00:00:17 demmel [n=demmel@dslb-094-216-077-147.pools.arcor-ip.net] has joined #lisp 00:00:23 1.) Any arbitrary Lisp expression. 00:00:41 2.) I don't know why it would want access, it probably wouldn't, but on the off-chance it does, I want the user to be able to do so. 00:01:11 3.) I dunno, just exploring ideas I guess. 00:01:13 but, you know what will be in the lexical environment of your EVAL-MY-SQL-STRING function .. right? 00:01:58 That's why that function will be a macro that expands into code that a) evaluates the code in the current lexical environment and b) does the querying and such. 00:02:03 -!- syamajala [n=syamajal@97-95-190-124.dhcp.oxfr.ma.charter.com] has quit ["leaving"] 00:02:17 -!- abeaumont [n=abeaumon@47.Red-213-97-139.staticIP.rima-tde.net] has quit [Remote closed the connection] 00:02:43 why step a .. what is in this lexical environment that can't simply be reconstructed in the eval form? 00:03:56 Hold on a second. 00:04:59 gh7d395pi69wd [n=asdf@unaffiliated/gh7d395pi69wd] has joined #lisp 00:05:03 Dumb question time, how come this works: (let ((y 2)) (eval '(+ y 3))) 00:05:16 y is a lexical variable, but I thought eval ignored lexical variables? 00:05:24 blitz_ [n=blitz@2001:6f8:10f6:200:21b:77ff:fe41:11ab] has joined #lisp 00:05:48 Captain_Thunder: it shouldn't work, and if it does either your lisp is broken or you've declared y special 00:06:11 I get an unbound variable error for y when I try that in SBCL 00:06:12 Oh, duh, I declare short variables like that all the time when I'm playing around at the REPL :) 00:06:17 Sorry. 00:06:21 Captain_Thunder: don't do that... now you see why 00:06:28 Lesson learned. 00:06:31 Captain_Thunder: _always_ use *earmuffs* :) 00:06:35 heh 00:06:35 or use *y* syntax 00:06:53 Phoodus: *y* is a convention, not syntax 00:07:20 details :) 00:07:30 Anyway, the user may have some variables floating around in the lexical environment that I'm unaware of, and I don't know whether or not they'd want to use them in the string-code. 00:08:06 Captain_Thunder: how could there possibly be variables in the lexical environment that you are not aware of ... it's in the bloody program text surrounding your call to EVAL 00:08:25 that's what lexical variables are! 00:09:02 (let ((y 3)) (eval (read-from-string "(+ y 1)"))) <-- Is there any magic eval that will make this produce 4? 00:09:22 That's what I want my library to be able to do. 00:09:25 you're playing with fire in terms of packages, too 00:09:35 That's the user's problem :) 00:09:38 Captain_Thunder: (let ((y 3)) (eval `(let ((y ,y)) (read-from-string "(+ y 1)")))) 00:09:52 chessguy_work [n=chessguy@pool-96-255-111-135.washdc.fios.verizon.net] has joined #lisp 00:10:05 or explicitly declare a package solely for the eval environment, and declare your variables in there 00:10:13 To do that would require knowing that there's a variable called y, and my library code can't know what the user's variables are. 00:10:24 So I either need to examine the environment, or eval in it. 00:10:43 Captain_Thunder: how is the user going to create a LET form in the lexical environment surrounding your call to EVAL? 00:10:43 chessguy_ [n=chessguy@67-130-43-2.dia.static.qwest.net] has joined #lisp 00:11:03 how are the "user's variables" declared? 00:11:32 (let ((my-variable 3)) (captain-thunder's-magic-function "(+ my-variable 3)")) 00:11:38 -!- chessguy_ is now known as chessguy 00:11:41 As one possible scenario. 00:11:49 I mean, you've got defvar et al, which it should see as long as your packages are kosher, and let, which it obviously won't 00:12:03 that can't work 00:12:07 Captain_Thunder: Lisp in Small Pieces is one book that explains how that's not feasible. 00:12:10 Captain_Thunder: i don't understand .. i thought the user was supplying the string to be passed to your magic function... 00:12:12 unless it's a macro 00:12:31 drewc, they are, that's the users code I just typed. 00:12:53 Xach, which is why I asked about SBCL's non-portable extensions for evaling in a non-null lexical environment. 00:12:57 Captain_Thunder: why would the user jump through those hoops instead of writing what they mean? 00:13:26 -!- fvw [n=sdfpme@113.77.216.232] has quit ["leaving"] 00:13:33 hefner, kind of a long story. Hopefully I'll be able to post a version of this on common-lisp.net soon :) 00:13:46 It may only run on CLISP at the rate things are going though... 00:14:06 Captain_Thunder: It's a pretty good book. Some of these questions indicate you haven't really thought the process through. LiSP will help clarify things in your mind. 00:14:17 lol 00:14:28 I'd prefer things to remain murky. 00:14:33 Captain_Thunder: I'd say you should make a macro and return the read list as the body 00:14:36 -!- rouslan [n=Rouslan@unaffiliated/rouslan] has quit [Read error: 54 (Connection reset by peer)] 00:15:01 rouslan [n=Rouslan@unaffiliated/rouslan] has joined #lisp 00:15:08 that also means that the string must be a literal 00:15:12 -!- demmel_ [n=demmel@dslb-094-216-221-237.pools.arcor-ip.net] has quit [Read error: 110 (Connection timed out)] 00:15:26 I'd like to avoid that, too. 00:15:29 though you could play more eval tricks to get around that 00:15:57 can you make a shorter story from teh "kind of a long story"? 00:15:58 -!- dstatyvka [i=ejabberd@pepelaz.jabber.od.ua] has left #lisp 00:16:05 -!- Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has quit [] 00:16:06 There's probably a much easier way of doing this 00:16:20 (let ((foo 1)) (macrolet ((foo (&env env) (Arnesi::lexical-variables env)))) (foo)) 00:16:40 Why thank you. 00:16:40 the only reason I can imagine you'd want to do this is if you wanted to create the surrounding environment once and substitute different things into it for evaluation 00:17:00 That &env thing, is that standard syntax, or specific to some library/implementation? 00:17:04 I've seen it before. 00:17:28 If only there were some way to easily check. 00:18:24 clhs 3.4.4 00:18:24 http://www.lispworks.com/reference/HyperSpec/Body/03_dd.htm 00:18:54 Oh, heh :) 00:19:00 Xach: sssh, he prefers things remain murky. 00:19:11 Not *too* murky... 00:19:12 *p_l* decides that linux VM subsystem is crazy beyond understanding and stops looking at memory usage data 00:20:00 p_l: you really can't extract much useful information from it, can you? (to be fair, with all the sharing between processes, I don't even know what to count) 00:22:57 p_l: what were you searching for? 00:23:19 -!- Pepe_ [n=ppjet@78.113.3.173] has quit [Remote closed the connection] 00:23:41 madnificent: I'm keeping watch so that OOM Killer won't kill my work 00:23:58 2G core + 2G swap is not enough for Firefox 00:24:07 LoL 00:24:38 though my problems had gone down noticeably after switching to 3.1b3-pgo 00:25:45 Oh God the full lexical environment printout it burns my eyes O_O 00:26:02 In a happy way :) 00:26:11 still, not so long ago, I had ~900 MB of memory used and ~400MB swap (that's only resident and shared, excluding cache). I ran "swapoff -a && swapon -a" and left to make tea... 00:26:15 p_l: does 3.5 boast better memory consumption too? (besides the speed increase) 00:26:31 when I returned, I had <800MB memory used and no swap usage 00:26:49 p_l: and many many processes killed+ 00:26:55 s/\+/?/ 00:27:05 madnificent: no, before running that combo I always check memory status 00:27:30 Ppjet6 [n=ppjet@78.113.3.173] has joined #lisp 00:27:32 borism_ [n=boris@195-50-199-19-dsl.krw.estpak.ee] has joined #lisp 00:27:51 still, at least now I don't have any app in swap 00:28:14 before, moving in mopintro.pdf and even workspace switching was slow 00:28:44 linux is a gamble wrt. that stuff .. i just save often now .. autosave or whatever in emacs 00:30:11 -!- athos [n=philipp@92.250.250.68] has quit ["leaving"] 00:30:50 -!- Ginei_Morioka [i=irssi_lo@78.112.63.135] has quit [Nick collision from services.] 00:30:57 Ginei_Morioka [i=irssi_lo@78.112.63.135] has joined #lisp 00:32:17 -!- rouslan [n=Rouslan@unaffiliated/rouslan] has quit [Read error: 110 (Connection timed out)] 00:32:50 -!- borism [n=boris@195-50-200-218-dsl.krw.estpak.ee] has quit [Read error: 145 (Connection timed out)] 00:34:59 -!- chessguy_work [n=chessguy@pool-96-255-111-135.washdc.fios.verizon.net] has quit [Success] 00:35:45 -!- chris2 [n=chris@dslb-094-216-195-178.pools.arcor-ip.net] has quit ["Leaving"] 00:37:41 -!- fe[nl]ix [n=algidus@88-149-210-139.dynamic.ngi.it] has quit ["Valete!"] 00:38:48 fe[nl]ix [n=algidus@88-149-210-139.dynamic.ngi.it] has joined #lisp 00:41:01 I have an idea to solve DDOSing 00:41:31 But I'm afraid that I won't be able to express it precisely and people will misunderstand, so bear with me 00:41:32 is it color-related? 00:43:31 1=yes, 0=no. Make it so that the site only serves other locations to visit. The visitors who pick yes (Do you want to enter the site/location) are monitored. The slower visitors are regarded as normal, the faster as DDOS-bots. The more bots you use so that you can flood my system by having a bunch of different IP/unique visitors, the more you DDOS yourself. 00:44:05 It would not make any sense for a gigantic boost in traffic to occur independent of some gigantic event (MJ's death for instance) 00:44:33 A flood of unique new visitors at a speed that far exceeds average traffic would be suspicious, no? 00:44:50 how do you end up ddosing yourself 00:45:04 Where will you get that many bots? 00:45:30 -!- ikki [n=ikki@201.155.75.146] has quit [Remote closed the connection] 00:45:38 It becomes a game of who has more unique locations (Me, to serve information... You, to deny my service/ability to serve information) 00:45:39 from the 99% clueless Windows users out there? 00:45:57 I think I do misunderstand. 00:46:05 -!- Yuuhi [i=benni@84.131.183.34] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 00:46:27 Here's my point: How many times do you say "Yes" to enter a site? 00:46:30 What I got from that is that you defend a site by putting up a binary captcha. 00:46:34 -!- wubbster [n=wubb@doc-24-206-235-153.kw.tx.cebridge.net] has quit [Remote closed the connection] 00:46:46 If you say "Yes" multiple times too quickly, I will take that as a "No" 00:46:54 hmm 00:46:57 Since no human user is going to say "Yes" 500 times in a millisecond 00:47:03 That eliminates the individual capability. 00:47:04 niche: I'd say that such a site is doomed 00:47:08 Wait 00:47:12 We haven't gotten to botnets. 00:47:18 Just becomes a problem of botnet size 00:47:19 yeah 00:47:22 sorry, continue 00:47:23 Just like I said 00:47:31 That's all it will devolve to 00:47:41 So, I already have regular visitors (whitelisted) 00:47:54 All these new visitors are spam visitors until they prove otherwise (blacklisted) 00:48:00 It's not that I'm blocking you, but slowing you down 00:48:04 Anyone who moves to fast... 00:48:05 BAN THEM 00:48:19 A unique visitor can only make so many moves before they are banned and not given new locations 00:48:39 It's as if I'm giving hashes to real people in the form of pointing them to a secret, or generated location 00:48:47 fvw [n=sdfpme@113.77.216.232] has joined #lisp 00:49:00 fisxoj [n=fisxoj@cpe-24-59-253-23.twcny.res.rr.com] has joined #lisp 00:49:13 It makes sense. Because the point of denying service is to prevent me from getting information to actual visitors 00:49:18 If you don't know where my actual visitors are 00:49:21 And you can't get there 00:49:24 You can't deny my service 00:49:25 Simple 00:49:42 niche: I'd say this is a doomed scheme... 00:49:44 Adlai`: I see you. 00:49:46 Even if you do... if you move too fast (trying to outdo my processor speed or bandwidth) I will keep you from finding the new location 00:50:02 I have a hard time seeing why, but not in arrogance, in ignorance... 00:50:20 brnhck [n=hrk@acurwa002169.adsl.ppp.infoweb.ne.jp] has joined #lisp 00:50:24 sykopomp: I do not see you, but hello. 00:50:28 The way I see it, it's solved, but I could be completely wrong with my intuition 00:50:47 -!- brnhck [n=hrk@acurwa002169.adsl.ppp.infoweb.ne.jp] has left #lisp 00:50:58 :P 00:51:09 All I'm trying to express is that if you could slow down the speed at which your attackers send you junk, then they would not be able to DDOS you 00:51:27 Yes. 00:51:29 QinGW [n=wangqing@203.86.81.2] has joined #lisp 00:51:42 syamajala [n=syamajal@97-95-190-124.dhcp.oxfr.ma.charter.com] has joined #lisp 00:51:49 Pointing actual whitelisted visitors to unique locations where actual information is, with fresh and non-DDOS'd resources is like showing them the light 00:51:58 This is a pretty common idea. 00:52:03 Well, that's good 00:52:07 Hiding a server farm behind some kind of gatekeeper. 00:52:08 -!- Captain_Thunder [n=jimmymil@c-24-35-95-148.customer.broadstripe.net] has quit [] 00:52:12 Yes... a dummy server 00:52:14 niche: Such methods exist already, with combined IDS/IPS systems and "blackholes". If you're crazy, you can even order those for counterattack 00:52:25 I'm just worried that your IP ban lists will be vulnerable 00:52:32 and your gatekeeper will become the target 00:52:35 Hmm. Is this a novice idea or a somewhat more intermediate idea? 00:52:42 I'm a complete novice 00:52:43 words without meaning 00:52:48 you're looking for praise 00:53:09 I have a feeling that all lispers are connected a human-machine interface 00:53:17 As if you've all programmed the computer to integrate into your mind 00:53:24 You all have computer-like memories and logic 00:53:39 Exactly what I'm trying to achieve 00:53:46 Most IDS I know about use statistics more heavily than blacklists or whitelists 00:54:02 How many of you are talking into a microphone to speak to me with near perfect spelling? 00:54:12 niche: I'm not infosec specialist, but that scheme looks impractical to me, even detrimental to website in the end 00:54:17 could be wrong, but most of the time, if you're keeping around a list the attacker can control, he'll break you 00:54:29 Yes, but it's all a game of cat and mouse, is it not? 00:54:39 If you can make a protection scheme, someone can break it... 00:54:41 somehow 00:54:42 doesn't have to be 00:54:44 well 00:54:51 At what point does the mouse win? 00:54:52 you don't want to make it easy for the attacker :P 00:55:12 Security by confusion? 00:55:19 is a terrible idea 00:55:28 but I like your style 00:55:32 It's advanced obscurity (reddit) 00:56:04 I don't think I'll ever be a Schneier, so I'm not going to pretend I'm saying anything profound 00:56:26 I really think it's a problem of who's more intelligent... it's a game, rather than a problem 00:56:33 security through obscurity is bad, however, keeping away certain information is basic practice 00:56:33 ^nonsensical 00:56:36 paradoxical 00:56:39 i think it's just brute force in the end; if the attacker is "bigger" than you or your protector ("gateway") i don't think there is any way to protect yourself from ddosing at all 00:56:47 What if we could create a paradoxical system? 00:57:07 Hmmm.... I want to develop my ideas... investigate my own questions 00:57:44 I don't like the idea that we can *never* protect from DDOS'ing. It feels so defeatist. 00:58:03 If the entire internet was really interconnected, then I could just distribute the DDOS to a bunch of sites 00:58:14 chessguy_work [n=chessguy@pool-96-255-111-135.washdc.fios.verizon.net] has joined #lisp 00:58:24 If IP addresses weren't sites, but bits and pieces of multiple sites, then this would be easy 00:58:43 -!- blackened` [n=blackene@89.102.28.224] has quit [] 00:58:43 doesn't make sense to imagine a total "defeat" of dos attacks 00:58:50 We should fragment more... it sounds outrageous now, but each IP address could be like a collection of a segmented rar files 00:58:54 the class of attacks isn't bounded by anything really 00:58:59 -!- ASau [n=user@193.138.70.52] has quit [Read error: 110 (Connection timed out)] 00:58:59 Instead of one rar file for one site 00:59:02 you mean tcp over bittorrent? 00:59:10 Yes 00:59:13 (over udp, I imagine) 00:59:16 ... 00:59:18 Could you DDOS a bittorrent? 00:59:20 hes taking the mick 00:59:25 only conclusion 00:59:39 Who are you? How are you so insightful? 00:59:41 *p_l* backs off 00:59:46 How are you able to...? 01:00:38 -!- chessguy [n=chessguy@67-130-43-2.dia.static.qwest.net] has quit [Nick collision from services.] 01:00:52 -!- chessguy_work is now known as chessgiuy 01:01:01 have I walked into a particularly bad anime? 01:01:05 -!- chessgiuy is now known as chessguy 01:01:09 Does anyone else not notice that gh7d395pi69wd is reading me like an open book with absolutely no problem? Are you guys not seeing that? 01:02:12 hefner: I think you walked into particularly bad western hacker movie 01:02:57 I want to pm gh7d395pi69wd but I don't know if that would be a good idea or a terrible mistake 01:03:01 more like a bad hacker romantic comedy 01:03:21 Yes, I have to admit fascination with someone who can see right through me 01:03:31 slyrus_ sounds like the story of my life 01:03:56 I want to learn... not love 01:04:09 The meaning of life is to understand 01:04:14 Love is just a program 01:04:46 I would like a mentor... someone with the discipline to take another person under their wing 01:05:00 *lnostdal* is apparently in need of more drugs 01:06:08 slyrus_: does it even have "romantic" part? You have it good... 01:06:41 *drewc* stepping into something very weird here. 01:06:57 i struggle to see the connection to lisp.... 01:07:10 -!- fisxoj [n=fisxoj@cpe-24-59-253-23.twcny.res.rr.com] has quit ["Ex-Chat"] 01:09:16 drewc: me too 01:10:27 -!- chessguy [n=chessguy@pool-96-255-111-135.washdc.fios.verizon.net] has quit ["Leaving"] 01:11:11 i think i agree with lnostdal .... drugs might help to gain some insight here.... good ones at that .... 01:12:00 just stay away from codeine, the only thing I got from it was lack of taste and being in limbo for a week 01:12:02 even you guys are talking about ddos 01:12:26 I was reading about SCTP earlier this morning, I was wondering if that might be something of a solution 01:12:46 though I just heard about it, and haven't had much chance to read about it yet 01:13:13 -!- Jasko2 [n=tjasko@75-149-33-105-SFBA.hfc.comcastbusiness.net] has quit ["Leaving"] 01:13:24 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [] 01:15:03 -!- amnesiac [n=amnesiac@p3m/member/Amnesiac] has quit ["Leaving"] 01:16:28 ... please, no... 01:16:29 -!- syamajala [n=syamajal@97-95-190-124.dhcp.oxfr.ma.charter.com] has quit ["leaving"] 01:16:51 -!- ski [n=slj@c-e113e055.1149-1-64736c10.cust.bredbandsbolaget.se] has quit [Read error: 110 (Connection timed out)] 01:18:48 it's like TCP I guess, except it uses a 4-way handshake, so sending a fake source ip doesn't work nearly as well 01:19:41 Jasko [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has joined #lisp 01:19:43 and supports switching IPs mid-connection, IIRC 01:22:00 fvw_ [n=sdfpme@113.77.216.106] has joined #lisp 01:22:03 -!- jyujin [n=mdeining@82.113.121.139] has quit [] 01:25:55 BTW, does anyone here have experience with clsql/cl-perec ORMs with regards to DB views and stored procedures? 01:26:27 -!- lambda-avenger [n=roman@adsl-63-197-150-112.dsl.snfc21.pacbell.net] has left #lisp 01:29:22 -!- demmel [n=demmel@dslb-094-216-077-147.pools.arcor-ip.net] has quit [] 01:31:24 -!- dys`` [n=andreas@p5B3155CF.dip.t-dialin.net] has quit [Connection timed out] 01:32:27 -!- niche [n=haf@ool-18bd3ac7.dyn.optonline.net] has quit [] 01:33:52 -!- fvw [n=sdfpme@113.77.216.232] has quit [Read error: 110 (Connection timed out)] 01:34:35 -!- mrsolo [n=mrsolo@nat/yahoo/x-c569055b570f849b] has quit ["Leaving"] 01:36:17 -!- slyrus_ [n=slyrus@dsl092-019-253.sfo1.dsl.speakeasy.net] has quit [Read error: 110 (Connection timed out)] 01:37:12 ikki [n=ikki@201.155.75.146] has joined #lisp 01:39:30 demmel [n=demmel@dslb-094-216-077-147.pools.arcor-ip.net] has joined #lisp 01:40:42 -!- ikki [n=ikki@201.155.75.146] has quit [Client Quit] 01:43:33 -!- fvw_ is now known as fvw 01:45:10 -!- JAS415 [n=jon@ip24-250-13-137.ri.ri.cox.net] has quit ["Leaving."] 01:51:33 wubbster [n=wubb@doc-24-206-235-153.kw.tx.cebridge.net] has joined #lisp 01:51:38 dalton [n=user7994@187.35.196.144] has joined #lisp 01:58:26 dys`` [n=andreas@p5B316D1F.dip.t-dialin.net] has joined #lisp 01:59:27 -!- xan [n=xan@29.Red-83-59-49.dynamicIP.rima-tde.net] has quit [Read error: 113 (No route to host)] 02:01:10 -!- dialtone [n=dialtone@unaffiliated/dialtone] has quit [Read error: 110 (Connection timed out)] 02:03:26 -!- fvw [n=sdfpme@113.77.216.106] has quit [Remote closed the connection] 02:06:24 -!- demmel [n=demmel@dslb-094-216-077-147.pools.arcor-ip.net] has quit [] 02:06:24 ace4016 [i=ace4016@cpe-76-168-248-118.socal.res.rr.com] has joined #lisp 02:07:18 nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 03:27:31 ccl-logbot [n=ccl-logb@master.clozure.com] has joined #lisp 03:27:31 03:27:31 -!- names: ccl-logbot JuanDaugherty foobar__ ace4016 fvw serichse1 hawkbill rtoym_ jlf ahaas xinming_ dys``` stassats nvoorhies younder dialtone ausente wubbster Jasko QinGW fe[nl]ix Ginei_Morioka borism_ Ppjet6 gh7d395pi69wd ltriant Adrinael NNshag kpreid envi^home X-Scale2 mathrick mbishop Pegazus stepnem rread lispm CrazyEddy lemoinem foom replaca__ pjb aja dto pstickne legumbre_ rotty benny lnostdal bgs100 gonzojive commmmodo billstclair sbahra 03:27:31 -!- names: Phoodus joshe kmcorbett1 Drakeson ajhager ryepup Quadrescence synic schwinn434 yango lisppaste marcoecc Adlai daniel__ bobf ia hyperboreean sytse pem ianmcorvidae Adlai` jrockway slyrus maxote beach eno rtoym cmm s0ber_ mkfort Summersault a-s` mikezor joast Taggnostr kahmalo p_l free_thinker vy tic spiaggia srcerer boyscared kleppari Riastradh koft fullets1 alexbobp tsuru` sellout wlr bdowning deepfire lichtblau dalkvist Modius nowhereman 03:27:31 -!- names: Ringo48 rapacity peddie Patzy djinni` antifuchs CuriousTrain ``Erik [df]_ froydnj proq Swordsman kefka matimago Buganini schme KingNato rey_ keithr mgr xristos Gertm kidd kuhzoo Soulman__ dmiles_afk erg Xach nicktastic tessier arbscht_ kei nasloc__ ecraven piso housel drewc A_anekos enn rbancroft michaelw specbot kuwabara p8m jsnell guaqua smoofra araujo Guest49718 JeLuF dostoyevsky herbieB sad0ur rsynnott Bucciarati authentic Qsource spacebat_ 03:27:31 -!- names: PissedNumlock hefner ramus` Ralith meingbg bob_f phadthai pragma_ mtd Khisanth ineiros AntiSpamMeta madnificent Zhivago trebor_dki easyE erk cp2 sykopomp Draggor leo2007 _3b` johs cods vsync mornfall joga cYmen minion z0d felipe sciendan Kirklander r0bby fnordus yahooooo Xof azanar luis ilitirit tarbo egn zbigniew vcgomes tcoppi acieroid Chris Fade sepisultrum cavelife^ bohanlon sjbach thijso clog retupmoca djkthx galdor gz tvaalen Axioplase 03:27:31 -!