00:00:23 -!- marioxcc is now known as marioxcc-ocupado 00:00:33 *madnificent* goes back to searching for his mistake 00:01:33 you can post your code 00:02:05 -!- varjag [n=eugene@103.80-202-117.nextgentel.com] has quit ["Leaving"] 00:02:41 ah right, I somehow missed the way evaluation was done. I need a macro to make my example work (stupid stupid stupid) 00:02:57 thanks for the pointers, I was looking at the wrong spot 00:03:38 benny [n=benny@87.122.15.255] has joined #lisp 00:05:52 quek [n=read_eva@router1.gpy1.ms246.net] has joined #lisp 00:06:44 Hrm... I think the comment above gc_stop_the_world() in src/runtime/thread.c might be a little inaccurate... 00:07:12 chavo_ [n=user@c-71-195-63-115.hsd1.mn.comcast.net] has joined #lisp 00:07:15 *stassats* wouldn't mind a stop_the_world() function 00:07:46 "Stop the world, I want to get off." 00:08:05 -!- Hun [n=hun@pd956be5d.dip0.t-ipconnect.de] has quit [Remote closed the connection] 00:08:23 i just wanted to get some rest and then start it again 00:08:46 Recreating all of this mess for Win32 is going to be... interesting... 00:09:04 Urfin [i=foobar@85.65.93.74.dynamic.barak-online.net] has joined #lisp 00:09:14 *madnificent* gives stassats one 'stop the world game-card' 00:09:22 -!- envi_office [i=envi@203.109.25.110] has quit ["Leaving"] 00:09:47 nyef: is this because of win32 or because sbcl design isn't particularly suited for win32 00:09:57 stassats: use only if you're certain that you're not part of the world! 00:10:15 stassats: A little of both. 00:10:22 madnificent: it's ok, i'm just a part of the universe 00:11:39 stassats: Basically, Win32 has no asynchronous events, but there's a way to abuse the debug APIs to provide them. But because there's no support for asynchronous events -none- of the existing code is going to apply but most of the logic still does. 00:13:26 And, of course, there's the further complication that Win32 doesn't have the same set of thread synchronization primitives as pthreads, which means that even the low-level locking doesn't really translate across. 00:14:01 -!- Edward_ [n=Ed@AAubervilliers-154-1-29-185.w90-3.abo.wanadoo.fr] has quit ["L'oignon fait la farce."] 00:14:38 envi^office [i=envi@203.109.25.110] has joined #lisp 00:14:55 lmr [n=lemurian@unaffiliated/lemurian] has joined #lisp 00:15:02 slyrus_ [n=slyrus@75.36.222.205] has joined #lisp 00:17:38 -!- marioxcc-ocupado is now known as marioxcc 00:17:55 Okay, going offline for a bit. Back in perhaps half an hour. 00:18:02 -!- nyef [n=nyef@static-64-222-188-16.man.east.myfairpoint.net] has quit ["Gone for a bit."] 00:18:11 nyef: please see pm 00:20:14 lmr: let minion know about it 00:20:34 minion can't keep secrets 00:21:39 madnificent: ? 00:21:44 -!- Nshag [i=user@Mix-Orleans-106-4-104.w193-248.abo.wanadoo.fr] has quit [Remote closed the connection] 00:22:14 minion: help memos 00:22:15 To send a memo, say something like ``minion: memo for nick: the memo''. I'll remember the memo for any nick which is the same as the given nick, +/- differences in punctuation, and any nick which is an alias for it, and give it to them when they next speak. 00:22:31 -!- |male_terran| [n=KVIrc@ip98-162-161-191.pn.at.cox.net] has quit [] 00:23:04 -!- daniel__ [i=daniel@unaffiliated/daniel] has quit [Read error: 110 (Connection timed out)] 00:25:55 nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 00:29:01 -!- slyrus_ [n=slyrus@75.36.222.205] has quit [Read error: 145 (Connection timed out)] 00:29:40 slyrus_ [n=slyrus@adsl-75-36-222-205.dsl.pltn13.sbcglobal.net] has joined #lisp 00:32:26 -!- ignas [n=ignas@ctv-79-132-160-221.vinita.lt] has quit [Read error: 110 (Connection timed out)] 00:46:03 madnificent, cl-cont uses (declare (special ...)) a lot; if you want to see an alternative idea http://rondam.blogspot.com/2009/08/global-variables-done-right.html 00:46:22 -!- ve [n=a@94-193-95-252.zone7.bethere.co.uk] has quit [Read error: 110 (Connection timed out)] 00:47:17 alternatively see defglobal by nikodemus in sbcl 00:47:33 -!- c|mell [n=cmell@211.189.255.253] has quit ["to the embassy"] 00:48:10 ve [n=a@94-193-95-252.zone7.bethere.co.uk] has joined #lisp 00:48:56 -!- hypnosis [n=hypnosis@unaffiliated/hypnosis] has quit ["leaving"] 00:51:38 -!- slyrus_ [n=slyrus@adsl-75-36-222-205.dsl.pltn13.sbcglobal.net] has quit [Connection timed out] 01:04:32 -!- badeye [n=badeye@63-255-213-130.ip.mcleodusa.net] has quit [Client Quit] 01:05:57 -!- blandest [n=user@79.112.99.50] has quit [Read error: 60 (Operation timed out)] 01:09:10 minion: memo for seangrove: how's life? 01:09:10 Remembered. I'll tell seangrove when he/she/it next speaks. 01:09:11 seangrove, memo from seangrove: how's life? 01:09:20 minion: memo for seangrove: Not bad, thanks! 01:09:21 Remembered. I'll tell seangrove when he/she/it next speaks. 01:09:21 seangrove, memo from seangrove: Not bad, thanks! 01:09:46 minion: help 01:09:46 There are multiple help modules. Try ``/msg minion help kind'', where kind is one of: "lookups", "helping others", "adding terms", "aliasing terms", "forgetting", "memos", "avoiding memos", "nicknames", "goodies", "eliza", "advice", "apropos", "acronyms". 01:11:34 rnadom [n=user@72.170.156.242] has joined #lisp 01:16:45 -!- rnadom [n=user@72.170.156.242] has quit [Remote closed the connection] 01:17:29 c [n=user@72.170.156.242] has joined #lisp 01:17:57 -!- c is now known as Guest52122 01:19:38 slyrus_ [n=slyrus@adsl-75-36-222-205.dsl.pltn13.sbcglobal.net] has joined #lisp 01:20:17 -!- Guest52122 [n=user@72.170.156.242] has quit [Remote closed the connection] 01:24:35 QinGW [n=wangqing@203.86.89.226] has joined #lisp 01:24:58 -!- CrazyEddy [n=CrazyEdd@wrongplanet/CrazyEddy] has quit [Read error: 110 (Connection timed out)] 01:26:40 -!- tmh [n=thomas@pdpc/supporter/sustaining/tmh] has left #lisp 01:27:45 -!- slash_ [n=Unknown@p4FF0B24F.dip.t-dialin.net] has quit [Client Quit] 01:29:45 -!- billstclair [n=billstcl@unaffiliated/billstclair] has quit [Read error: 60 (Operation timed out)] 01:30:23 CrazyEddy [n=CrazyEdd@wrongplanet/CrazyEddy] has joined #lisp 01:33:06 -!- seangrove [n=user@173-11-104-25-SFBA.hfc.comcastbusiness.net] has quit [Read error: 113 (No route to host)] 01:36:38 hm. What's a clean way to create a 2d array and fill it with a bunch of different values (defined by what a function called repeatedly would return)? 01:37:35 -!- blackened` [n=blackene@ip-89-102-28-224.karneval.cz] has quit [] 01:39:03 hugod [n=hugod@bas1-montreal50-1279441929.dsl.bell.ca] has joined #lisp 01:39:15 sykopomp: make-array then loop and fill it 01:39:27 heh. 01:40:39 -!- stassats [n=stassats@wikipedia/stassats] has quit [Read error: 113 (No route to host)] 01:41:32 -!- ruediger [n=ruediger@91-115-30-128.adsl.highway.telekom.at] has quit ["Leaving"] 01:42:02 nyef [n=nyef@pool-71-161-71-17.cncdnh.east.myfairpoint.net] has joined #lisp 01:42:54 -!- salva [n=salva@105.11.117.91.dynamic.mundo-r.com] has quit [] 01:47:22 -!- minion [n=minion@common-lisp.net] has quit [Read error: 60 (Operation timed out)] 01:48:55 That's... ominous. 01:52:52 drewc: herep 01:54:22 herpes 01:56:00 ignas [n=ignas@ctv-79-132-160-221.vinita.lt] has joined #lisp 01:56:16 -!- ignas [n=ignas@ctv-79-132-160-221.vinita.lt] has quit [Client Quit] 02:01:01 -!- lisppaste [n=lisppast@common-lisp.net] has quit [Read error: 110 (Connection timed out)] 02:01:02 -!- specbot [n=specbot@common-lisp.net] has quit [Read error: 110 (Connection timed out)] 02:04:05 gko [n=gko@211.22.47.2] has joined #lisp 02:04:42 -!- jeti` [n=jeti@p548EFDFC.dip.t-dialin.net] has left #lisp 02:06:42 mrsolo [n=mrsolo@adsl-68-126-198-233.dsl.pltn13.pacbell.net] has joined #lisp 02:07:46 -!- ikki [n=ikki@189.247.82.198] has quit ["Leaving"] 02:08:19 salva [n=salva@105.11.117.91.dynamic.mundo-r.com] has joined #lisp 02:11:49 ikki [n=ikki@189.247.82.198] has joined #lisp 02:12:13 -!- quek [n=read_eva@router1.gpy1.ms246.net] has quit [Read error: 60 (Operation timed out)] 02:16:21 -!- bgs100 [n=ian@unaffiliated/bgs100] has quit [] 02:16:26 OmniMancer [n=OmniManc@111.69.240.226] has joined #lisp 02:17:57 emma [n=em@unaffiliated/emma] has joined #lisp 02:19:14 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [Read error: 104 (Connection reset by peer)] 02:20:07 coderdad [n=coderdad@ip72-200-214-240.ok.ok.cox.net] has joined #lisp 02:22:43 Adamant [n=Adamant@c-68-51-134-164.hsd1.ga.comcast.net] has joined #lisp 02:23:27 emmy [n=em@cpe-66-65-80-94.nyc.res.rr.com] has joined #lisp 02:23:57 -!- chavo_ [n=user@c-71-195-63-115.hsd1.mn.comcast.net] has quit ["ERC Version 5.2 (IRC client for Emacs)"] 02:30:40 -!- CrazyEddy [n=CrazyEdd@wrongplanet/CrazyEddy] has quit [Success] 02:31:49 -!- wlr [n=walt@c-65-96-92-150.hsd1.ma.comcast.net] has quit [Remote closed the connection] 02:34:28 wlr [n=walt@c-65-96-92-150.hsd1.ma.comcast.net] has joined #lisp 02:34:40 CrazyEddy [n=CrazyEdd@wrongplanet/CrazyEddy] has joined #lisp 02:35:56 -!- coderdad [n=coderdad@ip72-200-214-240.ok.ok.cox.net] has quit [Remote closed the connection] 02:39:35 -!- dreish [n=dreish@minus.dreish.org] has quit [] 02:40:35 -!- emma [n=em@unaffiliated/emma] has quit [Read error: 110 (Connection timed out)] 02:41:29 -!- saikat_ [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has quit [] 02:41:55 saikat_ [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has joined #lisp 02:42:33 Aszarsha_ [n=aszarsha@91-164-152-252.rev.libertysurf.net] has joined #lisp 02:44:06 Yay, my tiny address book works 02:44:07 halfway 02:44:11 02:45:47 -!- emmy [n=em@cpe-66-65-80-94.nyc.res.rr.com] has quit [Connection timed out] 02:45:52 -!- mrsolo [n=mrsolo@adsl-68-126-198-233.dsl.pltn13.pacbell.net] has quit [Read error: 104 (Connection reset by peer)] 02:46:04 -!- wlr [n=walt@c-65-96-92-150.hsd1.ma.comcast.net] has quit [Remote closed the connection] 02:50:04 chavo_ [n=user@c-71-195-63-115.hsd1.mn.comcast.net] has joined #lisp 02:51:52 -!- pr [n=pr@unaffiliated/pr] has quit ["leaving"] 02:59:01 -!- Aszarsha [n=aszarsha@91-164-134-169.rev.libertysurf.net] has quit [Read error: 101 (Network is unreachable)] 03:02:36 wlr [n=walt@c-65-96-92-150.hsd1.ma.comcast.net] has joined #lisp 03:03:44 -!- saikat_ [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has quit [Read error: 110 (Connection timed out)] 03:07:37 -!- Aszarsha_ is now known as Aszarsha 03:08:04 -!- Aszarsha is now known as Aszarsha_ 03:08:25 -!- Aszarsha_ is now known as Aszarsha 03:08:35 -!- marioxcc is now known as marioxcc-ocupado 03:11:10 -!- chavo_ [n=user@c-71-195-63-115.hsd1.mn.comcast.net] has quit [Remote closed the connection] 03:11:51 fusss [n=chatzill@60-241-1-206.static.tpgi.com.au] has joined #lisp 03:13:03 legumbre_ [n=user@r190-135-14-34.dialup.adsl.anteldata.net.uy] has joined #lisp 03:13:14 -!- Fade [i=fade@outrider.deepsky.com] has quit [Remote closed the connection] 03:13:29 Cowmoo [n=Cowmoo@c-68-55-111-252.hsd1.va.comcast.net] has joined #lisp 03:14:46 -!- lmr [n=lemurian@unaffiliated/lemurian] has quit [Client Quit] 03:14:47 seangrove [n=user@98.248.37.94] has joined #lisp 03:16:40 -!- ehu [n=chatzill@82-170-33-173.ip.telfort.nl] has quit [Read error: 110 (Connection timed out)] 03:17:24 nixeagle [n=nixeagle@Wikimedia/Nixeagle] has joined #lisp 03:17:42 ehu [n=chatzill@82-170-33-173.ip.telfort.nl] has joined #lisp 03:18:27 wingo [n=wingo@cpe-98-154-52-199.socal.res.rr.com] has joined #lisp 03:20:08 -!- Dawgmatix [n=dman@c-68-32-44-191.hsd1.nj.comcast.net] has quit ["Ex-Chat"] 03:25:34 male_terran [n=KVIrc@ip98-162-161-191.pn.at.cox.net] has joined #lisp 03:26:18 -!- Urfin [i=foobar@85.65.93.74.dynamic.barak-online.net] has quit [Read error: 113 (No route to host)] 03:26:23 rullie [n=rullie@bas4-toronto47-1176152692.dsl.bell.ca] has joined #lisp 03:28:17 QinGW1 [n=wangqing@203.86.89.226] has joined #lisp 03:30:40 -!- legumbre [n=user@r190-135-35-79.dialup.adsl.anteldata.net.uy] has quit [Read error: 110 (Connection timed out)] 03:33:25 -!- ikki [n=ikki@189.247.82.198] has quit ["Leaving"] 03:33:55 dralston [n=dralston@S010600212986cca8.va.shawcable.net] has joined #lisp 03:34:34 coderdad [n=coderdad@ip72-200-214-240.ok.ok.cox.net] has joined #lisp 03:36:42 emacs-dwim [n=user@cpe-74-71-11-230.twcny.res.rr.com] has joined #lisp 03:45:15 -!- QinGW [n=wangqing@203.86.89.226] has quit [Success] 03:45:40 -!- OmniMancer [n=OmniManc@111.69.240.226] has quit [Read error: 113 (No route to host)] 03:50:15 -!- male_terran [n=KVIrc@ip98-162-161-191.pn.at.cox.net] has quit ["Changing server..."] 03:52:26 male_terran [n=KVIrc@ip98-162-161-191.pn.at.cox.net] has joined #lisp 03:54:13 Good morning! 03:54:35 Vonunov: congratulations! 03:54:38 good evening! 03:55:04 eno__ [n=eno@70.137.152.199] has joined #lisp 03:55:14 Eek. It's about five minutes to midnight. 03:55:20 Hello beach. 03:55:42 -!- eno [n=eno@nslu2-linux/eno] has quit [Read error: 131 (Connection reset by peer)] 03:56:30 QinGW [n=wangqing@203.86.89.226] has joined #lisp 03:59:11 lpolzer [n=lpolzer@dslb-088-073-207-020.pools.arcor-ip.net] has joined #lisp 04:13:04 -!- QinGW1 [n=wangqing@203.86.89.226] has quit [Connection timed out] 04:14:32 -!- lpolzer_ [n=lpolzer@dslb-088-073-201-082.pools.arcor-ip.net] has quit [Read error: 110 (Connection timed out)] 04:20:54 what search engine does cliki use? 04:21:10 actually, can't we get details on the cliki implementation? 04:21:41 traffic details would be nice too (along with a pony, of course) 04:22:11 -!- davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 04:24:23 Cliki is Common Lisp wiki software. I believe, for search, Cliki uses Cliki. The precise implementation, I do not know. 04:24:29 However, I think it's open source. 04:24:32 -!- marioxcc-ocupado is now known as marioxcc 04:25:00 http://www.cliki.net/CLiki 04:25:07 QinGW1 [n=wangqing@203.86.89.226] has joined #lisp 04:25:20 I wouldn't bet on any public source release being up-to-date. 04:25:40 And I'm not sure if cliki is still running the original software, or if it's now running drewc's reimplementation. 04:25:43 Yeah, looking at it, the source seems to have gone AWOL. 04:26:15 You're honestly not missing anything. 04:27:05 dan?b has done some amazing work, but some of the things he's written, cliki included, have been absolute hacks. 04:29:03 DrunkTomato [n=DEDULO@ext-gw.wellcom.tomsk.ru] has joined #lisp 04:30:01 schemer999 [n=tmilford@76.90.137.46] has joined #lisp 04:30:02 Demosthenes [n=demo@64.208.221.2] has joined #lisp 04:31:35 -!- QinGW [n=wangqing@203.86.89.226] has quit [Read error: 145 (Connection timed out)] 04:33:56 -!- mattrepl [n=mattrepl@ip68-100-82-124.dc.dc.cox.net] has quit [] 04:38:38 -!- coderdad [n=coderdad@ip72-200-214-240.ok.ok.cox.net] has quit [Remote closed the connection] 04:40:24 poor drewc fell for the reddit worm 04:40:52 ericholscher [n=eric@ericholscher.com] has joined #lisp 04:45:45 Okay, I'm off to bed before my head explodes from trying to figure out condition variables. 04:45:48 -!- nyef [n=nyef@pool-71-161-71-17.cncdnh.east.myfairpoint.net] has quit ["G'night all."] 04:46:55 -!- cpc26 [n=cpc26@72.170.156.242] has quit [Read error: 104 (Connection reset by peer)] 04:46:57 -!- TDT [n=dthole@254.91.248.216.dyn.southslope.net] has quit [] 04:47:01 -!- marioxcc [n=user@200.92.168.65] has quit ["http://putolinux.wordpress.com/2009/05/09/como-instalar-el-puto-nucleo-linux-libre-en-debian-gnulinux-5-0/ <= poniendo linux-] 04:47:22 cpc26 [n=cpc26@72.170.156.242] has joined #lisp 04:48:23 -!- dstatyvka [i=ejabberd@pepelaz.jabber.od.ua] has left #lisp 04:57:03 wow, hard to tell if this xah lee person is a flaming troll or a useful howto writer :P 04:57:50 Demosthenes: Ah, you don't know, do you? 04:57:54 -!- redblue [i=star@ppp106.108-253-207.mtl.mt.videotron.ca] has quit [Read error: 110 (Connection timed out)] 04:58:08 Demo: A suitable answer might be 'both, at various times.' 04:58:35 ? 04:59:00 Put him down for "well intentioned troll". 04:59:41 -!- dralston [n=dralston@S010600212986cca8.va.shawcable.net] has quit ["Leaving"] 05:00:43 -!- xristos [n=x@research.suspicious.org] has quit [Read error: 110 (Connection timed out)] 05:01:35 Makoryu: something specific? 05:02:08 -!- vishsingh [n=vsingh@CPE0014bf4af6c2-CM000a739caee2.cpe.net.cable.rogers.com] has quit [] 05:06:01 I absolutely do not understand how package and asdf work together or relate to each other. I've read everything I can find on this, but I need a real (but simple) working example, not pseudo-code. 05:06:47 You might be confusing packages with systems or modules. 05:07:01 A package is just a namespace for mapping strings to symbols. 05:13:16 lexa_ [n=lexa_@seonet.ru] has joined #lisp 05:13:38 -!- lexa_ is now known as Guest63076 05:13:47 Zhivago, I understand the difference. I just can't get it to work. 05:13:53 drafael [n=tapio@ip-118-90-129-87.xdsl.xnet.co.nz] has joined #lisp 05:16:19 -!- fusss [n=chatzill@60-241-1-206.static.tpgi.com.au] has quit [Read error: 104 (Connection reset by peer)] 05:18:53 -!- ski [n=slj@c-2111e055.1149-1-64736c10.cust.bredbandsbolaget.se] has quit [Read error: 110 (Connection timed out)] 05:20:48 -!- coyo[asleep] is now known as coyo 05:22:41 hey guise, i do not understand PUSH 05:22:47 how does PUSH work? 05:23:08 i do not understand what it means by "place" 05:23:23 Do you understand pointers in C? 05:23:38 i have an idea of the concept 05:23:43 it's a reference 05:23:52 ta-da! 05:24:19 in c, it's a reference to a place in memory 05:24:22 In C *p is place that you can store a value. Lisp doesn't have pointers, but it does have a similar idea. 05:24:29 and haunts the nightmares of many 05:24:43 okay 05:24:55 (first x) is a place where we can store a value -- (setf (first x) 1) 05:25:03 ah 05:25:07 -!- ASau [n=user@83.69.227.32] has quit ["off"] 05:25:13 i heard push uses setf inside it 05:25:33 (push 1 (first x)) will use (first x) as a place, too. 05:26:05 (macroexpand '(push 1 x)) might be educational. 05:26:15 so what is returned by (first x) is not actually a value, but a reference? 05:26:52 coyo: Neither. In this case we go and look up "how to store a value in somewhere referred to via first". 05:26:53 -!- a-s [n=user@nat-240.ro.66.com] has quit [Remote closed the connection] 05:27:07 coyo: This will be some code like replca. 05:27:48 coyo: setf will then arrange for this to be called properly. 05:28:02 Davidbrcz [n=david@212-198-78-230.rev.numericable.fr] has joined #lisp 05:28:24 coyo: So there is no first class pointer or reference -- just a way of finding out how to get a suitable mutator. 05:29:34 um, i cant paste my code, because "proxy error, cannot connect to upstream server, connection refused" 05:29:40 Fade [i=fade@outrider.deepsky.com] has joined #lisp 05:30:11 -!- Guest63076 [n=lexa_@seonet.ru] has left #lisp 05:31:13 i'll just use another paste :( 05:32:21 mrSpec [n=Spec@unaffiliated/mrspec] has joined #lisp 05:32:44 hello 05:33:14 coyo: Gist rarely goes down 05:33:20 http://gist.github.com/ 05:33:50 redblue [i=star@ppp013.108-253-207.mtl.mt.videotron.ca] has joined #lisp 05:35:29 http://paste.pocoo.org/show/141855/ 05:37:00 SandGorgon [n=OmNomNom@122.162.3.112] has joined #lisp 05:38:40 KingNatoG5_ [n=patrik@84-217-6-126.tn.glocalnet.net] has joined #lisp 05:39:18 Quadrescence [n=quad@unaffiliated/quadrescence] has joined #lisp 05:39:36 #lisp pop quiz!! Whoever answers first correctly, WINS!! Who is this man: http://img32.imageshack.us/img32/6492/lololfp.jpg 05:42:56 -!- pixel5 [n=pixel@copei.de] has quit ["Lost terminal"] 05:43:03 I have hin on my front door. 05:43:32 -!- nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has quit [Read error: 104 (Connection reset by peer)] 05:43:45 nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 05:47:21 -!- KingNatoG5 [n=patrik@84-217-7-130.tn.glocalnet.net] has quit [Read error: 110 (Connection timed out)] 05:47:21 -!- KingNatoG5_ is now known as KingNatoG5 05:50:24 -!- wingo [n=wingo@cpe-98-154-52-199.socal.res.rr.com] has quit [Read error: 60 (Operation timed out)] 05:51:53 OmniMancer [n=OmniManc@222-154-179-43.jetstream.xtra.co.nz] has joined #lisp 05:52:29 huangjs [n=user@192.51.54.129] has joined #lisp 05:54:19 -!- kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 05:54:42 -!- Davidbrcz [n=david@212-198-78-230.rev.numericable.fr] has quit [Read error: 110 (Connection timed out)] 05:54:42 Hey all, I can't quite understand why (require :sb-md5) works, but not (defpackage #:qdsl (:use #:cl :sd-md5)) - it tells me that :sb-md5 doesn't refer to a package 05:55:10 -!- SandGorgon [n=OmNomNom@122.162.3.112] has quit [Read error: 104 (Connection reset by peer)] 05:55:15 Or rather, "The name "SB-MD5" does not designate any package." 05:56:06 -!- fnordus [n=dnall@70.70.0.215] has quit [Read error: 104 (Connection reset by peer)] 05:56:30 fnordus [n=dnall@70.70.0.215] has joined #lisp 05:58:32 fusss [n=chatzill@60-241-1-206.static.tpgi.com.au] has joined #lisp 05:58:40 Ah, I think it's a problem in my .asd 06:01:00 what is a .asd, anyway? 06:01:04 *coyo* looks curious 06:01:18 system definition file 06:01:33 o.o 06:01:36 what does it do? 06:02:14 how is it different from a .lisp ? 06:02:30 -!- xinming [n=hyy@218.73.137.216] has quit [Read error: 110 (Connection timed out)] 06:04:45 coyo, it defines the order .lisp files are to be compiled. 06:04:49 -!- peddie [n=peddie@c-67-169-146-99.hsd1.ca.comcast.net] has quit [Read error: 110 (Connection timed out)] 06:04:55 tcr [n=tcr@host146.natpool.mwn.de] has joined #lisp 06:05:25 -!- Cowmoo [n=Cowmoo@c-68-55-111-252.hsd1.va.comcast.net] has left #lisp 06:05:54 lat: thanks :D 06:07:29 coyo, that is about all I know about it. I haven't been able to get it to work for me. 06:07:54 awww 06:08:02 *coyo* cuddles the lat sympathetically 06:09:19 :>) 06:09:50 -!- wlr [n=walt@c-65-96-92-150.hsd1.ma.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 06:09:59 wlr [n=walt@c-65-96-92-150.hsd1.ma.comcast.net] has joined #lisp 06:11:42 freiksenet [n=freiksen@193.166.13.253] has joined #lisp 06:11:45 SandGorgon [n=OmNomNom@122.162.50.31] has joined #lisp 06:12:01 ski [n=slj@c-c110e055.1149-1-64736c10.cust.bredbandsbolaget.se] has joined #lisp 06:13:00 Hello! I have sort of a matrix data-type - a list of lists. Is it possible to iterate over it (using iterate?) to make a list of (column-number row-number value)? 06:14:04 I tried (iter (for row in matrix) (for cell in row) (collect cell)) but it seems that iteration does not work over stuff created inside of iterate %) 06:15:38 dmelani [n=dmelani@c83-253-52-86.bredband.comhem.se] has joined #lisp 06:17:31 Calling an asdf expert. http://pastebin.com/d236ce923 06:17:54 -!- rstandy [n=rastandy@net-93-144-187-130.t2.dsl.vodafone.it] has quit [Read error: 110 (Connection timed out)] 06:21:15 -!- xan-afk_ is now known as xan 06:31:22 daniel [i=daniel@unaffiliated/daniel] has joined #lisp 06:32:28 mishoo [n=mishoo@79.112.50.83] has joined #lisp 06:32:52 lexa_ [n=lexa_@seonet.ru] has joined #lisp 06:33:10 -!- dv_ [n=dv@85-127-111-103.dynamic.xdsl-line.inode.at] has quit ["Verlassend"] 06:33:20 -!- lexa_ is now known as Guest24560 06:34:22 is there a ceiling function that returns the higher of two numbers? 06:34:54 -!- abeaumont [n=abeaumon@84.76.48.250] has quit [Read error: 110 (Connection timed out)] 06:35:03 -!- Guest24560 [n=lexa_@seonet.ru] has left #lisp 06:35:42 clhs max 06:36:06 gah, no specbot? 