14:32:19 ccl-logbot [~ccl-logbo@setf.clozure.com] has joined #lisp 14:32:19 14:32:19 -!- names: ccl-logbot madnificent rme galaxywatcher kpreid Xach billitch tritchey asarch Shaftoe simplechat Dawgmatix nyef nha dlowe m`` gko relcomp fiveop lusory LiamH zomgbie revel0 kejsaren legumbre stettber` grouzen_ [mzm] Dodek ZabaQ splittist Harag Krystof Yuuhi antoszka lpolzer swilde Edward_ Joreji HG` mbohun argiopeweb_ stis leo2007 jdz Axioplase_ hlavaty lvillani tfb Guest22022 blandest mathrick attila_lendvai cataska zoldar bozhidar joast zophy trebor_dki 14:32:19 -!- names: aerique pookleblinky levente_meszaros martin2 mcsontos kleppari lhz ASau OmniMancer mvilleneuve hdurer`` Kolyan araujo redline6561 mindCrime_ Euthydemus Snamich emma Draggor tychoish Ambiguity ljames x-ip housel antifuchs stepnem setheus_ mtd borism ace4016 Phoodus abend Taggnostr devon p_l|backup Intensity sellout kephas HDurer bandu daniel_ srcerer qebab gz` specbot frodef peddie_ Amadiro djm Tanami erk_ z0d DrForr ennen aoh hc_e cYmen nasloc__ burton 14:32:19 -!- names: cpt_nemo sjbach MrWGW turbo24prg franki^ nullman ineiros moeffju[Away] codemonkeyx johanbev tic spacebat Yamazaki-kun fda314925 hanneso [df] Buganini vsync foom yacin schmrkc kencausey ri4a fe[nl]ix hypno joshe johs sid3k jsnell az tsuru ramus reb` drewc derrida Tordek boyscared christoph_debian rapacity knobo` Aisling_ stray_hound__ xristos yahooooo p8m hohum loomer prip abeaumont zeroish carrl TDT rootzlevel clop joga Tasunteld mjonsson @slyrus sykopomp 14:32:19 -!- names: fihi09 rtoym hugod thunk rahul sad0ur erg c|mell adeht Odin- ``Erik Patzy MetalDust nuba Salamander df_aldur arbscht pkhuong_ m4thrick tomaw rtra varjag felipe mgr xinming tcr gzip4 rdd barcon332 ikki beach kloeri phadthai rread delYsid lonstein_ bzzbzz_ benny copec billstclair pchrist dto Ralith rgrau HET2 BrianRice spoofy lemoinem pjb lichtblau bytecolor eno dmiles_afk Tabmow pr lnostdal amaron Adrinael sytse bgs000 Anarch inklesspen mornfall djinni` Fade 14:32:19 -!- names: zbigniew cods slyrus_ andreer metaperl deepfire ianmcorvidae ski _3b herbieB galdor tltstc Helheim kuwabara clog krappie kom_ CrazyEddy ojuice levene rsynnott ecraven scode tessier eldragon Zhivago Ginei_Morioka ve koollman luis Khisanth bobbysmith007 _3b` froydnj PissedNumlock acieroid bfein dcrawford Obfuscate dejones qsun dym Aperculum l_a_m fgtech^ Pepe_ rlpowell rotty tvaalen pok bougyman jrockway @Xof bakkdoor tmitt mal__ 14:33:52 billitch1 [~billitch@men75-12-88-183-197-206.fbx.proxad.net] has joined #lisp 14:33:57 tcr: great! Thanks. 14:35:48 tayloj [~tayloj@indigo-15.dynamic2.rpi.edu] has joined #lisp 14:39:19 aw [~aw@147.81-166-214.customer.lyse.net] has joined #lisp 14:39:39 Davse_Bamse [~davse@82.103.143.161] has joined #lisp 14:40:25 gigamonkey [~user@adsl-99-155-195-250.dsl.pltn13.sbcglobal.net] has joined #lisp 14:40:29 -!- madnificent [~madnifice@83.101.62.132] has quit [Ping timeout: 248 seconds] 14:42:00 -!- HG` [~HG@xdsl-92-252-52-220.dip.osnanet.de] has quit [Quit: Leaving.] 14:43:52 sepult [~user@xdsl-87-79-140-143.netcologne.de] has joined #lisp 14:45:48 -!- sepult [~user@xdsl-87-79-140-143.netcologne.de] has quit [Client Quit] 14:45:56 pjb: What are you ecl using for? 14:50:22 -!- Joreji [~thomas@82-238.eduroam.RWTH-Aachen.DE] has quit [Ping timeout: 240 seconds] 14:50:42 pjb: if you look into swank-ecl.lisp, there's GDB-INITIAL-COMMANDS which will be executed when attaching gdb. What's in there at the moment is what I needed for basic needs. In case you come up with fancy gdb macros, consider sending those upstream 14:50:43 How does one submit a patch to a CL project? 14:50:53 email the author with a diff file and an explanation? 14:51:10 Shaftoe: that's a decent way. 14:51:13 Shaftoe: If the project has a development mailing list use that 14:51:21 alrighty. thanks. 14:51:29 hohoho [~hohoho@ntkngw227224.kngw.nt.ftth.ppp.infoweb.ne.jp] has joined #lisp 14:51:40 Shaftoe: sometimes authors have other requirements, e.g. http://weitz.de/patches.html 14:52:01 fisxoj [~fisxoj@HSI-KBW-095-208-111-226.hsi5.kabel-badenwuerttemberg.de] has joined #lisp 14:52:19 as an aside, mailing lists make me unhappy. it's one of those things =) 14:52:27 some cl projects get really angry if you submit source patches... mostly those whose source you're not supposed to have (-; 14:52:34 For the greater good, I close my eyes and think of the queen! 14:52:59 *Xach* is hoping someone patches cffi before he does 14:54:33 Xach: fwiw, I'm just changing some conversions from uffi datatypes to cl datatypes in clsql-odbc 14:54:37 very minor and specific stuff 14:56:51 gigamonkey: will having PCL on Safari have any impact on your life? 14:58:22 astoon [~astoon@213.141.244.246] has joined #lisp 14:58:22 so, O'Reilly are publishing a lisp book after all (: 15:01:20 manic12 [~andrew@99-100-67-123.lightspeed.sntcca.sbcglobal.net] has joined #lisp 15:03:36 milanj [~milanj_@178.223.151.188] has joined #lisp 15:05:03 t3eblinder [~wolfgang@p5B2013DC.dip0.t-ipconnect.de] has joined #lisp 15:05:27 -!- kpreid [~kpreid@c-67-180-21-133.hsd1.ca.comcast.net] has quit [Quit: kpreid] 15:07:18 Xach: I don't think there is much to get from putting PCL on Safari as it is already available on the net 15:08:44 -!- Guest22022 is now known as pkhuong 15:09:33 carlocci [~nes@93.37.193.236] has joined #lisp 15:10:08 p_l|backup: I'm curious about what "much" means. 15:10:49 -!- jdz [~jdz@193.206.22.97] has quit [Ping timeout: 264 seconds] 15:10:53 -!- gzip4 [~xxx@78.108.73.250] has quit [Ping timeout: 265 seconds] 15:11:12 Xach: well, as far as I know, Safari is subscription-based internet access to various books, right? PCL is already available on the net without going through the hassle 15:12:43 p_l|backup: If you're already familiar with and subscribed to Safari and search within it for "Lisp", it might be easier to find. 15:13:39 Xach: right. But other than that, I see no real use (though there might be some financial gains for gigamonkey - then go for it) 15:13:41 -!- ZabaQ [~john.conn@23.101-84-212.ippool.ndo.com] has quit [Read error: Connection reset by peer] 15:14:05 p_l|backup: that is, I'm wondering about people starting from safari and looking for lisp, not starting from "i want to read practical common lisp" and ending up in safari to do it. 15:14:15 the latter seems pretty unlikely 15:14:21 -!- mcsontos [~mcsontos@nat/redhat/x-oszsrkospfizwldv] has quit [Ping timeout: 245 seconds] 15:14:38 except, maybe, for some printability angle. 15:15:04 Joreji [~thomas@82-238.eduroam.RWTH-Aachen.DE] has joined #lisp 15:18:07 -!- zoldar [~zoldar@ip6.ima.pl] has quit [Quit: Ex-Chat] 15:23:25 -!- hohoho [~hohoho@ntkngw227224.kngw.nt.ftth.ppp.infoweb.ne.jp] has quit [Remote host closed the connection] 15:25:34 -!- martin2 [~user@159.92.64.121] has quit [Remote host closed the connection] 15:29:05 -!- aerique [euqirea@xs2.xs4all.nl] has quit [Quit: ...] 15:31:36 -!- bozhidar [~user@212.50.14.187] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 15:38:44 -!- Harag [~Harag@41.56.62.116] has quit [Ping timeout: 240 seconds] 15:38:45 -!- lvillani [~lvillani@fedora/lvillani] has quit [Ping timeout: 240 seconds] 15:38:57 -!- Shaftoe [~Moe111@bas1-montreal02-1096728247.dsl.bell.ca] has quit [Quit: Shaftoe] 15:39:44 -!- asarch [~asarch@189.188.145.217] has quit [Remote host closed the connection] 15:41:11 tcr: I'm on a project where ecl will be used. We will have to produce a library to be used from other programming languages. 15:43:23 -!- ASau [~user@77.246.231.201] has quit [Quit: off] 15:43:35 Is Allegro able to produce libraries and DLLs? 15:44:21 hohoho [~hohoho@ntkngw227224.kngw.nt.ftth.ppp.infoweb.ne.jp] has joined #lisp 15:44:21 asarch [~asarch@189.188.145.217] has joined #lisp 15:44:42 allegro can generate dlls 15:47:13 -!- stettber` is now known as stettberger 15:47:21 madnificent [~madnifice@83.101.62.132] has joined #lisp 15:49:12 manic12: and libraries on linux? 15:49:29 Lis [~Lis@business-092-079-130-087.static.arcor-ip.net] has joined #lisp 15:52:27 -!- simplechat [~simplecha@unaffiliated/simplechat] has quit [Quit: Leaving] 15:52:28 -!- Joreji [~thomas@82-238.eduroam.RWTH-Aachen.DE] has quit [Ping timeout: 260 seconds] 15:53:22 Here's a puzzler from me to you. Say I have every CL project available to load via ASDF and they all load fine. I'd like to answer the question "What packages and external symbols does loading system introduce?" With the intent of isolating 's stuff, not stuff loaded in support of . 15:53:47 -!- borism [~boris@237.80.191.90.dyn.estpak.ee] has quit [Ping timeout: 252 seconds] 15:55:03 Bonus points if it doesn't involve extracting any metadata from system definitions. 15:55:20 -!- bgs000 is now known as bgs100 15:56:36 -!- abeaumont [~abeaumont@84.76.48.250] has quit [Remote host closed the connection] 15:57:45 Xach: you could hook some methods around load-system (or whichever primitive is actually called there) and compare loaded packages 15:59:13 krzysz00 [~user@pool-71-252-233-89.dllstx.fios.verizon.net] has joined #lisp 16:00:40 -!- rtra [~rtra@unaffiliated/rtra] has quit [Ping timeout: 265 seconds] 16:01:40 I just released version 0.4 of my common lisp roguelike Menace of the Mines (motm), available at http://motm.sourceforge.net/ . The major new features are items and replay mode. (Lisp is so great you can roll a major release in 5 committed days) 16:02:26 -!- pchrist [~spirit@gentoo/developer/pchrist] has quit [Quit: leaving] 16:03:02 pchrist [~spirit@gentoo/developer/pchrist] has joined #lisp 16:03:37 adu [~ajr@pool-71-191-173-125.washdc.fios.verizon.net] has joined #lisp 16:04:06 -!- madnificent [~madnifice@83.101.62.132] has quit [Ping timeout: 240 seconds] 16:05:11 -!- benny [~user@i577A229F.versanet.de] has quit [Ping timeout: 245 seconds] 16:07:02 Xach: I agree with p_l|backup, an around method on system load that does a before-and-after comparison of the package state seems about right. 16:07:05 timor [~timor@port-92-195-75-147.dynamic.qsc.de] has joined #lisp 16:07:31 I'll give that a try, thanks. 16:07:40 rat [~quassel@c-24-22-21-120.hsd1.or.comcast.net] has joined #lisp 16:09:27 benny [~user@i577A12CE.versanet.de] has joined #lisp 16:09:36 -!- nha [~prefect@imamac13.epfl.ch] has quit [Ping timeout: 260 seconds] 16:09:50 -!- tayloj [~tayloj@indigo-15.dynamic2.rpi.edu] has left #lisp 16:10:17 holycow [~new@poco208-2.fredcanhelp.com] has joined #lisp 16:10:37 so I read this piece of lisp code on the net "(define (map proc items) (if (null? items); nil; (cons (proc (car items))" I dont really know lisp is this a definition of map(ie map head ++map tail)? 16:10:49 kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has joined #lisp 16:10:58 hello 16:11:00 -!- abend [~alx@076-076-146-016.pdx.net] has quit [Ping timeout: 252 seconds] 16:11:10 can someone help me with this please ? 16:11:13 http://paste.lisp.org/display/111988 16:11:59 rat: looks like you need to ask #scheme. 