00:00:38 Joreji [~thomas@77-066.eduroam.RWTH-Aachen.DE] has joined #lisp 00:01:01 mathrick [~mathrick@users177.kollegienet.dk] has joined #lisp 00:05:12 -!- kwabbles [~mike@h-67-101-178-243.lsanca54.static.covad.net] has quit [Quit: kwabbles has no reason] 00:05:14 mbohun [~mbohun@202.124.72.31] has joined #lisp 00:05:30 -!- ignas [~ignas@ctv-79-132-160-221.vinita.lt] has quit [Ping timeout: 240 seconds] 00:06:56 rbarraud [~rbarraud@118-93-191-215.dsl.dyn.ihug.co.nz] has joined #lisp 00:13:21 l_n [~not_you@tuxhacker/lordnothing] has joined #lisp 00:13:33 -!- srolls [~user@167.216.131.126] has quit [Remote host closed the connection] 00:14:34 -!- rvirding [~chatzilla@h19n4c1o1034.bredband.skanova.com] has quit [Remote host closed the connection] 00:15:39 -!- davazp [~user@84.122.72.152.dyn.user.ono.com] has quit [Remote host closed the connection] 00:16:52 rgrau_ [~user@201.Red-79-150-62.dynamicIP.rima-tde.net] has joined #lisp 00:19:31 -!- Makoryu [~vt920@pool-71-174-191-10.bstnma.fios.verizon.net] has quit [] 00:28:41 -!- rbarraud [~rbarraud@118-93-191-215.dsl.dyn.ihug.co.nz] has quit [Ping timeout: 260 seconds] 00:28:50 -!- cisticola [~daddy@202.134.251.153] has quit [Read error: Connection reset by peer] 00:29:05 -!- coyo [~unf@pool-71-164-234-173.dllstx.fios.verizon.net] has quit [Quit: KVIrc Insomnia 4.0.0, revision: , sources date: 20090520, built on: 2010/07/07 00:46:30 UTC http://www.kvirc.net/] 00:29:19 cisticola [~daddy@202.134.251.153] has joined #lisp 00:31:05 konr` [~user@201.82.129.247] has joined #lisp 00:31:39 coyo [~unf@pool-71-164-234-173.dllstx.fios.verizon.net] has joined #lisp 00:32:55 -!- konr [~user@201.82.129.247] has quit [Ping timeout: 255 seconds] 00:33:36 xyxxyyy [~xyxu@58.41.13.12] has joined #lisp 00:35:43 insanux2 [~hola@83.50.180.208] has joined #lisp 00:36:37 -!- nus [~nus@unaffiliated/nus] has quit [Ping timeout: 260 seconds] 00:37:04 -!- incandenza [~djm@ip68-231-109-244.ph.ph.cox.net] has quit [Remote host closed the connection] 00:38:01 -!- sentry [~sentry@151.sub-69-98-145.myvzw.com] has quit [Ping timeout: 260 seconds] 00:38:41 *Xach* looks around for willing quicklisp testers, encourages them to join #quicklisp 00:38:48 incandenza [~djm@ip68-231-109-244.ph.ph.cox.net] has joined #lisp 00:39:01 -!- insanux [~hola@83.54.63.183] has quit [Ping timeout: 265 seconds] 00:41:14 rbarraud [~rbarraud@121.90.214.46] has joined #lisp 00:45:21 -!- rgrau_ [~user@201.Red-79-150-62.dynamicIP.rima-tde.net] has quit [Remote host closed the connection] 00:48:49 LiamH [~nobody@pool-72-75-117-208.washdc.east.verizon.net] has joined #lisp 00:48:52 davazp [~user@99.Red-88-25-185.staticIP.rima-tde.net] has joined #lisp 00:57:23 -!- pizzledizzle [~pizdets@pool-96-250-215-244.nycmny.fios.verizon.net] has quit [Ping timeout: 272 seconds] 00:58:30 dnolen [~dnolen@pool-96-224-26-15.nycmny.east.verizon.net] has joined #lisp 00:58:58 fusss [~chatzilla@cpe-184-59-201-74.new.res.rr.com] has joined #lisp 00:59:00 greetings 00:59:04 -!- Joreji [~thomas@77-066.eduroam.RWTH-Aachen.DE] has quit [Ping timeout: 240 seconds] 00:59:38 yo fussssss 00:59:38 yo 00:59:42 can anyone summarize to me how to use &allow-other-keys in the generic function signature, without forcing the specific methods to include it? 00:59:51 -!- dnolen [~dnolen@pool-96-224-26-15.nycmny.east.verizon.net] has quit [Client Quit] 00:59:55 hey guys 01:00:01 -!- vp8dmh [~clarkema@bs0176.nerc-bas.ac.uk] has quit [Ping timeout: 252 seconds] 01:01:01 I have a GF that ought to take on required argument, the rest of the arguments are all optional and might include keywords known to the applicable specific methods 01:01:35 *p_l* figures he isn't in best condition to answer that after reading GF as "girlfriend" >_< 01:01:40 -!- prip [~foo@host65-194-dynamic.17-79-r.retail.telecomitalia.it] has quit [Read error: Operation timed out] 01:01:41 clhs 7.6.4 01:01:42 http://www.lispworks.com/reference/HyperSpec/Body/07_fd.htm 01:02:44 -!- coyo [~unf@pool-71-164-234-173.dllstx.fios.verizon.net] has quit [Changing host] 01:02:44 coyo [~unf@unaffiliated/bandu] has joined #lisp 01:04:05 thanks stassats` 01:04:25 #5 was what I was asking 01:05:26 Oh my lord, tab completion at the repl has significantly become better since I last used it over a year ago. :D 01:05:58 funny, because it hasn't changed 01:06:28 Maybe I just forgot then. 01:06:43 maybe you used it wrong 01:07:01 Maybe, but no time for speculating on the past. 01:07:19 ok, you'll use it wrong sooner or later! 01:08:03 stassats`: No no, completion had to have changed, I don't remember any of these ~completion scores~ or anything. And I assure you, my memory is faultless! 01:08:28 then used a different kind of completion, there are three of them 01:08:42 stuff "you" somewhere in between 01:12:30 -!- timor [~timor@port-92-195-102-61.dynamic.qsc.de] has quit [Remote host closed the connection] 01:12:56 Mesh [~Mesh@216.201.34.14] has joined #lisp 01:13:26 aidalgol [~user@114-134-6-5.rurallink.co.nz] has joined #lisp 01:13:40 -!- TraumaPony [~TraumaPon@202.173.162.181] has quit [] 01:17:40 prip [~foo@host229-9-dynamic.7-79-r.retail.telecomitalia.it] has joined #lisp 01:25:32 Avisch [~Avisch@70-101-99-64.dsl1-field.roc.ny.frontiernet.net] has joined #lisp 01:27:41 -!- dfkjjkfd [~paulh@96-13-ftth.onsnetstudenten.nl] has quit [Quit: Lost terminal] 01:29:39 -!- bgs100 [~ian@unaffiliated/bgs100] has quit [Quit: Leaving] 01:31:13 neoesque [~neoesque@210.59.147.232] has joined #lisp 01:32:20 -!- rbarraud [~rbarraud@121.90.214.46] has quit [Ping timeout: 255 seconds] 01:36:03 Elench [~user@unaffiliated/elench] has joined #lisp 01:40:10 xach: you here? 01:41:58 Yep. 01:44:11 The suspense is killing me! 01:46:58 -!- adu [~ajr@pool-71-191-154-37.washdc.fios.verizon.net] has quit [Read error: Connection reset by peer] 01:47:00 Are you trying to send a private message and it's not working because you're not identified? Or are you just wondering out loud if I'm here? 01:47:24 adu [~ajr@pool-71-191-154-37.washdc.fios.verizon.net] has joined #lisp 01:48:00 xach: no, i just wandered away absentmindedly. 01:48:21 Cruel. 01:48:27 xach: can you go to symbolicspace.com in chrome and see if you have your prob anymore? 01:48:33 tayloj [~tayloj@cpe-67-248-22-159.nycap.res.rr.com] has joined #lisp 01:49:20 Seems to work fine. 01:51:19 -!- hohoho [~hohoho@ntkngw229253.kngw.nt.ftth.ppp.infoweb.ne.jp] has quit [Remote host closed the connection] 01:51:40 cool thanks 01:51:45 put some team info up 01:52:34 Shiro Kawai! 01:52:40 *Xach* fondly remembers his Final Fantasy paper 01:55:26 final fantasy paper? 01:56:12 http://www.lava.net/~shiro/articles-e.html 01:56:20 -!- Modius [~Modius@cpe-24-28-30-165.austin.res.rr.com] has quit [Ping timeout: 276 seconds] 02:00:29 psilord2 [~psilord@adsl-99-154-14-142.dsl.mdsnwi.sbcglobal.net] has joined #lisp 02:00:56 he's so cool, got to meet him in honolulu last week 02:01:32 kaveh, of course, is already like a brother from another mother 02:01:42 he was shiro's manager on ff 02:01:53 the movie, i dont think ff7 02:03:21 everybody on the team is smarter and more experienced than i 02:03:24 it's awesome 02:03:28 Good morning everyone! 02:03:50 beach: I need to go to sleep, good morning 02:04:32 madnificent: Feel free! :) 02:07:35 pizzledizzle [~pizdets@pool-96-250-215-244.nycmny.fios.verizon.net] has joined #lisp 02:11:14 -!- insanux2 [~hola@83.50.180.208] has quit [Quit: Ex-Chat] 02:13:30 hey beach! 02:13:46 Edward__ [~ed@AAubervilliers-154-1-20-27.w90-3.abo.wanadoo.fr] has joined #lisp 02:15:06 -!- Edward_ [~ed@AAubervilliers-154-1-45-106.w90-3.abo.wanadoo.fr] has quit [Ping timeout: 240 seconds] 02:16:07 -!- bzzbzz [~franco@modemcable240.34-83-70.mc.videotron.ca] has quit [Quit: leaving] 02:16:24 galaxywatcher [~galaxywat@ppp-58-8-52-3.revip2.asianet.co.th] has joined #lisp 02:18:30 jsfb [~jon@unaffiliated/jsfb] has joined #lisp 02:21:32 -!- sdsds [~sdsds@dyn-16.sub2.cmts01.cable.TORON10.iasl.com] has quit [Ping timeout: 258 seconds] 02:22:13 -!- debiandebian_ [~chatzilla@ntszok008244.szok.nt.adsl.ppp.infoweb.ne.jp] has quit [Remote host closed the connection] 02:23:51 -!- Mesh [~Mesh@216.201.34.14] has quit [Quit: HydraIRC -> http://www.hydrairc.com <- Po-ta-to, boil em, mash em, stick em in a stew.] 02:24:15 at what point in the build process for sbcl is Makefile.features generated (and from what is it generated)? 02:27:12 Genesis? 02:28:25 -!- jsfb [~jon@unaffiliated/jsfb] has quit [Quit: Leaving] 02:29:09 debiandebian [~chatzilla@ntszok008244.szok.nt.adsl.ppp.infoweb.ne.jp] has joined #lisp 02:29:48 -!- moocow [~new@64.151.208.2] has quit [Remote host closed the connection] 02:30:01 qbomb [~Q@pool2-60.teleclipse.net] has joined #lisp 02:30:08 -!- LiamH [~nobody@pool-72-75-117-208.washdc.east.verizon.net] has quit [Ping timeout: 245 seconds] 02:30:27 stassats`: Any ever think about having slime do filename completion on logical pathnames or (gasp!) cmucl search-lists? 02:30:53 hargettp [~anonymous@pool-71-184-181-149.bstnma.east.verizon.net] has joined #lisp 02:31:42 i don't use logical pathnames, so no, i haven't thought about it 02:32:10 and the current filename completion is fully done by comint 02:32:14 I don't use logical pathnames much either. I do use search-lists a bit. 02:32:26 Kerrick [~Kerrick@e40-1.nat.iastate.edu] has joined #lisp 02:33:16 doing completion on the lisp side may be better, for ssh and such 02:34:21 anyway to generate colorized HTML Lisp code from an emacs buffer? 02:34:23 I have seen it 02:34:30 i've been meaning for a long time to improve fuzzy completion in some obvious ways 02:34:41 -!- jaiball [~justin@c-76-119-131-178.hsd1.ct.comcast.net] has quit [Ping timeout: 260 seconds] 02:34:57 fusss: M-x htmlfontify-buffer 02:35:28 *stassats`* just did M-x html 02:35:37 there's also colorize or something like that 02:35:42 stassats`: excellent! 02:42:04 frvx [~marc@c-76-106-92-89.hsd1.md.comcast.net] has joined #lisp 02:44:24 -!- pdelgallego [~pdelgalle@1503031474.dhcp.dbnet.dk] has quit [Ping timeout: 240 seconds] 02:46:39 -!- frvx [~marc@c-76-106-92-89.hsd1.md.comcast.net] has quit [Quit: fnord] 02:48:24 hahaha 02:51:15 -!- owned [~owned@93-96-43-166.zone4.bethere.co.uk] has quit [Quit: owned] 02:54:47 rbarraud [~rbarraud@118-93-191-215.dsl.dyn.ihug.co.nz] has joined #lisp 03:00:12 -!- carlocci [~nes@93.37.220.29] has quit [Quit: eventually IE will rot and die] 03:05:20 -!- pizzledizzle [~pizdets@pool-96-250-215-244.nycmny.fios.verizon.net] has quit [] 03:10:25 -!- rbarraud [~rbarraud@118-93-191-215.dsl.dyn.ihug.co.nz] has quit [Remote host closed the connection] 03:10:29 -!- Raynes [~macr0@unaffiliated/raynes] has quit [Ping timeout: 276 seconds] 03:10:45 rbarraud [~rbarraud@118-93-191-215.dsl.dyn.ihug.co.nz] has joined #lisp 03:15:12 -!- Edward__ [~ed@AAubervilliers-154-1-20-27.w90-3.abo.wanadoo.fr] has quit [] 03:20:52 -!- hargettp [~anonymous@pool-71-184-181-149.bstnma.east.verizon.net] has quit [Quit: hargettp] 03:26:35 -!- jhalogen [~jake@cpe-98-154-250-117.socal.res.rr.com] has quit [Quit: jhalogen] 03:27:33 -!- rbarraud [~rbarraud@118-93-191-215.dsl.dyn.ihug.co.nz] has quit [Quit: Konversation terminated!] 03:27:39 rbarraud [~rbarraud@118-93-191-215.dsl.dyn.ihug.co.nz] has joined #lisp 03:28:46 -!- qbomb [~Q@pool2-60.teleclipse.net] has left #lisp 03:30:13 -!- peterhil [peterhil@a91-153-120-132.elisa-laajakaista.fi] has quit [Ping timeout: 260 seconds] 03:33:56 -!- l_n [~not_you@tuxhacker/lordnothing] has quit [Quit: Vision[0.9.7-H-090423]: i've been blurred!] 03:37:54 -!- rbarraud [~rbarraud@118-93-191-215.dsl.dyn.ihug.co.nz] has quit [Quit: Konversation terminated!] 03:38:13 rbarraud [~rbarraud@118-93-191-215.dsl.dyn.ihug.co.nz] has joined #lisp 03:41:19 -!- dreish [~dreish@minus.dreish.org] has quit [Quit: dreish] 03:41:30 -!- bougyman [bougyman@bougyman.com] has quit [Changing host] 03:41:30 bougyman [bougyman@pdpc/supporter/gold/bougyman] has joined #lisp 03:44:07 -!- rbarraud [~rbarraud@118-93-191-215.dsl.dyn.ihug.co.nz] has quit [Read error: Connection reset by peer] 03:55:46 holycow [~new@poco208-2.fredcanhelp.com] has joined #lisp 04:04:27 -!- aidalgol [~user@114-134-6-5.rurallink.co.nz] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 04:10:32 -!- bougyman is now known as tj 04:10:37 -!- tj is now known as bougyman 04:11:35 -!- PuffTheMagic [~PuffTheMa@unaffiliated/puffthemagic] has quit [Remote host closed the connection] 04:19:19 -!- tayloj [~tayloj@cpe-67-248-22-159.nycap.res.rr.com] has left #lisp 04:19:52 -!- davazp [~user@99.Red-88-25-185.staticIP.rima-tde.net] has quit [Remote host closed the connection] 04:24:59 PuffTheMagic [~PuffTheMa@unaffiliated/puffthemagic] has joined #lisp 04:25:01 -!- PuffTheMagic [~PuffTheMa@unaffiliated/puffthemagic] has quit [Remote host closed the connection] 04:32:36 PuffTheMagic [~PuffTheMa@unaffiliated/puffthemagic] has joined #lisp 04:33:19 kami`` [~user@unaffiliated/kami-] has joined #lisp 04:35:35 -!- kami` [~user@unaffiliated/kami-] has quit [Ping timeout: 276 seconds] 04:39:48 -!- ace4016 [~jmarcelin@adsl-225-199-163.mia.bellsouth.net] has quit [Ping timeout: 240 seconds] 04:41:04 -!- TomJ [~tomj@78.151.111.133] has quit [Ping timeout: 258 seconds] 04:42:12 spradnyesh [~pradyus@nat/yahoo/x-sjtnmzklcghxrgjh] has joined #lisp 04:43:49 -!- PuffTheMagic [~PuffTheMa@unaffiliated/puffthemagic] has quit [Remote host closed the connection] 04:45:01 ace4016 [~jmarcelin@adsl-32-120-9.mia.bellsouth.net] has joined #lisp 04:45:05 -!- _s1gma [~d.d.derp@77.107.164.131] has quit [Quit: irl ragequit] 04:46:04 benny [~user@i577A1B86.versanet.de] has joined #lisp 04:48:50 thom_logn [~thom@pool-74-100-140-188.lsanca.fios.verizon.net] has joined #lisp 04:49:21 -!- loomer [~loomer@unaffiliated/loomer] has quit [Read error: Operation timed out] 04:50:01 loomer [~loomer@unaffiliated/loomer] has joined #lisp 04:54:26 -!- Avisch [~Avisch@70-101-99-64.dsl1-field.roc.ny.frontiernet.net] has quit [Quit: This computer has gone to sleep] 04:57:42 PuffTheMagic [~PuffTheMa@unaffiliated/puffthemagic] has joined #lisp 05:06:29 OEP [~OEP@c-76-27-129-70.hsd1.al.comcast.net] has joined #lisp 05:10:11 -!- thom_logn [~thom@pool-74-100-140-188.lsanca.fios.verizon.net] has quit [Quit: Leaving] 05:12:51 khrum42 [~khrum42@2001:a60:f05d:1::dead] has joined #lisp 05:13:55 -!- ikki [~ikki@201.155.75.146] has quit [Ping timeout: 265 seconds] 05:15:01 thom_logn [~thom@pool-74-100-140-188.lsanca.fios.verizon.net] has joined #lisp 05:16:59 milanj [~milanj_@109-92-124-204.dynamic.isp.telekom.rs] has joined #lisp 05:21:47 -!- stassats` [~stassats@wikipedia/stassats] has quit [Ping timeout: 276 seconds] 05:24:43 -!- prip [~foo@host229-9-dynamic.7-79-r.retail.telecomitalia.it] has quit [Ping timeout: 245 seconds] 05:26:56 ikki [~ikki@200.95.162.199] has joined #lisp 05:31:22 -!- holycow [~new@poco208-2.fredcanhelp.com] has quit [Remote host closed the connection] 05:31:27 aidalgol [~user@114-134-6-5.rurallink.co.nz] has joined #lisp 05:33:40 vu3rdd [~vu3rdd@164.164.250.10] has joined #lisp 05:35:45 -!- jleija [~jleija@user-24-214-122-46.knology.net] has quit [Quit: leaving] 05:37:30 prip [~foo@host254-123-dynamic.35-79-r.retail.telecomitalia.it] has joined #lisp 05:37:34 sea4ever [~sea@unaffiliated/sea4ever] has joined #lisp 05:38:26 maybe lisp/CLOS is the right tool to express very complex datatype relationships, which I never really realized until I saw this graph http://i.imgur.com/mrOzU.gif 05:40:18 -!- Kerrick [~Kerrick@e40-1.nat.iastate.edu] has quit [Quit: This system is going down for poweroff RIGHT FREAKING NOW!!!] 05:40:20 constraint networks are pretty easy to build, too 05:41:20 anyone seen an even moderately usable web-based programming text editor? something that at least handle highlighting and indentation? 05:41:40 i don't care what language, just that it exists 05:42:00 even better if it does auto-indentation 05:42:36 a web-based IDE? Now that would be nifty..never heard of one. 05:42:39 there was one built as an extension to google wave 05:42:53 it did indentation and syntax highlighting 05:43:22 http://ecco.sourceforge.net <- This looks like it is what you want. 05:44:03 sea4ever: woah 05:44:20 9ne is a web-based emacs clone too, along with kodengine 05:44:24 jewel [~jewel@196-215-16-132.dynamic.isadsl.co.za] has joined #lisp 05:45:16 OliverUv: I've been debating on a language to implement my computer algebra system, and first I was thinking haskell etc., but those were too restrictive; scheme, but that's too minimal/inconsistent (with other schemes); and I decided C. But upon my resurgence in interest in lisp processors (and lisp machines (and reading the SmugLispWeenies on cliki))), perhaps CL is worth trying. Though I desperately need to read up on CLOS and incorporating 05:45:16 types 05:45:25 ramble ramble 05:45:39 ldunn [~user@d110-32-145-146.sun800.vic.optusnet.com.au] has joined #lisp 05:46:33 -!- ldunn [~user@d110-32-145-146.sun800.vic.optusnet.com.au] has quit [Changing host] 05:46:33 ldunn [~user@unaffiliated/baddog144] has joined #lisp 05:46:33 CLOS is very easy to use if you want a Java-like object system 05:47:42 zomgbie [~jesus@h081217133138.dyn.cm.kabsi.at] has joined #lisp 05:47:47 Quadrescence: ummm, Macsyma? 05:47:56 fusss: What about it? 05:47:59 that's what I do now, but I kinda want to learn how to use its more advancd features 05:48:09 Quadrescence: google it 05:48:15 fusss: I know what it is. 05:48:31 But I'm asking why you brought it up. 05:49:27 -!- sea4ever [~sea@unaffiliated/sea4ever] has quit [Remote host closed the connection] 05:50:02 Quadrescence: there is hardly any reason not to think "Lisp" first, when it comes to implementing CAS apps. 05:50:38 gonzojive [~red@128.12.169.254] has joined #lisp 05:51:31 sea4ever [~sea@unaffiliated/sea4ever] has joined #lisp 05:51:50 Quadrescence: the whole thread is interesting, specially Dr. Fateman's input, but KMP lays down some history here: http://groups.google.com/group/comp.lang.lisp/msg/f3b93140c2f2e922?pli=1 05:51:53 If you want to write a CAS as a library, I agree with you. If you're going to be writing a language on top of a language anyway, and a runtime, then I'd disagree (that there's hardly any reason) 05:53:54 alright, good luck! 05:54:04 *fusss* back to fun 05:54:39 fusss: I'm not saying I'm going to use C (I *was*). I'm just saying bla bla there are arguments against it. 05:54:54 bla bla got it 05:56:17 -!