00:06:01 -!- Nshag [i=user@Mix-Orleans-106-2-151.w193-248.abo.wanadoo.fr] has quit [Read error: 60 (Operation timed out)] 00:07:11 -!- apo [n=apo@pD9E7F1F0.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 00:07:13 -!- Odin- [n=sbkhh@s121-302.gardur.hi.is] has quit [] 00:09:47 nvoorhies [n=nvoorhie@adsl-75-36-207-191.dsl.pltn13.sbcglobal.net] has joined #lisp 00:10:07 -!- pstickne [n=pstickne@c-24-21-76-57.hsd1.wa.comcast.net] has quit [Read error: 110 (Connection timed out)] 00:10:56 <_3b> is there an easy way to see the args to a function in ldb? 00:11:07 -!- chessguy [n=chessguy@67-130-43-2.dia.static.qwest.net] has quit ["Leaving"] 00:11:58 blbrown [n=Berlin@c-71-236-25-127.hsd1.ga.comcast.net] has joined #lisp 00:12:36 Probably not. 00:13:33 I would not be surprised if we could get most of the benefit of ldb from automatically dumping all the interrupt contexts and a backtrace and quitting. 00:14:09 Okay, my bisect is directing me to build .30, and says there are 4 commits left to test. 00:14:44 <_3b> sounds like you will get same answer i did 00:16:04 <_3b> my guess is that it is the stuff for patching in source locations after the bit that needed to be reproducible was done 00:16:29 Quite possibly. 00:18:10 -!- bobbysmith007 [n=russ@one.firewall.gnv.acceleration.net] has quit ["Leaving."] 00:18:42 <_3b> not sure if file-write-date is failing for some reason and then failing to print an error, or calling format and that fails 00:19:06 -!- gonzojive [n=red@fun.Stanford.EDU] has quit [Remote closed the connection] 00:19:13 It's probably calling stat(). 00:19:17 bombshelter13_ [n=bombshel@209-161-233-254.dsl.look.ca] has joined #lisp 00:19:25 mokogobo [n=mokogobo@pcp075595pcs.unl.edu] has joined #lisp 00:19:26 Which means it needs a "real" filename. 00:19:38 Well, maybe. 00:19:51 -!- apo_ is now known as apo 00:22:53 -!- slyrus_ [n=slyrus@dsl092-019-253.sfo1.dsl.speakeasy.net] has quit [Read error: 110 (Connection timed out)] 00:23:18 -!- bombshelter13_ [n=bombshel@209-161-233-254.dsl.look.ca] has left #lisp 00:23:39 antifuchs: that was the easy part of the debate ;) I do prefer numbers myself 00:23:58 -!- oudeis [n=oudeis@92.61.225.131] has quit ["This computer has gone to sleep"] 00:25:45 hmm... interesting. For (time (sleep 2)), SBCL and CLISP didn't cons anything, while ECL consed 80 bytes and did GC once. CLISP on the other hand registered real runtime longer compared to SBCL and ECL.. 00:26:44 mrsolo [n=mrsolo@adsl-68-126-205-8.dsl.pltn13.pacbell.net] has joined #lisp 00:27:06 -!- The-Kenny [n=moritz@p5087AD31.dip.t-dialin.net] has quit [Read error: 60 (Operation timed out)] 00:28:23 p_l: Have you considered repeated trials rather than a single data point from each implementation? 00:28:54 oudeis [n=oudeis@bzq-82-81-239-4.cablep.bezeqint.net] has joined #lisp 00:29:32 nyef: I did it out of complete curiosity (aka "doing everything except doing the fucking java assignment") 00:29:38 It might also be instructive to (time (eval '(time (sleep 2))))... 00:30:26 <_3b> time on sbcl measures consing fairly coarsely also 00:31:26 <_3b> as long as none of them give the same result as 1.0.18 sbcl :) 00:31:46 envi^laptop [n=envi@115.94.144.228] has joined #lisp 00:33:14 -!- tomoyuki28jp [n=tomoyuki@w221062.ppp.asahi-net.or.jp] has quit [Remote closed the connection] 00:33:21 *p_l* decided that he might get commonqt to work as another form of procrastination 00:33:35 -!- Samy [n=sbahra@c-76-26-144-229.hsd1.dc.comcast.net] has quit [Remote closed the connection] 00:34:05 tritchey_ [n=tritchey@c-68-58-88-241.hsd1.in.comcast.net] has joined #lisp 00:34:53 at least CommonQT doesn't have the promise of comments "why didn't you use subclasses and inheritance to implement this" 00:35:10 p_l: So, I can't distract you by asking if you have any ideas for a better project name for "lameulator"? 00:35:19 hmm... suddenly append can be found again... yay? 00:35:26 -!- tritchey [n=tritchey@c-68-58-88-241.hsd1.in.comcast.net] has quit [Read error: 60 (Operation timed out)] 00:35:27 nyef: oh, you can, can ! :) 00:35:39 most things are better than writing Java 00:35:48 For my hello lisp I'm trying to implement bubble sort 00:35:57 and falling on my face, for the most part 00:36:37 Though I'm enjoying myself... 00:37:23 Last build in the bisection... 00:37:30 *p_l* facepalms after his first try at name nearly ended as "N.S.D.A.P." 00:37:52 And I fully expect it to tell me what _3b already knows. 00:38:42 http://pastebin.com/m76f88380 <-- infinite recursion by mistake. The only way I could make sense of it would be that pop is either not mutating 'unsorted' or I have a parenthesis off somewhere... 00:39:08 Or the fact that I've been vomitting and having hot/cold flashes for the past three days is catching up to me... 00:39:10 So, it occurred to me today that, should I ever write a Go-playing program, the perfect name for it would be TAGBODY. 00:39:22 nyef: why? 00:39:30 p_l: GO. 00:39:35 haha 00:39:43 ROTFL 00:40:33 -!- nvoorhies [n=nvoorhie@adsl-75-36-207-191.dsl.pltn13.sbcglobal.net] has quit [] 00:40:40 chessguy_work [n=chessguy@67-130-43-2.dia.static.qwest.net] has joined #lisp 00:41:27 tritchey [n=tritchey@c-68-58-88-241.hsd1.in.comcast.net] has joined #lisp 00:42:09 I guess I could step into it and see what exactly is going on 00:42:10 cipher [n=cipher@pool-173-48-136-239.bstnma.fios.verizon.net] has joined #lisp 00:42:30 -!- chessguy_work is now known as chessguy 00:42:34 -!- tritchey_ [n=tritchey@c-68-58-88-241.hsd1.in.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 00:43:02 stylistically though, does anyone have any suggestions? Not asking someone to debug my code (not yet ;-)) 00:43:15 -!- cads [n=max@c-76-122-89-218.hsd1.ga.comcast.net] has quit [Read error: 110 (Connection timed out)] 00:43:34 -!- sykopomp [n=sykopomp@unaffiliated/sykopomp] has quit [Excess Flood] 00:43:48 Though I think perhaps I should just take a nap 00:44:01 sykopomp [n=sykopomp@unaffiliated/sykopomp] has joined #lisp 00:44:08 loz- [n=loz@203.181.dsl.syd.iprimus.net.au] has joined #lisp 00:44:11 It's unusual to have the arglist on a separate line. Unsupplied arguments default to nil anyway. 00:44:11 _dulouz [n=ross@dsl254-119-219.nyc1.dsl.speakeasy.net] has joined #lisp 00:44:25 haha 00:44:29 right, duh 00:44:42 -!- _dulouz is now known as dulouz 00:44:47 Cleaner already :-) 00:44:47 You might do better to have your recursion termination condition -first- in your COND. 00:45:12 *p_l* recalls GNU style seeing separate line for arglist 00:45:43 nyef: Wouldn't that make for extra logic? The condition is the inverse of the other two 00:46:27 ... And the big thing is that you're calling POP as if you expect it to mutate the copy of UNSORTED in the caller, which simply doesn't work. 00:46:46 At least, that's the impression I'm getting. 00:47:19 nyef: That's where I figured the failure was. Is there a cleaner solution to that chunk that doesn't require mutation? 00:47:32 nyef: Though I don't understand why it's failing 00:47:33 And 1.0.27.31 is listed as the first bad commit... As expected. 00:49:22 Well, you're actually not using the return value of bubble-sort outside of the toplevel invocation, which is also kindof wierd... 00:50:11 The call (bubble-sort sorted) seems, at first glance, to be less than useful. 00:50:24 -!- oudeis [n=oudeis@bzq-82-81-239-4.cablep.bezeqint.net] has quit ["This computer has gone to sleep"] 00:51:07 nyef: I think you're right. Having the termination case in mind when someone starts interpreting the function gives them an idea of what is being worked towards 00:51:12 Fixing... 00:51:17 I would consider attempting to reformulate the problem so as to allow for tail-recursion, as that would require the various bits of state to be more explicit... 00:51:39 oudeis [n=oudeis@bzq-82-81-239-4.cablep.bezeqint.net] has joined #lisp 00:51:49 Balooga [n=luke@208.87.19.36] has joined #lisp 00:51:58 And I won't give you grief about trying to implement one of the lamer sort algorithms. 00:52:11 -!- joshe [n=aurum@opal.elsasser.org] has left #lisp 00:52:36 <_3b> does seem an odd choice to implement recursively though 00:52:48 hehe, the names of algorithms were sorted alpha-numerically on Wikipedia 00:53:00 octoberdan: lisp generally works best in a functional manner. Where possible, try to implement something new in a functional manner and make it more efficient later :) (functional doesn't imply purely functional) 00:53:04 So... bogo-sort should come first, shouldn't it? 00:53:05 _3b: Seemed like it'd be easier that way 00:53:21 madnificent: Gotcha 00:53:52 Balooga pasted "An OpenRM/OpenGL backend for SDL in just a few lines of Common Lisp code." at http://paste.lisp.org/display/79165 00:53:57 -!- spradnyesh [n=pradyus@117.192.6.193] has quit [Connection timed out] 00:54:43 *octoberdan* rewrites his function 00:55:28 octoberdan: Please consider using lisppaste instead of that other paste service for lisp code. 00:55:40 nyef: Will do 00:56:51 sellout [n=greg@c-24-128-50-176.hsd1.ma.comcast.net] has joined #lisp 00:58:17 -!- grkz [n=qsvans@pdpc/supporter/active/grkz] has left #lisp 00:58:25 Wow 00:58:43 nyef: Things really cleaned up when I move the termination 00:58:49 -!- ejs [n=eugen@196-194-124-91.pool.ukrtel.net] has quit ["This computer has gone to sleep"] 00:58:50 *moved 00:58:58 *terminal? 00:59:37 -!- Balooga [n=luke@208.87.19.36] has quit ["Balooga has no reason"] 00:59:46 down two lines of code ;-) 01:00:45 octoberdan: you can make an annotation of your paste with the new version :) 01:02:05 -!- envi^laptop [n=envi@115.94.144.228] has quit ["Leaving"] 01:02:17 envi^laptop [n=envi@115.94.144.228] has joined #lisp 01:03:04 -!- mathrick [n=mathrick@users177.kollegienet.dk] has quit [Read error: 104 (Connection reset by peer)] 01:03:13 _mathrick [n=mathrick@users177.kollegienet.dk] has joined #lisp 01:03:51 grkz [n=qsvans@pdpc/supporter/active/grkz] has joined #lisp 01:04:00 -!- SirNick [n=SirNick@mail.colortechnology.com] has quit [] 01:04:48 I would I get all the elements of the list except for the first two elements? 01:05:03 Is there some kind of section function I can specify a range to? 01:05:35 -!- chessguy [n=chessguy@67-130-43-2.dia.static.qwest.net] has quit ["Leaving"] 01:05:37 octoberdan: cddr 01:05:40 you can use subseq, but why not just say (rest (rest list)) or (cddr list) 01:05:49 cool 01:07:45 <_3b> antifuchs: does build bot build in a completely clean dir every time, or is stuff left from previous builds? 01:09:12 Pegazus [n=awefawe@host11.190-31-40.telecom.net.ar] has joined #lisp 01:09:57 -!- oudeis [n=oudeis@bzq-82-81-239-4.cablep.bezeqint.net] has quit ["This computer has gone to sleep"] 01:10:20 -!- thom_logn [n=thom@pool-96-229-99-100.lsanca.dsl-w.verizon.net] has quit ["Leaving"] 01:10:48 I guess it would be cheating using the sort function in my sorting algorithm... 01:15:32 <_3b> nyef: apparently it is trying to get file-write-date of a nonexistant file 01:16:25 Hunh. What fil... something via an LPT that isn't set up, isn't it? 01:16:54 <_3b> ? 01:17:05 <_3b> you mean the nonexistant file? 01:17:08 Yeah. 01:17:32 Hrm, that's not quite right, as the LPTs are initialized just before the debug info. 01:18:04 <_3b> ah, that LPT 01:18:13 *_3b* wondered what printers had to do with it :) 01:18:37 No, no... This is files, not streams. :-P 01:18:50 <_3b> any idea how to print out a pathspec that early in init? 01:19:41 It is theoretically a namestring, not a pathspec... 01:20:27 pelzel000 [n=user@dynamic-acs-72-23-19-139.zoominternet.net] has joined #lisp 01:21:22 I'm just going to try the obvious thing. 01:21:22 -!- tritchey [n=tritchey@c-68-58-88-241.hsd1.in.comcast.net] has quit [] 01:21:27 <_3b> what's that? 01:21:42 PRINT, of course. 01:22:09 Theoretically, it's a string, so there shouldn't be much hassle from the pprinter if it's not set up. 01:22:09 <_3b> ah, princ seemed to be breaking for me, is print safer? 01:22:25 If that doesn't work, I'll break out sb-show. 01:24:02 -!- pelzel000 [n=user@dynamic-acs-72-23-19-139.zoominternet.net] has quit [Client Quit] 01:24:17 nyef: is SBCL on Windows ready for prime time these days? 01:24:47 (Leaving aside the cygnus/emacs/SLIME issue we were talking about the other day) 01:25:38 No, the three major areas of concern (at least one of which escapes me at the moment) have yet to be addressed. 01:25:51 Fixed memory map is still an issue, for example. 01:27:28 -!- CrazyEddy [n=cabob@wrongplanet/CrazyEddy] has quit [Read error: 110 (Connection timed out)] 01:28:12 phadthai [i=mmondor@ginseng.pulsar-zone.net] has joined #lisp 01:29:34 I don't have the card that I wrote down the big problems on in any of the decks I have with me, it must still be in New Hampshire. 01:32:15 CrazyEddy [n=decatize@wrongplanet/CrazyEddy] has joined #lisp 01:32:26 <_3b> not working with slime without hacks, no way to interrupt it from slime 01:32:31 -!- Soulman [n=kvirc@42.84-48-88.nextgentel.com] has quit [Read error: 110 (Connection timed out)] 01:34:25 Well, if we get threading working then it should be a better situation all around. 01:35:14 <_3b> yeah, that would be nice in addition to anything it might fix as a side effect :) 01:36:33 -!- blbrown [n=Berlin@c-71-236-25-127.hsd1.ga.comcast.net] has quit ["Ex-Chat"] 01:36:58 Anyway, SLIME is -usable- with the new-serve-event hack, and we need something like that at some point anyway. 01:37:20 -!- zbrown [n=rufius@unaffiliated/zbrown] has left #lisp 01:39:05 manic12 [n=manic12@c-76-29-88-103.hsd1.il.comcast.net] has joined #lisp 01:43:32 hmmm... another try with compiling smoke 01:44:29 Oh just get the java assignment out of hte way already. :-P 01:45:48 -!- kjbrock [n=kevinbro@66.166.232.134] has left #lisp 01:48:06 danlei` [n=user@pD9E2C4E6.dip.t-dialin.net] has joined #lisp 01:50:16 pelzel000 [n=user@dynamic-acs-72-23-19-139.zoominternet.net] has joined #lisp 01:51:00 chessguy_work [n=chessguy@67-130-43-2.dia.static.qwest.net] has joined #lisp 01:51:23 -!- chessguy_work is now known as chessguy 01:51:53 -!- tmh [n=thomas@pdpc/supporter/sustaining/tmh] has left #lisp 01:52:51 Okay, take three on getting usable information from the build... 01:53:41 ikki [n=ikki@201.155.75.146] has joined #lisp 01:54:47 <_3b> hmm, possibly it is that SYS:OUTPUT file 01:55:23 -!- _mathrick [n=mathrick@users177.kollegienet.dk] has quit [Remote closed the connection] 01:56:01 -!- pelzel000 is now known as TimDalySr 01:57:22 Possibly. If I get a good build this time I'll have a good chance at figuring things out. 01:57:30 <_3b> nyef: you aren't building in a path with /src/ in it anywhere, are you? 01:57:37 <_3b> or rather "src" 01:57:41 ... Yes. Yes, I am. 01:57:44 nyef: oh, the java assignment is probably farther along than most other students :P 01:57:53 I'm in ~/src/lisp/sbcl/sbcl-git/. 01:58:03 <_3b> (position "src" dir :test #'string= ... 01:58:16 Of all the... 01:58:34 Odin- [n=sbkhh@s121-302.gardur.hi.is] has joined #lisp 01:59:19 -!- dulouz [n=ross@dsl254-119-219.nyc1.dsl.speakeasy.net] has quit [Remote closed the connection] 01:59:58 So, random thought: It'd be nice if boinkor's gitweb did hyperspec lookup. 02:01:01 -!- amnesiac [n=amnesiac@p3m/member/Amnesiac] has quit ["Leaving"] 02:01:41 Yeah, that looks like it'd do it. 02:01:59 Do you want to fix it, or shall I? 02:02:06 <_3b> i'll let you 02:03:58 tritchey [n=tritchey@c-68-58-88-241.hsd1.in.comcast.net] has joined #lisp 02:04:32 <_3b> might be nice if that bit could be made more resistant to file errors in general 02:05:38 mathrick [n=mathrick@users177.kollegienet.dk] has joined #lisp 02:05:41 I was thinking that he's already testing based on pathname-directory equal to "output", why not see if either of the last two elements are "src" or if the next-to-last is "compiler"... 02:05:51 -!- danlei [n=user@pD9E2FD2A.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 02:06:04 <_3b> i mean the part where it checks the file-write-date 02:06:16 Anyway, I'll get a reasonable solution soon enough... 02:06:30 Um. Wasn't thinking about making that sane. 02:07:17 <_3b> right, making it build is probably enough for during the freeze :) 02:07:46 I thought we had just less than three hours left before the freeze? 02:08:01 <_3b> could be 02:08:16 *_3b* has no idea how to fix it though 02:08:31 I have a couple ideas on making it at least build. 02:08:37 -!- rolly1975 [n=rory@5ac959a9.bb.sky.com] has quit [Read error: 113 (No route to host)] 02:08:45 legumbre` [n=user@r190-135-8-99.dialup.adsl.anteldata.net.uy] has joined #lisp 02:09:37 <_3b> could check for output later than src in the string 02:09:53 It's actually not a string, it's a list of strings. 02:10:08 <_3b> ah, right... same thing though 02:10:08 But yeah, I can see that as an option. 02:11:11 I think limiting the search to the last three components of the directory name should be sufficient. 02:11:27 ... maybe. 02:11:44 I'd better double-check. src/compiler/assembly might be a problem. 02:12:19 There we go. A real error message. 02:12:36 (SB!IMPL::SIMPLE-FILE-PERROR "failed to find the WRITE-DATE of ~A" "SYS:SRC;LISP;SBCL;SBCL-GIT;OUTPUT;STUFF-GROVELED-FROM-HEADERS.LISP" 2) 02:13:22 And, yes, I'm going to mention that pathname in my commit message. 02:13:52 *p_l* finally built smokeqt 02:15:08 Damn. src/compiler/assembly/target/ 02:15:52 Plus src/sbcl-whatever/output/ means that my initial plan wouldn't have worked anyway... 02:17:59 *_3b* wonders if src/output/... wouldn't be the easiest solution :p 02:20:50 Okay, take two, this one actually likely to build... 02:21:20 I can now see why Xof missed this. 02:22:00 -!- TimDalySr [n=user@dynamic-acs-72-23-19-139.zoominternet.net] has quit ["ERC Version 5.2 (IRC client for Emacs)"] 02:23:38 TimDalySr [n=user@dynamic-acs-72-23-19-139.zoominternet.net] has joined #lisp 02:23:40 <_3b> yeah, just made an old bug more obvious 02:25:50 -!- HET2 [i=diman@xover.htu.tuwien.ac.at] has quit ["This computer has gone to sleep"] 02:26:19 -!- legumbre [n=user@r190-135-53-122.dialup.adsl.anteldata.net.uy] has quit [Read error: 110 (Connection timed out)] 02:26:53 Take three, this time with the conditional the right way around (I hope)... 02:27:09 Clearly, I either have too much or not enough beer in me. 02:28:18 -!- dysinger [n=tim@cpe-75-80-200-182.hawaii.res.rr.com] has quit [] 02:29:48 good night (I'm postponing finishing with commonqt and java assignment ;-) ) 02:30:46 "This time for sure!" 02:30:52 p_l: Sleep well. 02:33:46 -!