00:03:11 blitz_ [n=blitz@2001:6f8:10f6:200:21b:77ff:fe41:11ab] has joined #lisp 00:03:35 Now I had to use three backquotes 00:03:35 http://paste.lisp.org/display/79103#2 00:03:50 they're probably redundant 00:04:36 At least, it's quite readable. So don't worry. 00:05:30 -!- apo_ [n=apo@pD9E7F10E.dip.t-dialin.net] has quit [Read error: 101 (Network is unreachable)] 00:05:31 Notice that in scheme, (define (g f x) (if (< x 0) 1 (f (- x 1)))) (g g 3) works, but not with your macro :-( 00:05:50 *pjb* is evil >:-> 00:06:26 xinming [n=hyy@125.109.254.237] has joined #lisp 00:07:25 what? this works? 00:07:34 In scheme, yes. 00:07:37 But f is called with 1 argument and you're passing g for f 00:07:56 Oops. I mean (f f (- x 1)) 00:08:13 ah 00:08:26 p0a: Don't worry too much about it, it's because scheme is a lisp-1 and CL a lisp-2. Something could be done with define, but it would be an incomplete hack. You would have to write another macro to transform lisp-1 expressions into lisp-2. 00:09:26 The conclusion is that you have to be careful with superficial macros such as define: they may look the same as in another language, but if the fundament semantics are different, it's misleading. 00:11:00 I see why it does not work now 00:12:00 durka42 [n=durka@d81.wireless.swarthmore.edu] has joined #lisp 00:12:54 I could modify the cars of lists that are parameter names to funcalls. Make (f f (- x 1)) to (funcall f f (- x 1)) 00:13:21 Yes. 00:14:26 -!- badkins [n=user@adsl-068-209-204-112.sip.rdu.bellsouth.net] has left #lisp 00:14:36 You would also have to handle (let ((f (lambda ...))) (f ...)), and be careful with shadowing: (lambda (f) (f 3) #|ok|# (let ((f 42)) (f 2) #|error|#)) 00:15:37 p0a: notice there is a whole R4RS scheme implementation written in CL: pseudo-scheme. 00:16:46 -!- dstatyvka [i=ejabberd@pepelaz.jabber.od.ua] has left #lisp 00:16:49 tomoyuki28jp [n=tomoyuki@w221062.ppp.asahi-net.or.jp] has joined #lisp 00:16:50 heh, that is cool 00:17:48 Well, enough IRCing, I have to go 00:17:49 -!- p0a [n=user@athedsl-380118.home.otenet.gr] has quit ["bye"] 00:17:57 ftp://ftp.cs.indiana.edu/pub/scheme-repository/imp/pseudo212.tar.gz 00:19:34 -!- HET2 [i=diman@xover.htu.tuwien.ac.at] has quit ["This computer has gone to sleep"] 00:21:18 What can I do with a swank server? I am currently use a tramp from emacs or use ssh+screen to edit remote file. Can I do more stuff with a swank server? 00:23:32 Well, you can use slime. It might prove helpful. 00:24:03 pjb: I have been using slime. 00:24:34 that means I have already been using a swank server? 00:25:01 I am not quite sure what the swank server is. 00:25:13 I don't think swank helps in accessing remote sources. The preconised way would be to share a nfs volume. But for remote development, tramp or ssh are probably better. 00:25:29 swank is the backend of slime. slime runs in emacs, swank in CL. 00:25:44 Yes, you've been using swank. 00:25:48 pjb: I see. thanks for the info. 00:26:43 The protocol is defined, so theorically you could use swank with another front-end than slime (something have been written for vim IIRC). 00:27:11 I see. 00:27:25 pjb: How do you run lisp program in a remote server? What I can think of is open emacs in a remote server inside screen and suspend screen to keep it running. 00:27:51 -!- sepult [n=buggarag@xdsl-87-78-130-246.netcologne.de] has quit ["leaving"] 00:28:00 Yes, that's what I do. There's also detachtty which is lighter than screen. 00:28:42 -!- dagnachew [n=dagnache@modemcable207.114-201-24.mc.videotron.ca] has quit [Client Quit] 00:28:43 pjb: uhm, do you run emacs inside a screen? 00:29:06 Or, with swank, you can just run sbcl (or whatever implemetnation) in screen, and connect to it with an emacs in your workstartion. 00:29:30 It really makes sense. 00:29:38 Yes, I also run emacs inside screen. But for server processes I don't start them from emacs. 00:30:46 In a remote server, if you don't open emacs, how do you edit running programs? Use swank and edit it from a local emacs? 00:31:10 Yes. 00:31:21 I see. 00:31:36 How can I run lisp program with a swank server? 00:31:56 -!- jbmigel [n=jbm@S010600179a220e57.cg.shawcable.net] has quit [Read error: 104 (Connection reset by peer)] 00:32:02 http://common-lisp.net/project/slime/doc/html/Connecting-to-a-remote-lisp.html 00:32:06 -!- nekobaka [n=baka@c-76-29-163-77.hsd1.ca.comcast.net] has quit [Read error: 54 (Connection reset by peer)] 00:32:22 pjb: thanks! 00:32:44 so, yay. a happy user of cxml-rpc just wrote to me that he got 110 transactions / second with it. that is a testament to the quality of cxml's klacks parser (: 00:32:44 -!- tmh [n=thomas@pdpc/supporter/sustaining/tmh] has left #lisp 00:33:53 -!- blitz_ [n=blitz@2001:6f8:10f6:200:21b:77ff:fe41:11ab] has quit [Read error: 60 (Operation timed out)] 00:34:54 JAS415 [n=jon@c-24-34-16-25.hsd1.ma.comcast.net] has joined #lisp 00:35:12 -!- slyrus_ [n=slyrus@dsl092-019-253.sfo1.dsl.speakeasy.net] has quit [Read error: 110 (Connection timed out)] 00:35:12 -!- milanj [n=milan@93.87.143.245] has quit ["Leaving"] 00:35:23 blitz_ [n=blitz@2001:6f8:10f6:200:21b:77ff:fe41:11ab] has joined #lisp 00:35:32 dulouz [n=ross@dsl254-119-219.nyc1.dsl.speakeasy.net] has joined #lisp 00:37:11 kinnetica [n=kinnetic@208-58-64-106.c3-0.129-ubr4.lnh-129.md.cable.rcn.com] has joined #lisp 00:37:21 Good night! 00:37:39 Good night! 00:37:46 thanks for your help, pjb. 00:43:10 -!- pstickne [n=pstickne@69.166.35.201] has quit [Read error: 110 (Connection timed out)] 00:44:01 -!- Odin- [n=sbkhh@s121-302.gardur.hi.is] has quit [] 00:53:43 nyef [n=nyef@24.220.182.229] has joined #lisp 00:53:49 Hello all. 00:53:49 nyef, memo from p_l: Been looking through your temp/. Nice MKR pic (Hikaru's notes xD) 00:54:04 Heh. 00:54:17 p_l: Herep? 00:55:17 the "?" seems redundant :) 00:55:37 Indeed. It's more of a pronunciation guideline than anything else. 00:57:19 So, I was surprised at SBCL 1.0.27.19, as I was under the impression that that was actually undesireable functionality. 00:57:31 -!- mrSpec [n=NoOne@82.177.125.6] has quit ["G'night!"] 00:57:38 hugod [n=hugod@bas1-montreal50-1279442133.dsl.bell.ca] has joined #lisp 01:00:05 -!- mjonsson [n=mjonsson@66-234-42-75.nyc.cable.nyct.net] has quit [Client Quit] 01:02:09 -!- metasyntax [n=taylor@pool-71-127-85-87.aubnin.fios.verizon.net] has quit [""Nichts mehr.""] 01:03:05 -!- blitz_ [n=blitz@2001:6f8:10f6:200:21b:77ff:fe41:11ab] has quit [sendak.freenode.net irc.freenode.net] 01:03:05 -!- nooper [i=nooper@2001:41c8:0:866:21c:c0ff:fe7f:7198] has quit [sendak.freenode.net irc.freenode.net] 01:03:05 -!- DrForr [n=drforr@eldwist.darkuncle.net] has quit [sendak.freenode.net irc.freenode.net] 01:03:05 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [sendak.freenode.net irc.freenode.net] 01:03:05 -!- rumbleca [n=rumble@174.0.46.123] has quit [sendak.freenode.net irc.freenode.net] 01:03:05 -!- qebab [n=finnrobi@eros.orakel.ntnu.no] has quit [sendak.freenode.net irc.freenode.net] 01:03:05 -!- alexbobp [n=alex@adsl-75-63-0-19.dsl.austtx.sbcglobal.net] has quit [sendak.freenode.net irc.freenode.net] 01:03:05 -!- Dazhbog [n=sampo@geek.fi] has quit [sendak.freenode.net irc.freenode.net] 01:04:12 blitz_ [n=blitz@2001:6f8:10f6:200:21b:77ff:fe41:11ab] has joined #lisp 01:04:12 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 01:04:12 qebab [n=finnrobi@eros.orakel.ntnu.no] has joined #lisp 01:04:12 Dazhbog [n=sampo@geek.fi] has joined #lisp 01:04:12 alexbobp [n=alex@adsl-75-63-0-19.dsl.austtx.sbcglobal.net] has joined #lisp 01:04:12 nooper [i=nooper@2001:41c8:0:866:21c:c0ff:fe7f:7198] has joined #lisp 01:04:12 rumbleca [n=rumble@174.0.46.123] has joined #lisp 01:04:12 DrForr [n=drforr@eldwist.darkuncle.net] has joined #lisp 01:04:38 -!- alexbobp [n=alex@adsl-75-63-0-19.dsl.austtx.sbcglobal.net] has quit ["leaving"] 01:04:51 alexbobp [n=alex@adsl-75-63-0-19.dsl.austtx.sbcglobal.net] has joined #lisp 01:05:04 HET2 [i=diman@xover.htu.tuwien.ac.at] has joined #lisp 01:08:37 -!- ramus` [n=ramus@adsl-69-209-223-131.dsl.chcgil.sbcglobal.net] has left #lisp 01:13:48 -!- nvoorhies_ [n=nvoorhie@adsl-75-36-207-191.dsl.pltn13.sbcglobal.net] has quit [] 01:21:27 -!- WarWeasle [n=brad@c-98-220-168-14.hsd1.in.comcast.net] has left #lisp 01:24:26 gigamonkey [n=user@adsl-99-35-219-234.dsl.pltn13.sbcglobal.net] has joined #lisp 01:24:40 Any happy PCL readers here? 01:25:14 -!- hyperboreean [n=none@unaffiliated/hyperboreean] has quit [Network is unreachable] 01:25:23 If so, how big a revision would it take for you to consider buying a 2nd edition, if you already have the 1st edition? 01:25:47 sbahra [n=sbahra@c-98-218-239-172.hsd1.dc.comcast.net] has joined #lisp 01:27:10 Looking for a cost-effectiveness argument for revision vs. writing a new book / accepting a book proposal for another author (depending on the desired target audience for the hypothetical argument)? 01:27:36 i never bought the first edition, but i read a good part of the second online 01:27:42 er 01:27:44 i mean first 01:28:11 *dulouz* smacks self to wake up abit 01:28:23 i haven't read the first one well enough to actually have an opinion 01:29:03 I'm sorry to say that I've never managed to spare the cash to purchase it either... But then, I also get the impression that I'm not quite the target audience for it anyway. 01:29:11 but if you must know what to write about i'd say, i'd like to see more about conditions 01:29:15 nyef: Thinking a bit about seeing if Apress is interested in doing a 2nd edition. (Especially now that O'Reilly is going -- perhaps -- to be entering the Lisp book market.) 01:29:33 O'Reilly is -what-? 01:29:40 Haven't you heard? 01:29:48 nyef: seriously, don't you have a twitter account? 01:29:48 No, I tend not to pay attention. 01:30:01 After ILC they specifically started soliciticing Lisp book proposals. 01:30:05 Does this mean that "Head First Scheme" has a chance to exist? 01:30:18 I heard from nearly a half dozen people asking my advice. 01:30:39 i'd buy a copy of it if I saw it on a bookshelf at B&N, just in the hopes that they start carrying more than java/ruby/python books 01:31:01 nyef: Could be. Though I think, interestingly, the editor sees Common Lisp as the dominant Lisp dialect. 01:31:24 Most lisp book I've read, I've read online, as it makes it easier to search them later. 01:31:37 guaqua: yeah, conditions are something I might try to expand on. 01:31:39 I'd buy it and put it in B&N then use the online version! 01:31:44 lol 01:32:15 -!- bobf [n=bob@unaffiliated/bob-f/x-6028553] has quit [Read error: 60 (Operation timed out)] 01:33:02 online is great, but i do a fair amount of reading on the subway. i did print out chapter of PCL once to read on my commute. 01:33:24 and PAIP is just too heavy to lug around 01:33:25 -!- postamar [n=postamar@x-132-204-252-135.xtpr.umontreal.ca] has quit [] 01:33:25 i have to say, as a newbie on lisp, i have no idea how much focus you should put on conditions. it's just one of the things i've had trouble getting right after reading the book 01:34:15 It does seem to be something, that despite my best efforts, people are still confused by. So if I did a new edition, I should see if I can improve it in some way. 01:34:41 but sorry, i think this wasn't the main thing you were asking of in here :) 01:35:48 guaqua: Conditions are one of those things that you can ignore for a while if you're just tooling around, but are incredibly useful as soon as you want to have stuff that can run without programmer interaction or provide a usable interface for detecting and recovering from various undesirable events and circumstances. 01:36:34 nyef: generally, when you're not just fooling around in the REPL :) 01:36:38 -!- Soulman [n=kvirc@42.84-48-88.nextgentel.com] has quit [Read error: 60 (Operation timed out)] 01:37:14 *hefner* wonders if you can render into an FBO from one process and use it as a texture in another 01:38:12 Okay, I just checked my first major lisp project (nevermore), and it doesn't use define-condition, handler-bind, handler-case, or anything with the word "restart" in it. 01:38:34 I don't think i've even read the documentation on that stuff 01:38:43 on the newbie note, I'm trying to get some slime stuff setup from cvs. i've got my .emacs set up to load things in slime/contrib/ (i'm using slime-fancy as my idicator that things are working). however, when trying to add slime-tramp to my slime-setup list, i dont seem to have any of the functions defined in slime-tramp available to use. any ideas? are my expectations incorrect? 01:38:55 *hefner* seldom uses the condition system to full advantage 01:39:05 slime setup would actually be a really good section 01:39:05 conditions/etc are one of those sort of features which are a lot less relevant if you're not writing a library for unspecified someone elses to use 01:39:13 in a lisp book 01:39:18 There are five calls to error, all of which have a single parameter of "Bogus width". 01:40:02 slime setup seems like a black art, i got mine to work, but i don't know *how* I did it 01:40:38 And of the four occurances of the word "signal", three are in comments and one is in a docstring. 01:40:43 JAS415: yeah. That's something I skimped on originally because it was a bit less stable then. 01:40:44 perhaps such a book could even be included with slime. they could call it.. a manual! ;) 01:41:18 Should be able to deal with it a bit more reasonably these days. 01:43:20 You might consider including a warning about using SBCL + SLIME + cygwin emacs. 01:43:41 Or SBCL/Win32 + SLIME in general. 01:44:25 -!- blitz_ [n=blitz@2001:6f8:10f6:200:21b:77ff:fe41:11ab] has quit [Read error: 101 (Network is unreachable)] 01:44:37 nyef: what's the warning? 01:44:58 Anyone know how well ClozureCL works on Windows? 01:47:20 -!