00:03:46 -!- manuel_ [n=manuel@port-92-205-28-78.dynamic.qsc.de] has quit [] 00:09:12 -!- Liempt [n=Bevinvan@unaffiliated/liempt] has quit [Connection timed out] 00:09:21 Liempt [n=Bevinvan@unaffiliated/liempt] has joined #lisp 00:10:43 manuel_ [n=manuel@port-92-205-28-78.dynamic.qsc.de] has joined #lisp 00:13:27 -!- Liempt [n=Bevinvan@unaffiliated/liempt] has quit [Connection reset by peer] 00:16:11 -!- vanLiempt [n=Bevinvan@unaffiliated/liempt] has quit [Success] 00:16:38 -!- Nshag [i=user@Mix-Orleans-105-3-89.w193-250.abo.wanadoo.fr] has quit ["Quitte"] 00:19:06 -!- manuel_ [n=manuel@port-92-205-28-78.dynamic.qsc.de] has quit [] 00:20:42 dayid [n=dayid@unaffiliated/dayid] has joined #lisp 00:24:56 manuel_ [n=manuel@port-92-205-28-78.dynamic.qsc.de] has joined #lisp 00:25:07 -!- a-s [i=root@93.112.71.47] has quit [Remote closed the connection] 00:27:36 -!- Yuuhi [n=user@p5483D710.dip.t-dialin.net] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 00:28:30 -!- fe[nl]ix [n=algidus@88-149-212-223.dynamic.ngi.it] has quit ["Valete!"] 00:28:41 kpreid: you were right about the spurious conflicts, they were a result of eval'ing things while inside common-lisp-user, thanks 00:29:10 -!- manuel_ [n=manuel@port-92-205-28-78.dynamic.qsc.de] has quit [Client Quit] 00:30:04 *araujo* the more he uses RoR, the more he wonders ... how the hell a framework like this never made it on Lisp before 00:30:41 larrytheliquid: minor point that will help avoid confusion later: it is not eval that interns symbols, it is read 00:30:53 araujo: can't want everything right away. Have you heard of Lisp on Lines? 00:31:10 sykopomp, no ... something like it? 00:31:15 it's just that ... 00:31:34 araujo: Lisp on Lines does things in a really neat way. 00:31:41 I feel that framework so Lisp'er .... 00:31:43 and it's meant to be sort of like Rails 00:31:59 mm.. it came after or before Rails? 00:32:06 after 00:32:28 oh well, that's what I mean.. but, I will check it out right away 00:32:38 before 00:32:48 the name came after though 00:32:53 oh. Nevermind then. That answers that. 00:32:58 that's really cool :D 00:33:17 drewc, really? ... good to know.. I guess 00:33:18 it's actually nothing like rails, as far as i understand it. 00:33:33 i've never used rails personally. 00:33:47 drewc, pretty much a DSL 00:34:23 ha 00:34:37 I don't know if it's just me .. but .. heck, I find it an approach so suitable for a Lisp application .... 00:34:43 does it have web 2.0 synergy? 00:34:54 drewc: tongue-in-cheek?.... 00:34:56 That I keep wondering why lisp didn't have anything like this .... 00:35:05 "DSL" 00:35:09 yes drewc 00:35:31 this word lost its meaning, mostly because of ruby folk abusing it IMO. 00:35:53 ruby folks love their buzzwords. Possibly even more than java folks. 00:36:17 drewc, If you ask me ... I think that the ruby community wasn't ready for such an application 00:36:45 araujo: i'm not sure what you mean, but lets talk about lisp. 00:36:58 RoR is a Lisp application written on Ruby 00:37:16 That's how I have felt during these last weeks I have been playing with it 00:37:19 obviously not a lisp application. It's written in Ruby. 00:37:30 drewc, that's what I meant ^^ 00:37:33 although I'd be interested in putting my finger on what makes RoR really nice to use. 00:37:39 so we can take that 00:37:39 :) 00:37:58 lisp: 'oooh, we like this little thing you can do, it's a good idea. I think I'll have it now' 00:38:18 from what i remember about rails, it has a whole bunch of template crap, and code generation, and blah blah REST this... yawn. 00:38:51 drewc: I'm not quite clear on what LoL does that makes things so easy. 00:38:56 :\ 00:39:00 sykopomp, I really wouldn't mind to get my hands dirty coding some Lisp web framework 00:39:13 I actually would pretty much like to see one 00:39:28 araujo: *points at drewc* talk to this guy. I'm trying to help, too. 00:39:31 There are plenty of room to improve over the RoR framework 00:39:39 That would suit better on Lisp 00:39:49 amnesiac [n=amnesiac@p3m/member/Amnesiac] has joined #lisp 00:40:04 I don't know if I'd like to see a copy of rails. I think we can do better than that, eh? ;) 00:40:07 drewc, The template part .. being one of those things to improve upon 00:40:42 sykopomp, pretty much what I said ... but RoR follows a Lisp philosophy .. imho 00:40:47 araujo: have you tried all the existing lisp web frameworks and found them lacking, or is this all just hyperbole?? 00:40:56 ths_ [n=ths@X745a.x.pppool.de] has joined #lisp 00:41:04 I guess that a reason for a framework to be RoR like would be so previous RoR users may more easily migrate, but is that a good enough incentive? 00:41:19 drewc, there is such a thing like Lisp framework? 00:41:24 er, web framework 00:41:50 minion: web? 00:41:51 web: Web clients (web client), servers, application servers, browsers, etc. http://www.cliki.net/web 00:42:03 minion: ucw? 00:42:04 ucw: UnCommon Web is a Common Lisp web application development framework. http://www.cliki.net/ucw 00:42:11 minion: hunchentoot? 00:42:12 hunchentoot: Hunchentoot is a web server written in Common Lisp and at the same time a toolkit for building dynamic websites. http://www.cliki.net/hunchentoot 00:42:23 minion: bknr? 00:42:24 bknr: bknr is an object datastore, a template system, a web framework and support for images, blogs, billboards, etc. http://www.cliki.net/bknr 00:42:37 minion: minion: lisp-on-lines? 00:42:38 i like lisp... i'm written in it 00:42:45 minion: lisp-on-lines? 00:42:46 lisp-on-lines: Lisp-on-lines is a web application framework built on top of CLSQL and UCW and provides an application development model similar in many ways to Ruby on Rails. http://www.cliki.net/lisp-on-lines 00:42:53 minion: cl-weblocks? 00:42:54 cl-weblocks: cl-weblocks is a web framework written in Common Lisp. http://www.cliki.net/cl-weblocks 00:43:03 wow plenty of them ... :-P 00:43:09 araujo: ever try cliki? or google? ;) 00:43:12 JustWhie [n=chatzill@c-71-60-93-1.hsd1.pa.comcast.net] has joined #lisp 00:43:18 *araujo* wonders why none of them are popular enough 00:43:26 :| 00:43:30 enough for what? 00:43:43 housel, for at least a chehap oreilly book to exist 00:43:46 cheap* 00:43:53 housel: didn't you hear that good things are always popular? 00:44:02 *drewc* shakes his head and moves on. 00:44:19 lisp is also not popular enough :) 00:45:03 phadthai, there are good lisp applications ... I even know some commercial privates ones 00:45:08 no web framework though 00:45:11 araujo: of course 00:45:26 well, let's see if some of them are even better (or worse ..) than RoR then 00:46:14 _cheerios [n=Jack@dsl-hkibrasgw3-fe74fb00-140.dhcp.inet.fi] has joined #lisp 00:46:49 sykopomp, oh, hah, yeah Lisp-on-Lines might be what I am looking for 00:47:39 erm... trying to build latest release of sbcl with clisp-2.43 ... anyone interested in seeing error mess ;?) i can always try binaries... 00:47:44 araujo: I don't think they'll be a book any time soon... LoL is still for .. self-starters right now. 00:48:42 drewc, LoL seems .. new from the repo logs , and actively developed ... 00:48:55 araujo: drewc is the developer 00:49:14 new ... 4 years old now i think. 00:49:15 ah, awesome :-D 00:49:23 haha 00:49:27 actively developed though, yes. 00:49:33 drewc, and then .. what happens? 00:49:38 you the only developer? 00:50:24 araujo: at the moment yes. 00:50:30 ok 00:50:37 -!- lukego [n=lukegorr@72.37.205.4] has quit [] 00:51:25 drewc, the web page says it implements a development model similar to RoR? 00:51:30 -!- lemonodor [n=lemonodo@66.43.112.62] has quit [Read error: 113 (No route to host)] 00:52:18 jtoy [n=jtoy@58.61.213.234] has joined #lisp 00:53:24 drewc: Did you see my email about rsync on common-lisp? 00:54:52 gigamonkey` [n=user@adsl-99-179-47-191.dsl.pltn13.sbcglobal.net] has joined #lisp 00:55:08 fe[nl]ix [n=algidus@88-149-212-223.dynamic.ngi.it] has joined #lisp 00:56:12 -!- ths [n=ths@X4c7f.x.pppool.de] has quit [Read error: 110 (Connection timed out)] 00:57:31 http://paste.lisp.org/display/68293 00:58:22 drewc: lol, roflmao. cute :) 00:59:04 larrytheliquid: (:export first-function) should be (:export #:first-function) -- the first interns a useless symbol in whatever the current package is 00:59:33 larrytheliquid: and cl:use-package at the bottom there takes effect only at runtime, so is probably not what you want -- use the :use option to defpackage instead 01:00:01 -!- schasi [n=schasi@p54A26CAC.dip.t-dialin.net] has quit ["Verlassend"] 01:00:23 lemonodor [n=lemonodo@adsl-76-214-15-172.dsl.lsan03.sbcglobal.net] has joined #lisp 01:00:27 araujo: there is no web page 01:00:58 araujo: there is a cliki page. i didn't make it. 01:02:42 rtoym: i see that now .. what exactly is the error? 01:02:43 10 times faster. This looks promising. 01:02:54 drewc: Hold on.... 01:03:12 -!- NoorDextor [n=NoorDext@unaffiliated/noordextor] has left #lisp 01:03:20 drewc: It says failed to connect. Connection refused. 01:05:30 drewc, 0_o ok 01:05:34 kpreid: if i just make the first change you mentioned i still get the same error, is that due to the runtime use-package? 01:06:10 larrytheliquid: I'm not sure, but when working with packages remember that there is still the *current* state of the package system, not just what's written in the files 01:06:15 araujo: there is also very little for documentation :) 01:06:58 kpreid: do you mean like the *package* that refers to common-lisp-user in my repl? 01:07:20 rtoym: try now? 01:07:26 larrytheliquid: no, I mean in the packages that exist, what symbols they have interned, what the use-list is, and so on 01:08:02 drewc: Works fine now. Thanks! 01:08:09 w00t! 01:08:22 larrytheliquid: take a look at http://www.lispworks.com/documentation/HyperSpec/Body/c_packag.htm for tools to see what is 01:08:23 *drewc* is teh yoonicks master. 01:08:39 What happened? Forgot to enable the port through the firewall? 01:09:10 kpreid: im still confused about the package/symbol system but ill take a look at the link, thanks 01:09:47 rtoym: forgot to set it to 'true' in /etc/defaults/rsync 01:10:50 -!- sylvander_ [n=sylvande@92.22.101.35] has quit ["leaving"] 01:11:05 rtoym: re: the new cvs, if you notice anything weird let me know .. we ran into a silly bug and i had to patch it. 01:11:06 -!- larrytheliquid [n=larrythe@nmd.sbx10362.wintefl.wayport.net] has quit [] 01:11:39 drewc: Seems to be working ok so far. I've done quite a few checkins and stuff already. 01:11:44 good to know. 01:12:12 Although, if I had know there were bugs, I might have hesitated. :-) 01:12:56 the bug was in the configuration file parsing, so it shouldn't touch the important bits 01:13:09 that said... cvs is one hell of an ugly code base. 01:17:18 -!- jajcloz [n=jaj@209-6-216-149.c3-0.smr-ubr3.sbo-smr.ma.cable.rcn.com] has quit [] 01:19:53 -!- parodyoflanguage [n=klh@mmds-216-19-34-118.twm.az.commspeed.net] has quit [Read error: 104 (Connection reset by peer)] 01:21:05 parodyoflanguage [n=klh@mmds-216-19-34-118.twm.az.commspeed.net] has joined #lisp 01:21:48 -!- lemonodor [n=lemonodo@adsl-76-214-15-172.dsl.lsan03.sbcglobal.net] has quit [] 01:22:37 -!- bhyde [n=bhyde@c-66-30-202-56.hsd1.ma.comcast.net] has quit [] 01:25:38 -!- mulligan` [n=user@e178031213.adsl.alicedsl.de] has quit [Remote closed the connection] 01:29:57 rickardg` [n=user@c-e51ae455.027-77-6c756e10.cust.bredbandsbolaget.se] has joined #lisp 01:31:28 cheapmachine [n=chatzill@97-81-224-137.dhcp.lds.al.charter.com] has joined #lisp 01:46:20 -!- rickardg [n=user@c-e51ae455.027-77-6c756e10.cust.bredbandsbolaget.se] has quit [Read error: 110 (Connection timed out)] 02:04:06 -!- JustWhie [n=chatzill@c-71-60-93-1.hsd1.pa.comcast.net] has left #lisp 02:06:33 larrytheliquid [n=larrythe@169.157.33.65.cfl.res.rr.com] has joined #lisp 02:15:30 -!- cheapmachine [n=chatzill@97-81-224-137.dhcp.lds.al.charter.com] has quit ["ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]"] 02:21:41 -!- gigamonkey` [n=user@adsl-99-179-47-191.dsl.pltn13.sbcglobal.net] has quit [Read error: 110 (Connection timed out)] 02:22:25 -!- seelenquell_ [n=seelenqu@pD9E469B1.dip.t-dialin.net] has quit ["Konversation terminated!"] 02:24:19 gigamonkey` [n=user@adsl-99-179-47-191.dsl.pltn13.sbcglobal.net] has joined #lisp 02:27:19 lukego [n=lukegorr@cpe-76-87-84-57.socal.res.rr.com] has joined #lisp 02:35:05 Elby [n=Elby@pool-138-88-9-110.res.east.verizon.net] has joined #lisp 02:35:15 hey 02:36:30 hi 02:36:54 I am a beginner at lisp. I am looking for someone who can patiently help me get down the subtlties of it. 02:37:10 or, should I say, the subtlties necessary to begin programming in it 02:39:07 I just typed up my first function that I wrote by hand on paper 02:40:13 http://pastebin.com/d90a287c 02:40:22 Please tell me how my code looks. 02:40:38 try out gigamonkey's book, Practical Common Lisp (freely available) or find an open source lisp package that you fancy and try to hack on it yourself 02:40:50 but I am going to fall over from sleepiness, so bye bye 02:40:54 -!- b4|hraban [n=b4@0brg.xs4all.nl] has quit ["zzzZZZzzzZZZ"] 02:42:10 araujo: none of them are popular enough because everyone is too busy writing web frameworks to actually write websites with them 02:43:07 Elby585 [n=Elby@pool-138-88-9-110.res.east.verizon.net] has joined #lisp 02:43:10 -!- Elby [n=Elby@pool-138-88-9-110.res.east.verizon.net] has quit ["http://irc2go.com/"] 02:43:16 -!- Elby585 is now known as Elby 02:43:18 S11001001, ? 02:43:26 -!- Elby [n=Elby@pool-138-88-9-110.res.east.verizon.net] has quit [Client Quit] 02:43:43 Elby: the code looks like you don't quite know what you are doing yet :) 02:44:06 that's the power of lisp - every web programmer has their own framework. it's easy! 02:44:26 hefner FSVO easy. it's easy to do the easy part. 02:44:42 -!- dialtone [n=dialtone@adsl-99-136-101-166.dsl.pltn13.sbcglobal.net] has quit [Read error: 110 (Connection timed out)] 02:44:58 luckily, you're miles ahead at that point, and the hard part can figure itself out over the years. 02:45:06 I wasn't being too serious, but what is the hard part? 02:45:08 Ash [i=aaron@facestab.org] has joined #lisp 02:45:52 * araujo wonders why none of them are popular enough 02:46:03 S11001001, right .. ok , that explains it :-P 02:46:15 Elby [n=Elby@pool-138-88-9-110.res.east.verizon.net] has joined #lisp 02:46:19 hefner: well, IMO its making your web-dsl descriptive and flexible enough so that you actually use the framework rather than hand-coding 50% of it. 02:46:33 what's a web dsl? 02:47:39 hehehe. In my case, it's a few forms that define a declarative language for describing how lisp objects will be displayed. 02:48:10 maybe showing makes sense ... 02:48:37 can anyone help me write a basic function? 02:48:54 or, rather, would anyone kindly help me write a basic function. 02:48:59 We can perhaps assist by answering questions 02:49:08 lisppaste: url? 02:49:09 To use the lisppaste bot, visit http://paste.lisp.org/new/lisp and enter your paste. 02:49:23 -!- jao [n=user@129.Red-83-42-110.dynamicIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)] 02:49:26 Alright. I typed up some code in Notepad and pasted into my terminal that is running lisp. 02:49:37 Elby: check out "The little schemer" 02:49:42 Elby: BASIC doesn't support functions; however, you can use line numbers and GOSUB to simulate them. 02:49:45 jao [n=user@129.Red-83-42-110.dynamicIP.rima-tde.net] has joined #lisp 02:49:50 drewc pasted "my 'web dsl'" at http://paste.lisp.org/display/68295 02:49:55 i wondered if anyone would make the joke 02:50:01 http://pastebin.com/d90a287c 02:50:21 S11001001: (tagbody 10 (print foo) 20 (go 10)) :) 02:50:38 this is supposed to, given a list of characters, print out all subsets of size 3 from that character set, using loops 02:50:49 first, no to the setf, use let instead 02:51:09 okay. what's the difference? 02:51:23 minion: tell Elby about pcl 02:51:25 Elby: look at pcl: 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). 02:51:29 setf alters an existing place; you didn't make a place yet 02:51:58 let will make a variable for you, which is a kind of place 02:51:58 Elby: it looks like you'd benefit from reading that book.. at least the first couple chapters to get the basics down. 02:52:00 salex [n=user@216.80.143.240] has joined #lisp 02:52:14 second: use an editor that indents lisp right 02:52:37 drewc: i'm sorry. I know how I learn, and I really learn best from a little guidance. I read the first few chapters of my friends lisp book yesterday 02:52:58 so i have a sense of it, but working in the IDE and getting results is a little different 02:53:03 Elby: was your friends lisp book PCL? 02:53:04 hmm, I never realized a class metaclass could change the parsing of slots 02:53:07 hi S11001001. I'm thinking of making the w-r-i generator return the non-terminating flag of the macro char, too. 02:53:20 hefner: oh hell yeah, and then some. 02:53:33 drewc: there seem to be mails stuck in the cdr-discuss mailinglist on c-l.net. 02:53:35 S11001001: I tried to install the Lisp plugin for Eclipse but it wasn't compiling. I don't know what text editor to use in the terminal, so I'm using Windows notepad and pasting it into the program called alisp 02:53:48 drewc: I think the book was called ANSI Lisp or something 02:54:07 tcr: that property has little to do with character macrology; I think it would be more confusing than helpful 02:54:10 Elby: it would really help you to read pcl. it is available online 02:54:24 drewc: time is also a concern of mine. 02:54:38 Elby: as it is to the people in this channel. 02:54:58 that's why I'm asking for volunteers, not demanding service. 02:55:01 Elby: then find another solution. For one thing, your own indentation versus my reindentation shows that your parens aren't where you think they are 02:55:19 -!- eno [n=eno@nslu2-linux/eno] has quit [Read error: 60 (Operation timed out)] 02:55:22 Could I see your reindentation 02:55:23 Elby: go get LispWorks, or *deity* forbid, Emacs 02:55:34 Elby: Asking basic questions that are available with a simple google is considered somewhat.. is rude the word? at least inconsiderate. 02:55:38 And ask, did you do it by hand or use some sort of auto-intenter 02:55:38 elby: 1) download lisp in a box or a demo from franz or lispworks if you ar on windows 02:55:41 2) read PCL 02:55:46 S11001001: Basically there are two common use cases for iterating through a readtable: a) getting at the macro chars, copying a macro char from the rt being traversed into another rt; and for the latter case, the terminating flag is required. 02:55:52 eno [n=eno@nslu2-linux/eno] has joined #lisp 02:55:59 Elby: any decent editor will indent for you 02:56:12 Elby: no one does it by hand, that's what I'm saying. You'll be lost until you get an editor that can do it right. 02:56:21 tcr: damnit. 02:56:30 tcr: stuck how? 02:56:42 Elby: the reindentation also shows that all your DO forms are incorrectly constructed. 02:56:46 drewc: H4ns confirmed that they're in exim's maillog or something 02:57:17 -!- bombshelter13 [n=bombshel@209-161-240-8.dsl.look.ca] has quit [Client Quit] 02:57:20 tcr: whoa boy. The new mailing list setup is not my forte yet, but i'll see what i can do. 02:57:31 tcr: can you set a char to be terminating without it being a macro char? 02:57:41 I am downloading Lispworks. It should take around 5 minutes. 02:58:05 S11001001: Not portably, I think. 02:58:44 S11001001: In fact, terminating is not a valid character syntax type, but a "flavor" of the macro-char syntax type 02:58:45 download pcl while you are at it 03:00:22 tcr: as a matter of fact .. i don't even have access to admin the lists back yet . youd be better of sending a ticket in to admin@common-lisp.net so gwking can have a look.. he's the one who's supposed to be making this work again. 03:00:27 I found the pcl website. Is that sufficient? 03:00:59 drewc: Will this also reach H4ns? 03:01:07 tcr: which doesn't make sense to me, but if that's how it is... 03:01:33 elby: erm, actually i was thinking of the source that goes with. sorry, unclear 03:01:43 what is the source? 03:01:46 and why do i need it? 03:01:57 S11001001: what should a terminating non-macro-char entail? 03:02:04 examples in the book. you can follow along (you said you wanted guidance, no?) 03:02:29 okay. i downloaded the source. 03:02:34 anyway work your way through first half dozen chapters or so to get a feel for it 03:02:37 have fun :) 03:02:40 lol 03:03:38 tcr: when you set s to be terminating, reading "notaswe" sans quotes twice yields the symbols NOTA and SWE 03:03:49 S11001001: If it's supposed to make the char end a token, you can use (set-syntax-from-char to-char #\Space to-rt (copy-readtable nil)) 03:03:57 i.e. to make the syntax type of TO-CHAR be whitespace 03:04:07 tcr: that would yield NOTA and WE 03:04:29 tomoyuki28jp [n=tomoyuki@EM119-72-25-210.pool.e-mobile.ne.jp] has joined #lisp 03:04:29 true 03:04:38 Is there a way to write a macro which works for both of this cases? (macro "param1" (body...)) and (macro ("param1" "param2") (body...)) 03:04:50 since you can read (a)(b) twice and get (A) and (B)... 03:04:50 Sure 03:05:59 Well, you can get that behaviour by installing a reader fn which just invokes READ recursively 03:06:23 -!