- ikki [n=ikki@201.155.75.146] has quit ["Leaving"] 02:34:05 -!- kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 02:34:05 _3b: So, for the really twisted, consider if somone has both "src" and "output" in the pathname for where they unpacked SBCL... 02:35:16 <_3b> well, theoretically there is only that one file that doesn't have src in the sbcl supplied part of the path 02:35:24 -!- chessguy [n=chessguy@67-130-43-2.dia.static.qwest.net] has quit ["Leaving"] 02:35:38 <_3b> and it has output after any user supplied path, so that would catch it as an exception 02:35:59 That's a point. If we check for the special-case -first-... 02:36:01 <_3b> nyef: might be reasonable to verify that the created path works before storing it though 02:36:25 lacedaemon [n=algidus@88-149-208-178.dynamic.ngi.it] has joined #lisp 02:36:28 We can't. LPNs aren't portable enough to set up a corresponding translation on the host. 02:36:53 <_3b> fair enough 02:37:10 eno__ [n=eno@adsl-70-137-142-84.dsl.snfc21.sbcglobal.net] has joined #lisp 02:37:15 -!- fe[nl]ix [n=algidus@88-149-210-173.dynamic.ngi.it] has quit [Read error: 110 (Connection timed out)] 02:37:32 Hrm. And we can't put the special-case first without adding extra error handling afterwards. 02:37:42 CyberBlue [n=yong@60.26.109.84] has joined #lisp 02:38:27 So, (and src (or (not output) (> src output))) or (and src (> src (or output -1))) ? 02:39:06 -!- eno [n=eno@nslu2-linux/eno] has quit [Nick collision from services.] 02:39:11 -!- eno__ is now known as eno 02:40:11 <_3b> or maybe (and src (not (string= "output" (car (last dir)))) 02:40:25 <_3b> which ended up a =bit longer than i though, assuming it is right to start with 02:40:37 <_3b> string-equal i guess 02:40:41 -!- mrsolo_ [n=mrsolo@nat/yahoo/x-e3d5e770336ca48b] has quit ["This computer has gone to sleep"] 02:41:15 <_3b> i'd probably pick your second option, possibly with the (or ... -1) moved to the calculation of output if it isn't used later 02:41:46 That's a fair point. The use of POSITION involves repeated string=, so it'd be faster to avoid it. 02:43:16 <_3b> true, and only checking last element fits with Xof's changes to make it more specific about that particular file 02:43:51 If this builds, I expect I'll run with it. 02:49:28 <_3b> original bug appears to have been there since 0.8.5.26 :p 02:50:09 nixzs [n=nonamme@cpe-74-64-125-220.nyc.res.rr.com] has joined #lisp 02:50:15 hi 02:50:23 I seriouly need help 02:51:16 hello? 02:51:29 <_3b> nixzs: works better if you specify with what you need help 02:51:34 *nyef* hopes it's something simple, like help with the latest SBCL from CVS not building with a wierd internal error... 02:51:53 I want to learn lisp. 02:51:59 Oh. 02:52:02 That's a bit harder. 02:52:14 minion: Tell nixzs about that-dead-sexy-book 02:52:15 nixzs: please see that-dead-sexy-book: pcl-book: "Practical Common Lisp", an introduction to Common Lisp by Peter Seibel, available at http://www.gigamonkeys.com/book/ and in dead-tree form from Apress (as of 11 April 2005). 02:52:25 And that's your starting place if you already know a bit of programming. 02:52:44 I know I have it already, I got all sorts of books from torrents. 02:52:48 If you are a layman, however, the best book is "Alice in Wonderland". 02:53:08 I just need examples on how to do stuff 02:53:20 -!- CrazyEddy [n=decatize@wrongplanet/CrazyEddy] has quit [Success] 02:53:23 Examples like... ID3 tag parsing, maybe? 02:54:10 I can certainly understand needing examples. I learned C and Unix network programming by reading and synthesizing from MUD engine sources. 02:54:45 Ooh! Build worked. Now I just need to clean up and I can commit. 02:54:54 <_3b> yay :) 02:56:03 <_3b> presumably now M-. will work on a bunch of constants nobody cares about too :) 02:56:19 do those books walk you through what code enter and so on? I mean something like 'Casting Spells' 02:56:52 PCL certainly walks you through it in places... As does PAIP. 02:57:11 wow network programing. that's like iptables right? 02:58:38 what does MUD do? 02:58:42 That's like sockets and select(2). 02:59:21 A MUD is an old-school MMORPG, from back before they had graphics and mouse input and such. 03:00:14 mrsolo_ [n=mrsolo@adsl-68-126-205-8.dsl.pltn13.pacbell.net] has joined #lisp 03:00:14 one thing I find it funny is the \ and / denoting progress in cli programs. Can I do this with lisp? 03:00:22 CrazyEddy [n=rational@wrongplanet/CrazyEddy] has joined #lisp 03:00:26 I don't know what name has it. 03:00:29 It depends. 03:00:58 I have one program that does a file transfer and displays its progress in the emacs minibuffer. 03:01:04 I mean that looks simple 03:02:00 The spinny \|/- thing is only really possible when you can do a backspace and replace operation, which isn't possible with some interfaces. 03:02:03 oh yes even a progress bar with | or # or anything. 03:02:57 interfaces meaning shells? 03:03:48 A shell window will -usually- allow it. SLIME tends not to, so I ended up with the emacs minibuffer thing. If you're doing a graphical application you almost certainly want something else anyway. 03:03:56 And so it goes. 03:04:39 err.. no graphical. 03:06:45 by the way, in order to write a GUI are mono, gtk, qt stuff required? I see some ppl using png or jpeg to make buttons and so on. Is it possible to do it in a neutral way? for instance using the RGB notation or the hex color codes 03:07:02 Again, it depends. 03:07:43 There's CLX, which is equivalent to XLib. There's a few things built on top of CLX. There are bindings to Gtk, Qt, etc. There's at least a few bindings for Win32. 03:08:19 droogie [n=user@88.238.206.216] has joined #lisp 03:08:26 -!- mrsolo_ [n=mrsolo@adsl-68-126-205-8.dsl.pltn13.pacbell.net] has quit [Read error: 60 (Operation timed out)] 03:08:37 bindings is cheating 03:08:39 If you're sufficiently determined, you could even write bindings for the linux framebuffer and input device interfaces and roll something entirely from scratch. 03:09:17 mm.. 03:09:50 I think I am a bit far from that though 03:10:00 If you want to go lower-level than even that, it's not that hard to get direct I/O port access to mess with the CRTC directly. 03:10:17 (Well, at least on Linux it isn't.) 03:12:05 mrsolo_ [n=mrsolo@nat/yahoo/x-02e176552c943961] has joined #lisp 03:12:19 as I understand bindings is a way to manipulate objects and code for another programming language. Se essentially the final product isn't the original 03:12:27 _3b: committed to CVS. 03:13:20 nixzs: Given that the Linux kernel is written in C, you have to have -some- bindings. 03:13:44 Unless you're willing to write an entire LispOS. 03:14:03 (And if you are, good luck. It's not a small project.) 03:14:59 eh.. 03:16:44 do reverse binding exist? instead of lisp using the binding the other program using its own bindings for lisp 03:17:00 X-Scale [i=email@2001:470:1f08:b3d:0:0:0:2] has joined #lisp 03:17:47 Yes. But it can sometimes be difficult to load a lisp system as an extension to a C program. 03:18:42 ok I'll try spells again 03:19:18 sbahra [n=sbahra@c-76-21-209-249.hsd1.dc.comcast.net] has joined #lisp 03:19:30 that was fun 03:23:19 ... There's also an NYC lisp group of some sort, if memory serves. If you start getting into using Lisp, it might be interesting to see if they have anything going on. 03:23:30 lispNYC 03:23:52 lispnyc.org 03:24:17 <_3b> nyef: cool, i'll try it out once it propagates to the git mirrors 03:24:52 interesting 03:28:00 _3b: Thanks for tracking it down, btw. 03:28:03 -!- Odin- [n=sbkhh@s121-302.gardur.hi.is] has quit [] 03:28:09 ignas [n=ignas@ctv-79-132-160-221.vinita.lt] has joined #lisp 03:28:22 <_3b> nyef: debugging is fun :) 03:28:46 Debugging can be fun. It can also be horribly demoralizing. 03:29:05 <_3b> yeah, helps if it is someone else's bug 03:29:07 since i've spent the last 3 days debugging, i'm with nyef today 03:29:22 It also helps if it's not something you're consistently overlooking. 03:29:37 <_3b> right, being stuck indefinitely makes it less fun 03:29:39 *nyef* remembers the D flag problem with SBCL/Win32 all too well. 03:30:03 well, considering that axiom is about 1 million "things of code" there is probably something i'm overlooking :-) 03:30:09 beach` [n=user@ABordeaux-158-1-23-60.w90-55.abo.wanadoo.fr] has joined #lisp 03:30:12 (And the problem he had with getting his netbsd/mac68k booter to work...) 03:30:30 -!- beach [n=user@86.201.109.97] has quit [Read error: 110 (Connection timed out)] 03:31:06 TimDalySr: Yeah, but when it's some part of the system that you check forwards, backwards, up and down and still don't see the bug for months on end? 03:32:13 war stories of programming, chapters 16-117 03:33:01 it is certainly easier to find them in lisp though. gdb is painful 03:33:14 what's google of code? 03:33:34 I mean why google? can ppl just code it up themselves? 03:34:28 <_3b> nixzs: did you mean google summer of code? 03:34:43 haven't looked at google of code... too busy making this round thing 03:35:01 aja [n=aja@S01060018f3ab066e.ed.shawcable.net] has joined #lisp 03:37:53 -!- CyberBlue [n=yong@60.26.109.84] has quit ["Leaving"] 03:38:07 -!- kreuter [n=kreuter@pool-96-252-14-107.bstnma.fios.verizon.net] has quit [Read error: 60 (Operation timed out)] 03:38:15 -!- yahooooo [n=yahooooo@c-76-104-183-185.hsd1.wa.comcast.net] has quit [Read error: 110 (Connection timed out)] 03:43:01 yes google of code 03:43:38 i participated in the google summer of code in the first year 03:43:50 the google of code though seems to be a code search tool 03:44:41 <_3b> 'google code search' is a code search, 'google code' is a code hosting site... 03:47:17 I've got a problem summer of code, do ppl go to googleplex to code? 03:48:05 is the word in between ** such as *objects* a lisp syntax/semmantics? 03:48:18 pure syntax 03:48:24 its a naming convention 03:48:26 whatt I mean if I can change objects to something else random 03:48:42 (defvar anything nil) 03:48:49 -!- sledge [n=chris@pdpc/supporter/student/sledge] has quit [Read error: 110 (Connection timed out)] 03:49:08 but the convention is (defvar *anything* nil) 03:49:10 so it has to be the word objects? 03:49:10 <_3b> lisp doesn't care, but the ** make it easier forthe programmer 03:49:36 sledge [n=chris@pdpc/supporter/student/sledge] has joined #lisp 03:49:46 <_3b> the * are part of the name 03:50:02 oh ok, Casting SPELs is using (setf *objects* etc) 03:50:05 "the word between **" .... as in the symbol-name of the substring between **, can be anything 03:50:18 thanks 03:50:31 great! 03:50:50 the convention makes it easier to see that you're referencing something external to the binding context 03:50:51 -!- mokogobo [n=mokogobo@pcp075595pcs.unl.edu] has quit [Read error: 110 (Connection timed out)] 03:51:26 <_3b> TimDalySr: i'd say it is more about special than external, but i don't think nixzs is worried about that yet :) 03:51:39 And perhaps more important, that you're getting a special binding. 03:51:43 Er, what _3b said. 03:51:44 +1 point to _3b 03:52:19 is telling me 'this variable is undefined *OBJECTS*' 03:52:41 <_3b> nixzs: if you are using sbcl, you can ignore that for now 03:52:43 (defvar *objects* nil) at the command prompt 03:52:54 <_3b> or that :) 03:52:54 or at the top level of your file 03:53:04 +1 for me :-) 03:53:27 yes I am using SBCL 03:53:32 <_3b> many tutorials and sample code skip that step to keep things short, but technically it is required so sbcl reminds you 03:53:46 i never ignore compiler.... they always get me back later 03:54:01 how do I reset or start from scratch? 03:54:49 (setq *global-heap-pointer* nil) .... just joking 03:55:10 -!- droogie [n=user@88.238.206.216] has quit [Remote closed the connection] 03:55:17 alright i'll ignore it 03:55:29 ,restart-inferior-lisp (if you're in slime) 03:55:31 i don't know how to answer that question except to say restart the program..... 03:55:54 lisp is very dynamic and you can just keep piling up changes... 03:56:05 my sessions sometimes last days 03:56:14 so everytime that shows up I type 0? 0: [ABORT] Exit debugging, returning to top level, 03:56:41 HET2 [i=diman@xover.htu.tuwien.ac.at] has joined #lisp 03:56:50 you'll fall into the debugger a lot.... mostly you just want to exit to the top level again 03:56:52 <_3b> ah, if you actually got there, possibly we were guessing the problem wrong :) 03:57:07 but how do I tell is not using the wrong line? 03:57:17 electric-sheep [n=user@88.238.206.216] has joined #lisp 03:57:18 -!- X-Scale [i=email@2001:470:1f08:b3d:0:0:0:2] has quit [Read error: 104 (Connection reset by peer)] 03:57:31 -!- danlei` is now known as danlei 03:57:44 can't help there. i don't use slime. 03:57:54 the actuall message is longer 03:58:05 -!- TimDalySr [n=user@dynamic-acs-72-23-19-139.zoominternet.net] has quit ["ERC Version 5.2 (IRC client for Emacs)"] 03:58:11 I am not using slime, SBCL 03:58:32 <_3b> slime is a lisp IDE for emacs 03:58:40 X-Scale [i=email@2001:470:1f08:b3d:0:0:0:2] has joined #lisp 03:58:49 nyef: i saw an interesting project on your website, SBCL on the metal, what happened to that? 03:58:53 <_3b> which can talk to SBCL, or other lisps 03:59:10 that's it, I killed the terminal with sbcl 03:59:13 electric-sheep: My attention moved elsewhere. 03:59:20 TimDalySr [n=user@dynamic-acs-72-23-19-139.zoominternet.net] has joined #lisp 03:59:49 nyef: i see ;) 04:00:19 -!- yango [n=yango@unaffiliated/yango] has quit [Read error: 104 (Connection reset by peer)] 04:00:51 electric-sheep: If you want to pick up where I left off, feel free. 04:01:18 nixzs: usually, if you're in the debugger, you just choose one of the restarts by typing a shortcut like :r1 or something. in slime you can choose by picking a number, or hitting q (or c). restarts are offers how to continue, you are the handler. 04:01:58 sebell [n=sctb@S01060016cbc2d41a.cg.shawcable.net] has joined #lisp 04:04:11 yahooooo [n=yahooooo@c-76-104-183-185.hsd1.wa.comcast.net] has joined #lisp 04:05:11 nyef: thanks, i would love to, but i really don't have the background to do that, but it'ld be really good to have a look at it, is it uploaded somewhere? 04:06:20 There's a patch on the website, and a disk image for the bootstrap Forth. 04:06:33 I'm not sure how up-to-date they are. 04:06:47 I know that one of the last things I got running was demand paging of the core from disk. 04:07:27 did forth try to write to track 0, sector 0 to save the file? 04:08:50 The Forth image only writes to disk if you try to save any changes to its blockfile or if you try to recompile the Forth system itself. 04:09:25 ISTR that the Lisp image had to be mapped to the start of the hard drive, though. 04:09:30 ah, then it is not the one true forth, grasshopper 04:10:14 <_3b> the concept of 'one true forth' seems pretty un-forthlike 04:10:18 i want to write a forth-based lisp system so that the entire file system is actually a single list object, that is, the whole disk is a single list 04:10:22 There is no one true forth. Any forth programmer worth his stacks writes his or her own. 04:10:29 forth true one? 04:10:56 how do I go onto a new line without yet entering the code? 04:11:06 if the entire filesystem is a single list object then you can "format the drive" with (setq *disk* nil) 04:12:08 nixzs: As you should expect by now, "it depends". 04:12:25 surely you're joking, Mr. Daly 04:12:53 nixzs: as long as there is one paren open, just press return. if you want to execute several toplevel forms at the repl, you could use (progn (foo) (bar ... 04:13:17 of course 04:13:46 ok 04:13:52 nixzs: And if you're defining functions at a REPL, don't. That way lies madness. 04:13:53 it was this (setf *map* '((living-room (you are in the living-room of a wizards house. there is a wizard snoring loudly on the couch.) 04:14:01 so there are more lines below 04:14:42 by the way it's here http://www.lisperati.com/data.html 04:15:17 do I indent? 04:15:40 <_3b> for best results, use an editor that indents for you 04:16:17 ......oooooo, _3b starts a religious war! 04:16:45 the war has already been won by emacs+slime 04:17:05 I see the pattern 04:18:06 blbrown [n=Berlin@c-71-236-25-127.hsd1.ga.comcast.net] has joined #lisp 04:18:35 -!- konr [n=karkeej@201.82.139.214] has quit [Remote closed the connection] 04:18:36 <_3b> TimDalySr: do people still argue in favor of editors that can't indent? 04:19:02 -!- blitz__ [n=blitz@2001:6f8:10f6:200:21b:77ff:fe41:11ab] has quit [Read error: 113 (No route to host)] 04:19:31 i use emacs and run lisp in a shell buffer 04:19:57 but i use emacs in fundamental mode so it doesn't do autoindent 04:20:07 _3b: are there editors that can't indent? i've heard some stories about them but never saw one. 04:20:17 i write "literate lisp" so the buffer is more than lisp code, it also contains latex 04:20:39 <_3b> TimDalySr: sounds like you need to make your editor smarter to deal with your workflow :) 04:21:02 -!- tritchey [n=tritchey@c-68-58-88-241.hsd1.in.comcast.net] has quit [] 04:21:11 emacs has lisp-mode, emacs has latex-mode 04:21:14 <_3b> electric-sheep: ed, notepad.exe, probably others 04:21:28 <_3b> TimDalySr: emacs has mmm-mode (or whatever replaced it) 04:21:43 but the modes change what i wrote and i'm pretty anal about saying what i mean, where i mean it 04:22:18 <_3b> well, that's whatprogrammability is for :) 04:22:24 since the literate lisp has latex chunks that contain lisp code (but not necessarily for execution) i don't want them changed 04:22:45 -!- blbrown [n=Berlin@c-71-236-25-127.hsd1.ga.comcast.net] has quit [Remote closed the connection] 04:22:49 i'm not man enough to write a literate.el 04:23:01 _3b: all of them has indention - kbd is several whitespaces afaik 04:24:37 <_3b> nyef: guess what, i'm 4secting sbcl again :/ it builds with your patch, but apparently some contribs died on win32 between .27.30 and head 04:24:37 but i'd seriously suggest that you look into writing literate lisp. its one of the few things that have deeply changed my view of programming 04:25:06 QinGW [n=user@218.241.188.34] has joined #lisp 04:25:36 -!- awayekos is now known as anekos 04:25:54 -!