06:36:26 coyo: the max function takes an arbitrary number of arguments, and returns the maximum. 06:37:47 coyo: ceiling, on the other hand, does something totally different. 06:38:08 okay 06:38:16 max was what i was looking for 06:38:21 thank you very much <3 06:38:23 *coyo* hugs 06:39:35 freiksenet: I don't know iter, but I would guess you would have to nest, i.e., (iter (for row in matrix) ... (iter (for cell in row) ...)) 06:40:54 freiksenet: with loop it would be something like: (loop for row in matrix for r from 0 append (loop for cell in row for c from 0 collect (list r c cell))) 06:41:18 http://l1sp.org/cl/max works too. 06:41:29 oh, right 06:41:32 thanks! 06:41:44 ASau [n=user@host183-230-msk.microtest.ru] has joined #lisp 06:47:20 mvilleneuve [n=mvillene@LLagny-156-36-4-214.w80-14.abo.wanadoo.fr] has joined #lisp 06:47:37 good morning 06:47:38 hello mvilleneuve 06:50:54 Tordek_ [n=tordek@host162.200-82-15.telecom.net.ar] has joined #lisp 06:51:55 *beach* is leaving in a few minutes to go pick up his repaired home computer, and will probably spend the rest of the day installing it. 06:53:24 beach: thanks, it works :) 06:53:35 freiksenet: no problem 06:58:33 HET2 [n=diman@cpc1-cdif12-2-0-cust125.5-1.cable.virginmedia.com] has joined #lisp 06:58:40 What is wrong with this? http://pastebin.com/d236ce923 07:02:44 -!- tcr [n=tcr@host146.natpool.mwn.de] has quit ["Leaving."] 07:03:26 there is a substitute function to return list with replaced elements. is there such function for replacing items at index? 07:03:40 -!- coyo is now known as coyo[sleeping] 07:03:41 -!- Tordek [n=tordek@host96.200-82-21.telecom.net.ar] has quit [Read error: 110 (Connection timed out)] 07:03:52 freiksenet: (setf (elt *my-list* index) new-value) :) 07:04:04 sykopomp: with no side-effects 07:04:53 (setf (elt (copy-list *my-list* index) index) new-value) >:3 07:05:20 attila_lendvai [n=ati@catv-89-134-66-153.catv.broadband.hu] has joined #lisp 07:05:35 sykopomp: this is ugly cause I will have to then return new-value :( 07:05:36 rstandy [n=rastandy@212.189.140.32] has joined #lisp 07:05:47 pixel5 [n=pixel@copei.de] has joined #lisp 07:05:54 as setf does not return modified list 07:06:17 (let ((new-list (copy-list *my-list*))) (setf (elt new-list index) new-value) new-list) c'mon 07:06:23 GIMME SOME MO' 07:07:32 sykopomp: lol :) thanks, seems there is no such function. 07:07:38 jdz [n=jdz@85.254.193.149] has joined #lisp 07:08:01 what do you mean there's no such function? 07:08:07 oh, to do that. 07:08:07 -!- antoszka [n=antoszka@unaffiliated/antoszka] has quit ["+++ killed by SIGSEGV +++"] 07:08:18 free_thinker [n=willijar@134.151.144.246] has joined #lisp 07:09:54 abeaumont [n=abeaumon@85.48.202.13] has joined #lisp 07:10:50 -!- kwertii [n=kwertii@67.180.202.175] has quit ["bye"] 07:11:59 Hun [n=hun@p50993726.dip0.t-ipconnect.de] has joined #lisp 07:15:41 -!- gigamonkey [n=user@adsl-99-58-29-142.dsl.pltn13.sbcglobal.net] has quit [Read error: 110 (Connection timed out)] 07:16:43 licoresse [n=peropaal@cm-84.209.69.164.getinternet.no] has joined #lisp 07:19:19 -!- SandGorgon [n=OmNomNom@122.162.50.31] has quit [Read error: 54 (Connection reset by peer)] 07:28:32 there's something perverse about running ABCL, and it feels fantastic :) 07:29:41 -!- BrianRice [n=briantri@c-76-115-44-87.hsd1.or.comcast.net] has quit [] 07:32:13 -!- attila_lendvai [n=ati@catv-89-134-66-153.catv.broadband.hu] has quit ["..."] 07:34:00 -!- ltriant [n=ltriant@lithium.mailguard.com.au] has quit ["leaving"] 07:34:12 -!- DrunkTomato [n=DEDULO@ext-gw.wellcom.tomsk.ru] has quit [] 07:35:07 attila_lendvai [n=ati@catv-89-134-66-153.catv.broadband.hu] has joined #lisp 07:38:42 saikat_ [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has joined #lisp 07:41:58 kingless [n=user@adsl-144-174-240.rmo.bellsouth.net] has joined #lisp 07:45:32 ljames [n=ln@unaffiliated/ljames] has joined #lisp 07:47:28 -!- benny [n=benny@87.122.15.255] has quit [Read error: 110 (Connection timed out)] 07:49:49 varjag [n=eugene@122.62-97-226.bkkb.no] has joined #lisp 07:49:54 -!- beach [n=user@ABordeaux-158-1-12-86.w90-50.abo.wanadoo.fr] has left #lisp 07:51:04 -!- mishoo [n=mishoo@79.112.50.83] has quit [Read error: 113 (No route to host)] 07:52:27 mishoo [n=mishoo@79.112.50.83] has joined #lisp 07:52:55 fiveop [n=fiveop@g229085177.adsl.alicedsl.de] has joined #lisp 07:54:38 -!- HET2 [n=diman@cpc1-cdif12-2-0-cust125.5-1.cable.virginmedia.com] has quit ["This computer has gone to sleep"] 07:54:47 -!- lat [n=lat@125.167.140.159] has quit [Read error: 110 (Connection timed out)] 07:55:58 lat [n=lat@125.167.140.159] has joined #lisp 07:56:57 Joreji [n=thomas@134.61.43.159] has joined #lisp 07:59:43 -!- freiksenet [n=freiksen@193.166.13.253] has quit [Remote closed the connection] 08:01:25 emma [n=em@unaffiliated/emma] has joined #lisp 08:03:41 stassats [n=stassats@wikipedia/stassats] has joined #lisp 08:03:54 dv_ [n=dv@83-64-248-68.inzersdorf.xdsl-line.inode.at] has joined #lisp 08:12:05 -!- younder [n=jpthing@113.14.202.84.customer.cdi.no] has quit [Read error: 60 (Operation timed out)] 08:16:49 ejs [n=eugen@77.222.151.102] has joined #lisp 08:20:10 -!- QinGW1 [n=wangqing@203.86.89.226] has quit [Read error: 131 (Connection reset by peer)] 08:21:23 -!- araujo [n=araujo@gentoo/developer/araujo] has quit ["Leaving"] 08:21:27 QinGW [n=wangqing@203.86.89.226] has joined #lisp 08:22:45 jsoft [n=user@unaffiliated/jsoft] has joined #lisp 08:26:01 ejs1 [n=eugen@nat.ironport.com] has joined #lisp 08:28:32 SimonH [n=SimonH@78.148.132.114] has joined #lisp 08:30:00 nowhere_man [i=c1fb3fe8@gateway/web/freenode/x-cbnkbmxzxffunjbq] has joined #lisp 08:32:37 Soulman__ [n=kae@Gatekeeper.vizrt.com] has joined #lisp 08:32:45 -!- SimonH [n=SimonH@78.148.132.114] has left #lisp 08:32:51 -!- Soulman [n=kae@Gatekeeper.vizrt.com] has quit [Read error: 104 (Connection reset by peer)] 08:33:39 -!- ejs [n=eugen@77.222.151.102] has quit [Read error: 110 (Connection timed out)] 08:37:35 -!- ehu [n=chatzill@82-170-33-173.ip.telfort.nl] has quit [Read error: 110 (Connection timed out)] 08:37:43 icecube [n=icecube@p549C3509.dip.t-dialin.net] has joined #lisp 08:37:44 -!- nowhere_man [i=c1fb3fe8@gateway/web/freenode/x-cbnkbmxzxffunjbq] has quit [Ping timeout: 180 seconds] 08:38:11 -!- icecube is now known as Guest47821 08:38:19 trebor_dki [n=user@130.83.156.186] has joined #lisp 08:39:04 thehcdreamer [n=thehcdre@94.198.78.26] has joined #lisp 08:39:47 HET2 [n=diman@w283.engin.cf.ac.uk] has joined #lisp 08:39:48 -!- HET2 [n=diman@w283.engin.cf.ac.uk] has quit [Remote closed the connection] 08:40:45 blandest [n=user@softhouse.is.ew.ro] has joined #lisp 08:41:54 -!- Guest47821 [n=icecube@p549C3509.dip.t-dialin.net] has quit [Client Quit] 08:42:06 icecube [n=icecube@p549C3509.dip.t-dialin.net] has joined #lisp 08:42:13 -!- icecube is now known as icecub_ 08:42:20 -!- trebor_dki [n=user@130.83.156.186] has left #lisp 08:42:36 asd123123 [n=Unregist@183.66.50.60.brf01-home.tm.net.my] has joined #lisp 08:42:59 trebor_dki [n=user@mail.dki.tu-darmstadt.de] has joined #lisp 08:44:05 http://17fc5c36.linkbucks.com 08:44:20 -!- asd123123 [n=Unregist@183.66.50.60.brf01-home.tm.net.my] has quit [Client Quit] 08:45:28 SimonH [n=simonh@78.148.132.114] has joined #lisp 08:45:39 aerique [i=euqirea@xs2.xs4all.nl] has joined #lisp 08:49:30 SandGorgon [n=OmNomNom@122.162.50.31] has joined #lisp 08:50:42 QinGW1 [n=wangqing@203.86.89.226] has joined #lisp 08:52:04 -!- blandest [n=user@softhouse.is.ew.ro] has quit [Remote closed the connection] 08:53:40 SimonH_ [i=4e948472@gateway/web/freenode/x-qpbtrwlcmplxbltx] has joined #lisp 08:54:00 -!- SimonH_ [i=4e948472@gateway/web/freenode/x-qpbtrwlcmplxbltx] has quit [Client Quit] 08:54:28 -!- ChanServ has set mode +o antifuchs 08:56:06 -!- ia [n=ia@89.169.161.244] has quit [Read error: 60 (Operation timed out)] 08:57:15 drafael1 [n=tapio@118.90.132.209] has joined #lisp 08:58:47 -!- ve [n=a@94-193-95-252.zone7.bethere.co.uk] has quit [Read error: 113 (No route to host)] 09:02:04 QinGW2 [n=wangqing@203.86.89.226] has joined #lisp 09:04:40 phao [n=phao@189.13.223.29] has joined #lisp 09:05:28 -!- QinGW1 [n=wangqing@203.86.89.226] has quit [Read error: 60 (Operation timed out)] 09:05:30 what is the Practical Common Lisp's writer nick? 09:05:36 ops 09:05:46 phao: gigamonkey 09:06:00 sykopomp, thank you 09:06:02 simplechat [n=simple@unaffiliated/simplechat] has joined #lisp 09:06:08 That book is amazing 09:06:42 it's dead sexy 09:06:54 "officially" dead sexy... 09:07:08 But between you and me, indeed it's amazing. 09:07:11 -!- QinGW [n=wangqing@203.86.89.226] has quit [Read error: 110 (Connection timed out)] 09:09:11 ia [n=ia@89.169.161.244] has joined #lisp 09:12:51 -!- thehcdreamer [n=thehcdre@94.198.78.26] has quit [] 09:13:12 -!- drafael [n=tapio@ip-118-90-129-87.xdsl.xnet.co.nz] has quit [Read error: 110 (Connection timed out)] 09:18:17 fusss_ [n=chatzill@60-241-1-206.static.tpgi.com.au] has joined #lisp 09:18:17 -!- fusss [n=chatzill@60-241-1-206.static.tpgi.com.au] has quit [Read error: 104 (Connection reset by peer)] 09:18:19 -!- fusss_ is now known as fusss 09:19:05 sykopomp: what are you using abcl for? 09:19:42 ve [n=a@94-193-95-252.zone7.bethere.co.uk] has joined #lisp 09:19:59 fusss: at this point, nothing. I was tempted to try running something on the android sdk 09:20:17 but abcl seems pretty god-awful slow, and I'm not sure it would be all that practical to use. I might be wrong. 09:20:37 get the Java ME SDK and see if you can run abcl on the mobile phone emulator 09:20:52 does it generate standlone binaries? 09:21:12 I'm not even sure. Documentation was confusing, and I didn't care enough to look too deeply. 09:21:27 I just wanted to screw around with it, but it sounds like you're stuck writing some java code no matter what you do... 09:21:38 (I might be wrong. It would be very nice to be wrong.) 09:22:02 -!- phao [n=phao@189.13.223.29] has left #lisp 09:22:12 -!- saikat_ [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has quit [] 09:23:34 fusss: considering how much of the android documentation stresses the need to write optimized, lightweight java code, it sounds like trying to run abcl would be a very bad idea. 09:23:55 oh, ok 09:24:22 it almost tempts me to use clojure, but just reading clojure makes my eyes bleed. 09:24:55 couldn't one write clojure in more idiomatic ways? 09:24:58 I cannot understand why he didn't extend CL (or a subset) instead of writing something with an entirely different syntax... 09:25:06 it doesn't have to suck 09:25:19 clojure isn't CL 09:25:25 That's the point. 09:25:33 yes 09:25:48 anyways, it's the whole "CL is too broken to try to fix" mindset, which I find absurd. 09:25:52 We had a big effort around 1984 to define a common lisp, because having all these variant was a PITA. 09:26:02 Why do we create new variants again? 09:26:03 sykopomp: not CL, let's call it the MacLisp family of programming languages. (Mac|LML|Zeta|Spice|Common|ISO|)Lisp. 09:26:38 fusss: clojure isn't CL, and you shouldn't try to write it as if it were. 09:26:39 -!- lat [n=lat@125.167.140.159] has quit ["Ex-Chat"] 09:26:56 I can't write CL as if it was clojure! 09:27:05 I do can write CL as if it was clojure! I mean. 09:27:06 sykopomp: clojure is no scheme either; it's just unidiomatic lisp code through and through, just like Arc and NewLisp 09:27:18 man, no kidding. 09:27:18 matimago: but it won't work as if it were (: 09:27:31 I could write libraries to make it work similarly. 09:27:37 then go ahead (: 09:27:41 (And even betterm aref is O(1), not O(log(N)). 09:27:48 I've been trying to translate rhickey's ants demo to using ChanL, and it reaches epic levels of unreadability. 09:28:08 not to mention, there's some apparently retarded constructs, and I don't understand why he does it. 09:28:20 like what? *curious* 09:28:34 for example, there's this shit: {0 foo 1 bar 2 baz 3 quux} <-- THIS IS RETARDED 09:28:51 why so? 09:28:54 why the fuck is he using a map, and manually writing out indices? 09:28:56 seriously 09:29:00 Oh, like that. 09:29:02 he has vectors 09:29:08 why not [foo bar baz quux]? 09:29:11 Thought you meant the map syntax itself. 09:29:11 same fucking effect 09:29:29 I find {} to be retarded, because you really don't need literal syntax. 09:29:41 hickey might be a great hacker, but man, aesthetics! 09:29:45 well. that shouldn't make things reach epic levels of unreadability, though. 09:29:47 (make-map 0 foo 1 bar 2 baz 3 quux) is the same shit. 09:29:50 sykopomp, what else is there? 09:29:59 tic: functions. 09:30:01 agreed on the make-map. 09:30:03 you know, instead of reader macros? :) 09:30:19 but the syntax isn't my rant at all. 09:30:22 that's a matter of taste 09:30:23 sykopomp, yes yes. but I wonder what's causing the headaches. that was one example. 09:30:38 what I'm wondering is why the fuck he uses a hash table to pretend he has indices, when he already has them. 09:30:47 it's not a clojure rant. It's more a wtf moment I had. 09:31:07 well.. still nothing causing epic levels of unreadability. stil waiting for it. :) 09:31:20 oh, you're asking for that. 09:31:23 Yes. 09:31:31 -!- cpc26 [n=cpc26@72.170.156.242] has quit [Read error: 54 (Connection reset by peer)] 09:31:35 let me crawl through the ants demo again and see if I find anything good. 09:31:38 thanks! 09:31:46 tic: oh come on 09:31:50 fusss, ? 09:31:52 I'm trying to find something that wasn't just a matter of me going "oh, that's different" 09:31:54 loop was a good example 09:31:57 that's just sophistry 09:32:15 you're allowed to bitch about something you don't fully grok just by the look of it 09:32:41 i don't know clojure and i can't stand its syntax :-) 09:32:43 sophistry is a word I'm not yet familiar with! 09:32:47 Heh. 09:33:20 I'll buy that it's Different(tm), but I have a hard time imagining there actually being language constructs that simply don't translate into equivalent CL code 09:33:30 (([move #(turn % -1) #(turn % 1)] 09:33:30 (wrand [(if (:ant @ahead) 0 (ranks ahead)) 09:33:30 (ranks ahead-left) (ranks ahead-right)])) 09:33:33 eyeburn right there. 09:33:44 it feels like someone read the first two chapters of SICP and the original Erlang papers each, then went to town with NetBeans 09:33:53 fusss, :)) 09:34:25 tic: there's some constructs in clojure that are a little bit of a pain in the ass to use in CL 09:34:27 sykopomp, (([move ... whatnow? 09:34:37 tic: exactly 09:34:40 he has a love for reader macros. what's # and @? 09:34:46 that's literal lambda syntax 09:35:02 #(turn % -1) => (lambda (_) (turn _ -1)) 09:35:10 in other words, rcurry 09:35:17 cpc26 [n=cpc26@72.170.156.242] has joined #lisp 09:35:19 the @, I still have no idea about. 09:35:24 *fusss* thinks hickey should be upstaged by writing a source to source translator that takes out a sane lisp and spits out clojure. 09:35:32 the really painful part about these reader macros is that they're incredibly hard to search for in the documentation. 09:35:38 I can't just look up {} (I tried) 09:35:43 haha. :-) 09:36:06 I sort of knew {} was for maps, but I couldn't believe it when I ran into his cute little indexing example. 09:36:31 I also am having mixed feelings about clojure's destructuring lambda lists. 09:37:20 -!- nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has quit [] 09:37:32 the [] stuff? 09:37:41 er. no. 09:37:56 no, the part where you take a lambda-list and do destructuring-bind within it. 09:38:53 like ((lambda ((x y) &rest foo) (append (list x y) foo)) (list 1 2) 3 4 5) 09:39:02 xinming [n=hyy@125.109.254.35] has joined #lisp 09:41:17 the -> is particularly cute, too. 09:41:47 piping with macros (in a functional language!) for fun and profit. 09:47:12 Yuuhi [n=user@p5483C84A.dip.t-dialin.net] has joined #lisp 09:47:19 It is called o usually... 09:47:36 (define fog (o f g)) 09:47:52 -!- schemer999 [n=tmilford@76.90.137.46] has quit [] 09:49:22 no, it's not just function composition 09:49:32 because it doesn't take functions as arguments 09:49:40 it has another function for that (comp, I think) 09:49:52 -> actually crawls code and rewrites it so it pipes. 09:50:36 -!- SimonH [n=simonh@78.148.132.114] has left #lisp 09:50:55 Well, it exchanges the order of the arguments, but otherwise that's the same as o. 09:51:00 -!- Fade [i=fade@outrider.deepsky.com] has quit [Read error: 110 (Connection timed out)] 09:51:15 (-> x h g f) == (o f g h x) 09:51:31 nope 09:51:33 no 09:51:35 (considering that x is a function returning x). 09:51:43 o is compose. 09:51:46 Perhaps we're not talking about the same -> 09:51:49 correct? 09:51:58 (-> x h g f) == (f (g (h x))) 09:52:06 I'm talking about clojure's -> 09:52:10 (o f g h x) == (f (g (h x))) 09:52:17 yes, that is not correct. 09:52:27 that's composition 09:52:43 -> is a macro that rewrites code so that it sort of has composition, but it's not a higher order function (like compose) 09:53:14 Well (o f g h) is a function, but (o f g h x) is a short hand for ((o f g h) x) which is not possible to write in CL. 09:53:42 matimago: alexandria:compose :) 09:54:13 matimago: and that would depend on what types f g and h have, right? 09:54:18 (funcall (compose 'f 'g 'h) x) 09:54:33 uglyp 09:54:37 nil 09:54:39 (defun append-concat (separator &rest strings) 09:54:41 (apply #'concatenate 'string (mapcar (lambda (str) 09:54:41 matimago: it might as well be shorthand for ((o f g) h x)? 09:54:42 SimonH [i=c0abc49f@gateway/web/freenode/x-pbqotqgkpghpqpes] has joined #lisp 09:54:42 (concatenate 'string str separator)) 09:54:44 strings))) 09:55:03 sorry 09:55:04 -!- rstandy [n=rastandy@212.189.140.32] has quit [Read error: 145 (Connection timed out)] 09:55:33 mrsolo [n=mrsolo@adsl-68-126-198-233.dsl.pltn13.pacbell.net] has joined #lisp 09:55:35 Well, you're right there's syntaxtic differences. 09:56:01 -!- SimonH [i=c0abc49f@gateway/web/freenode/x-pbqotqgkpghpqpes] has quit [Client Quit] 09:56:25 fusss: you can do that with format much more concisely. 09:56:35 -!- Aszarsha [n=aszarsha@91-164-152-252.rev.libertysurf.net] has quit [Read error: 113 (No route to host)] 09:56:53 sykopomp: no can format or loop 09:57:00 i am big-macro impaired 09:57:09 princ-to-string is my best friend 09:57:19 let me check one thing and I can give you a pretty simple format.... 09:57:33 isnt format slow anyway? 09:57:47 not necessarily. 09:57:49 format is fast 09:57:57 It is designed so that most of the work can be done at compilation time. 09:58:03 doing I/O might be slow, though 09:58:06 While you have literal control strings... 09:58:48 -!- jsoft [n=user@unaffiliated/jsoft] has quit [Read error: 60 (Operation timed out)] 09:58:48 fusss: you can also use FORMATTER, and have pre-compiled format functions 09:59:01 clhs formatter 09:59:07 (Good implementations call it automatically). 09:59:24 RIP specbot? 09:59:28 (format nil "~{~A~^ ~}" '(a b c)) 09:59:37 ah, % it was... 09:59:45 err, ^ 10:00:39 ~* is also pretty neat 10:00:43 jdz: hm. What's a nice way to insert the separator? 10:00:59 sykopomp: what do you mean? his example does that 10:01:13 sykopomp: positional parameter maybe? 10:01:18 Adlai: not when you're trying to pass the separator as an argument. 10:01:28 ahh. hmm. 10:01:55 pr [n=pr@unaffiliated/pr] has joined #lisp 10:01:57 jdz: what do you mean? 10:02:13 sykopomp: look at ~* 10:02:24 ~0@* will go back to the 0th argument 10:02:35 eh, you can leave out the 0 10:02:37 format makes me cry sometimes. 10:02:38 ~@* 10:02:42 -!- QinGW2 [n=wangqing@203.86.89.226] has quit [Read error: 104 (Connection reset by peer)] 10:02:55 Worse is better :) 10:02:57 Adlai: what do you mean? 10:03:46 hahahah it's GOTO 10:03:51 clhs 22.3.7.1 10:03:59 yeah, I found it :) 10:04:06 looks like positional arguments won't work in ~{ 10:04:30 hmmm 10:04:31 ~@* is for positional arguments. What doesn't work? 10:04:37 jdz: of course, the fallback is to use FORMAT to build a format string. 10:04:38 ;) 10:05:02 matimago: positions are relative to the list being processed when used within ~{ 10:05:11 ahhhh look at this 10:05:20 you can pass an empty iteration string 10:05:23 Right :-) 10:05:25 -!- kingless [n=user@adsl-144-174-240.rmo.bellsouth.net] has quit [Read error: 110 (Connection timed out)] 10:05:35 I can absolutely recommend Montezuma for production user 10:05:46 lazy standards guys, they did not format turing complete... 10:05:50 never been more confident with something i was so suspicious of at first 10:06:09 s/did not/ did not make 10:06:23 (format nil "~{~}" (format nil "~~A~~^~A" separator) list) 10:06:58 (format nil "~{~}" (format nil "~~A~~^~A" (escape-tildes separator)) list) 10:07:03 jdz: I know, right!? TeX is turing-complete and yet FORMAT isn't!? 10:07:08 i wonder if montezuma's index files are platform independent like sqlite3 databases 10:07:20 -!- Baconizer [n=baconize@unaffiliated/baconizer] has quit [Read error: 110 (Connection timed out)] 10:07:27 matimago: ah, true, that's important too. 10:07:29 lisp paste is dead! 10:07:30 (defun separate-lulz (separator &rest strings) 10:07:30 (format nil (format nil "~~{~~A~~^~A~~:}" separator) strings)) 10:07:39 bah, you guys beat me to it while I was fussing with lisppaste :( 10:07:42 sykopomp: that's old news! 10:07:52 also, you fail -- look at matimago's latest. 10:07:57 yes, I saw that. 10:08:21 And you need something inside ~{~} 10:08:44 Perhaps s/~{~}/~?/ 10:09:03 or you could just build the full format string and pass it straight to format 10:09:15 matimago: no, you don't need anything inside ~{~} 10:09:19 as in ^ 10:09:33 -!- zophy [n=sy@host-242-6-111-24-static.midco.net] has quit [Read error: 110 (Connection timed out)] 10:09:50 What would ~{~} format? Nothing. 10:10:17 it uses the next argument as the formatter for the loop 10:10:38 -!- dv_ [n=dv@83-64-248-68.inzersdorf.xdsl-line.inode.at] has quit ["Verlassend"] 10:11:04 (format nil "~{~}" "~A~^, " '(a b c)) => "A, B, C" 10:11:13 fascinating 10:11:22 this makes me both love and hate format. 10:11:29 benbos67 [n=ben@CPE-124-177-162-112.qld.bigpond.net.au] has joined #lisp 10:11:46 I'm surprized. 10:12:02 Where is it specified that ~{~} should be equivalent to ~? ? 10:12:10 at the bottom of clhs 22.7.3.4 10:12:14 matimago: me too 10:12:17 sorry, 22.3.7.4 10:12:31 Adlai: nice. how'd you come across that? careful reading of the spec? 10:12:42 Xach: blood! sweat! tears! 10:12:49 and yes, careful reading of the spec :) 10:12:56 good on you! 10:13:02 *Adlai* glows 10:13:12 Indeed... 10:13:26 does he get a sticker or something? 10:13:28 at least? 10:13:42 Fufie [n=poff@Gatekeeper.vizrt.com] has joined #lisp 10:13:59 We should incf language laywer points on common-lisp.net :-) 10:14:22 matimago: what are those? 10:14:29 An invention. 10:14:34 Adlai: cll epeen 10:14:43 heh 10:14:52 There are People pages on common-lisp.net, we could track LL points for each person. 10:15:54 might actually be a nice resource for browsing 10:16:04 hmm, my only People page is on CLiki. 10:16:18 haiwei [n=haiwei@192.9.202.3] has joined #lisp 10:16:32 Adlai: you have a cliki page. you know what the means right? you exist :-) 10:16:45 -!- haiwei [n=haiwei@192.9.202.3] has quit [Client Quit] 10:16:53 A more interesting question should be, given that we have compilers these days, is there a good justification for format? 