- mrsolo [n=mrsolo@nat/yahoo/x-d930476f981eeeea] has left #lisp 01:47:35 aja [n=aja@S01060018f3ab066e.ed.shawcable.net] has joined #lisp 01:47:48 gigamonkey: http://www.lisphacker.com/temp/article-drafts/sbcl-win32-slime-and-cygwin-emacs.txt is an initial draft of what I have to say about the matter. I learned a decent chunk since then, so the article is badly in need of revision, though. The important bits are that using cygwin emacs means you need path translation and using SBCL/Win32 at all means you need a fixed SERVE-EVENT. 01:48:19 It'll probably all be sorted out in a year, but the situation is still bad now. 01:49:14 Maybe a couple of blurbs about different editors and useful extensions in general rather that focusing on emacs and slime (then a bit more detailed about how awesome emacs and slime and paredit are :-P ). 01:49:59 nyef: I think i ended up using windows emacs when i was on windows. 01:50:10 Mmm... I finally started using paredit in my default configuration. I still have occasional problems with M-(, but other than that it's not bothered me so far. 01:50:46 JAS415: Yeah, I'm experimenting with that as well, along with building a .emacs that supports SLIME and SBCL on both. 01:50:53 droogie [n=user@88.238.44.78] has joined #lisp 01:52:03 does SBCL on windows still compile stuff on C:\ ? 01:52:19 i installed ubuntu and never looked back tbh... 01:52:25 -!- amnesiac [n=amnesiac@p3m/member/Amnesiac] has quit ["Leaving"] 01:52:44 I have no idea, TBH... And that's more of a slime decision than an SBCL decision anyway. 01:53:19 I was pleasantly surprised when it turned out that with careful handling of pathnames I only needed to condition out three consecutive toplevel forms when not on cygwin to have everything "work". 01:53:58 <_3b> hefner: rendering to FBO in other threads should work, i think you need to handle synchronization yourself though 01:55:08 -!- mejja [n=user@85.229.182.246] has quit [Remote closed the connection] 01:55:32 -!- HET2 [i=diman@xover.htu.tuwien.ac.at] has quit ["This computer has gone to sleep"] 01:58:05 dulouz: about your problems with slime-tramp, are you using the cvs version of slime? In which OS? 01:58:20 rstandy: yes and linux 01:58:49 dulouz: could you paste your config on lisppaste? 01:58:58 my .emacs stuff? 02:00:27 dulouz: yes 02:01:30 -!- droogie [n=user@88.238.44.78] has quit [Remote closed the connection] 02:02:22 dulouz: btw, if you try to see the documentation for `slime-filename-translations' with C-h v slime-filename-translations 02:02:33 dulouz: Emacs will complain there isn't a var named like that? 02:03:05 dulouz pasted "slime-tramp problems" at http://paste.lisp.org/display/79114 02:04:24 yes, i do see it 02:04:38 documentation for that variable, that is. 02:04:44 dulouz: well, then slime-tramp is loaded ;-) 02:04:49 so i guess i'm just not understanding how to then use it 02:05:02 b4|hraban [n=b4@a83-163-41-120.adsl.xs4all.nl] has joined #lisp 02:05:11 dulouz: ok, I'll paste you an example 02:05:13 http://common-lisp.net/project/slime/doc/html/Setting-up-pathname-translations.html was not working for me 02:06:30 tritchey [n=tritchey@c-68-58-88-241.hsd1.in.comcast.net] has joined #lisp 02:06:56 well, now i get a different error on that, so I must have changed my .emacs in the last 30 minutes in just the right way to make some progress 02:07:27 legumbre` [n=user@r190-135-24-68.dialup.adsl.anteldata.net.uy] has joined #lisp 02:07:40 rstandy pasted "slime-filname-translations example" at http://paste.lisp.org/display/79115 02:07:55 rpg [n=rpg@216.243.156.16.real-time.com] has joined #lisp 02:08:59 I'll give that a whirl 02:08:59 dulouz: let me know if it works for you 02:09:31 Quick question (I hope) on my mac, sbcl guesses the architecture is x86. Should I tell it it's really x86-64, or just stick with the 32-bit build? 02:10:01 ... Wait, we still have slime-filename-translations? 02:10:03 maybe your mac is really x86 and sbcl uncovered the lie? 02:10:18 I thought that got ditched at some point? 02:10:26 S11001001: I'll just have to go tell those BASTARDS at the mac store! ;-) 02:10:31 jfactor [n=jfactor@student166-252.hampshire.edu] has joined #lisp 02:10:38 Seriously, I have a reason to prefer 64 bit if I can get it. 02:10:48 rpg: SBCL_ARCH=x86-64 then 02:11:13 pkhuong: thanks. I figured that was right, but wasn't sure if there was some pitfall I didn't know about. 02:11:14 nyef: why should we don't? :-) 02:12:01 I think the port is slightly more suspect still ;) I try to avoid looking at the runtime. 02:12:21 _3b: the use case I'm thinking of is a display server, where they'd be separate processes (rather than threads). 02:12:29 rstandy: I don't know, but when I tried getting my original SBCL/Win32 filename translation thing for cygwin emacs running it didn't work, and I couldn't find evidence of the existence of slime-filename-translations and so went with the slime-{from,to}-lisp-filename-function pair instead. 02:13:21 <_3b> hefner: hmm, might still work, what platform? 02:13:33 linux 02:13:46 (probably varies wildly depending on drivers) 02:14:06 some day I'll catch up with all the exciting post-1998 developments in OpenGL :) 02:14:32 <_3b> heh 02:14:46 *_3b* needs to get cl-opengl caught up too one of these days 02:14:47 -!- kinnetica [n=kinnetic@208-58-64-106.c3-0.129-ubr4.lnh-129.md.cable.rcn.com] has quit [] 02:15:07 nyef: slime-{from,to}-lisp-filename-function use `slime-filename-translations' to setup filname translators 02:15:19 Hunh. 02:15:28 Clearly, I need to do more digging at some point. 02:15:41 Which is the preferred interface? 02:15:52 (And is there a contrib involved?) 02:15:53 -!- deech [n=deech@71-10-166-226.dhcp.stls.mo.charter.com] has quit [Read error: 104 (Connection reset by peer)] 02:16:12 _3b: just don't burn any bridges. OpenGL ES is probably the future, not this stillborn OpenGL 3 silliness. =p 02:16:13 deech [n=deech@71-10-166-226.dhcp.stls.mo.charter.com] has joined #lisp 02:16:14 nyef: I don't know, I found simpler to use directly the variable 02:16:51 (I guess ES is more like 3 than 1.1 or so anyway) 02:16:52 <_3b> hefner: ES would probably look more like the gl3 stuff than like the old GL :p 02:16:54 heh 02:17:08 <_3b> and the biggest problem is just figuring out a nice way to support both of them 02:17:29 <_3b> with a nice minimal interface for use with gl3, but keeping the old stuff available for people who need it 02:17:50 I'm biased, I live in linuxland, where the opengl drivers on my laptop still suck and resemble something from the late 80s 02:17:51 90s. 02:18:13 meh 02:18:15 doom 3 works 02:18:18 don't care otherwise 02:18:25 pkhuong: Sorry, one more. Any reason I should get an error about missing sbcl_arch directory? 02:18:26 <_3b> hefner: ok, looks like you would need to either be in 1 process, or '[both contexts] nondirect using the same server' 02:18:27 hraban_ [n=b4@a83-163-41-120.adsl.xs4all.nl] has joined #lisp 02:18:33 nyef: about the contrib: yes, you should load the slime-tramp contrib 02:18:33 -!- b4|hraban [n=b4@a83-163-41-120.adsl.xs4all.nl] has quit [Read error: 104 (Connection reset by peer)] 02:18:43 -!- hraban_ is now known as b4|hraban 02:18:46 _3b: lame! 02:19:09 So noted. Thank you. 02:19:23 x86-64, with a lowercase x and a hyphen? 02:19:36 <_3b> hefner: (and that is assuming FBO itself doesn't add limitations) 02:19:39 (Clearly, I need to update my article on SBCL/Win32 and SLIME... and my .emacs.) 02:19:49 pkhuong: You are right and I am an idiot. Underscore :-( 02:20:01 more reasons it should have stayed amd64 02:20:14 S11001001: incf :) 02:20:18 ah, yeah I forgot i uninstalled slime and everything on my remote server... this may be a while before i actually get to really try that, rstandy. 02:20:59 dulouz: take your time ;-) 02:21:02 For some reason that underscore sticks in my head -- there must have been some linux building I did in the old day where I used the underscore. 02:21:15 I'm going to sleep, good night dear lispers 02:21:25 rpg: I think all the docs and most linuxes use x86_64. 02:21:28 later, thanks for your help 02:21:46 dulouz: you're welcome 02:22:15 ace4016 [i=ace4016@cpe-76-168-248-118.socal.res.rr.com] has joined #lisp 02:23:05 <_3b> hefner: wouldn't be surprised if there is some way to work around that limitation though, since the fancy window managers probably need to do something similar 02:23:38 _3b: hmm, good point. 02:24:25 <_3b> been intending to look into that sort of thing at some point, would simplify things if i could grab a running emacs frame as a texture in a full screen GL app :) 02:24:30 -!- legumbre [n=user@r190-135-62-34.dialup.adsl.anteldata.net.uy] has quit [Connection timed out] 02:24:47 -!- dialtone [n=dialtone@adsl-99-136-101-166.dsl.pltn13.sbcglobal.net] has quit [Read error: 110 (Connection timed out)] 02:24:48 -!- awayekos is now known as anekos 02:24:56 -!- b4|hraban [n=b4@a83-163-41-120.adsl.xs4all.nl] has quit ["Leaving"] 02:30:43 jbmigel [n=jbm@S010600179a220e57.cg.shawcable.net] has joined #lisp 02:34:14 gonzojive [n=red@fun.Stanford.EDU] has joined #lisp 02:36:00 -!- nyef [n=nyef@24.220.182.229] has quit ["G'night all."] 02:37:09 -!- durka42 [n=durka@d81.wireless.swarthmore.edu] has quit [] 02:38:02 -!- benny [n=benny@i577A1CD9.versanet.de] has quit [Read error: 113 (No route to host)] 02:41:00 -!- dinendalelanesse [n=dinendal@c-66-229-16-216.hsd1.fl.comcast.net] has quit [] 02:44:14 rme [n=rme@pool-70-105-122-215.chi.dsl-w.verizon.net] has joined #lisp 02:49:05 kreuter [n=kreuter@pool-96-252-14-107.bstnma.fios.verizon.net] has joined #lisp 02:49:09 -!- tomoyuki28jp [n=tomoyuki@w221062.ppp.asahi-net.or.jp] has quit [Remote closed the connection] 02:49:23 <_3b> kreuter: did you find a compilation from method call example yet? 02:49:38 <_3b> (if that is what you wanted) 02:49:54 _3b: not one that trips the bug I'm trying to reproduce. 02:50:04 have you got a suggestion? 02:50:16 <_3b> might try http://paste.lisp.org/display/46531 02:51:09 hm. interesting. thanks. 02:51:54 (trace compile :break t)? 02:52:00 f8f9k [i=orly@gateway/gpg-tor/key-0x5E4720FD] has joined #lisp 02:53:33 -!- huangjs [n=user@watchdog.msi.co.jp] has left #lisp 02:54:06 jdv79 [n=jdv79@u10570642.ul.warwick.net] has joined #lisp 02:54:44 anyone know of good examples of heavy uses of CLOS? 02:55:03 <_3b> mcclim? 02:55:07 i'm just curious to see if it was a "success"? 02:55:18 yes, definitely. 02:55:20 i admit i know very little about anything lisp 02:55:30 <_3b> it seems as successful as any OO 02:56:01 CLOS is used pervasively in Common Lisp code, and in varying degrees. 02:56:13 not just OO. i'm wondering if the MOP was as useful as the paper purportted. 02:56:21 (lately, I routinely write programs that use generic functions but don't define any classes) 02:56:38 -!- rpg [n=rpg@216.243.156.16.real-time.com] has quit [] 02:56:44 yes, i'm looking for the "most meta" type of CLOS usage 02:57:40 on the meta side, look at the libraries for persistent objects, like Elephant 02:57:41 <_3b> MOP probably isn't as common, since the built in stuff covers the common cases 02:57:54 S1100100` [n=sirian@74-137-151-39.dhcp.insightbb.com] has joined #lisp 02:58:12 -!- S11001001 [n=sirian@pdpc/supporter/active/S11001001] has quit [Remote closed the connection] 02:58:24 the MOP is also pretty well suited to constructing custom object systems, which some people do. 03:00:02 ah 03:00:39 there was a presentation at this year's ILC about a security-oriented object system, which bottomed out at the MOP. 03:02:06 spradnyesh [n=pradyus@nat/yahoo/x-3bbe956b9492d367] has joined #lisp 03:02:08 -!- manuel_ [n=manuel@HSI-KBW-078-043-184-124.hsi4.kabel-badenwuerttemberg.de] has quit [] 03:02:41 kreuter: a link? 03:02:46 -!- spradnyesh [n=pradyus@nat/yahoo/x-3bbe956b9492d367] has left #lisp 03:02:53 i'm been told ContextL could be a cool MOP usage 03:02:54 hm 03:04:19 jdv79: I don't know if the proceedings are public right now. the talk was by Howard Shrobe, titled "Towards a Secure Programming Lanuage: An Access Control System for Common Lisp" 03:04:46 thanks 03:06:04 -!- jbmigel [n=jbm@S010600179a220e57.cg.shawcable.net] has quit ["Ex-Chat"] 03:07:47 SirNick [n=SirNick@mail.colortechnology.com] has joined #lisp 03:14:18 -!- gigamonkey [n=user@adsl-99-35-219-234.dsl.pltn13.sbcglobal.net] has quit [Read error: 54 (Connection reset by peer)] 03:15:59 manic12 [n=manic12@c-76-29-88-103.hsd1.il.comcast.net] has joined #lisp 03:23:13 durka42 [n=durka@d161.mertza.swarthmore.edu] has joined #lisp 03:25:15 Pegazus [n=awefawe@host194.190-136-98.telecom.net.ar] has joined #lisp 03:28:01 -!- jmbr [n=jmbr@87.223.68.107] has quit [Read error: 110 (Connection timed out)] 03:30:34 aquagnu [n=aquagnu@85.118.228.172] has joined #lisp 03:40:17 eno__ [n=eno@adsl-70-137-155-169.dsl.snfc21.sbcglobal.net] has joined #lisp 03:41:03 -!- JAS415 [n=jon@c-24-34-16-25.hsd1.ma.comcast.net] has left #lisp 03:42:31 -!- dreish [n=dreish@minus.dreish.org] has quit [] 03:44:34 -!- eno [n=eno@nslu2-linux/eno] has quit [Read error: 104 (Connection reset by peer)] 03:46:13 tomoyuki28jp [n=tomoyuki@EM114-51-191-119.pool.e-mobile.ne.jp] has joined #lisp 03:48:32 I am editing remote files from my local emacs with connecting to the remote swank server. I can evaluate each sexp with C-x C-e, but I cannot compile the file with C-c C-k. How can I compile a remote file? 03:48:35 weee! i think i have filename translation working for editing remote files. 03:48:51 wow, talk about coincidence. lol 03:49:13 dulouz: Are we talking about the same thing at the same time!? 03:49:20 hah, pretty close atleast :D 03:49:33 wow :D 03:50:14 let me see if compiling works for m 03:50:16 me* 03:50:23 dulouz: thanks! 03:51:30 -!- tritchey [n=tritchey@c-68-58-88-241.hsd1.in.comcast.net] has quit [] 03:51:46 yep, i can 03:52:15 oh really? 03:52:21 let me paste the error message 03:53:32 the error message is this: failed to find the TRUENAME of /ssh:remote-server.com:/usr/lib/sbcl/site/package/file.lisp: 03:53:53 I use tramp 03:54:01 I do as well 03:54:05 oh really 03:54:07 strange. 