- dayid [n=dayid@unaffiliated/dayid] has left #lisp 03:06:55 (not quite that easy, I think the stream argument is positioned forwardly one char) 03:07:21 but we recently had the issue if it was legit to invoke unread-char on the stream passed to a reader fn 03:08:43 anyway, I'm just saying that terminatyness is independent of macroness 03:09:55 I'll raise the issue on the mailing list once cdr-discuss works again 03:13:28 tcr: yup, all of us read the admin@t tickets 03:16:26 tomoyuki28jp: what do you think, using what you know about lisp? 03:17:28 -!- tcr [n=tcr@host145.natpool.mwn.de] has quit ["Leaving."] 03:17:54 drewc: In scheme, I can do that with syntax-case, but I have no idea how I can do that in lisp. 03:17:58 do do do 03:18:07 clhs defmacro 03:18:07 http://www.lispworks.com/reference/HyperSpec/Body/m_defmac.htm 03:18:34 Jasko2 [n=tjasko@65.243.193.195] has joined #lisp 03:18:34 Elby: fwiw, most lispers i know and work with do not use DO. 03:18:36 okay, got lisp works up and running 03:19:32 drewc, could you please help me translate a simple java function to lisp, just to help me over the initial hurdle of familiarizing myself with syntax and rules? 03:19:35 here is the real trick, implement a function DO-ONCE that accepts a no-arg function, calls it, and *always* answers NIL. 03:20:16 Elby: no, but if you have a specific question i'd be happy to answer it. The best way to familiarize yourself is to read PCL and play with things at the repl. 03:20:46 (besides, i know very little java) 03:21:32 repl? 03:21:38 *rtoym* uses do. 03:22:14 Elby: Read Eval Print Loop .. the lisp listener. 10 minutes reading PCL will at least get you that far. 03:23:14 rtoym: as a habit or in those few cases where do is actually the best thing? :) 03:23:14 alright, so why not use DO 03:23:17 what do you suggest 03:23:22 what is closest to a for loop 03:23:26 Elby: LOOP 03:23:27 or is there a for loop 03:23:30 i see 03:23:47 Elby: 10 minutes with pcl etc etc 03:24:37 drewc: Mostly habit. And because I can never remember all the options for loop. But I do use for loops. 03:24:55 i looked up the loop 03:24:59 and it still uses the word do 03:25:03 is this a problem? 03:25:27 drewc: yes, thank you for your recommendation. No more need to refer me to pcl. 03:25:47 -!- birdsbite [n=mark@75.110.164.248] has quit ["Lost terminal"] 03:26:22 Elby: ah good. Enjoy the book, and lisp. 03:27:07 so, the for loop uses the word do. is this a problem? 03:28:03 Elby: read pcl already? 03:28:21 -!- anekos is now known as awayekos 03:28:22 alright. fuck off good sir. k thx. 03:28:57 ahem. 03:29:13 -!- ChanServ has set mode +o drewc 03:29:27 FUCK THE POLICE 03:29:38 SMOKE MARIJUANA EVERY DAY 03:29:40 FUCK YOU DREW C 03:29:44 EAT MY DICK 03:30:12 -!- drewc has set mode +b *!*n=Elby@*.res.east.verizon.net 03:30:12 -!- drewc [n=drewc@89.16.166.162] has been kicked from #lisp 03:30:51 php programmers .. all the same 03:30:55 -!- ChanServ has set mode -o drewc 03:31:12 i did not want to do that at all.. i feel as if i've been trolled succesfully. 03:31:35 Holy smokes 03:32:56 whoa. What the hell. 03:33:08 also, he misspelled "DA" 03:37:34 -!- rickardg` [n=user@c-e51ae455.027-77-6c756e10.cust.bredbandsbolaget.se] has quit [Read error: 110 (Connection timed out)] 03:40:13 ah man .. i cvs up'ed my slime yesterday to test cvs on cl-net ... now i have to restart my emacs.. cause i just bounced my lisp now :( 03:40:22 dnm [n=dnm@cpe-67-246-46-208.nycap.res.rr.com] has joined #lisp 03:41:51 -!- Chrononaut [n=bjorn@195.20.207.210] has quit [Read error: 113 (No route to host)] 03:42:43 envi^home [n=envi@220.121.234.156] has joined #lisp 03:43:15 drewc: I have been reading the link you gave me, but I still haven't find a way to write a macro which will match both of atom and list parameters. Could you give me some more hint? 03:44:05 (defmacro foo (atom-or-list) (if (atom atom-or-list) ...) 03:44:37 -!- Hun [n=Hun@p4FCF5E37.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 03:49:52 drewc: I had been confusing on something, but now I got the idea. Thanks a lot for your help! 03:50:26 tomoyuki28jp: no problemo! 03:50:41 drewc: :) 03:51:19 that is too bad about elby; I was just thinking that some people might have personalities better suited to learning through interaction with other people, and considered using Elby to experiment in that regard. 03:52:16 e.g. I couldn't learn how to save my life from a lecture since kindergarten, so it was always a good thing that books covered most of the material 03:52:21 but oh well, troll 03:55:31 S11001001: oddly enough, there are these things called `courses' 03:55:48 -!- BrianRice-mb [n=briantri@c-24-18-253-20.hsd1.wa.comcast.net] has quit [] 03:55:54 which is interaction with other people, no? 03:56:04 anyway, nm 03:56:47 salex: courses in Common Lisp? 03:57:21 sure, why not? 03:57:27 I mean, where are they? 03:57:27 ;) 03:57:34 -!- _cheerios [n=Jack@dsl-hkibrasgw3-fe74fb00-140.dhcp.inet.fi] has quit ["sleep"] 03:58:18 -!- parodyoflanguage [n=klh@mmds-216-19-34-118.twm.az.commspeed.net] has quit [] 03:58:26 that i can't help you with. anyway nothing wrong in principle with what elby was initially trying to do. but starting off with a sense of entitlement and then turning into a prick won't help anyone 03:58:33 well, indeed 03:58:42 I was absent for the interim, and thinking about it 03:59:14 anyway, the CL community selects for those who can learn it through books, largely discarding the other sort of learners 03:59:20 -!- tomoyuki28jp [n=tomoyuki@EM119-72-25-210.pool.e-mobile.ne.jp] has quit [Remote closed the connection] 04:00:38 I don't think that's true at all 04:00:50 many learn from example code, mostly 04:01:14 it is true that #lisp selects for newbies who will put independent effort in 04:01:20 but beyond that, is pretty helpful 04:02:59 okay, but could it benefit through providing mechanisms for the sort who can't/won't learn CL that way? 04:03:18 Could such people usefully expand the community, or just be a drain on limited newbie-aid resources? 04:03:42 what mechanisms do you feel are missing? 04:03:43 hmm is viewcvs known to be broken on common-lisp.net? 04:04:21 something like a "course" as you mentioned 04:04:28 i.e. "http://common-lisp.net/cgi-bin/viewcvs.cgi/zip/gray.lisp?rev=1.8&root=zip&view=markup" 04:04:33 that includes human interaction as a primary learning tool 04:04:57 rather than reading the first 6 chapters of PCL first, as valuable as that may be 04:04:58 isomer`` [n=isomer@CPE001310e6cb31-CM0011aec5e684.cpe.net.cable.rogers.com] has joined #lisp 04:04:58 -!- larrytheliquid [n=larrythe@169.157.33.65.cfl.res.rr.com] has quit [] 04:05:15 oh reading is only the start. you have to do. 04:05:20 always. 04:05:38 (i mean, that's the only way to learn anything, regardless of other aids) 04:06:34 Well, for example, constructivism seems to take the opposite approach, having you skip the reading part in favor of "doing" with the aid of a high-feedback interactive programming environment 04:06:44 S11001001: the problem is that there is a lot of material to cover, and IRC is not the medium for it. 04:06:52 That's fair. 04:07:01 but what you describe (a course) is expensive. how do you expect to make that work? 04:07:16 and, as drewc says, irc is a bad medium 04:07:21 I don't know, that's why I'm throwing it out here rather than just going to implement it myself :) 04:07:39 waht i'm tring to figure out is what part of this is cl-specific? 04:08:01 not to mention using irc it's hard to beat someone over the head with a lisp manual... 04:08:10 I care about CL, specifically :) 04:08:15 S11001001: PCL costs about half of what an hour of my time is worth... and is also available for free. I'm more than willing to give my time to someone who has made an effort to avoid wasting it :) 04:08:17 education & learning requires effort; you can't avoid that 04:08:39 of course 04:08:52 if you're going to ask people for their time gratis, you really should look to spend at *least* 10x the time yourself for starters, otherwise it doesn't make any sense 04:09:35 asking people to tell you things that you could trivially have looked up, or asking questions that make it clear you haven't tried/learned much --- this will just annoy generous people 04:09:44 *even generous 04:10:15 OT, and long shot: anyone here remember a .bst file for reverse-chron bibliographies in latex? 04:10:39 salex: it seems to be common among some language advocates to spent a fair amount of time convincing poor souls to use their language. i personally find it annoying to see that, as the effort is usually in vain. it usually creates admirers who are bad programmers. 04:10:56 indeed 04:11:08 H4ns: that's a good point. 04:11:13 if you have a bit of a `potential well', it's not all bad 04:11:33 -!- isomer [n=isomer@CPE001310e6cb31-CM0011aec5e684.cpe.net.cable.rogers.com] has quit [Connection timed out] 04:11:48 salex: yes. it is also a question of manners and politeness. all not really cl or irc specific in that respect :) 04:12:07 lispers have never seemed the 'please use our language type'. Then we complain that it's not popular... :) 04:12:24 'please use our language' type 04:13:06 I suspect you really only want 25%ish; how to target to them ? 04:13:16 :) this reminds me the "NetBSD - Hype free" slogan heh 04:14:14 salex: I think we (i) have to put some effort into making the path a little less resistant. 04:14:35 yes, i think so 04:14:44 I'm sure there are those who can 'help themselves' but simply get annoyed and have work to do. 04:15:06 yes. making the path easy for serious attempts is good 04:15:32 drewc: i think a well-maintained lisp distribution that includes loads of libraries, is used for real-world things and maintained could help. 04:15:39 no need to spend effort on attracting 50,000 people --- for 2 weeks each 04:15:54 H4ns: like franz, you mean? 04:15:56 ;) 04:16:01 H4ns: i think that everybody knows that, but we've been trying to avoid it. 04:16:11 salex: exactly. only with reasonable licensing term. 04:16:32 one real problem with that is sytem integration and UI integration 04:16:47 it sounds like it'll have to be a one-person effort 04:16:53 all these efforts for CPAN-like mechanisms and we've only got maybe 25-50 libraries actually worth using :) 04:16:53 drewc: openmcl seems to become more and more attractive for such a project because it supports all platforms. 04:16:55 not the easiest thing, and would require a load of work from a pretty small group of people 04:17:20 H4ns: i have been feeling that way myself. 04:17:28 yeah, but an awful lot of CPAN etc. is crap 04:17:41 so raw numbers isn't really the way to look at things 04:17:47 i think there are key areas, more. 04:17:48 salex: an awful lot of the lisp libraries that exist is crap, too, no? 04:18:07 salex: allegro actually is not such a bad model. 04:18:17 I'm sure. catching up to the amount of crap perl has isn't a great goal, though 04:18:58 I've been toying with the name 'fusion'. 04:19:13 for my lisp distro with 'batteries included' 04:19:16 the question is, do `we' want/need an all-singing, all-dancing free distribution that is cross platform? at least, enought to put the effort in to get there 04:19:40 it's true that another obstacle to this could be licensing 04:19:45 salex: i don't know if we need it, but i think i'm in a position to make money through it, so i'm willing to give it a go. 04:19:49 Chrononaut [n=bjorn@195.20.207.210] has joined #lisp 04:19:51 salex: i do want a lisp that i can use on "linux" and that can be deployed on "windows" 04:20:40 H4ns: do you need this to be non-commercial though? 04:20:43 salex: obviously, many want to replace "linux" by "osx" and "windows" by "linux" above :) 04:20:55 cross platform is hard 04:21:08 salex: i could go for lispworks, i know. 04:21:13 cross platform with good system & particularly UI integration is hard 04:21:24 H4ns: right. that's what I meant. You have options 04:21:26 couldn't a "posix" oriented one also work on windows with posix compat libraries though? 04:21:31 salex: but then, as i am a windows user and deploy on freebsd, lispbox is rather expensive. 04:21:38 salex: not terribly so. you provide a lowest-common-denominator interface, and people will use it. 04:21:50 this excludes UI details of course 04:22:05 drewc: ok, hard to do well in terms of UI stuff particularly 04:22:19 salex: well, clim. *cough* 04:22:19 salex: cross platform is not so hard for me as i am a server person. i'm not envisioning a cross-platform lisp gui toolkit. i'd be happy if we had one acceptable free gui toolkit for any platform, for that matter. 04:22:58 salex: i mean lispWORKS is expensive for me because i use two platforms 04:22:59 part of it is interest. sbcl is a good compiler, but putting it on windows has been a slog 04:23:16 probably because few sbcler's care 04:24:13 salex: yeah. part of the problem also is that sbcl's threading is terribly ad-hoc, making it a pain to get it to run on windows. 04:24:32 salex: ccl is much nicer there, because it had a multiprocessing model to start with. 04:24:45 I've also not looked into the details but was initially surprised that it wouldn't work on other platforms with pthreads either 04:25:22 i didn't have the impression it was ad-hoc so much as a different model; from the cmucl stuff 04:25:34 never looked that carefully, as I don't care much about threads 04:25:56 but yeah, there are road bumps 04:26:11 salex: neither do i, but i do care about multiprocessing. 04:26:17 but they'll exist for any of them, going cross platform. and someone has to care enough to do the work 04:26:18 cmu's threads are cool, but green. i think people want proper OS threads these days. 04:26:54 salex: It seems like clozure has been doing the work. 04:26:57 drewc: cmu does not have threads, which is what make its multiprocessing facilities a lot more pleasant to use. so i think. 04:27:00 drewc: right. i just meant it was an older system, good for what it was, not necc. what person X wants. But person X (or some proxy) has to shift it. 04:27:06 drewc: agreed, or you'd just use a proper cffi layer over non-blocking I/O operations 04:27:22 phadthai: "just" 04:27:32 it comes along, slowly. but i can see the problem 04:27:35 "only would have to" ? 04:27:47 phadthai: iolib is nice for that, but yeah it requires turning things upsidedown 04:27:50 fundamentally, if I want to use sbcl, i just put it on a box that it works well on, etc. 04:27:59 phadthai: it is the implication of smallness of effort that made me put quotes around the just. 04:28:01 i'm sure lots of people are vaguely in that position 04:28:42 agree in principle it would be nice, but have easier ways to sort things out than doing the hard work themselves (if they could) :) 04:28:43 H4ns: I've used non-blocking I/O in C with BSD sockets and postgresql at least, and it didn't seem that bad, but I admit I didn't play with cffi yet heh 04:29:11 salex: coming from the newbie's perspective, i don't think that cl will be very attractive to newbies if it requires a change of platforms. :) 04:29:37 well, it doesn't. there are cl's on loads of platforms 04:29:43 phadthai: i used cmucl mp a lot, and found it very pleasant to work with. 04:30:01 sellout [n=greg@pool-138-88-27-104.res.east.verizon.net] has joined #lisp 04:30:01 salex: that again being the issue. :) 04:30:15 drewc: well at least there is a standard 04:30:27 -!- sellout [n=greg@pool-138-88-27-104.res.east.verizon.net] has quit [Read error: 104 (Connection reset by peer)] 04:30:31 it's not like were talking about an implmentation-as-standard language 04:30:32 phadthai: the only real-world downside with cmucl mp is that you can't use foreign libraries that do blocking i/o, i.e. mysql or oracle :) 04:30:34 H4ns: does that use a pool of processes? I added a note to check multiprocessing out but didn't use it yet either 04:30:36 salex: the patch of least resistance probably shouldn't start with a 10-pronged fork in the road.. 04:30:39 path 04:30:53 in theory it shouldn't matter much 04:30:59 sellout [n=greg@pool-138-88-27-104.res.east.verizon.net] has joined #lisp 04:31:00 H4ns: oh, so I assume pth-like user threads then 04:31:03 i agree in practice this may not be true :) 04:31:27 it would be really nice to be able to say 'grab foolisp and go'. where foolisp is our mythical lisp distro 04:32:08 right now it's 'grab sbcl if you're on linux, clisp on windows or ccl on mac' 04:32:12 or whatever. 04:32:33 yeah, i understand where you're coming from 04:32:36 this is not a terrible situation, but to an outsider it seems fragmented. 04:32:43 otoh, why don't people say that for gcl? 04:33:14 it's something the drscheme people did nicely but 04:33:30 a) small language and b) they had some resources to throw at it 04:34:43 heh. figured out how to flip the bibliographies around. yikes but bibtex language in opaque 04:36:24 hmm what was drscheme's GUI built around? 04:36:51 drewc: what I mean is, the existence of a portable lisp is only part of that battle 04:37:10 people won't be happy with only, say, perl/pyton performance 04:37:16 salex: true enough. 04:37:17 or lack of native GUI hooks 04:37:23 or ... or .. or... 04:37:24 :) 04:38:06 i thought the lispbox idea was valuable to get by the oh-my-god-fragmentation issue for newbies, myself 04:38:17 lisp-in-a-box, whatever it was 04:39:13 phadthai: you mean the IDE or the PLT toolkit? 04:40:11 salex: the toolkit (since it seems to be available to multiple platforms) 04:40:48 right. I think they wrote that internally, with only basics hooked from the various systems 04:40:55 to be consistent 04:41:02 I never tried gtk on osx but it did run well on posix and windows, it possibly could be a contender if any of the various cl-*gtk* projects works well enough 04:41:05 istr some hooks for other sytems too 04:41:15 salex: ok 04:41:34 hans [n=H4ns@p57A0F7B3.dip.t-dialin.net] has joined #lisp 04:41:49 -!- hans is now known as H4nsX 04:41:52 anything on OSX that is X11 based is somewhat problematic 04:42:46 so gtk on osx doesn't have hooks to the native ui? 04:43:08 not afaik. i'm not current though. 04:43:55 wasn't one of mcl's issues the effort needed to rewrite for cocoa? 04:45:07 I think Apple put too much effort into using GNU, but avoid Linux and creating vendor lock hooks. 04:45:12 MCL's problem was the switch to Intel. 04:45:33 hmm if gtk wasn't a suitable portable toolkit, would using a c++ (i.e. qt or wxwindow) be any more trouble than a C one considering cffi hooks? 04:45:37 rme: yeah, that was the big one. 04:46:03 i know there is a native gtk effort for OSX, i have no idea how well developed 04:46:09 Bzek [n=SK_sj@mcc-dyn-30-186.kosnet.ru] has joined #lisp 04:46:10 MCL's user interface is based on Carbon, which still is supported on 32-bit platforms, but not 64-bit. 04:46:21 phadthai: i'm not sure if i understand your question right, but if you ask if one can use c++ libraries from lisp, then the general answer is "no" 04:46:45 H4nsX: ok this indeed answers greatly :) 04:46:50 rme: right, and they were running up against issues that way before ppc was abandoned, no? 04:47:06 H4nsX: the general answer is the same as for every other language 04:47:13 "c++ sucks for libraries" 04:47:36 qt would be possible phadthai 04:47:43 except perhaps ones tieing over some COM or XPCOM or equivalent 04:47:47 i don't know how close native gtk is on osx 04:47:51 salex: indeed. but there are solutions for other languages (e.g. perl) which are not available for lisp. 04:48:12 salex: I think so. AFAIK, there's wasn't any way to use Cocoa from MCL, but I could be wrong about that. 04:48:25 rme: right, that was my impression, but i don't really know 04:48:32 in any case, intel was the last straw 04:49:03 And they needed 32-bit Intel, too, and that's a hard port. 04:49:18 issue with OSX, moreso that windows, is that users won't be happy with gtk/qt/wxwindows/whatever, for the most part 04:49:30 rme: yeah, although now you could blow that off 04:49:41 rme: "he said, knowingly". congrats for having that done! :) 04:49:42 but they were screwed at the time 04:50:40 *rme* accidentally fishes for compliments 04:51:27 -!- O_4 [n=souchan@ip-118-90-67-180.xdsl.xnet.co.nz] has quit [] 04:51:44 salex: hmm then what alternative would there be to a native cocoa, since it must anyhow be portable thus abstracted through some generic toolkit... 04:52:02 phadthai: there isn't really. which is part of the problem 04:52:20 if you want to do mac UI `right', it's not going to be portable 04:52:34 which mac users seem more sticky about than others 04:52:41 osx users of free software wouldn't accept the compromise? 04:52:56 yeah, they do. 04:53:01 and it's less prevalent. 04:53:18 but if you were writing something commercial, it could be a problem. 04:54:05 many popular free software systems are ported, then immediately fork a `mac like' version 04:54:12 eg. gimp, emacs, etc. 04:54:20 yes. basically GTK is really jarring and ugly on the Mac while it looks fine on Linux and Windows. 04:54:42 mac users also don't like java interfaces, similar reasons 04:54:46 i don't believe that ported free software is very popular among mac users. 04:55:00 some is 04:55:08 at least i, with my mac user hat on, use only native stuff. 04:55:10 but shareware has a long successfull history there too 04:55:15 yeoh [n=chatzill@24.48.48.60.kmr03-home.tm.net.my] has joined #lisp 04:55:17 it's a different market 04:55:23 it's a market. 