00:00:09 -!- beach [n=user@58.186.166.9] has quit [Remote closed the connection] 00:00:15 hefner: Wouldn't help if it's obtained from splitting a string in the first place. 00:00:17 another way is rethinking what you are trying to do 00:00:21 ISTR we had that one yesterday. 00:00:23 -!- BrianRice [n=briantri@c-98-225-51-246.hsd1.wa.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 00:01:08 BrianRice [n=briantri@c-98-225-51-246.hsd1.wa.comcast.net] has joined #lisp 00:01:44 the alternative is to have somekind of case statement 00:02:05 Nonsense. 00:02:29 You're doing an assignment, and a string is a perfectly reasonable :equal hash table key. 00:02:54 CHJA [n=christia@90-231-255-161-no145.tbcn.telia.com] has joined #lisp 00:02:59 Or you could do something with INTERN or FIND-SYMBOL... 00:03:43 which probably will use hash-table under the hood too 00:03:51 Or some other lookup, such as an ASSOC, FIND, GETF, etc. 00:04:16 Or we could get sillier and involve the filesystem. 00:04:21 cads [n=max@c-71-56-62-166.hsd1.ga.comcast.net] has joined #lisp 00:05:14 -!- matley [n=matley@83.224.196.8] has quit [Read error: 110 (Connection timed out)] 00:05:22 But one of the more basic questions is, "what are you trying to acomplish here?" 00:06:29 being able to change the values of symbols from an irc line, that's been split up by spaces, without having to write something for each symbol 00:06:49 Okay, -why- do you want that? 00:06:59 to set his nick :) 00:07:04 *nyef* blinks. 00:07:06 nick of the bot 00:07:12 Ahh, you're building a bot. 00:07:54 mornin' :) 00:08:02 Hello schme. 00:08:29 Hey nyef. 00:08:45 Cronos: How general do you wish this facility to be? Any symbol, or just some symbols? 00:08:51 any 00:09:00 Cronos: Or do you want the ability to run arbitrary code? 00:10:00 have tried to get something like (when (symbolp (nth 5 explode)) before setting it 00:10:16 Never evaluates to T, does it? 00:10:29 not that i've found 00:10:50 This is a bit dangerous, but what about (read-from-string (nth 5 explode))? 00:11:22 (Those of you who immediately thought "#.(die)" may now pipe up.) 00:11:27 -!- jlouis [n=jlouis@port1394.ds1-vby.adsl.cybercity.dk] has quit [Remote closed the connection] 00:12:12 (let (*read-eval*) ...) 00:12:21 And what's with this (nth 5 explode) stuff? Establish a binding, either via LET or DESTRUCTURING-BIND. 00:12:30 stassats: Yeah, but that's on the advanced side. 00:14:26 can't find anyway on *read-eval* 00:14:32 Besides, what about "nonexistent-package:who-cares"? 00:15:10 Or "common-lisp::nonexistent-symbo", for those implementations with package locks... 00:15:12 -!- retupmoca [n=retupmoc@adsl-67-36-58-203.dsl.klmzmi.ameritech.net] has quit [Read error: 104 (Connection reset by peer)] 00:15:32 or polluting packages with symbols 00:16:46 retupmoca [n=retupmoc@adsl-67-36-58-203.dsl.klmzmi.ameritech.net] has joined #lisp 00:17:07 Is it just me or is clim-player not quite working? 00:17:26 it's probably not you 00:17:44 arrgh. 00:18:08 *schme* moves it further up the TODO list. 00:18:11 you're saying there are some issues on current mcclim, or that it is otherwise broken? 00:18:56 I'm not sure. clim-player fires up just fine, but when I browse around my file hierarchy I descend ~/audio/music library/ 00:19:06 and bam. that comes up as empty. with just .. 00:19:18 Maybe it's the space in the path? 00:19:30 maybe it's pathnames! 00:19:44 (grr.. pathnames) 00:20:03 no that can't be .. pathnames never cause trouble. 00:20:11 That's troublesome if it does not handle spaces though. :) I'll try to dumb some mp3 in a non space place. 00:20:24 -!- tcr [n=tcr@host145.natpool.mwn.de] has quit [Remote closed the connection] 00:20:32 -!- Jabberwockey [n=jens@dslb-082-083-111-117.pools.arcor-ip.net] has quit [Read error: 110 (Connection timed out)] 00:21:06 I'm betting on pathnames. I'd trace cl:directory. 00:21:22 not getting a wild pathname, perhaps 00:21:40 hoh. now it won't even start . 00:22:18 wedgeV [n=wedge@cm56-238-229.liwest.at] has joined #lisp 00:23:12 Weird. I think I broke something. 00:23:56 hmm. when I run it, I see directories, but no files in them. 00:24:09 but they are in fact being returned by the directory call. 00:25:23 pathnames 1, hefner 0 00:26:17 yes yes.. I get things like that too. 00:27:47 Maybe the presentation for files is messed? 00:28:11 Hrrrm.. See I am in need of a music manager, seeing how the only good one around is crazy with braindeadness, so I'm thinking I should build it around climplayer, but I dunno now if it is broken or not :) 00:28:20 the only way to win is to not play the game. 00:28:53 ianmcorvidae|alt [n=ianmcorv@fsf/member/ianmcorvidae] has joined #lisp 00:29:28 schme: I like Quod Libet (but I like ye olde XMMS better). 00:29:30 It's no game, hefner, I need music playing here :) 00:29:47 *schme* looks it up. 00:29:49 Meh. That's why we have mpg123. 00:30:19 See what I really need is something that manages my music collection just nice, and lets me sort in good ways. And as it seems only one lets me sort by Composer. 00:30:26 So it wins, but it's crazy in other ways. 00:30:28 aaargh 00:30:43 -!- bdowning [n=bdowning@c-98-212-138-194.hsd1.il.comcast.net] has quit ["Disconnecting from stoned server."] 00:30:47 dash__ [n=dash@dslb-084-057-004-244.pools.arcor-ip.net] has joined #lisp 00:30:58 bdowning [n=bdowning@c-98-212-138-194.hsd1.il.comcast.net] has joined #lisp 00:31:00 So I figured hacking up climplayer, and also adding some libmtp support for handling mp3 players with it.. well it'll just be sweet. 00:31:56 this is an important step on your road to disillusionment 00:32:26 -!- Cronos [n=a@5ad5b20b.bb.sky.com] has quit [] 00:32:36 -!- fooquux [n=fooquux@udp265832uds.hawaiiantel.net] has quit [Read error: 110 (Connection timed out)] 00:32:38 hahah 00:33:42 I say it will just be schweet. 00:33:46 fooquux [n=fooquux@udp265832uds.hawaiiantel.net] has joined #lisp 00:34:03 -!- _magnus_ [n=magnus@83.209.81.249] has quit ["Leaving"] 00:34:09 fooquux: How's the database stuff coming? 00:37:50 -!- syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has quit ["Leaving..."] 00:38:10 -!- drwhen [n=d@209-112-181-104.static.acsalaska.net] has quit ["\(^^) LET THE GOOD TIMES ROLL! "\(^^)""] 00:38:23 Atomsk [i=ace4016@cpe-76-168-248-118.socal.res.rr.com] has joined #lisp 00:38:45 -!- Atomsk is now known as _ace4016_ 00:38:50 schme: there seems to be some interesting stuff in climplayer such that it might be preferable to fix it 00:39:00 -!- ace4016 [i=ace4016@cpe-76-168-248-118.socal.res.rr.com] has quit [Nick collision from services.] 00:39:08 -!- _ace4016_ is now known as ace4016 00:39:15 hefner: Ah cools :) 00:39:53 (I wouldn't, but do as I say, not as I do..) 00:40:04 Hahahaha. 00:40:36 What cool things though? 00:40:55 I don't know. I couldn't follow the code without some effort, so I assume it must be up to something really interesting. ;) 00:40:59 -!- blitz__ [n=julian@77.64.176.217] has quit ["leaving"] 00:41:00 *schme* looks for more painkillers. 00:41:04 Ah good thinking. 00:41:20 blitz_ [n=julian@77.64.176.217] has joined #lisp 00:41:31 -!- dash_ [n=dash@dslb-084-057-004-095.pools.arcor-ip.net] has quit [Read error: 110 (Connection timed out)] 00:41:39 -!- hfoo [n=h@p5B17E075.dip.t-dialin.net] has quit ["Leaving"] 00:41:51 Well if the browsing thing gets fixed it should be ok. Then add some sql magic to it to keep track of music collection, and tag browsing. and voila. 00:42:16 Caffeine and I will actually get poking with this :) 00:42:38 I've long wanted to write a music player (or a frontend to control a music player), but as I mentioned the other day, I still haven't figured out the right interface. 00:42:51 Ah yes. 00:43:01 schme: Remind me one of these days to sit down and write that SQL database engine I've been pondering for the past year and a half. 00:43:17 I like the amarok2 interface actually. 00:43:22 nyef: something like clsql? 00:43:42 z0d: took the words out of my mouth there :) 00:43:57 Umm... No. Something like sqlite, maybe. 00:44:03 -!- Phoodus [i=foo@ip68-231-38-131.ph.ph.cox.net] has quit [Read error: 104 (Connection reset by peer)] 00:44:12 I assume they're just trying to show off, but the first few screenshots on the amarok site melt my brain 00:44:12 Or... SQL Anywhere, actually. 00:45:09 nyef: I don't know SQL Anywere. would it be an interface to only SQLite? 00:45:14 nyef: Oh. You mean like not the interaction layer, but an actual sql engine :) 00:45:37 -!- nxt [n=nxt@77.207.25.109] has quit [Read error: 113 (No route to host)] 00:45:42 -!- ianmcorvidae [n=ianmcorv@fsf/member/ianmcorvidae] has quit [Read error: 110 (Connection timed out)] 00:45:46 schme: Right. Query optimizer, storage engine, the rest. 00:46:17 hefner: I installed it just some day ago. I really like it. 'cept certain braindead bits about it. Like how in amarok1.4 you could go into preferences and tell it where to send the output. No more! Now you need to go look up KDE systemsettings. 00:46:27 sounds cool 00:46:32 and it lets me sort by composer, so I am happy. 00:46:48 doh, /me finally built gauche scheme for an arm device 00:47:02 nyef: Sounds way cool :) 00:47:24 nyef: btw, what about continuing porting sbcl to arm? 00:47:48 stassats: I know what I need to do next, I just have a pile of other stuff I'm doing right now. 00:47:59 Feel free to give me an incentive, though. :-P 00:48:23 nyef: how quickly will you be working on that? 00:48:31 mogunus: On what? 00:48:47 At my current rate of digesting SBCL documentation, I estimate that I would be helpful for an arm port in about 1.5 months. 00:48:47 Two projects mentioned in the past... five minutes or so. 00:48:51 clsql is a bit ugly to work with. and it has all the [] and shit. What would be really neat is a sql all in lisp so one could get away from that. 00:49:30 *schme* passes out the incentive. 00:50:03 nyef: "that" being an arm port 00:50:25 schme: see postmodern and cl-rdbms/cl-perec for alternatives 00:50:32 luis: Ya I know :) 00:50:44 luis: postmodern only works with the postgresql though. 00:51:02 It'd be neat with some small embedded thing like sqlite, for sure. 00:51:02 -!- ThomasI [n=thomas@unaffiliated/thomasi] has quit ["Bye Bye!"] 00:51:05 without the sqlite 00:52:47 -!- Yuuhi [i=benni@p5483CA4A.dip.t-dialin.net] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 00:53:02 though not SQL, I like BerkeleyDB too 00:53:05 jollygood [n=jollygoo@pool-72-65-153-129.chrlwv.east.verizon.net] has joined #lisp 00:53:26 schme: If you like, I have some 579 lines of hacking-and-documentation for accessing some version of an "adaptive server anywhere" (formerly "watcom sql") database file. 00:53:43 Hrm. September 2007. Not quite a year and a half. 00:54:07 Aaah.. some other year :) 00:54:18 Plus I think this format might be implemented in some other open-source software as well... 00:54:18 vacation is over, I have no time for stuff :( 00:54:25 Aww. 00:54:26 jcowan [n=jcowan@cpe-74-68-154-18.nyc.res.rr.com] has joined #lisp 00:55:02 Licensing question: does the CL community strongly favor or disfavor any particular permissive license? 00:55:19 We disfavor GPL and its derivatives. 00:55:20 There is no CL community ;) 00:55:26 That too. 00:55:38 -!- njsg [n=njsg@unaffiliated/njsg] has quit ["gone!!"] 00:55:52 In that case there should be no objection to the Apache 2.0 license (which is incompatible with GPL 2.0)? 00:56:14 I wouldn't use it 00:56:22 jcowan: If you're going to do something like that, you might consider your user scenarios. 00:56:25 -!- rdd [n=user@c83-250-142-219.bredband.comhem.se] has quit [Read error: 104 (Connection reset by peer)] 00:56:32 I mean a library incompatible with GPL 00:57:01 Hmm. It would still be possible to execute it on Clisp, though 00:57:09 schme: postmoderns sexp syntax need not be used only with postgres. 00:57:18 jcowan: I like the MIT license 00:57:37 jcowan: But is it legal to execute it in clisp? 00:57:49 (More accurately, the Apache Foundation claims that Apache 2.0 *is* compatible with GPL 2.0; the FSF denies it. All agree that Apache 2.0 is compatible with GPL 3.0.) 00:57:58 jcowan: the Apache license is fine by me. 00:58:12 schme: Yes. To clisp, Lisp code is just data 00:58:28 i don't ship any GPL or LGPL lisp code ever. 00:58:36 LLGPL is ok-ish 00:58:48 jcowan: I'm not convinced the issue is quite so crystal clear :) 00:58:50 i prefer MIT/BSD-sans-advertising 00:58:51 rdd [n=user@c83-250-142-219.bredband.comhem.se] has joined #lisp 00:58:51 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 00:59:13 *schme* shrugs. 00:59:26 Now I need to go coffee hunting. 01:00:14 It's part of my Blesity project (the ISLisp in CL emulator), and I'd like to use a single license throughout. 01:00:33 that is, the part I'm asking about is the emulator 01:00:41 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 01:00:46 which can be used independently of the rest 01:00:49 MIT 01:01:02 -!- davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 01:01:03 if you don't want to use GPL or other copyleft, why not a GPL-compatible non-copyleft? 01:01:06 such as MIT 01:06:10 -!- Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has quit [] 01:08:51 -!- alley_cat [i=AlleyCat@sourcemage/elder/alleycat] has quit [Remote closed the connection] 01:09:05 -!- chris2 [n=chris@p5B16B78C.dip0.t-ipconnect.de] has quit ["Leaving"] 01:11:57 thom_logn [n=thom@pool-96-229-99-100.lsanca.dsl-w.verizon.net] has joined #lisp 01:12:44 l_n [n=shawn@tuxhacker/lordnothing] has joined #lisp 01:14:14 QUICK! blog. cl. does not suck. 01:14:17 recommendations? 01:14:33 Not bothering? 01:14:51 rottcodd [n=user@ppp59-167-52-157.lns1.cbr1.internode.on.net] has joined #lisp 01:16:03 slyrus_ [n=slyrus@207.189.195.44] has joined #lisp 01:16:24 not an option 01:16:35 it's an ad agency. they _need_ a blog. 01:17:29 i could half ass something in php, but no two projects use the same php. anybody else notice that? 01:17:58 -!- _8david` [n=user@port-83-236-3-235.dynamic.qsc.de] has quit [Read error: 104 (Connection reset by peer)] 01:18:01 _8david` [n=user@port-83-236-3-235.dynamic.qsc.de] has joined #lisp 01:18:04 drupal will rape my time, for sure. themeing it to fit in with the rest of the site is gonna be pain .. 01:18:14 Hell, I've noticed one company with four websites, all of which ran X-Cart, and no two of which ran the same version of -anything-. 01:18:50 every semi-major php project uses a half-assed version of smarty or lookalike 01:19:42 fusss: The really clever ones try to cover that up by using both, but for some reason they end up with two -left- halves? 01:19:44 nyef: i have been stripping down ZenCart, we might have a working CL shopping cart if i stop dicking around with crappy web gigs 01:19:56 hhaha 01:20:26 -!- jcowan [n=jcowan@cpe-74-68-154-18.nyc.res.rr.com] has left #lisp 01:20:40 syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has joined #lisp 01:21:41 "Paste too large." boooooo 01:22:23 I 'liked' it when the company upgraded PHP from x.y.z to x.y.(z+1) and web sites broke. it turned out that the PHP guys changed to argument list of the sorting function 01:22:25 no clsql data-model of the ZenCart database for you, i guess :-P 01:23:00 z0d: On a -point release-? 01:23:27 Something like that is worth a minor version bump, at least. 01:23:45 Yes. Though the sysadmins/programmers should have read the changes list 01:24:03 Should have tried it on a staging system first. 01:24:05 if only they'd standardized their language in 1994 01:24:23 hefner: if only they kept it to themselves in 1994 01:25:05 if only they kept it to themselves forever... 01:26:01 people in my country say "I should've slept that night" to their children when scolding them :-P 01:27:05 frosty [n=Frostysn@c-76-115-11-91.hsd1.wa.comcast.net] has joined #lisp 01:27:19 Yeah, there are numerous variations on that theme 01:29:27 i know someone who should've slept the night he conceived Araneida (*tsk* *tsk*) 01:30:08 I don't know about the araneida internals, but I liked to use it 01:31:51 *fusss* thinks GNU screen _really_ needs an option to set mnemonics for open screens 01:31:52 I -do- know about the araneida internals. And the cliki internals. Yeek! 01:32:19 screen -r 319281 sucks. screen -r mysite much better. 01:34:32 -!- jsoft [n=Administ@unaffiliated/jsoft] has quit [Read error: 110 (Connection timed out)] 01:38:15 martyrdom: http://cyrusharmon.org/blog/display?id=96 01:41:21 fusss: http://git.cyrusharmon.org/cgi-bin/gitweb.cgi?p=nuclblog.git do you know it? 01:43:09 do you guys know if there exists a C extension that provides a lisp-like macro facility for generating c code and abstracting away a lot of the routine repetitive syntax that you see in C loops and data structure manipulation/ 01:43:10 ? 01:43:10 araneida .. yowza. 01:43:14 z0d: that url seems familiar. hmmmm 01:43:18 sellout [n=greg@c-24-128-50-176.hsd1.ma.comcast.net] has joined #lisp 01:43:57 cads: why not write a syntax for C ala parenscript, and do it in lisp. 01:43:59 fusss: screen -R mysite when you first start a new session what you are looking for? 01:44:01 cads: Yes. It's called "The C Preprocessor". It's not quite lisp-like, but it can be workable... 01:44:02 drewc: got a CL based blog thinggie laying around somewhere? 01:44:32 -!- antoni [n=antoni@57.pool85-53-7.dynamic.orange.es] has quit [Read error: 148 (No route to host)] 01:44:46 fusss: no ... but i'll help you build one in UCW + rucksack if you like :D 01:44:56 *drewc* wants such a beast for himself 01:45:00 nyef, is the C preprocessor expressive enough to write, say, a map function over an array? 01:45:03 bryte: woah! i see your manpages don't emit cryptonite like mine. 01:45:20 cads: I've never actually tried that. 01:45:51 drewc: just for a weird 11th hour subproject i just got. need it up by tomorrow noon anyway :-S 01:46:10 Is there some jack library floating around somewhere? 01:46:18 fusss: in that case .. we did run cl-blog at one point with some success. 01:46:38 -!- jrockway [n=jrockway@stonepath.jrock.us] has quit [Read error: 104 (Connection reset by peer)] 01:46:46 -!- wedgeV [n=wedge@cm56-238-229.liwest.at] has quit [] 01:47:20 nyef, I think C could be a little more friendly to me if it didn't give me the option of shooting myself in the foot all the time :D 01:47:45 string manipulation in C is not sane 01:47:57 string manipulation is not sane 01:47:59 cads: there are some safer C dialects 01:48:36 I bet I could write a few emacs macros that'd take care of most of my headaches 01:48:59 -!- schoppen1auer [n=css@unaffiliated/schoppenhauer] has quit [Remote closed the connection] 01:49:12 drewc: doesn't cl-blog depend on araneida? 01:49:53 cads: maybe http://sourceforge.net/projects/thinlisp/ is what you're looking for 01:50:03 yes, but that's easy to remedy. It took me about 5 mins to do hunchencliki 01:50:29 z0d, I've been looking for either an extension provided by a compiler, or some metaprogramming environment, that'd basically help me with some of the tedious and error prone chores of filling in the blanks between all the braces in a c program 01:50:56 fusss: i still had to include araneida for the html generation etc .. but iirc cl-blog is a little cleaner than cliki 01:51:17 fusss: there are a few others around as well ... nuclog i think is more modern. 01:51:45 I'm just trying nunclog 01:51:56 drewc: will hunt later. thanks all. 01:52:51 -!- gz [n=gz@209.6.18.72] has left #lisp 01:53:27 gz [n=gz@209.6.18.72] has joined #lisp 01:58:21 -!- l_n [n=shawn@tuxhacker/lordnothing] has quit [Read error: 60 (Operation timed out)] 02:03:23 schme: if you're looking for sound output, libao looks very easy to write a binding to 02:04:51 TDT [n=TDT@143.108.177.207.dyn.southslope.net] has joined #lisp 02:05:06 And doing a set of OSS bindings should be easy as well, if anybody still wants to use OSS. 02:05:18 l_n [n=shawn@tuxhacker/lordnothing] has joined #lisp 02:05:18 *nyef* hasn't looked at ALSA yet. 02:05:52 nyef: you did well, then 02:06:19 *hefner* fears he will end up doing his first ALSA programming very soon 02:06:33 -!- jollygood [n=jollygoo@pool-72-65-153-129.chrlwv.east.verizon.net] has quit [] 02:07:13 hefner: I'll look into it, but I don't think it is what I want :) 02:07:41 What I want is to have thou output going to jackd ;) 02:07:42 maybe not, but these audio servers are the work of the devil 02:07:49 Ok. 02:07:54 jackd seems great 02:07:58 what's it do? 02:08:21 Though google gave me some thread saying that there were issues with sbcl + jack. 02:08:25 *nyef* has heard horror stories about jackd, but hasn't seriously investigated it. 02:08:33 Hmm.. well basically it's a patchbay. 02:08:48 and it gives very good latency. 02:09:04 for patching output from one thing to the input of another, and so forth. 02:09:16 hmm, okay. sounds interesting. I thought it was another pulseaudio/esd-style abomination. 02:09:49 The only real issue I have at the moment is that it does not play nice with the dmix in ALSA. It just locks up the whole hw:0 interface, so I can't really watch videos on youtube. 02:09:52 (no great loss) 02:09:58 Hmm.. no idea about pulseaudio/esd :) 02:10:23 alsa developers should be burnt alive 02:10:26 jack is only really useful if one likes to send audio output from software to other software. Which I do. 02:11:58 actually, I was just wondering how to do that the other day (would've been handy to pull something into a waveform editor) 02:12:39 -!- kzar` [n=kzar@hardwick.demon.co.uk] has quit [Remote closed the connection] 02:12:50 Hmm.. pulling into a waveform editor I'm not sure. It seems to me you would actually have to save the audio off to a .wav somewhere. Tohough I guess waveform editors can record from live input. 02:14:55 The post I found mentions something about sbcl's GC not being designed for RT :) 02:15:47 use big buffers :) 02:16:53 -!- lambda is now known as davirus 02:17:16 I guess. 02:17:58 what post are you referring to? 02:18:01 I was hoping someone had the ready made bindings and a nice library. 02:18:03 hmmm. 02:18:13 -!- Spec[afk] [n=NoOne@82.177.125.6] has quit [] 02:18:16 http://ccrma-mail.stanford.edu/pipermail/cmdist/2008-October/005094.html 02:18:21 ok, I have an architectural question 02:18:49 (no idea if this is a real issue yet.) 02:18:58 sohum: Shoot :) 02:19:16 I like the CLOS's method combination system, and can see it being useful 02:19:24 loz [n=loz@163.c.003.ncl.iprimus.net.au] has joined #lisp 02:19:49 in particular, I can see :after methods and such being used to great effect to create an effective method in response to certain input 02:20:25 *nyef* can sense a "but" coming. 02:21:19 schme: I find that the effect of the non-realtime aspect of garbage collection in SBCL can be reduced quite a bit by not consing. 02:21:29 nyef: Probably. 02:21:29 the problem here is that would essentially, in the system I'm thinking of, require me to define a large number of classes to hold the separate method-parts, and then /every/ /single/ possible subcombination of them, at least the DEFCLASSes 02:21:57 nyef: This is just what came at me when I googled "jack +common lisp". I guess I'll see what happens. 02:22:47 which is obviously not feasible, because combinatorial explosion ahoy! Is there a better way to do something like this, or am I just missing something obvious? 02:22:54 sohum: Either you're doing something wrong, conceptually, with the method combination, or your problem just isn't a good fit for a CLOS solution. 02:23:16 sohum: I think dynamic-class was made exactly for that use case 02:23:22 -!- davirus [n=lambda@c-24-5-198-113.hsd1.ca.comcast.net] has quit [Client Quit] 02:23:57 minion: tell sohum about dynamic-classes 02:23:58 sohum: direct your attention towards dynamic-classes: Another instance of the ubiquitous pages ASDF-Install uses to keep the universe glued together. http://www.cliki.net/dynamic-classes 02:24:48 -!- kzar [n=kzar@hardwick.demon.co.uk] has quit [Remote closed the connection] 02:25:31 sohum: although I smell that what you're looking for is a trait-based object system 02:26:14 minion, fe[nl]ix: ooh, dynamic classes look nifty. Need to read documentation now.. 02:26:15 please stop playing with me... i am not a toy 02:26:24 fe[nl]ix: trait-based? 02:26:34 -!- drdo [n=psykon@167.111.54.77.rev.vodafone.pt] has quit [Remote closed the connection] 02:26:42 owait, minion is a bot 02:26:43 ha 02:28:47 sohum: And a good one, too. Manages to pass the turing test for a while with some people. 02:29:27 nyef: ha, apparently. pretty awesome, is he OS? 02:29:41 It's a modified cliki-bot from cl-irc. 02:29:49 *stassats* remembers himself asking "is minion a bot?" 02:29:58 minion: Are you a bot? 02:29:58 i'm not a bot. i prefer the term ``electronically composed''. 02:30:06 heee 02:30:07 drdo [n=psykon@167.111.54.77.rev.vodafone.pt] has joined #lisp 02:30:38 minion: Just how composed are you? 02:30:39 why do you want to know? 02:30:50 minion: I am doing research. 02:30:50 Would you /please/ stop playing with me? 3 messages in 52 seconds is too many. 02:31:01 sohum: traits are a lot like mixins, but spare you from having to define all combinations of your base classes 02:32:08 sohum: you would just create an instance of class FOO with traits A and B at runtime 02:32:18 -!- crod [n=cmell@cad4e7-094.dynamic.tiki.ne.jp] has quit [Read error: 60 (Operation timed out)] 02:33:17 actually, it's essentially an ircbot I'm writing. I want each line to be scanned by each applicable module, and then construct an applicable method (probably composed of print statements and such) from any modules that say that they can do something with it 02:33:26 traits sounds really useful 02:33:49 -!- syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has quit ["Leaving..."] 02:34:42 oSand [n=heartles@222-154-255-19.adsl.xtra.co.nz] has joined #lisp 02:34:54 -!- oSand [n=heartles@222-154-255-19.adsl.xtra.co.nz] has left #lisp 02:35:14 lambda [n=lambda@c-24-5-198-113.hsd1.ca.comcast.net] has joined #lisp 02:36:38 -!- runenes [n=runenes@proxy-gw.uib.no] has quit [Read error: 110 (Connection timed out)] 02:37:02 runenes [n=runenes@proxy-gw.uib.no] has joined #lisp 02:39:18 heh, is everyone writing an IRC bot today? 02:39:58 I'm not. 02:40:20 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 02:40:47 me neither 02:41:11 In fact, I'm not even not working on one today. 02:41:46 O_4 [n=souchan@125-236-190-161.broadband-telecom.global-gateway.net.nz] has joined #lisp 02:42:46 *luis* blinks 02:43:23 (That means that I don't even have an IRC bot project to work on.) 02:43:28 I was trying to run an cl-irc-based irc bot today 02:43:42 but the problem I was having had to do with JVM versions, so it's not exactly relevant 02:43:43 Technically, I could probably hack on minion/specbot/lisppaste, but they're not -mine-. 02:44:12 trebor_h` [n=user@dslb-084-058-216-182.pools.arcor-ip.net] has joined #lisp 02:45:32 -!- netaust1n_ [n=austinsm@68-191-51-69.dhcp.nwtn.ct.charter.com] has quit [] 02:45:37 crod [n=cmell@cb8a69-009.dynamic.tiki.ne.jp] has joined #lisp 02:48:03 i just noticed how well LW has done at emulating some nuanced emacs details 02:49:15 even though it's an MDI application, with a classic win32 look and feel, still C-x-o takes you to the the "other" window. The figured if you know what you're doing, you're likely to split the client area between the code window and the listener. 02:49:32 -!- oudeis_ [n=oudeis@bzq-79-179-109-100.red.bezeqint.net] has quit ["This computer has gone to sleep"] 02:50:40 i have repeatedly typed code into LW by mistake not noticing it wasn't emacs, because they nailed the color scheme down :-) 02:53:35 I once built an X window manager sufficiently close in looks to windows classic theme that I got frustrated that some of the keyboard shortcuts hadn't been implemented yet. 02:55:08 haha 02:55:55 Ok. nuclblog has some rough edges 02:56:08 z0d: the good hack is underway :-) 02:56:29 z0d: This is Lisp. Almost everything has some rough edges. 