03:54:18 yeah, i just got it working tonight for the first time 03:54:24 did you happen to set up remote filename translations? if i tried even editing a remote file, it would say something like "no translation found for hostname" 03:55:24 fielname translation? you mean complement? 03:57:46 someone in here earlier told me to check out C-h v slime-filename-translations 03:58:18 so i did, adding that to my .emacs when slime loads, and then it seems to have worked 03:58:42 oh really. let me try that one. thanks a lot for the info. 04:00:16 -!- dysinger [n=tim@cpe-75-80-200-182.hawaii.res.rr.com] has quit [] 04:00:51 that was after making sure slime-tramp was loaded, otherwise, that variable will be undefined 04:01:10 this one, right? http://sage.math.washington.edu/home/palmieri/emacs/slime/contrib/slime-tramp.el 04:01:51 *dulouz* ponders linux as Open Office launched on that link 04:02:15 yeah, that looks about right 04:03:56 It seems like it comes with slime under the contrib dir. 04:04:59 yeah 04:05:01 it does 04:05:52 a little searching last night, i found http://bc.tech.coop/blog/070927.html which talks nicely about contrib 04:06:41 which was new to me, i had been using slime from linux packages instead of from cvs. so it was all a black box 04:06:50 -!- f8f9k [i=orly@gateway/gpg-tor/key-0x5E4720FD] has left #lisp 04:07:17 -!- S1100100` is now known as S11001001 04:07:39 brb 04:07:43 -!- tomoyuki28jp [n=tomoyuki@EM114-51-191-119.pool.e-mobile.ne.jp] has quit [Remote closed the connection] 04:10:09 tomoyuki28jp [n=tomoyuki@EM114-51-191-119.pool.e-mobile.ne.jp] has joined #lisp 04:11:31 chavo_ [n=user@c-66-41-11-10.hsd1.mn.comcast.net] has joined #lisp 04:12:41 -!- Pegazus [n=awefawe@host194.190-136-98.telecom.net.ar] has left #lisp 04:12:53 dulouz: Which argument did you pass for slime-filename-translations? 04:13:45 i'll show past my updated .emacs file... it might not be the *right* place to put it all, but it seems to be working. 04:14:21 dulouz: thanks for your help, I really appreciate it. 04:14:56 dulouz pasted "updated .emacs with filename translation" at http://paste.lisp.org/display/79119 04:16:07 so that "push (list ... " part in eval-after-load 04:16:45 dulouz: yeah, thanks a lot. let me try that one. 04:19:43 i need to sleep, good luck! i think i've told all i know on this 04:20:01 dulouz: thanks a lot for your help, I really appreciate it. 04:20:05 do you often come here? 04:20:56 tomo: my pleasure. i come here in phases, not super regular. 04:21:16 I see. I am looking forward to see you next time. 04:21:19 have a good nite 04:21:30 same to you 04:21:38 later! good nite :) 04:21:39 -!- dulouz [n=ross@dsl254-119-219.nyc1.dsl.speakeasy.net] has left #lisp 04:22:32 <_3b> yay for overly tolerant parsers :/ 04:22:38 -!- davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 04:23:15 <_3b> finally figured out why my flash files weren't loopingg, i was putting in incorrect value for the size fields 04:23:38 <_3b> neither flash player, the flex decompiler, or my parser cared, aside from flash player not looping 04:24:01 yay for silent failure 04:24:08 <_3b> that too 04:25:38 <_3b> loopy mario : http://www.3bb.cc/tmp/svg-test-mario3.swf 04:26:13 -!- legumbre` is now known as legumbre 04:26:33 straight gangsta' ! 04:27:13 it's like compiz in my web browser 04:27:20 <_3b> heh 04:28:07 spradnyesh [n=pradyus@nat/yahoo/x-03f3f23abd7fb0ec] has joined #lisp 04:28:35 -!- tomoyuki28jp [n=tomoyuki@EM114-51-191-119.pool.e-mobile.ne.jp] has quit [Remote closed the connection] 04:29:13 <_3b> hopefully this time i can manage to get to gradients without trying out something else that uncovers a random hard to find bug 04:30:04 -!- spradnyesh [n=pradyus@nat/yahoo/x-03f3f23abd7fb0ec] has left #lisp 04:35:06 -!- SirNick [n=SirNick@mail.colortechnology.com] has quit [] 04:42:55 -!- existentialmonk [n=carcdr@64-252-67-89.adsl.snet.net] has quit [Remote closed the connection] 04:44:20 amnesiac [n=amnesiac@p3m/member/Amnesiac] has joined #lisp 04:44:41 -!- eno__ is now known as eno 04:50:54 thom_logn [n=thom@pool-96-229-99-100.lsanca.dsl-w.verizon.net] has joined #lisp 04:54:17 huangjs [n=user@watchdog.msi.co.jp] has joined #lisp 04:56:35 -!- chavo_ [n=user@c-66-41-11-10.hsd1.mn.comcast.net] has quit [Remote closed the connection] 05:01:10 -!- gonzojive [n=red@fun.Stanford.EDU] has quit [] 05:03:21 SirNick [n=SirNick@c-24-20-213-118.hsd1.wa.comcast.net] has joined #lisp 05:09:07 fusss [n=chatzill@pool-70-108-97-40.res.east.verizon.net] has joined #lisp 05:09:14 lnostdal: herep 05:09:37 fusss, t 05:10:04 can I message you? 05:10:23 yeah 05:12:00 mrsolo [n=mrsolo@adsl-68-126-205-8.dsl.pltn13.pacbell.net] has joined #lisp 05:14:17 -!- doxtor [n=doxtor@unaffiliated/mitja] has quit [Read error: 110 (Connection timed out)] 05:16:15 is it rude, wrong or offensive to post job ads here? for an advertising startup. 05:16:18 Good morning. 05:16:28 hey beach 05:22:42 -!- huangjs [n=user@watchdog.msi.co.jp] has quit [Remote closed the connection] 05:22:50 *fusss* is juggling the conflicting feelings of a minor business success with the upset of death in the family 05:23:37 -!- rme [n=rme@pool-70-105-122-215.chi.dsl-w.verizon.net] has left #lisp 05:32:51 xan-afk_ [n=xan@cs78225040.pp.htv.fi] has joined #lisp 05:33:20 fusss: I think it is allowed if it fits in the realm of Lisp discussion 05:35:25 i need a few people to come on board with me on a startup. deferred pay and equity, but with a monthly commission. i really feel dirty for saying that to your faces, #lispers, sorry. bigthingist@gmail.com for more info. 05:36:33 not just employees and partners, i need mentors. people who are willing to subscribe to our private mailing list and see how our company actually operates. 05:38:09 i'm probably just panicking for no reason, but in less than 12 days i will have to serve 20k hits a day off of the stuff you all know. hunchentoot, sbcl, postfix and few other glue thingies. 05:38:16 fusss: you should also see http://lispjobs.wordpress.com/about/ 05:38:59 -!- xan-afk_ is now known as xan 05:39:54 i know, but there are lispers and there are #lispers. the people here have proven themselves to me to be both competent, accessible and courteous. i just don't wanna go public with the ANN until i have someone else to talk to about what i'm doing. 05:40:06 Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has joined #lisp 05:41:50 does anyone know what "adversarial web serving" is? it's a phrase I coined which is mildly related to adversarial information retrieval. 05:42:45 brb 05:43:22 if you coined the phrase how could we know ;P 05:43:55 -!- xan-afk [n=xan@cs78225040.pp.htv.fi] has quit [Read error: 110 (Connection timed out)] 05:44:44 it's basically configuring each page to the user's historic behaviour profile. except in this case it's targeted at marketing decision makers. 05:45:13 you slip an exec a link to a landing page, and from then on every page he sees will be different than what normal viewers would see. 05:45:20 -!- tsuru [n=user@c-69-245-36-64.hsd1.tn.comcast.net] has quit [Read error: 110 (Connection timed out)] 05:45:23 like i said, cutting edge stuff 05:45:25 and evil. 05:46:28 ehu` [n=chatzill@82-170-33-173.ip.telfort.nl] has joined #lisp 05:46:39 behaviour targeting + semantic latent analysis + information retrieval 05:46:44 brb, cigarette 05:54:36 -!- SirNick [n=SirNick@c-24-20-213-118.hsd1.wa.comcast.net] has quit [] 06:01:45 sunwukong [n=vukung@ortros.den.rcast.u-tokyo.ac.jp] has joined #lisp 06:08:15 nvoorhies [n=nvoorhie@adsl-75-36-207-191.dsl.pltn13.sbcglobal.net] has joined #lisp 06:12:54 -!- mrsolo [n=mrsolo@adsl-68-126-205-8.dsl.pltn13.pacbell.net] has quit [Read error: 110 (Connection timed out)] 06:14:50 drwhen [n=d@216-67-73-247-rb1.fai.dsl.dynamic.acsalaska.net] has joined #lisp 06:18:24 -!- mokogobo [n=mokogobo@pcp075595pcs.unl.edu] has quit [Read error: 110 (Connection timed out)] 06:18:32 jao [n=jao@132.Red-83-42-111.dynamicIP.rima-tde.net] has joined #lisp 06:23:00 bombshelter13_ [n=bombshel@209-161-233-254.dsl.look.ca] has joined #lisp 06:26:07 joshe` [n=aurum@opal.elsasser.org] has joined #lisp 06:29:23 -!- amnesiac [n=amnesiac@p3m/member/Amnesiac] has quit ["Leaving"] 06:30:31 -!- joshe [n=aurum@opal.elsasser.org] has quit [Read error: 104 (Connection reset by peer)] 06:30:38 -!- slyrus [n=slyrus@adsl-68-121-172-169.dsl.pltn13.pacbell.net] has quit [Remote closed the connection] 06:31:46 -!- ehu` [n=chatzill@82-170-33-173.ip.telfort.nl] has quit [Read error: 110 (Connection timed out)] 06:37:05 dwave [n=ask@pat-tdc.opera.com] has joined #lisp 06:37:45 Jabberwockey [n=Tumnus_@port-5114.pppoe.wtnet.de] has joined #lisp 06:39:45 -!- bombshelter13_ [n=bombshel@209-161-233-254.dsl.look.ca] has quit [Client Quit] 06:39:59 mvilleneuve [n=mvillene@ABordeaux-253-1-126-133.w86-201.abo.wanadoo.fr] has joined #lisp 06:40:31 good morning 06:40:32 hello mvilleneuve 06:42:14 mrsolo [n=mrsolo@adsl-68-126-205-8.dsl.pltn13.pacbell.net] has joined #lisp 06:46:03 -!- cracki [n=cracki@sglty.kawo2.RWTH-Aachen.DE] has quit ["The funniest things in my life are truth and absurdity."] 06:49:39 -!- ikki [n=ikki@201.155.75.146] has quit ["Leaving"] 06:51:39 spradnyesh [n=pradyus@nat/yahoo/x-0b0054b770fe0a16] has joined #lisp 06:52:24 tcr [n=tcr@host145.natpool.mwn.de] has joined #lisp 06:53:31 alinp [n=alinp@86.122.9.2] has joined #lisp 06:54:49 good morning 06:58:57 -!- rread [n=rread@nat/sun/x-fd6b1435b71319d5] has quit [Read error: 110 (Connection timed out)] 07:00:40 gemelen [n=shelta@shpd-78-36-166-34.static.vologda.ru] has joined #lisp 07:01:19 -!- spradnyesh [n=pradyus@nat/yahoo/x-0b0054b770fe0a16] has left #lisp 07:01:41 return is a loop keyword outside finally still, right? 07:01:59 all I seem to be able to internalize is that it was once allowed in finally, but was confusing so removed 07:02:02 S11001001: not anymore. 07:02:13 It was in cltl1, but not in ANSI CL. 07:02:18 MrSpec [n=NoOne@82.177.125.6] has joined #lisp 07:02:27 meh 07:02:35 I usually write do (return anyway 07:02:44 (loop :finally (return result)) (loop :named x :finally (return-from x result)) 07:02:52 mega1 [n=mega@53d8275f.adsl.enternet.hu] has joined #lisp 07:04:01 <_3b> (loop return 1) looks OK if that is what you were asking though 07:10:02 blandest [n=blandest@softhouse.is.ew.ro] has joined #lisp 07:11:00 pjb: yeah, I meant outside finally 07:11:13 as _3b 07:11:30 rread [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has joined #lisp 07:11:34 envi^home [n=envi@220.121.234.156] has joined #lisp 07:16:08 rread_ [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has joined #lisp 07:16:46 daniel_ [i=daniel@unaffiliated/daniel] has joined #lisp 07:17:30 rread__ [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has joined #lisp 07:17:30 -!- rread_ [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 07:18:04 ejs [n=eugen@77.222.151.102] has joined #lisp 07:18:05 -!- rread__ [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has quit [Read error: 54 (Connection reset by peer)] 07:18:18 rread_ [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has joined #lisp 07:19:21 -!- mrsolo [n=mrsolo@adsl-68-126-205-8.dsl.pltn13.pacbell.net] has quit [Read error: 60 (Operation timed out)] 07:23:29 Hello! 07:25:29 -!- sunwukong [n=vukung@ortros.den.rcast.u-tokyo.ac.jp] has quit ["bye"] 07:26:46 -!- aja [n=aja@unaffiliated/aja] has quit [Client Quit] 07:29:26 kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 07:29:38 mrsolo [n=mrsolo@adsl-68-126-205-8.dsl.pltn13.pacbell.net] has joined #lisp 07:31:13 l_a_m [n=nlamirau@194.51.71.190] has joined #lisp 07:32:07 -!- daniel [i=daniel@unaffiliated/daniel] has quit [Read error: 110 (Connection timed out)] 07:32:09 -!- rread [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has quit [Read error: 113 (No route to host)] 07:34:07 -!- stassats` [n=stassats@wikipedia/stassats] has quit [Read error: 110 (Connection timed out)] 07:38:08 Ragnaroek [i=54a67a2c@gateway/web/ajax/mibbit.com/x-f7158dd7b12cc62e] has joined #lisp 07:40:28 stassats [n=stassats@wikipedia/stassats] has joined #lisp 07:43:13 Nshag [i=user@Mix-Orleans-106-2-151.w193-248.abo.wanadoo.fr] has joined #lisp 07:43:29 jdz [n=jdz@85.254.193.149] has joined #lisp 07:44:07 jdz_ [n=jdz@85.254.211.133] has joined #lisp 07:48:14 ejs1 [n=eugen@nat.ironport.com] has joined #lisp 07:49:30 -!- kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 07:54:30 Soulman [n=kvirc@42.84-48-88.nextgentel.com] has joined #lisp 07:56:56 -!- ejs [n=eugen@77.222.151.102] has quit [Read error: 110 (Connection timed out)] 07:57:56 -!- sohail [n=Sohail@unaffiliated/sohail] has quit [Read error: 113 (No route to host)] 08:00:38 -!- jdz [n=jdz@85.254.193.149] has quit [Read error: 110 (Connection timed out)] 08:00:48 bobf [n=bob@unaffiliated/bob-f/x-6028553] has joined #lisp 08:04:50 -!- alinp [n=alinp@86.122.9.2] has quit ["Leaving."] 08:05:04 alinp [n=alinp@86.122.9.2] has joined #lisp 08:05:43 Hun [n=Hun@dial18-107.RZ.FH-Augsburg.DE] has joined #lisp 08:06:22 -!- MrSpec [n=NoOne@82.177.125.6] has quit [Remote closed the connection] 08:07:45 hyperboreean [n=none@unaffiliated/hyperboreean] has joined #lisp 08:08:35 ejs2 [n=eugen@nat.ironport.com] has joined #lisp 08:09:26 kib2 [n=kib2@bd137-1-82-228-159-28.fbx.proxad.net] has joined #lisp 08:15:30 -!- blandest [n=blandest@softhouse.is.ew.ro] has quit ["Leaving."] 08:15:53 blandest [n=blandest@softhouse.is.ew.ro] has joined #lisp 08:17:03 cracki [n=cracki@43-029.eduroam.RWTH-Aachen.DE] has joined #lisp 08:17:33 -!- blandest [n=blandest@softhouse.is.ew.ro] has quit [Read error: 104 (Connection reset by peer)] 08:19:23 -!- ejs1 [n=eugen@nat.ironport.com] has quit [Read error: 110 (Connection timed out)] 08:21:55 aerique [i=euqirea@xs2.xs4all.nl] has joined #lisp 08:22:49 blandest [n=blandest@softhouse.is.ew.ro] has joined #lisp 08:24:41 -!- l_a_m [n=nlamirau@194.51.71.190] has quit ["Lost terminal"] 08:27:57 -!