- konr` [~user@201.82.129.247] has quit [Ping timeout: 260 seconds] 05:58:07 I find it unfortunate that just because lisp is suitable to write a differentiation program naturally, that general symbolic computation and deeper, more intricate algorithms can run along the same paradigm as the differentiation program, and therefore Lisp = Natural for CASs. 05:58:37 s/unfortunate that/unfortunate that people think that/ 05:59:25 fengshaun [~armin@blk-138-43-181.eastlink.ca] has joined #lisp 05:59:45 -!- fengshaun [~armin@blk-138-43-181.eastlink.ca] has quit [Quit: leaving] 06:00:29 holycow [~new@poco208-2.fredcanhelp.com] has joined #lisp 06:03:51 -!- milanj [~milanj_@109-92-124-204.dynamic.isp.telekom.rs] has quit [Quit: Leaving] 06:08:38 Quadrescence: Why do you think that is unfortunate? 06:09:41 [df] [~df@163.8.187.81.in-addr.arpa] has joined #lisp 06:11:07 beach: Because algebraic algorithms don't always follow lisp's "path of evaluation" like that, furthermore you can't just attach a few simple simplification rules after to get the desired result. 06:11:49 Quadrescence: I don't understand. Are you saying Lisp is less suited than other languages for such applications? 06:12:47 No. I am just saying that purporting lisp is good for algebra simply because there are simple examples, like differentiation, isn't a good reason to say lisp is a good language to use for the task. 06:13:38 simple examples (of symbolic computation) 06:13:39 Quadrescence: OK I see. The right reason is that Lisp is simply more suited than most other languages for most applications. 06:13:58 Including CAS systems. 06:14:06 And including C. 06:15:33 -!- jewel [~jewel@196-215-16-132.dynamic.isadsl.co.za] has quit [Ping timeout: 245 seconds] 06:16:48 I agree. The only reason I was considering C is because of its portability to old systems, a very tight structure (building a house with toothpicks), etc. I no longer think that C is the right choice to begin writing algebra stuff in, but at the time, I figured I would just implement the base CAS language in C, and then write the rest in that language 06:18:06 That sounds like Greenspunning to me. You would end up re-implementing a large part of CL. 06:18:44 trebor_dki [~user@mail.dki.tu-darmstadt.de] has joined #lisp 06:20:23 die already, sourceforge 06:22:40 -!- kami`` is now known as kami- 06:22:45 -!- kami- is now known as kami 06:23:30 beach: Yeah, that's how I feel now about it. 06:23:49 -!- daniel_ is now known as daniel 06:24:16 rbarraud [~rbarraud@118-92-27-142.dsl.dyn.ihug.co.nz] has joined #lisp 06:25:59 mrSpec [~Spec@unaffiliated/mrspec] has joined #lisp 06:25:59 found the motherlode, Re: web-based code editors. http://en.wikipedia.org/wiki/Comparison_of_Javascript-based_source_code_editors 06:39:18 gonzojive1 [~red@condi.Stanford.EDU] has joined #lisp 06:40:09 -!- quasisane [~sanep@c-76-24-80-97.hsd1.nh.comcast.net] has quit [Ping timeout: 260 seconds] 06:41:38 -!- mbohun [~mbohun@202.124.72.31] has quit [Remote host closed the connection] 06:42:18 Quadrescence: regarding greenspunning... have you ever seen IRAF? 06:42:31 No. 06:43:57 -!- Guest33861 is now known as seejay 06:44:08 -!- seejay [~seejay@plexyplanet.org] has quit [Changing host] 06:44:08 seejay [~seejay@unaffiliated/seejay] has joined #lisp 06:45:17 zbeasnyy [~mornfall@ip-89-102-11-250.net.upcbroadband.cz] has joined #lisp 06:45:38 Quadrescence: it greenspuns an operating system 06:45:52 in FORTRAN 77 06:45:55 i'm running the SBCL tests and redirecting stderr and stdout to files, but SBCL still manages to print output. I'm trying to debug what appears to be errors, and then errors occuring while errors are reported, and so on. my terminal overflows with the text. any idea why 2> .. 1> .. is not enough? 06:45:58 for a scientific soft 06:46:18 mbohun [~mbohun@202.124.74.146] has joined #lisp 06:53:19 -!- mathrick [~mathrick@users177.kollegienet.dk] has quit [Ping timeout: 258 seconds] 06:56:36 -!- aidalgol [~user@114-134-6-5.rurallink.co.nz] has quit [Remote host closed the connection] 06:58:08 -!- rbarraud [~rbarraud@118-92-27-142.dsl.dyn.ihug.co.nz] has quit [Remote host closed the connection] 06:58:27 aidalgol [~user@114-134-6-5.rurallink.co.nz] has joined #lisp 06:58:53 rbarraud [~rbarraud@118.92.27.142] has joined #lisp 07:03:50 loxs [~loxs@213.169.45.106] has joined #lisp 07:04:24 astoon [~astoon@213.141.244.246] has joined #lisp 07:06:53 mathrick [~mathrick@emp.nat.sdu.dk] has joined #lisp 07:09:58 quasisane [~sanep@pool-71-161-71-222.cncdnh.east.myfairpoint.net] has joined #lisp 07:12:32 -!- loomer [~loomer@unaffiliated/loomer] has quit [Quit: Leaving] 07:13:11 -!- SegFaultAX [~SegFaultA@c-98-234-1-162.hsd1.ca.comcast.net] has quit [Quit: Lost terminal] 07:14:23 -!- astoon [~astoon@213.141.244.246] has quit [Ping timeout: 240 seconds] 07:16:35 gonzojive: because sbcl also writes directly to the terminal 07:22:29 nha [~prefect@imamac13.epfl.ch] has joined #lisp 07:24:14 astoon [~astoon@213.141.244.246] has joined #lisp 07:26:52 jdz [~jdz@193.206.22.97] has joined #lisp 07:32:29 -!- rbarraud [~rbarraud@118.92.27.142] has quit [Excess Flood] 07:32:44 rbarraud [~rbarraud@118-92-27-142.dsl.dyn.ihug.co.nz] has joined #lisp 07:32:49 -!- zbeasnyy [~mornfall@ip-89-102-11-250.net.upcbroadband.cz] has quit [Quit: the old ways are lost] 07:34:29 jan247 [~jan247@unaffiliated/jan247] has joined #lisp 07:36:43 Beetny [~Beetny@ppp118-208-30-116.lns20.bne1.internode.on.net] has joined #lisp 07:38:08 mk2 [~user@evanspc2.rai.umds.ac.uk] has joined #lisp 07:47:40 Krystof: Hi, do you know why primitive-type is dropped from load TNs? 07:50:30 tcr [~tcr@81-233-246-97-no37.tbcn.telia.com] has joined #lisp 07:50:50 -!- Beetny [~Beetny@ppp118-208-30-116.lns20.bne1.internode.on.net] has quit [Read error: Connection reset by peer] 07:50:53 bozhidar [~user@212.50.14.187] has joined #lisp 07:54:37 no 07:55:07 to start finding out, I'd undrop it, build, and see what breaks 07:59:48 well, I've built it succesfully, but don't know how to see what broke beyond that... 08:06:20 -!- rbarraud [~rbarraud@118-92-27-142.dsl.dyn.ihug.co.nz] has quit [Remote host closed the connection] 08:07:45 <_8david> if it passes its own test suite and the ansi tests, the compiler can't be that buggy 08:08:04 how can I run the tests? 08:08:13 rbarraud [~rbarraud@118-92-27-142.dsl.dyn.ihug.co.nz] has joined #lisp 08:09:30 TraumaPony [~TraumaPon@202-173-162-181.dyn.iinet.net.au] has joined #lisp 08:09:32 ah, I see - run-tests.sh 08:12:11 -!- rbarraud [~rbarraud@118-92-27-142.dsl.dyn.ihug.co.nz] has quit [Remote host closed the connection] 08:12:47 rbarraud [~rbarraud@118-92-27-142.dsl.dyn.ihug.co.nz] has joined #lisp 08:13:29 ejs [~eugen@77.222.151.102] has joined #lisp 08:14:36 kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has joined #lisp 08:15:32 -!- gzip4 [~xxx@78.108.73.250] has quit [Remote host closed the connection] 08:15:58 -!- BrianRice [~water@c-98-246-165-205.hsd1.or.comcast.net] has quit [Ping timeout: 255 seconds] 08:16:03 good morning 08:16:45 levente_meszaros [~levente_m@4d6f5d3b.adsl.enternet.hu] has joined #lisp 08:16:57 BrianRice [~water@c-98-246-165-205.hsd1.or.comcast.net] has joined #lisp 08:16:57 good morning kiuma 08:17:55 Sumpen [~Sumpen@81-232-77-93-no46.tbcn.telia.com] has joined #lisp 08:18:17 -!- Sumpen [~Sumpen@81-232-77-93-no46.tbcn.telia.com] has quit [Client Quit] 08:19:55 legumbre_ [~leo@r190-135-22-212.dialup.adsl.anteldata.net.uy] has joined #lisp 08:21:05 Sumpen [~Sumpen@81-232-77-93-no46.tbcn.telia.com] has joined #lisp 08:21:30 Davidbrcz [~david@212-198-91-47.rev.numericable.fr] has joined #lisp 08:22:08 -!- legumbre [~leo@r190-135-55-133.dialup.adsl.anteldata.net.uy] has quit [Ping timeout: 272 seconds] 08:24:24 slash_ [~unknown@p4FF0A0EE.dip.t-dialin.net] has joined #lisp 08:25:06 -!- Amadiro [~whoppix@ti0021a380-dhcp1747.bb.online.no] has quit [Quit: A subtle thought that is in error may yet give rise to fruitful inquiry that can establish truths of great value.] 08:27:08 fiveop [~fiveop@erft-d932fa93.pool.mediaWays.net] has joined #lisp 08:27:50 -!- vandemar [spin@2001:470:1f10:56b::4] has quit [Ping timeout: 272 seconds] 08:28:11 vandemar [syndicate@2001:470:1f10:56b::4] has joined #lisp 08:28:55 varjag [~eugene@122.62-97-226.bkkb.no] has joined #lisp 08:30:45 -!- Krystof [~csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has quit [Ping timeout: 260 seconds] 08:32:20 -!- khrum42 [~khrum42@2001:a60:f05d:1::dead] has quit [Quit: Leaving.] 08:32:39 tfb [~tfb@92.40.183.158.sub.mbb.three.co.uk] has joined #lisp 08:34:53 Thorium [~opera@188.26.233.203] has joined #lisp 08:35:46 hello 08:36:10 how can I construct and return a list from a loop like this one 08:36:15 (loop while (>= (/ num 10) (/ 1 10)) do 08:36:18 ? 08:37:34 collect 08:37:48 doesn't work 08:37:54 it says collect is undefined 08:38:05 (loop while ... collect ...) 08:38:10 jgracin [~jgracin@dh111-186.xnet.hr] has joined #lisp 08:38:21 this is my function 08:38:25 (defun getdigits (num) 08:38:25 (loop while (>= (/ num 10) (/ 1 10)) do 08:38:25 (append (list (mod num 10))) 08:38:25 (setq num (floor (/ num 10))) )) 08:38:34 if i use collect instead of append 08:38:35 homie [91fd03c5@gateway/web/freenode/ip.145.253.3.197] has joined #lisp 08:38:42 it says collect is undefined 08:39:02 and with append the function returns nill 08:39:17 thats because collect is a loop keyword not a function. Say (loop ... collect ...) 08:39:17 Thorium: collect is a loop keyword like DO 08:40:49 xan_ [~xan@112.158.62.134] has joined #lisp 08:40:50 -!- homie [91fd03c5@gateway/web/freenode/ip.145.253.3.197] has quit [Client Quit] 08:40:56 jan247_ [~jan247@unaffiliated/jan247] has joined #lisp 08:41:01 -!- jan247 [~jan247@unaffiliated/jan247] has quit [Read error: Connection reset by peer] 08:41:02 -!- jan247_ is now known as jan247 08:41:09 I wrote the function like this 08:41:14 (defun getdigits (num) 08:41:14 (loop while (>= (/ num 10) (/ 1 10)) do 08:41:14 (collect (mod num 10)) 08:41:14 (setq num (floor (/ num 10))) )) 08:41:18 antoszka [~antoszka@unaffiliated/antoszka] has joined #lisp 08:41:29 -!- Salamander [~Salamande@ppp121-45-85-224.lns20.adl6.internode.on.net] has quit [Ping timeout: 260 seconds] 08:41:35 EVAL: undefined function COLLECT 08:41:44 can you tell me what am I doing wrong ? 08:42:43 Thorium: you are using COLLECT like a function 08:42:55 it is a loop keyword, like tfb and I have been trying to tell you (: 08:43:15 try http://www.gigamonkeys.com/book/loop-for-black-belts.html - this explains loop pretty nicely. 08:43:18 how it should be ? like this? 08:43:20 (loop while (>= (/ num 10) (/ 1 10)) 08:43:20 collect (mod num 10) 08:43:34 yes 08:43:41 it doesnt work! 08:44:04 it loops forever? 08:44:14 (that's because you don't modify NUM 08:44:16 -!- adu [~ajr@pool-71-191-154-37.washdc.fios.verizon.net] has quit [Quit: adu] 08:44:34 try the FOR loop keyword 08:45:00 pdelgallego [~pdelgalle@1503031474.dhcp.dbnet.dk] has joined #lisp 08:46:13 (defun getdigits (num) 08:46:13 (loop while (>= (/ num 10) (/ 1 10)) 08:46:13 collect (mod num 10) 08:46:13 (setq num (floor (/ num 10))) )) 08:46:24 attila_lendvai [~attila_le@apn-94-44-14-138.vodafone.hu] has joined #lisp 08:46:36 and the error 08:46:40 LOOP: illegal syntax near (SETQ NUM (FLOOR (/ NUM 10))) in 08:46:40 (LOOP WHILE (>= (/ NUM 10) (/ 1 10)) COLLECT (MOD NUM 10) (SETQ NUM (FLOOR (/ NUM 10)))) 08:47:24 I suggest reading that link 08:47:24 -!- Axioplase is now known as Axioplase_ 08:47:35 do you know what this error means ? 08:47:39 I already read it 08:47:51 there is no example there 08:48:00 with a loop while to return a list 08:48:16 I'm trying to use loop while and return a list from this loop 08:48:21 To execute lisp forms in a LOOP you have to use DO 08:48:23 what is the proper way to do it ? 08:50:29 can you show me an example with loop while do which uses collect to return a list 08:50:30 ? 08:50:33 -!- OEP [~OEP@c-76-27-129-70.hsd1.al.comcast.net] has quit [Ping timeout: 245 seconds] 08:51:55 -!- cmm [~cmm@bzq-79-181-203-193.red.bezeqint.net] has quit [Ping timeout: 252 seconds] 08:52:29 cmm [~cmm@bzq-79-181-203-193.red.bezeqint.net] has joined #lisp 08:53:44 -!- Khisanth [~Khisanth@pool-96-250-28-106.nycmny.east.verizon.net] has quit [Quit: Leaving] 08:54:19 Khisanth [~Khisanth@pool-96-250-28-106.nycmny.east.verizon.net] has joined #lisp 08:55:47 I can't ind any example which uses while on that link 08:55:58 Salamander [~Salamande@ppp121-45-27-7.lns20.adl2.internode.on.net] has joined #lisp 08:56:17 do you know some example which uses loop while do and collect to return a list from a loop 08:56:18 ?? 08:56:31 is this even possible ? 08:58:40 _8david: For some reason, 'running impure tests' hangs on my system even if I check out pure 1.0.41.56 08:58:41 gemelen [~shelta@shpd-92-101-155-254.vologda.ru] has joined #lisp 08:59:05 do you know some places where I can post some lisp questions ? 08:59:10 red1ynx [~Dzmitry@91.149.140.201] has joined #lisp 08:59:49 Thorium: (map 'list #'digit-char-p (princ-to-string 12345)) 09:00:26 man I'm not trying to write that function 09:00:41 I want just to understand how can I use loop with while and collect 09:01:47 Pretty much how you try to do, all you're missing is that you need the DO loop keyword to execute arbitrary lisp forms in a LOOP as I said before, man. 09:01:56 Athas [~athas@shop3.diku.dk] has joined #lisp 09:02:40 antifuchs: Is SBCL hanging on 'Running impure tests (#)' some kind of known problem? 09:02:57 Or is it my system? 09:05:03 tcr: can you rewrite this to work ? so I can see where is the mistake ... 09:05:03 -!- gonzojive [~red@128.12.169.254] has quit [Read error: Connection reset by peer] 09:05:05 gonzojive [~red@128.12.169.254] has joined #lisp 09:05:07 (loop while (>= (/ num 10) (/ 1 10)) do 09:05:07 collect (mod num 10) 09:05:07 (setq num (floor (/ num 10))) )) 09:05:25 -!- attila_lendvai [~attila_le@apn-94-44-14-138.vodafone.hu] has quit [Remote host closed the connection] 09:06:15 collect (mod ) do (setq ) 09:06:25 Thorium: http://paste.lisp.org/display/113888 09:06:29 for instance 09:07:19 (actually mine does not have do and collect, just collect, but you can obviously add do clauses in the obvious way) 09:08:30 that one uses for.... 09:09:25 I give up. LOOP is not that hard. 09:09:31 Hello, i have a list and i want to print it in a tree structure. Could someone help me do that? 09:09:33 :) 09:09:50 thanks tcr 09:09:53 it works now ... 09:14:32 antifuchs: It looks exactly like this bug report: https://bugs.launchpad.net/sbcl/+bug/578391 09:17:43 -!- levente_meszaros [~levente_m@4d6f5d3b.adsl.enternet.hu] has quit [Ping timeout: 272 seconds] 09:23:12 slash_1 [~unknown@p5DD1D483.dip.t-dialin.net] has joined #lisp 09:23:49 -!- slash_ [~unknown@p4FF0A0EE.dip.t-dialin.net] has quit [Ping timeout: 252 seconds] 09:23:56 Beetny [~Beetny@ppp118-208-30-116.lns20.bne1.internode.on.net] has joined #lisp 09:24:05 -!- slash_1 [~unknown@p5DD1D483.dip.t-dialin.net] has quit [Client Quit] 09:24:35 -!- Thorium [~opera@188.26.233.203] has left #lisp 09:27:09 -!- tcr [~tcr@81-233-246-97-no37.tbcn.telia.com] has quit [Read error: Operation timed out] 09:28:30 tcr [~tcr@81-233-246-97-no37.tbcn.telia.com] has joined #lisp 09:30:11 -!- TraumaPony [~TraumaPon@202-173-162-181.dyn.iinet.net.au] has quit [Read error: Connection reset by peer] 09:31:03 Beetny` [~Beetny@ppp118-208-30-116.lns20.bne1.internode.on.net] has joined #lisp 09:31:20 HG` [~HG@xdsl-92-252-89-161.dip.osnanet.de] has joined #lisp 09:31:26 TraumaPony [~TraumaPon@202-173-162-181.dyn.iinet.net.au] has joined #lisp 09:32:27 debugger invoked on a SB-BSD-SOCKETS:UNKNOWN-PROTOCOL in thread # "initial thread" RUNNING 09:32:28 {1002AC2F71}>: 09:32:28 Protocol not found: "tcp" 09:32:36 when running slime 09:33:07 does somebody already encountered that error ? 09:33:21 I'm on a very classic linux machine 09:33:22 -!- Beetny [~Beetny@ppp118-208-30-116.lns20.bne1.internode.on.net] has quit [Ping timeout: 255 seconds] 09:33:30 it worked yesterday, before I update my system 09:37:41 -!- neoesque [~neoesque@210.59.147.232] has quit [Remote host closed the connection] 09:37:44 removing ~/.slime and starting over didn't change anything 09:39:11 galdor: check out /etc/protocols - does it list "tcp" 09:39:13 ? 09:40:01 oh crap this file doesn't exist (anymore) 09:40:09 -!- Phoodus [foo@174-22-199-91.phnx.qwest.net] has quit [Read error: Connection reset by peer] 09:40:12 weeeeelll (: 09:40:14 why do I have the feeling that the last system update broke something 09:40:26 without it, I suspect getprotobyname will fail 09:42:43 you were right, reinstalling the package providing /etc/services worked 09:42:44 thank you :) 09:42:49 attila_lendvai [~attila_le@apn-94-44-14-138.vodafone.hu] has joined #lisp 09:43:14 minion: memo for gigamonkey: what's the licensing status of PCL code samples? 09:43:15 Remembered. I'll tell gigamonkey when he/she/it next speaks. 09:45:15 -!- gonzojive [~red@128.12.169.254] has quit [Quit: gonzojive] 09:50:08 Joreji [~thomas@77-066.eduroam.RWTH-Aachen.DE] has joined #lisp 09:50:21 Why are sbcl sources using packages like sb!c, and how do they turn into stuff like sb-c in the end result? 09:52:13 angavrilov: to separate the host and target environment, i think. i'm not sure how they turn. 09:52:29 angavrilov: I wrote a paper describing the build process and its rationale 09:52:38 dfkjjkfd [~paulh@143-13-ftth.onsnetstudenten.nl] has joined #lisp 09:52:38 -!