:09 --- join: Herkamire (~jason@wsip68-15-54-54.ri.ri.cox.net) joined #forth 07:21:04 Right 07:21:09 Hi Herkamire 07:21:23 tach Herkamire 07:21:39 moin moin Herkamire 07:23:08 hi all :) 07:24:09 gilbertdeb: I'm starting to want to try mod_python 07:24:35 oh? 07:24:46 what is wrong with mod_perl :P 07:26:42 for cgi ? 07:32:51 gilbertdeb: heh :) 07:33:47 Speuler: yeah basically. I haven't read anything much about mod_python. but mod_perl is an apache module that (amung many other things) runs cgi scripts without (much) startup overhead 07:39:34 i've put a forth cgi interface together 07:39:58 isn't completed, but most common stuff is easy to do with it 07:40:07 including javascript mouseover 07:40:27 any chance it might be useful for you ? 07:42:57 got up to the point where i can define dialogs, which return input to server, but isn't passed through to script yet 07:43:51 dialogs? 07:44:07 you mean popup windows in the browser? 07:44:21 dialog box, to enter data into, editable text fields 07:44:44 yes 07:44:51 you are evil 07:45:10 that's standard stuff. nothing extraordinary 07:45:20 everybody hat's popup windows 07:45:24 but forth syntax looks ok 07:45:29 except sometimes for a small help page 07:45:59 example: "enter your comments" .... "send" 07:46:39 just one or several fields you can edit, and submit 07:47:47 forms are cool 07:48:04 popup windows are not cool 07:48:21 there's an example output (but no images on the server, they remain blank) on http://logilan.info/~l/forthcgi/index1.html, the input script is there too, as well as the cgo lib 07:48:41 w/o the images it doesn't look so good though, and misses parts of the functionality 07:49:52 forth is the script. index1.html is the output 07:50:06 i can see whether i can put some images on so it looks nicer 07:50:21 mean to use it for forthfreak.net 07:50:59 links of the images 3rd row also are not right 07:51:09 just copied it from another machine 07:51:43 where's the source? 07:51:49 same directory 07:52:04 still a bit rough 07:52:12 cgi.f 07:52:33 thats the lib. forth is the source 07:52:40 forth includes cgi.lib 07:56:11 cool 07:56:24 I'm scared to try it in Mozilla for fear of javascript, but I like it in general 07:56:49 only mouseover uses javascript 07:56:53 rest is plain html 07:57:06 I thought you said there were dialogs 07:57:09 w/o the images, mouseover doesn't even show 07:57:33 yes, saying dialog i mean form 07:58:08 but opening links, excuting forth word for example, works 07:58:09 ahhhh :) 07:58:27 that is, using a form, with link on button which opens another page 07:58:56 again, you need images to see the effect, have put that on the demo images only 07:59:34 I get links to shuttle.logil.... which can't be found 07:59:35 see "third image is special" in source 07:59:51 right. that's where the index.html has been generated. 07:59:54 no dns entry on the net 08:00:18 i'll move the stuff to the internet server, so it is more complete there 08:00:49 was first meant as an examplle how the forth script source looks like 08:01:01 why do you have obsolute links 08:01:27 to not have to bother about the actual location of the images 08:01:38 lazyness 08:02:16 got the images in docroot on the server i made it 08:02:57 all paths in script source. 08:04:28 what happens with the images at gnome, dark-gnome etc is, if you press one, page rebuld, andimage selection is shown in 3rd soaring.gif 08:04:59 there's a forth word associated to each of them 08:05:07 the demo1 demo2 demo3 ... 08:05:46 which is executed by commandline demo5 08:06:03 commandline is the forth word taken from paramter in url, 08:06:19 demo5 is the word executed if no parameter given after url 08:06:27 the default, you may say 08:06:48 each of demo1...5 just generates the html text to show an image 08:07:29 120 120 0 imagelink http://shuttle.logilan/cgi-bin/forth?demo1 /gnome.jpg 08:07:42 this is how "demo1" is passed to "commandline" 08:09:46 so if you press the first image, from the url containing demo1 the demo1 is extracted, and executed by commandline 08:09:47 need to be careful with that though 08:09:51 sealed voc, for example 08:09:53 because it opens the forth interpreter ... 