- tsuru [n=user@c-69-245-36-64.hsd1.tn.comcast.net] has quit [Read error: 110 (Connection timed out)] 04:27:11 <_3b> wonder if it is faster to distribute builds evenly over the range, or if there is a more efficient distribution 04:27:11 _3b: are you using win32 sbcl for your flash interface project or do you generally use it? i have a question if you have experianca with win32 sbcl 04:27:34 <_3b> electric-sheep: yeah, i am usually too lazy to boot into linux :( 04:27:51 <_3b> i try to avoid platform specific code though 04:28:00 -!- dreish [n=dreish@minus.dreish.org] has quit [] 04:28:08 -!- sebell [n=sctb@S01060016cbc2d41a.cg.shawcable.net] has left #lisp 04:28:31 That's... impressive, given that nothing changes the contrib source over that time. 04:28:32 oO 04:28:41 _3b: so do you know about this gencgc.c file error of sbcl on win? 04:28:45 <_3b> also wonder if i could get away with more than 4 builds at a time without killing speed 04:29:06 gencgc.c file error? 04:29:26 <_3b> electric-sheep: nyef is the authority, but i'll help if i can 04:29:39 On the GC? I'm most certainly -not- the authority. 04:29:43 afaik its something related to the garbage collector but i've found nothing else 04:30:19 believe me, for my level, you are both the authority :) 04:30:31 <_3b> electric-sheep: helps if you specify the error and context 04:31:23 <_3b> well, thats .35-38 ok, .40 bad 04:31:32 <_3b> and .39 bad 04:31:51 it is really not a fresh error that i get since i couldn't have any answers when i asked before, but i end up with it several times while requiring a package 04:31:56 SIGCHLD? Interesting. 04:32:06 I can easily believe that it wasn't tested on win32, though. 04:32:08 <_3b> added a test, wonder if that broke contrib tests? 04:32:42 spradnyesh [n=pradyus@117.192.13.197] has joined #lisp 04:33:29 Shouldn't. 04:33:45 I think more likely is that it broke run-program completely, thus killing sb-grovel. 04:33:52 <_3b> ah, could be 04:34:11 And without the groveller you lose sb-bsd-sockets, sb-posix, etc. 04:34:15 usually something like "couldn't blah nth line bla bla in file gencgc.c" but its really fuzzy for me, i think once i had that while i was trying to use clx. 04:34:28 <_3b> that would fit the description in the sbcl-dev mail i was testing it in order to respond to :) 04:34:47 Liempt [n=Bevinvan@S01060014bfef2498.va.shawcable.net] has joined #lisp 04:35:03 <_3b> breaks asdf-install, sb-bsd-sockets, sb-posix in addition to the usual sb-simple-streams 04:35:07 and once again with another package that i don't understand but if it will be useful i can download clx again and write the exact error that i get 04:36:48 Hrm. 04:36:59 -!- spradnyesh [n=pradyus@117.192.13.197] has left #lisp 04:37:39 _3b: It's late, and I need to see about getting some sleep. I'll take a look at it tomorrow if I don't see a fix report by then. 04:38:15 <_3b> ok, i'll send it to the list 04:38:49 electric-sheep: I'm afraid that the problem you're having sounds like it should be addressed (with considerably more detail as to the actual error messages involved) to sbcl-devel or one of the GC gurus around here. 04:39:03 Meanwhile, I'm going to make myself scares for a few hours. 04:39:06 -!- nyef [n=nyef@24.220.182.229] has quit ["G'night all."] 04:41:07 ok, thanks :) 04:42:34 -!- existentialmonk [n=carcdr@64-252-67-89.adsl.snet.net] has quit [Read error: 104 (Connection reset by peer)] 04:43:45 can make aliases for things such as setf? 04:45:05 and defun 04:45:20 what are these called? 04:47:34 durka42 [n=durka@d81.wireless.swarthmore.edu] has joined #lisp 04:52:11 argh, how do I edit a previous block of code? 04:52:29 I mean how can I invoke it 04:54:12 -!- dto [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has quit [Read error: 110 (Connection timed out)] 04:56:56 never mind, starting all over 04:57:14 *hefner* wishes there were a real ivory emulator, instead of this VLM silliness 04:57:33 <_3b> nixzs: you should use an editor that can talk to sbcl 04:57:50 or, hell, a 3600 emulator. that'd be different. 04:58:10 such as? 04:58:24 <_3b> nixzs: if you know emacs at all, slime is good 05:00:35 <_3b> nixzs: or for something simpler to get started with, you could try able ( http://phil.nullable.eu/ ) 05:02:40 -!- HET2 [i=diman@xover.htu.tuwien.ac.at] has quit ["This computer has gone to sleep"] 05:08:41 electric-sheepp [n=user@88.238.206.216] has joined #lisp 05:12:05 _3b: i found the error, it's not clx, its mcclim, not a problem while asdfing mcclim itself but it drops to lowlevel debugger while loading mcclim-examples 05:13:38 fatal error encountered in SBCL pid 1460: GC invariant lost, file "gencgc.c" line 860 05:14:24 -!- ia [n=ia@89.169.189.230] has quit [Client Quit] 05:14:47 <_3b> most likely, it means sbcl's heap is corrupted somehow 05:14:54 i mean when i try (asdf:oos 'asdf:load-op :clim-examples) i end up with this 05:15:00 hmm 05:15:11 <_3b> any idea what mcclim backend it is using? or what the example it was trying to run does? 05:15:39 i set it to use clx, but i really don't know the example 05:15:59 i never achieved the honor of seeing it :) 05:17:42 <_3b> does it give you an ldb prompt, or just exit completely? 05:17:54 ldb promt 05:18:18 <_3b> try backtrace at that prompt, and lisppaste the results 05:20:43 how about aliases, can make aliases for commands? 05:20:47 -!- mrsolo_ [n=mrsolo@nat/yahoo/x-02e176552c943961] has quit [Read error: 104 (Connection reset by peer)] 05:23:01 ok, i'll do that, by the way, i see that i also get an error when (asdf:oos 'asdf:load-op :clx) first, i remember seeing "if it gives an error try 'accept' at debugger" on manual and it gave that error and continued when i accept. could this error be the consequence of the clx error? 05:24:08 <_3b> probably depends on the error 05:25:06 -!- octoberdan [n=user@c-65-96-169-141.hsd1.ma.comcast.net] has quit [Read error: 110 (Connection timed out)] 05:27:23 debugger invoked on a ASDF:COMPILE-WARNED: error while invoking # on # 05:28:28 but that's doesn't seem a very useful error :) 05:30:36 ehu` [n=chatzill@82-170-33-173.ip.telfort.nl] has joined #lisp 05:31:32 DeusExPikachu [n=DeusExPi@ip98-164-199-153.oc.oc.cox.net] has joined #lisp 05:32:56 <_3b> yeah, you need to scroll back and see what the warning or error was that caused the compilation to fail 05:34:49 no way does CLX cause (directly) SBCL to die in the GC when you haven't even run an example yet 05:34:59 i think the best choice will be to use mcclim directly in linux :) it seems pretty unreliable in win on xming bla bla, even if it works :) 05:36:19 -!- gigamonkey [n=user@adsl-99-35-219-234.dsl.pltn13.sbcglobal.net] has quit [Read error: 110 (Connection timed out)] 05:37:06 well, here's some new user tips that maybe aren't documented anywhere. sometime aftering load the mcclim system, load the mcclim-freetype system, and also do (climi::use-pixie) 05:37:46 hefner: yes i was just loading examples not running one, at first it seemed probably unrelated to me too, but i'm not sure because i don't exacly know how it works 05:38:44 also there were not even one error while loading mcclim itself so it didn't seem like a mcclim error too, so i'm lost :) 05:39:54 hefner: hmm, i'll try that soon and inform you if it works 05:40:00 -!- yahooooo [n=yahooooo@c-76-104-183-185.hsd1.wa.comcast.net] has quit [Remote closed the connection] 05:42:58 SandGorgon [n=user@122.162.233.53] has joined #lisp 05:44:24 hefner: how can i load freetype system? 05:47:34 oh, you need to load the asd first. symlink it wherever asdf searches for systems, or load it (load mcclim/Experimental/freetype/mcclim-freetype.asd") 05:52:13 yahooooo [n=yahooooo@c-76-104-183-185.hsd1.wa.comcast.net] has joined #lisp 05:52:14 hmm, i loaded it but then while asdfing it gave several errors, and it debugger itself is totally locked at the last error 05:53:19 what sort of errors? (also, don't try this on windows) 05:54:20 you mean running mcclim on windows? 05:57:55 eno__ [n=eno@adsl-70-137-146-126.dsl.snfc21.sbcglobal.net] has joined #lisp 05:58:56 sure, but specifically freetype. I seriously doubt it will work without luck and some hand holding. 05:59:08 -!- eno [n=eno@nslu2-linux/eno] has quit [Nick collision from services.] 05:59:13 -!- eno__ is now known as eno 05:59:33 -!- drwhen [n=d@216-67-73-247-rb1.fai.dsl.dynamic.acsalaska.net] has quit ["Ex-Chat"] 06:00:45 yes you're right, and also i seriously doubt xp is worth that pain :) 06:01:00 (split-sequence:split-sequence "." "0.6.6") -- that doesn't do anything even vaguely like what I'd expect. Am I missing something? Is there a better way to do that? 06:01:16 (using the CL utilities split-sequence there) 06:01:46 hefner: i'll switch ubuntu or whatever i think :) 06:02:06 Ah. #\. Nevermind. Thanks for listening; I'll shut up now. :) 06:02:54 dv_ [n=dv@85-127-117-181.dynamic.xdsl-line.inode.at] has joined #lisp 06:03:34 anyway, i need some sleep, thanks, bye everyone 06:03:44 -!- electric-sheepp [n=user@88.238.206.216] has quit [Remote closed the connection] 06:06:00 -!- vy [n=user@213.139.194.186] has quit [Remote closed the connection] 06:06:38 lat [n=lat@125.167.140.159] has joined #lisp 06:06:47 -!- lat [n=lat@125.167.140.159] has quit [Read error: 104 (Connection reset by peer)] 06:09:18 -!- davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 06:11:46 What's the CL noop? 06:15:54 nil? 06:16:27 No, I mean something that I can put in, for example, (case ('foo [here])) and not doing anything. 06:16:38 s/not doing/not do/ 06:16:55 i think there is a nothing symbol.... /me reaches for his book 06:17:44 This would be a nothing function, SLT. 06:17:50 OSLT. Can't type tonight. 06:18:03 rlpowell: identity 06:18:54 (let ((x 0)) (case x (0))) ;; just leave it empty 06:19:16 Oh, OK. 06:19:17 btw, you shouldn't quote the 'foo in (case .... (foo here)) 06:19:34 IIRC there's a general no-op, but I was in fact using case. Good to know. 06:19:41 the symbols in a case macro are unevaluated 06:19:55 OK, thanks. 06:21:19 *rlpowell* is working on a half-assed ASDF to mudballs converter, cuz he's weird like that. 06:27:32 Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has joined #lisp 06:34:56 -!- QinGW [n=user@218.241.188.34] has quit [Remote closed the connection] 06:43:23 ejs [n=eugen@94.179.120.179] has joined #lisp 06:54:29 xan [n=xan@cs78225040.pp.htv.fi] has joined #lisp 06:58:14 sykopomp` [n=sykopomp@unaffiliated/sykopomp] has joined #lisp 07:02:23 -!- sykopomp [n=sykopomp@unaffiliated/sykopomp] has quit [Read error: 111 (Connection refused)] 07:04:33 Soulman [n=kvirc@42.84-48-88.nextgentel.com] has joined #lisp 07:05:55 -!- electric-sheep [n=user@88.238.206.216] has quit [Read error: 104 (Connection reset by peer)] 07:08:31 luis` [n=user@r42.eu] has joined #lisp 07:09:14 -!- luis [n=user@r42.eu] has quit [Read error: 104 (Connection reset by peer)] 07:16:28 daniel_ [i=daniel@unaffiliated/daniel] has joined #lisp 07:17:41 stassats` [n=stassats@wikipedia/stassats] has joined #lisp 07:17:52 Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has joined #lisp 07:19:56 -!- sykopomp` is now known as sykopomp 07:20:56 -!- stassats [n=stassats@wikipedia/stassats] has quit [Read error: 60 (Operation timed out)] 07:22:52 dialtone [n=dialtone@adsl-64-175-33-111.dsl.pltn13.pacbell.net] has joined #lisp 07:24:58 thom_logn [n=thom@pool-96-229-99-100.lsanca.dsl-w.verizon.net] has joined #lisp 07:25:44 I noticed theres a lot more discussion in the mailing list for SBCL. I good sign of progress I think. 07:26:54 <_3b> commits to cvs are a better sign :) 07:27:22 true 07:27:45 but its just I wake up, check my mail, boom 50+ emails 07:27:49 most from sbcl emailing list 07:33:37 -!- daniel [i=daniel@unaffiliated/daniel] has quit [Read error: 110 (Connection timed out)] 07:37:42 dwave [n=ask@084202073105.customer.alfanett.no] has joined #lisp 07:41:53 _3b: the boinkmarks build starts from an entirely clean working directory at al times (: 07:42:32 it does git clean -(all sorts of stuff-removing flags) and git reset --hard to the revision 07:43:12 <_3b> antifuchs: yeah, it turned out to be the path causing the problem :) 07:43:23 very good 07:45:44 <_3b> (for the record, don't try to bisect trhough .27.* in a path containing a dir named src :) 07:46:12 <_3b> though i guess you don't need to build to bisect, so it shouldn't matter 07:47:09 vy [n=user@213.139.194.186] has joined #lisp 07:47:20 heh, that is a cute failure (: 07:47:34 SirNick [n=SirNick@c-24-20-213-118.hsd1.wa.comcast.net] has joined #lisp 07:47:43 <_3b> yeah, wouldn't be so bad if it could print the error when it fails 07:49:45 -!- aja [n=aja@unaffiliated/aja] has quit [Client Quit] 07:50:53 gemelen [n=shelta@shpd-78-36-164-77.static.vologda.ru] has joined #lisp 07:52:58 -!- SirNick [n=SirNick@c-24-20-213-118.hsd1.wa.comcast.net] has quit [] 08:00:09 ehu`_ [n=chatzill@82-170-33-173.ip.telfort.nl] has joined #lisp 08:05:02 jao [n=jao@131.Red-79-156-141.staticIP.rima-tde.net] has joined #lisp 08:09:53 -!- manuel___ [n=manuel@HSI-KBW-078-043-184-124.hsi4.kabel-badenwuerttemberg.de] has quit [] 08:10:01 schoppenhauer [n=css@unaffiliated/schoppenhauer] has joined #lisp 08:10:07 Jacob_H [n=jacob@92.4.30.12] has joined #lisp 08:10:25 tombom [n=tombomp@wikipedia/Tombomp] has joined #lisp 08:10:32 tcr [n=tcr@host145.natpool.mwn.de] has joined #lisp 08:11:37 <_3b> current sbcl CVS does pass all (or as many as usual) contrib tests on normal platforms right? 08:12:02 -!- Qsource [i=dima@torch.blackened.com] has left #lisp 08:12:12 *_3b* is too lazy to actually start up a VM and build it 08:14:31 -!- rdd [n=rdd@c83-250-153-45.bredband.comhem.se] has quit [Remote closed the connection] 08:15:38 rdd [n=user@c83-250-153-45.bredband.comhem.se] has joined #lisp 08:18:31 -!- ehu` [n=chatzill@82-170-33-173.ip.telfort.nl] has quit [Read error: 110 (Connection timed out)] 08:19:11 mrsolo_ [n=mrsolo@adsl-68-126-205-8.dsl.pltn13.pacbell.net] has joined #lisp 08:19:41 tombom_ [n=tombomp@wikipedia/Tombomp] has joined #lisp 08:19:57 -!- DeusExPikachu [n=DeusExPi@ip98-164-199-153.oc.oc.cox.net] has quit ["Leaving"] 08:26:29 -!- sledge [n=chris@pdpc/supporter/student/sledge] has quit [Read error: 110 (Connection timed out)] 08:27:19 sledge [n=chris@pdpc/supporter/student/sledge] has joined #lisp 08:29:35 <_3b> ah, nevermind, i think i see it now... 08:31:17 _3b, http://rafb.net/p/T9BcUp13.html 08:31:54 that's what i got just now .. i don't run these tests very often so i don't what's normal 08:32:26 <_3b> i mean just the normal ones on building 08:33:56 <_3b> doesn't look like any of the parts i was wondering about though 08:34:48 <_3b> now i just need to try to form coherent thoughts to send to sbcl-devel :) 08:35:08 fiveop [n=fiveop@pD9E6C5C5.dip.t-dialin.net] has joined #lisp 08:35:40 -!- loz- [n=loz@203.181.dsl.syd.iprimus.net.au] has quit ["Leaving"] 08:36:02 -!- tombom [n=tombomp@wikipedia/Tombomp] has quit [Read error: 110 (Connection timed out)] 08:36:02 -!- tombom_ is now known as tombom 08:51:41 -!- envi^laptop [n=envi@115.94.144.228] has quit [Read error: 104 (Connection reset by peer)] 08:52:45 athos [n=philipp@92.250.250.68] has joined #lisp 08:55:52 oudeis [n=oudeis@bzq-82-81-239-4.cablep.bezeqint.net] has joined #lisp 08:59:24 kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 09:00:33 dstatyvka [i=ejabberd@pepelaz.jabber.od.ua] has joined #lisp 09:00:41 ruediger [n=the-rued@62-47-158-122.adsl.highway.telekom.at] has joined #lisp 09:02:21 -!- schoppenhauer [n=css@unaffiliated/schoppenhauer] has quit ["Lost terminal"] 09:04:19 -!- Jacob_H [n=jacob@92.4.30.12] has quit [Read error: 110 (Connection timed out)] 09:04:48 -!- yahooooo [n=yahooooo@c-76-104-183-185.hsd1.wa.comcast.net] has quit [Read error: 110 (Connection timed out)] 09:04:51 Jacob_H [n=jacob@92.2.126.235] has joined #lisp 09:06:29 sepult [n=buggarag@xdsl-87-78-74-72.netcologne.de] has joined #lisp 09:07:32 Jabberwockey [n=Tumnus_@port-16305.pppoe.wtnet.de] has joined #lisp 09:12:24 -!- SandGorgon [n=user@122.162.233.53] has quit [Remote closed the connection] 09:13:07 SandGorgon [n=user@122.162.233.53] has joined #lisp 09:15:26 -!- kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit ["leaving"] 09:22:59 kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 09:24:25 -!- ejs [n=eugen@94.179.120.179] has quit [Read error: 110 (Connection timed out)] 09:28:29 nostoi [n=nostoi@80.31.175.82] has joined #lisp 09:29:50 -!- beach` is now known as beach 09:30:41 joachifm [n=joachim@bjo1-1x-dhcp143.studby.uio.no] has joined #lisp 09:35:52 -!- benny [n=benny@i577A16B1.versanet.de] has quit [Read error: 110 (Connection timed out)] 09:37:11 jmbr [n=jmbr@87.223.68.107] has joined #lisp 09:37:42 -!- mrsolo_ [n=mrsolo@adsl-68-126-205-8.dsl.pltn13.pacbell.net] has quit [Read error: 110 (Connection timed out)] 09:40:12 -!- ruediger [n=the-rued@62-47-158-122.adsl.highway.telekom.at] has quit ["This computer has gone to sleep"] 09:41:45 CyberBlue [n=yong@60.26.109.84] has joined #lisp 09:42:29 ejs [n=eugen@191-202-124-91.pool.ukrtel.net] has joined #lisp 09:45:34 -!- dv_ [n=dv@85-127-117-181.dynamic.xdsl-line.inode.at] has quit ["Verlassend"] 09:46:49 decaf [n=mehmet@88.232.178.8] has joined #lisp 09:47:10 ruediger [n=the-rued@62-47-158-122.adsl.highway.telekom.at] has joined #lisp 09:51:09 yahooooo [n=yahooooo@76.104.183.185] has joined #lisp 09:51:43 rindolf [n=shlomi@bzq-219-139-216.static.bezeqint.net] has joined #lisp 09:54:23 wg1024 [n=wg1024@dslb-084-058-181-192.pools.arcor-ip.net] has joined #lisp 09:58:05 -!- wg1024 [n=wg1024@dslb-084-058-181-192.pools.arcor-ip.net] has left #lisp 09:58:10 wg1024 [n=wg1024@dslb-084-058-181-192.pools.arcor-ip.net] has joined #lisp 09:59:37 -!- xan is now known as xan-afk 09:59:43 nvoorhies [n=nvoorhie@adsl-75-36-207-191.dsl.pltn13.sbcglobal.net] has joined #lisp 10:00:38 chris2_ [n=chris@p5B168793.dip0.t-ipconnect.de] has joined #lisp 10:00:56 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [] 10:01:28 Adamant [n=Adamant@c-76-29-188-22.hsd1.ga.comcast.net] has joined #lisp 10:02:59 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [Client Quit] 10:03:03 -!- wg1024 [n=wg1024@dslb-084-058-181-192.pools.arcor-ip.net] has left #lisp 10:10:23 kidd [n=kidd@60.Red-79-150-112.dynamicIP.rima-tde.net] has joined #lisp 10:13:03 roidrage [n=roidrage@dslb-084-058-181-192.pools.arcor-ip.net] has joined #lisp 10:13:31 -!- ejs [n=eugen@191-202-124-91.pool.ukrtel.net] has quit ["This computer has gone to sleep"] 10:16:28 jewel [n=jewel@dsl-242-183-106.telkomadsl.co.za] has joined #lisp 10:17:26 -!- CyberBlue [n=yong@60.26.109.84] has quit ["Leaving"] 10:23:00 -!- Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has quit [Remote closed the connection] 10:26:08 -!- ehu`_ [n=chatzill@82-170-33-173.ip.telfort.nl] has quit [Read error: 110 (Connection timed out)] 10:26:51 -!- ruediger [n=the-rued@62-47-158-122.adsl.highway.telekom.at] has quit ["This computer has gone to sleep"] 10:31:00 LostMonarch [n=roby@host205-182-dynamic.57-82-r.retail.telecomitalia.it] has joined #lisp 10:38:54 -!- lacedaemon is now known as fe[nl]ix 10:41:17 milanj [n=milan@212.200.194.238] has joined #lisp 10:42:42 -!- akcom [i=akcom@rrcs-67-79-132-132.se.biz.rr.com] has quit [Read error: 110 (Connection timed out)] 10:43:29 ruediger [n=the-rued@62-47-158-122.adsl.highway.telekom.at] has joined #lisp 10:44:17 -!- antgreen1 [n=green@CPE0014bf0b631a-CM001e6b1858fa.cpe.net.cable.rogers.com] has left #lisp 10:47:09 -!- nostoi [n=nostoi@80.31.175.82] has quit ["Verlassend"] 10:47:37 -!- mathrick [n=mathrick@users177.kollegienet.dk] has quit [Remote closed the connection] 10:49:46 apo_ [n=apo@217.231.197.78] has joined #lisp 10:57:17 -!- plutonas [n=plutonas@c-83-233-152-13.cust.bredband2.com] has quit [Remote closed the connection] 10:58:59 puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has joined #lisp 10:59:13 -!- apo [n=apo@pD9E7E4B2.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 11:02:32 can I let clbuild fetch a branch of a git 11:02:48 repository (why is so close to - on dvorak?) 11:02:53 -!- kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 11:03:38 -!- abeaumont [n=abeaumon@84.76.48.250] has quit [Remote closed the connection] 11:03:59 Spyderco [n=nash@194.45.110.65] has joined #lisp 11:04:05 ThomasI [n=thomas@unaffiliated/thomasi] has joined #lisp 11:05:24 -!- jeremiah [n=jeremiah@31.Red-213-98-123.staticIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)] 11:11:56 -!- tcr [n=tcr@host145.natpool.mwn.de] has quit ["Leaving."] 11:12:04 Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has joined #lisp 11:12:43 -!- tombom [n=tombomp@wikipedia/Tombomp] has quit ["Peace and Protection 4.22.2"] 11:13:57 mathrick [n=mathrick@users177.kollegienet.dk] has joined #lisp 11:16:10 -!- deech [n=deech@71-10-166-226.dhcp.stls.mo.charter.com] has quit [Read error: 104 (Connection reset by peer)] 11:16:45 deech [n=deech@71-10-166-226.dhcp.stls.mo.charter.com] has joined #lisp 11:17:22 -!- billstclair [n=billstcl@unaffiliated/billstclair] has quit [] 11:18:50 -!- dkcl [n=dkcl@unaffiliated/dkcl] has quit ["leaving"] 11:19:16 dkcl [n=dkcl@unaffiliated/dkcl] has joined #lisp 11:19:38 not yet, but it should be be easy to add a "if test -n "$3"; then git checkout -b $3 origin/$3; fi" below the "git clone" in the clbuild script 11:20:23 nikodemus [n=nikodemu@cs181229041.pp.htv.fi] has joined #lisp 11:20:33 hoy 11:20:35 howdy 11:20:38 benny [n=benny@i577A1A68.versanet.de] has joined #lisp 11:20:40 hi nikodemus 11:20:40 long time, etc 11:21:40 i feel inspired to work on defglobal again 11:22:16 lichtblau: that would work, but it would'nt allow a checkout of tags etc (which might be nice too, no?) 11:23:20 jeremiah [n=jeremiah@31.Red-213-98-123.staticIP.rima-tde.net] has joined #lisp 11:24:03 surely the git way would be "git reset --hard $3" :-) 11:25:52 if I redefine a function running in a thread, I don't need to restart the thread to get the new function do I? 11:26:18 yes. though if the function has been inlined, you need to recompile the callers 11:26:32 ok thanks 11:27:02 -!- anekos is now known as awayekos 11:27:32 Cel: remeber that this is not necessairly true for macros tho. (i wasted a shitload of time on that one) 11:27:57 -!- drdo` [n=psykon@167.111.54.77.rev.vodafone.pt] has quit [Remote closed the connection] 11:28:35 yes, I've already had some non-threaded 'problems' with macros.. once I realised they live in 'the land of names' things went better 11:28:36 Krystof: just doing git checkout $3 could work, but then you'd have to write origin/my-branch to pull that particular branch 11:30:38 -!- awayekos is now known as anekos 11:33:10 -!- Patzy [n=somethin@coa29-1-88-174-11-170.fbx.proxad.net] has quit [Read error: 104 (Connection reset by peer)] 11:33:25 Patzy [n=somethin@coa29-1-88-174-11-170.fbx.proxad.net] has joined #lisp 11:33:34 anyone work with net-telent-date's parser? 11:33:52 hey, i didn't remember my old branch was almost done. i wonder if there is a snafu here that i didn't make a note of, or if it's really as done as it looks like... 11:33:53 the result I get from (date:parse-time "now") is 1hr ahead of (get-universal-time) 11:35:30 does it correlate with your time zone? 11:36:28 it's a universal time, and therefore lacks a time zone 11:36:44 however, my zone is CDT atm, which may be messing things up in the parser 11:37:31 (- (parse-time "now") (get-universal-time)) => 3600 11:38:11 HET2 [i=diman@xover.htu.tuwien.ac.at] has joined #lisp 11:42:01 -!- LostMonarch [n=roby@host205-182-dynamic.57-82-r.retail.telecomitalia.it] has quit ["raise RuntimeError"] 11:42:03 Universal time is the number of seconds commenced since January first 1970 an 0:0 time in Greewich England 11:42:15 So it sort of has one ;) 11:43:48 tsuru [n=user@c-69-245-36-64.hsd1.tn.comcast.net] has joined #lisp 11:44:17 well, one that is as meaningless as the variance in values of type '(integer 0 0) 11:45:22 elurin [n=user@81.213.201.55] has joined #lisp 11:49:50 ah, there is no way to specify to encode-universal-time whether to account for daylight savings time 11:49:53 how inconvenient 11:50:00 beach` [n=user@ABordeaux-158-1-96-102.w86-201.abo.wanadoo.fr] has joined #lisp 11:50:16 -!- beach [n=user@ABordeaux-158-1-23-60.w90-55.abo.wanadoo.fr] has quit [Read error: 60 (Operation timed out)] 11:53:38 Hmm. Maybe I can encode to universal, then decode back, and if daylight-p=>t, subtract 3600 sec 11:54:10 I'm not sure whether it would be right around the transition times, but should work most of the time 11:55:43 envi^home [n=envi@220.121.234.156] has joined #lisp 11:59:26 ehu` [n=chatzill@82-170-33-173.ip.telfort.nl] has joined #lisp 12:03:07 oh great, "The daylight saving time flag is nil if time-zone is supplied." How...useful. 12:04:57 hmm 12:05:19 beach`` [n=user@ABordeaux-158-1-69-62.w90-60.abo.wanadoo.fr] has joined #lisp 12:05:23 -!- ruediger [n=the-rued@62-47-158-122.adsl.highway.telekom.at] has quit ["This computer has gone to sleep"] 12:05:33 instead of trying to support a feature that X3J13 was apparently bent on making difficult 12:05:55 how about I code up a time machine, go back in time, and *kill* whoever thought DST was a good idea. 12:06:05 well, that's logical, how it will decide in which actual timezone do you leave? 12:06:24 ruediger [n=the-rued@62-47-158-122.adsl.highway.telekom.at] has joined #lisp 12:06:42 live 12:07:01 well it gets that right 12:07:04 zoneinfo 12:07:12 -!- beach` [n=user@ABordeaux-158-1-96-102.w86-201.abo.wanadoo.fr] has quit [Read error: 60 (Operation timed out)] 12:07:32 S11001001: i mean when you specify timezone explicitly 12:07:54 only one timezone is relevant at a time for a particular date 12:08:08 in my case, US Central 12:09:44 ztzg_ [n=dash@dslb-084-057-060-168.pools.arcor-ip.net] has joined #lisp 12:11:16 -!- roidrage [n=roidrage@dslb-084-058-181-192.pools.arcor-ip.net] has quit [] 12:12:02 hmm, well this sucks 12:12:12 I'm just going to cheat and assume the world lives in my time zone 12:14:29 hi, in my webapp, errors invoke debugger, e.g.: "debugger invoked on a SIMPLE-ERROR in thread # as it is supposed to run in the background (as a server), it is inconvenient 12:14:55 -!- jeremiah [n=jeremiah@31.Red-213-98-123.staticIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)] 12:15:07 can I do something about it, like let the thread die automatically? 12:15:12 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 12:15:37 -!- ztzg [n=dash@dslb-088-067-009-156.pools.arcor-ip.net] has quit [Read error: 110 (Connection timed out)] 12:16:08 roidrage [n=roidrage@dslb-084-058-181-192.pools.arcor-ip.net] has joined #lisp 12:17:55 talking about SBCL run via screen 12:19:01 clhs handler-case 12:19:02 http://www.lispworks.com/reference/HyperSpec/Body/m_hand_1.htm 12:20:52 -!- drafael [n=tapio@ip-118-90-133-204.xdsl.xnet.co.nz] has quit ["Leaving."] 12:21:58 puchacz: What do you use as webserver? If its hunchentoot, it should be able to ignore errors. handler-case can allow you to ignore anything (and redirect to a nice error-page) 12:23:02 madnificent: I use ucw as an app framework and mod_lisp with apache 12:23:20 except for return-from :) 12:24:19 handler-case is very local.... I thought you could modify the behaviour of the whole SBCL 12:24:27 puchacz: now add drewc or fade to that comment, and someone will be here soon (drewc probably doesn't have time, but fade resides in #lispweb) 12:28:27 puchacz: you can replace the debugger 12:28:38 uhmm, anyone using apache2+mod_proxy+weblocks here? 12:28:45 yes 12:28:54 S11001001: how to replace it, pls? 12:29:10 puchacz: apropos debug in clhs 12:29:20 S11001001: thx 12:29:21 rstandy: www.hfsbo.com runes on such an arrangement 12:29:43 runs 12:30:07 S11001001: ohhh, you are Stephen Compall :-) 12:30:08 (I am writing a date parser with net.telent.date for it atm :) 12:30:15 S11001001: thanks for your work on weblocks 12:30:47 S11001001: and pleased to meet you 12:31:03 you're welcome 12:31:17 are you on dev or stable? 12:31:49 S11001001: I'm on dev. I just now changed my code to use your :default-store parameter for webapps :-) 12:31:56 S11001001: it's very usful 12:32:04 btw do you have one-store-per-webapp? 12:32:14 S11001001: but there seems to be som problems with the elephant store though 12:32:39 jeremiah [n=jeremiah@31.Red-213-98-123.staticIP.rima-tde.net] has joined #lisp 12:32:41 S11001001: yes, I have a class hierarchy of webapps 12:32:44 dto [n=user@pool-98-118-1-212.bstnma.fios.verizon.net] has joined #lisp 12:32:48 ah 12:32:57 -!- dwave [n=ask@084202073105.customer.alfanett.no] has quit ["Be back later"] 12:33:07 is your problem elephant-specific, though? 12:33:13 S11001001: with various subtrees of the hierarchy using a different store 12:33:28 Because I only deal with CLSQL myself, my one elephant hack was to change it to use dynamic-transactions 12:33:40 nice 12:33:46 S11001001: yes, I think the problem is related to the elephant:*store-controller* dynamic var 12:33:51 dwave [n=ask@84.202.73.105] has joined #lisp 12:34:08 is there a group thread about this? 12:34:33 S11001001: I whould really refactor the code to offer an example webapp 12:35:23 S11001001: I was just starting a thread on weblocks-dev yesterday before reading about your new code with about webapp specific store :-) 12:35:48 S11001001: then I stopped and tried your code 12:35:59 S11001001: but I got the same result with elephant 12:36:11 S11001001: so I'll end up writing the post 12:36:35 so, i'm adding an ALWAYS-BOUND declaration 12:36:41 S11001001: surely Ian Eslick will have something to suggest about that 12:36:45 -!- decaf [n=mehmet@88.232.178.8] has left #lisp 12:36:49 if *store-controller* is like clsql:*default-database*, then you will still need an :around method on handle-client-request 12:36:50 (for specials and globals-to-be) 12:37:34 S11001001: yeah, wrote that and it works when for example using a datagrid to show stored objects 12:38:02 S11001001: but for some reason, it is not working for a dataform when inserting objects in the store 12:38:05 -!- Phoodus [i=foo@ip68-231-38-131.ph.ph.cox.net] has quit [Read error: 54 (Connection reset by peer)] 12:38:25 i'm inclined to make it work only in PROCLAIM/DECLAIM -- not as a local declaration: as a global thing it can be made safe by making it inhibit makunbound and requiring the symbol to be bound when proclaimed as such 12:38:36 S11001001: I should write some example code to show the problem in my post 12:38:39 (sans people deliberately shooting their feet off) 12:39:36 but is prohibiting it in local declarations bad style? it could be made to work there as well 12:39:37 S11001001: anyway, I also have some problems with the apache2+weblocks arrangement, but before wasting your time, I'll check the weblocks-dev archive for answers ;-) 12:39:51 -!- mathrick [n=mathrick@users177.kollegienet.dk] has quit [Remote closed the connection] 12:40:06 rstandy: are you inserting during GET requests? 12:40:07 mathrick [n=mathrick@users177.kollegienet.dk] has joined #lisp 12:40:12 S11001001: btw, www.hfsbo.com is a nice site :-) 12:41:07 thanks 12:41:15 S11001001: uhmm, I didn't wrote the code to insert in the datastore. I let the subimt button code of the dataform do it 12:41:17 my code-fu is better than my design-fu though 12:41:59 S11001001: there is alos the elephant store specific proxy code involved 12:42:42 S11001001: I use something like this: 12:43:00 ah yes, a complicated arena I thankfully never had to deal with thanks to weblocks-clsql :) 12:43:24 rstandy pasted "datafor widget example with proxy class" at http://paste.lisp.org/display/79179 12:43:55 S11001001: yeah, maybe I should go with some other store backend 12:43:56 -!- dwave [n=ask@84.202.73.105] has quit ["Be back later"] 12:44:12 *dataform 12:44:13 S11001001: looks like our Alien Technology is taking over the real-estate web market, eh? www.bansvik.com is also CL. :) 12:44:17 I also highly recommend adding f-underscore to your deps and importing f_% 12:44:22 -!- Jacob_H [n=jacob@92.2.126.235] has quit ["Leaving"] 12:44:33 it makes lots of my hooks prettier 12:44:58 S11001001: oh, thanks for the suggestion, I didn't know about f-underscore 12:46:49 kreuter [n=kreuter@pool-96-252-14-107.bstnma.fios.verizon.net] has joined #lisp 12:48:52 S11001001: uhm my problems with mod_proxy are discussed here: http://groups.google.com/group/weblocks/browse_thread/thread/2af4b55b4e481e35/6d3fb42245291ed2?lnk=gst&q=mod_proxy#6d3fb42245291ed2 12:49:06 S11001001: ups, sorry for the ugly link 12:50:17 S11001001: how do you solved the missing styles problems links with your site? 12:50:41 billstclair [n=billstcl@unaffiliated/billstclair] has joined #lisp 12:52:07 S11001001: my action links are also broken when behind mod_proxy 12:52:32 tritchey [n=tritchey@c-68-58-88-241.hsd1.in.comcast.net] has joined #lisp 12:52:45 rstandy: I put Alias lines for /hfsbo/pub and /pub that point to the the pub dir in my tree 12:52:47 -!- beach`` is now known as beach 12:53:04 Good afternoon. 12:53:21 beach: hi 12:53:41 I use ProxyPassMatch with ^(/hfsbo)? at the beginning to accept both sorts (of course, I can't then have a static-selector match anything beginning with hfsbo, but I don't care about that) 12:54:02 Playing with format: http://paste.lisp.org/display/79180 12:54:25 and what's with your action links? 12:54:48 Neet trick accessing format arguments in random order. 12:54:52 S11001001: uhmm, could you please paste your weblocks related apache configs? I'm a little confused 12:55:22 S11001001: I'm an apache newbie 12:56:52 -!- jeremiah [n=jeremiah@31.Red-213-98-123.staticIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)] 12:56:52 S11001001: IIUC, your setup must be replcated for all webapps that needs custom "pub" resources 12:57:12 rstandy: http://home.nocandysw.com/~sirian/lisp/hfsbo.com.conf 12:58:02 and yes 12:58:15 I'm not sure the pub redirect is necessary 12:58:24 but it's more efficient, so might as well 12:58:27 S11001001: thank you very, very much, I'll thake a look 12:58:42 wraith0x2b [n=wraith@89.136.174.25] has joined #lisp 12:58:55 save a copy, that's the machine on which I type this, and goes away often 12:59:09 This might sound noobish but how do you enable C syntax highlighting in emacs?! 12:59:15 S11001001: I think it's necessary. Here is what happens without: http://www.cmcc.it/apps/hunchentoot/bfm-download 12:59:36 S11001001: no stylesheets, no javascript files :-) 13:00:23 it kind of works but doesn't highlight variables etc, only definitions and function names 13:00:50 if you express the extended absolute path in Weblocks, which I think webapp objects support, the URLs generated would be correct 13:01:17 in my case, I simply aliased and proxied for all possible urls generated, and didn't worry about Weblocks getting it right for me 13:02:28 S11001001: aha! I should look more carefully at the code. I didn't notice the webapp parameter 13:02:46 S11001001: again, many thanks for precious help 13:02:50 it may be enough just to set :public-files-uri-prefix or specialize compute-webapp-public-files-uri-prefix 13:03:00 davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #lisp 13:03:27 the latter could wrap the next-method by adding /apps/hunchentoot to the beginning, and therefore work for all inherited apps 13:03:28 wraith0x2b: you might ask that in #emacs. This channel is more for Common Lisp. 13:03:40 everything that will interest you is in src/application.lisp 13:04:12 S11001001: yeah, that's a good plan 13:04:15 -!- wraith0x2b [n=wraith@89.136.174.25] has left #lisp 13:04:35 S11001001: I may need some modifications for the render-link function also 13:04:56 S11001001: my submit button seems to suffer the url problem 13:05:08 -!- ehu` [n=chatzill@82-170-33-173.ip.telfort.nl] has quit [Read error: 110 (Connection timed out)] 13:05:38 Odin- [n=sbkhh@s121-302.gardur.hi.is] has joined #lisp 13:05:40 -!- chris2_ is now known as chris2 13:05:59 -!- nvoorhies [n=nvoorhie@adsl-75-36-207-191.dsl.pltn13.sbcglobal.net] has quit [Read error: 104 (Connection reset by peer)] 13:07:12 it might be easier just to pass the whole path through to hunchentoot 13:07:34 that way the action and link renderer would see the right path 13:08:25 S11001001: uhm, any pointers on how to do that? 13:09:01 apache config: well, pass the whole path instead of dropping part of it 13:09:36 S11001001: ah ok, so I must read the mod_proxy docs 13:10:36 then pass :prefix to defwebapp/specialize/wrap initialize-instance with a version that computes :prefix and passes to call-next-method 13:12:05 S11001001: uhmm, I feel like the pieces of the puzzle are going to be putted together 13:13:07 S11001001: thank you :-) 13:13:42 no problem 13:14:10 blitz__ [n=blitz@2001:6f8:10f6:200:21b:77ff:fe41:11ab] has joined #lisp 13:14:27 jeremiah [n=jeremiah@31.Red-213-98-123.staticIP.rima-tde.net] has joined #lisp 13:14:35 dwave [n=ask@pat-tdc.opera.com] has joined #lisp 13:16:31 attila_lendvai_: ping 13:26:00 -!