- dfkjjkfd [~paulh@143-13-ftth.onsnetstudenten.nl] has quit [Client Quit] 09:52:45 dfkjjkfd [~paulh@143-13-ftth.onsnetstudenten.nl] has joined #lisp 09:52:54 angavrilov: you can get it from 09:53:42 If I want to add a new sb-sse package, I should do it in package-data-list.lisp-expr? 09:55:36 yes. (Do you really need a new package? How many symbols are we talking about? Would it fit better in sb-vm?) 09:58:48 My ECL contrib exports 291 09:59:42 Here is where most of them are defined: http://github.com/angavrilov/ecl-sse/blob/master/contrib/sse/sse-core.lsp 10:00:56 mulander [mulander@078088239019.lubin.vectranet.pl] has joined #lisp 10:03:38 Salamander_ [~Salamande@ppp118-210-132-215.lns20.adl6.internode.on.net] has joined #lisp 10:04:02 nus [~nus@unaffiliated/nus] has joined #lisp 10:05:22 Xof: incidentally, since pkhuong appears to be very busy, maybe you could comment on the symbol naming scheme re possibility of a compatible implementation being accepted into sbcl? here is my mail detailing my decisions: http://sourceforge.net/mailarchive/message.php?msg_name=AANLkTikaxvna_%2BUfE7dBfdbsec6_pu23Hp8UESxEpZua%40mail.gmail.com 10:05:44 -!- Salamander [~Salamande@ppp121-45-27-7.lns20.adl2.internode.on.net] has quit [Ping timeout: 252 seconds] 10:06:22 well, one thing to suggest might be to structure your implementation as a contrib/ rather than something that is automatically included 10:06:22 -!- rbarraud [~rbarraud@118-92-27-142.dsl.dyn.ihug.co.nz] has quit [Ping timeout: 240 seconds] 10:06:44 Salamander__ [~Salamande@ppp121-45-101-184.lns20.adl6.internode.on.net] has joined #lisp 10:06:46 then you don't have to worry so much about the sbcl build 10:07:19 and also it will be obvious when someone reviews your patch what in the core code needs changing to support your work 10:07:35 by nature it would mostly consist of VOPs and compiler transforms; how does that match the contrib paradigm in sbcl? 10:08:52 also, I'm just trying to see if I can continue from pkhuong's prototype code, and I definitely cannot do it from scratch... 10:09:02 -!- Salamander_ [~Salamande@ppp118-210-132-215.lns20.adl6.internode.on.net] has quit [Ping timeout: 252 seconds] 10:10:03 -!- jmcphers [~jmcphers@218.185.108.156] has quit [Remote host closed the connection] 10:10:03 Krystof [~csr21@howells.doc.gold.ac.uk] has joined #lisp 10:10:46 here's what I have at this moment: http://github.com/angavrilov/sbcl/commits/sse/ 10:14:58 angavrilov: for an example of a contrib which is similarly mostly vops, see sb-rotate-byte 10:15:16 (I can't review code now) 10:16:13 ignas [~ignas@78-60-73-85.static.zebra.lt] has joined #lisp 10:16:39 -!- xyxxyyy [~xyxu@58.41.13.12] has quit [Quit: Leaving.] 10:22:47 -!- Salamander__ [~Salamande@ppp121-45-101-184.lns20.adl6.internode.on.net] has quit [Ping timeout: 252 seconds] 10:23:29 Salamander__ [~Salamande@ppp118-210-250-122.lns20.adl6.internode.on.net] has joined #lisp 10:25:22 -!- salva [~salva@105.11.117.91.dynamic.mundo-r.com] has quit [Quit: salva] 10:26:11 mstevens [~mstevens@fsf/member/pdpc.active.mstevens] has joined #lisp 10:34:51 Salamander_ [~Salamande@ppp121-45-27-195.lns20.adl2.internode.on.net] has joined #lisp 10:35:22 Yuuhi [benni@p5483D768.dip.t-dialin.net] has joined #lisp 10:37:05 -!- Salamander__ [~Salamande@ppp118-210-250-122.lns20.adl6.internode.on.net] has quit [Ping timeout: 252 seconds] 10:38:28 -!- xan_ [~xan@112.158.62.134] has quit [Ping timeout: 245 seconds] 10:41:53 ziarkaen [~ziarkaen@87.113.173.90] has joined #lisp 10:42:18 -!- ssideris [~sideris@93-97-178-232.zone5.bethere.co.uk] has quit [Ping timeout: 240 seconds] 10:47:15 levente_meszaros [~levente_m@4d6f5d3b.adsl.enternet.hu] has joined #lisp 10:49:34 -!- mbohun [~mbohun@202.124.74.146] has quit [Quit: Leaving] 10:51:25 -!- ldunn [~user@unaffiliated/baddog144] has quit [Remote host closed the connection] 10:51:46 At which stage of compilation does (declaim (inline)) style inlining happen? I mean relative to transforms, VOP expansion etc 10:51:49 -!- mstevens [~mstevens@fsf/member/pdpc.active.mstevens] has quit [Quit: leaving] 10:52:04 very early 10:52:23 xyxxyyy [~xyxu@114.93.152.11] has joined #lisp 10:53:33 -!- nus [~nus@unaffiliated/nus] has quit [Ping timeout: 260 seconds] 10:54:42 -!- krl [~krl@port-87-193-235-133.static.qsc.de] has quit [Remote host closed the connection] 10:54:53 krl [~krl@port-87-193-235-133.static.qsc.de] has joined #lisp 10:57:09 urandom__ [~user@p548A57CB.dip.t-dialin.net] has joined #lisp 11:09:54 -!- aidalgol [~user@114-134-6-5.rurallink.co.nz] has quit [Quit: zZzZzZz] 11:11:51 gravicappa [~gravicapp@ppp85-141-167-179.pppoe.mtu-net.ru] has joined #lisp 11:12:20 -!- vandemar [syndicate@2001:470:1f10:56b::4] has quit [Read error: Operation timed out] 11:12:28 -!- jan247 [~jan247@unaffiliated/jan247] has quit [Quit: jan247] 11:14:29 Is it possible for a transform body to know if some of its arguments are constants, without specifying it as a dispatch constraint? And can it determine if some of arguments are guaranteed to be the same, e.g. when the call is (foo a a a a)? 11:14:42 vandemar [nonserviam@2001:470:1f10:56b::4] has joined #lisp 11:15:50 tfb_ [~tfb@94.197.70.214.threembb.co.uk] has joined #lisp 11:15:55 -!- tfb [~tfb@92.40.183.158.sub.mbb.three.co.uk] has quit [Disconnected by services] 11:16:02 -!- tfb_ is now known as tfb 11:19:18 nus [~nus@unaffiliated/nus] has joined #lisp 11:20:56 slash_ [~unknown@p5DD1D8DD.dip.t-dialin.net] has joined #lisp 11:23:12 angavrilov: a compilermacro-function gets the form that was input. You could eq on the symbols and see if they are the same that way 11:29:19 -!- attila_lendvai [~attila_le@apn-94-44-14-138.vodafone.hu] has quit [Ping timeout: 258 seconds] 11:31:17 -!- levente_meszaros [~levente_m@4d6f5d3b.adsl.enternet.hu] has quit [Ping timeout: 255 seconds] 11:31:29 Jabberwockey [~Jens@port-12485.pppoe.wtnet.de] has joined #lisp 11:42:40 -!- Jasko [~tjasko@c-174-59-223-208.hsd1.pa.comcast.net] has quit [Quit: Leaving] 11:42:46 mbohun [~mbohun@ppp115-156.static.internode.on.net] has joined #lisp 11:49:30 -!- astoon [~astoon@213.141.244.246] has quit [Quit: astoon] 11:58:30 Bronsa [~bronsa@host152-175-dynamic.7-79-r.retail.telecomitalia.it] has joined #lisp 12:00:34 -!- varjag [~eugene@122.62-97-226.bkkb.no] has quit [Remote host closed the connection] 12:07:50 hargettp [~anonymous@pool-71-184-181-149.bstnma.east.verizon.net] has joined #lisp 12:11:29 Jasko [~tjasko@209.74.44.225] has joined #lisp 12:12:51 -!- spradnyesh [~pradyus@nat/yahoo/x-sjtnmzklcghxrgjh] has left #lisp 12:14:49 -!- red1ynx [~Dzmitry@91.149.140.201] has quit [Quit: red1ynx] 12:15:20 -!- Elench [~user@unaffiliated/elench] has quit [Read error: No route to host] 12:15:30 -!- zeroish [~zeroish@135.207.174.50] has quit [Ping timeout: 240 seconds] 12:16:01 ziarkaen_ [~ziarkaen@87.112.140.185] has joined #lisp 12:18:41 Elench [~user@unaffiliated/elench] has joined #lisp 12:19:18 -!- ziarkaen [~ziarkaen@87.113.173.90] has quit [Ping timeout: 245 seconds] 12:19:29 moocow [~new@poco208-2.fredcanhelp.com] has joined #lisp 12:19:50 -!- holycow [~new@poco208-2.fredcanhelp.com] has quit [Ping timeout: 265 seconds] 12:21:35 Adlai [~adlai@unaffiliated/adlai] has joined #lisp 12:28:34 -!- epoxy [~epoxy@dialin.inttek.net] has quit [Quit: Leaving] 12:32:42 -!- fusss [~chatzilla@cpe-184-59-201-74.new.res.rr.com] has quit [Ping timeout: 240 seconds] 12:32:44 xyxxyyy1 [~xyxu@58.41.11.200] has joined #lisp 12:34:27 hohoho [~hohoho@ntkngw229253.kngw.nt.ftth.ppp.infoweb.ne.jp] has joined #lisp 12:35:01 petercoulton [~petercoul@cpc4-midd16-2-0-cust402.11-1.cable.virginmedia.com] has joined #lisp 12:35:38 -!- xyxxyyy [~xyxu@114.93.152.11] has quit [Ping timeout: 258 seconds] 12:39:30 levente_meszaros [~levente_m@4d6f5d3b.adsl.enternet.hu] has joined #lisp 12:41:01 -!- Beetny` [~Beetny@ppp118-208-30-116.lns20.bne1.internode.on.net] has quit [Ping timeout: 272 seconds] 12:41:29 jaiball [~justin@c-76-119-131-178.hsd1.ct.comcast.net] has joined #lisp 12:42:22 dose [~tristan@offyourtrol.li] has joined #lisp 12:42:56 -!- tcr [~tcr@81-233-246-97-no37.tbcn.telia.com] has quit [Ping timeout: 265 seconds] 12:48:14 vp8dmh [~clarkema@bs0176.nerc-bas.ac.uk] has joined #lisp 12:51:32 -!- vp8dmh [~clarkema@bs0176.nerc-bas.ac.uk] has quit [Client Quit] 12:54:57 Avisch [~Avisch@70-101-99-64.dsl1-field.roc.ny.frontiernet.net] has joined #lisp 12:55:20 -!- Avisch [~Avisch@70-101-99-64.dsl1-field.roc.ny.frontiernet.net] has quit [Client Quit] 12:58:05 -!- vu3rdd [~vu3rdd@164.164.250.10] has quit [Remote host closed the connection] 12:58:41 HarryS [H@harry.lu] has joined #lisp 13:01:09 -!- ejs [~eugen@77.222.151.102] has quit [Quit: This computer has gone to sleep] 13:01:15 -!- tankrim [~user@pdpc/supporter/active/tankrim] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 13:03:11 -!- mathrick [~mathrick@emp.nat.sdu.dk] has quit [Remote host closed the connection] 13:04:13 LiamH [~none@pdp8.nrl.navy.mil] has joined #lisp 13:05:16 -!- kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has quit [Ping timeout: 265 seconds] 13:06:22 kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has joined #lisp 13:07:29 -!- Tasunteld [~jsz@rps312.ovh.net] has quit [Ping timeout: 260 seconds] 13:07:50 tritchey [~tritchey@c-98-226-81-194.hsd1.in.comcast.net] has joined #lisp 13:07:53 Tasunteld [~jsz@rps312.ovh.net] has joined #lisp 13:07:57 -!- acieroid [~acieroid@ks23738.kimsufi.com] has quit [Ping timeout: 260 seconds] 13:08:36 -!- adeht [void@91.121.18.93] has quit [Ping timeout: 258 seconds] 13:08:44 _death [void@91.121.18.93] has joined #lisp 13:10:24 ssideris [~sideris@93-97-178-232.zone5.bethere.co.uk] has joined #lisp 13:10:43 angavrilov: You mean deftransform or some other transform? 13:11:29 I don't yet know what is the exact difference between e.g. deftransform and defoptimizer... 13:12:28 defoptimizer is primarily used for type propagation. deftransform transforms the code into something else, presumably simpler or better. 13:13:24 And yes, deftransforms can tell if the argument is a constant, and it can figure out value of that constant. 13:14:09 -!- sellout [~greg@c-24-61-13-161.hsd1.ma.comcast.net] has quit [Quit: sellout] 13:14:45 _s1gma [~d.d.derp@77.107.164.131] has joined #lisp 13:15:07 -!- _s1gma [~d.d.derp@77.107.164.131] has quit [Max SendQ exceeded] 13:15:44 I'm considering something like partial constant folding for a certain set of functions, i.e. mutating the set of primitive operations used to build a complex result based on which args are constant or provably equal to each other 13:17:33 The functions to look at is constant-continuation-p and continuation-value to see if it's constant and get the value. At least for cmucl. 13:18:45 same-leaf-ref-p (and non-const-same-leaf-ref-p) will tell you if the args are the "same". 13:20:22 -!- galaxywatcher [~galaxywat@ppp-58-8-52-3.revip2.asianet.co.th] has quit [Ping timeout: 240 seconds] 13:20:33 sounds good :) 13:24:39 -!- bhyde [~Adium@c-66-30-201-212.hsd1.ma.comcast.net] has left #lisp 13:25:24 acieroid [~acieroid@ks23738.kimsufi.com] has joined #lisp 13:26:57 sebyte [~sebyte@213.229.74.8] has joined #lisp 13:27:07 galaxywatcher [~galaxywat@ppp-58-8-47-103.revip2.asianet.co.th] has joined #lisp 13:27:34 stassats` [~stassats@wikipedia/stassats] has joined #lisp 13:28:18 -!- antoszka [~antoszka@unaffiliated/antoszka] has quit [Ping timeout: 240 seconds] 13:28:25 antoszka [~antoszka@unaffiliated/antoszka] has joined #lisp 13:30:05 stassats`: How do I use slime's disassemble-frame? There are some compiler warnings there that I'd like to fix. (And maybe make it work on something other than cmucl/x86.) 13:32:33 Are higher bits of xmm double-reg and single-reg in x86_64 SBCL guaranteed to be 0, or can they contain random garbage? 13:33:48 I'd look to see if realpart and imagpart clear out the high bits. 13:34:04 I think that's the only way to get stuff in the high bits. 13:34:17 ziarkaen__ [~ziarkaen@87.113.250.130] has joined #lisp 13:35:30 rtoym: by pressing D on a frame in SLDB 13:36:10 Cool. That seems to work for a Lisp frame. 13:36:49 The doc string for sb-ext:run-program reads "The program arguments and the environment are encoded using the default external format for streams." How do you alter this default external format and are there any issues with using UTF-8? 13:36:51 stipet [~user@ua.blixtvik.net] has joined #lisp 13:37:03 looks like it uses GDB for other frames 13:37:48 rtoym: it does seem that it clears them out, and for single-float that relies on the high part being zero 13:38:01 sellout [~greg@2002:4855:eb9a:0:21d:4fff:fefe:c504] has joined #lisp 13:38:05 sebyte: sb-impl::*default-external-format* 13:38:08 -!- ziarkaen_ [~ziarkaen@87.112.140.185] has quit [Ping timeout: 265 seconds] 13:38:20 though it's not exported, with all kinds of strings attached 13:39:01 -!- loxs [~loxs@213.169.45.106] has quit [Read error: No route to host] 13:40:35 stassats`: Ok. D doesn't work for a frame call_into_lisp. I'll look into it. 13:41:49 -!- galaxywatcher [~galaxywat@ppp-58-8-47-103.revip2.asianet.co.th] has quit [Ping timeout: 272 seconds] 13:42:00 -!- psilord2 [~psilord@adsl-99-154-14-142.dsl.mdsnwi.sbcglobal.net] has left #lisp 13:42:29 -!- gemelen [~shelta@shpd-92-101-155-254.vologda.ru] has quit [Ping timeout: 260 seconds] 13:42:41 stassats`: thanks - i'm running into an issue with cl-sendmail and non-latin characters in full names. if I wrap my (with-email ... ) forms in a let which binds sb-impl::*default-external-format* to utf-8, rather than changing its value globally, i should be OK, no? 13:43:02 -!- stipet [~user@ua.blixtvik.net] has left #lisp 13:43:13 do you know any work/paper related to fexprs *and* compilation through partial evaluating the fexpr based interpreter? 13:43:46 sebyte: it should be, but being unexported it can disappear in the next release or whatnot 13:44:57 stassats`: what's the Right Thing to do then, IYO? 13:45:12 I built a toy interpreter supporting quote, if, progn, lambda (limited), applicative primitive functions and macros based on *fexprs* only 13:45:26 sebyte: just use it 13:45:43 stassats`: good answer :))) 13:45:44 galaxywatcher [~galaxywat@ppp-61-90-20-10.revip.asianet.co.th] has joined #lisp 13:45:52 -!- sonnym [~evissecer@rrcs-184-74-137-167.nys.biz.rr.com] has quit [Quit: Leaving.] 13:45:57 levente_meszaros: look at picolisp http://www.software-lab.de/down.html 13:46:00 I'm compiling programs with partially evaluating the CL interpreter 13:47:20 you can control evaluation of args as you please in picolisp 13:48:09 hlavaty, that's nice, with fexprs you can even do even more 13:48:09 dlowe [~dlowe@ita4fw1.itasoftware.com] has joined #lisp 13:48:17 *levente_meszaros* looking at picolisp 13:48:22 levente_meszaros: what even more? 13:48:56 hlavaty, sure, evaluate any parts of arguments in any order, any number of times, restructure them upside-down, etc. 13:49:17 whatever weird idea you can come up with 13:49:52 and the interesting thing I would like to know how far can one get with partially evaluating such an interpreter 13:49:59 i don't follow, what other idea could you come up with other then evaluating or not evaluating args? 13:50:11 now I'm compiling to CL and I get some nice results for some simple examples 13:51:27 in picolisp you can do the stuff you described, e.g. (de foo L ...) where ... is the function body that can do whatever with the unevaluated args in L 13:52:24 hlavaty, that's an fexpr then 13:52:41 i see, didn't know that scientific name ;-) 13:52:45 so you don't need to turn evaluation on/off per arg, because you can always call back eval if you wanted to 13:53:13 mathrick [~mathrick@users177.kollegienet.dk] has joined #lisp 13:53:21 and you can implement IF, PROGN, LET, macros, strict functions, lazy functions, etc. on top of that 13:53:49 except that it is extremely hard if not impossible to compile such a language where you are allowed to write fexprs 13:53:51 something like that, picolisp have some conventions for a few useful cases, look at http://www.software-lab.de/doc/ref.html#ev 13:54:18 sebyte: in sbcl at least if your shell does 'export LANG=en_US.UTF-8' in the rc and the appropriate locale is set, sbcl will pick up utf-8 when you start. 13:54:59 I think ccl exhibits similar behaviour. 13:55:39 levente_meszaros: well, you need things to be static to reasonable degree for compiler, picolisp is an interpreter and completely dynamic (although fast and extremely practical) 13:56:05 -!- galaxywatcher [~galaxywat@ppp-61-90-20-10.revip.asianet.co.th] has quit [Quit: galaxywatcher] 13:56:07 hlavaty, I don't agree with the static stuff 13:56:16 why not? 13:56:33 galaxywatcher [~galaxywat@ppp-61-90-20-10.revip.asianet.co.th] has joined #lisp 13:56:56 depends on what static means to you 13:57:21 in my interpreter the "special" form IF can be shadowed 13:57:58 -!- jgracin [~jgracin@dh111-186.xnet.hr] has quit [Remote host closed the connection] 13:58:00 but the compiler can still compile IF to CL:IF if that's not the case 13:59:00 compiling a form is done by partially evaluating the interpreter wrt that form in a certain environment 13:59:11 -!- pdelgallego [~pdelgalle@1503031474.dhcp.dbnet.dk] has quit [Ping timeout: 276 seconds] 14:00:38 -!- _death is now known as adeht 14:00:56 levente_meszaros: hmm, i'd have to think more what "static" means ;-) also what does a "compiler" mean? 14:01:08 -!- mjonsson [~mjonsson@cpe-98-14-173-5.nyc.res.rr.com] has quit [Ping timeout: 276 seconds] 14:01:10 xan_ [~xan@220.118.197.183] has joined #lisp 14:01:25 heh, for that matter explain me "interpreter" too (-; 14:02:15 it's easy to give an interpretation for those :) which will be argued for hours.... 14:02:24 levente_meszaros: do you have hygienic macros? 14:02:36 gonzojive1: Do you ever commit to cl-twitter? 14:02:59 pkhuong_, well, since fexprs get the environment as an argument, I guess it's doable, right? 14:03:09 mstevens [~mstevens@fsf/member/pdpc.active.mstevens] has joined #lisp 14:03:15 for example, the rough idea in picolisp is that you have a virtual machine which interprets its datastructures (say cons trees). roughly speaking you don't need a compiler because you write in "bytecode" ;-) 14:03:37 pkhuong_: Xof above suggested implementing SSE intrinsics as a contrib. What do you think? 