16:12:31 sorry 16:13:38 silenius [~silenius@c-24-130-172-194.hsd1.ca.comcast.net] has joined #lisp 16:14:11 kiuma: do you understand how, and more importantly when, the lisp reader works? 16:14:23 not so much 16:14:30 rat: No problem. (: 16:14:31 kiuma: eval-when has an implicit progn... Eval-when at non-toplevel with the :execute keyword -is- progn... 16:14:33 -!- OmniMancer [~OmniMance@219-89-109-120.jetstart.xtra.co.nz] has quit [Quit: Leaving.] 16:14:55 kiuma: well, you want to compile that form, yes? so first, it must be read. 16:15:20 So you're already badly confused. 16:15:33 kiuma: at read time, the string "cl-api:api-gen" will attempt to intern the API-GEN symbol into the package named CL-API 16:15:43 kiuma: can you see the problem? 16:15:47 nyef, progn was dirty code 16:16:44 the reader will attempt to read the string by interning the API-GEN symbol rather. 16:16:46 drewc, yes I see, but I wouldn't like to load the package if not needed and I'wouldn't like to createe another asdf system if possible 16:17:09 I knew it was wrong 16:17:11 kiuma: i would like world peace 16:17:24 but I don't know how to make it correctly 16:17:33 -!- aw [~aw@147.81-166-214.customer.lyse.net] has quit [Ping timeout: 248 seconds] 16:17:41 the correct way is to make an ASDF system. 16:17:59 ah, ok 16:18:10 regardless, the package has to exist before you try reading symbols into it.. non? 16:18:42 you could use the (funcall (find-symbol ....)) trick here, but that's not a good idea. 16:18:50 -!- blandest [~user@85.204.33.242] has quit [Remote host closed the connection] 16:19:22 yea I understand, but isn't to bloated a new asdf for api gen ? 16:22:04 kiuma: do you intend for users to be able to load your API generator? 16:22:33 normally, the way we provide loadable systems to our end-users is via a system definition facility of some sort. 16:23:02 one could use a load.lisp, but as you don't understand how lisp read/compile/load's things, you probably want to use a system written by those who do :) 16:23:45 -!- astoon [~astoon@213.141.244.246] has quit [Ping timeout: 260 seconds] 16:24:20 (or: define bloat?) 16:24:58 drewc, (I'll use cl-api for clonsigna api generation) usually a user will go to the site do get it. 16:25:15 bloat kiuma's dictionary = too much 16:25:25 that makes no sense to me... go to the site to get what? 16:25:37 what site? what does that have to do with loading and compiling lisp code? 16:25:40 read the api 16:26:12 Davidbrcz [~david@212-198-91-47.rev.numericable.fr] has joined #lisp 16:26:19 drewc, nothing (it's mainly for my use) 16:26:41 I'll put clonsigna on common-lisp.net 16:26:48 and cl-api too 16:26:55 none of this has anything to do with my question 16:27:22 mrSpec [~Spec@unaffiliated/mrspec] has joined #lisp 16:27:59 what package does the generate-api symbol live in 16:28:39 how is _that_ code loaded? 16:28:57 jewel [~jewel@196-210-134-12.dynamic.isadsl.co.za] has joined #lisp 16:29:25 now it lives in :clonsigna 16:29:33 the output is http://www.wingstech.it/kiuma/clonsigna/clonsigna.html 16:29:48 (manually generated) 16:29:49 -!- billitch [~billitch@2a01:e35:8b7c:5ce0:1e4b:d6ff:fe5b:afd3] has quit [Ping timeout: 248 seconds] 16:29:56 how is the :clonsigna code loaded? you _must_ be using an ASDF system somewhere, non? 16:30:06 if you're not, you're doing it wrong. 16:30:33 yes consigna uses asdf 16:30:59 but cl-api asdf must stay optional 16:31:52 so, create a (defsystem clonsigna.api ...) form in your .asd, and load it after loading clonsigna... is that bloat? 16:31:56 I'd like something like (asdf:oos 'asdf:document-op :clonsigna) as for test-op 16:32:04 Does someone use putty with ssh? 16:32:08 howdy drewc 16:32:20 how's life? You're on a boat trip? 16:32:20 How can I download a file with putty from the remote server to my local drive? 16:32:26 Lis: you can't 16:32:47 yes... the documentation says its possible with get 16:32:48 Lis: use an scp program.. i believe putty has one on their web site 16:33:21 your wrong... 16:33:26 get is an FTP command, are you sure you're reading the correct part of the documentation? 16:33:29 I try another channel 16:33:33 Lis: the question is off topic. 16:33:55 drewc, I try, but the do I have to (asdf:oos 'asdf:load-op :clonsigna.api) ? 16:34:23 kiuma: no, of course not.. you don't have to do anything. However, if you'd like to use the code, it's a good idea to load it first. 16:34:37 pjb: hello 16:34:38 I'm i just being an elitist or should this be obvious? 16:34:48 -!- christoph_debian [christoph@cl-1281.dus-01.de.sixxs.net] has quit [Ping timeout: 260 seconds] 16:34:53 tcr: I was on vacation for a few weeks, but back now 16:34:56 abend [~alx@67.136.131.11] has joined #lisp 16:35:28 borism [~boris@237.80.191.90.dyn.estpak.ee] has joined #lisp 16:35:29 drewc, ok I'll leave the code commented as a remind. 16:35:31 (was on eastern time for the start of the world cup... thank $diety :)) 16:35:43 just to clear things up: http://the.earth.li/~sgtatham/putty/0.60/puttydoc.txt 6.2.9 The `get' command: fetch a file from the server To download a file from the server and store it on your local PC, you use the `get' command. 16:35:44 -!- rat [~quassel@c-24-22-21-120.hsd1.or.comcast.net] has quit [Read error: Connection reset by peer] 16:36:39 Lis: this is the psftp section 16:37:08 Ah 16:37:13 drewc: I'm in kuala lumpur at the moment 16:37:29 Lis: I suppose if the file is small enough you can use uuencode + copy/paste :p 16:38:54 tcr: cool, the city of dreams! My sister was just there. 16:39:27 Lis: "Chapter 5: Using PSCP to transfer files securely".. now you're off topic and not listening to advice... so shhh :) 16:39:56 hey I didnt say anything you where the last one to say something ;) 16:40:17 Lis: shhh 16:40:31 drewc: what? 16:41:38 -!- borism [~boris@237.80.191.90.dyn.estpak.ee] has quit [Ping timeout: 252 seconds] 16:41:44 Oh sorry drewc did you say something i accidently put you on ignore... 16:42:01 -!- ChanServ has set mode +o drewc 16:42:18 Lis: please try and keep it on topic 16:42:45 drewc, thanks for help, I have to go home now :) 16:43:20 Yes if you let me to... 16:43:26 kiuma: no worries, the whole read/compile/load/execute thing can be confusing if you're used to interpreted languages 16:44:40 -!- mvilleneuve [~mvilleneu@LLagny-156-36-4-214.w80-14.abo.wanadoo.fr] has quit [Quit: Lost terminal] 16:44:52 or compiled languages that don't work on an image 16:45:30 yeah, or anything without INTERN and packages, really. 16:45:53 christoph_debian [christoph@sf-ogame.de] has joined #lisp 16:47:05 -!- kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has quit [Quit: Bye bye ppl] 16:47:17 kejsaren_ [~kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 16:48:20 docl [~luke@216-161-87-126.ptld.qwest.net] has joined #lisp 16:50:10 *drewc* scrolls up in erc attempting to find the c.l.l thread he was following. think more coffee might be needed. 16:50:36 -!- jewel [~jewel@196-210-134-12.dynamic.isadsl.co.za] has quit [Ping timeout: 245 seconds] 16:51:01 Hun [~hun@95-89-68-85-dynip.superkabel.de] has joined #lisp 16:54:14 I *so* forgot about c.l.l best indication that I just have no leisure time anymore :-) 16:55:01 -!- varjag [~eugene@122.62-97-226.bkkb.no] has quit [Read error: Connection reset by peer] 16:55:45 -!- swilde [~wilde@aktaia.intevation.org] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 16:56:47 c.l.l. is like a garbage dump. You might find some treasure, but you'll have spent a ton of time and will still smell like garbage 16:56:51 R3cur51v3 [~Recursive@173-29-151-107.client.mchsi.com] has joined #lisp 16:57:09 force of habit really... i skim through it every morning 16:57:24 hasn't been of any use in years really. 16:59:29 nyef, p_l|backup: ok, what is the most sane gf to which to add the around method? 17:00:26 I try to install hunchentoot on sbcl. I use clbuild to update and install the needed packages. If I try to recompile everything after installing I get this error: http://codepad.org/muTk2dp1 and if I try to install the cl-yacc package manually I get this error: http://codepad.org/L2RRL8JQ does somoene know what causes this problem? Thank you. 17:00:53 astoon [~astoon@213.141.244.246] has joined #lisp 17:01:08 Lis: the system file for cl-yacc is named "yacc" 17:01:15 Beetny [~Beetny@ppp118-208-129-153.lns20.bne1.internode.on.net] has joined #lisp 17:01:26 Thanks 17:01:43 Lis: that doesn't really fix your problem, but I think it might explain the cause a little bit 17:02:31 -!- tfb [~tfb@212.183.140.18] has quit [Quit: tfb] 17:03:47 -!- zophy [~sy@host-242-6-111-24-static.midco.net] has quit [Remote host closed the connection] 17:05:02 -!- timor [~timor@port-92-195-75-147.dynamic.qsc.de] has quit [Ping timeout: 240 seconds] 17:05:31 *Xach* found out all the many varieties of project names, system file names, and system definition names when getting started with quicklisp 17:08:18 Shaftoe [~Moe111@bas1-montreal02-1096728247.dsl.bell.ca] has joined #lisp 17:08:36 Xach: how many projects you have found so far? 17:08:59 p_l|backup: 255 17:09:08 What's the easiest way to test some uffi foreign string stuff? i.e. I need to create a uffi-foreign string so I can use it as a test harness 17:09:53 p_l|backup: not all of them load for me, though. of all the systems defined by those 255 projects, 362 compile and load, 117 fail. 17:10:16 madnificent [~madnifice@83.101.62.132] has joined #lisp 17:10:18 though gwking just fixed lift and log5, so i think i'll be able to move a few of those 117 fails into the win column. 17:10:36 *Xach* also has a number of fails due to a too-old gtk on his distro 17:10:57 dstatyvka [ejabberd@pepelaz.jabber.od.ua] has joined #lisp 17:11:30 ASau [~user@83.69.227.32] has joined #lisp 17:13:32 jdz [~jdz@host38-222-static.48-79-b.business.telecomitalia.it] has joined #lisp 17:14:12 Joreji [~thomas@82-238.eduroam.RWTH-Aachen.DE] has joined #lisp 17:16:10 http://xach.com/tmp/wins.txt has some info 17:16:22 leo2007: Hi! 17:16:30 lolsuper_ [~super_@unaffiliated/lolsuper-/x-9881387] has joined #lisp 17:16:38 argh, stupid locale sorting 17:18:04 -!- Krystof [~csr21@158.223.51.76] has quit [Ping timeout: 276 seconds] 17:18:10 Shaftoe: I recommend switching to CFFI where possible, though :D (seems better supported) 17:18:17 pjb: you seem to write lisp code in an unbelievable speed. How long it took you to feel lisp like mother tongue? 17:18:44 p_l: I'm actually working on CLSQL (which internally uses UFFI) 17:18:53 -!- gko [~gko@122-116-15-138.HINET-IP.hinet.net] has quit [Ping timeout: 240 seconds] 17:19:25 Hm chaning the whole project name from cl-yacc to yacc in clbuild removed the error. 17:21:57 Hmm I'd like to have a :USE-FROM clause in defpackage which basically is exactly like :IMPORT-FORM except it will barf if the symbol named is not actually external 17:21:58 TR2N [email@89.180.165.150] has joined #lisp 17:21:59 bozhidar [~user@93-152-185-88.ddns.onlinedirect.bg] has joined #lisp 17:22:40 I grew the habit to explicitly name symbols I use from libraries in case I use only very few symbols from that library, or the library is a misc bag like alexandria 17:22:49 -!