10:16:58 haiwei [n=haiwei@192.9.202.3] has joined #lisp 10:17:24 Zhivago: FORMAT is for rendering payroll sheets 10:17:33 matimago: where are the people pages on c-l.net? 10:17:35 and for 99 bottles of beer. 10:18:08 http://99-bottles-of-beer.net/language-common-lisp-114.html My favorite format string ever. 10:18:08 There are simpler ways to do that :) 10:18:19 cliki.net is down? 10:18:46 attila_lendvai: it looks like everything CL-related is down right now, even specbot is gone. 10:18:58 the apocalypse is upon us. 10:19:02 chls 22.3.7.4 10:19:03 C has finally won. 10:19:10 ... no specbot! 10:19:12 planet lisp struggles on 10:19:17 l1sp.org remains! 10:19:35 -!- haiwei [n=haiwei@192.9.202.3] has quit [Client Quit] 10:19:37 Xach: Row row fight da powah. 10:19:39 yay! 10:20:58 Xach: I have l1sp.org in my Conkeror webjumps 10:21:08 (just put it in very recently) 10:21:12 -!- yango [n=yango@unaffiliated/yango] has quit [Remote closed the connection] 10:21:24 -!- gko [n=gko@211.22.47.2] has quit ["rcirc on GNU Emacs 23.1.1"] 10:21:26 yango [n=yango@unaffiliated/yango] has joined #lisp 10:22:31 Adlai: you know you have clhs as a webjump, right? 10:22:48 C-u g clhs print 10:22:49 i don't like in cliki that there are no subjective "try this one first" suggestions... i'm looking for a parser generator lib to parse http dates specified by an rfc grammar and i have no clue which lib to choose... 10:23:03 anyone with some experience? 10:23:03 sykopomp: yes, but now I can go to "lisp pcl/eval-when" 10:23:28 hm 10:23:37 interesting. 10:23:39 attila_lendvai: hmm, somebody recommended one a few days ago, I think it was cl-yacc but I'm not certain. 10:24:49 *attila_lendvai* is trying meta currently 10:25:09 sykopomp: also, I never use conkeror's CLHS webjump, because I have slime-hyperspec-lookup that searches my local copy of the spec and pops it up using a text-only browser -- thus, I have instant CLHS access 10:25:22 meh! 10:25:51 :P 10:26:36 -!- ace4016 [i=ace4016@cpe-76-168-84-210.socal.res.rr.com] has quit ["night"] 10:33:29 *attila_lendvai* is at cl-yacc 10:40:49 -!- licoresse [n=peropaal@cm-84.209.69.164.getinternet.no] has quit [Remote closed the connection] 10:40:55 -!- s0ber [n=s0ber@118-168-239-148.dynamic.hinet.net] has quit [Remote closed the connection] 10:44:03 what do you call it when you chase the wrong bug an mutilate some perfectly good code in the process 10:44:18 *fusss* cross fingers and hopes he doesn't run out of undo levels 10:44:21 I call that git reset --hard 10:44:39 fusss: lern2git. Make small commits. Love reset and branching. 10:45:05 i use the fucker for work and do about one commit command every friday 10:45:18 fusss: bad idea, imo. 10:45:20 you should commit all the fucking time :P 10:45:33 make very frequent commits, and then squash them together when it's time to "present" 10:45:52 it keeps asking me to input a message, meh 10:46:04 put it on the commantline 10:46:07 honestly, let's not git there 10:46:09 Adlai likes committing all the fucking time... me, on the other hand... I prefer doing a medium-sized blob of work that doesn't go too far, and then splitting the diff into several commits later. 10:46:26 fusss: well, you're digging your own grave here, is all. 10:46:29 s0ber [i=pie@114-45-226-57.dynamic.hinet.net] has joined #lisp 10:46:34 `git commit -m "fix bug in foo/bar/baz.lisp"' 10:46:47 ^ 10:47:02 -!- SandGorgon [n=OmNomNom@122.162.50.31] has quit [Remote closed the connection] 10:47:08 when I have commits that I plan on squashing, sometimes I even just do 'git commit -m squash' 10:47:10 ouch, i was gonna say the win32 port is hard to install but it's already installed on my machine 10:47:17 nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 10:47:32 Adlai: gitvangelism hi-5 o/ 10:47:36 C-_ is my git 10:47:48 fusss: sounds painful. 10:47:52 fusss: try C-/ 10:47:52 -!- redblue [i=star@ppp013.108-253-207.mtl.mt.videotron.ca] has quit [Read error: 104 (Connection reset by peer)] 10:47:53 *Adlai* gives sykopomp the dolphin 10:48:20 i wonder if there's a list of regression tests that are expected to fail on sbcl win32 10:48:20 sweet, you guys are dropping pills now? 10:48:33 dolphin was a great ecstasy pill 10:48:38 git with the times. 10:48:42 what? no clue what that is. 10:48:51 "The dolphin" is when you pretend to hi5 somebody 10:48:57 Adlai: we need to have a discussion. 10:49:01 then at the last moment, you duck your arm down like a dolphin 10:49:39 kerimbasol_ [n=kerim@85.107.158.2] has joined #lisp 10:49:41 *whoosh* 10:51:11 nevermind... 10:51:37 adlai gets a free pass for the day due to ~{~} 10:51:45 fusss: the WP article on X is hilarious, btw. 10:51:59 Xach: free pass on what? 10:52:09 Adlai: on being lame. 10:52:13 to clarify: I know what X is, I just don't know what "the dolphin" is in the context of X. 10:52:31 -!- benbos67 [n=ben@CPE-124-177-162-112.qld.bigpond.net.au] has quit ["Ex-Chat"] 10:52:34 *sykopomp* figured it's one of the pill names. 10:52:50 rstandy [n=rastandy@host193-174-static.89-94-b.business.telecomitalia.it] has joined #lisp 10:54:55 Jafet [n=Jafet@unaffiliated/jafet] has joined #lisp 10:58:45 -!- nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has quit [] 11:03:13 this new Adlai-pomp needs to hatch a new elaborate web platform to rival core server, dwim and bknr 11:03:28 something that's unlike all others, but does the same thing, more or less 11:03:28 -!- bohanlon [n=bohanlon@pool-173-48-104-141.bstnma.fios.verizon.net] has quit [] 11:03:33 the web sucks, though. 11:04:04 I've found it pretty hard to actually get interested in web-related crap :\ 11:04:21 you guys can do it over message-passing sheeple, drop the continuations fad 11:04:45 why not just use CLOS? 11:04:56 sykopomp: i am being a wiseass 11:05:09 why do you say that you are being a wiseass? 11:07:12 ok, you too get a pass for asking that question, by induction 11:07:34 why do you say that ok, you too get a pass for asking that question, by induction? 11:07:42 goodnight, #lisp :) 11:07:48 Adlai-pomp? 11:07:49 ok, Eliza Minelli 11:08:20 Adlai: has a ring to it, don't it? 11:09:05 sykopomp: how do you feel when has a ring to it, don't it? 11:12:18 .oO( not buddha's recommended discussion style ) 11:12:24 p0a [n=user@athedsl-391806.home.otenet.gr] has joined #lisp 11:12:27 Hello 11:12:34 How can I write a CFFI for system calls? 11:12:53 Should I first wrap around those system calls into a C library? I hope there's an easier way 11:12:54 p0a: by writing FFIs for the functions and their arguments 11:13:06 no C library needed 11:13:09 fusss: but how do I load the "library"? There's no library for system calls 11:13:14 you can do it from the repl if you want 11:13:24 p0a: which system calls? Osicat or your implementations might have it wrapped already. 11:13:53 Adlai: inotify system calls 11:13:56 p0a: all executable binaries implicitly link against system calls 11:14:18 fusss: how will I DEFINE-FOREIGN- 11:14:24 LIBRARY (cont.) 11:14:41 p0a: as long as you pass a valid name as a string, the linking will take places 11:15:11 Which string must that be 11:15:13 for system calls 11:15:13 ... 11:15:17 weirdo [n=sthalik@c142-5.icpnet.pl] has joined #lisp 11:15:20 the function's name 11:15:46 oh yeah 11:15:52 just use foreign-function and defcfun then 11:15:53 tahnks! 11:16:12 for arguments, you will need to use their corresponding primitive values, and for more complex stuff like structs, you will need to duplicate them; a grovel tool might help with this 11:16:17 yes 11:16:40 Reav__ [n=Reaver@212.88.117.162] has joined #lisp 11:18:30 I just didn't realize how obvious it is that system calls are available to every executable 11:19:14 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 11:22:04 -!- p0a [n=user@athedsl-391806.home.otenet.gr] has quit ["bye"] 11:22:29 p0a: iolib has everything for system calls 11:23:42 p0a: system calls are not invoked by the same function call semantics 11:23:56 did he leave? 11:24:18 He did. 11:26:34 beach [n=user@ABordeaux-158-1-49-56.w90-16.abo.wanadoo.fr] has joined #lisp 11:26:38 Good afternoon! 11:26:53 good afternoon beach 11:27:33 Edico [n=Edico@unaffiliated/edico] has joined #lisp 11:27:51 Puppster [n=jared@99-23-192-153.lightspeed.austtx.sbcglobal.net] has joined #lisp 11:28:17 Hmm, so I guess the Gnome GUI interface is going to be useless for configuring my graphics card an monitors. 11:28:48 silenius [n=jl@yian-ho03.nir.cronon.net] has joined #lisp 11:30:07 scottmaccal [n=scott@169.244.94.54] has joined #lisp 11:30:08 Most likely. 11:34:51 sanxiyn [n=tinuviel@110.76.72.94] has joined #lisp 11:35:41 What happened to www.cliki.net? 11:35:43 I can't resolve it. 11:36:19 beach: why? 11:37:44 madnificent: I wish I knew. It doesn't know the monitor type and doesn't even give me all the resolution options. 11:39:33 which card? 11:39:41 Some GeForce something 11:40:00 sanxiyn: http://www.squish.net/dnscheck/ says it's highly nonexistent 11:40:05 -!- fusss [n=chatzill@60-241-1-206.static.tpgi.com.au] has quit ["ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]"] 11:40:07 ThomasI [n=thomas@unaffiliated/thomasi] has joined #lisp 11:40:25 is it still supported? 11:40:42 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 11:40:51 beach: have you tried the "device drivers" program? (in System / Administration, on Ubuntu) 11:41:05 mvilleneuve: Nope 11:41:09 beach: are you using binary nvidia drivers? 11:41:29 antifuchs: So how do I get to Cliki? 11:41:32 stassats: I don't know. I just installed this thing a few minutes ago. 11:41:43 sanxiyn: I don't know, sorry )-: 11:41:47 beach: on my PC it did find the correct nvidia driver, and everything worked fine the first time I used it. 11:41:54 What a pity 11:42:27 beach: if you are ok with proprietary blob tainting your kernel, then install them 11:43:49 stassats: I think I am about to. 11:44:42 bohanlon [n=bohanlon@66.170.231.48] has joined #lisp 11:44:52 -!- bohanlon [n=bohanlon@66.170.231.48] has quit [Remote closed the connection] 11:45:18 OK, I need to reboot... 11:45:22 -!- beach [n=user@ABordeaux-158-1-49-56.w90-16.abo.wanadoo.fr] has left #lisp 11:45:33 -!- clop2 [n=jared@99-23-192-153.lightspeed.austtx.sbcglobal.net] has quit [Read error: 110 (Connection timed out)] 11:45:59 thehcdreamer [n=thehcdre@94.198.78.26] has joined #lisp 11:46:14 -!- Jasko2 [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has quit ["Leaving"] 11:50:19 -!- drafael1 [n=tapio@118.90.132.209] has quit [Client Quit] 11:53:20 -!- Quadrescence [n=quad@unaffiliated/quadrescence] has left #lisp 11:56:15 HET2 [n=diman@w283.engin.cf.ac.uk] has joined #lisp 11:57:01 -!- PissedNumlock [n=resteven@igwe19.vub.ac.be] has quit [Read error: 113 (No route to host)] 11:59:39 beach [n=user@ABordeaux-158-1-49-56.w90-16.abo.wanadoo.fr] has joined #lisp 11:59:52 Ah, that's better. At least one screen with full resolution. 12:01:02 Dawgmatix_ [n=dman@c-68-32-44-191.hsd1.nj.comcast.net] has joined #lisp 12:01:30 -!- OmniMancer [n=OmniManc@222-154-179-43.jetstream.xtra.co.nz] has quit ["Leaving."] 12:03:39 benny` [n=benny@i577A1199.versanet.de] has joined #lisp 12:04:16 hi i m searching good lisp book for AI 12:04:22 which one is best 12:04:29 PAIP 12:04:29 i m beginner at LISP 12:04:51 http://norvig.com/paip.html 12:05:05 thanx 12:05:07 ;) 12:05:12 tic : thanks 12:05:22 tic : ;) 12:05:52 Yay! Two screens working! 12:06:32 -!- redline6561 [n=redline@c-66-56-16-250.hsd1.ga.comcast.net] has quit [Read error: 113 (No route to host)] 12:06:45 -!- benny` is now known as benny 12:06:52 -!- emma [n=em@unaffiliated/emma] has quit [Read error: 110 (Connection timed out)] 12:09:51 emma [n=em@unaffiliated/emma] has joined #lisp 12:10:52 beach, yay 12:11:56 kerimbasol_, grab a torrant for PAIP, no need to pay for it 12:12:02 But with the change of drivers, the workspace manager is useless :( 12:12:09 And the panels moved to the other screen. 12:12:41 weirdo thats strange advice 12:12:43 Jasko [n=tjasko@209.74.44.225] has joined #lisp 12:12:52 beach: enable xinerama if you didn't. it let's you move windows between the screens... 12:12:55 you can have panels on both screens. 12:13:17 attila_lendvai: That's already the case. 12:13:22 Dawgmatix_, why bother paying 50$ for a dead-tree book, then waiting for it to ship, if you can get a PDF for free? 12:13:24 tic: I am sure I can. I just have to figure out how. 12:13:42 drag them with the mouse? 12:13:48 right-click, new panel, move it. 12:14:14 attila_lendvai: My problem is that I have 50 workspaces, and they used to be named. Now they are so tiny (and they are unnamed) that I can't reasonably select one. 12:14:34 beach: alt+drag panels to move them 12:14:53 weirdo : thanks . I am doing now your advice 12:14:53 attila_lendvai: Works! Thanks! 12:15:08 a) you can read a dead-tree book sitting on the john b) you want to have incentive for people to write great books, and paip is a great book 12:15:37 hello. i am using sbcl 1.0.28 under linux - and i somehow have problems to use asdf-install. i did (require 'asdf) (require 'asdf-install) and (asdf-install 'ltk) -> personal installation. i always get an error: name service error in 'getaddrinfo" -2 (sb-bsd-sockets:host-not-found-error). what am i doing wrong? 12:15:46 lisppaste [n=lisppast@208.72.159.207] has joined #lisp 12:16:14 you should (asdf-install:install :ltk) 12:16:22 Ah, much better when I removed it and then put it back. 12:17:05 Dawgmatix_: oh, sorry i mistyped -> i surely did (asdf-install::install 'ltk) 12:17:08 minion [n=minion@common-lisp.net] has joined #lisp 12:17:14 specbot [n=specbot@common-lisp.net] has joined #lisp 12:17:21 -!- hefner_ [n=hefner@ppp-61-90-103-217.revip.asianet.co.th] has quit [Read error: 113 (No route to host)] 12:18:00 trebor i think thats because cliki is down 12:18:25 so asdf-install wont work as of now. you can use clbuild 12:18:48 a-s [n=user@nat-240.ro.66.com] has joined #lisp 12:19:56 gmlk [n=gmlk@alicia.demon.nl] has joined #lisp 12:20:02 Dawgmatix_: thanks for the hint (now i see that browsing http://www.cliki.net/LTK does not work either). 12:22:33 cliki works for me when using telnet(1) but not when using an actual web browser 12:25:40 carlocci [n=nes@93.37.197.170] has joined #lisp 12:26:53 weirdo: Does DNS resolve for you? 12:27:01 sanxiyn, it does 12:27:10 What is its IP address? 12:27:10 cliki.net has address 80.68.85.134 12:27:15 (I want to put that on /etc/hosts) 12:28:15 *Xach* smacks forehead 12:28:20 Nice, Cliki works fine now. 12:29:16 someone forgot to renew the domain *again*? 12:30:21 weirdo: i think the DNS host has had some problems with the transition from dan_b to drewc 12:31:39 Aszarsha [n=aszarsha@91-164-152-252.rev.libertysurf.net] has joined #lisp 12:31:52 -!- thehcdreamer [n=thehcdre@94.198.78.26] has quit [] 12:32:15 -!- legumbre_ is now known as legumbre 12:32:54 coderdad [n=coderdad@ip72-200-214-240.ok.ok.cox.net] has joined #lisp 12:35:43 hefner_ [n=hefner@ppp-58-11-44-79.revip2.asianet.co.th] has joined #lisp 12:37:34 pr_ [n=pr@p579CA861.dip.t-dialin.net] has joined #lisp 12:43:49 araujo [n=araujo@gentoo/developer/araujo] has joined #lisp 12:44:07 Fade [i=fade@outrider.deepsky.com] has joined #lisp 12:50:40 sanxiyn: yeah, now that we've got TB HD, we could fill /etc/hosts with all the hosts :-) 12:51:04 heh .. /etc/hosts reminds me of 1994 12:51:14 "we should trade host entries" 12:51:16 :) 12:51:27 -!- pixel5 [n=pixel@copei.de] has quit ["Lost terminal"] 12:54:14 -!- pr [n=pr@unaffiliated/pr] has quit [Read error: 101 (Network is unreachable)] 12:54:48 disumu [n=disumu@p54BCC651.dip.t-dialin.net] has joined #lisp 12:55:13 thehcdreamer [n=thehcdre@94.198.78.26] has joined #lisp 12:58:58 Nshag [i=user@Mix-Orleans-106-3-85.w193-248.abo.wanadoo.fr] has joined #lisp 13:00:30 paip.pdf link shortcut http://www.mediafire.com/?gy4tqqnbdwg 13:00:53 kerimbasol_: Don't do that. 13:01:08 -!- thehcdreamer [n=thehcdre@94.198.78.26] has quit [] 13:01:23 and it looks like a crappy photopdf 13:02:45 You already downloaded it? 13:02:51 i think we're able to find paip on the interwebz ourselves :) (i pirated it too, then bought it) 13:02:56 no, i looked at the size of the file 13:03:01 17.5MB 13:03:25 Didn't you hear, OCR tools still can't handle parentheses well 13:03:32 No, I mean, the time needed to ponder the ethical questions! :-) 13:03:44 Jafet: Tsk tsk. 13:03:45 i own the hardcopy 13:03:47 I still need to buy a copy... so far I've just been reading th esource code 13:03:54 *matimago* too 13:04:22 i was just wondering maybe there is a .pdf i have missed for times when i'm not bringing the book with myself :) 13:04:27 -!- Demosthenes [n=demo@64.208.221.2] has quit ["leaving"] 13:04:43 Adlai: me too 13:06:02 happily enough for him I suspect norvig is indifferent to book income nowadays :) 13:06:19 the publisher might care, however 13:06:26 'tis true 13:07:11 and finding publishers willing to publish books on niche topics is probably as hard as finding authors willing to write about them 13:07:59 SandGorgon [n=OmNomNom@122.162.50.31] has joined #lisp 13:15:47 antgreen [n=user@CPE0014bf0b631a-CM0018c0b3c272.cpe.net.cable.rogers.com] has joined #lisp 13:17:56 -!- ejs1 [n=eugen@nat.ironport.com] has quit [Read error: 60 (Operation timed out)] 13:19:35 jleija [n=jleija@user-24-214-122-46.knology.net] has joined #lisp 13:21:12 Hmm, why are suddenly a lot of sites unreachable from my home computer, whereas they are reachable from work? 13:21:45 google.com for instance, or sbcl.sourceforge.net 13:22:16 beach: do you have cats? 13:22:17 dns servers? 13:22:40 jdz: as in the domestic animal? No. 13:23:04 anyway, i did not have anything to add except the usual isp/dns messups 13:23:06 ljames: it resolves the name, but let me check that it is resolved the same way as at work. 13:23:53 beach: where can I find information for erasmus students in bordeaux? 13:24:14 madnificent: In CS? Write to carrere@labri.fr 13:24:34 beach: there's no page of information? 13:24:41 thanks btw! 13:24:41 I don't know 13:24:51 we're looking for information on some places for next year :) 13:24:54 beach: it's usually dns 13:24:59 though it could be routing 13:25:19 rsynnott: Can it be DNS if a ping actually seems to resolve to an IP address? 13:25:45 if ping gets an IP address, then DNS is working 13:25:54 ping uses systems resolving facilities... 13:26:17 madnificent: That's what I thought. 13:26:46 So I am having problems with IP connectivity, but freenode works fine, and so do some other sites. 13:26:58 -!- hefner_ is now known as hefner 13:27:02 beach: yep, probably not 13:27:33 (there used to be an annoying linux problem where some software used a dns lib which would break differently to other dns libs) 13:27:36 hefner: Hey, where are you? 13:27:53 so ping would be fine bit not netscape navigator, for instance; I'm pretty certain it's long gone tho 13:28:30 rsynnott: I see. Well, I can't ping either :( 13:28:33 beach: Bangkok. Spending a few months here (two down, one to go). 13:28:44 hefner: Nice! Work? 13:29:10 beach: if you're in doubt about the DNS server, try to use one of opendns... 13:29:22 beach: no, just visiting. 13:29:43 hefner: Three months? That's a pretty nice visit. 13:29:58 madnificent: Thanks, I'll remember that. For now, that doesn't seem to be the case. 13:30:48 beach: otherwise you could try to use a proxy... then see if all content gets in... if so then your ISP is toying around with blocking software... and it isn't doing it very well 13:31:29 beach: one of the perks of unemployment. 13:31:42 blackened` [n=blackene@ip-89-102-28-224.karneval.cz] has joined #lisp 13:32:07 madnificent: Thing is, I just hooked up a new computer to the same wire where I have had a different one for a long time. I have had no problems in the past, so it would be a strange coincidence if it were the ISP. 13:32:38 hefner: Hmm, yes. 13:32:40 (and youngish and free of obligations) 13:33:05 hefner: joy! 13:33:11 splittist [n=dmurray@85.4.220.156] has joined #lisp 13:33:13 morning 13:33:25 hello splittist 13:33:53 TDT [n=user@206.196.111.202] has joined #lisp 13:34:38 beach: is your new computer using the same IP address as the old one? 13:35:34 arbscht: I don't know. 13:36:17 manuel__ [n=manuel@pD9E6F199.dip.t-dialin.net] has joined #lisp 13:36:21 arbscht: How does that make a difference? 13:37:01 beach: if not, I would have asked whether there was a firewall along the way that might be filtering traffic based upon IP addresses 13:37:09 beach: it _could_ if your ISP allows you to connect only X computers and you've connected X+1 computers in the past Y hours 13:37:34 -!- attila_lendvai [n=ati@catv-89-134-66-153.catv.broadband.hu] has quit [Read error: 110 (Connection timed out)] 13:37:36 madnificent: Possibly, but then NO IP address would work, and freenode does, and some others as well. 13:37:52 arbscht: I see, but it is dynamic IP address allocation. 13:38:00 beach: I'm including the fact that your isp is messing up a bit somehow :) 13:38:11 s/fact/presumption/ 13:38:23 I guess I should go reboot the modem and see whether that helps. 13:38:38 -!- salva [n=salva@105.11.117.91.dynamic.mundo-r.com] has quit [] 13:39:40 beach: we'll tell you if anything exciting happens while you're away... 13:39:45 madnificent: how would an isp even trace that? I've never heard of one doing it 13:39:56 (unless they controlled the modem or something, I suppose) 13:40:06 redline6561 [n=redline@gate-20.spsu.edu] has joined #lisp 13:40:17 splittist: Thanks! 13:40:37 beach: they haven't been assigned the same ip address, have they? 13:41:08 -!- patmatch [n=wmbot@dsl093-216-036.aus1.dsl.speakeasy.net] has quit [Read error: 60 (Operation timed out)] 13:42:12 rsynnott: there are multiple ways. One way telenet uses is by giving each computer an IP address themselves. That means that if you change it too often, it will not work. Our campus here only redirects the DNS requests. Belgacom provides you with routers which can be 'limited' to a certain number of connections (eg: 4). You can clearly work around it by putting your own router behind the provided router, but NAT doesn't follow the 13:42:12 services principle of networks protocols, so you may want to *not* have that too. 13:42:21 -!- SandGorgon [n=OmNomNom@122.162.50.31] has quit [Read error: 60 (Operation timed out)] 13:43:16 beach` [n=user@ABordeaux-158-1-79-46.w90-60.abo.wanadoo.fr] has joined #lisp 13:43:20 hmm, that sounds exceedingly annoying 13:43:25 rsynnott: as a side-note, Belgium is the most fcked up country with respect to any form of digital communications 13:43:31 rsynnott: 'yes' 13:43:51 rsynnott: most of the belgians are capped at 20Gb per month too :) 13:43:52 -!- xan is now known as xan-afk 13:43:56 could you not just use your own modem/router? 13:44:00 -!- beach [n=user@ABordeaux-158-1-49-56.w90-16.abo.wanadoo.fr] has quit [Nick collision from services.] 13:44:03 -!- beach` is now known as beach 13:44:15 rsynnott: you can place your own router behind the router they gave you... that clearly works 13:44:19 Edward [n=Ed@81.249.193.178] has joined #lisp 13:44:37 rsynnott: Only one of them is connected. 13:45:04 (our big telecom briefly attempted to force people to use provided modems only around 2001; they desisted when the govenment threatened to fine them) 13:45:23 madnificent: Sounds like New Zealand. 