02:57:02 nyef: For some reason yes. Though e.g. Edi's libraries are generally good. 02:57:13 fusss: are you Cyrus? 02:57:18 His distribution policy has some rough edges. 02:57:39 z0d: Harmon? That'd be slyrus. 02:57:43 z0d: no, but i'm writing a blog engine in lisp as we speak 02:58:35 -!- blitz_ [n=julian@77.64.176.217] has quit ["Lost terminal"] 02:59:36 nyef: So, why are there so much rough edges? Too few Lisp hackers? When I try out something it almost always request some hacking and/or not maintained anymore. Just look at the cl.net projects list 02:59:49 fusss pasted "30 minutes into it" at http://paste.lisp.org/display/73103 03:00:10 z0d: The cl.net projects list suffers from the sourceforge effect. 03:00:33 -!- trebor_home [n=user@dslb-084-058-245-119.pools.arcor-ip.net] has quit [Read error: 110 (Connection timed out)] 03:01:19 As for rough edges, I've been hacking Lisp since 2003(?)ish. In that time, SBCL has made a release -every month-. There are still rough edges in it, even on a major target platform like x86 linux. 03:01:53 (And, oh god, have you seen what it comes up with if anything goes wrong with CLOS-using code?) 03:02:17 z0d: get into hacking it, then they can be your own rough edges :) 03:03:23 S11001001: what a lovely scenario! 03:03:41 nyef: the sourceforge effect ? 03:04:14 ltbarcly [n=jvanwink@nc-76-0-128-24.dhcp.embarqhsd.net] has joined #lisp 03:04:20 fe[nl]ix: It's free project hosting, so there are a lot of people who sign up and then don't do anything, or start something and then vanish. 03:05:16 It's obvious on sourceforge, because they have so bloody many such projects, because they've been around for so long and have such a high profile. 03:05:50 b4|hraban [n=b4@0brg.xs4all.nl] has joined #lisp 03:06:18 -!- runenes [n=runenes@proxy-gw.uib.no] has quit [Read error: 110 (Connection timed out)] 03:06:34 runenes [n=runenes@proxy-gw.uib.no] has joined #lisp 03:06:43 nyef: it's somehow better than sourceforge 03:07:03 In absolute or relative terms? 03:07:09 relative 03:07:24 z0d: i can give you this blog when i finish it. do you mind sqlite3 for database? it's all clsql anyway, but i don't have time to test it with anything else. 03:07:41 c-l.net hasn't yet reached the point where 85+% of projects are abbandonware 03:07:49 ManateeLazyCat [n=user@222.212.133.86] has joined #lisp 03:08:01 fusss: Ok. I don't mind sqlite 03:08:15 catch me here if you're not around later 03:09:11 now you mention it... I should probably head for bed. it's 4:10 am 03:10:07 z0d: Sleep well. 03:10:42 Good night. happy hacking. 03:13:35 -!- iasc [n=\0@varenka.cime.net] has quit [Read error: 104 (Connection reset by peer)] 03:15:07 awwwww, bad ajax, bad bad bad! I wrote a crappy ajax clock last night which has been running since then. needless to say, my server logs got fat and it messed up my google analytics :-S 03:15:14 -!- blbrown [n=Berlin@c-71-236-25-127.hsd1.ga.comcast.net] has quit [Client Quit] 03:15:40 60 requests per minute * about 25 hours 03:15:59 Sooner or later, someone's going to abbreviate that to "googlytics". 03:16:39 if i served ads they would've yanked my accounts by now 03:17:23 younder [n=jpthing@062016236162.customer.alfanett.no] has joined #lisp 03:18:18 -!- CHJA [n=christia@90-231-255-161-no145.tbcn.telia.com] has quit [Connection timed out] 03:18:32 <_3b> serving ads on an ajax request seems a bit odd 03:19:34 rottcodd_ [n=user@ppp59-167-52-157.lns1.cbr1.internode.on.net] has joined #lisp 03:20:16 not so much the ajax stuff, just the heavy page "refresh" thing 03:20:37 -!- ManateeLazyCat [n=user@222.212.133.86] has quit [Remote closed the connection] 03:20:44 glad it's just a text fetch which gets stuffed into a div 03:21:26 <_3b> yeah, just wondering why google would even see that? 03:21:44 -!- younder [n=jpthing@062016236162.customer.alfanett.no] has left #lisp 03:21:45 younder [n=jpthing@062016236162.customer.alfanett.no] has joined #lisp 03:21:56 -!- younder [n=jpthing@062016236162.customer.alfanett.no] has left #lisp 03:22:15 syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has joined #lisp 03:22:17 younder [n=jpthing@062016236162.customer.alfanett.no] has joined #lisp 03:22:21 -!- younder [n=jpthing@062016236162.customer.alfanett.no] has left #lisp 03:22:23 younder [n=jpthing@062016236162.customer.alfanett.no] has joined #lisp 03:22:47 younder: Desist and refrain. 03:24:25 sorry if i misspoke. google doesn't see your server logs. it's my crappy awstats setup. 03:24:43 calling it "logs" doesn't sound very web 2.0, analytics otoh 03:24:55 <_3b> yeah, effect on local logs would be more annoying :) 03:25:49 (list-databases *db-spec* :database-type :sqlite3) <-- redundant :-) 03:29:01 -!- rottcodd_ [n=user@ppp59-167-52-157.lns1.cbr1.internode.on.net] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 03:29:05 do i have to run POSTed data through escape-string if they're to go a database, to be retrieved later for html rendering? 03:29:29 in case spammers wanna post links 03:29:51 <_3b> need to at least make sure they won't break the DB (which may or may not involve escaping) 03:30:17 sqlite3. what could possibly go wrong? 03:30:34 amnesiac [n=amnesiac@p3m/member/Amnesiac] has joined #lisp 03:30:52 <_3b> no idea, just make sure you aren't building SQL out of untrusted strings 03:31:37 -!- bryte [n=user@c-24-21-34-206.hsd1.wa.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 03:32:14 <_3b> then once you pull it out of the db, watch out for xss and such if sending it to browsers as html 03:33:49 bryte [n=user@c-24-21-34-206.hsd1.or.comcast.net] has joined #lisp 03:34:01 -!- LiamH [n=nobody@pool-72-75-101-81.washdc.east.verizon.net] has quit [Read error: 110 (Connection timed out)] 03:34:16 -!- bryte [n=user@c-24-21-34-206.hsd1.or.comcast.net] has left #lisp 03:34:32 bryte [n=user@c-24-21-34-206.hsd1.or.comcast.net] has joined #lisp 03:35:01 -!- rottcodd [n=user@ppp59-167-52-157.lns1.cbr1.internode.on.net] has quit [Read error: 110 (Connection timed out)] 03:37:23 hmmm 03:40:17 simplechat [n=simplech@unaffiliated/simplechat] has joined #lisp 03:42:34 -!- mejja [n=user@c-4bb5e555.023-82-73746f38.cust.bredbandsbolaget.se] has quit [Remote closed the connection] 03:46:38 -!- runenes [n=runenes@proxy-gw.uib.no] has quit [Read error: 110 (Connection timed out)] 03:47:03 runenes [n=runenes@proxy-gw.uib.no] has joined #lisp 03:49:05 -!- syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has quit ["Leaving..."] 03:51:03 Damn.. can't sleep 03:53:58 -!- dreish [n=dreish@minus.dreish.org] has quit [] 03:54:49 fusss_ [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 03:54:52 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 03:55:02 -!- fusss_ is now known as fusss 03:56:13 z0d: try consing sheep 03:57:30 Hehe. 03:59:03 I decided to try rucksack instead. 04:00:20 fusss: use prepared statements and sql injection is not a worry 04:00:45 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 04:01:16 then just make sure you use something like yaclml's AS-HTML to render to the browser and you avoid xss 04:01:33 z0d: rucksack is teh cool 04:02:15 drewc: I skimmed through the Elephant doc as well, but rucksack seems lighter and easier. 04:02:50 z0d: elephant is cool as well, but i don't like the reliance on external db engines 04:03:19 i've used elephant in a project, and it was fun to work with. 04:03:27 i can't wait for an excuse to use rucksack 04:05:38 clhs #@ 04:05:39 http://www.lispworks.com/reference/HyperSpec/Body/02_dh.htm 04:06:41 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 04:06:46 Ah, okay, no wonder it wasn't familiar. 04:06:56 -!- greyface [n=greyface@248-177.77-83.cust.bluewin.ch] has quit [] 04:08:16 even Perl doesn't have #@ 04:08:52 Yeah, #@ is undefined by the standard, but is available for implementation extensions. 04:09:02 Or user extensions, for that matter. 04:09:44 slyrus_: ping 04:09:53 slyrus: ping2 04:10:04 Silly C people and their varargs. 04:12:22 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 04:12:29 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 04:12:35 -!- amnesiac [n=amnesiac@p3m/member/Amnesiac] has quit ["Leaving"] 04:15:39 do you guys know of an article which considers what the world might be like had lisp been used to design a commercially successful operating system, rather than C? 04:16:08 ... I didn't know that lisp was used to design C. 04:16:27 cads: you mean some kind of sci-fi? 04:16:28 -!- runenes [n=runenes@proxy-gw.uib.no] has quit [Read error: 110 (Connection timed out)] 04:16:41 because it's hard to predict 04:16:50 runenes [n=runenes@proxy-gw.uib.no] has joined #lisp 04:17:27 what the hell? Symbolics carved a nice niche for itself making Lisp OSes 04:17:43 cads: the entire history of cll can be taken as a "what if LispOS" tearfest. 04:17:53 fusss: I think the argument is "they're dead now, so it wasn't successful". 04:18:55 too bad the export/import API of rucksack is not yet implemented 04:18:55 fusss, really I'd like to find a usable and comfortable non-unix non-c operating system in which to do my every day desktop tasks as well as neat computing experiments 04:19:35 z0d: it's not? oh 04:19:51 cads: Have you considered System 6.0.7 with MultiFinder? 04:19:54 cads: have you seen emacs? :-P 04:20:38 fusss: that's just for dumping the objects into s-exp format, so not really mission critical. but would be good for backing up the data 04:21:09 I am currently of the belief that the state of the lisp world is such that it is unworthy of building a LispOS, and that a new LispOS would be a unix-type system anyway. 04:21:23 cads: BeOS was C++ and not very unixy. Amiga was Amiga. RiscOS was pascal. 04:21:43 -!- dkcl [n=danderse@unaffiliated/dkcl] has quit [Remote closed the connection] 04:21:44 BeOS wasn't very nice, programmatically, either. 04:22:01 nyef: the future LispOS has dynamically generated flash GUIs fed to a browser with a non-broken tag 04:22:09 The few times I tried to do anything with it were unpleasant. 04:22:37 there are some ongoing(?) Lisp OS projects 04:22:55 nyef: back in 1998, it was better than sex (and i would've really appreciated sex back then, at 17 :-P) 04:23:02 z0d: Check their activity rates over the past year, see if they're really ongoing. 04:23:13 Lice and movitz 04:23:32 LICE? nice! 04:23:44 http://www.emmett.ca/~sabetts/ 04:23:46 *fusss* goes back to scratch an itch 04:24:07 http://common-lisp.net/project/movitz/ 04:24:32 z0d: Last LiCE news item was from march '07. 04:24:53 I see 04:25:26 *cads* sadly continues to use ubuntu 04:26:16 I think there's not enough energy (time, money) to build a decent Lisp OS. 04:26:38 Last Movitz CVS mailing list traffic was last August. 04:27:01 On -devel there are 2 messages from December 04:27:16 z0d, are lisp compilers refined enough that you'd want to write kernel code in lisp? 04:27:18 I think that the classic LispOS model is in direct conflict with the modern notion of a decent OS. 04:28:17 Yes, I can't imagine a casual user using a Lisp OS <-: 04:28:19 cads: your OS will not have a running compiler, that would be back imo 04:28:28 Even for just lisp work, I want separate address spaces. 04:28:43 you use the lisp compiler to generate binaries, which all share a common runtime 04:29:05 nyef: that seems less fun, somehow. 04:29:10 (jack-get-sample-rate) => 48000 I'll celebrate with some coffee. 04:29:17 I'm ok with Emacs + SLIME + SBCL. 04:29:41 ramkrsna [n=ramkrsna@unaffiliated/ramkrsna] has joined #lisp 04:30:17 zilt [n=user@74.210.19.218] has joined #lisp 04:32:22 -!- aja [n=aja@S01060018f3ab066e.ed.shawcable.net] has quit [Client Quit] 04:34:12 dagnachew [n=dagnache@modemcable207.114-201-24.mc.videotron.ca] has joined #lisp 04:35:06 hehe, I feel like it would increase the least common denominator in kernel hackers, if they were using lisp 04:35:33 nyef_ [n=nyef@pool-70-20-51-23.man.east.verizon.net] has joined #lisp 04:35:44 hefner: I don't know about fun, but I have some frightening project ideas for you. Starting with "user-mode-lispnux", which is basically user-mode-linux but with a "kernel" written in lisp. 04:37:09 Wouldn't xen be a better target? 04:37:34 Depends on what you're trying to do, doesn't it? 04:37:55 The advantage to user-mode-lispnux is the ability to attach a debugger. 04:38:49 That might indeed be easier within xen in the way. 04:38:56 er without. 04:41:07 The upside? It's lisp! The downside? It's still unix, too. 04:42:09 yoshinori [n=read_eva@router1.gpy1.ms246.net] has joined #lisp 04:43:19 -!- sohail [n=Sohail@unaffiliated/sohail] has quit [Remote closed the connection] 04:43:42 unix, the ultimate bootloader 04:44:06 sohail [n=Sohail@unaffiliated/sohail] has joined #lisp 04:44:08 not only unix, but linux even heh 04:44:11 needs moar grub 04:44:13 nyef, i once read a review of a book called the unix hater's handbook which presented complaints from users of more advanced systems which had been forced to use unix because of its commercial success 04:44:42 I imagine the contents would be more interesting than the review. 04:45:06 -!- sohail [n=Sohail@unaffiliated/sohail] has quit [Remote closed the connection] 04:45:10 binarycodes [n=sujoy@59.93.169.189] has joined #lisp 04:45:17 well, it was 20 years ago 04:45:26 -!- binarycodes [n=sujoy@59.93.169.189] has quit [Client Quit] 04:45:33 Ok here's a q for you sbcl wizzes. 04:45:39 sohail [n=Sohail@unaffiliated/sohail] has joined #lisp 04:45:40 _sohail_ [n=Sohail@d207-81-121-15.bchsia.telus.net] has joined #lisp 04:45:50 I need to set up a callback that is called from the C world and the documentation goes like this: 04:45:51 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 04:46:01 The code in the supplied function must be suitable for real-time execution. That means that it cannot call functions that might block for a long time. This includes malloc, free, printf, pthread_mutex_lock, sleep, wait, poll, select, pthread_join, pthread_cond_wait, etc, etc. 04:46:03 -!- _sohail_ [n=Sohail@d207-81-121-15.bchsia.telus.net] has quit [Connection reset by peer] 04:46:05 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 04:46:07 -!- sohail [n=Sohail@unaffiliated/sohail] has quit [Read error: 104 (Connection reset by peer)] 04:46:25 Soo.. with sbcl I'll do what to make it all safe and pretty? :) 04:46:39 a big complaint that they had at that time was x windows, but that's rather matured 04:47:03 another was the security model, but no one else really had a better one 04:47:11 -!- nyef [n=nyef@64.222.164.52] has quit [Read error: 110 (Connection timed out)] 04:47:13 I'm thinknig maybe turn off the GC. 04:47:26 (what could go wrong!) 04:47:29 schme: That's one option... sortof. 04:47:46 What could go wrong? Well, if SBCL didn't create the thread the callback is invoked on, plenty! 04:47:57 of course :) 04:47:58 (GC or no GC.) 04:47:59 so the whole thing ends up reading like carping which was mostly all invalidated by the rise of linux and open source 04:48:12 nyef_: How do you mean? 04:48:21 except for the /etc directory :P 04:48:46 I mean, SBCL fondly imagines that it creates all of the threads that can run lisp code. 04:48:47 sohail [n=Sohail@unaffiliated/sohail] has joined #lisp 04:49:00 cads: different security models do exist. the only problem is that they require lots of configuration 04:49:11 nyef_: Well ok. But the callback is lisp code. 04:49:21 a lot of* 04:49:32 gigamonkey [n=user@adsl-99-50-126-237.dsl.pltn13.sbcglobal.net] has joined #lisp 04:49:48 So, let's see... first, I'd try to not cons at all in the callback. 04:49:58 nyef_: Here's what goes on. I register a callback with jackd, that jackd can call "when there is work to be done", but the whole callback is in sbcl. 04:50:21 <_3b> sounds like writing the callback in C would be easiest 04:50:48 Actually, first I'd see if it was coming in on a lisp-created thread or a thread created by jackd. 04:50:49 _3b: But basically it just sets a few vars around lisp space :) 04:50:52 <_3b> (or write a c-like low level compiler for sbcl) 04:51:24 oh it is always called by jackd. 04:51:28 If it's on a non-lisp thread, I'd need to break out my evil bag of tricks. 04:51:39 Modius [n=Modius@davidp4.lnk.telstra.net] has joined #lisp 04:52:23 <_3b> schme: yeah, but if you move the lisp-c boundary somewhere you control, you can set your own rules which are easier to be sure of from the lisp side 04:52:44 Hmmm 04:52:55 If it's on a lisp thread, it's easy enough, just have to make sure hte callback doesn't cons, or doesn't cons sufficiently to exhaust its alloc region, ever, and possibly have the actual function code-object pinned. 04:53:26 -!- dagnachew [n=dagnache@modemcable207.114-201-24.mc.videotron.ca] has quit ["Leaving"] 04:53:28 <_3b> nyef_: no risk of the GC in another thread stopping the callback? 04:53:31 fusss_ [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 04:53:33 May dig out the evil bag of tricks to get a fresh alloc region every time the callback is invoked, though... 04:53:52 _3b: Oh, that's easy enough to finagle. Just run without threading enabled. :-P 04:54:09 hehe 04:54:25 Right well. I'm sure this will work out just fine. 04:54:30 I'd also run without threading if jackd is creating its own threads. 04:55:19 Sure, the callback would be running in a thread that the runtime had no idea about, but that's all workable... 04:55:26 -!- kpreid [n=kpreid@cpe-67-242-4-64.twcny.res.rr.com] has quit [] 04:55:57 kpreid [n=kpreid@cpe-67-242-4-64.twcny.res.rr.com] has joined #lisp 04:56:35 Hmm.. 04:56:44 Yes I'll build a sbcl without the threads for now. 04:57:44 Oh, if jackd isn't creating its own threads you can run on a threaded sbcl, just that actually making threads is risky. 04:58:35 z0d: yo 04:58:57 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 04:59:21 That's a thought, though... I already believe that when a thread is executing foreign code it shouldn't have to stop for GC. 04:59:29 Does sbcl (1.0.19) exiting with code 256 mean anything in particular? 04:59:36 nyef_: I'm not quite sure how it all works. The url they provide with the "see more here!" 404's :) 04:59:43 slyrus_: hello. I tried running nuclblog and noticed some things: 04:59:47 Heh. 04:59:47 I seem to be unable to asdf load large libraries that used to load fine. 04:59:52 exit codes > 255 don't come from the process 04:59:58 Ah. 05:00:19 you mean process exit code, or something else? 05:00:30 Wait, wait. Code 256? 05:00:39 What's your host platform? 05:00:41 z0d: oh, nice. I don't think I've had a new victim^H^Huser in some time 05:00:51 Debian running on a tech.coop VPS. 05:01:08 Something's wrong. You can't -get- that as an exit code on linux. 05:01:18 slyrus_: hunchentoot-vhost HEAD contains references to hunchentoot::dispatch-table, but it's not called virtual-host-dispatch-table 05:01:20 It's an 8-bit field. 05:01:27 Hmm, that's what Slime is telling me it was. 05:01:35 -!- araujo [n=araujo@gentoo/developer/araujo] has quit [Read error: 60 (Operation timed out)] 05:01:52 that may mean it's failing to exec sbcl or the like? 05:01:56 -!- fusss_ [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 05:02:08 Sorry, did I say it was an 8-bit field? I was wrong. 05:02:14 It's a 7-bit field. 05:02:14 fusss__ [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 05:02:14 -!- fusss__ is now known as fusss 05:02:17 No, sbcl starts. Then I try to load-system something like hunchentoot or cl-pdf and it barfs after a while. 05:02:21 The 8th bit is for indicating a core dump. 05:02:23 hm, yeah shells return 127 :p 05:02:38 slyrus_: I get errors like: "The function HUNCHENTOOT-AUTH::REQUEST-URI* is undefined.". I failed to fixed this one. 05:02:39 araujo [n=araujo@gentoo/developer/araujo] has joined #lisp 05:03:13 z0d: you're using the git heads? It will probably have to wait until i return tomorrow before I can track these down. 05:03:21 Err... Scratch that again, it's an 8-bit field for a normal exit. 05:03:27 slyrus_: I get it when I try to open a page which requires authentication, e.g. /login 05:03:28 Still, 256 is out of range. 05:03:38 slyrus_: Yes 05:03:49 So I'm trying it in sbcl run from the shell to get SLIME out of the loop. 05:04:47 z0d: ah, you must be using the release hunchentoot instead of the dev version 05:04:49 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 05:04:55 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 05:05:15 z0d: I made a poor decision not to put the hunchentoot-dev support on a branch 05:05:29 slyrus_: I'm using what clbuild downloaded. Isn't that HEAD too? 05:05:36 -!- ikki [n=ikki@189.228.229.54] has quit ["Leaving"] 05:05:41 I was expecting a hunchentoot release, but it has been a while 05:06:02 z0d: couldn't tell you. The one that works with nuclblog is h4ns' ediware svn repo 05:06:02 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 05:06:36 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 05:06:44 Ugh. It's gone midnight. 05:07:22 this is what clbuild downloaded: http://common-lisp.net/~loliveira/ediware/hunchentoot/ 05:07:51 seems like a release 05:08:40 ok. you can either try the svn dev hunchentoot, or I can consider backporting things to the release hunchentoot. 05:08:40 slyrus_: so if I used HEAD hunchentoot everything will work fine? 05:08:45 G'night all. 05:08:46 yeah, it should 05:08:48 night nyef 05:08:53 Good night nyef_. 05:09:00 slyrus_: Ok, thanks. I'll try it 05:14:20 slyrus_: where is the hunchentoot svn repo? I can't find it 05:14:23 tiesje [n=user@202.63.242.211] has joined #lisp 05:14:23 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 05:14:26 fusss_ [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 05:14:29 -!- fusss_ is now known as fusss 05:14:37 -!- Modius [n=Modius@davidp4.lnk.telstra.net] has quit [Read error: 110 (Connection timed out)] 05:15:03 mmmmm, can't get that plastic toy taste of sqlite from my mouth 05:15:28 fusss: try rucksack 05:15:36 no i wont 05:15:38 :-P 05:16:59 z0d: i play with it from time to time 05:18:25 z0d: try svn://bknr.net/svn/ediware 05:19:04 slyrus_: thanks 05:19:33 -!- l_n [n=shawn@tuxhacker/lordnothing] has quit ["foo."] 05:21:29 -!- segv_ [n=mb@p4FC1DD0F.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 05:21:41 clsql pwns. really. good stuff there. 05:22:49 segv_ [n=mb@p4FC1E09C.dip.t-dialin.net] has joined #lisp 05:23:27 Modius [n=Modius@davidp4.lnk.telstra.net] has joined #lisp 05:23:46 bpt [n=bpt@cpe-071-070-209-067.nc.res.rr.com] has joined #lisp 05:33:00 fusss: you have got to be kidding 05:33:21 i cannot stand clsql. 05:33:31 then again .. i had to actually work with it for 4 years or so 05:33:46 clsql? 05:34:14 eno__ [n=eno@adsl-70-137-148-198.dsl.snfc21.sbcglobal.net] has joined #lisp 05:34:14 -!- Modius [n=Modius@davidp4.lnk.telstra.net] has quit [Read error: 60 (Operation timed out)] 05:34:25 yes. it got to the point that i wrote my own just to get away from it. 05:34:41 i have at least 5 clsql applications out there in the wild 05:35:38 why hammer a perfectly column oriented data to an object store? any dumbass can take care of a SQL database when you move on. no need to interrupt holidays or nights out cuz the site went down nobody can bring it back up because they can't load rucksack 05:36:23 It's a disease like that of transparent distribution or persistence. 05:37:17 it's gonna make me look bad afterwards if they can't back or migrate their database, imo 05:37:20 i use postgresql, don't get me wrong. 05:37:31 and now a break. 05:37:34 my clients would not accept rucksack 05:37:44 but i do not use clsql. 05:38:13 drewc: transparent access to sqlite, mysql and berkeley db. sounds like a win to me :-) 05:38:57 fusss: umm ... that last one? 05:39:11 bdb... you can do that with clsql? 05:39:15 i know, nor rdbms 05:39:28 yes you can 05:39:33 i did not know that. 05:39:41 sqlite i 05:39:57 oh shite! 05:40:11 never used .. but i would sooner add support for it to my ROFL than use clsql 05:40:12 you're right. i use Elephant atop clsql and it's elephant that does bdb. 05:40:35 mysql, otoh, has no good use case. 05:40:44 haha. 05:40:58 *schme* just spent yesterday building mysqlembedded to get the audio player working :) 05:41:18 aja [n=aja@S01060018f3ab066e.ed.shawcable.net] has joined #lisp 05:41:36 Which seems to be sqlite, but not. 05:42:35 who on earth uses mysql as an embedded db for an audio player? 05:42:50 oh.. the KDE people. 05:42:50 lots of people, several years ago 05:43:07 amarok uses MySQLe 05:43:25 fucken nice player too, but I am starting to doubt it is worth the 500MB KDE that comes with it :) 05:44:47 -!- eno [n=eno@nslu2-linux/eno] has quit [Read error: 110 (Connection timed out)] 05:45:09 I am out of coffee, but I have audio going in one side of SBCL and out the other. 05:45:12 eexcellent 05:45:24 you certainly like doing things the hard way. 05:47:08 schme: what exactly did you do? 05:47:09 Tha hard way? 05:47:15 sure, using jack 05:47:22 I can't think of an easier way. 05:47:32 fusss: I have it running as a jack client. 05:47:47 heh 05:47:56 doing any filtering? 05:48:15 I'm not doing anything with it at the moment. Just passing through. 05:48:52 -!- xinming [n=hyy@125.109.242.28] has quit [Read error: 110 (Connection timed out)] 05:49:13 hefner: I'd love to know of an easier way that using libjack for actually getting stuff to work with jack. 05:49:19 than 05:49:29 Dunno.. seems there is no other good way. 05:49:41 Well maybe rewire, but that is a bit pricey, and I can't do what I want with it. 05:50:04 I suppose if you need to work with jack, you're stuck. I'm free to waive that requirement while claiming you're making your life difficult, particularly when I have no idea what you're doing. 05:50:18 aha :) 05:50:23 haha :) I mean. 05:50:57 I plan to use sbcl for various audio tasks. I imagine it'll make a great soft synth. 05:51:09 I still need to get the output -> jackd. 05:51:15 -!- jbjohns [n=chatzill@52-45.3-213.fix.bluewin.ch] has quit ["ChatZilla 0.9.84 [Firefox 3.0.5/2008120122]"] 05:52:01 one of my proposed christmas hacks was to write toy little drum machine app, but I never got around to it. 05:52:15 Aha. 05:52:21 Well you need cl-jack for that. 05:52:33 No one will use it if you can't have it send output to jackd :) 05:52:44 I'll use it, because I don't use jack. 05:52:54 :) 05:53:01 anyway, I was going to write it in C, and run it on the nokia n800 05:53:06 Ah cool. 05:53:15 *schme* googles up the n800 05:53:26 instead, I'm spending the evening rewriting the timing of my nes emulator for the third or fourth time this week 05:54:05 That looks kinda nice that nokia. 05:54:37 yeah, they're neat. 05:54:49 Seriously though. JACK is the easy way. There is no other way to get a solid low latency audio system running on the linux :) 05:55:05 and it's hackable that nokia? 05:55:18 pardon me if I don't see how adding another layer of software helps me avoid latency :) 05:55:37 woah. 05:55:42 It says here it runs debian. 05:55:53 xinming [n=hyy@125.109.244.103] has joined #lisp 05:55:58 hefner: Ya that's ok. I didn't see it at first either. 05:56:06 *schme* pets his 4ms 05:56:35 I wouldn't call it debian, but it uses .deb packages. It's not hard to compile for, if you can get the environment installed (I couldn't, but people have made vmware images of it). 05:56:35 jsoft [n=Administ@unaffiliated/jsoft] has joined #lisp 05:58:09 Oh ok. 05:58:46 I think people have booted debian/arm on it, come to think of it. 06:00:08 It looks neat though :) 06:02:48 -!- eno__ [n=eno@adsl-70-137-148-198.dsl.snfc21.sbcglobal.net] has quit ["leaving"] 06:02:53 I'd stop work on the nes and get right on the drum machine though :) 06:02:58 everyone needs a good drum sequencer! 06:04:00 I never said I wanted to write a *good* one 06:04:12 holycow [n=bite@S01060016b6b53675.vf.shawcable.net] has joined #lisp 06:04:21 cheatcountry [n=cheatcou@cpe-72-177-217-220.satx.res.rr.com] has joined #lisp 06:04:29 more like another fun experiment in underengineering I could finish with in a few evenings 06:04:33 hi guys 06:04:39 couple of noob questions 06:04:46 hefner: hahaha. "that's the spirit" 06:04:51 holycow: Shoot :) 06:05:20 paranthases ... i don't really mind them at all, mostly they just add a few recursive parantheses in comparison to something like python or ruby 06:05:48 but ... is there a way to get simpler 'typing' experience with fewer paranths? 