- madnificent [~madnifice@83.101.62.132] has quit [Ping timeout: 264 seconds] 17:23:18 -!- christoph_debian [christoph@sf-ogame.de] has quit [Read error: Operation timed out] 17:23:36 -!- galaxywatcher [~galaxywat@ppp-58-8-50-8.revip2.asianet.co.th] has quit [Quit: galaxywatcher] 17:25:04 Xach: where is lose.txt? :) <---- my hobby, trying to guess where Xach keeps his data hidden 17:25:12 -!- hc_e [~hc@pdpc/supporter/active/hc-e] has quit [Ping timeout: 260 seconds] 17:25:14 sepult [~user@xdsl-87-79-140-143.netcologne.de] has joined #lisp 17:25:35 -!- leo2007 [~leo@cpc1-cmbg13-0-0-cust596.cmbg.cable.ntl.com] has quit [Quit: rcirc on GNU Emacs 23.2.9] 17:25:36 tayloj [~tayloj@indigo-15.dynamic2.rpi.edu] has joined #lisp 17:25:58 leo2007 [~leo@cpc1-cmbg13-0-0-cust596.cmbg.cable.ntl.com] has joined #lisp 17:26:38 christoph_debian [christoph@cl-1281.dus-01.de.sixxs.net] has joined #lisp 17:26:47 -!- sepult [~user@xdsl-87-79-140-143.netcologne.de] has quit [Client Quit] 17:26:50 hc_e [~hc@pdpc/supporter/active/hc-e] has joined #lisp 17:26:53 -!- Joreji [~thomas@82-238.eduroam.RWTH-Aachen.DE] has quit [Ping timeout: 240 seconds] 17:27:19 tcr: yeah, i've thought of that need as well, only i don't want to import the symbol at all, just make sure it exists and is external to that package i assume it is. 17:27:33 -!- zomgbie [~jesus@chello062178135013.2.14.vie.surfer.at] has quit [Ping timeout: 260 seconds] 17:28:14 oh right import will actually change the symbol's status? 17:28:44 (though i guess a toplevel form like '(:use-external-symbols foo:bar foo:baz) somewhere after the defpackage would 'work') 17:29:18 timor [~timor@port-92-195-75-147.dynamic.qsc.de] has joined #lisp 17:29:32 Writing a wrapper around defpackage is trivial, I just did that yesterday to make it include :depends-on 17:29:45 but I settled with a toplevel (depends-on ) form 17:30:00 yeah, i think that's the better way really 17:30:34 -!- kejsaren_ [~kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Ping timeout: 276 seconds] 17:30:51 Iirc the clhs entry for defpackage explicitly allows implementators to add new clauses to defpackage 17:32:35 -!- revel0 [~revel0@h15a2.n2.ips.mtn.co.ug] has quit [Quit: ~ Trillian Astra - www.trillian.im ~] 17:33:20 chiiph [~quassel@gentoo/developer/chiiph] has joined #lisp 17:33:45 -!- tritchey [~tritchey@205.233.9.181] has quit [Quit: tritchey] 17:34:07 -!- R3cur51v3 [~Recursive@173-29-151-107.client.mchsi.com] has quit [Read error: Operation timed out] 17:34:21 davazp [~user@83.55.180.105] has joined #lisp 17:35:32 kejsaren_ [~kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 17:35:37 -!- docl [~luke@216-161-87-126.ptld.qwest.net] has left #lisp 17:36:35 -!- relcomp [~chatzilla@pc92-e03-3.igf.Uni-Osnabrueck.DE] has quit [Quit: ChatZilla 0.9.85 [Iceape 2.0.5/20100623161114]] 17:38:43 tritchey [~tritchey@205.233.9.181] has joined #lisp 17:43:09 sepult [~user@xdsl-87-79-140-143.netcologne.de] has joined #lisp 17:44:59 Zephyrus [~emanuele@unaffiliated/zephyrus] has joined #lisp 17:45:51 Harag [~Harag@iburst-41-213-32-218.iburst.co.za] has joined #lisp 17:46:46 stassats [~stassats@wikipedia/stassats] has joined #lisp 17:48:34 drewc: fail.txt is actually a bunch of logfiles ending in backtraces 17:52:41 e.g. http://xach.com/tmp/elephant 17:52:54 -!- Lis [~Lis@business-092-079-130-087.static.arcor-ip.net] has quit [Quit: Nettalk6 - www.ntalk.de] 17:54:10 -!- attila_lendvai [~attila_le@4d6f5d3b.adsl.enternet.hu] has quit [Quit: Leaving.] 17:54:37 -!- Shaftoe [~Moe111@bas1-montreal02-1096728247.dsl.bell.ca] has quit [Quit: Shaftoe] 17:55:19 aw [~aw@147.81-166-214.customer.lyse.net] has joined #lisp 17:55:28 -!- Hun [~hun@95-89-68-85-dynip.superkabel.de] has quit [Ping timeout: 260 seconds] 17:55:48 Xach: a few projects need to be configured before use 17:56:29 fe[nl]ix: yes, a few. 17:56:44 -!- Kolyan [~nartamono@89-178-240-16.broadband.corbina.ru] has quit [] 17:59:03 zomgbie [~jesus@mk092248003031.a1.net] has joined #lisp 18:02:32 -!- Snamich [~Snamich@166.189.147.49] has quit [Quit: Snamich] 18:03:03 damn, some craziness in lisppaste... 18:03:22 http://paste.lisp.org/+2EF6 <--- any opinions on this (method dispatch optimization) 18:05:17 single dispatch? 18:06:31 p_l|backup: you'll have to specify how much staticness you introduced in your "CLOS" first. 18:07:34 -!- zomgbie [~jesus@mk092248003031.a1.net] has quit [Ping timeout: 264 seconds] 18:09:09 frito [~keithmant@cpc2-sout4-0-0-cust13.sotn.cable.ntl.com] has joined #lisp 18:09:12 nice, i'm tying this on an old-style symbolics keyboard :) 18:09:25 typing.. 18:09:56 -!- holycow [~new@poco208-2.fredcanhelp.com] has quit [Remote host closed the connection] 18:10:02 -!- frito [~keithmant@cpc2-sout4-0-0-cust13.sotn.cable.ntl.com] has quit [Client Quit] 18:12:07 Soulman [~knute@79.80-202-237.nextgentel.com] has joined #lisp 18:13:41 Snamich [~Snamich@166.189.147.49] has joined #lisp 18:16:01 pkhuong: not much - if the data doesn't fit, there would be extra code to recompute the reference in literal vector 18:16:22 p_l|backup: "the class". 18:16:41 stassats: just an example - in this case it was single dispatch, but I was looking to at most two argument specialized 18:17:18 -!- davazp [~user@83.55.180.105] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 18:18:02 -!- bandu [~coyotama@coyotama-1-pt.tunnel.tserv8.dal1.ipv6.he.net] has quit [Ping timeout: 240 seconds] 18:18:33 pkhuong: "class" in this case came from the original system, what I meant was "object this argument was specialized on", though I suspect stuff that isn't based on standard-class (or rather, isn't a classoid) might be impossible to inline this way 18:19:35 I was simply thinking of subsequent defclass here. 18:21:46 -!- aw [~aw@147.81-166-214.customer.lyse.net] has quit [Read error: Connection reset by peer] 18:22:10 aw [~aw@147.81-166-214.customer.lyse.net] has joined #lisp 18:22:11 pkhuong: personally I was thinking of a) placing subsequent defclass-created classes at different addresses than original, thus forcing recomputation or b) registering each optimized function & running fixups 18:22:34 it's directed more towards long-running, less changing code 18:23:16 while trying to avoid breaking CLOS dispatch 18:23:32 -!- Edward_ [edward@AAubervilliers-154-1-41-231.w90-3.abo.wanadoo.fr] has quit [Ping timeout: 260 seconds] 18:23:36 roygbiv [~JohnRambo@pdpc/supporter/active/roygbiv] has joined #lisp 18:23:43 (it *is* based on static, single-dispatch system, though) 18:24:20 There are lower-hanging fruits... annotation coming. 18:26:02 pkhuong: it was a spur-of-the-moment thing - I just read an example of how it is done in two different Java implementations (Sun's JVM, Azul's Java-specific cpu) 18:28:24 madnificent [~madnifice@83.101.62.132] has joined #lisp 18:29:24 p_l|backup: does it make more sense to compile it normally first, and then optionally re-compile with static dispatch once a class has been sealed? 18:31:21 drewc: well, this code assumes no sealing available - adding sealed class support would be trivial 18:31:35 s/code/idea/ 18:31:39 zomgbie [~jesus@h081217133138.dyn.cm.kabsi.at] has joined #lisp 18:31:41 can't call it code :) 18:32:51 daniel__ [~daniel@p5082F4C1.dip.t-dialin.net] has joined #lisp 18:33:44 p_l|backup: right, but what i'm saying is that sealing is a well known way to implement the optimisations you seem to want, so you might want to investigate that. 18:34:09 drewc: I am going to investigate sealing as well 18:34:15 p_l|backup: otherwise, i can't see the advantage over, say, just using defun over defgeneric 18:34:21 drewc: it's more like a step toward being able to implement the above. 18:34:44 pkhuong: yeah, fair enough :) 18:35:01 p_l|backup: annotated 18:36:06 pkhuong: thanks 18:36:37 -!- daniel_ [~daniel@p5082F3E3.dip.t-dialin.net] has quit [Ping timeout: 264 seconds] 18:36:51 The whole thing happened rather randomly when I encountered a blog that described how Java manages to avoid constant lookups for all method calls (as all are "virtual") 18:37:15 "constant" as in "incessant"? 18:37:52 does java have dynamic typing, generic functions, and run-time class/gf/method redefinition? 18:38:17 drewc: everything except multiple dispatch. 18:38:19 drewc: nope, but I just seen that code and started to think if similar thing can be applied to CLOS 18:38:32 (nope as in "dunno" - I don't care too much about Java) 18:39:32 abugosh [~Adium@207-172-111-88.c3-0.tlg-ubr4.atw-tlg.pa.cable.rcn.com] has joined #lisp 18:39:41 HotSpot modifies the cache at runtime, at least 18:40:59 Joreji [~thomas@82-238.eduroam.RWTH-Aachen.DE] has joined #lisp 18:41:09 -!- Davse_Bamse [~davse@82.103.143.161] has quit [Quit: leaving] 18:41:14 astalla [~astalla@93-36-230-5.ip62.fastwebnet.it] has joined #lisp 18:41:17 well, shows what i know about java, which is nil. I do know a bit about CLOS though, which is probably where i'm getting confused... this is not suppose to be a full CLOS, is it? 18:41:25 -!- zomgbie [~jesus@h081217133138.dyn.cm.kabsi.at] has quit [Ping timeout: 264 seconds] 18:41:41 -!- madnificent [~madnifice@83.101.62.132] has quit [Ping timeout: 240 seconds] 18:42:07 drewc: full clos, just trying to get the most probable case in dispatch handled fast 18:42:21 brb 18:46:10 -!- thunk is now known as thunk|smoking 18:47:10 is it stupid to ask why the _class_ would be sealed, in CLOS, as opposed to the generic function? 18:47:41 or did I misinterpret the convo? 18:48:18 Good evening! 18:48:22 beach: 'lo 18:48:25 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 260 seconds] 18:48:27 zomgbie [~jesus@h081217133138.dyn.cm.kabsi.at] has joined #lisp 18:48:34 *beach* doesn't see the point of sealing. 18:48:56 holycow [~new@poco208-2.fredcanhelp.com] has joined #lisp 18:49:05 sykopomp: no, not stupid at all really.. it really amounts to the same thing... but you have to seal classes before you can seal things that dispatch on them... non? 18:49:19 sykopomp: you need both for a lot of interesting stuff. 18:49:26 drewc: not necessarily. 18:49:43 drewc: Well, I was thinking that if you 'seal' a genfun, the class won't be garbage collected. 18:49:58 and the genfun is where all the juicy stuff happens, anyway. 18:50:19 beach: i don't either... any time method dispatch has been a bottleneck, i reach for DEFUN. This has only ever happened once though, so what do i know :). 18:50:29 pkhuong: what can you add by sealing the class itself (as far as _method_ dispatch goes, not stuff like slot access) 18:51:01 -!- leo2007 [~leo@cpc1-cmbg13-0-0-cust596.cmbg.cable.ntl.com] has quit [Ping timeout: 245 seconds] 18:51:20 drewc: Sure, but that's not quite what I meant. I meant, in Lisp, since we always have total knowledge of the entire program, why is it that not everything is sealed, until it is no longer? 18:52:12 beach: Probably bookkeeping overhead and cache invalidation cost. 18:52:35 nyef: I am not convinced, but you might be right. 