13:46:30 rsynnott: we have more fun! One of the companies (telenet) has actually received a monopoly right for the next (10?) years wrt internet over cable-television lines. The other big company (using phone-lines/adsl) is not a 'monopoly' as they only serve half of the country... nice huh? 13:46:51 beach: yes, they have had their mess too 13:46:54 Rebooting the modem worked. Now I can install SBCL. Thanks to everyone. 13:46:59 *madnificent* finds it frustrating 13:47:07 good luck! 13:47:24 -!- kiko_ [n=user@67.207.130.53] has left #lisp 13:48:20 how does postmodern reuse database-connections. Will it reuse it when I use (with-connection (connection-parameters) body) ? 13:51:33 ejs [n=eugen@220-31-178-94.pool.ukrtel.net] has joined #lisp 13:51:52 -!- ejs [n=eugen@220-31-178-94.pool.ukrtel.net] has quit [Client Quit] 13:53:24 Clicking on the green square for Linux/AMD64 on the SBCL download page, I don't get a .bz2 file the way I get if I click on the Linux/x86 square. Does anyone else have that problem? 13:54:16 Blaay [i=Blay@BSN-142-112-74.dial-up.dsl.siol.net] has joined #lisp 13:54:36 rudi [n=rudi@129.27.202.101] has joined #lisp 13:55:14 beach: the link doesn't point to the right place 13:55:27 beach: http://sourceforge.net/projects/sbcl/files/sbcl/1.0.29/sbcl-1.0.29-x86_64-linux-binary-r2.tar.bz2/download 13:55:39 postamar [n=postamar@75-119-241-70.dsl.teksavvy.com] has joined #lisp 13:55:57 Right, I noticed the difference between the `-' and the `_'. 13:59:23 Odin- [n=sbkhh@s121-302.gardur.hi.is] has joined #lisp 13:59:46 ejs [n=eugen@94.178.31.220] has joined #lisp 14:02:05 wol [n=wol@67.174.222.215] has joined #lisp 14:03:24 sadly, sourceforge changed its release system a couple of months ago, breaking all links in the process 14:03:41 which reminds me, if any sbcl releaser has any spare cycles to do a release this month: I don't think I do 14:03:53 hm, that means jsnell of people here 14:04:05 davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #lisp 14:06:21 Hmm, perhaps I should use clbuild this time. Does it still work? 14:07:07 beach: it's been working fine for me (as of yesterday) 14:07:08 yes 14:07:19 minion: clbuild 14:07:20 clbuild: I can't be expected to work when CLiki doesn't respond to me, can I? 14:07:36 True, I'll find it myself. 14:08:30 Well, I'd be expecting more from a simili-AI such as minion. It could have its own memory... 14:09:25 ignas [n=ignas@ctv-79-132-160-221.vinita.lt] has joined #lisp 14:09:33 *hefner* hands minion a hash table 14:09:39 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 14:10:14 nyef [n=nyef@pool-71-161-71-17.cncdnh.east.myfairpoint.net] has joined #lisp 14:10:22 G'morning all. 14:10:58 what's wrong with cliki.net? 14:11:49 hello nyef 14:12:00 What does clbuild do when SBCL is already installed? 14:13:22 jleija: Looks like a DNS fault from here, which is definately wierd. 14:13:38 Hrm. Or, rather, a partial DNS fault. 14:14:09 beach: if no SBCL is sitting in clbuild/target, it defaults to the one in $PATH 14:14:39 beach: for "clbuild compile-implementation sbcl", it will also use the exsiting one to bootstrap. Afterwards, the newly compiled one takes precedence. 14:17:24 nyef: yeah, weird. I think it's been like this for a couple days 14:18:10 It was working for me yesterday, FWIW. 14:18:12 lichtblau: Thanks! 14:18:23 it's been doing it at least occasionally for the last few days 14:18:37 lat [n=lat@125.167.140.159] has joined #lisp 14:18:48 Sounds like a bad zone file somewhere and an incoherent cache. 14:18:58 on the subject of clbuild, how do i stop skipping updating a project? 14:19:22 I'd dig further, but I appear to have misplaced my memory of where the master copy of the DNS information resides. 14:19:37 kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 14:20:10 nyef: i recently asked drewc to update planet.cliki.net and he said the dns provider had some confusion about whether dan_b's records or drewc's records were the right ones. 14:20:17 nyef: maybe in resolving that confusion some more confusion sprang up 14:20:32 exponential confusion 14:20:56 Ah. 14:21:05 How fun. 14:21:23 here's a choice quote 14:21:37 Xach: FFS... i just got an email back from our DNS provider telling me they don't have 'diki.net' on their severs. I told him to get a proper font. 14:21:51 Heh. 14:22:03 that was on september 24th 14:23:34 ramus`: "skip" should have an effect on "update --resume". Can you elaborate on your situation? 14:23:46 mattrepl [n=mattrepl@129.174.97.34] has joined #lisp 14:24:11 it skips said project every update, resume or not 14:26:34 gah. What's the magical incantation to get slime to load sbcl on Windows? 14:26:43 So, from this fiasco, my takeaway is "sharing a single DNS server across all of your customers can lead to problems". 14:26:48 splittist: M-x slime ? 14:26:58 what do people think about picolisp? I just came across it 14:27:21 splittist: Or were you looking for something that sets up the SBCL_HOME environment variable as well? 14:27:25 Adlai: seems more interesting than some of the toy lisps i've seen, but i've not looked into it in detail 14:27:33 it seems a bit silly because it's all interpreted cons cells, but the minimalism is interesting. 14:27:44 it looks to me like Arc Done Right 14:28:06 and I have the same opinion about "Arc Done Right" that Linus Torvalds has about "CVS Done Right" 14:28:14 splittist: http://www.lisphacker.com/temp/article-drafts/sbcl-win32-slime-and-cygwin-emacs.txt should be taken with a grain of salt, but might be helpful. 14:28:15 but it's still a bit intriguing. 14:28:36 nyef: perhaps. I want to stop the fatal error "can't find core file at e:\Steel Bank Common Lisp\1.0.29\/sbcl.core". I could swear the setup in my .emacs (equiv) worked before... 14:28:41 Whe 14:28:44 ramus`: ls clbuild/source/*/.clbuild-skip-update #? 14:29:16 Ah, -that- I can't really help with. 14:30:08 lichtblau: that file doesn't exist in the project it's skipping 14:30:54 ruediger [n=ruediger@91-115-181-187.adsl.highway.telekom.at] has joined #lisp 14:34:04 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 14:34:12 -!- mattrepl [n=mattrepl@129.174.97.34] has quit [] 14:34:29 *splittist* realises the change was emacs 22 -> 23, goes back to clisp 14:35:57 -!- Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has quit [Read error: 113 (No route to host)] 14:38:15 When I type C-c ~ to synchronize, I get: The assertion (PACKAGEP SWANK::P) failed. What is wrong? How to fix? 14:39:17 lat: do you have an (in-package ...) form in this buffer? 14:39:43 does package in referred in in-package exist? 14:39:59 michaelw, yes. 14:40:08 mattrepl [n=mattrepl@129.174.97.34] has joined #lisp 14:40:31 -!- icecub_ [n=icecube@p549C3509.dip.t-dialin.net] has quit [Remote closed the connection] 14:41:33 lat: so, does it exists? 14:41:56 stassats, I don't know for sure. I defined it. 14:42:03 Adlai: fwiw, I find hedgehog lisp nicer than picolisp, in that it's more CL-like in some ways. however, I've not yet found a use for it where ecl or clisp didn't suffice 14:42:31 lat: what is the result of (find-package 'package) ? 14:42:54 arbscht: hmm, I'll look it up... google? 14:43:06 *Adlai* googles 14:43:58 stassats, NIL 14:44:26 that means it doesn't exist 14:47:11 -!- simplechat [n=simple@unaffiliated/simplechat] has quit [Remote closed the connection] 14:48:35 Err...quick dumb question, is there a complete list of built-in reader macros lying around somewhere? 14:48:40 *redline6561* just noticed #x 14:49:19 redline6561: Yes, in the CLHS. 14:49:29 clhs # 14:49:30 http://www.lispworks.com/reference/HyperSpec/Body/02_dh.htm 14:49:35 Thanks. Got to run... 14:49:38 clhs 2.4 14:49:38 http://www.lispworks.com/reference/HyperSpec/Body/02_d.htm 14:50:31 *Xach* wonders about a lisp-critic extension that warn when using #r unnecessarily 14:50:41 ThomasIl [n=thomas@unaffiliated/thomasi] has joined #lisp 14:50:59 do you use lisp-critic? 14:51:27 no. 14:51:54 Xach: can lisp-critic even handle that, or does it READ its input? 14:52:13 pkhuong: i don't think it can handle it. 14:52:35 maybe it could go alongside Xof's unnecessary :: warning... 14:53:05 pkhuong: could use matimago's reader... 14:54:47 Lisp-critic reads its input. Complains a lot about '() (it reads it as 'nil). 14:55:03 That would be the first thing to do on it. 14:56:00 Then it complains a lot on silly things, probably because of limits set too low such as function size... 14:56:20 *michaelw* didn't know until now that lisp critic is 22 years old 14:56:27 (Might be good for an educational setup, but a alot of care would probably be needed for industrial code). 14:57:00 -!- disumu [n=disumu@p54BCC651.dip.t-dialin.net] has quit ["..."] 14:57:39 -!- jdz [n=jdz@85.254.193.149] has quit ["Somebody booted me"] 14:58:26 -!- ASau [n=user@host183-230-msk.microtest.ru] has quit ["off"] 14:58:46 Hmm, does anyone have any particularly good argument for (call-next-method ...) with different arguments to the arguments received? 14:59:16 Zhivago: commutative operations 14:59:52 Well, remember that it's not allowed to change anything that might have affected dispatch. 14:59:52 An argument for a good argument for different arguments to the arguments! 14:59:58 Zhivago: (defclass circle () ((r :initarg :radius))) (defclass elipse (circle) ((a :initarg :long-axis) (b :initarg :short-axis))) 15:00:45 (not arguing on the place in the hiearchy of circle vs. ellipse). 15:01:27 Zhivago: well, ok, I'm talking about &key arguments that are not taken into account for dispatching. 15:02:11 That's a good point. 15:02:53 stassats, I'm sending a paste. 15:03:04 There was another example, about constrained sizes. If you must hold your screen-rectangle inside the screen, (set-width r w) might change w when calling the next method. 15:03:11 lat: what for? 15:03:13 lat pasted "package problem" at http://paste.lisp.org/display/87827 15:03:46 you need to define a package before using C-c ~ 15:04:14 matimago: Sounds like it should be delegation there. 15:04:53 Again, I'm not arguing the modelization. You might have to work with legacy code... 15:05:00 *Xach* wishes it would just sync *d-p-d* if the package does not exist (and maybe warn) 15:05:32 stassats, Isn't the package defined in the file I pasted? 15:05:55 did you try to evaluate defpackage form? 15:06:01 it's malformed 15:06:23 -!- ThomasI [n=thomas@unaffiliated/thomasi] has quit [Read error: 110 (Connection timed out)] 15:06:31 Xach: i agree, i'll do this a little bit later 15:06:54 thehcdreamer [n=thehcdre@94.198.78.26] has joined #lisp 15:08:11 mattrepl_ [n=mattrepl@129.174.97.34] has joined #lisp 15:08:55 LiamH [n=none@common-lisp.net] has joined #lisp 15:10:11 redblue [i=star@ppp077.108-253-207.mtl.mt.videotron.ca] has joined #lisp 15:11:13 -!- mathrick [n=mathrick@users177.kollegienet.dk] has quit [Read error: 110 (Connection timed out)] 15:11:15 -!- mattrepl [n=mattrepl@129.174.97.34] has quit [Connection timed out] 15:11:16 -!- mattrepl_ is now known as mattrepl 15:11:22 -!- Lycurgus [n=Ren@pool-71-186-141-217.bflony.east.verizon.net] has quit [Read error: 110 (Connection timed out)] 15:11:32 attila_lendvai [n=ati@catv-89-134-66-153.catv.broadband.hu] has joined #lisp 15:11:51 Lycurgus [n=Ren@pool-71-186-137-230.bflony.east.verizon.net] has joined #lisp 15:12:09 stassats, How is it misformed? I'm been trying for several days to learn how to use packages. I must be missing something very fundamental. 15:13:06 when you evaluate it, you'll get an error message 15:13:10 clhs defpackage 15:13:11 http://www.lispworks.com/reference/HyperSpec/Body/m_defpkg.htm 15:13:47 legumbre_ [n=user@r190-135-10-140.dialup.adsl.anteldata.net.uy] has joined #lisp 15:14:16 and you should define your package in package.lisp file, not in asd 15:14:24 minion: tell lat about xach-asdf 15:14:24 lat: please see xach-asdf: Xach's article "Making a small Common Lisp project" can be found at http://xach.livejournal.com/130040.html 15:14:48 lat: defpackage options must be keywords, the option "shadowing-import" doesn't exist and it wouldn't take a list as sole argument anyway 15:15:04 it doesn't evaluate things in that position 15:15:13 package.lisp or packages.lisp. I've seen (and used) both. Any logical preferences? 15:15:39 splittist: i think i would use packages.lisp if i was defining multiple packages for a project. 15:15:43 splittist: if there's one package, the former, if there are more, the later. 15:15:44 *Xach* has not yet done that 15:15:51 *Adlai* does same as Xach and matimago 15:16:14 *fe[nl]ix* likes pkgdcl.lisp 15:16:16 -!- rudi [n=rudi@129.27.202.101] has quit ["Client exciting"] 15:16:23 Moe111 [n=Moe111@modemcable067.226-21-96.mc.videotron.ca] has joined #lisp 15:16:26 wndrfl! 15:16:27 pkgdcl.lsp, surely 15:16:36 Anyone know of a good tutorial for using SSL with hunchentoot? 15:16:48 Thanks, all. 15:17:02 Moe111: i use SSL with hunchentoot by using SSL in a proxy in front of it. 15:17:12 Moe111: i haven't tried doing lisp-side SSL yet. 15:17:15 series has s-package.lisp or something 15:17:25 -!- Joreji [n=thomas@134.61.43.159] has quit [Read error: 148 (No route to host)] 15:17:38 Xach: you haven't tried because you don't need to or don't want to? 15:17:56 Xach: I'm running Hunchentoot that is already proxied by a different server in the DMZ 15:18:01 i prefer crptcnme.lisp :) 15:18:04 I find it a bit excessive to proxy it again 15:18:16 Moe111: you can add SSL to that dmz proxy. 15:18:29 Moe111: i have one nginx instance that proxies both SSL and plain traffic to hunchentoot 15:19:26 Xach: Hmmm. I will see about that then. Just to be clear though: are you implying SSL on Hunchentoot isn't mature enough to handle this? 15:19:45 sbcl master is broken on win32, is this a known issue? 15:19:53 attila_lendvai: is that hungarian? :P 15:20:11 Hungarian Notation, of course. Hungarian Notation for packages... 15:20:18 Moe111: No. I'm implying that in my situation it was easier not to bother trying it. 15:20:36 Moe111: Adding SSL to the existing proxy was 5 lines of configuration. 15:20:40 Xach: ok. I will look into it then. Thanks. 15:20:41 Failure in the very beginning of make-target-1.sh 15:20:43 ryepup1 [n=ryan@216.155.105.115] has joined #lisp 15:20:47 deepfire: What sort of "broken"? 15:20:49 Adlai: that would be CRyPTiCalNaME, I suppose 15:20:55 nyef, sec 15:21:12 *attila_lendvai* stays away from hungarian notation... 15:21:32 I've got 1.0.31.19 built as of a couple days ago, along with some custom stuff since. 15:21:51 deepfire pasted "win32 build failure in make-target-1.sh" at http://paste.lisp.org/display/87829 15:22:09 Omg, sorry, ignore my report. 15:22:27 pthread.h? 15:22:38 I've sub-consciously added :sb-thread in c-t-f.lisp... 15:22:48 And I recognize that error in x86-assem.S. 15:22:56 :-D 15:22:57 Yeah, it'd be nice if that worked though, wouldn't it? 15:23:06 I'm trying to get SBCL to talk to my arduino via USB on /dev/ttyUSB0. variants of with-open-file hang or crash, although I can "less -f /dev/ttyUSB0" just fine. my google skills are failing me, any advice? Ideally I'd like to read/write to /dev/ttyUSB0 like any other stream 15:24:12 ryepup1: I'd bet that select() doesn't work on device files as SBCL's fd-streams expect 15:24:12 nyef, are there any remote chances of you stepping up, if people collect enough of a donation pool? 15:24:17 ryepup1: I'd imagine that you want suitable :direction and :if-exitsts / :if-does-not-exist options, and a good :element-type to start with... 15:24:29 I realise this is a touchy question, so feel free to ignore it.. 15:24:37 deepfire: Yes, the chance exists. 15:24:44 What's the pool at now, 500 euro? 15:25:07 nyef: what pool ? where ? 15:25:08 And what constitutes success in this case? 15:25:18 fe[nl]ix: The SBCL/Win32 thread pool. 15:25:34 phf [n=user@host.icnfull.com] has joined #lisp 15:25:38 (Carefully abusing the notion of a "thread pool" here.) 15:26:06 I noticed 15:26:20 -!- sanxiyn [n=tinuviel@110.76.72.94] has left #lisp 15:28:51 -!- Lycurgus [n=Ren@pool-71-186-137-230.bflony.east.verizon.net] has quit [Read error: 110 (Connection timed out)] 15:29:18 nyef: to continue on the touchy subject: what would be an interesting amount of money for you to step in? 15:29:43 -!- legumbre [n=user@r190-135-14-34.dialup.adsl.anteldata.net.uy] has quit [Read error: 110 (Connection timed out)] 15:30:26 -!- Fufie [n=poff@Gatekeeper.vizrt.com] has quit ["Leaving"] 15:30:43 ryepup1 pasted "reading /dev/ttyUSB0" at http://paste.lisp.org/display/87830 15:30:48 It rather depends on how much work is involved, you know? 15:31:12 nyef: that combo worked, but is it safe to :supersede /dev/ttyUSB0? 15:31:13 nyef: yeah i know :) 15:31:45 ryepup1: I'd rather use :append 15:31:49 ryepup1: Umm... I don't know that :supersede is the right option. I might be using :overwrite for my own device access. 15:31:50 or :direction :io 15:32:23 nyef: perhaps for defining succes a simple test case should be made 15:32:23 aerique: And the amount of work involved at least partly depends on what "success" or "completion" is in this case. 15:32:24 Well with :direction :io you don't need to specify anything for :if-exists, the default is good. 15:32:47 matimago: :direction :io needs an :if-exists, if fails with the default 15:32:53 *stassats* recently used :append for ttyS0 15:33:53 I presume that part of success is actually having the threading changes in mainline source control, right? 15:33:56 ryepup1: the :append or :overwrite, for a device they should do the same. 15:34:06 overwrite is definitely the correct option 15:34:57 :APPEND is defined by the spec to seek (can you do that on a USB device?) and is misimplemented as O_APPEND by SBCL (pointless in this case). 15:35:23 :overwrite and :append both work, in my case I think it's serial-over-USB 15:35:34 :SUPERSEDE is defined by the spec to make a new file, i.e. delete the old one first. This one is misimplemented as O_TRUNC by SBCL (pointless again). 15:35:50 Yeah, that's a serial device, not a USB device (and I'd know, given lh-usb). 15:35:59 at any rate, to interact with it in python I use the serial module, so I assume it's doing some serial over USB munging 15:35:59 nyef: i personally wouldn't mind loading some extra files on windows to get it working 15:36:32 Another important question: is it worth to waste effort on win32/x86, as opposed to looking abit into the future /and/ simplifying life by going with register-rich x86-64 15:36:48 aerique: The thing is, if it's not in mainline you'd have to either apply a patch and build or pull from an alternate source repository and build. 15:37:07 :supersede, :append, and :overwrite all work for reading, I'll stick with :overwrite and see how it goes when I start writing from lisp to arduino 15:37:13 thanks for your help everyone 15:37:16 people are not /yet/ fleeing win32/x86 en masse, though 15:37:27 nyef: i really would mind that 15:37:31 deepfire: Have you looked into what is involved in getting SBCL running on Windows x64 yet? 15:37:36 rme [n=rme@pool-70-105-93-115.chi.dsl-w.verizon.net] has joined #lisp 15:37:49 IL32P64 15:37:52 Look into the future and see what low-power netbooks will be running. 15:37:54 deepfire: yeah good point, i could life with it working on just that platform if it would make things easier 15:38:02 personally, I'd prefer paying for more stable threads on Darwin/OSX 15:38:07 piso: Not even that, the exception handling stuff. 15:38:09 nyef: i really *wouldn't* mind that 15:38:18 piso: thanks for the key. 15:38:36 nyef: will that be harder on Win64? 15:38:48 deepfire: I wouldn't be surprised if a lot of people make the jump from 32-bit XP to 64-bit windows 7 (that's my plan for my work computer, at least) 15:39:13 aerique: So, you'd be happy enough if there were a fork from, say, 1.0.32 that had threading support on win32, but wasn't necessarily checked into mainline? 15:40:00 piso: Yes, because it's based on a stack-walker similar to C++ exception handling/unwind on most systems these days, but with more restrictions on function prologue/epilogue. 15:40:01 nyef: sure that would suffice 15:40:22 piso: Hence the "table-based unwind" thing that I keep looking at every so often. 15:40:41 dstatyvka [i=ejabberd@pepelaz.jabber.od.ua] has joined #lisp 15:41:00 -!- HET2 [n=diman@w283.engin.cf.ac.uk] has quit ["This computer has gone to sleep"] 15:41:29 Levenson [n=Levenson@92.46.86.8] has joined #lisp 15:41:40 gigamonkey [n=user@adsl-99-58-29-142.dsl.pltn13.sbcglobal.net] has joined #lisp 15:41:57 -!- davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 15:42:09 aerique: Okay, and you want the various thread synchronization structures described in the SBCL manual to work as well? (mutexes, semaphores and waitqueues) 15:42:17 spilman [n=spilman@ARennes-552-1-98-108.w92-139.abo.wanadoo.fr] has joined #lisp 15:42:58 Possibly with further restrictions (I know some of the mutex stuff is listed as not recommended and possibly to be deprecated). 15:42:59 aerique, did anyone set up a sbcl-win32-threads donation site yet? 15:43:16 deepfire: no i was just probing for interest 15:43:42 Xach: ok... got the DNS right this time... should be good after TTL 15:43:58 nyef: i actually don't know yet, i'd have to take some time and take a good look at that 15:44:09 drewc: did that affect the rest of cliki.net too? 15:44:10 envi^home [n=envi@220.121.234.156] has joined #lisp 15:44:43 Xach: what's diki.net? :D 15:45:29 cliki.net 15:45:36 with cl glued together = d 15:46:13 drewc: for example, the disappearance of the A record for www.cliki.net 15:46:15 tcr [n=tcr@host146.natpool.mwn.de] has joined #lisp 15:46:29 drewc: or maybe it's just that cliki.net has a bad A record 15:47:17 Does anybody know of an especially good open-source-oriented donation hub? 15:47:57 Xach: the www. A record was missing... i just uploaded a fix 15:48:25 drewc: Does that include the sbcl-internals. and mcclim (whatever the actual prefix is) ones? 15:48:33 -!- slyrus_ [n=slyrus@adsl-75-36-222-205.dsl.pltn13.sbcglobal.net] has quit [Read error: 110 (Connection timed out)] 15:49:11 -!- gigamonkey [n=user@adsl-99-58-29-142.dsl.pltn13.sbcglobal.net] has quit [Read error: 54 (Connection reset by peer)] 15:49:37 -!- lat [n=lat@125.167.140.159] has quit [Read error: 60 (Operation timed out)] 15:50:31 drewc: what happened, and how much can be blamed on bytemark? 15:51:24 -!- Moe111 [n=Moe111@modemcable067.226-21-96.mc.videotron.ca] has quit [] 15:51:33 Hmm, FOSS-oriented donation hubs don't make themselves easily apparent. 15:52:17 nyef: yeah, it will 15:52:21 deepfire: what is a donation hub? 15:53:11 Xach: it's partially my fault... when it couldn't make DNS changes stick i uploaded a minimal record in an attempt to debug things. When bytemark finally sorted things out, that was the record that was still on the server. 15:53:23 -!- weirdo [n=sthalik@c142-5.icpnet.pl] has quit [Read error: 60 (Operation timed out)] 15:53:39 ah 15:53:52 Xach, a site with purpose of facilitating independent donation efforts. 15:54:00 Xach: From context, I'm guessing a service that provides escrow accounts for donations towards specific goals with open-source software. 15:54:28 nyef: gotcha 15:55:51 I guess FOSS orientation isn't a blocker feature. 