04:55:24 :) 04:55:37 and, like i said, not one very freindly to lowest-common-denom UI 04:55:45 whereas "free software" is more marketing than market 04:55:54 anyway, if you want max Mac support you need to be able to use Cocoa on your Lisp running on the Mac. 04:56:06 no, i mean mac is different from other platforms, as a market 04:56:14 Adamant_: or at least carbon 04:56:27 salex: Carbon is a dead end 04:56:32 they're killing it 04:56:44 with the move to 64-bit 04:56:53 all singing all dancing all Cocoa all the time 04:57:08 just think what would have been possible if the gtk folk at the beginning had not only avoided c++ but also avoiding rolling their own half-assed object system 04:57:24 Adamant_: that's the roadmap yes. mcl did nicely on carbon for a while 04:57:27 -!- lukego [n=lukegorr@cpe-76-87-84-57.socal.res.rr.com] has quit [] 04:57:28 salex: what would they have done instead? 04:57:36 (re gtk: and used objective-c instead) 04:57:48 would have made OSX integration easier ;) 04:57:59 objective C wasn't much used anywhere except on next and mac though :) 04:58:13 we could all be living in a *Step compatible GUI work 04:58:13 yes. but it was a decent design 04:58:24 unlike gtk objects ;) 04:58:45 but i'll hand it to them for avoiding the c++ swamp 04:58:51 yeah I'm not really advocating gtk either, I find it bloated actually 04:59:04 Obj-C is OK 04:59:12 i think that (the object design) was part of the problem, phadthai 04:59:18 it's got a beat and you can dance to it 04:59:32 fostered a "lets do everything in house" feel that went a bit overboard 04:59:35 salex: possibly 04:59:51 fuck it. we'll do it live. 05:00:13 what would people think about a really decent CLIM targetted to various back ends? 05:00:19 yhara [n=yhara@84.215.12.221.megaegg.ne.jp] has joined #lisp 05:00:24 other than every non lisper thinking it was insane, i mean 05:00:24 salex: \o/ 05:00:33 :)) 05:03:22 I think anyone developing a GUI toolkit in this era should have some really snappy design and usability people working with them. 05:03:36 lukego [n=lukegorr@cpe-76-87-84-57.socal.res.rr.com] has joined #lisp 05:03:38 you'd think. doesn't seem to be the case 05:03:45 -!- lukego [n=lukegorr@cpe-76-87-84-57.socal.res.rr.com] has quit [Client Quit] 05:03:52 -!- isomer`` is now known as isomer 05:04:07 even apple, who were miles ahead, are in many ways doing a worse job of that than 10 years ago 05:04:38 everyone else is at best stuck around the same time, but with more sparkly bits 05:04:48 salex: I tend to be skeptical of this. 05:05:05 of which? 05:05:19 that's doing a job that is in any meaningful way worse 05:06:06 Adamant_: i can't see how the mac os x ui is significantly "better" than others 05:06:30 OS 9 lacked a certain bit of usability called "crashing less than Windows ME" 05:06:43 Adamant_: same desktop metaphore, same lack of model for asynchronous activities, same overlapping window mess, same desktop metaphor. 05:06:51 that was a technical issue, not a design issue 05:07:17 Adamant_: and for me, the lack of decent default keyboard bindings is something that makes mac a lot less usable than windows, specifically. 05:07:24 OS 9 still followed the HIG and was very consistent 05:07:40 salex: it ultimately was a usability issue. 05:07:54 you can disagree with the design choices, but it was a design at least 05:08:03 windows was a mess at the same time 05:08:07 an linux, forget it 05:08:08 Windows XP was dramatically more useable just by not falling over all the time. 05:08:16 i agree uptime is good 05:08:30 but the point is, nobody is making an effort that way really, these days 05:08:34 for better or worse 05:08:40 salex: for uptime? 05:08:51 no, for interface and designe 05:08:53 ah 05:09:02 I think OS X is good, personally. 05:09:03 apple is less consistent than it was 05:09:15 honestly I never saw why people loved OS 9. 05:09:38 windows has picked up the mid 90s tricks 05:09:50 across all platforms it's just a bit mushy 05:10:11 O_4 [n=souchan@ip-118-90-67-180.xdsl.xnet.co.nz] has joined #lisp 05:10:18 Adamant_: people loved it because of the consistency. Pick up a new app, you already knew how to make it do a bunch of stuff 05:10:22 -!- creddy [n=CrazyEdd@wrongplanet/CrazyEddy] has quit [Connection timed out] 05:10:41 to a degree never realized in windows or unix. 05:11:26 I guess. I never saw it as all that dramatically much more usable then Win95 and beyond. 05:11:54 everyone has got a bit better in some direction since then, of course. no arguing. i was just reacting to your idea that people these days should have snappy design and UI people working with them 05:12:02 yeah 05:12:02 they *should* 05:12:12 but there is no evidence of it 05:12:15 well, basically 05:12:22 if you're going to do a CLIM or someone is going to 05:12:24 actually, to be more fair -- the evidence is they don't listen to them 05:12:28 coming from the amiga back then, pre-osx macos seemed primitive to me but I did see the great consistency between applications however 05:12:33 they've got to compete on something. 05:12:45 right now it's just "works with CL" 05:12:50 phadthai: yeah, but 8 or 9 it was pretty creaky 05:12:51 i.e. a standard menu editor would work with all applications and all 05:13:11 -!- yeoh [n=chatzill@24.48.48.60.kmr03-home.tm.net.my] has left #lisp 05:13:24 there are good interface people working on particular applications, obviously 05:13:30 but that has it's own problems 05:13:36 divergence, and wierd trends 05:13:53 phadthai: I spent time on the Amiga. it's too bad the Amiga fanboys weren't as rabid as the Apple ones. 05:13:59 remember when all of a sudden all the apps were going to internal windowing? 05:14:08 and then back out again? 05:14:28 amiga fanboys were at least as rabit as apple ones 05:14:38 it's just hte company had less going for it, is all 05:14:41 nice hardware though 05:14:45 salex: not really, if they were Amiga would still be alive 05:14:46 -!- Chrononaut [n=bjorn@195.20.207.210] has quit [Remote closed the connection] 05:14:49 Chrononaut [n=bjorn@195.20.207.210] has joined #lisp 05:14:55 I still wasn't very gui centric though, my startup basically opened a fullscreen shell heh 05:14:59 Apple would have died if not for the fanatics 05:14:59 Adamant_: no, it wasn't the lack of fanboys 05:15:11 apple had a couple of good commercial niches 05:15:20 that were getting demolished 05:15:35 lasted out there more that amiga (that had what, the video toaster?) 05:15:41 yup 05:15:46 yeah 05:15:49 oh, and they had a better CEO once he got back ;) 05:15:57 and a few other custom hardware solutions with decent software but which never were marketted properly 05:15:58 but really, apple survied on pros for a long time 05:16:01 not fanboys 05:16:08 and commodore never helped heh 05:16:09 and then it was saved by the ipod 05:16:32 salex: by the mid-90's you could do pretty much anything you could on a Mac, on a Windows machine 05:16:33 phadthai: right. just never came together for them 05:16:36 I used to work for a local amiga store and the store wasn't even helped when it had to advertize on tv 05:16:46 Adamant_: doesn't matter. 05:17:05 salex: it does, the pros were abandoning the faith in droves 05:17:09 you had design studios and post production etc. 05:17:13 that didn't want to leave 05:17:22 they slowly bled out, but they didn't want to 05:17:32 switching platforms is a dead loss for them 05:17:39 nah 05:17:44 they can be standard 05:17:51 it's only when mac actually fell behind notably on hardware that this started to happen 05:18:02 nobody gives a shit if their a grand or two more 05:18:11 lukego [n=lukegorr@cpe-76-87-84-57.socal.res.rr.com] has joined #lisp 05:18:17 they cared when they were 1/3 the speed though 05:19:00 also, I debate that the "pros" did not have a mega contingent of fanboys 05:19:24 `standard' my ass. I talked to a manager with a room full of graphic design people, who figured the switch to windows for mostly photoshop cost him a hundred grand or so 05:19:28 same software 05:19:53 so the attitude is don't do it unless you have too 05:19:58 *snicker* Haha windows 05:20:08 except Adobe software went to Windows and they keep the latest and greatest there and have for a while. 05:20:11 anyway, shops like that got apple through a bunch of lean years 05:20:27 Adamant_: right, which is why these guys moved over 05:20:47 but like I said, in theory it's no change, in practice it cost them a hundred grand or mor 05:20:56 nobody does that for fun 05:21:14 how much would buying new Apples and new licenses for the new Photoshop cost? 05:21:21 no, no no 05:21:23 I'm guessingmore. 05:21:36 i meant a hundred grand in lost hours 05:21:40 ahhh 05:21:48 hardware + software not counted 05:21:52 ah ok 05:21:54 that was his dead loss from the switch 05:21:55 transition cost 05:22:05 yup, i should have been more clear 05:22:17 so these shops knew it would be a bite, 05:22:26 how much were his costs after not paying the Apple tax? 05:22:40 that was factored in 05:22:44 ah ok 05:23:01 hmm so about clim, it would help to unify the ui and the porting effort, yet would newbies consider the learning curve for it too high to use it? Or wouldn't it make them feel that they're using a too old system? 05:23:23 design studios, recording studios, etc. loads of them stayed on apple long after others gave up on them. thats what amiga didn't have 05:23:30 not that I care about "gui fashion", but considering all those projects redoing their site mainly for look to not seem outdated :) (part of PR I guess :) 05:23:47 phadthai: Isn't CL a too old a system in the first place? ;) 05:23:53 phadthai: yeah, that's a question. squeak faces the same think, i guess 05:23:57 schme_: :) 05:24:24 salex: people used Video Toaster for fucking ever, but there weren't enough of them, 05:24:43 Squeak's main problem is not playing nice with the OS 05:24:51 yeah, that's what I said. Apple had a bunch of niches, amiga had the toaster 05:24:56 it wasn't enough 05:25:11 no I do not want to use your nonsense GUI VM that only interoperates with itself 05:26:15 your language is nice, beautiful even, but I do not want to subscribe to your newsletter even if you are a genius, Mr. Kay 05:26:24 I have seen custom hardware + software 24-bit graphics editing software solutions at the time other than video toaster also, and at least one expensive software competing with autocad, a few good desktop publishing applications 05:26:41 but none of them survived either 05:28:12 Adamant: it would be a bit boring if every programming environment in the world adapted itself to Unix, though :) 05:28:18 imho at least :) 05:28:53 phadthai: it's easy to underestimate the effect of a good piece of niche softwar 05:28:59 lukego: true, but I think Smalltalk would have done a lot better if there was a more conventional compliment to Squeak with the same level of push. 05:29:09 they're trying to do that right now IIRC 05:29:24 someone is doing an in-browser thingy. could be interesting 05:29:27 iirc 05:29:44 Smalltalk to Javascript? 05:30:08 smalltalk-like in javascript, iirc. i forget the details. 05:30:31 Squeak was a research system so bums on seats at the expense of novelty would've been a lose for them. VisualWorks etc could be another story 05:30:49 mmhmm 05:30:50 mulligan [n=user@e178031213.adsl.alicedsl.de] has joined #lisp 05:31:16 Smalltalk hackers seem to enjoy Javascript itself to a fair extent 05:31:25 it's lispy :) 05:31:30 lukego: would it have killed them to have core libs someone could have stripped to make a more conventional Smalltalk? 05:31:46 it's a treat between bouts of writing PHP 05:32:09 doc, it hurts when I do that 05:32:15 *this 05:32:58 javascript is exciting because you get to hack at fx (and make pretty things) 05:33:08 I've used spidermonkey for a few applications and liked it, comming from C, after writing the necessary C-JS interface code I needed 05:33:18 I haven't used it for client-side web though 05:34:03 I can't say it was high performance though, but the tracing jit they're introducing might help 05:35:15 I need to mess around with parenscript. The idea of having to use DO is a bit scary. Heh. 05:35:35 loop junkie ;) 05:35:47 actually, map/reduce junkie >_> 05:36:21 *S11001001* is becoming series junkie 05:36:31 series? 05:36:45 minion: series+ 05:36:46 Sorry, I couldn't find anything in the database for ``series+''. 05:36:50 minion: series 05:36:51 series: Series is a Library for operating on series, a data structure similar to a sequence. http://www.cliki.net/series 05:37:18 hmm is that to cover the problem of extending the sequence class? :) 05:37:27 or rather subclassing sequence 05:37:32 I might find that useful 05:37:34 -!- leogeo [n=greg@ip68-11-196-50.br.br.cox.net] has quit ["rcirc on GNU Emacs 23.0.60.1"] 05:37:53 no, it's main neat thing is transforming mapping operations into pure loops 05:38:42 Can you use them with parenscript? >_> 05:38:44 <_< 05:38:47 no 05:38:51 sucks 05:39:44 so, for example, (map-fn t #'print (catenate (scan 'list '(1 2 3)) (scan-range :from 4 :upto 6))) in a non-valuing context transforms to (progn (mapc #'print '(1 2 3)) (loop for n from 4 to 6 do (print n))) 05:39:57 well the expansion is way more complicated but that's the general idea 05:41:07 ah 05:41:55 of course the real wild part is when it transforms (let ((fourplus (scan-range :from 4 :upto 6))) (map-fn t #'print (catenate (scan 'list '(1 2 3)) fourplus))) to exactly the same code 05:43:32 biz897 [n=badon@c-71-195-215-153.hsd1.ut.comcast.net] has joined #lisp 05:43:54 vmCodes [n=vmCodes@122.166.154.177] has joined #lisp 05:53:08 vy [n=user@213.139.194.86] has joined #lisp 05:54:49 -!- yhara [n=yhara@84.215.12.221.megaegg.ne.jp] has quit [Read error: 113 (No route to host)] 05:56:35 arrenlex [n=em@S01060040f43406d4.ed.shawcable.net] has joined #lisp 05:56:50 What does the reduce special form do? 05:58:37 clhs reduce 05:58:37 http://www.lispworks.com/reference/HyperSpec/Body/f_reduce.htm 05:58:44 (not that it is a special form) 06:00:00 haiwei [n=haiwei@221.216.67.56] has joined #lisp 06:01:16 beach [n=user@ABordeaux-158-1-62-126.w90-16.abo.wanadoo.fr] has joined #lisp 06:01:23 good morning 06:02:04 Thank you, H4nsX. 06:04:33 arrenlex: `reduce' is an ordinary function. 06:08:19 xjrn [n=jim@c-69-181-213-99.hsd1.ca.comcast.net] has joined #lisp 06:08:47 yhara [n=yhara@84.215.12.221.megaegg.ne.jp] has joined #lisp 06:08:55 mrsolo_ [n=mrsolo@adsl-68-126-209-194.dsl.pltn13.pacbell.net] has joined #lisp 06:08:59 CrazyEddy [n=CrazyEdd@220-253-27-65.VIC.netspace.net.au] has joined #lisp 06:09:53 -!- jtoy [n=jtoy@58.61.213.234] has quit [] 06:17:33 http://bauhh.dyndns.org:8000/mcclim/screenshots/mcclim-multilingual.png clim can look better than I initially envisionned :) 06:18:41 phadthai: You must have very limited imagination. 06:20:06 heh 06:20:30 -!- papermachine [n=ahoman@61.152.106.169] has left #lisp 06:21:12 jtoy [n=jtoy@74.85.13.56] has joined #lisp 06:21:23 sudoer [n=jtoy@58.61.213.234] has joined #lisp 06:24:49 papermachine [n=ahoman@61.152.106.169] has joined #lisp 06:25:10 phadthai: I've seen some nice pictures of McClim in action (Climacs uses McClim, right?) 06:25:21 You bet! 06:25:52 -!- mrsolo [n=mrsolo@nat/yahoo/x-62a6bad59f62654b] has quit [Read error: 110 (Connection timed out)] 06:26:05 yes I checked a number of various screenshots just now 06:26:16 hm. i can't seem to find the screenshots I remember... 06:26:57 my initial impression was that clim was quite limited because of the little exposure I had seen, which looked like a somewhat enhanced curses heh 06:27:17 phadthai: http://mcclim.cliki.net/Screenshot 06:27:26 lots of nice stuff around that page 06:27:29 here is another example: http://common-lisp.net/project/gsharp/rapsoden-sjunger.png 06:27:35 yes this is what I looked at 06:27:49 (the http://mcclim.cliki.net/Screenshot list that is) 06:28:24 http://jlr.freeshell.org/data/mcclim/screenshots/2007-03-27-listener-dark-classgraph-context-menu.png That's really sexy. 06:28:42 I had seen screenshots of another toolkit named garnet or such before also which seemed decent 06:28:44 I hope climacs gets more support. I'd really like to help, as well. 06:29:06 sykopomp: you can probably just dig in and improve it. 06:30:12 beach: I'll probably try to as soon as I find time. I'm already working on 4 projects (sort of), and taking 3 classes :( 06:30:20 sykopomp: also http://bauhh.dyndns.org:8000/mcclim/screenshots/clim-rocks.png 06:30:24 anyone here used Xach's vecto and/or ttf loader on a mac? 06:30:44 doesn't Xach develop on a mac laptop?... 06:31:02 -!- arrenlex [n=em@S01060040f43406d4.ed.shawcable.net] has left #lisp 06:31:12 could be .. .but he's not here looks like 06:33:15 anyway, i can't get it to load a font 06:34:14 beach: can you tell me some nice things about mcclim? Anything in particular about it that's really nice besides being written in lisp? 06:35:33 sykopomp: I can tell you plenty. For one thing it's one of the few toolkits that does not make it necessary to use event-driven programming, because CLIM gives the initiative back to the program. This means you can execute commands in environments other than just the global one. 06:35:52 This in turn makes things like CLIM presentations possible, which is another unique features. 06:36:10 hmmm >_> 06:36:32 Presentations make your code more modular becuase you don't have to thing about *how* the objects are going to be used by other modules, just what their presentation types are. 06:37:28 Then you have other very nice stuff like incremental redisplay, graph drawing, etc, etc. 06:37:38 sounds nifty... 06:37:45 It really is. 06:37:51 probably nicer than dealing with cells or such, I imagine. 06:37:58 This is why I started working on McCLIM after reading the CLIM spec. 06:38:10 (and also because I needed a good toolkit in Lisp for Gsharp) 06:39:10 -!- jtoy [n=jtoy@74.85.13.56] has quit [Connection timed out] 06:39:39 is Gtkairo still troublesome? 06:40:47 actually, bedtime. 06:41:04 beach: I'll play with mcclim at some point, I imagine. Thanks for the info, it looks and sounds interesting :) 06:41:04 'night sykopomp 06:41:20 Good! 06:42:50 -!- ths_ is now known as ths 06:42:53 huh, that was nice. Took about 10 min to port my little chord drawing thingy over to Xach's vecto, including time to find and install the packages and mess about with fonts 06:45:33 can't really complain about that! 06:46:33 what did you use before? 06:46:38 cl-pdf 06:46:45 i wondered how similar they were 06:46:50 so i tried it 06:46:50 morning 06:46:53 salex: you're up late 06:46:56 hello Krystof 06:47:03 Krystof: yeah, just heading to bed 06:47:17 working on annoying application package 06:47:31 jgracin [n=jgracin@82.193.208.195] has joined #lisp 06:47:42 though i just took 10 min to move that chord diagram thingy over to Xach's vecto package 06:48:32 I added in my notes to read the CLIM spec, but now also have to head to bed 06:48:33 bbl 06:48:50 as a final note: http://web.archive.org/web/20070127022728/http://lists.midnightryder.com/pipermail/sweng-gamedev-midnightryder.com/2005-August/003804.html 06:48:55 i really hate cascading parens 06:49:33 Krystof: are you planning on mucking about with the hmm stuff this weekend? 06:49:41 "plan" is too strong a word 06:49:53 understood 06:50:00 I might implement my sampling. I still haven't hooked up brand's prior, either 06:50:58 k. i'm probably stuck with writing research statements etc. (it's due early next week). If i have a chance to play with it, i'll find you here or email if i come up with anything interesting 06:51:57 i forget, do you have baum-welch implemented? 06:52:41 yes 06:53:02 I recently implemented multiple observation sequences for training, too 06:53:10 maybe you'd like an up-to-date copy 06:53:33 thought so. ah, good. yes please, but no rush. i'm about to collapse really 06:54:18 so i'll call it a night. if you have a chance, email it? 06:54:25 okay 06:54:45 -!- Adamant_ [n=Adamant@unaffiliated/adamant] has quit [] 06:54:52 thanks. 06:54:54 night all 06:55:00 'night salex 06:55:07 -!- salex [n=user@216.80.143.240] has quit ["ERC Version 5.2 (IRC client for Emacs)"] 06:55:38 -!- H4nsX [n=H4ns@p57A0F7B3.dip.t-dialin.net] has quit [] 06:58:27 Adamant [n=Adamant@c-98-244-152-196.hsd1.ga.comcast.net] has joined #lisp 07:01:38 ehu [i=52aa21ad@gateway/web/ajax/mibbit.com/x-2af466fa5604bffe] has joined #lisp 07:01:48 -!- rme [n=rme@pool-70-105-121-210.chi.dsl-w.verizon.net] has quit [] 07:02:54 -!- pstickne [n=pstickne@c-71-236-177-238.hsd1.wa.comcast.net] has quit [Read error: 110 (Connection timed out)] 07:06:35 jtoy [n=jtoy@74.85.13.60] has joined #lisp 07:09:16 -!- sohail [n=Sohail@unaffiliated/sohail] has quit [Read error: 113 (No route to host)] 07:16:02 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [No route to host] 07:16:56 -!- sudoer [n=jtoy@58.61.213.234] has quit [Read error: 110 (Connection timed out)] 07:21:10 -!- lukego [n=lukegorr@cpe-76-87-84-57.socal.res.rr.com] has quit [Read error: 110 (Connection timed out)] 07:21:30 -!- ehu [i=52aa21ad@gateway/web/ajax/mibbit.com/x-2af466fa5604bffe] has quit ["http://www.mibbit.com ajax IRC Client"] 07:23:18 dialtone [n=dialtone@adsl-67-121-124-157.dsl.pltn13.pacbell.net] has joined #lisp 07:34:22 elurin [n=user@85.99.132.149] has joined #lisp 07:39:45 -!- papermachine [n=ahoman@61.152.106.169] has quit [Client Quit] 07:43:58 -!- jgracin [n=jgracin@82.193.208.195] has quit ["Leaving"] 07:47:06 -!- enigmus [n=e@S0106001d7e52d1d9.vc.shawcable.net] has quit [Read error: 110 (Connection timed out)] 07:48:00 hey, I found a practical use for change-class! 