18:52:42 beach: i've often asked wondered that myself. I assume that the performance issues associated with doing it like that outweigh the benifits, but i am not a compiler writer :) 18:53:10 xan_ [~xan@213.47.69.75] has joined #lisp 18:53:10 drewc: I am not convinced, but you might be right. :) 18:53:21 beach: i'm not convinced either, FWIW :) 18:53:29 sykopomp: I was replying to drewc. 18:53:37 "You just redefined the macro FOO, we're now going to recompile every function that used it." 18:53:43 pkhuong: oh, sorry 18:53:59 nyef: that would be damn nice! 18:54:09 sykopomp: unless it's not what you want. 18:54:25 sykopomp: Expensive, especially since it invalidates the concept of minimal compilation, and thus fasls. 18:54:39 beach: also, hard to get right with threads. 18:54:54 pkhuong: you're right actually, i can see a lot of benifit to sealing a GF by itself 18:55:42 how it's better than DEFUN + cond/case/typecase/whatever remains to be seen. 18:56:02 drewc: you don't have to do it yourself, and it maintains modularity in the source code. 18:56:05 In case anyone is wondering, I am not going to argue in favor of my suggestion; I am just listening to arguments for or against. Then I'll think about it. 18:56:38 sepult` [~user@xdsl-87-79-174-21.netcologne.de] has joined #lisp 18:57:05 -!- thunk|smoking is now known as thunk 18:57:14 You could write a pseudo-defgeneric/defmethod pair of macros that accumulate source at compile-time and dumps everything when you're done defining methods, I suppose. 18:57:23 pkhuong: yeah, i'll give it that... but 18:57:26 beach: have you compiled a haskell program before? :D 18:57:28 well what you just said actually 18:57:36 (especially a bigger chunk of code) 18:57:52 p_l|backup: takes forever?... 18:57:53 p_l|backup: Not very many. 18:58:02 it shows how expensive "total knowledge of program" optimization can be if you want to still play with changes in code 18:58:10 -!- mrSpec [~Spec@unaffiliated/mrspec] has quit [Ping timeout: 276 seconds] 18:58:11 Krystof [~csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has joined #lisp 18:58:35 p_l|backup: Haskell has very different semantics though. 18:58:37 -!- bgs100 is now known as bgs000 18:58:53 -!- sepult [~user@xdsl-87-79-140-143.netcologne.de] has quit [Ping timeout: 240 seconds] 18:59:52 sealing should be a pre-deployment optimization; say, a :seal t option to save-image. 19:00:10 astalla: And why is that? 19:00:11 doesn't make much sense to have it during development 19:00:24 -!- stis [~stis@1-1-1-39a.veo.vs.bostream.se] has quit [Ping timeout: 252 seconds] 19:00:39 astalla: And I was just arguing that it doesn't make sense at all. 19:00:41 astalla: well, it's kinda what I was investigating when I wrote that little idea on inline caching 19:01:16 beach: oh, I got a different impression, sorry 19:01:40 p_l|backup: interesting, I must have missed it 19:02:03 kejsaren1 [~kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 19:02:24 -!- lonstein_ is now known as rlonstein 19:02:42 LiamH: why are there those symlinks in gsd? 19:03:15 Xach: because clbuild wants asds at top level, otherwise it ignores them (see e.g. static-vectors) 19:04:12 interesting. 19:04:49 -!- kejsaren [~kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Ping timeout: 276 seconds] 19:05:19 -!- fisxoj [~fisxoj@HSI-KBW-095-208-111-226.hsi5.kabel-badenwuerttemberg.de] has quit [Ping timeout: 276 seconds] 19:08:25 -!- Joreji [~thomas@82-238.eduroam.RWTH-Aachen.DE] has quit [Ping timeout: 264 seconds] 19:10:21 mrSpec [~Spec@unaffiliated/mrspec] has joined #lisp 19:13:15 -!- sepult` [~user@xdsl-87-79-174-21.netcologne.de] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 19:14:24 Joreji [~thomas@82-238.eduroam.RWTH-Aachen.DE] has joined #lisp 19:15:06 stis [~stis@1-1-1-39a.veo.vs.bostream.se] has joined #lisp 19:17:44 borism [~boris@237.80.191.90.dyn.estpak.ee] has joined #lisp 19:24:15 beach: I'm with you. Though I'm also with nyef in that I'm not sure you could really do what you're talking about in CL, as specified. 19:25:02 Even laying aside the whole macro-redefinition or type-redeclaration thing, limiting it only to class hierarchies, that leaves... 19:25:57 Well, that's actually a lot easier, since it can more easily be arranged for the invalidated sections to be both reasonably small and well-specified. 19:26:13 chee [~chee@unaffiliated/chee] has joined #lisp 19:27:00 -!- timor [~timor@port-92-195-75-147.dynamic.qsc.de] has quit [Ping timeout: 260 seconds] 19:28:01 -!- kleppari [~spa@bitbucket.is] has quit [Quit: leaving] 19:29:26 -!- bgs000 is now known as bgs100 19:30:09 gravicappa [~gravicapp@ppp85-140-64-235.pppoe.mtu-net.ru] has joined #lisp 19:31:51 -!- jdz [~jdz@host38-222-static.48-79-b.business.telecomitalia.it] has quit [Quit: Leaving] 19:33:07 sepult [~user@xdsl-87-79-174-21.netcologne.de] has joined #lisp 19:33:08 frito [~keithmant@cpc2-sout4-0-0-cust13.sotn.cable.ntl.com] has joined #lisp 19:33:38 -!- frito [~keithmant@cpc2-sout4-0-0-cust13.sotn.cable.ntl.com] has quit [Client Quit] 19:40:28 francogrex [~user@109.130.156.131] has joined #lisp 19:41:57 well good to see that some people are not glued in front of their TVs watching soccer! anyway, what is the easiest way to programatically extract the number 122 buried within this: "tgf12rf2#" ? 19:42:22 (parse-integer (remove-if-not #'digit-char-p "tgf12rf2#")) 19:42:41 lovely 19:42:53 timor [~timor@port-92-195-27-237.dynamic.qsc.de] has joined #lisp 19:44:56 *drewc* is coding and watching the football 19:45:27 -!- zomgbie [~jesus@h081217133138.dyn.cm.kabsi.at] has quit [Ping timeout: 240 seconds] 19:47:19 zomgbie [~jesus@h081217133138.dyn.cm.kabsi.at] has joined #lisp 19:47:49 drewc: BZZZ 19:49:43 madnificent [~madnifice@83.101.62.132] has joined #lisp 19:49:56 heh.... every bit of white noise i hear since the start of the tourney now sounds like the vuvuzela 19:50:06 engine + tire noise while driving is the big one 19:50:55 stassats [~stassats@wikipedia/stassats] has joined #lisp 19:51:47 *p_l|backup* thinks a visit to a shrink is necessary cause he starts to spew startup ideas 19:52:47 Did anyone get hierarchical packages http://tfeb.org/lisp/hax.html#HIERARCHICAL-PACKAGES work on SBCL> 19:52:50 *? 19:52:59 mstevens [~mstevens@fsf/member/pdpc.active.mstevens] has joined #lisp 19:54:03 jleija [~jleija@user-24-214-122-46.knology.net] has joined #lisp 19:54:12 dysinger [~dysinger@173-13-81-254-NewEngland.hfc.comcastbusiness.net] has joined #lisp 19:54:13 -!- dysinger [~dysinger@173-13-81-254-NewEngland.hfc.comcastbusiness.net] has quit [Excess Flood] 19:55:13 dysinger [~dysinger@173-13-81-254-NewEngland.hfc.comcastbusiness.net] has joined #lisp 19:56:29 (as for startup, I thought of making Access/FoxPro thingy in CL, with CL as programming language exported) 19:58:07 p_l|backup: That's not a startup idea... no money in writing programming tools. think about things one might write using this access/foxpro system, in there there is a startup idea. 19:58:37 drewc: I wasn't planning on selling just the programming tools 19:59:42 I was thinking of hosted solution 20:00:09 who is your market... lispers? or are you just doing another dabbledb? 20:00:47 -!- madnificent [~madnifice@83.101.62.132] has quit [Ping timeout: 240 seconds] 20:01:21 *drewc* has done a good many startups and would like to save others some pain :) 20:01:33 drewc: people who otherwise haven't done much programming, and aren't looking for it as a constant job (so a simplified language available for making small, simple apps) 20:02:16 "with CL as programming language exported" <---- that's going to attract casual users? 20:02:41 drewc: that was a flop in my thinking - Iwas going for making it available under the hood, if someone wanted to 20:03:08 *Xach* wonders whatever happened to Dave Bakhash's cl product 20:03:15 *Xach* spelled that name wrong, probably 20:03:17 so, what's the difference between your idea and the 100's of similar solutions? 20:03:50 p_l|backup: because i simply don't see a large market there, and the existing solutions are quite complete as it is 20:04:07 drewc: let him try, he allways can opensource it 20:04:13 amaron: haha 20:04:19 haha 20:04:23 :D 20:04:30 good one :) 20:04:46 -!- dlowe [~dlowe@ita4fw1.itasoftware.com] has quit [Quit: Leaving.] 20:04:49 ysph [~user@24-181-93-165.dhcp.leds.al.charter.com] has joined #lisp 20:05:09 *p_l|backup* wasn't planning on going into it without research though, just got a little thinking about availability of "simple" programming outside of database tools 20:05:22 there is no such thing as simple programming 20:05:43 drewc: how about "simple" programming? 20:05:54 drewc: I did quote it :P 20:06:10 then it's so worthless a term as to be useless, non? :P 20:06:18 it's simple "programming" 20:06:51 anyway I support it, go for it p_l|backup :) 20:07:45 *p_l|backup* was thinking more towards "simple app" thing rather than "database tool", just Access and similar were an example 20:08:45 funnily enough, I think Shoes got a good idea there :) 20:08:51 there is no much inovative things lately, so if you are original, you can actually make something useful 20:09:18 regarding database oriented apps 20:09:58 -!- gravicappa [~gravicapp@ppp85-140-64-235.pppoe.mtu-net.ru] has quit [Ping timeout: 264 seconds] 20:10:42 is there a way to check if my SBCL supports 'native callbacks'? 20:11:02 you could say that I was thinking of heroku.com for that guy in small company that wants to automate few things, not learn how to program :) 20:11:19 p_l|backup: i'm for it too... i make a good portion of my living turning "simple" database apps written by non-programmers into proper databases once they realize they are over their heads :) 20:11:53 so the more of those you create, the bigger the market for consultants! :D 20:12:09 sykopomp: (find :alien-callbacks *features*) 20:12:10 heh 20:12:24 badipod [~badipod@unaffiliated/baddog144] has joined #lisp 20:12:26 T1750 [~T1750@unaffiliated/t1750] has joined #lisp 20:12:30 drewc: well, for consultants there would be the full power of CL under the hood :P 20:12:39 hey guys i was in here earlier about the waitpid error when trying to compile 20:12:47 did anyone have any ideas whats happening? 20:13:57 T1750: waitpid isn't linked in the runtime. 20:14:11 pkhuong: can you explain more? 20:14:21 as in why isn't it 20:14:31 p_l|backup: thanks :) 20:14:36 or how to fix it is even better :) 20:15:00 add "waitpid" to the list in tools-for-build/ldso-stubs.lisp 20:15:08 <_3b> T1750: did you see https://bugs.launchpad.net/sbcl/+bug/492783 ? 20:15:19 no i'm a n00b 20:16:43 where do i get waitpid 20:16:49 -!- [mzm] [~ziggy@86.99.31.212] has quit [Quit: Leaving] 20:16:54 you type it? 20:17:02 lol 20:17:08 [mzm] [~ziggy@86.99.31.212] has joined #lisp 20:18:31 ziarkaen [~ziarkaen@87.113.18.37.plusnet.pte-ag1.dyn.plus.net] has joined #lisp 20:18:34 just below "wait3" in stubs? 