- jao [n=jao@132.Red-83-42-111.dynamicIP.rima-tde.net] has quit [Read error: 110 (Connection timed out)] 08:29:43 -!- mrsolo [n=mrsolo@adsl-68-126-205-8.dsl.pltn13.pacbell.net] has quit [Read error: 60 (Operation timed out)] 08:31:29 abeaumont [n=abeaumon@84.76.48.250] has joined #lisp 08:32:04 kreuter: are you satisfied with the latest patch in the run-program thread? 08:38:12 -!- Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has quit [Read error: 113 (No route to host)] 08:41:58 oudeis [n=oudeis@80.250.159.240] has joined #lisp 08:44:10 HG` [n=wells@91.111.13.16] has joined #lisp 08:47:29 Davidbrcz [n=david@ANantes-151-1-156-148.w86-199.abo.wanadoo.fr] has joined #lisp 08:48:53 manby-ace [n=manby-ac@88-96-24-54.dsl.zen.co.uk] has joined #lisp 08:49:43 pkhuong: around? 08:49:46 loxs [n=loxs@83.228.122.198] has joined #lisp 08:51:38 -!- Davidbrcz [n=david@ANantes-151-1-156-148.w86-199.abo.wanadoo.fr] has quit [Client Quit] 09:00:26 mega1 annotated #79013 "comment" at http://paste.lisp.org/display/79013#1 09:00:44 Athas [n=athas@192.38.109.188] has joined #lisp 09:02:24 fiveop [n=fiveop@pD9E6C1DA.dip.t-dialin.net] has joined #lisp 09:05:39 hkBst [n=hkBst@253pc223.sshunet.nl] has joined #lisp 09:06:03 kami- pasted "compile-file warned" at http://paste.lisp.org/display/79125 09:06:44 how can I get rid of this. 09:06:46 ? 09:06:52 -!- hkBst [n=hkBst@gentoo/developer/hkbst] has quit [Client Quit] 09:07:17 I mean: how can I avoid the debugger. Do I have to write a handler around the loading code? 09:07:42 hkBst [n=hkBst@gentoo/developer/hkbst] has joined #lisp 09:07:45 Or would a change to closer-mop be more appropriate? 09:08:02 huh, did you load closer-mop before? 09:08:29 antifuchs: you mean the reason is my loading c-mop twice? 09:08:32 that sounds like the error you get if something makes a package, exports symbols from it, then defines that package again. 09:08:45 antifuchs: thank you. will check for that. 09:08:53 ok (: 09:09:52 normally, this kind of warning can be skipped by using the ACCEPT restart, if it happens in development. at least on sbcl, it's not really serious, but something to keep in mind before shipping. 09:10:06 antifuchs: I don't load it twice, explicitly. It is a defsystem with many :depends-on's 09:10:46 could be that this happens twice (although a system that has been loaded once should not get loaded again if it didn't change) 09:10:47 antifuchs: if closer-mop is needed transitively, should it be loaded twice? 09:11:09 maybe something loads another system that defines the same package, again? 09:11:15 -!- oudeis [n=oudeis@80.250.159.240] has quit [Read error: 110 (Connection timed out)] 09:11:22 the path says cl-dwim/closer-mop/pcl/closer-mop.fasl 09:11:36 is cl-dwim loading another closer-mop than the one that has been loaded before? 09:12:08 antifuchs: not possible. I have only one in the load-path (and start with --no-userinit) 09:12:31 antifuchs: or I just _assume_ this (and should check twice :) 09:15:15 sounds like a plan ((: 09:16:43 I get that all the time when trying to compile mcclim 09:17:02 as a result I haven't ever been able to compile mcclim smoothly 09:17:37 -!- fusss [n=chatzill@pool-70-108-97-40.res.east.verizon.net] has quit ["ChatZilla 0.9.84 [Firefox 3.0.9/2009040821]"] 09:18:37 well, you can set asdf:*compile-file-failure-behaviour* to :warn, instead of sbcl's default :error 09:18:37 09:18:39 kami- annotated #79125 "content of closer-mop-utility-packages.lisp" at http://paste.lisp.org/display/79125#1 09:18:56 benny` [n=benny@i577A16B1.versanet.de] has joined #lisp 09:19:05 stassats: thank you. this would already help (but maybe I would miss something else) 09:19:09 Fufie [n=poff@80.203.160.34] has joined #lisp 09:19:12 -!- benny` is now known as benny 09:19:37 does the programmatic re-export of the symbols leads to the problem? 09:19:46 not by itself 09:20:00 but defining the package, exporting the symbols, and then defining the package again does it 09:20:10 that's what in the vanilla asdf: (defvar *compile-file-failure-behaviour* #+sbcl :error #-sbcl :warn) 09:20:12 i wonder why 09:21:24 to adhere to sbcl's as many errors as possible policy? 09:22:08 antifuchs: ok. I will restart the image dump thing with tee and see whether it were really loaded twice 09:22:15 could be that other implementations don't take the "FAILURE" return value of c-f as seriously 09:23:02 kami-: I wonder if compiling, then loading the file could do it... but the re-exporting doesn't happen on the top level and doen't happen in an eval-when 09:23:03 so hm. 09:23:38 antifuchs: let me first retry and check the output 09:23:39 nope, it's not that. 09:23:44 ok 09:26:57 grc [n=user@217.33.170.226] has joined #lisp 09:27:25 athos [n=philipp@92.250.250.68] has joined #lisp 09:28:37 deliana [n=deliana@147.210.246.189] has joined #lisp 09:30:23 -!- Jabberwockey [n=Tumnus_@port-5114.pppoe.wtnet.de] has quit [Remote closed the connection] 09:32:39 -!- rstandy [n=rastandy@net-93-144-246-189.t2.dsl.vodafone.it] has quit [Remote closed the connection] 09:33:39 kami- annotated #79125 "loaded twice" at http://paste.lisp.org/display/79125#2 09:34:08 I now probably have to find out _why_ it gets loaded twice, right? 09:34:31 Odin- [n=sbkhh@s121-302.gardur.hi.is] has joined #lisp 09:35:18 jao [n=jao@74.Red-80-24-4.staticIP.rima-tde.net] has joined #lisp 09:38:30 morning 09:39:13 l_a_m [n=nlamirau@194.51.71.190] has joined #lisp 09:40:27 -!- thom_logn [n=thom@pool-96-229-99-100.lsanca.dsl-w.verizon.net] has quit ["Leaving"] 09:41:04 Davidbrcz [n=david@ANantes-151-1-156-148.w86-199.abo.wanadoo.fr] has joined #lisp 09:41:23 morning Xof 09:41:29 regarding sbcl release schedule, I have my repeatable stuff ready for merging, I think. I want to test every rebased revision, so it might take me most of the day to get it all in. Then freeze and release on Thursday? 09:43:56 -!- plutonas [n=plutonas@c-83-233-152-13.cust.bredband2.com] has quit [Remote closed the connection] 09:45:50 -!- Odin- [n=sbkhh@s121-302.gardur.hi.is] has quit [] 09:46:49 oudeis [n=oudeis@bzq-82-81-239-4.cablep.bezeqint.net] has joined #lisp 09:49:09 -!- envi^home [n=envi@220.121.234.156] has quit [Read error: 104 (Connection reset by peer)] 09:50:38 -!- oudeis [n=oudeis@bzq-82-81-239-4.cablep.bezeqint.net] has quit [Client Quit] 09:51:55 hypno [n=hypno@195.43.248.100] has joined #lisp 09:55:31 puchacz [n=puchacz@87.194.5.99] has joined #lisp 09:59:31 lichtblau: do you remember what your impression of the state of the work you did on the generic specializers protocol was? 09:59:40 -!- ejs2 [n=eugen@nat.ironport.com] has quit [Read error: 60 (Operation timed out)] 09:59:44 yvdriess [n=yvdriess@progpc35.vub.ac.be] has joined #lisp 09:59:49 I seem to remember a discussion between you and Pascal but not its conclusion 10:00:18 -!- yvdriess [n=yvdriess@progpc35.vub.ac.be] has quit [Client Quit] 10:02:31 ejs2 [n=eugen@77.222.151.102] has joined #lisp 10:03:25 I think his specializer-precedence-list was simpler than my specializer<, and someone (perhaps you) pointed out that some of my MOP methods were superfluous, demonstrating my lack of MOP experience, but that's about it. 10:04:07 Unless someone has actual code doing the same thing in a better way, I'd probably continue where I started (if I needed this sort of thing currently). 10:04:20 lichtblau: also, I don't know if you saw, but I just got a mail from a happy cxml-rpc user. he got 110 transactions / seconds, saturating his xml-rpc client. that's a testament to the excellent klacks parser, I'm sure (: 10:04:22 thanks. I'd quite like to use something like this in my ELS tutorial 10:05:04 -!- yango [n=yango@unaffiliated/yango] has quit [Remote closed the connection] 10:05:15 yango [n=yango@unaffiliated/yango] has joined #lisp 10:05:36 silenius [n=jl@yian-ho03.nir.cronon.net] has joined #lisp 10:07:20 antifuchs: couldn't rpc calls be done by sending strings to the repl too? 10:07:36 antifuchs: (besides interoperability issues) 10:08:03 using swank or some other protocol, sure 10:09:07 the xml-rpc advantage is that it's already adopted in so many places, and that its spec is mostly static nowadays. 10:09:47 yes, I was simlpy seeing if I was overlooking something important 10:10:37 Krystof [n=csr21@158.223.51.76] has joined #lisp 10:10:40 -!- jrockway [n=jrockway@stonepath.jrock.us] has quit [Remote closed the connection] 10:19:14 oudeis [n=oudeis@89-139-94-206.bb.netvision.net.il] has joined #lisp 10:19:42 -!- jdz_ [n=jdz@85.254.211.133] has quit [] 10:19:47 eirik [i=eirikald@tvilling.pvv.ntnu.no] has joined #lisp 10:21:00 G'day! 10:21:59 hi spiaggia 10:26:33 -!- Athas [n=athas@192.38.109.188] has quit [Remote closed the connection] 10:27:01 jrockway [n=jrockway@stonepath.jrock.us] has joined #lisp 10:36:38 -!- Hun [n=Hun@dial18-107.RZ.FH-Augsburg.DE] has quit [Remote closed the connection] 10:40:13 ejs1 [n=eugen@nat.ironport.com] has joined #lisp 10:41:08 -!- dv_ [n=dv@85-127-117-181.dynamic.xdsl-line.inode.at] has quit ["Verlassend"] 10:41:28 tombom [n=tombomp@wikipedia/Tombomp] has joined #lisp 10:41:58 ManateeLazyCat [n=user@119.127.188.203] has joined #lisp 10:45:04 -!- ejs2 [n=eugen@77.222.151.102] has quit [Connection timed out] 10:46:15 -!- aquagnu [n=aquagnu@85.118.228.172] has quit [Remote closed the connection] 10:55:26 jdz [n=jdz@85.254.211.133] has joined #lisp 10:57:23 ejs [n=eugen@77.222.151.102] has joined #lisp 10:57:28 -!- ejs1 [n=eugen@nat.ironport.com] has quit ["Leaving"] 10:58:59 jewel [n=jewel@dsl-242-183-106.telkomadsl.co.za] has joined #lisp 11:02:55 -!- holycow [n=rtaylor@S01060016b6b53675.vf.shawcable.net] has quit [Remote closed the connection] 11:03:23 ThomasI [n=thomas@unaffiliated/thomasi] has joined #lisp 11:04:30 yvdriess [n=yvdriess@progpc35.vub.ac.be] has joined #lisp 11:08:13 iListenU [n=name@78-57-141-87.static.zebra.lt] has joined #lisp 11:08:33 -!- iListenU [n=name@78-57-141-87.static.zebra.lt] has quit [Excess Flood] 11:08:46 iListenU [n=name@78.57.141.87] has joined #lisp 11:09:09 -!- iListenU [n=name@78.57.141.87] has quit [Excess Flood] 11:10:07 kejsaren_ [n=kejsaren@p41-n239.kthopen.kth.se] has joined #lisp 11:15:53 -!- yvdriess [n=yvdriess@progpc35.vub.ac.be] has quit [] 11:15:57 -!- deech [n=deech@71-10-166-226.dhcp.stls.mo.charter.com] has quit [Read error: 54 (Connection reset by peer)] 11:16:43 deech [n=deech@71-10-166-226.dhcp.stls.mo.charter.com] has joined #lisp 11:17:30 sepult [n=buggarag@xdsl-87-78-75-195.netcologne.de] has joined #lisp 11:18:32 -!- Davidbrcz [n=david@ANantes-151-1-156-148.w86-199.abo.wanadoo.fr] has quit [Remote closed the connection] 11:23:13 -!- kpreid [n=kpreid@216-171-189-59.northland.net] has quit [Read error: 110 (Connection timed out)] 11:23:48 plutonas [n=plutonas@nomad51162.netlogon.lu.se] has joined #lisp 11:25:42 -!- cracki [n=cracki@43-029.eduroam.RWTH-Aachen.DE] has quit ["The funniest things in my life are truth and absurdity."] 11:25:50 Yuuhi [i=benni@p5483EAC3.dip.t-dialin.net] has joined #lisp 11:26:21 somewhat offtopic, but is there a format of that mop book that could be suitable to read on an ebook reader? 11:26:55 ignas [n=ignas@ctv-79-132-160-221.vinita.lt] has joined #lisp 11:27:48 you mean AMOP? 11:28:12 yeshzors 11:28:13 madnificent: yes 11:28:30 oudeis_ [n=oudeis@89-139-94-206.bb.netvision.net.il] has joined #lisp 11:28:39 (for which google doesn't give me a good link) :( 11:28:43 sepult: ah, where? 11:29:37 I'm not finding it on the ebook selling sites I know 11:30:30 ah, it doesn't really matter where it comes from. It reeds pdfs etc, I just can't find them 11:30:34 you could try to check the legally questionable sites for it 11:30:45 isn't the book freely available? 11:30:51 never heard of a PDF. Google books has it, if that ebook reader of yours can render that. 11:30:51 no 11:30:54 chapters are... 11:30:55 and thus valid to copy on the ebook reader... 11:31:08 oh 11:31:17 if it's available in html, you can convert it with Stanza and other applications. 11:31:42 http://www.lisp.org/mop/contents.html << this is the book, right? 11:31:45 also, most ebook formats are just packed html with index pages, anyway 11:31:49 A couple of pages are missing on google books, but only a few IIANM. 11:32:35 sigh, sorry /me is a bit ill 11:33:12 madnificent: that's the mop specification, section II of amop 11:33:20 apparantly 11:33:31 and amop is the advised book to read, right? 11:33:46 it's certainly worth reading 11:35:02 -!- oudeis [n=oudeis@89-139-94-206.bb.netvision.net.il] has quit [No route to host] 11:35:11 -!- ManateeLazyCat [n=user@119.127.188.203] has quit [Success] 11:37:01 -!- addled [n=adlirc@mail.andrewlawson.org] has quit [Remote closed the connection] 11:37:27 Odin- [n=sbkhh@s121-302.gardur.hi.is] has joined #lisp 11:37:35 -!- nvoorhies [n=nvoorhie@adsl-75-36-207-191.dsl.pltn13.sbcglobal.net] has quit [] 11:38:47 and reading the specification probably wouldn't get me on the same level, right? 11:39:13 envi^home [n=envi@220.121.234.156] has joined #lisp 11:40:31 it depends if you're already comfortable with metacircularity 11:40:39 HET2 [i=diman@xover.htu.tuwien.ac.at] has joined #lisp 11:40:44 and perhaps also CLOS implementation in general 11:41:08 if you are, then the other chapters are mostly redundant or confusing; if you aren't, then you'll probably need them to get anything out of the spec 11:41:29 Xof: doesn't scare me if it's the same thing smalltalk does (but perhaps it doesn't and I can't say that I really see any problems with it, so maybe I simply don't grasp it) 11:41:59 I haven't seen anything on the implementation of CLOS though 11:43:11 dash_ [n=dash@88.67.9.156] has joined #lisp 11:44:38 -!- dash_ is now known as ztzg 11:44:50 tsuru [n=user@c-69-245-36-64.hsd1.tn.comcast.net] has joined #lisp 11:45:09 madnificent: there is only one pdf which holds only the chapters 5 and 6 which is available of the mop 11:45:22 -!