07:48:15 sorry if a little excited, I never expected it to happen 07:48:21 outside metaprogramming situations 07:51:20 -!- araujo [n=araujo@gentoo/developer/araujo] has quit ["Leaving"] 07:55:12 x6j8x [n=x6j8x@77.21.43.186] has joined #lisp 08:00:26 travisjeffery [n=eatsleep@user64-159.vicres.utoronto.ca] has joined #lisp 08:01:36 edon [n=edon@82.114.94.2] has joined #lisp 08:01:52 sudoer [n=jtoy@58.61.213.234] has joined #lisp 08:02:59 manuel_ [n=manuel@HSI-KBW-085-216-060-145.hsi.kabelbw.de] has joined #lisp 08:03:53 esden___ [n=esdentem@91-67-156-25-dynip.superkabel.de] has joined #lisp 08:06:38 -!- x6j8x [n=x6j8x@77.21.43.186] has quit [] 08:07:32 kiuma [n=kiuma@ip-126-26.sn1.eutelia.it] has joined #lisp 08:08:29 hello lispers 08:08:40 'morning kiuma 08:09:33 lukego [n=lukegorr@cpe-76-87-84-57.socal.res.rr.com] has joined #lisp 08:13:47 -!- O_4 [n=souchan@ip-118-90-67-180.xdsl.xnet.co.nz] has quit [] 08:18:00 Ogedei [n=user@78.52.230.47] has joined #lisp 08:19:58 -!- jtoy [n=jtoy@74.85.13.60] has quit [Connection timed out] 08:21:46 -!- dmiles_afk [n=dmiles@c-67-168-159-175.hsd1.wa.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 08:22:58 schasi [n=schasi@p54A255FD.dip.t-dialin.net] has joined #lisp 08:24:31 lukego_ [n=lukegorr@cpe-76-87-84-57.socal.res.rr.com] has joined #lisp 08:24:36 -!- lukego [n=lukegorr@cpe-76-87-84-57.socal.res.rr.com] has quit [Read error: 104 (Connection reset by peer)] 08:24:46 slash_ [n=Unknown@p4FEAFFA0.dip.t-dialin.net] has joined #lisp 08:27:02 -!- gigamonkey` [n=user@adsl-99-179-47-191.dsl.pltn13.sbcglobal.net] has quit [Read error: 110 (Connection timed out)] 08:27:54 -!- jmbr_ [n=jmbr@87.223.197.81] has quit [Read error: 104 (Connection reset by peer)] 08:28:07 Hun [n=Hun@p4FCF5364.dip.t-dialin.net] has joined #lisp 08:28:25 besiria [n=user@ppp083212084224.dsl.uom.gr] has joined #lisp 08:38:33 -!- vmCodes [n=vmCodes@122.166.154.177] has left #lisp 08:40:18 -!- manuel_ [n=manuel@HSI-KBW-085-216-060-145.hsi.kabelbw.de] has quit [Read error: 110 (Connection timed out)] 08:47:04 -!- besiria [n=user@ppp083212084224.dsl.uom.gr] has quit [Remote closed the connection] 08:47:52 -!- yhara [n=yhara@84.215.12.221.megaegg.ne.jp] has quit [Read error: 104 (Connection reset by peer)] 08:47:56 yhara [n=yhara@84.215.12.221.megaegg.ne.jp] has joined #lisp 08:51:06 Fufie [n=punchbal@86.80-203-225.nextgentel.com] has joined #lisp 08:52:52 jmbr [n=jmbr@87.223.197.81] has joined #lisp 09:00:07 S11001001: you can use it for something like FREE-INSTANCE, http://groups.google.com/group/comp.lang.lisp/msg/26fbc8ff52fc02e4 09:00:13 S11001001: what's your use case? 09:00:23 -!- Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has quit [Read error: 113 (No route to host)] 09:02:35 planet lisp is now featuring financial analysis? 09:10:47 obsethryl [n=llort@unaffiliated/obsethryl] has joined #lisp 09:11:40 chris2 [n=chris@p5B16BFE0.dip0.t-ipconnect.de] has joined #lisp 09:13:51 dmiles_afk [n=dmiles@c-67-168-159-175.hsd1.wa.comcast.net] has joined #lisp 09:14:47 -!- sudoer [n=jtoy@58.61.213.234] has quit [] 09:15:31 Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has joined #lisp 09:15:47 H4ns1 [n=hans@p57A0FB73.dip.t-dialin.net] has joined #lisp 09:19:17 ecraven [n=nex@plc31-103.linzag.net] has joined #lisp 09:19:17 -!- dash_ [n=dash@dslb-084-057-019-145.pools.arcor-ip.net] has quit [Read error: 104 (Connection reset by peer)] 09:21:43 name [n=name@sburn/devel/name] has joined #lisp 09:22:49 dash_ [n=dash@dslb-088-065-136-090.pools.arcor-ip.net] has joined #lisp 09:28:10 birdsbite [n=mark@75.110.164.248] has joined #lisp 09:34:26 -!- H4ns [n=hans@p57A0F7B3.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 09:38:19 tst__ [n=Tim@p4FD2F399.dip.t-dialin.net] has joined #lisp 09:40:00 -!- edon [n=edon@albalug/edon] has quit [Read error: 60 (Operation timed out)] 09:41:38 -!- elurin [n=user@85.99.132.149] has quit [Remote closed the connection] 09:41:54 elurin [n=user@85.99.132.149] has joined #lisp 09:43:54 O_4 [n=souchan@ip-118-90-67-180.xdsl.xnet.co.nz] has joined #lisp 09:44:05 -!- awayekos is now known as anekos 09:44:29 hans_ [n=H4ns@p57A0FB73.dip.t-dialin.net] has joined #lisp 09:46:01 ivanst [i=ivans@78-0-58-77.adsl.net.t-com.hr] has joined #lisp 09:46:48 ahem... is there a way to do "order by foo descending" with clsql? googling fails to find anything relevant... 09:47:18 it seems that there was previously an :order-by-descending keyword for select, but it was removed. can it be done in any other fashion? 09:50:49 -!- pchrist [n=spirit@gentoo/developer/pchrist] has quit ["leaving"] 09:51:42 pchrist [n=spirit@gentoo/developer/pchrist] has joined #lisp 09:52:32 nikodemus [n=nikodemu@cs181229041.pp.htv.fi] has joined #lisp 09:58:47 Adamant [n=Adamant@c-98-244-152-196.hsd1.ga.comcast.net] has joined #lisp 09:59:34 kib2 [n=chatzill@bd137-1-82-228-159-28.fbx.proxad.net] has joined #lisp 10:01:35 -!- elurin [n=user@85.99.132.149] has quit [Read error: 110 (Connection timed out)] 10:04:34 sqvirt [n=sqvirt@c-24-16-244-102.hsd1.wa.comcast.net] has joined #lisp 10:05:47 malumalu [n=malu@hnvr-4dbb946e.pool.einsundeins.de] has joined #lisp 10:06:02 Is there a function like an input version of format that can read structured data and assign values to variables? (like scanf() in C) 10:06:37 sqvirt: Can you given an example? 10:06:38 read + destructuring-bind? 10:06:50 (which is almost but not completely unlike scanf) 10:07:18 Hun: cl-ppcre can be more handy, in terms of scanf. 10:07:36 lesho [n=uzivatel@lesho.intrak.tuke.sk] has joined #lisp 10:07:41 hi 10:07:47 true, but that depends on what `structured' is 10:08:39 vy: I have a file on which each line is a record, the first value, (a special one) is separated from the others by a comma, but all the rest of the values on the line are separated by spaces. I'm trying to figure out a way to separate the values. The fact that there are 2 kinds of delimiters in each line is making it harder. 10:09:17 sqvirt: See CL-PPCRE:SPLIT. 10:09:26 vy: ahh, OK 10:09:31 vy: thanks! 10:09:35 ,cl-ppcre 10:09:39 ??cl-ppcre 10:09:41 or even cl-ppcre:scan-to-s-trings 10:09:45 ?cl-ppcre 10:09:54 -!- hans_ is now known as H4nsX 10:10:22 edon [n=edon@82.114.94.0] has joined #lisp 10:12:38 mathrick [n=mathrick@users177.kollegienet.dk] has joined #lisp 10:12:46 seelenquell [n=seelenqu@pD9E4792B.dip.t-dialin.net] has joined #lisp 10:13:29 ? cl-ppcre 10:13:31 ?? cl-ppcre 10:13:52 vy: can you explain what you are trying to achive? 10:14:00 maybe you want to talk to minion? 10:14:03 minion: cl-ppcre 10:14:04 cl-ppcre: CL-PPCRE is a portable, Perl-compatible regular expression library by Edi Weitz. http://www.cliki.net/cl-ppcre 10:15:30 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [No route to host] 10:19:48 -!- sqvirt [n=sqvirt@c-24-16-244-102.hsd1.wa.comcast.net] has quit ["Leaving"] 10:27:03 -!- O_4 [n=souchan@ip-118-90-67-180.xdsl.xnet.co.nz] has quit [] 10:28:12 -!- kib2 [n=chatzill@bd137-1-82-228-159-28.fbx.proxad.net] has quit ["ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]"] 10:28:57 O_4 [n=souchan@ip-118-90-67-180.xdsl.xnet.co.nz] has joined #lisp 10:31:09 sqvirt: Xof's (setf format) can probably be considered the equivalent of scanf 10:32:52 tcr [n=tcr@host145.natpool.mwn.de] has joined #lisp 10:34:46 -!- yhara [n=yhara@84.215.12.221.megaegg.ne.jp] has quit [Read error: 113 (No route to host)] 10:37:09 michaelw: You refer to this: http://jcsu.jesus.cam.ac.uk/~csr21/format-setf.lisp 10:37:27 _deepfire [n=deepfire@80.92.100.69] has joined #lisp 10:38:50 Odin- [n=sbkhh@s121-302.gardur.hi.is] has joined #lisp 10:40:07 stassats` [n=stassats@ppp78-37-151-232.pppoe.avangarddsl.ru] has joined #lisp 10:40:41 Yuuhi [n=user@p5483BFE3.dip.t-dialin.net] has joined #lisp 10:43:03 michaelw: http://failex.blogspot.com/2008/10/practical-use-for-change-class.html 10:43:24 (setf stol '((PETER ((A H) (K D)))(IVAN ((7 S) (9 D) (9 H))) (JOZEF ((7 D) (A S))))) 10:43:25 ivanst: yes, just a sec 10:43:29 sry 10:44:04 ivanst: do you have the sql [ syntax installed? 10:44:09 S11001001, indeed 10:44:30 okay, pass :order-by '([column-name] :desc) to select 10:44:35 you can also do like :limit 1 10:44:41 ah... 10:44:44 i didn't try that one! 10:44:59 i tried [* -1 [table field]] and similar :-] 10:45:12 *ivanst* goes to try S11001001's solution 10:45:45 there are more, like multiple columns; the CLSQL documentation includes examples of all this on the SELECT page 10:46:07 -!- schasi [n=schasi@p54A255FD.dip.t-dialin.net] has quit ["Verlassend"] 10:46:09 well I guess you'd have to use [slot-value ...] for multiple tables 10:46:21 if there are column name conflicts 10:46:26 -!- O_4 [n=souchan@ip-118-90-67-180.xdsl.xnet.co.nz] has quit ["Come alive!"] 10:47:25 at the cursory examination of the select documentation, i noticed no mention of that way of avoiding "conflicts" 10:48:10 it's part of the SQL [ syntax, not SELECT 10:49:01 whether you need it depends on what joins you're using 10:53:30 -!- hsaliak [n=hsaliak@cm73.sigma230.maxonline.com.sg] has quit [Read error: 110 (Connection timed out)] 10:53:42 envi^laptop [n=envi@220.121.234.156] has joined #lisp 10:54:35 -!- envi^laptop [n=envi@220.121.234.156] has quit [Client Quit] 10:56:25 -!- abeaumont [n=abeaumon@253.pool85-54-99.dynamic.orange.es] has quit [Read error: 110 (Connection timed out)] 10:57:05 abeaumont [n=abeaumon@166.pool85-54-98.dynamic.orange.es] has joined #lisp 10:59:02 -!- stassats [n=stassats@wikipedia/stassats] has quit [Read error: 110 (Connection timed out)] 11:01:25 darx [n=darx@82-37-17-145.cable.ubr03.wolv.blueyonder.co.uk] has joined #lisp 11:01:46 kib2 [n=chatzill@bd137-1-82-228-159-28.fbx.proxad.net] has joined #lisp 11:02:03 hi, can someone suggest some resources on recursion removal. i'm specifically looking for a general overview of all the techniques. thanks.. 11:04:09 you want a general algorithm for transforming a recursive algorithm into an iterative one? 11:05:13 darx: Did you check graph theory literature stuff? 11:06:06 nope 11:07:27 should i? 11:07:29 That's a common problem in shortest path and travelling salesman algorithms. 11:07:33 froog_ [n=david@87.192.28.247] has joined #lisp 11:08:29 hmm 11:08:39 will do thanks.. 11:09:23 elurin [n=user@85.99.132.149] has joined #lisp 11:09:27 But maybe you better should tell us more about what exactly you need. 11:12:55 I don't need anything specific. It's common knowledge that recursive functions are resource hungry and that refactoring for performance requires their removal. 11:13:25 heh 11:13:31 Ah you mean, recursive function calls... 11:13:44 Yup. 11:13:48 Nshag [i=user@Mix-Orleans-106-4-243.w193-248.abo.wanadoo.fr] has joined #lisp 11:13:51 darx: i'd rather say that's an urban myth :) 11:14:03 have you ever heard of "tail call optimization"? 11:14:33 I have. But I'm asking where can I read more about such? 11:14:58 library.readscheme.org has some papers, i'd guess 11:15:03 I'm not a native English speaker. I hope i'm making sense. 11:15:41 thanks 11:15:49 yhara [n=yhara@84.215.12.221.megaegg.ne.jp] has joined #lisp 11:16:57 darx: also check out LtU (lambda-the-ultimate.org 11:20:12 froog__ [n=david@87.192.28.247] has joined #lisp 11:22:30 b4|hraban [n=b4@0brg.xs4all.nl] has joined #lisp 11:23:10 -!- froog [n=david@87.192.28.247] has quit [Read error: 110 (Connection timed out)] 11:24:10 -!- kiuma [n=kiuma@ip-126-26.sn1.eutelia.it] has quit ["Bye bye ppl"] 11:24:32 darx: and take care; when you feel the need to prefix a statement with "it's common knowledge", there's a good chance that you've fallen into or are perpetrating the ad populum fallacy 11:29:09 -!- edon [n=edon@albalug/edon] has quit ["Konversation terminated!"] 11:33:14 -!- Hun [n=Hun@p4FCF5364.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 11:34:21 -!- Odin- [n=sbkhh@s121-302.gardur.hi.is] has quit [] 11:35:28 -!- froog_ [n=david@87.192.28.247] has quit [Read error: 110 (Connection timed out)] 11:36:14 Is this channel logged? 11:37:31 minion: logs 11:37:32 logs: #lisp logs are available at http://ccl.clozure.com/irc-logs/lisp/ ; older logs may be available at http://tunes.org/~nef/logs/lisp/ (down as of 2008/09/24) 11:38:27 nice, thanks 11:38:27 darx: recursive FFT is faster than iterative FFT because of locality. 11:38:55 darx: When I wote a recursive bit-reversal algorithm for the FFT, I beat the previous record (which was iterative) by a factor 5. 11:39:36 darx: recursive quicksort is very likely faster than iterative quicksort, also for reasons of locality. 11:39:38 davazp [n=user@77.Red-83-54-164.dynamicIP.rima-tde.net] has joined #lisp 11:39:58 ah, right, darx: ecraven is right, tail recursion is not expensive at all 11:40:04 -!- rdd [n=rdd@c83-250-142-219.bredband.comhem.se] has quit [Read error: 104 (Connection reset by peer)] 11:40:15 darx: Perhaps you, like many of my colleagues and students, think that computers still look sort of like the VAX? 11:42:24 -!- obsethryl [n=llort@unaffiliated/obsethryl] has quit [Read error: 104 (Connection reset by peer)] 11:44:18 (heheh and S11001001 is right, too :P; many things considered "common knowledge" in CS are actually bollocks... like global variables being inherently bad, for example) 11:46:13 k4cp [n=k4@196-209-111-98-wblv-esr-3.dynamic.isadsl.co.za] has joined #lisp 11:46:40 -!- k4cp is now known as Guest36464 11:46:58 schme_: planet lisp readership is probably more apt to recognize the worth of this paper than any other, given its preference for a 50-years old programming language ;-) 11:47:22 -!- Guest36464 is now known as k4cp 11:49:32 -!- Ogedei [n=user@78.52.230.47] has quit [Read error: 110 (Connection timed out)] 11:49:36 darx: In general, in order to remove recusion, you would have to manage an explicit stack. This is more costly than using the ordinary execution stack. Furthermore, for data structures that usually do not require recursion (such as lists) it is better to write the program iteratively in the first place. 11:50:36 darx: In summary, then, recursion removal just isn't worth it, and sometimes it is a bad idea. 12:02:00 elurin` [n=user@85.106.147.160] has joined #lisp 12:02:31 hugo [n=hugo@unaffiliated/hugo] has joined #lisp 12:02:58 jtoy [n=jtoy@58.63.218.55] has joined #lisp 12:03:21 beach, I finally got started with the floating-point directives today. Written a stub for ~F that "float" for floats, ~D for other numbers and ~A otherwise, just to test. I've had problems with fatigue lately, but on the bus from Lkpg I pulled myself together. 12:03:21 drdo [n=psykon@167.111.54.77.rev.vodafone.pt] has joined #lisp 12:05:08 tic: excellent! 12:05:21 ehu [i=52aa21ad@gateway/web/ajax/mibbit.com/x-d6a20224748b7bed] has joined #lisp 12:06:48 I snarfed code from src/target-format.lisp, that was the plan right? 12:07:03 froog___ [n=david@87.192.28.247] has joined #lisp 12:07:26 tic: only the implementation of the algorithm from the paper. Krystof is not sure the code for the formatters is correct. 12:07:29 besiria [n=user@ppp083212084214.dsl.uom.gr] has joined #lisp 12:07:49 beach, alright. 12:09:26 -!- elurin [n=user@85.99.132.149] has quit [Read error: 110 (Connection timed out)] 12:12:14 froog____ [n=david@87.192.28.247] has joined #lisp 12:20:33 -!- Jasko2 [n=tjasko@65.243.193.195] has quit [Read error: 60 (Operation timed out)] 12:22:19 -!- froog__ [n=david@87.192.28.247] has quit [Read error: 110 (Connection timed out)] 12:25:35 malumalu_ [n=malu@hnvr-4dbb8a0c.pool.einsundeins.de] has joined #lisp 12:27:18 -!- froog___ [n=david@87.192.28.247] has quit [Read error: 110 (Connection timed out)] 12:29:56 -!- sykopomp [n=sykopomp@unaffiliated/sykopomp] has quit [Remote closed the connection] 12:30:59 sykopomp [n=user@unaffiliated/sykopomp] has joined #lisp 12:41:27 -!- davazp [n=user@77.Red-83-54-164.dynamicIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)] 12:42:52 -!- malumalu [n=malu@hnvr-4dbb946e.pool.einsundeins.de] has quit [Read error: 110 (Connection timed out)] 12:43:35 bhyde [n=bhyde@c-66-30-202-56.hsd1.ma.comcast.net] has joined #lisp 12:44:14 emacs-dwim [n=user@cpe-69-202-149-234.twcny.res.rr.com] has joined #lisp 12:47:07 besiria` [n=user@ppp083212085040.dsl.uom.gr] has joined #lisp 12:47:52 man i need to get back into cl....think i've forgotten most of whatever little i knew :P 12:48:07 ace4016, PCL, rinse, lather, repeat? 12:49:09 heh, that's what i used the first time around 12:49:20 good book 12:49:40 might understand more of it going back the second time around too 12:50:18 -!- k4cp [n=k4@196-209-111-98-wblv-esr-3.dynamic.isadsl.co.za] has quit [] 12:50:22 -!- kpreid [n=kpreid@cpe-67-242-12-135.twcny.res.rr.com] has quit [] 12:50:51 kpreid [n=kpreid@cpe-67-242-12-135.twcny.res.rr.com] has joined #lisp 12:52:17 dkcl [n=dkcl@unaffiliated/dkcl] has joined #lisp 12:55:33 vasa [n=vasa@mm-126-93-84-93.dynamic.pppoe.mgts.by] has joined #lisp 12:56:39 -!- ace4016 [i=ace4016@cpe-76-168-253-20.socal.res.rr.com] has quit ["night"] 13:02:28 manuel_ [n=manuel@bigmother.querfunk.de] has joined #lisp 13:04:43 x6j8x [n=x6j8x@77.21.43.186] has joined #lisp 13:04:50 -!- kib2 [n=chatzill@bd137-1-82-228-159-28.fbx.proxad.net] has quit ["ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]"] 13:05:01 -!- rtoym [n=chatzill@user-0c8hpll.cable.mindspring.com] has quit [Remote closed the connection] 13:05:32 -!- jtoy [n=jtoy@58.63.218.55] has quit [] 13:10:27 manuel__ [n=manuel@port-92-205-28-78.dynamic.qsc.de] has joined #lisp 13:10:46 Hun [n=Hun@p4FCF55AE.dip.t-dialin.net] has joined #lisp 13:14:41 Kawa is a really interesting project, also appears to be quite mature. Thoughts? 13:15:52 oudeis [n=oudeis@p11811112.orange.net.il] has joined #lisp 13:16:45 -!- besiria [n=user@ppp083212084214.dsl.uom.gr] has quit [Read error: 113 (No route to host)] 13:16:45 i think i haven't seen it. 13:17:48 cmatei [n=cmatei@85.186.180.45] has joined #lisp 13:18:08 vy: would use it if forced to target JVM 13:18:09 -!- spiderbyte [n=dcl@unaffiliated/dcl] has quit [Remote closed the connection] 13:18:23 Odin- [n=sbkhh@s121-302.gardur.hi.is] has joined #lisp 13:18:30 mulligan` [n=user@e178039172.adsl.alicedsl.de] has joined #lisp 13:19:06 vy: is it very difference from clojure? 13:19:23 recursion removal is a good idea when you cannot guarantee that your dataset won't overflow the execution stack 13:19:27 of course scheme v common lisp, but I mean in the goals/what you can do with it. 13:19:53 bah, haven't scrolled all the way down :) 13:19:59 S11001001: Why not? I think it appears to be quite mature than clojure and has a really consistent java interaction base. 13:20:25 jtoy [n=jtoy@58.63.218.55] has joined #lisp 13:24:07 LiamH [n=nobody@138.88.126.7] has joined #lisp 13:24:57 c|mell [n=cmell@v113243.ppp.asahi-net.or.jp] has joined #lisp 13:25:38 Cool: (make size: 10) 13:26:58 jgracin [n=jgracin@82.193.208.195] has joined #lisp 13:27:45 -!- manuel_ [n=manuel@bigmother.querfunk.de] has quit [Read error: 113 (No route to host)] 13:32:20 my sbcl 1.0.21 gets stuck trying to compile a regexp from cl-irregsexp 13:32:36 (it works on ccl 1.2) 13:32:47 now it is at 1.3gb of memory 13:33:01 and it seems to be doing a lot of memory accesses 13:33:31 -!- mulligan [n=user@e178031213.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)] 13:33:32 anybody know how to debug what is going wrong? 13:34:07 vy pasted "kawa fuss" at http://paste.lisp.org/display/68310 13:34:46 -!- x6j8x [n=x6j8x@77.21.43.186] has quit [] 13:35:34 myrkraverk [n=johann@unaffiliated/myrkraverk] has joined #lisp 13:37:29 vy: what's the point? 13:37:55 -!- besiria` [n=user@ppp083212085040.dsl.uom.gr] has quit [Read error: 104 (Connection reset by peer)] 13:38:19 stassats`: There is generally no point in fuss. 13:38:44 -!- oudeis [n=oudeis@p11811112.orange.net.il] has quit [Read error: 110 (Connection timed out)] 13:38:53 ok, --script support is now in cvs 13:38:59 vy: so why do you post it? 13:41:13 alec [n=alec@pool-71-174-175-161.bstnma.east.verizon.net] has joined #lisp 13:42:19 besiria [n=user@ppp083212085040.dsl.uom.gr] has joined #lisp 13:42:20 davazp [n=user@77.Red-83-54-164.dynamicIP.rima-tde.net] has joined #lisp 13:50:03 -!- malumalu_ [n=malu@hnvr-4dbb8a0c.pool.einsundeins.de] has quit ["Verlassend"] 13:51:49 vy: and how would you interpret my phrase as arguing with that? If I don't have to JVM, I'd rather use Common Lisp. 13:52:01 -!- emacs-dwim [n=user@cpe-69-202-149-234.twcny.res.rr.com] has quit [Read error: 110 (Connection timed out)] 13:52:08 -!- jtoy [n=jtoy@58.63.218.55] has quit [] 13:52:49 i've added some help for installing sbcl from source with common-lisp-controller and not in /usr/local to http://www.cliki.net/SBCL 13:52:56 hope that was the right place 13:53:47 S11001001: Actually nobody wants to mess with JVM as long as he/she has an alternative. But the vast population of available libraries for Java and the need for agile development drives people the other way. 13:54:15 bombshelter13 [n=bombshel@209-161-240-8.dsl.look.ca] has joined #lisp 13:56:56 icebox [n=chatzill@83.225.181.184] has joined #lisp 13:58:13 pdeb [n=user@pool-68-161-179-70.ny325.east.verizon.net] has joined #lisp 13:58:27 okay my function is 89k lines 13:58:37 guess sbcl cannot handle that 13:59:37 myrkraverk` [n=johann@157-157-185-223.dsl.dynamic.simnet.is] has joined #lisp 13:59:46 Spune [n=Spune@161.sub-75-196-176.myvzw.com] has joined #lisp 14:01:06 -!