20:19:07 *T1750* doesn't know where to put it 20:19:35 T1750: the bug report is clear enough, did you read it? 20:19:53 no 20:19:57 i scanned it 20:20:06 seems it was the right place too 20:20:08 thank you :) 20:21:40 amaron: anyway, the idea was to make simple (non-fancy) UI, printing, db and few other bits so that quick weekend job apps can be easily done by untrained - if they want it professional, deliver them to someone willing to do it :) 20:21:50 drewc: why does it say between wait3 and write? p comes after e in the alphabet 20:21:59 i can add it after write and it's fine yeah? 20:22:18 no im dumb 20:22:19 n/m 20:23:03 T1750: http://en.wikipedia.org/wiki/Rubber_duck_debugging <--- learn it and use it! 20:24:01 ocne the package builds then i will learn it :) 20:24:14 i have some understanding of list, just a little, i used emacs for a while 20:24:28 *T1750* thought learning a real lisp could be useful 20:25:45 jgracin [~jgracin@vipnet486.mobile.carnet.hr] has joined #lisp 20:26:19 Edward_ [~edward@AAubervilliers-154-1-13-78.w86-212.abo.wanadoo.fr] has joined #lisp 20:26:26 they were the better team for sure, first time i've seen them look in form. 20:26:31 sorry! wrong channel that 20:27:24 attila_lendvai [~attila_le@89.135.200.140] has joined #lisp 20:28:37 leo2007 [~leo@cpc1-cmbg13-0-0-cust596.cmbg.cable.ntl.com] has joined #lisp 20:29:49 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 248 seconds] 20:30:53 -!- zomgbie [~jesus@h081217133138.dyn.cm.kabsi.at] has quit [Ping timeout: 240 seconds] 20:31:38 -!- ziarkaen [~ziarkaen@87.113.18.37.plusnet.pte-ag1.dyn.plus.net] has quit [Remote host closed the connection] 20:33:09 -!- krzysz00 [~user@pool-71-252-233-89.dllstx.fios.verizon.net] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 20:34:49 -!- mrSpec [~Spec@unaffiliated/mrspec] has quit [Ping timeout: 264 seconds] 20:36:19 p_l|backup: this can help you make instant apps if you have proper clients 20:36:47 p_l|backup: what UI you had in mind? 20:38:04 I mean, what library/environment you would use? 20:38:36 -!- Zephyrus [~emanuele@unaffiliated/zephyrus] has quit [Read error: Operation timed out] 20:40:31 -!- splittist [~joe@30-245.62-188.cust.bluewin.ch] has quit [] 20:41:00 -!- fiveop [~fiveop@g229165006.adsl.alicedsl.de] has quit [Quit: humhum] 20:43:40 amaron: depends on actual real-world usability, probably something in JS for web browsers :) 20:44:13 -!- ikki [~ikki@201.155.75.146] has quit [Ping timeout: 248 seconds] 20:44:19 (GTK, Qt and others have their uses, but I'm not sure if desktop deployment would pay for itself) 20:44:27 -!- CrazyEddy [~CrazyEddy@wrongplanet/CrazyEddy] has quit [Ping timeout: 240 seconds] 20:44:30 mrSpec [~Spec@unaffiliated/mrspec] has joined #lisp 20:47:07 -!- francogrex [~user@109.130.156.131] has quit [Remote host closed the connection] 20:47:17 p_l|backup: a cairo-based layout with wysiwyg using x11/win32 for display and ps/pdf backend for printing might be nice to have, especially if you'll provide a form editor of sorts 20:47:32 p_l|backup: if it's web based, make it fancy: html5/css3 20:47:34 I have a similar pipe dream project but other than thinking about it I never could start it heh 20:48:25 otherwise the equivalent might be svg+js perhaps for similar flexibility 20:50:09 phadthai: I was actually thinking of doing the GUI part in .NET (through interop) 20:51:18 at least for windows desktop 20:51:46 web-based would be HTML5/CSS3 (where possible), probably with ExtJS or similar 20:52:09 mon_key [~user@unaffiliated/monkey/x-267253] has joined #lisp 20:53:17 -!- ace4016 [~jmarcelin@adsl-10-21-101.mia.bellsouth.net] has quit [Ping timeout: 248 seconds] 20:53:47 p_l|backup: ccl then? 20:54:24 p_l|backup: abcl + swing? :) 20:54:38 on windows, CCL probably 20:54:48 -!- Beetny [~Beetny@ppp118-208-129-153.lns20.bne1.internode.on.net] has quit [Ping timeout: 252 seconds] 20:54:51 astalla: I don't have much faith in JVM installation :P 20:55:03 -!- mstevens [~mstevens@fsf/member/pdpc.active.mstevens] has quit [Quit: mstevens] 20:55:38 rat [~quassel@c-24-22-21-120.hsd1.or.comcast.net] has joined #lisp 20:56:41 for gui I like gtk most, but qt is ok too 20:56:53 p_l|backup: it's probably not much different than any Lisp installation, but I have not experience there 20:56:54 and it's portable 20:58:03 ziggy_ [~ziggy@de2-as23404.alshamil.net.ae] has joined #lisp 20:58:59 amaron: the issue in my case is Java, not ABCL 20:59:17 and Java is as portable as the program written in it :) 20:59:20 ace4016 [~jmarcelin@adsl-10-101-36.mia.bellsouth.net] has joined #lisp 21:00:42 -!- ASau is now known as ASau` 21:01:03 yes, I mean, installing the JVM is like installing, say, SBCL 21:01:20 ASau [~user@83.69.227.32] has joined #lisp 21:01:27 astalla: Except I find distribution of a dumped lisp image easier :) 21:01:37 though it is usable path 21:01:47 -!- [mzm] [~ziggy@86.99.31.212] has quit [Ping timeout: 265 seconds] 21:02:00 executable image you mean? 21:02:41 that would be like the JVM + your app + a shell script to start it all, but Lisp probably wins in this case :) 21:02:57 I mean, non-JVM lisps ;) 21:04:12 strange thing, isn't it? people always complain that Lisp is hard to deploy, yet Java is not that better 21:05:14 I don't find lisp all that hard to deploy, except that the binaries are like 50mb :| 21:05:44 really ? 21:05:47 that big ? 21:05:50 sykopomp: by lisp you mean SBCL? 21:05:51 well, no. It's bigger. 21:05:55 that small? 21:06:01 drewc: SBCL and CCL both dump things around that size. 21:06:11 although they compress down to ~10mb 21:06:18 sykopomp: the set of common lisp implementations is significantly larger than 2, non? 21:06:19 but that's still very large. 21:06:38 Stattrav [~Stattrav@117.192.130.49] has joined #lisp 21:06:38 madnificent [~madnifice@83.101.62.132] has joined #lisp 21:06:41 compared to what? 21:06:43 -!- dstatyvka [ejabberd@pepelaz.jabber.od.ua] has left #lisp 21:06:51 drewc: the set of common lisp implementations that are free and perform acceptably is not very big. 21:07:13 drewc: compared to the 'standalone' binaries other languages can compile. 21:07:26 ccl's double-clickable osx execs aren't wildly larger than other osx execs 21:07:47 sykopomp: so, the same old bullshit then... when you deploy to a system that already has your runtime installed, you don't have to include a runtime. 21:07:53 but that may be because every osx exec seems to include a copy of every library it uses. 21:07:59 exactly 21:08:10 drewc: I am aware that this is mostly irrelevant for server-side stuff, which is probably what you concern yourself most with. 21:08:13 sb-heapdump was able to produce such small binaries if you installed the right runtime beforehand 21:08:21 how big would be a Java app if it had to include the JVM? 21:08:29 or a C app if it included glibc? 21:08:31 -!- badipod [~badipod@unaffiliated/baddog144] has quit [Quit: Colloquy for iPod touch - http://colloquy.mobi] 21:08:31 on the other hand, I think it's unreasonable to expect people to install a lisp compiler on their platform to download a widely-distributed application. 21:08:32 and C#/.net, python, ... 21:09:03 drewc: they don't have to be that big. Really. 21:09:21 well, a treeshaker would be interesting. 21:09:23 sykopomp: if you say so. 21:09:26 the JVM is > 100MB. 21:09:27 -!- tayloj [~tayloj@indigo-15.dynamic2.rpi.edu] has quit [Quit: Leaving.] 21:09:34 the pain of requiring an installation is somewhat eased for java because of how widespread the JVM is. 21:09:38 somebody, somewhere/when actually did some exploratory coding into it, as I recall. 21:09:49 sykopomp: is that from experience? because mine has been the opposite. 21:10:06 drewc: no, but I don't think Java does very well in this case, either. 21:10:13 it's far more widespread than Lisp, but not ubiquitous at all 21:10:21 on windows it doesn't come preinstalled 21:10:23 but I can see how the situation may not be _as_ bad when it comes to expecting a client-side JVM. 21:10:36 on OSX it can't be updated, afaik 21:10:49 so, it's not a very good situation either 21:10:54 distributing 50mb command-line tools is absurd. :\ 21:11:06 actually java in my experience is a PITA when it comes to getting the runtime available 21:11:06 less absurd all the time. 21:11:07 why do you need SBCL for command line tools? 21:11:28 ECL is a great lisp if you want to play unix, and gnu clisp is also excellent 21:11:28 I'd concurr with that, p_l|backup 21:11:33 #!/usr/bin/sbcl --script ? 21:11:39 drewc: why do I have to use slow lisps that don't even play nicely with SLIME? 21:11:40 True, ECL gives tiny binaries. 21:11:52 sykopomp: you attach slime to your command line tools? 21:12:03 drewc: no, but I develop with SLIME. 21:12:05 i'm suddenly confused. 21:12:09 sykopomp: Not sure what you're doing but I developed a dumb script with SLIME+ECL. 21:12:14 *Xach* uses sbcl for command-line tools all the time, via buildapp -- it's been great for getting a predictable image loaded for analyzing systems and their relationships 21:12:44 sykopomp: so... you have to develop and deploy on the same lisp, and you want that lisp to be both fast, and small, and still include all the features that make it large... 21:12:53 ikki [~ikki@201.155.75.146] has joined #lisp 21:13:03 that's kinda hard :D 21:13:06 drewc: Too much to ask? :) 21:13:17 drewc: I don't always need a compiler at runtime :) 21:13:32 Nor do I need a vast number of features SBCL cooks into your images. 21:13:32 sykopomp: i know that.. do you? 21:13:48 drewc: I'd like to keep the precompiled code, tyvm. 21:13:51 you do if you want to deploy under sbcl 21:13:58 indeed 21:14:08 I do, and that's not something I like. 21:14:18 so don't deploy command line tools with SBCL if you don't want SBCL 21:14:44 right. I deploy command line tools in SlowLisp, because SBCL happens to not support features that would make a final binary significantly smaller. 21:14:47 there aren't any lisp tools to export standalone binaries? 21:14:49 -!- Stattrav [~Stattrav@117.192.130.49] has quit [Quit: Well, the machine might have gone down. Brb after a reboot.] 21:14:52 bougyman: many 21:15:04 drewc: any of them compile to a native binary? 21:15:08 bougyman: 'standalone' meaning 'include everything and the kitchen sink in the final binary' 21:15:09 bougyman: most 21:15:10 bougyman: yes 21:15:32 sykopomp: well that's the only way to accomplish it, i'm guessing. 21:15:40 sykopomp: so, you've tried ECL for your deployed command line apps and found it too slow to be acceptible, or is this more speculation? 21:15:45 have you looked at how haxe does their output of binaries? 21:16:10 nyef mentioned building "crippled" SBCL images 21:16:18 *drewc* has ECL apps deployed and found them quite acceptable. 21:16:22 I think a tree shaker would be 'nice', but image size hasn't been a huge hassle for me. 21:16:26 drewc: specifically, I'd like to ship games. Back when I was working on SquirL with a few others, even CCL couldn't really handle that kind of number crunching. 21:16:37 running even a simple demo on clisp was a bloody joke. 21:16:44 "even CCL"? 