15:55:59 *deepfire* finds pledgie.com 15:57:21 gigamonkey [n=user@99.58.30.58] has joined #lisp 15:58:17 nyef: do we have any callee-save regs on x86oid? 15:59:00 Lycurgus [n=Ren@pool-71-186-133-56.bflony.east.verizon.net] has joined #lisp 15:59:31 pkhuong: I don't know about x86-64, but on x86 all of the regs are involved in the calling convention. 15:59:58 nyef, do you feel ok about paypal? 16:00:19 (Stack pointer, frame pointer, arg count, three arg-passing regs, incoming function, and... Well, the eight might have been a scratch register, I forget.) 16:00:20 danlei [n=user@pD9E2C1CC.dip.t-dialin.net] has joined #lisp 16:00:21 All caller save on x86-64 (and elsewhere too, I believe, since we can then just exploit the spill mechanism) 16:00:26 deepfire: I'm not sure what I feel about paypal. 16:00:35 namor [n=namor@dslb-088-072-230-074.pools.arcor-ip.net] has joined #lisp 16:00:37 alley_cat [i=AlleyCat@sourcemage/elder/alleycat] has joined #lisp 16:00:38 pkhuong: Yeah, that sounds familiar. 16:02:11 -!- attila_lendvai [n=ati@catv-89-134-66-153.catv.broadband.hu] has quit [Read error: 110 (Connection timed out)] 16:02:53 Trying to see how value cells could be avoided. 16:03:16 lat [n=lat@125.167.140.159] has joined #lisp 16:03:34 -!- gz_ [n=gz@209-6-18-72.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has quit [] 16:03:50 -!- aerique [i=euqirea@xs2.xs4all.nl] has quit ["..."] 16:04:25 So, this is kindof random, but... The runtime allocates SAPs on the heap when it needs to call into Lisp. I think I know how to stack-allocate them so that they are always aligned properly, at least on 32-bit arches. 16:05:21 -!- mvilleneuve [n=mvillene@LLagny-156-36-4-214.w80-14.abo.wanadoo.fr] has quit ["Lost terminal"] 16:05:30 *deepfire* discovers kickstarter.com, chipin.com and fundable.also-com 16:06:06 -!- mattrepl [n=mattrepl@129.174.97.34] has quit [] 16:09:48 Hmm, my mingw as didn't like -O3 (binutils 2.18.50.20080109), while asked to assemble x86-assem.S 16:11:31 I wonder if it is at all appropriate to tell -O3 to as... 16:14:44 e|f [n=elf@ppp121-44-87-23.lns10.syd6.internode.on.net] has joined #lisp 16:15:19 -!- e|f [n=elf@ppp121-44-87-23.lns10.syd6.internode.on.net] has left #lisp 16:17:03 Maybe my mingw/msys-1.0.11 are too new. 16:18:20 schoppenhauer [n=senjak@unaffiliated/schoppenhauer] has joined #lisp 16:18:26 I'd consider arguing in favor of ditching the C runtime on Win32, but I -know- how much of a pain it'd be. 16:19:00 -!- varjag [n=eugene@122.62-97-226.bkkb.no] has quit ["Leaving"] 16:19:23 _deepfire pasted "another win32 build failure" at http://paste.lisp.org/display/87839 16:19:55 -!- namor [n=namor@dslb-088-072-230-074.pools.arcor-ip.net] has left #lisp 16:19:59 namor [n=namor@dslb-088-072-230-074.pools.arcor-ip.net] has joined #lisp 16:19:59 -!- redline6561 [n=redline@gate-20.spsu.edu] has quit [Read error: 110 (Connection timed out)] 16:20:01 -!- namor [n=namor@dslb-088-072-230-074.pools.arcor-ip.net] has quit ["Verlassend"] 16:20:12 Yeah, I'd argue that -O3 isn't really applicable to assembly code anyway. 16:20:21 namor [n=namor@dslb-088-072-230-074.pools.arcor-ip.net] has joined #lisp 16:22:46 as I understand it, O3 is usually considered best avoided (in the GNU toolchain) 16:24:05 So... why are we using it all over the place in SBCL? 16:24:20 no idea :) 16:24:44 *rsynnott* has seen a lot of posts and articles by gcc people more or less saying 'please don't do this', though 16:25:30 If it's so highly disrecommended, then, why is it still there? 16:25:31 AIUI, -Os is best for branchy code, due to decreased icache footprint. 16:25:31 rsynnott: if you can debug the asm output and downgrade to -O2 on buggy hosts, why not? 16:26:17 deepfire: it's even less stable than O3. I don't see how packing branches together helps though; You're only putting more pressure on the predictors (especially the BTB). 16:27:15 pkhuong, afaik people routinely build the linux kernel with -Os, so should be stable enough.. 16:27:30 No personal experience, though. 16:28:17 Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has joined #lisp 16:28:53 how many of them would be able to recognize a kernel bug? How much of kernel's code is executed on their boxes? 16:29:28 -!- Blaay is now known as Blay 16:29:32 I don't have any specific knowledge about gcc, so shutting up :-) 16:30:39 pkhuong, the -Os thing is about the embedded community, and these people are usually fairly competent. 16:30:40 -!- rstandy [n=rastandy@host193-174-static.89-94-b.business.telecomitalia.it] has quit [Read error: 110 (Connection timed out)] 16:30:54 -Os is routine on embedded linux targets, but there memory is more important than branch prediction and stuff. can't see why would anyone use it on x86 16:31:25 it has at various points in time generated faster code on x86 too 16:31:30 deepfire: and hopefully very much able to detect and fix compiler bugs. 16:31:33 (although I'm sure someone somewhere does) 16:32:42 foom, yeah I was about to say that too, but more assertions from me at this point would've been detrimental.. 16:33:16 Memory latencies are only going to increase. 16:33:23 -!- schoppenhauer [n=senjak@unaffiliated/schoppenhauer] has quit ["Leaving."] 16:33:31 So icache footprint is really really important. 16:33:32 deepfire: basically I don't find the argument that -Os is fire and forget because experts use it very convincing. 16:33:37 SandGorgon [n=OmNomNom@122.162.50.31] has joined #lisp 16:33:46 ASau [n=user@83.69.227.32] has joined #lisp 16:34:14 namor_ [n=namor@dslb-088-074-053-204.pools.arcor-ip.net] has joined #lisp 16:34:23 deepfire: caches too are getting larger. Moreover, if you're focusing on code size, it's really *bandwidth* that you're using better. 16:34:36 -!- namor [n=namor@dslb-088-072-230-074.pools.arcor-ip.net] has quit [Nick collision from services.] 16:34:44 -!- namor_ is now known as namor 16:34:46 Nuclear fission is perfectly safe: Experts use it, after all... 16:34:52 demmel [n=Adium@dslb-094-216-093-225.pools.arcor-ip.net] has joined #lisp 16:35:31 -!- danlei [n=user@pD9E2C1CC.dip.t-dialin.net] has quit [Remote closed the connection] 16:35:46 pkhuong, if your code fits 2 cachelines instead of three, then you won't have to reload that third cacheline when you need it. 16:35:59 So it's about latency as well. 16:36:15 only if you defeat prefetch. 16:36:20 on branchy code? 16:37:04 That's the _whole_ point why amd cpus with embedded memory controllers are good for compilers. 16:37:14 Meh. Just fit your entire program into the I-cache. Problem solved. 16:37:26 Because for branchy code your prefetch is near useless. 16:37:55 nyef: which keeps getting easier with larger caches. 16:38:20 at this point those discussions start resembling political ones, as far as avalable experimental evidence and applicability thereof is concerned :) 16:38:22 Also gets easier with smaller programs. :-p 16:38:51 cmm-, the stuff I'm talking about is very well known, btw. 16:38:56 -!- ljames [n=ln@unaffiliated/ljames] has quit [Remote closed the connection] 16:39:48 To the point that it's just all common sense, tbh. 16:40:24 it's common sense that code size can be an issue. What's not is that optimising for size will help on branchy code. 16:40:35 Numlock [n=resteven@igwe19.vub.ac.be] has joined #lisp 16:40:35 this common sense is not quite universal, see 16:40:45 fe[nl]ix, does this look correct: (:shadowing-import-from #:collate . #. (collate::external-symbols)) 16:40:45 the_unmaker [n=g@w005.z209031033.sjc-ca.dsl.cnc.net] has joined #lisp 16:40:58 so i am trying out stumpwm 16:41:11 hi gavino 16:41:15 not sure how happy I am with the default keybindings 16:41:18 hey 16:41:23 -!- the_unmaker is now known as gavino 16:41:25 what sup 16:42:05 lat: i don't think it is a good idea. i would prefer an explicit list. 16:42:27 that was a service announcement, not a greeting :) 16:42:28 we're looking for a program such that the instructions' working set fits in I$ with Os but not with O3, and such that the access patterns are not prefetchable. 16:44:36 pkhuong, that's not the point 16:44:44 They should be able to write such a thing. 16:44:49 Could someone give me a hint why my esa using mcclim/Backends/gtkairo initially shows a blank white window until some user interaction takes place? 16:45:17 Even going so far as to specify a fixed branch pattern that explicitly defeats the prefetcher. 16:45:32 pkhuong, you are overconstraining 16:45:44 Xach, ok. Thanks. 16:46:32 -!- thehcdreamer [n=thehcdre@94.198.78.26] has quit [] 16:47:04 -!- Reav__ [n=Reaver@212.88.117.162] has quit [Read error: 104 (Connection reset by peer)] 16:47:11 pkhuong, fit icache vs. not fit icache by a significant margin == huge difference 16:47:47 pkhuong, what you are aiming at is a meaningful quantitative difference 16:48:04 so just "fits better" is good enough to be concerned about it 16:48:17 emacs-dwim: a delicious and easily fixed mcclim bug, if I had to guess (which I am) 16:48:21 dv_ [n=dv@85-127-111-103.dynamic.xdsl-line.inode.at] has joined #lisp 16:49:07 -!- trebor_dki [n=user@mail.dki.tu-darmstadt.de] has quit [Read error: 60 (Operation timed out)] 16:49:14 -!- mrsolo [n=mrsolo@adsl-68-126-198-233.dsl.pltn13.pacbell.net] has quit [Read error: 110 (Connection timed out)] 16:52:25 hefner: How much money would you charge to fix it? 16:56:14 are there any plans to implement the Web backend for McClim? 16:56:40 -!- seisatsu [n=seisatsu@adsl-63-198-106-143.dsl.snfc21.pacbell.net] has quit [Read error: 110 (Connection timed out)] 16:56:47 emacs-dwim: Of course! The question is when, how, and who. 16:57:19 emacs-dwim: In fact, the web backend already exists; written by the SRI people. It's just a matter of integrating it into McCLIM. 16:57:27 OmniMancer [n=OmniManc@222-154-179-43.jetstream.xtra.co.nz] has joined #lisp 16:57:33 isn't that web backend based on image maps? 16:57:41 no idea. 16:57:43 I have a quarter-finished canvas backend 16:57:50 and we have an eighth-finished svg backend here too somewhere 16:58:19 beach: if it's the one that's been around since the mid 1990s, it is (because clients then weren't smart enough to do anything else) 16:58:27 I think that makes it somewhat hard to deal with things like keystrokes 16:58:29 slyrus_ [n=slyrus@dsl092-019-253.sfo1.dsl.speakeasy.net] has joined #lisp 16:58:38 possibly 16:59:16 -!- silenius [n=jl@yian-ho03.nir.cronon.net] has quit [] 16:59:21 beach: it's hard to say without investigating whether it would take an hour or a whole day (but in the latter case, I'd probably learn something useful along the way). I'll pull the figure of $50 USD out of a hat. 16:59:37 hefner: Deal. I'll pay you. 16:59:43 -!- SandGorgon [n=OmNomNom@122.162.50.31] has quit [Read error: 60 (Operation timed out)] 17:00:36 redline6561 [n=redline@gate-20.spsu.edu] has joined #lisp 17:00:47 hefner: Send me your bank data (IBAN, SWIFT, holder name, etc) and I'll transfer as soon as I get around to it. 17:00:54 deepfire: reducing code size has two major effects: asymptotically, it's about improving bandwidth use; at the low-end, it's about fitting in I$ (especially L1, which tends to have decoding hints). 17:01:32 *beach* is told by wife to come have dinner. BBL. 17:01:49 icecube [n=icecube@p549C3509.dip.t-dialin.net] has joined #lisp 17:02:17 -!- icecube is now known as Guest17231 17:02:59 A minor effect is that of fitting a larger portion of the code in cache. On bad access patterns, multiplying a cache's size by k tends to improve cache hits by ~sqrt(n). Thus, assuming hard to prefetch code, getting code to be x times as large will increase hits by ~1/sqrt(x). The sqrt quickly decreases the effect. For example, if code is 75% as large, you're looking at ~16% more hits... Bandwidth, is virtually increased by 33% (in fraction of code/time). 17:03:00 -!- Guest17231 is now known as icecube_ 17:03:40 -!- Ralith [n=ralith@69.90.49.189] has quit [Read error: 113 (No route to host)] 17:03:56 can you make asdf load a file if it is existent but ignore it if it is not? 17:04:25 I can't, but it's probably -possible-. 17:04:30 Not necessarily easy, though. 17:04:35 darn, I left my cl.net key on the wrong side of the planet. 17:05:47 salva [n=salva@105.11.117.91.dynamic.mundo-r.com] has joined #lisp 17:06:15 nyef: fair enough, I'll note that as a 'no' 17:06:40 madnificent: you can extend the file component, shouldn't be too hard. 17:06:48 -!- ryepup1 [n=ryan@216.155.105.115] has left #lisp 17:06:58 madnificent: perhaps you're looking for :weakly-depends-on? 17:07:16 tcr: ah, that could be the thing :) thanks 17:07:37 luis: that too, but it somehow feels like overkill 17:07:56 Basically :weakly-depends-on '(:foo), then #+foo inside your code (if FOO pushes its name onto *features*) 17:08:14 (without the quote, of course) 17:08:16 danlei [n=user@pD9E2C1CC.dip.t-dialin.net] has joined #lisp 17:09:11 -!- Addled [n=addled@209.20.68.236] has left #lisp 17:09:23 rstandy [n=rastandy@net-93-144-193-128.t2.dsl.vodafone.it] has joined #lisp 17:10:05 -!- lichtblau [n=user@port-92-195-97-60.dynamic.qsc.de] has quit [Read error: 110 (Connection timed out)] 17:10:38 Yoy, seems like cliki is online again. 17:10:39 tcr: it would be to allow for user modifications. But yes, I could use features if it would provide me with extra services. 17:10:40 -!- emacs-dwim [n=user@cpe-74-71-11-230.twcny.res.rr.com] has quit [Remote closed the connection] 17:11:55 Ñ/me throws in http://lkml.indiana.edu/hypermail/linux/kernel/0302.0/1068.html 17:12:00 blah 17:12:21 (yes, only because I have nothing specific to say) 17:12:23 -!- cpc26 [n=cpc26@72.170.156.242] has quit [Read error: 104 (Connection reset by peer)] 17:13:13 And it introduces interesting new terminology, "repeat rate". 17:13:13 disumu [n=disumu@p54BCE44B.dip.t-dialin.net] has joined #lisp 17:13:17 emacs-dwim [n=user@cpe-74-71-11-230.twcny.res.rr.com] has joined #lisp 17:13:35 -!- emacs-dwim [n=user@cpe-74-71-11-230.twcny.res.rr.com] has left #lisp 17:13:39 emacs-dwim [n=user@cpe-74-71-11-230.twcny.res.rr.com] has joined #lisp 17:14:30 mrsolo [n=mrsolo@nat/yahoo/x-qxhcccbohyktpsst] has joined #lisp 17:14:45 funkenblatt [n=user@adsl-69-238-246-201.dsl.pltn13.pacbell.net] has joined #lisp 17:14:49 -!- redline6561 [n=redline@gate-20.spsu.edu] has quit [Read error: 104 (Connection reset by peer)] 17:15:06 -!- funkenblatt [n=user@adsl-69-238-246-201.dsl.pltn13.pacbell.net] has left #lisp 17:15:21 That thread is interesting.. 17:15:32 redline6561 [n=redline@gate-20.spsu.edu] has joined #lisp 17:16:02 Krystof [n=csr21@158.223.51.76] has joined #lisp 17:16:47 deepfire: the start-up cost effect is related to bandwidth (be it from disk or main memory) more than latency. Bandwidth is getting much better. 17:17:11 "How do you get two piccolo players to play in sync?" "Shoot one." <- I'm reminded of this joke any time I have any thoughts of lispers cooperating to advance a cause. 17:17:24 -!- envi^home [n=envi@220.121.234.156] has quit ["Leaving"] 17:17:27 KingNatoG5_ [n=patrik@84-217-13-106.tn.glocalnet.net] has joined #lisp 17:17:30 pkhuong, I agree, the whole usage of latency was a sloppiness on my side 17:18:00 CallToPower [n=CallToPo@s15229144.onlinehome-server.info] has joined #lisp 17:18:06 The problem is probably better formulated using the bandwidth term. 17:18:51 It's /related/ to reducing the effect of memory latency, but only that. 17:18:57 stassats, thanks for the link, but it still doesn't work for me. I've read all the docs I have found about packages and asdf about 10 times. There are many contradictions and inconsistencies. There are no real examples with real file names and real code. I think I understand the concepts, but ... 17:19:00 lhz [n=shrekz@c-b9aa72d5.021-158-73746f34.cust.bredbandsbolaget.se] has joined #lisp 17:19:11 cpc26 [n=cpc26@72.170.156.242] has joined #lisp 17:19:18 -!- CallToPower [n=CallToPo@s15229144.onlinehome-server.info] has left #lisp 17:19:30 lat: My example used real file names and real code. 17:19:35 lat: Sorry it didn't help. 17:20:22 Interesting, I see a reference of Fedora using -O. 17:20:26 -Os 17:21:43 davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #lisp 17:24:06 -!- Adlai [n=adlai@unaffiliated/adlai] has quit [Remote closed the connection] 17:24:07 -!- xan-afk is now known as xan_ 17:24:17 Adlai [n=adlai@unaffiliated/adlai] has joined #lisp 17:24:27 2005, for kernels, but: http://search.luky.org/linux-kernel.2005/msg97638.html 17:25:58 Davidbrcz [n=david@212-198-78-230.rev.numericable.fr] has joined #lisp 17:26:05 -!- KingNatoG5 [n=patrik@84-217-6-126.tn.glocalnet.net] has quit [Read error: 110 (Connection timed out)] 17:26:05 -!- KingNatoG5_ is now known as KingNatoG5 17:27:12 adeht [n=death@bzq-84-110-245-184.red.bezeqint.net] has joined #lisp 17:27:19 -!- Krystof [n=csr21@158.223.51.76] has quit [Read error: 60 (Operation timed out)] 17:27:56 -!- abeaumont [n=abeaumon@85.48.202.13] has quit [Connection timed out] 17:28:25 ITLB misses are in the picture as well, but not for sbcl-runtime-sized code, indeed. 17:29:37 Ok, there's the 2004 RedHat paper about -Os: http://people.redhat.com/wcohen/ncsu2004/Space%20Optimizations.pdf 17:30:08 "Overall space optimizations proved to be very effective in improving application" 17:30:10 performance. 17:32:11 ljames [n=ln@unaffiliated/ljames] has joined #lisp 17:33:18 emacs-dwim: does this happen with just your app, or with the mcclim demos as well? 17:34:29 deepfire: If a 2.68% reduction in time is very effective. 17:34:42 nha [n=prefect@85.4.174.31] has joined #lisp 17:35:03 -!- ``Erik [i=Here@c-69-140-109-104.hsd1.md.comcast.net] has quit [Read error: 113 (No route to host)] 17:35:18 -!- Jafet [n=Jafet@unaffiliated/jafet] has quit [Read error: 60 (Operation timed out)] 17:35:51 -!- Adlai [n=adlai@unaffiliated/adlai] has quit [Remote closed the connection] 17:36:22 Adlai [n=adlai@unaffiliated/adlai] has joined #lisp 17:36:27 hefner: seems to happen with clim-listener 17:36:58 ska` [n=user@ppp-58-8-239-254.revip2.asianet.co.th] has joined #lisp 17:37:12 Xach, Sorry. I was referring to all the docs I've been reading. I came to your article already very confused. I am thinking that there may be something wrong with my installation. I'll read though all the docs (including yours) once more, and try again tomorrow. 17:37:16 ``Erik [i=Here@c-69-140-109-104.hsd1.md.comcast.net] has joined #lisp 17:40:24 hefner: the demo has some initial display problems too, it seems i can fix my problem by using a top-level method in my application-frame that calls #'redisplay-frame-panes on it. 17:42:38 Fare [n=Fare@ita4fw1.itasoftware.com] has joined #lisp 17:44:32 hefner: similarly, resizing the application frame (with stumpwm's `fullscreen') leaves the painting inconsistent, so i'll probably have to handle the resize event to redisplay also. 17:45:47 johnny_z [n=anonymou@217.24.250.130] has joined #lisp 17:48:09 -!- nha [n=prefect@85.4.174.31] has quit [Remote closed the connection] 17:49:51 -!- ejs [n=eugen@94.178.31.220] has quit [Read error: 145 (Connection timed out)] 17:53:27 -!- lat [n=lat@125.167.140.159] has quit [Read error: 145 (Connection timed out)] 17:56:26 ejs [n=eugen@152-231-124-91.pool.ukrtel.net] has joined #lisp 17:58:14 -!- ericholscher [n=eric@ericholscher.com] has left #lisp 17:59:56 FareWell [n=Fare@ita4fw1.itasoftware.com] has joined #lisp 18:00:12 mattrepl [n=mattrepl@c-76-104-2-182.hsd1.va.comcast.net] has joined #lisp 18:00:57 mathrick [n=mathrick@87.58.113.80] has joined #lisp 18:02:19 -!- demmel [n=Adium@dslb-094-216-093-225.pools.arcor-ip.net] has left #lisp 18:02:35 -!- coyo[sleeping] is now known as coyo 18:02:56 abeaumont [n=abeaumon@84.76.48.250] has joined #lisp 18:03:09 Hmm, updating binutils to 2.19.1 and fixing ASFLAGS didn't get me past x86-assem.S 18:03:47 maybe my mingw installation is bonkers after all 18:04:38 What error are you getting now? 18:05:19 line 91, junk at the end of line, first bad token: '(', something like that 18:05:46 sorry closed the window and gone fishing for a fresh mingw install 18:05:54 Line 91, though? 18:06:08 yes, got me wondering, as there's no ( there 18:06:10 lat [n=lat@125.167.140.159] has joined #lisp 18:06:24 right before the GNAME reference 18:06:25 -!- Edward [n=Ed@81.249.193.178] has quit ["L'oignon fait la farce."] 18:06:33 hey is there a builtin for getting the majority of a collection of values? something like, (majority '(1 2 2 3 3 3)) => 3 18:06:50 Yeah, bet it's not handling .globl right, or... Maybe it's not preprocessing correctly? 18:07:22 stassats` [n=stassats@wikipedia/stassats] has joined #lisp 18:07:28 DeusExPikachu: I don't think so. 18:07:30 nyef, what version of mingw do you have and what do you install, current or candidate? 18:07:47 should be fairly straightforward to write with LOOP, though. 18:07:52 They have recently updated the mingw installer to 5.1.6 18:08:02 sykopomp: yeah, just checking 18:08:03 deepfire: nope, but Boyer & Moore have a very nice algorithm; google for MJRTY. 18:08:18 DeusExPikachu, not deepfire. 18:08:57 mm, wait, what do you want to return when there's no majority value? 18:09:04 I have some version of mingw that I got via the cygwin installer. 18:09:12 -!- johnny_z [n=anonymou@217.24.250.130] has quit [Read error: 113 (No route to host)] 18:09:20 ho hum. 18:09:21 pkhuong: maybe break ties by returning the first mentioned 18:09:48 -!- ejs [n=eugen@152-231-124-91.pool.ukrtel.net] has quit ["This computer has gone to sleep"] 18:10:00 -!- coderdad [n=coderdad@ip72-200-214-240.ok.ok.cox.net] has quit [Remote closed the connection] 18:10:12 My binutils claims to be 2.18.50.20080625. 18:10:35 coderdad [n=coderdad@ip72-200-214-240.ok.ok.cox.net] has joined #lisp 18:10:55 -!- namor [n=namor@dslb-088-074-053-204.pools.arcor-ip.net] has quit [Remote closed the connection] 18:11:15 Urfin [i=foobar@85.65.93.74.dynamic.barak-online.net] has joined #lisp 18:11:26 I had both 2.18.50.20080109 and 2.19.1.something fail on me in the same way. 18:11:37 What about gcc? 18:11:52 namor [n=namor@dslb-088-074-053-204.pools.arcor-ip.net] has joined #lisp 18:12:02 saikat_ [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has joined #lisp 18:12:05 pkhuong: what do you think of thread pooling as a way to handle code that's meant to be heavily-threaded? 18:12:07 gcc (GCC) 3.4.4 (cygming special, gdc 0.12, using dmd 0.125) 18:12:28 sykopomp: what do you mean by "heavily threaded"? 18:12:36 Again, I'm running cygwin and rely on -mno-cygwin to get me a mingw compilation. 18:13:10 pkhuong: code where you would potentially be comfortable wanting to do 10-100k tasks 18:13:27 Umm... While I'm thinking about such things, is anyone here working on the sb-posix build error on snow leopard? 18:13:34 a thread pool won't help. Pooling only makes spawning threads cheaper. 18:13:42 -!- stassats [n=stassats@wikipedia/stassats] has quit [Read error: 145 (Connection timed out)] 18:14:06 nyef: wfm, unfortunately. 18:14:10 I meant more in the sense of limiting the thread pool and queueing up different tasks, to not have to spawn a ton of threads. 18:14:29 'cause we had similar problems with win32 on cygwin because the groveller would grab the cygwin headers instead of the mingw headers if it wasn't headed off properly. 18:14:46 sykopomp: sure, if the taks are mostly independent. 