14:04:34 levente_meszaros: without it, it becomes much harder to write correct macros... for instance, if I write CL' on top of CL, without hygiene, it'll end up being circular (CL' on top of CL')... 14:05:20 (I suppose you could move *all* the logic to the expander, if you're working with fexprs) 14:05:33 pkhuong_, that's exactly what I wanted to answer :) 14:06:22 angavrilov: some of it must be cooked in at build-time. For the user-facing stuff, like package and function definitions, though, it'll definitely be a contrib. 14:06:33 the thing that interest me in fexprs is that how difficult to compile programs written with them 14:06:39 rrice [~rrice@oh-67-76-18-254.sta.embarqhsd.net] has joined #lisp 14:06:39 -!- kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has quit [Ping timeout: 265 seconds] 14:06:45 and I think that conventional compilers cannot be used 14:07:01 -!- HG` [~HG@xdsl-92-252-89-161.dip.osnanet.de] has quit [Quit: HG`] 14:07:06 pkhuong_: did you do anything on the topic after your sse-pack-sbcl10 branch on repo? 14:07:06 what are fexprs? 14:07:19 sorry to butt in.. 14:07:22 kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has joined #lisp 14:07:22 dose, http://en.wikipedia.org/wiki/Fexpr 14:07:42 im in the process of writing a lisp to javascript compiler.. 14:07:49 the interpreter is working ... 14:08:02 i've also written eval and a compiler in itself.. 14:08:15 but i managed to hit the stack limit of javascript and can't get it to compile itself.. doh! 14:08:36 ah.. ok... then my macros support fexprs... 14:08:38 levente_meszaros: could you define what you mean by compiler? 14:08:39 angavrilov: nope. 14:08:49 from where to what? 14:09:00 the lisp interpreter is written in javascript.. 14:09:02 hlavaty, in my particular case or in general? 14:09:13 then i wrote a lisp interpreter in lisp running on the javascript implementation.. 14:09:13 pkhuong_: I was afraid I was tinkering with something very obsolete :) 14:09:22 and a compile in lisp running on the javascript.. 14:09:26 levente_meszaros: in your case, from fexp to common lisp functions? 14:09:40 partial evaluation sounds interesting... is that like what pypy does? 14:10:02 hlavaty, I try 14:10:08 pkhuong_: This tweak improves stability of dispatch on primitive-type inside VOPs, but I don't know if it is safe, or what was the reason for losing the type in load TNs: http://github.com/angavrilov/sbcl/commit/99801dd1cba0be1170ad882f903d37fa48c37615 14:10:13 cvandusen [~user@68-90-30-246.ded.swbell.net] has joined #lisp 14:11:22 in eval there's a case analysis for self evaluating atoms, variables that need to be looked up in the environment, and combinations 14:11:33 ziarkaen_ [~ziarkaen@87.112.87.249.plusnet.ptn-ag2.dyn.plus.net] has joined #lisp 14:12:17 combinations evaluate the first element and call the results as a function without evaluating the operands 14:12:27 -!- redline6561 [~redline@c-66-56-55-169.hsd1.ga.comcast.net] has quit [Ping timeout: 265 seconds] 14:12:35 the functions that are called in those cases are called fexprs 14:12:47 ok 14:12:54 they are free to return any valid value, and it will be returned by eval too 14:13:07 or they are free to call eval any number of times with any form they want to 14:13:11 like what scheme does, except without argument evaluation. 14:13:45 so I wrote a couple of fexprs in CL (since the language is not bootstrapped) 14:14:06 namely: quote, if, progn, lambda 14:14:12 pkhuong_: Do you want to have polymorphic intrinsics like sse-xor which generate an appropriate instruction depending on args, or is a xor-ps xor-pd xor-pi triplet (like I did for ecl) fine for you? 14:14:31 levente_meszaros: that's the interpreter part, right? 14:14:37 I also wrote an fexpr maker for primitive CL functions that will first evaluate all operands and call the CL function with the results 14:14:42 hlavaty, sure 14:14:47 I'm getting at the compiler 14:14:50 pkhuong_: I may have misunderstood your intention 2 days ago 14:14:52 -!- ziarkaen__ [~ziarkaen@87.113.250.130] has quit [Ping timeout: 265 seconds] 14:15:14 there is also a macro fexpr maker that takes an expander function and calls eval on its result 14:15:29 Can I use a declaration to tell the compiler that a parameter PARAM of a function contains a list with single-float elements? Right now I do (assert (eq 'single-float (type-of (first param)))) at the beginning of the function. 14:15:30 OEP [~OEP@c-76-27-129-70.hsd1.al.comcast.net] has joined #lisp 14:16:00 sure it's not hygienic as pkhuong_ pointed out, but the environment is there and the expansion can be factored in a way so that it is 14:16:09 so much for the interpreter 14:16:35 atoms are NIL, T, integers and strings 14:16:57 since we have lambda I could easily write a factorial function 14:17:06 isn't nil a list? 14:17:17 ugly, isn't it? (lambda (n) ((lambda (g) (g g n)) (lambda (f n) (if (= n 0) 1 (* n (f f (- n 1))))))) 14:17:18 an empty one.. 14:17:39 dose, I'm talking about the guest language, not CL 14:17:42 dose: At least in CL, atoms and lists are not mutually exclusive. 14:17:44 ah 14:17:45 ok 14:17:47 there it is a boolean value 14:17:59 angavrilov: I want to be able to write a polymorphic xor as simply as I can in asm. 14:18:00 I could just as well chose +false+ and +true+ 14:18:11 and inject to-lisp, to-lisp* everywhere to convert 14:18:26 yep 14:18:30 pkhuong_: in asm you have to choose which one of the float double or int to use 14:18:40 but they work the same. 14:18:57 the compiler takes a form and calls the CL partial evaluator on `(eval* ,form) 14:19:32 pkhuong_: pretending that the types really exist, it can be rephrased as if the intstructions implicitly and silently cast their args to their correct return type 14:19:36 the results is obviously CL code that does the same thing as evaluating (eval* form) 14:19:46 except that there are no calls to eval* in it 14:19:49 angavrilov: right. 14:20:13 pkhuong_: I've implemented such a cast for ecl by inserting calls to _mm_cast* intrinsics where necessary 14:20:13 at least, that's the idea, that works somewhat 14:20:53 pkhuong_: but it still has to track these types in order to know what should it cast to what 14:21:20 hlavaty, does that answer you question about the compiler? 14:21:22 that's the difference between msvc's weird support and asm. 14:22:25 levente_meszaros: i got lost here "there is also a macro fexpr maker that takes an expander function and calls eval on its result" 14:22:28 pkhuong_: to my knowledge at the moment there is no portable way to write a cast that converts any unknown input sse type to a specific one 14:22:38 pkhuong_: I think Intel began it with their compiler 14:22:48 angavrilov: unions? pointer cast? 14:23:20 loxs [~loxs@213.169.45.106] has joined #lisp 14:23:30 levente_meszaros: what is an expander function and what is a macro fexpr maker? 14:24:00 (I just emit the correct instruction; no casting needed. The other option I explored was having typed intrinsics, but with dummy cast-to-{float,int} functions... and then I realized that having to cast before using, e.g., streaming stores would piss me off) 14:24:13 pkhuong_: the part of ecl's compiler that can insert implicit casts can only produce what amounts to a function call 14:24:17 hlavaty, a macro is an fexpr that takes its arguments unevaluated (just as all the other fexprs) and returns a form 14:24:17 -!- legumbre_ is now known as legumbre 14:24:25 that is supposed to be evaluated in place 14:24:33 pkhuong_: the argument is very likely to be something you cannot take a pointer of 14:24:40 I was vague 14:24:43 angavrilov: that doesn't exist. 14:25:26 pkhuong_: what about if intrinsics were typed as something like (function (sse-pack sse-pack) (sse-pack single-float))? 14:25:39 angavrilov: that's what I currently do. 14:25:48 -!- Aferlak12 [~Aferlak12@64.120.233.114] has quit [Read error: Connection reset by peer] 14:25:49 so taking that macro expander fexpr (e.g. WHEN) and wrapping it into another that calls eval on the results you will have something similar what we call macros 14:25:49 levente_meszaros: why is the return value being evaluated? 14:26:08 pkhuong_: except that you refuse don't want to distinguish float and double 14:26:16 in the result type 14:26:49 hlavaty, that's the point of macros, they transform forms into another form that will be evaluated in the original environment 14:27:03 angavrilov: what's that? 14:27:25 have (sse-pack single-float) and (sse-pack double-float) be really distinct primitive types and so on 14:27:25 -!- ziarkaen_ [~ziarkaen@87.112.87.249.plusnet.ptn-ag2.dyn.plus.net] has quit [Ping timeout: 265 seconds] 14:28:15 so that when boxing the heap value can be tagged, and print can use the right representation for FP data 14:28:34 so when you have fexprs on the board, macros become easy 14:28:53 I only see the use of sse-pack subtypes to hint move functions, and maybe a couple polymorphic VOPs. I'm not ready to add overhead for that; it's just gravy. 14:28:58 in one word, best common lisp graph library 14:29:19 pkhuong_: I think that most data that is known FP is really FP and not raw bits - and it would be easier to debug if it was printed as such 14:29:41 pkhuong_: Would real production code contain any heap boxing at all? 14:29:58 pkhuong_: (I don't know how SBCLs compiler works) 14:30:04 hlavaty, the interesting thing about the partial evaluation based compilation is how the CL primitives remain there in the resulting form 14:30:19 anybody? lisp graphics? 14:30:31 minion: vecto? 14:30:32 vecto: Vecto is a graphics library that uses cl-vectors and ZPNG to draw vector graphics to PNG files. http://www.cliki.net/vecto 14:30:33 sonnym [~evissecer@singlebrookvpn.lightlink.com] has joined #lisp 14:30:35 levente_meszaros: why can't your fexpr return the value directly, why does it need to be evaluated out of that fexpr? 14:30:37 sabalaba: you'll have to decide, between graphs and graphics. 14:30:59 pkhuong_, graphicz 14:31:07 pkhuong_, graphs---as in mathematical 14:31:18 effectively doing the same thing as the original (eval* form) call while the partial evaluation eliminated all the eval* and friends 14:31:34 pkhuong_, i want to draw functions as well as create generative art 14:31:53 hlavaty, it does not need 14:32:15 it is done that way so that I can call those wrapped functions macros 14:32:20 nothing interesting there 14:32:36 pkhuong_: also, even if SF and DF move ops are equivalent on the current CPUs, I personally feel it is tidier to properly track which is supposed to be used where 14:33:02 levente_meszaros: those wrapped functions macros you mean cl:defmacro or your fexprs? 14:33:13 angavrilov: it might be tidier, using that information is a pessimisation. 14:34:15 pkhuong: you can decide to use the same instruction as a concious decision at the point of generation; not having information is a different matter 14:34:35 tracking the data more finely is a tiny change. 14:35:18 pkhuong_: do you actually expect production code to contain enough of heap boxing, so that one additional mov during it would matter? won't heap allocation screw performance enough already? 14:35:29 -!- cvandusen [~user@68-90-30-246.ded.swbell.net] has left #lisp 14:35:42 hlavaty, http://paste.lisp.org/display/113901 14:35:52 angavrilov: no, but I definitely expect to perform the same whether things get boxed or not. 14:35:57 unfortunately partial evaluation does not support closures properly 14:36:01 Perform as in perform the same operations. 14:36:10 so the interpreter makes closures by hand :( 14:36:39 hlavaty, see make-macro and evaluate-macro (not so good names) and how they are used 14:36:50 levente_meszaros: why do you want a compiler for fexprs in the first place? wouldn't it be better to write an efficient interpreter, with time critical parts implemented directly in common lisp? basically, fexprs cannot make any assumptions about the args, so there's really very little useful stuff a compiler can do i think 14:37:27 pkhuong_: I want repl to print the result of (add-ps foo bar) as # and not some hexadecimal 14:37:30 Currently, I'm not sure I can guarantee that happens. A boxed value of type sse-pack will still retain its integer/floatiness, even under assignment; when unboxed, that disappears. 14:37:43 angavrilov: you can have both. 14:37:55 ? 14:38:19 Just change the printing function. You could even parameterize it with a special variable. 14:39:43 what about the debugger? 14:40:11 will SBCL unbox stuff if it doesn't want to do any operations with it? 14:40:44 not likely. 14:41:02 hlavaty, there's a misunderstanding there 14:41:22 the compiler does not compile fexprs, but the programs that are using them 14:41:36 the fexprs are already compiled because they are written in CL 14:41:45 the language is not bootstrapped 14:41:52 Amadiro [~whoppix@1x-193-157-193-242.uio.no] has joined #lisp 14:41:54 stassats`: Found a bug in gdb-command in swank-cmucl.lisp for cmucl/darwin. The prompt is not correct. It should be "~%^done,thread-id=\"1\"~%(gdb) ~%". Not sure about the thread-id, but it's probably ok since cmucl doesn't do threads. 14:42:22 -!- Joreji [~thomas@77-066.eduroam.RWTH-Aachen.DE] has quit [Ping timeout: 240 seconds] 14:43:09 pkhuong_: btw, how do you think EQL EQUAL EQUALP should compare sse packs? 14:43:46 EQL, at some point, right after I make it extensible. 14:44:35 when the CL partial evaluator works on (eval* form) it inlines the fexprs' code 14:44:50 -!- stassats` [~stassats@wikipedia/stassats] has quit [Ping timeout: 265 seconds] 14:44:51 levente_meszaros: do you have examples for that? i guess it's the partial-eval-evaluate function 14:44:53 and also eval* when it is called back 14:45:00 b-man_ [~b-man@189.34.54.49] has joined #lisp 14:45:19 hlavaty, that's the part how the partial evaluator is parameterized 14:45:46 pkhuong_: why not say EQUAL? 14:45:49 I still do have open question (how surprising?!) that's why I was looking for papers 14:45:55 angavrilov: that'll fall implicitly from EQL. 14:46:18 I mean sse aren't quite numbers, and EQL is supposed to handle numbers 14:46:42 hlavaty, I annotated that page 14:46:48 -!- tfb [~tfb@94.197.70.214.threembb.co.uk] has quit [Ping timeout: 240 seconds] 14:46:48 with an example 14:47:39 gonzojive [~red@2002:800c:a95a:4:fa1e:dfff:fee3:34e8] has joined #lisp 14:47:42 number and characters are immutable, and as such, completely equivalent values are allowed not to be EQ. The situation is similar for SSE packs (or SAPs). 14:47:49 looking at your evaluate-if, that's how picolisp does it 14:47:54 levente_meszaros: looking at your evaluate-if, that's how picolisp does it 14:48:34 but the evaluator works probably a bit different 14:48:39 -!- mbohun [~mbohun@ppp115-156.static.internode.on.net] has quit [Quit: Leaving] 14:48:40 hlavaty, and can you imagine how the CL:IF escapes from there into the result of COMPILEX through the partial evaluation? 14:49:30 Joreji [~thomas@77-066.eduroam.RWTH-Aachen.DE] has joined #lisp 14:50:29 levente_meszaros: The function PARTIAL-EVAL is undefined. 14:51:10 is that what you are after? 14:51:10 Modius [~Modius@cpe-24-28-30-165.austin.res.rr.com] has joined #lisp 14:51:20 that's in hu.dwim.partial-eval 14:51:24 but I must leave now 14:51:43 -!- levente_meszaros [~levente_m@4d6f5d3b.adsl.enternet.hu] has quit [Quit: ...] 14:53:02 angavrilov: right. That's what was bugging me about tracking type information at runtime: it won't be consistent, and I'd prefer not to insert new ways for code to randomly act differently. 14:53:23 TomJ [~tomj@78.151.111.133] has joined #lisp 14:53:29 those hu.dwim guys have some interesting stuff... I just wish it wasn't written in hungarian lisp. 14:54:07 Fade: ;-) 14:54:18 sdsds [~sdsds@dyn-16.sub2.cmts01.cable.TORON10.iasl.com] has joined #lisp 14:54:21 pkhuong_: did you actually define any intrinsics at all? I don't see any in that repo branch 14:55:40 pkhuong_: also, how exactly are you going to address reads and stores to memory? 14:56:22 fsmunoz [~fsmunoz@a85-138-208-156.cpe.netcabo.pt] has joined #lisp 14:57:39 -!- mstevens [~mstevens@fsf/member/pdpc.active.mstevens] has quit [Quit: leaving] 15:00:14 -!- loxs [~loxs@213.169.45.106] has quit [Remote host closed the connection] 15:01:05 angavrilov: SAPs. And the intrinsic definitions are just pedestrian code that'll go in a contrib. 15:01:17 sap=? 15:01:57 C-style pointers. 15:02:11 -!- zomgbie [~jesus@h081217133138.dyn.cm.kabsi.at] has quit [Ping timeout: 276 seconds] 15:02:24 how does that look when used? 15:03:14 (sap-ref-sse-pack sap offset). 15:03:16 and how difficult would it be to implement syntactic sugar like (setf (row-major-aref-ps foo idx) ...) in sbcl? 15:03:19 zomgbie [~jesus@h081217133138.dyn.cm.kabsi.at] has joined #lisp 15:03:30 -!- Madsy [~madman@fu/coder/madsy] has quit [Ping timeout: 240 seconds] 15:03:33 easy, with SAPs. 15:04:00 how do you get that sap, and does it require any kind of strange stuff like locking down gc or something 15:04:50 just pinning. It's how we pass, e.g., array data between lisp and C. 15:05:59 what about a vop that directly references a vector without all the pinning mess 15:06:34 -!- kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has quit [Ping timeout: 265 seconds] 15:06:47 sure, later. All that stuff is really easy and can happen extremely late once the VM-level functionality has been cooked in. 15:07:05 Mesh [~Mesh@216.201.34.14] has joined #lisp 15:07:33 -!- Amadiro [~whoppix@1x-193-157-193-242.uio.no] has quit [Quit: A subtle thought that is in error may yet give rise to fruitful inquiry that can establish truths of great value.] 15:07:47 jewel [~jewel@196-215-16-132.dynamic.isadsl.co.za] has joined #lisp 15:08:02 Basically, nobody can tell the right way to do it in CL. I'll provide the building blocks and a minimal interface, and build on top of that with time. 15:09:34 drewc: so... now how do I got about getting a virtual box? 15:10:10 I don't see where that "nobody can tell" can come from. Because all you have are the instructions, and their set is fixed. 