08:10:01 forth searches for word1 in its dictionary, and executes it if found 08:10:19 don't want to allow to have forth exexute any word which is passed with the url ... 08:10:59 several ways possible to protect against execution of unwanted words 08:11:43 not a structural problem. more, the case now because of the experimental state of the forthcgi 08:19:48 just put a prefix on the word before you search the dict 08:20:23 that's one, yes 08:20:44 I want to write a forth apache module 08:20:52 but I can't figure out how to limit the execution time 08:22:15 maybe I'll do it with CGI 08:22:17 instead 08:22:19 watchdog the code. i.e. saves time, periodically checks for elapsed time. 08:24:05 I don't know how 08:25:31 my current forth is native PPC, and my webserver is not PPC, so I don't know if I'll bother for a long time (if ever 08:25:54 I'd like to see a mod_forth with a nice VM 08:29:22 I had a forth that would have worked nicely, and I started making such a module a while back. but I couldn't get the watchdog to work. 08:29:56 I followed the example and stuff from the apache docs, but it wouldn't kill my forth after the time was up 08:30:27 --- join: I440r (~mark4@1Cust15.tnt1.bloomington.in.da.uu.net) joined #forth 08:30:37 Hi I440r 08:30:48 * Robert plays with IsForth.. messing around with some file I/O 08:32:56 Herkamire: it needs a 'bye' 08:32:57 :D 08:34:18 Herkamire: you got "pause" in your forth ? 08:35:08 called by, for example, key, key? ms , all polling loops, etc ? 08:35:25 emit may be a good candiate for pause as well 08:36:03 (pause is the word forth multitaskers tend to hook themselves into the system) 08:36:29 often, it is a deferred word 08:36:37 If I make an apache module, it needs protection agains infinate loops 08:37:03 with that, you could plug in a forth word which check elapsed time, and does bye after timeout 08:38:00 : a begin again ; a 08:38:11 pause requires discipline. if you write an infinite loop, you'd call pause from it 08:38:20 --- join: tathi (~josh@wsip68-15-54-54.ri.ri.cox.net) joined #forth 08:38:29 : a begin pause again ; 08:38:51 Speuler: no. I need protection for my webserver against : a begin again ; a 08:39:16 I don't need pause in order to write code that exits. 08:39:25 I need a webserver that doen't die when you give it bad code 08:39:54 Don't let untrusted users run code. Don't run untested code on production servers. 08:40:12 besides, putting pause in there doesn't help. I want the script to output a page and exit. putting pause in there doesn't make it exit. 08:41:15 TreyB: good policy, but still I don't want it to be that easy to take apache down. 08:41:28 Herkamire: check out stopafter 08:41:46 stopafter - kill commands after a given time 08:41:46 Run the given command, killing it with a SIGHUP or other signal after a specified amount of time. It forks a child process to kill the parent process 08:41:46 and any children if still present. Package also includes the pushafter utility, which waits for a specified time before running the given command. 08:45:38 a bit complex, but maybe I can find something in the source 08:47:01 you may compile pause into the word, after begin 08:47:08 by begin 08:47:31 doesn't save you from users just redefining begin 08:48:01 : foo r> recurse ; 08:48:13 : foo rdrop recurse ; 08:49:18 --- quit: ChanServ (Shutting Down) 08:49:25 --- join: ChanServ (ChanServ@services.) joined #forth 08:49:25 --- mode: brunner.freenode.net set +o ChanServ 08:49:25 --- mode: ChanServ set +l 83 08:49:35 wb chanserv 08:49:41 It died! 08:50:10 ChanServ: wb 08:50:17 hmm 08:56:12 --- quit: skylan (Read error: 104 (Connection reset by peer)) 08:59:31 --- join: skylan (~sjh@Sprint287.tbaytel.net) joined #forth 09:15:11 --- quit: I440r ("Reality Strikes Again") 10:05:22 hmmm. 10:07:26 Hi gilbert. 