18:14:53 yeah :\ 18:15:27 makes me wish there was a lisp with preemptive microprocs, but that sounds like gigantic mess 18:15:31 So if the structures are different sizes between x86 and x86-64, and the groveller isn't kept in the loop, it could be troublesome. 18:15:50 grrr, don't want to mentally parse the mjrty fortran code, but from the description, it sounds like its implemented how I had it in my mind 18:16:18 DeusExPikachu: it only returns something meaningful if there's a strict majority (> 50%) value. 18:17:34 -!- ThomasIl [n=thomas@unaffiliated/thomasi] has quit ["Bye Bye!"] 18:17:34 seisatsu [n=seisatsu@adsl-63-198-106-143.dsl.snfc21.pacbell.net] has joined #lisp 18:17:44 mcspiff [n=user@DC51C.WPA.Dal.Ca] has joined #lisp 18:18:32 pkhuong: oh wait, that's true... 18:18:40 weirdo [n=sthalik@c142-5.icpnet.pl] has joined #lisp 18:18:54 pkhuong: I was only thinking of the tie case, I guess I should be askign for plurality 18:19:08 [Jackal] [n=jkl2k5@118.94.86.192] has joined #lisp 18:19:20 *hugod* wonders if there is anything like cilk for cl 18:19:49 hugod, you could write one yourself, I suppose 18:19:53 hefner: furthermore, everything works cleanly using clx. and now i notice that gtkairo is not clipping draw- to the drawn pane. 18:20:04 a semi-efficient version should be easy 18:20:15 now to make it super-efficient, is something else... 18:20:35 -!- rstandy [n=rastandy@net-93-144-193-128.t2.dsl.vodafone.it] has quit [Remote closed the connection] 18:20:47 hefner: but the text-size is ignored by clx. 18:21:00 Fare: you have reservations about it's usefulness? or the amount of work to implement it? 18:21:30 hugod, should be 1- useful, and 2- easy enough to implement if you don't care about super-efficiency 18:21:40 emacs-dwim: what are you supplying as a text-size? 18:21:50 to make superefficient, you might need some to add some assembly VOPs to SBCL 18:22:10 or then again, maybe not 18:22:58 The scheduler should be pretty straight forward, but I would need to scratch my head about the spawn/join part 18:23:18 especially if you only target CPU architectures with a lot of memory consistency such as Intel CPUs 18:23:25 hefner: 10 14 :huge 18:23:46 emacs-dwim: it should certainly honor :huge, unless the fonts are completely screwed up. 18:24:11 Addled [n=addled@209.20.68.236] has joined #lisp 18:24:32 for the simple implementation, you have a macro that puts the work to be done in a lambda, packages it with a semaphore to be grabbed by threads that look at this job, and a flag to indicate status. 18:24:50 it's a simple job queuing / dequeueing thing 18:24:57 Hmm. "xprop" gives the WM_NAME of "foobar.lisp" for this window, but (get-properties window :wm_name) shows just "emacs". Do you know why that might be the case? 18:25:34 now for a super-efficient implementation, you to optimize the "normal" path a lot... 18:27:04 What's the "normal" path? I thought it was essentially divide and conquer. 18:27:04 Xach: cache? 18:27:33 hugod, the normal path is when noone did your work for you and you have to do it yourself. 18:27:34 hugod: the normal path is that you don't have to pass the task to another thread. 18:27:50 does anyone want to steal from me work to do? 18:27:54 puh-lease! 18:28:25 hmm. gtkairo does not work as well as I remembered. 18:28:26 Fare: I think meta-hacking has to be subtler than that. 18:28:32 -!- cpc26 [n=cpc26@72.170.156.242] has quit [Read error: 54 (Connection reset by peer)] 18:29:01 *gigamonkey* has to go buy a USB headset--going to be on this week's StackOverflow podcast 18:29:14 nyef: My ultimate goal is to get a list of all as-displayed-by-the-window-manager window titles, and I'm getting stuck because I don't know much about X. clx makes it easy to poke around, but it doesn't seem to be a good exploration/learning tool fully on its own. 18:29:14 lichtblau [n=user@port-92-195-97-60.dynamic.qsc.de] has joined #lisp 18:29:18 cpc26 [n=cpc26@72.170.156.242] has joined #lisp 18:29:28 gigamonkey: good investment, I have one myself..I use it a lot 18:29:31 -!- cpc26 [n=cpc26@72.170.156.242] has quit [Read error: 54 (Connection reset by peer)] 18:29:46 TDT: any [dis]recommendations? 18:29:49 varjag [n=eugene@103.80-202-117.nextgentel.com] has joined #lisp 18:30:05 Xach: Mmm. CLX is just that bit too low-level for that sort of exploration, unfortunately. 18:30:17 gigamonkey: I got the plantonics one...let me get a link 18:30:32 nyef: any suggestions for this particular goal? 18:30:55 gigamonkey: why usb? 18:31:07 gigamonkey: http://www.amazon.com/Plantronics-Foldable-USB-Optimized-470/dp/B000VVXO7E/ref=sr_1_1?ie=UTF8&s=electronics&qid=1254162627&sr=8-1-spell 18:31:11 ... Where's get-properties defined? 18:31:30 stassats`: dunno. It's done over skype; I was told I needed a skype account and a reasonable quality USB headset. 18:31:53 Ah, get-property? 18:32:01 If you use a laptop, the mic ports are many times not powered from my understanding...I tried a non-usb one and it iddn't work for that reason. 18:33:11 Fare: eager-futures has what seems to be a kinda nice thread pooling thing that queues up lambdas for you and shoves them inthreads. 18:33:23 cpc26 [n=cpc26@72.170.156.242] has joined #lisp 18:33:37 hefner: the fonts work fine for gtkairo, emacs, conkeror, xterm, etc. 18:34:02 hugod, looks like eager-futures may already do the basic work for you, then. Now you have to super-optimize it :-/ 18:34:03 silenius [n=jl@dslb-088-073-069-161.pools.arcor-ip.net] has joined #lisp 18:34:36 -!- FareWell [n=Fare@ita4fw1.itasoftware.com] has quit ["Leaving"] 18:35:13 mmm, and add the analysis for determining granularity... 18:35:33 nyef: ah, i think i found what i need. 18:35:39 Xach: Oh! How are you getting the window to examine? 18:36:02 nyef: query-tree. too shallow. 18:36:06 Just thinking that you might be querying the window manager frame and not the application window. 18:36:21 yeah, i think that's the case. 18:36:26 hugod, instead of a per-object lock, you could have a per-thread lock. 18:36:28 hugod: cheap task creation & stealing will get you most of the way, and doesn't require much static analysis 18:37:09 -!- mcspiff [n=user@DC51C.WPA.Dal.Ca] has quit [Read error: 60 (Operation timed out)] 18:37:28 on Intel CPUs you might look at the memory model, and could possibly get the usual case to work without any lock at all. 18:37:57 what more, looking at how CILK does it itself might tell you how to do it with SBCL. 18:38:09 Athas` [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has joined #lisp 18:38:13 emacs-dwim: just wanted to make sure you weren't seeing everything in one generic font ("fixed" or whatever), given the trend toward sabotaging old X11 apps by senseless screwing with the selection of installed fonts 18:38:27 Ralith [n=ralith@69.90.49.189] has joined #lisp 18:38:33 *Fare* files a trivial bug on the sbcl-devel mailing-list. 18:39:17 pkhuong: yes, controlling the granularity would be icing on the cake 18:40:20 Fare: Your workaround, btw, is to locally notinline compile-file and compile-file-pathname. 18:41:33 nyef, oh thanks 18:41:43 Sadly, in 1.0.11 mingw reverted to windows's native console from the rxvt terminal, which is a 10x-times tty output speed regression. 18:42:19 fare: I'm also looking at jsr166 18:42:21 my workaround is to call (apply 'compile-file ... (when use-cfasl `(:emit-cfaslt ...))) 18:42:43 hugod, conceptually it's not that hard, but for extreme performance, the devil is in the details. 18:43:01 (I mean, performance without ugly race condition) 18:43:06 deepfire: also a big loss of usability, I'd guess. I recall having a reasonable terminal as my main reason for using mingw over cygwin (given I had both installed) 18:44:43 hefner, I can't remember if I stumbled upon any difference, when I went 1.0.10->1.0.11 18:45:15 There might have been, but I don't remember.. 18:45:24 well, the windows console is a hassle to resize, and to select and paste text in. 18:45:30 -!- redline6561 [n=redline@gate-20.spsu.edu] has quit [Success] 18:46:30 johnny_zetta [n=anonymou@217.24.250.130] has joined #lisp 18:47:38 OK, I give up; how do you invoke recent firefox to create a new profile? 18:48:11 it used to be firefox -ProfileManager, but that no longer works. 18:48:13 beach: you temporarily mv you .mozilla away? 18:48:29 Oh, dear. 18:48:53 stumpwm kinda cool 18:48:55 gettign hang of it 18:49:15 -ProfileManager looks like it's working here (but I'm using 3.0.6) 18:50:16 attila_lendvai [n=ati@adsl-89-132-6-198.monradsl.monornet.hu] has joined #lisp 18:50:17 hefner, and the scrollbar stalling output, and no shift+pgup/pgdn... my memory is flakey 18:51:21 Ok, it appears that MSYS 1.0.11 was the problem, 1.0.10 builds sbcl fine here. 18:51:40 Fare: is your meta parser able to parse something like the HTTP-date grammar in this rfc: http://www.lyberty.com/meta/date_format.html ? i mean, should i keep on reading the docs or i badly miss something and it's not the right tool for that? 18:51:59 <[Jackal]> hefner: http://sourceforge.net/projects/console/ 18:52:00 i want to get rid of the net-telent-date dependency 18:52:22 <[Jackal]> deepfire: MSYS 1.0.11 works fine here with latest updates. 18:53:06 <[Jackal]> Except for doc/manual. 18:53:08 Jackal, maybe it was that my installation regressed in some ways, then. 18:53:31 -!- Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has quit [Success] 18:53:39 <[Jackal]> deepfire: Quite possible. It's a mess, if I'm allowed to speak honestly. 18:53:47 attila_lendvai, it's just a simple LL(1) parser 18:54:37 that you could extend to LL(k) without too much difficulty by implementing some look-ahead buffering. More than that would require backtracking, which is something else. 18:54:39 Fare: and do you have experience with the parser generators, any recommendations? 18:54:55 anyway, Now I'm a happy user of an de-kitten-of-death-ified SBCL! 18:55:03 -!- seangrove [n=user@98.248.37.94] has quit [Read error: 145 (Connection timed out)] 18:55:06 btw, the code is initially by Jochen Schmidt 18:55:33 apart from the general suckiness of parser generation, no I don't have much to recommend. 18:55:42 Packrat parsers were all the hype recently 18:55:42 i was looking at cl-yacc, but to be honest i was expecting a tool to which i just reformat the grammar and emits a parser... 18:55:55 LALR were the previous hype 18:56:07 oh, right, nikodemus has some parser code... 18:56:13 a packrat thingie 18:56:23 you could write a parser for the other grammar, that dumps a cl-yacc grammar for you... 18:56:31 *deepfire* is somewhat sorry for having to strip out that nyef's piece of art, but our product placement folks want it that way. 18:56:32 (hiding) 18:56:46 if what you have is LL(1) by all means use META 18:57:01 Fare: err, originally i allocated about 30 mins for this... seems like i was too optimistic 18:57:06 deepfire, what did you do to it? 18:57:33 Fare, you won't believe it -- I gave it cookies and it went away! 18:57:34 *attila_lendvai* looks up nikodemus' parser 18:57:41 attila_lendvai: quick perec question: is there a way to declare unique indexes on more than one column? 18:58:11 These were special kitton-of-death cookies, mind you, hard to come by.. 18:58:19 *kitten 18:58:31 i.e. a page has a list of files (attachments); file names must be unique through that page, so I'd specify UNIQUE INDEX (page, filename) 18:59:52 mishoo: don't know, my perec knowledge is more on the user level... i just have the commit bit... :) 19:00:47 deepfire, you're placing sbcl??? 19:00:52 attila_lendvai: ouch... I thought you're one of the main developers :) 19:01:07 mishoo: cl-rdbms supports creating indexes and there's a way to get the rdbms column names from perec, but i'm not sure there's a ready made feature for that 19:01:20 gz [n=gz@static-72-85-235-154.bstnma.east.verizon.net] has joined #lisp 19:01:29 Fare, more than just sbcl, of course. 19:01:44 mishoo: i am part of the 3 guy team, but that's not my territory. i only take part in the occasional brainstormings and some smallscale fixes 19:01:57 Erm, sorry, I don't, I just write stuff. 19:02:01 -!- alley_cat [i=AlleyCat@sourcemage/elder/alleycat] has quit ["Ex-Chat"] 19:02:06 attila_lendvai: ok thanks, I'll investigate cl-rdbms and/or send an email to the list.. 19:02:19 ruediger_ [n=ruediger@188-23-95-154.adsl.highway.telekom.at] has joined #lisp 19:02:55 nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 19:03:59 HET2 [n=diman@cpc1-cdif12-2-0-cust125.5-1.cable.virginmedia.com] has joined #lisp 19:04:21 slash_ [n=Unknown@p5DD1C6BC.dip.t-dialin.net] has joined #lisp 19:06:05 this is really sad. i've though that CS is old enough to have tools for which i just pipe in a grammar and they emit an optimized parser... maybe i should write a clean and simple grammar interpreter and use our partial evaluator to generate an optimized parser. will not fit in the 30 min time frame i was planning to spend on it though... 19:08:38 -!- kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 19:09:16 what will you use it for? 19:09:35 at the moment http date parsing 19:10:56 -!- emacs-dwim [n=user@cpe-74-71-11-230.twcny.res.rr.com] has quit [Remote closed the connection] 19:12:17 sayyestolife [n=jot_n@h-60-147.A163.priv.bahnhof.se] has joined #lisp 19:12:44 -!- splittist [n=dmurray@85.4.220.156] has quit ["ETOOMUCHGAVINO"] 19:14:18 emacs-dwim [n=user@cpe-74-71-11-230.twcny.res.rr.com] has joined #lisp 19:15:18 http date, hmm 19:15:47 -!- mattrepl [n=mattrepl@c-76-104-2-182.hsd1.va.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 19:15:57 dandersen [n=dkcl@metabug/dandersen] has joined #lisp 19:15:57 mattrepl [n=mattrepl@c-76-104-2-182.hsd1.va.comcast.net] has joined #lisp 19:16:09 you're using a full-blown parser-generator for http date parsing? yikes. :P 19:16:10 -!- disumu [n=disumu@p54BCE44B.dip.t-dialin.net] has quit ["..."] 19:16:50 -!- dandersen is now known as dkcl 19:17:14 no, i'm doing The Right Thing, because i'm in that mood now... and it's fun, too 19:17:42 attila_lendvai: a classic demo of "perfect superoptimisation" is the extraction of boyer-moore from the naive O(n^2) description, specialised on a given needle. 19:17:54 -!- Dawgmatix_ [n=dman@c-68-32-44-191.hsd1.nj.comcast.net] has quit [Remote closed the connection] 19:18:00 *perfect supercompilation 19:18:10 Dawgmatix_ [n=dman@c-68-32-44-191.hsd1.nj.comcast.net] has joined #lisp 19:18:35 -!- mathrick [n=mathrick@87.58.113.80] has quit [Read error: 145 (Connection timed out)] 19:18:41 more duct tape 19:19:01 pkhuong: i need google to read you... :) 19:20:06 -!- ruediger [n=ruediger@91-115-181-187.adsl.highway.telekom.at] has quit [Read error: 110 (Connection timed out)] 19:20:15 pkhuong: so, if you were in a "The Right Thing" mood, how would you parse the http-date grammar in this: http://www.lyberty.com/meta/date_format.html ? just a sketch... 19:20:32 -!- Adlai [n=adlai@unaffiliated/adlai] has quit [Read error: 148 (No route to host)] 19:20:47 and i'd prefer a solution that results in reusable tools 19:23:06 -!- dstatyvka [i=ejabberd@pepelaz.jabber.od.ua] has left #lisp 19:24:11 http://labs.core.gen.tr/repos/core-server/src/rfc/2616.lisp 19:24:19 I'd have written it like: http://twistedmatrix.com/trac/browser/trunk/twisted/web2/http_headers.py#L163 but that's probably not The Right Thing. :) 19:24:39 parsec like approach 19:25:02 impulse32 [n=impulse@bas3-toronto48-1176440857.dsl.bell.ca] has joined #lisp 19:25:05 fxr: :) 19:25:06 actually we don't know about parsec when we were writing it. 19:25:51 v0|d: galneryus - new legend ;) 19:25:59 fxr: everlasting. 19:26:28 fxr: 19:26:30 ops 19:26:40 fxr: http://www.youtube.com/watch?v=GZlIJn6OoVo see the costume. 19:27:08 v0|d: queued 19:27:39 -!- [Jackal] [n=jkl2k5@118.94.86.192] has quit [Read error: 104 (Connection reset by peer)] 19:28:23 redline6561 [n=redline@gate-20.spsu.edu] has joined #lisp 19:29:44 v0|d: are there any parser-generators in cl world except core-server? 19:30:06 attila_lendvai: I'd use CL-PPCRE or CL-IRREGSEXP. 19:30:34 uhm s/generators/combinators/ 19:30:50 hefner: thanks. i had to install some basic X fonts and restart stumpwm. now clx works nicely. 19:31:46 fxr: don't know. 19:31:50 fxr: maybe someday. 19:33:31 pkhuong: hrm, well, i wanted to use a cannon for that bird... but thanks, maybe i should really just take the easy path... 19:33:50 v0|d: any functional markup combinators? 19:34:31 foom: you duct-taper! 19:35:35 SandGorgon [n=OmNomNom@122.162.50.31] has joined #lisp 19:35:39 fxr: ppl generally use macros. 19:35:44 Xach: I'll take that as a compliment. :) 19:37:13 -!- fiveop [n=fiveop@g229085177.adsl.alicedsl.de] has quit ["humhum"] 19:37:19 Adlai [n=Adlai@unaffiliated/adlai] has joined #lisp 19:38:37 -!- Makoryu [n=vt920@pool-74-104-123-150.bstnma.fios.verizon.net] has quit ["Absquatulandus sum"] 19:38:56 fxr: http://www.infoq.com/interviews/armstrong-peyton-jones-erlang-haskell 19:40:03 antoszka [n=antoszka@chello087206024243.chello.pl] has joined #lisp 19:40:33 do you know that the message passing style in erlang is similar to units? 19:41:54 am I blind or is there no instruction on how to get sbcl from their version control system on sbcl.org anymore? 19:42:02 -!- gigamonkey [n=user@99.58.30.58] has quit [Remote closed the connection] 19:43:19 v0|d: this video starts with a laugh :) you know, witty spj :) 19:43:25 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 19:43:27 -!- davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 19:43:29 fxr: love the hair style. 19:44:25 is this the official SBCL git repo? http://repo.or.cz/w/sbcl.git or where can I fetch the newest code? 19:45:23 ruediger_: I wish there will be a bot for this kind of questions... 19:45:24 ruediger_, it's official enough AFAICT - it follows the official CVS 19:45:40 the official is still the same CVS on SF.net. 19:45:51 oh doh :D 19:45:52 ruediger_: I use git://sbcl.boinkor.net/sbcl.git 19:46:04 I thought they moved to git a while ago 19:46:06 the git repos generally sync that after every change, so you can use git if it strikes your fancy 19:46:07 marioxcc [n=user@200.92.168.65] has joined #lisp 19:46:49 fxr: you don't need a bot. the SBCL guys should just put the checkout instructions on their homepage (again) :p 19:47:31 ruediger_: that will be fine :) what about the future questions? 19:47:49 which future questions? 19:47:50 fxr: Perhaps you can answer. 19:49:14 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [] 19:50:58 gigamonkey [n=user@adsl-99-58-30-58.dsl.pltn13.sbcglobal.net] has joined #lisp 19:51:17 lukego: herep 19:51:23 kib2 [n=kib2@bd137-1-82-228-159-28.fbx.proxad.net] has joined #lisp 19:51:26 howdy 19:51:37 You still unhappy with your Mac? 19:51:48 Want to trade. This Ubuntu POS just Doesn't Work Right. 19:52:08 yes, but I've ordered some thermal paste and been clued into how to get under the heat sink, so I'm hopeful of fixing it up 19:52:33 francogrex [n=user@91.177.17.237] has joined #lisp 19:52:37 Xach: I already did. 19:52:39 which is your laptop? 19:52:45 Actually I wonder if my machine has a similar problem--sometimes it's super snappy but a lot of the time it's not. And the fan runs a lot. 19:52:50 A System76 Pangolin. 19:53:22 schoppenhauer [n=christop@unaffiliated/schoppenhauer] has joined #lisp 19:53:34 It also kernel panics a fair bit. Which is not so cool. 19:53:47 -!- Levenson [n=Levenson@92.46.86.8] has quit [Remote closed the connection] 19:54:23 Isn't the whole point of owning a Mac that you can just take it to the Apple store and let them screw around with thermal paste, etc. 19:54:30 Or give you a new one? 19:54:55 I've ignored this problem for a long time, well outside warranty 19:55:07 Ah. 19:55:09 I thought the point of owning a Mac is so that you can play "artsier than thou" with the PC-users? 19:55:43 nyef: I thought it was about having a unix-enough mit sound. 19:55:58 also I've had pretty crappy experience of apple store support. in the USA I had to wait for ages while the "geniuses" walked around in baggy jeans high-fiving each other. in france they didn't even have a screwdriver, offered to post my laptop off to Holland for someone to open 19:56:40 pkhuong: I think it's Unix-enough and general stuff Just Works. 19:56:48 besides if I screw it up it's the perfect excuse to get a new one.... 19:56:50 For example Skype with my new USB headset. 19:57:53 Anyone on skype want to chat so I can test things out. gigamonkey on skype. 19:58:37 gigamonkey: I can 19:58:37 sure 19:58:50 -!- fe[nl]ix [n=algidus@88-149-211-151.dynamic.ngi.it] has quit ["Valete!"] 19:59:52 has a new clisp 2.48 install on archlinux with 2.6.31-bfe kernel 2838m free ram on 2xp4 3.0ghz cpu 65g free disk and some free time! 20:00:10 -!- SandGorgon [n=OmNomNom@122.162.50.31] has quit [Read error: 60 (Operation timed out)] 20:00:12 bfe for those in the crowd is the new brainfuck scheduler 20:02:04 -!- scottmaccal [n=scott@169.244.94.54] has quit ["Leaving."] 20:04:19 -!- francogrex [n=user@91.177.17.237] has quit ["ERC Version 5.2 (IRC client for Emacs)"] 20:06:16 gavino, gni? 20:07:36 -!- gz [n=gz@static-72-85-235-154.bstnma.east.verizon.net] has quit [Read error: 110 (Connection timed out)] 20:10:02 gni? 20:10:06 what is that? 20:11:02 has anyone made a lisppress like wordpress but with files and a lisp interpreter? 20:11:55 fe[nl]ix [n=algidus@88-149-211-151.dynamic.ngi.it] has joined #lisp 20:12:08 Hrm... Who was the last person to overhaul src/runtime/interrupt.c? 20:13:44 Ah. Gabor Melis was the last to make significant changes. 20:14:46 -!- sayyestolife [n=jot_n@h-60-147.A163.priv.bahnhof.se] has quit [] 20:14:53 -!- spilman [n=spilman@ARennes-552-1-98-108.w92-139.abo.wanadoo.fr] has quit ["Quitte"] 20:16:11 -!- male_terran [n=KVIrc@ip98-162-161-191.pn.at.cox.net] has quit [Read error: 104 (Connection reset by peer)] 20:20:32 GrayMagiker [n=steve@c-68-54-2-128.hsd1.nm.comcast.net] has joined #lisp 20:21:40 What does SBCL use to derive the external format for standard I/O streams? 20:23:38 -!- Jasko [n=tjasko@209.74.44.225] has quit [Read error: 145 (Connection timed out)] 20:23:51 deepfire: i believe it goes by your locale environment. 20:23:57 Edward_ [i=Ed@AAubervilliers-154-1-79-36.w81-249.abo.wanadoo.fr] has joined #lisp 20:24:12 Xach, oh, I forgot to add a peculiar detail: I'm on win32.. 20:24:19 -!- hefner [n=hefner@ppp-58-11-44-79.revip2.asianet.co.th] has quit [Read error: 110 (Connection timed out)] 20:24:32 I grepped through $(env), found nothing.. 20:25:10 SBCL helpfully provides me with a cp866 default :-) 20:25:19 -!- xan_ is now known as xan-afk 20:25:39 mathrick [n=mathrick@130.226.70.177] has joined #lisp 20:26:05 SandGorgon [n=OmNomNom@122.162.50.31] has joined #lisp 20:26:40 -!- antoszka [n=antoszka@unaffiliated/antoszka] has quit [Read error: 104 (Connection reset by peer)] 20:26:48 at least setting LANG doesn't seem to affect anythign 20:27:10 -!- dkcl [n=dkcl@metabug/dandersen] has quit ["leaving"] 20:27:13 probably on win32 it's using the machine's default codepage 20:27:17 not that I've looked at the code 20:27:23 but that would be the sensible thing to do 20:28:43 -!- Ralith [n=ralith@69.90.49.189] has quit [Read error: 113 (No route to host)] 20:28:45 ikki [n=ikki@200.95.162.30] has joined #lisp 20:31:26 kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 20:31:28 davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #lisp 20:32:55 I believe that, yes, on Win32 it does look up the default codepage. 