06:06:05 for example to quit a lisp interpreter i type in (quit) 06:06:10 no biggie just curious 06:06:17 Oh you want to have less parens. 06:06:33 Sorry, no. Don't go there, it's a bad place. :) 06:06:34 yeah i'm just a simple bash / python scripter / sysadmin 06:06:35 holycow: use paredit, to insert parens semi-automatically 06:06:45 thats an emacs extension? 06:06:48 holycow: yes 06:06:54 minion: tell holycow about paredit 06:06:55 holycow: please look at paredit: a set of Emacs commands, with a minor mode for convenient access to them, for editing balanced S-expressions and a number of higher-level operations on S-expressions, at ; see also the #paredit channel 06:06:55 k. i'll google it 06:06:58 not trying to stir shit btw 06:07:02 it relates to my next question 06:07:22 holycow: It's really just how you call functions in lisp (quit) calls the function QUIT. like you'd do halt. with the . in prolog. 06:07:31 *schme* stirs. 06:07:32 i want to start learning more lisp by using it for system admin tasks ... so evaluating various things on the file system like file names, file counts, file sizes, etc. 06:08:01 so my laziness is showing :) i'll google paredit 06:08:01 holycow: I press "one" key to get to parens in emacs. Alt-shift-( 06:08:14 oh hehe :) k that might work 06:08:33 Good night. 06:08:37 the other q i have is if i want to start thinking about my os/filessystem in lisp terms i guess i'm going to haveto use a lisp library for this, yes? 06:08:39 holycow: unix sys admin stuff and CL is perhaps not the best mix of things. 06:08:46 night z0d thx 06:08:49 left thumb on left alt. left ring-finger on left shift, right index on ( 06:09:04 fusss: Just do like normal people and map ( on the home row :) 06:09:13 btw, I swapped [ and ] with ( and ) in Emacs. it's easier to type 06:09:30 no Shift needed 06:09:36 welll true i mean one would not want to learn java try ing to manipulate the file system ... too ... fine grained 06:09:46 i lisp at the speed of thought. typing is the last of my problems. 06:09:46 lisp too i guess as everything has to be turned into functions 06:09:53 holycow: Though if you're interested in paranthesis in general there's the scheme shell (scsh) to replace all your bash needs. 06:09:57 but making those functions is what i hope to get practice in 06:10:22 aha, googling 06:10:36 holycow: stuff like paren-mode makes lisp typing feel like speaking. paren-mode is fine, but when you do the same for javascript (a la js2-mode) it pisses me off. 06:10:52 oh hehe 06:10:53 holycow: no googling required. you're looking for cl-fad 06:11:02 fusss: funny you mention that, when i read lisp thats what i like about it 06:11:06 its seems like i'm talking 06:11:08 in a way 06:11:38 okay so if i want to do stuff with my file system in lisp 06:11:49 what library/ies should i be looking at? 06:11:56 sb-posix 06:12:04 danke 06:12:19 oh that's a good save. 06:12:38 holycow: this has basic path stuff http://weitz.de/cl-fad/ 06:12:42 k. 06:13:36 <_3b> watch out for encoding problems, since lisp likes to deal with actual characters instead of arbitrary octets of unspecified encoding like unix tends to use :) 06:13:58 i'm sure i'll have lots of stupid q's :) 06:14:12 okay cool thanks guys, bookmarks and googling to go 06:15:10 holycow: try to play with the following libraries in any combination; bordeaux-threads, usocket, cl-ppcre, cl-fad, drakma, cl-mime, and closure-html. they do threads, sockets, regexes, web client, email mime parsing, and html parsing respectively. 06:15:19 Jabberwockey [n=jens@dslb-082-083-103-177.pools.arcor-ip.net] has joined #lisp 06:16:32 (all the stuff no one wants to do in one bang ;) ) 06:16:50 also osicat might have a few more of the stuff you asked for 06:17:05 schme: there is a good chunk of perl in cliki 06:17:28 k. added to my NFO for this self learning 06:17:40 minion: tell holycow about pcl-book 06:17:40 holycow: look at pcl-book: "Practical Common Lisp", an introduction to Common Lisp by Peter Seibel, available at http://www.gigamonkeys.com/book/ and in dead-tree form from Apress (as of 11 April 2005). 06:17:44 i just figure that plain exposure to simple things i need to get done on a regular basis anyway is a good start 06:17:57 Sure is. 06:17:59 what lisp are you using anyway? 06:18:02 oh yes right, they have a section on file systems there right 06:18:19 well common lisp from the ubuntu repos, lots of people insist on sbcl 06:18:32 i'm not entirely sure whatdifference it would make to me right now 06:18:36 holycow: Yes. It's the section named "I'm so sorry. the CL file system stuff really is this ugly. Please don't hate us :(" 06:18:38 sbcl IS a common lisp. did you mean clisp? 06:18:52 oh clisp yes 06:19:10 schme 06:19:13 lol 06:19:18 -!- aja [n=aja@S01060018f3ab066e.ed.shawcable.net] has quit [Client Quit] 06:19:23 thats mostly because of the libraries, not lisp tho right? 06:19:32 holycow: sb-posix will not work for you with clisp. 06:19:32 someone has to maintain that stufff and its frickin boring 06:19:40 oh so sbcl? 06:19:49 Well I'm sure clisp has something much like it. 06:20:09 But this is #sbcl, so it is probably easier to get help with it. 06:20:11 (: 06:20:25 holycow: everyone who uses or maintains sbcl is in the channel right now. 06:21:07 let me get this straight ... so common lisp is a specification, yes? 06:21:16 sorry i thought i understood this 06:21:17 yes 06:21:27 and sbcl and clisp are one of several implementations 06:21:29 k. 06:21:30 SBCL is to Common Lisp what GCC is to C 06:21:38 oh! 06:21:43 where does clisp fit? 06:21:44 sbcl == GCC, clisp == LCC 06:21:48 aha 06:21:53 heh, nice 06:21:59 interesting that paints a more clear picture 06:22:00 my thx 06:22:01 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [] 06:22:04 clisp: tiny, ultra portable, and not so smart 06:22:12 ohhhh 06:22:17 so like tcl vs. lisp 06:22:23 or scheme vs lisp or something 06:22:27 no 06:22:41 clisp is a complete common lisp, but sbcl has more non-standard stuff 06:22:44 <_3b> sbcl has a nice native compiler, but produces large executables, clisp compiles to bytecode 06:22:45 nm the original comparison seem smore apt :) 06:22:46 lol 06:23:23 k. i'm not worried about delivering anything usefull right now anyway 06:23:26 cool 06:23:32 <_3b> they should agree on behavior for any conformant CL program, but have different libraries bult in for things beyond the CL spec 06:24:12 cmucl is the mack daddy though, it has a builtin GUI :-) through motif bindings 06:24:23 oh interesting 06:24:41 <_3b> clisp comes with a CLX or 2, doesn't it? 06:24:55 _3b: yeah, on unix. 06:24:56 2 indeed. 06:25:22 holycow: CLX is the lisp bindings for Xlib. Yep! you get the whole lot there. 06:25:32 <_3b> clx != bindings 06:25:33 -!- trebor_h` [n=user@dslb-084-058-216-182.pools.arcor-ip.net] has quit [Read error: 60 (Operation timed out)] 06:25:39 socket comm 06:25:42 <_3b> clx is a lisp implementation of the X protocol 06:25:44 am i right to understand that lisp compilation is fast? meaning that say you have a large'ish program or a component or something and you run it and want to make some sort of upgrade 06:26:05 <_3b> holycow: depends on the implementation and omcpilation settings 06:26:07 you can make your changes and the compiler 'compiles on the parts of the code required' ... did i read that right out there on the googlenet? 06:26:19 fusss: googling 06:26:27 holycow: Yes. You can compile parts of stuff. 06:26:30 sbcl compiles even slower than gcc, which is really saying something 06:26:32 <_3b> incremental compilation makes compilation speed matter less though 06:26:38 holycow: if you're compiling code as you type it, then yes. 06:26:42 but you rarely recompile your whole program 06:26:55 i'm just trying to build a mental picture here 06:27:05 so if someone was running a large open source app coded in lisp 06:27:13 and if you're using sbcl, any expression you evaluate is compiled immediately 06:27:13 <_3b> clozure CL is frequently cited as a good compiler for compilation speed 06:27:23 -!- anekos is now known as awayekos 06:27:37 you could pop open the source and make some changes and if you are using the right implementation you could have a recompiled binary in seconds if the changes are small enough? 06:27:40 Compilation in CL may also mean something other than what you expect. 06:28:05 my only frame of reference here are gcc and pythong/ruby sorts of things and lisp seems to do a whole bunch of things that fit in between 06:28:16 Zhivago: heh, its baking my noodle 06:28:48 One way to think about compliation in CL is that you are adding a bunch of invariance to some code. 06:29:17 Yay Zhivago speaks. 06:29:24 Compiled or uncompiled code in lisp may or may not be translated to native code. 06:29:29 -!- bhall [n=bhall@ext.hallgroup.org] has left #lisp 06:29:33 aha 06:29:47 (e.g., sbcl translates uncompiled code to native code) 06:29:48 holycow: you can always say (disassemble 'foo) to see the code generated for a given function :-) 06:29:53 hrm... how do I allow for strings like "a\b" to be entered at the repl and to be processed as the string consisting of #\a #\\ #\b? 06:30:17 but that uncompiled code must take into account that bits of it might be replaced or changed or so on, which compiled code doesn't need to. 06:30:24 <_3b> holycow: with lisp it is common to modify a program as it is running, instead of building a new binary for each change 06:30:46 <_3b> slyrus_: write your own " reader macro? 06:30:53 (well, for varoius kinds of changes) 06:31:11 aha right 06:31:20 <_3b> slyrus_: (assuming the answer you were looking for isn't type "a\\b") 06:31:23 Quad: I downgraded to ircii which makes it easier to be in several places at once. 06:31:42 _3b: right, that's not the answer I'm looking for :) 06:32:02 Zhivago: Have you used xchat? 06:32:13 Does your example include the quotes? 06:32:40 Quad: I don't think so. 06:33:05 Zhivago: I'll screenshot for the sake of...screenshotting. 06:33:31 Zhivago: I'd like to be able to enter something like: (parse "a-string-with-a-\") at the repl 06:33:47 holycow: i'm editing the source code of my webserver right now, and whenver i make a change in text and evaluate the changed form my server behaves differently. no restarting :-) 06:33:56 whoa 06:33:57 Well, you could consider replacing the read macro for " 06:33:59 thats awesome 06:35:44 instead of compling a whole file, think of it as interacting with GCC via a shell, compiling each function, struct or typedef at a time, and the compiled code is saved to a database. once you're happy everything works as you expect, you're done. you tidy up the text and stash it in VCS. 06:35:46 that kinda blows my mind still, i've read that before, very cool 06:35:57 right right 06:36:36 I like an analogy with unix, where packages are directories, functions are executables, and the repl is a shell. 06:36:41 <_3b> sort of like livecoding if you are familiar with that, but for normal programs :) 06:37:13 the difference between, say, sbcl and python is that whatever you typed into sbcl is converted to native processor binary. in python it's a parse tree waiting to be interpreted by the VM. 06:37:20 well web apps sorta behave like that so yeah 06:37:27 hefner: *nod* neato 06:38:03 Zhivago: http://img514.imageshack.us/img514/1417/screenshot2yj6.png 06:38:17 i feel like making some disparaging comments about c++ and java :) but i'm just a crapy scripter so ill shutup 06:38:19 hehe 06:38:26 neat 06:39:40 Aankhen`` [n=Aankhen@122.162.169.62] has joined #lisp 06:40:33 <_3b> hmm, wonder if i should shove some more stuff into my avm2 assembler, or add a linker layer too 06:40:34 i really hope the stereotype about lisp being for "elite smart people" just goes away. all you need to grok it is good taste. 06:40:50 thats just mareting 06:40:59 I don't think you even need good taste. 06:41:00 _3b: oh, forgot about that one. far you coming? 06:41:14 if people can be convinced that ror is the next comming of jesus than it should be easy to show how life can be nicer with lisp 06:41:22 i "got" lisp because C++ and win32 COM screwed me up, royally 06:41:23 <_3b> fusss: rewriting the compiler to try to make it smart enough for closures 06:41:27 what you described makes coding in c++ seem .... well horrible 06:41:42 <_3b> holycow: it is :( 06:41:48 it's not that bad. 06:42:11 <_3b> holycow: one of the risks of learning lisp is that it can make using other languages more painful :) 06:42:12 depends how you do it. 06:42:18 C++ is good if you're following a strict regimented plan and don't need to be very creative. 06:42:26 lol _3b 06:42:45 fusss: Now that's not true. :> 06:43:47 anyone know what is the new ^ keyword MS added to ceeplusplus? 06:44:01 <_3b> fusss: currently trying to decide how to handle putting references to compiled lambdas into other compiled lambdas, which either needs the asm to do more, or needs a linker step 06:44:13 it looks like smalltalk instantiation, but it's use like a pointer dereference 06:44:51 okay i'm about to ask something completely stupid because i don't understand it 06:45:31 c libraries seem like they can be used by pretty much any other language/app but c++ has problems in that area. can lisp libraries/apps be used by other languages? 06:45:43 so say i develop some app that does some neat file monitoring say in lisp 06:46:05 and i want to modify a python app to use a lisp library developed for that lisp app ... is that even conceivable? 06:46:12 <_3b> lisp tends to want a lisp runtime, so it can be messy... some implementations are better than others though 06:46:19 _3b: alright. 06:46:20 i.e. do lisp programs only live in the lisp world? 06:46:24 in general, no one does this. 06:46:25 -!- runenes [n=runenes@proxy-gw.uib.no] has quit [Read error: 110 (Connection timed out)] 06:46:40 holycow: It is doable, but it sounds pretty painful :) 06:46:42 okay, that would be why its stupid 06:46:43 <_3b> ecl and some of the commercial lisps in particular put effort into making that sort of thing work 06:46:44 Not when my lisp is done! :D 06:46:45 holycow: without any win :) 06:46:53 fusss: looks like an extension for GC-friendly references in managed C++ 06:47:11 holycow: you can call nearly any C code from lisp. vice versa is not usually straightforward. 06:47:26 pkhuong: oh, thank you :-) 06:47:29 aha 06:48:12 holycow: the CFFI library helps you call C code from Lisp. I use Gtk from Lisp, for example. 06:49:01 holycow: You can also look at lisps which compile to C. 06:49:09 ejs [n=eugen@77-109-26-148.dynamic.peoplenet.ua] has joined #lisp 06:49:12 Ah, ecl has been mentioned already. 06:49:15 in python you can call C with a swig binding. but for C to call pythong you have to compile your binary against python to "embed" it, and not as straight forward. same thing. 06:49:17 oh right if one needed that ... interesting 06:49:41 you get the benefit of lisp environment + export of functionality to something more accessible 06:49:44 k. 06:49:44 ecl is quite nice -- it does dynamic compilation via gcc and dynamically loaded shared libraries. 06:50:00 It isn't a particularly clever compiler, but you can't have everything. 06:50:14 netaust1n [n=austinsm@68-191-51-69.dhcp.nwtn.ct.charter.com] has joined #lisp 06:50:28 yeah, i don't think i will ever get good enough to use it ... was just curious about an edge use case 06:50:37 i suggest the other way. collect the C libraries you wanna use and drive them from Lisp. CL is the more accessible platform. 06:50:57 holycow: why wouldn't you get good enough at lisp? 06:51:16 are you a deathrow inmate? 06:51:21 fusss: lol! 06:51:40 holycow: Once you know what CAR and CDR are, you know all of lisp! 06:51:43 its funny that you guys don't consider programming one of those 'you either have it or dont' things 06:51:50 (oh, DEFUN, can't forget about DEFUN) 06:51:50 marrying that amish girl soon ;-) 06:52:05 i'll tell you why i'm interested in lisp 06:52:26 i've come to the conclusion that the reason i can script at all is basically through exposure to computers and linux in particular 06:52:28 holycow: lisp => if you don't have it, then you can have it soon with a few button presses of the keyboard 06:52:29 holycow: There seems to have been a lot of things on reddit claiming that.. "either you have it or you don't". I say it is bullcrap. 06:52:38 Zhivago: that seems a little bit like playing with fire. I think I'll use a #" reader macro when I want my special #\\ parsing. 06:52:58 and i've come to realize that most people are kept from actually understanding computers by not having a command line interface ... so the mac windows thing is sort of a jail 06:53:21 holycow: (a lot of lispers love their macs ;) ) 06:53:31 and their ajax GUIs :-) 06:53:34 heh 06:53:38 A lot of people don't want to 'actually understand' their computers. 06:53:50 i just think if kids had a linux box available from the time they can walk 06:53:58 Like cars.. I'm happy with not understanding 'em. 06:54:04 they just would kind of learn to script by virtue of exposure 06:54:19 schme: I'm not. Car service is a huge ripoff. 06:54:39 so if there is a transitional behavriou patern there one can extend that a bit furtehr to get normal people to get computers to do interesting things beyond click buttons and buy iphone apps that shake boobs and such 06:54:41 heh 06:54:58 and lisp seems accessible to me from a 'plain reading' perspective 06:55:08 thats pretty well my inspiration 06:55:09 holycow: My pseudo-lisp interpreter compiled on the iphone before I went to bed last night. 06:55:11 hefner: Oh. I don't really even have a car, so maybe if I did.. though I doubt it. I hear they are much like computers. Iremember how one got the huge reference manual and code poke the sucker with a welding iron. now.. you just can't. I hear cars are like that too. 06:55:12 And it worked. :D I was happy. 06:55:29 whoa, really? 06:55:35 hehe just how 'speudo'? 06:55:43 Not very pseudo. 06:55:51 Just not featured. 06:55:56 neat 06:56:02 schme: i have a car and i have thrown my keys in the attic. 45 days without driving so far. 06:56:17 fusss: lol, thats hillarious 06:56:20 holycow: Personally I see no great glory in getting people interested in getting computers to do interesting stuff. But if it motivates you, great stuff there, mate. 06:56:43 fusss: That's it. Get a bike like a normal person :) 06:56:47 brb 06:56:47 schme: well maybe you are right, its more of an insight than something i care to accomplish 06:56:54 i'm hardly responsible for other peoples lives 06:56:56 *_3b* doesn't drive much, but couldn't not drive at all 06:57:06 i walked quarter a mile to buy my cigarettes today 06:57:15 That's crazy. 06:57:19 *_3b* doesn't want to walk or bike on roads with 55+mph speed limits and no shoulder/sidewalk 06:57:34 oooh. 06:57:44 tangential question guys ... but related to lisp anyway 06:58:05 holycow: I think you should start hacking on some lisp projects. It's a ball of fun. 06:58:16 i've made web apps, right now i'm working openacs/tcl/aolserver which is fun, somewhat lisp like and smallish scripting language 06:58:30 holycow: I hear some amazing guy even got his sbcl to register as a JACK client with jackd. 06:58:43 lol, what? thats crazy talk 06:58:45 :) 06:58:57 Ya guess so. 06:59:10 holycow: he did that earlier :-P 06:59:20 common lisp is shit for scripting your unix admin stuff though. :) 06:59:24 now what web apps suffer from is the fact they are shoving really crappy frontends on them using technologies never meant for that 06:59:25 schme: trolling for some lovin' arent' we? ;-) 06:59:31 haha. 06:59:55 javascript/forms/html/xhtml were never meant for that sort of thing, just the usual example of a simple technology being abused by those that don't know shit 06:59:56 ok store is open. I can go buy coffee. 07:00:09 get me another pack of marlies 07:00:23 is there anything in the lisp world where the interface is separate from the logic which is separate from some sort of server? 07:00:33 What wasn't javascript meant for? :) 07:00:55 so that an app can either be deployed locally like a normal static binary app, and then deployed over the net with the server sitting over here and the interface sitting over there? 07:00:57 javascript = 4chan trolling 07:01:09 Zhivago: lol, there are people writing webservers in javascript, it kinda cool 07:01:22 holy: There has been some talk of using clim like that. 07:01:40 aha 07:01:46 I'm kind of of the opinion that you might as well just start with html, these days. 07:01:53 <_3b> most people just go straight to web, haven't heard too many worrying about local + web 07:02:02 holycow: Aptana Ajaxer is all javascript 07:02:13 well i'm of the opinion that the app interface needs to be redefined just a little bit 07:02:20 In terms of half-arsed scripting languages javascript is pretty nice. 07:02:35 I'm hoping that it puts up some competition to python. 07:02:38 the xaml thing and the uxl kind of point the way to how to build interfaces 07:02:56 throw an ide on top of those and you can get non coders to design uis and hook them into binaries without knowing any coding 07:02:58 holycow: you can try Itanium, or Google Native Client, or Adobe AIR; nothing Lispy there yet, but we have parenscript to generate javascript from a lispy syntax :-P 07:03:45 kdevel has this great thing where i can point the ui to 'activators' or whatever they are called ... kind of an easy way to give away power to specialists 07:03:48 but i digress 07:04:04 XAML, UXL and Flex are broken imo. and i say that as an old school gui hacker. 07:04:20 fusss: aha, in what way? ... just curious 07:04:31 visual basic did that in 1994, nothing new 07:04:40 okay right yeah 07:05:21 I think that html's dom model with javascript actually does a pretty good job for most UIs. 07:05:35 they do, they are just incomplete 07:05:38 holycow: they're in XML. a declarative document structuring language. since you can't express actual code inline, you have to plug in magic identifier to signal where the code goes in. hardly the mark of clarity you want it to be. 07:05:50 in comparison to what you can do with more sophisticated toolkits like qt 07:06:03 -!- awayekos is now known as anekos 07:06:43 fusss: well the idea is to abstract that task enough to not require a programmer ... unless a programmer wants to be involved 07:07:00 those things always sit on top of a mechanism that handles all of that ... so if a coder wants to get involved they can 07:08:05 stassats` [n=stassats@ppp78-37-25-243.pppoe.avangarddsl.ru] has joined #lisp 07:08:27 holycow: you would have a better idea of where computing should've headed, specially with respect to beginner friendly tools, if you looked at many many languages. along with Lisp, look at LOGO, smalltalk and Self. 07:08:54 hehe :) i admit knowing enough to strangle my self and everyone around me 07:08:58 plenty of rope around :) 07:09:19 okay anyway at least i undersand clim has had some discussion around that sort of thing 07:09:33 common lisp is a tool for software engineering; a messy workshop for the busy professional. 07:10:15 Where are the #ifdef and #endif statements in common lisp 07:10:32 fusss, Zhivago, everyone, thank you for answering my questions 07:10:37 Quadrescence: not quite the same, but #+ and #- might do what you want 07:10:47 fusss: I was attempting to troll. 07:10:55 :{ 07:10:57 np holycow 07:10:58 :) 07:10:59 -!- frosty [n=Frostysn@c-76-115-11-91.hsd1.wa.comcast.net] has quit [] 07:11:04 good morning 07:11:22 Quadrescence: blokha! (or however you say "bad" in parusky :-) 07:11:41 hey mvilleneuve 07:11:44 fusss: ploxo 07:11:46 i gotta go crash myself 07:12:06 "plo-ha" (hard h) 07:12:07 ;) 07:12:33 -!- mrsolo [n=mrsolo@adsl-68-126-200-32.dsl.pltn13.pacbell.net] has quit [Remote closed the connection] 07:12:59 Quadrescence: i know 07:12:59 mrsolo [n=mrsolo@adsl-68-126-200-32.dsl.pltn13.pacbell.net] has joined #lisp 07:13:07 fusss: Oh. ;) 07:14:48 Ok. The store is not open for another hour. 07:15:08 I have started my 10th hour without caffeine. 07:15:17 btw, for anyone interested, i found my demo of mirai, a 3d app from about 7 years ago that was coded entirely in franz lisp 07:15:19 :) 07:15:28 would need nt4 to run it i think 07:15:30 maybe 2k 07:16:45 -!- gigamonkey [n=user@adsl-99-50-126-237.dsl.pltn13.sbcglobal.net] has quit [Read error: 110 (Connection timed out)] 07:18:45 MrSpec [n=NoOne@82.177.125.6] has joined #lisp 07:18:58 Hi 07:23:49 -!- stassats [n=stassats@wikipedia/stassats] has quit [Read error: 110 (Connection timed out)] 07:25:41 beach [n=user@serveur5.labri.fr] has joined #lisp 07:25:46 Good afternoon. 07:25:48 Hi. 07:27:54 hello bea 07:27:55 hello beach 07:29:53 oh hehe 07:30:04 inline comments in lisp ... *sniff* beautiful 07:30:09 oh god how i hate the hash mark 07:30:28 sweet i didn't know you could actually have a sensible way to write out comments 07:31:11 -!- rcy [n=rcy@S01060002553240a8.vc.shawcable.net] has quit [Read error: 110 (Connection timed out)] 07:33:49 http://www.international-lisp-conference.org/2009/index 07:33:55 lisp conferences? 07:33:57 neat 07:36:53 -!- crod [n=cmell@cb8a69-009.dynamic.tiki.ne.jp] has quit [Read error: 110 (Connection timed out)] 07:37:15 crod [n=cmell@cb8a6a-113.dynamic.tiki.ne.jp] has joined #lisp 07:40:40 for the bargain price of your firstborn, you too can attend the illustrious International Lisp Conference 07:41:03 Draggor [n=Draggor@216-80-120-145.alc-bsr1.chi-alc.il.static.cable.rcn.com] has joined #lisp 07:43:31 hefner: what do these things cost? 07:43:34 i don't see pricing up there 07:43:41 -!- slyrus_ [n=slyrus@207.189.195.44] has quit [Read error: 110 (Connection timed out)] 07:44:15 I don't know offhand, but if you aren't a student, they really bend you over 07:44:23 lol 07:45:34 any idea what the student price is? 07:45:40 ... hmm 07:45:43 -!- ianmcorvidae|alt is now known as ianmcorvidae 07:46:15 hefner: $500 or so, right? 07:46:36 holy f 07:46:37 really? 07:46:51 thats an awesome way to keep students out 07:46:55 Tonight, I dreamt I had started to use Emacs. :/ Probably why I was all shaky this morning. 07:49:05 Morning #lisp. What's up? 07:50:05 mega1 [n=mega@53d82bc8.adsl.enternet.hu] has joined #lisp 07:50:34 -!- daniel_ is now known as daniel 07:51:42 Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has joined #lisp 07:52:49 fusss_ [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 07:53:06 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 07:53:15 -!- fusss_ is now known as fusss 07:54:56 -!- MrSpec [n=NoOne@82.177.125.6] has quit [Read error: 110 (Connection timed out)] 07:55:29 mikezor_ [n=mikael@c-75ec70d5.04-404-7570701.cust.bredbandsbolaget.se] has joined #lisp 07:55:52 tic: what a nightmare! 07:56:09 beach, worse yet, I enjoyed it! :/ 07:56:11 tic: if you don't do anything, next time it will be worse: Climacs. 07:56:21 lol 07:56:27 all the horror of emacs, plus your meta key won't work 07:56:31 oh hey beach you are one of the guys working on climacs right? 07:56:44 -!- Jabberwockey [n=jens@dslb-082-083-103-177.pools.arcor-ip.net] has quit [Read error: 110 (Connection timed out)] 07:56:51 holycow: I started it, but havn't done anything for a while. 07:56:57 oh neat 07:57:12 hefner, that's the other Swede, schme! On my end ISO_Level_3_Shift doesn't work. :) 07:57:40 holycow: Everyone said that it couldn't be done, or that it would take decades. I think we proved them wrong. 07:57:52 beach: i used to be a graphic designer 07:57:56 it might take decades to get a working meta key everywhere 07:58:00 (morning) 07:58:01 i can help on the website or something 07:58:08 Working out a Vim emulation mode for Climacs would probably make me feel better inside, if nothing else I don't know elisp. 