10:08:24 --- join: wossname (wossname@HSE-QuebecCity-ppp82041.qc.sympatico.ca) joined #forth 10:10:52 hi 10:16:22 forthfreak.de has been brought over, and is active again now 10:16:27 .net still not 10:16:32 guess, tomorrow 10:16:51 currenly, forthfreak.de mirrors logilan.info/~l 10:17:54 provided, the dns you use has not still cached the old ip address 10:18:03 $1400000 var heap-size \ minimum size of heap to allocate (20 megs) 10:18:07 Heh. 10:18:18 20 MB... that's alot of RAM. 10:18:56 Robert: should only be allocated if actually used 10:19:28 Hmm.. are you sure? 10:19:30 i mean, you can reserve lots of heap, but only the amount of mem used goes away 10:19:36 It sounds like if he allocates it immediatley. 10:19:45 with gforth, yes. iforth, i don't know. but i suppose too 10:19:51 Hrm. 10:19:59 isforth i mean 10:20:01 You never know with i440r :) 10:20:12 easy to find out 10:20:49 ps -aux | grep isforth | cut -f 6 -d " " 10:21:52 hmm 10:21:58 ps aux | grep isforth 10:23:08 robert 9891 0.0 0.1 21708 100 pts/4 S 19:21 0:00 ./isforth 10:23:20 oops :) 10:24:48 even doing 10000000 allot with gforth doesn't take up ram. only when writing to ram is it alloted 10:25:23 :) 10:26:34 and then only the amount of ram actually written to, not the whole chunk 10:27:03 i.e. 10000000 allot here 1000000 - 50000 bl fill costs 50 kb ram 10:27:13 --- quit: gilbertdeb (""Monk has left the building"") 10:27:56 great for arrays with low fill grades 10:28:03 for hashing, for example 10:28:13 Nice. 10:28:31 work on is440r 10:30:03 btw, bashforth behaves similarily, but that's not bashforth's merit, but bash's 10:30:40 can allot 200 mbyte, and fill some cells there. hardly extra ram used 10:31:23 but would take about 50 hours to scan the mem :) 10:31:35 Heh 10:33:06 hmm. more like 25...30 10:33:31 less ! 10:33:40 i'm calculating a high-level loop 10:33:55 but scan is a ("quicker") primitive ! 10:34:08 *g*... etc 10:35:12 hey, speuler, what's your project? 10:35:12 --- nick: Speuler -> SpeedManiac 10:35:28 hi woss 10:35:41 none really going on. toying around a bit 10:36:30 i've geen bringing people to desperation with my bashforth thingy 10:37:09 ah, crazy :l 10:37:21 :) 10:37:42 got about all important things. 10:37:54 catch/throw, create does> and the like 10:38:05 include 10:38:18 but no restoring system from "binary" image 10:38:30 saving works, restoring not, as far 10:38:54 also, no key?, no multitasking 10:38:57 catch/throw, how are those used? 10:39:26 : foo key esc = if 12345 throw then ; 10:40:04 : bar ['] foo catch ?dup if ." oops" then ; 10:41:04 : newcli ['] quit catch ." interpreter aborted with error " . ; 10:41:49 -3 throw ( "stack overflow" ) 10:42:03 --- nick: SpeedManiac -> Speuler 10:42:05 Hmm. 10:42:51 catch works a bit like execute 10:43:06 throw roughly like exit 10:43:53 but stacks pointers are restored to value they had before catch, and the throw value is returned 10:44:17 Oh, okay. 10:44:53 so you can have all kind of garbage on either stack, when doing throw 10:45:08 don't need to bother about cleaning up 10:46:54 : foo >r >r >r 10 0 do strange = if boo throw then loop r> r> r> ; 10:49:28 nestable, btw 10:50:47 wossname: bashforth is on www.forthfreak.de . require bash 2.05 . maybe, just maybe, 2.04 ok untested. 10:50:55 2.03 definitely not 10:52:29 i'm looking 10:53:17 good god 10:53:58 sweet cheeses 10:54:43 this is cool :) 10:55:12 that's a qualification for a change 10:55:32 i'm used to "insane", "sick", "stupid" and the like 10:55:56 Haha 10:56:02 * Robert wonders why. 10:56:46 Die domain "www.forthfreak.net" ist gesperrt. :D 10:57:14 your dns cached the prev ip 10:57:37 Robert: it is http:/logilan.info/~l for you 10:57:56 Okay.. well, .de worked. 10:58:29 ok 10:59:28 doc word is a nice feature 11:00:06 --- join: semmelkant43 (~knoppix@h80n2fls31o270.telia.com) joined #forth 11:00:13 but for most words, it just shows stack effect right now 11:00:21 : test ." Moo." cr ; 11:00:21 test 11:00:21 ok 11:00:21 test 11:00:23 Moo. 11:00:26 ok 11:00:28 Why does it print "test" there? 11:00:35 It's SLOW :) 11:00:46 semmelkant43, heh. 11:01:27 it is 11:02:00 but it is twice as fast as it once was :) 11:02:15 :) 11:02:21 does anyone know of an ANSforth (extended)core word equivalent to fig-Forths P! and P@, for port i/o? 11:02:27 ahem .. 