20:32:59 -!- adeht [n=death@bzq-84-110-245-184.red.bezeqint.net] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 20:33:10 Any StackOveflow readers here? I need to pick a SO question to talk about in the podcast tomorrow. 20:33:28 Feel free to email or pm me. 20:33:47 Also note that SBCL invariably uses the "A" versions of various functions instead of the "W" versions. 20:34:04 nyef, what's A vs W? 20:34:18 ASCII vs. Wide-char, presumably. 20:34:40 gigamonkey: what is this podcast? 20:34:51 Basically, the "A" versions take 8-bit characters and the "W" versions take 16-bit characters. 20:35:05 Jeff Spolsky and Jeff Atwood record a weekly 1-hour podcast which they post on Stack Oveflow. 20:35:15 The A versions take characters in the default codepage, and the W versions take UTF-16. 20:35:16 Sometimes it's just them. Sometimes they have a guest. 20:35:25 And note that UTF-8 is never the default codepage 20:35:44 (so, basically, the A versions shouldn't ever be used.) 20:36:21 oh - so SBCL is doing the wrong thing because it was easier to bootstrap that way, but should eventually be fixed? 20:36:46 You end up having two versions of command-line argument processing; the posix stuff, and then the windows stuff with the W stuffix. At least that's what happened with ccl. 20:37:05 if someone wants to fix things up 20:37:16 the easiest way would be to transcode from "W" variant to utf-8 20:37:25 and then tell sbcl that utf-8 is the system charset always. :) 20:38:15 you can of course go directly from utf-16, if you're a tiny bit more adventurous. 20:38:17 Honestly, SBCL/Win32 has sufficiently-worse problems that fixing this up isn't even on the radar. 20:38:21 what about making utf-16le (or be, whichever it is) the system charset enocding? 20:39:00 nyef, apart from threads, what are the problems? the fixed vs mobile memory map? 20:39:09 nyef: sure, but maybe someone really wants to fix this. :) 20:39:51 Fare: Threads, fixed memory map, line ending control in external-formats, the ability to take an interactive interrupt at console or via SLIME... 20:39:57 Umm... 20:40:03 I'm sure there was something else as well. 20:40:37 Oh, right. Serve-event. 20:41:15 yeah, I suppose that's damning enough. But who's gonna fund sbcl for windows? 20:41:27 maybe if you promise a .NET backend, you can get money from Microsoft? 20:41:39 heh. 20:41:42 Apparently there's already 500 euros pledged towards thread support. 20:41:45 levy_ [n=levy@94.44.21.20] has joined #lisp 20:41:48 -!- postamar [n=postamar@75-119-241-70.dsl.teksavvy.com] has quit [] 20:42:05 and then sneak in a LLVM backend at the same time as the .NET one. :) 20:42:18 -!- silenius [n=jl@dslb-088-073-069-161.pools.arcor-ip.net] has quit ["Leaving"] 20:43:33 for the non-enforcing Makefile target of XCVB, should I use POIU, or extend the XCVB driver to do like POIU does? 20:44:33 if people in general or interested in using POIU, I'd say POIU. But if no one is interested, I might as well work on the XCVB driver. 20:45:27 -!- levy_ [n=levy@94.44.21.20] has quit [Remote closed the connection] 20:46:18 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 20:46:36 Can i call RETURN from within a PROGN? 20:46:37 btw, I cleaned up POIU, and you should be able to asdf:load-system it. 20:46:53 slash_, use BLOCK 20:47:17 (block nil ... (return 42) ...) 20:47:30 -!- redline6561 [n=redline@gate-20.spsu.edu] has quit [Read error: 110 (Connection timed out)] 20:48:20 slash_: you can also use PROG, but you'll probably get arrested. 20:48:46 heh 20:48:50 -!- namor [n=namor@dslb-088-074-053-204.pools.arcor-ip.net] has quit [Read error: 110 (Connection timed out)] 20:49:34 well, block has an implicit PROGN, what's the problem? 20:49:55 lispm [n=joswig@e177157223.adsl.alicedsl.de] has joined #lisp 20:50:12 -!- Addled [n=addled@209.20.68.236] has left #lisp 20:50:25 I have something like: (do (...) (if "some condition" (progn (format t ...") (return))... 20:50:44 Sigh, the function with cyrillic characters piped through iconv --from utf8 --to cp866 still cannot print to *stdout* with :cp866 exformat. 20:50:51 levy_ [n=levy@89.223.239.162] has joined #lisp 20:51:06 (do () (condition 10) (format t "")) ? 20:51:13 slash_: why are you calling RETURN at the end of a PROGN? 20:51:17 *deepfire* headwalls 20:51:24 "~{~}" 20:51:32 be aware that DO establishes an implicit block named NIL 20:51:37 Because i want to leave the loop? 20:51:55 ah, ok. So that does make sense. 20:52:10 slash_: do has already a condition for leaving 20:52:13 well, your DO syntax is bad, though. 20:52:16 clhs do 20:52:16 http://www.lispworks.com/reference/HyperSpec/Body/m_do_do.htm 20:52:21 like i showed 20:52:30 but learn LOOP instead 20:52:47 you could just do (do (...) ("some condition" (format t ..."))), to translate your example "literally" 20:52:55 The problem with DO is that it checks the condition at the beginning. 20:53:10 after binding the variables, though. 20:53:19 many DO loops are designed to do most of the iteration through variables. 20:53:23 I need something like "do{} while();" 20:53:59 (loop while (foo-p) do (baz)) 20:54:02 (loop do (print "HELLO!") while not-tired) 20:54:07 (loop do ... while ) 20:54:20 Doh. stassats` wins. 20:54:47 Heh. I always though i could avoid learning LOOP. :P 20:54:50 thought* 20:54:57 slash_: bookmark it! 20:54:59 why bother? LOOP is fairly nice. 20:55:05 slash_, even better: learn ITERATE 20:55:16 -!- seisatsu [n=seisatsu@adsl-63-198-106-143.dsl.snfc21.pacbell.net] has quit ["Ex-Chat"] 20:55:18 and you really don't have to use the really advanced constructs most of the time. 20:55:30 Adlai: I have the book on my desk. ;) 20:55:45 slash_: so bookmark :P 20:56:06 http://common-lisp.net/project/iterate/doc/index.html 20:56:40 Addled [n=addled@209.20.68.236] has joined #lisp 20:57:06 deepfire: LOOP is pretty good at a lot of simpler constructs with relatively little syntactic garbage. 20:57:06 Many pros vs. the only con I know of: it incurs an additional dependency. Pkhuong also says that it's careless about using macroexpansion on its body forms. 20:57:34 but iterate is very good at iterating over hash tables (compared to LOOP) 20:57:48 -!- SandGorgon [n=OmNomNom@122.162.50.31] has quit [Connection timed out] 20:58:31 Also, LOOP has an extremely high chance of misindented. 20:58:38 use goto instead 20:58:46 (let ((i 0)) (tagbody start (print i) (incf i) (when (< i 10) (go start)))) 20:58:48 deepfire: really? Emacs seems to handle it fine for me. 20:58:48 lispm: you mean tagbody ;) 20:58:50 Jasko [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has joined #lisp 20:59:05 gigamonkey: ehhhh 20:59:20 well, Adlai already proposed using PROG 20:59:24 I think the indentation issue is fairly minor, but it's true that indentation is pretty shitty. 20:59:49 emacs + recent slime works fine on loop indentation. 20:59:51 (prog ((i 0)) start (print i) (incf i) (when (< i 10) (go start))) 20:59:58 gigamonkey, as I drive through emacs versions and setups the way emacs indents my LOOP statements change ever so slightly. 21:00:01 if your slime isn't indenting loop correctly, it's really old. 21:00:08 sykopomp: hmmm. Not my experince. Or maybe I'm just less picky. Or maybe I've been trained to subconsciously avoid constructs that don't get indented properly. 21:00:08 welcome to 70s 21:01:02 gigamonkey: two examples are: to get all keywords lined up properly, you need to have a newline after the LOOP call, and when you use :do, multiple forms get indented wrong 21:01:14 it's... very minor, but it's still pretty shitty indentation. 21:01:18 bah, a missed chance to pitch SERIES 21:01:29 michaelw: I hate the series API 21:01:52 michaelw, don't! people advertising series tend to disappear! 21:01:56 well, hate is a strong way of saying it. I dislike it. 21:01:58 suck it up :) 21:02:06 but I desperately want something like SERIES. 21:02:24 I think half the time, when I use LOOP, it's to get a numerical index :( 21:02:57 in my experience series reads better in CLtL2 than it is to use actually 21:03:01 being able to do (map nil *my-list* (series 0 .. 100)) would be pretty fantastic 21:03:49 efficiently 21:04:05 well, yes. I could use IOTA 21:04:13 lispm: it's pretty okay actually. what's missing is a series-ext library of sorts 21:05:24 chris2 [n=chris@p5B169F33.dip0.t-ipconnect.de] has joined #lisp 21:05:36 Pervasive, extensible sequences are missing, actually. 21:05:40 michaelw: there -is- such a thing as type inference these days, too, you know. And there's all those pesky declarations... and built-in sequence functions, etc. 21:05:49 deepfire: SEQUENCE 21:06:11 which might actually make it into cltl3 21:06:12 wooo 21:06:17 sykopomp, I refer to Krystof's stuff 21:06:37 deepfire: what stuff? 21:07:02 http://www.doc.gold.ac.uk/~mas01cr/talks/2007-04-03%20Cambridge/ilc-talk.pdf 21:08:14 deepfire: oh. Yeah, that's what I'm talking about too. 21:08:15 -!- lhz [n=shrekz@c-b9aa72d5.021-158-73746f34.cust.bredbandsbolaget.se] has quit ["Leaving"] 21:08:31 -!- salva [n=salva@105.11.117.91.dynamic.mundo-r.com] has quit [] 21:08:40 for some reason, I didn't realize Krystof/Xof was actually crhodes 21:08:58 Extend it with a library of things such as (seq 0 100) producing generative sequences. 21:09:24 yeah. I'd really like that. 21:11:17 See how much of series could be replaced with extending that. 21:11:31 how much of series couldn't? 21:11:33 Speaking of, Xof, you here? 21:11:35 stassats`: PROG is better than PROGV, at least... 21:11:46 that's a bad excuse 21:11:52 (why is it called PROGV, btw? Any ideas?) 21:11:59 well, prog and progv are totally different 21:12:05 "V" for variables, maybe? 21:12:08 francogrex [n=user@91.177.17.237] has joined #lisp 21:12:10 (That's just a pure guess.) 21:12:11 -!- redblue [i=star@ppp077.108-253-207.mtl.mt.videotron.ca] has quit [Connection timed out] 21:12:30 gigamonkey: that's mine too, but it's still a weak naming. 21:12:51 besides cliki what major CL sites are there (those with important projects contributed)? 21:12:57 rme: they're different, and yet they both smell of the same mothballs. 21:13:12 wigflip 21:13:20 francogrex: common-lisp.net 21:13:41 stassats`: wigflip is way more important than c-l.net! 21:13:51 heheh 21:14:40 francogrex: cliki doesn't contain projects, it's just a wiki 21:14:52 -!- johnny_zetta [n=anonymou@217.24.250.130] has quit [] 21:14:52 weitz.de :) 21:15:12 francogrex: most CL project are on c-l.net, github.com and code.google.com 21:15:24 ah ok; well the common-lisp.net seems good because it refers to others 21:15:55 cons.org 21:17:06 O Great #Lisp 21:17:30 I offer to thee my first script 21:17:39 *Vonunov* lays http://pastebin.org/28616 before #lisp 21:18:10 wow, with PORN ads 21:18:11 -!- phf [n=user@host.icnfull.com] has quit [Read error: 110 (Connection timed out)] 21:18:16 Microsoft's edit.exe is laughably broken wrt. cp866: symbols entered through cat are misrendered by it. And it totally ignores the current keymap. So fail. 21:18:18 [Latter half of vital functions coming soon] 21:18:27 Vonunov: you made a database (-: 21:18:41 Vonunov: paste.lisp.org 21:18:46 -!- GrayMagiker [n=steve@c-68-54-2-128.hsd1.nm.comcast.net] has quit ["Leaving"] 21:18:49 Oh, look at that, just for lisp. 21:19:00 Vonunov: you might not want to replace 'load' 21:19:08 clhs load 21:19:08 http://www.lispworks.com/reference/HyperSpec/Body/f_load.htm 21:19:13 somehow I like paste.lisp.org better, no porn ads, SFW 21:19:24 bgs100 [n=ian@unaffiliated/bgs100] has joined #lisp 21:19:28 Oh, oop.s 21:19:35 lispm: ad blockers make a huge chunk of the internet safe for work (-: 21:19:41 At some point I could have sworn I was told that there are no reserved words, but maybe it was something I misunderstood. 21:19:42 indeed. 21:19:44 I'll rename that, then. 21:19:46 (I didn't even notice there are any) 21:20:16 Vonunov: nope, functions in the cl package are dangerous (and in some implementations, impossible) to override 21:20:22 antifuchs: ad blockers are getting more attractive every day 21:20:28 Oh, okay. 21:20:30 antifuchs: lately, I've been using http://add-art.org/ instead of just plain adblock 21:20:37 antifuchs, hi! 21:20:54 lispm: if you're on a mac, I recommend glimmer blocker; it's a nice proxy that filters out ads without the need for a browser plugin and stuff 21:21:01 I used to use a CL-HTTP proxy to filter ads 21:21:10 should do it again... 21:21:45 newcmr [i=5811c0af@gateway/web/freenode/x-lamjqjyykupehnwo] has joined #lisp 21:22:15 sounds good. it shouldn't be too hard to integrate the relevant known-ad lists 21:22:29 antifuchs, I just de-rustified poiu. 21:22:35 oh yay (: 21:22:44 must have been a bit of work; asdf changed a lot lately 21:23:04 how can you write code that has at the same time such pretty bits and such ugly ugly hacks? 21:23:23 I am a very balanced person (-: 21:23:29 I was thinking of trying to merge as much as possible of POIU into ASDF itself. 21:23:46 to minimize the maintenance hassle. 21:23:50 that would be great 21:24:04 sepult [n=levgue@xdsl-87-78-171-31.netcologne.de] has joined #lisp 21:24:06 I think there were a couple of bug fixes to asdf that I never got around to sending 21:24:18 on the other hand, I'm also trying to kill ASDF. But to kill it when it is at its best, not at its worst 21:24:18 (the normalization of module/component names is one, I think) 21:24:23 -!- TDT [n=user@206.196.111.202] has quit [Remote closed the connection] 21:24:36 -!- Nshag [i=user@Mix-Orleans-106-3-85.w193-248.abo.wanadoo.fr] has quit [Remote closed the connection] 21:24:49 I sent a few fixes to ASDF recently. Not this one, I think 21:25:20 -!- kib2 [n=kib2@bd137-1-82-228-159-28.fbx.proxad.net] has quit [Read error: 104 (Connection reset by peer)] 21:25:23 can you explain this normalization to me? 21:25:38 (interestingly, XCVB also has some normalization inside) 21:26:04 I admit I previously only looked at the concurrency aspects of POIU (and got it to work with CCL) 21:26:17 ISTR there was a problem where redefining a system would remove a module and add it, which leads to the components in the module to be non-object-equal to the ones that asdf knows have had an operation "done" 21:26:32 so it would re-do that operation on those "new" components (which are just the old components) 21:26:54 by normalizing the components' names, this can be avoided 21:27:06 ...as asdf has that precaution built-in, just not working for every case 21:27:17 kib2 [n=kib2@bd137-1-82-228-159-28.fbx.proxad.net] has joined #lisp 21:27:19 but I'll be damned if I find where I put this code 21:27:26 anyone here in a lisp startup? 21:27:43 there's those hungarians! 21:27:52 they're in a startup? 21:27:56 gavino? 21:27:58 here? 21:28:00 again? 21:28:00 isn't that what they're doing? 21:28:07 gavino, yes 21:28:18 levy_: what's the startup? 21:28:38 Adlai, our company? 21:28:41 maybe 21:28:41 *antifuchs* waits for the "is lisp ready for the enterprise" question 21:28:52 levy_: is dwim proprietary!? 21:28:57 Baconizer [n=baconize@unaffiliated/baconizer] has joined #lisp 21:29:08 Adlai, ?? 21:29:14 *Adlai* is really confused 21:29:28 -!- Davidbrcz [n=david@212-198-78-230.rev.numericable.fr] has quit [Read error: 60 (Operation timed out)] 21:29:43 Any c-l.net admins around? 21:29:49 *levy_* is confused too 21:30:05 antifuchs: "can lisp be used to write 'foo', only better and faster without learning to program?" :) 21:30:08 -!- Addled [n=addled@209.20.68.236] has left #lisp 21:30:10 levy_: ah, I see -- from the cliki page, "the result of our work in a startup" 21:30:12 *rtoym* wonders why c-l.net seems to redirect through localhost, at least according to noscript. 21:30:13 dwim is as free as it can be 21:30:16 rtoym: yeah, what's up 21:30:20 rtoym: what happens ? 21:30:36 but still we are in a startup 21:30:39 levy_: it's not GPLd, though, is it? 21:30:39 levy_: what's the startup/company? 21:30:43 sykopomp: BSD 21:30:45 we just give all our stuff for free 21:30:48 Adlai: pffft 21:30:48 http://common-lisp.net/project/cl-dwim/ 21:30:50 to anybody interested 21:30:56 BSD 21:31:01 all of our codebase 21:31:02 rtoym: what do you mean? 21:31:20 Adlai, I don't think the name means to you much 21:31:26 we are not international 21:31:29 yet :-) 21:31:38 drewc, fe[nl]ix: trace (cmucl) was unreachable earlier today from work. There was a warning about the proxy or something not accepting connections to http://localhost/cmucl or something. (sorry, I'm not at work any more, so I don't remember.) 21:31:39 is there a page I can feed to google translate? 21:31:43 just discovered that holger schauer lives in a city that I visited a few weeks ago. if only I'd known earlier! 21:31:47 lol enterprise is such a scam term, I get the shovel out when someone says it, and its main reason I HATE redaht linux, they are enterprise zombies, what an inetersting competitive marketing word 21:32:25 antifuchs, how about using those @ powers for good? 21:32:38 Adlai, no, there should be dwim.hu but that you already know of I guess 21:32:38 rtoym: you mean the trac? 21:32:44 oh my, I seem to have grown wings 21:32:53 drewc: yes. CMUCL's trac in particular. 21:32:57 *antifuchs* swivels the laser eyes 21:33:05 drewc: Seems to be working ok now from home. 21:33:08 rtoym: it was down temporarily, and you were getting an error from squid 21:33:09 Addled [n=addled@209.20.68.236] has joined #lisp 21:33:23 levy_: btw, maybe you should link to dwim.hu from the c-l.net page? 21:33:24 we put trac behind squid to avoid the performance issues 21:33:52 drewc: how did you know which pages could be cached? 21:33:57 rtoym: it should be okay now ... the power supply blew up on the box that hosts cl-net yesterday. 21:34:20 Adlai, sure, we are just not there yet, I thinks it's generally a bad idea to advertise this more before it gets some real content and becomes mature 21:34:23 -!- OmniMancer [n=OmniManc@222-154-179-43.jetstream.xtra.co.nz] has quit ["Leaving."] 21:34:37 drewc: Hopefully! The cvs server at c-l.net tends to be unbearably slow. 21:34:37 er, I mean the site 21:34:39 foom: i'm assuming that trac has proper cache-control headers. otherwise i don't care and will wait for complaints about improper caching. 21:35:08 tcr: i hope we've licked that... it seem somewhat better now anyway. 21:35:09 drewc: But when I upgraded noscript plugin for firefox, I now get warnings from noscript's ABE (Applications Boundary Enforcer) about http://localhost/cmucl for login. Perhaps ABE is broken. 21:35:36 levy_: ok, probably a good idea. 21:35:37 rtoym: odd. 21:36:10 noscript is known to be broken 21:36:12 drewc: Oooh, a bad power supply. Sounds bad. 21:36:31 wow web gui on clos 21:36:45 rtoym: it could have been... but my hosting guy was down at the datacentre within 10 minutes, and had a spare temp fix in less than 1/2 hour. 21:37:03 I would love it if someone made a web gui witout clos and without using sql as persistance. 21:37:04 :) 21:37:10 the advantages of leasing hardware 21:37:13 fe[nl]ix: In general, or ABE in particular? Perhaps it's time to stop using noscript, but it's kind of nice not having all kinds of javascript running. 21:37:23 just get string from broswer and use all of lisp to sort seach and assemble compenent based pages. 21:37:42 that is my dream 21:37:46 gavino: i'd love it if you gained a clue, or learned to program. 21:38:36 *rtoym* starts some cmucl 20a builds for darwin. 21:38:49 gavino, use cl-quasi-quote and lambda, and you are freed from clos 21:38:58 not that you win anything with that 21:39:10 rtoym: in general. blocks too much, way too aggressive, it used to interfere with other plugins(adblock+), etc... 21:39:18 levy_, you have heard of Gavino before? 21:39:27 lispm, not much 21:39:35 that explains ;-) 21:39:40 ah, I see 21:39:41 lispm: :D 21:39:44 lispm: is this a c.l.l. troll that just made its way in here? 21:39:53 lasers! 21:39:56 http://paste.lisp.org/display/87860 <-- fixed load, and also changed (load-db (addresses)) to (load-db (filename)) -- I had taken a moment to understand that I was naming the argument, not what its value may be :P 21:40:02 -!- jleija [n=jleija@user-24-214-122-46.knology.net] has quit ["leaving"] 21:40:04 -!- jamesmkukla [n=jmk@74.207.227.87] has left #lisp 21:40:06 hehe 21:40:16 If anyone notices further errors or faux pas I'd be glad to hear before I learn wrong. :P 21:40:56 does anyone care to see how can I partial evaluate a make-instance called from SBCL source? 21:40:58 Vonunov: http://gigamonkeys.com/book 21:41:07 s/called/call 21:41:15 Yes, that's where I butchered this from. :P 21:41:22 minion: give Vonunov pcl 21:41:23 Vonunov: have a 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). 21:41:33 I don't care to catalogue MP3s so I made an address book instead. :3 21:41:41 Thanks. 21:41:43 *Adlai* is so glad that minion is live again 21:41:44 -!- schoppenhauer [n=christop@unaffiliated/schoppenhauer] has quit ["Verlassend"] 21:41:51 Vonunov: PCL is about much more than that 21:41:52 behold, the power of lisp. You, too can build your own s-exp database. 21:41:54 :P 21:42:02 sykopomp: LoL 21:42:08 sykopomp: a bit late -- pg does that in Arc already! 21:42:14 minion: glad your brain is back 21:42:15 i agree - glad my brain is back 21:42:18 Adlai: oh god, you can't be serious. 21:42:25 but of course 21:42:29 I imagine I will discover the arcane wonders of PCL once I progress beyond the first few chapters. 21:42:37 It's been busy here, though, might be a while. ;) 21:42:45 Vonunov: nothing arcane about it. Just awesome sauce. 21:42:59 levy_: i was just playing with the idea of writing a parser grammar evaluator and using partial-eval to optimize it... but then i ended up writing a few regexps... 21:43:14 sykopomp: Even better. :D 21:43:19 attila_lendvai: so you have two problems now? 21:43:28 bah, I just lost my first user :-( 21:43:43 levy_: ? 21:44:09 Adlai: my only problem right now is that i'm thirsty... :) 21:44:38 *Adlai* is referring to a nugget of jwzdom 21:44:41 madnificent: you can not expect too many users for a partial-evaluator lib... he just lost me when i turned to cl-ppcre instead 21:44:53 madnificent, I mean attila_lendvai could have been the first user of my somewhat toy partial-eval lib 21:45:06 -!- LiamH [n=none@common-lisp.net] has quit ["Leaving."] 21:45:20 I'm going to kick his *** next time we meet... ;-) 21:45:32 ah 21:46:10 *madnificent* hopes youtube will be blessed with imagery of that 21:46:16 attila_lendvai, you are absolutely wrong on that point 21:47:15 -!- rtoym [n=chatzill@user-0c8hpll.cable.mindspring.com] has quit [Remote closed the connection] 21:47:18 come on, these regexps are fast'n'easy... :) 21:47:25 Xof: ping 21:47:40 write once and pray that you should never see them again 21:47:41 -!- saikat_ [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has quit [] 21:47:58 attila_lendvai: did you at least use the CL-PPCRE s-exp syntax? 21:48:13 schoppenhauer [n=christop@unaffiliated/schoppenhauer] has joined #lisp 21:48:36 -!- ruediger_ [n=ruediger@188-23-95-154.adsl.highway.telekom.at] has quit ["Leaving"] 21:49:44 Vonunov: I think that Chapter 9 is one of the great bits in PCL 21:49:49 -!