- rindolf [n=shlomi@bzq-219-139-216.static.bezeqint.net] has quit ["Leaving"] 13:30:12 _mathrick [n=mathrick@users177.kollegienet.dk] has joined #lisp 13:31:04 -!- mathrick [n=mathrick@users177.kollegienet.dk] has quit [Read error: 104 (Connection reset by peer)] 13:35:54 postamar [n=postamar@x-132-204-253-215.xtpr.umontreal.ca] has joined #lisp 13:36:26 -!- davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 13:37:35 ehu` [n=chatzill@82-170-33-173.ip.telfort.nl] has joined #lisp 13:38:42 -!- joachifm [n=joachim@bjo1-1x-dhcp143.studby.uio.no] has quit ["Leaving"] 13:40:00 -!- roidrage [n=roidrage@dslb-084-058-181-192.pools.arcor-ip.net] has quit [] 13:42:34 davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #lisp 13:42:52 -!- puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has quit [Remote closed the connection] 13:46:08 shmho [n=user@58.142.15.103] has joined #lisp 13:48:06 ejs [n=eugen@149-113-112-92.pool.ukrtel.net] has joined #lisp 13:53:40 -!- Spyderco [n=nash@194.45.110.65] has quit [Read error: 104 (Connection reset by peer)] 13:53:54 -!- Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has quit [Read error: 113 (No route to host)] 13:54:02 Spyderco [n=nash@194.45.110.65] has joined #lisp 13:55:45 is there still any activity on the cl-gardeners list? 13:56:15 kreuter: nope 13:57:29 why, there certainly appears to be. not that it's any kind of gardening activity, though 13:57:57 more fertilizing than planting? :) 13:57:59 hrr4 [n=hrr4@p5081C771.dip0.t-ipconnect.de] has joined #lisp 13:58:27 heh 13:58:54 -!- blitz__ [n=blitz@2001:6f8:10f6:200:21b:77ff:fe41:11ab] has quit ["Leaving"] 13:59:01 lol 14:00:42 rolly1975 [n=rory@5ac959a9.bb.sky.com] has joined #lisp 14:03:44 HG` [n=wells@91.111.13.16] has joined #lisp 14:05:55 -!- jeremiah [n=jeremiah@31.Red-213-98-123.staticIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)] 14:07:07 The-Kenny [n=moritz@p5087FBEB.dip.t-dialin.net] has joined #lisp 14:07:51 -!- ruediger [n=the-rued@62-47-158-122.adsl.highway.telekom.at] has quit ["This computer has gone to sleep"] 14:09:34 ruediger [n=the-rued@62-47-158-122.adsl.highway.telekom.at] has joined #lisp 14:12:27 Nshag [i=user@Mix-Orleans-106-4-84.w193-248.abo.wanadoo.fr] has joined #lisp 14:13:07 -!- vy [n=user@213.139.194.186] has quit [Remote closed the connection] 14:15:39 ManateeLazyCat [n=user@116.4.151.192] has joined #lisp 14:16:47 kreuter: there is a channel #cl-gardeners . Not active at all, but when questions are asked, answers are given, so... 14:18:02 rindolf [n=shlomi@bzq-219-139-216.static.bezeqint.net] has joined #lisp 14:23:17 -!- sepult [n=buggarag@xdsl-87-78-74-72.netcologne.de] has quit [Read error: 60 (Operation timed out)] 14:23:30 sepult [n=buggarag@xdsl-87-78-29-178.netcologne.de] has joined #lisp 14:23:37 4\\\\\\\\\\\\\\\ 14:23:41 jeremiah [n=jeremiah@31.Red-213-98-123.staticIP.rima-tde.net] has joined #lisp 14:24:18 -!- dwave [n=ask@pat-tdc.opera.com] has quit ["Be back later"] 14:26:46 -!- dkcl [n=dkcl@unaffiliated/dkcl] has quit ["leaving"] 14:27:10 this can't be right... 14:27:38 serge_ [n=serge@ppp91-78-124-73.pppoe.mtu-net.ru] has joined #lisp 14:27:46 CLHS doesn't seem to prohibit locally declaring constants as special. Someone please point to me where it says this is not legal! 14:27:53 *headache* 14:27:58 beach` [n=user@ABordeaux-158-1-102-174.w86-201.abo.wanadoo.fr] has joined #lisp 14:28:39 does anybody use emacs 23? installed from cvs it does not start =( 14:28:47 *p_l* raises hand 14:28:54 says: "cannot open load file" 14:29:04 -!- jeremiah [n=jeremiah@31.Red-213-98-123.staticIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)] 14:29:05 does it mention which file? 14:29:09 nyef [n=nyef@24.220.182.229] has joined #lisp 14:29:13 G'morning all. 14:29:17 kinnetica [n=kinnetic@dhcp-226-32.wireless.american.edu] has joined #lisp 14:29:17 serge_: Also, do you have any stale *.elc files? 14:29:20 compiled, but not installed version runs awesome 14:29:35 cyril-util 14:29:36 this is not #emacs 14:29:43 beach`` [n=user@86.201.112.142] has joined #lisp 14:29:45 nyef: Good Afternoon :) 14:30:06 stassats`: sorry, #emacs is all dead for now =( 14:30:28 -!- sepult [n=buggarag@xdsl-87-78-29-178.netcologne.de] has quit ["leaving"] 14:31:00 -!- legumbre` [n=user@r190-135-8-99.dialup.adsl.anteldata.net.uy] has quit [Read error: 110 (Connection timed out)] 14:31:23 sepult [n=buggarag@xdsl-87-78-29-178.netcologne.de] has joined #lisp 14:31:50 p_l: all just from cvs 14:32:28 it's from cvs, it can be broken 14:33:07 -!- beach` [n=user@ABordeaux-158-1-102-174.w86-201.abo.wanadoo.fr] has quit [Read error: 60 (Operation timed out)] 14:33:20 stassats`: oO 14:33:39 stassats`: what should i do? 14:34:21 serge_: this channel is about common lisp, which really doesn't have much to do with emacs :) 14:34:21 maybe check some supposedly more stable tags like EMACS_PRETEST_23_0_92 14:34:23 serge_: write a bug report/try debugging it :) 14:34:31 -!- beach [n=user@ABordeaux-158-1-69-62.w90-60.abo.wanadoo.fr] has quit [Read error: 60 (Operation timed out)] 14:34:41 dkcl [n=dkcl@unaffiliated/dkcl] has joined #lisp 14:35:50 *nikodemus* decides to for now consider declaring constants special a spec bug 14:36:04 unless it really is prohibited somewhere, of course... 14:36:12 > A constant defined by defconstant can be redefined with defconstant. However, the consequences are undefined if an attempt is made to assign a value to the symbol using another operator 14:36:47 and we could say that a local special declaration would be assigning a different value to the symbol 14:37:01 hm, i can work with that 14:37:40 also, since constants cannot be bound the declaration would be a no-op anyways 14:37:53 error it is, then 14:37:55 oh, even better: The consequences are undefined when constant symbols are rebound as either lexical or dynamic variables. 14:38:30 oh, i missed the word dynamic. was scanning for "special"... 14:43:48 -!- serge_ [n=serge@ppp91-78-124-73.pppoe.mtu-net.ru] has quit ["Ex-Chat"] 14:43:58 konr [n=karkeej@201.82.139.214] has joined #lisp 14:45:32 beach [n=user@ABordeaux-158-1-47-62.w90-55.abo.wanadoo.fr] has joined #lisp 14:45:56 -!- dfox [n=dfox@r11jn246.net.upc.cz] has quit [Remote closed the connection] 14:46:19 jeremiah [n=jeremiah@31.Red-213-98-123.staticIP.rima-tde.net] has joined #lisp 14:46:27 dfox [n=dfox@r11jn246.net.upc.cz] has joined #lisp 14:46:37 mattrepl [n=mattrepl@68.100.82.124] has joined #lisp 14:46:50 nikodemus: If you're in the mood for concurrency madness, I'm not completely convinced is race-free yet (: 14:47:36 apo__ [n=apo@pD9E7E8A5.dip.t-dialin.net] has joined #lisp 14:48:23 *nyef* tries a win32 build of 1.0.27.46. 14:50:09 IListenU [n=IListenU@78-57-141-87.static.zebra.lt] has joined #lisp 14:50:35 -!- beach`` [n=user@86.201.112.142] has quit [Read error: 110 (Connection timed out)] 14:50:46 -!- IListenU [n=IListenU@78-57-141-87.static.zebra.lt] has quit [Remote closed the connection] 14:50:58 are there any lisp packages to work with the financial market 14:51:07 *market? 14:51:19 plutonas [n=plutonas@c-83-233-152-13.cust.bredband2.com] has joined #lisp 14:51:20 In what sense of "work with the financial market"? 14:51:46 making money ? 14:52:19 printer + postscript files? :D 14:52:29 haha 14:52:42 (Almost everything I can think of for the term would fall under the category of too specialized to release, and considered a competitive advantage anyway.) 14:53:11 pkhuong: you need to define thread-count-limit or some such so that you know XADD cannot overflow. :P 14:53:27 I want to analyze the stocks' data, build models and get higher profit margins 14:53:30 -!- apo__ [n=apo@pD9E7E8A5.dip.t-dialin.net] has quit ["leaving"] 14:53:33 nikodemus: just in case we have 2G readers? :p 14:53:49 blbrown [n=Berlin@c-71-236-25-127.hsd1.ga.comcast.net] has joined #lisp 14:54:06 seriously speaking, looks interesting 14:54:30 konr: There are apps in CL for that, but you won't get them, unless you have money to pay :) 14:54:57 konr: which leaves you to parsing the reuters site or any other resource you have (that is possible) 14:55:00 i like the odd/even trick. very clever! 14:55:16 p_l: hmmm, which ones? Are they free (as in freedom) to study the code/ 14:55:44 nikodemus: If I only have to worry about x86-64, I'd probably have 2 32-bit counters, one for writers the other for readers. 14:55:59 *nyef* would be surprised if they were. Again, "competitive advantage". 14:56:01 konr: nope, they are proprietary as hell, because that code not falling into other's hand is the competetive advantage. Sorry, Open source doesn't work everywhere :D 14:56:05 -!- apo_ [n=apo@217.231.197.78] has quit [Read error: 110 (Connection timed out)] 14:56:06 -!- blbrown [n=Berlin@c-71-236-25-127.hsd1.ga.comcast.net] has quit [Client Quit] 14:56:44 ... "fortune passes everywhere"? 14:57:23 konr: if you're paying for a direct market data feed, your feeder will give you a spec and likely a java reader. 14:57:44 but the cost of such feeds means that there aren't (m)any free decoders of such data. 14:58:01 i get a warm and fuzzy feeling seeing ATOMIC-INFC getting use :) 14:59:46 Work queues soon. Might even help spectral norm on the shootout. 15:00:02 eh? 15:00:07 konr: to put it short: there isn't any ready-to-go code. You might have access to some parsing code, but the core logic is up to you (and reading papers) 15:00:11 .oO(who the hell is spectral norm?) 15:00:17 hefner: the multithreaded version, 15:00:43 okay if you say so 15:01:12 -!- ruediger [n=the-rued@62-47-158-122.adsl.highway.telekom.at] has quit ["This computer has gone to sleep"] 15:03:52 p_l: sounds really amazing :) 15:05:00 konr: if you like challenges... 15:08:23 the upside is that if you write a good package, you don't even have to sell it to make money -- of course you can still possibly make money selling even a mediocre package 15:08:41 like someone said: "daytraders will pay for anything" :P 15:09:04 Ragnaroek [i=d951c8b4@gateway/web/ajax/mibbit.com/x-787eb7d51fad97d3] has joined #lisp 15:09:31 if you're good at writing trading logic with such data, it's a lucrative career. :) 15:10:15 konr: build it, launch it, wait for lispers to make it better => rich lisp community => more lispers => ??? => profit \o/ (go back a bit for the profit part) 15:11:23 ejs1 [n=eugen@149-113-112-92.pool.ukrtel.net] has joined #lisp 15:11:48 nvoorhies [n=nvoorhie@adsl-75-36-207-191.dsl.pltn13.sbcglobal.net] has joined #lisp 15:12:32 hahaha 15:13:04 My friends are really into it... If I could earn a big percentage, it'd be my dream job <3 15:15:13 -!- ehu` [n=chatzill@82-170-33-173.ip.telfort.nl] has quit [Read error: 110 (Connection timed out)] 15:15:15 konr: if you'd be building something like that, and you create an open platform for ideas/creation. I think you could get a lot of input, from a lot of different parts of the socio-economically organised world) 15:15:44 clhs plusp 15:15:45 http://www.lispworks.com/reference/HyperSpec/Body/f_minusp.htm 15:15:55 abeaumont [n=abeaumon@84.76.48.250] has joined #lisp 15:15:55 konr: the design could be non-trivial, as many people will want to work with different brokers 15:17:34 -!- tsuru [n=user@c-69-245-36-64.hsd1.tn.comcast.net] has quit [Remote closed the connection] 15:18:26 -!- antgreen [n=Anthony@CPE0014bf0b631a-CM001e6b1858fa.cpe.net.cable.rogers.com] has left #lisp 15:18:30 madnificent: hmmm, interesting... Wouldn't people have to know quite a bit of programming to use it? I think the main contributors would be mathematicians, economists etc that could build additional models for analyzing the data 15:19:07 i'd be surprised if anybody contributed trading logic. that stuff is closely guarded as trade secret. 15:19:34 dwave [n=ask@212251218147.customer.cdi.no] has joined #lisp 15:20:20 konr: they would probably write a spec about an algorithm they want in there, then the coders would ensure that what is needed to make the algorithm work is available. It may even be that the mathematitians write it in psuedo-code (as lisp isn't *that* popular) 15:20:35 Now, what would really be impressive is a system to automatically create usable trading logic... 15:21:20 Fade: I know a couple close friends that are interested in those topics. Even though it is at an amateur level, many of them define their own simple rules and want to work with them (especially during these times). 15:21:34 a system for doing genetic culling of dynamic agents against historical data would be very interesting. 15:21:37 -!- dwave [n=ask@212251218147.customer.cdi.no] has quit [Client Quit] 15:21:47 for a lot of uses, it isn't really needed to have the best system in the world, as long as it can be configured by those that want to use it on a per-stock basis 15:25:07 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 15:26:09 beach` [n=user@ABordeaux-158-1-130-186.w90-60.abo.wanadoo.fr] has joined #lisp 15:26:29 the last time I looked at this, the feeds cost between 10,000 and 15,000 per month with 12 month contracts, so the barrier to entry is very high. 15:26:41 $USD 15:27:08 -!- beach [n=user@ABordeaux-158-1-47-62.w90-55.abo.wanadoo.fr] has quit [Read error: 60 (Operation timed out)] 15:27:23 puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has joined #lisp 15:30:07 kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 15:30:14 Fade: ? 15:30:27 hrmn? 15:30:30 Fade: regular trading is quite cheap 15:30:41 sure, but real time market data is very very expensive. 15:30:45 Fade: I think its about 15euros per transaction here 15:31:13 yes, however I think that for the hobby market, you'd be fine with hourly data (or even slower) 15:31:17 -!- sledge [n=chris@pdpc/supporter/student/sledge] has quit ["Ex-Chat"] 15:31:44 depends on the instruments you're trading. 15:32:08 blue chips? sure. but the slower the instrument the less valuable autonomous agents become. 15:32:47 their use is largely confined to fx and commodity exchanges. 15:32:52 I think it wouldn't be wise to focus on the big players to join the development of a system like that, certainly not at a stage that shouts development 15:33:05 where unit gains are often no larger than .001 cent. 15:33:16 and volumes are enormous. 15:33:39 -!- rread [n=rread@nat/sun/x-1a45eec256f2cade] has quit [Read error: 110 (Connection timed out)] 15:33:49 and humans still dominate strategic play. 15:33:59 rdd` [n=rdd@c83-250-153-45.bredband.comhem.se] has joined #lisp 15:34:17 rhclopes [n=user@bodega.brunel.ac.uk] has joined #lisp 15:36:07 -!- ThomasI [n=thomas@unaffiliated/thomasi] has quit ["Bye Bye!"] 15:36:24 -!- rhclopes [n=user@bodega.brunel.ac.uk] has quit [Remote closed the connection] 15:38:08 -!- durka42 [n=durka@d81.wireless.swarthmore.edu] has quit [] 15:39:18 -!- rdd [n=user@c83-250-153-45.bredband.comhem.se] has quit [Connection timed out] 15:40:20 -!- rdd` is now known as rdd 15:41:00 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 15:41:16 -!- beach` is now known as beach 15:42:01 -!- HG` [n=wells@91.111.13.16] has quit [Client Quit] 15:42:40 -!- athos [n=philipp@92.250.250.68] has quit ["leaving"] 15:44:35 -!- rolly1975 [n=rory@5ac959a9.bb.sky.com] has quit ["The computer fell asleep"] 15:44:59 nurv101 [n=nurv101@wifi.ist.utl.pt] has joined #lisp 15:45:41 wedgeV [n=wedgeV@80.122.51.198] has joined #lisp 15:47:19 I haven't heard of any parts of a realtime trading system written in Lisp. 15:47:36 hmmm... I looked over some graphs generated from clbuild deps... it looks like getting Babel and trivial-features as something prepackaged could be an advantage :D 15:47:53 usually 100% java, or java + analytics library in C++ 15:48:06 puchacz: have a look at http://www.ravenpack.com/ 15:48:36 puchacz: Isn't that was Espen Vestre is doing? 15:48:37 puchacz: I heard of some written in OCaml, F# and other, but I[m not sure about realtime trading... financial analysis, sure 15:48:49 puchacz: netfonds? ie primetrader? 15:49:53 Looks like changing (when (plusp child) ...) to (unless (minusp child) ...) in run-program does fix the win32 build. 15:49:57 lisp is well suited to the domain. 15:50:03 I meant something used at investment banks. 15:50:12 nyef: minusp is not (not plusp) 15:50:22 Fufie [n=innocent@86.80-203-225.nextgentel.com] has joined #lisp 15:50:24 the commercial off the shelf packages do all seem to be built in java. 15:50:34 pjb: Yes, I know. Had to look it up first, though. 15:50:43 errr.... what about garbage collection? even if you use Java, you have to avoid major gc. 15:51:17 realtime is not "fast". 15:51:47 puchacz: commercial, enterprise and java. I guess the solution to GC is "here is our offer of prepackaged powerful servers" 15:51:49 "Payroll is a realtime system. If you haven't finished your payroll calculations before the next pay cycle, you're screwed." 15:51:53 realtime wrt financial markets isn't synonymous with the kind of hard realtime associated with operating systems for applications like missile guidance. 15:52:29 not the same as "hard realtime", but still you don't want pauses. can sbcl go as predictably smooth as java? i.e. no major gc during the day. 15:52:41 -!- davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 15:52:44 and you want reliable threading 15:53:06 well, the system would have to be designed to minimise consing, and you'd have a ridiculous amount of ram in such a system. 15:53:07 puchacz: why would you limit yourself to sbcl for such an important application? 15:53:22 which one then, if not sbcl? allegro? 15:53:40 durka42 [n=durka@d81.wireless.swarthmore.edu] has joined #lisp 15:53:49 puchacz: search for the article about implementation of ITA Soft.'s system 15:53:58 ITA is not realtime 15:54:18 puchacz: They did however have info about avoiding GC 15:54:19 it is as realtime as your trading system needs to be 15:54:22 LiamH [n=nobody@pool-68-239-77-4.res.east.verizon.net] has joined #lisp 15:54:26 if you want to live without gc pauses you can 15:54:30 puchacz: well, allegro? lispworks? whatever floats your boat. 15:54:35 you just need to think about consing 15:54:44 lispworks also offers a realtime CL 15:54:52 i like the way i say "just" :) 15:54:56 but it's on individual, case-by-case way :) 15:55:19 p_l: realtime? lispworks? huh? 15:55:45 LostMonarch [n=roby@host205-182-dynamic.57-82-r.retail.telecomitalia.it] has joined #lisp 15:56:03 what is the behaviour of sbcl at the outter edge of load? how pronounced is GC? 15:56:19 nikodemus: there's a separate implementation, not available by default (there's some info about it on webpage) 15:56:33 lispworks does not have real threads 15:56:53 the generic lispworks, I mean 15:57:29 depends on what "real threads" mean. as i understood it, they have native threads, but the scheduler switch between all available threads. 