- names: DrForr scode Orest^bnc dfox azuk` _3b Bootvis l_a_m jkantz Borbus Aisling dcrawford bfein rlonstein guenthr noptys krappie pok chii 03:32:28 -!- legumbre_ [n=user@r190-135-61-58.dialup.adsl.anteldata.net.uy] has quit [Read error: 54 (Connection reset by peer)] 03:32:41 -!- schwinn434 [n=schwinn4@cpe-75-81-202-25.we.res.rr.com] has quit [Remote closed the connection] 03:34:12 -!- foobar__ [n=_prip@host66-122-dynamic.32-79-r.retail.telecomitalia.it] has quit [Read error: 110 (Connection timed out)] 03:34:26 foobar__ [n=_prip@host137-194-dynamic.17-79-r.retail.telecomitalia.it] has joined #lisp 03:35:03 legumbre [n=user@r190-135-34-181.dialup.adsl.anteldata.net.uy] has joined #lisp 03:40:15 ^authentic [n=authenti@85-127-20-145.dynamic.xdsl-line.inode.at] has joined #lisp 03:44:11 -!- authentic [n=authenti@unaffiliated/authentic] has quit [Connection timed out] 03:49:01 QinGW1 [n=wangqing@203.86.81.2] has joined #lisp 03:49:39 -!- nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has quit [] 03:54:21 prip_ [n=_prip@host99-195-dynamic.17-79-r.retail.telecomitalia.it] has joined #lisp 03:54:32 -!- foobar__ [n=_prip@host137-194-dynamic.17-79-r.retail.telecomitalia.it] has quit [Read error: 110 (Connection timed out)] 04:07:14 -!- dialtone [n=dialtone@unaffiliated/dialtone] has quit ["leaving"] 04:12:17 SandGorgon [n=OmNomNom@122.162.51.25] has joined #lisp 04:12:54 -!- prip_ [n=_prip@host99-195-dynamic.17-79-r.retail.telecomitalia.it] has quit [Read error: 110 (Connection timed out)] 04:13:07 prip_ [n=_prip@host19-123-dynamic.32-79-r.retail.telecomitalia.it] has joined #lisp 04:15:15 slyrus_ [n=slyrus@adsl-76-241-19-133.dsl.pltn13.sbcglobal.net] has joined #lisp 04:17:23 -!- QinGW [n=wangqing@203.86.81.2] has quit [Success] 04:23:49 -!- bgs100 [n=ian@unaffiliated/bgs100] has quit [] 04:23:51 -!- araujo [n=araujo@gentoo/developer/araujo] has quit [Read error: 113 (No route to host)] 04:26:42 good morning 04:26:47 *p_l* starts to see clearly why Lispers so often exhibit cases of NIH 04:26:59 p_l: why is that? 04:27:13 beach: it's way too easy to roll your own 04:27:20 hmm 04:27:27 and btw, morning 04:27:42 heh 04:28:13 I'm plagued by thoughts of making yet-another-orm 04:28:50 NIH? 04:29:17 Not Invented Here. 04:29:29 lexa_ [n=lexa_@seonet.ru] has joined #lisp 04:29:37 minion: What does NIH stand for? 04:29:38 Neofetal Intercostohumeral Hemoculture 04:29:45 ... 04:29:46 splittist [n=dmurray@63-55.5-85.cust.bluewin.ch] has joined #lisp 04:29:47 morning 04:29:49 hello splittist 04:29:54 hi splittist 04:29:57 -!- lexa_ is now known as Guest11225 04:31:14 in my case, it's even worse now that I'm reading mopintro... 04:31:45 it starts with databases as example :/ 04:32:27 rread_ [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has joined #lisp 04:34:12 mrsolo [n=mrsolo@adsl-68-126-215-139.dsl.pltn13.pacbell.net] has joined #lisp 04:35:25 Demosthenes [n=demo@173.86.192.171] has joined #lisp 04:41:52 -!- Guest11225 [n=lexa_@seonet.ru] has left #lisp 04:43:31 p_l: before you do another ORM, please have a look at my ROFL library ... it's a good ORM, it's mature and in production, etc. 04:43:59 (and i had a lot of experience with the MOP _before_ i wrote it.... this is important :)) 04:44:11 -!- Khisanth [n=Khisanth@pool-141-157-237-196.ny325.east.verizon.net] has quit [Read error: 110 (Connection timed out)] 04:44:40 drewc: I wasn't planning on starting completely from scratch anyway :) 04:45:35 p_l: it's 04:45:49 'relational-objects-for-lisp' in clbuild if you want to check it out. 04:46:04 drewc: I believe I have it on disk, actually :) 04:46:15 p_l: there is also a small tutorial available ... 04:46:24 lambda-avenger [n=roman@adsl-63-197-150-112.dsl.snfc21.pacbell.net] has joined #lisp 04:46:30 darcs get http://www.deepsky.com/~fade/roflcopter/ 04:46:45 and I'm reading on MOP anyway, just in case I have to augment whatever library I end up using 04:47:07 indeed .. learn your MOP! great stuff in there :) 04:47:09 how is darcs doing these days 04:47:34 darcs is dead enough that i'm switching to git when i get the chance. 04:47:48 smoofra: the new format seems to be much faster than the old one, but git seems to be winning :) 04:47:50 ack drewc you have something called LoL. i knew i shouldn't have named my firefox addon that :/ 04:48:22 smoofra: you can always rename it .. i was here first! :P 04:49:30 -!- rread [n=rread@nat/sun/x-ea76ac5dc7868a2e] has quit [Read error: 110 (Connection timed out)] 04:50:11 drewc: I see you have much nicer referencing than clsql:d-v-c :) 04:50:31 p_l: CLSQL is teh suck 04:50:50 i hate the syntax, hate the ORM, and hate digging through the code. 04:51:14 that said, CLSQL is complete and works with most major DB engines. ROFL is postgres only atm. 04:51:47 well, I'm going with Postgres for main storage anyway 04:52:12 p_l: good... it is the best IMO :) 04:52:21 what I'm worried about is how hard it would be to add support for additional field types etc. 04:52:26 trivial 04:53:03 as well as making special support for building SELECT etc. slightly different (for PostGIS) 04:53:12 so sad, darcs was abosolutely the best SCM availabe for a little while 04:53:51 well, actually it depends really.... rofl makes no effort to ensure the right types of slots etc, and basically relies on postmodern to translate from lisp objects to DB types. adding types to postmodern is easy enough. 04:54:12 p_l: what do you mean by 'building select etc' etc? 04:54:39 khisanth_ [n=Khisanth@pool-141-157-237-196.ny325.east.verizon.net] has joined #lisp 04:55:02 drewc: for geographical indexes 04:55:17 btw, (requests :referenced-from request :on application-user-id :accessor requests) <--- made my day. 04:56:48 I'll be also looking forward for a method to do caching, I guess :) 04:57:27 yeah, there is some caching support... i forget where... 04:58:33 -!- prip_ [n=_prip@host19-123-dynamic.32-79-r.retail.telecomitalia.it] has quit [Read error: 110 (Connection timed out)] 04:59:06 prip_ [n=_prip@host16-197-dynamic.17-79-r.retail.telecomitalia.it] has joined #lisp 04:59:20 well, I suspect PostgreSQL is smart enough for views to speed up some queries 04:59:31 p_l: ahh .. there it is. FUNCALL-WITH-FIND-OBJECT-CACHE is the ticket 04:59:53 drafael [n=tapio@ip-118-90-137-9.xdsl.xnet.co.nz] has joined #lisp 05:01:56 how to cut the first 256 rows of a 2D array to form a new array? 05:02:48 araujo [n=araujo@gentoo/developer/araujo] has joined #lisp 05:03:39 leo2007: subseq ? 05:03:49 drafael: on 2d array? 05:04:29 stassats: depends how the 2d array is done doesn't it? - could be that every row is an element in a 1d array 05:04:49 no, 2d array is 2d array 05:07:11 i guess (adjust-array (make-array '(256 other-dim) :displaced-to array) '(256 other-dim)) 05:08:32 -!- SandGorgon [n=OmNomNom@122.162.51.25] has quit [Success] 05:09:06 stassats: that doesn't form a 'new array' though... it just indirects... not sure what the OP wanted of course. 05:09:26 only what he asked, which makes little sense to me :) 05:09:27 drewc: adjust creates new contents 05:09:38 oh .. you adjusted.. my bad 05:09:49 i read make-array for some reason... getting tired i guess 05:09:59 leo2007: it's probably overkill for what you need, but my CLEM matrix stuff has operations for taking subsets of matrices and the like 05:10:52 -!- ahaas [n=ahaas@neptune.pettomato.net] has quit ["leaving"] 05:10:53 dialtone [n=dialtone@c-69-181-124-148.hsd1.ca.comcast.net] has joined #lisp 05:13:06 -!- khisanth_ is now known as Khisanth 05:18:16 slyrus_: thanks, it is overkill. I use loop instead 05:19:38 well, when you decide you want a relatively fast discrete convolution, or affine transformation, you'll come crawling back... :) 05:21:06 -!- fvw [n=sdfpme@113.77.216.106] has quit [Read error: 110 (Connection timed out)] 05:21:19 drewc: if I have an object from database in ROFL, it doesn't get saved till I call update-object, right? 05:21:53 -!- NNshag [i=user@Mix-Orleans-106-4-184.w193-248.abo.wanadoo.fr] has quit [Remote closed the connection] 05:23:02 *p_l* might just add ROFL support to Weblocks 05:23:10 p_l: correct. The alternative ('live' objects) doesn't work well with how i use the objects in lisp-on-lines 05:23:31 p_l: you'd be better off just using lisp-on-lines and UCW ;) 05:23:47 drewc: I'm experimenting at the moment :) 05:24:33 willb [n=wibenton@h69-129-204-3.mdsnwi.broadband.dynamic.tds.net] has joined #lisp 05:24:45 ROFL was designed to work with LoL, so that's really the path of least resistance. Having said that, LoL is a bit more complex than ROFL with a steeper learning curve. 05:26:40 *drewc* is writing proper docs for LoL right now 05:27:09 well, i'm finished for the day and watching torchwood, but i _did_ work on the docs today. 05:29:20 -!- ltriant [n=ltriant@lithium.mailguard.com.au] has quit [Read error: 104 (Connection reset by peer)] 05:29:34 ltriant [n=ltriant@lithium.mailguard.com.au] has joined #lisp 05:35:47 jmbr_ [n=jmbr@128.33.220.87.dynamic.jazztel.es] has joined #lisp 05:40:40 SandGorgon [n=OmNomNom@122.162.51.25] has joined #lisp 05:44:20 dcorking [n=dcorking@82.152.248.212] has joined #lisp 05:45:00 tltstc [n=tltstc@cpe-76-90-92-154.socal.res.rr.com] has joined #lisp 05:52:28 fvw [n=sdfpme@113.77.216.106] has joined #lisp 05:52:45 saikat_ [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has joined #lisp 05:55:47 -!- yahooooo [n=yahooooo@c-76-104-183-185.hsd1.wa.comcast.net] has quit [Read error: 110 (Connection timed out)] 05:56:17 -!- jmbr_ is now known as jmbr 05:59:54 -!- dcorking [n=dcorking@82.152.248.212] has quit ["Client Exiting"] 06:00:17 Holcxjo [n=holly@ronaldann.demon.co.uk] has joined #lisp 06:02:36 -!- Pegazus [n=awefawe@host250.190-224-109.telecom.net.ar] has left #lisp 06:03:43 Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has joined #lisp 06:04:58 -!- younder [n=jthing@165.244.251.212.customer.cdi.no] has left #lisp 06:05:34 Muld [i=wr23@88-196-38-67-dsl.noe.estpak.ee] has joined #lisp 06:07:51 -!- dys``` is now known as dys 06:08:13 yahooooo [n=yahooooo@c-76-104-183-185.hsd1.wa.comcast.net] has joined #lisp 06:11:22 ahaas [n=ahaas@neptune.pettomato.net] has joined #lisp 06:12:45 -!- stassats [n=stassats@wikipedia/stassats] has quit [Read error: 113 (No route to host)] 06:19:05 -!- jmbr [n=jmbr@128.33.220.87.dynamic.jazztel.es] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 06:21:05 Davse_Bamse [n=davse@4306ds4-soeb.0.fullrate.dk] has joined #lisp 06:23:08 -!- jlf [n=user@unaffiliated/jlf] has quit [Read error: 110 (Connection timed out)] 06:25:15 not lisp, but i like web+lisp -- and this is interesting wrt. that; http://www.reddit.com/r/programming/comments/8z6f7/google_officially_announces_chrome_os/ .. whoha 06:26:28 I hope they won't go the braindamaged way for system "security", like PolicyKit 06:26:49 *lnostdal* whacks policykit with a stick 06:27:29 oh, god. I didn't even know about PolicyKit. that's horrifying. I *knew* the freedesktop clowns were trying to take over my computer! 06:28:39 I don't run D-Bus on systemwide level, so it doesn't apply to me, but I consider PolicyKit an abomination 06:31:10 luckily i'm a nihilist when it comes to my own operating system. ...as long as it doesn't really bug me 06:32:40 Soulman [n=kvirc@154.80-202-254.nextgentel.com] has joined #lisp 06:34:00 I have difficulty reconciling my feeling that unix is a trainwreck with my intense distrust (okay, hatred) of these people regardless of the technical merits of their work. 06:34:10 blandest [n=blandest@softhouse.is.ew.ro] has joined #lisp 06:34:53 text-file based configuration with crummy cli based applications as front-end manipulators of said text-files does not appeal 06:34:59 I would have thought that distrust and trainwrecks were easily reconciled ... 06:35:06 oh, and the text-file formats are totally random ofc. 06:36:11 alinp [n=alinp@86.122.9.2] has joined #lisp 06:37:06 maybe I just resent them obsoleting whatever accumulated knowledge I have about how to administer my linux system :) 06:37:55 the entire all-is-c and ffi thing is a mess too; i'm sure you've noticed, hefner :) .. on that note i wonder how remote ffi'ing (even on localhost..) by using dbus (+gobject or qt's dbus api) or something would work out .. there is at least one (incomplete) dbus implementation for cl out there 06:38:19 What unix and C get right is that representing data as arbitrary byte sequences allows you to handle data that you don't understand. What they do wrong is to take advantage of this to be lazy and use really bad data formats that no-one else is expected to understand ... 06:38:25 -!- X-Scale2 is now known as TR2N 06:39:57 However, Unix started out right with having single authentication/security system and simple namespace covered by it. Now, not only we have sockets (which by themselves don't give too many problems, but don't fit semantics), we have weirdos like D-Bus where we need special, separate and not really integrated security systems 06:40:12 -!- vy [n=user@nbvyazici.cs.bilkent.edu.tr] has quit [Read error: 104 (Connection reset by peer)] 06:40:15 vy [n=user@nbvyazici.cs.bilkent.edu.tr] has joined #lisp 06:40:32 *hefner* forgot what problem d-bus solves 06:41:08 Jarvellis [n=jarv@dsl-217-155-101-22.zen.co.uk] has joined #lisp 06:41:09 hefner: in beginning it was supposed to be a common inter-component/application bus for desktop (Desktop-Bus) 06:41:38 as in "use one, single IPC system in place of two different" 06:42:00 and it certainly looks simpler than CORBA 06:43:23 I guess I can stretch my imagination and imagine a use for that. It doesn't seem like anything on my desktop communicates with anything else in a useful way, but hey, maybe that's because I don't run d-bus. =p 06:43:41 DCOP didn't look so bad though, and had quite simple usecase (even if having similar problem as windows regarding messages) 06:44:05 though I guess it's easier to avoid said problems in DCOP than in Win32 messaging model 06:44:16 oh, no, I take that back. I do run dbus. Naughty debian. 06:45:13 one example is inserting an usb-stick, inserting a cd .. etc. .. causes "stuff on the desktop" to pop up etc., hefner so dbus hooks low-level linux event stuff up with desktop level stuff 06:45:52 or network related events .. "connected, not connected .. new connection" .. or "new hardware detected" 06:45:53 lnostdal: except those actions can be done without D-Bus (actually, HAL duplicates much stuff that is done on lower levels) 06:46:30 lnostdal: yeah, well, I hate that crap. I don't run any file manager, and anyway using dbus for that sounds like you're just inventing the need for a running daemon to generate the notifications instead of monitoring the OS events directly. 06:47:00 hefner: in the beginning it might have looked fine, but udev supports 90% of said functionality 06:47:17 the general idea seems sound to me, hefner 06:47:18 I don't think most applications have any good reason to care when the network is connected/disconnected either. In windows, that's usually a misfeature. 06:47:31 .._some_ indirection is useful here i think 06:48:19 lnostdal: yes, but why does such a simple thing requires XML on the way? 06:48:32 oh, plugging in a photo camera is another example .. that's nice (for "normal" users at least :)) 06:49:07 isn't that the same as inserting a usb stick? anyway, it should be the job of the application that intends to do something with it. 06:49:33 lnostdal: yes, but it still doesn't require a daemon written in python keeping a freaky XML database, with another daemon playing security which is oblivious to kernel 06:49:58 it's just over engineered and brittle to start breaking things into servers and talking to them over sockets when you don't need to. 06:50:12 i don't know enough about dbus vs. hal vs. udev etc. to say anything about that, p_l 06:50:44 lnostdal: HAL, PolicyKit and ConsoleKit (and upcoming DeviceKit) are all based around D-Bus 06:51:05 *hefner* wonders why 83132 disappeared from lisppaste 06:51:07 lnostdal: for obvious reasons, D-Bus doesn't fit with OS security model 06:51:22 (it's all done over standard unix domain socket) 06:51:42 jdz [n=jdz@85.254.193.149] has joined #lisp 06:51:56 udev is a simple system of rules executed on events from kernel, mostly used for managing /dev tree 06:53:37 ok 06:53:49 -!- blandest [n=blandest@softhouse.is.ew.ro] has quit [Read error: 110 (Connection timed out)] 06:55:05 -!- beach [n=user@ABordeaux-158-1-36-70.w90-55.abo.wanadoo.fr] has quit [Read error: 110 (Connection timed out)] 06:55:55 however, nothing stops you from making use of that system to implement more. No need for frakking XML database (I found writing HAL's FDI files much harder than udev rules) 07:04:06 -!- lambda-avenger [n=roman@adsl-63-197-150-112.dsl.snfc21.pacbell.net] has left #lisp 07:04:13 mvilleneuve [n=mvillene@ABordeaux-253-1-86-153.w81-250.abo.wanadoo.fr] has joined #lisp 07:04:49 good morning 07:05:06 -!- drafael [n=tapio@ip-118-90-137-9.xdsl.xnet.co.nz] has quit [Read error: 110 (Connection timed out)] 07:05:11 tcr [n=tcr@host145.natpool.mwn.de] has joined #lisp 07:05:30 -!- gonzojive [n=red@condi.Stanford.EDU] has left #lisp 07:05:37 drafael [n=tapio@ip-118-90-139-235.xdsl.xnet.co.nz] has joined #lisp 07:08:22 morning 07:09:35 -!- Muld [i=wr23@88-196-38-67-dsl.noe.estpak.ee] has quit [Success] 07:09:35 -!- hawkbill [n=pradyus@nat/yahoo/x-7f73d8292b4a8110] has quit ["Leaving."] 07:09:44 -!- A_anekos is now known as anekos 07:10:18 minion: diff? 07:10:19 diff: DIFF is a simple asdf-installable Text Library which can compute unified-style or context-style diffs between two files. http://www.cliki.net/diff 07:12:16 I was going to ask if anyone had a copy of that, but I've since found it. 07:13:36 -!- pstickne [n=pstickne@c-24-21-76-57.hsd1.wa.comcast.net] has quit [Read error: 110 (Connection timed out)] 07:15:04 Hm. I just discovered slime's profiler. 07:15:07 \o/ 07:20:26 wbraun [n=chatzill@91.32.59.210] has joined #lisp 07:20:27 ski [n=slj@85.224.17.144] has joined #lisp 07:23:23 While implementing a typical recursive accumulator via (labels ((iter (accum ...) (if ... (iter (cons ... accum) ...) (nreverse accum))) ...) ...), is there any simple way to avoid using `NREVERSE' and still have the same effect? 07:24:23 sure: keep the accumulator's last pair around and cons onto it 07:24:33 keep a pointer to the tail and nconc with ... yeah 07:24:58 see how Norvig does it in his queue implementation, for example 07:25:04 -!- Gertm [n=user@mail.dzine.be] has quit [Read error: 60 (Operation timed out)] 07:25:15 or try macroexpanding a loop with collect clause 07:25:54 cmm: Krystof: To be honest, I was looking for a more simpler way (i.e. some builtin function similar to `COLLECT' of `LOOP'), but it seems that appears to be the only possible way of doing that. 07:26:12 -!- fvw [n=sdfpme@113.77.216.106] has quit [Remote closed the connection] 07:27:10 Gertm [n=user@mail.dzine.be] has joined #lisp 07:27:13 well, maybe you should have asked that question rather than the one you asked 07:27:19 you're welcome 07:27:26 -!- wbraun [n=chatzill@91.32.59.210] has quit ["ChatZilla 0.9.85 [Firefox 3.5/20090624012820]"] 07:28:57 -!- saikat_ [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has quit [] 07:31:57 athos [n=philipp@92.250.250.68] has joined #lisp 07:38:49 saikat_ [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has joined #lisp 07:39:16 vy: MAPCAR collects the results of each loop iteration. 07:40:09 -!- ltriant [n=ltriant@lithium.mailguard.com.au] has quit ["leaving"] 07:44:16 -!- saikat_ [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has quit [] 07:48:16 Quadre` [n=quad@24.118.241.200] has joined #lisp 07:49:45 vy: also, the head-tail-list data structure let you do that. You could write a one-liner collect function too. (labels ((make-ht-list () (let ((prefix (cons nil nil))) (cons prefix prefix))) (collect-into (x htlist) (push x (cddr htlist)) (setf (cdr htlist) (cddr htlist)))) (let ((res (make-ht-list))) (collect-into 1 res) (collect-into 2 res) (collect-into 3 res) (cdar res))) 07:50:03 pjb: that's quite a one-lines... *squints* 07:50:36 I gave an example with it. Consider just collect-into. 07:52:48 Good morning. 07:52:48 vy: Use my PUSHEND macro 07:53:17 tcr: is it not O(n)? 07:53:21 Yes 07:53:34 He wants to call it in a loop... 07:53:48 (let ((list nil) (tail nil)) (dotimes (i 1000) (pushend i list tail)) list) 07:54:02 hkBst [n=hkBst@gentoo/developer/hkbst] has joined #lisp 07:54:05 Like this is can be O(1). 07:54:34 Oh It's O(1), I thought you were talking about a loop being O(n) with it 07:54:55 ok 07:55:01 vy: it's on paste.lisp.org; search through it via google 07:55:54 Quadre_ [n=quad@24.118.241.200] has joined #lisp 07:57:08 -!- azanar [n=azanar@dsl231-050-107.sea1.dsl.speakeasy.net] has quit [Client Quit] 08:00:09 -!- Phoodus [i=foo@ip68-231-38-131.ph.ph.cox.net] has quit [Read error: 54 (Connection reset by peer)] 08:03:01 xan [n=xan@29.Red-83-59-49.dynamicIP.rima-tde.net] has joined #lisp 08:04:09 Is DEFINE-SYMBOL-MACRO the way to go to define an alias for a function? 08:05:28 blandest [n=blandest@softhouse.is.ew.ro] has joined #lisp 08:06:05 No 08:06:14 -!- Quadre` [n=quad@24.118.241.200] has quit [Read error: 110 (Connection timed out)] 08:06:35 Quadre` [n=quad@24.118.241.200] has joined #lisp 08:07:36 tcr: What is then? 08:08:03 QinGW [n=wangqing@203.86.81.2] has joined #lisp 08:08:05 And what's wrong with DEFINE-SYMBOL-MACRO? Debugging mess? 08:08:43 mega1 [n=mega@pool-03102.externet.hu] has joined #lisp 08:10:41 won't work on #' 08:10:42 -!- dys [n=andreas@p5B316C65.dip.t-dialin.net] has quit [No route to host] 08:12:15 You can use (setf (fdefinition 'bar) #'(lambda (&rest args) (apply 'foo args))) 08:13:14 won't (setf (symbol-function 'alias) #'existing-function) work? it's globalish, but so is define-symbol-macro 08:13:26 how about (setf (fdefinition 'bar) (fdefinition 'foo)) 08:14:07 Think about what the difference between that and mine is 08:15:10 (1) your one works for macros, (2) works if the function is not defined or redefined, (3) loses the function-lambda-expression 08:16:24 how do you come to conclude (1)? 08:16:40 -!- lemoinem [n=swoog@modemcable110.189-201-24.mc.videotron.ca] has quit [Read error: 110 (Connection timed out)] 08:17:05 wrongly (doh) 08:18:10 i was thinking about this (defmacro alias (from to) `(setf (fdefinition ',from) (lambda(&rest args) (,to ,@args))) 08:18:21 lexa_ [n=lexa_@seonet.ru] has joined #lisp 08:18:42 Yeah I was thinking of redefinition; (3) is valid in so far as the arglist of BAR will be suboptimal. You'd have to reinsert the arglist in the lambda form. Or use some macro which retrieves the arglist of FOO, and uses that. 08:18:49 -!