11:02:30 half as slow ... 11:02:37 ;) 11:02:59 Does ANS use any machine-specific words? 11:03:26 semmelkant43: under linux, direct port access is for most ports forbidden 11:03:45 as user, i mean 11:04:33 Speuler: true, but not for root 11:05:09 got asm in your forth ? 11:05:23 ioperm... 11:05:33 I do, but Id rather find a way to do it without asm 11:05:55 Lycka till ;P 11:06:00 fett med tack 11:06:08 god dag 11:06:15 :) 11:06:37 swedish voc exhausted 11:07:19 fett met tack = "all right, thank you" ? 11:08:02 "fett med tack" ="thanks alot" 11:08:05 slang 11:08:47 fat with thanks ? 11:08:53 exactly 11:09:00 or kindof 11:09:03 maybe not 11:09:04 blargh 11:09:27 plenty of thank you 11:09:43 yes 11:09:47 something like that 11:10:12 got to go, will return later, in a couple of week. lumpen suger, slutovning vareld. 11:10:20 --- quit: semmelkant43 (Remote closed the connection) 11:12:11 Robert: you can try out catch/throw now :) 11:12:33 bashforth's catch/throw is ans conform 11:12:45 works as explained above 11:14:12 Okay. 11:14:31 Poor semmelkant :| 11:15:47 bai~ 11:16:09 --- quit: wossname ("!@$") 11:29:20 just notices i've put in a bug with negative numbers 11:29:24 noticed 11:38:59 found it 11:43:40 uploaded. v42 fixed negative number bug, introduced with v0.40. also added time&date 13:13:33 --- quit: flyfly ("using sirc version 2.211+KSIRC/1.2.4") 13:14:05 * Herkamire reads the backlog, and laughs at "sweet cheeses" 13:14:15 --- join: flyfly (~marekb@ip164.ktvprerov.cz) joined #forth 13:15:26 --- join: I440r (~mark4@1Cust122.tnt1.bloomington.in.da.uu.net) joined #forth 13:50:54 --- quit: I440r ("Reality Strikes Again") 14:29:08 --- quit: tathi ("leaving") 14:36:04 --- quit: Herkamire ("leaving") 15:15:03 --- join: Herkamire (~jason@ip68-9-68-123.ri.ri.cox.net) joined #forth 16:11:39 --- quit: flyfly ("using sirc version 2.211+KSIRC/1.2.4") 16:39:36 --- quit: Speuler (Read error: 60 (Operation timed out)) 16:47:34 --- join: Speuler (~Speuler@mnch-d9ba4d42.pool.mediaWays.net) joined #forth 17:27:32 --- join: I440r (~mark4@1Cust212.tnt1.bloomington.in.da.uu.net) joined #forth 17:44:37 --- join: tathi (~josh@ip68-9-58-105.ri.ri.cox.net) joined #forth 17:49:07 --- quit: Klaw (Read error: 104 (Connection reset by peer)) 18:01:36 --- quit: I440r ("Reality Strikes Again") 18:21:31 --- quit: tathi ("leaving") 18:50:15 --- join: gilbertdeb (~gilbert@fl-nked-ubr2-c3a-29.dad.adelphia.net) joined #forth 18:50:36 . 18:51:39 there's a speck on my screen 18:51:48 oh 18:51:51 it moves 18:52:17 hahah 18:52:53 www.forthfreak.de is online again 18:53:08 www.forthfreak.net not yet 18:56:26 forthfreak.de is good enough no? 18:57:20 .net sounds nicer 18:57:26 forthfreak net 18:58:00 i'm not so much of a fan of country tlds 18:58:44 got a substitution set up with the irc client: 18:58:45 http://www.forthfreak.de 18:58:52 it works :) 19:00:31 disadvantage is, i can't easily type ff 19:20:44 hit me just now... 19:20:50 what did? 19:20:52 was typing abbrev for flopflop 19:20:58 huh? 19:21:02 what client are you using? 19:21:06 http://www.forthfreak.de 19:21:08 shit ... 19:21:12 http://www.forthfreak.de 19:21:15 argh 19:22:10 ff 19:22:14 ah 19:22:17 xchat 20:24:38 questions about bashforth can be sent to bashforth@forthfreak.de 21:22:51 --- quit: Herkamire ("leaving") 21:44:37 --- join: Herkamire (~jason@ip68-9-68-123.ri.ri.cox.net) joined #forth 21:44:58 * Herkamire feels much better after doing a little work on his forth 22:03:26 --- quit: Speuler (Read error: 60 (Operation timed out)) 22:12:54 --- join: Speuler (~Speuler@mnch-d9ba4168.pool.mediaWays.net) joined #forth 22:51:11 Speuler: how do you get part of a string in bash again? 23:12:55 never mind. I found it. 23:13:23 Speuler: (in the forth log) so I still thank you for your discovery :) I'm not using it in a script. 23:13:32 --- quit: Herkamire ("zzz") 23:35:34 --- quit: Speuler (Read error: 110 (Connection timed out)) 23:54:52 --- quit: gilbertdeb (""Monk has left the building"") 23:59:59 --- log: ended forth/03.03.12