07:58:11 what state is it in now? 07:58:19 and here I was going to make a joke about it taking decades to redisplay, but that's really not fair anymore 07:58:28 no, Athas fixed that one 07:58:34 Krystof, I think it's a combination of crappy keymap and some oddities of CLX. Because it does work, at least on my end. 07:59:39 it's a symptom of CLX not supporting XKEYBOARD and the whole of the rest of the world assuming that XKEYBOARD is supported 07:59:52 XKEYBOARD is meant to be backwards-compatible, but it isn't (quite) 07:59:58 beach: was the idea to remain consistent with emacs? 08:00:10 so gnome sets up your keyboard using XKEYBOARD, and the almost not quite compatible server translation layer stuffs up modifiers 08:01:01 there's an idea - report non-backward compatibility as a bug. I'm sure they'll hop right on that. 08:01:46 I have had discussions with people who actually do stuff with X.org 08:01:47 Krystof, I set up my keyboard using xmodmap, actually, but I guess that's roughly equally broken nowadays -- I do run gnome-panel, but otherwise I try to stay away from it. 08:02:12 essentially, every time I say "I'm not using " they say "well, use " 08:02:16 so I have given up that fight 08:02:16 -!- arbscht_ [n=arbscht@unaffiliated/arbscht] has quit [Operation timed out] 08:02:19 you're welcome to try 08:02:34 tic: hm, that's marginally more interesting a case, then 08:02:34 mishok13 [n=gdmfsob@dm.sonopia.com] has joined #lisp 08:02:38 arbscht [n=arbscht@unaffiliated/arbscht] has joined #lisp 08:03:01 hefner, he-he. "uhm, I have a totally custom keymap, and I can't type parens, ie. ISO_Level_Shift_3+Aring, plz hlp?" 08:03:09 Krystof, the xmodmap part? 08:03:58 yeah. assuming that the xmodmap you're using doesn't call xkbstuff behind the scenes, I would have hoped that clx and mcclim would work 08:04:39 Oh. It _does_ work, but CLX (same problem in StumpWM) doesn't recognize my iso_level_3_shift. 08:04:40 it might be something silly like the remove modifiers code in either mcclim's clx backend or esa's keybindings (I forget which) 08:05:02 Ok, that means it's probably my fault 08:05:06 or responsibility, whatever 08:05:22 i.e. when I press AltGr-ÅÄÖ I should get ?() in StumpWM, but I just get ÅÄÖ. 08:06:04 Mod5 is Mode_switch (0x5d), ISO_L3_S (0x71), ISO_L3_S (0x7c); according to xmodmap -pm 08:06:19 excellent, sounds like an excuse for me to try to page X keyboard handling back into my brain 08:07:18 -!- r2q2 [n=user@c-24-7-212-60.hsd1.il.comcast.net] has quit [Connection timed out] 08:07:34 -!- mikezor [n=mikael@c-75ec70d5.04-404-7570701.cust.bredbandsbolaget.se] has quit [Read error: 110 (Connection timed out)] 08:08:52 Yay! In case you need it, here's http://mikael.jansson.be/static/config/Xmodmap and http://mikael.jansson.be/static/config/Xmodmap.effective for what xmodmap -pm -pp -pke tells me after it's been loaded. 08:09:37 holycow: nice! Perhaps you could design a native look for McCLIM applications? 08:10:12 i could give it a shot 08:10:29 holycow: excellent. 08:10:36 although in true lazy gpl fashion, re-using kde's oxygen would be a great way to start 08:11:19 holycow: consistency with Emacs is a goal only when the choice doesn't matter. The main goal is to improve on Emacs, such as the concept of syntax modules which is better than that of modes. 08:11:27 Someone write MAPCAR in lisp 08:11:30 3 2 1 go 08:11:36 beach: interesting 08:11:43 -!- loz [n=loz@163.c.003.ncl.iprimus.net.au] has quit ["Leaving"] 08:14:04 <_3b> Quadrescence: (defun mapcar (&rest a) (apply 'cl:mapcar a)) :p 08:14:14 heh! 08:14:20 :< 08:14:29 :D 08:14:32 *_3b* probably should have specified cl:apply 08:14:41 or perhaps using LOOP? (the implementation of LOOP left as an exercise) 08:14:48 tic: hehehe 08:14:52 exercise. 08:14:57 <_3b> yeah, that was more or less my point 08:15:14 <_3b> you need to say which parts of lisp are available 08:15:23 _3b, mhm, I was just trying to be more annoying. Guess that wasn't very funny when you had already given the above solution. 08:15:33 function application, car, cdr, if, cond 08:15:36 ;) 08:15:37 If not much of Lisp is available, a simple recursion then? 08:15:43 <_3b> Quadrescence: no tco? 08:15:47 Nope. 08:15:52 Ouch. 08:15:55 <_3b> if not, you probably want some looping construct too 08:16:06 Wait, what's tco? :| 08:16:12 well, mapcar w/o tco is okay, too, just don't give it large lists. 08:16:15 Tail Call Optimization. 08:16:18 <_3b> tail call optimization (you probably don't have it) 08:16:18 Oh. 08:16:37 I don't. I wonder how that optimization is done specifically, anyway 08:17:08 <_3b> depends on how you compile 08:17:10 Actually, I have WHILE 08:17:11 :D 08:17:39 <_3b> usually just compile it to a jump of some sort 08:18:25 -!- dto [n=user@68-187-211-226.dhcp.oxfr.ma.charter.com] has quit [Remote closed the connection] 08:20:07 dto [n=user@68-187-211-226.dhcp.oxfr.ma.charter.com] has joined #lisp 08:27:49 *_3b* wonders how much if any performance benefit there is for calling method directly instead of doing a lookup by name in avm2 08:30:59 finally I have my coffee. 08:32:03 Darkness3477 [n=maveric@122-148-16-200.static.dsl.dodo.com.au] has joined #lisp 08:34:13 ManateeLazyCat [n=Andy@222.212.133.86] has joined #lisp 08:37:13 -!- sohail [n=Sohail@unaffiliated/sohail] has quit [Remote closed the connection] 08:37:41 toddoon_ [n=guillaum@mar92-11-82-245-210-60.fbx.proxad.net] has joined #lisp 08:38:30 -!- crod [n=cmell@cb8a6a-113.dynamic.tiki.ne.jp] has quit [Read error: 110 (Connection timed out)] 08:38:37 crod [n=cmell@cb8a16-213.dynamic.tiki.ne.jp] has joined #lisp 08:38:43 are there any "literate programming" modes for Lisp source code, along the lines of Literate Haskell where code begins with > in the first column? 08:39:34 tic: literate programming is for nubs. Just comment well, docstring well, and write a manual in TeX on the side 08:39:35 :D 08:40:00 Where would one put the >'s ? 08:40:08 also good mornin', tic 08:40:30 schme: Obviously, you use a '>' special form 08:40:39 > is not a form. 08:40:45 I know. 08:40:50 schme, lines beginning with > means "here be code", otherwise assume comment. 08:41:02 You might use some horrible reader macrology, though. 08:41:04 I guess one could hack the reader. 08:41:07 tic: I know how it works in the haskell. I'm just curious how it would work for lisp. 08:41:16 I mean haskell source has a concept of lines. 08:41:42 You could just use #| |# 08:41:59 #> <# for source! 08:42:07 (discarding everything else) 08:42:21 #{ }# 08:42:33 Quadrescence: that sent shivers down my spine. 08:42:39 >:D 08:43:58 -!- andrerav [i=andrerav@tvilling2.pvv.ntnu.no] has left #lisp 08:44:12 sohail [n=Sohail@unaffiliated/sohail] has joined #lisp 08:46:08 pstickne [n=pstickne@c-71-236-177-238.hsd1.wa.comcast.net] has joined #lisp 08:48:14 -!- beach [n=user@serveur5.labri.fr] has left #lisp 08:51:48 -!- lambda [n=lambda@c-24-5-198-113.hsd1.ca.comcast.net] has quit [] 08:52:05 tiesje` [n=user@202.63.242.211] has joined #lisp 08:53:00 -!- ManateeLazyCat [n=Andy@222.212.133.86] has quit [Read error: 104 (Connection reset by peer)] 08:53:05 lambda [n=lambda@c-24-5-198-113.hsd1.ca.comcast.net] has joined #lisp 08:55:25 how about (comment whatever I'm typing here is for the literate programming part in will probably not be used anywhere) (some code is standing here) 08:55:36 just write a macro for it... 08:55:48 Use '(' and ')' to denote comments 08:57:11 bob_f [n=bob_f@mail.phgroup.com] has joined #lisp 09:00:48 ManateeLazyCat [n=user@222.212.133.86] has joined #lisp 09:01:24 l_a_m_ [n=lam@194.51.71.190] has joined #lisp 09:03:12 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit ["ChatZilla 0.9.84 [Firefox 3.0.5/2008120122]"] 09:05:54 nxt [n=nxt@77.207.25.109] has joined #lisp 09:09:06 -!- sohail [n=Sohail@unaffiliated/sohail] has quit [Read error: 113 (No route to host)] 09:09:28 ejs1 [n=eugen@92.49.232.3] has joined #lisp 09:11:11 Phoodus [i=foo@ip68-231-38-131.ph.ph.cox.net] has joined #lisp 09:13:37 -!- ejs [n=eugen@77-109-26-148.dynamic.peoplenet.ua] has quit [Read error: 110 (Connection timed out)] 09:20:49 -!- pchrist [n=spirit@gentoo/developer/pchrist] has quit ["leaving"] 09:21:26 pchrist [n=spirit@gentoo/developer/pchrist] has joined #lisp 09:21:42 Well what a bugger. What's a good alternative to memcpy() ? 09:21:51 Nothing. 09:21:54 Why? 09:22:00 What are you doing? 09:22:23 copying data from one place to another place. 09:22:31 then memcpy 09:22:32 ;) 09:22:32 But I would like to do it in lisp of course. 09:22:41 oh 09:22:44 :\ 09:22:49 And it's from one pointer location to another. 09:22:58 eh. 09:22:58 ffi and memcpy. 09:23:05 Ya that's beautiful. 09:23:13 :) 09:23:25 antoni [n=antoni@48.pool85-53-16.dynamic.orange.es] has joined #lisp 09:23:37 I don't really need a *good* alternative. I just can't think of anything that I could even use here other than the ffi. 09:23:39 aerique [i=euqirea@xs2.xs4all.nl] has joined #lisp 09:24:05 schme: In which lisp? 09:24:06 you have pointers. You're already in FFI land. 09:24:23 Zhivago: Well I'm in sbcl. 09:24:29 pkhuong: yup that's right. 09:24:37 schme: cmucl/sbcl have pointery things. 09:25:04 pkhuong: (let ((foo (the-c-func-that-returns-the-pointer)) (bar (same thing))) (memcpy foo bar))) :) 09:25:05 See SAP -- system area pointer. 09:25:22 Yeah, or use the ffi, which will be more portable. 09:25:35 But remember that memcpy takes three arguments. 09:25:38 Ya I'll ffi. 09:25:40 yup. 09:26:11 there's sb-kernel:system-area-ub*-{copy,fill} 09:26:56 I'm starting to think that maybe lisp is not so good for this stuff :) 09:27:12 But that should never stop one. 09:27:27 lisp+C is my preference. 09:27:41 One day I'll be as good as Zhivago. :3 09:28:25 -!- ia [n=ia@89.169.165.188] has quit [Read error: 60 (Operation timed out)] 09:28:44 -!- yangsx [n=yangsx@218.247.244.25] has quit [Read error: 110 (Connection timed out)] 09:29:03 Yes C is dead sexy. 09:30:40 -!- ejs1 [n=eugen@92.49.232.3] has quit ["This computer has gone to sleep"] 09:31:23 (defun copy-memory (src dst size) (loop repeat size do (poke dst (peek src)) (incf src) (incf dst))) 09:31:35 Now define PEEK and POKE. 09:33:06 matimago: That's not half bad :) 09:33:53 (defun peek (address) (dotimes (address) (setf address (incf (car address))))) 09:33:56 omg PEEK and POKE.... 09:34:20 the memories :) 09:34:29 brings back memories of reading various books on BASIC... 09:34:46 good thing I never ended using it :) 09:34:54 aah 09:35:15 Beket [n=Beket@athedsl-430690.home.otenet.gr] has joined #lisp 09:35:16 Quadrescence: (defvar *memory* (make-array *memory-size* :element-type '(unsigned-byte 8) :initial-element 0)) (defun peek (address) (aref *memory* address)) (defun poke (address byte) (setf (aref *memory* address) byte)) 09:35:31 oh my god 09:36:12 Well no :) 09:36:21 They're pointers to C-space ;) 09:36:36 nostoi [n=nostoi@122.Red-79-146-64.staticIP.rima-tde.net] has joined #lisp 09:36:53 now, wasn't there some assembler available, at least from inside SBCL? Corman also had interface to use direct assembly in lisp code... 09:38:36 hoh. 09:38:43 Have a look at SAPs. 09:38:49 I think I should be able to use MEM-REF from CFFI anyway. 09:38:57 They do most of the horrible things you're talking about. 09:39:05 :) 09:39:18 again, there's sb-kernel:system-area-ub*-{copy,fill}. 09:39:33 pkhuong: But .. the portability :) 09:39:34 schme: implementation details. I'd debug the algorithm in lisp first. Then you can substitute your peek and poke... 09:39:38 pkhuong: IF you're using sbcl. 09:39:43 matimago: he is. 09:40:03 matimago: debug what algorithm though? I just want to move stuff from foo to bar. 09:40:25 I'll just call a debugged and tuned implementation instead of rolling my own. 09:41:05 Well there's the smart way and my way ;) 09:41:28 Seriously though, I already have this running just swell, and it is just a toy app to see if stuff works. 09:41:52 Well, it depends on if you are planning to operate on lisp objects or not, I guess. 09:42:07 I'm not. 09:42:57 *foo = somewhere; *bar=somwhere else; memcpy(foo,bar,lots); that's all. You crazy lispers :) 09:43:18 -!- sohum [n=sohum@114.73.29.15] has quit [Read error: 54 (Connection reset by peer)] 09:43:22 I think that quote I saw in another channel just now applies. Asking #lisp for help is like trying to drink from a fire hose 09:43:35 Raynes [n=Raynes@ACA2AD25.ipt.aol.com] has joined #lisp 09:44:35 ia [n=ia@89.169.165.188] has joined #lisp 09:44:58 schme: Just stick your tongue in very very slowly. 09:45:13 arf. how I wish there was s-exp syntax for Python-the-language. 09:45:29 schme: actually, in ecl, you can just write (c-line "memcpy(foo,bar,lots);") 09:45:33 or something like that. 09:45:33 tic: Can't you do that sort of madness in the cl python? 09:45:40 -!- bob_f [n=bob_f@mail.phgroup.com] has quit [Nick collision from services.] 09:45:47 wedgeV [n=wedge@cm56-238-229.liwest.at] has joined #lisp 09:46:03 bob_f [n=bob_f@mail.phgroup.com] has joined #lisp 09:46:18 Here's a funny one. sizeof(float) ? 09:46:24 4? 09:46:31 Well I dunno. :) 09:48:01 -!- mrsolo [n=mrsolo@adsl-68-126-200-32.dsl.pltn13.pacbell.net] has quit ["Leaving"] 09:48:11 schme, I need it the other way around for compatibility. A pure-Python "Common Lisp" that'd compile down to Python code. 09:48:35 I see that the system-area-ub's are reverse of memcpy. That is good. 09:48:47 tic: I see. 09:49:17 schme, mostly a 1:1 translation, save for some gensymming required due to those thigns called statements. 09:49:47 gnnh, must...do...actual...work...now 09:50:09 tic: You need some good coding music and to close the irc :) 09:51:54 -!- tiesje` [n=user@202.63.242.211] has quit ["ERC Version 5.2 (IRC client for Emacs)"] 09:52:52 schme, any recommendations? Oldfield makes me sleepy. 09:52:59 np: Paco De Lucia - Patio Custodio (Buleria) 09:53:40 You just need to get really into it and imagine paco's fingers on the guitar == your fingers on the keyboard. 09:53:50 (: 09:55:51 -!- crod [n=cmell@cb8a16-213.dynamic.tiki.ne.jp] has quit [Read error: 110 (Connection timed out)] 09:55:59 crod [n=cmell@cb8ac8-056.dynamic.tiki.ne.jp] has joined #lisp 09:56:36 -!- locklace [i=locklace@gateway/tor/x-faa77e0cd2acb1e8] has quit [Remote closed the connection] 09:57:09 locklace [i=locklace@gateway/tor/x-e64465b25000c3d6] has joined #lisp 09:57:18 tic: if oldfield makes you sleepy, I won't recommand Klaus Schultze then... 09:57:34 Try Kraftwerk or Tangerine Dreams. 09:58:06 Otherwise, Mozarts, Bach, Bethoven, Schubert, even Shostakivitch. 09:58:12 s/ts/t/ 09:58:31 Oh, Tangerine Dreams! I listened to that yesterday. I should do that again. Thanks for the hint on Klaus Schultze, I'll have to look that up. 09:58:45 quad|iphone [n=quad@c-71-193-68-220.hsd1.mn.comcast.net] has joined #lisp 09:59:12 tcr [n=tcr@host145.natpool.mwn.de] has joined #lisp 10:00:02 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 10:00:13 jewel [n=jewel@dsl-242-170-119.telkomadsl.co.za] has joined #lisp 10:00:32 Zhivago: would you ever consider writing an essay on your thoughts or philosophy of programming? 10:02:23 vy [n=user@213.139.194.186] has joined #lisp 10:04:59 neomage [n=anon@c-98-216-105-88.hsd1.ma.comcast.net] has joined #lisp 10:05:09 MrSpec [n=NoOne@82.177.125.6] has joined #lisp 10:05:15 -!- neomage [n=anon@c-98-216-105-88.hsd1.ma.comcast.net] has quit [Client Quit] 10:05:29 neomage [n=anon@c-98-216-105-88.hsd1.ma.comcast.net] has joined #lisp 10:12:27 Now how on earth do I get my SLIME to indent TAGBODY nicely? 10:13:06 Get on an iphone. :o 10:13:24 ? 10:13:33 schme: Try the slime-indent contrib 10:13:46 *schme* looks. 10:13:51 I don't know if it changes the indentation of TAGBODY if it does, I'd be interested in knowing that 10:14:02 Ya lemme just figure out how to load it here :) 10:14:27 Or is this crazy indentation of it just there to prevent people from using it? hrrrr 10:14:45 H4ns [n=hans@p57A0FFE4.dip.t-dialin.net] has joined #lisp 10:15:33 -!- O_4 [n=souchan@125-236-190-161.broadband-telecom.global-gateway.net.nz] has quit ["Come alive!"] 10:15:38 I don't use TAGBODY often, but I regularly use PROG 10:16:19 oh. 10:16:25 slime-indentation is the same? 10:17:18 right 10:18:22 naah. Not doing it for me. 10:18:28 -!- ManateeLazyCat [n=user@222.212.133.86] has quit [Read error: 104 (Connection reset by peer)] 10:20:09 -!- H4ns [n=hans@p57A0FFE4.dip.t-dialin.net] has quit ["Leaving."] 10:21:02 H4ns [n=hans@p57A0FFE4.dip.t-dialin.net] has joined #lisp 10:22:36 -!- lambda [n=lambda@c-24-5-198-113.hsd1.ca.comcast.net] has quit [] 10:22:52 lambda [n=lambda@c-24-5-198-113.hsd1.ca.comcast.net] has joined #lisp 10:23:09 malumalu [n=malu@hnvr-4dbb4cd1.pool.einsundeins.de] has joined #lisp 10:23:21 appletizer [i=a@82-32-123-68.cable.ubr04.hawk.blueyonder.co.uk] has joined #lisp 10:24:01 ManateeLazyCat [n=user@222.212.133.86] has joined #lisp 10:24:47 -!- nostoi [n=nostoi@122.Red-79-146-64.staticIP.rima-tde.net] has quit ["Verlassend"] 10:24:55 -!- quad|iphone [n=quad@c-71-193-68-220.hsd1.mn.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 10:27:09 brill [n=brill@0x57386060.hrnqu2.dynamic.dsl.tele.dk] has joined #lisp 10:29:09 kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 10:33:21 What CFFI magic is there for free() ? 10:40:07 -!- appletizer [i=a@82-32-123-68.cable.ubr04.hawk.blueyonder.co.uk] has quit [Remote closed the connection] 10:41:22 -!- crod [n=cmell@cb8ac8-056.dynamic.tiki.ne.jp] has quit [Read error: 110 (Connection timed out)] 10:41:41 crod [n=cmell@cb8ac8-056.dynamic.tiki.ne.jp] has joined #lisp 10:43:14 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 10:45:44 sbahra [n=sbahra@77.31.29.117] has joined #lisp 10:48:31 -!- cracki [n=cracki@sglty.kawo2.RWTH-Aachen.DE] has quit ["The funniest things in my life are truth and absurdity."] 10:49:27 -!- prip [n=_prip@host202-132-dynamic.43-79-r.retail.telecomitalia.it] has quit [Read error: 110 (Connection timed out)] 10:49:52 prip [n=_prip@host51-134-dynamic.42-79-r.retail.telecomitalia.it] has joined #lisp 10:54:10 -!- wedgeV [n=wedge@cm56-238-229.liwest.at] has quit [] 10:54:33 sunkencity_ [n=sunkenci@h121n2c1o1036.bredband.skanova.com] has joined #lisp 10:57:25 hefner: here? 11:03:26 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 11:04:58 -!- tiesje [n=user@202.63.242.211] has quit [Read error: 113 (No route to host)] 11:06:12 -!- segv_ is now known as segv 11:07:27 robyonrails [n=roby@host181-178-dynamic.57-82-r.retail.telecomitalia.it] has joined #lisp 11:08:24 -!- netaust1n [n=austinsm@68-191-51-69.dhcp.nwtn.ct.charter.com] has quit [] 11:12:13 matley [n=matley@matley.imati.cnr.it] has joined #lisp 11:12:29 -!- MrSpec is now known as spec[away] 11:14:07 -!- robyonrails [n=roby@host181-178-dynamic.57-82-r.retail.telecomitalia.it] has quit [Remote closed the connection] 11:15:18 -!- kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 11:16:02 _magnus_ [n=magnus@83.209.81.249] has joined #lisp 11:17:54 poor sbcl dropping off to ldb 11:20:12 -!- bob_f [n=bob_f@mail.phgroup.com] has quit ["moving to my new desk"] 11:23:44 abeaumont [n=abeaumon@170.pool85-49-170.dynamic.orange.es] has joined #lisp 11:27:28 Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has joined #lisp 11:27:35 I have a format function that should print "&" only if the next argument to format is a non-nil list. This should be easy, but I can't find out how to do it... any ideas (I tried ~^& first, but that clearly doesn't work) 11:27:39 chris2 [n=chris@p5B16B5F7.dip0.t-ipconnect.de] has joined #lisp 11:28:44 madnificent: that can't be done in general. Can you assume that the argument is a list of some kind? 11:29:07 I may assume that the argument is existant and that it is either nil, or a linked list 11:29:17 * fatal error encountered in SBCL pid 19710(tid 3054828432): 11:29:18 GC invariant lost, file "gencgc.c", line 834 11:29:23 the argument is processed right after Krystof 11:29:30 Is this related to my ffi'ing ? :) 11:29:49 madnificent: then try ~@[&~] 11:29:52 Krystof: I could put the condition outside format and write two format calls for it, but it felt dirty 11:31:04 Krystof: well, that consumes the argument, and it should actually leave it alone, since it is consumed right after 11:31:46 ~:* 11:35:15 Krystof: I can't get that to work, what is it supposed to do? (I couldn't *find* the info on format-strings in the clhs either) 11:35:33 it goes back one argument in the format argument list 11:35:37 clhs ~* 11:35:37 http://www.lispworks.com/reference/HyperSpec/Body/22_cga.htm 11:35:51 Krystof: it seems to go back through all arguments 11:36:00 or use the boolean form of ~[ 11:36:17 Krystof: I don't think he can do the "non-nil _list_" predication via FORMAT. 11:36:29 vy: you know what? I actually said that 11:36:32 I can get it running with that data, I think :) 11:36:55 Krystof: Oops! Sorry. 11:37:03 -!- bombshelter13 [n=bombshel@209-161-235-98.dsl.look.ca] has quit ["Where is the glory in complying with demands?"] 11:37:38 <_3b> couldn't you fit the original spec by erroring on non-nil atoms? 11:38:54 ~:* behaves badly under sbcl... it seems to backup all arguments 11:40:31 madnificent: Does this do the work: (format t "FOO: ~A, ~:[~;&~:*~S~], BAR: ~A~%" 'foo '(1 2) 'bar) (Try to replace '(1 2) with nil.) 11:40:35 (format nil "~A~@[?~]~1@*~{~A=~A~^&~}" "foo" (:bar "baz"} does work however... 11:41:04 vy: it works the way Krystof suggested, but I think there might be a fault in sbcl 11:44:02 "If it is true, then the argument is not used up by the ~[ command but remains as the next one to be processed, and the one clause consequent is processed." 11:44:21 perhaps there is a fault in your reading comprehension 11:44:50 ironChicken [n=richard@158.223.51.84] has joined #lisp 11:45:49 Krystof: no, ~:* should backup *one* argument, but it seems to backup to the beginning of the format 11:46:29 Krystof: nvm, must be me 11:46:33 -!- benny [n=benny@i577A0953.versanet.de] has quit [Read error: 104 (Connection reset by peer)] 11:47:10 madnificent: ~:* is working correctly. ~@[ does not do what you think it does 11:47:44 binarycodes [n=sujoy@59.93.242.21] has joined #lisp 11:48:39 benny [n=benny@i577A0953.versanet.de] has joined #lisp 11:49:07 -!- binarycodes [n=sujoy@59.93.242.21] has left #lisp 11:49:13 Krystof: apparantly 11:49:33 apparently even 11:51:21 minion: memo for hefner: I think this (http://article.gmane.org/gmane.lisp.slime.devel/8144) is one of the major annoyance you have with Slime. If that's the case, could I ask you to reply and say so? 11:51:21 Remembered. I'll tell hefner when he/she/it next speaks. 11:52:43 <_3b> tcr: which buffers is that referring to? 11:55:43 shmho [n=user@58.142.15.103] has joined #lisp 11:58:22 -!- ace4016 [i=ace4016@cpe-76-168-248-118.socal.res.rr.com] has quit ["night"] 11:59:10 -!- Maddas [n=Maddas@tardis-b23.ee.ethz.ch] has quit [Remote closed the connection] 11:59:51 envi^home [n=envi@220.121.234.156] has joined #lisp 12:01:44 ThomasI [n=thomas@unaffiliated/thomasi] has joined #lisp 12:01:48 bob_f [n=bob_f@mail.phgroup.com] has joined #lisp 12:04:20 -!- simplechat [n=simplech@unaffiliated/simplechat] has quit [Remote closed the connection] 12:05:41 robyonrails [n=roby@host181-178-dynamic.57-82-r.retail.telecomitalia.it] has joined #lisp 12:10:44 -!- jsoft [n=Administ@unaffiliated/jsoft] has quit [Remote closed the connection] 12:11:47 -!- robyonrails [n=roby@host181-178-dynamic.57-82-r.retail.telecomitalia.it] has quit ["raise RuntimeError"] 12:16:48 silenius [n=jl@yian-ho03.nir.cronon.net] has joined #lisp 12:29:37 -!- kidd [n=kidd@80.31.143.150] has quit ["WeeChat 0.2.6"] 12:35:08 -!- zilt [n=user@74.210.19.218] has quit [Read error: 60 (Operation timed out)] 12:35:24 r2q2 [n=user@c-24-7-212-60.hsd1.il.comcast.net] has joined #lisp 12:37:21 -!- ManateeLazyCat [n=user@222.212.133.86] has quit [Read error: 60 (Operation timed out)] 12:38:19 Oh my. This time I even got a segfault and not the ldb :) 12:38:45 Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has joined #lisp 12:39:19 -!- Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has quit [] 12:41:22 -!- yoshinori [n=read_eva@router1.gpy1.ms246.net] has left #lisp 12:41:45 -!- Jasko [n=tjasko@c-98-235-21-22.hsd1.pa.comcast.net] has quit ["Leaving"] 12:42:43 good afternoon. 12:43:31 -!- lambda [n=lambda@c-24-5-198-113.hsd1.ca.comcast.net] has quit [] 12:44:08 tic: Good morning 12:46:01 what are you up to? 12:46:33 yoshinori [n=read_eva@router1.gpy1.ms246.net] has joined #lisp 12:46:36 eh waking up. 12:47:23 Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has joined #lisp 12:47:39 wedgeV [n=wedge@cm56-238-229.liwest.at] has joined #lisp 12:48:32 _3b: Basically each buffer popping up on a slime command. 12:48:58 _3b: But the compilation notes buffer in particular because compilation is not momentaneous 12:49:18 <_3b> tcr: would that make it pop up even when there were no notes? 12:50:01 Yow! 12:50:10 Yes 12:50:57 <_3b> hmm, not sure if that would be more or less annoying than the current behavior :) 12:51:20 eevar2 [n=jalla@195.1.147.242] has joined #lisp 12:51:24 *_3b* doesn't like it currently, so that isn't a vote against changing it though 12:51:26 It could kill itself if there were no notes 12:51:30 bartiosze [i=bartiosz@nintendos.pl] has joined #lisp 12:51:31 sohail [n=Sohail@unaffiliated/sohail] has joined #lisp 12:51:34 -!- yoshinori [n=read_eva@router1.gpy1.ms246.net] has left #lisp 12:52:04 -!- wlr [n=walt@c-65-96-92-150.hsd1.ma.comcast.net] has quit [Remote closed the connection] 12:52:05 jollygood [n=jollygoo@129.71.215.161] has joined #lisp 12:52:15 -!- Beket [n=Beket@athedsl-430690.home.otenet.gr] has quit [] 12:52:40 -!- Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has quit [Remote closed the connection] 12:57:24 <_3b> not sure if killing itself would be good or not, since then it would run into the window configuration problem 12:58:04 <_3b> probably less likely to be changing window config during a compile than other situations though, so might not matter 12:59:29 I can see it being annoying for C-c C-c. 12:59:38 which is probably fast anyway 13:00:46 -!- r2q2 is now known as Dr_Adventure 13:00:58 -!- Dr_Adventure is now known as r2q2 13:04:08 wlr [n=walt@c-65-96-92-150.hsd1.ma.comcast.net] has joined #lisp 13:04:26 <_8david`> is it prohibitively hard to make parts of a buffer behave as if they were in different modes, so that keybindings depend on the line the point is in? 13:04:58 you ask because of docstrings? 13:04:59 ejs [n=eugen@92-49-227-196.dynamic.peoplenet.ua] has joined #lisp 13:05:54 <_3b> mmm-mode? 13:06:22 <_8david`> no, this was more in reply to the whole "opening of buffers" thing. Personally I'd much prefer having the debugger and inspector show up directly in the REPL buffer (while offering the abbreviated stack trace and full interactivity the slime debugger has). 13:06:35 <_8david`> kind of like an accepting-values dialog in a CLIM listener 13:07:14 <_8david`> _3b: ah, that sounds interesting 13:07:28 z0d: How are your subjects? 13:07:43 tic: What are you up to in the afternoon? 13:07:52 Jasko [n=tjasko@209.74.44.225] has joined #lisp 13:07:55 r2q2: What do you mean? 13:08:10 z0d: you aren't general zod are you? 13:08:15 z0d: Nevermind bad joke. 13:08:54 I am thinking about using common lisp acl2. 13:09:01 -!- fooquux [n=fooquux@udp265832uds.hawaiiantel.net] has quit [Read error: 110 (Connection timed out)] 13:09:07 I was wondering if there was an scheme embedded in acl2 that I could use. 13:09:23 r2q2, some Python hacking, actually. Wondering if I can easily invent an s-exp syntax for Python. 13:09:56 I believe that in the case of general zod they were minions. 13:10:20 Oh yea. 13:10:36 sorry to disappoint you, but my name doesn't come from general zod <-: 13:10:39 http://script.hu/zodsnow.jpg 13:13:36 _8david`: Why would you prefer that? 13:16:03 cavelife_ [n=cavelife@211.201.172.41] has joined #lisp 13:16:44 beach [n=user@58.186.166.9] has joined #lisp 13:16:50 Good evening. 