- Spune [n=Spune@161.sub-75-196-176.myvzw.com] has quit [Read error: 104 (Connection reset by peer)] 14:01:30 Spune [n=Spune@68-245-203-6.area3.spcsdns.net] has joined #lisp 14:01:34 c|mell: how is it not handling it? heap exhaustion? just too slow at compiling it? 14:01:48 xMilesTegx [n=Spune@161.sub-75-196-176.myvzw.com] has joined #lisp 14:01:50 -!- xMilesTegx [n=Spune@161.sub-75-196-176.myvzw.com] has quit [Read error: 104 (Connection reset by peer)] 14:02:12 -!- Spune [n=Spune@68-245-203-6.area3.spcsdns.net] has quit [Read error: 104 (Connection reset by peer)] 14:02:14 Hun` [n=Hun@p4FCF55AE.dip.t-dialin.net] has joined #lisp 14:04:56 abend_ [n=sasha@sub26-151.member.dsl-only.net] has joined #lisp 14:04:57 moghar [n=user@unaffiliated/moghar] has joined #lisp 14:08:55 chandler, i have left it for ten minutes but it does not finish 14:09:06 what optimize settings are you using? 14:09:17 and how fast is your computer? 14:09:26 it isn't overly surprising to me that a huge function would take a long time to compile 14:09:30 some parts are optimize (speed 3) (safety 0) 14:09:42 and I'd expect it to be slower than if it was broken up into multiple functions 14:09:52 with the most critical parts declared inline 14:10:14 i am on a core2 l7700 at 1.8 with 2GB ram 14:10:18 instruction cache hates big functions 14:10:34 well i have simplified the monster macros a bit and it works now 14:11:01 just wondering how to get into sbcl to find out what is blowing it up 14:11:06 c|mell: sbcl is very slow for huge functions. this is another reason to split out macro bodies into helper functions. 14:11:16 re 14:11:33 yes but helper functions destroy my speediness 14:11:36 c|mell: I don't think "blowing up" is the right description. I wouldn't necessarily expect the compiler to be O(n) in the size of the function. 14:11:46 ... did you profile? 14:11:57 sb-sprof? 14:12:00 (and in fact, it's not linear) 14:12:02 Yes. 14:12:31 You can selectively inline the helper functions you care about with inline declarations, but there's no easy way to selectively outline code from a huge function. 14:13:20 it seemed to me that the type inference works better if you make things macros not inline functions 14:13:31 "seemed"? 14:13:31 maybe i was wrong? 14:13:46 Does Helmut Eller hang around in here? 14:13:52 c|mell: yes, you were 14:14:03 well i was trying to get cl-irregsexp fast enough to load in the netflix dataset 14:14:04 vy: I have not seen him here in a while. 14:14:14 -!- vasa [n=vasa@mm-126-93-84-93.dynamic.pppoe.mgts.by] has quit ["I am not vasya, i am vasa"] 14:14:16 enigmus [n=e@S0106001d7e52d1d9.vc.shawcable.net] has joined #lisp 14:14:27 and making things into macros gave me the boost i needed 14:14:37 syamajala2 [n=syamajal@140.232.182.214] has joined #lisp 14:14:39 -!- Hun [n=Hun@p4FCF55AE.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 14:14:48 (like instead of hours less than 20mins) 14:15:08 there isn't any reason to turn a function into a macro just to inline it 14:15:09 but maybe i should try to see exactly what was going on 14:15:16 so, ask yourself: is that the only possible thing that could have changed your runtime? 14:16:27 the major other thing i did was put a typecase in front so that strings, byte-vectors, and other things each have their own copy of the regexp 14:16:54 you can't run a good experiment by changing more than one thing at a time! 14:17:08 true 14:17:17 that's why i said "seemed" 14:17:42 woot 14:17:42 chandler: What is his nickname? 14:17:47 chandler: oh yeah? You have 12 balls, one of which is either lighter or heavier than the others. (etc.) 14:17:50 oops wrong channel 14:17:58 vy: heller 14:18:04 -!- myrkraverk [n=johann@unaffiliated/myrkraverk] has quit [Read error: 110 (Connection timed out)] 14:18:22 Krystof: yeah yeah :-) insert appropriate qualifications and weasel words above 14:18:29 don't confuse generalizations with truth! 14:20:18 -!- bobrown`` [n=user@dsl081-198-234.nyc2.dsl.speakeasy.net] has quit [Remote closed the connection] 14:20:57 the reason i said it blew up is because i tried killing sbcl with sigterm and sigint but they don't pop me into slime 14:21:06 in fact slime just ignores we 14:21:33 What is showing in the *inferior-lisp* buffer? 14:21:36 lisppaste: url? 14:21:36 To use the lisppaste bot, visit http://paste.lisp.org/new/lisp and enter your paste. 14:21:42 its like, you gave me an 89k function, and I'm not going to talk to you anymore 14:23:38 -!- abend [n=sasha@sub26-151.member.dsl-only.net] has quit [Read error: 110 (Connection timed out)] 14:24:26 yes 14:24:47 i will give it a few minutes and let you know 14:24:54 "yes"? 14:25:15 Is there anything of interest in the *inferior-lisp* buffer? Like, say "heap exhausted, game over"? 14:25:56 x6j8x [n=x6j8x@77.21.43.186] has joined #lisp 14:27:30 -!- besiria [n=user@ppp083212085040.dsl.uom.gr] has quit [Remote closed the connection] 14:27:59 -!- b4|hraban [n=b4@0brg.xs4all.nl] has quit [Remote closed the connection] 14:28:03 i have often seen that error message :) 14:28:12 -!- manuel__ [n=manuel@port-92-205-28-78.dynamic.qsc.de] has quit [Read error: 104 (Connection reset by peer)] 14:28:24 but not in this case 14:28:40 manuel_ [n=manuel@port-92-205-28-78.dynamic.qsc.de] has joined #lisp 14:30:22 at the moment i have it sort of paused because i pressed C-c C-c in the slime repl 14:31:00 malumalu [n=malu@hnvr-4dbb8a0c.pool.einsundeins.de] has joined #lisp 14:32:13 one thing: i start it with sbcl --dynamic-space-size 2000 so that it won't set off the oom killer 14:32:21 but now it is up to 1.2GB 14:33:00 "Optimization: you're doing it wrong." 14:33:23 okay 14:34:38 replor [n=replor@ntkngw375028.kngw.nt.ftth.ppp.infoweb.ne.jp] has joined #lisp 14:35:06 so at the moment my regexp macro is like this `(typecase ,string (byte-vector (macrolet (...) ,@body)) (simple-string (macrolet (...) ,@body) etc.) 14:35:20 it works on clozure cl 14:35:34 so let me know how i should massage it to work on sbcl 14:36:43 spiderbyte [n=dcl@unaffiliated/dcl] has joined #lisp 14:36:54 at the moment slime and swank seem to have stopped talking to each other 14:37:45 I wouldn't start from there 14:38:10 any clues? 14:38:46 macdice [n=macdice@78-86-162-220.zone2.bethere.co.uk] has joined #lisp 14:43:46 b4|hraban [n=b4@0brg.xs4all.nl] has joined #lisp 14:44:39 is it a bad idea to use the reader for a network protocol? 14:44:58 oudeis [n=oudeis@p11811120.orange.net.il] has joined #lisp 14:45:22 i mean, passing around simple s-expressions over sockets 14:45:35 c|mell: not with that amount of detail that you're giving me. Generally: write small functions, specialized to a given type 14:45:47 glue them together appropriately, in other small functions 14:46:04 also: profile first, identify the hotspot, and optimize that and only that 14:46:25 argh 14:47:37 do you understand why i am doing the typecase? 14:47:52 -!- xjrn [n=jim@c-69-181-213-99.hsd1.ca.comcast.net] has quit [Read error: 110 (Connection timed out)] 14:47:55 elindio [n=badiss@gar13-5-88-161-23-155.fbx.proxad.net] has joined #lisp 14:49:00 hello , i'm new on this channel , somebody can explain me the differences between lisp , common lisp ,emacs lisp 14:49:03 there is something funny about the part ",string (byte-vector" ... :) 14:49:09 ? 14:49:27 elindio: No, that would take too long. Look it up on Wikipedia. 14:49:28 besiria [n=user@ppp083212085040.dsl.uom.gr] has joined #lisp 14:49:42 -!- b4|hraban [n=b4@0brg.xs4all.nl] has quit ["Leaving"] 14:50:21 macdice: if you're in control of both sides of the network protocol, it might not be that bad, for prototyping purposes. 14:50:42 ok thanks ! 14:50:48 : ) 14:51:01 S11001001: it won't be too long. 14:51:12 S11001001, i want to be able to use the regexp matcher on strings, vectors of unsigned byte 8 and general sequences 14:51:33 elindio: Lisp: sometimes used for a fairly large family of languages, sometimes just as a synonym for Common Lisp. 14:52:10 S11001001, but i want it to go fast, and i know i can't afford type dispatch per character 14:52:13 as i tried that 14:52:27 elindio: Common Lisp a standardized langauge, discussed in this channel. 14:52:52 elurin [n=user@88.224.40.6] has joined #lisp 14:52:53 elindio: Emacs Lisp a dialect of Lisp written by RMS specifically in order to implement Emacs. 14:52:53 ok it's a standardisation of lisp 14:53:08 elindio: why do you want to know? 14:53:24 thank you beach 14:54:09 there is a link between "lambda calculus" and lisp (i'm reading a pdf on this) 14:55:11 elindio: Yes, very early versions of Lisp were inspired by the lambda calculus. Nowadays, Lisp is a general-purpose programming language, and you won't recongnize much of the lambda calculus in it. 14:55:19 elindio: but you didn't answer my question. 14:55:30 c|mell: the reason for the typecase is clear. however, it is generally better to do (typecase x ((simple-bit-vector (foo-on-simple-bit-vectors x)) ...) 14:55:58 c|mell: yes, but you named the parameter "string", when it clearly can be a bit-vector ;) 14:56:01 aiur [n=Jan@218.109.91.183] has joined #lisp 14:56:13 nikodemus, thanks! yes actually i am doing that, because i do need it to be simple 14:56:21 -!- elurin` [n=user@85.106.147.160] has quit [Read error: 110 (Connection timed out)] 14:56:23 matley [n=matley@83.224.133.248] has joined #lisp 14:56:31 S11001001, it is really called "target" in my code 14:56:46 c|mell: note the out-of-line function that does the actual work 14:57:04 out of line functions are also more amnenable to profiling 14:57:46 nikodemus: inventing new terminology? 14:57:55 unless you are already doing it, you may want to generate the typecase and the specialized functions and the typecase with a macro(let), to reduce manual duplication of code 14:58:10 i hope not. what did i say that was funny? 14:58:19 amnenable 14:58:36 nikodemus, unfortunately the actual function body has to be generated by the macro as it is the regexp matcher itself 14:58:41 s/mn/m/ 14:58:50 so? 14:59:40 is it good practice to cook up a lot of gensym functions 14:59:59 actually that sounds like a really helpful suggestion! 15:01:58 at the moment i am trying out shoving the worker-functions into an flet (the compile hasn't freaked out yet but it is taking ages) 15:02:01 -!- aiur [n=Jan@218.109.91.183] has quit [Remote closed the connection] 15:02:44 aiur [n=Jan@218.109.91.183] has joined #lisp 15:03:15 uh, are you expanding the whole regex inline? 15:03:58 -!- oudeis [n=oudeis@p11811120.orange.net.il] has quit ["This computer has gone to sleep"] 15:04:00 yeah, you probably don't want gensym defuns 15:05:38 -!- vy [n=user@213.139.194.86] has quit [Remote closed the connection] 15:06:11 c|mell: yes, I do understand, and my suggestion stands 15:06:15 nikodemus, yes i am 15:06:32 if you need three versions of helper functions, so be it 15:06:57 it's better than one triply-expanded huge function body 15:07:02 looks like boosting the three cases into flet has made it work 15:07:11 new CDR proposal: nine more N+ variants of 1+ when 1 isn't enough 15:07:39 -!- elurin [n=user@88.224.40.6] has quit [Remote closed the connection] 15:09:40 elurin [n=user@88.224.40.6] has joined #lisp 15:10:22 dabd [n=dabd@85.139.98.38] has joined #lisp 15:12:19 -!- name [n=name@sburn/devel/name] has quit ["Lost terminal"] 15:12:58 but if you have access to the exact regexp when generating the function, surely you should be able to tell if it is going to match strings, byte-vectors, or what? 15:17:17 -!- c|mell [n=cmell@v113243.ppp.asahi-net.or.jp] has quit [Read error: 104 (Connection reset by peer)] 15:18:21 Soulman [n=knute@62.80-202-238.nextgentel.com] has joined #lisp 15:19:45 -!- segv_ [n=mb@p4FC1D3F0.dip.t-dialin.net] has quit [Read error: 60 (Operation timed out)] 15:20:25 mrd-_ [n=mrd@c-24-61-70-86.hsd1.ma.comcast.net] has joined #lisp 15:20:46 xjrn [n=jim@c-69-181-213-99.hsd1.ca.comcast.net] has joined #lisp 15:21:26 c|mell [n=cmell@v113243.ppp.asahi-net.or.jp] has joined #lisp 15:21:48 sorry sbcl apparently took down my whole computer 15:21:52 !!! 15:22:54 -!- elurin [n=user@88.224.40.6] has quit [Read error: 60 (Operation timed out)] 15:24:12 Aankhen`` [n=pockled@122.162.154.37] has joined #lisp 15:24:13 hi could someone help me with the sbcl 1.0.20 and the latest slime from cvs 15:24:29 I'm getting the following compilation error on swank-sbcl.lisp 15:24:32 Symbol "DEBUG-SOURCE-FROM" not found in the SB-DI package 15:26:49 Is there a way to ask asdf what the path your were installed into was? (I want to read some data files in) 15:28:07 herbieB: asdf:system-definition-pathname 15:29:42 also (defvar *foo* #.*compile-file-pathname*) 15:29:51 S11001001: Better question, how did you find that? 15:30:54 herbieB: Weblocks uses it to tell Hunchentoot where a webapp's public files are. 15:30:55 i'm pretty sure you are not actually running the latest CVS Slime 15:31:07 (for dabd) 15:32:04 -!- mrd- [n=mrd@c-24-147-145-191.hsd1.ma.comcast.net] has quit [Read error: 110 (Connection timed out)] 15:32:28 S11001001: Hm, do I need to follow symlink after going there? It just gives me the site-systems dir that everything gets simlinked to. 15:33:35 sure, truename does it portably (except LW but their loss) 15:33:59 nikodemus: I followed the instructions for installing slime found here: http://www.cliki.net/SLIME-HOWTO 15:34:07 I checked out the latest slime 15:34:10 segv_ [n=mb@p4FC1D404.dip.t-dialin.net] has joined #lisp 15:34:26 S11001001: Le awesome, thanks so much :) 15:34:35 ah hah! heller broke SLIME stuff on 10/4 15:35:44 nunb [n=nunb@host97-228-dynamic.30-79-r.retail.telecomitalia.it] has joined #lisp 15:36:24 changed the swank-compile-file interface without all the implementations 15:36:28 nikodemus:why do you say I'm not using the latest CVS slime? 15:36:35 -!- froog____ is now known as froog 15:37:29 because of the symptom you reported: that happens when your slime is too old 15:38:20 -!- Odin- [n=sbkhh@s121-302.gardur.hi.is] has quit ["Am I missing an eyebrow?"] 15:38:22 current slime tests for the existence of a couple of alternative symbols at read-time, and and deals with them appropriately 15:39:33 dabd: are you on debian or ubuntu by any chance? 15:40:11 ubuntu 15:40:27 double-check that you don't have the slime.deb installed 15:40:32 nikodemus: i thought mentioning unicode support as a feature would be good, but the more i think about it the less i think it would be good 15:40:52 I removed slime with synaptic 15:40:56 also, some emacsen (iirc aquaemacs) bundle slime -- which tends to break things 15:41:03 dabd: you might need to purge it 15:41:16 check that there's no slime file in /etc/emacs/site-start.d or similar 15:42:09 Xach: i thought about putting "buzzword enabled" somewhere there ... 15:43:00 what's bad about its unicode support? 15:43:09 I have a 50slime.el in /etc/emacs/site-start.d 15:43:27 dabd, if you want to use cvs slime and you have slime installed from ubuntu, make sure CVS is first in you emacs load path and (setf slime-backend "swank-loader.lisp") in .emacs 15:43:45 nuke it, and report it as a but against the slime package in ubuntu 15:43:48 tcr: non-exportedness, non-discoverable, undocumented, incomplete 15:44:03 annoyingly ubuntu sets slime-backend to the full path of their installed version 15:44:08 some bits are implemented over streams and some over octets (and fortunately most over both) 15:44:22 doesn't support the latest version of the Unicode standard 15:44:22 c|mell: I don't know this slime-backend variable; is it new? 15:44:29 no built-in support for collation, normalization 15:44:30 etc 15:44:48 no idea, but i had that problem today and that's how i solved it 15:45:07 wow i have got it all to compile! 15:45:19 cjmell:I already fixed that bc it was bound to the previous slime installation 15:45:32 Krystof: Ok, thanks. 15:46:26 -!- drdo [n=psykon@167.111.54.77.rev.vodafone.pt] has quit [Remote closed the connection] 15:46:29 dabd, if you go to *inferior-lisp* you can see which swank it loads 15:47:17 -!- jlouis [n=jlouis@port1394.ds1-vby.adsl.cybercity.dk] has quit [Read error: 104 (Connection reset by peer)] 15:47:42 yes it is loading the correct swank from the slime CVS 15:48:03 I already deleted the 50slime.el from /etc/emacs/site-start.d 15:48:12 For my sb-reader patch, I moved some conditions into the sb-reader package which were previously exported from sb-int; should I make those symbols be reexported from sb-int for backwards compatibility? (sb-int is a private package after all.) 15:48:40 I also did apt-get remove --purge slime 15:48:47 but I still get the same error 15:49:20 tcr: are you aware of anything using those symbols? 15:49:37 kreuter: alexandria 15:49:48 aren't you a maintainer of alexandria? 15:50:06 dabd: did you shut down emacs after doing all the cleanups? 15:50:08 kreuter: You can easily grep for them through your source archive; that's the only offence I could find this way. 15:50:12 sohail [n=Sohail@unaffiliated/sohail] has joined #lisp 15:50:29 kreuter: No, I'm a mere contributor; but nikodemus is. 15:50:37 jlouis [n=jlouis@port1394.ds1-vby.adsl.cybercity.dk] has joined #lisp 15:50:48 jazen3 [n=blah@77.20.240.183] has joined #lisp 15:51:10 well, then nikodemus gets to fix it :) 15:51:26 i was thinking that most of the things that sb-reader exports the in patch should be exported from either sb-ext or sb-int 15:51:33 nikodemus: I shut down emacs. 15:51:56 nikodemus: why not have sb-reader be a public package? 15:52:17 It's a semi-public package. 15:52:23 if you're willing to document and maintain it, sure :) 15:53:04 ISTM that there's not a lot of value putting condition names in sb-ext. 15:53:12 ? 15:53:32 it implies it is a supported condition type that you can handle and introspect 15:53:43 What does "document it" entail? 15:53:57 (which is what alexandria does, although the condition is in sb-int) 15:54:00 Docstrings for the stuff that is supposed to be used by a user? 15:54:08 nikodemus: doesn't exporting the condition name from its home package suffice for that? 15:54:22 if the home-package is not public, no 15:55:14 why not? 15:55:38 because export from a non-public package implies an internal interface, not a public one 15:55:41 AFAICT, the set of symbols we're talking about from sb-reader would be pretty small right now. 15:55:45 elurin [n=user@88.224.40.6] has joined #lisp 15:55:57 so I don't see why we shouldn't make it the public interface to our reader extensions 15:56:06 which is why i'm not sure exporting them from a separate package would be a great win 15:56:18 i'm not adamantly opposed either, though 15:56:21 well, sb-ext is used by cl-user 15:57:18 15:58:03 maybe a Great Package Reorganization page on sbcl-internals? 15:58:06 if it's a small number of symbols, what's the point of a whole pacakge for it? 15:58:49 I could think of two possibilities 15:59:22 I'm still in trouble with slime... 16:00:18 (1) there's some suggestion that one way to modularize the build process would be to carve up the big packages into small ones. this could be part of that. 16:01:18 nikodemus: http://git.boinkor.net/gitweb/sbcl.git?a=blobdiff;f=src/runtime/runtime.c;h=79dbcbf0e956ead73f30515116a0694c53bfbce0;hp=02898e3d16d0b7959339fbfcb4dd564949731242;hb=ee222567ee95eaac8f6f4c877242dd116bfb8337;hpb=f691243c787fb7a06091807a1297a130eb561b2a has got a superfluous addition, it seems 16:02:18 dabd, why not try sbcl 1.0.21 with CVS slime; i installed both today and it is working as well as usual 16:02:35 (2) there might be more reader extensions eventually, and so we might as well introduce a home package for them now . 16:02:48 this is the error log from *inferior-lisp* http://paste.lisp.org/display/68314 16:02:52 kpreid: good catch, thanks 16:03:12 -!- elurin [n=user@88.224.40.6] has quit [Remote closed the connection] 16:03:26 elurin [n=user@88.224.40.6] has joined #lisp 16:04:00 c|mell: I'll try that 16:04:28 -!- mishok13 [n=gdmfsob@92.112.65.184] has quit [Read error: 60 (Operation timed out)] 16:04:41 -!- jfrancis [n=jfrancis@72.14.224.1] has quit [] 16:04:46 dabd: nuke ~/.slime/fasl 16:05:16 kreuter: Yes, I had both points in mind. (Though not modularizing the build process, but "perceived" modularization in a self-documentation sense of way); I'd also like to export READ-TOKEN at some point, and a hook into symbol interning. 16:05:31 i don't understand what is going on there, but somehow two slime versions seem to getting mixed up 16:05:33 is /home/dabd/software/slime/ the CVS version? 16:05:53 tcr: I don't overly much like that idea, but that's a different fight. :) 16:06:25 kreuter: The hook thingie? 16:06:28 name [n=name@sburn/devel/name] has joined #lisp 16:06:34 um 16:06:46 the hook might be fine. I think exporting read-token is problematic. 16:07:09 nikodemus: I already nuked ~/.slime/fasl and yes home/dabd/software/slime is the CVS version 16:07:38 OTOH, I don't see why people who need a custom reader don't just implement it themselves. 16:08:02 kreuter: Oh ok. Well, I find the hook issue problematic, because it feels like a kludge to remedy one symptom. But a generic reader protocol needs some serious thought. 16:08:46 nikodemus pasted "dabd: do lines 903-915 in /home/dabd/software/slime/swank-sbcl.lisp look like this" at http://paste.lisp.org/display/68315 16:10:19 wait, what? your reader error occurs on line 755? that is definitely not CVS slime 16:10:22 -!- enigmus [n=e@S0106001d7e52d1d9.vc.shawcable.net] has quit [Read error: 110 (Connection timed out)] 16:11:10 lisppaste:no they don't look like that 16:11:34 lisppaste:my version doesn't have the conditional compilation 16:11:46 "cvs up -dPC" in the directory 16:13:15 kreuter: What is *source-location-namestring-hook* about? 