- lexa_ is now known as Guest7976 08:20:21 Sikander [n=soemraws@wirenat-eld.strw.leidenuniv.nl] has joined #lisp 08:20:41 vy` [n=user@nbvyazici.cs.bilkent.edu.tr] has joined #lisp 08:21:22 i was thinking about this (defmacro alias (from to) `(defmacro ,to (&rest args) `(,',from ,@args))) 08:21:44 as well, but that is quite unfortunate as it destroys the functionlikeness 08:22:16 -!- vy [n=user@nbvyazici.cs.bilkent.edu.tr] has quit [Read error: 113 (No route to host)] 08:22:31 frozsyn [n=frozsyn@blm93-2-82-229-63-104.fbx.proxad.net] has joined #lisp 08:23:40 -!- xan [n=xan@29.Red-83-59-49.dynamicIP.rima-tde.net] has quit [Read error: 113 (No route to host)] 08:24:07 -!- QinGW1 [n=wangqing@203.86.81.2] has quit [Connection timed out] 08:27:12 blandest1 [n=blandest@85.204.33.242] has joined #lisp 08:28:38 -!- blandest [n=blandest@softhouse.is.ew.ro] has quit [Read error: 104 (Connection reset by peer)] 08:29:20 jdz_ [n=jdz@85.254.211.133] has joined #lisp 08:29:21 -!- Quadre_ [n=quad@24.118.241.200] has quit [Connection timed out] 08:29:32 aerique [i=euqirea@xs2.xs4all.nl] has joined #lisp 08:30:57 is this actually a problem? .. heh 08:31:05 just wondering .. i haven't had a use for this, yet 08:31:27 I never understood the recurring desire for this, either. 08:31:50 yeah, but still .. aliasing a frickin' function .. in CL? .. that should be trivial, right? .. :P 08:31:56 hm 08:31:57 Why would you want to alias a name, if the order of the arguments is left untouched? 08:32:18 the order of the arguments, or their default values etc... 08:32:53 ahyeah .. there is actually a couple of things going on there 08:33:47 Sikander: backwards compatibility 08:34:14 Sikander: You may also want to define a compiler-macro for the old function which signals a warning at compilation-time that the function is deprecated 08:34:21 xan [n=xan@193.145.130.250] has joined #lisp 08:34:40 tcr: That latter part makes more sense, I agree 08:35:54 tcr: but just aliasing makes no sense imho 08:36:32 -!- Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has quit [Read error: 113 (No route to host)] 08:36:36 unless you want to participate in some code obfuscation contest 08:36:47 But for that, there was a whole language invented: Perl! 08:36:48 What part about backwards compatibility did you not understand? 08:37:18 the part where you renamed the function without changing the interface at all =p 08:37:34 exactly 08:37:44 makes no sense! 08:38:33 Has happened quite a lot in Emacs. 08:39:36 And you can change the interface as long as it's not incompatible 08:39:41 how about localizing the language? 08:40:07 Heh, may make sense for a DSL 08:40:54 *hefner* thinks about how to portably toggle between readable and unreadable printing of a structure. :around method on print-object, perhaps? 08:41:21 hefner, when do you want to toggle between the two modes? *curious* 08:41:29 blandest [n=blandest@softhouse.is.ew.ro] has joined #lisp 08:42:38 -!- blandest1 [n=blandest@85.204.33.242] has quit [Read error: 104 (Connection reset by peer)] 08:42:43 tic: normally, I like the unreadable representation, because otherwise every time I print one, it prints a whole huge graph (or if I forget to set *print-circle*, explodes). but I'd like to write them to a file, and the #S syntax would give me that for free. 08:42:58 hefner, clever. 08:44:53 hefner: You have to use (:print-object nil) for a print-object method to be generated 08:45:31 -!- jdz [n=jdz@85.254.193.149] has quit [Read error: 110 (Connection timed out)] 08:45:44 hefner: Sure you just don't want to set *PRINT-LEVEL* to some smallish value? 08:46:04 -!- jdz_ is now known as jdz 08:46:28 attila_lendvai [n=ati@89.132.189.132] has joined #lisp 08:47:03 oh, of course. I not only misread something on the defstruct page (I thought it always generated a print-object method), but neglected to read what print-object does by default. 08:47:13 jdz: ... please, no localising of language... it still gives me shivers ... 08:47:51 p_l: only for French then? ;) 08:48:46 Sikander: ... I had to deal with two localised variants of language. The second one had been slightly easier to swallow due to being simply a localised bastardisation of Pascal. 08:49:07 i think i'd appreciate if i could create a package with localized aliases of standard functions, for educational purposes (like teaching a kid a bit of programming) 08:49:20 But when the localised language is basically a Lisp without parentheses, aka Logo, it's really bad 08:49:30 but aliases might be the wrong approach anyway 08:49:31 tcr: no, even a print level of 2 would be too deep, but setting it to 1 would get irritating quickly. that's the trouble with *print-level*, it can't read your mind. 08:49:32 Isn't that what basic is for? 08:49:47 Sikander: Basic wasn't designed for that 08:49:56 Logo is an excellent choice, IMHO 08:50:06 hefner: Too deep for what? 08:50:18 tcr: to need not to fill my screen and be irritating. 08:50:21 s/need/deep/ 08:50:26 p_l: I meant basic was for teaching kids... I know I started with basic when I was 8 08:50:33 jdz: I consider drilling english terminology in programming languages a good thing 08:50:48 my kid knows english from playing games anyway 08:50:54 so the whole point is moot 08:51:02 Sikander: BASIC started out on some big, early multiuser system 08:51:24 i was just trying to come up with a "roasonable" excuses for the aliasing thing 08:51:25 hefner: Well you could also customize *print-length*, the downside is that it would apply to any datum 08:51:26 Sikander: then it got popular with PCs 08:51:28 qamikaz [n=qamikaz@79.123.145.8] has joined #lisp 08:51:29 p_l: Sure, I was just making a stupid observation. Didn't mean to offend anyone 08:52:14 Sikander: no offence taken. Just that LOGO, when taught properly, is a powerful educational tool. And possibly not only educational 08:52:16 hefner: So perhaps you want an :around method which binds *print-level* and *print-length* :) But if you're as far as that then you can just roll your own, too! 08:52:51 tcr: no, I think I just want to check *print-readably* somewhere and print accordingly 08:53:20 I think it's better to have an teach programming in one language. Else, before you know it, you have localised code everywhere, no one understands a word of it (and of course the comments will be incomprehensible as well) 08:53:21 koollman [n=samson_t@sd-10510.dedibox.fr] has joined #lisp 08:53:31 s/an/and/ 08:53:35 The fact is that the CL standard is not entirely nice for retrospection: it doesn't give a lot of help to redefine the CL system itself. eg. you cannot redefine functions in the CL package. 08:53:39 delYsid [n=user@chello084115136207.3.graz.surfer.at] has joined #lisp 08:53:47 Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has joined #lisp 08:53:48 hefner: Oh yeah it seems like structures are always printed readably 08:54:15 However, you can easily enough define a layer above standard CL that is entirely retrospectable. In this layer you may even provide a CL package. 08:54:17 matimago: I suspect that giving such freedom would bode badly for implementations 08:54:35 stassats [n=stassats@wikipedia/stassats] has joined #lisp 08:54:37 p_l: indeed, the CL standard cares as much about implementations as users. 08:54:49 hefner: Do your structures have distinct names, or IDs? 08:55:10 matimago: it has to. Otherwise you end up with PL/I 08:55:19 :-) 08:55:19 structs are printed readably unconditionally, yeah. I wonder if it's somehow dictated by the spec or is a sort of universal accident 08:55:31 I should learn PL/I one day... 08:55:52 matimago: afaik there isn't a single implementation that covers complete PL/I 08:56:11 cmm: 22.1.3.12 08:56:29 clhs 22.1.3.12 08:56:30 http://www.lispworks.com/reference/HyperSpec/Body/22_acl.htm 08:56:31 *stassats* thought that's a version of emacs 08:56:42 the newer ones possibly no longer care about some obscure features, but the earlier ones simply never got it done 08:56:57 tcr: thank you 08:57:19 -!- tcr [n=tcr@host145.natpool.mwn.de] has quit ["Leaving."] 08:58:05 It doesn't say how slot values are printed. (defstruct s x) (defclass a () ()) (let ((*print-readably* t)) (print (make-s :x (make-instance 'a)))) ==> error 08:58:16 tcr: they have filenames associated with them, but it isn't necessary contained in a slot 09:00:42 kmcorbett [n=Keith@76.119.215.57] has joined #lisp 09:05:34 that's another point them, I'm usually not interested in most of the slots, just the name. 09:06:04 *hefner* begins to question his grip on the English language.. 09:09:06 hefner: because you have grips on it? Language generator (speach centers) are compiled out once you've learned the language. Your thinking is converted to speach (or typing) without any control or "grips"... All you can do, is to spend time to consciously regenerate correct forms, several times, and see if a new compilation occurs... 09:12:17 matimago: I'd say it's rather more complicated 09:12:48 jewel_ [n=jewel@dsl-247-203-169.telkomadsl.co.za] has joined #lisp 09:13:45 but can always be simplified by application of tasty alcohol-containing drinks 09:13:57 I think the thought->typing path has been optimized over the years to run quickly and effortlessly, but it tends to discard or mangle the language a bit.. 09:14:18 *stassats* can't type after alcohol-containing drinks 09:14:48 -!- mrsolo [n=mrsolo@adsl-68-126-215-139.dsl.pltn13.pacbell.net] has quit [Read error: 110 (Connection timed out)] 09:14:53 stassats: interesting. my typing remains the same, I just think I'm funny when I'm not 09:14:56 stassats: well, that does make things simpler, doesn't it? 09:15:38 -!- kmcorbett1 [n=Keith@76.119.215.57] has quit [Read error: 110 (Connection timed out)] 09:15:44 Chris: it's certainly funny to read logs in the morning 09:20:11 mrsolo [n=mrsolo@adsl-68-126-215-139.dsl.pltn13.pacbell.net] has joined #lisp 09:22:56 -!- Guest7976 [n=lexa_@seonet.ru] has left #lisp 09:25:52 lemoinem [n=swoog@modemcable110.189-201-24.mc.videotron.ca] has joined #lisp 09:28:35 Yuuhi [i=benni@p5483E5C6.dip.t-dialin.net] has joined #lisp 09:30:08 rdd [n=user@c83-250-157-93.bredband.comhem.se] has joined #lisp 09:31:04 -!- athos [n=philipp@92.250.250.68] has quit ["leaving"] 09:31:10 -!- stassats [n=stassats@wikipedia/stassats] has quit [Remote closed the connection] 09:31:36 stassats [n=stassats@wikipedia/stassats] has joined #lisp 09:38:14 existentialmonk [n=carcdr@64-252-32-194.adsl.snet.net] has joined #lisp 09:39:02 -!- kahmalo [n=Kalle@2002:5517:2040:1:230:84ff:fe3e:cfef] has quit [Remote closed the connection] 09:39:03 -!- mrsolo [n=mrsolo@adsl-68-126-215-139.dsl.pltn13.pacbell.net] has quit [Connection timed out] 09:42:15 abeaumont_ [n=abeaumon@193.145.130.250] has joined #lisp 09:46:43 ASau [n=user@193.138.70.52] has joined #lisp 09:46:45 kahmalo [n=Kalle@2002:5517:2040:1:230:84ff:fe3e:cfef] has joined #lisp 09:47:01 |koft| [n=kvirc@adsl-221-71-194.rmo.bellsouth.net] has joined #lisp 09:48:40 -!- KingNato [n=patno@fw.polopoly.com] has quit [] 09:50:03 -!- wubbster [n=wubb@doc-24-206-235-153.kw.tx.cebridge.net] has quit [Remote closed the connection] 09:55:17 -!- daniel__ [i=daniel@unaffiliated/daniel] has quit [Read error: 104 (Connection reset by peer)] 09:55:55 ejs [n=eugen@220-82-135-95.pool.ukrtel.net] has joined #lisp 09:56:01 daniel [i=daniel@unaffiliated/daniel] has joined #lisp 10:01:32 -!- xan [n=xan@193.145.130.250] has quit [Read error: 110 (Connection timed out)] 10:02:54 zophy [n=sy@host-242-6-111-24-static.midco.net] has joined #lisp 10:05:07 -!- koft [n=kvirc@adsl-221-79-167.rmo.bellsouth.net] has quit [Read error: 113 (No route to host)] 10:05:56 -!- ejs [n=eugen@220-82-135-95.pool.ukrtel.net] has quit ["This computer has gone to sleep"] 10:06:35 xan [n=xan@193.145.130.250] has joined #lisp 10:07:10 stassats` [n=stassats@wikipedia/stassats] has joined #lisp 10:07:36 -!- stassats` [n=stassats@wikipedia/stassats] has quit [Client Quit] 10:07:41 okflo [n=user@188-23-82-85.adsl.highway.telekom.at] has joined #lisp 10:07:57 -!- stassats [n=stassats@wikipedia/stassats] has left #lisp 10:07:59 stassats [n=stassats@wikipedia/stassats] has joined #lisp 10:09:09 hawkbill [n=pradyus@117.192.7.21] has joined #lisp 10:09:37 -!- QinGW [n=wangqing@203.86.81.2] has quit [Read error: 104 (Connection reset by peer)] 10:13:16 -!- okflo [n=user@188-23-82-85.adsl.highway.telekom.at] has quit [Remote closed the connection] 10:13:47 okflo [n=user@188-23-82-85.adsl.highway.telekom.at] has joined #lisp 10:16:14 -!- lispm [n=joswig@e177121219.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)] 10:18:51 -!- aja [n=aja@S01060018f3ab066e.ed.shawcable.net] has quit [Read error: 110 (Connection timed out)] 10:19:13 -!- ski [n=slj@85.224.17.144] has quit [Read error: 104 (Connection reset by peer)] 10:25:14 ski [n=slj@85.224.17.144] has joined #lisp 10:26:28 Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has joined #lisp 10:26:47 hypno [n=hypno@195.43.248.100] has joined #lisp 10:27:13 azanar [n=azanar@dsl231-050-107.sea1.dsl.speakeasy.net] has joined #lisp 10:28:49 -!- borism_ [n=boris@195-50-199-19-dsl.krw.estpak.ee] has quit [Read error: 145 (Connection timed out)] 10:32:08 mstevens [n=mstevens@zazen.etla.org] has joined #lisp 10:35:16 -!- ia [n=ia@89.169.189.230] has quit [Read error: 110 (Connection timed out)] 10:36:16 ia [n=ia@89.169.189.230] has joined #lisp 10:43:33 -!- Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has quit ["Am I missing an eyebrow?"] 10:43:53 ruediger [n=ruediger@62-47-143-82.adsl.highway.telekom.at] has joined #lisp 10:45:39 -!- ASau [n=user@193.138.70.52] has quit [Read error: 54 (Connection reset by peer)] 10:48:06 -!- Gertm [n=user@mail.dzine.be] has quit [Read error: 60 (Operation timed out)] 10:48:20 -!- rdd [n=user@c83-250-157-93.bredband.comhem.se] has quit [Remote closed the connection] 10:50:08 fvw [n=sdfpme@113.77.216.106] has joined #lisp 10:51:33 manuel_ [n=manuel@213.144.1.98] has joined #lisp 10:54:38 ThomasI [n=thomas@unaffiliated/thomasi] has joined #lisp 10:54:49 i have a list of objects with slots s1, s2 and s3. from this list i want to find all those objects for which s1=1 and s2=2. how can i do this? 10:54:56 -!- qamikaz [n=qamikaz@79.123.145.8] has quit [Remote closed the connection] 10:56:09 -!- kuwabara [n=kuwabara@84.14.121.138] has quit [Read error: 60 (Operation timed out)] 10:56:16 (remove-if-not (lambda (object) (and (equal (slot-1 object) 1) (equal (slot-2 object) 2))) list) 10:56:55 Jabberwockey [n=jens@port-3757.pppoe.wtnet.de] has joined #lisp 10:58:00 hefner: ping 10:59:08 jmbr [n=jmbr@guest245.gti.ssr.upm.es] has joined #lisp 11:00:01 kuwabara [n=kuwabara@84.14.121.138] has joined #lisp 11:02:14 -!- abeaumont_ [n=abeaumon@193.145.130.250] has quit [Remote closed the connection] 11:02:53 Edico [n=Edico@unaffiliated/edico] has joined #lisp 11:02:55 stassats: thanks. i'm still not using lambda enough. need to try and change the way i think 11:03:01 -!- mstevens [n=mstevens@zazen.etla.org] has quit ["leaving"] 11:03:23 lambda is not the point 11:04:10 i know that that's not the point 11:04:22 what i meant was thinking the 'functional programming' way 11:04:47 and could it be written as (with-slots (s1 s2) object (and (eql s1 1) (eql s2 2))) 11:05:40 didn't know about that. but the lambda way, i should've figured that one out myself 11:07:51 levy [n=levy@apn-89-223-154-233.vodafone.hu] has joined #lisp 11:09:46 -!- levy [n=levy@apn-89-223-154-233.vodafone.hu] has quit [Read error: 104 (Connection reset by peer)] 11:10:15 levy [n=levy@apn-89-223-154-233.vodafone.hu] has joined #lisp 11:13:21 hefner, I like the interface of Shuffletron. Have you looked into Spotify? 11:15:09 elderK [n=elderK@222-152-93-33.jetstream.xtra.co.nz] has joined #lisp 11:15:49 -!- elderK [n=elderK@222-152-93-33.jetstream.xtra.co.nz] has left #lisp 11:21:18 ignas [n=ignas@78-60-73-85.static.zebra.lt] has joined #lisp 11:24:12 nikodemus [n=nikodemu@cs181229041.pp.htv.fi] has joined #lisp 11:24:20 hoy 11:24:54 just when i was going to commit a last fix and start cutting the release, i have a new regression, i think 11:25:06 (need to check on thing still) 11:25:26 -!- blandest [n=blandest@softhouse.is.ew.ro] has quit [Read error: 60 (Operation timed out)] 11:25:56 -!- Axioplase is now known as Axioplase_ 11:26:47 -!- jmbr [n=jmbr@guest245.gti.ssr.upm.es] has quit [Read error: 113 (No route to host)] 11:29:42 -!- Jasko [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has quit ["Leaving"] 11:30:55 milaz [n=user@85.175.41.248] has joined #lisp 11:35:54 -!- levy [n=levy@apn-89-223-154-233.vodafone.hu] has quit ["..."] 11:37:07 nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 11:39:32 ASau [n=user@193.138.70.52] has joined #lisp 11:39:34 Guest497` [n=user@72.14.228.129] has joined #lisp 11:40:15 -!- Guest49718 [n=user@72.14.228.129] has quit [Read error: 104 (Connection reset by peer)] 11:42:31 krumholt [n=krumholt@port-92-193-20-71.dynamic.qsc.de] has joined #lisp 11:42:31 -!- rtoym_ [n=chatzill@user-0c8hpll.cable.mindspring.com] has quit [Read error: 104 (Connection reset by peer)] 11:45:56 KingNato [n=patno@fw.polopoly.com] has joined #lisp 11:46:10 daniel__ [i=daniel@unaffiliated/daniel] has joined #lisp 11:46:30 -!- hawkbill [n=pradyus@117.192.7.21] has quit ["Leaving."] 11:49:26 yvdriess [n=yvdriess@134.184.43.185] has joined #lisp 11:50:13 antoszka [n=antoszka@unaffiliated/antoszka] has joined #lisp 11:50:26 silenius [n=jl@yian-ho03.nir.cronon.NET] has joined #lisp 11:50:26 Tordek [n=tordek@host157.190-137-190.telecom.net.ar] has joined #lisp 11:51:12 ok, it's a regression since 1.0.29.42 at least 11:52:19 -!- Quadrescence [n=quad@unaffiliated/quadrescence] has quit [SendQ exceeded] 11:52:41 pathnames again? :) 11:52:45 nikodemus pasted "regression (fp related miscompilation on x86-64)" at http://paste.lisp.org/display/83226 11:53:31 i think this is either SSE2 or unboxed constants 11:53:42 pkhuong!! 11:53:55 how on earth did you find that? 11:54:39 blandest [n=blandest@softhouse.is.ew.ro] has joined #lisp 11:54:59 mstevens [n=mstevens@zazen.etla.org] has joined #lisp 11:54:59 my raytracer broke -- i realized that it makes a good test-case for floating point only at the last minute 11:56:39 -!- yvdriess [n=yvdriess@134.184.43.185] has quit [] 11:56:45 bombshelter13p_ [n=bombshel@76-10-149-209.dsl.teksavvy.com] has joined #lisp 11:56:53 -!- bombshelter13p_ [n=bombshel@76-10-149-209.dsl.teksavvy.com] has quit [Remote closed the connection] 11:58:36 speaking of regressions, I can't seem to make clisp 2.44.1 build sbcl on Linux x86 11:59:14 of course it could be me rather than clisp or sbcl :) 11:59:42 yvdriess [n=yvdriess@soft85.vub.ac.be] has joined #lisp 12:00:39 piso: hrm. What's the failure mode? 12:00:39 -!- ASau [n=user@193.138.70.52] has quit [Read error: 110 (Connection timed out)] 12:01:06 piso: I think no-one but me tests exotic hosts, and I have had negative time this month to do anything at all 12:01:40 1.0.29.53 is good 12:01:55 -!- daniel [i=daniel@unaffiliated/daniel] has quit [Read error: 110 (Connection timed out)] 12:02:05 (35 more exam scripts to go, 35 scripts piled high on my desk. 35 scripts that need green ink scribbled all over them, and then I will be finished with these damn scripts) 12:02:29 the build usually ends up dropping into ldb (I've tried a few recent versions of sbcl) 12:02:43 what, so you get as far as target-2 and then it dies? 12:02:45 that's lame 12:03:02 actually, sometimes make-host-1 fails 12:03:32 clisp warns on (declare (index foo)), for example 12:03:42 (declare (type index foo)) is ok 12:03:54 piso: yes, that I know about 12:03:59 I can fix those and get a bit further 12:04:14 rswarbrick [n=rswarbri@user-5447c311.wfd92.dsl.pol.co.uk] has joined #lisp 12:04:29 clhs declare 12:04:30 http://www.lispworks.com/reference/HyperSpec/Body/s_declar.htm 12:04:43 -!- manuel_ [n=manuel@213.144.1.98] has quit [] 12:04:43 since nikodemus is holding up the release a bit anyway, let me see if I can fix things a bit 12:04:43 I only think you guys might care because INSTALL says "CLISP (only some versions: 2.44.1 is OK, 2.47 is not)" 12:04:50 piso: yes, I care 12:05:02 -!- mstevens [n=mstevens@zazen.etla.org] has quit ["leaving"] 12:07:11 piso: oh, I should perhaps say that it's possible that I only tested thoroughly and properly on x86-64 and was going to deal with problem reports on other architectures as they came in 12:07:16 hmm, why doesn't CLISP like the (declare (typespec var)) form? 12:07:26 joachifm [n=joachim@bjo1-1x-dhcp154.studby.uio.no] has joined #lisp 12:07:48 luis: there's a contradiction in the spec 12:08:31 but cvs clisp likes at least non-compound typespecs again 12:08:35 -!- hkBst [n=hkBst@gentoo/developer/hkbst] has quit [Read error: 104 (Connection reset by peer)] 12:09:08 yep. it's the inline constants 12:10:18 Jasko [n=tjasko@209.74.44.225] has joined #lisp 12:10:39 nikodemus annotated #83226 "inline constants indeed" at http://paste.lisp.org/display/83226#1 12:11:50 -!- xan [n=xan@193.145.130.250] has quit [Read error: 110 (Connection timed out)] 12:12:12 ASau [n=user@193.138.70.52] has joined #lisp 12:18:16 -!- Quadre` is now known as Quadrescence 12:18:49 -!- ignas [n=ignas@78-60-73-85.static.zebra.lt] has quit [Read error: 110 (Connection timed out)] 12:19:41 ASau` [n=user@193.138.70.52] has joined #lisp 12:19:55 -!- Jasko [n=tjasko@209.74.44.225] has quit ["Leaving"] 12:21:07 got a fix, it was just a single register confusion 12:21:11 hkBst [n=hkBst@gentoo/developer/hkbst] has joined #lisp 12:21:29 ignas [n=ignas@78-60-73-85.static.zebra.lt] has joined #lisp 12:22:52 demmel [n=demmel@94.216.77.147] has joined #lisp 12:23:20 clisp 2.44.1 doesn't build sbcl cvs on x86-64 here either 12:24:34 FAILURE-P was set when creating "obj/from-host/src/compiler/ir1util.lisp-obj" 12:24:45 Unknown declaration COMBINATION. 12:25:56 TuxPurple [n=TuxPurpl@unaffiliated/tuxpurple] has joined #lisp 12:26:54 we could just mandate (type foo var) style for all non-standard types, i guess 12:27:55 imo it is good style in any case, since otherwise macros cannot really hope to understand type declarations properly (which variables they refer to, in case the macro has to splice them in different places) 12:28:09 ...not that we have that problem internally 12:28:13 -!- okflo [n=user@188-23-82-85.adsl.highway.telekom.at] has quit [Read error: 110 (Connection timed out)] 12:30:13 (type foo var) is a bit more readable, in any case, if foo isn't a standard type 12:31:45 ** - Continuable Error 12:31:46 "SB!XC" is already a package name. 12:31:59 -!