13:17:11 -!- wedgeV [n=wedge@cm56-238-229.liwest.at] has quit [] 13:17:26 <_8david`> for one thing, I want the contents of the buffers to be preserved in the scrollback of the REPL. If a debugger popped up because of an error in the REPL, the debugger interaction belongs to the workflow of the REPL conceptually. When I scroll back to that point, I also want to see the stacktrace. 13:17:33 <_8david`> The same goes for compilation notes. 13:18:00 Nshag [i=user@Mix-Orleans-105-1-86.w193-250.abo.wanadoo.fr] has joined #lisp 13:18:07 *_3b* tends to not use the repl that much while working 13:19:01 <_8david`> People try to paste things from SLIME all the time, and instead of pasting (a) compilation results (b) the eventual error or warning (c) the stack trace, and in particular, all of those in the right order (!), they paste a weird combination of the above from different buffers, making it impossible to figure out what error they actually got when. 13:21:23 <_8david`> And as a personal preference, I find buffers and windows extremely annoying when they pop up for any reason other than because I asked for it with a keystroke. Your proposal would lessen the impact of that a little, but I would much prefer the problem of multiple buffers to not arise in the first place. 13:21:31 The drawback that I see with such a scheme that it's awkward to look at stuff simultaneously 13:22:16 I'm thinking of different scheme, make buffers never pop up, but introduce a keybinding which pop ups the buffer created by the last command 13:22:41 Graghh [n=anon@93-97-116-19.zone5.bethere.co.uk] has joined #lisp 13:23:01 <_8david`> In ELI you just open multiple REPLs if you want to do several things at once. While SLIME has many more features now than ELI ever had, it's workflow was way better and more robust than SLIME. 13:23:08 <_8david`> argh, *its 13:23:12 ManateeLazyCat [n=user@222.212.133.86] has joined #lisp 13:23:40 Slime supports multiple listeners since a few days 13:23:49 robyonrails [n=roby@host181-178-dynamic.57-82-r.retail.telecomitalia.it] has joined #lisp 13:25:27 syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has joined #lisp 13:25:31 -!- alpheus` [n=user@vpn.cashnetusa.com] has quit [Remote closed the connection] 13:28:35 kidd [n=kidd@80.31.143.150] has joined #lisp 13:28:49 Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has joined #lisp 13:29:08 <_8david`> okay, I'll stop gloating now. Perhaps I need to start embracing SLIME, i.e. start hacking it into submission myself. 13:33:35 Anyone here play with acl2? 13:33:46 -!- cavelife_ is now known as cavelife 13:33:50 _8david`, then you can make a movie about how you stopped worrying and started .... 13:37:05 <_8david`> ... to stop worrying and love the SLIME? 13:37:54 -!- Darkness3477 [n=maveric@122-148-16-200.static.dsl.dodo.com.au] has quit [Read error: 110 (Connection timed out)] 13:38:20 Darkness3477 [n=maveric@122-148-16-226.static.dsl.dodo.com.au] has joined #lisp 13:40:06 _8david`, exactly. 13:40:07 -!- abeaumont [n=abeaumon@170.pool85-49-170.dynamic.orange.es] has quit [Read error: 104 (Connection reset by peer)] 13:40:47 abeaumont [n=abeaumon@170.pool85-49-170.dynamic.orange.es] has joined #lisp 13:42:11 slash_ [n=Unknown@p4FF0A9E7.dip.t-dialin.net] has joined #lisp 13:43:25 -!- robyonrails [n=roby@host181-178-dynamic.57-82-r.retail.telecomitalia.it] has quit ["raise RuntimeError"] 13:43:36 hkBst [n=hkBst@gentoo/developer/hkbst] has joined #lisp 13:44:08 -!- abeaumont [n=abeaumon@170.pool85-49-170.dynamic.orange.es] has quit [Read error: 104 (Connection reset by peer)] 13:44:49 abeaumont [n=abeaumon@170.pool85-49-170.dynamic.orange.es] has joined #lisp 13:46:02 Hey party people. Any ideas on debugging ffi madness, with the outside world firing off callbacks in sbcl land? 13:47:49 wormilwork [n=Miranda@adsl-070-155-056-058.sip.bct.bellsouth.net] has joined #lisp 13:48:48 -!- AntiSpamMeta [n=MetaBot@unaffiliated/afterdeath/bot/antispambot] has quit [Read error: 104 (Connection reset by peer)] 13:49:33 AntiSpamMeta [n=MetaBot@unaffiliated/afterdeath/bot/antispambot] has joined #lisp 13:50:57 -!- moesenle [n=moesenle@atradig124.informatik.tu-muenchen.de] has quit ["Leaving"] 13:51:13 -!- r2q2 [n=user@c-24-7-212-60.hsd1.il.comcast.net] has quit [Remote closed the connection] 14:07:14 ccl-logbot [n=ccl-logb@master.clozure.com] has joined #lisp 14:07:14 14:07:14 -!- names: ccl-logbot schoppenhauer kiuma AntiSpamMeta ignas dlowe malumalu_ drdo` moesenle wormilwork abeaumont hkBst slash_ Darkness3477 Athas kidd syamajala ManateeLazyCat Graghh Nshag beach cavelife Jasko ejs wlr jollygood sohail bartiosze eevar2 Odin- silenius bob_f ThomasI envi^home shmho benny ironChicken chris2 _magnus_ matley kejsaren sunkencity_ prip sbahra crod brill malumalu H4ns neomage spec[away] vy jewel tcr locklace ia Raynes aerique antoni 14:07:14 -!- names: pchrist Phoodus nxt l_a_m_ pstickne toddoon_ dto arbscht mishok13 mikezor_ mega1 Draggor stassats` Aankhen`` cheatcountry holycow xinming bpt segv araujo kpreid nyef_ ramkrsna bryte younder b4|hraban ltbarcly drdo TDT gz _8david` thom_logn rdd bdowning dash__ ianmcorvidae retupmoca cads BrianRice mtd_ pon][ Khisanth ypsa ``Erik uriel09 manic12 Dextor daniel lemonodor anekos tsuru archangelpetro koning_r1bot chii turbo24prg Xof Guest53748 michaelw 14:07:14 -!- names: xristos _3b dublpaws PanGoat Thumper_ _dd Kirklander luis srcerer Atherton wgl dwave cmm olejorge1b maxote Balooga sledge pkhuong Ginei_Morioka rey_ cods slyrus l4ndfo scode mornfall p8m djinni` egn Riastradh matimago ehu` rotty hsaliak authentic |Soulman| lisppaste specbot sad0ur_ yango_ patmatch sjbach kuwabara andrewy Thas jsnell lucca tessier_ technik boyscared eirik johs fnordus Fade j_king lemoinem minion replaca phadthai deepfire nasloc__ 14:07:14 -!- names: simonb leadnose rumbleca dostoyevsky guenther1_ azuk e271 Quadrescence Dazhbog ASau bobrown`` mqt df_aldur cky Balooga_ erg tltstc maskd mvilleneuve keithr vsync jajcloz_ joast JuanDaugherty wol Bucciarati Patzy qebab abend Zhivago CrEddy Soulman__ Aisling Paraselene__ gaja PissedNumlock bohanlon akhilleus l_a_m mogunus pjb mcxx dfox_ housel weirdo hugod Tristam adeht Jaearess bit` xan kmkaplan Cel myrkraverk S11001001 madnificent Eno_ brickhazel 14:07:14 -!- names: yahooooo eliasm mathrick borism galdor oklopol ecraven dnm spiderbyte schme kleppari r0bby cddr wasabi__ fe[nl]ix mgr rlpowell albino billstclair V-ille @drewc cYmen tic yoonkn lexclose Tordek Martinp23 lnostdal bfein dcrawford djkthx z0d Chrononaut p_l emma tarbo hefner sepisultrun danderson joga Partyzant felipe rsynnott krappie meingbg kuhzoo vcgomes ahaas Wombat2 mr_uggla ineiros zbigniew thijso Adrinael Buganini clog aking pok herbieB 14:07:14 -!- names: @antifuchs larstobi 14:08:05 I don't quite understand how you mean with this whole new backend thing :) 14:10:00 -!- drdo [n=psykon@167.111.54.77.rev.vodafone.pt] has quit [Read error: 110 (Connection timed out)] 14:12:06 schme: _8david` says, that you should try eliminating the possibility that wrong data is sent from C to sbcl, which could corrupt sbcl 14:12:32 Oh. 14:12:42 -!- malumalu [n=malu@hnvr-4dbb4cd1.pool.einsundeins.de] has quit [Read error: 113 (No route to host)] 14:12:46 schme: thus, if you'd replace the backend, you could first see what will be included in your lisp-core before actually throwing it in. That might indeed put you forward 14:13:29 does anybody know what happened to Xach? 14:13:59 hmmm.. 14:14:02 rme [n=rme@pool-68-238-2-212.chi.dsl-w.verizon.net] has joined #lisp 14:14:05 I'm trying to wrap my head around this here. 14:14:21 Jabberwockey [n=jens@boo.dfn-cert.de] has joined #lisp 14:14:33 tomoyuki28jp [n=tomoyuki@w221062.ppp.asahi-net.or.jp] has joined #lisp 14:14:50 Is there a way to know what slots a certain class have? 14:15:46 clhs class-slots 14:15:47 Sorry, I couldn't find anything for class-slots. 14:15:56 tomoyuki28jp: CLASS-SLOTS is part of the mop 14:16:14 hi lispers 14:16:52 H4ns: seems like class-slots is undefined 14:16:53 madnificent, got my baby @ home just yesterday :) 14:17:09 tomoyuki28jp: use closer-mop 14:17:20 minion: tell tomoyuki28jp about closer-mop 14:17:21 tomoyuki28jp: please see closer-mop: Closer to MOP is a compatibility layer that rectifies many of the absent or incorrect MOP features as detected by MOP Feature Tests. http://www.cliki.net/closer-mop 14:17:53 H4ns: I will take a look at it. thanks! 14:19:04 r2q2 [n=user@c-24-7-212-60.hsd1.il.comcast.net] has joined #lisp 14:19:23 kiuma: she still sleeping long? 14:19:25 H4ns, I'm considering to write a cms with claw, is there a hort way to checkout all the exported symbols of a package that designate a subclass of claw-cms-template ? 14:19:45 madnificent, into the normality. 14:20:17 xinming_ [n=hyy@125.109.246.58] has joined #lisp 14:20:38 madnificent, I mean this night she started crying and made my wife to wakeup while I continued sleeping :P 14:21:04 kiuma: short way? there is class-direct-subclasses 14:21:14 -!- wormilwork [n=Miranda@adsl-070-155-056-058.sip.bct.bellsouth.net] has quit [Read error: 104 (Connection reset by peer)] 14:21:29 kiuma: good enough :) 14:21:47 dash [n=washort@ampere.divmod.com] has joined #lisp 14:22:09 madnificent, hehe 14:22:14 hi. anybody know if there's been recent work on garbage collecting tls indexes in sbcl? 14:22:20 H4ns, thx 14:22:38 i am trying to run some code that causes SBCL to crash with "Thread local storage exhausted". 14:22:45 H4ns, I can't lisp now, or my wife kills me 14:23:00 kiuma: i can't lisp for you, sorry 14:23:55 H4ns, lol, the suggestion was just enough 14:24:18 (this is on sbcl 1.0.23) 14:24:29 I don't think it is any wrong data being sent.. The function does not really get sent any data. It is just called, does nothing, returns.. and after it has been called 159 times or so sbcl goes down. 14:24:54 Hrrm. 14:25:19 kiuma: I created a micro-blog earlier to see how my mini-framework is going. I was rather pleased with the result of development. I am going to change some things though. I'll notify you about what I learned, perhaps claw could benefit from it some day 14:25:49 *schme* goes back to pokin' around 14:27:23 perhaps 14:28:31 lenst [n=user@90-227-134-8-no52.tbcn.telia.com] has joined #lisp 14:28:58 -!- slash_ [n=Unknown@p4FF0A9E7.dip.t-dialin.net] has quit [Client Quit] 14:30:38 LiamH [n=none@common-lisp.net] has joined #lisp 14:31:39 -!- schoppenhauer [n=css@unaffiliated/schoppenhauer] has quit [Remote closed the connection] 14:33:40 oh got it. My lisp is broken (: 14:34:52 omg911 [n=omg@CPE00402b407f68-CM0011aec50bd8.cpe.net.cable.rogers.com] has joined #lisp 14:35:11 schme: not much occasions in which that can be a relief, lucky you 14:35:51 -!- spec[away] is now known as mrSpec 14:36:16 pkhuong: do you have a patch for the "revisiting vm tricks" blog entry? 14:36:38 madnificent: the occasion is usually after you say "either my lisp is broken, or I am going totally insane" ;-) 14:37:23 dash: sadly, most of the times, the person in question will be going insane 14:38:10 madnificent: it happens 14:38:42 -!- xinming [n=hyy@125.109.244.103] has quit [Read error: 110 (Connection timed out)] 14:38:51 robyonrails [n=roby@host181-178-dynamic.57-82-r.retail.telecomitalia.it] has joined #lisp 14:38:58 I don't think I'm crazy to want SBCL to gc TLS indexes though :) 14:40:13 l8r 14:40:16 -!- kiuma [n=kiuma@81-208-106-75.ip.fastwebnet.it] has quit ["Bye bye ppl"] 14:40:34 -!- ejs [n=eugen@92-49-227-196.dynamic.peoplenet.ua] has quit [Read error: 110 (Connection timed out)] 14:40:58 willb [n=wibenton@wireless107.cs.wisc.edu] has joined #lisp 14:43:37 <_8david`> dash: http://article.gmane.org/gmane.lisp.steel-bank.devel/11244 has the only discussion of this that I'm aware of 14:45:18 -!- Darkness3477 [n=maveric@122-148-16-226.static.dsl.dodo.com.au] has quit ["Ex-Chat"] 14:46:34 yeah 14:46:45 people talk about it, but nothing has been done AFAICT =| 14:46:49 -!- cavelife [n=cavelife@211.201.172.41] has quit ["Ex-Chat"] 14:46:55 there was a mention of it in august as well 14:48:47 -!- abeaumont [n=abeaumon@170.pool85-49-170.dynamic.orange.es] has quit [Read error: 104 (Connection reset by peer)] 14:49:45 zimbu668 [n=zimbu668@pool-71-248-17-189.bltmmd.east.verizon.net] has joined #lisp 14:50:32 -!- mrSpec [n=NoOne@82.177.125.6] has left #lisp 14:51:06 What are TLS index? 14:51:32 luis: I think that change is unnecessary, ECL is going to have :unix on *features* for NetBSD in future. 14:52:15 brickhazel_ [n=brickhaz@63.144.132.79] has joined #lisp 14:52:51 4096 dynamic bindings ought to be enough for everyone 14:52:55 ASau: it'd be nice to export it (and :bsd as well) for older versions anyway. 14:53:17 luis: I don't think it is that necessary. 14:53:37 luis: ECL is in beta stage AFAIUI. 14:54:23 ASau: call it what you want but there are older versions being used. 14:54:35 Which ones? 14:54:49 Maxima can run only with 0.9l. 14:55:03 FriCAS requires post-0.9l. 14:55:52 weirdo: yeah, I wish. 14:55:53 In 3 months we will deprecate 8.12.0 (I hope), and all's fine. 14:55:54 -!- kpreid [n=kpreid@cpe-67-242-4-64.twcny.res.rr.com] has quit [] 14:56:08 dash__: true enough, I guess running out of tls indices is fairly easy to avoid, usually. 14:56:20 kpreid [n=kpreid@cpe-67-242-4-64.twcny.res.rr.com] has joined #lisp 14:56:27 ... and the fix would make specials slower. 14:56:34 there's also the option of running a non-threaded build 14:56:53 weirdo: that's probably what i'm going to do 14:57:08 mega1: preferring speed over correctness? that sounds like what a C programmer would do ;) 14:57:47 not really, I'm only describing why it's low priority. 14:57:59 sure 14:58:00 SBCL-related question: are there any plans to support bootstrapping with ECL or CLISP? 14:58:18 dash__: but patches are welcome 14:58:20 mega1: it just goes against the grain of the attitudes of many lisp users. :) 14:58:39 mega1: Understood. I looked at this a while back and got lost pretty rapidly 14:58:55 I'm probably just going to use a unithread build for the moment 14:58:56 gentle leg pulling doesn't work, though. Sponsorship on the other hand ... :-) 14:59:00 dash: isn't it always an optimisation function of which the best result is influenced by the extremeness of the preference someone has for a specific feature (feature being speed versus correctness here) 14:59:34 sellout [n=greg@guest-fw.dc4.itasoftware.com] has joined #lisp 14:59:37 mega1: oh, well. not trying that as a motivational technique really, that's more of a reflex for me :) 15:00:11 madnificent: Mostly I'm reflecting on the stereotype C-programmer mindset of optimizing by sacrificing correctness for speed. :) 15:00:20 *dash* has been writing a lot of stuff in C lately. :) 15:01:27 my condolences dash ;) 15:01:39 -!- vy [n=user@213.139.194.186] has quit [Remote closed the connection] 15:02:45 salex [n=user@216.80.143.240] has joined #lisp 15:03:11 antgreen [n=green@CPE0014bf0b631a-CM001e6b1858fa.cpe.net.cable.rogers.com] has joined #lisp 15:05:41 abeaumont [n=abeaumon@170.pool85-49-170.dynamic.orange.es] has joined #lisp 15:11:06 ASau: CLISP works sometimes 15:12:53 dash, you could emulate dynamic bindings by, say, making them symbol-macros 15:13:07 pointing to a thread-local alist of bindings 15:13:27 though that breaks SYMBOL-VALUE 15:13:31 weirdo: well, i'm running a program that has this problem, not writing it, unfortunately 15:13:47 unithread is probably my best best for now 15:16:16 madnificent: Do I get any condolences? 15:16:46 -!- galdor [n=galdor@bur91-2-82-231-160-213.fbx.proxad.net] has left #lisp 15:17:34 luis: last time it worked for me was 1.0.9 or so. 15:17:38 -!- dash [n=washort@ampere.divmod.com] has left #lisp 15:19:12 nyef__ [n=nyef@pool-70-20-58-237.man.east.verizon.net] has joined #lisp 15:19:36 -!- nyef_ [n=nyef@pool-70-20-51-23.man.east.verizon.net] has quit [Read error: 101 (Network is unreachable)] 15:19:41 Quadrescence: what is your problem? 15:19:55 Quadrescence: ^ that was not agressive 15:19:55 I've been writing in C too to implement this lisp. :D 15:20:51 ah certainly then! You're even working on something I may like :D 15:21:00 -!- rme [n=rme@pool-68-238-2-212.chi.dsl-w.verizon.net] has left #lisp 15:21:16 how is it going? 15:22:02 sh10151 [n=user@132-174-113-48.ip.oclc.org] has joined #lisp 15:22:14 madnificent: Very well. :) 15:23:06 I remember someone here was developing a lisp library for reading the wiimote but I forget the nick. Is he around? ;) 15:24:31 joga: could you send me a message (minion) if yo get that running? I seem to remember that that existed for DrScheme (or something like that) 15:24:53 joga: allso, do we have a bluetooth library in general? 15:25:52 ejs [n=eugen@77-109-28-160.dynamic.peoplenet.ua] has joined #lisp 15:25:58 -!- benny [n=benny@i577A0953.versanet.de] has quit [Read error: 113 (No route to host)] 15:26:23 madnificent, I'm not sure how the library worked and if it was using foreign interfaces, but when I find out who was coding it I can ask :) 15:27:29 cracki [n=cracki@47-252.eduroam.RWTH-Aachen.DE] has joined #lisp 15:28:04 -!- brill [n=brill@0x57386060.hrnqu2.dynamic.dsl.tele.dk] has quit ["Lost terminal"] 15:28:12 blitz_ [n=julian@77.64.176.217] has joined #lisp 15:30:16 -!- ejs [n=eugen@77-109-28-160.dynamic.peoplenet.ua] has quit [Client Quit] 15:34:45 joga: I'm not the one you're looking for but I've made some ad-hoc bindings to OIS (http://sourceforge.net/projects/wgois/) in the past which is said to support the wiimote as well. 15:38:28 luis: ** - Continuable Error 15:38:28 luis: EVAL: undefined function HOST-CLOAD-STEM 15:38:47 Like in almost all previous versions up to 1.0.9 or so. 15:39:13 slyrus_ [n=slyrus@207.189.195.44] has joined #lisp 15:40:16 aerique: wiimote and lirc, looks handy 15:43:02 <_8david`> ASau: you can help with this! A good start would be to scroll up to find the actual error or warning causing the problem. 15:43:43 fixing whatever that problem is won't make it reliably build with clisp 15:44:12 _8david`: I can paste the whole build log. %] 15:44:29 I need some at least vague understanding what's going on. 15:44:46 -!- nyef__ is now known as nyef 15:45:15 if figuring out why the cold core is non-deterministically invalid was as easy as scrolling through the build log, it would've been done years ago :-) 15:45:17 ismith [n=chatzill@ip67-152-34-218.z34-152-67.customer.algx.net] has joined #lisp 15:46:02 milanj [n=milan@93.86.112.129] has joined #lisp 15:47:30 drdo`` [n=psykon@167.111.54.77.rev.vodafone.pt] has joined #lisp 15:49:01 <_8david`> I'm hoping for nyef's core file parser to make that investigation easy. 15:49:06 <_3b> joga: i have some win32 code for wiimote, don't remember how useable it is though 15:49:17 I can upload the whole build tree as well. 15:50:11 Umm... I have a question. 15:50:29 Are the -fasls- the core is built from supposed to be deterministic? 15:50:34 _8david`: weren't you and xophe doing something on making fasl build determinism? that seems like a better debugging approach than poking in the core 15:50:39 (That is, modulo timestamps and gensym names.) 15:51:04 <_8david`> nyef: Xof has a patch for "more deterministic" cold fasls. (Possible not entirely deterministic yet.) 15:51:16 What's nondeterministic about them? 15:52:13 -!- matimago [n=user@cac94-10-88-170-236-224.fbx.proxad.net] has quit [Read error: 104 (Connection reset by peer)] 15:52:37 Or should I just do two consecutive builds in different trees and compare them myself to see? 15:52:43 <_8david`> jsnell: yes, but I got distracted. Personally I will need to compare core files anyway to investigate relocation issues, not just fasls. 15:53:12 wait, that mail says sbcl doesn't GC TLS indices? 15:53:36 so binding a single gensym to a dynamic value in a loop will crash it after 4096 - something iterations? 15:53:37 matimago [n=user@cac94-10-88-170-236-224.fbx.proxad.net] has joined #lisp 15:53:52 weirdo: SBCL also doesn't dynamically allocate more TLS space if it runs out of slots, can't vary the stack sizes for a new thread, etc. 15:53:56 <_8david`> nyef: I could forward Xof's patch to you, assuming he doesn't mind 15:54:00 For functions that only do side-effects, does CL generally return T or NIL? 15:54:09 _8david`: I'm more just curious at this point. 15:54:46 Quadrescence: (values), effectively NIL. 15:55:05 Ogedei [n=user@78.52.233.130] has joined #lisp 15:55:15 sellout: (values)? 15:55:34 (defun foo () (print "side effect") (values)) 15:55:47 <_8david`> nyef: the patch does s/gensym/make-symbol/, s/format-universal-time/machine-instance/, and nulls out debug-source-created/-compiled 15:56:23 Sounds like it effectively ruins some debugability, though. 15:56:41 -!- lemonodor [n=lemonodo@adsl-76-214-16-250.dsl.lsan03.sbcglobal.net] has quit [] 15:56:50 -!- _magnus_ [n=magnus@83.209.81.249] has quit ["Leaving"] 15:56:58 IIRC the differences between sbcl and clisp host-produced xc fasls are a lot bigger than can be accounted for that, though 15:57:09 If you have any patches to try, don't hesitate. %] 15:57:30 -!- myrkraverk [n=johann@unaffiliated/myrkraverk] has quit [Read error: 60 (Operation timed out)] 15:57:33 like sbcl host producing a 80kB fasl while clisp host produces a 100kB one 15:57:33 (loop for foo = (gensym) do (eval `(let ((,foo 42)) (declare (special ,foo)) ,foo))) 15:57:38 that crashes for me after a while 15:57:49 pretty scary 15:57:57 -!- eevar2 [n=jalla@195.1.147.242] has quit ["This computer has gone to sleep"] 15:58:12 <_8david`> jsnell: sure, I think this is just to rule out the differences that are unrelated to clisp. 15:58:21 not allowing > 4096 dynamic bindings at the same time would be less scary 15:59:03 right 15:59:06 Quadrescence: a multiple-value return, but with 0 values. 15:59:22 mmmm 15:59:32 weirdo: You know, it probably wouldn't be too hard to think up a hack to GC TLS slots without having to patch SBCL to do so... But it'd be nasty. 15:59:43 greyface [n=greyface@188-144.1-85.cust.bluewin.ch] has joined #lisp 15:59:51 Quadrescence: stassats` gave an example. 16:00:15 stassats`: strictly speaking, (defun foo () (print "side effect") (values)) is side-effect-free. It's PRINT that has side effects... 16:00:38 rvirding [n=rvirding@h40n5c1o1034.bredband.skanova.com] has joined #lisp 16:00:52 -!- Graghh [n=anon@93-97-116-19.zone5.bethere.co.uk] has quit [Read error: 110 (Connection timed out)] 16:01:06 matimago: I'd say that's false. 16:01:12 matimago: but calling that function results in a side effect 16:01:16 -!- Jabberwockey [n=jens@boo.dfn-cert.de] has quit [Read error: 110 (Connection timed out)] 16:02:16 gigamonkey [n=user@adsl-99-50-126-237.dsl.pltn13.sbcglobal.net] has joined #lisp 16:02:18 matimago: what if print just calls princ internally ... is print then side-effect-free? 16:02:41 nyef, patching sbcl isn't a problem, it would be optimal to do it without inducing terrible slowness and getting the patch into the tree 16:02:51 but didn't someone say it's of low-priority? 16:03:18 Given the number of complaints about it, it certainly -seems- to be low-priority. 16:03:40 -!- drdo` [n=psykon@167.111.54.77.rev.vodafone.pt] has quit [Read error: 110 (Connection timed out)] 16:04:14 avdi [n=avdi@216.230.102.194] has joined #lisp 16:04:59 matimago: how is your foo side-effect free? by calling it, there is a side-effect. It simply doesn't matter what caused it (for in the end, you'll have to say that only machine-code can have side-effects) 16:05:24 nullwork [n=kyle@c-24-245-23-122.hsd1.mn.comcast.net] has joined #lisp 16:07:02 nyef, i'm interested because it affects my use-case. previously thought it simply disallows 4096 bindings at the same time... 16:07:53 -!- ramkrsna [n=ramkrsna@unaffiliated/ramkrsna] has quit [Remote closed the connection] 16:08:39 schoppenhauer [n=css@unaffiliated/schoppenhauer] has joined #lisp 16:08:56 -!- r2q2 [n=user@c-24-7-212-60.hsd1.il.comcast.net] has quit [Remote closed the connection] 16:09:27 rread [n=rread@c-98-234-51-39.hsd1.ca.comcast.net] has joined #lisp 16:09:44 -!- cracki [n=cracki@47-252.eduroam.RWTH-Aachen.DE] has quit ["The funniest things in my life are truth and absurdity."] 16:09:49 lemonodor [n=lemonodo@adsl-69-234-128-47.dsl.irvnca.pacbell.net] has joined #lisp 16:13:08 is it ok to have a macro without a body? or is it considered bad style/error in design? 16:13:58 <_3b> you mean one that doesn't return any code? or that just doesn't have a block of code passed into it as the 'body' ? 16:14:07 mcxx, depends on the use-case 16:14:31 _3b: the second one 16:14:58 <_3b> yeah, those seem reasonably common 16:15:47 <_3b> need more details to say anythinga bout a specific case though 16:15:53 ok then 16:15:59 -!- pstickne [n=pstickne@c-71-236-177-238.hsd1.wa.comcast.net] has quit [Read error: 110 (Connection timed out)] 16:16:20 -!- tomoyuki28jp [n=tomoyuki@w221062.ppp.asahi-net.or.jp] has quit [Remote closed the connection] 16:16:27 -!- lenst [n=user@90-227-134-8-no52.tbcn.telia.com] has quit [Remote closed the connection] 16:18:10 deximer [n=deximer@168.203.117.66] has joined #lisp 16:18:14 Well ok, I want a more discriminating terminology here. FOO doesn't by itself implement any side-effect, but indeed it calls functions that do. If you changed PRINT to make it side effect free, FOO would automatically become side effect free. 16:18:58 I'm not even sure PRINT has side effects. What if (prog1 (values) (with-output-to-string (*standard-output*) (foo))) ? 16:19:44 matimago: then can you make an example of a function with side-effects in your definition? 16:19:56 <_3b> matimago: does a function calling setf have side effects? 16:20:34 can you call macros? 16:20:36 Yes, we need to define the boundary to be able to define side effects. 16:20:56 mega1: My VM tricks tree is mirrored at . 16:21:07 Roo [n=anon@93-97-116-19.zone5.bethere.co.uk] has joined #lisp 16:21:11 eno [n=eno@nslu2-linux/eno] has joined #lisp 16:21:13 <_3b> in the w-o-t-s case, i'd say it does have side effects, you are just isolating them 16:21:20 _3b: i can make setf expansion which doesn't have SE 16:22:17 lhz [n=shrekz@c-db43e455.021-158-73746f34.cust.bredbandsbolaget.se] has joined #lisp 16:22:41 nyef: re TLS slots and GC, (: 16:24:26 <_3b> you also need to decide if you are asking about 'has side effects now' or 'can be safely assumed to have no side effects even if other code is modified' 16:24:46 rpg [n=rpg@216.243.156.16.real-time.com] has joined #lisp 16:24:52 -!- abeaumont [n=abeaumon@170.pool85-49-170.dynamic.orange.es] has quit [Read error: 110 (Connection timed out)] 16:25:02 beaumonta [n=abeaumon@120.pool85-54-87.dynamic.orange.es] has joined #lisp 16:25:09 -!- beaumonta is now known as abeaumont 16:25:45 pkhuong: I'm thinking that without safe points development is too hard, it's way too easy to add async signal unsafe code. 16:26:13 I found some examples in a quick review of x86 code. 16:26:27 matimago: and the boundary should clearly be: if a function has side-effects (which means anywhere << because that is the only thing the user of the function cares about, the implementation should not matter), then it is not side-effect-free 16:27:11 ikki [n=ikki@201.155.75.146] has joined #lisp 16:27:41 does funcall have side-effects? 16:28:40 -!- envi^home [n=envi@220.121.234.156] has quit ["Leaving"] 16:28:54 stassats`: depends on the function being called 16:28:58 In rucksack is there a 1:1 mapping between rucksack directories and classes or I can use a directory to store multiple classes? 16:29:15 mega1: there's only what was needed for a build and a couple tests and no more. Anything involving the debugger or IO usually makes it crash, iirc. 16:29:20 jeremiah [n=jeremiah@31.Red-213-98-123.staticIP.rima-tde.net] has joined #lisp 16:29:34 stassats`: nice one. I beleive the answer is yes. Because it may have side effects. (if (some-rare-condition) (have-side-effects) (no-side-effect)) should definitely be considered to have side effects. 16:30:23 why not work with three-value-logic then: yes/no/maybe 16:31:45 -!