15:10:37 Aside the aref stuff, all it amounts to is a translation between a lisp name and the underlying single instruction 15:11:01 So the only thing one could argue about is how to name the function, and the type handling 15:12:03 Are there any compilers at all, which contain a full semantic model of SSE, allowing for it to be used like regular stuff? 15:12:03 angavrilov: sure. It's a couple lines of pure CL code, given SAPs. I don't feel a pressing need to decide what should and shouldn't be provided. 15:12:46 deepfire: what do you mean by regular stuff 15:13:22 Well, the level at which plain integer/float values are integrated. 15:13:43 deepfire: that's what we're working on. 15:13:50 pkhuong_: what about providing everything in the C intrinsic API, supported by all of ICC GCC MSVC 15:13:57 -!- sebyte [~sebyte@213.229.74.8] has quit [Remote host closed the connection] 15:14:33 AIUI the SSE-like shenanigans mostly happen in compilers in some kind of hardwired "autovectorisation" schemes. 15:14:42 basically, I'll define the VOP intrinsics themselves, and a minimal set of VOPs to move data between packs and other lisp values. 15:14:58 deepfire: http://msdn.microsoft.com/en-us/library/y0dh78ez%28VS.71%29.aspx 15:15:06 deepfire: I'd say it's the reverse; except for a couple of "easy" patterns, you have to use intrinsics explicitly. 15:15:22 kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has joined #lisp 15:15:23 stis [~stis@1-1-1-39a.veo.vs.bostream.se] has joined #lisp 15:15:54 -!- xyxxyyy1 [~xyxu@58.41.11.200] has quit [Ping timeout: 240 seconds] 15:15:55 angavrilov: if the C API is decently portable, then I will. Unlike all the VM-level stuff, it only needs regular CL definitions. 15:16:38 I have defined wrapping for almost all of it for ECL: http://github.com/angavrilov/ecl-sse/blob/master/contrib/sse/sse-core.lsp 15:16:53 Interesting, thanks! 15:17:47 What I'm really looking after is some kind of agreement on an API, so that I can update my ECL code to it, and then get on with using it while being reasonably confident that when SBCL version is implemented it won't deliberately deviate from it. 15:17:52 Great stuff you're working on, folks! 15:18:23 angavrilov: again, any difference will be really easy to fix on my end, anyway. 15:18:43 jan247 [~jan247@unaffiliated/jan247] has joined #lisp 15:19:10 the C api seems to be copied from the original intel documentation, except a few functions that are named differently on msvc (gcc has both versions for compat) 15:19:13 I'm focusing on the building blocks; after that, the only thing missing will be tons of short boilerplate definitions. 15:19:15 search for #+msvc in that file 15:19:39 Raynes [~macr0@unaffiliated/raynes] has joined #lisp 15:19:51 plus the cast intrinsics appeared later than the bulk, so older compilers don't have them and you have to define them yourself 15:20:22 the type stuff is sort of important; how the functions will be named, not so much. As for data access, well, we can both do pretty much anything that's possible in C (or in asm, for SBCL) 15:21:05 felideon [~user@12.228.15.162] has joined #lisp 15:22:29 I'm going to stick to the AREF-PS style syntactic sugar, since on ECL all pointers are boxed for some reason, except in explicit inline C code, while the sugar can be made to expand to just array->dataptr[index] 15:22:47 under (safety 0) (speed 3) 15:24:05 angavrilov: sure, and I can just define aref-ps in straight CL. 15:24:14 I'll try changing my ftype definitions to accept any sse-pack and think more whether I absolutely need typed boxed constants for ECL 15:24:32 I define it in CL to, as a macro :) What's magic is row-major-aref-* 15:25:53 a macro? Why? 15:27:42 because in ECL inlining sucks (hopefully only now), and defining only a macro is shorter than a function + a compiler macro 15:28:07 of course the right thing is probably function + macro anyway 15:28:16 and yet function + compiler maro is the right way. 15:28:22 -!- jmbr [~jmbr@28.33.220.87.dynamic.jazztel.es] has quit [Remote host closed the connection] 15:28:45 especially with variadic functions, being able to use apply is a big boon. 15:28:45 I was just being lazy :) 15:29:00 -!- dnm [~dnm@c-68-34-57-154.hsd1.va.comcast.net] has quit [Ping timeout: 272 seconds] 15:29:52 although since sse is supposed to be used for performance, apply would be somewhat beside the point 15:30:43 (and at that performance at the level of using the very best CPU instructions) 15:30:46 not with compiler macros to get the common case right. 15:34:13 dnm [~dnm@c-68-34-57-154.hsd1.va.comcast.net] has joined #lisp 15:34:41 -!- bozhidar [~user@212.50.14.187] has quit [Quit: Asta la vista] 15:34:42 not really arguing, just rationalizing my laziness... :) 15:37:17 kwabbles_ [~mike@h-67-101-178-243.lsanca54.static.covad.net] has joined #lisp 15:37:22 ok, re typed sse packets: the problem is with assignments and unboxed values. If I want to track pack types consistently, I need to do so at the ptype level. However, ptypes are static, while boxed packs are typed dynamically. Most of the time, the type is trivial to deduce... except with assignments to variables that accept more than one pack type. 15:38:11 serichsen [~user@f049005165.adsl.alicedsl.de] has joined #lisp 15:38:14 Good evening! 15:40:33 -!- nus [~nus@unaffiliated/nus] has quit [Ping timeout: 260 seconds] 15:41:30 -!- Jasko [~tjasko@209.74.44.225] has quit [Quit: Leaving] 15:43:55 -!- jan247 [~jan247@unaffiliated/jan247] has quit [Quit: jan247] 15:47:27 -!- Bronsa [~bronsa@host152-175-dynamic.7-79-r.retail.telecomitalia.it] has quit [Quit: Lost terminal] 15:50:00 Bronsa [~bronsa@host152-175-dynamic.7-79-r.retail.telecomitalia.it] has joined #lisp 15:50:37 I am playing around with bordeaux-threads a bit, and I have some questions. I have SBCL 1.0.40 and bordeaux-threads 0.7.0. 15:51:18 -!- Joreji [~thomas@77-066.eduroam.RWTH-Aachen.DE] has quit [Ping timeout: 240 seconds] 15:52:24 CL-USER> (bordeaux-threads:make-thread (lambda () (print "Hi!"))) 15:52:36 -!- Jabberwockey [~Jens@port-12485.pppoe.wtnet.de] has quit [Read error: Connection reset by peer] 15:52:49 Jabberwockey [~Jens@port-12485.pppoe.wtnet.de] has joined #lisp 15:52:53 X-02 [~x_02@p2208-ipbf203kyoto.kyoto.ocn.ne.jp] has joined #lisp 15:52:57 This does not print anything. Do the threads have a different standard output? 15:53:48 serichsen: try adding a (force-output) after that print statement 15:54:00 or (terpri) 15:54:30 dlowe: I tried force-output, with no change. 15:55:03 <_3b> check in *inferior-lisp* buffer 15:55:15 ah, yeah. good call. 15:55:35 Heh. OK, there it is. :) 15:56:00 I wonder why it does not turn up in the REPL buffer, but OK. 15:56:18 serichsen: IO is redirected via dynamic binding. 15:57:13 -!- ikki [~ikki@200.95.162.199] has quit [Ping timeout: 255 seconds] 15:57:35 Ah, I see where that goes, I think. 15:57:48 Madsy [~madman@ti0207a340-0265.bb.online.no] has joined #lisp 15:57:48 -!- Madsy [~madman@ti0207a340-0265.bb.online.no] has quit [Changing host] 15:57:48 Madsy [~madman@fu/coder/madsy] has joined #lisp 15:57:52 asdf2 can scan a dir recursively for asd files right? 15:58:28 PuffTheMagic: so can asdf, but asdf2 has that by default. 15:58:33 redline6561 [~redline@c-66-56-55-169.hsd1.ga.comcast.net] has joined #lisp 15:58:34 ohh 15:58:42 is the syntax the same 15:58:50 or do i need #+asdf2 15:58:59 Ok, next question: why should user code not use thread-join? In fact, it seems that my bordeaux-threads implementation does not even have it. Is manually setting up a condition-variable somehow cleaner? 15:59:17 serichsen: you can save the current streams and rebind them in the new thread. e.g. (let ((out *standard-output*)) (make-thread (lambda () (let ((*standard-output* out)) ...))) 16:00:04 peterhil [peterhil@a91-153-120-132.elisa-laajakaista.fi] has joined #lisp 16:00:30 pkhuong_: I guess that I could also use the optional initial-bindings for that. 16:01:10 srolls [~user@167.216.131.126] has joined #lisp 16:01:15 serichsen: re thread-join, the issue is probably with thin wrappers for pthread, in which it's unspecified what happens if a thread is joined multiple times. 16:04:59 adding support for join-thread is invasive 16:08:38 meaning that make-thread has to cooperate 16:09:03 if the lisp you are using has no support for join-thread built in 16:09:07 -!- kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has quit [Quit: Bye bye ppl] 16:09:23 mega1: I see what you mean. 16:09:28 ok, then I stop. 16:10:56 Phoodus [foo@174-22-199-91.phnx.qwest.net] has joined #lisp 16:11:04 -!- benny [~user@i577A1B86.versanet.de] has quit [Remote host closed the connection] 16:14:53 pdelgallego [~pdelgalle@1503031474.dhcp.dbnet.dk] has joined #lisp 16:15:17 ikki [~ikki@201.155.75.146] has joined #lisp 16:16:35 -!- Jabberwockey [~Jens@port-12485.pppoe.wtnet.de] has quit [Read error: Connection reset by peer] 16:16:49 Jabberwockey [~Jens@port-12485.pppoe.wtnet.de] has joined #lisp 16:16:49 hello, is it possible to save hash tables in files using sbcl? 16:17:16 yes, but I don't know how 16:17:18 X-02: you could perhaps use something like cl-store. 16:17:54 OliverUv: no, actually. I know its okay with clisp, though 16:18:12 serichsen: thank you, I'll google it 16:18:15 X-02: You can program the printer (using print-object) to save it in some form that you can then read back (by program the reader to recognize it). I do things like that all the time, but not necessarily for hash tables. 16:19:20 milanj [~milanj_@109-92-124-204.dynamic.isp.telekom.rs] has joined #lisp 16:21:33 thank you all, I found cl-store and print-object. appreciate it. 16:21:51 make-load-form might work out better for you. 16:22:19 -!- jdz [~jdz@193.206.22.97] has quit [Read error: Operation timed out] 16:22:26 -!- pdelgallego [~pdelgalle@1503031474.dhcp.dbnet.dk] has quit [Quit: Leaving] 16:22:47 pdelgallego [~pdelgalle@1503031474.dhcp.dbnet.dk] has joined #lisp 16:26:57 pkhuong_: I think I'll have to think this over. If all fails, maybe typed packs may be left as a ecl-specific implementation detail. 16:27:05 loxs [~loxs@85-130-37-174.2073285806.ddns.cablebg.net] has joined #lisp 16:27:42 angavrilov: how will you implement a variable that can take any type of pack? 16:28:16 currently in ECL that means always boxed 16:28:34 recommend using cl-serializer instead of cl-store 16:28:38 I'd rather not have that (: 16:28:40 i.e. in sbcl terms, a generic sse-pack has primitve type boxed object 16:29:17 with casts the need for precisely typed constants is definitely reduced a lot 16:29:54 carlocci [~nes@93.37.214.169] has joined #lisp 16:30:00 slash_1 [~unknown@p4FF0A7A0.dip.t-dialin.net] has joined #lisp 16:30:01 -!- slash_ [~unknown@p5DD1D8DD.dip.t-dialin.net] has quit [Ping timeout: 240 seconds] 16:32:24 I think I'll have to sleep on this and come back tomorrow :) 16:33:05 -!- loxs [~loxs@85-130-37-174.2073285806.ddns.cablebg.net] has quit [Quit: Leaving] 16:34:43 -!- redline6561 [~redline@c-66-56-55-169.hsd1.ga.comcast.net] has quit [Ping timeout: 240 seconds] 16:36:10 -!- xan_ [~xan@220.118.197.183] has quit [Quit: leaving] 16:36:35 kpreid_ [~kpreid@128.153.212.232] has joined #lisp 16:39:49 -!- cmm [~cmm@bzq-79-181-203-193.red.bezeqint.net] has quit [Ping timeout: 258 seconds] 16:40:41 cmm [~cmm@bzq-79-181-203-193.red.bezeqint.net] has joined #lisp 16:40:49 e66 [~chatzilla@180.149.12.65] has joined #lisp 16:41:44 owned [~owned@93-96-43-166.zone4.bethere.co.uk] has joined #lisp 16:42:42 hello, we are translating sbcl codes to c by chestnut. does anyone know do I improve the chest nut runtime library? 16:43:04 magius_pendragon [~magius@cpe-071-070-210-101.nc.res.rr.com] has joined #lisp 16:43:22 I mean is there any way to tune it ? 16:44:15 -!- magius_pendragon [~magius@cpe-071-070-210-101.nc.res.rr.com] has left #lisp 16:44:34 what's chestnut? 16:44:50 sbcl to c translator. 16:45:03 pkhuong_: http://groups.google.com/group/comp.lang.lisp/msg/88c50739f36e0b7e 16:45:12 -!- owned [~owned@93-96-43-166.zone4.bethere.co.uk] has quit [Read error: Connection reset by peer] 16:45:40 e66: chestnut is not a public project, so i don't know if anyone here has used it or knows much about it. 16:46:20 Xach: Oh! 16:47:10 and I'm fairly certain its source language isn't sbcl. It's more probably common lisp, or a restricted dialect of CL. 16:48:23 -!- Krystof [~csr21@howells.doc.gold.ac.uk] has quit [Ping timeout: 240 seconds] 16:48:52 SBCL = Steel Bank Common Lisp 16:49:04 pkhuong_: I take it you meant something other than sbcl there? 16:49:06 gigamonkey [~user@c-98-248-194-46.hsd1.ca.comcast.net] has joined #lisp 16:50:07 e66: if chestnut's RTL is a dynamic library, you can try to shadow some of its symbols. ThinLisp might work better for you, as it's similar in spirit, and is distributed under a BSD license. 16:50:29 -!- OEP [~OEP@c-76-27-129-70.hsd1.al.comcast.net] has quit [Ping timeout: 265 seconds] 16:51:35 -!- arbscht [~arbscht@unaffiliated/arbscht] has quit [Ping timeout: 276 seconds] 16:52:44 doesn't gcl also work as a CL-to-C translator? 16:53:10 (though I'd not characterize its output as clean C) 16:53:28 dlowe: right, the KCL family doesn't have the same design goals. 16:53:33 pkhuong_: chestnuts RTL is written in C. And normal .lisp files are translated to C file. 16:53:43 vu3rdd [~vu3rdd@122.167.69.10] has joined #lisp 16:54:03 dlowe: a translator can not translate to clean C code. 16:54:34 tcr1 [~tcr@81-233-246-97-no37.tbcn.telia.com] has joined #lisp 16:54:36 e66: there are levels of clean code 16:54:52 I see all the types are void * when translated by chestnut 16:57:15 SegFaultAX [~SegFaultA@c-98-234-1-162.hsd1.ca.comcast.net] has joined #lisp 16:57:31 arbscht [~arbscht@unaffiliated/arbscht] has joined #lisp 16:58:24 nus [~nus@unaffiliated/nus] has joined #lisp 17:00:44 stassats` [~stassats@wikipedia/stassats] has joined #lisp 17:00:52 -!- sabalaba [~sabalaba@c-76-118-76-200.hsd1.vt.comcast.net] has quit [Quit: Leaving] 17:02:32 -!- Jabberwockey [~Jens@port-12485.pppoe.wtnet.de] has quit [Read error: Connection reset by peer] 17:02:46 Jabberwockey [~Jens@port-12485.pppoe.wtnet.de] has joined #lisp 17:07:43 -!- slash_1 [~unknown@p4FF0A7A0.dip.t-dialin.net] has quit [Remote host closed the connection] 17:08:28 slash_ [~unknown@p4FF0A7A0.dip.t-dialin.net] has joined #lisp 17:14:08 -!- nha [~prefect@imamac13.epfl.ch] has quit [Ping timeout: 276 seconds] 17:16:22 slyrus: around? i just emailed you 17:16:44 thanks drewc 17:17:51 hiya, drewc 17:18:36 benny [~user@i577A3474.versanet.de] has joined #lisp 17:19:23 hey nus 17:19:38 drewc, I've been asking around for some real stats for websites running on CL (in terms of VM/CPU pressure in a VPS under load). Do you happen to have something at hands? 17:21:54 -!- ignas [~ignas@78-60-73-85.static.zebra.lt] has quit [Ping timeout: 240 seconds] 17:22:34 nus: well, it really depends on the site... here's some cliki.net info : 17:23:08 top - 13:23:26 up 554 days, 19:00, 3 users, load average: 0.01, 0.02, 0.00 17:23:08 17:23:16 Mem: 127112k total, 124072k used, 3040k free, 8740k buffers 17:23:16 17:23:33 Swap: 262136k total, 9692k used, 252444k free, 29320k cached 17:23:33 17:25:55 so the VM has growed up to the hard limit and sits there... 17:26:00 -!- incandenza [~djm@ip68-231-109-244.ph.ph.cox.net] has quit [Quit: leaving] 17:26:16 loxs [~loxs@85-130-37-174.2073285806.ddns.cablebg.net] has joined #lisp 17:26:35 salva [~salva@105.11.117.91.dynamic.mundo-r.com] has joined #lisp 17:26:44 incandenza [~djm@ip68-231-109-244.ph.ph.cox.net] has joined #lisp 17:27:15 the VM doesn't grow or shrink, it's Xen. 17:28:15 eh? the commited pages count surely did change? 17:29:47 yes, it does actually use memory. 17:30:01 davazp [~user@84.122.72.152.dyn.user.ono.com] has joined #lisp 17:31:14 1206 cliki 16 0 865m 56m 9948 S 3.0 45.3 4933:46 sbcl 17:31:48 -!- slash_ [~unknown@p4FF0A7A0.dip.t-dialin.net] has quit [Ping timeout: 252 seconds] 17:31:53 ^ PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 17:32:37 -!- e66 [~chatzilla@180.149.12.65] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.8/20100722155716]] 17:33:06 jdz [~jdz@host10-106-dynamic.14-87-r.retail.telecomitalia.it] has joined #lisp 17:33:46 -!- milanj [~milanj_@109-92-124-204.dynamic.isp.telekom.rs] has quit [Ping timeout: 272 seconds] 17:34:48 -!- b-man_ [~b-man@189.34.54.49] has quit [Ping timeout: 240 seconds] 17:36:05 HG` [~HG@85.8.72.229] has joined #lisp 17:38:15 Deesl [~bsdboy@unaffiliated/deesl] has joined #lisp 17:38:16 milanj [~milanj_@109-92-113-128.dynamic.isp.telekom.rs] has joined #lisp 17:38:30 pavelludiq [~user@83.222.175.142] has joined #lisp 17:39:16 -!- hohoho [~hohoho@ntkngw229253.kngw.nt.ftth.ppp.infoweb.ne.jp] has quit [Remote host closed the connection] 17:40:55 that'd be araneida on sbcl running for ~ 3 days using 56 megs of physmem out of 128 available, right? 17:41:16 what's pagehit count for the 3 days? 17:42:30 dreish [~dreish@minus.dreish.org] has joined #lisp 17:44:14 -!- dreish [~dreish@minus.dreish.org] has quit [Read error: Connection reset by peer] 17:44:41 on an unrelated note, what's up with Bill Clementson? (I guess we all miss his lisp blogging) 17:50:11 -!- fsmunoz [~fsmunoz@a85-138-208-156.cpe.netcabo.pt] has quit [Read error: Operation timed out] 17:53:04 slash_ [~unknown@p5DD1C8E3.dip.t-dialin.net] has joined #lisp 17:53:04 -!- vandemar [nonserviam@2001:470:1f10:56b::4] has quit [Read error: Connection reset by peer] 17:56:29 -!- slash_ [~unknown@p5DD1C8E3.dip.t-dialin.net] has quit [Client Quit] 17:59:53 -!- HG` [~HG@85.8.72.229] has quit [Quit: Leaving.] 18:03:02 bozhidar [~user@93-152-185-88.ddns.onlinedirect.bg] has joined #lisp 18:03:49 Amadiro [~whoppix@ti0021a380-dhcp1747.bb.online.no] has joined #lisp 18:07:09 udzinari [~user@209.158.broadband13.iol.cz] has joined #lisp 18:08:11 udzinari` [~user@209.158.broadband13.iol.cz] has joined #lisp 18:12:02 -!- udzinari [~user@209.158.broadband13.iol.cz] has quit [Ping timeout: 276 seconds] 18:12:30 -!