- sytse [i=sytse@speedy.student.ipv6.utwente.nl] has quit [Read error: 113 (No route to host)] 12:32:22 sbcl should be incorporated in clisp test cases... And sbcl should incorporate clisp in its bootstrapping procedure test cases :-) 12:32:39 ;; Loading file obj/from-host/src/compiler/main.lisp-obj ... 12:32:39 WARNING: Found non-STANDARD-CHAR 12:32:41 blandest1 [n=blandest@softhouse.is.ew.ro] has joined #lisp 12:33:36 -!- blandest [n=blandest@softhouse.is.ew.ro] has quit [Read error: 104 (Connection reset by peer)] 12:37:36 cracki [n=cracki@134.130.183.101] has joined #lisp 12:37:51 -!- antoszka [n=antoszka@unaffiliated/antoszka] has quit ["+++ killed by SIGSEGV +++"] 12:37:56 El piso es mojado. 12:39:50 hah, this is the same as the complex atanh bug 12:39:55 small favors, etc 12:41:55 blandest [n=blandest@85.204.33.242] has joined #lisp 12:42:00 -!- blandest1 [n=blandest@softhouse.is.ew.ro] has quit [Read error: 104 (Connection reset by peer)] 12:47:07 antoszka [n=antoszka@unaffiliated/antoszka] has joined #lisp 12:48:37 piso: I've just built 1.0.29.54.rc3 on x86-64/linux 12:49:01 piso: oh, I bet I know -- I use 'clisp -ansi -on-error abort' 12:49:13 -on-error abort actually means "continue if it's a cerror, or else abort" 12:51:09 oh, let me try that 12:51:22 tcr [n=tcr@host145.natpool.mwn.de] has joined #lisp 12:52:17 pkhuong [n=pkhuong@x-132-204-255-22.xtpr.umontreal.ca] has joined #lisp 12:52:32 nikodemus: thanks for fixing that (: What sort of register issue? 12:54:08 Jasko [n=tjasko@75-149-33-105-SFBA.hfc.comcastbusiness.net] has joined #lisp 12:54:20 nikodemus annotated #83226 "the fix" at http://paste.lisp.org/display/83226#2 12:54:43 just tmp / r confusion 12:55:40 right doh (: 12:56:38 nikodemus: your test cases declaims an ftype for rotate-around-BAD. 12:56:58 -s 12:58:11 bytbox [n=chatzill@AC84B606.ipt.aol.com] has joined #lisp 12:58:56 Common Lisp doesn't have support for sockets, right? 12:59:11 bytbox: it is not part of the standard, no. 12:59:25 bytbox: there are some libraries that are widely ported to CL implementations. usocket is one. 12:59:31 every self-respectable implementation does support them 12:59:34 Xach: thanks for the pointer 12:59:57 also, what is the differencle between GCL and CLisp? They both seem to be common lisp as implemented by gnu... 13:00:00 spilman [n=spilman@ARennes-552-1-123-234.w92-139.abo.wanadoo.fr] has joined #lisp 13:00:20 bytbox: GNU CLISP is only GNU because of readline. 13:00:38 and GCL is old and buggy 13:00:44 pkhuong: Oh, ok. 13:00:49 stassats: oops. I'll switch 13:00:56 bytbox: very different histories, implementation strategies, performance characteristics, etc 13:01:13 well, thanks all! 13:01:48 So, what about MIT Lisp and BSD Lisp? 13:01:55 ;-) 13:02:23 matimago: no lisp on the west coast. 13:03:38 UCI Lisp? 13:03:44 BSD Lisp aka Franz Lisp 13:03:58 pkhuong: isn't there some CL implementation delivered with a BSD license? 13:03:58 bytbox: if you're looking for closest "geneologically" to GCL implementation, then there's ECL (apparently both of the have Kyoto? Common Lisp as ancestor) 13:04:29 matimago: MIT Lisps are obviously Maclisp and Scheme :D 13:04:33 p_l: no, I had just gotten confused. 13:04:54 bytbox: but if you're looking for a very good implementation, look at SBCL. :) 13:04:56 matimago: SBCL? CMUCL? at least partially i think. 13:05:11 cmucl is carnegie mellon - what's sb? 13:05:19 steel bank 13:05:44 aka "sanely bootstrappable"? :) 13:05:57 that's what Carnegie and Mellon were fond of respectively 13:06:01 Captain_Thunder [n=jimmymil@c-24-35-95-148.customer.broadstripe.net] has joined #lisp 13:06:05 Yes, CMUCL and SBCL are a mix of BSD-license and public domain. We could call them BSD lisps, like GCL and CLISP are GPL lisps. 13:06:34 jmbr [n=jmbr@128.33.220.87.dynamic.jazztel.es] has joined #lisp 13:06:44 ABCL is under GPL too 13:07:07 Gertm [n=user@mail.dzine.be] has joined #lisp 13:07:08 matimago: do you think there should also be an X11 Lisp and an Artistic Lisp? 13:07:16 :-) 13:07:34 EULA Lisp? 13:07:43 pkhuong: oops, thanks 13:08:11 Creative Commons Lisp? 13:08:30 Microsoft Lisp(tm) 13:08:42 With clip, you know 13:09:29 xan [n=xan@conference/gcds/x-29c0b28be1d70076] has joined #lisp 13:09:34 Wow guys, I just got an email saying that a Mr. Halim Mama would like to transfer $7.2 million into my bank account :D 13:10:15 Xach: MIT and X11 license are iirc the same thing 13:10:21 With that money, you can not only buy Lisp in Small Pieces, you will have the leisure time to read it thoroughly. 13:11:17 I have taken the channel's advice, and will more thoroughly investigate why my idea is frowned upon. 13:11:22 But I don't have to like it :P 13:11:42 schoppenhauer [n=senjak@unaffiliated/schoppenhauer] has joined #lisp 13:12:03 Xach: IIRC X11 has MIT license due to its origins, with said license being "preferred" method licensing for "technology transfer" or some other legal mumbo jumbo 13:12:26 -!- bytbox [n=chatzill@AC84B606.ipt.aol.com] has left #lisp 13:13:21 rread [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has joined #lisp 13:13:22 -!- rread_ [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has quit [Read error: 54 (Connection reset by peer)] 13:14:30 What's the reasoning behind the decision to have lexical variables disappear at runtime? 13:14:44 (Obviously it's possible to retain them, since some implementations do so for you) 13:14:47 Captain_Thunder: they're not needed anymore. Optimization. 13:15:15 (disassemble (compile nil '(lambda (x) (let ((y (* x x))) (/ y 2))))) 13:15:28 there's no need for the lexical variable Y. 13:15:35 No need to store it in any memory location. 13:15:38 -!- demmel [n=demmel@94.216.77.147] has quit [] 13:15:48 LET is only a naming mechanism, not a storage management one. 13:15:50 Captain_Thunder: that's the point of lexical scoping. Scope is essential to encapsulation. Otherwise your programs become impossible to reason about locally. 13:16:05 -!- bdowning [n=bdowning@mnementh.lavos.net] has quit [Read error: 60 (Operation timed out)] 13:16:28 The retaining is only done for debugging purpose. Then optimizations are disabled. 13:16:44 pkhuong, I don't see how optimizing out lexical variables helps you reason about programs locally. 13:17:02 Captain_Thunder: no, that's the reverse. 13:17:05 matimago: is y then replaced with a constant in a register or what? 13:17:13 Captain_Thunder: the ability to reason locally allows the optimizations. 13:17:18 Captain_Thunder: If you think you want them at runtime, you have not misunderstood the meaning of lexical scope 13:17:26 heh 13:17:32 * -not 13:17:38 matimago: erm, duh. nevermind. 13:17:39 erm of course 13:17:46 hypno: yes, the variable is replaced by some transient temporary expression results in registers. 13:18:21 Captain_Thunder: If you want them at runtime, perhaps you want a funcallable instance instead. 13:18:47 So does Lisp provide some sort of local variables that still exist at runtime? 13:19:14 Captain_Thunder: variables are purely a linguistic abstraction. Your question doesn't make sense. 13:19:46 Captain_Thunder: dynamic variables. 13:20:05 Dynamic variables don't have to be global? 13:20:13 (defvar *a* 1) (let ((*a* 2)) (symbol-value '*a*)) -> 2 13:20:21 Captain_Thunder: no, they don't have. 13:20:38 Well, that's rather nice :) 13:20:48 (let ((*b* 1)) (declare (special *b*)) (let ((*b* 2)) (symbol-value '*b*))) --> 2 13:21:20 Captain_Thunder: Well, they're useful in some cases, but they're not so nice. They break tail calls... 13:21:25 matimago: they're still global; CL conflates symbols and variables for specials. 13:21:38 Nshag [i=user@Mix-Orleans-106-4-22.w193-248.abo.wanadoo.fr] has joined #lisp 13:21:45 matimago: i'd expect 1 13:22:03 Good catch 13:22:03 stassats: you expect wrong. 13:22:05 True, I forgot a declare. 13:22:17 (let ((*b* 1)) (declare (special *b*)) (let ((*b* 2)) (declare (special *b*)) (symbol-value '*b*))) -> 2 13:22:41 Well, depends whether pjb's first snippet was evaluated first ;) 13:22:44 And none of the code outside that expression will know that *b* is even defined? 13:23:06 Yes, if you try to get *b* outside, you'll get an unbound variable error. 13:23:08 Captain_Thunder: Every code that is run in the dynamic extent of the LET. 13:23:31 outside of the dynamic scope, yes. 13:24:15 God are we confusing! 13:24:24 No, I understand. 13:25:50 bombshelter13_ [n=bombshel@toronto-gw.adsl.erx01.mtlcnds.ext.distributel.net] has joined #lisp 13:26:16 -!- rread [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 13:26:28 rread [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has joined #lisp 13:26:37 I tried typing in *b* outside the block it was declared, and I just get its value back... 13:27:04 what value? 13:27:10 (let ((z 1)) (declare (special *b*)) (setf *b* 1) (eval '(+ *b* 1))) => 2 13:27:13 *b* => 1 13:27:32 AFAIK, this is implementation specific behavior. 13:27:47 If you used let instead of *b*, it would have been revered to it's unbound status. 13:27:58 reverted. 13:28:28 and the declaration above is meaningless, no? 13:28:33 Oh, I guess it was important to DECLARE the same variable I was LETting. 13:28:36 (let ((z 1)) (let ((*b* 41)) (declare (special *b*)) (eval '(+ *b* 1)))) -> 42 13:28:39 I get it now. 13:28:47 the *b* does not refer to any "variable" in scope, right? 13:28:53 Yes. 13:29:52 So, the point of lexical variables is to act as little symbol-macros type things that don't live past the compilation phase? 13:30:35 no, the point of lexical scoping is to have programs one can understand. 13:31:13 Captain_Thunder: have you ever written any program in BASIC? 13:31:25 I know why we *need* them. 13:31:37 I was just trying to clarify the rationale as to why they don't exist very long. 13:32:00 that's what makes lexical scoping lexical. 13:32:02 The variable names are not necessary for the binary code to execute. 13:32:25 TDT [i=dthole@dhcp80ff869b.dynamic.uiowa.edu] has joined #lisp 13:32:48 You can write: (+ (* a x x) (* b x) c) or (let ((q (* a xx)) (l (* b x))) (+ q l c)), they're equivalent. 13:33:01 Thanks to the lexical scoping of q and l, we know for sure. 13:33:01 Variables might "exist" forever (whatever you mean by exist), you can't observe the difference if the scoping rules are obeyed. 13:33:11 variables are not necessary for the Turing completeness 13:33:28 shmho [n=user@58.142.15.103] has joined #lisp 13:33:35 If you write the LET form, the compiler also knows for sure that it's equivalent to the former form, and if it finds it more efficient to compile the former form than to store data in memory, it will be able to. 13:34:00 Writing (declare (special)) still feels ugly when the variable is already up there in the LET block, is there anything in the standard like a dynamic-let macro that does the declarations for you? 13:34:17 Captain_Thunder: there is not, but you can easily write such a macro. 13:34:23 lacedaemon [n=algidus@88-149-209-121.dynamic.ngi.it] has joined #lisp 13:34:32 Also, ISLISP has a different syntax for dynamic variables. 13:34:40 In Scheme they say that (let ((x 1)) ...) is equivalent to ((lambda (x) ...) 1) 13:34:41 -!- fe[nl]ix [n=algidus@88-149-210-139.dynamic.ngi.it] has quit [Nick collision from services.] 13:34:42 But ISLISP is not CL. 13:34:51 -!- lacedaemon is now known as fe[nl]ix 13:34:55 Well, I would never want to write that, I don't know if users of my library ever would. 13:35:01 But thanks for the help everyone :) 13:35:06 milaz: In CL it's also equivalent. 13:35:15 indeed :) 13:35:21 trebor pasted "error in thinking or colored-noisy (random)?" at http://paste.lisp.org/display/83229 13:35:55 milaz: but in CL, special variables are also processed in parameter lists: (defvar *c* 1) (defun f (*c*) (g)) (defun g () (print *c*)) (f 2) ==> prints 2 13:36:56 hello. i thought folding (random ) with any oszillating function would result in a sum about 0.0 - what am i thinking wrong? 13:36:56 -!- rread [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 13:36:59 rread [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has joined #lisp 13:37:17 trebor_dki: if you sum a very large number of times 0, you might reach 1. But only for a very very large number of 0. 13:37:34 matimago: Cool. You mean that during the call to f, *c* will all the time be bound to 2? 13:38:28 milaz: yes. 13:38:30 -!- blandest [n=blandest@85.204.33.242] has quit [Read error: 110 (Connection timed out)] 13:38:34 trebor_dki: random is not really random. 13:39:05 trebor_dki: what do "folding" and "oszillating" mean here? 13:39:14 pkhuong: it's clear from the code ;-) 13:39:18 blandest [n=blandest@softhouse.is.ew.ro] has joined #lisp 13:39:56 pkhuong: folding as i understand it is integrating a product of two functions 13:40:16 look like trebor_dki integrates a product of sinus and random function 13:41:21 pkhuong: i meant oscillating, like sinus or cosinus ... 13:41:46 periodic? 13:42:33 stassats: and chnging it's sign 13:42:37 *changing 13:43:46 trebor_dki: also, you're looping TO, instead of BELOW, so you have one more term for the 0 angle. 13:45:00 matimago: yes, sure .. (1- value-num) or below, you are right, but what irritates me, is, that the sum increases with increasing value-num. 13:45:19 trebor_dki: what increases is the standard deviation of your statistic. 13:45:42 -!- Davse_Bamse [n=davse@4306ds4-soeb.0.fullrate.dk] has quit ["leaving"] 13:45:44 trebor_dki: If you repeat your loop several times, the mean will be 0. 13:45:51 will tend to. 13:46:03 pkhuong: yes, but the standard deviation should decrease if value-num increases 13:46:14 trebor_dki: only if you divide by the number of repetitions. 13:46:22 matimago: increasing value-num should be equivalent to repeating the loop 13:47:29 jan247 [n=jan247@120.28.131.253] has joined #lisp 13:47:30 trebor_dki: you're summing, not taking the mean. The central limit theorem doesn't apply. The variance of a finite sum of independent variates is the sum of the variances. 13:47:48 introductory physics called; it wants its random walk back 13:48:13 -!- ignas [n=ignas@78-60-73-85.static.zebra.lt] has quit ["Ex-Chat"] 13:48:36 Xof: something about R ~ sqrt(N)? 13:48:55 Jasko2 [n=tjasko@209.74.44.225] has joined #lisp 13:49:07 *trebor_dki* googling about random walk.... 13:49:25 -!- Captain_Thunder [n=jimmymil@c-24-35-95-148.customer.broadstripe.net] has quit [] 13:50:21 that's the right idea. If physics is easier for you than statistics, you can think of adding together sines of multiple random numbers as being the projection onto one axis of a 2d random walk with fixed step size 13:50:30 -!- spilman [n=spilman@ARennes-552-1-123-234.w92-139.abo.wanadoo.fr] has quit [Read error: 110 (Connection timed out)] 13:50:55 giving you an expected distance from your starting point of l\times\sqrt(N) 13:51:06 spilman [n=spilman@ARennes-552-1-89-252.w92-139.abo.wanadoo.fr] has joined #lisp 13:51:38 oh my, as i am reading about random-walk, i start to remember ... about relative and absolute error, and about standard derivation .... you are so right Xof... 13:51:48 pkhuong is also right! 13:52:06 surely he is... 13:52:56 I was about to suggest adding cos and drawing the points to realize what's happening... 13:53:00 -!- Jabberwockey [n=jens@port-3757.pppoe.wtnet.de] has quit [Remote closed the connection] 13:54:56 *trebor_dki* admits that "random walk" flashed a red light, while "independent variates" didn't ;) 13:55:21 salex [n=user@216.80.147.206] has joined #lisp 14:00:55 -!- joachifm [n=joachim@bjo1-1x-dhcp154.studby.uio.no] has quit [Client Quit] 14:01:06 -!- free_thinker [n=willijar@134.151.144.246] has quit ["Leaving."] 14:01:18 Xof: using 'clisp -ansi -on-error abort' did the trick for me on both x86 and x86-64 14:01:26 you might want to mention that in INSTALL 14:01:36 -!- Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has quit [Remote closed the connection] 14:02:26 piso: nah, I want people to send me patches fixing the bugs that we have without -on-error abort :-) 14:02:35 ok 14:02:38 heh 14:02:58 besides, where would I get my air of omniscience from if it weren't for this kind of thing? 14:03:12 on x86-64: 14:03:15 Failure: float.pure.lisp / (SCALE-FLOAT-OVERFLOW BUG-372) 14:03:15 Failure: float.pure.lisp / (ADDITION-OVERFLOW BUG-372) 14:03:38 (no unexpected failures on x86) 14:04:06 the good news is, xcl is clearly faster on x86-64, and the same fast on x86 :) 14:04:06 Fare [n=Fare@98.216.111.110] has joined #lisp 14:05:15 *Fare* releases command-line-arguments on qitab. Meh. 14:05:32 piso: oh, really? Shows what my memory knows 14:05:46 maybe x will slow down once you implement the pretty printer :-) 14:06:03 maybe, but there's plenty of room for improvement in the compiler 14:06:32 style question: I've been asked to package xcvb with all its dependencies in a single tarball -- what should the directory tree look like? 14:07:10 Fare: i suggest to create a "dependencies" dir 14:07:11 the top being xcvb, with dependencies in direct subdirectories? the top being something different, with xcvb and dependencies each in its own? 14:07:42 *attila_lendvai* can't decide between the two 14:07:55 *Fare* thinks that xcvb is usable enough to deserve being used around, but that the difficulty of pulling it + dependencies is a barrier to adoption. 14:08:03 -!- Jasko [n=tjasko@75-149-33-105-SFBA.hfc.comcastbusiness.net] has quit [Read error: 110 (Connection timed out)] 14:08:03 the latter sounds cleaner, but then the toplevel dir is not xcvb really 14:09:14 what would I name the toplevel directory? xcvb-install ? 14:09:52 xcvb/ xcvb/README xcvb/lib/ xcvb/src/ 14:10:01 -!- tcr [n=tcr@host145.natpool.mwn.de] has left #lisp 14:10:44 cannot answer, so i guess i'd go with mega's setup. but i prefer dependencies to lib 14:11:12 what about xcvb/README xcvb/xcvb xcvb/cl-launch xcvb/asdf xcvb/asdf-dependency-grovel xcvb/closer-mop xcvb/command-line-arguments ? 14:11:31 skipping the lib directory 14:11:44 eni4ever [n=admin@79.114.1.251] has joined #lisp 14:11:48 Fare: i don't like that. doesn't make it clear that those stuff are not necessary, just there for convenience 14:12:10 *Fare* replaced calls to (directory ...**..) to run-program of find and sped up XCVB by a factor 10, making it usable. 14:12:18 I like skipping the lib directory, but I think the top-level directory in the tarball should have the version number: xcvb-1.2.3/ 14:12:38 piso: OK 14:13:04 is it "normal" that directory is so slow in SBCL? 14:13:22 (is it trying to do too much?) 14:13:35 it has to truename everything 14:13:45 I use :resolve-symlinks nil, though 14:14:02 dunno, then. look at the source, profile, and see what it does? 14:14:10 -!- jdz [n=jdz@85.254.211.133] has quit [Read error: 110 (Connection timed out)] 14:14:35 probably should. Adding that to the TODO list (but low) 14:15:43 -!- sellout [n=greg@75-25-126-88.lightspeed.sjcpca.sbcglobal.net] has quit [] 14:16:24 -!- lemoinem [n=swoog@modemcable110.189-201-24.mc.videotron.ca] has quit [Read error: 113 (No route to host)] 14:18:11 -!- frozsyn [n=frozsyn@blm93-2-82-229-63-104.fbx.proxad.net] has quit [Read error: 113 (No route to host)] 14:18:15 attila_lendvai, well, all the dependencies are necessary -- the fact that I bundle them together in a tarball would be for convenience 14:18:55 Fare: i mean, they are not part of the thing... if one has them in their workspace then they are free to not use them 14:19:11 sure 14:20:40 athos [n=philipp@92.250.250.68] has joined #lisp 14:21:55 mattrepl [n=mattrepl@204.194.78.3] has joined #lisp 14:23:00 -!- blandest [n=blandest@softhouse.is.ew.ro] has quit [Read error: 110 (Connection timed out)] 14:23:28 blandest [n=blandest@85.204.33.242] has joined #lisp 14:24:27 fisxoj [n=fisxoj@149.43.252.13] has joined #lisp 14:25:42 -!- fvw [n=sdfpme@113.77.216.106] has quit [Remote closed the connection] 14:26:56 so xcvb-0.4/{xcvb,dependencies/{asdf,asdf-dependencies-grovel...}} ? 14:28:13 Fare: that seems fine 14:28:25 -!- ThomasI [n=thomas@unaffiliated/thomasi] has quit ["Bye Bye!"] 14:31:40 what compression for the tarball? gzip? bzip2? lzma? 14:31:45 -!- nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has quit [Read error: 104 (Connection reset by peer)] 14:31:57 nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 14:32:00 gzip unless it's huge (then bzip) 14:32:26 -!- anekos is now known as A_anekos 14:32:54 LiamH [n=none@208.72.159.207] has joined #lisp 14:33:00 -!- Quadrescence [n=quad@unaffiliated/quadrescence] has quit [Read error: 104 (Connection reset by peer)] 14:33:16 Quadrescence [n=quad@unaffiliated/quadrescence] has joined #lisp 14:35:25 ok -- what about installation paths? 14:35:29 to /usr/local ? 14:35:32 tmh [n=thomas@pdpc/supporter/sustaining/tmh] has joined #lisp 14:35:36 Greetings. 14:36:07 tmh: hi 14:36:25 -!- eni4ever [n=admin@79.114.1.251] has quit [Client Quit] 14:36:51 overridable with INSTALL_DIR=foo ? 14:37:37 nah, with PREFIX=/usr/local/stow/xcvb-0.4 14:38:30 I'd prefer something like SBCL: just untar and use it 14:40:28 milanj [n=milan@93.87.193.108] has joined #lisp 14:41:16 rstandy [n=rastandy@net-93-144-170-242.t2.dsl.vodafone.it] has joined #lisp 14:41:47 -!- cracki [n=cracki@134.130.183.101] has quit ["If technology is distinguishable from magic, it is insufficiently advanced."] 14:42:46 milaz: SBCL installs in /usr/local/bin by default. 14:44:03 -!- daniel__ is now known as daniel 14:44:08 -!- Zhivago [n=zhivago@li49-59.members.linode.com] has quit [Read error: 60 (Operation timed out)] 14:44:10 "just untar and use it" -- you mean, usable without installation? That would suppose standalone executables, which I guess is possible with clisp or sbcl. 14:44:10 Fare: of course, I mean that it would be nice to be able to avoid writing to system directories in some cases, as SBCL allows to do. 14:44:26 *Fare* hasn't tried compiling xcvb with clisp in a while... 14:45:21 xcvb doesn't have any executables in it, does it? 14:46:37 once xcvb is built on the local system, can you just LOAD it? 14:47:25 sytse [i=sytse@speedy.student.ipv6.utwente.nl] has joined #lisp 14:47:53 Xof: is it normal that fasl still contain the universal time of production? 14:48:16 isn't that both redundant with filesystem timestamp and a PITA for reproducibility? 14:48:22 bgs100 [n=ian@unaffiliated/bgs100] has joined #lisp 14:48:26 determinism / whatever you call it 14:48:50 fvw [n=sdfpme@113.77.216.106] has joined #lisp 14:49:30 -!- shmho [n=user@58.142.15.103] has quit [Remote closed the connection] 14:49:39 sbcl/src/compiler/dump.lisp:326:#-sb-xc-host (format-universal-time nil (get-universal-time)) 14:49:59 same for machine-instance 14:50:09 sepult [n=user@xdsl-87-78-24-142.netcologne.de] has joined #lisp 14:50:14 -!