- aerique [i=euqirea@xs2.xs4all.nl] has quit ["..."] 16:32:13 runenes [n=runenes@proxy-gw.uib.no] has joined #lisp 16:32:19 madnificent: can you do anything with the difference between yes and maybe ? 16:33:35 <_3b> implementations can cheat for funcall since it is in the CL package, which might count as doing something with maybe 16:34:37 kuwabara: as a terminology, it would be more correct, thus it could make arguments more sound 16:34:47 kuwabara: only statistically .... 16:34:53 REPLeffect [n=REPLeffe@69.54.115.254] has joined #lisp 16:36:28 Tordek_ [n=tordek@host166.201-253-15.telecom.net.ar] has joined #lisp 16:36:33 -!- Tordek [n=tordek@host14.190-137-244.telecom.net.ar] has quit [Read error: 145 (Connection timed out)] 16:38:41 athos [n=philipp@92.250.204.223] has joined #lisp 16:40:11 jewel_ [n=jewel@dsl-242-132-215.telkomadsl.co.za] has joined #lisp 16:40:27 dreish [n=dreish@minus.dreish.org] has joined #lisp 16:41:29 -!- Tordek_ is now known as Tordek 16:41:32 seelenquell [n=seelenqu@tmo-105-101.customers.d1-online.com] has joined #lisp 16:42:04 |stern| [n=seelenqu@tmo-105-101.customers.d1-online.com] has joined #lisp 16:42:15 -!- |stern| [n=seelenqu@tmo-105-101.customers.d1-online.com] has quit [Read error: 104 (Connection reset by peer)] 16:42:19 davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has joined #lisp 16:43:03 <_3b> wonder if 'probably side-effect free' for an auto-paralellizing compiler... distribute the 'probably safe' code among cores, and block on side effects or something 16:43:11 <_3b> *could be useful 16:43:55 i can has know how many TLS dynamic bindings exist? 16:44:09 anuj [n=anuj@210.212.55.3] has joined #lisp 16:44:48 -!- anuj [n=anuj@210.212.55.3] has left #lisp 16:45:06 -!- fe[nl]ix [n=algidus@88-149-209-209.dynamic.ngi.it] has quit [Read error: 113 (No route to host)] 16:46:16 fe[nl]ix [n=algidus@88-149-208-150.dynamic.ngi.it] has joined #lisp 16:46:19 _3b: you also need to decide which functions to paralellize, because overhead of the paralellization can be more then calling that function sequentially 16:46:20 -!- Raynes [n=Raynes@ACA2AD25.ipt.aol.com] has quit [Read error: 60 (Operation timed out)] 16:46:38 Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has joined #lisp 16:46:39 -!- antoni [n=antoni@48.pool85-53-16.dynamic.orange.es] has quit [Client Quit] 16:46:55 "more than" 16:47:21 <_3b> stassats`: yeah, but trying things that might be pure instead of only the provably pure functional gives you more options 16:48:26 even haskell has some challenges with automatic paralellization 16:48:54 <_3b> not planning on trying it any time soon, just suggesting it as a use for the yes/no/maybe idea :) 16:49:48 _3b: this sounds related to a lot of recent MPI research. I don't know much about the grubby details, but it can be pretty subtle 16:51:39 (not quite the same, but the issues are similar) 16:52:18 -!- rvirding [n=rvirding@h40n5c1o1034.bredband.skanova.com] has left #lisp 16:55:10 -!- dreish [n=dreish@minus.dreish.org] has quit [] 16:56:41 -!- jewel [n=jewel@dsl-242-170-119.telkomadsl.co.za] has quit [Read error: 110 (Connection timed out)] 16:57:40 -!- lemonodor [n=lemonodo@adsl-69-234-128-47.dsl.irvnca.pacbell.net] has quit [] 16:57:51 Cronos [n=a@5acc7297.bb.sky.com] has joined #lisp 16:58:39 weirdo: a bit less than 4096 by default on SBCL. 16:58:45 -!- robyonrails [n=roby@host181-178-dynamic.57-82-r.retail.telecomitalia.it] has quit ["raise RuntimeError"] 16:59:16 -!- mtd_ is now known as mtd 16:59:21 _3b: there's some work in STM-style speculative parallelisation. Not extraordinary performance, but better than nothing when you don't want to rewrite everything. 16:59:27 guess i'll have to do something inelegant to allow O(1) variable access 17:00:18 asdas [n=Beket@athedsl-251830.home.otenet.gr] has joined #lisp 17:00:19 -!- yango_ is now known as yango 17:01:18 galdor [n=galdor@bur91-2-82-231-160-213.fbx.proxad.net] has joined #lisp 17:02:34 -!- schoppenhauer [n=css@unaffiliated/schoppenhauer] has quit [Remote closed the connection] 17:02:46 -!- zimbu668 [n=zimbu668@pool-71-248-17-189.bltmmd.east.verizon.net] has quit ["Quit"] 17:04:43 schoppenhauer [n=css@unaffiliated/schoppenhauer] has joined #lisp 17:06:06 mooglenorph [n=marco@173.9.7.10] has joined #lisp 17:06:16 -!- toddoon_ [n=guillaum@mar92-11-82-245-210-60.fbx.proxad.net] has quit [Client Quit] 17:09:22 -!- schoppenhauer [n=css@unaffiliated/schoppenhauer] has quit [Remote closed the connection] 17:09:38 benny [n=benny@i577A0953.versanet.de] has joined #lisp 17:09:52 -!- avdi [n=avdi@216.230.102.194] has quit ["Leaving."] 17:10:25 -!- nxt is now known as lasts 17:12:17 chaitanya [n=chaitany@122.162.15.109] has joined #lisp 17:12:36 Hi all. I'd like to comment whether this code http://paste.lisp.org/display/73121 abuses macros' concept or if it is legitimate. 17:12:44 s/like/like you 17:13:26 antares_ [n=antares_@77.108.193.227] has joined #lisp 17:14:15 sterne [n=user@65.211.14.131] has joined #lisp 17:14:28 asdas: i'd say it is 'legitimate', but not particularily stylish or idiomatic. 17:16:10 asdas: but as you pasted it, it is not clear why you use a macro and what you really want to do, so it is hard to judge or give advice. destructive operations on lists or trees are often not the best way of achieving things. 17:16:19 (and that is true even without macros) 17:16:21 H4ns: thanks you bothered to check & comment. Is there any other established methodology that enables a recursive operation to alter persistently some non-local data ? 17:16:26 Hm 17:17:29 Ok. 17:17:39 asdas: i often use a local function to do the recursive operation and expose a non-recursive api. 17:19:51 Would you please be so kind as to provide me with a minimal example, as I am new to Lisp ? If you are busy, no problem at all. 17:20:48 Hi all. I know nothing about relational databases. I want to fix that, and do it in lisp. Can people recommend some lisp software to practice in, and a couple good books? 17:21:00 http://www.gigamonkeys.com/book/ 17:21:05 asdas: i don't think i fully understand your question, so it is hard to come up with an answer or example code. 17:21:36 Cronos: i know that pcl is a good book, but i did not know that it teaches relational databases. 17:21:45 -!- ManateeLazyCat [n=user@222.212.133.86] has quit [Read error: 104 (Connection reset by peer)] 17:21:57 still on the first chapter 17:22:01 mooglenorph: you first need to learn something about relational databases before you can meaningfully apply that knowledge from lisp applications. 17:22:22 H4ns, can you point me at a language-agnostic book for that? 17:22:38 mooglenorph: no, sorry. 17:22:51 H4ns: I'd like to replace the "most-car" element of a nested list with some element e. It really is meaningless from a usefulness standpoint. I just want to understand how can be elegantly done, since functions are call-by-value. Anyway, don't mind. You've helped already. Thanks 17:22:54 H4ns: though, it has chapters about databases 17:23:15 cavelife [n=cavelife@211.201.172.41] has joined #lisp 17:23:54 asdas: in common lisp, the idiom is to pass the data structure to a function and return a copied or modified version. see, for example, REMOVE 17:24:10 -!- cavelife [n=cavelife@211.201.172.41] has quit [Client Quit] 17:24:25 cavelife [n=cavelife@211.201.172.41] has joined #lisp 17:24:27 That's too bad. I could use a recommendation for a language-specific book, then? I don't mind translating examples to lisp. 17:24:44 asdas: it is also common to not modify data structures in place. for what you want, i'd say a function that returned a freshly consed list with the "most car" element being replaced by something else would be idiomatic 17:25:16 -!- ecraven [n=nex@140.78.42.103] has quit ["bbl"] 17:25:49 mooglenorph: no, sorry again :) i acquired my knowledge in passing years ago, by reading database manuals basically. maybe pcl is not such a bad recommendation after all, for getting started 17:26:09 amnesiac [n=amnesiac@p3m/member/Amnesiac] has joined #lisp 17:26:14 H4ns: What can I say ? Highly appreciate your insights. Thanks :) 17:26:25 I've read PCL, and I think that the database chapters hand-roll basic columns and tables from CLOS. 17:26:34 asdas: call by value nothing to do with in-place modification of data structures. 17:26:41 I'd like to learn how to use SQL in a competent fashion, from inside lisp ;\ 17:27:23 mooglenorph: best have some understanding of SQL first, then 17:27:52 (mind you, 90% of web apps, including big ones like wordpress, seem to have been written without any understanding of SQL) 17:27:59 pkhuong: how can a function then change some data passed to it and have the modification be persistent ? 17:28:32 rsynnott, this is what I have heard, which is what makes me worried about the quality of SQL-learning resources 17:28:47 nah, that's not really the problem 17:29:07 mooglenorph: in many cases, it is not bad at all to write "incompetent sql" and treat the database as a row store 17:29:11 I think it's more a cultural problem; people get the idea, incorrectly, that writing webapps is very,very easy 17:29:30 writing webapps is horrible. 17:29:41 mooglenorph: but then, you may ask yourself if you really need a relational database or if you could as well just load all your data into main memory and operate on that. 17:30:00 if a site or app ever becomes big and successful, then either they hire someone who knows about databases, or, if it's open source, people who commit to mysql make snide comments on their blogs until things change 17:30:14 (that's how wordpress's SQL became less horrid) 17:30:35 mooglenorph: it is probably at least worth having some idea about transactions if you want to use a relational database 17:30:42 I mean, in the sense that it seems really hard. You need to deal with many layers of abstraction represented in about three or four different languages. 17:30:57 there are also non-relational options, like H4ns says, and also object databases like elephant 17:31:07 Dynetrekk [n=Dynetrek@193.216.225.9] has joined #lisp 17:31:08 asdas: even if you were passed references, you would only change the binding of the variable (or constant) in the caller's scope. Changing data and returning a different datum that you'll then use to modify a variable's binding are two different things. 17:31:17 rsynnott: isn't elephant an object relational mapping? 17:31:17 (also ORMs, which are somewhere in between, in that they abstract out some of the SQL stuff) 17:31:27 H4ns: nope 17:31:41 it can use postgres as a backend, but basically fakes btrees with it 17:31:44 Hmm, okay. Since I am unlikely to generate something big and successful, I think that I will just try and learn as much as possible, which is really the point. 17:31:48 simple question, is there a way to export in a defpackage all the functions created by a defstruct ? 17:32:01 rsynnott: ah, ok. so it abuses a relational database as a row store. fair enough. 17:32:13 I mean other than manually export each function 17:32:27 mooglenorph: if you want to learn common lisp, i'd recommend that you avoid sql :) 17:32:31 galdor: no. 17:32:32 indeed it does :) 17:32:42 H4ns: thank you 17:32:50 (the best-supported store is bdb, which isn't relationAL to begin with, of course 17:32:51 Oh, I already "know" CL. I've read PCL and generated some simple programs. 17:32:55 H4ns: that might be extreme 17:32:59 galdor: writing a defstruct* macro that does that automatically may be a cute exercise, not that i'd recommend using it too much. 17:33:04 there are things for which proper SQL does make a lot of sense 17:33:32 pkhuong: I don't fully understand, but thanks for the hint. I'll google for the details. 17:33:32 rsynnott: sure. sql is great for processing relational, tabular data 17:33:41 H4ns: just thought than there're slots which shouldn't been exported, so it's ok 17:33:44 but it shouldn't be the first thing you leap for, necessarily 17:33:45 yep 17:33:55 I'm sticking with lisp, seems there's a reason for everything 17:33:59 Mostly I munge XML around and deal with large things of my own data. 17:34:16 mooglenorph: I'd hardly call that 'knowing' CL. You know _of_ CL ;) 17:34:26 And I wrote a simple interperter. 17:34:27 mooglenorph: xml is tree structured, so sql would not help you much with processing it. 17:34:35 mooglenorph: the data store for this: http://bknr.net/html/home.html might be of interest 17:34:43 mooglenorph: you propably know that, i only mean to reassure you :) 17:34:48 H4ns: unless it be nightmarishh SQL Server 2008 stuff :) 17:35:51 drewc, oh, yeah, I'm very much a noob. But I like CL much more than python or java (the other two languages I know) 17:35:53 robyonrails [n=roby@host181-178-dynamic.57-82-r.retail.telecomitalia.it] has joined #lisp 17:35:54 rsynnott: there is all kinds of nightmare out there in the it world. everything i wrote above would be downright denied by a proper oracle zealot of course :) 17:36:29 jjong [n=user@203.246.179.177] has joined #lisp 17:36:29 *rsynnott* suspects that the current craze of magic XML-stuff-in-relaational-database is more marketing than anything else 17:36:32 mooglenorph: well .. you've got good taste, i'll give you that :) 17:37:18 rsynnott: *shudder* 17:38:08 Oh, and I wrote a thing that takes sexper descriptions of a role-playing game's skill trees to a png from s-dot. 17:38:30 THat is probably my favorite program. Very easy stuff. But I'm interested in doing more. 17:39:39 xinming [n=hyy@125.109.254.17] has joined #lisp 17:39:49 ah, we have another generating-large-graphs enthusiast, then? :) 17:39:49 mooglenorph: it takes time to learn CL, yet you can use it whilst not knowing everything there is about it. As for SQL, I found learning how databases work (meaning the way they store things on disc) and then learning the SQL-syntax as an 'extra' to be very effective. As for web-apps, lisp makes them easier to manage 17:40:07 mooglenorph: and give us images! :D 17:40:45 pitui [n=pitui@doh.research.att.com] has joined #lisp 17:41:59 madnificent, sure :-) 17:42:01 http://rapidshare.com/files/180088682/mayhem.pdf.html 17:42:03 minion: chant! 17:42:03 MORE PR0N 17:42:16 (that's the rulebook that it generates) 17:42:28 Anyone else using CVS Emacs? 17:42:29 (the tables and the graphs are generated in lisp) 17:42:43 decafbad [n=mehmet@85.104.94.135] has joined #lisp 17:43:08 -!- drdo`` [n=psykon@167.111.54.77.rev.vodafone.pt] has quit [Connection timed out] 17:43:09 (tables are LaTeX syntax from lisp (very easy) graphs use s-dot, they're both generated from the same sexper syntax) 17:43:17 Seems to slow things down like mad. 17:43:31 I'm in CVS emacs from about a month ago, and it is just as fast as normal. 17:43:47 mooglenorph: games builder? 17:44:42 xinming__ [n=hyy@125.109.241.37] has joined #lisp 17:44:46 rtra [n=rtra@unaffiliated/rtra] has joined #lisp 17:44:47 mooglenorph: Are you using the XFT backend? 17:45:01 (Not sure "backend" is the right word there, but can't remember the specific.) 17:45:05 Specifics, even. 17:45:13 the trees look good, very professional 17:46:24 Aankhen``, yes 17:46:33 madnificent, thanls :-) 17:46:45 It is my scary homebrew RPG. 17:47:00 mooglenorph: Strange, I wonder why it was so slow for me. 17:47:04 The magic system is actually a programming language based on the lambda calculus. 17:47:06 Anyway, I'm reverting to Emacs 22 for now. 17:47:15 Aankhen``, I have CVS from about a month or two ago though. 17:47:32 *Aankhen``* shrugs. 17:47:37 jso [n=user@151.159.200.8] has joined #lisp 17:47:50 Aankhen``, if you're willing to bother to get xft fonts you could roll back by a week or so until it is fast 17:49:17 mooglenorph: That's okay, I can live without 'em till it's officially released (I'm, perhaps incorrectly, assuming the slowdown will be noticed and fixed by then if it's not specific to my setup). 17:49:24 jlf` [n=user@unaffiliated/jlf] has joined #lisp 17:50:22 -!- xinming [n=hyy@125.109.254.17] has quit [Read error: 60 (Operation timed out)] 17:50:44 Aankhen``: Xft is indeed somewhat slower than the bitmap backend, but IMO it's not a big problem 17:52:14 I am probably going to set up a webpage for the next major revision of that RPG. Hope to do it in lisp. (hunchentoot/cl-who?) 17:52:39 The code that makes all that stuff will be open source, of course. As will the game content. Whee. 17:52:44 -!- xinming_ [n=hyy@125.109.246.58] has quit [Read error: 110 (Connection timed out)] 17:54:01 nl: It was enough of a problem here to affect my workflow (e.g. compilation of Lisp files took ages because printing the notes was so slow). 17:55:59 That was for fe[nl]ix, sorry. 17:58:07 Hmm, for some reason SLIME is spewing error information in the REPL buffer. 17:58:16 "11 nested errors", wow. 17:59:28 ac97c23e [n=robb@83-71-31-97-dynamic.b-ras1.srl.dublin.eircom.net] has joined #lisp 17:59:57 -!- Dynetrekk [n=Dynetrek@193.216.225.9] has quit [Read error: 110 (Connection timed out)] 18:00:21 mulander [n=opera@nat-4.interq.pl] has joined #lisp 18:03:28 Dynetrekk [n=Dynetrek@193.216.225.9] has joined #lisp 18:03:47 Aankhen``: do you have a recent slime? 18:09:58 -!- gigamonkey [n=user@adsl-99-50-126-237.dsl.pltn13.sbcglobal.net] has quit [Read error: 110 (Connection timed out)] 18:11:27 -!- mega1 [n=mega@53d82bc8.adsl.enternet.hu] has quit [Read error: 110 (Connection timed out)] 18:12:35 nested errors often points to a PRINT-OBJECT problem 18:12:55 Aankhen``: do you use ,l to load ASDF systems ? 18:13:20 fe[nl]ix: Yes. 18:13:25 rsynnott: Yup, from CVS. 18:15:11 Aankhen``: my solution to that problem was modifying swank::operate-on-system, to bind *compile-verbose* and *compile-print* to nil 18:15:22 -!- chaitanya [n=chaitany@122.162.15.109] has left #lisp 18:15:38 no more compiler output 18:15:50 -!- runenes [n=runenes@proxy-gw.uib.no] has quit [Remote closed the connection] 18:16:05 -!- Cronos [n=a@5acc7297.bb.sky.com] has quit [] 18:17:12 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 18:17:18 afternoon 18:18:14 *fusss* has to learn XPath and XSLT in the next day. whatever those maybe. 18:18:53 persi [n=user@242.sub-75-252-181.myvzw.com] has joined #lisp 18:19:52 i also have a blog login thing that uses openID, and i think something like that could confuse newbs and lead to them to severe phishing, imo 18:20:33 fusss: what for? 18:20:53 XPath and XSLT 18:21:12 madnificent: it is coming up allot in the web 2.0 crap i'm reading 18:21:21 Is it possible to have the keyword prefix printed ":" but no the package prefix? (write-to-string '(pack:foo :baz)) -> "FOO :BAZ" ? 18:21:25 fusss: xslt's great. just remember to avoid 18:22:22 blitz__ [n=julian@77.64.176.217] has joined #lisp 18:22:37 ironChicken: where is it implemented anyway? the browser? third party tools? 18:23:35 mozilla has a reasonable implementation. but mainly it's in libraries: libxslt (from gnome), Saxon (in Java), Xalan (in Java from Apache), etc. 18:23:57 i think there's even a cl one in cxml 18:24:24 $ ./clbuild list xslt 18:24:41 -!- matley [n=matley@matley.imati.cnr.it] has quit [Read error: 110 (Connection timed out)] 18:24:46 ultimately i want to create a full internationalized site; where content are submitted through my home-made CMS. XML, i heard, is good with i18n. 18:25:03 er 18:25:21 did you hear what in particular made XML good with i18n? 18:25:28 i see. if you want to transform on the server-side, then cocoon is kind of ok. 18:26:58 Krystof: you made me curious, tell me 18:27:17 Krystof: i have an application window (a page) that has several holes in it, and each hole is filled through ajax. sorta like a dashboard. i want to be able to plug locale-aware data in there, depending on browser language or user setting. but i wanna do it as standard and common way as possible. 18:27:34 madnificent: I don't think that was what Krystof meant. 18:27:57 -!- rtra [n=rtra@unaffiliated/rtra] has quit [Remote closed the connection] 18:27:57 Aankhen``: ahhh, I was waiting for some xml-bashing :( 18:28:06 i try to avoid making things only I can understand. i should be able to hand the database to some php weenie and have him create another interface for the app. 18:28:07 -!- persi [n=user@242.sub-75-252-181.myvzw.com] has left #lisp 18:28:25 -!- antares_ [n=antares_@77.108.193.227] has quit [Read error: 60 (Operation timed out)] 18:28:29 madnificent: It's not really XML bashing to just ask what it was you heard. :-) 18:28:31 fusss: the first thing you do: xml -> sexps 18:28:46 http://burningbird.net/writing/the-parable-of-the-languages/ 18:28:51 ShereKahn [n=ajourez@91.176.62.142] has joined #lisp 18:28:58 Aankhen``: no, but XML isn't very 'good' in general, so I was assuming something funny would come up :) 18:29:42 madnificent: Well, from what I've seen, XML has no effect one way or another on I18N. 18:29:59 -!- ironChicken [n=richard@158.223.51.84] has left #lisp 18:30:24 Aankhen``: doesn't it help you structure a document, and make a clear way to keep several versions of the same document? 18:30:39 -!- mooglenorph [n=marco@173.9.7.10] has quit [Read error: 110 (Connection timed out)] 18:30:42 -!- syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has quit [Remote closed the connection] 18:30:48 -!- silenius [n=jl@yian-ho03.nir.cronon.net] has quit [] 18:31:33 ehu [n=chatzill@82-170-33-173.ip.telfort.nl] has joined #lisp 18:31:38 i'm just trying to see if there is a clearer way than the lisp equivalent of {$LANG}/index.html or whatever 18:31:59 fusss: It's just a data interchange format. 18:32:00 mrsolo [n=mrsolo@nat/yahoo/x-ab6fd99243b57919] has joined #lisp 18:32:22 About the only effect it has is to (theoretically) make sure you can handle UTF-8. 18:32:40 fusss: How did you think it helps you structure a document and make a clear way to keep several versions of the same document? 18:33:09 so what exactly is profound about it? I don't remember 500 books and a thousand conferences not to mention a whole industry being made around XDR, etc. 18:33:31 After all, with XML, you create your own vocabulary; presumably with an alternative format you'd either create your own vocabulary in a similar way or else use an existing vocabulary which actually helps in I18N. 18:33:48 There is nothing profound about it. 18:34:02 -!- Dynetrekk [n=Dynetrek@193.216.225.9] has quit [Read error: 54 (Connection reset by peer)] 18:34:10 Aankhen``: because my database can export to XML. Every web tool i have can export to XML. Ajax has innerXML. At this point it's like avoiding ASCII, imo. 18:34:23 Dynetrekk [n=Dynetrek@193.216.225.9] has joined #lisp 18:34:31 I didn't say avoid it. Like I said, it's a data interchange format. There's nothing special about it. 18:34:46 <_8david`> In addition to XML taking care of i18n in the sense of encoding stuff at least, XSLT in particular has various l10n features to format numbers and do sorting correctly. 18:34:49 Well, there's something special about it in that you probably shouldn 18:34:53 -!- blitz_ [n=julian@77.64.176.217] has quit [Read error: 113 (No route to host)] 18:34:59 shouldn't roll your own format to do the same sorts of thing 18:35:09 _8david`: SEE! that's what i read and was under the impression of 18:35:44 <_8david`> Mind you, those l10n features are optional (because they don't mandate support for any language in particular) and Xuriella doesn't support anything except naive english language stuff. Other implementations have this support though. 18:35:59 froydnj [n=froydnj@gateway.codesourcery.com] has joined #lisp 18:36:15 sh10151: Sure, you should probably just use what exists. I was trying to answer fusss's question and my own. 18:36:16 antares_ [n=antares_@77.108.193.227] has joined #lisp 18:36:22 <_8david`> As for implementations, Xuriella is a full XSLT 1.0 implementation in CL. Tarball coming RSN, but it should already be usable. 18:36:25 -!- ShereKahn [n=ajourez@91.176.62.142] has quit ["Leaving"] 18:37:36 _8david`: mind if I link to Xuriella from cliki/XML? it's not there. 18:37:59 <_8david`> fusss: don't bother -- I'm going to package it up for asdf-install and set up the cliki page next weekend 18:38:11 _8david`: great! thanks :-) 18:38:13 Does CXML still require that one parse the document into one representation to use DOM methods, another to use STP, etc.? 18:38:28 -!- asdas [n=Beket@athedsl-251830.home.otenet.gr] has quit [] 18:39:10 <_8david`> Aankhen``: yes, although for read-only needs, you could use our CLOSy XPath protocol, which can walk both models. (And any other model you write generic functions for.) 18:39:32 _8david`: Ah, I see. 18:40:03 Davidbrcz [n=david@nsc.ciup.fr] has joined #lisp 18:41:55 -!- Roo [n=anon@93-97-116-19.zone5.bethere.co.uk] has quit [Read error: 60 (Operation timed out)] 18:43:41 antoni [n=antoni@48.pool85-53-26.dynamic.orange.es] has joined #lisp 18:44:56 Roo [n=anon@93-97-116-19.zone5.bethere.co.uk] has joined #lisp 18:45:03 do you know the name of the tool used to generate cffi and asdf manuals? 18:45:16 does it parse docstrings or is it just a texinfo document? 18:45:35 weirdo: the latter 18:45:51 the same for SBCL? 18:46:47 <_8david`> SBCL's docstrings.lisp parses docstrings. 18:47:20 <_8david`> luis has a portable version of that called texinfo-docstrings, which I'm currently working on to modularize it little. 18:47:35 thank you 18:47:50 already downloaded luis' code, but i don't know texinfo yet :) 18:48:09 i don't like tinaa, because it lumps everything together 18:48:20 edi weitz also have a simple doc making thing 18:48:23 hi guys! how can I check if a character is a number? along the lines of (is-char (x) (eql x "3")) 18:48:24 weirdo: C-h i then `m', `Texinfo RET' 18:48:36 (the one he uses for his own libs, it seems) 18:48:42 clhs digit-char-p 18:48:42 http://www.lispworks.com/reference/HyperSpec/Body/f_digi_1.htm 18:48:43 see for instance cl-containers, how tinaa makes it unreadable 18:48:46 Dynetrekk: See that. 18:48:55 Aankhen``: thanks! 18:49:19 Hurray, my brilliant idea worked. I altered my testing macro to group them into tests of 100 tokens each, so now it doesn't have monolithic STEFIL:DEFTEST forms to process. 18:49:23 <_8david`> weirdo: I think you'll find that most people here agree with you on tinaa. 18:50:00 weirdo: texinfo sucks... a little bit less than the alternatives 18:50:15 I like Scribe syntax 18:50:34 cracki [n=cracki@sglty.kawo2.RWTH-Aachen.DE] has joined #lisp 18:50:36 I wonder how complete Fare's Escribe is 18:50:36 sbcl's docstrings.lisp is great in that aspect, both extracting stuff from the image and not lumping everything together 18:50:54 luis, hmm, how about a markup language similar to gigamonkey's? 18:51:08 weirdo: I stole cffi's manual texinfo file to create the documentation for parse-declarations 18:51:15 http://www.gigamonkeys.com/lisp/markup/ 18:52:08 lemonodor [n=lemonodo@adsl-76-214-16-250.dsl.lsan03.sbcglobal.net] has joined #lisp 18:52:30 -!- sterne [n=user@65.211.14.131] has left #lisp 18:52:35 -!- H4ns [n=hans@p57A0FFE4.dip.t-dialin.net] has quit [Connection timed out] 18:53:24 People should realize that documentation means more than lumping docstrings together. Period. 18:54:25 tcr: fwiw I think Java has that problem a lot more than Lisp does 18:54:40 tcr: considering e.g. the CLISP and scsh manuals 18:55:48 tcr: that sort of documentatiion can be marginally useful, though 18:55:49 <_8david`> in Java, people tend to separate strictly between javadoc-generated "API documentation" and hand-written "tutorials". I think in the Lisp world, most people tend to prefer a mixture of the two, where per-function documentation is "picked" from the lisp image into the manual. 18:56:14 -!- benny [n=benny@i577A0953.versanet.de] has quit [Read error: 60 (Operation timed out)] 18:56:47 rtra [n=rtra@unaffiliated/rtra] has joined #lisp 18:56:53 sh10151, except that java has separate namespaces for classes 18:57:09 there it can probably make /some/ sense 18:57:15 <_8david`> Much good documentation in the Lisp world still separates the two parts of the manual though. E.g. CLHS, MOP, and Xach's documentation all come with a hand-written description followed by a "dictionary" of functions. 