15:57:36 puchacz: many realtime apps don't use native threads 15:57:41 p_l: you mean Liquid Common Lisp? afaik it's not realtime any more than lispworks 15:57:57 and it's not like you /have/ to use threads either for that matter. what is the real issue here? 15:58:09 nikodemus: not liquid. There's a mention "we have an in-house, not available for normal sale CL implementation with realtime GC" 15:58:13 'realtime' is a mercurial word. 15:58:23 "for some value of 'realtime'" 15:58:23 ah 15:58:33 lispworks has native threads these days 15:58:56 nikodemus: it was apparently developed for some contracts and is kept current, but you have to talk to them if you want it - it's completely outside normal licensing 15:58:57 is there high level information about sbcl's GC? 15:59:09 Jacob_H [n=jacob@92.2.126.235] has joined #lisp 15:59:17 or documentation 15:59:22 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 16:00:02 beach` [n=user@ABordeaux-158-1-124-161.w90-60.abo.wanadoo.fr] has joined #lisp 16:00:11 what sort of answer are you looking for? 16:00:12 generational copying, conservative on stack and registers, uses a wp barrier 16:00:21 vande [n=sdfpme@116.4.163.85] has joined #lisp 16:00:22 that's the executive summary :) 16:00:34 -!- abeaumont [n=abeaumon@84.76.48.250] has quit [Remote closed the connection] 16:00:38 pkhuong: pretty much what nikodemus said 16:00:42 realtime is pretty well defined in trading / market making application. constant throughput within the range of 100 ms at hundreds of events per second, reliable threading, major gc considered catastrophic etc. 16:00:57 nikodemus: thanks 16:01:11 Fade: "how pronounced is gc" depends entirely on your application 16:01:41 writing non-consing code is not that hard -- so you don't need to gc at all 16:01:58 so, with this definition of realtime as above, can any current lisp implementation deal with it? 16:02:30 Hello everyone I want to learn lisp, can recommend a good book for me? 16:02:40 i'd say that with careful code, most modern lisp implementations would be fine for the kind of realtime required for financial software. 16:02:43 minion: tell vande about that dead-sexy-book 16:02:45 vande: i don't know what you're referring to 16:02:48 if you write your application "smartly", all native threading implementations can deal with it 16:02:51 minion: pcl 16:02:51 damn 16:02:51 pcl: pcl-book: "Practical Common Lisp", an introduction to Common Lisp by Peter Seibel, available at http://www.gigamonkeys.com/book/ and in dead-tree form from Apress (as of 11 April 2005). 16:03:17 vande: also: "Paradigms of Artificial Intelligence Programming". contrary to the title, excellent Lisp textbook. 16:03:43 vande: PCL assumes you have already programmed in other languages 16:04:17 p_l: i use c lua 16:04:34 if you've become curious about lisp without any prior programming experience, you qualify as a beautiful and unique snowflake. 16:04:54 -!- beach [n=user@ABordeaux-158-1-130-186.w90-60.abo.wanadoo.fr] has quit [Read error: 60 (Operation timed out)] 16:08:06 HG` [n=wells@91.111.13.16] has joined #lisp 16:09:27 -!- Tordek [n=tordek@host87.190-137-248.telecom.net.ar] has quit [Read error: 113 (No route to host)] 16:09:43 -!- Ragnaroek [i=d951c8b4@gateway/web/ajax/mibbit.com/x-787eb7d51fad97d3] has quit ["http://www.mibbit.com ajax IRC Client"] 16:12:12 Hmm, I'm getting lots of these messages with SLIME/CCL: "; Compiler warning: In a toplevel form: Nonspecific warning". Anyone else get that? 16:13:11 -!- kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 16:15:12 rolly1975 [n=rory@5ac959a9.bb.sky.com] has joined #lisp 16:18:02 no, latest slime and ccl here 16:18:55 -!- luis` is now known as luis 16:19:03 blbrown [n=Berlin@c-71-236-25-127.hsd1.ga.comcast.net] has joined #lisp 16:19:25 Hmm, seems to be some issue with Stefil. 16:19:27 rhcltest [n=rhcltest@bodega.brunel.ac.uk] has joined #lisp 16:20:20 Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has joined #lisp 16:20:34 -!- rhcltest [n=rhcltest@bodega.brunel.ac.uk] has quit [Remote closed the connection] 16:21:30 -!- hrr4 [n=hrr4@p5081C771.dip0.t-ipconnect.de] has quit ["Lost terminal"] 16:21:48 mvilleneuve [n=mvillene@che33-1-82-66-18-171.fbx.proxad.net] has joined #lisp 16:23:00 good evening 16:23:05 hrr4 [n=hrr4@p5081C771.dip0.t-ipconnect.de] has joined #lisp 16:23:40 good evening mvilleneuve 16:24:58 the cl implementation benchmarks are quite interesting 16:25:16 guaqua: have you read the book? 16:25:30 benchmark book? 16:26:08 guaqua: hold on, you're talking about cl-bench right? 16:26:22 no idea, this page: http://www.cliki.net/Performance%20Benchmarks 16:26:50 bgs100 [n=ian@unaffiliated/bgs100] has joined #lisp 16:27:17 How would you convert a string to a number? 16:27:28 clhs parse-integer 16:27:29 http://www.lispworks.com/reference/HyperSpec/Body/f_parse_.htm 16:27:38 minion: parse-number? 16:27:38 parse-number: parse-number is a Library of functions which accept an arbitrary string and attempt to parse it into one of the standard Common Lisp number types, if possible, or else it signals an error of type invalid-number. http://www.cliki.net/parse-number 16:28:04 ikki [n=ikki@189.228.217.125] has joined #lisp 16:28:21 guaqua: many of those benchmarks are described here: www.dreamsongs.com/Files/Timrep.pdf 16:28:26 (read-from-string "#c(123 4/5)") 16:28:41 wow, SBCL 0.8.13. that's a piece of history :) 16:29:12 Sussex Poplog Common Lisp 2.0 sounds interesting :) 16:29:24 dv_ [n=dv@85-127-117-181.dynamic.xdsl-line.inode.at] has joined #lisp 16:29:31 How can I refer to the whitespace character? #\whitespace? 16:29:40 #\space 16:29:41 davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #lisp 16:30:01 #\Space, #\Newline, #\Tab 16:30:17 thanks! 16:30:34 guaqua: http://sbcl.boinkor.net/bench/ has pretty graphs for SBCL, CMUCL, and Clisp 16:31:59 -!- nurv101 [n=nurv101@wifi.ist.utl.pt] has quit [Remote closed the connection] 16:32:19 Thanks for the str to num help 16:33:21 luis` [n=user@r42.eu] has joined #lisp 16:33:58 -!- luis [n=user@r42.eu] has quit [Read error: 104 (Connection reset by peer)] 16:35:02 nikodemus: getting other implementations to that graph would be nice :/ 16:36:38 i'm guessing that's for benchmarking sbcl version by version 16:37:41 it would be nice to have ECL included in those graphs 16:38:02 antifuchs: do you have something to do with http://sbcl.boinkor.net/bench/ ? if so, would it be possible to use different colors from red and green? 16:38:14 ECL 9.04 promises some speedups, I heard 16:39:31 p_l: I'm just curious as to where it would be good at, and where it would probably be failing 16:39:34 When I get back my windows environment, I'll have to test why it claims not to work with unicode/threads on win32 :) 16:40:26 *madnificent* never really tried to run it, just interested 16:40:38 it would be cool if clbuild would allow ecl to be selected :) 16:40:41 beach`` [n=user@ABordeaux-158-1-122-235.w90-60.abo.wanadoo.fr] has joined #lisp 16:40:53 -!- beach`` is now known as beach 16:41:31 -!- beach` [n=user@ABordeaux-158-1-124-161.w90-60.abo.wanadoo.fr] has quit [Read error: 60 (Operation timed out)] 16:43:09 hmm.. for some reason, I can't get :unicode in ecl 9.4.1 16:43:19 -!- davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 16:43:39 my bad, just noticed it 16:45:23 beach` [n=user@ABordeaux-158-1-118-131.w90-60.abo.wanadoo.fr] has joined #lisp 16:45:43 madnificent: that should be a simple matter of someone doing the work -- possibly even the person who thinks it'd be cool 16:45:50 -!- beach [n=user@ABordeaux-158-1-122-235.w90-60.abo.wanadoo.fr] has quit [Nick collision from services.] 16:45:54 -!- beach` is now known as beach 16:45:55 guaqua: antifuchs is the guy to talk to (and offer benchmarking resources, etc) 16:46:02 I need to test ECL more, especially wrt. to CommonQT and Win32 :) 16:47:26 -!- vande [n=sdfpme@116.4.163.85] has quit ["leaving"] 16:52:22 Krystof: my clbuild experience is virtualy inexistant and I've never ran ecl. Besides that, I do think it's cool. 16:52:41 -!- ejs1 [n=eugen@149-113-112-92.pool.ukrtel.net] has quit ["This computer has gone to sleep"] 16:53:03 jbjohns [n=chatzill@52-45.3-213.fix.bluewin.ch] has joined #lisp 16:56:13 roidrage [n=roidrage@dslb-084-058-187-100.pools.arcor-ip.net] has joined #lisp 16:56:33 -!- Odin- [n=sbkhh@s121-302.gardur.hi.is] has quit [] 16:58:22 guaqua: are you also colourblind? 16:58:35 Fade: quite :) 16:58:43 me too. i'm a monochromat. :) 16:59:20 i've basically given up looking at complex graphs. 16:59:21 i can tell the lines from each other, but taking 5 seconds for deducing at 5 inch distance...is not very gratifying :) 16:59:34 -!- Spyderco [n=nash@194.45.110.65] has quit [] 17:00:02 Spyderco [n=nash@194.45.110.65] has joined #lisp 17:00:18 guaqua: perhaps you could introduce a layout that allows you to show the graph of the subject you're hovering over with your mouse? 17:00:32 dysinger [n=tim@cpe-75-80-200-182.hawaii.res.rr.com] has joined #lisp 17:00:58 -!- ffx` [n=ffx@60-241-74-240.static.tpgi.com.au] has quit [Read error: 104 (Connection reset by peer)] 17:00:59 madnificent: those would be nice 17:01:53 it would actually be a good standard to have on the web: graphs are given in a way that colorblind people can understand them too 17:02:22 that would be nice, but I've basically given up. 17:02:52 except where I can vote with my wallet. I try to buy gear that has only blue/white LEDs 17:03:38 you need to ensure that you are heard. many webdevs don't care because you never hear from those that have limitations in what they can see 17:04:01 I used to write emails and letters, but it never actually accomplished anything. 17:04:11 now I just try to acquire the numbers instead of the graphs. 17:05:55 -!- postamar [n=postamar@x-132-204-253-215.xtpr.umontreal.ca] has left #lisp 17:06:41 deat [n=deat@fac34-8-88-172-174-215.fbx.proxad.net] has joined #lisp 17:06:49 how long would it take to build ecl (a very rough idea) 17:07:30 ffx` [n=ffx@60-241-74-240.static.tpgi.com.au] has joined #lisp 17:07:55 nvm, it just finished :) 17:08:40 legumbre [n=user@r190-135-33-26.dialup.adsl.anteldata.net.uy] has joined #lisp 17:09:24 blitz_ [n=blitz@2001:6f8:10f6:200:21b:77ff:fe41:11ab] has joined #lisp 17:10:00 akcom [i=akcom@rrcs-67-79-132-132.se.biz.rr.com] has joined #lisp 17:14:01 hi, how can I entirely disable GPG checks with sbcl/asdf-install? 17:14:08 disable/skip* 17:15:39 by using clbuild :) 17:16:12 (that answer is not entirely frivolous) 17:16:44 hah k 17:17:01 xuanwu [i=xuanwu@140-182-146-197.dhcp-bl.indiana.edu] has joined #lisp 17:17:18 asdf-install:*verify-gpg-signatures* 17:18:39 excuse me, when I load file under clisp it gives "undefined function defsystem" error. Is this a version problem? 17:20:09 Odin- [n=sbkhh@s121-302.gardur.hi.is] has joined #lisp 17:20:43 xuanwu: hmm. did you load the defsystem package first? 17:21:18 files with defsystems in them are not normally ment to be manually loaded 17:21:40 oh I don't know about defsystem package yet 17:21:54 you're ment to use whatever command the system definition facility in question provides to load them 17:21:56 xuanwu: That message indicates that the file you are loading is using some other facility (in this case `defsystem') that you haven't loaded yet. 17:23:19 ok, accessing globals is almost 50% faster than specials -- on a threaded build. so this was not wasted time in my books 17:24:06 Yet I just followed the instructions in readme of the package... anyway I'll check out the defsystem. Thanks. 17:24:25 loading asdf files works well enough, if you don't feel like fussing with symlinks and such 17:24:29 xuanwu: what is the system you are trying to load? 17:24:36 what's the package in question? 17:25:44 this clisp version of ConceptNet: http://web.media.mit.edu/~hugo/conceptnet/omcsnetcl-v0.9.tar.gz 17:27:05 danlei: thanks 17:27:10 xuanwu: are you aware that `clisp' is an implementation of Common Lisp, and that Common Lisp is abbreviated CL? 17:27:24 yes. 17:27:49 xuanwu: So this is software is clisp-specific? 17:27:57 dreish [n=dreish@minus.dreish.org] has joined #lisp 17:28:05 davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #lisp 17:28:15 what makes you think this is the clisp version? the readme seems to indicate it is allegro only 17:28:42 and i'm pretty sure the defsystem there is for allegro's defsystem facility 17:29:06 oh that's good to know. Thanks! 17:29:17 ehu` [n=chatzill@82-170-33-173.ip.telfort.nl] has joined #lisp 17:29:18 well, make that "i'm guessing" instead of "i'm pretty sure" -- but still 17:29:37 readme says: For a simple test run in Allegro try the following: 17:29:38 (load "defsys") [in omcsd directory] 17:30:09 hrr4_ [n=hrr4@p5081A6B4.dip0.t-ipconnect.de] has joined #lisp 17:31:01 sorry I thought the implementations won't diff much... 17:31:22 xuanwu: If you thought that, why did you make a reference to a particular one? 17:32:18 that software seems to use at least some allegro specific packages: EXCL and NET.ASERVE 17:32:54 and DBI.MYSQL 17:33:43 -!- oudeis [n=oudeis@bzq-82-81-239-4.cablep.bezeqint.net] has quit ["This computer has gone to sleep"] 17:34:08 to get it up and running, especially if you don't have experience with cl, your best bet is to grep the allegro trial edition 17:34:30 yes. I'm doing that. 17:34:58 thanks for the guidance. 17:38:33 *beach* still wonders why xuanwu thought it was a clisp version. 17:39:18 wedgeV_ [n=wedge@80.122.51.198] has joined #lisp 17:40:31 perhaps he took clisp as a synonym for "common lisp", like everyone else who hasn't already been corrected? 17:40:43 yeah, i was just about to say that. 17:40:58 hefner: which is why I asked that question, and he said that he knew the difference. 17:41:39 (assuming xuanwu is a `he') 17:41:40 so you did! double-dumbass on me. 17:41:46 -!- davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 17:42:13 no I knew the difference, but overloaded the symbol "clisp". sorry about that. 17:43:33 dinnertime, methinks 17:44:51 -!- bgs100 [n=ian@unaffiliated/bgs100] has quit ["DON'T PANIC"] 17:44:55 -!- slyrus [n=slyrus@adsl-68-121-172-169.dsl.pltn13.pacbell.net] has quit [Read error: 60 (Operation timed out)] 17:45:44 davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #lisp 17:46:00 xan [n=xan@cs78225040.pp.htv.fi] has joined #lisp 17:47:07 slyrus [n=slyrus@adsl-68-121-172-169.dsl.pltn13.pacbell.net] has joined #lisp 17:48:01 mega1 [n=mega@3e44ad1e.adsl.enternet.hu] has joined #lisp 17:48:28 JAS415 [n=jon@c-24-34-16-25.hsd1.ma.comcast.net] has joined #lisp 17:48:54 clhs logical-pathname-translations 17:48:55 http://www.lispworks.com/reference/HyperSpec/Body/f_logica.htm 17:50:48 oudeis [n=oudeis@bzq-84-109-137-43.red.bezeqint.net] has joined #lisp 17:51:11 Does anyone know if format calls formatter if the format string is known at run-time in SBCL? 17:51:40 nurv101 [n=nurv101@bl14-80-121.dsl.telepac.pt] has joined #lisp 17:53:24 -!- rindolf [n=shlomi@bzq-219-139-216.static.bezeqint.net] has quit ["Leaving"] 17:54:58 hi, what's Lispspeak for ANY or ALL functions taking unspecified number of arguments, and maybe a predicate? 17:55:06 -!- hrr4 [n=hrr4@p5081C771.dip0.t-ipconnect.de] has quit [Read error: 110 (Connection timed out)] 17:55:41 err 17:55:52 tsuru [n=user@c-69-245-36-64.hsd1.tn.comcast.net] has joined #lisp 17:55:56 -!- hrr4_ [n=hrr4@p5081A6B4.dip0.t-ipconnect.de] has left #lisp 17:56:04 &key (predicate #'default-fn) &rest rest 17:56:16 i believe if im getting the question right 17:57:05 JAS415: I was after (or ...) and (and ....) but as functions, not macros, so OK to evaluate all arguments. 17:57:18 right, I can go with remove-if and remove-if-not 17:57:52 so 17:57:59 every/any and #'identity 17:58:23 ah i see, i answered wrong question :-) 17:58:41 pkhuong: that's it, yes :-) 17:59:17 JAS415: for answering such questions you need telepathic abilities 17:59:47 errr, is there ANY function? 17:59:53 clhs any 17:59:54 Sorry, I couldn't find anything for any. 18:00:16 clhs some 18:00:16 http://www.lispworks.com/reference/HyperSpec/Body/f_everyc.htm 18:00:17 stassats`: or knowing what language is being referred to for ANY and ALL. 18:00:51 beach: English :-) 18:01:06 I did not have any specific programming language in mind 18:01:16 anyway, thanks, guys 18:01:32 puchacz: then it would have been better to specify what those functions are supposed to do. 18:01:40 -!- xan-afk [n=xan@cs78225040.pp.htv.fi] has quit [Read error: 110 (Connection timed out)] 18:01:49 "Lispspeak" confused me completely 18:01:57 try 'some' 18:02:01 yup 18:02:07 dwave [n=ask@062016246159.customer.alfanett.no] has joined #lisp 18:03:37 -!- dwave [n=ask@062016246159.customer.alfanett.no] has quit [Client Quit] 18:04:14 dlowe [n=dlowe@c-66-31-201-117.hsd1.ma.comcast.net] has joined #lisp 18:08:06 slash_ [n=Unknown@whgeh0138.cip.uni-regensburg.de] has joined #lisp 18:12:43 -!- roidrage [n=roidrage@dslb-084-058-187-100.pools.arcor-ip.net] has quit [] 18:14:04 -!- jeremiah [n=jeremiah@31.Red-213-98-123.staticIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)] 18:15:21 -!- Spyderco [n=nash@194.45.110.65] has quit [Read error: 104 (Connection reset by peer)] 18:15:26 Spyderco2 [n=nash@194.45.110.65] has joined #lisp 18:16:46 -!- dkcl [n=dkcl@unaffiliated/dkcl] has quit ["leaving"] 18:16:59 dkcl [n=dkcl@unaffiliated/dkcl] has joined #lisp 18:19:00 -!- ejs [n=eugen@149-113-112-92.pool.ukrtel.net] has quit [Read error: 104 (Connection reset by peer)] 18:19:16 -!- _mathrick is now known as mathrick 18:19:35 -!- nvoorhies [n=nvoorhie@adsl-75-36-207-191.dsl.pltn13.sbcglobal.net] has quit [] 18:20:53 nvoorhies [n=nvoorhie@adsl-75-36-207-191.dsl.pltn13.sbcglobal.net] has joined #lisp 18:21:16 -!- billstclair [n=billstcl@unaffiliated/billstclair] has quit [Read error: 110 (Connection timed out)] 18:25:01 rap424 [n=ryan@c-67-182-172-128.hsd1.ca.comcast.net] has joined #lisp 18:27:34 is there a library to do nice charts and graphs? 18:28:18 beach` [n=user@ABordeaux-158-1-43-249.w90-55.abo.wanadoo.fr] has joined #lisp 18:28:28 -!