- udzinari` [~user@209.158.broadband13.iol.cz] has quit [Ping timeout: 252 seconds] 18:12:58 konr [~user@201.82.129.247] has joined #lisp 18:19:41 -!- vu3rdd [~vu3rdd@122.167.69.10] has quit [Remote host closed the connection] 18:20:01 -!- zomgbie [~jesus@h081217133138.dyn.cm.kabsi.at] has quit [Quit: Lost terminal] 18:26:42 timor [~timor@port-92-195-49-195.dynamic.qsc.de] has joined #lisp 18:27:40 varjag [~eugene@4.169.249.62.customer.cdi.no] has joined #lisp 18:29:42 -!- ace4016 [~jmarcelin@adsl-32-120-9.mia.bellsouth.net] has quit [Disconnected by services] 18:29:43 _ace4016_ [~jmarcelin@adsl-32-120-9.mia.bellsouth.net] has joined #lisp 18:35:33 ace4016 [~jmarcelin@adsl-32-120-9.mia.bellsouth.net] has joined #lisp 18:35:33 redline6561 [~redline@gate-20.spsu.edu] has joined #lisp 18:37:06 -!- _ace4016_ [~jmarcelin@adsl-32-120-9.mia.bellsouth.net] has quit [Ping timeout: 272 seconds] 18:37:29 aidalgol [~user@114-134-6-5.rurallink.co.nz] has joined #lisp 18:38:55 seelenquell [~kvirc@pD9E42CA4.dip.t-dialin.net] has joined #lisp 18:39:42 ziarkaen [~ziarkaen@87.115.156.134] has joined #lisp 18:40:21 -!- metasyntax [~taylor@72.86.89.174] has quit [Read error: Operation timed out] 18:41:40 Does anyone know a good alternative for CL + SLL? 18:41:46 -!- moocow [~new@poco208-2.fredcanhelp.com] has quit [Read error: Operation timed out] 18:43:15 SLL? 18:43:19 minion: SLL? 18:43:19 Sorry, I couldn't find anything in the database for ``SLL''. 18:43:22 moocow [~new@poco208-2.fredcanhelp.com] has joined #lisp 18:43:27 SSL, I believe he means 18:43:40 seelenquell: why is an alternative needed? 18:44:31 dlowe: yes, ssl, it doesnt work with ACL 18:44:33 ehu [~ehuels@ip118-64-212-87.adsl2.static.versatel.nl] has joined #lisp 18:45:13 seelenquell: allegro seems to have its own SSL support. 18:46:13 seelenquell: http://www.franz.com/support/documentation/7.0/doc/socket.htm#ssl-1 18:47:56 -!- Deesl [~bsdboy@unaffiliated/deesl] has quit [Ping timeout: 276 seconds] 18:49:20 sykopomp: I Know, but iI am looking for a system which also runs on other implementations. 18:50:49 -!- davazp [~user@84.122.72.152.dyn.user.ono.com] has quit [Remote host closed the connection] 18:51:38 daniel__ [~daniel@p5B327CA0.dip.t-dialin.net] has joined #lisp 18:54:17 -!- redline6561 [~redline@gate-20.spsu.edu] has quit [Ping timeout: 276 seconds] 18:55:05 -!- daniel [~daniel@p5B3279AD.dip.t-dialin.net] has quit [Ping timeout: 276 seconds] 18:55:12 seelenquell: *features* runs on other implementations. 18:55:32 metasyntax [~taylor@72.86.89.174] has joined #lisp 18:56:11 -!- ehu [~ehuels@ip118-64-212-87.adsl2.static.versatel.nl] has left #lisp 18:56:39 oh, I dosnt know, thanks 18:57:29 seelenquell: you can write a simple protocol for your application, and use #+ and #- to conditionalize it based on whether you're running ABCL, or a CL+SSL-compatible implementation. 19:00:40 sykopomp: Yes, that I have has already seen. 19:00:46 Sikander [~soemraws@5356ECA7.cable.casema.nl] has joined #lisp 19:00:53 good evening 19:03:41 Is there a portable way to run an external program in CL? Say, something similar to the C system() function? 19:04:01 minion: tell Sikander about trivial-shell 19:04:02 Sikander: look at trivial-shell: Trivial shell is a simple platform independent interface to the underlying Operating System. http://www.cliki.net/trivial-shell 19:04:14 thanks 19:04:22 adu [~ajr@pool-71-191-154-37.washdc.fios.verizon.net] has joined #lisp 19:05:39 -!- Bronsa [~bronsa@host152-175-dynamic.7-79-r.retail.telecomitalia.it] has quit [Quit: Lost terminal] 19:06:23 LiamH: Hi, did you get any further? 19:06:33 -!- timor [~timor@port-92-195-49-195.dynamic.qsc.de] has quit [Quit: Leaving.] 19:07:19 sabalaba [~sabalaba@c-76-118-76-200.hsd1.vt.comcast.net] has joined #lisp 19:08:09 -!- Aisling [ash@blk-222-192-36.eastlink.ca] has quit [Quit: leaving] 19:08:12 Aisling [ash@blk-222-192-36.eastlink.ca] has joined #lisp 19:10:15 -!- gonzojive [~red@2002:800c:a95a:4:fa1e:dfff:fee3:34e8] has quit [Ping timeout: 252 seconds] 19:10:57 Krystof [~csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has joined #lisp 19:14:48 sykopomp: ssl is in the express-verison of ACl not available: http://www.franz.com/support/documentation/7.0/doc/installation.htm#install-openssl-1 19:15:14 Beetny [~Beetny@ppp118-208-129-226.lns20.bne1.internode.on.net] has joined #lisp 19:15:24 -!- aidalgol [~user@114-134-6-5.rurallink.co.nz] has quit [Ping timeout: 265 seconds] 19:17:21 seelenquell: You might consider using a different product in that case. 19:20:21 Salamander__ [~Salamande@ppp118-210-41-194.lns20.adl2.internode.on.net] has joined #lisp 19:21:52 Immer das gleiche problem. Soething Does not run. 19:22:27 sorry, wrong channel 19:22:54 -!- Salamander_ [~Salamande@ppp121-45-27-195.lns20.adl2.internode.on.net] has quit [Ping timeout: 252 seconds] 19:24:01 Xach: Always the same problems, Something doesn't work 19:25:08 yep. That's computers for you. 19:25:35 hypno [~hypno@impulse2.gothiaso.com] has joined #lisp 19:26:23 -!- jewel [~jewel@196-215-16-132.dynamic.isadsl.co.za] has quit [Ping timeout: 245 seconds] 19:27:52 Salamander_ [~Salamande@ppp118-210-5-192.lns20.adl2.internode.on.net] has joined #lisp 19:28:54 -!- adu [~ajr@pool-71-191-154-37.washdc.fios.verizon.net] has quit [Quit: adu] 19:30:03 -!- Salamander__ [~Salamande@ppp118-210-41-194.lns20.adl2.internode.on.net] has quit [Ping timeout: 252 seconds] 19:32:50 wbooze [~user@xdsl-78-34-232-86.netcologne.de] has joined #lisp 19:32:57 homie [~user@xdsl-78-34-232-86.netcologne.de] has joined #lisp 19:34:20 zomgbie [~jesus@h081217133138.dyn.cm.kabsi.at] has joined #lisp 19:34:53 Hm. 19:35:01 mutoga [~Shiva@3615mylife.foreverlovingjah.fr] has joined #lisp 19:35:47 -!- mk2 [~user@evanspc2.rai.umds.ac.uk] has quit [Remote host closed the connection] 19:37:46 I want to define a struct which contains a couple of lists. These lists may or may not be used by the various instances of that struct, although at least one will probably always be used (otherwise I don't really have any reason for that instance to exist) 19:38:12 -!- petercoulton [~petercoul@cpc4-midd16-2-0-cust402.11-1.cable.virginmedia.com] has quit [Ping timeout: 265 seconds] 19:38:14 -!- galaxywatcher [~galaxywat@ppp-61-90-20-10.revip.asianet.co.th] has quit [Quit: galaxywatcher] 19:38:23 are you saying that you want weak references? 19:38:51 I guess I'm asking if the empty lists would potentially incur some huge memory cost without any benefit or if they'd just be null values that didn't really matter either way 19:39:37 Kenjin [~josesanto@bl19-227-50.dsl.telepac.pt] has joined #lisp 19:40:38 well, empty list is NIL 19:41:31 -!- mutoga [~Shiva@3615mylife.foreverlovingjah.fr] has left #lisp 19:43:37 that's one machine word right there 19:44:04 *Xach* has only so many mega-octets available for his logiciels 19:44:37 and don't forget that NIL is immediate (on sane implementations) 19:44:54 Xach: don't your octects go into giga-octets territory? 19:46:03 jdz: up to 11 giga-octets 19:46:15 Hmm. Maybe I should send patches to slime-devel instead of code.... 19:46:27 Xach: yeah, 11 is not a big number 19:46:56 stassats: So having a bunch of empty lists floating around isn't going to hurt me? 19:47:36 Mesh: You could profile, and find out. 19:48:41 Mesh: there is only one empty list 19:49:13 ToGgG [~Shiva@3615mylife.foreverlovingjah.fr] has joined #lisp 19:49:25 your only overhead is an additional slot in a structure, which is word-wide 19:50:04 but that can become many zeroes and ones! 19:51:15 stassats: Is that overhead significant? 19:51:32 Mesh: that depends 19:51:32 Mesh: if you have 3 billion structs, yes! 19:51:38 (two-slot structs) 19:51:52 struct instances* 19:52:03 it's 4 bytes on 32-bit and 8-bytes on 64-bit, times the number of instances 19:55:28 Okay, so what if I condensed those three lists into one list and wrote my own accessor functions? 20:00:52 Sikander: I tried to using a copy of the vector in vector/length instead of altering the original, thinking that might solve the problem you identified, but it didn't help. 20:01:54 ignas [~ignas@ctv-79-132-160-221.vinita.lt] has joined #lisp 20:02:45 Edward__ [~ed@AAubervilliers-154-1-30-161.w90-3.abo.wanadoo.fr] has joined #lisp 20:03:20 aidalgol [~user@114-134-7-235.rurallink.co.nz] has joined #lisp 20:05:08 Mesh: what's the point? Are you sure it's going to affect performance? 20:05:26 Mesh: Premature optimization is the fastest way to write slow, buggy, impenetrable code. 20:05:39 the only thing you'll optimize is your own suffering. 20:06:29 Is (SB-KERNEL:%COERCE-CALLABLE-TO-FUN '(setf car)) supposed to err? 20:06:50 -!- sonnym [~evissecer@singlebrookvpn.lightlink.com] has quit [Quit: Leaving.] 20:06:56 deepfire: they don't have to be functions. 20:07:30 sykopomp, well, the point is I get a type error. 20:08:13 -!- felideon [~user@12.228.15.162] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 20:08:53 -!- ignas [~ignas@ctv-79-132-160-221.vinita.lt] has quit [Ping timeout: 245 seconds] 20:09:57 Also, #'(setf car) => #<... 20:10:10 yeah, this is something else. 20:10:27 it seems %coerce-callable-to-fun only accepts function designators. 20:10:53 which seems reasonable enough. 20:11:22 Salamander__ [~Salamande@ppp118-210-170-4.lns20.adl6.internode.on.net] has joined #lisp 20:11:34 deepfire: http://www.lispworks.com/documentation/HyperSpec/Body/26_glo_f.htm#function%20designator 20:12:46 Yeah, I know what isn't an extended function designator :-) 20:13:27 However, if you M-. down the call tree, you'd see the thing dealing with list cases. 20:13:30 -!- Salamander_ [~Salamande@ppp118-210-5-192.lns20.adl2.internode.on.net] has quit [Ping timeout: 252 seconds] 20:13:41 Which is what got me intrigued. 20:13:52 Kerrick [~Kerrick@b11wi-1.nat.iastate.edu] has joined #lisp 20:14:08 ah 20:15:14 Joreji [~thomas@77-066.eduroam.RWTH-Aachen.DE] has joined #lisp 20:16:24 Salamander_ [~Salamande@ppp121-45-124-123.lns20.adl6.internode.on.net] has joined #lisp 20:16:44 It ends up calling SB-IMPL:valid-function-name-p, which deals with lists. 20:17:14 However, this is the function form of %c-c-t-f, there are source transform forms. 20:18:01 http://paste.lisp.org/display/113929 20:18:06 it does not look like fun. 20:18:28 udzinari [~user@209.158.broadband13.iol.cz] has joined #lisp 20:18:34 i'm not even sure how should i resolve it. 20:18:37 redline6561 [~redline@c-66-56-55-169.hsd1.ga.comcast.net] has joined #lisp 20:18:42 -!- Salamander__ [~Salamande@ppp118-210-170-4.lns20.adl6.internode.on.net] has quit [Ping timeout: 240 seconds] 20:18:43 Dodek: don't use-package. 20:18:45 -!- legumbre [~leo@r190-135-22-212.dialup.adsl.anteldata.net.uy] has quit [Read error: Connection reset by peer] 20:18:58 Xach: i'm building weblocks via clbuild 20:19:13 i didn't even write any code yet. 20:19:24 Salamander [~Salamande@ppp118-210-79-92.lns20.adl2.internode.on.net] has joined #lisp 20:19:31 ouch 20:20:34 Odd. I haven't had any trouble loading weblocks lately. (but i use quicklisp, not clbuild) 20:20:52 legumbre [~leo@r190-135-73-190.dialup.adsl.anteldata.net.uy] has joined #lisp 20:21:01 quicklisp? never heard of. 20:21:03 jesusito [~user@193.pool85-49-235.dynamic.orange.es] has joined #lisp 20:21:12 -!- Salamander_ [~Salamande@ppp121-45-124-123.lns20.adl6.internode.on.net] has quit [Ping timeout: 252 seconds] 20:21:17 Xach: SBCL will throw that error even if you use DEFPACKAGE. 20:21:34 -!- Kenjin [~josesanto@bl19-227-50.dsl.telepac.pt] has quit [Quit: Get MacIrssi - http://www.sysctl.co.uk/projects/macirssi/] 20:21:45 Does weblocks depend on parenscript? 20:21:53 deepfire: yes. 20:22:05 sykopomp: It wasn't meant to suggest defpackage with :use either. 20:22:27 Dodek: quicklisp.org has a small bit of info 20:22:50 -!- Kerrick [~Kerrick@b11wi-1.nat.iastate.edu] has quit [Ping timeout: 264 seconds] 20:23:51 Dodek: I don't see that cl-who and parenscript have any conflicting symbols, either. maybe clbuild is pulling from some place different than me? 20:23:54 *Xach* checks 20:24:17 hmm 20:24:25 maybe i should have updated clbuild. 20:24:39 ahh, i'm getting parenscript from http://common-lisp.net/project/parenscript/release/parenscript-latest.tgz 20:24:40 no, there are no changes in darcs. 20:24:45 Kerrick [~Kerrick@b11wi-1.nat.iastate.edu] has joined #lisp 20:24:56 Xach: is that horribly outdated? :) 20:25:03 sykopomp: maybe...but it builds! 20:25:21 I like it when things build! 20:25:24 *Xach* will try from git 20:26:24 sykopomp: parenscript-latest is from february 2010, so not *too* ancient 20:27:23 i too would appreciate it when it builds. 20:28:23 there's a lot of appreciation going around for quicklisp 20:28:43 adu [~ajr@64.134.69.82] has joined #lisp 20:28:49 too bad the website lacks any info about downloading, repo etc. 20:28:58 or maybe i was unable to find it. 20:29:22 Dodek: It's not released yet. 20:30:04 -!- Kerrick [~Kerrick@b11wi-1.nat.iastate.edu] has quit [Ping timeout: 240 seconds] 20:30:18 -!- urandom__ [~user@p548A57CB.dip.t-dialin.net] has quit [Remote host closed the connection] 20:30:33 I just mention it to people with problems to make them feel bad! 20:31:09 heh, i knew there is a conspiracy. 20:31:34 Dodek: if you'd like to try it out, hop on #quicklisp and someone can give you a link 20:31:37 *Xach* is about to head out 20:32:23 Dodek: I get the same failure as you for parenscript from git. 20:32:40 -!- hargettp [~anonymous@pool-71-184-181-149.bstnma.east.verizon.net] has quit [Quit: hargettp] 20:32:58 Dodek: Vladimir Sedach did that just yesterday. Perhaps he could be persuaded to undo it. 20:33:05 Author: Vladimir Sedach 20:33:08 Rename stringify to str (clojure influence). 20:33:16 -!- Edward__ [~ed@AAubervilliers-154-1-30-161.w90-3.abo.wanadoo.fr] has quit [Ping timeout: 258 seconds] 20:33:49 haha. 20:34:20 Dodek: drop him an email, i bet he'd reconsider. 20:34:31 Weird, HANDLER-CASE with a T clause doesn't catch my type error. 20:34:49 i was actually wondering who should i send an email to. 20:35:19 Perhaps I need a newer SBCL. 20:35:28 Salamander_ [~Salamande@ppp121-45-99-102.lns20.adl6.internode.on.net] has joined #lisp 20:36:59 Hmm, doesn't look like there were changes potentially affecting this. 20:37:10 -!- mega1 [~quassel@catv4E5CABA2.pool.t-online.hu] has quit [Ping timeout: 265 seconds] 20:38:15 -!- Salamander [~Salamande@ppp118-210-79-92.lns20.adl2.internode.on.net] has quit [Ping timeout: 252 seconds] 20:39:44 -!- stassats` [~stassats@wikipedia/stassats] has quit [Ping timeout: 276 seconds] 20:40:56 qbomb [~qbomb@firewall.gibsonemc.com] has joined #lisp 20:41:00 -!- qbomb [~qbomb@firewall.gibsonemc.com] has left #lisp 20:41:06 qbomb [~qbomb@firewall.gibsonemc.com] has joined #lisp 20:44:34 -!- aidalgol [~user@114-134-7-235.rurallink.co.nz] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 20:44:40 Ok, I've got a testcase. 20:45:03 -!- adu [~ajr@64.134.69.82] has quit [Quit: adu] 20:45:21 LiamH: It would appear I fixed the issue 20:45:57 -!- daniel__ is now known as daniel 20:46:18 -!- rrice [~rrice@oh-67-76-18-254.sta.embarqhsd.net] has quit [Quit: Leaving.] 20:47:22 deepfire: for t-not-a-supertype? 20:47:48 Hm, the pastebot didn't yell. 20:48:04 udzinari` [~user@209.158.broadband13.iol.cz] has joined #lisp 20:48:07 lisppaste doesn't announce in here anymore, I think. 20:48:11 http://paste.lisp.org/display/113930 20:48:26 lichtblau [~user@port-92-195-188-50.dynamic.qsc.de] has joined #lisp 20:48:42 -!- Joreji [~thomas@77-066.eduroam.RWTH-Aachen.DE] has quit [Remote host closed the connection] 20:48:44 Joreji_ [~thomas@77-066.eduroam.RWTH-Aachen.DE] has joined #lisp 20:48:45 or he's muted or something 20:48:54 *deepfire* pasted "funcall fails to call '(SETF FOO)" at http://paste.lisp.org/display/113930 20:49:15 -!- _8david [~user@port-92-195-188-50.dynamic.qsc.de] has quit [Ping timeout: 265 seconds] 20:49:29 Someone, please, tell me it is indeed a bug and I'll stuff it into launchpad. 20:50:11 Alternatively, tell me I'm on crack and that I need to stop my delusions creeping into this island of sanity. 20:50:55 LiamH: Ah, dammit, no, it's not fixed. I apparently by chance got 0 errors a few times in a row. 20:51:38 -!- udzinari [~user@209.158.broadband13.iol.cz] has quit [Ping timeout: 264 seconds] 20:51:43 Sikander: well, that's better than what I was seeing. 20:52:09 Okay, I'm calling it a day and homeward I go. 20:52:38 -!- ssideris [~sideris@93-97-178-232.zone5.bethere.co.uk] has quit [Ping timeout: 245 seconds] 20:52:41 -!- kwabbles_ [~mike@h-67-101-178-243.lsanca54.static.covad.net] has quit [Quit: kwabbles_ has no reason] 20:53:16 Oh, it's not a bug, indeed. FUNCALL doesn't take extended function designators. 20:54:18 ssideris [~sideris@93-97-178-232.zone5.bethere.co.uk] has joined #lisp 20:55:06 Interestingly, (let ((a '(1))) (funcall '(setf car) 2 a) a) => 2 20:55:14 -!- Joreji_ [~thomas@77-066.eduroam.RWTH-Aachen.DE] has quit [Ping timeout: 264 seconds] 20:55:32 Which is what misled me into thinking it does accept them. 20:56:23 -!- Beetny [~Beetny@ppp118-208-129-226.lns20.bne1.internode.on.net] has quit [Ping timeout: 245 seconds] 20:58:17 -!- varjag [~eugene@4.169.249.62.customer.cdi.no] has quit [Quit: Ex-Chat] 20:59:07 LiamH: Dammit... 20:59:16 Actually, the above LET snippet confuses the hell out of me. 20:59:35 Sikander: ? 20:59:37 LiamH: Ok, so as it turns out, gsl/fft/test.c (macro-expanded) tests wether out-of-stride elements are left alone 21:00:06 LiamH: But it tests this ONLY for complex FFT. For real FFT, it doesn't, as it apparently doesn't expect those to be left alone 21:00:27 LiamH: In other words, the failures are a result of the lisp-unit test checking out-of-stride as well 21:00:29 jmcphers [~jmcphers@218.185.108.156] has joined #lisp 21:01:28 -!- ToGgG [~Shiva@3615mylife.foreverlovingjah.fr] has left #lisp 21:01:35 LiamH: So for a copy of the GSL test functions, we need two other compare functions. 