16:13:48 tcr: so that we can record the source file's name with stuff compiled during slime-compile-string . 16:14:09 kreuter: Isn't there some magic currently to store emacs buffer information? 16:14:13 drewc: in case you were wondering: I'm still seeing errors. 16:14:26 kreuter: did you checkout :source-plist in with-compilation-unit? 16:14:42 nikodemus:the cvs update didn't bring anything but contrib and doc 16:14:54 strange 16:15:15 how did you check it out, and wherefrom? 16:15:19 tcr: nikodemus: yes, but I didn't want to bake SLIME-specific stuff into the source-location machinery. 16:15:35 have a look at the second attachment in the email I sent. 16:17:55 nikodemus: I did cvs -d :pserver:anonymous@common-lisp.net:/project/slime/cvsroot checkout slime 16:18:00 ok problem solved 16:18:11 I nuked software/slime and checked out again 16:18:15 now it all works 16:18:36 the cvs checkout was messed up 16:18:39 -!- jazen [n=blah@77.20.240.183] has quit [Connection timed out] 16:18:48 sorry for the trouble guys 16:18:49 thanks 16:20:36 kreuter: With the hook, and it being installed into swank-sbcl; isn't the only difference that the :namestring of the definition-source-location does not point to a tmp file? 16:20:46 yes 16:21:13 mishok13 [n=gdmfsob@92.112.254.10] has joined #lisp 16:21:15 kreuter: right. what about adding :source-namestring to with-compilation-unit? or specifying that :original-namestring in the :source-plist is magical? 16:21:56 kreuter: I can't see the point of it; it'd make more sense to me if you transformed the :plist information away, i.e. by not only updating :namestring, but also :character-offset 16:23:15 -!- elurin [n=user@88.224.40.6] has quit [Connection timed out] 16:23:41 alternatively, the hook could be *source-location-hook* -- so slime could be as accurate as it cared to 16:23:58 -!- syamajala2 [n=syamajal@140.232.182.214] has quit [] 16:24:16 I don't really care about the details. IIRC, nothing in SBCL actually uses the source-plist, which I consider to be a feature. 16:24:44 but as long as the hook is not public, i'm not opposed to the patch as it stands :) 16:24:44 Also, it may make more sense to have a stream-source-location, and a file-source-location; C-c C-c being "compile temp file" is a kludge anyway, and ideally it there should be a COMPILE-FROM-STREAM function. 16:25:11 -!- mozzyb [n=mozzyb@248.80-203-57.nextgentel.com] has quit ["Lost terminal"] 16:25:20 dagnachew [n=dagnache@modemcable207.114-201-24.mc.videotron.ca] has joined #lisp 16:25:26 tcr: as soon as someone tells me how the semantics of that are supposed to differ from LOAD 16:25:32 re. eval-when 16:25:47 I don't agree about compile-from-stream. 16:25:47 nikodemus: it should have the same semantics as compile-file 16:25:58 so :load-toplevel doesn't happen? 16:26:17 (I also think having LOAD work on streams is a bug.) 16:26:23 (defun foo ...) doesn't define foo? 16:26:50 it's a required bug that i just found a use for, so i don't complain :) 16:26:59 nikodemus: ecl's compile-from-stream has a :load keyarg 16:27:31 brr 16:28:43 kreuter: why do you consider it a bug? 16:32:09 b4|hraban [n=b4@0brg.xs4all.nl] has joined #lisp 16:32:15 -!- besiria [n=user@ppp083212085040.dsl.uom.gr] has quit [Remote closed the connection] 16:32:31 well, mostly because the way it's specified, it seems the implementation is supposed to support things like (load *terminal-io*), and that strikes me as wacky. 16:33:00 if it said that the stream had to be a file-stream, I think I wouldn't mind. 16:33:21 kiuma [n=kiuma@ip-126-26.sn1.eutelia.it] has joined #lisp 16:36:21 -!- cmatei [n=cmatei@85.186.180.45] has quit [Remote closed the connection] 16:36:26 Ok. 16:37:02 hm 16:37:43 i don't agree 16:37:44 mozzyb [n=mozzyb@248.80-203-57.nextgentel.com] has joined #lisp 16:37:57 Back to the issue: I like a *source-location-hook* more than the *source-location-namestring-hook*; I think it should also be commented what the original motivation for it is, i.e. making redefinition warning give proper context. 16:38:12 i don't see why loading from string-streams or network streams would be bogus by definition 16:39:02 I'd have thought that compile-from-stream returns a fasl-stream, and load should accept these. 16:39:07 nikodemus: for one thing, LOAD can't return from network streams. 16:39:15 ? 16:39:22 well, in general. 16:39:35 they might not end. 16:40:03 if you're loading from a network stream, the sender never closing is the least of your worries, i would think :) 16:40:11 -!- ecraven [n=nex@plc31-103.linzag.net] has quit ["bbl"] 16:40:37 that's my point: I'm claiming that loading from a network stream doesn't make a lot of sense. 16:40:57 but string streams and random gray streams make a lot of sense in context of things like slime 16:41:08 oh, an IDEA! 16:41:13 kill it 16:41:32 Xach: ? 16:42:22 drewc: ping 16:42:31 I mean, it's not as if the user can't write (loop (eval (read))) himself. 16:42:50 in fact, I just wrote it! 16:42:52 tcr: ack 16:42:52 add support for compile-file on streams, but require the output-file argument 16:43:00 kreuter: sorry. mostly random. 16:43:03 which could be a stream as well :P 16:43:20 drewc: I share michaelw's interest in your putative reader protocol, fwiw. 16:43:43 drewc: And if it's just a draft you don't dare to publish yet. 16:44:05 kreuter: I disagree that LOAD on a network stream doesn't make sense. 16:44:24 *tcr* <3 #lisp 16:44:38 tcr: I'm going to write something up in the near future... i keep finding use cases in my day to day work. 16:44:50 vy [n=user@88.230.190.201] has joined #lisp 16:45:13 drewc: I've got some, too. Hence my appeal to joint venture. :) 16:45:54 tcr: sounds good to me. :) have you got anything tangible yet? 16:46:26 lispm [n=joswig@e177155196.adsl.alicedsl.de] has joined #lisp 16:46:33 drewc: No, and I'm afraid I won't have time until christmas breaks. 16:46:53 kpreid: so, my mental model for LOAD is "(loop (eval (read))) plus some required variable bindings plus implementation-dependent magic". allowing the argument to be an arbitrary stream makes it impossible to always set up the required variable bindings, and in general invalidates implementation-dependent magic, too. 16:46:56 drewc: Perhaps some wiki would be good where we could put ideas on? 16:46:59 papermachine [n=Eien@61.152.106.169] has joined #lisp 16:47:14 kpreid: er, that's LOAD on a source file. 16:47:39 -!- gigamonkey [n=user@adsl-99-179-47-191.dsl.pltn13.sbcglobal.net] has quit [Read error: 104 (Connection reset by peer)] 16:47:44 -!- Patzy [n=somethin@coa29-1-88-174-11-170.fbx.proxad.net] has quit [Read error: 60 (Operation timed out)] 16:47:55 Patzy [n=somethin@coa29-1-88-174-11-170.fbx.proxad.net] has joined #lisp 16:48:19 since the user can just as well write "(loop (eval (read)))" himself, ISTM unfortunate that the spec for LOAD seems to require it to support any stream. 16:48:23 -!- dkcl [n=dkcl@unaffiliated/dkcl] has quit [Remote closed the connection] 16:48:52 -!- Patzy [n=somethin@coa29-1-88-174-11-170.fbx.proxad.net] has quit [Client Quit] 16:52:10 -!- kiuma [n=kiuma@ip-126-26.sn1.eutelia.it] has quit ["Bye bye ppl"] 16:52:25 kreuter: sure, as long as we give them the hooks into source location machinery :P 16:52:46 Patzy [n=somethin@coa29-1-88-174-11-170.fbx.proxad.net] has joined #lisp 16:53:22 eh? I think it's uncontroversial that LOAD should DTRT on a file. 16:53:49 for other sorts of streams, there's no meaningful location anyway. 16:53:54 (in general) 16:53:55 -!- sellout [n=greg@pool-138-88-27-104.res.east.verizon.net] has quit [] 16:54:18 kij [n=user@pc.tv2.dk] has joined #lisp 16:55:00 hey #lisp how do i go from "12345" -> (1 2 3 4 5) ? 16:55:06 coerce 16:55:08 -!- nunb [n=nunb@host97-228-dynamic.30-79-r.retail.telecomitalia.it] has quit [Read error: 104 (Connection reset by peer)] 16:55:14 er 16:55:42 (mapcar #'digit-char-p (coerce "12345" 'list)) 16:56:17 (map 'list #'digit-char-p "12345") 16:56:18 (map 'list #'digit-char-p "12345") 16:56:23 second! 16:56:24 -!- Hun` [n=Hun@p4FCF55AE.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 16:56:56 -!- manuel_ [n=manuel@port-92-205-28-78.dynamic.qsc.de] has quit [] 16:57:34 thanks! 16:58:10 -!- c|mell [n=cmell@v113243.ppp.asahi-net.or.jp] has quit [Read error: 104 (Connection reset by peer)] 16:58:54 c|mell [n=cmell@v113243.ppp.asahi-net.or.jp] has joined #lisp 17:01:02 syamajala2 [n=syamajal@68-116-203-246.dhcp.oxfr.ma.charter.com] has joined #lisp 17:01:55 oudeis [n=oudeis@p11811112.orange.net.il] has joined #lisp 17:02:16 Q: are there some graph drawing examples for McCLIM? 17:03:19 ramus`_ [n=ramus@adsl-75-34-3-31.dsl.chcgil.sbcglobal.net] has joined #lisp 17:03:21 funebre [n=mfunebre@c-71-227-176-54.hsd1.wa.comcast.net] has joined #lisp 17:03:42 -!- rpg [n=rpg@216.243.156.16.real-time.com] has quit [Read error: 110 (Connection timed out)] 17:03:55 -!- lispm [n=joswig@e177155196.adsl.alicedsl.de] has quit ["Leaving..."] 17:04:18 gigamonkey [n=user@adsl-99-2-148-202.dsl.pltn13.sbcglobal.net] has joined #lisp 17:05:59 -!- syamajala2 [n=syamajal@68-116-203-246.dhcp.oxfr.ma.charter.com] has quit [Client Quit] 17:06:12 -!- macdice [n=macdice@78-86-162-220.zone2.bethere.co.uk] has quit ["Leaving"] 17:09:02 -!- Zen_Clar` is now known as Zen_Clark 17:09:36 tcr: i think we can use cliki for that. 17:10:08 tcr: my goal today is to play with clojure, as he's got an interesting reader :) 17:12:41 lispm [n=joswig@e177155196.adsl.alicedsl.de] has joined #lisp 17:13:08 Planet Lisp down? 17:13:20 Nm. My system acting up. 17:14:01 enough lisping for today -- bye 17:14:03 -!- nikodemus [n=nikodemu@cs181229041.pp.htv.fi] has quit ["Leaving"] 17:14:37 -!- ramus` [n=ramus@75.34.7.77] has quit [Read error: 110 (Connection timed out)] 17:14:58 plutonas [n=plutonas@h-188-232.A189.priv.bahnhof.se] has joined #lisp 17:16:56 -!- bombshelter13 [n=bombshel@209-161-240-8.dsl.look.ca] has quit [Client Quit] 17:21:55 hugopt [n=hugo@unaffiliated/hugo] has joined #lisp 17:22:02 -!- hugo [n=hugo@unaffiliated/hugo] has quit [Read error: 104 (Connection reset by peer)] 17:22:45 ehu: hopefully we will get this fixed today :| 17:24:15 -!- hugopt is now known as hugo 17:25:35 -!- dagnachew [n=dagnache@modemcable207.114-201-24.mc.videotron.ca] has quit ["Leaving"] 17:27:37 gz [n=gz@209-6-158-10.c3-0.smr-ubr3.sbo-smr.ma.cable.rcn.com] has joined #lisp 17:29:53 sellout [n=greg@pool-138-88-27-104.res.east.verizon.net] has joined #lisp 17:29:53 -!- gigamonkey [n=user@adsl-99-2-148-202.dsl.pltn13.sbcglobal.net] has quit [Read error: 110 (Connection timed out)] 17:29:54 am I really so bad with google? can't for the life of me make my Mac's Command key act as Meta in Emacs in VMware (linux) 17:30:06 if anyone knows please put me out of my misery :) 17:30:55 gigamonkey [n=user@adsl-99-184-205-122.dsl.pltn13.sbcglobal.net] has joined #lisp 17:32:48 drewc: no hurry. just keeping you informed. 17:35:42 -!- sellout [n=greg@pool-138-88-27-104.res.east.verizon.net] has quit [] 17:35:47 -!- davazp [n=user@77.Red-83-54-164.dynamicIP.rima-tde.net] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 17:37:01 lukego_ : maybe this helps : http://tylerkieft.com/archives/2006/10/05/redefine-the-x11-meta-key-in-mac-os-x/ 17:37:41 -!- Bzek [n=SK_sj@mcc-dyn-30-186.kosnet.ru] has quit ["1 0 \/ 3"] 17:38:01 badon [n=badon@c-71-195-215-153.hsd1.ut.comcast.net] has joined #lisp 17:38:05 mozzyb: thanks, checking it out 17:40:00 -!- biz897 [n=badon@c-71-195-215-153.hsd1.ut.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 17:40:00 -!- badon is now known as biz897 17:40:36 lukego_ : np. If you search for "mac OS X meta key emacs" in google you may get a few interesting hits. 17:40:52 lam_fbx [n=nicolas@aqu33-5-82-245-96-5.fbx.proxad.net] has joined #lisp 17:41:04 no luck. it's always coming through as Control for me. xev reports this suspiciously "remapping to control" line: 17:41:04  17:41:07 XKeysymToKeycode returns keycode: 64 17:41:22 I've googled quite a bit, the issue's specific to running in vmware from what I can see 17:41:35 manuel_ [n=manuel@HSI-KBW-085-216-060-145.hsi.kabelbw.de] has joined #lisp 17:41:43 if XKeysymToKeycode would please not do that then life would probably be just fine 17:42:03 since the previous line is a friendly: state 0x0, keycode 115 (keysym 0xffe9, Alt_L), same_screen YES, 17:42:05 did you look at the preference window of VMWare? 17:42:16 it has a tab for keyboard stuff 17:42:45 Fusion 2.0 that is 17:43:14 user___ [n=user@p54925F5E.dip.t-dialin.net] has joined #lisp 17:44:02 2.0? this is 1.1.3 but I only bought it a few weeks ago 17:44:10 -!- gz` [n=gz@209-6-158-10.c3-0.smr-ubr3.sbo-smr.ma.cable.rcn.com] has quit [Read error: 110 (Connection timed out)] 17:44:21 2.0 is a free update 17:44:45 download from vmware 17:44:47 ah. thanks lispm this sounds very promising! 17:47:40 badon [n=badon@c-71-195-215-153.hsd1.ut.comcast.net] has joined #lisp 17:48:33 -!- icebox [n=chatzill@83.225.181.184] has quit ["ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]"] 17:49:01 -!- biz897 [n=badon@c-71-195-215-153.hsd1.ut.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 17:49:02 -!- badon is now known as biz897 17:50:13 BrianRice-mb [n=briantri@c-24-18-253-20.hsd1.wa.comcast.net] has joined #lisp 17:52:29 vasa [n=vasa@mm-126-93-84-93.dynamic.pppoe.mgts.by] has joined #lisp 17:53:07 -!- Tordek [n=tordek@host172.190-137-247.telecom.net.ar] has quit [Remote closed the connection] 17:53:12 -!- kij [n=user@pc.tv2.dk] has quit [Read error: 110 (Connection timed out)] 17:53:16 pstickne [n=pstickne@c-71-236-177-238.hsd1.wa.comcast.net] has joined #lisp 17:53:44 -!- biz897 [n=badon@c-71-195-215-153.hsd1.ut.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 17:53:51 biz897 [n=badon@c-71-195-215-153.hsd1.ut.comcast.net] has joined #lisp 17:54:32 thom_logn [n=thom@pool-96-229-99-100.lsanca.dsl-w.verizon.net] has joined #lisp 17:56:25 -!- ia [n=ia@89.169.165.188] has quit [Read error: 110 (Connection timed out)] 17:57:41 ia [n=ia@89.169.165.188] has joined #lisp 17:59:12 -!- aiur [n=Jan@218.109.91.183] has left #lisp 17:59:23 lemonodor [n=lemonodo@adsl-76-214-15-172.dsl.lsan03.sbcglobal.net] has joined #lisp 18:03:32 badon [n=badon@c-71-195-215-153.hsd1.ut.comcast.net] has joined #lisp 18:04:06 -!- xjrn [n=jim@c-69-181-213-99.hsd1.ca.comcast.net] has quit [Read error: 113 (No route to host)] 18:04:07 -!- vy [n=user@88.230.190.201] has quit [Remote closed the connection] 18:06:27 -!- lispm [n=joswig@e177155196.adsl.alicedsl.de] has quit ["Leaving..."] 18:08:09 ehu: looking at the commit script problem 18:08:23 badon2 [n=badon@c-71-195-215-153.hsd1.ut.comcast.net] has joined #lisp 18:11:35 -!- borism [n=boris@195-50-206-125-dsl.krw.estpak.ee] has quit [Read error: 145 (Connection timed out)] 18:11:38 -!- pdeb [n=user@pool-68-161-179-70.ny325.east.verizon.net] has quit [Remote closed the connection] 18:12:49 H4nsX: great! 18:13:12 H4nsX: http://tinyurl.com/467875 <- is this upgrade breakage? 18:13:55 michaelw: yes. i can look at that tomorrow, would you be so kind and open a ticket? just email the link to rt@common-lisp.net. 18:14:30 ok 18:15:12 -!- Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has quit [Read error: 113 (No route to host)] 18:15:32 ehu: it should be fixed, can you try? 18:16:51 elurin [n=user@88.224.40.6] has joined #lisp 18:17:21 -!- lam_fbx [n=nicolas@aqu33-5-82-245-96-5.fbx.proxad.net] has quit ["leaving"] 18:18:06 H4nsX: indeed it has! Thanks! 18:18:25 ehu: sure. open tickets for other problems that you may have 18:19:03 H4nsX: when will ViewCVS be back? 18:19:21 and the other viewers 18:19:29 chandler: i only learned about the problem today, will look at that tomorrow. 18:19:59 chandler: opening tickets helps us learning about and keeping track of problems :) 18:20:30 jcowan [n=jcowan@dsl-74-209-19-190.taconic.net] has joined #lisp 18:21:10 What typically happens in compiled implementations when the programmer violates a DYNAMIC-EXTENT declaration? 18:21:22 -!- jmbr [n=jmbr@87.223.197.81] has quit [Read error: 104 (Connection reset by peer)] 18:21:28 jcowan: corruption? 18:21:29 nasal demons 18:21:32 H4nsX: I assumed that breakage in the essential services was already known and didn't need to be reported. 18:21:51 michaelw: I know that *can* happen, but I doubt it actually *does* happen. 18:22:02 jcowan: Ideally, in safe code, an error should be signalled. 18:22:04 a-s [i=root@93.112.68.42] has joined #lisp 18:22:26 besiria [n=user@ppp083212085040.dsl.uom.gr] has joined #lisp 18:22:30 -!- biz897 [n=badon@c-71-195-215-153.hsd1.ut.comcast.net] has quit [Read error: 110 (Connection timed out)] 18:22:31 -!- badon2 is now known as biz897 18:22:33 lispm [n=joswig@e177155196.adsl.alicedsl.de] has joined #lisp 18:22:54 jcowan: in sbcl, you'll likely have pointers to noise on the stack (that is after all the point of dx). 18:22:59 jcowan: I would guess that if the object has been stack-allocated, uses outside that dynamic-extent will refer to something from the stack, which could be treated as anything. 18:23:29 chandler: everything is working fine! 18:23:32 chandler: :) 18:23:47 ISTR that slime and documentation-template both include code for extracting function and GF lambda lists (and methods), but is there a dedicated library for that which paints over some of theimplementation details? 18:23:53 wouldn't it require checking every pointer write to detect whether a pointer to into the stack was being stored somewhere? 18:24:00 -!- badon [n=badon@c-71-195-215-153.hsd1.ut.comcast.net] has quit [Read error: 110 (Connection timed out)] 18:24:20 kreuter: Every write, that is, within the dynamic extent. 18:24:21 H4nsX: By the way, I have noticed a significant increase in spam to lisppaste-requests in the past few days. Do you know of any change which might be responsible for this? Was there a spam filter installed on the old cl.net which is no longer installed on the new one? 18:24:37 Still, write barriers are hardly unknown to science. 18:25:01 michaelw: No, I meant to write it as part of a lambda list frobbing library, but I haven't done so far 18:25:04 How would such a write barrier work? I'm not sure I understand. 18:25:24 tcr: where is the lambda list frobbing? in editor-hints? 18:25:25 jcowan: right, but consider (defun foo () (let ((x ...)) (declare (dynamic-extent x)) (bar x))) (bar (x) (setf x)) 18:25:28 chandler: correct. that is something which indeed is on my list - we'll fix everything eventually, but it will take time. if you have some time to spare, just let us know 18:25:34 -!- emma [n=emma@unaffiliated/emma] has quit [Read error: 104 (Connection reset by peer)] 18:25:39 michaelw: No lambda list frobbing yet 18:25:52 er, the (bar ...) form is supposed to be (defun bar ...) 18:26:09 jcowan: you'd have to check whether you need a write barrier pretty much everywhere... That's almost as expensive as (if not more than) a write barrier already. 18:26:18 tcr: oh, I see 18:26:25 The only checking scheme which springs to mind for me is to precede the stack-allocated object with a magic cookie, which is checked before accesses to the object are allowed to succeed. You'd have to zero out the magic cookie on leaving the dynamic extent of the object. 18:26:28 michaelw: (It's meant to be a standalone library; I wanted to write it after parse-declarations, but continued to work on named-readtables instead, and run into the with-readtable-iterator issue which I spent all my time for.) 18:26:29 and anyway, the location you're writing into might itself be on the stack. 18:26:52 kreuter: Right. That's why I don't understand how this write barrier would work. 18:26:57 -!- darx [n=darx@82-37-17-145.cable.ubr03.wolv.blueyonder.co.uk] has quit [Read error: 110 (Connection timed out)] 18:27:06 yeah. I favor nose goblins. 18:27:17 tcr: do you have notes on it already, or just the idea? 18:27:27 also, you don't care about pointers to the stack being stored in the heap, but only about pointers escaping the extent, so you'd have to check return values and every location that's been written to when you leave the declaration's extent. 18:27:29 I actually like the magic cookie idea :-) But it is still a non-negligible amount of overhead. 18:27:47 Crudely but effectively, you could write-protect the heap and then check every write. (Of course, that works only on processors with adequate instruction restart.) 18:28:15 jcowan: it's certainly possible, but probably not practical for any real work. 18:28:20 jcowan: I am still not sure how that solves the problem wrt stack-allocated objects, and I'm not sure what dynamic-extent declarations do for you if not that. 18:28:24 Whether the cookie idea is negligible or not depends on wider issues, certainly. 18:28:27 michaelw: No, I don't. I've got a few use cases. There are lambda list libraries by pjb, in drei, and in swank. 18:28:31 emma [n=emma@unaffiliated/emma] has joined #lisp 18:30:01 (mind you, some users recently asked me if SBCL could have a mode that monitored at runtime whether their code were violating dynamic-extent semantics.) 18:30:14 -!- H4nsX [n=H4ns@p57A0FB73.dip.t-dialin.net] has quit [] 18:31:23 kreuter: in safe code and in situatiosn where SBCL can actually stack-allocate a dynamic-extent object, I assume? 18:31:34 yeah 18:31:42 why the latter constrain? 18:31:51 implementation issue? 