- wedgeV [n=wedgeV@80.122.51.198] has quit [] 18:28:36 -!- beach [n=user@ABordeaux-158-1-118-131.w90-60.abo.wanadoo.fr] has quit [Nick collision from services.] 18:28:40 -!- beach` is now known as beach 18:28:57 -!- wedgeV_ is now known as wedgeV 18:29:04 Krystof: ecl seems to be included in clbuild's implementations, yet it fails on me whilst trying to use it ( Filesystem error with pathname #P"/dev/null" ) 18:30:34 Pandala [n=Pandala@LSt-Amand-152-33-12-243.w82-127.abo.wanadoo.fr] has joined #lisp 18:31:02 konr: I once built something on top of vecto (but it woludn't be worth taking that), and there is some information on gnuplot on cliki.net 18:31:24 jeremiah [n=jeremiah@31.Red-213-98-123.staticIP.rima-tde.net] has joined #lisp 18:32:02 -!- mrsolo [n=mrsolo@adsl-68-126-205-8.dsl.pltn13.pacbell.net] has quit ["This computer has gone to sleep"] 18:32:17 -!- nurv101 [n=nurv101@bl14-80-121.dsl.telepac.pt] has quit [Remote closed the connection] 18:36:23 -!- jeremiah [n=jeremiah@31.Red-213-98-123.staticIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)] 18:37:44 -!- davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has quit [Read error: 113 (No route to host)] 18:39:23 ejs [n=eugen@92-161-124-91.pool.ukrtel.net] has joined #lisp 18:42:21 Modius_ [n=Modius@adsl-68-92-148-234.dsl.austtx.swbell.net] has joined #lisp 18:45:09 So, is the only known real badness on current SBCL HEAD the contrib build failures from run-program on win32? 18:45:36 -!- rap424 [n=ryan@c-67-182-172-128.hsd1.ca.comcast.net] has quit [] 18:45:52 -!- Pandala [n=Pandala@LSt-Amand-152-33-12-243.w82-127.abo.wanadoo.fr] has quit ["Leaving"] 18:46:33 -!- dlowe [n=dlowe@c-66-31-201-117.hsd1.ma.comcast.net] has quit ["Leaving."] 18:46:51 Hun` [n=Hun@pd956be5d.dip0.t-ipconnect.de] has joined #lisp 18:47:47 Modius__ [n=Modius@adsl-68-92-148-234.dsl.austtx.swbell.net] has joined #lisp 18:49:11 -!- ejs [n=eugen@92-161-124-91.pool.ukrtel.net] has left #lisp 18:49:24 ejs [n=eugen@92-161-124-91.pool.ukrtel.net] has joined #lisp 18:54:00 jeremiah [n=jeremiah@31.Red-213-98-123.staticIP.rima-tde.net] has joined #lisp 18:54:40 -!- Modius_ [n=Modius@adsl-68-92-148-234.dsl.austtx.swbell.net] has quit [Read error: 60 (Operation timed out)] 18:56:11 ia [n=ia@89.169.189.230] has joined #lisp 18:57:07 Phoodus [i=foo@68.231.38.131] has joined #lisp 18:58:57 Pandala [n=Pandala@LSt-Amand-152-33-12-243.w82-127.abo.wanadoo.fr] has joined #lisp 18:59:14 ehu`_ [n=chatzill@82-170-33-173.ip.telfort.nl] has joined #lisp 18:59:27 -!- Modius [n=Modius@adsl-68-92-148-234.dsl.austtx.swbell.net] has quit [Read error: 110 (Connection timed out)] 19:01:35 postamar [n=postamar@x-132-204-253-215.xtpr.umontreal.ca] has joined #lisp 19:01:50 vasa [n=vasa@80.94.234.105] has joined #lisp 19:04:20 starchild [n=ali@213.207.221.73] has joined #lisp 19:04:58 -!- Pandala [n=Pandala@LSt-Amand-152-33-12-243.w82-127.abo.wanadoo.fr] has quit ["This computer has gone to sleep"] 19:06:09 Pandala [n=Pandala@LSt-Amand-152-33-12-243.w82-127.abo.wanadoo.fr] has joined #lisp 19:06:15 athos [n=philipp@92.250.250.68] has joined #lisp 19:06:26 -!- Modius__ is now known as Modius 19:09:48 ehu`__ [n=chatzill@82-170-33-173.ip.telfort.nl] has joined #lisp 19:13:02 I'm trying to use gnoplot on sbcl via the cgn package, yet I'm getting "Couldn't write to : Broken pipe" error. Any idea of what could be happening? 19:13:36 rread [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has joined #lisp 19:14:32 mrsolo [n=mrsolo@adsl-68-126-205-8.dsl.pltn13.pacbell.net] has joined #lisp 19:15:24 -!- oudeis [n=oudeis@bzq-84-109-137-43.red.bezeqint.net] has quit ["This computer has gone to sleep"] 19:15:30 -!- starchild [n=ali@213.207.221.73] has left #lisp 19:17:21 -!- ehu` [n=chatzill@82-170-33-173.ip.telfort.nl] has quit [Read error: 110 (Connection timed out)] 19:18:29 dlowe [n=dlowe@c-24-91-154-83.hsd1.ma.comcast.net] has joined #lisp 19:19:43 I believe there are "Git" users here. It is often the case in git code where a change alters the indent level of a whole block of code, which at the line level shows up as subtraction and read of (potentially) a whole function's "guts". This would be less of a factor in C code where indentation isn't so much in play. I was wondering if anybody was using any good toys to mitigate this in regular workflow, or if it is just somethi 19:19:43 ng to live with 19:20:00 konr: The other day, I just used a CLIM window created from the listener to plot lots of stuff. I found it very nice to have all of CL (and all of CLIM) at my disposal for plotting. 19:20:17 -!- durka42 [n=durka@d81.wireless.swarthmore.edu] has quit [] 19:20:40 *hefner* writes things to a text file and plots them in octave 19:20:44 Modius: i don't see how that's any different to any vc tool 19:21:39 nikodemus: I didn't mean to imply that it was, I just wanted to know if anybody was using any handy tool, potentially git-specific, to work around it - even if it's just a lisp-friendly diff tool. 19:21:47 i at least just live with it 19:21:53 -!- LostMonarch [n=roby@host205-182-dynamic.57-82-r.retail.telecomitalia.it] has quit ["raise RuntimeError"] 19:22:03 oudeis [n=oudeis@bzq-84-109-137-43.red.bezeqint.net] has joined #lisp 19:22:43 -!- basic` [n=basic@osuosl/staff/basic] has quit [Read error: 104 (Connection reset by peer)] 19:22:52 hefner: you might be underestimating the value of the tools you created for us all. 19:23:18 maybe. if I load up mcclim, I'll get sidetracked fixing bugs. 19:23:23 ok, let's see if this worked. /me tries to align vector data on 16 byte boundaries 19:23:54 -!- oudeis [n=oudeis@bzq-84-109-137-43.red.bezeqint.net] has quit [Client Quit] 19:24:02 SSE for x86? 19:24:39 rread_ [n=rread@nat/sun/x-8c721a9981300896] has joined #lisp 19:24:43 yeah. nothing very far reaching, though. i just want to see if i can hack up a simd matrix multiply for my raytracer 19:24:55 ok, at least it builds 19:25:40 now, to get the gc to preserve the alignment... 19:26:39 -!- ehu`__ [n=chatzill@82-170-33-173.ip.telfort.nl] has quit ["ChatZilla 0.9.84 [Firefox 3.0.8/2009032609]"] 19:26:55 athos_ [n=philipp@92.250.250.68] has joined #lisp 19:27:19 -!- rstandy [n=rastandy@net-93-144-153-177.t2.dsl.vodafone.it] has quit [Read error: 104 (Connection reset by peer)] 19:27:27 -!- athos [n=philipp@92.250.250.68] has quit [Read error: 104 (Connection reset by peer)] 19:27:32 *beach* would very much like for hefner to be sidetracked. 19:27:40 -!- mrsolo [n=mrsolo@adsl-68-126-205-8.dsl.pltn13.pacbell.net] has quit ["This computer has gone to sleep"] 19:27:51 Tordek [n=tordek@host87.190-137-248.telecom.net.ar] has joined #lisp 19:28:47 -!- ehu`_ [n=chatzill@82-170-33-173.ip.telfort.nl] has quit [Read error: 110 (Connection timed out)] 19:29:31 -!- Pandala [n=Pandala@LSt-Amand-152-33-12-243.w82-127.abo.wanadoo.fr] has quit ["Leaving"] 19:29:38 kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 19:31:14 dwave [n=ask@212251242133.customer.cdi.no] has joined #lisp 19:31:26 Modius: git diff -w 19:32:45 -!- sbahra [n=sbahra@c-76-21-209-249.hsd1.dc.comcast.net] has quit [Remote closed the connection] 19:32:50 -!- rread [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has quit [Read error: 110 (Connection timed out)] 19:36:43 nikodemus: are you using amd, or intel? (and how big are these matrices, just 4x4 or big?) 19:37:00 -!- dlowe [n=dlowe@c-24-91-154-83.hsd1.ma.comcast.net] has quit ["Leaving."] 19:37:20 4x4 19:37:26 on intel 19:37:26 ah 19:37:33 nikodemus: are you still working on GRT? 19:37:36 -!- TimDalySr [n=user@dynamic-acs-72-23-19-139.zoominternet.net] has quit ["ERC Version 5.2 (IRC client for Emacs)"] 19:37:44 is it a nehelem? 19:37:48 nope. this is a different toy 19:37:53 oudeis [n=oudeis@89-139-94-206.bb.netvision.net.il] has joined #lisp 19:37:53 ah 19:38:08 -!- vasa [n=vasa@80.94.234.105] has quit ["I am not vasya, i am vasa"] 19:38:11 manic12: nope. why do you ask? 19:38:43 there is a lisp hacker at UC Berkeley that does a lot of work in that area 19:39:56 for his particular application he does not use sse though 19:41:00 personally, I would try CUDA if I were doing a raytracer 19:41:12 Excuse me, is there a way to "redirect" what a lisp program prints out by (format t...) into a list of strings? 19:41:24 format nil 19:41:38 <_3b> or with-output-to-string 19:42:12 <_3b> converting to a list not included in either case 19:43:20 rindolf [n=shlomi@bzq-219-139-216.static.bezeqint.net] has joined #lisp 19:43:50 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 19:43:52 manic12: sure, but i'm not going to write a CL->GPU thing first 19:43:55 nikodemus: you could post to lisp-matrix-devel@googlegroups.com if it's still active 19:44:09 manic12: i think you misunderstand my goals 19:44:23 HET3 [n=diman@mars.htu.tuwien.ac.at] has joined #lisp 19:44:26 cads [n=max@c-76-122-89-218.hsd1.ga.comcast.net] has joined #lisp 19:44:40 *manic12* scrolls back 19:44:41 _3b: in either case I got to add code everywhere (format t) is called, right? 19:44:46 no toy for you :P 19:45:48 <_3b> xuanwu: you might not for with-output-to-string 19:46:11 HET4 [n=diman@mars.htu.tuwien.ac.at] has joined #lisp 19:46:14 -!- HET4 [n=diman@mars.htu.tuwien.ac.at] has quit [Remote closed the connection] 19:46:16 (1) have a non-sbcl lisp project that has no users except me and will never be finished (2) work on easy SIMD bits related to SBCL 19:46:36 HET4 [n=diman@mars.htu.tuwien.ac.at] has joined #lisp 19:46:42 <_3b> xuanwu: (with-output-to-string (*standard-output*) (call-stuff-that-calls (format t ...))) 19:46:51 ok, i see 19:48:02 you don't have to contribute to a project to ask smart people on lisp-matrix-devel though 19:49:40 sure, but the only hard bit here is the SBCL side of the equation -- getting vectors aligned properly 19:49:59 writing a matrix-multiply vop that uses SIMD once that is done is the easy bit 19:51:58 i'm pretty sure my alignment code in allocate-vector-on-heap is suboptimal, though 19:52:02 rstandy [n=rastandy@net-93-144-233-134.t2.dsl.vodafone.it] has joined #lisp 19:52:20 -!- kinnetica [n=kinnetic@dhcp-226-32.wireless.american.edu] has quit [] 19:54:00 -!- Spyderco2 [n=nash@194.45.110.65] has quit [Read error: 104 (Connection reset by peer)] 19:55:36 nikodemus: sounds like a fun challenge 19:55:53 are there SIMD vops available already? 19:56:04 -!- ejs [n=eugen@92-161-124-91.pool.ukrtel.net] has quit ["This computer has gone to sleep"] 19:56:18 it would be interesting to (ab)use SSE for other tasks :D 19:57:34 -!- HET2 [i=diman@xover.htu.tuwien.ac.at] has quit [Remote closed the connection] 19:57:49 Pandala [n=Pandala@LSt-Amand-152-33-12-243.w82-127.abo.wanadoo.fr] has joined #lisp 19:58:49 it has been interesting working on linking in a large C++ library to lisp 19:59:06 vasa [n=vasa@80.94.234.105] has joined #lisp 20:00:21 planet lisp wants to know! linking c++ to lisp has not been written about nearly enough 20:00:43 ejs [n=eugen@92-161-124-91.pool.ukrtel.net] has joined #lisp 20:00:46 i was all cozy in my lisp world, then i had to actually learn something 20:01:11 I always imagined SSE vops would be hacked in as a quick kludge that operated within specialized arrays, but pkhuong seems to be doing it the right way 20:01:21 write to planet lisp even if i'm not done? 20:01:31 SSE could be used for a lot of more things than just arrays 20:01:32 -!- HET3 [n=diman@mars.htu.tuwien.ac.at] has quit [Remote closed the connection] 20:01:53 Davidbrcz [n=david@nsc.ciup.fr] has joined #lisp 20:02:13 -!- oudeis [n=oudeis@89-139-94-206.bb.netvision.net.il] has quit ["This computer has gone to sleep"] 20:02:18 and unlike other, lower-level languages, you could exploit them better if you have type information 20:02:45 -!- vasa [n=vasa@80.94.234.105] has quit [Client Quit] 20:02:46 -!- shmho [n=user@58.142.15.103] has quit [Remote closed the connection] 20:03:04 oh, i'd missed that! (paul's stuff) 20:04:04 -!- SandGorgon [n=user@122.162.233.53] has quit [Read error: 110 (Connection timed out)] 20:05:02 _3b: thanks! it works. Now I just need to seperate the whole strings into a list of lines. 20:05:19 oh wow, that's cool, machine instructions for adding vectors 20:05:23 oudeis [n=oudeis@89-139-94-206.bb.netvision.net.il] has joined #lisp 20:05:48 -!- nvoorhies [n=nvoorhie@adsl-75-36-207-191.dsl.pltn13.sbcglobal.net] has quit [Read error: 104 (Connection reset by peer)] 20:05:55 nvoorhies [n=nvoorhie@adsl-75-36-207-191.dsl.pltn13.sbcglobal.net] has joined #lisp 20:05:55 SSE includes some other interesting stuff, afaik including some cache control operations 20:06:35 -!- araujo [n=araujo@gentoo/developer/araujo] has quit ["Leaving"] 20:06:44 i'm suddenly tempted to just use paul's SSE-VALUE object as matrices - they are just the right size after all... 20:07:23 boyscare1 [n=bm3719@c-68-32-124-6.hsd1.va.comcast.net] has joined #lisp 20:09:17 -!- jeremiah [n=jeremiah@31.Red-213-98-123.staticIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)] 20:09:29 hefner: well, I'm loading from specialised arrays into sse-values... Bulk storage of SSE values is probably a sign that you're doing it wrong ;) 20:10:15 -!- boyscared [n=bm3719@c-68-32-124-6.hsd1.va.comcast.net] has quit [Read error: 110 (Connection timed out)] 20:12:07 the loading isn't a noticeable hit? 20:12:33 it's a movqda into an xmm register, usually. 20:13:19 not as nice as directly passing an EA to the operation, but that's an issue throughout python. 20:13:35 right. i was thinking of sse-values on the heap 20:16:16 and on x86-64 the data is already aligned. neat'o 20:16:39 -!- anekos is now known as awayekos 20:17:47 some people used vector operations for speeding up things like copying etc. Liberal use of them might be quite interesting :) 20:18:52 pkhuong: is your stuff up somewhere? 20:20:25 tombom [i=tombom@wikipedia/Tombomp] has joined #lisp 20:20:33 Pushing to repo right now 20:20:40 jbmigel [n=jbm@S010600179a220e57.cg.shawcable.net] has joined #lisp 20:21:58 It's up at . 20:24:21 jeremiah [n=jeremiah@31.Red-213-98-123.staticIP.rima-tde.net] has joined #lisp 20:26:30 gracias! 20:26:46 do you have any tests/example code that uses it? 20:28:12 -!- akcom [i=akcom@rrcs-67-79-132-132.se.biz.rr.com] has quit [Read error: 110 (Connection timed out)] 20:30:40 -!- oudeis [n=oudeis@89-139-94-206.bb.netvision.net.il] has quit ["This computer has gone to sleep"] 20:31:13 -!- hugod [n=hugod@bas1-montreal50-1279442133.dsl.bell.ca] has quit [] 20:34:16 http://paste.lisp.org/display/77541, maybe? 20:37:39 -!- ejs [n=eugen@92-161-124-91.pool.ukrtel.net] has quit [Read error: 110 (Connection timed out)] 20:38:12 niftiness 20:39:08 if i get the alignment on x86 right, do you foresee any significant hurdles in porting the whole sse-values stuff to 32 bits? 20:39:20 -!- dwave [n=ask@212251242133.customer.cdi.no] has quit ["Be back later"] 20:39:25 mm.. We'd need to port the register definitions too. 20:39:45 should that not be mostly copy-paste? 20:40:23 Probably, but I've encountered a lot of frustrating build failures hacking around that area ;) 20:40:39 ah 20:41:18 well, i'll see how the gc hacking goes first. if i get bored i'll just run on x86-64 :) 20:41:54 (which, given immediate single floats, i should probably be doing anyways) 20:42:33 might be time for x86/SSE. 20:42:51 -!- Pandala [n=Pandala@LSt-Amand-152-33-12-243.w82-127.abo.wanadoo.fr] has quit ["Leaving"] 20:43:12 tombom_ [i=tombom@wikipedia/Tombomp] has joined #lisp 20:44:02 -!- rread_ [n=rread@nat/sun/x-8c721a9981300896] has quit [Read error: 110 (Connection timed out)] 20:44:37 -!- jbmigel [n=jbm@S010600179a220e57.cg.shawcable.net] has quit ["Ex-Chat"] 20:45:16 -!- rindolf [n=shlomi@bzq-219-139-216.static.bezeqint.net] has quit ["Leaving"] 20:46:29 -!- 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."] 20:48:17 rread [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has joined #lisp 20:48:33 dwave [n=ask@212251242133.customer.cdi.no] has joined #lisp 20:50:23 -!- xan [n=xan@cs78225040.pp.htv.fi] has quit ["Lost terminal"] 20:51:26 -!- mega1 [n=mega@3e44ad1e.adsl.enternet.hu] has quit [Read error: 110 (Connection timed out)] 20:52:31 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 20:59:13 -!- tombom [i=tombom@wikipedia/Tombomp] has quit [Read error: 101 (Network is unreachable)] 20:59:13 -!- tombom_ is now known as tombom 20:59:43 -!- dwave [n=ask@212251242133.customer.cdi.no] has quit ["Be back later"] 21:01:08 chessguy_work [n=chessguy@pool-173-79-200-142.washdc.fios.verizon.net] has joined #lisp 21:02:31 durka42 [n=durka@d81.wireless.swarthmore.edu] has joined #lisp 21:02:52 schoppenhauer [n=css@unaffiliated/schoppenhauer] has joined #lisp 21:06:41 -!- durka42 [n=durka@d81.wireless.swarthmore.edu] has quit [Client Quit] 21:07:30 rread_ [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has joined #lisp 21:10:11 -!- Jabberwockey [n=Tumnus_@port-16305.pppoe.wtnet.de] has quit [Remote closed the connection] 21:12:51 how can I make (/ 32 3) return a floating point? 21:13:00 or which function should i run over it? 21:13:14 <_3b> (float (/ 32 3)) or (/ 32 3.0) 21:13:34 thanks! 21:13:54 <_3b> or use coerce 21:14:27 -!- HET4 is now known as HET2 21:14:47 -!- chessguy_work is now known as chessguy 21:14:47 what advantages it will give comparing to float? 21:14:55 sbahra [n=sbahra@c-76-26-144-229.hsd1.dc.comcast.net] has joined #lisp 21:17:29 -!- manic12 [n=manic12@c-76-29-88-103.hsd1.il.comcast.net] has quit [Read error: 110 (Connection timed out)] 21:19:20 -!- kleppari [n=spa@bitbucket.is] has quit [Remote closed the connection] 21:19:24 kleppari [n=spa@bitbucket.is] has joined #lisp 21:22:49 pkhuong: quite possibly -- those needing sse-less can build without the feature 21:23:41 i would guess 99% of x86 machines on which sbcl is run support SSE these days 21:24:01 nikodemus: I wouldn't be so sure :D 21:24:11 -!- rread [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has quit [Read error: 113 (No route to host)] 21:24:15 x86 is a mess 21:24:31 if that turns out to be the case, the sse feature can be off by default 21:25:09 cmucl has two cores and loads one on start-up 21:25:32 -!- fe[nl]ix [n=algidus@88-149-208-178.dynamic.ngi.it] has quit [Read error: 145 (Connection timed out)] 21:26:23 fe[nl]ix [n=algidus@88-149-208-224.dynamic.ngi.it] has joined #lisp 21:26:25 feh 21:27:17 doesn't sbcl already use sse? 21:27:29 on x86-64, where is is always available 21:27:45 i see 21:28:13 -!