- kpreid [n=kpreid@cpe-67-249-58-190.twcny.res.rr.com] has quit [] 14:50:16 would it be OK for me to just remove those fields from the FASL? 14:50:42 tmh pasted "Summing Array Elements" at http://paste.lisp.org/display/83231 14:50:45 kpreid [n=kpreid@cpe-67-249-58-190.twcny.res.rr.com] has joined #lisp 14:50:58 fiveop [n=fiveop@g229076131.adsl.alicedsl.de] has joined #lisp 14:51:38 That paste was motivated by a c.l.l post by Francogrex 14:53:07 Zhivago [n=zhivago@li49-59.members.linode.com] has joined #lisp 14:53:07 -!- fiveop [n=fiveop@g229076131.adsl.alicedsl.de] has quit [Client Quit] 14:53:11 HET2 [i=diman@128.131.95.22] has joined #lisp 14:53:14 -!- ruediger [n=ruediger@62-47-143-82.adsl.highway.telekom.at] has quit ["Leaving"] 14:53:19 fiveop [n=fiveop@g229076131.adsl.alicedsl.de] has joined #lisp 14:53:29 Fare: Well, i can see usage of such data (maybe as optional thing?) in case of commercial deployments 14:53:31 As a side note, having multiple CL implementations in this case is particularly useful for sanity checking. 14:53:53 blandest1 [n=blandest@softhouse.is.ew.ro] has joined #lisp 14:53:58 -!- Chris [n=chris@unaffiliated/chris] has quit ["User pressed ^C five times."] 14:54:06 -!- blandest [n=blandest@85.204.33.242] has quit [Read error: 104 (Connection reset by peer)] 14:54:22 so BUILD.lisp is xcvb's replacement for an .asd file? 14:54:39 nha [n=prefect@137-64.105-92.cust.bluewin.ch] has joined #lisp 14:55:20 p_l: I can't -- if it's a commercial deployment, you know what the compiling machine was, and the original timestamp, and even the md5sum -- you have saved that in your archive. 14:55:25 Blkt [n=Blkt@dynamic-adsl-94-37-247-157.clienti.tiscali.it] has joined #lisp 14:55:32 and I'd trust these more than whatever the FASL says 14:55:57 lichtblau, yes -- unless you convince me otherwise in the next few weeks 14:56:02 lexa_ [n=lexa_@seonet.ru] has joined #lisp 14:56:08 Okay, just checking my understanding. 14:56:09 Fare: What if you are releasing small, incremental patches in form of fasls that are loaded during startup, with rest of the application stored in image? 14:56:09 Is there reverse compatibility, i.e. would it be possible to write an ASDF extension that makes it possible for ASDF users to load a project the old way through ASDF, even if the project maintains only a BUILD.lisp definition xcvb? 14:56:25 p_l, I still trust the md5sum more than whatever the FASL itself says 14:56:30 -!- lexa_ is now known as Guest66056 14:56:50 Fare: just playing devil's advocate :) 14:56:51 -!- smoofra [n=user@cthulhu.elder-gods.org] has left #lisp 14:56:58 smoofra [n=user@cthulhu.elder-gods.org] has joined #lisp 14:57:07 lichtblau, it's planned, but not currently implemented (an old implementation has bitrotten heavily) 14:57:51 *lichtblau* looks forward to "make -j 2 -C clbuild/source/mcclim" working out-of-the-box 14:57:57 lichtblau, and not very high on my priority list 14:57:58 i'm not sure i like the full capitals of BUILD.lisp 14:58:15 attila_lendvai, I had to make it stand out somehow 14:58:51 leading non-breaking spaces 14:58:55 (that is not a serious suggestion) 14:58:58 clearly, there should be colourful pathnames 14:59:15 gah 14:59:47 windows with its non-case-sensitivity, etc... it doesn't help much. if people know xcvb, then they'll know build.lisp. if they don't know xcvb, then looking into BUILD.lisp will not make their lives much easier... (but i shut up until i actually start using xcvb, which may happen soonish) 14:59:55 to make attila_lendvai happy, use small caps instead 15:00:12 -!- kuwabara [n=kuwabara@84.14.121.138] has quit [Read error: 54 (Connection reset by peer)] 15:00:29 __build__.lisp 15:00:36 (I'm not serious) 15:00:53 Although, in Python, there is __init__.py for a package 15:01:10 build.xcvb? 15:01:14 saikat_ [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has joined #lisp 15:01:19 .lisp 15:01:36 build!.lisp 15:01:39 rjbs [n=rjbs@cancer.codesimply.com] has joined #lisp 15:01:42 kuwabara [n=kuwabara@84.14.121.138] has joined #lisp 15:01:52 Is Brucio's Lisp at Light Speed archived anywhere? 15:02:12 per lisp preferred punctuation it should be --build--.lisp :-) 15:02:19 i like build.xcvb, and having proper annotations for emacs can turn it into lisp mode for the lazy 15:03:07 kpreid: or *build*.lisp, but such filename may be impossible in windows 15:03:39 and fun to use from shell 15:04:16 *mega1* blends in with the crowd 15:04:27 BUILD = bad, build = good 15:05:45 .lisp 15:06:56 Unicode is really big, we could go on for ages. 15:07:05 "-- *.lisp" 15:07:29 Seriously, maybe "build.lisp" is okay. People using ASDF know to look for package definitions in "packages.lisp", and this is just a convention. They learned it, and as well it won't be hard to learn about "build.lisp" 15:08:04 -!- HET2 [i=diman@128.131.95.22] has quit ["Leaving"] 15:08:14 someone should make a screencast with "learn build.lisp" 15:08:15 the problem with that is if someone out there actually wants to call one of their source files "build.lisp", they can't any more 15:08:26 what are reasons for build.lisp against BUILD.lisp ? 15:08:42 -!- rjbs [n=rjbs@cancer.codesimply.com] has left #lisp 15:08:44 Xof: what do you propose? 15:08:44 build.xcvb wins, really. 15:08:51 I agree with luis 15:08:58 .xcvb makes for a poor extension 15:09:08 aesthetics 15:09:09 -!- xan [n=xan@conference/gcds/x-29c0b28be1d70076] has quit [Read error: 110 (Connection timed out)] 15:09:14 Captain_Thunder [n=jimmymil@c-24-35-95-148.customer.broadstripe.net] has joined #lisp 15:09:20 the main problem with BUILD.lisp is that someone is going to go with that, create a build.lisp source file, and get screwed when their filesystem is HFS 15:09:27 serichsen [n=harleqin@xdsl-81-173-189-82.netcologne.de] has joined #lisp 15:09:39 Granted. 15:09:54 you need something that does not have a potential collision with an actual lisp source file 15:09:57 can't run into trouble with logical pathnames? 15:09:58 build.xcvb is fine 15:10:02 yup 15:10:12 "makes for a poor extension" is a really stupid argument 15:10:21 ok 15:10:26 I don't even think it makes for a poor extension 15:10:32 in true asdf tradition it's going to be build.xcv to drive me crazy 15:10:34 demmel [n=demmel@e058.tum.vpn.lrz-muenchen.de] has joined #lisp 15:10:40 mega1: haha 15:10:59 in any case -- if it has to change from BUILD.lisp, now is the right time to do it 15:11:33 -!- TR2N [i=email@89.180.203.189] has left #lisp 15:11:44 In are several patches for paredit in the form of elisp files to load after paredit to modify its behaviour. Anyone interested is welcome to test these; I wish to solicit feedback. 15:12:42 __build__.lisp has indeed the "appeal" of being like python's __init__.py (if that's any good) 15:12:56 At the top of each elisp file in that directory is a blurb explaining what it does. 15:14:04 so, does everyone agree with BUILD.xcvb ? 15:14:12 Why uppercase? 15:14:16 lower case, please 15:14:26 "no" 15:14:29 does everyone agree with build.xcvb ? 15:14:32 yes 15:14:34 i like it 15:14:39 me three. 15:14:51  15:15:03 buIld.lisp 15:15:16 slyrus_: with a zero-width nonbreaking space in the middle? 15:15:18 I mean buIld.xcvb 15:15:20 Fare: __build__.lisp has a grain of sanity -- it doesn't look bad and it doesn't look like a filename that someone else is likely to collide with, and it stands out without problem with special symbols 15:15:20 heh 15:15:21 the nice thing about .xcvb (instead of __build__.lisp, for example) is that it sends the user off to google for xcvb, if all else fails 15:15:23 KingNatoG5 [n=patrik@84-217-4-121.tn.glocalnet.net] has joined #lisp 15:15:24 what possible benefit of upper case do you see? 15:15:42 trolling 15:15:42 salex: standing out. As for Makefile 15:16:02 "here's where to start reading the source code" 15:16:07 ah. I guess I never liked "Makefile" either ;) 15:16:08 can't we just use .xcvb? 15:16:20 milaz: __build__.lisp looks extremely bad (: 15:16:35 I think it should accept any synonym for `make' or `build': do, construct, prepare, compile, create, &c. Maybe also accept faire, hacer, &c. 15:16:35 schme: tastes differ 15:16:36 is there a unicode char yet for the CONS cell symbol? 15:16:36 slyrus_: and then everyone can refer to it as "the file formerly known as BUILD.lisp" 15:16:50 milaz: Indeed :) 15:17:22 salex: ex-build.lisp ? 15:17:34 xcvbuild.lisp ? 15:17:59 if you want it to stand out, .xcvb seems to be a nice way. 15:18:03 *tmh* regrets forgetting to (setf *print-array* nil) 15:18:14 __xcvb__.lisp 15:18:32 Xof: I second that! 15:18:52 x.cvb 15:18:54 e.g. SBCL does have Python, so... ,) 15:19:11 and the argument of piso of googling works in this case 15:19:16 yes, just call the file ".xcvb" -- nice way to stand out 15:19:30 "Where did my file go?!" 15:19:32 Hopefully there are no newcomer in this channel now given this awesome display expertise on our part. 15:19:38 Heh, not what I meant, but that's funny. 15:19:39 Fare: and a good way to stress cl:directory and friends! 15:19:40 -!- serichse1 [n=harleqin@xdsl-81-173-185-169.netcologne.de] has quit [Read error: 110 (Connection timed out)] 15:19:42 .xcvb isn't bad 15:19:48 :xcvb maybe? 15:19:57 erm, source files should not be invisible. 15:19:59 (or does that screw things up in OS X land?) 15:20:04 piso, tic: they will remember it good, after they will finally find the file 15:20:14 -!- blandest1 [n=blandest@softhouse.is.ew.ro] has quit ["Leaving."] 15:20:16 tic: : is permitted but confusing 15:20:21 who likes __xcvb__.lisp ? 15:20:23 \xcvb/ 15:20:28 (or dislikes it) 15:20:31 no :'s or *'s in filenames please 15:20:33 Fare: it's better than a lot of bad options 15:20:34 milaz, meanwhile they'll (read: I) will be wanting to bash someone's head for picking the dot.. 15:20:34 I stand with piso, build.xcvb 15:20:36 *stassats* dislikes __ 15:21:03 build.xcvb is my final answer 15:21:03 xcvb is a good extension, but does 'build' really represent everything it does? 15:21:10 build.xcvb has the lead so far. 15:21:11 *milaz* finds __ the best in practice 15:21:16 __ is horrible. 15:21:20 __.xcvb ? 15:21:24 (definition.xcvb or something) 15:21:29 maybe xcvb-build.lisp 15:21:32 milaz: go gack to python! 15:21:42 -!- Guest66056 [n=lexa_@seonet.ru] has left #lisp 15:21:52 (I don't care; this is the sum total of the effort I will spend thinking about xcvb 15:21:54 (.xcvb) ;) 15:21:58 xcvb.lisp? 15:22:07 Xof: sorry for the noise 15:22:33 luis: no, thanks :) 15:22:34 *Fare* would rather make a bit of noise now than a lot after people try to use it 15:22:43 who likes xcvb.lisp ? 15:22:54 or dislikes it 15:23:09 is an xcvb file a series of lisp *forms*? 15:23:11 *slyrus_* prefers build.xcvb 15:23:23 kpreid, not really. It's a sexp language 15:23:25 xcvb.lisp works for me 15:23:38 Fare: then it should not be named .lisp, IMO 15:23:46 if it's not Lisp, then it should not be named .lisp 15:23:49 lisp.xcvb ;) 15:23:56 .lisp-expr as used in SBCL or .sexp 15:23:57 what kpreid said 15:24:09 xcvb.sexp ? 15:24:22 feels funny, but sensible 15:24:26 How about simply `xcvb'? 15:24:40 extensions are nice for various tools 15:24:58 OTOH, in ASDF they call it .asd 15:25:11 why not .xcvb 15:25:24 is a terrible name 15:25:39 tmh annotated #83231 "Summing Array Elements in SBCL, CCL & LispWorks" at http://paste.lisp.org/display/83231#1 15:25:44 is meta-variable 15:25:49 Yeah, is a much better name. 15:25:55 Muld [n=wr23@88-196-36-122-dsl.noe.estpak.ee] has joined #lisp 15:26:09 lol 15:26:17 Riastradh: you mean <quelque%20chose> ? 15:26:20 tmh: is 10000 important? what if it's 1000? 15:26:34 piso: 1000 works, the dimensions matter. 15:26:48 where does it break? 15:26:54 xcvb.sexp communicates well 15:27:01 -!- Captain_Thunder [n=jimmymil@c-24-35-95-148.customer.broadstripe.net] has quit [] 15:27:15 tmh: I'm trying to play with it, but I don't seem to have enough memory for 10000 15:27:18 milaz: the lesson ASDF taught us is that a `system.asd' scheme fails in a subtle but important manner if it doesn't enforce that the .asd file contains the definition for system _and only_ that definition. 15:27:25 Fare, looks even better! But maybe we ought to use guillemets, rather than comparison signs. 15:28:01 piso: I inserted a format statement and it simply stops incrementing the sum a 1.6777216e7. 15:28:03 tmh: with the loop version, you end up adding 10000.0 + 10000.0 + 10000.0 10000 times 15:28:17 with the other versions, you add 1.0 + 1.0 + 1.0 100000000 times 15:28:30 *milaz* thinks that what in Common Lisp is sexp, it Scheme would be sex? 15:28:32 now go read "What every computer scientist should know about floating point" 15:28:52 milaz: sxp perhaps. 15:29:07 Xof: Ah, okay, I have read that. 15:29:19 *Sikander* wonders if xcvb also works without using make. 15:29:52 tmh: then my two-line explanation should make it obvious 15:30:04 Xof: most-positive-single-float 3.4028235e38 15:30:15 xcvb.sexp? 15:30:18 Sikander, it will, not yet though; make is the first "target" 15:30:21 tmh: go and read it again, then 15:30:33 *Sikander* sighs in relief 15:30:55 tmh: (log 1.6777216e7 2) might be enlightening. 15:31:36 dcrawford: will it always support make? 15:31:41 may I suggest build.xcvb? 15:32:18 pkhuong: Ah, yes. 15:32:45 CL-USER(20): (+ 1.6777216e7 1.0) 15:32:46 1.6777216e7 15:33:05 ok, sold for build.xcvb 15:33:15 Fare: excellent! 15:33:18 +1 for build.xcvb 15:33:39 Kahan's summation algorithm and/or double floats may be helpful. 15:33:43 tmh: single-float-epsilon is the constant that you were looking for 15:34:31 must this file have only one top-level form in it, and that must be MODULE form? 15:36:47 legumbre_ [n=user@r190-135-69-89.dialup.adsl.anteldata.net.uy] has joined #lisp 15:38:00 OK. Assuming I'm trying to get the CFASL patch committed upstream, where do I add documentation for it? 15:38:46 -!- fisxoj [n=fisxoj@149.43.252.13] has quit [Read error: 113 (No route to host)] 15:39:19 *antifuchs* falls into javascript's NIF trap 15:39:24 Xof: pkhuong: Thanks, that makes it clear. I was expecting an exception of some sort. 15:39:31 In that situation. 15:39:31 -!- stassats [n=stassats@wikipedia/stassats] has quit [Remote closed the connection] 15:39:52 Xach: brucio paints an accurate picture of the future of programming 15:39:53 stassats [n=stassats@wikipedia/stassats] has joined #lisp 15:40:03 there will be only war? 15:40:25 and NIF 15:41:21 antifuchs: what is NIF? 15:41:26 minion: what does NIF stand for? 15:41:26 Nonentres Ivorytype Fovea 15:41:52 introducing magic files with user-usable extensions is unhygienic. so xcvb.lisp is obviously bad 15:42:06 but what do I know 15:42:19 milaz: "natural" if, a bad idea that Brucio once wrote about 15:42:51 unfortunately, bruce deleted his weblog 15:42:57 he must have thought it's embarrassing (: 15:42:57 blackened` [n=blackene@89.102.28.224] has joined #lisp 15:43:36 ah, maybe something left in web.archive.org ? 15:43:46 possibly. it's a blogspot blog 15:43:49 "was" 15:44:01 antifuchs: do you remember the url? 15:44:10 http://brucio.blogspot.com/ 2004/12/beauty-in-motion.html was it, I think 15:44:33 Fare: How usable is XCVB right now? 15:44:34 -!- alinp [n=alinp@86.122.9.2] has left #lisp 15:44:35 cmm: what do you mean? 15:44:43 http://web.archive.org/web/20070524180812/http://brucio.blogspot.com/2004/12/beauty-in-motion.html 15:44:59 p_l: will be better tomorrow. But usable today if you care to download the dependencies, configure asdf and build 15:45:19 http://lemonodor.com/archives/2005/03/brucio_unleashe.html also has paul graham's take 15:45:33 Fare: it infringes on the user's namespace, depriving him of one possible (and very meaningful) file name 15:45:40 I'm working on packaging a pre-bootstrapped version of xcvb with dependencies that doesn't require you to configure asdf. 15:45:41 hahaha 15:45:50 *stassats* shivers at GIF 15:46:14 and here's a description of what is in javascript: http://www.sitepoint.com/blogs/2009/07/01/javascript-truthy-falsy/ <- notice the similarities /-: 15:46:19 Bigshot_ [n=BIG_SHOT@CPE002129abc864-CM001ac35cd4d0.cpe.net.cable.rogers.com] has joined #lisp 15:47:03 *sigh* Ah, arc's cond-like if... 15:48:52 might make it simpler to write lock-free programs on SBCL/x86-64. 15:49:47 -!- sytse [i=sytse@speedy.student.ipv6.utwente.nl] has quit [Read error: 60 (Operation timed out)] 15:50:01 fisxoj [n=fisxoj@149.43.252.13] has joined #lisp 15:50:21 cmm: isn't that reasoning valid whichever way I name the file? What should I use instead? directory meta-information stored in a central database? 15:50:27 -!- sepult [n=user@xdsl-87-78-24-142.netcologne.de] has quit ["ERC Version 5.2 (IRC client for Emacs)"] 15:51:11 -!- aerique [i=euqirea@xs2.xs4all.nl] has quit ["..."] 15:52:24 -!- legumbre [n=user@r190-135-34-181.dialup.adsl.anteldata.net.uy] has quit [Connection timed out] 15:52:59 ... 15:53:03 *p_l* had just seen GIF 15:53:05 -!- fvw [n=sdfpme@113.77.216.106] has quit [Read error: 110 (Connection timed out)] 15:53:16 -!- demmel [n=demmel@e058.tum.vpn.lrz-muenchen.de] has quit [] 15:54:02 I wonder if I should pluck my eyes out right now or start by torturing its author 15:54:10 pitui [n=pitui@135.207.174.197] has joined #lisp 15:54:24 ha-ha 15:54:26 xan [n=xan@193.145.150.120] has joined #lisp 15:54:32 brucio was ahead of his time. some of his ideas showed up in arc. 15:54:41 milanj- [n=milan@93.87.193.108] has joined #lisp 15:54:48 Brucio's blog is cruel. The comments are heart breaking. 15:55:12 Fare: I think cmm's point was that the .lisp extension is looking for trouble, since the user will think he knows what that means. The .xcvb extension should be safe. 15:55:16 luis: was cruel. 15:55:21 GIF :-) :-) 15:55:25 Xach: you meant he was running so fast that he didn't notice the slaughterhouse? 15:55:51 at the speed of light 15:55:58 rouslan [n=Rouslan@unaffiliated/rouslan] has joined #lisp 15:56:12 p_l: what's GIF? 15:56:30 a trafalmadorian macro. 15:56:38 rsynnott: check brucio's archive url above. 15:57:02 rsynnott: something used to torture sensible programmers 15:57:51 Here's a better IF: (altif cond then else then else then else ...) So you don't need to count the characters (altif cond (print 'hi) (do-something-else-with-a-longer-name) (print 'lo) nil (print 'done)) :-) 15:58:21 -!- drafael [n=tapio@ip-118-90-139-235.xdsl.xnet.co.nz] has quit [Connection timed out] 15:58:34 oh, dear 15:58:40 If you tab the code, you can even put the then expressions in the left column and the else expressons on the right! :-) 15:58:45 heheh, an empty hash table 15:59:01 *milaz* has to think about something scarrier 15:59:10 demmel [n=demmel@c215.tum.vpn.lrz-muenchen.de] has joined #lisp 15:59:27 Swords2 [n=kohii@pool-71-112-25-59.sttlwa.dsl-w.verizon.net] has joined #lisp 15:59:37 for example, conditions must start in odd column numbers, and the corresponding operators from even ones. 15:59:40 though how is he finding the line length of the terms? :) 15:59:52 rsynnott: format nil, i suppose 15:59:53 princ-to-string 15:59:56 Hun [n=hun@p50993726.dip0.t-ipconnect.de] has joined #lisp 16:00:00 or prin1. 16:00:32 and spaces in between? 16:00:43 *p_l* would prefer simple thing like (milf (values) ((cond)(form))((cond)(form))) but it's still ridiculous 16:01:23 milf? 16:01:45 luis: I couldn't help myself :> 16:02:13 I recall someone putting in comment a series of names like if, gif, nif, milf, wtf ;_) 16:02:13 minion: what does milf stand for? 16:02:14 Megalodactylism Irrubrical Loafer Ferryway 16:03:09 megalodactilysm, megalodactilysm... 16:03:10 okflo [n=user@91-115-89-64.adsl.highway.telekom.at] has joined #lisp 16:03:23 -!- milanj [n=milan@93.87.193.108] has quit [Read error: 110 (Connection timed out)] 16:03:50 postamar [n=postamar@x-132-204-254-42.xtpr.umontreal.ca] has joined #lisp 16:04:00 rread_ [n=rread@nat/sun/x-a776eec7d0f16262] has joined #lisp 16:05:11 Xach: well, yes, but he'd still need to detect line breaks :) 16:06:42 -!- mvilleneuve [n=mvillene@ABordeaux-253-1-86-153.w81-250.abo.wanadoo.fr] has quit ["leaving"] 16:06:51 can such reader macro be defined that starts working from one opening bracket, i.e. #\( ? 16:06:54 no, that's why he's using length instead of line-breaks 16:07:02 -!- Riastradh [n=rias@pool-141-154-207-110.bos.east.verizon.net] has quit [Read error: 60 (Operation timed out)] 16:07:11 doh 16:07:12 Riastradh [n=rias@pool-141-154-207-110.bos.east.verizon.net] has joined #lisp 16:07:14 :) 16:07:24 -!- Swordsman [n=kohii@pool-71-112-25-59.sttlwa.dsl-w.verizon.net] has quit [Read error: 110 (Connection timed out)] 16:07:28 that's wasn't addresed to you 16:07:33 s/'s// 16:08:44 oh, so really he should take that approach? 16:09:07 blandest [n=blandest@softhouse.is.ew.ro] has joined #lisp 16:10:47 tic: no, I don't know anything about Spotify. 16:10:55 is sbcl currently frozen or open? 16:11:10 frozen 16:11:17 -!- rread [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has quit [Read error: 110 (Connection timed out)] 16:11:53 -!- milaz [n=user@85.175.41.248] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 16:14:11 -!- nikodemus [n=nikodemu@cs181229041.pp.htv.fi] has quit ["This computer has gone to sleep"] 16:14:54 beach [n=user@ABordeaux-158-1-36-70.w90-55.abo.wanadoo.fr] has joined #lisp 16:14:57 Good evening. 16:15:17 lexa_ [n=lexa_@83.222.5.112] has joined #lisp 16:15:45 -!- lexa_ is now known as Guest69368 16:16:00 -!- Muld [n=wr23@88-196-36-122-dsl.noe.estpak.ee] has quit [Connection timed out] 16:17:08 odin_ [n=mark@nom27669d.nomadic.ncsu.edu] has joined #lisp 16:18:16 -!- marcoecc [i=me@gateway/gpg-tor/key-0x9C9AAE7F] has quit [Remote closed the connection] 16:19:55 -!- AntiSpamMeta [n=MetaBot@unaffiliated/afterdeath/bot/antispambot] has quit [Read error: 60 (Operation timed out)] 16:20:35 marcoecc [i=me@gateway/gpg-tor/key-0x9C9AAE7F] has joined #lisp 16:20:47 AntiSpamMeta [n=MetaBot@unaffiliated/afterdeath/bot/antispambot] has joined #lisp 16:21:25 danlei [n=user@pD9E2ED7C.dip.t-dialin.net] has joined #lisp 16:22:20 -!- odin_ [n=mark@nom27669d.nomadic.ncsu.edu] has quit [] 16:22:32 amnesiac [n=amnesiac@p3m/member/Amnesiac] has joined #lisp 16:25:53 -!- Demosthenes [n=demo@173.86.192.171] has quit ["leaving"] 16:27:04 masm [n=masm@213.22.191.93] has joined #lisp 16:27:28 cracki [n=cracki@sglty.kawo2.RWTH-Aachen.DE] has joined #lisp 16:27:33 ikki [n=ikki@201.155.75.146] has joined #lisp 16:29:21 -!- nha [n=prefect@137-64.105-92.cust.bluewin.ch] has quit [Remote closed the connection] 16:29:26 nikodemus [n=nikodemu@cs181144086.pp.htv.fi] has joined #lisp 16:30:02 -!