21:16:49 I mean... even for client code, just about everybody I know or deal with has at least a megabit class broadband connectio these days. 21:16:49 if you're shipping a game i'd think you'd also want to ship SBCL with it. 21:17:03 Yeah, building a "crippled" SBCL core is possible in theory... But not so useful in practice, yet. 21:17:04 bougyman: not necessarily. 21:17:06 just like complex java apps will include their own JVM. 21:17:06 Fade: manual lists of "droppable" components would work, I guess 21:17:33 almost every commercial java app i've used shipped its own jvm, as a matter of fact. 21:17:42 Fade: I believe I can be a poster child of how bad modern links can be. Especially now that cellular data is growing area 21:17:43 I haven't bought a computer with less than two gigabytes of memory since 2005 or something. 21:17:52 drewc: it turns out that yes, I would like to be able to ship binaries that are hopefully less than 10mb, without losing the performance of the code SBCL already compiled. 21:18:04 I don't consider than an unreasonable wish. 21:18:05 a game in under 10mb? 21:18:10 bougyman: the binary. 21:18:19 p_l|backup: you don't live in a city? 21:18:31 and we're not necessarily talking about crysis here. :) 21:18:46 sykopomp: if you were to work on a treeshaker, I doubt the work would be unappreciated. 21:18:55 Fade: I would if I could. 21:19:10 sykopomp: have you tried lispworks? 21:19:33 -!- madnificent [~madnifice@83.101.62.132] has quit [Ping timeout: 252 seconds] 21:19:44 Fade: right now, I'm in a city, and getting good network requires hiking few kilometers to university :/ 21:19:56 :\ 21:20:03 drewc: I'm considering it, yes, but I'd need to guarantee a source of income from the applications i'm going to distribute if I were ever even willing to try it. 21:20:12 badipod [~badipod@unaffiliated/baddog144] has joined #lisp 21:20:17 I'm spoiled, maybe. I even have broadband at my cottage. 21:20:19 (I live in rented room, getting internet might be hard) 21:20:20 and I don't even know if LW would be fast enough to not choke on simple physics simulations. 21:20:32 and that would be yet another implementation that I'd need to test code on. 21:20:39 (but hey, if it works...) 21:20:42 Fade: at home, for a long time, it was 48kbit/s and nothing more 21:20:50 gack. 21:21:06 I've had at least bonded ISDN since the early nineties. 21:21:40 i remember when I used to long for a T1. now my phone gets 15mbit and feels sluggish, still. 21:22:04 Fade: where my family lives, the telephone provider decided that hauling extra wire (~500m in straight line) isn't necessary and dumped everyone with GSM terminals 21:22:20 ehu [~ehuels@ip118-64-212-87.adsl2.static.versatel.nl] has joined #lisp 21:22:21 lol 21:22:56 ouch 21:23:00 said terminals can at most hook to good old modem or allow basic GPRS, but you're billed premium 21:23:05 though now i'm wishing i would have waited for the Epic. I'm going to convince my wife to take my Evo so I can get the Epic, i think. 21:23:23 the most I did with data access on them was to send a fax :) 21:23:46 you live in a very different world than me, p_l|backup 21:23:48 :) 21:23:55 -!- kejsaren_ [~kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Ping timeout: 276 seconds] 21:24:04 -!- jgracin [~jgracin@vipnet486.mobile.carnet.hr] has quit [Ping timeout: 265 seconds] 21:25:06 Fade: right now we've got better connection, cause I find a small ISP that uses 802.11a links while looking for a provider to use (I was going to start the ISP business myself due to lack of options) 21:25:51 802.11_a_ 21:25:52 ? 21:26:26 Fade: You didn't think the numbering started at 802.11b, did you? 21:27:09 no, I actually remember deploying 802.11a when I was at UUnet, but I haven't seen or heard of anybody actually using it in... long time. 21:27:35 Probably means that the hardware is cheap on the used market... 21:27:38 Fade: 802.11a is quite common for "backhaul" links between basestations, it seems 21:27:42 christop` [christoph@cl-1281.dus-01.de.sixxs.net] has joined #lisp 21:28:08 nyef: many full 11n cards also support 11a 21:28:20 (as 11n includes frequencies used by 11a) 21:28:23 Ah. Fair enough. 21:28:24 the radio base stations that are used to connect my cottage to a tower about 2.5 km away are built in poland. 21:28:38 they aren't using any of the 802.11 stuff, though. 21:28:49 as for 11a hardware, microtik boards are very common 21:28:53 -!- christop` [christoph@cl-1281.dus-01.de.sixxs.net] has quit [Client Quit] 21:28:59 (You know you're fucked when your wireless networking, your telephone, and your microwave oven all operate on the same set of frequencies.) 21:29:09 heh 21:29:35 nyef: aka (big chunk of) GSM, microwaves and common wifi hw? :D 21:30:02 well, GSM slightly less, due to amount of 800/850/900 MHz hw 21:30:24 -!- christoph_debian [christoph@cl-1281.dus-01.de.sixxs.net] has quit [Remote host closed the connection] 21:30:28 1800/1900 is rather close and low-quality microwaves might interact with it 21:30:36 christoph_debian [christoph@cl-1281.dus-01.de.sixxs.net] has joined #lisp 21:30:53 -!- aw [~aw@147.81-166-214.customer.lyse.net] has quit [Ping timeout: 240 seconds] 21:32:33 p_l|backup: I was more thinking of the 2.4GHz cordless handsets that are far too common around here. 21:32:59 nyef: btw, at home we are still thinking of putting an ethernet cable outside building to connect AP in attic and one in office, to get the signal usable :) 21:33:12 -!- Joreji [~thomas@82-238.eduroam.RWTH-Aachen.DE] has quit [Ping timeout: 265 seconds] 21:33:31 tayloj [~tayloj@cpe-67-248-22-159.nycap.res.rr.com] has joined #lisp 21:33:56 Joreji [~thomas@82-238.eduroam.RWTH-Aachen.DE] has joined #lisp 21:35:25 it's too late to add cables and we found the hard way that WiFi doesn't deliver :D 21:37:36 some of the contrib modules have failed to build asdf-install sb-bsd-sockets sb-posix sb-simple-streams 21:38:03 T1750: What version? 21:38:03 *T1750* is trying to work out how much of this needs to be pastebinned for help 21:38:07 1.0.39 21:38:18 source, built with 1.0.39 21:38:22 ... Hunh. 21:38:26 *T1750* is trying to package sbcl for OpenSUSE 21:38:35 T1750: what libc do you use? 21:38:39 6 21:39:06 libc-2.10.1.so 21:39:16 nyef: there was an issue with waitpid needing an ldso stub, earlier, fwiw. 21:39:29 glibc-2.10.1-10.5.1.i686 21:39:56 if my package is good they mighr even make it part of proper suse as well as clisp 21:40:33 unfortunately not knowing a lot of lisp isn't helping ;D but you guys can do that i'm just like a middleman 21:40:37 and sbcl spreads 21:41:06 OmniMancer [~OmniMance@219-89-109-120.jetstart.xtra.co.nz] has joined #lisp 21:42:00 T1750: At this point, I'm going to suggest trying the latest SBCL from CVS/git, as we're actually in the code-freeze period for a new release. 21:42:13 1.0.40 should be out "soon". 21:42:58 lpolzer_ [~lpolzer@dslb-088-073-222-035.pools.arcor-ip.net] has joined #lisp 21:43:32 bombshelter13b [~bombshelt@76-10-149-209.dsl.teksavvy.com] has joined #lisp 21:43:59 well can i get this one working first it shouldn't be too hard to update to the next one when it comes out 21:44:05 http://pastebin.com/AXsV5tKR 21:44:11 does this mean anything to you? 21:45:06 i could probably turn off that disable debugger thing so i can interact with it 21:45:19 No, no... My point is that the new version might not have whatever problem you're running into. 21:45:25 oh 21:45:26 i see 21:45:36 right :) 21:45:42 does it have the waitpid problem? 21:45:58 Although, find-dynamic-foreign-symbol-address seems rather important... 21:46:31 -!- lpolzer [~lpolzer@dslb-088-073-202-010.pools.arcor-ip.net] has quit [Ping timeout: 276 seconds] 21:46:36 Same issue, different symbol. "getaddrinfo" this time. 21:46:47 legumbre_ [~leo@r190-135-45-50.dialup.adsl.anteldata.net.uy] has joined #lisp 21:48:58 -!- legumbre [~leo@r190-135-44-46.dialup.adsl.anteldata.net.uy] has quit [Ping timeout: 264 seconds] 21:50:07 Athas [~athas@82.211.209.162] has joined #lisp 21:50:56 -!- Aisling_ [ash@blk-222-192-36.eastlink.ca] has quit [Read error: No route to host] 21:51:51 -!- abend [~alx@67.136.131.11] has quit [Ping timeout: 245 seconds] 21:52:54 well i'll just wait until you release 1.0.40 21:53:11 but you see i'm trying to get a big IT contract and get off welfare and i'm getting somewhere 21:53:14 -!- asarch [~asarch@189.188.145.217] has quit [Quit: Leaving] 21:53:26 if by the time i get it if i do, i'd like to know enough lisp i don't end up doing it in python :) 21:53:35 and if i can't package it in opensuse then that wont happen 21:54:18 *T1750* wants a fully working no bugs sbcl then goto the "i am a lisp newbie" tutorial do it all 21:54:31 thats the workflow i don't want to learn to debug lisp before i know lisp 21:55:12 hadronzoo [~hadronzoo@64.134.146.171] has joined #lisp 21:55:17 -!- astalla [~astalla@93-36-230-5.ip62.fastwebnet.it] has quit [Quit: Sto andando via] 21:55:25 -!- tritchey [~tritchey@205.233.9.181] has quit [Quit: tritchey] 21:55:48 -!- c|mell [~cmell@188-220-238-74.zone11.bethere.co.uk] has quit [Remote host closed the connection] 21:56:05 -!- ehu [~ehuels@ip118-64-212-87.adsl2.static.versatel.nl] has quit [Ping timeout: 240 seconds] 21:56:08 Heh. A "fully working no bugs sbcl"? I'd -love- one of those. 21:56:20 -!- milanj [~milanj_@178.223.151.188] has quit [Ping timeout: 260 seconds] 21:56:25 oh 21:56:29 i'll take one of those too! 21:56:34 the whole "it compiles to native code" sold me on it 21:56:57 well i can probably fix bits of it myself once i get the damn thing working without bugs at stage 1 21:57:04 *T1750* does both bug report and patch 21:57:09 Hell, I'd be happy for now with working threads on all of my hardware, and that bloody nonlinear linear-variable problem gone. 21:57:39 you are making me want to stick to python stop saying heresy 21:58:05 -!- ziggy_ [~ziggy@de2-as23404.alshamil.net.ae] has quit [Quit: Leaving] 21:58:19 Nah, it's just that we've used SBCL for so long (or in my case hacked on it for so long) that we know where some of the nastier bugs are. 21:58:28 -!- timor [~timor@port-92-195-27-237.dynamic.qsc.de] has quit [Remote host closed the connection] 21:58:32 well, please put a list of them up 21:58:36 Like the two-unwind-protects-one-frame thing for Win32... 21:58:39 for newbies so we don't run into them :) 21:58:54 [mzm] [~ziggy@de2-as23404.alshamil.net.ae] has joined #lisp 21:58:59 wiki wiki 21:59:06 Does SBCL not have a bugtracker? 21:59:18 T1750: i'd be pretty suprised if you manage to trigger any of these bugs while learning CL 21:59:35 badipod: We use launchpad. 21:59:49 I see 22:00:17 https://bugs.launchpad.net/sbcl/ 22:00:17 -!- attila_lendvai [~attila_le@89.135.200.140] has quit [Read error: Connection reset by peer] 22:00:36 drewc: i'm one of those "slow starter, then learns with logarithmic increases" type people 22:00:45 kleppari [~spa@bitbucket.is] has joined #lisp 22:00:53 it's not that unlikely i have a habit of poking dark corners to gain understanding 22:01:54 -!