18:57:17 rindolf [n=shlomi@bzq-219-139-216.static.bezeqint.net] has joined #lisp 18:57:26 <_8david`> So I think that there a lots of useful styles between the two extremes. 18:57:46 could we agree on an acceptable markup? :) 19:00:04 <_8david`> markup in the sense of beautiful output or in the sense of the markup language in the source code? 19:00:54 -!- Dynetrekk [n=Dynetrek@193.216.225.9] has quit [] 19:01:27 well, there's always going to have to be hand-written documentation for anything non-trivial 19:01:29 weirdo: There has been some very recent discussion wrt documentation: http://thread.gmane.org/gmane.lisp.editors.hints.devel/17 19:01:34 some libs have a small BOOK 19:01:43 the syntax (i.e. the source) should be usable 19:02:13 weirdo: why? 19:02:41 -!- S11001001 [n=sirian@pdpc/supporter/active/S11001001] has quit ["bored now"] 19:02:41 I can't assume that an x86 with enough RAM to run SBCL has CMOV?! (c.f. Via C3) 19:04:34 I made up a backend subfeature for that many years ago 19:04:38 -!- jjong [n=user@203.246.179.177] has quit ["ERC Version 5.2 (IRC client for Emacs)"] 19:04:53 pkhuong: I still have a samuel2 laying around, please let it be blessed with sbcl :) 19:05:38 tcr, i think docstrings could be usable, when defining subpages to contain docstrings of selected symbols 19:05:54 -!- mulander [n=opera@nat-4.interq.pl] has quit [Remote closed the connection] 19:06:34 i'd rather avoid having to write documentation twice, for both PDF/HTML output and docstrings 19:06:55 pkhuong, that's 686, isn't it? 19:06:58 -!- milanj [n=milan@93.86.112.129] has quit ["This computer has gone to sleep"] 19:07:06 guess you can. no one uses 586 and lower nowadays 19:07:23 weirdo: nope, 686 doesn't imply CMOV :) 19:07:26 -!- leadnose [i=leadnose@xob.kapsi.fi] has left #lisp 19:07:44 weirdo: the intel implementations do have it, but it is specified as being optional 19:08:11 i feel like defining my own markup :| 19:08:41 in all fairness, even gcc had that one wrong (perhaps it still has) 19:09:05 weirdo: for markup only, tend to like textile (hint hint) 19:09:09 fabrizio [n=fabrizio@host129-251-dynamic.11-79-r.retail.telecomitalia.it] has joined #lisp 19:09:16 -!- fabrizio [n=fabrizio@host129-251-dynamic.11-79-r.retail.telecomitalia.it] has left #lisp 19:10:29 Krystof: yes, it's just a bit annoying having to add a check for that. 19:11:12 weirdo: you know there's always literate programming :) 19:11:24 doxtor [n=doxtor@unaffiliated/mitja] has joined #lisp 19:11:34 nah, it involves preprocessing lisp code 19:11:58 that's too invasive, especially in the context of lisp code 19:13:15 mulander [n=opera@nat-4.interq.pl] has joined #lisp 19:14:02 drdo [n=psykon@167.111.54.77.rev.vodafone.pt] has joined #lisp 19:14:17 hmm, couldn't you write some special syntax for it to avoid always preprocessing? 19:14:22 mindCrime [n=chatzill@74.213.159.129] has joined #lisp 19:14:34 docstrings but outside of any defun 19:15:15 -!- blitz__ [n=julian@77.64.176.217] has quit ["Lost terminal"] 19:15:55 i could write a reader macro that returns no values 19:16:06 -!- cads [n=max@c-71-56-62-166.hsd1.ga.comcast.net] has quit [Remote closed the connection] 19:17:24 weirdo: (with-doc (documentation-syntax (be here)) (defun blerg)) then let the macro store the markup somewhere so you can use it. Quick, clear, abstract 19:18:13 want to create something new about which you want to be able to write documentation, then define some way of abstracting data out of it. (with-documentation (blergh) (defwidget (foo) (bar baz))) 19:18:49 binarycodes [n=sujoy@59.93.242.21] has joined #lisp 19:19:01 -!- binarycodes [n=sujoy@59.93.242.21] has left #lisp 19:19:06 blitz_ [n=julian@77.64.176.217] has joined #lisp 19:20:25 madnificent, what about multiple files? it's not a documentation group could fit in one file 19:21:26 maybe assigning global identifiers with an option of including another one would make sense 19:21:27 dfox__ [n=dfox@r5cv134.net.upc.cz] has joined #lisp 19:22:33 puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has joined #lisp 19:23:52 mjf [n=mjf@r9fk63.net.upc.cz] has joined #lisp 19:24:16 weirdo: you could allow auxiliary documentation to reside in different files. Perhaps users can define where documentation comes from somewhat like with latex-sources. (weirdo.doc:build-documentation (markup containing frontpage) (introduction markup be here) (source "/home/weirdo/code/lisp/weirdo-doc/documentation.lisp" "/the/sources/will/be/searched/for/data/and/the/information/from/them/may/be/inserted.lisp") (more sources be here) 19:24:16 (some markup may be here)) 19:25:04 -!- dfox_ [n=dfox@r5cv134.net.upc.cz] has quit [Read error: 104 (Connection reset by peer)] 19:25:53 you could have a look at http://www.cliki.net/cl-typesetting as it *could* be helpful 19:32:14 madnificent, i'd rather name all nodes and allow xref/inclusion 19:33:00 but your approach violates POLA 19:33:06 pola? 19:33:13 principle of least astonishment 19:33:27 when will people be astonished? 19:33:44 when they meta-point my symbol and see the markup 19:33:59 then they'll wonder, "are these toplevel forms or now?" 19:34:05 "what is the code and what is the markup?" 19:34:13 s/now/not 19:34:36 I was assuming it to be rather clear. But fair enough 19:34:37 Dynetrekk [n=Dynetrek@193.216.225.9] has joined #lisp 19:35:13 I assumed it to be the most lispy approach and the closest to the code itself (which makes it easier to document the code itself 19:35:45 my approach will probably be defining nodes to be a list of (or symbol node) 19:36:11 where 'symbol' will be a symbol to slurp docstrings from 19:36:33 -!- ThomasI [n=thomas@unaffiliated/thomasi] has quit ["Bye Bye!"] 19:36:34 maybe i'll also allow freeform markup in nodes, so make it (or symbol node string) 19:37:43 maybe even specific stuff to be documented, for instance, 'class or 'function'... (or symbol node string (cons documentation-designator (cons symbol null))) 19:38:19 weirdo: Have you seen pbook.el? 19:38:38 i did not 19:39:13 Have a look, it's an interesting approach. 19:39:16 i'm not interested in including the source in the documentation 19:39:51 not for any proprietary-software reason, just taste 19:40:19 z` [n=tltstc@cpe-76-90-92-154.socal.res.rr.com] has joined #lisp 19:40:40 weirdo: Excluding the code would just mean commenting out two lines. 19:40:51 That was phrased poorly. 19:41:00 To exclude the code, you'd just have to comment out two lines in pbook.el. 19:41:53 Hmm, actually, you wouldn't even need to do that. Just replace the call to PBOOK-DO-CODE in PBOOK-FORMAT-BUFFER with a call to PBOOK-GOTO-END-OF-CODE. 19:42:20 so, did Xach quit IRC completely? 19:43:56 apparently he's still connected to freenode 19:45:45 -!- Davidbrcz [n=david@nsc.ciup.fr] has quit ["Ex-Chat"] 19:46:11 finis [n=mkr@dh207-78-33.xnet.hr] has joined #lisp 19:49:36 roby_ [n=roby@host234-206-dynamic.49-82-r.retail.telecomitalia.it] has joined #lisp 19:49:52 ManateeLazyCat [n=user@222.212.133.86] has joined #lisp 19:50:37 chinawhite [n=ben@asa01-ext.utc.backmo.com] has joined #lisp 19:53:28 -!- tltstc [n=tltstc@cpe-76-90-92-154.socal.res.rr.com] has quit [No route to host] 19:54:09 khisanth_ [n=Khisanth@pool-141-157-238-161.ny325.east.verizon.net] has joined #lisp 19:56:44 fisxoj [n=fisxoj@ool-4356ce50.dyn.optonline.net] has joined #lisp 20:01:07 fnpod [n=irchon@116.58.16.121] has joined #lisp 20:01:15 hi 20:02:02 fnpod: hello 20:02:03 hey fnpod 20:02:07 -!- amnesiac [n=amnesiac@p3m/member/Amnesiac] has quit ["Leaving"] 20:02:23 -!- chris2 [n=chris@p5B16B5F7.dip0.t-ipconnect.de] has quit ["Leaving"] 20:02:45 ircing from my wife's ipod touch 20:03:15 -!- robyonrails [n=roby@host181-178-dynamic.57-82-r.retail.telecomitalia.it] has quit [Connection timed out] 20:03:22 -!- madnificent [n=user@83.101.62.132] has quit [Remote closed the connection] 20:03:35 madnificent [n=user@83.101.62.132] has joined #lisp 20:04:00 custom app or something browser based? 20:05:14 ugh. sldb broke 20:05:24 free app: irchon. supports multiple servers, doesn't seem to support private messages 20:05:31 "wrong number of arguments: nil, 3" 20:06:36 oh, it's not sldb that broke, sbcl died on stack exhaustion 20:06:55 -!- ManateeLazyCat [n=user@222.212.133.86] has quit [Read error: 60 (Operation timed out)] 20:07:58 -!- antares_ [n=antares_@77.108.193.227] has quit [] 20:09:40 josemanuel [n=josemanu@85.1.222.87.dynamic.jazztel.es] has joined #lisp 20:11:11 milanj [n=milan@91.150.101.132] has joined #lisp 20:11:17 ok, this may seem like a dumb question (since i would expect the answer to be 'yes') but... you can create applications like, webservers/browsers etc. in LISP, right? 20:11:46 yes 20:11:52 -!- vsync [n=vsync@220-27.202-68.tampabay.res.rr.com] has quit [Read error: 113 (No route to host)] 20:12:27 in Lisp, too. ;) (LISP usually refers to the older family of s-exp based languages, whereas Lisp usually means Common Lisp) 20:12:41 but only AI applications composed entirely of lists 20:13:16 archangelpetro, have a look at http://cliki.net/web and http://common-lisp.net 20:13:24 aaah! you've got lisp in my web browser! 20:13:25 yes, lists are the only data type 20:13:36 ? 20:13:56 ok thanks, that was my next question 'is there an archytypal tome of Lisp wisdom, equivalent to the K&R book for C?' :D 20:14:05 archangelpetro, I really like Practical Common Lisp 20:14:12 minion, tell archangelpetro about that-dead-sexy-book 20:14:13 archangelpetro: have a look at that-dead-sexy-book: pcl-book: "Practical Common Lisp", an introduction to Common Lisp by Peter Seibel, available at http://www.gigamonkeys.com/book/ and in dead-tree form from Apress (as of 11 April 2005). 20:14:16 CLHS is the standard 20:14:23 minion: paip 20:14:23 paip: Paradigms of Artificial Intelligence Programming: Case Studies in Common Lisp by Peter Norvig. http://www.cliki.net/paip 20:14:38 it's quite readable for an ANSI standard, at that 20:14:40 archangelpetro: there really isn't a "K&R" for lisp, but those are good suggestions 20:14:43 and keene, and amop, and ... 20:14:57 stassats`, yes, throw the AMOP at him! It will make perfect sense! 20:15:01 stassats`, keene? 20:15:09 weirdo, Object-oriented programming in Common Lisp. 20:15:12 minion: tell weirdo about keene 20:15:13 weirdo: direct your attention towards keene: "Object-Oriented Programming in Common Lisp: A Programmer's Guide to CLOS" by Sonya E Keene. http://www.cliki.net/keene 20:15:17 -!- Khisanth [n=Khisanth@pool-141-157-251-219.ny325.east.verizon.net] has quit [Connection timed out] 20:15:25 -!- khisanth_ is now known as Khisanth 20:15:27 tic, anything AMOP doesn't mention? 20:15:45 weirdo, it has a CLOS tutorial, so yes, I think so. 20:15:48 archangelpetro, read PCL then AMOP and CLHS when you need an authoritative source 20:15:57 hyperspec.el helps 20:15:57 -!- decafbad [n=mehmet@85.104.94.135] has quit [Connection reset by peer] 20:16:24 -!- fnpod [n=irchon@116.58.16.121] has quit [] 20:16:25 (unless you're into meta-programming, you don't /really/ need to read AMOP) 20:16:34 and then write programs, write programs, and write programs 20:17:39 (on lisp is also good for macros) 20:18:18 -!- hugod [n=hugo@bas1-montreal50-1279439877.dsl.bell.ca] has quit [] 20:18:28 Hm. I think I actually have "all" Lisp books on my desk. Except for LiSP, but is that really about Lisp? 20:18:50 it is not about Common Lisp 20:18:57 -!- rtra [n=rtra@unaffiliated/rtra] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 20:19:03 Hm. Then I shall not require it. 20:19:36 but it is good, as far as i heard 20:20:43 -!- finis [n=mkr@dh207-78-33.xnet.hr] has quit ["..."] 20:20:52 -!- drdo [n=psykon@167.111.54.77.rev.vodafone.pt] has quit [Remote closed the connection] 20:21:49 Hm. Now, should I cheat on this problem and borrow PAIP's Prolog, or should I reinvent the wheel myself? The first option leads to me learning Prolog, the second to learn Lisp. :) 20:23:04 -!- Dynetrekk [n=Dynetrek@193.216.225.9] has quit [Read error: 110 (Connection timed out)] 20:23:14 hmm SUBTYPEP doesn't know about classes? 20:23:22 weirdo: sure. 20:23:24 tic: prolog isn't too hard to implement, I built a basic (didn't include everything) in it in the beginning of th year :) 20:23:30 pkhuong, bummer. 20:23:31 netaust1n [n=austinsm@160.79.78.72] has joined #lisp 20:23:43 weirdo: that is, sure it does. It even has to. 20:25:02 drdo [n=psykon@167.111.54.77.rev.vodafone.pt] has joined #lisp 20:25:06 pkhuong, bah. PEBKAC :) 20:25:26 divinebovine [n=new@mail.wjsgroup.com] has joined #lisp 20:25:53 canuck11716 [n=cliffr@66.183.147.55] has joined #lisp 20:26:09 madnificent, that's what I'm considering too. I just need backtracking and full (i.e. non-destructive) pattern matching. 20:26:19 madnificent, it's a substitution crypto w/ a fixed dictionary. 20:27:40 heh, was just using stumpwm 20:27:51 tic: I used a lisp-syntax for the socend one (which made pattern matching trivial) and I used restarts to do the backtracking :) 20:28:00 neat knowing you can hack away at a wm while its running 20:28:10 user_ [n=user@p54926436.dip.t-dialin.net] has joined #lisp 20:28:52 Jacob_H [n=jacob@92.5.76.16] has joined #lisp 20:30:38 madnificent, hm, that'd be a nice way to learn restarts! with manual intervention or just automatic? 20:32:49 jlouis [n=jlouis@port1394.ds1-vby.adsl.cybercity.dk] has joined #lisp 20:33:08 Dynetrekk [n=Dynetrek@193.216.225.9] has joined #lisp 20:33:16 automatic, but after a first result a manual restart with the request if more solutions needed to be found 20:33:28 okay. 20:34:03 Agiboo [n=pietjepr@d54C6AFEA.access.telenet.be] has joined #lisp 20:34:27 wolfboy22 [i=baumgold@tithonus.cs.brandeis.edu] has joined #lisp 20:34:40 tic: just for the record, that was actually very nice and fun to work it. I used it because I had no internet and needed to solve some prolog exercises for school (well, I needed the solutions) 20:34:41 what does it mean to have #P in front of a string? 20:34:58 -!- rdd [n=user@c83-250-142-219.bredband.comhem.se] has quit [Remote closed the connection] 20:35:31 wolfboy22: that it's really a pathname 20:35:36 -!- Roo [n=anon@93-97-116-19.zone5.bethere.co.uk] has quit [] 20:35:40 divinebovine, another thing to put on the to-do list 20:36:04 dlowe: is there a macro in sbcl I can use to turn a string into a pathname? 20:36:16 evenin' 20:36:31 -!- ignas [n=ignas@office.pov.lt] has quit [Read error: 113 (No route to host)] 20:36:42 wolfboy22: pathname 20:36:45 clhs pathnae 20:36:45 Sorry, I couldn't find anything for pathnae. 20:36:48 clhs pathname 20:36:48 http://www.lispworks.com/reference/HyperSpec/Body/a_pn.htm 20:37:10 thanks 20:38:40 -!- antoni [n=antoni@48.pool85-53-26.dynamic.orange.es] has quit [Read error: 148 (No route to host)] 20:41:37 madnificent, a reason as good as any! 20:41:46 but PATHNAME is a function, not a macro! 20:41:47 :-) 20:41:56 the correct answer is "no" :-))) 20:43:36 -!- madnificent [n=user@83.101.62.132] has quit [Remote closed the connection] 20:43:50 madnificent [n=user@83.101.62.132] has joined #lisp 20:44:36 gigamonkey [n=user@adsl-99-50-126-237.dsl.pltn13.sbcglobal.net] has joined #lisp 20:45:15 -!- mindCrime [n=chatzill@74.213.159.129] has quit [Read error: 104 (Connection reset by peer)] 20:46:18 -!- tsuru [n=user@c-69-245-36-64.hsd1.tn.comcast.net] has quit [Remote closed the connection] 20:46:32 silenius [n=jl@fuckup.club.berlin.ccc.de] has joined #lisp 20:48:35 fooquux [n=fooquux@udp265832uds.hawaiiantel.net] has joined #lisp 20:51:17 tsuru [n=user@c-69-245-36-64.hsd1.tn.comcast.net] has joined #lisp 20:52:34 -!- Nshag [i=user@Mix-Orleans-105-1-86.w193-250.abo.wanadoo.fr] has quit ["Quitte"] 20:53:07 tic: If you use PAIP's prolog, you can learn something about compiler optimizations for prolog, which is pretty cool. 20:53:39 mindCrime [n=chatzill@74.213.159.129] has joined #lisp 20:53:47 -!- Aankhen`` [n=Aankhen@122.162.169.62] has quit ["It's never too late to have a happy childhood"] 20:54:01 gottesmm [n=gottesmm@c007h037.acad.reed.edu] has joined #lisp 20:54:53 paip's prolog sucks 20:55:20 it simply checks one matching clause against another, even when they're simple symbols 20:57:09 alley_cat [i=AlleyCat@sourcemage/elder/alleycat] has joined #lisp 20:57:36 weirdo: I don't think paip's prolog was there to be some superb implementation. I believe it is there to show some features of lisp (and how te use them) 20:57:55 -!- Dynetrekk [n=Dynetrek@193.216.225.9] has quit [] 20:57:56 Is this #prolog now? 20:59:22 Davidbrcz [n=david@nsc.ciup.fr] has joined #lisp 20:59:24 weirdo: You mean it doesn't do clever hashing (like a good prolog compiler)? 20:59:27 amnesiac [n=amnesiac@p3m/member/Amnesiac] has joined #lisp 21:00:16 Someone really needs to implement a lisp in the paip prolog. 21:00:45 "sucks" isn't really helpful. 21:01:02 hfoo [n=h@p5B17CA2B.dip.t-dialin.net] has joined #lisp 21:01:27 Jacob_H_ [n=jacob@92.5.76.16] has joined #lisp 21:01:34 -!- Jacob_H_ [n=jacob@92.5.76.16] has quit [Read error: 104 (Connection reset by peer)] 21:01:40 -!- silenius [n=jl@fuckup.club.berlin.ccc.de] has quit ["Leaving"] 21:01:50 waterh [n=waterh@114.143.36.66] has joined #lisp 21:02:44 prip, AFAIK it doesn't 21:02:55 maybe paiprolog now does, but the one in paip.pdf 21:03:03 s/prip/rpg 21:05:25 There's a pdf now? 21:05:35 I seem to remember the PAIP one not being so prology :) 21:06:03 Like setof/3 actually was implemented in lisp and not in the prolog. 21:06:27 -!- sh10151 [n=user@132-174-113-48.ip.oclc.org] has quit [Remote closed the connection] 21:06:43 schme: that's how I'd expect an embeddable prolog to work. 21:07:33 pkhuong: Well I just meant it didn't quite strike me as a very "full" prolog where one could hack shit up in prolog :) 21:07:33 schme, there is, but only available illegally 21:07:49 weirdo: Well I don't need it. I'm just surprised someone has scanned the sucker. 21:07:52 i can't provide links on the channel due to the possible resulting use of banhammer 21:08:12 rpg, thatks for the hint. 21:08:14 code is formatted badly and there's even a missing page in the prologue 21:08:29 Not in my copy :P 21:08:53 code from paip is available from norvig's site 21:10:06 schoppenhauer [n=css@unaffiliated/schoppenhauer] has joined #lisp 21:10:17 sounds like someone complaining about the quality of his bootleg. 21:10:34 I like prolog.. it has that sexy code == data thing going for it :) 21:10:51 schme: So does machine code. :-P 21:11:04 rdd [n=user@c83-250-142-219.bredband.comhem.se] has joined #lisp 21:11:14 schme: Actually, I think Franz used Peter's code as the basis for the prolog they integrated with allegro cache. 21:11:14 Cronos [n=a@5ad01fe7.bb.sky.com] has joined #lisp 21:11:16 nyef: Hey that's cool. So it's lisp, prolog and machine code :) 21:11:52 rpg: Well that's great. :) I'm not saying it is all bad, I'm just saying it is not what I would chose if I personally wanted to do some prolog hacking :) 21:11:54 There is a small set of things for which Prolog is The Bomb. Alas, it is horrifically bad for system-building. 21:12:02 And then there's the metalinguistic possibilities inherent in Forth... 21:12:04 Heh, no shit. 21:12:31 Prolog in Lisp isn't very difficult to imagine. Would Forth in Lisp make much sense? 21:12:41 schme: You should ask some more active prolog folks. I haven't found a great one lately . SICStus is fabulous, but unless you are at a university, it's punishingly expensive. 21:12:50 tic: I've considered writing one. 21:12:58 *rpg* is possibly giving badly outdated information. 21:13:12 nyef, a bunch of symbols basically? 21:13:21 -!- Ogedei [n=user@78.52.233.130] has left #lisp 21:13:27 rpg: I'm fond of swi myself. But sadly enough it dun really compile everywhere so one needs to check out the others. And ya sicstus is great :P 21:13:35 I will now stop my prolog rant. 21:13:41 tic: Well, it depends on how involved you want to get. 21:13:51 But I blame you people. It is you who turned this into #prolog :) 21:14:00 nyef, I guess. And I'm also a bit confused as the last stack-based code I looked at was Factor, yesterday. 21:14:05 Jabberwockey [n=jens@dslb-082-083-103-177.pools.arcor-ip.net] has joined #lisp 21:15:27 Arf! Who sped up time for me? :( 21:16:43 -!- pon][ [n=pon@c-d531e253.516-1-64736c10.cust.bredbandsbolaget.se] has quit ["leaving"] 21:17:11 -!- jollygood [n=jollygoo@129.71.215.161] has quit [Read error: 110 (Connection timed out)] 21:18:23 -!- rindolf [n=shlomi@bzq-219-139-216.static.bezeqint.net] has quit ["Leaving"] 21:18:32 -!- gigamonkey [n=user@adsl-99-50-126-237.dsl.pltn13.sbcglobal.net] has quit ["ERC Version 5.2 (IRC client for Emacs)"] 21:18:56 ecraven [n=nex@140.78.42.103] has joined #lisp 21:19:27 decafbad [n=mehmet@78.168.59.196] has joined #lisp 21:19:31 -!- Agiboo [n=pietjepr@d54C6AFEA.access.telenet.be] has quit [] 21:19:50 anyone tried Lisa? 21:19:58 forward chaining closy goodness 21:20:39 you might need to keep the CLIPS and Jess manuals nearby though 21:20:40 *tic* only did Lisa Spaces. 21:20:46 oudeis [n=oudeis@89-138-15-229.bb.netvision.net.il] has joined #lisp 21:21:11 -!- araujo [n=araujo@gentoo/developer/araujo] has quit [Read error: 110 (Connection timed out)] 21:21:45 schme, that's why I come here. mostly interesting topics, even if OT 21:21:57 What!? 21:22:04 fusss: Lisa is sitting here right next to me :P 21:22:15 No! There is only one Lisa. 21:22:18 she's in Louvren. 21:22:20 antoni [n=antoni@208.pool85-53-19.dynamic.orange.es] has joined #lisp 21:22:28 Apple Lisa! 21:22:38 schme: good dad :-) 21:22:47 hfoo: Also there is a good quantity of low quality jokes here. 21:23:08 I will google Apple Lisa and see who that is. 21:23:23 schme: What. 21:23:27 schme: you serious? 21:23:29 vsync [n=vsync@220-27.202-68.tampabay.res.rr.com] has joined #lisp 21:23:32 schme: It was a machine. 21:23:43 fuck. 21:23:45 Well. 21:23:54 see back in those days I was a sinclaire man. 21:24:20 Nothing that fugly :) 21:24:33 Now I am going to bed here, so goodnight. 21:24:34 i was an apple IIg guy, in the late 90s :-P 21:24:35 schme: I'm not sure what relevance your petrol preference has. ;-) 21:24:35 -!- _8david` [n=user@port-83-236-3-235.dynamic.qsc.de] has quit [Remote closed the connection] 21:24:45 there is a lisa at my school, top floor :D 21:24:46 _8david` [n=user@port-83-236-3-235.dynamic.qsc.de] has joined #lisp 21:25:27 fusss: not late 80's? 21:25:49 late 80s i was busy making kites and fishing barefoot in africa :-) 21:27:49 araujo [n=araujo@gentoo/developer/araujo] has joined #lisp 21:27:53 fair enough. nothing wrong with retro :) 21:29:25 loz [n=loz@163.c.003.ncl.iprimus.net.au] has joined #lisp 21:32:16 nice 21:32:30 -!- nullwork [n=kyle@c-24-245-23-122.hsd1.mn.comcast.net] has quit ["Leaving"] 21:32:33 lisp does seem intruiging 21:33:08 archangelpetro: ahhh, welcome to the dreamworld of coding 21:33:28 archangelpetro: this is where computers are, when they're asleep. Nirvana of code 21:33:30 -!- ecraven [n=nex@140.78.42.103] has quit ["bbl"] 21:34:52 a path to enlightenment, eh? 21:35:38 -!- schoppenhauer [n=css@unaffiliated/schoppenhauer] has quit [Remote closed the connection] 21:36:05 it is the purest form of code I know, so I'd say yes. You are reading a book, I presume 21:39:54 wow, keyword arguments don't have to be keywords 21:40:04 too bad i never knew this 21:40:13 weirdo: eg? 21:40:26 ((lambda (&key ((foo foo))) foo) 'foo 42) 21:40:53 makes sense though :) 21:40:56 where the first 'foo' is the keyword name and second 'foo' is the variable name 21:41:31 it's great for CLOS, as a way to avoid keyword namespace pollution for MAKE-INSTANCE 21:42:27 madnificent: i'm reading practical common lisp 21:42:48 manuel_ [n=manuel@HSI-KBW-091-089-250-148.hsi2.kabel-badenwuerttemberg.de] has joined #lisp 21:43:03 though i'm having a hard time conceptualising how to program standard things, like.. a file reader etc. 21:43:15 but that's prolly cuz i'm not much into the first part of the book 21:43:58 -!- Jacob_H [n=jacob@92.5.76.16] has quit [Read error: 110 (Connection timed out)] 21:44:23 weirdo, didn't know that either, nice! 21:44:28 -!- slyrus_ [n=slyrus@207.189.195.44] has quit [Read error: 110 (Connection timed out)] 21:44:55 tic, did you learn from PCL like me? :) 21:45:07 Jacob_H [n=jacob@92.6.253.157] has joined #lisp 21:45:20 weirdo, yup! (and later Keene) 21:45:45 you guys are already doing well, having just learned from pcl 21:45:57 *archangelpetro* is a newbie :D 21:46:03 archangelpetro: altough the writing style doesn't hint it, the book is packed with data. It might take you slightly longer than initially planned, to read that book thorougly 21:46:09 *weirdo* AMOP instead of keene 21:46:29 AMOP is dry. hardcore of you, weirdo! 21:46:44 weirdo, it's not an xor situation, really. I felt Keene gave me much more feel for CLOS. AMOP is all about specs and implementation. 21:46:44 sonya keene seems to be female, judging by the name. how could i trust a programming book written by a female? 21:46:52 madnificent: ok :) 21:46:56 weirdo: wtf? 21:47:14 weirdo, good point. 21:47:25 keene and AMOP are very different books with very different intent 21:47:26 weirdo: huh? 21:47:31 tic: wtf? 21:47:42 weirdo: boobs! 21:47:55 creepville! 21:47:58 females are way more interested in shopping and fashion than "hard science" 21:48:04 salex, you know. they're more about cooking and such. 21:48:04 -!- waterh [n=waterh@114.143.36.66] has quit ["later"] 21:48:06 statistically, of course 21:48:14 yeah, and cooking 21:48:16 Uh, now weirdo is taking this too far. I was just being ironic. 21:48:18 then you don't understand statistics 21:48:23 weirdo: you trusted them when you needed feeding as a child, why not trust them now? 21:48:25 (wierdo) 21:48:56 :D 21:48:57 *tic* 's main interests are cooking and computing, guess that makes him a female.. 21:49:09 seriously, i didn't know keene's book existed by the time i read AMOP 21:49:15 weirdo, you should get it. 21:49:32 if nothing else to catch them all. 21:49:37 tic: does that mean i can somehow convince you to enrich my life thorugh gastronomy? 21:49:39 i have it lying on the disk somewhere, obviously downloaded illegally 21:50:14 archangelpetro, that I do not know. But I've been known to enrich people's lives by feeding them delicious food. Anyway, that's for #food. 21:50:18 -!- _8david` [n=user@port-83-236-3-235.dynamic.qsc.de] has quit [Remote closed the connection] 21:50:28 _8david` [n=user@port-83-236-3-235.dynamic.qsc.de] has joined #lisp 21:51:45 I thought this -was- #food? 21:53:02 no, this is #juvenille-fart-jokes-and-stale-beer-breath 21:54:07 wedgeV [n=wedge@cm56-238-229.liwest.at] has joined #lisp 21:55:48 :) 21:58:06 -!- Jasko [n=tjasko@209.74.44.225] has quit [Read error: 110 (Connection timed out)] 21:58:26 -!- LiamH [n=none@common-lisp.net] has quit ["Leaving."] 21:59:08 fusss, or the other way around 21:59:58 -!- user_ [n=user@p54926436.dip.t-dialin.net] has quit ["leaving"] 22:00:15 what do linkage-table has to do with gencgc ? 22:01:46 so if i (setf (find-class foo) ...), all newly created objects inherit from it if they class has that name somewhere in the inheritance chain? 22:01:52 do i need to FINALIZE-INHERITANCE too? 22:03:26 daniel_ [i=daniel@unaffiliated/daniel] has joined #lisp 22:03:46 -!- jewel_ [n=jewel@dsl-242-132-215.telkomadsl.co.za] has quit [Read error: 148 (No route to host)] 22:07:01 -!- seelenquell [n=seelenqu@tmo-105-101.customers.d1-online.com] has quit [Read error: 110 (Connection timed out)] 22:08:53 weirdo: make-instance is enough. 22:09:05 thank you! 22:12:05 -!- tessier_ is now known as all 22:12:14 -!- all is now known as tessier_ 22:13:56 -!- wedgeV [n=wedge@cm56-238-229.liwest.at] has quit [] 22:14:28 trittweil [n=rittweil@rayhalle1.informatik.tu-muenchen.de] has joined #lisp 22:14:39 mooglenorph [n=marco@173.9.7.10] has joined #lisp 22:17:44 repnop [n=Mage@adsl-69-225-13-63.dsl.skt2ca.pacbell.net] has joined #lisp 22:19:00 -!