- rouslan [n=Rouslan@unaffiliated/rouslan] has quit [Remote closed the connection] 16:31:21 xan_ [n=xan@conference/gcds/x-d09f1ec400f9a52d] has joined #lisp 16:31:43 how pretentious to announce good evening 16:32:05 -!- okflo [n=user@91-115-89-64.adsl.highway.telekom.at] has quit [Remote closed the connection] 16:32:40 -!- xan [n=xan@193.145.150.120] has quit [Read error: 60 (Operation timed out)] 16:32:51 although sbcl on os x is single threaded, it seems to use both cores evenly 16:33:46 or you seem to think it uses both cores 16:34:01 leo2007: as in, it jumps back and forth between both cores? 16:35:40 REPLeffect [n=REPLeffe@69.54.115.254] has joined #lisp 16:36:01 or you think it is single-threaded 16:36:17 leo2007: if it's single threaded and jumps around, that's not necessarily good :D 16:36:32 Fare: aroundp 16:36:52 when I run a long simulations, the cpu usage showed up on both cores 16:37:24 it is single threaded as shown in the Activity Montitor 16:37:30 letexpx [n=letexpx@LPuteaux-156-16-28-243.w82-127.abo.wanadoo.fr] has joined #lisp 16:37:54 I wonder if most users enable mutlti-threading on os x? 16:38:48 *slyrus_* does 16:39:15 if i had os x, i'd do too 16:40:07 leo2007: shows up on both cores but not simultaneously, right? 16:40:52 -!- attila_lendvai [n=ati@89.132.189.132] has quit ["..."] 16:41:05 lispm [n=joswig@e177123134.adsl.alicedsl.de] has joined #lisp 16:42:18 -!- pjb [n=t@85-169-63-25.rev.numericable.fr] has left #lisp 16:42:45 pjb [n=t@85-169-63-25.rev.numericable.fr] has joined #lisp 16:43:00 gh7d395pi69wd: it's not an announce, it's a wish. 16:43:09 good morning all 16:43:18 :) 16:43:53 hahahah (reading scrollback), brucio how i miss you. 16:44:33 luis: I forgot to pay attention whether they are simultaneous or not, I will do that in a moment. 16:44:44 Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has joined #lisp 16:45:33 mrsolo [n=mrsolo@adsl-68-126-215-139.dsl.pltn13.pacbell.net] has joined #lisp 16:45:56 leo2007: my guess is that the OS scheduler is migrating your process from one core to another from time to time. 16:46:20 t 16:46:26 luis: how to see if they are simultaneous or not? 16:46:59 that'd be quite strange if they are 16:48:06 -!- willb [n=wibenton@h69-129-204-3.mdsnwi.broadband.dynamic.tds.net] has quit [Success] 16:48:30 leo2007: I don't remember how the Activity Monitor works, but doesn't it show one bar per cpu/core? 16:49:14 -!- Edico [n=Edico@unaffiliated/edico] has quit [Read error: 60 (Operation timed out)] 16:49:44 fe[nl]ix: pong 16:50:35 luis: yes, I use istat to watch it, when one core is high the other is low 16:50:55 -!- letexpx [n=letexpx@LPuteaux-156-16-28-243.w82-127.abo.wanadoo.fr] has quit ["Leaving"] 16:51:12 leo2007: so the process is sometimes running one one core somtimes on the other. 16:51:20 drewc: I think I'll go with UCW/LOL after all... good thing I've got SICP here, I haven't yet done anything with continuations :D 16:51:26 deafmacro [n=user@59.96.207.140] has joined #lisp 16:54:28 -!- dialtone [n=dialtone@unaffiliated/dialtone] has quit [Read error: 60 (Operation timed out)] 16:54:49 -!- nikodemus [n=nikodemu@cs181144086.pp.htv.fi] has quit [Read error: 110 (Connection timed out)] 16:55:26 leo2007: the OS schefuler controls this, not SBCL. 16:55:52 -!- silenius [n=jl@yian-ho03.nir.cronon.NET] has quit [Remote closed the connection] 16:56:00 nikodemus [n=nikodemu@cs181144086.pp.htv.fi] has joined #lisp 16:56:20 how exactly do dynamic and lexical bindings play together for one symbol? 16:57:20 if there is a lexical binding and a dynamic binding, the lexical binding takes precedence? 16:57:35 unless there was a (declare (x special)) 16:57:37 p_l: cool! 16:57:47 demmel: they're exclusive one of the other in a given scope. 16:58:03 p_l: for the most part, you don't actually need to know or understand continuations to use UCW. 16:58:04 the default is lexical 16:58:30 p_l: it helps though :) 17:00:31 drewc: still, I find it hilarious that parts of the application that definitely don't need UCW will load it anyway due to ROFL xD 17:00:35 Edico [n=Edico@unaffiliated/edico] has joined #lisp 17:00:38 abdullah` [n=user@122.164.242.140] has joined #lisp 17:02:22 syamajala [n=syamajal@140.232.179.15] has joined #lisp 17:04:11 -!- abdullah` [n=user@122.164.242.140] has quit [Remote closed the connection] 17:05:13 which forms can establish lexical variable bindings? let, let*, prog, labels, flet, lambda (the latter 3 only for the bodies of the functions) 17:05:32 dialtone [n=dialtone@adsl-99-136-101-166.dsl.pltn13.sbcglobal.net] has joined #lisp 17:05:45 (I'm looking for all the ways a symbol macro could be shaddowed) 17:06:39 -!- mrsolo [n=mrsolo@adsl-68-126-215-139.dsl.pltn13.pacbell.net] has quit [Read error: 110 (Connection timed out)] 17:06:59 special forms? 17:07:44 stassats: I'm looking for a comprehensive list. I dont care about implementation specific special forms 17:08:06 -!- slyrus_ [n=slyrus@adsl-76-241-19-133.dsl.pltn13.sbcglobal.net] has quit [Read error: 110 (Connection timed out)] 17:08:07 special forms are standardized 17:08:19 okflo [n=user@91-115-89-64.adsl.highway.telekom.at] has joined #lisp 17:08:46 stassats: so did i forget any? 17:09:08 clhs 3.1.2.1.2.1 17:09:09 http://www.lispworks.com/reference/HyperSpec/Body/03_ababa.htm 17:09:25 -!- yvdriess [n=yvdriess@soft85.vub.ac.be] has quit [] 17:09:52 p_l: the dependencies are somewhat coarsely defined right now ... that's something i need to work on 17:10:56 stassats: I'm very familiar with that list. But I'm curious if i missed anything when saying: let, let*, prog, flet, labels and (function (lambda ..)) are the only special forms to establish lexical bindings? 17:11:31 well, block too 17:12:42 and catch 17:13:10 uh... when I have a variable *var* holding a list like '(foo bar baz) and want to supply it to a function demanding a &rest args like (foo (&rest var) ...), how can I get it spliced to get (foo 'foo 'bar 'baz) instead of (foo '(foo bar baz))? 17:13:14 stassats: oh. I should have said variable bindings. 17:13:15 and progv, that's all i think 17:13:37 stassats: not name or tag 17:13:44 stepnem: apply 17:13:47 Captain_Thunder [n=jimmymil@c-24-35-95-148.customer.broadstripe.net] has joined #lisp 17:13:52 Xach: oh, thanks! 17:14:06 *sigh* 17:14:08 sepult [n=user@xdsl-87-78-25-189.netcologne.de] has joined #lisp 17:14:43 stassats: progv only establishes dynamic variable bindings 17:15:17 demmel: desctructuring-bind, m-v-b 17:15:18 oh, you asked lexical, then yes, progv and catch are crossed out 17:15:48 drewc: those are not special forms 17:15:50 varjag [n=eugene@59-125-183-79.HINET-IP.hinet.net] has joined #lisp 17:16:48 anyone here who had success cross compiling ecl for arm? 17:16:49 demmel: if you want only variable, then flet and lables do not suit too 17:17:03 -!- okflo [n=user@91-115-89-64.adsl.highway.telekom.at] has left #lisp 17:17:41 stassats: but the bind variables in the bodies of the defined functions. 17:17:43 -!- rswarbrick [n=rswarbri@user-5447c311.wfd92.dsl.pol.co.uk] has quit [Remote closed the connection] 17:17:46 they* 17:19:06 demmel: in that case, PROG is a macro as well... not a special form. 17:19:39 why you need that, anyway? 17:20:17 phf [n=phf@host.icnfull.com] has joined #lisp 17:20:32 -nil- pasted "recursive compress" at http://paste.lisp.org/display/83239 17:20:32 drewc: you're right. I got that mixed up. 17:21:19 stassats: code walking. I need to notice if something shadows a symbol-macro 17:21:33 oops it should be (a 3 b 2 c 1) 17:21:40 drewc: btw, UCW breaks with newest swank 17:21:58 -!- Quadrescence is now known as AntonioG 17:22:10 can anyone have a look at my paste above? ^^' 17:22:13 stassats: whats with block? How can that establish lexical variable bindings? 17:22:16 p_l: oh? 17:22:24 p_l: oh, right... the backtrace thing 17:22:26 demmel: not variable 17:22:35 stassats: ah ok good. 17:22:56 -!- AntonioG is now known as AnthonyG 17:22:59 drewc: yeah, change from swank-backend:f-s-l-f-e to swank-backend:f-s-l 17:23:06 -!- AnthonyG [n=quad@unaffiliated/quadrescence] has quit [Nick collision from services.] 17:23:12 milan [n=milan@93.87.193.108] has joined #lisp 17:23:15 stassats, drewc: Thanks! 17:23:21 AnthonyG [n=quad@24.118.241.200] has joined #lisp 17:23:25 -!- AnthonyG [n=quad@24.118.241.200] has quit [Nick collision from services.] 17:23:40 metabowl [n=quad@24.118.241.200] has joined #lisp 17:23:52 stassats: the recursion master have a look at my lisppaste please 17:24:08 Bigshot_: that's a home work, isn't it? 17:24:09 Bigshot_: it's -nil- who pasted... 17:24:22 i'll include my name later on 17:24:52 that looks like the exercise from ANSI Common Lisp 17:25:06 Bigshot_: what about (a a b a a c) ? 17:25:16 (a 2 b 1 a 2 c 1) 17:25:24 -!- metabowl is now known as Quadrescence 17:25:26 ok. 17:25:37 that doesn't look compressed 17:25:39 -!- phf [n=phf@host.icnfull.com] has quit ["Leaving..."] 17:25:46 :-) 17:25:48 RLE 17:26:04 man i've spent 2 hours after it 17:26:33 *Fare* tries to build a release tarball of xcvb... 17:26:35 http://lib.store.yahoo.net/lib/paulgraham/acl2.lisp 17:26:37 Isn't rle becoming more interesting only at 2 repetitions (i.e. 3 or more identical characters) 17:27:10 it certainly will look more compressed 17:27:19 :D 17:27:34 Yeah, I can see it in the book 17:28:10 *p_l* monkey patches swank to avoid API change 17:28:40 what are you patching? 17:28:56 Phoodus [i=foo@ip68-231-38-131.ph.ph.cox.net] has joined #lisp 17:29:26 Bigshot_: I would do it in two phases. 17:29:48 Bigshot_: you can split the problem in two: count the length of a run, and collect the results of each runs. 17:30:27 Bigshot_: just because graham using shite variables names doesn't mean you should... see http://www.cs.northwestern.edu/academics/courses/325/readings/graham/graham-notes.html for what not to do. ;) 17:30:34 uses* 17:30:35 i will have to use an array then pjb? 17:30:48 Bigshot_: can you write a simple (recursive if you need to) function to count the number of "a"s in (a a a a b ...) ? 17:31:09 hold on 17:31:14 stassats: swank changed api and breaks UCW 17:31:25 -!- mega1 [n=mega@pool-03102.externet.hu] has quit [Read error: 110 (Connection timed out)] 17:32:13 pjb why "if you need" to? 17:32:14 -!- milanj- [n=milan@93.87.193.108] has quit [Read error: 110 (Connection timed out)] 17:32:22 isn't that required for the task? 17:32:40 oh nvm 17:33:31 stassats: still, it's nice how I could simply... coerce the API :P 17:36:21 p_l: no need to patch swank ... just change the one form in UCW to use the correct name for the function 17:36:44 or dump the function entirely ... i don't think it's used anymore. 17:38:30 drewc: still, I wanted to test how it will work 17:38:59 p_l: fair enough. what's called 'monkey patching' elsewhere is usually just called 'developing' in lisp ;) 17:39:45 watch out for gigamonkeys 17:40:20 -!- schoppenhauer [n=senjak@unaffiliated/schoppenhauer] has left #lisp 17:40:58 Bigshot_: so? Any difficuly in counting the elements of a list? 17:41:21 gigamonkey-patching ... i like that. 17:41:43 *Xach* learns about COPY-SYMBOL 17:41:44 pjb hold on i was doing something else 17:42:00 ok 17:42:59 knobo [n=user@ti0073a340-0120.bb.online.no] has joined #lisp 17:43:11 -!- Jasko2 [n=tjasko@209.74.44.225] has quit [Read error: 110 (Connection timed out)] 17:43:34 Any idea how I can make an error in hunchentoot end up in the debuger in the repl? 17:44:41 SBCL is released about monthly, right? 17:44:54 demmel: yes 17:45:08 pjb pasted "reporting-error for hunchentoot" at http://paste.lisp.org/display/83241 17:45:19 Xach: Hot stuff! 17:45:27 knobo: at least, I get a backtrace with that macro... 17:46:51 -!- milan [n=milan@93.87.193.108] has quit [Read error: 110 (Connection timed out)] 17:49:02 -!- dto [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has quit [Read error: 113 (No route to host)] 17:49:16 eni4ever [n=admin@79.114.1.251] has joined #lisp 17:49:22 dto [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has joined #lisp 17:49:28 -!- varjag [n=eugene@59-125-183-79.HINET-IP.hinet.net] has quit ["Leaving"] 17:49:31 -!- eni4ever [n=admin@79.114.1.251] has quit [Remote closed the connection] 17:49:34 Jasko [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has joined #lisp 17:51:44 manuel_ [n=manuel@HSI-KBW-091-089-172-094.hsi2.kabel-badenwuerttemberg.de] has joined #lisp 17:51:48 hefner: please try http://common-lisp.net/~sionescu/cffi-grovel.patch 17:52:06 pjb annotated #83241 "with-debugger" at http://paste.lisp.org/display/83241#1 17:52:18 mrsolo [n=mrsolo@nat/yahoo/x-05ff785e4105721c] has joined #lisp 17:52:23 knobo: you could wrap the forms you want to debug with that with-debugger macro. 17:52:39 Bigshot_ pasted "count length" at http://paste.lisp.org/display/83242 17:52:57 Bigshot_: remember, you can annotate your previous paste. 17:53:08 oh ok 17:53:38 thank you :) 17:54:06 Bigshot_: what does count-length return? 17:54:17 len 17:54:22 Bigshot_: did you look at the link Xach pasted? 17:54:24 of what type? 17:54:34 integer? 17:55:02 and what is (car list)? 17:55:48 first element of the list 17:55:59 stassats: doesn't seem likely. 17:56:07 i know that code is rubbish but that 's what i tried 17:56:07 What sense does it make to compare the first element of the list with the length of the run? 17:57:05 Try to think. count-length must return the number of elements that are equal to the first element of the list. So already, you see that you need one more argument: the first element of the list. 17:57:26 ok 17:57:28 (defun count-length (first-element list length-of-run) ...) 18:00:01 -!- Adrinael [i=adrinael@rid7.kyla.fi] has quit [Client Quit] 18:00:06 Adrinael [i=adrinael@rid7.kyla.fi] has joined #lisp 18:00:08 Odin- [n=sbkhh@s121-302.gardur.hi.is] has joined #lisp 18:00:16 -!- Adrinael [i=adrinael@rid7.kyla.fi] has quit [Client Quit] 18:00:39 Adrinael [n=adrinael@barrel.rolli.org] has joined #lisp 18:03:58 -!- xan_ [n=xan@conference/gcds/x-d09f1ec400f9a52d] has quit [Read error: 110 (Connection timed out)] 18:04:21 drewc: if I understand ROFL code correctly, If I specify (:table . 'table-name) the class will be operating on table_name table, right? 18:04:31 adeht [i=death@host-90-233-250-184.mobileonline.telia.com] has joined #lisp 18:04:45 bdowning [n=bdowning@mnementh.lavos.net] has joined #lisp 18:04:52 roygbiv [n=blank@pdpc/supporter/active/roygbiv] has joined #lisp 18:05:05 pjb: how can i extract the first element? 18:05:11 dysinger [n=tim@71-20-231-3.war.clearwire-wmx.net] has joined #lisp 18:05:24 if i do car x then won't first element change everytime 18:05:34 since i am also taking cdr x 18:05:52 -!- Sikander [n=soemraws@wirenat-eld.strw.leidenuniv.nl] has quit ["Leaving"] 18:06:00 p_l: the . notation and the quote shouldn't be needed .... if they are it's my fault and a bug... but besides that yeah, that's the idea 18:06:43 drewc: I'm just taking it from examples :) 18:06:50 (rtracker.lisp) 18:06:54 rouslan [n=Rouslan@unaffiliated/rouslan] has joined #lisp 18:06:56 dstatyvka [i=ejabberd@pepelaz.jabber.od.ua] has joined #lisp 18:07:02 and checking it in source 18:07:10 -!- Ppjet6 is now known as Pepe_ 18:08:28 -!- mattrepl [n=mattrepl@204.194.78.3] has quit [] 18:08:34 p_l: that's probably correct then... i have never used that feature outside of the examples :) 18:08:57 p_l: patches for a better syntax welcome :P 18:12:47 willb [n=wibenton@compsci-wifi-39.cs.wisc.edu] has joined #lisp 18:12:50 -!- envi^home [n=envi@220.121.234.156] has quit ["Leaving"] 18:13:28 knobo annotated #83241 "my way" at http://paste.lisp.org/display/83241#2 18:13:59 -!- rouslan [n=Rouslan@unaffiliated/rouslan] has quit [Remote closed the connection] 18:16:25 drewc: class-table-name should return symbol, right? not a list? :D 18:21:17 (count-length (first list) list 0) 18:21:29 Bigshot_: (count-length (first list) list 0) 18:22:28 -!- deafmacro [n=user@59.96.207.140] has left #lisp 18:22:42 benny` [n=benny@i577A2D82.versanet.de] has joined #lisp 18:23:43 -!- Captain_Thunder [n=jimmymil@c-24-35-95-148.customer.broadstripe.net] has quit [] 18:24:14 -!- benny [n=benny@i577A3184.versanet.de] has quit [Nick collision from services.] 18:24:16 -!- benny` is now known as benny 18:26:17 -!- fisxoj [n=fisxoj@149.43.252.13] has quit [Remote closed the connection] 18:29:02 -!- syamajala [n=syamajal@140.232.179.15] has quit ["Leaving..."] 18:30:57 tcr [n=tcr@host145.natpool.mwn.de] has joined #lisp 18:32:17 -!- ace4016 [i=ace4016@cpe-76-168-248-118.socal.res.rr.com] has quit ["When there's nothing left to burn, you have to set yourself on fire."] 18:32:55 Bigshot_ pasted "count length" at http://paste.lisp.org/display/83246 18:33:27 pjb: 18:35:26 Captain_Thunder [n=jimmymil@c-24-35-95-148.customer.broadstripe.net] has joined #lisp 18:36:16 drewc: ok, I got lost in the code - is there a way to specify type of slot on lisp side? (integer, text, etc.) 18:37:42 orlandov [n=orlando@gateway.gossamer-threads.com] has joined #lisp 18:38:07 (defclass foo (superclasses...) ((slot-name :type type))) 18:38:17 brandini [n=brandini@metabug/brandon] has joined #lisp 18:38:29 xX_IM_LOCO_Xx [n=xX_IM_LO@pool-173-59-129-198.nrflva.east.verizon.net] has joined #lisp 18:38:38 what up 18:38:57 anyone in here that doesn't want to try an OS I started in lisp? 18:38:57 Hello 18:39:04 p_l: depends what you mean 18:39:32 do u guys play Halo 3???? 18:39:45 We talk about Lisp here, xX_IM_LOCO_Xx 18:39:46 xX_IM_LOCO_Xx: wrong channel. 18:39:57 Halo! 18:40:00 k!!!!!!!!!!!!!!!! 18:40:02 p_l: the short answer is no, i just use that approach. The slightly longer answer is yes, you can specify the lisp types of the slot if you want type-checking. the longer answer yet is of course, it's lisp, you can do whatever you want :) 18:40:04 -!- xX_IM_LOCO_Xx [n=xX_IM_LO@pool-173-59-129-198.nrflva.east.verizon.net] has quit [Client Quit] 18:40:06 brandini: i don't. 18:40:08 hey - i'm just getting started with learning lisp (and pretty excited about it) but I was wondering - what is the most stand out popular lisp interpreter/compiler? 18:40:21 -!- Guest69368 [n=lexa_@83.222.5.112] has left #lisp 18:40:23 i've been using clisp, but was wondering if there was somethign else i should check out 18:40:24 orlandov: there isn't a single one. there are several really good ones for different needs. 18:40:41 orlandov: i really like to use sbcl on linux. i write programs with emacs and slime and sbcl. 18:40:59 i mostly make web-related stuff, and some text processing stuff. 18:41:04 are there any resources that contrast them? 18:41:08 brandini, I would give it a try and load it on my Lisp Machine 18:41:13 I work for Clozure. I like our lisp. :) 18:41:22 orlandov: yes, dan weinreb did a survey. just a sec, i'll find it. 18:41:24 #ccl 18:41:28 billstclair: I'd be a little worried if you didn't 18:41:33 i like the idea of clojure but the jvm thing turns me off :) 18:41:33 indeed 18:41:37 orlandov: http://common-lisp.net/~dlw/LispSurvey.html is it 18:41:41 i was looking at clozure earlier today 18:41:42 clojure is not clozure 18:41:47 i know :) 18:41:49 k 18:41:52 actually, CCL isn't that bad ;-) 18:42:04 Xach++ # ty 18:42:08 -!- brandini [n=brandini@metabug/brandon] has left #lisp 18:42:25 hmm, brandini left 18:42:50 maybe his os crashed... 18:42:56 hers/its 18:43:11 p_l: I want to hack ECL so it stores arglists even for the functions implemented in C. 18:43:25 salex: but he sure would have a restart 18:43:42 p_l: I don't know what the @(defun ...) preprocess stuff should expand to, given that you do not have the concepts of toplevel-forms in C. 18:43:43 one would think.... 18:44:05 How do you incrementally define data in C? 18:44:25 tcr: unfortunately, I'm not familiar enough with ECL internals to help you 18:44:35 Oh It's a C question 18:44:55 drewc: so, let me guess, what I need to do is to implement normal lisp typechecking and add apriopriate translators to postmodern? 18:46:31 -!- stassats [n=stassats@wikipedia/stassats] has quit [Remote closed the connection] 18:47:54 -!- legumbre_ is now known as legumbre 18:48:46 Bigshot_: sorry for the delay. The second one clause is not correct. The test is ok, but you must count the elements of the list, not only the first one. So you will have to call count-length on the rest of the list. 18:49:05 p_l: yeah, that's the easy way. The reason i don't do this is that LoL likes to store things in slots temporarily... like lists of things that could potentially belong in that slot. 18:49:10 k 18:49:20 Bigshot_: and about the third clause, if the first-element is not equal to (first list), it means the run is done, so the result should be 0. 18:49:46 Actually, in both 1st and 3rd clauses, the result should be len, not 0, since we're in the accumulator pattern. 18:50:00 p_l: all my typechecking etc is handled through lol, so i've kept ROFL as lightweight as possible. 18:50:44 quek [n=read_eva@router1.gpy1.ms246.net] has joined #lisp 18:50:56 peddie_ [n=matthew@67.169.49.37] has joined #lisp 18:51:04 drewc: Well, I have need for some specialised types for which I'd rather not like to contain something else, least it shall blow up in my face like holy hand grenade 18:51:20 sebell [n=sctb@mail.arcurve.com] has joined #lisp 18:52:25 I want to pass a gis:point class for example and do all interaction on its contents through methods etc. 18:54:47 p_l: the postmodern + simple-date stuff (included with postmodern) is a useful example of how to do that. SBCL will treat :type declarations as assertions at a certain SAFETY setting, and that should be all you need. 18:55:10 schoppenhauer [n=christop@unaffiliated/schoppenhauer] has joined #lisp 18:55:20 Bigshot_ pasted "count length - done pjb" at http://paste.lisp.org/display/83250 18:55:54 drewc: sounds ok 18:55:56 Bigshot_: the third clause is not correct. 18:56:44 When the first-element is not (first list), it means that you've eaten all the run: (count-run 'a '(b b c)) --> 0 18:56:59 When the first-element is not (first list), it means that you've eaten all the run: (count-run 'a '(b b c) 3) --> 3 18:57:21 drewc: when I get PostGIS integration code into somewhat usable shape, I'll post it on GitHub 18:57:28 -!- orlandov [n=orlando@gateway.gossamer-threads.com] has left #lisp 18:58:29 p_l: cool! 18:58:43 konr [n=konrad@201.82.132.33] has joined #lisp 18:58:49 stassats [n=stassats@wikipedia/stassats] has joined #lisp 18:59:45 drewc: I was also thinking of integrating schema generation into it, so I won't have to separately write SQL :> 19:00:30 Sikander [n=soemraws@oemrawsingh.xs4all.nl] has joined #lisp 19:01:08 Bigshot_ pasted "count length - done pjb" at http://paste.lisp.org/display/83251 19:01:39 -!