00:00:00 --- log: started retro/06.08.29 01:39:29 How do you use syscalls to use a shared library? 01:40:27 I've only used Windows kernel32 calls so far from rf. 03:10:14 --- join: virl (n=virl@chello062178085149.1.12.vie.surfer.at) joined #retro 06:12:33 --- join: timlarson_ (n=timlarso@65.116.199.19) joined #retro 06:16:05 --- join: Ray_work (n=Raystm2@199.227.227.26) joined #retro 09:53:11 --- nick: Raystm2 -> nanstm 11:11:10 --- join: Quartus_ (n=Quartus_@209.167.5.1) joined #retro 11:12:09 hey Quartus! good morning. 11:16:23 Hey Ray_work 11:16:42 Do you happen to know how do I bind libraries to retro? 11:20:56 in Windows, for example: from kernel32.dll import 11:21:48 what about linux? Quartus_: I knew that. ;) 11:22:44 is it perhaps the same? from libsomething.a instead? 11:23:02 or .so, rather 11:25:57 Cheery, then you need the generic version. 11:26:25 because the linux version doesn't have that really implemented yet. 11:26:43 oow. 11:26:47 Oh well. :) 11:27:14 it should be possible through the syscalls, but that's a kind of magic which I don't know. 11:27:56 I assume it works through the load lib syscall and then jumping to the adresses which are defined in the lib file.. 11:29:00 and so you would need an elf reader.. I think that's the reason why crc didn't do that, it's a hard work. 11:29:32 --- join: snoopy_1711 (i=snoopy_1@dslb-088-068-001-157.pools.arcor-ip.net) joined #retro 11:30:21 --- quit: Snoopy42 (Nick collision from services.) 11:30:57 --- nick: snoopy_1711 -> Snoopy42 11:39:25 aah, elf reader... 11:39:50 * Cheery just drooled the floor. 11:40:19 I feel I love doing such things... sometimes. :) 11:41:11 But I wonder, then crc would propably already put a golden statue for me or something. ;P 11:41:44 And that wouldn't be too remote, and I partially know how it is done already. 11:42:17 Only I'd need would be some structure reading functionality, which I shouldn't need to look from far away. 12:26:13 --- part: Cheery left #retro 14:04:36 --- quit: timlarson_ ("Leaving") 14:35:12 --- quit: crc ("http://retroforth.org") 14:37:18 --- join: crc (n=crc@pdpc/supporter/active/crc) joined #retro 14:37:18 --- mode: ChanServ set +o crc 14:37:38 hey crc :) 14:37:43 hi Ray 14:37:44 long time. :) 14:37:49 been busy? 14:37:52 yes 14:38:07 thougth so. how's eveybody's health over there? 14:38:09 We had a record month at work ($705k) 14:38:12 not too bad 14:38:28 I've been getting headaches more often than normal, but otherwise good 14:38:31 yikes excellent. how many employees? 14:38:46 20 14:38:48 oooh headaches, sorry to read that. 14:38:57 man that's excellent. 14:39:14 How much does your outfit do annually. 14:39:35 About 5.8mil last year 14:39:43 we expect to get 6.5-7 mil this year 14:39:55 I see. All parts or do you do installs as well? 14:40:18 We just handle parts and materials 14:40:24 electrical, right? 14:40:45 man that's phenominal. What's your edge, imports? 14:42:08 we mostly do smaller things, primarily in the residential market 14:42:32 we're starting to get more involved in the commercial market, which is much bigger 14:42:41 sure. 14:43:32 We only do 2mil a year each year these last 12 years. we've got more customers but the price keeps dropping. 14:45:32 the trip was excellent. We drove thru Philly on the way up and Scranton on the way back. Philly at night and Scranton by day. 14:51:57 cool 15:05:38 --- quit: Ray_work ("User pushed the X - because it's Xtra, baby") 15:27:56 --- quit: virl (Remote closed the connection) 16:23:12 --- quit: Quartus_ ("jmIrc destroyed by the OS") 19:35:59 --- join: nighty (n=nighty@CPE00119576a9c5-CM0012c90d36fc.cpe.net.cable.rogers.com) joined #retro 20:38:41 --- quit: nighty (Read error: 104 (Connection reset by peer)) 21:07:12 CRC! You have a word named eb you madman, that's a hex value. I just spent ten minutes figuring that out. 21:40:58 --- join: Cheery (n=Cheery@a81-197-19-23.elisa-laajakaista.fi) joined #retro 22:26:00 crc, somebody: how do I find out the position of my terminal caret and the size of my terminal window? 