- Davidbrcz [n=david@nsc.ciup.fr] has quit ["Ex-Chat"] 21:29:35 (and even there only for fp) 21:34:47 -!- dkcl [n=dkcl@unaffiliated/dkcl] has quit ["leaving"] 21:34:59 dkcl [n=dkcl@unaffiliated/dkcl] has joined #lisp 21:39:09 -!- Fufie [n=innocent@86.80-203-225.nextgentel.com] has quit ["Leaving"] 21:40:24 If I am using CLOS and I am trying to a generic function with defmethod; and I specify my own type, should I be able to use that defined type in my defmethod specifications? 21:40:39 hrr4 [n=hrr4@p5081A1BD.dip0.t-ipconnect.de] has joined #lisp 21:42:12 CLOS only works on classes. 21:42:38 bummer, would be cool if it worked on other stuff... 21:45:17 Note that some things you can define that have classes, even if they aren't called such, but things defined via deftype are not such. 21:46:17 ace4016 [i=ace4016@cpe-76-168-248-118.socal.res.rr.com] has joined #lisp 21:46:35 -!- Hun` [n=Hun@pd956be5d.dip0.t-ipconnect.de] has quit [Read error: 104 (Connection reset by peer)] 21:46:48 -!- elurin [n=user@81.213.201.55] has quit [Read error: 60 (Operation timed out)] 21:48:00 elurin [n=user@81.213.201.55] has joined #lisp 21:48:03 I guess I'm stuck with case statements then 21:48:04 hum 21:49:54 typecase? 21:49:58 JAS415: what kind of types of your own are you talking about? 21:50:20 deftype? defstruct? defclass? define-condition? 21:50:42 nikodemus: While you're here and we're talking about types, do you know anything about SB-ALIEN's type system? 21:50:46 you can dispatch on all of those _except_ deftype 21:50:54 well they are keywords relating to directions 21:51:21 You know that symbols can be eql-specialized on, right? 21:51:27 (defmethod go ((direction (eql :north))) ...) 21:51:33 -!- Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has quit [Remote closed the connection] 21:51:36 go, eh? 21:52:04 JAS415 pasted "untitled" at http://paste.lisp.org/display/79190 21:52:31 ah 21:52:47 so i'd do an eql 21:52:50 pstickne_ [n=pstickne@c-24-21-76-57.hsd1.wa.comcast.net] has joined #lisp 21:53:09 For DIRECTIONP, have a look at the function MEMBER. 21:53:20 For your deftype, consider a MEMBER type. 21:53:39 -!- mvilleneuve [n=mvillene@che33-1-82-66-18-171.fbx.proxad.net] has quit ["Lost terminal"] 21:53:46 and unless you really need to use keywords, you can just as well do 21:53:56 (defclass direction () ()) 21:54:17 (defvar *north* (make-instance 'direction)) -- etc 21:54:33 depending on the whole thing, keywords may be better -- or not 21:54:44 nyef: uh. maybe. i've poked at it, but i don't really have a good overall view 21:55:08 Oh well. 21:56:10 I noticed not so long ago that SB-ALIEN doesn't handle "tagged" data very well. 21:56:40 you mean the dreaded "this will make your computer look really slow" compiler notes? 21:56:58 No, I mean something like wrapping a windows HGDIOBJ. 21:57:02 oh, right 21:57:21 Essentially, there's an API that, given a random HGDIOBJ, will tell you what type it is. 21:57:36 interfacing with ad-hoc c object systems with tags, basically? 21:57:38 Ok i think i have something that will work, thanks guys 21:57:57 Yeah, or non-C object systems with tags or other run-time type information. 21:58:35 and you'd like the type manifest on lisp side with better accuracy? 21:58:41 Yes. 21:59:04 The problem is that TYPEP on a local alien operates solely in terms of its declared type. 21:59:09 -!- white-rabbit-obj [n=cpc5@cpe-67-241-171-153.buffalo.res.rr.com] has quit ["Leaving."] 21:59:23 Which leads me to wonder what the heck else I missed. 21:59:47 so you cast to the accurate type, but TYPEP doesn't believe it? 21:59:58 I don't want to have to cast. 22:00:12 -!- yahooooo [n=yahooooo@76.104.183.185] has quit [Connection timed out] 22:00:55 but rather? would you want to have sb-alien do the api calls and figure out the type, or are you willing to do that in your own code? 22:01:08 I'd rather have sb-alien do the API calls. 22:01:09 chessguy_work [n=chessguy@pool-173-79-200-142.washdc.fios.verizon.net] has joined #lisp 22:01:26 I already have it for naturalize, it's just that it won't happen -unless- the handle is naturalized. 22:01:35 there isn't really any provision for that in sb-alien that i know of 22:01:37 http://www.lisphacker.com/temp/hgdiobj.lisp if you want to see the code. 22:04:15 yahooooo [n=yahooooo@c-76-104-183-185.hsd1.wa.comcast.net] has joined #lisp 22:04:51 iirc sb-alien specifically tries to avoid naturalizing things as much as possible, because it is inefficient 22:05:00 Right. 22:05:11 Though the warning about when it does so sucks as well. 22:05:22 so i think you'd be really better off doing your own naturalization in a layer above sb-alien 22:05:48 -!- tombom [i=tombom@wikipedia/Tombomp] has quit ["Peace and Protection 4.22.2"] 22:06:05 I'd rather see sb-alien suck less, though, you know? 22:06:15 quite 22:06:20 *if* cffi's translator stuff would work for this (i haven't used it really, so i'm pretty hazy there), then sb-cffi might be worth a go... 22:07:08 AIUI, CFFI doesn't support not naturalizing objects, and doesn't tie into the compiler's notion of the type system anywhere near as well. 22:07:19 ok 22:07:30 (And we can leave aside the "another blasted dependency" argument for now.) 22:07:44 i meant as sb-alien replacement :) 22:08:09 Yeah, right. And you were going to get that to run in a cold-core -how-? 22:08:24 ah 22:09:08 maybe just a more plain bones FFI that doesn't cons up additional containers 22:10:16 like cffi in that respect -- because i *hate* it when i try to avoid naturalization and sb-alien insists on it and kills performance 22:11:08 iirc sb-posix:stat is one place that i've tried to make not suck so badly several times 22:11:32 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 22:11:48 sb-alien could then be a "safe" layer on top of that 22:11:50 Yeah, there are a couple things like that I could want from sb-alien. One is a good explanation of exactly when and why its performance profile starts sucking, and another is better integration with the host lisp. 22:12:15 For example, I'd like to be able to use LET instead of WITH-ALIEN and still get unboxed stack storage. 22:12:20 -!- chessguy [n=chessguy@pool-173-79-200-142.washdc.fios.verizon.net] has quit [No route to host] 22:12:47 actually, if we had a lower-level FFI, we could just use it internally, and implement the fancy stuff with clos... 22:13:30 That we could, and it might even be a good idea. 22:13:31 because sb-alien is one object system too many right now (in addition to the type system, streams, and the compiler...) 22:13:43 Heh. 22:13:57 Isn't it -part- of the type system, though? 22:14:55 Hrm. Guess not quite. 22:15:05 not really. tied to it, but not part of it 22:15:33 The ideal, of course, would be cold-load PCL. 22:15:50 yeah. that would make a lot of things a lot easier 22:16:12 X-Scale2 [i=email@2001:470:1f08:b3d:0:0:0:2] has joined #lisp 22:16:29 hmph. why does this not work? 22:16:47 Why does what not work? 22:16:54 aligning vector data 22:17:13 nikodemus: "the fancy stuff with clos" would be more or less CFFI, which only uses SAPs 22:17:14 1. add 2 words padding to the allocated storage 22:17:24 That I do not know... 22:17:36 2. add 15 to result 22:17:47 2. and with #xfffffff0 22:17:54 3. or with #x8 22:18:26 works fine when i hotpatch it, breaks cold init when i build it 22:18:41 ... Breaks in what sense? 22:18:48 It's just not aligned at that point? 22:19:01 -!- chessguy_work is now known as chessguy 22:19:02 no, what looks like random corruption 22:19:26 unbound variable in the package system, etc 22:19:36 my logic is broken 22:20:45 -!- jao [n=jao@131.Red-79-156-141.staticIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)] 22:23:31 -!- pstickne_ [n=pstickne@c-24-21-76-57.hsd1.wa.comcast.net] has quit [Read error: 60 (Operation timed out)] 22:23:37 nyef: one thing you could do would be to initially have the objects as undifferentiated HGDIOBJ instances, and call the GetObjectType when you need to dispatch on it, change-class to the right class, and go 22:23:52 bah. 22:23:56 beach` [n=user@ABordeaux-158-1-73-138.w90-60.abo.wanadoo.fr] has joined #lisp 22:23:58 this works without the OR 22:24:18 so if i align the _whole_ vector to 16 bytes, no trouble. 22:24:27 so my math is wrong. i suck 22:24:52 Again, doesn't work unless the object gets naturalized, and I'm looking specifically at the case where it doesn't. 22:25:10 oh, right 22:25:16 Oh well, I'll figure something out. 22:25:27 I've already had to patch sb-alien once for this project. 22:25:29 you're master of bits and logical ops 22:25:33 -!- beach [n=user@ABordeaux-158-1-43-249.w90-55.abo.wanadoo.fr] has quit [Read error: 60 (Operation timed out)] 22:25:39 Uh-oh. 22:26:51 Oh, yeah, I see where the math you posted earlier was wrong besides it having two step 2s. 22:26:52 nikodemus pasted "my problem" at http://paste.lisp.org/display/79192 22:27:09 -!- HG` [n=wells@91.111.13.16] has quit [Client Quit] 22:27:17 yay, tell 22:27:47 i pad too little? 22:27:54 Yeah. 22:28:11 You're shy two words on your padding. 22:28:21 ... I think. 22:28:26 Adamant [n=Adamant@c-76-29-188-22.hsd1.ga.comcast.net] has joined #lisp 22:28:58 You align to 16, mask to 16, and then add 8, so if you were in the second 8 to begin with you overflow, don't you? 22:29:10 Or was it if you were in the first 8? 22:29:15 One or the other. 22:29:41 hm 22:30:19 but since i need quadword alignment, and allocation is double-word alignment the max padding needed should be 2 words? 22:31:04 Your allocation SIZE is padded, right? 22:31:32 yes, the lines commented with ;;; Pad in the paste 22:31:45 -!- Jacob_H [n=jacob@92.2.126.235] has quit ["Leaving"] 22:31:52 -!- gemelen [n=shelta@shpd-78-36-164-77.static.vologda.ru] has quit ["I wish the toaster to be happy, too."] 22:32:20 Why are you padding relative to the size of vector-data-offset? 22:32:20 oudeis [n=oudeis@77.124.71.204] has joined #lisp 22:32:32 -!- ffx` [n=ffx@60-241-74-240.static.tpgi.com.au] has quit [Connection timed out] 22:32:33 And why the bare "4"s? 22:32:47 -!- X-Scale [i=email@2001:470:1f08:b3d:0:0:0:2] has quit [Read error: 110 (Connection timed out)] 22:33:01 because i didn't find an appropriate constant, and didn't define one yet 22:33:17 And isn't that going to pad the wrong direction? 22:33:35 YES! 22:33:43 that's it. i suck. 22:33:54 no, wait. what? 22:34:23 mokogobo [n=mokogobo@pcp075595pcs.unl.edu] has joined #lisp 22:35:01 What's vector-data-offset? 2? 22:35:09 yes 22:35:27 And 4 is the number of words you need to align to... 22:35:37 Hrm. 22:36:12 yes. and it's the data that should be aligned, not the header. 22:37:07 <_3b> should that be - #x8 instead of or #x8? 22:38:00 _3b: No, because it's already masked down to the start of the paragraph if the allocation is paragraph-aligned. 22:38:44 <_3b> nyef: not start of next paragraph? 22:39:05 Not if it's aligned to start with. 22:39:15 manic12 [n=manic12@c-76-29-88-103.hsd1.il.comcast.net] has joined #lisp 22:39:15 -!- blbrown [n=Berlin@c-71-236-25-127.hsd1.ga.comcast.net] has quit ["Ex-Chat"] 22:39:25 If it's disaligned, -then- it's at the start of the next paragraph. 22:39:34 Err... 22:39:42 Yeah, that's shy two words of padding in that case. 22:40:09 mxb [n=mxb@bealbywm.plus.com] has joined #lisp 22:40:16 -!- mxb [n=mxb@unaffiliated/mxb] has quit [Remote closed the connection] 22:40:39 i don't really follow. maybe this is too late for me 22:40:54 If it's aligned to start with, the OR soaks up two words of padding. 22:41:19 If it's disaligned to start with, the LEA + AND soaks up two words of padding and -then- the OR soaks up two words of unallocated storage. 22:41:37 i think i see 22:41:56 And the OR can be elided. Add the 8 to the displacement when you tag. 22:42:09 gotcha 22:42:25 but can i do this with 2 words of padding without adding a branch? 22:42:55 -!- milanj [n=milan@212.200.194.238] has quit ["Leaving"] 22:43:17 because 2 words should be enough adjustment for any doubleword aligned allocation to get quadword allocation 22:43:48 hm, padding with 4 words loses still -- heap corruption again 22:43:54 mxb [n=mxb@bealbywm.plus.com] has joined #lisp 22:43:55 -!- mxb [n=mxb@unaffiliated/mxb] has quit [Remote closed the connection] 22:44:54 p0a [n=user@athedsl-385381.home.otenet.gr] has joined #lisp 22:45:00 Hello, is always '(foo) equivalent to (quote foo) 22:45:20 no. it's equivalent to (quote (foo)) 22:45:27 'foo is (quote foo) 22:46:08 -!- jewel [n=jewel@dsl-242-183-106.telkomadsl.co.za] has quit [Read error: 113 (No route to host)] 22:46:09 so '(1 2 3) is equivalent to (quote (1 2 3)) 22:46:19 Well, your allocation is always qword-aligned, isn't it? 22:46:21 Ah. thanks 22:46:35 no, allocation is double-word aligned 22:46:46 So, every 4 bytes then? 22:47:06 uh, overload error 22:47:11 let's talk bytes :) 22:47:16 -!- p0a [n=user@athedsl-385381.home.otenet.gr] has quit [Client Quit] 22:47:16 allocation is 8 byte aligned 22:47:21 Okay, your allocation is 8-octet aligned. 22:47:34 i need data at 16 byte alignment 22:48:20 if the allocation is 16-byte aligned already, i need to jump ahead 8 bytes so that data ends up on the 16 byte boundary 22:48:44 if the allocation is just 8, not 16-byte aligned, i'm good to go 22:49:17 And you've padded for the extra. 22:49:27 Don't mask the pointer down, then. 22:49:32 Just OR in an 8. 22:49:54 At which point you can't fold it into the tagging, because that is an ADD not an OR. 22:50:41 Although, lowtags are 3 bits, which means that you could OR it in instead of adding... 22:51:27 So, (allocation result size) (inst or result (logior 8 other-pointer-lowtag)) (sc-case type ... 22:51:44 i think i see it. i'll need to play at the repl + pen and paper tomorrow to drive home the lesson 22:51:45 And that means you only need the two words of padding that you already have. 22:51:52 thanks! 22:51:59 No problem. 22:52:04 -!- nyef [n=nyef@24.220.182.229] has left #lisp 22:52:09 nyef [n=nyef@24.220.182.229] has joined #lisp 22:52:18 Oops. Killed the wrong window. :-/ 22:52:39 <_3b> can you get the address before you decide the size? 22:53:20 if i'm lucky and there is still space in the thread-local allocation area 22:53:39 oh, wait 22:54:19 hrm 22:55:01 i might be able to teach ALLOCATION about padding... 22:55:42 Hey, what happens with the uninitialized space you skipped over? Is it guaranteed to be 0-fill, or can it have garbage tags? 22:56:34 zero filled 22:57:13 should be, at least 22:59:20 Be nasty if it wasn't. At least zero-fill reads as (0 . 0), and not anything of more significance to the GC. 22:59:48 -!- oudeis [n=oudeis@77.124.71.204] has quit ["This computer has gone to sleep"] 22:59:49 except that my build still breaks -- but identically to the way it did with 16 bytes of padding 23:00:08 so perhaps i am getting garbage indeed 23:00:24 Are you actually using SSE instructions at this point, or are you just trying to get the alignment correct? 23:00:46 working on the alignment 23:01:06 -!- danlei [n=user@pD9E2C4E6.dip.t-dialin.net] has quit [Remote closed the connection] 23:01:09 this VOP is the only change in the tree at the moment 23:01:23 So disaligned arrays from genesis aren't likely to cause problems yet... 23:01:33 nope 23:02:00 and in this tree GC doesn't preserve array alignment either 23:02:24 -!- elias` [n=me@unaffiliated/elias/x-342423] has quit ["leaving"] 23:03:08 Is the allocation macro okay with the result and size parameters being the same TN? 23:03:40 should be -- especially since that is the same way in HEAD 23:03:57 Fair enough. 23:04:32 -!- xuanwu [i=xuanwu@140-182-146-197.dhcp-bl.indiana.edu] has quit [] 23:06:29 Well, if it's uninitialized heap space that's the problem, then it could be at either end. 23:06:44 And clearing it at the beginning would be easier than clearing it at the end. 23:07:55 Beyond that, I'm rather at a loss. 23:08:00 yes -- but garbage there sounds like a gc bug to me. it's been a few months, but i'm reasonably sure we should be zeroing fresh pages 23:08:48 i'm doing a final build with a minimized diff -- then to bed. tomorrow i'll check if the space has junk in it 23:09:02 Okay. Sleep well, then. 23:11:24 night 23:11:25 -!- nikodemus [n=nikodemu@cs181229041.pp.htv.fi] has quit ["Leaving"] 23:13:44 -!- fiveop [n=fiveop@pD9E6C5C5.dip.t-dialin.net] has quit ["humhum"] 23:15:13 -!- ianmcorvidae [n=ianmcorv@fsf/member/ianmcorvidae] has quit [Remote closed the connection] 23:16:56 -!- dfox [n=dfox@r11jn246.net.upc.cz] has quit [Remote closed the connection] 23:17:27 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [] 23:20:45 ianmcorvidae [n=ianmcorv@student167-193.hampshire.edu] has joined #lisp 23:22:28 davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #lisp 23:23:26 -!- postamar [n=postamar@x-132-204-253-215.xtpr.umontreal.ca] has left #lisp 23:25:38 durka42 [n=durka@d81.wireless.swarthmore.edu] has joined #lisp 23:29:46 danlei [n=user@pD9E2C4E6.dip.t-dialin.net] has joined #lisp 23:29:55 -!- elurin [n=user@81.213.201.55] has quit [Read error: 110 (Connection timed out)] 23:33:15 -!- mokogobo [n=mokogobo@pcp075595pcs.unl.edu] has quit [Read error: 110 (Connection timed out)] 23:34:36 armon [n=armon@114.28.192.170] has joined #lisp 23:35:02 existentialmonk [n=carcdr@64-252-67-89.adsl.snet.net] has joined #lisp 23:41:13 wedgeV_ [n=wedge@80.122.51.198] has joined #lisp 23:42:38 araujo [n=araujo@gentoo/developer/araujo] has joined #lisp 23:44:44 -!- wedgeV [n=wedge@80.122.51.198] has quit [Read error: 110 (Connection timed out)] 23:47:09 -!- wedgeV_ [n=wedge@80.122.51.198] has quit [] 23:49:02 -!- Modius [n=Modius@adsl-68-92-148-234.dsl.austtx.swbell.net] has quit [Read error: 110 (Connection timed out)] 23:49:34 -!- Odin- [n=sbkhh@s121-302.gardur.hi.is] has quit [] 23:51:02 can I make aliases for commands? 23:52:14 commands? 23:52:18 what commands? 23:53:29 nixzs: aliases for functions ? 23:53:37 if you mean functions: (setf (symbol-function 'my-alias) #'target) 23:59:05 alias for stuff such setf setenv, second, assoc <--- those 23:59:15 dfox [n=dfox@r11jn246.net.upc.cz] has joined #lisp 23:59:23 nixzs: these are not in the same category. 23:59:34 setf is different, since it's a special form 23:59:37 that's how the Casting SPELs tutorial calls them 23:59:47 setf is a macro. setenv is unknown, second is an accessor. asssoc is a function.