- konr [n=konrad@201.82.132.33] has quit [Client Quit] 19:01:47 Bigshot_: Please, annotate! 19:01:52 p_l: i tend to do it the other way, and generate base classes from the sql schema. Schema generation is hard to get right in my experience. 19:01:55 konr [n=konrad@201.82.132.33] has joined #lisp 19:02:11 will it expire in 1 hour if i annotate it? 19:02:24 Bigshot_: For the last clause, you can keep T as condition, It's necessarily (NOT (OR all the previous clauses)) 19:02:39 p_l: there is some support for schema migrations in the code, but it's not complete. 19:02:45 drewc: I quite like the way DataMapper does it 19:02:57 Bigshot_: In the case where first-element is not (first list), it means that we have finished counting, so we must return the result, len. That's all. 19:07:19 -!- peddie [n=matthew@67.169.49.37] has quit [Read error: 113 (No route to host)] 19:08:12 milaz [n=user@85.172.104.151] has joined #lisp 19:08:51 -!- SandGorgon [n=OmNomNom@122.162.51.25] has quit [Read error: 60 (Operation timed out)] 19:09:21 xan [n=xan@29.Red-83-59-49.dynamicIP.rima-tde.net] has joined #lisp 19:10:19 Bigshot_ annotated #83251 "untitled" at http://paste.lisp.org/display/83251#1 19:10:25 pjb 19:10:36 -!- manuel_ [n=manuel@HSI-KBW-091-089-172-094.hsi2.kabel-badenwuerttemberg.de] has quit [] 19:11:02 pjb hope i've not made any mistake now 19:13:31 Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has joined #lisp 19:14:55 Bigshot_: you're still calling count-length in the third case. Why? 19:15:23 When you reach this third case, you have finished counting and the result is len. So you just need to return len, nothing more. 19:15:42 so it should be (t len)? 19:15:46 Yes. 19:15:57 Bigshot_: you have two case when you're done counting. 1) when you reach the end of the list, and when you reach an element that is not equal. 19:16:16 s/and when/and 2) when/ 19:17:12 p_l: DataMapper looks interesting... some good ideas in there. 19:18:18 Bigshot_ annotated #83251 "doesn't work gives 0" at http://paste.lisp.org/display/83251#2 19:18:31 drewc: it's IMHO one of better ORMs, especially in Ruby world :) 19:18:32 pjb 19:19:11 pjb hold on 19:19:27 Should be good. Try it. 19:20:17 ((equal first-element (car list)) (count-length (first list) (cdr list) (1+ len))) 19:20:18 mattrepl [n=mattrepl@ip68-100-82-124.dc.dc.cox.net] has joined #lisp 19:20:30 this doesn't work 19:20:56 Indeed, there's a bug with (count-length 'a '(a a a) 0) 19:21:03 Probably the first clause is wrong. 19:21:18 peddie [n=matthew@67.169.49.37] has joined #lisp 19:21:18 (list (count-length 'a '(a a a b c) 0) (count-length 'a '(a a a) 0) (count-length 'a '(b b ) 0) (count-length 'a '() 0)) should return (3 3 0 0) 19:22:21 sytse [i=sytse@speedy.student.ipv6.utwente.nl] has joined #lisp 19:22:26 p_l: indeed. The minimizing of round-trips to the db was something i had in mind when i started ROFL, but didn't get around to doing. the find-object cache takes care of part of that, but it could be a lot better. 19:22:51 -!- peddie_ [n=matthew@67.169.49.37] has quit [Read error: 60 (Operation timed out)] 19:25:11 -!- nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has quit [] 19:27:16 lujz [n=lujz@cpe-92-37-14-9.dynamic.amis.net] has joined #lisp 19:28:20 drewc: it's also a surprisingly nice method to build database schemas :) 19:30:19 -!- quek [n=read_eva@router1.gpy1.ms246.net] has quit [Remote closed the connection] 19:32:16 Bigshot_ annotated #83251 "pjb what's the bug here?" at http://paste.lisp.org/display/83251#3 19:32:39 -!- nikodemus [n=nikodemu@cs181144086.pp.htv.fi] has quit ["Leaving"] 19:33:28 -!- zophy [n=sy@host-242-6-111-24-static.midco.net] has quit [Remote closed the connection] 19:33:42 Bigshot_: if you want to deal with proper-lists, then use endp instead of null. But otherwise, it wasn't the test of the first clause that was erroneous, but the result. You returned 0 instead of len. 19:34:13 p_l: my object model often differs somewhat from my database schema, so i don't mix them often. I can, however, see why one might want to :) 19:34:14 -!- bgs100 [n=ian@unaffiliated/bgs100] has quit [] 19:34:24 pjb annotated #83251 "applying the accumulator pattern" at http://paste.lisp.org/display/83251#4 19:35:26 Are there reasons other than shorthand/cosmetic for the existence of the let form? A friend says he does not get why not just use some set form (for creating bindings). 19:35:32 Bigshot_: have a look at my annotation. Perhaps it's simplier if you first make a recursive function, without the recursion being a tail-call. Then transform it into a recursive tail-call with the accumulator pattern. When the operation is associative, there's no difficulty. 19:36:46 pjb: and commutative. 19:36:56 antoszka: With SETF, we cannot know the scope of the variable. With LET, we have a lexical scope, so we can better reason about the program. 19:37:18 -!- ikki [n=ikki@201.155.75.146] has quit [Read error: 113 (No route to host)] 19:37:19 pkhuong: Yes, I didn't forget commutative in my annotation. 19:37:53 pjb: Right, that's how I feel. But I thought that creating a binding within, say, a function definition limits the scope to that definition. 19:37:59 rswarbrick [n=rswarbri@user-5447c311.wfd92.dsl.pol.co.uk] has joined #lisp 19:38:12 There could be some difficulty if the operation is associative but not commutative (e.g. append) 19:38:16 antoszka: setf changes the value of place to be newvalue. and since we don't know the scope, the compiler must modify the memory. In the case of LET, since it knows the lexical scope, it can prove to itself that no memory store is needed (in some cases) and therefore it can avoid using the memory for variable bound with LET. 19:38:29 antoszka: there's no faster code than the code that doesn't use the memory... 19:38:39 Mhm. Thx. 19:39:09 antoszka: Functions are better defined and debugged when they don't have side effect, that is when they use no global variable. 19:39:16 antoszka: i think you can consider the LET form like C's block, but immensly more cooler, practical and more powerful. 19:40:05 pkhuong: yes. In the annotation, I mentioned that it was easy when they were BOTH commutative and associative. Special treatment must be provided when they're not. 19:41:09 pjb, hypno: thx 19:41:13 DeusExPikachu [n=DeusExPi@pool-173-58-94-2.lsanca.fios.verizon.net] has joined #lisp 19:42:35 "DeusExPikachu" :-) 19:42:46 drewc: DM is rather flexible, IMHO, though when it comes to being completely detached from db schema, Sequel is quite interesting case 19:42:46 Hi 19:42:59 DeusExPikachu: funny nick! 19:43:15 *Xach* notes the universal-time 3456789012 is coming up later this month 19:43:15 pjb, yeah it sorta stuck with me since highschool 19:43:38 obviously I was into pokemon and the game Deus Ex at the time... 19:43:41 -!- ASau` [n=user@193.138.70.52] has quit [Read error: 104 (Connection reset by peer)] 19:44:11 -!- ASau [n=user@193.138.70.52] has quit [Read error: 54 (Connection reset by peer)] 19:44:45 ASau` [n=user@193.138.70.52] has joined #lisp 19:45:08 *p_l* was thinking of something more like "pulled wacky trick involving said electric furball, turning over situation like Deus Ex Machina" ;-) 19:45:30 -!- Captain_Thunder [n=jimmymil@c-24-35-95-148.customer.broadstripe.net] has quit [] 19:45:47 No, just a small god coming out of Pikachu leg... 19:49:58 Yeah its all of those things 19:52:33 puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has joined #lisp 19:53:09 -!- Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has quit [Remote closed the connection] 19:53:14 -!- puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has quit [Client Quit] 19:53:23 puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has joined #lisp 19:53:44 -!- adeht [i=death@host-90-233-250-184.mobileonline.telia.com] has quit [] 19:54:06 btw, anyone know if hunchentoot works with serve-event or some other io-multiplexing on sbcl? perhaps in trunk or somesuch? 19:56:58 -!- A_anekos is now known as anekos 20:03:10 milanj [n=milan@93.86.55.223] has joined #lisp 20:07:00 lemoinem [n=swoog@modemcable110.189-201-24.mc.videotron.ca] has joined #lisp 20:07:16 -!- roygbiv [n=blank@pdpc/supporter/active/roygbiv] has left #lisp 20:14:32 quek [n=read_eva@router1.gpy1.ms246.net] has joined #lisp 20:15:30 -!- tmh [n=thomas@pdpc/supporter/sustaining/tmh] has left #lisp 20:16:43 wubbster [n=wubb@doc-24-206-235-153.kw.tx.cebridge.net] has joined #lisp 20:18:54 Bigshot_: so? 20:19:09 diog3n3s [n=diog3n3s@97-118-148-240.hlrn.qwest.net] has joined #lisp 20:19:18 -!- sepult [n=user@xdsl-87-78-25-189.netcologne.de] has quit ["ERC Version 5.2 (IRC client for Emacs)"] 20:19:30 -!- Holcxjo [n=holly@ronaldann.demon.co.uk] has quit [Read error: 110 (Connection timed out)] 20:20:19 Holcxjo [n=holly@ronaldann.demon.co.uk] has joined #lisp 20:20:37 -!- willb [n=wibenton@compsci-wifi-39.cs.wisc.edu] has quit [Read error: 110 (Connection timed out)] 20:21:25 -!- stassats [n=stassats@wikipedia/stassats] has quit [Read error: 113 (No route to host)] 20:23:34 manuel_ [n=manuel@HSI-KBW-091-089-172-094.hsi2.kabel-badenwuerttemberg.de] has joined #lisp 20:31:19 pjb: i''ll continue later 20:31:36 ok. 20:31:50 I'll annotate indications for later. 20:32:28 -!- quek [n=read_eva@router1.gpy1.ms246.net] has quit [Read error: 110 (Connection timed out)] 20:32:44 Captain_Thunder [n=jimmymil@c-24-35-95-148.customer.broadstripe.net] has joined #lisp 20:33:30 -!- wubbster [n=wubb@doc-24-206-235-153.kw.tx.cebridge.net] has quit [Remote closed the connection] 20:34:10 wubbster [n=wubb@doc-24-206-235-153.kw.tx.cebridge.net] has joined #lisp 20:37:26 bgs100 [n=ian@unaffiliated/bgs100] has joined #lisp 20:37:44 francogrex [n=franco@91.176.41.93] has joined #lisp 20:38:14 hi anyone has experience with UFFI? 20:38:34 if you can, go for CFFI instead of UFFI francogrex 20:39:04 lnostdal, I have read something to that effect. But why? 20:39:05 francogrex: UFFI is deprecated for CFFI. 20:39:18 francogrex: UFFI supports less implementations than CFFI. 20:39:25 i see 20:39:31 -!- djkthx [n=yacin@glug.id.iit.edu] has quit ["leaving"] 20:40:09 francogrex: on the other hand, some implementation specific FFI are of a higher level than CFFI, and you may prefer use it (eg. CLISP FFI is much nicer). But you'd lose portability... 20:40:22 thing is uffi comes already in my system. while cffi i had to install the cffi;asdf system 20:40:27 HG` [n=wells@xdsleu037.osnanet.de] has joined #lisp 20:40:40 francogrex: don't be afraid of libraries! ;-) 20:40:45 Jacob_H [n=jacob@92.4.246.133] has joined #lisp 20:41:14 pjb, I had used ffi-clines in ecl, they are low level and quite nice to incorporate/embed any c code 20:41:19 snippets even 20:41:31 Indeed, they've got their point too. 20:41:41 Windows came with my laptop here 20:41:48 IE6 too 20:41:56 djkthx [n=yacin@glug.id.iit.edu] has joined #lisp 20:41:58 -!- Jasko [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has quit [Read error: 110 (Connection timed out)] 20:42:18 rouslan [n=Rouslan@unaffiliated/rouslan] has joined #lisp 20:43:09 -!- synic [n=squish@pdpc/supporter/student/synic] has quit [Read error: 110 (Connection timed out)] 20:44:13 francogrex: there's also ACLs FFI... oh how I wish we could get that part turned into open source library ^^; 20:46:13 Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has joined #lisp 20:46:25 -!- xan [n=xan@29.Red-83-59-49.dynamicIP.rima-tde.net] has quit [Read error: 113 (No route to host)] 20:46:41 p_l, are those ACL-ffi that good? what's special about them, 20:46:42 ? 20:48:07 *francogrex* is trying to implement a very old Fortran algorithm for exact CI odds ratio into CL 20:49:49 francogrex: ACL has very good support for C++, iirc 20:50:21 jyujin [n=mdeining@82.113.121.138] has joined #lisp 20:50:23 p_l, ecl should be as good for that, if not even better 20:51:33 macdice [n=macdice@78-86-162-220.zone2.bethere.co.uk] has joined #lisp 20:52:30 francogrex: that requires compiling ECL with C++ compiler and writing by hand C++ classes -> CLOS mappings :) 20:52:58 urk 20:53:22 but what does that have to do with fortran code? 20:53:59 salex: I kind of missed the fortran part, I jumped in only for FFI :P 20:55:14 salex, no that's not to do with anything yet; if works ok, nothing needed, otherwise is f2c then cffi or f2cl to get it to cl, but will try by hand first 20:55:46 ok, sure, but I was confused at how c++ got involved 20:55:46 but regardless, i'm always interetsed in embedding other language programs into common lisp 20:56:03 if it isn't too complex, and you don't need blinding speed, native will be nice 20:56:09 sometimes it's impractical 20:56:10 well that was p_l's 20:56:32 sure 20:56:35 c++ is pretty usually a pain to embed, period. 20:56:37 is all 20:56:54 anyway, i understand now 20:57:06 I know, i don't even like c++. My 2 preferred languages are lisp and C; That's all 20:57:36 -!- dstatyvka [i=ejabberd@pepelaz.jabber.od.ua] has left #lisp 20:59:59 i have been trying to integrate a C++ library into an SBCL app. it involves an awful lot of typing 21:00:35 c++ is generally happier as the integrator, not the integratee, ime 21:00:38 ikki [n=ikki@201.155.75.146] has joined #lisp 21:00:52 Lisp too. 21:00:59 I was about to say 21:01:00 is it possible to make comments in a format string? 21:01:01 Does that make C++ and Lisp incompatible? 21:01:07 lisp too surely 21:01:10 I pity the fool who triel to integrate lisp into their C app. 21:01:11 pjb: true 21:01:12 knobo: I'd bet yes. 21:01:18 sykopomp: ecl? 21:01:19 s/triel/tries. 21:01:26 pjb: how much? 21:01:29 salex: "I pity the fool..." 21:01:31 c++ often can't even integrate with other c++ 21:01:31 sykopomp, why? 21:01:33 (format t "abc~:[comment~]def" nil) 21:01:33 it does sometimes make for ugly solutions involving pipes, handles and RPC. 21:01:34 sykopomp: with ECL it isn't hard... 21:01:48 ah. Well, I haven't played with ecl that much >_> 21:01:58 and I guess there's that cl-python thing for python, too 21:02:09 -!- puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has quit [Remote closed the connection] 21:02:12 knobo: now of course, we'd want to avoid the argument... 21:03:00 sykopomp, p_l, well yes it's possible to mix some lisp code into a C then compile a exe or a dll 21:03:34 into a C ? 21:03:37 francogrex: ECL uses DLLs as its fasl/core format 21:03:38 what i've done was write a library in lips compile as a fas file and load 21:03:58 -!- HG` [n=wells@xdsleu037.osnanet.de] has quit [Client Quit] 21:04:32 p_l, yes and just one "annoying" thing is that your "hybrid" c/lisp dll or exe will always depend on ecl.dll being around 21:04:56 francogrex: you can compile it statically... 21:05:23 the dll? not possible 21:05:28 the exe maybe 21:05:59 francogrex: nothing stops you from making your own variant of ecl.dll that contains your image and link that into your app 21:06:43 right. it's hardly odd for an exe that is dynamically linked to depend on the dynamic libs it er, depends on.... 21:06:55 knobo: I've got a comment that needs one arguments without eating it... 21:06:59 p_l, I'm curious as to how? when using lisp into a c prog you'll need the whole runtime (that was an older discussion with juanjo) 21:07:08 (format t "abc~:[~:*~:[~;comment~]~;~]~:*def ~S" arg) 21:07:46 francogrex: when using C into a lisp program, you need the whole runtime too. libc is not small... 21:08:01 -!- mkfort [i=X8TQ91P4@68-189-164-209.dhcp.spbg.sc.charter.com] has quit [Read error: 110 (Connection timed out)] 21:08:15 francogrex: ECL runtime is quite small, that's one thing, two, you can play around with the way build scripts run to create a static ecl.a and combine it with your fasls 21:08:26 pjb, sure, but i's there already for almost all pc users, they don't even notice 21:08:39 francogrex: they won't notice libecl either. 21:08:46 however when you ship ur apps with a "foreign" dll, eyebrows are raised 21:08:58 by whom? 21:09:07 francogrex: how big is that dll? 21:09:36 4.5 mb maybe, but it's not just the size... 21:09:37 francogrex: seeing how many deps a simple app can have nowadays and no-one notices, I'd say ecl.dll would be "noise" 21:09:50 i'd have to agree with p_l 21:10:06 tcr, well, among others: regulatory agencies, company datamanagement departments etc 21:10:27 kmcorbett1 [n=Keith@c-76-119-215-57.hsd1.ma.comcast.net] has joined #lisp 21:10:33 *p_l* had experiences trying to install small app and see >300MB of deps 21:10:36 For an application, you may statically link with libecl. 21:10:37 francogrex: I don't see what it is you're actually looking for. Of course you need a runtime, unless you've got a complete treeshaker and can convinced anything unneeded is, in fact, unneeded 21:10:45 this isn't really lisp specific though 21:11:08 salex, my dream is to try using a tree shaker 21:11:14 why? 21:11:19 unfortunately lispworks is so expensive 21:11:29 well for experimenting 21:11:31 francogrex: I don't see what might make them interested. You put runtime library in app-related directory and nobody gives a damn 21:11:45 francogrex: Why are they wary on something shipping a 5mb .dll, and not be equally wary on a 5mb sized .exe? 21:11:49 i mean, in practice it seem like a very narrow set of cases when you really care 21:11:53 particularly these days 21:12:07 p_l, yes, but those are supposed to be validated by microsft, or whoever 21:12:22 your lisp/C code is open and can be validated, but 21:12:23 francogrex: nope 21:12:30 francogrex: I believe the answer I've seen before as far as tree-shaking goes is "there's other ways to save space that should probably be tried first" 21:12:38 who wants to go thu the whole runtime for vlidation? 21:12:53 sykopomp, like? 21:12:58 francogrex: validated by who? what context are you shipping this in in which the seat cost of lispworks is anything but trivial? 21:13:04 francogrex: ask the compiler people. 21:13:18 for example. 21:13:21 1) you can show them sourcecode to runtime 2) locking down to only system-provided runtimes is sure way to introduce more bugs and less validation due to NIH 21:13:22 -!- wubbster [n=wubb@doc-24-206-235-153.kw.tx.cebridge.net] has quit [Remote closed the connection] 21:14:05 3) said DLL is probably produced by you during deployment, you sign it 21:14:07 What do they validate of C run-time exactly? 21:14:11 salex, in regulated environements 21:14:27 wubbster [n=wubb@doc-24-206-235-153.kw.tx.cebridge.net] has joined #lisp 21:14:31 slash_ [n=Unknown@p4FF0BE5B.dip.t-dialin.net] has joined #lisp 21:14:34 We could produce a distribution with only lisp programs apart from the kernel. 21:14:59 francogrex: in truly regulated environments, they would simply pay for a team to go over the whole code, because an audit is *THAT* important 21:15:00 pjb, how? 21:15:02 does closer-mop allow you to add slots to a class definition without calling reinitiablize-instance? 21:15:41 francogrex: like this: http://www.informatimago.com/linux/emacs-on-user-mode-linux.html 21:15:56 francogrex: *what* regulated environments? I can't think of one where a lispworks license would be an issue 21:15:56 We could also use Movitz, but we'd still need the drivers... 21:16:06 I'm not saying such things don't exists 21:16:06 p_l, in theory yes; I don't know in practice, they would say; u have too systems, equally good, but one needs a lot of resources and effort, while the other is practcally free... hmm 21:16:28 I'm just getting cognitive dissonance on your mix of comments 21:16:33 after all, lispworks is cheap 21:16:34 francogrex: and in such environments, if they run anything from microsoft, they probably have a private repo of microsoft sourcecode or contacts with a team that does regular audits of said code 21:16:58 (to continue your LW example) 21:17:12 p_l, possibly true 21:17:15 madnificent: closer-mop is not another implementation of CLOS, it's just a portability library over what the various implementations provides of MOP. So if it's possible to add slots to a class in an implementation without calling reinitialize-instance, there... 21:17:46 -!- saikat_ [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has quit [] 21:17:55 pjb: Only because one implementation happens to support that, does not mean that that feature makes into closer-mop. 21:17:59 salex, I didn't say LW would pass easier than anything else for validation, was just a wish for me to experiment with the treeshaker 21:18:21 tcr: that would mean reinitialize-instance is called explicitely by closer-mop. 21:18:35 That's a different question. 21:18:41 *madnificent* waits for the discussion between tcr and pjb to settle 21:18:56 francogrex: no, i'm having trouble imaginging a scenario where you would care this much about actual .dll validation and yet worry about the cost of LW to play around with a treeshaker 21:19:02 francogrex: and then, after all those audits, you are *not* allowed to say that there's no warranty 21:19:11 I can't imagine one, so I asked you what it looked like. 21:19:28 also: it isn't really a problem if the class would call reinitialize-instance, but that requires it to get a complete list of slots (which I would like to skip) 21:19:35 I also think that I can work around it 21:19:41 pjb: Well, I interpreted madnificent whether it's possibly to rely on something like that quasi-portably 21:19:57 francogrex: or are you just holding up a vague shadow of auditing/validation that you don't actually work with? 21:19:58 +'s question 21:20:00 TR2N [i=email@89-180-145-28.net.novis.pt] has joined #lisp 21:20:04 -!- demmel [n=demmel@c215.tum.vpn.lrz-muenchen.de] has quit [] 21:20:34 madnificent: AFAIR (and I haven't finished reading the AMOP), MOP API isn't too fine, and you often have to deal with complete lists... 21:20:44 tcr, pjb: my question was the union of both your interpretations 21:20:48 :-) 21:21:18 -!- schoppenhauer [n=christop@unaffiliated/schoppenhauer] has quit [Read error: 104 (Connection reset by peer)] 21:21:28 salex, do different things, please dissocite: validation is work; treeshaker is for personal experimentation/pleasure 21:21:41 dissociate I meant to say 21:21:58 francogrex: ah. my confusion was due to discussion of treeshaker as a way to avoid excess runtime 21:22:19 saikat_ [n=saikat@c-98-210-13-214.hsd1.ca.comcast.net] has joined #lisp 21:22:27 salex, no was just expressing my wish to try it 21:22:38 francogrex: when you really have to work with regulatory requirements, you don't have time to grumble about runtimes :D 21:22:40 -!