- daniel [i=daniel@unaffiliated/daniel] has quit [Read error: 110 (Connection timed out)] 22:19:50 -!- Davidbrcz [n=david@nsc.ciup.fr] has quit ["Ex-Chat"] 22:20:26 Jasko [n=tjasko@c-98-235-21-22.hsd1.pa.comcast.net] has joined #lisp 22:22:06 -!- dlowe [n=dlowe@ita4fw1.itasoftware.com] has quit ["*poof*"] 22:22:14 user_ [n=user@p549260E6.dip.t-dialin.net] has joined #lisp 22:22:23 -!- Patzy [n=somethin@coa29-1-88-174-11-170.fbx.proxad.net] has quit [Remote closed the connection] 22:22:30 -!- ac97c23e [n=robb@83-71-31-97-dynamic.b-ras1.srl.dublin.eircom.net] has quit [Remote closed the connection] 22:22:34 Patzy [n=somethin@coa29-1-88-174-11-170.fbx.proxad.net] has joined #lisp 22:23:55 trebor_h` [n=user@dslb-084-058-216-182.pools.arcor-ip.net] has joined #lisp 22:28:23 -!- Patzy [n=somethin@coa29-1-88-174-11-170.fbx.proxad.net] has quit ["leaving"] 22:28:39 Patzy [n=somethin@coa29-1-88-174-11-170.fbx.proxad.net] has joined #lisp 22:29:38 -!- younder [n=jpthing@062016236162.customer.alfanett.no] has left #lisp 22:30:49 syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has joined #lisp 22:31:46 pstickne [n=pstickne@c-71-236-177-238.hsd1.wa.comcast.net] has joined #lisp 22:32:18 -!- josemanuel [n=josemanu@85.1.222.87.dynamic.jazztel.es] has quit ["Saliendo"] 22:32:58 hugod [n=hugo@bas1-montreal50-1279439877.dsl.bell.ca] has joined #lisp 22:35:58 O_4 [n=souchan@125-236-190-161.broadband-telecom.global-gateway.net.nz] has joined #lisp 22:36:10 oudeis_ [n=oudeis@89-138-15-229.bb.netvision.net.il] has joined #lisp 22:37:34 now this is strange 22:38:17 i changed (find-class 'foo) and now (subtypep some-descendant-of-foo 22:38:26 'foo) returns nil 22:38:41 hmm maybe ENSURE-CLASS will do the trick 22:38:50 -!- lhz [n=shrekz@c-db43e455.021-158-73746f34.cust.bredbandsbolaget.se] has quit ["Leaving"] 22:39:18 -!- shmho [n=user@58.142.15.103] has quit [Remote closed the connection] 22:39:28 nyef: of curiosity, what is the sense of implementing Forth in Lisp? 22:39:39 -!- O_4 [n=souchan@125-236-190-161.broadband-telecom.global-gateway.net.nz] has quit [Client Quit] 22:39:53 nyef: I understand the other way arount, but not this one. 22:40:09 O_4 [n=souchan@125-236-190-161.broadband-telecom.global-gateway.net.nz] has joined #lisp 22:40:13 The only reasonable scenario I could come up with was as a cross-build environment, actually. 22:40:25 -!- mindCrime [n=chatzill@74.213.159.129] has quit [Read error: 110 (Connection timed out)] 22:40:53 Hm. 22:40:59 -!- Jacob_H [n=jacob@92.6.253.157] has quit [Read error: 110 (Connection timed out)] 22:41:05 Exploring optimization? 22:41:49 Mmm... Plausible, maybe. It'd certainly be easier to write an optimizing compiler in lisp. 22:42:47 fusss_ [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 22:42:49 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 22:42:53 -!- fusss_ is now known as fusss 22:43:20 -!- netaust1n [n=austinsm@160.79.78.72] has quit [] 22:43:21 a js or php syntax for CL would be sweet. not a vm, not an interpreter, just syntax. 22:43:43 Feel free to implement one. 22:44:01 fusss: there was m-expressions 22:44:05 and there's Dylan 22:44:10 something like that? 22:44:10 weirdo: There were at least two women heavily involved in the standardization of Common Lisp. 22:44:18 Just... be aware that M-expressions didn't catch on for a reason. 22:44:25 i plan to cut corners and export lisp api as xml-rpc services. let the web minions use it anyway they see fit. 22:44:26 -!- fooquux [n=fooquux@udp265832uds.hawaiiantel.net] has quit [Read error: 60 (Operation timed out)] 22:45:36 xml-rpc is the _easiest_ and cleanest way to call lisp, that i know of. and certainly most useful. 22:45:52 -!- puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has quit [Read error: 60 (Operation timed out)] 22:45:54 fusss: why would PHP syntax be sweet for CL? the power of Lisp lies in its syntax 22:46:52 weirdo pasted "what am i doing wrong?" at http://paste.lisp.org/display/73141 22:47:02 z0d: it comes handy to give to non lispers to make use of the lisp app functionality. but xml-rpc is just as good. 22:47:06 fusss: don't say these things, they scare me 22:47:23 fusss: the syntax of lisp is the least of their worries I believe 22:47:42 -!- Cronos [n=a@5ad01fe7.bb.sky.com] has quit [] 22:47:49 any clos hackers around? :| 22:47:56 grasping the use of multiple inheritance, mapping lists, using keywords and their namespaces, closures, lambda-functions in general... these are the hard parts 22:48:45 nearly all "reusable" software is in components; instantiate this objects, call these advertised methods, and interpret the results in this format. xml-rpc wraps up lisp really well, and you can let nearly everyone use it anyway they see fit. 22:48:56 madnificent: I wouldn't say that syntax is the least of their worries. I think it accounts heavily for Lisp not being accepted 22:49:12 z0d: not being accepted: yes, not being understood: no 22:49:39 madnificent: it's not being understood, because it's not accepted <-: 22:49:43 hmm how about a web framework with per-session widgets, each of which can receive messages through HTTP GET/POST? 22:49:45 fusss: when I explain how keywords work to java-people, they need some time to realise what it means. 22:49:47 puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has joined #lisp 22:49:59 -!- Xof [n=crhodes@dunstaple.doc.gold.ac.uk] has quit [Remote closed the connection] 22:50:11 z0d: no, it is not being understood, because it is not thought and not learned by many. 22:50:30 z0d: for some (and I'd say a big) part, that is because people don't use it 22:50:54 in my univ, it is fairly easy to spot who knows who, by the programming languages they've heard of 22:51:17 madnificent: i don't wanna explain anything to them. just wanna let them use the software from java; here, instantiate this wrapper class, call this method which sends your constructor arguments down the wire, and the result is XML. they don't have to understand fuckall, and i'm not in the business of handing out enlightenment. 22:51:41 none seem to know smalltalk, nor C#. No one uses C++ (except for the engineers, but they tend to separate themselves anyways) 22:52:25 c# is fairly popular on the market 22:52:28 I think the main part of getting people to know lisp (and use it), is by telling them about it... Not by making up another strange syntax around the problem 22:52:32 as are all .net languages 22:52:47 weirdo: yes, but I'm talking about how people get to know things... that's why c# was a good example ;) 22:52:48 weirdo: like F# or L# ;) 22:53:10 madnificent, what's needed is corporate acceptance 22:53:24 people learn languages when they can make money using them 22:53:33 weirdo: I don't seem to need corprotate acceptance for people to learn lisp 22:53:52 you don't, but masses do 22:53:57 a minor amount of enthousiasm is enough to get them started... 22:54:01 could someone please help me with my clos problem? http://paste.lisp.org/display/73141 22:54:02 :) 22:54:36 madnificent, the corporate value is programmer reusability, so they can fire them when they e.g. ask for a sick leave 22:54:39 no, corporate acceptance is like a friend everyone knows... good luck trying to motivate corporations to move towards lisp 22:55:00 madnificent, thus B&D languages which annoy the programmer with errors when they try to do anything interesting 22:55:22 it means the code has to be understandable for every drone 22:55:38 -!- salex [n=user@216.80.143.240] has quit [Remote closed the connection] 22:55:40 on top of it, I don't see how another strange syntax is going to motivate corporations 22:55:50 other solution might be starting our own startups and allowing hired programmers to use lisp (or only lisp) 22:56:16 weirdo: when I do work for someone, they ask me to explain some code if they're wondering what I'm building. So understanding the code is mainly a bogus argument 22:56:34 weirdo: Works for me. Try with a fresh image. 22:56:44 weirdo: they don't accept lisp though, that is true... but there is no sound reason as to why they don't 22:56:54 madnificent, that's what made me quit my job, they said no one in poland programs lisp (not true, but try convincing a moron employer) 22:57:33 madnificent, the goal is getting acceptance, not merely proving that current lack of acceptance is fallacious 22:58:07 ha, my employer folded soon thereafter i quit. good for me 22:58:11 -!- oudeis_ [n=oudeis@89-138-15-229.bb.netvision.net.il] has quit ["Leaving"] 22:58:18 weirdo: I've went to the exact same troubles as lisp is having, with rails... I don't see any reason why lisp can't pull the same trick 22:58:45 madnificent, ruby is popular with unix hobbyist-programmer crew 22:58:46 certainly not during these economically rough times, in which any company is looking for a competitive andvantage (and is willing to try) 22:58:51 -!- deximer [n=deximer@168.203.117.66] has left #lisp 22:59:00 ruby is alright 22:59:04 in the hobbyist crew, there exists a need for instand gratification which lisp can't provide 22:59:18 weirdo: but it wasn't they that pushed it forward. It was the style around it. A couple of success stories 22:59:21 weirdo: what? That's exactly why I use CL. 22:59:27 madnificent: if your place of work uses rails then they're pretty experimental. last place i worked we had to sneak GCC undercovers because the the shop didn't allow unauthorized C compilers. 22:59:30 archangelpetro: yes, I used to like it rather much 22:59:50 pkhuong, just reading a few pages of a perl manuals lets one write (fsvo) useful programs 22:59:51 -!- sellout [n=greg@guest-fw.dc4.itasoftware.com] has quit [] 23:00:02 for CL, reading PCL is *a lot* harder 23:00:07 -!- oudeis [n=oudeis@89-138-15-229.bb.netvision.net.il] has quit [No route to host] 23:00:15 weirdo: cl gives instant grat. it's just the mindset people have about programming that makes it hard to notice. 23:00:32 fusss: rails is getting pretty common. And in any case, the amount of work that is contracted for rails is certainly a factor higher than the work lisp is getting (which feels like a skewed reality) 23:00:33 i had to read PCL (which is few dozen pages) and reread chapters because i didn't understand them at first 23:00:34 CL is harder than perl? yeah, for a unix sysadmin maybe. 23:00:37 compare that to perl 23:00:38 weirdo: most people want to write 'standalone' executables. Running cross-platform and compiled/obfuscated. 23:00:47 btw evening all 23:01:04 mulander, no they don't. neither perl, ruby, php or python provides executables 23:01:05 -!- roby_ [n=roby@host234-206-dynamic.49-82-r.retail.telecomitalia.it] has quit [Remote closed the connection] 23:01:08 weirdo: me too, many many great concepts, in a very nice syntax 23:01:38 weirdo: then explain the existance of perl2exe, rubyscript2exe etc. ? 23:01:43 that doesn't give instant gratification to unix-hobbyist crowd 23:01:55 mulander, perlcc? it breaks on shared objects 23:01:58 weirdo: ah you're talking about the learning curve. 23:02:00 mulander: 'yo', and I don't think that is an argument... it simply isn't distributed that way. Install the interpreter and run the code 23:02:10 mulander: I've even seen that with haskell... 23:02:14 pkhuong, i don't say CL is hard. it just doesn't provide *instant* gratification 23:02:27 there's a cll post where i elaborated on the issues 23:02:44 weirdo: because most hobbyists don't want to write 'systems' but scripts for parsing files/crawling websites 23:02:47 weirdo: which I do get, now that I know how to use my tools. 23:02:52 mulander: keep in mind that it isn't targetting old companies that soulely use Java or C... 23:02:58 http://42.pl/gg/giqr5c$2nie$1@opal.icpnet.pl 23:02:58 ** C and derivatives 23:03:03 dreish [n=dreish@minus.dreish.org] has joined #lisp 23:03:11 madnificent: I was reffering to the crowd outside of corporations. 23:03:23 the main "groups" are hobbyists or corporations 23:03:32 mulander: and how do they hire people? (or do you mean schools?) 23:03:39 maybe also academicians who like the hindley-milner type checker for some reason 23:04:23 netaust1n [n=austinsm@160.79.78.72] has joined #lisp 23:04:45 madnificent: most corps. are covered with concrete. They don't want to touch there development cycle if they don't have to. The only place You can plug in new technology is for utilities to help with the old process. 23:04:53 -!- malumalu_ [n=malu@hnvr-4dbb4c2e.pool.einsundeins.de] has quit ["Verlassend"] 23:05:30 it's easier to push lisp there then for the main products. 23:05:37 I think it's a bit pop culture too. most people like simple movies/music/etc. fewer people like more complex movies/music. that's the same with Lisp 23:05:54 don't push jack. go out there and take their bread, using the best technologies you can find. 23:05:55 lisp is not complex 23:05:57 slash_ [n=Unknown@p4FF0A9E7.dip.t-dialin.net] has joined #lisp 23:06:11 lisp is dead simple 23:06:20 Lisp is complex 23:06:32 you're not filling me with confidence 23:06:34 :( 23:06:34 mulander: nope, I don't think so. You push a language with success stories 23:06:38 Dead simple is complex. 23:06:54 Just consider the physical sciences. 23:06:55 mulander: you don't need to plug in here and there to start with. You need to be better than the ancient competition 23:07:15 -!- abeaumont [n=abeaumon@120.pool85-54-87.dynamic.orange.es] has quit [Read error: 54 (Connection reset by peer)] 23:07:43 what's there to learn, anyway? read syntax, CL special forms, CL functions and CL macros 23:07:48 many rails projects start for that reason. Some investors have 'heard' it to be better. They bought the story and they are willing to put their money where their mouth is... 23:07:54 maybe libraries 23:08:02 abeaumont [n=abeaumon@120.pool85-54-87.dynamic.orange.es] has joined #lisp 23:08:02 a compiler vendor needs to "push" a language. not me :-) 23:08:11 weirdo: symbols, packages 23:08:12 oh, and evaluation semantics 23:08:20 weirdo: compilation 23:08:29 z0d, which are covered by CL functions and read syntax :) 23:08:34 archangelpetro: it's a tear-fest, ignore and half fun hacking 23:08:44 madnificent: I mostly agree. I'm just saying that it isn't worth it to fight for the main product if You work for a major corporation. 23:08:47 maybe a killer app would do the trick? 23:08:54 Don't forget good taste, the performance model and gotchas of your particular implementation, etc. 23:08:57 -!- pstickne [n=pstickne@c-71-236-177-238.hsd1.wa.comcast.net] has quit [Read error: 110 (Connection timed out)] 23:08:58 weirdo: values, clos, the way symbols and packages work, mapping of values, reducing of values, lambda-functions, closures, keywords, functions with variable arguments, restarts, probably some more 23:08:58 another web framework or something? 23:09:11 madnificent: if they let You work on Your own tools in whatever language You want then this can lead to success stories that can boost Your tools for better usage scenarios. 23:09:13 weirdo: not a framework, an app 23:09:18 cadabra [n=cadabra@128.187.176.117] has joined #lisp 23:10:00 weirdo: the framework is just there to make the story possible :) (and by now, many lispers seem to have their own framework \o/ <- me's building too) 23:10:01 -!- pjb [n=t@81-66-196-92.rev.numericable.fr] has quit [Remote closed the connection] 23:10:09 weirdo: Common Lisp is very complex. 23:10:25 In example. I doubt that I could replace an application built for >7 years by an enormous team of developers. But I sure can provide great utilities for other tasks around that system :) 23:10:28 CL is simply 'more' than other languages :) 23:10:32 I'm not sure whether it's particular complex, but it's very complex in any case. 23:11:25 the subset explained in PCL isn't that hard 23:11:34 beats scheme with call/cc semantics 23:11:54 seems like no one wants to help me with my CLOS problem :| http://paste.lisp.org/display/73141 23:12:15 Adamant [n=Adamant@c-71-226-66-93.hsd1.ga.comcast.net] has joined #lisp 23:12:17 weirdo: I told you that it works for me. Try with a fresh imagine. 23:12:26 mulander: in some cases you could be surprised... Some Java-code we wrote I initially tested in lisp (to show my system was possible). Due to the lack in message-overloading (don't think it's the exact term for this but heck) and generics, the Java code was roughly 5 times as big (and this is not an exception). Not only do you need to write that, you must allso keep an overview of the system, dramatically slowing you down on slightly 23:12:26 bigger code-bases 23:12:32 pjb [n=t@81-66-196-92.rev.numericable.fr] has joined #lisp 23:13:00 weirdo: did you try it in a new lisp image? 23:13:00 weirdo: s/imagine/image/ 23:13:39 oh right. i did FIND-CLASS 23:13:42 i mean 23:13:48 (setf find-class) 23:14:12 which could break things, as opposed to ENSURE-CLASS 23:14:12 I also think that the barrier to most people learning lisp is that they are used to a different development model. And it's hard for them to map it in CL. Even though great examples like PCL exist. 23:14:29 l_n [n=shawn@tuxhacker/lordnothing] has joined #lisp 23:14:42 tcr, now it works, thanks a bunch! 23:15:04 mulander: thats my prob atm 23:15:19 You know what? Enough about why more people don't use lisp. Shut up and hack! 23:15:42 i am hacking in the other frame 23:15:54 TLS index limit doesn't help :( 23:16:02 nyef: that's the ticket 23:16:05 -!- sunkencity_ [n=sunkenci@h121n2c1o1036.bredband.skanova.com] has quit [Read error: 110 (Connection timed out)] 23:16:42 mulander: lisp can take any form you want actually. Graciously mixing functional and object oriented methodologies, where usefull. But key is to master them all, so you can see what you need 23:17:06 EtFb [n=etfb@mail.hatrix.com] has joined #lisp 23:17:43 Too few people actually use Lisp, because everybody just whines about why more people doesn't use it. <-: 23:17:56 madnificent: Don't forget declarative paradigms. 23:18:07 z0d: I tend to be a lisp-advocate 23:18:37 nyef: my apologies. You can implement other methodologies in it too though :D 23:18:46 weirdo: do you really need a different dynamic binding for all of your variables? 23:19:03 -!- ehu [n=chatzill@82-170-33-173.ip.telfort.nl] has quit [Read error: 110 (Connection timed out)] 23:19:06 archangelpetro: just don't do the mistake I did while learning lisp for the 1st time. Don't think about how to deploy code, choosing a gui toolkit, finding that lib for the app You think You will write. Just focus on learning the language and do it from the repl. Just explore the language ;) 23:19:12 Yeah use one hash-table 23:19:30 nyef: 100% right, less babbling more coding :) 23:19:52 repl? 23:20:00 archangelpetro: Read Eval Print Loop 23:20:13 mcorp [n=mike@ukchill.demon.co.uk] has joined #lisp 23:20:14 -!- netaust1n [n=austinsm@160.79.78.72] has quit [] 23:20:30 pstickne [n=pstickne@c-71-236-177-238.hsd1.wa.comcast.net] has joined #lisp 23:21:31 -!- manuel_ [n=manuel@HSI-KBW-091-089-250-148.hsi2.kabel-badenwuerttemberg.de] has quit [] 23:21:47 -!- pjb [n=t@81-66-196-92.rev.numericable.fr] has left #lisp 23:21:49 archangelpetro: that thing in which you can enter (+ 1 2) and that will return 3 to you 23:22:14 pkhuong, variables need to be bound to gensyms 23:22:24 Good morning. 23:22:26 ggbbgg [n=russd@willers.employees.org] has joined #lisp 23:22:30 Hello beach. 23:22:38 weirdo: yes, but do you really need a different dynamic binding for each variable? 23:22:39 weirdo: ... why? 23:22:53 -!- Atherton [n=atherton@mathesar.kwzs.be] has quit [Read error: 104 (Connection reset by peer)] 23:22:54 pkhuong, i do, for O(1) access 23:22:56 good mornig 'beach that must ruin the last bit of biological clock I had' 23:22:59 netaust1n [n=austinsm@160.79.78.72] has joined #lisp 23:23:20 oh wait 23:23:21 i'm stupid 23:23:25 i could just cons a bit 23:23:26 wow, from LISP only has lists to LISP only has symbols. 23:23:29 *beach* apologizes for upsetting everyone's schedule by being in a different time zone. 23:23:42 beach: apologies accepted 23:24:45 no, what am i thinking 23:25:03 a vector will do 23:25:22 pjb [n=t@81-66-196-92.rev.numericable.fr] has joined #lisp 23:25:25 Atherton [n=atherton@mathesar.kwzs.be] has joined #lisp 23:25:41 like, put the previous value in a lexical gensym and later restore it 23:26:09 weirdo: Don't forget your unwind-protect. :-P 23:26:21 nyef, no need for unwind-protect 23:26:57 there are no bound handlers 23:27:17 using that scheme, i could make everything a TAGBODY 23:27:48 -!- Jabberwockey [n=jens@dslb-082-083-103-177.pools.arcor-ip.net] has quit [Remote closed the connection] 23:28:08 too bad i haven't thought of it earlier 23:28:13 thanks a lot, guys :) 23:29:16 was bitching about xml guis yesterday, which forced me to take a second look at them for about an hour 23:30:11 nyone here use lisp on fedora? 23:30:11 today we took a second look at XUL and already found it useful 23:30:15 can SUBTYPEP fail when there's no type named after a symbol? 23:30:30 i mean, fail, like throw an exception 23:30:32 archangelpetro: yes. 23:30:50 was going to ask what packages to install, but i just found them :D 23:31:08 "SUBTYPEP ... Exceptional Situations: None." 23:31:15 thanks! 23:32:17 Jabberwockey [n=jens@dslb-082-083-103-177.pools.arcor-ip.net] has joined #lisp 23:32:17 -!- mcorp [n=mike@ukchill.demon.co.uk] has quit ["leaving"] 23:32:34 xinming [n=hyy@125.109.250.26] has joined #lisp 23:33:04 pkhuong: it might not throw an specific exception, but if one of the arguments is not a valid type it "signals" an error 23:33:29 X-Scale [i=email@89.180.199.141] has joined #lisp 23:33:34 clhs subtypep 23:33:35 http://www.lispworks.com/reference/HyperSpec/Body/f_subtpp.htm 23:34:13 i don't see "Exceptional Situations" in neither the LW nor the franz hyperspec 23:34:23 -!- user_ [n=user@p549260E6.dip.t-dialin.net] has quit ["leaving"] 23:34:45 fusss: How are you getting on with the blog? 23:34:56 Ah, I got it. When beach arrives, it's way past my bedtime. Nighty! 23:35:06 z0d: on it. still figuring out a way to authenticate remove user to post a story 23:35:13 lemonodor_ [n=lemonodo@adsl-76-240-179-154.dsl.lsan03.sbcglobal.net] has joined #lisp 23:36:19 nighty tic 23:36:19 rvirding [n=rvirding@h40n5c1o1034.bredband.skanova.com] has joined #lisp 23:36:32 nite tic 23:36:33 archangelpetro, I always install from source, except for the first time, grabbing a binary directly. 23:37:15 z0d: would you happen to know how to remove "dots" from html UL unordered lists? 23:37:26 -!- chinawhite [n=ben@asa01-ext.utc.backmo.com] has quit [] 23:37:35 fusss: Yes, just a moment 23:37:40 my submission form is an UL and subject, body, submit and reset elements are members. 23:37:42 hothoofs [n=new@burnaby.axiomnetworking.ca] has joined #lisp 23:37:56 -!- milanj [n=milan@91.150.101.132] has quit ["Leaving"] 23:38:16 fusss: ul { list-style: none; } 23:38:58 ahh, great 23:39:06 -!- puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has quit [Remote closed the connection] 23:41:31 (:div :id "blogcontainer" (:div :id "blogbody") (:div :id "blogrhs")) those are your CSS ids. the blog is contained in blogcontainer, so you can inject it into any html. "blogbody" is just the list of recent stories, and "blogrhs" has the tag cloud, recent posts and links to archives. all get populated with tweaked db queries. 23:42:05 -!- alley_cat [i=AlleyCat@sourcemage/elder/alleycat] has quit ["Ex-Chat"] 23:42:09 fusss: it might signal an error... It's not defined what happens; NIL, NIL is also a standard compliant return value in that case. 23:43:04 pkhuong: i wasn't sure where you got that "Exceptional Situations" string. do you have the ANSI spec? or is it from CLtL? 23:43:32 fusss: Look again. 23:45:01 GrayMagiker [n=steve@c-68-35-128-231.hsd1.nm.comcast.net] has joined #lisp 23:45:18 -!- xinming__ [n=hyy@125.109.241.37] has quit [Read error: 110 (Connection timed out)] 23:45:24 pkhuong: there isn't even an "Exc" string anywhere. neither franz nor LW hyperspec. open the clhs link above and see. 23:47:44 The Exceptional Situation subsection is found in all the function definitions, near the end, between Affected By and See Also. 23:49:14 sellout [n=greg@c-24-128-50-176.hsd1.ma.comcast.net] has joined #lisp 23:50:01 -!- brickhazel [n=brickhaz@ool-45717ef5.dyn.optonline.net] has quit ["Leaving"] 23:50:02 -!- lemonodor [n=lemonodo@adsl-76-214-16-250.dsl.lsan03.sbcglobal.net] has quit [Read error: 110 (Connection timed out)] 23:50:14 fusss: clisp is the package i want, right? 23:50:19 what's the repl for it? 23:51:08 -!- cadabra [n=cadabra@128.187.176.117] has quit [Remote closed the connection] 23:51:27 pkhuong: two things conspired against me. the franz clhs DOESN'T have that section. and 2, my friken firefox was set to case sensitive search. 23:51:51 archangelpetro: i'm not sure i remember what we talked about. You need a common lisp implementation? 23:52:27 pkhuong: Iirc at least clisp's subtypep signals errors on some circumstances. 23:53:05 -!- divinebovine [n=new@mail.wjsgroup.com] has quit [Read error: 110 (Connection timed out)] 23:53:10 pkhuong: http://i41.tinypic.com/earzt2.jpg :-S 23:53:46 trittweil: yes, clisp DOES signal an error and has one "abort" restart or some such 23:54:01 archangelpetro: the repl is built into any lisp. When you launch it (say: you're evil and just enter clisp in a terminal), you get the repl :) 23:54:11 fusss: yea, i need clisp on my machine, i installed 'clisp and clisp-devel' :) 23:54:12 ok 23:54:19 fusss: I'm not sure why you mention the abort restart, there's IGNORE-ERRORS, you know. 23:54:37 archangelpetro: just type "clisp" and awaaay you go. you will soon need emacs + slime though 23:54:53 slime aint in the repos, as far as i can tell 23:54:58 trittweil: Type specifier is actually precisely defined (and not just any random sexp). OTOH, it seems the authors didn't think it was necessary to include a note that the function should/might signal some type error. 23:55:02 archangelpetro: you'll probably want to connect it to emacs (hard to learn, but it's very nice for lisp development), or to able or another editor/ide of your choice 23:55:14 -!- Atherton [n=atherton@mathesar.kwzs.be] has quit [Read error: 60 (Operation timed out)] 23:55:50 -!- beach [n=user@58.186.166.9] has left #lisp 23:56:06 madnificent: you only need to learn 8 commands to hack lisp with emacs. an afternoon of fun, imo. 23:56:12 gotta get back to work 23:56:39 It seems reasonable to me that subtypep can signal an error under certain circumstances. 23:56:54 -!- hothoofs is now known as moocow 23:56:58 archangelpetro: the REPL is like the prompt of your shell 23:57:01 -!- djinni` [n=djinni`@ludios.net] has quit [Client Quit] 23:57:16 djinni` [n=djinni`@ludios.net] has joined #lisp 23:57:19 pkhuong: The syntax of type specifier is precisely defined. Not what constitutes a valid type specifier at any point of time. 23:57:24 -!- trebor_h` [n=user@dslb-084-058-216-182.pools.arcor-ip.net] has quit [Read error: 110 (Connection timed out)] 23:57:26 Its first two arguments are specified as type specifiers, and, per 1.4.4.3, the consequences of them not being so are undefined. 23:57:37 -!- willb [n=wibenton@wireless107.cs.wisc.edu] has quit [Read error: 110 (Connection timed out)] 23:57:38 -!- bit` [n=bit@c-67-171-211-187.hsd1.or.comcast.net] has quit [Remote closed the connection] 23:58:15 z0d: i got that, maybe i'm just 'that' bad at Lisp that i didnt realise i was at the interpreter/reader 23:58:28 fusss: you do have a point. Perhaps they should be on cliki or something like that (as to be able to point newcomers to them) 23:58:34 I would've liked it :) 23:58:52 fusss: have fun working :) 23:58:58 And 4.2.3 defines what a type specifier is and when and how new ones are created. 23:59:04 tcr, how's the timeline for merging my stuff to named-readtables? :) 23:59:09 yangsx [n=yangsx@218.247.244.25] has joined #lisp 23:59:10 archangelpetro: Take a look at the book Practical Common Lisp (also available online)