21:02:01 LiamH: One that tests in-stride only, and one that does the exact complement (i.e. _not_ testing in-stride, but out-of-stride only) 21:02:41 sonnym [~evissecer@rrcs-184-74-137-167.nys.biz.rr.com] has joined #lisp 21:04:33 LiamH: So that was a whole lot of digging for nothing; we were looking at the wrong lisp part. The problem lies in the fft-real-result-check for stride != 1 21:04:42 -!- Jabberwockey [~Jens@port-12485.pppoe.wtnet.de] has quit [Read error: Connection reset by peer] 21:04:59 Kerrick [~Kerrick@e40-1.nat.iastate.edu] has joined #lisp 21:05:16 Sikander: Oh, I can't remember what I was looking at. 21:05:43 -!- dlowe [~dlowe@ita4fw1.itasoftware.com] has quit [Quit: Leaving.] 21:06:05 -!- loxs [~loxs@85-130-37-174.2073285806.ddns.cablebg.net] has quit [Quit: Leaving] 21:07:53 LiamH: So when you only perform tests with stride = 1, then there's no problem 21:08:10 ejs [~eugen@92-49-213-218.dynamic.peoplenet.ua] has joined #lisp 21:08:24 holycow [~new@poco208-2.fredcanhelp.com] has joined #lisp 21:08:26 galaxywatcher [~galaxywat@ppp-61-90-20-10.revip.asianet.co.th] has joined #lisp 21:08:52 LiamH: But to come back to your initial problem, yes, testing all sizes from 1 to 99 (with only stride = 1) takes 55 seconds. Which is longer than gsl, which does a whole bunch of other tests as well. 21:09:05 -!- moocow [~new@poco208-2.fredcanhelp.com] has quit [Ping timeout: 260 seconds] 21:09:23 Sikander: but that's way faster than what I was seeing before. 21:10:09 LiamH: Even though I only test for stride = 1? Did you see more than a factor of 3 longer? Don't remember. 21:10:31 Sikander: 17 minutes is what I recall. 21:10:45 Oh, right 21:11:19 So factor of 3 improvement maybe; maybe better, considering that stride >1 should take less time than stride 1 for same size. 21:12:04 -!- seelenquell [~kvirc@pD9E42CA4.dip.t-dialin.net] has quit [Quit: When two people dream the same dream, it ceases to be an illusion. KVIrc 3.4.2 Shiny http://www.kvirc.net] 21:12:04 LiamH: I'm not sure about this since I don't have any experience with optimization, but could it be because most of the lisp testing functions (making vectors, generating noise, etc) are not compiled with optimization on? 21:13:07 stassats` [~stassats@wikipedia/stassats] has joined #lisp 21:14:04 LiamH: That would mean that it's at _least_ a factor of about 6 improvement. Less than 1 minute for more than a third of the tests that initially took 17 minutes. 21:14:29 Dodek: will you mail him, or shall i? 21:14:42 LiamH: How long does (all-fft-test-forms 99 1 (64 99)) take on your computer? 21:14:44 Yay, peephole optimiser for SBCL popped up! 21:14:53 can somebody explain to me what a DSL is and why it's powerful? 21:15:11 because it's specific 21:15:18 domain specific even 21:15:42 sabalaba, you can express your problem in terms of your choice. 21:15:59 And you can compose them any way you like. 21:16:21 deepfire, I've heard that one of the things that makes Lisp powerful is its ease of composing DSLs--I don't know what a DSL is or what it does 21:17:18 DSL stands for domain-specific language, but you could have found this much anywhere. 21:17:24 sabalaba: It is well known that the closer to your application domain a programming language is, the easier it is to express your problem in it. So, there is a software-engineering methodology that consists of first writing a domain-specific language, and then writing your application in it. 21:17:51 Sikander: started 21:18:35 beach, ok, now that I understand that. can you give me an example? 21:19:10 clhs loop 21:19:10 http://www.lispworks.com/reference/HyperSpec/Body/m_loop.htm 21:19:11 beach, it makes sense that a programming language that is close to your application domain will be easier to use for solving the problem--but I guess I don't understand how that might manifest itself 21:19:11 Sikander: if what you have compiles and runs, can you commit it? 21:19:13 sabalaba: Now, most people who write domain-specific languages use traditional technology from compiler theory like lex/yacc, etc. However, as Paul Graham points out in OnLisp, it is much more convenient to write such a language on top of an extensible language like Lisp. 21:19:20 sabalaba: see the link 21:19:36 sabalaba: I thought that http://www.cl-http.org:8002/mov/dsl-in-lisp.mov is a great intro 21:19:38 -!- cinch [~cinch@unaffiliated/cinch] has quit [Remote host closed the connection] 21:20:13 sabalaba: It also makes sense to write your DSL as an extension to your ordinary programming language, provided it allows for that, and Lisp uniquely does. 21:20:15 deepfire, I believe the problem is ordinary lambda list vs gf-lambda-list 21:20:48 Xach: i already did. 21:20:56 Sikander: 23.195 seconds of real time 21:21:23 petercoulton [~petercoul@cpc4-midd16-2-0-cust402.11-1.cable.virginmedia.com] has joined #lisp 21:21:26 Sikander: FAST-FOURIER-TRANSFORM: 1010 assertions passed, 0 failed. 21:21:37 sabalaba: That technique is called "embedded language" in that it is able to reuse all of the host language (say arithmetic, etc), and only the additional stuff that is specific to your DSL has to be implemented, whereas if you do it in a traditional language, you essentially have to re-implement all that. 21:21:38 sabalaba: you might take a look at http://www.gigamonkeys.com/book/practical-parsing-binary-files.html for an example of a DSL embedded in Lisp. 21:21:39 gigamonkey, memo from p_l: what's the licensing status of PCL code samples? 21:21:55 p_l: you around? 21:22:00 LiamH: Yep. Apparently I have a really old computer. 21:22:17 gigamonkey, beach, Sikander, stassats` thank you all very much 21:22:30 Sikander: still unreasonably long compared to what GSL does 21:22:43 minion: memo for p_l: BSD license on the code, IIRC. There should be a COPYING file in the download. 21:22:43 Remembered. I'll tell p_l when he/she/it next speaks. 21:22:51 Sikander: which is to fly through it so fast you can't even tell it ran :-( 21:23:01 sabalaba: Here is a great example that happened to me. I needed to get the functionality of Metafont (by Donald Knuth). Knuth, having only access to languages like Pascal, re-implemented an entire language. I was able to implement most of it on top of CL in a few weeks. 21:23:31 LiamH: Yes. Clearly, it can't delay on the ffi side. Could it be that the lisp layer in between should be compiled with optimizations on? 21:23:35 minion: memo for p_l: Actually it's called, LICENSE ;-) 21:23:36 Remembered. I'll tell p_l when he/she/it next speaks. 21:25:04 sabalaba, a fun (but of, admittedly, less importance) part is that your DSL programs largely compile down to machine code, because your DSL is "just" re-fashioned Lisp. 21:25:27 hargettp [~anonymous@pool-71-184-181-149.bstnma.east.verizon.net] has joined #lisp 21:25:49 -!- qbomb [~qbomb@firewall.gibsonemc.com] has left #lisp 21:27:12 LiamH: I also don't know how grid works internally; if there's any translation back and forth from grids to C arrays, could that provide a large overhead? 21:28:07 Of course nothing prevents you from coming up with an intepreted DSL, but people mostly use standard techniques to compose their DSL's so that they end up being Lisp. 21:28:30 Sikander: yeah, it might. But I really don't think it should be that large; if it is, it's saying something really bad about the implementation of grids. 21:29:11 On the topic of DSLs, I notice that when people make interactive programs (e.g. shuffletron or pjb's ed implementation) there's still quite a large (cond ...) to interpret the input. 21:29:22 -!- jesusito [~user@193.pool85-49-235.dynamic.orange.es] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 21:29:33 Is that the usual way of doing that type of thing? 21:29:46 Sikander, does that COND execute during the runtime? 21:29:55 gko [~gko@122-116-15-138.HINET-IP.hinet.net] has joined #lisp 21:30:03 Well, after macroexpansions. 21:30:41 *madnificent* wonders who works at Franz and is in #lisp regularly 21:31:04 madnificent: antifuchs for instance. 21:31:05 Isn't antifuchs working there since recently? 21:31:07 deepfire: I think so; it's directly in a defun 21:31:28 ah, it was antifuchs himself :) thanks 21:32:47 deepfire: is that not good? Is there a better way? 21:33:08 -!- stis [~stis@1-1-1-39a.veo.vs.bostream.se] has quit [Remote host closed the connection] 21:33:09 Anarch [~w3@elrond.hsl.washington.edu] has joined #lisp 21:33:25 LiamH: Ok, but where could the overhead come from if not from there? 21:33:28 Jasko [~tjasko@c-174-59-223-208.hsd1.pa.comcast.net] has joined #lisp 21:34:13 Sikander: you ask an excellent question 21:34:48 LiamH: Would a declare optimize here and there change anything? Mind you, I only did this in one other project, and that's only because similar functions already had optimize on as well 21:34:50 I smell some profiling in my future. Oh wait, I can't profile the C part, can I? 21:35:08 If you're talking about keyboard input, does keyboard input result in some sort of stuff that you could use as a key in a hash table, and then map the keys to functions that way or something..? 21:35:19 -!- sellout [~greg@2002:4855:eb9a:0:21d:4fff:fefe:c504] has quit [Quit: sellout] 21:35:33 Sikander: it might, but again, the non-optmized part shouldn't be THAT bad 21:35:46 mm, not part, I mean the non-optimized version 21:36:14 Ok 21:36:27 I should learn how to profile in lisp... 21:36:46 Sikander: so what is it that you fixed? 21:37:04 LiamH: Well, nothing, obviously. :( 21:37:05 Sikander, shuffletron isn't a traditional Lisp-style DSL at all. 21:37:26 Fare [~Fare@ita4fw1.itasoftware.com] has joined #lisp 21:37:40 LiamH: While poking around and rewriting/improving some functions, I found that the real problem lies in the comparison 21:37:45 Sikander: so we still haven't found the bug(s)? 21:38:34 LiamH: Well, if you want to call it a bug, then I can say that the bug is that the function that compares the results of real FFTs with stride > 1 do a wrong compare. 21:38:37 Sikander, if, as I assume, you refer to PARSE-AND-EXECUTE, it's a textline-oriented UI, not a programmer-oriented DSL. 21:39:00 -!- tcr1 [~tcr@81-233-246-97-no37.tbcn.telia.com] has quit [Read error: Operation timed out] 21:39:11 Sikander: OK, but we don't know more than that at this point? 21:39:12 deepfire: Yes, that's what I refer to. Is that the best you can do for a textline-oriented UI (similar also to ed, in that sense) 21:39:22 (The orientations I mentioned are of orthogonal senses.) 21:39:28 deepfire, hi! 21:39:42 Fare, hi :-) 21:39:45 LiamH: I'm 99% sure that that's the problem. The GSL tests specifically test only if the vectors are equal, taking the stride into account 21:39:49 Hum. rpg's and my article for ilc2010 is taking shape. 21:40:11 -!- jdz [~jdz@host10-106-dynamic.14-87-r.retail.telecomitalia.it] has quit [Ping timeout: 276 seconds] 21:40:18 if anyone is interesting in losing 1d6 SAN, he can read what we wrote about ASDF. 21:40:25 LiamH: In the tests for the real ffts, it skips the tests on the off-stride components, which it DOES perform for the complex ffts. 21:41:03 deepfire, it saddens me to see the same thing done over and over again, and good work getting wasted. 21:41:15 Sikander, I've not much experience with textline UIs, actually. The only one I did was for the clbuild-like package management thing, and I used reader-macros. 21:41:16 now Xach is reimplementing half of desire. 21:41:18 LiamH: So if lisp-unit has a way to compare only on-stride components, and also a way to compare only off-stride components, then I can figure out more 21:41:25 Sikander: to answer your initial question, do you know ASDF has a run-program in it? 21:41:30 I'm trying to write a macro that takes two arguments, args and body, and then would expand to (lambda args body), assuming that args was a list and body was a &body... list that was spliced in. Whatever. 21:41:41 Fare, I was loud enough on this channel about desire, btw. 21:41:51 I suppose. Sigh. 21:41:55 Xach surely heard the noises I made. 21:42:04 sure. 21:42:18 /nobody/ except juanjo and you ever showed any interest. 21:42:23 marketing is a difficult art. 21:42:36 I don't master it myself. 21:42:48 tcr1 [~tcr@81-233-246-97-no37.tbcn.telia.com] has joined #lisp 21:42:51 LiamH: ... No I didn't know that, I'll have a look. I found out that trivial-shell won't do the job, since I want to capture the output of the program. 21:43:20 This seems to work okay if I pass in the literal arguments arg (if I wanted x and y as arguments, I need to pass in (x y)), but if I try to extract the arguments from somewhere else (let's say I have a list of potential argument pairs I want and do (second possible-args)) I can't really pass that in since it'll use second and possible-args as the arguments instead of what that would otherwise evaluate to 21:43:38 Fare, deepfire perhaps nobody bothered to setup some collaboration? 21:43:44 deepfire: Which is this clbuild-like package management thing, and can I find the source somewhere to see what you did? 21:43:58 So is there any way to have the macro evaluate args at expansion time, and then insert whatever it evaluates to..? 21:44:02 deepfire: I mostly remember desire from fare asking about it. 21:44:07 http://www.feelingofgreen.ru/shared/src/desire/doc/overview.html 21:44:28 you know what they say: it's not enough to be better, if you're not ten times better, no one will bother to switch. 21:44:46 deepfire, f.e. there's no wiki to yell at you (-; 21:45:00 deepfire: Thanks :) 21:45:05 -!- s0ber [~s0ber@111-240-209-101.dynamic.hinet.net] has quit [Ping timeout: 255 seconds] 21:45:11 And the buildbot: http://feelingofgreen.ru/desire-waterfall 21:45:22 abugosh [~Adium@pool-173-66-13-115.washdc.fios.verizon.net] has joined #lisp 21:45:27 (and, btw, having common-lisp.net, why is everyone setting up something on the launchpad?) 21:45:34 nus, there's a launchpad project. 21:45:54 -!- pavelludiq [~user@83.222.175.142] has quit [Remote host closed the connection] 21:46:00 -!- coyo [~unf@unaffiliated/bandu] has quit [Quit: KVIrc Insomnia 4.0.0, revision: , sources date: 20090520, built on: 2010/07/07 00:46:30 UTC http://www.kvirc.net/] 21:46:19 https://launchpad.net/~desire 21:46:25 Hey, this desire thing seems pretty nice 21:46:28 Mesh: yes. 21:46:39 https://launchpad.net/~desire 21:46:46 Oh, sorry. 21:46:58 pjb: Anywhere I might want to look about that? 21:47:01 s0ber [~s0ber@111-240-215-37.dynamic.hinet.net] has joined #lisp 21:47:09 -!- gko [~gko@122-116-15-138.HINET-IP.hinet.net] has quit [] 21:47:16 I'm not sure the bootstrap script works anymore, actually. 21:47:16 Mesh: clhs eval 21:47:28 Last I tested it was half a year ago. 21:47:38 Sikander: oops, asdf:run-shell-command; see init.lisp in GSLL 21:47:40 See I was doing that but I read somewhere that using eval like that was a bad idea? 21:47:41 Mesh: notice the standard difficulties with eval, combined by the fact that you don't know when and how many times the macro is executed! 21:47:50 Heck, I even had an automated /buildbot/ deployment testing set up. 21:48:09 Remote deployment, even. 21:48:19 Mesh: so, while what you're asking is possible, you really don't want to do that. Here you have a hint you should do it otherwise. 21:48:29 Mesh: Therefore, what do you want to do really? 21:48:41 LiamH: Hey, nice 21:48:50 -!- hargettp [~anonymous@pool-71-184-181-149.bstnma.east.verizon.net] has quit [Quit: hargettp] 21:49:30 Mesh: more precisely, the problem is that while DEFMACRO, MACROLET, MACROEXPAND, etc take an &ENVIRONMENT argument, EVAL does not. 21:49:42 -!- gravicappa [~gravicapp@ppp85-141-167-179.pppoe.mtu-net.ru] has quit [Remote host closed the connection] 21:49:50 -!- Davidbrcz [~david@212-198-91-47.rev.numericable.fr] has quit [Read error: Operation timed out] 21:50:04 -!- ssideris [~sideris@93-97-178-232.zone5.bethere.co.uk] has quit [Ping timeout: 265 seconds] 21:50:11 The buildslave-oriented phases of the buildbot have been discontinued, as I never bothered to re-configure the buildslave machine after it got new hardware and OS install. 21:50:39 What I want to do is store a set of lambda functions in a hash table that maps said functions to a symbol name. 21:50:47 can a kind soul include 2.007 in SBCL? 21:51:01 LiamH: Why don't you use the same trick for fsbv to determine where to find the library via pkg-config? 21:51:02 Only instead of using the standard lambda syntax I'd like to have something that looks like (symbol (args) body) 21:51:26 Sikander: because pkg-config gives me bogus answers a lot 21:51:31 Mesh: first do it with functions, and then write a wrapper macro. 21:51:51 LiamH: ah, ok 21:51:57 ssideris [~sideris@93-97-178-232.zone5.bethere.co.uk] has joined #lisp 21:52:11 LiamH: how bogus ? 21:52:17 Even though the actions required are trivial, like setting up an ssh account. 21:52:22 LiamH: Because libffi still installs to some arbitrary location 21:52:23 Mesh: (defun register-function (name function) (setf (gethash name *funtable*) function)) 21:52:46 pjb: The hash-adding stuff isn't part of the macro. 21:52:49 fe[nl]ix: can't find the header files that are installed in the normal Debian place. 21:52:49 -!- abugosh [~Adium@pool-173-66-13-115.washdc.fios.verizon.net] has quit [Quit: Leaving.] 21:53:04 Mesh: (defmacro define-command (name arguments &body body) `(register-function ,name (lambda ,arguments ,@body))) 21:53:16 Sikander: are you on OS X? 21:53:38 pjb: That doesn't address the problem I was having, does it? 21:53:51 BTW, as the buildslave phases are separated, it should be fairly trivial to arrange the buildbot to do testing on multiple OS/arch combinations, as long as desire works there. 21:53:52 insanux [~hola@83.50.180.208] has joined #lisp 21:54:05 LiamH: then you should file a bug with debian 21:54:06 So far, though, it's only linux/CCL and linux/SBCL. 21:54:27 As I said, if you have a list of argument lists and try to feed that (second arg-lists) you'd end up with it expanding to 21:54:35 linux/ECL is almost there. 