- levente_meszaros [~levente_m@4d6f5d3b.adsl.enternet.hu] has quit [Ping timeout: 252 seconds] 22:02:00 attila_lendvai [~attila_le@adsl-89-134-26-115.monradsl.monornet.hu] has joined #lisp 22:02:18 anyways it won't be just CL i will be needing to build add-ins for it in C to do what i want 22:02:28 unless other already made them 22:02:46 T1750: you really should learn CL first before trying to build things in it. 22:02:55 (in my opinion) 22:03:03 i agree 22:03:21 i also agree that i should be able to download and have SBCL work without bugs to have a clean perfect place to learn 22:03:34 it was my intention to go and do that lisp book as soon as i got that stage 22:03:34 well, then you're fucked, SBCL will always have bugs 22:03:43 lol how come? 22:03:50 All software has bugs. 22:03:51 because it's software 22:03:52 is there any other lisp which will make native code? 22:03:52 SBCL is software. 22:03:58 Therefore SBCL has bugs. 22:04:06 (Or was it all software is SBCL?) 22:04:07 -!- xan_ [~xan@213.47.69.75] has quit [Ping timeout: 240 seconds] 22:04:43 did that pastebin mean anything to anyone? 22:04:50 -!- adu [~ajr@pool-71-191-173-125.washdc.fios.verizon.net] has quit [Quit: adu] 22:04:55 T1750: there are indeed, and they all have bugs too. new to software are you? ;) 22:04:59 Pohsul [Pohsul@83.231.17.70] has joined #lisp 22:05:31 which is my best choice LISP for making commercial code 22:05:44 stuff where i will sell them native binary as a service 22:06:14 T1750: LispWorks, I guess. Little pricy if you're on tight budget, though 22:06:27 so if i'm on a tight budget SBCL? 22:06:59 -!- thunk is now known as thunk|smoking 22:06:59 -!- stis [~stis@1-1-1-39a.veo.vs.bostream.se] has quit [Read error: Connection reset by peer] 22:07:10 *T1750* likes the way lisp is so configurable it will be "your own language" 22:07:10 tritchey [~tritchey@205.233.9.181] has joined #lisp 22:07:28 c|mell [~cmell@188-220-238-74.zone11.bethere.co.uk] has joined #lisp 22:07:45 T1750: the answer is not that simple... clisp, ecl, sbcl, ccl, lw, acl, abcl... all have advantages and disadvantages. 22:08:08 -!- tritchey [~tritchey@205.233.9.181] has quit [Client Quit] 22:08:17 T1750: http://common-lisp.net/~dlw/LispSurvey.html is a good resourse 22:08:20 resource 22:08:22 T1750: SBCL will fit nearly all of the tasks if you're on x86/amd64 linux 22:08:55 i can code C though i'm nothing special, if I have a reference website open i can make anything in C 22:09:04 so i can add things which it lack 22:09:15 so you want a lisp that is written in C? 22:09:31 no i mean if SBCL doesn't do something i need that is OS related i can add it 22:09:44 go ecl then 22:09:47 it has c 22:09:49 you can add it anyway. 22:09:54 by 'add it' you mean 'call it via FFI'? 22:10:16 cmsimon [~tific@unaffiliated/cmsimon] has joined #lisp 22:10:21 *T1750* recognises those lists we just patched of string 22:10:25 -!- t3eblinder [~wolfgang@p5B2013DC.dip0.t-ipconnect.de] has left #lisp 22:10:52 a lot of them are c function calls 22:11:04 waitpid must have been made with C 22:11:05 T1750: Yes, you can call C from SBCL. From a bunch of other lisp implementations, too. 22:11:23 T1750: enough that there's a portability layer library called CFFI that gives you a cross-impl interface. 22:12:15 T1750: by C function calls you mean 'functions that follow C calling conventions'? ;) 22:12:28 *drewc* is not going to get anywhere with this one, so goes back to work 22:12:33 drewc: to waitpid it must have accessed a syscall 22:12:51 *T1750* also knows a little linux asm 22:12:54 x86 22:13:02 -!- roygbiv [~JohnRambo@pdpc/supporter/active/roygbiv] has left #lisp 22:13:20 iirc SBCL goes through libc wrappers for syscalls, right? 22:13:26 you guys seem like you don't want me to make the sbcl work 22:13:34 exactly my point... waitpid could have been written in any language, including ASM... this has little to do with C 22:13:47 -!- Pohsul [Pohsul@83.231.17.70] has quit [] 22:13:48 maybe i will just use python this is giving me a headache 22:13:48 T1750: hey, don't generalize! Not all of us are jerks. 22:14:00 sorry 22:14:04 T1750: good idea! let people on IRC tell you what to do! :) 22:14:05 that's probably a good idea. #python is pretty friendly. :P 22:14:20 i don't need any much help with python :) 22:14:26 it's docs are amazing 22:15:05 -!- LiamH [~none@pdp8.nrl.navy.mil] has quit [Quit: Leaving.] 22:15:05 ok 22:15:19 ... "Error: Cannot find source location for: #". Gee, thanks, slime. 22:16:13 T1750: the answer to your question is "Yes, Lisp can get access to libraries written in C" 22:16:23 nyef: lots of that going on here. 22:17:41 drewc: have you been able to build lisp-on-lines w/ SBCL 1.0.39.21 yet? 22:18:10 sykopomp: http://pastebin.com/AXsV5tKR 22:18:11 mon_key: i don't think i've tried 22:18:14 can you make sense of this? 22:18:18 mon_key: what errors are you getting? 22:18:40 mon_key: are you using the lol from darcs or from git? 22:18:53 drewc: git I'm pretty sure. 22:19:03 ok, well that's the right one 22:19:11 drewc: its git 22:19:45 -!- franki^ [~franki@unaffiliated/franki] has quit [Ping timeout: 240 seconds] 22:20:27 frame 0: SB-PCL::FAST-METHOD ASDF:PERFORM (ASDF:COMPILE-OP ASDF:CL-SOURCE-FILE)) ..) 22:20:36 CrEddy [~CrazyEddy@wrongplanet/CrazyEddy] has joined #lisp 22:20:47 -!- badipod [~badipod@unaffiliated/baddog144] has quit [Quit: Colloquy for iPod touch - http://colloquy.mobi] 22:21:20 ok so The function SB-SYS:FIND-DYNAMIC-FOREIGN-SYMBOL-ADDRESS 22:21:37 what kind of error is this 22:21:45 one with SBCL or one with ASDD 22:21:47 ASDF 22:23:03 dullard [~user@188-223-137-38.zone14.bethere.co.uk] has joined #lisp 22:23:09 what are you really trying to accomplish? what is the goal? if you want sbcl, just download the binary and leave the whole building thing for now. 22:23:24 -!- hadronzoo [~hadronzoo@64.134.146.171] has quit [Quit: hadronzoo] 22:23:57 -!- hohoho [~hohoho@ntkngw227224.kngw.nt.ftth.ppp.infoweb.ne.jp] has quit [Ping timeout: 248 seconds] 22:24:29 mon_key: can you paste (at lisppaste) the actual error? If you're loading via ASDF, the error that causes the compilation error should be printed somewhere, either at your repl or in your *inferior-lisp* 22:24:44 -!- lhz [~shrekz@c-dba672d5.021-158-73746f34.cust.bredbandsbolaget.se] has quit [Quit: Leaving] 22:24:56 drewc: yes, I will paste in a moment. 22:25:01 hypno: two things, i want to build the suse package and 2, i want to know SBCL can be trusted because it will build correctly when you follow its build instructions 22:25:15 2 is failing, but if i knew why it's failing maybe it's not a big deal 22:25:15 T1750: that error doesn't look complete :) 22:25:26 drewc: still crawling over the backtrace 22:25:40 mon_key: Wait, wait... This isn't specifically a problem with an unexpected offset in one of the SB-VM::EA-FOR-C?F-????-STACK functions, is it? 22:25:55 T1750: probably everything will compile except sb-concurrency 22:26:09 nyef: :) not that i know of 22:26:15 Oh, okay. 22:26:22 and it's not even because of some inherent bugs in sb-concurrency, as far as I had understood it 22:26:27 the contribs dont work i guess being contrib i can do what fedora did and just disable the checks they work 22:26:32 (Good. This appears to be long-standing nastiness. As in April, 2009.) 22:26:34 but ... shouldn't they work? 22:27:02 p_l|backup: Inherent bugs in sb-concurrency? What? 22:27:21 nyef: I'd rather say "bad decisions in test failure cases" 22:27:44 Ah. 22:27:45 the only reason I don't have sb-concurrency in my SBCL is because it fails... on timeout 22:28:48 which is kind of stupid, as the timeouts can be (and probably are) caused by external issues 22:29:22 madnificent [~madnifice@83.101.62.132] has joined #lisp 22:30:37 -!- CrEddy [~CrazyEddy@wrongplanet/CrazyEddy] has quit [Ping timeout: 264 seconds] 22:30:44 Mmm. My own sb-concurrency failure mode is, of course, the memory-barrier thing, which I still haven't settled in to figure out an implementation plan for. 22:31:00 R3cur51v3 [~Recursive@173-29-151-107.client.mchsi.com] has joined #lisp 22:31:01 -!- astoon [~astoon@213.141.244.246] has quit [Quit: astoon] 22:32:09 is there an updated version of sb-heapdump? 22:32:41 -!- HET2 [~diman@chello084114129176.4.15.vie.surfer.at] has quit [Quit: This computer has gone to sleep] 22:34:17 Shaftoe [~Moe111@bas1-montreal02-1096728247.dsl.bell.ca] has joined #lisp 22:34:35 abend [~alx@076-076-146-016.pdx.net] has joined #lisp 22:35:41 could anyone tell me how I could convert a string that has been blindly imported in sbcl into a character encoded utf string. 22:35:56 so essentially convert a data buffer into utf. 22:38:03 Pohsul [Pohsul@83.231.21.84] has joined #lisp 22:38:13 drewc: http://paste.lisp.org/+2EFM 22:38:24 drewc: that one is from clbuild 22:38:54 drewc: I will add an annotation for one without slime/clbuild 22:39:36 Shaftoe: The correct procedure rather depends on what "blindly imported" means. And note that under some circumstances such an import process might fail with an error message or lose data required for accurate utf conversion. 22:40:40 nyef: my situation is like so: it seams the clsql-odbc implementation of clsql (and only that one) has a bug whereby BLOBs are not properly character encoded. The read-data-in-chunks method just reads the data. 22:41:02 Reads the data, how? 22:41:06 In a first step, I want to see if that data is actually the right data (i.e. at first, I'm looking to see if there's a single function that could just quickly do it) 22:41:53 mon_key: that's odd, my version just grabbed from git has the patch for that. Can you look in src/packages.lisp and see if there's a shadowing-import-from :closer-mop for those symbols? 22:42:14 as far as I can tell, using this: %sql-get-data 22:43:21 nyef: the thing I'm trying to avoid though is this: currently the entire scaffholding for this read-data-in-chunks is building a string. I can't just insert a conversion routine into the chunked reads because I could have a chunk that cuts a code point in half. 22:43:23 p_l|backup: i don't think anybody has working on sb-heapdump in a long time... it was dlichtblau's iirc 22:43:24 drewc: will do 22:43:33 ... How does the data get pulled off the wire? What transformations, if any, does it go through on its way into the lisp heap? How is it stored on the heap? 22:43:34 so I would need to rewrite the entire method. 22:43:53 -!- Athas [~athas@82.211.209.162] has quit [Remote host closed the connection] 22:44:05 ... And what on earth happens if you want to stash a -binary- blob? 22:44:40 Shaftoe: why are you keeping it in chunks if you need the entire string anyway? can't we make a long array first and then octets-to-string it? 22:44:40 nyef: what happens is that I don't do binary for now. This is a bug fix for a show-stopper right now... 22:45:37 nyef: to answer you, I've done a test where there's a stored proc that outputs static unicode data (via static select statement) plus ntext, text, and nvarchar columns. static data and nvarchar both properly output the data. 22:46:30 ntext and text do not. 22:46:36 test was as follows: http://paste.lisp.org/display/111985 22:46:42 *nyef* sighs. 