- Captain_Thunder [n=jimmymil@c-24-35-95-148.customer.broadstripe.net] has quit [] 21:22:41 ah. well it's not so terribly interesting 21:22:45 personally at home in the dark 21:22:47 pjb, tcr: guess I'll work around it then, thanks 21:22:55 but your delivery issues seem non-starters to me 21:23:09 relative to my experience of actual oversight, which isnt' huge 21:23:09 salex: I'd find it interesting how it deals with CLOS 21:23:10 salex, why not intresting? u tried it? 21:23:33 francogrex: well, what are you wanting to do with it? 21:23:55 see how it works, for a start 21:24:01 see the results 21:24:03 francogrex: when you have to deal with verification like EAL6+, then you can speak about having a lot of work :D 21:24:38 p_l: try FDA cert 21:24:51 for embedded/surgical 21:25:02 also fun 21:25:04 salex: I think EAL6+ is good enough. You don't want a strategic bomber running amok 21:25:16 isn't that what they're for? 21:25:20 ') 21:25:24 makes for great movies, too. 21:25:25 p_l, i'm not having work, i'm just enjoying myself now is relaxation time 21:25:58 FDA can wait 21:26:03 salex: "controlled" amok. Not "shit crashed mid flight and released nukes over your own city" :P 21:26:08 -!- kmcorbett [n=Keith@76.119.215.57] has quit [Read error: 110 (Connection timed out)] 21:26:50 willb [n=wibenton@h69-129-204-3.mdsnwi.broadband.dynamic.tds.net] has joined #lisp 21:26:50 francogrex: it just seemed like you were begging trouble about delivery issues, is all 21:27:11 p_l: well, either way, i'd rather work medical than defense 21:27:46 and FDA is enough of a pain (a bit different sort) 21:28:45 *francogrex* works medical/healthcare 21:29:19 ok guys, good chatting with you, need to get on with a few things 21:29:26 *salex* works more when he's not here ... so.... 21:29:27 -!- salex [n=user@216.80.147.206] has left #lisp 21:29:28 c u later 21:29:54 -!- francogrex [n=franco@91.176.41.93] has quit ["Leaving"] 21:31:06 demmel [n=demmel@d064.tum.vpn.lrz-muenchen.de] has joined #lisp 21:31:42 Adamant [n=Adamant@c-76-29-188-60.hsd1.ga.comcast.net] has joined #lisp 21:32:20 -!- foom [n=jknight@ita4fw1.itasoftware.com] has quit [Remote closed the connection] 21:33:18 On linux/ix85, libecl.a is less than half libc.a 21:34:51 Captain_Thunder [n=jimmymil@c-24-35-95-148.customer.broadstripe.net] has joined #lisp 21:35:32 -!- spilman [n=spilman@ARennes-552-1-89-252.w92-139.abo.wanadoo.fr] has quit ["Quitte"] 21:36:54 -!- LiamH [n=none@208.72.159.207] has quit ["Leaving."] 21:37:42 jlf` [n=user@unaffiliated/jlf] has joined #lisp 21:39:20 OK - here's a release of XCVB: http://common-lisp.net/project/xcvb/releases/ 21:39:28 contains all matching dependencies 21:39:29 Fare: awesome! 21:39:42 -!- DeusExPikachu [n=DeusExPi@pool-173-58-94-2.lsanca.fios.verizon.net] has quit ["Leaving"] 21:39:57 apparently, though, ADG needs some love to handle Edi's use of defconstant properly -- working on it. 21:44:04 Fare: why would I be interested in xcvb if asdf has been more than enough for me so far, as far as system-loading goes? 21:44:22 (as in, systems are relatively small and simple) 21:44:30 system definitions* 21:44:58 -!- bombshelter13_ [n=bombshel@toronto-gw.adsl.erx01.mtlcnds.ext.distributel.net] has quit [] 21:47:50 The CLHS references at e.g. http://www.cliki.net/Issue%20DEFINE-CONDITION-DOCUMENTATION seem broken. Can someone reconfigure the CLiki server to fix them? Alternatively, would it be useful to change the links to plain HTML and to which server should they then point? 21:48:23 sykopomp: yeah, asdf's :serial works fine for plenty of projects. 21:48:33 foom [n=jknight@ita4fw1.itasoftware.com] has joined #lisp 21:48:46 -!- demmel [n=demmel@d064.tum.vpn.lrz-muenchen.de] has quit [Read error: 110 (Connection timed out)] 21:50:15 -!- tcr [n=tcr@host145.natpool.mwn.de] has quit ["Leaving."] 21:52:56 sykopomp, only if you're interested in 1- incremental compilation, 2- reproducibility, 3- generating executables (or images) 21:53:25 if you're only writing libraries, stick to ASDF for now -- we'll get there before the end of summer 21:53:39 -!- pkhuong [n=pkhuong@x-132-204-255-22.xtpr.umontreal.ca] has quit [Read error: 110 (Connection timed out)] 21:53:41 mkay, thanks. 21:54:08 though we may be offering patches already 21:54:16 as far as xcvb goes... I'm a bit concerned about the whole having to put the module header into each file. Maybe that's not in the API anymore? 21:54:42 I kinda like being able to centralize that into a single .asd. 21:55:36 sykopomp I think a similarity to c or java include/import header lines is the idea 21:55:44 -!- TDT [i=dthole@dhcp80ff869b.dynamic.uiowa.edu] has left #lisp 21:55:45 having the local requirements local 21:56:06 dcrawford: and that's supposed to be a good thing? 21:56:10 dcrawford: that's a big turnoff for me. 21:57:04 dcrawford: a couple of days ago, I reorganized the entire file structure of a project I'm working on (about a dozen files or so?). It was originally toplevel and flat, and I put them into several folders, moved stuff between files, etc. All I had to do once the files were moved was tweak the .asd file and it worked. 21:57:35 it would've been horrible to have to go to each file and figure out what the relative path to each file was. 21:57:41 nightmarish, in fact. 21:57:55 mkfort [i=fiH5Hoht@68-189-164-209.dhcp.spbg.sc.charter.com] has joined #lisp 21:58:12 maybe slime could've done it for you. :) 21:58:16 -!- Jacob_H [n=jacob@92.4.246.133] has quit [Remote closed the connection] 21:58:35 but if not, sed -i makes easy work of that kind of thing. 21:58:48 Fare is it possible for most of this header requirements type stuff to be in the build.xcvb ? 21:59:00 -!- CrazyEddy [n=CrazyEdd@wrongplanet/CrazyEddy] has quit [Read error: 60 (Operation timed out)] 21:59:08 foom: why? It was really smooth and nice just using asdf. Why should I leave something like that to a dev environment? 21:59:30 what header requirements type stuff? 21:59:32 frankly, rearranging the entire project is not a common operation. 21:59:48 Fare .. see above lines by sykopomp 21:59:49 adding or removing requirements for a file is 21:59:57 oh. 22:00:11 Fare: putting +xcvb(:depends-on (:module ...)) on each file 22:00:25 I suppose we could have some syntax allowing to declare dependencies for files in the build 22:00:28 foom: that's a good point, yes. But it does happen. 22:00:38 So it makes a lot of sense to me for it to be optimized for the second. 22:00:40 not the first 22:00:46 foom: what do you think about that? 22:00:47 jmbr_ [n=jmbr@156.33.220.87.dynamic.jazztel.es] has joined #lisp 22:00:59 foom: adding files -is- a common operation, though. 22:01:12 and having a single central place to add information is still good, in that case. 22:01:33 Fare: I think people are complaining about something that isn't going to be an issue in real life 22:01:54 ok. I'll add such a feature at the bottom of my TODO list 22:02:02 Fare: yes, that's what I was getting at. :) 22:02:14 foom: I've already had the situation several times, when I've wanted to restructure a small-to-medium size library. 22:02:39 Fare: it's the big ball of asd vs local requirements lists back-and-forth 22:03:15 sykopomp: tags-query-replace or sed. 22:03:50 -!- milanj [n=milan@93.86.55.223] has quit ["Leaving"] 22:03:56 foom: I don't know sed, nor do I see why I should learn it just to do this. 22:04:33 I haven't heard of tags-query-replace 22:04:51 Fare what were teh advantages of the per-file (module blocks again? 22:04:53 sed is a good command to know, despite what you think of this issue. :) 22:05:07 that's probably true, yeah :) 22:05:24 (don't take it the wrong way, either. xcvb sounds nice, which is why I'm asking all this) 22:05:52 -!- bgs100 [n=ian@unaffiliated/bgs100] has quit [] 22:05:56 find,grep,sed,awk,paste,cut,sort,uniq are a great toolset 22:06:00 having an easy way to release software that can just be compiled into an image with a makefile is fantastic. 22:06:54 dcrawford: local changes are local. adding a dependency to a file doesn't require rebuilding the rest of the system. knowing what needs rebuilding is trivial. 22:07:12 dcrawford: even a dumb program like make can be used to build the system. :) 22:09:25 -!- masm [n=masm@213.22.191.93] has quit ["Leaving."] 22:09:35 -!- Adlai` [n=user@c-69-181-142-8.hsd1.ca.comcast.net] has quit [Read error: 110 (Connection timed out)] 22:09:38 dreish [n=dreish@minus.dreish.org] has joined #lisp 22:10:40 -!- Holcxjo [n=holly@ronaldann.demon.co.uk] has quit [Connection timed out] 22:12:55 -!- fiveop [n=fiveop@g229076131.adsl.alicedsl.de] has quit ["humhum"] 22:13:28 -!- jewel_ [n=jewel@dsl-247-203-169.telkomadsl.co.za] has quit [Read error: 113 (No route to host)] 22:14:26 afk - happy hacking ;-) 22:14:54 nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 22:15:45 roygbiv [n=blank@pdpc/supporter/active/roygbiv] has joined #lisp 22:16:21 -!- jmbr [n=jmbr@128.33.220.87.dynamic.jazztel.es] has quit [Success] 22:19:25 bobvs [n=bob@cpc2-nott8-0-0-cust610.nott.cable.ntl.com] has joined #lisp 22:19:46 -!- manuel_ [n=manuel@HSI-KBW-091-089-172-094.hsi2.kabel-badenwuerttemberg.de] has quit [] 22:22:34 -!- nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has quit [Read error: 104 (Connection reset by peer)] 22:22:37 nvoorhies_ [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 22:24:46 ozy` [n=vt920@97-95-183-15.dhcp.oxfr.ma.charter.com] has joined #lisp 22:24:46 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [] 22:24:49 -!- Edico [n=Edico@unaffiliated/edico] has quit ["Leaving"] 22:26:28 is it possible to do hot-loading of code in lisp similar to erlang? 22:28:18 bobvs: example? 22:28:22 bobvs: you mean loading definitions into a running lisp image? 22:28:25 bobvs, sure; (defun hello () (write-line "before")), (hello) => "before", (defun hello () (write-line "after")), (hello) => "after" 22:28:29 Jasko [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has joined #lisp 22:28:44 (the answer is yes) 22:28:44 bobvs, we do it all the time; it's how we develop .. it is normal 22:28:52 repl ftw 22:28:56 bobvs: it's standard procedure when working in slime :) 22:29:05 (in the modern interpretation of ftw) 22:29:08 and it makes it particularly fun to work on irc bots :D 22:29:16 uroboros [n=mjf@r11gz66.net.upc.cz] has joined #lisp 22:29:29 sykopomp: :P 22:29:31 madnificent: 4 devs total. This is nuts. 22:29:37 sykopomp: only with sheeple ;) 22:29:41 -!- Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has quit [Read error: 113 (No route to host)] 22:29:44 sykopomp: if I'd have the time, I'd help out too 22:29:46 :( why can't I get 4 devs with my MUD instead? :( 22:29:52 no, if I had the time, I'd work on sheeple 22:29:56 oh that 22:30:00 that'd be cool 22:30:09 does modern SBCL have a way to override SBCL_HOME from within? 22:30:27 oh, ok, I'll have a look at it. thanks. 22:31:09 WarWeasle [n=brad@98.220.168.14] has joined #lisp 22:31:33 -!- WarWeasle [n=brad@98.220.168.14] has left #lisp 22:32:40 not 1.0.28.70 at least 22:32:41 sigh 22:34:31 Fare: Two years ago, I used to SB-POSIX:PUTENV "SBCL_HOME=..." and it worked. 22:34:40 bobvs: how does erlang do it? 22:34:54 bobvs: its more intuitive than how erlang does it 22:35:20 -!- willb [n=wibenton@h69-129-204-3.mdsnwi.broadband.dynamic.tds.net] has quit [Read error: 110 (Connection timed out)] 22:35:29 meaning you dont have to code the migration yourself 22:35:51 manuel_ [n=manuel@HSI-KBW-091-089-172-094.hsi2.kabel-badenwuerttemberg.de] has joined #lisp 22:36:35 bobvs: do you mean with "c(module)."? 22:37:05 erlang keeps the old version of the code around too 22:37:18 and you have to plan for your processes to run the new, else they get killed 22:37:28 pjb: nice to know 22:37:45 ozy`: http://erlang.org/white_paper.html in section "Example 14 - Code replacement" 22:39:15 -!- diog3n3s [n=diog3n3s@97-118-148-240.hlrn.qwest.net] has left #lisp 22:41:02 slyrus_ [n=slyrus@dsl092-019-253.sfo1.dsl.speakeasy.net] has joined #lisp 22:41:03 -!- uroboros [n=mjf@r11gz66.net.upc.cz] has quit ["GNU is *indeed* No UNIX!"] 22:41:05 -!- commmmodo [n=commmmod@208.74.35.107] has quit [] 22:41:10 -!- lispm [n=joswig@e177123134.adsl.alicedsl.de] has quit [Read error: 104 (Connection reset by peer)] 22:41:10 uroboros [n=mjf@r11gz66.net.upc.cz] has joined #lisp 22:41:44 nvoorhies [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has joined #lisp 22:41:48 -!- nvoorhies_ [n=nvoorhie@adsl-75-36-204-63.dsl.pltn13.sbcglobal.net] has quit [Read error: 104 (Connection reset by peer)] 22:41:57 -!- wubbster [n=wubb@doc-24-206-235-153.kw.tx.cebridge.net] has quit ["Leaving"] 22:42:13 wubbster [n=wubb@doc-24-206-235-153.kw.tx.cebridge.net] has joined #lisp 22:42:54 Can someone point me to the most appropriate CLX repository/tarball for use with SBCL, CCL, and McClim? 22:43:15 http://common-lisp.net/~crhodes/clx 22:43:42 Thanks :) 22:43:54 -!- uroboros [n=mjf@r11gz66.net.upc.cz] has quit [Client Quit] 22:46:48 *hefner* notes that mcclim-freetype doesn't work with ccl 22:48:06 bobvs: yeah, that example is doable in CL, but I think you really have to be careful when you play with such things, and you can do them in many different ways: recompile a package, paste/compile forms into the repl, load a fasl, etc. 22:48:41 -!- djkthx [n=yacin@glug.id.iit.edu] has quit [Read error: 60 (Operation timed out)] 22:49:09 krumholt_ [n=krumholt@port-92-193-41-161.dynamic.qsc.de] has joined #lisp 22:50:27 -!- krumholt [n=krumholt@port-92-193-20-71.dynamic.qsc.de] has quit [Read error: 60 (Operation timed out)] 22:50:36 djkthx [n=yacin@glug.id.iit.edu] has joined #lisp 22:50:37 bobvs: in erlang though, would the compilation just error if you changed the semantics of any (in-running) function? 22:51:04 demmel [n=demmel@dslb-094-216-077-147.pools.arcor-ip.net] has joined #lisp 22:53:03 my specific gripe with CCL, now that I happen to remember it, is that the default value of the :sharing keyword to open ought to come from a special variable (or just not default to :private), otherwise you get screwed in the case of using code from outside your immediate control which happens to open and return a stream (or retain it in its own data structures) 22:53:24 -!- hkBst [n=hkBst@gentoo/developer/hkbst] has quit [Remote closed the connection] 22:53:39 meaning, for instance, that I'd have to hack the source code to zpb-ttf wherever it calls OPEN just to make mcclim-truetype work on ccl, which is obnoxious. 22:54:13 fvw [n=sdfpme@113.77.216.106] has joined #lisp 22:55:37 -!- foom [n=jknight@ita4fw1.itasoftware.com] has quit [Remote closed the connection] 22:56:51 -!- Sikander [n=soemraws@oemrawsingh.xs4all.nl] has quit ["Leaving"] 22:57:01 hypno: no, the running threads would hit an undefined function error 22:57:08 threads plus streams in that case? 22:57:36 I'm also curious how Lisp is expected to do hot code replacement when for isntance redefining functions that are lower down in your call stack, or currently in process in another thread 22:58:56 termite scheme can replace its own code with code received over the wire 22:59:14 right, but I'm talking about CL 22:59:17 by receiving a thunk and [tail] calling it 22:59:22 Phoodus: hmm. and if you were running a watchdog process to monitor said process that depends on the erroring function, wouldn't that go into some sort of infinite loop? or can the scheduler take measures/precaution in such cases? 22:59:58 Erlang's monitor stuff has a max number of restart attempts per time period 22:59:58 -!- Captain_Thunder [n=jimmymil@c-24-35-95-148.customer.broadstripe.net] has quit [] 23:00:05 to handle the case of things that just won't start 23:00:18 s/monitor/supervisor/ 23:00:18 -!- jmbr_ [n=jmbr@156.33.220.87.dynamic.jazztel.es] has quit [Read error: 110 (Connection timed out)] 23:00:22 i see. very fancy. 23:00:40 Phoodus: "redefining a function" only means changing the function object that is associated with a symbol 23:01:26 serichsen: hmm, true, so even if a package is deleted or whatever, the function on the stack will still remain? 23:02:17 also, I wonder if lisp optimizers are conservative enough not to hard-code in references to something that might be redefined later 23:03:04 Phoodus: if you don't inline functions, sure. 23:03:51 yeah, Erlang has a notion of "modules" where lisp simply has function objects floating around independently 23:04:25 Fare: for that matter asdf _should_ have it's own asdf definition, to be loaded when asdf is around, of course... 23:04:41 Phoodus: so you can only hotswap whole modules, I gather. 23:04:55 correct 23:05:23 and it deliberately holds 2 versions (an old and a new) in memory at once for migration purposes 23:06:03 -!- rswarbrick [n=rswarbri@user-5447c311.wfd92.dsl.pol.co.uk] has quit [] 23:07:06 that seems to be similar to the way CLOS objects are migrated when their class definition is changed 23:07:13 LiamH [n=nobody@pool-141-156-233-205.res.east.verizon.net] has joined #lisp 23:07:38 ltriant [n=ltriant@lithium.mailguard.com.au] has joined #lisp 23:08:07 serichsen: yeah, there are similarities there, especially with its state migration facilities 23:08:41 Xach: hmm? threads versus streams-owned-by-threads, if that was in reply to me. The font-loaders hang around indefinitely while apps may run in a new thread. 23:12:29 -!- macdice [n=macdice@78-86-162-220.zone2.bethere.co.uk] has quit ["zap"] 23:12:34 -!- Hun [n=hun@p50993726.dip0.t-ipconnect.de] has quit [Remote closed the connection] 23:12:50 slyrus: not possible, as I have demonstrated in my last email to asdf-devel 23:13:18 or VERY VERY confusing 23:13:41 ASDF, unlike XCVB, has a problem of incestuous relationship between builder and target 23:14:18 luis: i sent you a new patch 23:14:30 so either you (delete-package :asdf) and all the current asdf state is in limbo, unaccessible by the asdf extensions that try to extend it 23:14:35 xristos: yeah, I've skimmed it. Looks good. 23:14:51 or you don't, and then incompatible versions of asdf will fail your build 23:15:24 xcvb, on the other hand, has no problem building another xcvb for the same or a different lisp implementation 23:16:24 (as demonstrated in the recently released xcvb tarball) 23:17:28 (well, it does have one problem -- which is precisely loading a new asdf on a clisp that has an old asdf) 23:18:30 fe[nl]ix: that patch works. it seems I can now do the unload/reload dance entirely using CFFI, without dipping into sb-alien. 23:19:12 hefner: using (cffi:list-foreign-libraries :type :wrapper) ? 23:19:16 yeah. 23:19:41 good 23:20:10 *hefner* should probably see if any of this still works on CCL 23:20:22 good idea 23:21:21 dto` [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has joined #lisp 23:21:22 -!- dto [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has quit [Read error: 113 (No route to host)] 23:21:44 Error: value # is not of the expected type (OR STRING PATHNAME STREAM) 23:22:15 Jabberwock [n=jens@port-3757.pppoe.wtnet.de] has joined #lisp 23:22:16 maybe this is my fault. how do you get backtrace out of this thing? 23:23:17 hefner: until I commit all this, edit osicat.asd and add :soname "libosicat" to "wrappers" 23:23:24 -!- manuel_ [n=manuel@HSI-KBW-091-089-172-094.hsi2.kabel-badenwuerttemberg.de] has quit [] 23:23:42 mrsolo_ [n=mrsolo@adsl-68-126-215-139.dsl.pltn13.pacbell.net] has joined #lisp 23:26:34 bgs100 [n=ian@unaffiliated/bgs100] has joined #lisp 23:27:44 Fare: interesting... makes me think: "bootstrapping is hard ... let's go shopping!" 23:28:08 fe[nl]ix: anyway, that error is exactly what it sounds like. foreign-library-pathname tries to call pathname-directory on the foreign-library-handle, which is a SHLIB object, not a pathname or string. 23:28:30 darn 23:28:32 -!- kahmalo [n=Kalle@2002:5517:2040:1:230:84ff:fe3e:cfef] has quit [Remote closed the connection] 23:28:41 I see 23:29:18 -!- fvw [n=sdfpme@113.77.216.106] has quit [Remote closed the connection] 23:29:35 X-Scale2 [i=email@89.180.145.28] has joined #lisp 23:29:39 slyrus, bootstrapping is hard, but it's fun. 23:29:42 or is it 'fun 23:29:45 or ''fun ? 23:30:09 spilman [n=spilman@ARennes-552-1-89-252.w92-139.abo.wanadoo.fr] has joined #lisp 23:30:09 anyway. cl-launch was an exercise in a short bootstrap. xcvb is a long bootstrap 23:30:12 *hefner* dares to imagine a world without pathnames, and where CFFI doesn't bother swaddling libraries inside silly objects for whatever internal bookkeeping purpose 23:30:30 Fare: so since xcvb handles source/target issues well, one should be able to use xcvb to build, say, SBCL, no? 23:31:01 slyrus, certainly! 23:31:06 but I'm not going to do this one 23:31:47 hi 23:32:11 *Fare* has been working on XCVB for 12 hours now. Time to stop. 23:32:25 Thanks to all who contributed comments. 23:32:39 -!- Fare [n=Fare@98.216.111.110] has quit ["Leaving"] 23:33:26 -!- bobvs [n=bob@cpc2-nott8-0-0-cust610.nott.cable.ntl.com] has quit ["Miranda IM! Smaller, Faster, Easier. http://miranda-im.org"] 23:33:46 I have a problem with GNU Common Lisp, when I try to launch it (with "gcl" command), I have a Segmentation Error 23:34:12 (I'm using Fedora 10, and GCL 2.67) 23:34:28 spilman: unless you have a specific need for GCL, you should probably use some other implementation. 23:35:10 like what implementations? 23:36:06 spilman: my favorite is SBCL. 23:36:28 ok, I will test that, thanks 23:37:53 hefner: you'll thank those silly objects when shuffletron works on systems that have libfoo.so.3 instead of the libfoo.so you happen to have. (someday!) :-) 23:41:28 -!- ASau` [n=user@193.138.70.52] has quit [Read error: 113 (No route to host)] 23:42:52 -!- mrsolo_ [n=mrsolo@adsl-68-126-215-139.dsl.pltn13.pacbell.net] has quit [Read error: 60 (Operation timed out)] 23:42:52 -!- Blkt [n=Blkt@dynamic-adsl-94-37-247-157.clienti.tiscali.it] has quit [Read error: 60 (Operation timed out)] 23:44:02 luis: perhaps, but I've already learned that you can't depend on anything but glibc (and not even that, if you stretch your timeframe out to more than 6 or 7 years) 23:44:25 fvw [n=sdfpme@113.77.216.106] has joined #lisp 23:44:47 oh well, that's true. 23:45:06 Captain_Thunder [n=jimmymil@c-24-35-95-148.customer.broadstripe.net] has joined #lisp 23:46:01 davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #lisp 23:46:36 -!- TR2N [i=email@89-180-145-28.net.novis.pt] has quit [Connection timed out] 23:49:04 -!- ia [n=ia@89.169.189.230] has quit ["...may the Source be with you..."] 23:54:51 -!- spilman [n=spilman@ARennes-552-1-89-252.w92-139.abo.wanadoo.fr] has quit ["Quitte"] 23:55:14 -!- athos [n=philipp@92.250.250.68] has quit [Remote closed the connection] 23:56:41 -!- Jabberwock [n=jens@port-3757.pppoe.wtnet.de] has quit [Read error: 60 (Operation timed out)] 23:58:08 ia [n=ia@89.169.189.230] has joined #lisp