22:26:56 Nothing built into rf for that as far as I know. I build Windows stuff for it a couple of days ago. 22:28:48 Hmm 22:28:52 I think I found something... :) 22:28:54 ioctl 22:29:27 Not an rf word? 22:31:40 nop. 22:31:45 it's a syscall. 22:31:57 num. 54 22:39:28 Cheery: if you get it working with ioctl, please send me the word(s) 22:39:40 * crc has some console words for Linux/BSD at http://retroforth.net/paste/?id=164 22:40:34 Quartus: words that are also hex values (that I recall offhand) are: b0 e d f eb 22:42:06 mutter mutter 22:42:40 such astonishingly evil words should be in their own wordlist 22:42:57 Along with any words named 0 1 2 3 4 5 6 7 8 or 9 :) 22:43:42 crc: I wonder, didn't found the correct code for that ioctl 22:45:28 Quartus: you could just relink the dictionary to hide them (see http://retroforth.net/paste/?id=165 for 'hide') 22:45:43 aaw, now I hit the spot. :) 22:46:08 crc, I may need to do that. 22:46:21 "terminal resize asm" 22:50:43 I think I found how I can resize&get the size of terminal. 22:52:48 TIOCGWINSZ can be used to get the terminal width and height. 23:03:42 now yet, I should get the caret (or whatever is it called?) 23:03:56 cursor 23:17:32 Update available: http://quartus.net/retro/retro-ans.fs 23:18:24 now has 'needs' and 'hide'. I can't recall if I posted an update for the backtrace stuff, but that's in there too. 23:32:42 It's working. :) 23:33:12 I tested on the interpreter, yet I code it, then it is ready. 23:33:54 crc: what would you say if you'd start using steel bank common lisp stylish way to handle spaces and newlines? 23:34:12 instead of that C-stylish. 23:34:42 in SBCL they put newline and space when they need to. 23:35:24 instead you everytime put cr or space after a line. 23:48:56 I don't understand what you're saying there, ChanServ. 23:48:58 oops. Cheery. 23:50:16 http://retroforth.net/paste/?id=166 23:50:52 Quartus: I mean putting pretty printer to handle spaces and returns. :) 23:51:05 Which I'm just about to do. 23:51:15 Handle them how? 23:51:34 to put them when necessary, not when finished. 23:52:04 You mean when entering code at the rf console? 23:52:11 This is a feature, not a limitation, you could override the pretty printer any time. 23:52:27 Quartus: nop, when the program prints. 23:52:37 I'm puzzled. I'm sorry, I don't track what you're saying. Surely the program prints the way you typed it in. 23:53:08 well, maybe the example is the best. 23:53:28 wait a little. :) 23:56:56 http://retroforth.net/paste/?id=167 23:57:44 I know pretty printing would be more useful if it was done by the OS, but it would need definetly more work. 23:57:58 Oh, I see. That should be pretty simple, even if you hard-coded it to 80 cols it'd be pretty generally useful. 23:58:02 and a new interface for overriding the pretty printer. 23:58:34 Yes, I have something like that built into Quartus Forth. Word-wrap is automatic. 23:58:56 Quartus: I don't need to. :) But I'd need the cursor position to handle newlines. 23:59:10 CR produces a newline. 23:59:43 If you want a linux-only facility, you could build it around the snippet you just wrote. I'm sure crc would prefer a more general solution given that rf builds for multiple targets. 23:59:46 yeh, I could calculate the cursor position by hand, but if I'd get it by ioctl or sth. would be much simpler. 23:59:59 --- log: ended retro/06.08.29