22:46:55 -!- attila_lendvai [~attila_le@adsl-89-134-26-115.monradsl.monornet.hu] has quit [Quit: Leaving.] 22:47:06 -!- dullard [~user@188-223-137-38.zone14.bethere.co.uk] has quit [Read error: Connection reset by peer] 22:47:32 drewc: I don't store the chunks. I'm saying the current method coerces the chunks into a string. I can't just put a blind conversion on the chunks. 22:47:58 -!- dysinger [~dysinger@173-13-81-254-NewEngland.hfc.comcastbusiness.net] has quit [Quit: dysinger] 22:48:06 -!- madnificent [~madnifice@83.101.62.132] has quit [Ping timeout: 265 seconds] 22:48:15 -!- T1750 [~T1750@unaffiliated/t1750] has left #lisp 22:48:20 coerces the chunks to a string... without joining them first? that's not going to work. 22:48:46 so, my suggestion would be to not do that. 22:50:09 well... it could be made to work, but seems like more trouble than it's worth. 22:50:12 -!- abugosh [~Adium@207-172-111-88.c3-0.tlg-ubr4.atw-tlg.pa.cable.rcn.com] has quit [Quit: Leaving.] 22:50:41 drewc: clbuild/source/closer-mop/closer-mop-packages.lisp 22:51:05 has this #+(or allegro clisp clozure ecl lispworks sbcl) 22:51:05 (:shadow #:defgeneric #:defmethod #:standard-generic-function) 22:51:16 mon_key: nay, lisp-on-lines/src/packages.lisp is what i'm interested in 22:51:16 drewc: you are right. I'm looking at the code, and what it is doing is this: 22:51:32 drewc: not that i can see 22:51:53 -!- Joreji [~thomas@82-238.eduroam.RWTH-Aachen.DE] has quit [Ping timeout: 260 seconds] 22:51:56 http://paste.lisp.org/display/112019 22:51:57 mon_key: how odd. latest lol from git using latest clbuild? 22:52:34 I believe so. But before that I had it added by hand. 22:52:56 make sure with a fresh install before i go ahead and try it myself, please :) 22:53:18 That is... painful to look at. 22:53:53 What the heck is "%cstring-into-vector"? 22:53:58 drewc: http://common-lisp.net/pipermail/clbuild-devel/2010-May/000499.html 22:54:07 nyef: I know. 22:54:39 Shaftoe: ouch, that code is scary... i remember now why i stopped using cl-sql! :) 22:54:48 mon_key: yeah, so we should be up to date 22:55:00 Fade: lol works out of the box with clbuild these days, non? 22:55:16 http://paste.lisp.org/display/112019 22:55:24 drewc: I can do a fresh install but it may take a few hours. 22:55:39 drewc: quicklisp, too :) 22:55:50 Next question, what is "ensure-char-character"? 22:56:12 Xach: quicklisp FTW! :) 22:56:23 *drewc* should transition his projects to quicklisp 22:56:34 Xach: what is quicklisp? 22:56:53 Oh, lovely. Web browser just locked up. :-/ 22:57:00 mon_key: quicklisp is an unfinished idea for making it easy to get a CL library environment up and running quickly. 22:57:14 mon_key: it's a little like clbuild in that regard. 22:57:21 Xach: more finished than clbuild? 22:57:24 only better! 22:57:29 nyef: refresh paste 22:57:47 (I am using sbcl) 22:57:47 Hang on, recovering my browser session. 22:57:53 mon_key: I don't really know how finished clbuild is, sorry. 22:57:56 ok. 22:58:08 mon_key: quicklisp has the distinct advantage that it's written in lisp and doesn't rely on version control 22:58:17 drewc: lol 22:58:24 -!- bobbysmith007 [~russ@216.155.97.1] has quit [Quit: Leaving.] 22:58:49 -!- Davidbrcz [~david@212-198-91-47.rev.numericable.fr] has quit [Ping timeout: 264 seconds] 22:59:12 nyef: it is a defmacro for code-char 22:59:20 Yeah, I see now. 22:59:25 -!- MetalDust [~metaldust@75-32-203-61.lightspeed.ftwotx.sbcglobal.net] has quit [] 23:01:00 So, the simplest thing to do here is to use octets-to-string on the result of (map-into (make-array (length blob) :element-type (unsigned-byte 8)) #'char-code blob) with your desired encoding. 23:01:04 quicklisp is a glimmer of hope in these dark times. 23:01:07 mon_key: would you like to try an early proof-of-concept demo? 23:01:14 -!- tychoish [~tychoish@foucault.cyborginstitute.net] has quit [Remote host closed the connection] 23:01:29 tychoish [~tychoish@foucault.cyborginstitute.net] has joined #lisp 23:02:27 Xach: I feel like i should be waiting for the punchline to drop.. 23:02:27 nyef: what you call blob would be the vector from %cstring-into-vector 23:02:28 ? 23:02:53 mon_key: ok. 23:04:18 T1750 [~T1750@unaffiliated/t1750] has joined #lisp 23:04:44 Umm... Yeah, that might work. 23:04:51 Xach: is it your project? 23:05:04 mon_key: yes. 23:05:08 Oh, hell, the first mistake is %cstring-into-vector. 23:05:22 doesn't it just... convert bytes to chars? 23:05:24 Isn't there already some sort of converter for that in SBCL? 23:05:32 nyef: =) 23:05:41 nyef: I know, that's what I was kinda aiming for. 23:06:01 something which could take a byte array from one side and output a char encoded string from the other. 23:06:01 Xach: reading a lisp log from 2010-06-21 re quicklisp 23:06:09 nixeagle [~user@Wikimedia/Nixeagle] has joined #lisp 23:06:48 sb-ext:octets-to-string .. non? or babel:octets-to-string if you want to be portable. 23:06:56 nyef: if you were to supplant %cstring-into-vector, what would you recommend be the path I take? 23:07:11 Xach: so you distribut the client? 23:08:29 mon_key: yes. 23:08:41 Xach: from http://src.quicklisp.org/ 23:08:44 drewc: I see flexi-streams has an octets-to-string function too 23:09:15 There's bound to be an SBCL function that takes in a SAP/length pair and an external-format and returns a string. 23:09:24 Or similar. 23:09:35 Shaftoe: (loop :for i :upto (foreign-string-length data-ptr) :do (setf (aref str i) (deref-array ptr '(:array :unsigned-char) i)) :finally (octets-to-string str)) 23:10:23 or something like that. 23:10:33 drewc: ha ha. That' would be damn sweet if it just works. I'm starting the hackamachine. 23:10:36 The problem comes when the database returns chunks that aren't split on an encoded-character boundary and when the characters can be more than one octet wide anyway. 23:10:47 Xach: interesting to compare the dependency graph at http://src.quicklisp.org/systems.txt for lisp-on-lines 23:11:12 yeah. that's the biggest problem: codepoints split between chunks. 23:11:19 mon_key: http://xach.com/tmp/ql.pdf has an actual graph 23:11:28 quicklisp -> lisp-on-lines lisp-on-lines contextl arnesi alexandria parse-number yaclml trivial-garbage 23:11:33 clbuild/dependencies -> lisp-on-lines alexandria anaphora arnesi cl-ppcre closer-mop contextl iterate lw-compat parenscript parse-number puri trivial-garbage yaclml 23:11:57 also, I'm not sure it's a contiguous memory zone, as far as I know, consuming a blob requires repeated calls to a get-data style function 23:12:02 I'm looking into this right now. 23:12:14 -!- Yuuhi [benni@p5483B7C9.dip.t-dialin.net] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 23:12:21 mon_key: i think clbuild dependencies are maintained by hand 23:12:31 mon_key: quicklisp automates it. 23:12:39 <_3b> clbuld deps are automated as well 23:12:51 *_3b* suspects the difference is per-system vs per-project 23:13:40 <_3b> or possibly quicklisp doesn't list deps of deps the same way 23:13:46 jsfb [~jon@unaffiliated/jsfb] has joined #lisp 23:13:53 #nookdevs 23:14:13 oops 23:14:16 badipod [~badipod@unaffiliated/baddog144] has joined #lisp 23:14:25 _3b: looks like ql's is recursive? 23:15:52 <_3b> mon_key: if you mean ql only lists direct deps, that was the difference i was suspecting 23:16:06 _3b> e.g. lisp-on-lines-ucw lisp-on-lines lisp-on-lines ucw puri parenscript 23:16:24 <_3b> (which is possible due to tracking deps per-system instead of per-project) 23:16:55 *Xach* tracks deps per-system 23:17:22 <_3b> Xach: only direct deps, right? 23:18:23 AntiSpamMeta [~MetaBot@unaffiliated/afterdeath/bot/antispambot] has joined #lisp 23:19:06 drewc: FWIW lisp-on-lines from darcs on sbcl 1.0.39.18 builds ok. 23:19:17 kvsari [~kvsari@203.171.93.21.static.rev.aanet.com.au] has joined #lisp 23:19:22 _3b: yes. 23:19:47 mon_key: from darcs? that's odd, the darcs repository is somewhat lagging... 23:19:57 drewc: on a different machine. trying with lol from git before rebuilding a new clbuild. 23:20:15 -!- R3cur51v3 [~Recursive@173-29-151-107.client.mchsi.com] has quit [Ping timeout: 260 seconds] 23:20:18 <_3b> that would be the difference then, clbuild lists all dependencies, not just direct ones 23:20:47 clbuild also has the dependencies for the lisp-on-lines.test system 23:21:13 quicklisp's list does not include those 23:21:25 -!- Patzy [~something@coa29-1-88-174-11-170.fbx.proxad.net] has quit [Ping timeout: 260 seconds] 23:21:25 <_3b> right, clbuild tracks them by project rather than system 23:21:28 Patzy [~something@coa29-1-88-174-11-170.fbx.proxad.net] has joined #lisp 23:23:07 drewc: lisp-on-lines.test failed, i think. let me check my logs... 23:24:03 i'm not even sure it still builds with libraries in the wild 23:26:00 drewc: I can get lisp-on-lines from git to fail on sbcl 1.0.39.18 23:26:17 drewc: same error around `defgeneric' 23:31:19 -!- mrSpec [~Spec@unaffiliated/mrspec] has quit [Quit: mrSpec] 23:31:39 that's odd. 23:32:44 <_3b> possibly should :use c2cl instead of cl and closer ? 23:35:11 nyef and drewc: I have liftoff. thanks to both of you. 23:35:18 gzip4 [~xxx@78.108.73.250] has joined #lisp 23:35:27 abugosh [~Adium@207-172-111-88.c3-0.tlg-ubr4.atw-tlg.pa.cable.rcn.com] has joined #lisp 23:35:28 jcw [~Adium@madder.xs4all.nl] has joined #lisp 23:36:02 R3cur51v3 [~Recursive@173-29-151-107.client.mchsi.com] has joined #lisp 23:36:51 -!- kvsari [~kvsari@203.171.93.21.static.rev.aanet.com.au] has quit [Quit: Lost terminal] 23:37:20 -!- Adrinael [~adrinael@barrel.rolli.org] has quit [Read error: Operation timed out] 23:38:11 -!- thunk|smoking is now known as thunk 23:38:14 oh, i see 23:38:28 -!- rat [~quassel@c-24-22-21-120.hsd1.or.comcast.net] has quit [Ping timeout: 276 seconds] 23:39:15 python rules 23:39:16 -!- T1750 [~T1750@unaffiliated/t1750] has left #lisp 23:40:28 -!- abugosh [~Adium@207-172-111-88.c3-0.tlg-ubr4.atw-tlg.pa.cable.rcn.com] has quit [Quit: Leaving.] 23:41:06 mon_key: i have to run out, but i see what the problem is. I'll fix it when i get back 23:41:38 drewc: the idea corrupted me (the one with "simple programming" tool) :) 23:43:24 drewc: great sorry to be the bearer of bad news. 23:45:14 -!- jcw [~Adium@madder.xs4all.nl] has left #lisp 23:46:26 Adrinael [~adrinael@barrel.rolli.org] has joined #lisp 23:48:39 bohunm [~mbohun@202.124.73.129] has joined #lisp 23:49:38 -!- R3cur51v3 [~Recursive@173-29-151-107.client.mchsi.com] has quit [Ping timeout: 260 seconds] 23:50:31 pizzledizzle [~pizdets@pool-96-250-215-244.nycmny.fios.verizon.net] has joined #lisp 23:54:50 -!- Soulman [~knute@79.80-202-237.nextgentel.com] has left #lisp 23:55:00 madnificent [~madnifice@83.101.62.132] has joined #lisp 23:55:22 CrEddy [~CrazyEddy@wrongplanet/CrazyEddy] has joined #lisp 23:57:28 JD__ [~jd@5e091ddd.bb.sky.com] has joined #lisp 23:58:33 -!- JD__ [~jd@5e091ddd.bb.sky.com] has left #lisp 23:58:36 minor announcement :) http://dto.github.com/notebook/2010expo.html 23:58:45 of our new improved lispy game contest