- loxs [n=loxs@83.228.122.198] has quit ["Leaving"] 11:45:22 madnificent: on palo alto research centers home pages 11:45:26 sepult: ty 11:46:28 aaaaaaaaarg what's wrong with google today? It hasn't given me a sigle good link 11:46:59 madnificent: http://books.google.de/books?id=3X5Gnudn3k0C&printsec=frontcover&dq=arisia+xerox+mop&source=gbs_summary_r&cad=0#PPA318,M1 11:47:10 adlirc [n=adlirc@mail.andrewlawson.org] has joined #lisp 11:47:12 -!- adlirc is now known as addled 11:47:22 madnificent: this site shows you a pdf version of it, but nonethless it's not free 11:47:37 madnificent: i bought the print version 11:48:00 lyte [n=lyte@unaffiliated/neerolyte] has joined #lisp 11:48:31 sepult: I'm going to bookmark that. I'm a student, so money is (sadly) an issue. I'll try to read the specs first, if that fails the sponsoring agency may drop in :D 11:49:25 madnificent: it's not that it gives the full spec of the mop you'd have to get the amop specs for that i think 11:49:49 madnificent: it's just an introduction to the mop 11:50:03 madnificent: with closette a mini clos 11:50:20 -!- ztzg [n=dash@88.67.9.156] has quit ["leaving"] 11:50:25 http://www.lisp.org/mop/index.html < that is the complete spec right? and isn't that included in the book? (as the page I refer to says) 11:50:27 madnificent: every lisp has slight implementation differences 11:50:34 ztzg [n=dash@dslb-088-067-009-156.pools.arcor-ip.net] has joined #lisp 11:50:47 ah yes, unspecified behavior may differ :) 11:50:50 madnificent: yes it is included in the book, chapters 5 and 6 11:51:07 ah, so if I'd grasp that, then I'm safe 11:51:25 -!- blandest [n=blandest@softhouse.is.ew.ro] has quit ["Leaving."] 11:51:38 blandest [n=blandest@softhouse.is.ew.ro] has joined #lisp 11:51:44 -!- ztzg [n=dash@dslb-088-067-009-156.pools.arcor-ip.net] has quit [Client Quit] 11:52:11 ztzg [n=dash@dslb-088-067-009-156.pools.arcor-ip.net] has joined #lisp 11:52:25 Xof: nice work with those patches 11:52:56 oh, you haven't seen the good ones yet :-) 11:54:03 -!- kejsaren_ [n=kejsaren@p41-n239.kthopen.kth.se] has quit [Read error: 101 (Network is unreachable)] 11:54:48 how's sbcl doing on the ansi tests these days? 11:55:03 beach` [n=user@ABordeaux-158-1-56-193.w90-16.abo.wanadoo.fr] has joined #lisp 11:55:44 ~35 tests failing iirc 11:56:20 Spyderco [n=nash@194.45.110.65] has joined #lisp 11:56:45 -!- Ragnaroek [i=54a67a2c@gateway/web/ajax/mibbit.com/x-f7158dd7b12cc62e] has quit ["http://www.mibbit.com ajax IRC Client"] 11:57:28 is the pretty printer still the main offender? 11:57:43 I think so 11:58:10 -!- beach` [n=user@ABordeaux-158-1-56-193.w90-16.abo.wanadoo.fr] has quit [Read error: 60 (Operation timed out)] 11:58:15 beach` [n=user@86.201.109.97] has joined #lisp 11:59:10 plage [n=user@serveur5.labri.fr] has joined #lisp 11:59:14 Good afternoon. 12:00:03 sonja keenes clos book is available 12:00:26 -!- kreuter [n=kreuter@pool-96-252-14-107.bstnma.fios.verizon.net] has quit [Read error: 54 (Connection reset by peer)] 12:00:50 -!- Spyderco [n=nash@194.45.110.65] has quit [Read error: 54 (Connection reset by peer)] 12:00:52 -!- beach [n=user@ABordeaux-158-1-118-26.w90-60.abo.wanadoo.fr] has quit [Read error: 60 (Operation timed out)] 12:00:52 it emphasizes some things you won't find somewhere else i think 12:01:05 kreuter [n=kreuter@pool-96-252-14-107.bstnma.fios.verizon.net] has joined #lisp 12:01:14 buuuut i found the examples to be a little bit complex 12:01:43 up until page 150 or so 12:02:48 cause locking is not necceseraily something one can grasp without computer knowledge beforehand 12:03:45 -!- sledge [n=chris@pdpc/supporter/student/sledge] has quit [Read error: 60 (Operation timed out)] 12:04:50 good afternoon beach` (guess I should eat now) 12:05:15 madnificent: That happens at home when I get disconnected by the ISP 12:05:36 sepult: i am quite of the opposite opinion. Keenes book is so far the best lisp book i've read, and the examples (but more importantly the design and thought process) is very explicit. 12:05:51 -!- Jasko2 [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has quit ["Leaving"] 12:05:52 danlei [n=user@pD9E2FD2A.dip.t-dialin.net] has joined #lisp 12:06:58 hypno: yes i find it to be a good book 12:07:41 hypno: i just say that the examples could have been chosen different maybe 12:07:44 Jasko [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has joined #lisp 12:07:56 ejs1 [n=eugen@nat.ironport.com] has joined #lisp 12:08:14 It's very good, but the examples are dated or hard for some. 12:08:29 that's exactly what i wanted to say 12:08:38 but otherwise its good 12:08:52 hard for me maybe 12:09:52 *plage* vanishes to participate actively in the meeting for once. 12:09:52 i can agree with the locking example (but hey, isnt /that/ if anything highly relevant today with all the concurrency hype, etc), but the rest i think pretty nice. i especially like the network protocol implementation explaining specializers. 12:10:33 hypno: yes, it's just not something a beginner would comprehend at first glance maybe 12:11:01 antgree1 [n=Anthony@CPE0014bf0b631a-CM001e6b1858fa.cpe.net.cable.rogers.com] has joined #lisp 12:11:01 The good thing about beginners is that they don't last long. 12:13:51 lol 12:15:33 -!- antgree1 [n=Anthony@CPE0014bf0b631a-CM001e6b1858fa.cpe.net.cable.rogers.com] has quit [Read error: 104 (Connection reset by peer)] 12:15:38 mega1: ping 12:15:50 antgreen [n=Anthony@CPE0014bf0b631a-CM001e6b1858fa.cpe.net.cable.rogers.com] has joined #lisp 12:15:57 pkhuong: hello, I annotated the rw lock paste. 12:16:01 -!- addled [n=adlirc@mail.andrewlawson.org] has quit [Remote closed the connection] 12:16:17 can't say that it works, but it sure looks good to me. 12:16:34 -!- durka42 [n=durka@d161.mertza.swarthmore.edu] has quit [] 12:16:35 sepult: i think that is very minor annoyance to be honest. the great thing about the book is that it focus on proper clos design and production lisp code. it's not overly clever (ie, hacks) and it really illuminates how to think in clos-terms i think. 12:18:11 adlirc [n=adlirc@mail.andrewlawson.org] has joined #lisp 12:18:13 -!- adlirc is now known as addled 12:18:40 attila_lendvai_: is there somewhere an inspector in the code base which doesn't 'expand' a clicked slot, but renders the slot instead of the containing object (ideally adding a bread crumb navigation :) ? 12:18:41 -!- manby-ace [n=manby-ac@88-96-24-54.dsl.zen.co.uk] has quit [Read error: 54 (Connection reset by peer)] 12:19:02 sledge [n=chris@pdpc/supporter/student/sledge] has joined #lisp 12:19:09 manby-ace [n=manby-ac@88-96-24-54.dsl.zen.co.uk] has joined #lisp 12:19:19 mega1: thanks. 12:20:10 pkhuong: is it for hash tables? 12:22:32 no, just for fun. I was waiting for the cluster to handle all my jobs (: It could be a contrib, I guess. 12:23:16 postamar [n=postamar@x-132-204-253-118.xtpr.umontreal.ca] has joined #lisp 12:23:52 dlowe [n=dlowe@ita4fw1.itasoftware.com] has joined #lisp 12:24:21 -!- dkcl [n=danderse@unaffiliated/dkcl] has quit [Remote closed the connection] 12:25:26 -!- hugod [n=hugod@bas1-montreal50-1279442133.dsl.bell.ca] has quit [] 12:26:06 if it can be made async signal safe and it's fair and fast then hash tables could take advantage of it 12:26:27 have you done any performance evaluation? 12:28:04 no, not yet re perf. Re async signals, does it mostly involve avoiding recursive acquisitions? 12:29:02 also protecting against async unwinds which is usually harder. 12:31:05 which means w/o-interrupts everywhere. That'll be fun to check. 12:31:08 -!- puchacz [n=puchacz@87.194.5.99] has quit [Remote closed the connection] 12:31:55 pkhuong: it's even harder because you don't want to block in a without-interrupts 12:32:55 it's a ungodly mix of without-interrupts, with-interrupts and unwind-protect 12:33:34 The futex based mutex acquisition/release are two examples of how it can work. 12:34:13 But they rely on futexes not changing on being waked up. 12:34:33 Which is why the lutex based implementation is not async unwind safe. 12:37:54 What's the problem with the GC not stopping threads in foreign code, and each thread checking whether they should stop when returning/called back from foreign? 12:40:25 disumu [n=disumu@p54BCEF86.dip.t-dialin.net] has joined #lisp 12:41:06 pkhuong: foreign code can take an arbitrarily long time in one thread disabling gc in the whole image 12:41:29 why do you ask this? 12:41:47 hah, slightly lisp related: somebody ported the 2d platformer/shooter "abuse" to the iphone. 12:41:52 no, just not worry about foreign threads at all. 12:42:05 (it uses an elisp-like language to script the enemies and levels) 12:42:33 I'm thinking about this because pthread code is a special case of foreign code. 12:42:44 ecl maybe 12:43:43 pkhuong: any blas matrix multiplication on sufficiently big matrices would prevent gc for a long time 12:43:49 The-Kenny [n=moritz@p5087AD31.dip.t-dialin.net] has joined #lisp 12:44:34 mega1: no, the GC can scan that thread's stack without stopping it. 12:44:57 -!- frank_s_ [n=FrankS@41.145.115.144] has quit [Read error: 60 (Operation timed out)] 12:45:16 what about callbacks? 12:46:09 yvdriess [n=yvdriess@progpc35.vub.ac.be] has joined #lisp 12:47:02 what about async unwinding from foreign code? Is that to be disallowed? 12:47:03 They have to check whether they should stop when they're called from foreign code. 12:47:33 -!- billstclair [n=billstcl@unaffiliated/billstclair] has quit [Read error: 113 (No route to host)] 12:48:13 at least async unwinds from foreign code are the user's problem. 12:48:44 would this also imply that allocation of lisp object is not allowed in foreign code? 12:48:53 is it currently allowed now? 12:49:52 in the runtime, yes. I imagine to call callbacks allocation of argument may have to be done, too. 12:50:26 what are you trying to accomplish by this? 12:50:30 I imagine that could be hacked around. 12:50:30 -!- oudeis_ [n=oudeis@89-139-94-206.bb.netvision.net.il] has quit [No route to host] 12:50:40 yes, quite possibly. 12:51:25 I'm just trying to see if there's an alternative to disabling and enabling interrupts around anything that might block. 12:52:10 well, gc is not blocking. 12:52:35 you are considering safe points, aren't you? 12:52:45 frank_s_ [n=FrankS@41.145.115.144] has joined #lisp 12:53:27 not necessarily. Signal only for lisp code and safe points around foreign code seemed like it might work. 12:55:05 (without-interrupts 12:55:05 (setq fd (with-interrupts (open "sdfsa")))) 12:55:23 -!- sellout [n=greg@c-24-128-50-176.hsd1.ma.comcast.net] has quit [] 12:55:30 this is the most used example of async unwind badness 12:56:07 the window of fd lossage is between open returning and with-interrupts being left 12:56:22 right, I remember nyef fighting with that. 12:56:22 An async unwind there loses fd. 12:57:02 billstclair [n=billstcl@unaffiliated/billstclair] has joined #lisp 12:57:05 without safe points in Lisp I can't see how you can fix that. 12:58:52 this ought to work right?: (read-from-string "(\"a\" \"b\")") (it works for me in CLISP but not in SBCL 1.0.25.debian) 12:59:25 aerique: how does it not work? 12:59:47 pkhuong: I'm getting The variable |A\\\\| is unbound. 13:00:07 I can lisppaste the whole thing if you want 13:00:12 aerique: please, do that. 13:00:20 Ragnar [i=54a67a2c@gateway/web/ajax/mibbit.com/x-01661c1ea4a7f391] has joined #lisp 13:00:26 works with sbcl 1.0.18 13:00:55 -!- yvdriess [n=yvdriess@progpc35.vub.ac.be] has quit [] 13:01:29 pchrist [n=spirit@gentoo/developer/pchrist] has joined #lisp 13:01:59 mega1: I don't have the runtime chops to see that through yet. I'm also not sure the change would be unanymously approved. 13:02:54 ejs2 [n=eugen@77.222.151.102] has joined #lisp 13:03:24 existentialmonk [n=carcdr@64-252-67-89.adsl.snet.net] has joined #lisp 13:03:34 *tcr* tried to reduce some style-warnings in sbcl's build process of itself, but gives up due to so many inscrutable dependencies 13:04:12 I think moving to safe points on the Lisp side is almost unavoidable. I don't know how to handle foreign land. 13:04:40 -!- plutonas [n=plutonas@nomad51162.netlogon.lu.se] has quit [Read error: 104 (Connection reset by peer)] 13:05:35 I'd consider foreign threads as stopped, modulo allocation code. 13:05:47 pkhuong: I found the problem, since kuwabara said it worked for him. It's not SBCL but the Linedit package I'm using (http://www.cliki.net/Linedit). 13:05:48 There are no safe points in foreign land, so interrupts may be allowed to be handled on the spot. 13:06:30 oudeis [n=oudeis@77.124.71.204] has joined #lisp 13:07:36 -!- jewel [n=jewel@dsl-242-183-106.telkomadsl.co.za] has quit [Success] 13:08:51 jewel [n=jewel@41.242.183.106] has joined #lisp 13:11:23 -!- ejs1 [n=eugen@nat.ironport.com] has quit [Read error: 104 (Connection reset by peer)] 13:12:19 LiamH [n=none@common-lisp.net] has joined #lisp 13:12:41 blitz_ [n=blitz@2001:6f8:10f6:200:21b:77ff:fe41:11ab] has joined #lisp 13:14:40 krumholt [n=krumholt@port-92-193-18-193.dynamic.qsc.de] has joined #lisp 13:15:33 scottmaccal [n=scottmac@Sentry3.jay.k12.me.us] has joined #lisp 13:17:07 -!- oudeis [n=oudeis@77.124.71.204] has quit ["This computer has gone to sleep"] 13:17:41 Davidbrcz [n=david@86.199.10.148] has joined #lisp 13:18:53 antifuchs: are boinkmarks not updated these days? 13:19:13 mega1: until the new rounder corner version is up 13:19:24 davazp [n=user@79.153.148.56] has joined #lisp 13:19:51 hah, I thought he was fixing the tests themselves :-) 13:23:11 mega1: about run-program, I wonder why we're not taking advantage of the siginfo_t argument to the handler: on conforming POSIX implementations, the siginfo contains the pid and the status during SIGCHLD, so we shouldn't have to call any wait() function for each pid in the processes collection. 13:24:31 kreuter: I don't think you will get as many sigchld signals as there are exiting children. 13:24:53 really? 13:24:56 that's why it's recommended to wait in a loop 13:25:00 ok 13:25:30 hugod [n=hugod@modemcable086.138-70-69.static.videotron.ca] has joined #lisp 13:26:01 does what's on sbcl-devel look ok to you? 13:26:04 -!