18:32:07 huh? 18:32:14 H4ns1: 18:32:19 aroundp 18:32:38 tcr: if the object is declared dynamic-extent but heap-allocated anyway, there's no really good way to check on future accesses that it was originally declared dynamic-extent 18:32:44 iow, why would it only be possible to allocate certain object types on the stack? 18:33:05 ace4016 [i=ace4016@cpe-76-168-253-20.socal.res.rr.com] has joined #lisp 18:33:14 *chandler* will let someone else answer that question 18:33:17 jcowan: well for one thing, some objects have (conceptually) mutable size. 18:33:17 -!- lispm [n=joswig@e177155196.adsl.alicedsl.de] has quit [] 18:33:44 hash tables & adjustable arrays come to mind 18:33:59 CLOS instances' slot vectors too, I guess. 18:34:12 is there some way you could allocate the dx-able object on the heap and when its life ends, scribble over it in some way that would cause an error were it accessed again? 18:35:02 hefner: interesting suggestion. 18:35:04 how would you scribble over a cons? 18:35:35 for anything with a widetag, sure. just set the widetag to some invalid value 18:36:02 chandler: store pointers to something akin to the unbound marker? 18:36:20 lispm: thanks !! now it seems to be working fine in 2.0 18:36:20 That Sounds Hard. 18:36:26 I just disabled the "key mapping" feature entirely 18:36:31 kreuter: thus resulting in an obscure error even farther from the actual issue? :\ 18:36:33 -!- besiria [n=user@ppp083212085040.dsl.uom.gr] has quit [Remote closed the connection] 18:36:38 lukego_: he left a minute or two ago 18:37:03 pkhuong: it would at least let you know that you were violating dx requirements. 18:37:10 I think I'd rather have a set of should-be-dead objects checked by the gc. 18:37:27 that way you can even have an idea of where the bad pointers were stored. 18:37:30 vmCodes [n=vmCodes@122.167.93.113] has joined #lisp 18:37:44 pkhuong: if you're going to do that, you could just allocate all the dx objects in a special nursery 18:37:49 -!- elurin [n=user@88.224.40.6] has quit [Connection timed out] 18:37:51 chandler: indeed (: 18:37:53 (or maybe that's what you're suggesting) 18:37:55 chandler: oops, thanks :) 18:38:46 pkhuong: the problem is that I imagine (for no good reason, really) that most breaking of dx constraints occurs immediately after the dx scope is exited 18:39:35 (let (my-result) ... (let ((x (blah))) (declare (dynamic-extent x)) ... (setf my-result x) ...) ... (frob my-result)) 18:39:36 but it looks like a bit more work than just checking a couple pointers in a list. 18:39:37 rme [n=rme@pool-70-105-121-210.chi.dsl-w.verizon.net] has joined #lisp 18:39:59 at least, that's the kind of dx constraint breaking I've observed before 18:40:25 where the binding of the dx-declared variable is often hidden in a macro 18:41:39 The one I've seen is (defun foo (f &rest args) (declare (dynamic-extent args)) (apply f args)), where f could possibly returns its args. 18:42:25 and it's the one where such a check would be good because it only breaks in few cases 18:43:51 you could allocate each dx object in its own "nursery", and read+write-protect it on exit from the DX scope 18:44:02 and collect the nurseries after doing a gc-time check 18:44:09 *cough* regions *cough* 18:44:24 manuel__ [n=manuel@HSI-KBW-085-216-060-145.hsi.kabelbw.de] has joined #lisp 18:44:32 The stack is, in a sense, a special-purpose nursery. 18:44:46 it's a bit expensive for a dx cons cell, but as long as this is only for (safety 11) :-) 18:44:47 It's just that it commits infanticide on a regular basis. 18:45:00 I didn't understand how going the gc way would result in error more local to the problematic section; anyone mind elaborating on that a bit? 18:45:52 -!- manuel_ [n=manuel@HSI-KBW-085-216-060-145.hsi.kabelbw.de] has quit [Read error: 104 (Connection reset by peer)] 18:46:19 If it comes to that, an implementation that held all objects on the heap but eagerly discarded objects pointed to by dx variables when the extent ended would also be correct. 18:46:36 tcr: what I just suggested would result in an error immediately when the object is accessed or at gc time if it escapes but is not accessed 18:47:17 jcowan: how do you "discard" an object without doing a gc? 18:47:26 chandler: I was refering to pkhuong objections against kreuter's first idea 18:47:35 tcr: ah 18:47:45 tcr: you get to see the parents containing the faulty pointers 18:47:58 I still like the magic cookie idea, but it also Would Be Hard. 18:48:02 instead of just the first access to its contents. 18:48:11 rme_ [n=rme@pool-70-104-113-122.chi.dsl-w.verizon.net] has joined #lisp 18:48:23 pkhuong: What's parents here? 18:48:38 -!- rme [n=rme@pool-70-105-121-210.chi.dsl-w.verizon.net] has quit [Read error: 60 (Operation timed out)] 18:48:48 parent object(s). 18:49:25 -!- manuel__ [n=manuel@HSI-KBW-085-216-060-145.hsi.kabelbw.de] has quit [Read error: 104 (Connection reset by peer)] 18:49:53 the objects pointing to the dx obj? 18:49:57 yes. 18:51:35 the path to the dx obj must be rootless (not to be a dx violation), right? 18:52:15 tcr: well, at gc time, we know that if a live object pointers to a dx object that's out of scope, then there's a violation. 18:52:51 -!- mathrick [n=mathrick@users177.kollegienet.dk] has quit [Read error: 110 (Connection timed out)] 18:52:53 -!- m4thrick_ [n=mathrick@users177.kollegienet.dk] has quit [Read error: 110 (Connection timed out)] 18:53:25 there's a dozen ways to misuse dx and still pass that check, but at least it's easy to implement and gives some debugging info. 18:53:36 -!- rme_ [n=rme@pool-70-104-113-122.chi.dsl-w.verizon.net] has quit [Read error: 60 (Operation timed out)] 18:54:13 And the parents reflect the section where the dx object was created in so far as the closure (containing the questionable code section) also contains a pointer to that dx object? 18:54:33 gigamonk` [n=user@adsl-99-2-148-46.dsl.pltn13.sbcglobal.net] has joined #lisp 18:54:35 -!- gigamonkey [n=user@adsl-99-184-205-122.dsl.pltn13.sbcglobal.net] has quit [Read error: 110 (Connection timed out)] 18:55:14 tcr: it doesn't tell you where it was created, but it should give you a hint of the code in which the store was executed. 18:55:30 i can't seem to find definitive results on: is there a lisp implementation which will run on the CELL architecture. 18:56:33 bougyman: you'll find a couple that run fine on linux/ppc. For the SPUs themselves, no, and I doubt you want to be coding in anything like lisp for that sort of fiddly architecture. 18:57:38 pkhuong: we were considering http://www-03.ibm.com/systems/bladecenter/hardware/servers/qs22/index.html 18:57:57 but not if application support doesn't match the computing power. 18:58:03 -!- matley [n=matley@83.224.133.248] has quit [Read error: 113 (No route to host)] 18:58:41 i was told programming on them is for some reason best done in C++ 18:58:47 in which case, no thanks. 18:59:00 our Xeon blades are no slouches, those are just remarkable numbers. 18:59:19 -!- vmCodes [n=vmCodes@122.167.93.113] has left #lisp 18:59:43 -!- ramus`_ is now known as ramus` 19:00:46 bougyman: I don't think there are any C++ compilers that will automatically take care of the SPEs either. 19:01:32 They're DSP-like processing units, and you'll need to adjust your code to fit the architecture. You could certainly handle what's being run on the PPU with SBCL or OpenMCL. 19:01:40 chandler: i (nor my friend) know for sure, he was just hired for this position and told me he's going to use C++ because: "It's the fastest on CBEs" 19:01:51 and also mentioned what you just stated. 19:02:18 that the applicatio has to be written for CBE or it will be functionally slower than on a PC architecture. 19:02:24 I don't know what C++ has to do with it. 19:02:34 i didn't understand that part of it, either. 19:02:34 Except for that, everything you just said is true. 19:02:56 which is what got me looking of implementations of languages on that architecture. 19:02:59 I can't find many. 19:03:43 but then I see people putting debian on them. 19:03:56 -!- ivanst [i=ivans@78-0-58-77.adsl.net.t-com.hr] has quit [Read error: 110 (Connection timed out)] 19:03:59 i'm wondering if linux even gives access to what's needed to do it 'right?' 19:04:13 for now I'll just applaud their numbers and wait til they've been out longer. 19:04:14 The Cell is a dual-threaded 64-bit PowerPC combined with seven or eight "synergistic processing elements". 19:04:21 ivanst [i=ivans@89-172-12-188.adsl.net.t-com.hr] has joined #lisp 19:04:22 yes, I know what it's all about. 19:04:37 You can access the SPEs from Linux. 19:05:02 The Linux kernel runs on the PowerPC core, and your application is responsible for managing the SPEs. 19:05:43 it would seem the language choice wouldn't affect the speed, per se, then. 19:05:48 -!- chris2 [n=chris@p5B16BFE0.dip0.t-ipconnect.de] has quit [Read error: 145 (Connection timed out)] 19:05:55 Not the speed of what's being run on the SPEs, no. 19:06:26 If you don't want to write to the SPEs, you would be better off going with the POWER6 (or commodity x86). 19:06:31 but the language runtime or compiler would have to be optimized for that architecture, correct? 19:06:39 we're doing commodity now happily. 19:06:51 i was just intrigued by how adamant he was re: c++. 19:06:53 the SPEs have their own instruction set, and I know they modified gcc to generate the code 19:07:10 so perhaps that's why. 19:07:24 the compiler is more able. 19:07:26 there is not Power-based computers for home 19:07:35 -!- travisjeffery [n=eatsleep@user64-159.vicres.utoronto.ca] has quit [Success] 19:07:37 maxote: that's wildly irrelevant. 19:07:42 maxote: there is one company making laptops in em now. 19:07:42 it's a failure of IBM 19:07:43 bougyman: the "C++ is fasted" statement sounds fishy 19:07:48 but they're not official. 19:07:48 they also have very little memory, which is used for both code and data, so you'll have a hard time coming up with, say, a CL runtime + code + data in each 19:07:50 -!- ChanServ has set mode +o chandler 19:07:58 toshiba is releasing CELL laptops soon, though. 19:08:02 bougyman: the key to good performance on the Cell is good DMA handling 19:08:03 IBM is waiting on Microsoft. 19:08:22 bougyman: uh, Cell laptops? link? 19:08:32 I think it was either 256K or 512K per SPE 19:08:34 jackdaw [n=chris@cpe-076-182-043-222.nc.res.rr.com] has joined #lisp 19:08:38 "Cell laptop" is a bit misleading. 19:08:39 adeth: 256k 19:08:42 michaelw: there's a new qosmio w/ a CBE-based graphic thingy. 19:08:47 oh please 19:08:49 join #boingboing 19:08:52 (doh) 19:08:59 http://www.betanews.com/article/Toshiba_expected_to_sell_laptop_powered_by_Cell_processor/1210631076 19:09:03 http://www.engadget.com/2008/06/14/toshiba-qosmio-g55-features-spursengine-visual-gesture-controls/ 19:09:06 Cells only for gameconsoles and TVs, why not for PCs ? 19:09:11 here I always thought apple should've put a cell in an imac and made it play PS3 games 19:09:15 maxote: please stop. 19:09:23 anyway.. at work we looked into the Cell a year ago or so.. but ended up using nvidia with CUDA.. 19:09:41 ps3 has hardly been a good showcase for cell... 19:09:42 adeht: so only two coding gods needed instead of three? ;) 19:09:43 IIRC the SpursEngine has half the number of SPEs as the Cell in the PS3. 19:09:44 well i mean, hardly wowing 19:10:06 adeht: we're using supermicro superblades on an infiniband backplane. 19:10:16 pkhuong: actually - yeah.. another programmer and me were the only ones writing code for that 19:10:24 no way 19:10:30 bougyman: what kind of app? 19:10:37 chandler: Put it on a free list of some sort, I suppose, and take the risk of corruption if there are stray pointers to it. 19:10:43 adeht: how did cuda work out for you? 19:11:06 jcowan: I don't think that makes anything faster or safer, so I'm not sure I see the point. 19:11:07 hefner: many apps, financial, medical industry backend stuff. 19:11:16 I said it was correct, not faster or safer. 19:11:29 -!- QV [n=viosys@cpe-24-175-17-39.elp.res.rr.com] has quit [Connection timed out] 19:11:51 jcowan: so why would anyone use dx if it's just as slow but unsafe? 19:12:11 actually, that'd be slower than not using dx :-) 19:12:36 michaelw: very well.. we got about x12 speed-up for system.. though their SDK had some bugs (that they fixed) and no async until fairly recently.. nowadays I don't work on that code anymore, though. 19:12:42 Dx is obviously not about safety. 19:13:16 dkcl [n=dkcl@unaffiliated/dkcl] has joined #lisp 19:13:49 The designers of the language (its name is still confidential) are thinking about what, if anything, they can do about stack-allocated objects; I thought I'd collect information from a community with experience with them. 19:14:05 jcowan: it also obviously is not about making things slower. If you're using a free list (instead of, e.g., linear allocation in a thread-local nursery), what's the point? 19:14:35 adeth: interesting; why did you decide against the Cell? 19:15:14 Which will be the new Garbage Collector of the new Millenium? The answer will be one that has not Garbage Collector. 19:15:20 jcowan: theoretically, they can be automatically detected. In practice, I don't think it happens much/at all. Some JVM does it in a couple very restricted cases. 19:15:40 michaelw: business decision - we didn't even get a machine.. wrote code for IBM's simulator. 19:15:40 -!- fisxoj [n=fisxoj@149.43.108.10] has quit ["Ex-Chat"] 19:15:49 maxote: You're not making any sense. Please stop. 19:16:17 adeht: ah, guessed as much. well, that'll be not much of a problem for me. 19:16:23 froog_ [n=david@87.192.28.247] has joined #lisp 19:16:54 jcowan: regions generalise the concept of dx allocation, but are also a bit more work, especially with side effects. 19:16:57 I'd imagine there are a lot of cases where delivering on a CUDA is cheaper than using Cell. 19:17:05 er, CUDA-capable video card 19:17:23 chandler: sure, but as I said, that's not much of an issue to me 19:17:39 Sure. I'd certainly prefer writing to the Cell myself. 19:17:46 schasi [n=schasi@p54A255FD.dip.t-dialin.net] has joined #lisp 19:17:53 chandler: yes.. but it has its own problems, e.g. power consumption 19:17:54 chandler: that's why I decided for it :) 19:17:57 I'm cautiously optimistic about Larabee. 19:18:09 -!- chandler has set mode -o chandler 19:18:35 depending on how the details shake out, you really *could* run a Lisp on Larabee :-) 19:18:59 well it'd be interesting to write some kind of Lisp for the CUDA.. though iirc they didn't implement code unloading and such. 19:19:06 -!- dabd [n=dabd@85.139.98.38] has quit [Read error: 110 (Connection timed out)] 19:19:07 adeht: I'd like to see the numbers on that, actually. I know the PS3 is a bit of a monster. 19:20:29 obsethryl [n=llort@unaffiliated/obsethryl] has joined #lisp 19:21:27 chandler: aren't most of the PS3 GFLOPS from the GPU, though? 19:21:45 I've no idea. 19:21:59 (I meant "bit of a monster" in terms of power consumption.) 19:22:13 ah 19:22:27 chandler: I don't really know the numbers, because we haven't decided on a particular configuration.. and so far it seems we're gonna ditch nvidia and just use a plain Intel machine with faster but less accurate algos. 19:22:33 -!- Aankhen`` [n=pockled@122.162.154.37] has quit ["Powercut"] 19:22:53 -!- BrianRice-mb [n=briantri@c-24-18-253-20.hsd1.wa.comcast.net] has quit [] 19:24:03 *chandler* is less interested in heavily numeric applications that tend to favor stream processors and the Cell than in throughput-oriented applications that favor multicore SoCs like the Niagara 19:24:15 PS3 has 1 doctor, 7 nursers (instead if 8), 1 hidden hypervisor nurser and 1 sony propietary gpgpu accelerator 19:24:28 /if/of 19:24:45 has maxote ever uttered a single sentence that made sense? 19:24:57 -!- delYsid [n=user@debian/developer/mlang] has quit [Connection timed out] 19:25:05 not to the best of my recall, no. 19:25:15 minion: logs? 19:25:16 logs: #lisp logs are available at http://ccl.clozure.com/irc-logs/lisp/ ; older logs may be available at http://tunes.org/~nef/logs/lisp/ (down as of 2008/09/24) 19:25:42 Doctor and nurse may be a somewhat more palatable metaphor than master and slave, to some. 19:26:06 09:41:55 Apple is for gays 19:26:12 (from yesterday's logs) 19:26:26 jcowan: master/worker WFM 19:26:50 dear leader, heart of the people / proud members of the revolutionary state 19:27:00 great. tunes is down. I'll have to ask beach for a copy of his older logs. 19:27:07 -!- ChanServ has set mode +o chandler 19:27:22 -!- chandler has set mode +b maxote!*@* 19:27:52 -!- chandler [n=chandler@opendarwin/developer/chandler] has been kicked from #lisp 19:27:55 -!- chandler has set mode -o chandler 19:29:12 chandler: FWIW, unless maxote was harassing anyone, muting would've been enough 19:29:47 Fair enough. I don't often think about using the mute option. 19:29:51 -!- ChanServ has set mode +o chandler 19:29:58 -!- chandler has set mode -b maxote!*@* 19:30:04 -!- chandler has set mode +b %maxote!*@* 19:30:07 -!- chandler has set mode -o chandler 19:31:17 Back-on-topic question: Is there a limit to the number of arguments to a macro? 19:31:22 -!- froog [n=david@87.192.28.247] has quit [Connection timed out] 19:31:32 basant_ [n=basant@59.184.124.50] has joined #lisp 19:31:44 call-arguments-limit? 19:32:04 Supposing that the macro receives them via &rest, &body, or &whole, not a number of required or optional arguments that may exceed lambda-parameters-limit. 19:32:05 adeht: different concern. 19:32:06 -!- basant_ [n=basant@59.184.124.50] has quit [Read error: 104 (Connection reset by peer)] 19:32:08 at least that's the upper bound 19:32:12 basant_ [n=basant@59.184.124.50] has joined #lisp 19:32:30 hello 19:32:33 adeht: the macro's arguments aren't passed as individual arguments to the macro-function, so I'm not sure where that would come in. 19:33:49 I suppose it's academic, given that call-arguments-limit is sufficiently large on the implementations I'm concerned about. 19:34:09 pkhuong: I have a copy of the logs up to september, which I am (slowly) uploading to the web 19:34:39 chandler: you mean lie-to-the-user large? ;) 19:34:48 hefner: bzipped and all that, I assume. 19:35:03 pkhuong: yeah, it's about 100 MB. 19:35:22 pkhuong: large enough for something that starts as a list of XML tag attributes in the source code :-) 19:35:35 "more than fifty, and more than I'd want to type in" 19:36:46 mathrick [n=mathrick@users177.kollegienet.dk] has joined #lisp 19:36:47 m4thrick [n=mathrick@users177.kollegienet.dk] has joined #lisp 19:37:00 rme [n=rme@pool-68-238-8-171.chi.dsl-w.verizon.net] has joined #lisp 19:38:23 asdf25 [n=jeff@pool-96-241-127-243.washdc.fios.verizon.net] has joined #lisp 19:38:58 _cheerios [n=Jack@dsl-hkibrasgw3-fe74fb00-140.dhcp.inet.fi] has joined #lisp 19:39:06 -!- envi^home [n=envi@220.121.234.156] has quit ["Leaving"] 19:41:08 -!- jackdaw [n=chris@cpe-076-182-043-222.nc.res.rr.com] has quit ["Leaving"] 19:47:46 -!- lemoinem [n=swoog@modemcable125.83-81-70.mc.videotron.ca] has left #lisp 19:48:21 -!- tst__ [n=Tim@p4FD2F399.dip.t-dialin.net] has quit ["leaving"] 19:48:34 besiria [n=user@ppp083212085040.dsl.uom.gr] has joined #lisp 19:48:40 for all you aspiring cyberstalkers, http://vintage-digital.com/hefner/lisplogs.tar.bz2 (93 MB) is #lisp logs from 2001 to sometime last month 19:49:06 2000, even. 19:49:18 hefner: wow, good pipe. 19:49:19 *chandler* stalks hefner 19:50:09 pkhuong: 13 minutes? doesn't seem terribly fast to me 19:50:12 ecraven [n=nex@plc31-103.linzag.net] has joined #lisp 19:50:57 750KBps from there to here. 19:51:29 Oh, the download pipe. Yes, that's pretty good. 19:51:45 :-) 19:53:21 lispm [n=joswig@e177155196.adsl.alicedsl.de] has joined #lisp 19:57:03 -!- lukego_ [n=lukegorr@cpe-76-87-84-57.socal.res.rr.com] has quit [Read error: 60 (Operation timed out)] 19:57:26 rpg [n=rpg@216.243.156.16.real-time.com] has joined #lisp 20:00:02 dabd [n=dabd@85.139.98.38] has joined #lisp 20:00:49 -!- yhara [n=yhara@84.215.12.221.megaegg.ne.jp] has quit [Read error: 113 (No route to host)] 20:05:02 rdd [n=user@c83-250-142-219.bredband.comhem.se] has joined #lisp 20:05:16 hans_ [n=H4ns@p57A0FB73.dip.t-dialin.net] has joined #lisp 20:05:40 araujo [n=araujo@gentoo/developer/araujo] has joined #lisp 20:09:34 larrytheliquid [n=larrythe@nmd.sbx10362.wintefl.wayport.net] has joined #lisp 20:12:54 can anyone recommend an antispam package that i could install on the new common-lisp.net? 20:14:08 -!- hans_ is now known as H4nsX 20:14:32 I don't know. But I'm curious whether it's the for the new server or if it's a redesign? :) 20:14:43 tic: ? 20:15:24 "the new common-lisp.net" -- it's changed servers, right? and I also heard drewc planning redesigning it. 20:15:47 tic: drew wants to redesign the web appearance, but we'll not move to another server, no. 20:15:54 tic: painful enough to do it once. 20:16:26 yean new in this case was the new server, not the redesign, which is not ready. 20:16:36 s/yean// 20:16:47 alright, that answers it. 20:17:07 *H4nsX* installs spamassassin 20:17:38 H4nsX: we're using milter-sender with good results 20:17:47 (together with SA) 20:17:52 lukego [n=lukegorr@cpe-76-87-84-57.socal.res.rr.com] has joined #lisp 20:18:05 -!- basant_ [n=basant@59.184.124.50] has quit [] 20:18:08 michaelw: ok, will look at that, too. 20:18:30 <_8david> hans_: did anyone fix the default mailman configuration, which accepts any spam from a forged @common-lisp.net address? 