21:54:40 (lambda (second arg-lists) body stuff here) 21:54:42 fe[nl]ix: I should 21:54:54 Mesh: (defmacro define-commands (&rest definitions) `(progn ,@(mapcar (lambda (def) (destructuring-bind (name argument &body body) def `(register-function ,name (lambda ,argument ,@body))))))) 21:55:15 Mesh: you can write `(lambda ,(second arg-lists) ,@body stuff here) 21:55:20 deepfire, the overview doc is total mess: architecture, terminology, engineering all intermixed. 21:55:26 LiamH: Nope 21:55:51 pjb: That's assuming that I'll ALWAYS use (second arg-lists) there which isn't a very sound assumption 21:56:01 Sikander: what os, and how did you (it) install libffi? 21:56:10 Mesh: It's YOU who's programming. 21:56:16 deepfire, the section 6 'Overview of terms' shouldn't sit somewhere in the middle 21:56:26 LiamH: libffi headers are installed in a most terrible place: /usr/lib64/libffi-3.0.9/include 21:56:30 Mesh: clhs if 21:56:50 Sikander: does pkg-config give the right answer? 21:57:11 LiamH: I'm running gentoo. Install happens as upstream defined it. In that terrible directory with the version name 21:57:14 LiamH: Yes 21:57:17 nus, thanks for feedback! 21:57:55 LiamH: When I update fsbv (I think I did this only two times) I manually paste in the output of pkg-config 21:57:56 you're welcome. 21:58:05 pjb: Okay, let me try explaining this another way. 21:58:23 LiamH: pkg-config --cflags --libs libffi gives me -I/usr/lib64/libffi-3.0.9/include -lffi which is correct 21:58:44 _s1gma [~d.d.derp@77.107.164.131] has joined #lisp 21:58:45 I want whatever args evaluates to. 21:58:51 Not args itself. 21:58:55 As Fare said, it's a pity that so many efforts are duplicated and wasted 'cause of miscommunication. 21:58:57 This is not a good idea. 21:59:02 -!- _s1gma [~d.d.derp@77.107.164.131] has quit [Max SendQ exceeded] 21:59:02 And I don't want to use eval to evaluate it. 21:59:06 -!- billstclair [~billstcla@unaffiliated/billstclair] has quit [Ping timeout: 252 seconds] 21:59:12 pb: Why isn't it a good idea? 21:59:13 nus, honestly, I thought I've annoyed the heck of people. 21:59:17 symbole [~user@216.214.176.130] has joined #lisp 21:59:22 Mesh: (defvar *arg-list* '(x y z)) (whatever-macro *arg-list*) won't work cleanly. 21:59:40 *out of the 21:59:46 Sikander: it gives me only -lffi despite the existence of /usr/include/x86_64-linux-gnu/ffi.h 21:59:49 Because of the compilation-time / macroexpansion-time / run-time environments differences. 22:00:02 Okay, so what about this. 22:00:08 LiamH: ? Even when indicateing --cflags? 22:00:15 Mesh: what you can do, is to build the functions at run-time. Then you don't need macros. 22:00:24 Sikander: I gave it exactly the command you gave. 22:00:45 LiamH: Perhaps it is because that location is a default location? I don't know... 22:00:55 Mesh: (register-function 'example (compile nil `(lambda ,(rest *arg-list*) ...))) 22:01:14 Sikander: perhaps so, but then pkg-config becomes distinctly less helpful 22:01:18 LiamH: At least, when it's at that location, you (clearly) don't need to tell the compiler where to find the header file, since it'll look for it there by itself 22:01:25 LiamH: I agree 22:01:26 That's not actually what I want to do. 22:01:34 Sikander: if you're using gcc, yes 22:01:45 -!- Kerrick [~Kerrick@e40-1.nat.iastate.edu] has quit [Ping timeout: 265 seconds] 22:01:48 Here, what if I stick a let statement inside of a lambda function being used by a mapcar? 22:02:05 Right. 22:02:06 LiamH: What os? 22:02:16 Debian testing 22:02:17 Mesh: so you must know your argument list at compilation time. 22:02:24 deepfire, I believe it'd take at least twice as much time and efforts as were spent on development to properly document and "market" the thing 22:02:49 Mesh: notice that you can use (&rest arguments)... 22:02:51 coyo [~unf@unaffiliated/bandu] has joined #lisp 22:02:53 What exactly would happen there? If I have (let ((x (second args))) whatever), would that binding, like, die with every iteration of the... mapping thing? 22:02:55 -!- scode_ is now known as scode 22:03:13 nus, I have no skills or desire to do the marketing thing. As to documentation, well, I thought it was half-decent. 22:03:29 Mesh: indeed. 22:03:42 If you noticed, there's a "crash tutorial": http://www.feelingofgreen.ru/shared/src/desire/doc/tutorial.html 22:03:48 So i wouldn't be shadowing the original binding? The original binding wouldn't be there anymore? 22:04:01 Mesh: we need more context. 22:04:33 deepfire: I never heard of desire, I have to admit. It looks pretty decent. 22:04:37 khrum42 [~khrum42@2001:a60:f05d:1::dead] has joined #lisp 22:05:24 Sikander, half a year later it may be bit rot, due to dependency movement. 22:05:44 half a year since I last touched it, I mean. 22:05:50 (mapcar #'(lambda (x) (let ((y x)) (format t "~A" y))) number-list) 22:05:55 something like that. 22:05:59 _s1gma [~d.d.derp@77.107.164.131] has joined #lisp 22:06:02 There is no problem with this form. 22:06:06 -!- khrum42 [~khrum42@2001:a60:f05d:1::dead] has quit [Client Quit] 22:06:09 Alright, then. 22:06:37 khrum42 [~khrum42@2001:a60:f05d:1::dead] has joined #lisp 22:06:40 Why would you have thought otherwise? 22:07:25 billstclair [~billstcla@unaffiliated/billstclair] has joined #lisp 22:07:39 jan247 [~jan247@unaffiliated/jan247] has joined #lisp 22:08:41 deepfire: That's too bad... You won't revive it? 22:09:00 deepfire, have in mind that people that could make sence of the overview already have some infrastructure in place and their utmost concern is the migration. OTOH, your average CL newbie needs not only answers to "how?"s but "why?"s too. 22:09:32 -!- tcr1 [~tcr@81-233-246-97-no37.tbcn.telia.com] has quit [Quit: Leaving.] 22:09:39 dto [~user@pool-96-252-62-25.bstnma.fios.verizon.net] has joined #lisp 22:10:29 deepfire: I agree with nus. I'm a bit of a newbie, with a little over a year experience with CL and would like to see things like how it compares to e.g. asdf and clbuild or if it doesn't quite compare, how it fits in. 22:11:29 Sikander, I may revive it, let's put it this way. I'm just not sure if anyone would want it. 22:12:48 I have a serious project underway now, which keeps my thoughts when I'm falling asleep and catches them when I wake up. 22:13:13 LiamH: that would mean that debian testing messes around with /usr/lib/pkgconfig/libffi.pc. 22:14:16 Sikander: I think I agree with your previous statement, if it's in the default location, it doesn't give you a flag because that's presumed to be seen. 22:14:20 deepfire: The design seems pretty nice, from what I can see, but it's too bad it's not platform and lisp agnostic 22:14:42 deepfire, that's the curse of users. When you don't have them, you curse. When you have them, you curse. 22:15:11 Fare, at the time it seemed to me I would have begged them for feedback. 22:15:19 LiamH: Yes, that's it indeed. My pkg-config also doesn't give me any cflags with libmpg123. 22:15:31 I was doing desire daily, for /months/. 22:16:05 Sikander: but it makes me wonder, all cffi-grovel is doing is running gcc anyway, so shouldn't the default work just fine? 22:16:10 Heck, I was skipping dayjob for it. 22:16:23 :-/ 22:16:32 I know exactly how you feel 22:17:20 LiamH: That would mean that you should be able to run pkg-config --cflags, and if it gives output, use that as a cc-flags argument. If there's no output, you just don't run the cc-flags. 22:18:00 Sikander: --cflags-only-I I think 22:18:19 Salamander__ [~Salamande@ppp118-210-24-28.lns20.adl2.internode.on.net] has joined #lisp 22:18:25 Sikander: you might be right 22:18:36 -!- symbole [~user@216.214.176.130] has quit [Remote host closed the connection] 22:18:44 So the reason I didn't know about desire, is that by the time I started getting more into lisp, you already stopped working on it. 22:19:01 LiamH: sure, --cflags-only-I sounds good. 22:19:05 Sikander: I'd need several OS Xers to test; that was the platform that was causing all the problems. 22:19:17 LiamH: Ah, ok. 22:19:47 due to darwin ports, fink, OS X itself etc. being possible sources of libraries 22:20:08 LiamH: Ok, so it's late, and I'm going to turn in. What do we do about a lisp-unit comparison that works on-stride only and one that works off-stride only? 22:20:50 Sikander: I don't really understand what you've discovered, but I'll try to look at it tonight and see if I can figure out what's going on. 22:21:06 LiamH: Let me explain a bit better. 22:21:07 -!- Salamander_ [~Salamande@ppp121-45-99-102.lns20.adl6.internode.on.net] has quit [Ping timeout: 252 seconds] 22:21:38 -!- ejs [~eugen@92-49-213-218.dynamic.peoplenet.ua] has quit [Ping timeout: 264 seconds] 22:21:43 LiamH: The only tests that fail, as far as I can see, are the tests on real FFTs with a stride > 1 22:22:13 rbarraud [~rbarraud@118-93-169-153.dsl.dyn.ihug.co.nz] has joined #lisp 22:22:27 LiamH: In these, the GSLL tests verify that _all_ the components of the vectors are the same. 22:22:46 LiamH: However, in the equivalent GSL tests, only the on-stride components are checked. 22:23:00 LiamH: So, that's for the real FFTs with a stride > 1 22:23:19 -!- Salamander__ [~Salamande@ppp118-210-24-28.lns20.adl2.internode.on.net] has quit [Ping timeout: 252 seconds] 22:23:26 Sikander: OK, so if I modify that comparison test, it should all work? 22:23:42 LiamH: for complex FFTs, the GSL tests do something similar (not quite, though) to what GSLL tests do now already. 22:23:51 Ok, I'm going offline for a while. 22:24:00 bye deepfire 22:24:05 later on 22:24:13 LiamH: It would help seeing if there are _real_ problems. 22:24:19 -!- petercoulton [~petercoul@cpc4-midd16-2-0-cust402.11-1.cable.virginmedia.com] has quit [Ping timeout: 272 seconds] 22:25:00 Sikander: OK, I will see if I can make real more like complex in that regard. 22:25:23 LiamH: See, what I'm trying to say is that the GSL tests don't assume _all_ components will match, only the on-stride ones. So behaviour of off-stride ones (for _real_ FFTs) is according to tests in GSL undefined. 22:25:37 LiamH: No that won't help 22:25:51 Sikander: ? 22:25:51 LiamH: what we need is three compare functions. 22:26:11 LiamH: One that compares _every_ element. We have that already, and you are already using that. 22:26:27 aidalgol [~user@114-134-7-235.rurallink.co.nz] has joined #lisp 22:26:38 LiamH: We need a second compare that only compares on-stride elements. These should be used for all ffts with stride > 1 22:26:45 hohoho [~hohoho@ntkngw229253.kngw.nt.ftth.ppp.infoweb.ne.jp] has joined #lisp 22:27:19 LiamH: We then need a third compare that is the complement to the second compare, namely it only compares off-stride elements. These should be used (in specific cases only) for complex FFTs. 22:27:27 Edward_ [ed@AAubervilliers-154-1-32-98.w90-3.abo.wanadoo.fr] has joined #lisp 22:27:36 l4ndfo [~l4ndfo@S0106002129a187e9.ed.shawcable.net] has joined #lisp 22:27:44 LiamH: Do you see what I mean? 22:28:09 LiamH: GSLL just always uses the first compare method, while GSL uses all three, for different cases depending on the case. 22:28:32 LiamH: Clearly, GSLL tests may fail due to using the wrong compare method. 22:28:45 Sikander: I do, though I'm not entirely sure why. However it occurs to me that the first two could be folded into one, as in compare every-n-elements, and accept that n=1 is permissible. 22:29:01 LiamH: Sure. 22:29:21 hargettp [~anonymous@pool-71-184-181-149.bstnma.east.verizon.net] has joined #lisp 22:29:25 Sikander: I will take a look at it tonight. 22:30:19 LiamH: The reason lies, I expect, in the packing/unpacking, real to half-complex stuff. Even with a stride, these vectors are handled in such a way by GSL, that it cannot guarantee off-stride elements are maintained. I need to read the docs regarding that and see what they say. 22:30:34 LiamH: s/expect/suspect/ 22:30:58 Sikander: Ah, OK that makes some sense, but I don't understand why we get random agreement. 22:31:27 sellout [~greg@c-24-61-13-161.hsd1.ma.comcast.net] has joined #lisp 22:31:39 -!- stassats` [~stassats@wikipedia/stassats] has quit [Read error: Operation timed out] 22:31:44 LiamH: But it apparently _can_ guarantee it (sometimes?) for complex FFTs 22:32:23 LiamH: I think I did manage to fix that. Let me have a look at it tomorrow, to be sure there's no weirdness, and I'll send you a patch for that. 22:32:24 -!- bozhidar [~user@93-152-185-88.ddns.onlinedirect.bg] has quit [Read error: Connection reset by peer] 22:32:40 LiamH: with "fix that" I meant "fix that randomness". 22:33:45 LiamH: Ok, if we're good for now, I'm off. 22:34:10 Sikander: me too, got to go home. Goodnight. 22:34:33 Salamander [~Salamande@ppp118-210-157-21.lns20.adl6.internode.on.net] has joined #lisp 22:34:37 Goodnight all 22:34:46 -!- Sikander [~soemraws@5356ECA7.cable.casema.nl] has quit [Quit: Time to sleep...] 22:36:53 -!- LiamH [~none@pdp8.nrl.navy.mil] has quit [Quit: Leaving.] 22:37:17 -!- udzinari` [~user@209.158.broadband13.iol.cz] has quit [Ping timeout: 255 seconds] 22:39:53 \join org-babel 22:41:26 rbarraud_ [~rbarraud@118-93-169-153.dsl.dyn.ihug.co.nz] has joined #lisp 22:44:13 moocow [~new@poco208-2.fredcanhelp.com] has joined #lisp 22:44:18 -!- holycow [~new@poco208-2.fredcanhelp.com] has quit [Ping timeout: 240 seconds] 22:45:08 cmo-0 [~user@92.99.48.191] has joined #lisp 22:46:00 -!- rbarraud [~rbarraud@118-93-169-153.dsl.dyn.ihug.co.nz] has quit [Ping timeout: 272 seconds] 22:46:04 -!- tritchey [~tritchey@c-98-226-81-194.hsd1.in.comcast.net] has quit [Ping timeout: 240 seconds] 22:48:09 brnhack [~hrk@ntsitm087073.sitm.nt.ngn4.ppp.infoweb.ne.jp] has joined #lisp 22:49:15 -!- MetalDust [~metaldust@75-32-203-61.lightspeed.ftwotx.sbcglobal.net] has quit [] 22:49:29 -!- brnhack [~hrk@ntsitm087073.sitm.nt.ngn4.ppp.infoweb.ne.jp] has quit [Client Quit] 22:49:37 Kerrick [~Kerrick@e40-1.nat.iastate.edu] has joined #lisp 22:50:41 psilord2 [~psilord@adsl-99-154-14-142.dsl.mdsnwi.sbcglobal.net] has joined #lisp 22:52:36 Salamander_ [~Salamande@ppp118-210-250-167.lns20.adl6.internode.on.net] has joined #lisp 22:54:43 -!- wbooze [~user@xdsl-78-34-232-86.netcologne.de] has quit [Ping timeout: 245 seconds] 22:54:55 -!- homie [~user@xdsl-78-34-232-86.netcologne.de] has quit [Ping timeout: 265 seconds] 22:55:13 -!- Salamander [~Salamande@ppp118-210-157-21.lns20.adl6.internode.on.net] has quit [Ping timeout: 252 seconds] 22:56:44 -!- rbarraud_ [~rbarraud@118-93-169-153.dsl.dyn.ihug.co.nz] has quit [Remote host closed the connection] 22:57:02 rbarraud_ [~rbarraud@118-93-169-153.dsl.dyn.ihug.co.nz] has joined #lisp 23:01:19 deepfire: i wanted to find my favorite libraries in the waterfall, but it was very difficult due to the vertical name layout. 23:01:49 -!- Salamander_ [~Salamande@ppp118-210-250-167.lns20.adl6.internode.on.net] has quit [Ping timeout: 252 seconds] 23:04:20 Salamander_ [~Salamande@ppp118-210-134-73.lns20.adl6.internode.on.net] has joined #lisp 23:06:19 -!- legumbre [~leo@r190-135-73-190.dialup.adsl.anteldata.net.uy] has quit [Read error: Connection reset by peer] 23:06:35 legumbre [~leo@r190-135-73-190.dialup.adsl.anteldata.net.uy] has joined #lisp 23:06:42 -!- fiveop [~fiveop@erft-d932fa93.pool.mediaWays.net] has quit [Quit: humhum] 23:11:40 -!- Edward_ [ed@AAubervilliers-154-1-32-98.w90-3.abo.wanadoo.fr] has quit [Ping timeout: 240 seconds] 23:18:21 -!- galaxywatcher [~galaxywat@ppp-61-90-20-10.revip.asianet.co.th] has quit [Quit: galaxywatcher] 23:19:24 galaxywatcher [~galaxywat@ppp-61-90-20-10.revip.asianet.co.th] has joined #lisp 23:20:10 -!- _s1gma [~d.d.derp@77.107.164.131] has quit [Remote host closed the connection] 23:20:50 -!- ssideris [~sideris@93-97-178-232.zone5.bethere.co.uk] has quit [Ping timeout: 272 seconds] 23:22:02 ssideris [~sideris@93-97-178-232.zone5.bethere.co.uk] has joined #lisp 23:23:59 -!- Raynes [~macr0@unaffiliated/raynes] has quit [Ping timeout: 276 seconds] 23:25:29 Raynes [~macr0@unaffiliated/raynes] has joined #lisp 23:26:23 -!- galaxywatcher [~galaxywat@ppp-61-90-20-10.revip.asianet.co.th] has quit [Quit: galaxywatcher] 23:27:13 rvirding [~chatzilla@h19n4c1o1034.bredband.skanova.com] has joined #lisp 23:27:54 Salamander__ [~Salamande@ppp118-210-107-216.lns20.adl2.internode.on.net] has joined #lisp 23:28:10 -!- jan247 [~jan247@unaffiliated/jan247] has quit [Quit: jan247] 23:29:56 -!- aidalgol [~user@114-134-7-235.rurallink.co.nz] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 23:30:11 -!- gigamonkey [~user@c-98-248-194-46.hsd1.ca.comcast.net] has quit [Ping timeout: 272 seconds] 23:30:25 -!- Salamander_ [~Salamande@ppp118-210-134-73.lns20.adl6.internode.on.net] has quit [Ping timeout: 252 seconds] 23:30:58 -!- milanj [~milanj_@109-92-113-128.dynamic.isp.telekom.rs] has quit [Ping timeout: 245 seconds] 23:33:23 Salamander_ [~Salamande@ppp118-210-136-157.lns20.adl6.internode.on.net] has joined #lisp 23:33:44 Joreji [~thomas@77-066.eduroam.RWTH-Aachen.DE] has joined #lisp 23:35:55 -!- Salamander__ [~Salamande@ppp118-210-107-216.lns20.adl2.internode.on.net] has quit [Ping timeout: 252 seconds] 23:37:15 -!- Athas [~athas@shop3.diku.dk] has quit [Remote host closed the connection] 23:37:36 rich_holygoat [~rnewman@pdpc/supporter/student/rich-holygoat] has joined #lisp 23:37:42 Edward_ [ed@AAubervilliers-154-1-4-63.w86-212.abo.wanadoo.fr] has joined #lisp 23:39:37 Xach, acknowledged, Juanjo told me the same, but I was too enamored with this layout at the time. 23:40:10 *deepfire* ponders setting up KVM for buildslaves 23:40:14 -!- naiv [~naiv@ARennes-553-1-62-123.w81-48.abo.wanadoo.fr] has quit [Ping timeout: 264 seconds] 23:40:31 -!- ziarkaen [~ziarkaen@87.115.156.134] has quit [Quit: Leaving] 23:40:52 mbohun [~mbohun@202.124.72.127] has joined #lisp 23:41:49 So, my desire-related priorities, for now: 1. fixing short-term bitrot problems, 2. getting Juanjo satisfied, 3. port to ECL 23:42:32 udzinari` [~user@209.158.broadband13.iol.cz] has joined #lisp 23:44:01 -!- nus [~nus@unaffiliated/nus] has quit [Ping timeout: 260 seconds] 23:51:52 ysph [~user@24-181-93-165.dhcp.leds.al.charter.com] has joined #lisp 23:56:01 slyrus_ [~chatzilla@dsl092-019-253.sfo1.dsl.speakeasy.net] has joined #lisp