- scottmaccal is now known as foofoop 13:26:21 I basically only added comments and a test since. 13:27:11 it seems fine. 13:27:20 -!- foofoop is now known as scottmaccal 13:29:54 hm. git is a few commits behind CVS. is anybody in the middle of a commit right now? 13:29:59 yes 13:30:04 me too 13:30:17 ok, I'll hold off. 13:30:21 busily deleting CLOS 13:30:27 paul graham said it wasn't useful 13:30:28 well, right now right now no, but incrementally through the afternoon 13:30:36 dkcl [n=dkcl@unaffiliated/dkcl] has joined #lisp 13:30:57 then I grab the cvs rubberduck 13:31:19 durka42 [n=durka@d81.wireless.swarthmore.edu] has joined #lisp 13:31:20 mega1: could you squeak it when you're done? 13:31:27 binarin [n=user@62.105.145.214] has joined #lisp 13:32:34 not that I would dare advocating anything in the realm of VCS, but since switching to DVCSes is the hip thing with the hacker kids these days, how come SBCL hasn't followed suit? 13:32:40 -!- mega1 is now known as rubberduck 13:32:42 CyberBlue [n=yong@60.26.103.139] has joined #lisp 13:32:47 *rubberduck* squeaks 13:32:50 -!- rubberduck is now known as mega1 13:33:31 S11001001: cvs/sf.net seems to mostly work, c-l.net is down more often than s-f.net. Let's not go around fixing things that are already workable. 13:33:39 sure 13:34:02 you can also look through the archives for a couple bikeshed threads that boil down to the above. 13:35:00 S11001001: for my part, time spent getting hip comes out of the "time to do something productive" budget. 13:36:44 timor [n=martin@w4642.dip.tu-dresden.de] has joined #lisp 13:36:48 -!- timor [n=martin@w4642.dip.tu-dresden.de] has left #lisp 13:36:50 mega1: antifuchs: boinkmarks seem somewhat bogus, since they don't really execute for long enough to be meaningful 13:36:51 I really like git, but I'm satisfied with the git mirror 13:37:08 well, one thing I like about git/darcs is that they work for me, unlike CVS, which is always a PITA 13:37:25 but git mirror is good enough for a non-developer :) 13:37:40 froydnj: that's going to be fixed in the new shiny version 13:37:52 froydnj: I routinely discard/disregard the shortest running ones 13:37:53 sellout [n=greg@guest-fw.dc4.itasoftware.com] has joined #lisp 13:37:54 jsnell: oho! clearly I am behind the times. 13:38:51 -!- athos [n=philipp@92.250.250.68] has quit ["leaving"] 13:40:56 oudeis [n=oudeis@80.250.159.240] has joined #lisp 13:41:10 -!- alec [n=aberryma@ita4fw1.itasoftware.com] has quit ["Leaving"] 13:44:42 -!- daniel_ is now known as daniel 13:44:49 -!- blitz_ [n=blitz@2001:6f8:10f6:200:21b:77ff:fe41:11ab] has quit [Read error: 113 (No route to host)] 13:45:41 -!- sellout [n=greg@guest-fw.dc4.itasoftware.com] has quit [Read error: 104 (Connection reset by peer)] 13:46:59 what is the behaviour for setf when the variable does not exists 13:47:06 sellout [n=greg@guest-fw.dc4.itasoftware.com] has joined #lisp 13:47:16 PissedNumlock: undefined. some implementations make a special variable automatically 13:47:21 is it wrong to use it when the var isn't initialized yet 13:47:24 k, so it is :) 13:47:27 sbcl gives a warning 13:47:58 PissedNumlock: yes, it is. Use defvar or defparameter to define a special variable, LET to define a lexical variable 13:48:00 -!- billstclair [n=billstcl@unaffiliated/billstclair] has quit [] 13:48:17 I can't stand cvs, but svn is fine 13:49:02 (though the big problem with sourceforge is more the terrible mail archive interface, imo) 13:50:33 dlowe: LET can bind in the dynamic environment, too. :( 13:50:56 billstclair [n=billstcl@unaffiliated/billstclair] has joined #lisp 13:51:40 timor [n=martin@w4642.dip.tu-dresden.de] has joined #lisp 13:52:00 -!- timor [n=martin@w4642.dip.tu-dresden.de] has left #lisp 13:52:12 -!- Fufie [n=poff@80.203.160.34] has quit ["Leaving"] 13:52:28 kreuter: yeah. what should I say then? "local" doesn't really sound right either 13:52:47 non-top-level? 13:53:01 -!- billstclair [n=billstcl@unaffiliated/billstclair] has quit [Client Quit] 13:53:17 -!- ejs2 [n=eugen@77.222.151.102] has quit ["This computer has gone to sleep"] 13:53:42 -!- sad0ur [n=sad0ur@psi.cz] has quit [Remote closed the connection] 13:53:52 dunno. "local lexical"? 13:54:02 sad0ur [n=sad0ur@psi.cz] has joined #lisp 13:54:31 yvdriess [n=yvdriess@progpc35.vub.ac.be] has joined #lisp 13:54:54 -!- sbahra [n=sbahra@c-98-218-239-172.hsd1.dc.comcast.net] has quit [Read error: 110 (Connection timed out)] 13:55:06 *shrug* it's lexical by default. 13:56:40 ISTR playing with some icky code that played games with non-global specials, progv & symbol-value once. incomprefreakinghensible. 13:58:05 -!- manby-ace [n=manby-ac@88-96-24-54.dsl.zen.co.uk] has quit [Read error: 104 (Connection reset by peer)] 13:58:15 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [] 14:00:28 -!- durka42 [n=durka@d81.wireless.swarthmore.edu] has quit [] 14:00:45 -!- oudeis [n=oudeis@80.250.159.240] has quit ["This computer has gone to sleep"] 14:01:31 durka42 [n=durka@130.58.194.81] has joined #lisp 14:04:04 UnwashedMeme [n=nathan@216.155.97.1] has joined #lisp 14:08:06 legumbre` [n=user@r190-135-53-122.dialup.adsl.anteldata.net.uy] has joined #lisp 14:09:20 -!- ejs [n=eugen@77.222.151.102] has quit [Read error: 110 (Connection timed out)] 14:14:45 -!- xan [n=xan@cs78225040.pp.htv.fi] has quit ["leaving"] 14:14:48 timor [n=martin@w4642.dip.tu-dresden.de] has joined #lisp 14:14:57 -!- timor [n=martin@w4642.dip.tu-dresden.de] has left #lisp 14:15:50 tritchey [n=tritchey@c-68-58-88-241.hsd1.in.comcast.net] has joined #lisp 14:17:40 -!- durka42 [n=durka@130.58.194.81] has quit [] 14:17:56 -!- prip [n=_prip@host254-129-dynamic.36-79-r.retail.telecomitalia.it] has quit [Read error: 54 (Connection reset by peer)] 14:19:04 -!- Phoodus [i=foo@ip68-231-38-131.ph.ph.cox.net] has quit [Read error: 54 (Connection reset by peer)] 14:21:57 vy [n=user@213.139.194.186] has joined #lisp 14:22:46 prip [n=_prip@host254-129-dynamic.36-79-r.retail.telecomitalia.it] has joined #lisp 14:22:57 ikki [n=ikki@201.155.75.146] has joined #lisp 14:23:06 bobbysmith007 [n=russ@one.firewall.gnv.acceleration.net] has joined #lisp 14:25:53 -!- ia [n=ia@89.169.189.230] has quit [Connection timed out] 14:26:42 envi^laptop [n=envi@220.121.234.156] has joined #lisp 14:26:47 -!- envi^laptop [n=envi@220.121.234.156] has quit [Client Quit] 14:26:58 froydnj: yeah, http://sbcl.boinkor.net/boinkmarks is the new version, and almost everything is there already; the only thing missing is the atom feed of significant changes, really 14:27:17 The requested URL /boinkmarks was not found on this server. 14:27:24 ia [n=ia@89.169.189.230] has joined #lisp 14:27:26 oh, wait. /boinkmarks/index 14:27:32 oudeis [n=oudeis@bzq-82-81-239-4.cablep.bezeqint.net] has joined #lisp 14:27:37 yes, also, redirects for common cases of the mistypes (-: 14:28:02 sohail [n=Sohail@unaffiliated/sohail] has joined #lisp 14:28:58 antifuchs: ooo, shiny! 14:29:39 for one, it looks like mega's call/ret patch seems to have brought a really nice speed win on the likes of http://sbcl.boinkor.net/boinkmarks/index#TRIANGLE and #TRTAK 14:29:39 legends would be nice. maybe my firefox is just not displaying them, though. 14:29:48 nope, there is no legend right now 14:30:05 I'll work on making the floaty right bar a legend across the graphs 14:30:21 something did a number on FIB, too. 14:30:32 -!- legumbre [n=user@r190-135-24-68.dialup.adsl.anteldata.net.uy] has quit [Read error: 110 (Connection timed out)] 14:30:53 hovering and clicking gives you details, too (and you can zoom in on the last release and see what did the number (-:) 14:31:08 and yeah, it's 1.0.27.17 as well 14:31:36 -!- addled [n=adlirc@mail.andrewlawson.org] has quit [Remote closed the connection] 14:32:18 adlirc [n=adlirc@mail.andrewlawson.org] has joined #lisp 14:32:20 -!- adlirc is now known as addled 14:34:35 this is perhaps more of an emacs question, but i'm wondering if it is possible to "clip" (?) the repl-buffer .. i mean set a maximum for number of lines allowed before it starts deleting lines at the top .. i'm running long long tests that produce a lot of output and i don't care about the stuff at the start (it pauses when something interesting is detected though) 14:36:36 antifuchs: I like the by-date view, but the by-commit view would also be useful I think 14:38:54 cracki [n=cracki@sglty.kawo2.RWTH-Aachen.DE] has joined #lisp 14:39:37 yeah, especially with the tendency of commits to cluster 14:40:13 yes, that's true 14:40:22 I hope I can patch flot to allow that one day 14:42:10 -!- Ragnar [i=54a67a2c@gateway/web/ajax/mibbit.com/x-01661c1ea4a7f391] has quit ["http://www.mibbit.com ajax IRC Client"] 14:42:30 what is this flot and why is it not apt-gettable? 14:42:56 "Flot is a pure Javascript plotting library for jQuery." 14:43:06 No apt-get for you. 14:43:37 sepult_ [n=buggarag@xdsl-87-78-121-209.netcologne.de] has joined #lisp 14:43:54 -!- sepult [n=buggarag@xdsl-87-78-75-195.netcologne.de] has quit [Read error: 60 (Operation timed out)] 14:44:06 gigamonkey [n=user@adsl-99-35-219-234.dsl.pltn13.sbcglobal.net] has joined #lisp 14:44:48 The latter doesn't quite follow from the former, but nolo contendere 14:45:10 it's the plotting substrate I use for boinkmarks, yeah 14:45:47 actually, I don't understand why flot would need to be patched 14:45:54 don't you just need to give it different (x,y) data? 14:46:10 "Not all lisps think (log 2d0 10d0) is the same." 14:46:12 Hmm, I see jquery apt-gettable. Didn't expect that. 14:46:19 mokogobo [n=mokogobo@pcp075595pcs.unl.edu] has joined #lisp 14:46:56 chris2 [n=chris@p5B16A7DB.dip0.t-ipconnect.de] has joined #lisp 14:47:02 mega1: I apologise for the tortured grammar :-) 14:47:54 -!- plage [n=user@serveur5.labri.fr] has quit [Remote closed the connection] 14:48:00 Xof: that's true 14:48:27 luis: can't you get javascript repls these days? 14:48:28 I could just evenly distribute the timestamp values. very sensible 14:48:38 you can! 14:48:56 (and you should. js development is torture without 14:51:51 hm, flot. I had been using plotkit, but maybe flot is actually maintained... 14:53:21 -!- alinp [n=alinp@86.122.9.2] has left #lisp 14:55:27 Ragna [i=54a67a2c@gateway/web/ajax/mibbit.com/x-90deb61add58003f] has joined #lisp 14:55:37 Okay guys, I'm vaguely thinking about the possibility of a 2nd edition of PCL. If you have ideas of what I might add, let me know. (Here or by email.) 14:56:19 matley [n=matley@193.204.39.77] has joined #lisp 14:58:26 -!- davazp [n=user@79.153.148.56] has quit [Remote closed the connection] 14:58:55 Initial thoughts: More on CLOS (currently a good complementary book is Sonya's, but it could be made more-or-less obsolete by an addition to PCL). A more exhaustive list of FORMAT/LOOP directives, and more examples of conditions & restarts. Maybe more on ASDF 14:59:04 gigamonkey: hmm, ASDF, FFI, multiprocessor programming, compiler macros, MOP? 14:59:27 sql 15:00:22 -!- addled [n=adlirc@mail.andrewlawson.org] has quit [Remote closed the connection] 15:00:28 year, something about multiprocessor programming 15:00:38 gigamonkey: remove any mentions of asdf-install? :-) 15:00:46 Lycurgus [n=Ren@cpe-72-228-150-44.buffalo.res.rr.com] has joined #lisp 15:00:49 luis: done, I think. 15:00:54 add clbuild? 15:01:03 addled [n=adlirc@mail.andrewlawson.org] has joined #lisp 15:01:03 *gigamonkey* never really understood asdf-install. 15:01:26 Is there any chance ASDF is going to be obsoleted by one of the other build systems folks are working on? 15:01:46 xcvb and mudballs, is it? 15:01:51 -!- Lycurgus [n=Ren@cpe-72-228-150-44.buffalo.res.rr.com] has left #lisp 15:01:51 -!- jdz [n=jdz@85.254.211.133] has quit ["Somebody booted me"] 15:01:51 and oh, clbuild. 15:01:56 gigamonkey: hunchentoot? simple 3d graphics using cl-opengl and cl-glfw and cl newbies on windows always seem to want deliver executables 15:01:57 Yeah, those are the ones I was thinking of. 15:02:09 willb [n=wibenton@wireless74.cs.wisc.edu] has joined #lisp 15:02:17 Yeah, I'd definitely switch from allegroserve to hunchentoot. 15:02:27 gigamonkey, XCVB is more a replacement for ASDF, where the other two are more systems-installers that use ASDF... right? 15:02:49 tic_: I don't think mudballs uses ASDF. 15:03:03 tic_: that sounds right to me, from what I know. So the question is, is XCVB possibly going to supplant ASDF? 15:03:16 gigamonkey: unlikely 15:03:17 -!- aunwork [n=aunwork@mail.ionicsoft.com] has quit ["Leaving"] 15:03:18 Or would I need to write both. Or is ASDF going to fight off the upstart challenger. 15:03:23 gigamonkeys: i don't think we can say that at this moment 15:03:35 gigamonkey, depends on if you ask Fare or not. ;) 15:03:36 speaking of, has ASDF development moved to c-l.net? 15:03:45 gigamonkeys: asdf will be around for a while (couple of years at least) 15:03:47 I haven't seen mailing list traffic in a while. 15:04:03 user___ [n=user@p57A7C95F.dip.t-dialin.net] has joined #lisp 15:04:12 multiprocessor programming in lisp (given that it doesn't offer anything terribly unusual in the field) is possibly best left to specialist books on muliprocessor programming 15:04:15 The other way I'm looking at a 2nd edition is (semi cynically) what's the minimal set of changes I could make to PCL in a 2nd ed. that would cause a good chunk of the people who bought the 1st ed. to think it worthwhile to buy a new copy. 15:04:45 gigamonkey, that's a good way of thinking. 15:04:58 If I try to cover too much new stuff, then I should just write _More Practical Common Lisp_ 15:05:08 gigamonkeys: make it a supplement and release a corrected edition of PCL? 15:05:27 make only minor changes and call it Practically Practical Common Lisp? 15:05:28 -!