20:19:12 _8david: don't think so. that is gking's area. open a ticket if you care. 20:19:17 drdo [n=psykon@167.111.54.77.rev.vodafone.pt] has joined #lisp 20:19:52 rvirding [n=chatzill@h235n3c1o1034.bredband.skanova.com] has joined #lisp 20:19:57 michaelw: ah, milter-sender is something sendmail specific, right? we're using exim now, so we possibly don't need that. 20:20:24 H4nsX: yes, okay 20:21:51 *ehu* gets abcl below 50 failures on ANSI tests - not through the removal of tests, btw 20:22:01 what git gui do the cool kids use? 20:22:11 ehu: neat 20:22:32 lukego pasted "latex that uses too little page" at http://paste.lisp.org/display/68326 20:22:34 ehu: nice going. 20:22:35 njsg [n=njsg@unaffiliated/njsg] has joined #lisp 20:22:40 tcr: tcr: git citool and gitk for me 20:22:53 ehu: how is the quality (not speed) of CLOS right now? I don't think the ANSI tests cover that area as comprehensively as some other areas. 20:23:10 is there someone good with latex who can help me tweak that paste to make (much) more use of page space? 20:23:34 lukego: \usepackage{fullpage} 20:24:03 -!- elindio [n=badiss@gar13-5-88-161-23-155.fbx.proxad.net] has quit ["leaving"] 20:24:06 (also, bah, no pbook? ;) 20:24:15 michaelw: what do you use the first for, what the second? 20:24:23 chandler: that's right. V-ille is looking into using Closer tests to identify our quality level there. However, since we still don't support the long form of define-method-combination, I'm not sure the level is too good. 20:24:28 unfortunately. 20:25:18 I haven't had much time to look at the Closer tests yet, real life has required too much attention. 20:25:32 I wonder if just ripping PCL out of SBCL or CMUCL is hopelessly difficult. 20:25:55 -!- lemonodor [n=lemonodo@adsl-76-214-15-172.dsl.lsan03.sbcglobal.net] has quit [] 20:27:02 tcr: gitk for looking at branches and browsing; citool for (un)staging; oh, I also use gitsum (from emacs) 20:27:08 mejja [n=user@c-7d2472d5.023-82-73746f38.cust.bredbandsbolaget.se] has joined #lisp 20:27:25 chandler: I think that'd be a rather boring solution 20:27:26 b4 [n=b4@0brg.xs4all.nl] has joined #lisp 20:27:29 -!- b4 [n=b4@0brg.xs4all.nl] has quit [Read error: 104 (Connection reset by peer)] 20:27:29 chandler: abcl and sbcl use similar constructs to store class references in objects (through layouts), but the implementations are actually very different. 20:27:51 V-ille: but maybe very effective ? :-) 20:27:53 michaelw: I'm looking for something to cherry-pick conveniently 20:28:01 ehu: maybe, maybe not 20:28:12 pkhuong: that's pbook output 20:28:38 I'd much rather read AMOP etc. and do it from scratch, or nearly from scratch 20:28:43 lukego: k. For some reason I thought you were writing that by hand. 20:28:48 tcr: I seldomly do that. I merge a lot, though 20:30:51 -!- funebre [n=mfunebre@c-71-227-176-54.hsd1.wa.comcast.net] has quit [] 20:31:20 hmm clbuild has a problem when trying to ./cbuild update fare-matcher 20:31:36 fare.tunes.org|66.114.33.57|:80 won't respond 20:31:43 who do I contact about this? 20:32:14 ehu: Any progress on the other fronts, commit rights, scripting branch? 20:32:23 <_8david> H4nsX: okay... No, I fixed it for my own lists, and gwking knows about it. It is or was the number one reason (and reason #2 and reason #3 :-)) for spam on cl.net, so I thought I'd mention it. 20:34:29 pkhuong: thanks, that helps a LOT! 20:35:12 V-ille: commit rights may come up soon, as soon as drewc, gking or H4nsX get to it. 20:35:17 _8david: *sigh* i forgot about spam when i accepted responsibility for email on c-l.net - google mail collapses spam to one number for me, and i have not thought about it for more than a year 20:35:24 scripting branch is on its way. 20:35:28 H4nsX: I've learned more about what we actually run in cl-net in the last two days than i have in a year of admin. 20:35:39 ehu: is there an outstanding request for that? 20:35:42 pkhuong: I'd still like to use more of the left margin though, do you know a way? 20:35:44 yes. 20:35:49 drewc: yeah - painful for users, but helpful for us 20:36:05 ehu: i'll look it up, you know the subject? 20:36:20 ehu: did you submit a ticket or did it go the the old admin list? 20:36:34 old admin list. 20:36:40 *sigh* 20:36:41 commit rights for armed-bear. 20:36:44 <_8david> H4nsX: yeah, I didn't really notice the problem myself except when I got unsubscribed automatically from my own lists because my mail server rejected all the spam that came from them :-) 20:36:45 *H4nsX* don't have that. 20:36:57 lukego: once fullpage is loaded, you get to fiddle some more with the margins. shows how. 20:36:57 H4ns1: i should have it in my gmail, i'll do the work 20:37:08 drewc: ok, thanks. 20:39:04 ehu: actually i don't have it either.. commit rights for who? 20:39:06 actually I'm way over 80 columns so I should attack this problem at its source and fix that :) 20:41:05 ehu: please resend the request, and if at all possible, cc these guys on their personal email accounts. 20:41:41 Or if there's a fancy ticket system, submit there. Whatever. :) 20:42:25 -!- njsg [n=njsg@unaffiliated/njsg] has quit ["gone"] 20:42:57 -!- ivanst [i=ivans@89-172-12-188.adsl.net.t-com.hr] has quit ["Leaving"] 20:43:25 -!- oudeis [n=oudeis@p11811112.orange.net.il] has quit ["This computer has gone to sleep"] 20:44:01 O_4 [n=souchan@ip-118-90-101-9.xdsl.xnet.co.nz] has joined #lisp 20:45:38 -!- larrytheliquid [n=larrythe@nmd.sbx10362.wintefl.wayport.net] has quit [] 20:46:48 larrytheliquid [n=larrythe@nmd.sbx10362.wintefl.wayport.net] has joined #lisp 20:49:29 -!- mishok13 [n=gdmfsob@92.112.254.10] has quit [Read error: 110 (Connection timed out)] 20:49:57 mishok13 [n=gdmfsob@92.112.86.92] has joined #lisp 20:51:24 -!- c|mell [n=cmell@v113243.ppp.asahi-net.or.jp] has quit ["hiking"] 20:53:23 -!- plutonas [n=plutonas@h-188-232.A189.priv.bahnhof.se] has quit [Remote closed the connection] 20:54:00 kreuter: Are you here? 20:54:21 maxote [n=maxote@84.79.67.254] has joined #lisp 20:56:47 drewc: np. Ill resend. where's the ticketing system for filing? 20:57:03 it's admin@common-lisp.net now. 20:57:21 we probably lost your previous mail somewhere in the switch :) 20:57:23 ak70 [n=ak70@195.158.90.55] has joined #lisp 20:57:36 np. 20:58:04 drewc: so, what's rt@c-l.net then? 20:58:17 michaelw: the same thing 20:58:26 ah 20:59:32 -!- Jarvellis [n=andrew@dsl-217-155-101-22.zen.co.uk] has quit [Read error: 110 (Connection timed out)] 20:59:40 -!- stassats` [n=stassats@wikipedia/stassats] has quit [Remote closed the connection] 21:00:01 TravisD [n=travisdi@d199-126-159-163.abhsia.telus.net] has joined #lisp 21:00:06 stassats` [n=stassats@ppp78-37-151-232.pppoe.avangarddsl.ru] has joined #lisp 21:01:26 dagnachew [n=dagnache@modemcable207.114-201-24.mc.videotron.ca] has joined #lisp 21:06:24 -!- b4|hraban [n=b4@0brg.xs4all.nl] has quit ["Leaving"] 21:06:24 -!- dagnachew [n=dagnache@modemcable207.114-201-24.mc.videotron.ca] has quit ["Leaving"] 21:15:10 -!- lispm [n=joswig@e177155196.adsl.alicedsl.de] has quit [] 21:20:33 Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has joined #lisp 21:24:56 -!- vasa [n=vasa@mm-126-93-84-93.dynamic.pppoe.mgts.by] has quit ["I am not vasya, i am vasa"] 21:25:36 -!- mrsolo_ [n=mrsolo@adsl-68-126-209-194.dsl.pltn13.pacbell.net] has quit ["This computer has gone to sleep"] 21:25:37 plutonas [n=plutonas@h-188-232.A189.priv.bahnhof.se] has joined #lisp 21:25:55 -!- moghar [n=user@unaffiliated/moghar] has quit [Remote closed the connection] 21:28:39 if i have in-package at the top of my file and code underneath, is there a way to execute some code at the end of the file, whether that be a hook into existing the file or exiting the package? 21:28:54 into exiting* 21:29:00 -!- malumalu [n=malu@hnvr-4dbb8a0c.pool.einsundeins.de] has quit ["Verlassend"] 21:31:17 -!- jgracin [n=jgracin@82.193.208.195] has quit [Remote closed the connection] 21:31:23 larrytheliquid: i don't quite understand the question, can you re-phrase? what does the in-package have to do with evaluation of a file? 21:31:26 larrytheliquid: i'm not quite sure what you mean, but ASDF will allow you to do things like that. 21:34:07 sounds like larry wants (eval-when (:after-everything-else) ...) :) 21:34:11 daniel [i=daniel@unaffiliated/daniel] has joined #lisp 21:35:09 specifically, i have something like (in-package :my-unit-test-library), then some code defining tests, at the end of the file i normally type out (my-unit-test-library:run-tests), but it would be nice if that would happen automatically 21:35:41 well, if you type it once at the end of the file, that's what it will do. 21:35:52 can you write down an algorithm for how the system is meant to know without you typing it that you want that function run? 21:36:21 Krystof: an exit-file, or exit-package hook method would work 21:36:28 how? 21:36:40 how can you tell one file from another? 21:36:40 (with-magic (read-my-mind)) :) 21:36:58 larrytheliquid: a package is never "exited". 21:37:04 how can you tell one "exit-package" (not a useful concept) from another? 21:37:29 Krystof: by defining it as part of the package, on-exit do-this 21:37:30 Krystof: exit-file could be pretty well defined; consider that we already have e.g. *load-truename* as precedent for LOAD-extent things 21:37:42 (not saying it's a good idea, just that it's not incoherent) 21:37:52 -!- ecraven [n=nex@plc31-103.linzag.net] has quit ["bbl"] 21:38:06 -!- LiamH [n=nobody@138.88.126.7] has quit ["Leaving."] 21:38:44 larrytheliquid: describe what you want, fully 21:38:54 not just by waving your hands and saying "you could do it this way" 21:39:08 write it out, completely, describing when things happen and when they don't 21:39:14 it will be a learning experience for you 21:39:36 puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has joined #lisp 21:40:59 -!- wchogg [n=wchogg@h216-165-147-7.mdtnwi.dsl.dynamic.tds.net] has quit [Remote closed the connection] 21:42:13 something like this http://paste.lisp.org/display/68329 21:43:25 larrytheliquid: why do you want that? are you aware that your package is never "exited" when there is no second in-package form in your file? 21:44:22 *JuanDaugherty* wonders if larrytheliquid is a changeling/founder in the startrek/ds9 sense. 21:44:47 when you compile a file in common-lisp-user, does it switch back to common-lisp-user at the end of compilation? 21:45:06 H4nsX: ...or never entered, to start with, if there is never an in-package form 21:46:44 larrytheliquid: the compilation unit is separate from your repl session. i will propably be corrected by someone who knows better, but i think for my personal purposes, thinking of *package* be separately bound in the compilation unit is sufficiently precise 21:47:10 larrytheliquid: so think bindings of special variables, think *package* as being a special variable. 21:47:51 drewc, h4nsX: request resent. 21:47:53 H4ns1: actually, it is LOAD, specifically, that rebinds *package*, compilation units don't affect it 21:48:05 kpreid: thank you :) 21:48:15 ehu: ok, i got it. 21:48:43 good :-) 21:48:57 -!- rpg [n=rpg@216.243.156.16.real-time.com] has quit [Read error: 110 (Connection timed out)] 21:48:57 alright, so it sounds like the initial recommendation of specifying some code to be executed after compilation in asdf is the best way to go 21:49:43 -!- daniel_ [i=daniel@unaffiliated/daniel] has quit [Read error: 110 (Connection timed out)] 21:49:46 drewc, you got teh superior mind reading skillz! 21:50:06 H4nsX: i win! 21:50:17 thanks to all for clearing things up 21:50:18 BrianRice-mb [n=briantri@dsl081-166-030.sea1.dsl.speakeasy.net] has joined #lisp 21:53:58 lemoinem [n=swoog@dsl-67-204-19-122.acanac.net] has joined #lisp 21:54:02 anyone know if sbcl on x86-64 can use 32gb of ram? 21:55:20 sylvander_ [n=sylvande@92.19.63.201] has joined #lisp 21:55:28 So, I've been thinking about this for a little while now, but I'm kind of at a loss as to where I should start. I have been asked to implement an interpreter for a simple functional language which has no environment. Given a list of function definitions, and a single function application, I am evaluate it by "replacing equals with equals" 21:55:44 Does anyone have any general suggestions? 21:55:47 -!- dkcl [n=dkcl@unaffiliated/dkcl] has quit [Remote closed the connection] 21:56:47 asdf25: I managed to allocate 8G, but don't feel like getting to close to 16G. 21:57:00 I can't think of any hard limit 22:00:42 dkcl [n=dkcl@unaffiliated/dkcl] has joined #lisp 22:03:05 TravisD: I'm guessing but it it sounds a bit like inlining or macro expansion 22:03:58 -!- larrytheliquid [n=larrythe@nmd.sbx10362.wintefl.wayport.net] has quit [] 22:04:00 in the body of the function application, when another function is called, you look it up and replace the call with its definition 22:04:03 or something :) 22:04:18 hehe yeah 22:04:55 replacing equals with equals is bit like a zen koan 22:08:59 -!- hugo [n=hugo@unaffiliated/hugo] has quit [Read error: 110 (Connection timed out)] 22:09:02 asdf25: 34,400,046,672 bytes for 805,143 dynamic objects (space total.) 22:09:32 ah great, thanks 22:09:58 -!- ehu [i=52aa21ad@gateway/web/ajax/mibbit.com/x-d6a20224748b7bed] has quit ["http://www.mibbit.com ajax IRC Client"] 22:10:26 -!- tsuru [n=user@c-69-245-36-64.hsd1.tn.comcast.net] has quit [Remote closed the connection] 22:10:33 michaelw: that is a lot of bytes 22:12:18 Xach: *print-length* and *print-array* are very dear to me 22:12:34 drdo` [n=psykon@167.111.54.77.rev.vodafone.pt] has joined #lisp 22:16:30 -!- puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has quit [Remote closed the connection] 22:19:14 puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has joined #lisp 22:19:40 oudeis [n=oudeis@DSL217-132-37-185.bb.netvision.net.il] has joined #lisp 22:21:13 -!- dkcl [n=dkcl@unaffiliated/dkcl] has quit [Remote closed the connection] 22:21:22 -!- puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has quit [Remote closed the connection] 22:22:01 puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has joined #lisp 22:22:54 Are there any non-command line lisp interpreters available for mac? 22:24:48 TravisD: common lisp is mostly compiled today. lispworks is one implementation with a (mac) gui, albeit not free 22:25:26 Right now I'm using Eclipse, and it has a window called "repl". The only reason I am unhappy with the command line version that I have installed (sbcl), is that I cannot access previous commands 22:25:56 TravisD: SLIME has command line history 22:26:00 TravisD: most people in this channel use slime as their ide. 22:26:07 hmm 22:26:12 does that require emacs? 22:26:17 TravisD: yes 22:26:28 maybe SBCL should print a warning at starting telling people they're doing it wrong if they see this :) 22:26:32 *H4nsX* has a deja vu 22:26:47 *startup 22:27:01 "you're doing it wrong" 22:27:16 -!- drdo [n=psykon@167.111.54.77.rev.vodafone.pt] has quit [Connection timed out] 22:27:17 no doubt. 22:27:23 i want to nail, but i can't be made to use a hammer! 22:27:33 Cusp is available for Eclipse: http://www.bitfauna.com/projects/cusp/cusp.htm#download disclaimer: I don't know how well it works these days 22:34:03 any tips on getting slime working on OS X? 22:34:18 -!- besiria [n=user@ppp083212085040.dsl.uom.gr] has quit [Remote closed the connection] 22:34:51 TravisD: which problem are you having? 22:35:01 well, how do I activate it from emacs? haha 22:36:51 milanj [n=milan@77.46.251.69] has joined #lisp 22:36:52 -!- sylvander_ [n=sylvande@92.19.63.201] has quit ["leaving"] 22:36:59 TravisD: like described in the manual 22:37:46 -!- puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has quit [Remote closed the connection] 22:38:12 davazp [n=user@77.Red-83-54-164.dynamicIP.rima-tde.net] has joined #lisp 22:38:24 puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has joined #lisp 22:40:10 oudeis_ [n=oudeis@89-139-183-47.bb.netvision.net.il] has joined #lisp 22:40:11 -!- puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has quit [Remote closed the connection] 22:40:39 dkcl [n=dkcl@unaffiliated/dkcl] has joined #lisp 22:40:48 -!- antifuchs [n=asf@86.59.21.99] has quit ["Caught sigterm, terminating..."] 22:40:50 puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has joined #lisp 22:41:02 kij [n=user@x1-6-00-09-5b-fd-49-01.k891.webspeed.dk] has joined #lisp 22:41:21 -!- puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has quit [Remote closed the connection] 22:42:01 puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has joined #lisp 22:42:34 -!- puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has quit [Remote closed the connection] 22:43:14 puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has joined #lisp 22:45:31 does anyone know how to match a newline in an Elisp regexp? I want to match a multi-line block of text 22:46:22 someone in #emacs might know 22:46:59 -!- ak70 [n=ak70@195.158.90.55] has quit ["Leaving."] 22:47:32 Tordek [n=tordek@host172.190-137-247.telecom.net.ar] has joined #lisp 22:48:27 antifuchs [n=asf@baker.boinkor.net] has joined #lisp 22:48:33 lukego: C-q C-j 22:48:40 enigmus [n=e@S010600119505ec11.vs.shawcable.net] has joined #lisp 22:48:56 \n should work, too 22:52:13 -!- oudeis [n=oudeis@DSL217-132-37-185.bb.netvision.net.il] has quit [Read error: 110 (Connection timed out)] 22:52:31 sellout [n=greg@pool-138-88-27-104.res.east.verizon.net] has joined #lisp 22:52:51 hm yes I was using $ instead of \n which was no good 22:52:54 thanks lads 22:53:00 -!- dkcl [n=dkcl@unaffiliated/dkcl] has quit [Remote closed the connection] 22:53:30 -!- drdo` [n=psykon@167.111.54.77.rev.vodafone.pt] has quit [Remote closed the connection] 22:53:50 clog_ [n=nef@bespin.org] has joined #lisp 22:58:50 Balita [n=ase23azz@p7220-ipbfp1401fukuokachu.fukuoka.ocn.ne.jp] has joined #lisp 22:58:53 -!- name [n=name@sburn/devel/name] has quit ["Lost terminal"] 23:00:59 -!- schasi [n=schasi@p54A255FD.dip.t-dialin.net] has quit ["Verlassend"] 23:02:20 -!- dabd [n=dabd@85.139.98.38] has quit ["Ex-Chat"] 23:03:10 milanj- [n=milan@77.46.251.69] has joined #lisp 23:03:38 -!- milanj [n=milan@77.46.251.69] has quit [Read error: 60 (Operation timed out)] 23:04:05 -!- puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has quit ["Konversation terminated!"] 23:04:29 -!- jcowan [n=jcowan@dsl-74-209-19-190.taconic.net] has left #lisp 23:05:28 -!- tcr [n=tcr@host145.natpool.mwn.de] has quit ["Leaving."] 23:14:09 -!- lemoinem [n=swoog@dsl-67-204-19-122.acanac.net] has quit [Read error: 104 (Connection reset by peer)] 23:14:30 -!- sellout [n=greg@pool-138-88-27-104.res.east.verizon.net] has quit [] 23:14:35 -!- kij [n=user@x1-6-00-09-5b-fd-49-01.k891.webspeed.dk] has quit [Read error: 110 (Connection timed out)] 23:16:22 something tells me pbook should be 5% as much code as it is 23:16:37 lemoinem [n=swoog@dsl-67-204-19-122.acanac.net] has joined #lisp 23:18:23 dboswell [n=dave@208.177.146.111.ptr.us.xo.net] has joined #lisp 23:21:38 milanj [n=milan@77.46.251.69] has joined #lisp 23:25:27 -!- clog [n=nef@bespin.org] has quit [Connection timed out] 23:25:27 -!- clog_ is now known as clog 23:26:10 pbook? 23:26:41 -!- rme [n=rme@pool-68-238-8-171.chi.dsl-w.verizon.net] has quit [] 23:29:30 http://fresh.homeunix.net/~luke/misc/emacs/pbook.el / http://fresh.homeunix.net/~luke/misc/emacs/pbook.pdf 23:31:35 -!- milanj- [n=milan@77.46.251.69] has quit [Connection timed out] 23:31:38 pkhuong_ [n=pkhuong@modemcable125.83-81-70.mc.videotron.ca] has joined #lisp 23:33:15 -!- lemoinem [n=swoog@dsl-67-204-19-122.acanac.net] has quit [Read error: 110 (Connection timed out)] 23:33:31 if you just count the code and reduce it by 95% that wouldn't be much code at all :) 23:33:54 -!- spacebat_ is now known as spacebat 23:34:27 -!- jlouis [n=jlouis@port1394.ds1-vby.adsl.cybercity.dk] has quit ["Leaving"] 23:35:41 lemoinem [n=swoog@dsl-67-204-19-122.acanac.net] has joined #lisp 23:36:06 that would be appropriate because it hardly actually does anything :) 23:37:32 maybe pbook encourages wordy programming ;) 23:37:35 dkcl [n=dkcl@unaffiliated/dkcl] has joined #lisp 23:37:48 -!- rvirding [n=chatzill@h235n3c1o1034.bredband.skanova.com] has quit ["ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]"] 23:38:37 -!- dkcl [n=dkcl@unaffiliated/dkcl] has quit [Remote closed the connection] 23:39:28 dkcl [n=dkcl@unaffiliated/dkcl] has joined #lisp 23:41:02 -!- oudeis_ [n=oudeis@89-139-183-47.bb.netvision.net.il] has quit ["This computer has gone to sleep"] 23:42:38 -!- pkhuong [n=pkhuong@modemcable125.83-81-70.mc.videotron.ca] has quit [Read error: 110 (Connection timed out)] 23:44:12 -!- enigmus [n=e@S010600119505ec11.vs.shawcable.net] has quit [Read error: 110 (Connection timed out)] 23:45:48 hugo [n=hugo@unaffiliated/hugo] has joined #lisp 23:47:50 replor_ [n=replor@ntkngw375028.kngw.nt.ftth.ppp.infoweb.ne.jp] has joined #lisp 23:48:21 -!- Khisanth [n=Khisanth@151.204.138.236] has quit ["moving router"] 23:49:52 -!- obsethryl [n=llort@unaffiliated/obsethryl] has quit [Connection reset by peer] 23:52:41 -!- yango [n=yango@unaffiliated/yango] has quit [Read error: 104 (Connection reset by peer)] 23:54:23 -!- Nshag [i=user@Mix-Orleans-106-4-243.w193-248.abo.wanadoo.fr] has quit ["Quitte"] 23:54:42 -!- TravisD [n=travisdi@d199-126-159-163.abhsia.telus.net] has quit [] 23:57:42 yango [n=yango@unaffiliated/yango] has joined #lisp 23:58:33 Khisanth [n=Khisanth@151.204.138.236] has joined #lisp