- Edico [n=Edico@unaffiliated/edico] has quit ["Ex-Chat"] 21:49:52 (the one about the unit test framework) 21:50:12 ShadowChild [n=lukjadOO@unaffiliated/lukjad007] has joined #lisp 21:50:48 Adlai: I didn't... it made me want to use my own unit testing framework... 21:51:01 Adlai: nope. then i'd use http://common-lisp.net/project/cl-irregsexp/ but we already depend on cl-ppcre 21:51:07 it did serve as a great example 21:51:07 madnificent: I'm not saying it's a great test framework 21:51:24 Adlai: it is actually quite fair, as it is very easy to extend :) 21:51:27 it is, however, a great example of building up a little system 21:52:03 -!- varjag [n=eugene@103.80-202-117.nextgentel.com] has quit ["Leaving"] 21:52:16 typical lisp thing: "Oh wait, I need this thing that could take up a lot of coding effort, but has this lispy interface... quickly write it... win! 21:52:39 ^ insert " somewhere 21:52:42 -!- Athas` [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has quit [Remote closed the connection] 21:52:57 -!- ikki [n=ikki@200.95.162.30] has quit ["Leaving"] 21:53:21 merimus [n=merimus@c-67-171-83-6.hsd1.pa.comcast.net] has joined #lisp 21:54:35 -!- lukjad007 [n=lukjadOO@unaffiliated/lukjad007] has quit [Read error: 131 (Connection reset by peer)] 21:54:43 dralston [n=dralston@S010600212986cca8.va.shawcable.net] has joined #lisp 21:55:14 wow, 'generative programming' workshop 21:55:37 where? 21:55:38 'Generative programming is an emerging paradigm aimed at automating important tasks in software development, ...' 21:55:44 -!- ShadowChild is now known as lukjad007 21:55:45 emerging, good god 21:55:58 http://wgt2010.elte.hu/ 21:56:01 lispm: somewhat like rails's generations then, I guess 21:56:18 'compile-time and run-time code transformation' 21:56:21 wow! 21:56:25 strange 21:56:32 lispm, hungarians again? ;-) 21:56:35 'and the creation of domain-specific languages and flexible libraries.' 21:56:45 in cyprus 21:56:53 emerging, though 21:57:38 attila_lendvai, Zoltan Porkolab 21:57:51 attila_lendvai, and IP :-) 21:57:53 rtoym [n=chatzill@user-0c8hpll.cable.mindspring.com] has joined #lisp 21:58:38 levy_: heh, cyprus for the winter on company expenses? :) 21:58:47 drewc: Well, now I'm having trouble logging in. It asks for my password, and then says requested URL /cmucl was not found. (I disabled noscript.) 21:59:26 attila_lendvai, good idea, I wish we could afford that 21:59:32 drewc: But, in fact, I am logged in, as I can see when I go to cmucl's trac page. 21:59:35 maybe I talk to the boss ;-) 22:00:04 oh, what a surprise, he's sleeping 22:00:12 rtoym: hmm, I can reproduce this 22:00:50 fe[nl]ix: Oh good, it's not just me, then. :-( 22:01:07 Something has changed in the last week or two. I didn't have this problem then. 22:01:16 attila_lendvai_ [n=ati@adsl-89-134-1-46.monradsl.monornet.hu] has joined #lisp 22:01:26 -!- attila_lendvai [n=ati@adsl-89-132-6-198.monradsl.monornet.hu] has quit [Nick collision from services.] 22:01:30 -!- attila_lendvai_ is now known as attila_lendvai 22:03:15 -!- dralston [n=dralston@S010600212986cca8.va.shawcable.net] has quit ["Leaving"] 22:03:46 dreish [n=dreish@minus.dreish.org] has joined #lisp 22:03:59 -!- mishoo [n=mishoo@79.112.50.83] has quit [Read error: 113 (No route to host)] 22:04:51 I think I must prepare my neurons for tomorrow 22:06:07 hopefully I still have a few left willing to work in Java and/or C++ 22:06:15 timor [n=martin@port-87-234-97-27.dynamic.qsc.de] has joined #lisp 22:06:17 -!- merimus [n=merimus@c-67-171-83-6.hsd1.pa.comcast.net] has quit [Client Quit] 22:06:43 -!- timor [n=martin@port-87-234-97-27.dynamic.qsc.de] has left #lisp 22:06:47 bb 22:07:08 -!- levy_ [n=levy@89.223.239.162] has quit ["..."] 22:09:05 lispm, some paradigms take decades to emerge. Speak of a painful birth. 22:09:25 Monk1 [n=a@cpe-74-74-140-173.rochester.res.rr.com] has joined #lisp 22:09:41 saikat_ [n=saikat@adsl-76-228-82-245.dsl.pltn13.sbcglobal.net] has joined #lisp 22:10:02 the funny thing is that 'generative programming' has emerged several times now 22:10:09 'emerged' 22:10:44 the last time it was the C++ crowd with template-meta-programming, I think 22:10:59 sepult` [n=levgue@xdsl-87-78-26-158.netcologne.de] has joined #lisp 22:11:22 -!- sepult [n=levgue@xdsl-87-78-171-31.netcologne.de] has quit [Read error: 60 (Operation timed out)] 22:15:17 oh you mean there's a new emergency? 22:16:14 Adlai: I'll certainly look forward to chapter nine, then. :D 22:18:24 -!- kpreid [n=kpreid@216-171-189-244.northland.net] has quit [Remote closed the connection] 22:19:02 Guest38594 [n=cti@132.198.13.188] has joined #lisp 22:19:37 I don't really know lisp, so... 22:19:53 Anyone know of a good guide? I heard Practical Common Lisp (PCL) is pretty good... 22:20:02 yes! 22:20:34 Guest38594: yes, I recommend PCL 22:20:36 that's what we typically recommend here 22:20:42 if you have no programming experience at all, though... 22:20:45 isnt lisp been doing programs that write programs for long time? 22:20:47 minion: tell Guest38594 about gentle 22:20:48 Guest38594: direct your attention towards gentle: "Common Lisp: A Gentle Introduction to Symbolic Computation" is a smoother introduction to lisp programming. http://www.cs.cmu.edu/~dst/LispBook/ 22:20:55 I have a decent background with functional programming. 22:21:03 But never really dug into the way of the lispnik 22:21:11 generative 22:21:13 ok, so you can ignore Gentle Introduction (and gavino, too) 22:21:23 *Vonunov* has a decent background with starting to learn programming languages and then losing interest 22:21:30 Guest38594, PCL has a practical approach 22:21:33 Adlai: Great! Thanks for the suggestion! 22:21:37 hey not me 22:21:43 Guest38594: follow them both! 22:21:56 Will do. I shall start reading now. 22:22:07 Thanks so much for the feedback! You are all friendly and awesome. 22:22:08 so is ansi common lisp by graham not in vogue anymore, and his followup on lisp 22:22:21 *Vonunov* spreads the love 22:22:27 -!- Guest38594 [n=cti@132.198.13.188] has quit ["rcirc on GNU Emacs 22.2.1"] 22:22:46 Guest38594, Lisp is not so much about functional programming, it is more about multiple paradigms, so you might want to forget some stuff before you read a Lisp book 22:23:33 ... BTW, if anyone is near Peterborough, NH, the library book store there had a copy of Gentle Introduction for sale as recently as last friday. 22:23:41 :exit 22:23:44 -!- francogrex [n=user@91.177.17.237] has quit ["ERC Version 5.2 (IRC client for Emacs)"] 22:23:54 -!- Blay [i=Blay@BSN-142-112-74.dial-up.dsl.siol.net] has quit ["gnight"] 22:24:11 You can still do functional programming in Common Lisp, it's just not as natural. You can make it much easier by using Alexandria: 22:24:16 minion: alexandria for Guest69565 22:24:17 Guest69565: please look at alexandria: Alexandria is a collection of portable public domain utilities. http://www.cliki.net/alexandria 22:24:27 hmmm, wrong guest. sorry. 22:24:36 haha 22:24:39 (They also had a copy of the Dragon Book, by Aho and Ulman, but I grabbed that for myself.) 22:25:15 -!- Edward_ [i=Ed@AAubervilliers-154-1-79-36.w81-249.abo.wanadoo.fr] has quit ["L'oignon fait la farce."] 22:26:17 And if anyone's at CEGEP de Ste-Foy in Quebec city, the library has a copy of Touretzky. 22:26:17 cl-irregesexp is 6 times faster than cl-ppcre with the http date parsing 22:26:35 attila_lendvai: and it matters? 22:26:35 it just has that "non-standard" cl-utilities dependency... 22:26:47 -!- kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 22:26:54 pkhuong: not really, but i'm still in The Right Thing mode... :) 22:29:00 "maybe_save_gc_mask_and_block_deferrables()"? Eesh. 22:29:16 also nice, 'language oriented programming, 2003, 'This paper describes the concept of language oriented programming which is a novel way of organising the development of a large software system' 22:29:23 novel 22:29:50 mejja [n=user@85.229.182.73] has joined #lisp 22:30:25 one should hit them with a copy of SICP over their head 22:30:27 http://mitpress.mit.edu/sicp/full-text/sicp/book/node75.html 22:30:55 "This paper describes the concept of geographically oriented programming which is a novel, but unhelpful, way of..." 22:31:10 lispm: "novel" is just short for "we didn't know that before" 22:31:28 novel = 'we don't read old papers' 22:31:40 ahh, irregsexp is not that nice inside... (with-unique-names (s b r a f) ...) -- a few more and they are not unique anymore... :) 22:32:03 michealw: the paper then cites Emacs as an example 22:32:29 -!- nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has quit [] 22:33:10 plus the example software is a LISP program 22:33:15 note the uppercase 22:34:55 From time to time I get depressing how incredibly fast pasting our metier is (so fast pasting that you can't spend the time to master anything because it's redundant after you've mastered it), I try to cheer me up by saying that most of it is scap (to no great avail in that mood though.) 22:35:10 crap 22:35:32 tcr: "pace"? 22:35:46 am of course 22:36:02 -!- coderdad [n=coderdad@ip72-200-214-240.ok.ok.cox.net] has quit [Remote closed the connection] 22:36:11 running fast in circles 22:36:51 I'd like to think it's spirals, at least 22:36:58 rtoym: wgl did the trac setup, we'll have to bug him. 22:37:08 wgl: around? 22:38:54 -!- mejja [n=user@85.229.182.73] has quit [Remote closed the connection] 22:40:09 -!- Fare [n=Fare@ita4fw1.itasoftware.com] has quit ["Leaving"] 22:41:32 -!- gavino is now known as gws 22:42:14 -!- cpc26 [n=cpc26@72.170.156.242] has quit [Read error: 54 (Connection reset by peer)] 22:43:53 -!- mrSpec [n=Spec@unaffiliated/mrspec] has quit [] 22:44:22 -!- xinming [n=hyy@125.109.254.35] has quit [Remote closed the connection] 22:47:35 cpc26 [n=cpc26@72.170.156.242] has joined #lisp 22:50:22 -!- kib2 [n=kib2@bd137-1-82-228-159-28.fbx.proxad.net] has left #lisp 22:52:44 xinming [n=hyy@125.109.254.35] has joined #lisp 22:52:47 rajesh [n=rajesh@nylug/member/rajesh] has joined #lisp 22:52:51 kpreid [n=kpreid@216.171.189.244] has joined #lisp 22:57:30 redblue [i=star@ppp246.108-253-207.mtl.mt.videotron.ca] has joined #lisp 22:57:43 -!- fgtech [n=fgtech@193.219.39.203] has quit ["Disconnecting from stoned server."] 22:58:01 fgtech [n=fgtech@193.219.39.203] has joined #lisp 22:58:35 ltriant [n=ltriant@lithium.mailguard.com.au] has joined #lisp 22:59:59 Summermute66 [n=scott@c-68-34-67-216.hsd1.dc.comcast.net] has joined #lisp 23:01:47 fisxoj [n=fisxoj@149.43.98.34] has joined #lisp 23:01:58 Adlai` [n=adlai@unaffiliated/adlai] has joined #lisp 23:02:36 -!- Adlai [n=Adlai@unaffiliated/adlai] has quit [Nick collision from services.] 23:02:38 -!- Adlai` is now known as Adlai 23:05:11 nvoorhies [n=nvoorhie@32.156.72.148] has joined #lisp 23:05:28 -!- ska` [n=user@ppp-58-8-239-254.revip2.asianet.co.th] has quit [Read error: 110 (Connection timed out)] 23:07:42 ok my dear bit-twiddly friends, who has a neat solution for taking two 16-bit words A[0-15] and B[0-15] and creating 32-bit interleaved word C[A0,B0,A1,B1,...] ? 23:08:33 -!- tcr [n=tcr@host146.natpool.mwn.de] has quit ["Leaving."] 23:09:17 lukego: liked your TR quote on reddit. 23:09:21 Anyone here with write access to the sbcl code repository? 23:09:39 antgreen` [n=user@CPE0014bf0b631a-CM0018c0b3c272.cpe.net.cable.rogers.com] has joined #lisp 23:09:44 Guest69565: Yes. 23:09:56 -!- Yuuhi [n=user@p5483C84A.dip.t-dialin.net] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 23:09:57 (but not me) 23:10:10 gigamonkey: I liked your context. the internet seems to have missed the point of the whole interview, but with any luck they'll end up reading it :) 23:10:52 the place today where you can kind of do the whole "i build my os before i print out something" is embedded programming 23:10:57 Oh. I was thinking that I could get someone to fix a minor bug without going to the trouble of figuring out how to report it properly. 23:10:59 -!- antgreen [n=user@CPE0014bf0b631a-CM0018c0b3c272.cpe.net.cable.rogers.com] has quit [Read error: 60 (Operation timed out)] 23:11:05 I think the take-away from the jwz interview is not "you should write crappy code" but rather "you should quit your stupid enterprise job and try to have an impact on the world" :) 23:11:07 you don't get to optimize for drum memory :) 23:11:10 -!- Hun [n=hun@p50993726.dip0.t-ipconnect.de] has quit [Remote closed the connection] 23:12:25 Guest69565: Off the top of my head (and some commits in the SBCL repo), jsnell pkhuong Xof antifuchs and possibly a few others have commit access. 23:12:59 mega and nikodemus! 23:13:24 antifuchs: ah, but neither are online right now :) 23:13:28 lukego: Here you go in tweetable form: 'The take-away from the jwz C@W interview is not "write crappy code" but "quit your stupid enterprise job and have an impact on the world" :)' 23:13:31 manual__: and not just embedded either. that's the code I'm writing these days - laptop firmware 23:13:48 lukego: that's pretty much "embedded 23:13:52 though isn't it? 23:14:08 so tweeted :) 23:14:10 much under that umbrella though 23:14:17 Guest69565: I think the course of action that will most likely see your bug fixed is if you report it on launchpad, and send a message to the sbcl-devel list 23:14:17 what's your twitter username luke? 23:14:26 lukego: Heh. Good. Now I can quote you tomorrow on the StackOverflow podcast. 23:14:31 manual__: well kinda. I think of embedded as smaller computers. my embedded platform has a 1GHz CPU and 1GB of RAM :) 23:14:33 lukego 23:14:38 lukego: ah yeah 23:14:55 mine have from 512 bytes to 64 kB 23:15:31 nice just followed you 23:15:32 -!- Summermute [n=scott@c-68-34-67-216.hsd1.dc.comcast.net] has quit [Read error: 110 (Connection timed out)] 23:16:23 -!- Dawgmatix_ [n=dman@c-68-32-44-191.hsd1.nj.comcast.net] has quit ["Ex-Chat"] 23:18:03 -!- cpc26 [n=cpc26@72.170.156.242] has quit [Dead socket] 23:21:30 lukego: if A and B happen to be in memory in contiguous addresses you could simply swap A1 and B0 23:21:55 -!- Addled [n=addled@209.20.68.236] has left #lisp 23:22:02 maybe I wasn't clear. it's the bits that need to be interleaved 23:22:23 but honestly I don't have a bit-twiddly brain. I'm going to write a loop and do it all one bit at a time. 23:23:49 the spreading is awkward... 23:23:50 lukego: if performance matters, unroll the loop... other than that, I don't really have suggestions... maybe it's easier to do it by "expanding" each word separately and then adding them? 23:24:01 lukego: That problem sounds like something you'd find discussed in either HAKMEM or Hacker's Delight. 23:24:16 I think performance doesn't matter. this is to install a new mouse-pointer bitmap in a format that the Geode graphics controller understands 23:24:19 Umm... Or your CPU might have special instructions for it. 23:24:24 This for an ARM system? 23:24:30 x86 23:24:36 Ah. 23:24:52 make 16 masks, mask the bits, shift them, logand them 23:25:59 lukego: morton order? It's an important operation in graphic stuff, so you can probably google for good implementations. 23:26:36 you can optimize the number of swaps 23:26:43 hey c'mon guys I just wanted you to tell me there's no fast way and so I should just use a nested pixel-by-pixel loop :) 23:27:02 Sorry, it's four C statements according to Warren. 23:27:07 And that's the straightforward code. 23:27:14 What's Moreton order? 23:27:54 -!- rajesh [n=rajesh@nylug/member/rajesh] has quit ["leaving"] 23:27:55 nyef: that will interleave the bits of two half-words? 23:28:24 I want to splice A0A1A2A3 and B0B1B2B3 into A0B0A1B1A2B2A3B3 23:28:35 lukego: you could lobby Intel to add a specific instruction :D 23:28:48 fe[nl]ix: there's something like that in Larrabee. 23:29:12 Right, so do something like int32 x = (a << 16) | b;, and then shuffle x destructively. 23:30:16 -!- scode [n=scode@hyperion.scode.org] has quit [Remote closed the connection] 23:30:38 Then you get a few statements like "x = (x & 0x0000ff00) << 8 | (x >> 8) & 0x0000ff00 | x & 0xff0000ff;", and changing the mask and shift each time. 23:31:00 scode [n=scode@hyperion.scode.org] has joined #lisp 23:31:05 nyef: how about interleaving nybbles + LUT? 23:31:09 -!- icecube_ [n=icecube@p549C3509.dip.t-dialin.net] has quit ["Leaving"] 23:31:22 Spending D-cache to reduce I-cache? 23:31:49 Warren gives the straightforward implementation as 42 "basic RISC instructions". 23:31:49 and to save on the few shifters we have around. 23:32:30 -!- ignas [n=ignas@ctv-79-132-160-221.vinita.lt] has quit [Read error: 110 (Connection timed out)] 23:32:33 antifuchs: thanks 23:32:42 maybe that crazy pshufb x86 instruction would do this all in one go. 23:32:57 rme: no, it shuffles bytes ;) 23:33:04 you're trying to blit pixels optimally? 23:33:07 Especially if it's in an inner loop, D$ isn't an issue. 23:33:08 drewc: Might be system-specific. It fails with firefox on macosx, but firefox on linux seems ok. (Plugin sets might be different though.) 23:33:16 Or you could spread the two words first, then OR them together once they're spread. 23:33:50 You have sufficiently limited shifter power on an x86oid that a shift by 8 is difficult? 23:34:25 you can shift per-bit in either direction by register count on x86. 23:34:35 iirc 23:34:52 oh, bits. sorry. and geode doesn't have sse2 anyway. 23:34:55 nyef: shifts are cheap, but there aren't many of them so you can easily be throughput limited. 23:35:06 Ralith [n=ralith@69.90.49.189] has joined #lisp 23:35:11 rme: I'm sure there's something like that in MMX too ;) 23:35:39 seems better to just mask repeatedly. burn icache with an unrolled loop. 23:36:11 print the numbers to a string of bits, then interleave the two strings into one and read from there with base 2 23:36:12 You'd already be throughput-limited due to data availability from the shifter operation anyway, wouldn't you? 23:36:26 with masks and ors you do it without touching the shifter. 23:37:01 nyef: not necessarily. A shift has a latency of a couple cycles, you're in a well-behaved inner loop, and you have ~6(?) shifts/2*16 bit? 23:37:28 I think you can get it down to nine shifts total for the entire operation. 23:37:32 -!- HET2 [n=diman@cpc1-cdif12-2-0-cust125.5-1.cable.virginmedia.com] has quit [Read error: 110 (Connection timed out)] 23:37:58 So long as you have two shifters, you should be able to cover the latency. 23:38:29 nyef: how's that? The shifts are mostly dependent, no? 23:39:06 this doesn't have to be fast - it's updating a 16x16 cursor that will happen only a few times per minute - but it has to be easy to write because it's late :) 23:39:07 If you spread the two halfwords first, they can be processed independently in parallel until they need to be re-joined. 23:40:14 haha 23:41:27 -!- Aszarsha [n=aszarsha@91-164-152-252.rev.libertysurf.net] has left #lisp 23:42:45 we need a new cpu instruction bitfiddle 23:42:54 b = (b & 0x00ff) | ((b & 0xff00) << 8); b = ((b << 4) | b) & 0x0f0f0f0f; b = ((b << 2) | b) & 0x33333333; b = ((b << 1) | b) & 0x55555555;, do something similar with b in parallel, arranging for it to be one bit further left, and OR them together. 23:43:01 time for some microcode 23:43:16 That's four shifts per halfword, plus one extra to align them. 23:45:21 About sixteen instructions per half, assuming that everything is in registers, with some savings for running in parallel and thus not having to reload the shift counts... 23:48:06 hi, ignoring the undeniable fact that all you'll ever need when programming is Lisp :) ... What would be from your point of view the set of most interesting languages a programmer should know? 23:48:26 English, for starters. 23:48:53 Latin helps -- "Quidquid latine dictit, altus sonatur" 23:48:54 English is the important one. 23:48:57 or something along those lines. 23:48:59 ok. 23:49:29 splice(<<>>, <<>>) -> <<>>; 23:49:29 splice(<>, <>) -> <>. 23:49:30 ok, it's reasonable but I meant programming languages 23:49:31 In no particular order: Lisp, C, Haskell, FORTH, Prolog, APL (or maybe J or K or whatever the modern variants are called). Smalltalk. 23:49:31 .. I miss erlang sometimes. 23:49:39 Latin isn't as helpful as English, though. 23:49:43 Adlai: that would be "Quidquid latine dictum sit, altum sonatur" 23:49:54 fe[nl]ix: ok, thanks, I guess. :\ 23:49:54 obviously Adlai is still a novice :D 23:50:02 Ralith: long time no latin 23:50:16 lukego: dude, to what does that compile? :p 23:50:19 I still like the latin error messages in SBCL and CMUCL. 23:50:35 pkhuong: erlang guys don't worry about that sort of stuff :) 23:50:38 pkhuong: Are you reminded of the "Creators admit C, UNIX a hoax" article? 23:50:55 nyef: doesn't ring a bell. 23:51:16 pkhuong: It's at the back of the UNIX-Haters Handbook, among other places. 23:51:41 incidentally I am writing this in forth 23:51:50 gigamonkey: what's remarkable about forth? 23:52:15 Basically the story is that they put it all together as a joke, decided that the compiler was done after they got a clean compile on something that looked like line noise, were considering selling it to the russians to set them back a decade or two in computing, and then were blown away when many US companies and universities adopted it instead. 23:52:20 newcmr: relatively high-level for an embedded language or something? 23:52:30 lukego: can tell you better than me. But it's pretty different from all the rest. 23:52:32 ok, sykopomp 23:52:43 newcmr: Er, rather, lukego can tell you. 23:52:43 learn lambda calculus 23:52:45 I always thought the neat thing about forth was the whole write-a-bootloader thing. Stack-based languages are also interesting. 23:52:54 even if they're clearly turing tarpits 23:52:54 Forth is one of the "secret weapon" languages. It's bloody amazing how flexible it can be. 23:52:55 ;D 23:53:04 forth is like Lisp ported from church to von neuman :) 23:53:27 forth is Lispm assembler 23:53:52 The TI Explorer add-in boards had their startup diagnostics written in a FORTH-like bytecode. 23:53:57 lispm: not (D|R)NA? ;) 23:54:14 nyef: was it FCODE like sparc/ppc? 23:54:19 Umm... And I think one of the diagnostic tools was also written in it. 23:54:33 pkhuong, good idea 23:54:35 Might have been, though it might also have been a different set of opcodes. 23:54:48 I'm not sufficiently familiar with FCODE to be able to tell for certain. 23:55:32 FCODE was a standardised forth bytecode (basically, tokenized source code) that's an IEEE standard. the drivers that I'm writign today compile down to FCODE, so you could actually load them onto a 1980s Sun machine, or perhaps even a TI Explorer :) 23:55:34 (Though I did figure out what the TI diagnostic bytecode semantics were by disassembling the microcode that interpreted it.) 23:56:01 FCODE is IEEE1275 - spec is linked from wikipedia page 23:56:37 The Explorer didn't have a full environment, just the bytecode interpreter, some stuff in the add-in board declaration ROMs, and the GDOS diagnostics. 23:57:16 15-bit bytecode address space, all actual bytecodes had their high bit set, etc. 23:57:32 -!- frontiers [n=frontier@139.79-160-22.customer.lyse.net] has quit [Read error: 110 (Connection timed out)] 23:57:33 OmniMancer [n=OmniManc@202.36.179.65] has joined #lisp 23:58:43 -!- OmniMancer [n=OmniManc@202.36.179.65] has quit [Read error: 104 (Connection reset by peer)] 23:59:41 any sbcl experts willing to answer one or two questions?