- ThomasI [n=thomas@unaffiliated/thomasi] has quit ["Bye Bye!"] 15:05:30 MORE PRACTICAL 15:05:31 impractical common lisp! 15:06:08 gigamonkey: Revenge of Practical Common Lisp? 15:06:14 That's what they put on my copy of PCL here at work. 15:06:18 ("im") 15:06:23 aerique: I'm not sure I understand what you mean. 15:06:33 or make it "Real World Common Lisp" :D 15:06:50 p_l: I think O'Reilly's already working on that one. ;-) 15:06:51 gigamonkey, fix the known bugs in PCL. Then, make an additional book, like More PCL with the suggested changes. 15:06:56 mega1: one embarassing thing is that clisp's value for (log 2d0 10d0) is closer than sbcl's 15:07:04 gigamonkey: well you basically covered it in your remark before mine 15:07:23 Xof: another grammar killed :-) 15:07:23 have an appendix on 'how to use #lisp' :) 15:07:41 good idea 15:07:41 with a grain of salt 15:07:42 closer to the correct value? 15:07:50 tic_: Well, what's a "known bug". Is the fact that I used allegroserve when hunchentoot didn't exist a bug? 15:07:51 to the mathematically correct value, yes 15:07:54 how come? 15:07:56 gigamonkey, more like errata. 15:08:03 gigamonkey, or is ther no errata for the book? 15:08:15 mega1: clisp afaik uses GSL for its maths 15:08:18 tic_ there are some. But not very many big ones that I'm aware of. 15:08:44 Small errata (those that can be fixed without changing pagination) can be fixed in new printings. 15:08:54 didn't, hunchentoot exist, in lispworks-only form? 15:08:57 that is, clisp's value is accurate to double-float precision, whereas sbcl's is off by more than 0.5 ulp 15:08:58 gigamonkey, then it might not be worth it. This is just me translating what aerique said, by the way. I think you should either make large changes to PCL, or possibly write More PCL, if you can find enough information. 15:09:16 so what does GSL do differently? 15:09:40 I definitely would love to see another book like PCL out there 15:09:48 sbcl probably does (/ (log 2d0) (log 10d0)) 15:09:59 mega1: it has probably more people working on it ;-) 15:10:07 that goes more in depth on some stuff..I'm not a huge fan of Apress though :( well, for their digital content I should say. 15:10:19 I would imagine that gsl will have series for log to various bases and some method for interpolating 15:10:43 -!- jdv79 [n=jdv79@u10570642.ul.warwick.net] has left #lisp 15:10:47 we don't need to go all the way to gsl to find better numerics than in sbcl; netlib has a lot of this stuff covered too 15:11:01 Drives me nuts that stuff like the sony reader/kindle are out there and they don't have ebooks available on those formats (unless you buy through kindle store). Got the dead-tree version of PCL so far, but really want it on my ebook reader too, heh. 15:11:34 p_l: clisp doesn't use gsl 15:11:47 I wonder how you have to encode graphics for the Kindle. 15:11:52 fe[nl]ix: Doesn't? Then I was wrong... 15:11:58 TDT: doesn't the kindle thingy take pdfs? 15:11:59 gigamonkey: normal jpeg and png works, afaik 15:12:16 Hmmm. I could add a Kindle backend to my Markup system. 15:12:19 gigamonkey: Kindle uses Mobireader with some lockins in their files 15:12:42 *rsynnott* occasionally uses my mobile phone to read programming books :P 15:12:44 content to be fed into book generator can be standard HTML + some metadata 15:12:47 gigamonkey: It's actually not too bad, there are converters out right now that if you have PCL in one giant html file with graphics in the file (not linked elsewhere) you can convert it in one swoop. 15:12:58 rsynnott: Yeah,, it does, but apress PDFs are password protected 15:13:02 TDT: download the html version and convert it to azw. 15:13:20 azw is packaged html with an index document. it's really simple to generate 15:14:10 antifuchs: I don't know much about azw, I should look at that...but if I get it in HTML, one fileish or so, I can go from there easy enough, just separate files, so far meh. 15:14:18 TDT: that is, if you use the kindle (like I do). other readers should accept pdb or prc, which is basically the same format 15:14:25 ah, okay 15:14:31 separate files are not hard either. 15:14:40 kjbrock [n=kevinbro@66.166.232.134] has joined #lisp 15:14:44 The hard part is buying a Kindle... 15:14:54 actually, I have one apress pdf ebook; there was a bit of a bug with the password protection 15:15:04 (macos preview was able to happily print it to pdf) 15:15:08 Xof: this may be more severe: https://bugs.launchpad.net/sbcl/+bug/327192 15:15:19 luis: I have a sony reader, these devices..worth it, seriously 15:15:31 there are various conversion programs that work with multiple html input documents (IIRC the dpans2texi I used to convert the hyperspec was a conversion from multi-file html) 15:15:35 mega1: yes. Thanks to the marvellous chip designers at Intel 15:15:47 I find myself carrying it everywhere, and I just toss random web content converted on it..so some PCL stuff is there, but not everything. 15:16:25 Fufie [n=innocent@86.80-203-225.nextgentel.com] has joined #lisp 15:16:38 ooor you could just concatenate the html for each chapter. 15:17:35 I guess another question is: is there anything I could add to PCL that would cause any of you who haven't bought it, to buy a copy. 15:17:39 antifuchs: Yeah, that's pretty much what I think I'd have to do. The fantastic program I use is calibre for much of this, and it supports html directly to lrf. The conversion is a bit goofy at times though, so having it as PDF is fine too, as long as it's not password protected that is. 15:17:43 gigamonkey: You could think of adding exercises to the book (not an easy job, mind you.) People complain about that regularly. 15:17:56 scratch_ [n=scratch@adsl-232-6-140.asm.bellsouth.net] has joined #lisp 15:18:09 gonzojive [n=red@fun.Stanford.EDU] has joined #lisp 15:18:19 tcr: Hmmmm. I didn't write exercises mostly because I never do them when I'm reading a book. 15:18:26 Really? 15:18:27 But yes, I know people complain about that. 15:19:03 gigamonkey: Well, better coverage of restarts and condition handling would be nice. I had to supplement hte heck out of that chapter to finally get it :) An ebook format of some sort (I'm willing to help you with that if you wish), and some more exclusive content in the book vs what's on the web would work too. Say give a bit of all the information on the web, but more indepth in the book. 15:19:14 Really. I prefer to read books straight through and then, if I really want to learn the thing, I'll come up with stuff to do on my own. 15:19:23 how can I make a variable inside a function that won't be overwritten by recursive calls by that function on itself? 15:19:42 -!- Fufie [n=innocent@86.80-203-225.nextgentel.com] has quit [Success] 15:19:56 gigamonkey: Some people do exercises though..I didn't, but I honestly spent a lot of time coming up with my own exercises to reinforce the learning. I suppose that's why I spent, like 6 months in your book. 15:19:58 why would you want that? 15:20:10 gigamonkey: as for generating mobireader books, the easiest way is to create XHTML + some book related metadata (there's standard for that). Then load that into mobireader book generator, get .mobi or pdb/prc file back (they use PalmDB format for archives) 15:20:40 scratch_: you seem to be looking for a global variable, aren't you ? 15:20:40 p_l: so if you include IMG tags in your XHTML the generator does the right thing? 15:20:46 gigamonkey: yes 15:20:49 kuwabara, no the opposite 15:20:52 scratch_: what do you mean by "overwritten"? 15:21:01 I'm using "(let (... (true nil) (false nil) ...) (setf true (this-function...)) (setf false (this-function ...)))" with base cases 15:21:14 and at the end, true and false are set to the base cases and I get a one-depth tree 15:21:27 Fufie [n=innocent@86.80-203-225.nextgentel.com] has joined #lisp 15:21:33 -!- araujo [n=araujo@gentoo/developer/araujo] has quit [Read error: 110 (Connection timed out)] 15:21:34 gigamonkey: I found mobireader to be one of most "nice to publisher" systems (well with exception of webscriptions xD) 15:21:42 because when I call (this-function...) again it setfs the same variable name 15:22:04 scratch_: what do you mean "at the end"? 15:22:11 I need a very very local variable basically, one whose scope does not cross recursive calls 15:22:34 scratch_: you will have to make that yourself. 15:22:35 scratch_: I think you need to paste some code to http://paste.lisp.org/new/lisp 15:22:36 the calls to (this-function...) terminate at a specific base case that's mostly irrelevant, when a particular list has 0 size 15:22:51 -!- addled [n=adlirc@mail.andrewlawson.org] has left #lisp 15:23:05 gigamonkey: Once you get the book into one format, epub is a good format, you can get it into all the other formats with extreme ease. 15:23:11 it's HW and I don't feel comfortable pasting the whole thing :( maybe gen-syms would help 15:23:21 addled [n=adlirc@mail.andrewlawson.org] has joined #lisp 15:23:31 scratch_: your questions don't really make sense. 15:23:36 gensyms is for writing macros. 15:23:43 (Typically) 15:23:48 gensyms are for making new symbols dynamically right? 15:23:54 scratch_: you're asking for very local variable, that's what LET variables are, by default. 15:24:00 -!- blandest [n=blandest@softhouse.is.ew.ro] has quit ["Leaving."] 15:24:10 ok I will post some example code that shows what I am talking about 15:24:19 scratch_: not quite sure what you mean but maybe (let ((var value)) (defun ....))? 15:24:24 gigamonkey: I think you should show CFFI, and perhaps how to connect to a DB. If you don't want to make a whole practical chapter on them, at least give a very short primer in the Conclusion chapter. 15:24:28 scratch_ pasted "smashable let variable" at http://paste.lisp.org/display/79135 15:24:44 tcr: I suspect I'd write new practicals on both of those. 15:25:05 when you call that function on say 10, you get 100 15:25:12 if FFI, then probably a chapter on CFFI+SWIG would be a good thing. Also, network programming and concurrency 15:25:14 but the setf only happens on the recursive call with 5 15:25:18 maybe something about GUI programming? 15:25:41 gigamonkey, p_l: All that together sounds rather like a sequel than a second edition. 15:25:42 I want (test-smash 10) to have a reference to a different variable than (test-smash 5) does when it is called later 15:25:57 tcr: I'd prefer a sequel, too :P 15:26:16 *tic_* would like a squirrel. 15:26:41 tic_: Do you want nuts with that? 15:26:52 -!- fe[nl]ix [n=algidus@88-149-210-173.dynamic.ngi.it] has quit ["Valete!"] 15:27:05 p_l, I already have a stash, thank you. I'd be happy to share it with you. 15:27:07 Well, the thing is, I think if I wrote, say 200 pages of new stuff, that might make a new edition worth the price of admission, even for some people who have the 1st. But if I write a whole new sequel, then I have to write, say 400 pages. 15:27:10 *tic_* heads home 15:27:25 gigamonkey, I don't see the problem. ;-) 15:27:46 *p_l* doesn't see one either :D 15:27:50 gigamonkey: Why? Aren't the prospects good to publish a technical book with 200 pages? 15:27:53 gigamonkey: neither do I :P 400 page sof new content, going over things in more detail would be awesome. 15:27:54 does the example make sense? 15:28:04 tic_, p_l: Let me introduce you to my friend, The Family Budget. ;-) 15:28:48 gigamonkey: Perhaps you can get drewc on board, and he'll write the other 200pages about UCW. Harr! 15:28:49 gigamonkey: why not make it more a project then? 15:28:53 scratch_: Do you want (test-smash 10) to establish a binding, which will be used by the recursive calls, then be overriden at 5 for the next recursive calls, and return back to the first binding later ? If yes, this is "special". 15:29:07 gigamonkey, get him away from me, he's evil! :( 15:29:13 gigamonkey: there's a lot of really smart peoople here, if more than you write the content, and you release it as a group, it's less work and benefits the community leaps and bounds 15:29:28 scratch_: at your repl say (trace test-smash) 15:29:33 TDT, like real-world-haskell? 15:29:34 Then evaluate (test-smash 10) 15:29:54 tic_: I haven't read that, or heard of it, but if it follows this general idea then yeah :) 15:30:17 pjb annotated #79135 "static variables in lisp!" at http://paste.lisp.org/display/79135#1 15:30:24 scratch_: ^^^ 15:30:24 d'oh 15:30:31 rstandy [n=rastandy@net-93-144-153-177.t2.dsl.vodafone.it] has joined #lisp 15:30:42 chessguy_work [n=chessguy@67-130-43-2.dia.static.qwest.net] has joined #lisp 15:30:42 tic_: I just think getting more organized content about lisp would be helpful. I can't emphasize enough how much PCL really helped me get into this language. I know of some of the limitations of the book and can provide feedback since I literally read every line of it..but I'm definitely thirsty for more. 15:30:45 I did not capture the essence of my meaning in that example :( 15:31:11 gigamonkey: oh, oh, a SLIME tutorial, definitely. 15:31:13 TDT, *nod* 15:31:20 SirNick [n=SirNick@mail.colortechnology.com] has joined #lisp 15:31:20 *tic_* really heads home. 15:31:23 scratch_: of course, I don't advise using this macro. It's just to show you that you can get what you ask. But you should really use a normal lexical variable in a closure. 15:31:34 luis: I think Slime changes too quickly for that 15:31:34 TDT: well, email me any comments you have. They may be put to some use. 15:31:53 holycow [n=new@mail.wjsgroup.com] has joined #lisp 15:32:02 gigamonkey: k k 15:32:39 tic_: if you haven't yet really headed home, have you looked at Real World Haskell? 15:33:21 pjb annotated #79135 "example" at http://paste.lisp.org/display/79135#2 15:33:30 scratch_: (the first lacked a quote). 15:34:32 rullie [n=rullie@bas4-toronto47-1176151245.dsl.bell.ca] has joined #lisp 15:34:38 pjb annotated #79135 "smashed" at http://paste.lisp.org/display/79135#3 15:34:44 -!- KingNato [n=patno@fw.polopoly.com] has quit [] 15:34:46 doxtor [n=doxtor@unaffiliated/mitja] has joined #lisp 15:34:56 hi, i tried slime-connect to remote host and it says connection refused. but when I do ssh tunneling it's fine. why is that? 15:35:34 rullie: slime makes the socket listen locally only 15:35:35 rullie: swank is binding to the local interface, not the network interface 15:35:36 p0a [n=user@athedsl-385381.home.otenet.gr] has joined #lisp 15:35:41 Hello I have a question about lispbuilder-sdl 15:35:43 kuwabara annotated #79135 "opt arg ?" at http://paste.lisp.org/display/79135#4 15:36:01 dlowe: is it possible to have it listening to net.eth0? 15:36:08 pjb annotated #79135 "done correct, with a closure" at http://paste.lisp.org/display/79135#5 15:36:21 rullie: yes. it's not terribly secure to do so, though 15:36:25 Why can't I use both :MOUSE-MOTION-EVENT and :MOUSE-BUTTON-DOWN-EVENT in SDL:WITH-EVENTS? 15:36:26 scratch_: is it possible that you were just looking for this ? 15:36:39 dlowe: so it's preferred to do the tunnelling then connect? 15:36:44 rullie: Yes, it is 15:36:52 ok thanks tcr, dlowe :) 15:36:55 it's more secure 15:37:36 kuwabara, ah no haha I have really explained myself poorly 15:37:41 spradnyesh [n=pradyus@117.192.5.203] has joined #lisp 15:37:52 possibly because you guys know I couldn't be describing something impossible and I made a mistake causing me to believe the impossible has happened 15:37:56 scratch_: what functionality is it missing ? 15:39:01 scratch_: Can you post an example of some code that does something other than what you want *and* explain what you want it to do? 15:39:52 I am working on that gigamonkey thank you all for your help 15:40:04 Please do not waste any more time until I have explained myself again :) 15:40:06 envi^laptop [n=envi@220.121.234.156] has joined #lisp 15:40:48 -!- envi^laptop [n=envi@220.121.234.156] has quit [Client Quit] 15:40:52 so... anyone? :-( 15:41:11 p0a: sorry, no ideas here. 15:41:34 I'm going to post the source perhaps it'll help 15:41:37 -!- The-Kenny [n=moritz@p5087AD31.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 15:42:02 I took sdl-examples:mouse-painter and modified it 15:42:15 -!- l_a_m [n=nlamirau@194.51.71.190] has quit ["Lost terminal"] 15:42:30 http://paste.lisp.org/display/79136 15:42:34 slyrus [n=slyrus@adsl-68-121-172-169.dsl.pltn13.pacbell.net] has joined #lisp 15:42:41 the idea is to make a very simple mspaint-like progarm 15:43:11 I'm going to add the loadign image part and save image too (sdl provides such function, unfortunately only for BMP saves but I don't worry about that) 15:43:30 morning 15:43:37 but it only captures mouse button down events, and not motion events 15:43:47 morning slyrus 15:43:53 hello slyrus 15:45:29 p0a: movements when a button is pressed? 15:45:41 -!- moocow [n=new@mail.wjsgroup.com] has quit [No route to host] 15:46:32 stassats: I don't understand your question, what are you asking?\ 15:47:24 -!- kiuma [n=kiuma@85-18-55-37.ip.fastwebnet.it] has quit ["Bye bye ppl"] 15:47:25 pkhuong: I'm thinking about serve-event. Did you have some kind of continuation support for sbcl? 15:47:27 nvoorhies [n=nvoorhie@adsl-75-36-207-191.dsl.pltn13.sbcglobal.net] has joined #lisp 15:50:07 -!- aerique [i=euqirea@xs2.xs4all.nl] has quit ["more cowbell"] 15:50:23 -!- chessguy_work [n=chessguy@67-130-43-2.dia.static.qwest.net] has quit ["Leaving"] 15:51:14 p0a: nevermind 15:51:26 rpg [n=rpg@216.243.156.16.real-time.com] has joined #lisp 15:51:58 fe[nl]ix [n=algidus@88-149-210-173.dynamic.ngi.it] has joined #lisp 15:52:30 Xof: #!+long-float bits should just die 15:52:52 ljosa [n=ljosa@18.103.24.170] has joined #lisp 15:56:06 -!- antgreen1 [n=green@CPE0014bf0b631a-CM001e6b1858fa.cpe.net.cable.rogers.com] has quit [Read error: 104 (Connection reset by peer)] 15:56:08 antgreen1 [n=green@CPE0014bf0b631a-CM001e6b1858fa.cpe.net.cable.rogers.com] has joined #lisp 15:56:11 there's expressions that make the REPL go on forever (for example evaluating a circular list). Is a macro 'mute' worth writing, or there's a better method? 15:56:22 For example, (mute expr) ==> t always, while evaluating expr 15:56:43 p0a: (progn expr t) isn't that hard to write 15:57:01 dlowe: well that's what the macro would expand to 15:57:11 dlowe: i'd prefer prog1 there. 15:57:15 depends on how lazy you feel 15:57:18 p0a: no, bad idea 15:57:30 Why? 15:57:36 drewc: for repl use? 15:57:41 p0a: Set *print-circle* to T 15:57:42 this may have all been a big mistake of me interpreting output 15:57:49 dlowe: oh.. true 15:57:57 finally, a use for prog2 15:58:02 I have a lambda function with named variables in it, can I see the value of those variables somehow? 15:58:12 I'm rather wary of using trace on funcall or lambda 15:58:24 p0a: because as the reader of your code, i'd get angry at you for wasting my time to figure out mute when there is PROG1 15:58:35 scratch_: some debuggers let you do that 15:58:43 p0a: but, if it's for REPL use only and will never make it to a .lisp. go nuts 15:59:25 alright, it doesn't make sense to have mute to a .lisp file anyway :-) 16:00:14 p0a: hahaha.. true :) 16:00:37 well with some strategically placed formats I can tell I am getting an actual tree! 16:00:54 so, thanks for letting bang my head against your supple wall of wisdom guys 16:01:05 p0a: i'd also say it makes little sense to have in a .sbclrc too ... after all it's saves what .. 2-3 keystrokes once every few months? :) 16:01:49 shall I be release manager this month? 16:02:07 *froydnj* wonders how much easier Xof's patchset would be if sbcl had a >2-stage build process... 16:02:20 p0a: but in this case i'm just being consistently against needless macros.. there really is nothing wrong with MUTE for REPL use. 16:03:13 drewc: while you were contemplating how many keystrokes mute would save I implemented it. There's no turning back now 16:03:18 froydnj: I think about as hard, because if you don't do what I just did you would need to prove that any leakage into an intermediate stage does not and can never propagate into the final result 16:04:03 I have to go. I'll be back later for the sdl:with-events query. Bye 16:04:04 -!- p0a [n=user@athedsl-385381.home.otenet.gr] has quit ["bye"] 16:04:17 that seems to me harder than just making sure that nothing in the host environment leaks 16:05:48 anyway, that's the lot 16:07:00 minion: memo for Hun: Please ping me up when you're here. 16:07:00 Remembered. I'll tell Hun when he/she/it next speaks. 16:07:19 -!- mvilleneuve [n=mvillene@ABordeaux-253-1-126-133.w86-201.abo.wanadoo.fr] has quit ["Lost terminal"] 16:07:46 Is there a trick to getting probe-file to work with logical pathnames? It seems to want to stick on ".NEWEST" as a version --- do I just need to always set :version to NIL explicitly? 16:08:19 l_a_m [n=nlamirau@194.51.71.190] has joined #lisp 16:08:21 -!- sepult_ [n=buggarag@xdsl-87-78-121-209.netcologne.de] has quit [Client Quit] 16:09:13 *default-pathname-defaults* has :version NIL, so I'm not sure why a :version "newest" pops up... 16:10:03 sepult [n=buggarag@xdsl-87-78-121-209.netcologne.de] has joined #lisp 16:10:54 dwave_ [n=ask@pat-tdc.opera.com] has joined #lisp 16:11:22 -!- disumu [n=disumu@p54BCEF86.dip.t-dialin.net] has quit ["..."] 16:12:28 -!- dkcl [n=dkcl@unaffiliated/dkcl] has quit ["leaving"] 16:12:40 dkcl [n=dkcl@unaffiliated/dkcl] has joined #lisp 16:12:42 mega1: I have no idea that could efficiently work on win32. 16:13:19 pkhuong: and for posixoids? 16:13:22 -!- ljosa [n=ljosa@18.103.24.170] has quit [] 16:13:44 sbahra [n=sbahra@c-98-218-239-172.hsd1.dc.comcast.net] has joined #lisp 16:14:04 Does probe-file do an implicit merge-pathnames? 16:14:40 Because merge-pathnames takes a default-version which defaults to ".NEWEST" 16:14:49 er, :newest 16:15:04 every interaction with the filesystem does an implicit merge-pathnames 16:15:09 (so "yes") 16:15:19 That's what I thought, though I couldn't find where it mentioned it in PROBE-FILE 16:16:12 rpg: So if your default-pathname has a nil version and you don't specify a default version, you'd expect to get a version of :newest supplied by merge-pathnames. 16:16:30 gigamonkey: On a file system with no versioning. :-( 16:16:58 So (probe-file (merge-pathnames your-logical-pathname *default-pathnames-default* nil)) might do what you want. 16:17:04 gigamonkey: So I need to specify :unspecific when setting up *default-pathnames-default*? 16:18:28 or use an implementation which doesn't use the version component of a pathname on an unversioned filesystem 16:18:28 That might work. 16:18:46 mega1: yeah, I think I had something. Never coded anything complete up. Do we really want one-shot continuations in SBCL? (I don't) 16:18:47 Or just pretend logical-pathnames don't exist. (My favorite strategy.) 16:18:54 Xof: Yes. Porting from ACL to SBCL... 16:18:57 clhs 19.3.2.1 16:18:58 http://www.lispworks.com/reference/HyperSpec/Body/19_cba.htm 16:19:21 (don't use :unspecific in logical pathnames) 16:19:52 Xof: Ouch. So how does one get an unversioned logical pathname? 16:19:56 you don't 16:20:06 pkhuong: I thought that serve-event's interesting handler nesting could be rectified by reregistering a handler that's basically the continuation and throwing to the main event loop. 16:20:16 instead, you make sure that the translation throws away the version 16:20:18 -!- Davidbrcz [n=david@86.199.10.148] has quit ["Ex-Chat"] 16:20:21 Seems like the step of translating from logical to physical should treat :newest ... what Xof said. 16:22:20 * (setf (logical-pathname-translations "TMP") '(("TMP:**;*.*.*" "/tmp/**/*.*"))) 16:22:20 (("TMP:**;*.*.*" "/tmp/**/*.*")) 16:22:20 * (probe-file "TMP:SUBSTR.STY") 16:22:20 #P"/tmp/substr.sty" 16:24:24 in fact I can't get probe-file to complain about :version :newest in any way, which is good, because I remember writing code about 3 years ago to make the unix filesystem host agnostic to version 16:24:32 mega1: what do handlers look like? 16:25:25 pstickne [n=pstickne@c-24-21-76-57.hsd1.wa.comcast.net] has joined #lisp 16:25:51 currently? write-output-from-queue in fd-streams.lisp for example 16:26:24 are variables defined outside a lambda function set for that function when it is created? 16:26:28 -!- pstickne [n=pstickne@c-24-21-76-57.hsd1.wa.comcast.net] has quit [Remote closed the connection] 16:26:56 Xof: Ah. I see. I got the translations wrong. I was trying to do "**;*.*" to "**/*.*" to drop the version, but that's wrong. I must map "**;*.*.*" to "**/*.*" Thanks! 16:27:13 pkhuong: but the user's handler can be anything 16:27:35 -!- dwave_ [n=ask@pat-tdc.opera.com] has quit [Read error: 110 (Connection timed out)] 16:27:45 -!- dwave [n=ask@pat-tdc.opera.com] has quit [Read error: 110 (Connection timed out)] 16:27:48 Brilliant! Thanks again! 16:27:58 scratch_: free lexical variables will be captured in a closure 16:28:11 mega1: they get to handle that themselves (: 16:28:47 rpg: good-oh 16:28:53 -!- CyberBlue [n=yong@60.26.103.139] has quit ["Leaving"] 16:29:04 (let ((x 10)) (lambda () x)) will capture x each time this whole form will be evaluated 16:29:32 is there a way to evaluate the variable instead of capturing 16:29:49 what do you mean? 16:30:00 -!- phadthai [i=mmondor@ginseng.pulsar-zone.net] has quit [Read error: 54 (Connection reset by peer)] 16:30:04 one sec 16:30:13 that in place of x will be constant 10? 16:30:32 luke_ [n=luke@208.87.19.36] has joined #lisp 16:30:37 phadthai [i=mmondor@ginseng.pulsar-zone.net] has joined #lisp 16:30:47 -!- yvdriess [n=yvdriess@progpc35.vub.ac.be] has quit [] 16:31:14 *dlowe* doesn't see the point. 16:31:14 you can't get something from the lexical environment there, no 16:31:25 you might be after LOAD-TIME-VALUE, though 16:31:54 so you can do (lambda (x) (+ x (load-time-value ))) and similar 16:32:14 -!- luke_ is now known as Balooga 16:32:22 I forgot to put a channel in but http://paste.lisp.org/display/79140 16:32:22 amnesiac [n=amnesiac@p3m/member/Amnesiac] has joined #lisp 16:33:20 also, how does it work after a variable passes out of scope 16:33:47 pstickne [n=pstickne@c-24-21-76-57.hsd1.wa.comcast.net] has joined #lisp 16:33:52 for instance, I return a lambda function that used a let-variable declared inside the main function 16:34:30 it gets captured in a closure 16:35:12 capture meaning that if I ever setf that variable again it will change the interpretation of every lambda that refers to it? 16:35:36 durka42 [n=durka@d81.wireless.swarthmore.edu] has joined #lisp 16:36:02 -!- durka42 [n=durka@d81.wireless.swarthmore.edu] has quit [Client Quit] 16:36:21 yvdriess [n=yvdriess@progpc35.vub.ac.be] has joined #lisp 16:36:57 davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #lisp 16:37:26 scratch_: http://paste.lisp.org/display/79140#1 ? 16:37:48 it prints "6, 11" 16:38:10 hm, a new let? 16:38:25 that might just do the trick, thanks 16:39:00 oh 16:39:05 but I need to refer to the old x 16:39:41 or do I, hmm nevermind, thanks 16:39:58 old X bindings won't be affected 16:40:36 read the chapter about variables in practical common lisp 16:40:46 scratch_: Why do you think you need that? 16:40:54 tmh [n=thomas@pdpc/supporter/sustaining/tmh] has joined #lisp 16:41:00 Greetings. 16:41:27 does O_NONBLOCK work on all filesystems/platforms? 16:42:17 mega1: depends on what you mean by that 16:42:47 that a read/write will not block 16:42:48 mega1: O_NONBLOCK should work on all *modern* systems 16:43:32 I think there even is something like that on windows 16:44:08 what's a modern system? 16:44:15 dwave [n=ask@212251218147.customer.cdi.no] has joined #lisp 16:44:33 better yet, what's not modern among sbcl's supported platforms? 16:45:21 -!- spradnyesh [n=pradyus@117.192.5.203] has quit [Read error: 110 (Connection timed out)] 16:45:33 I guess nothing 16:45:49 O_NONBLOCK however probably doesn't exist on VMS (for different reasons) 16:46:19 as VMS doesn't have sync. I/O afaik 16:46:30 spradnyesh [n=pradyus@117.192.14.4] has joined #lisp 16:46:34 does sbcl work on sbcl? 16:46:55 -!- Krystof [n=csr21@158.223.51.76] has quit [Read error: 110 (Connection timed out)] 16:47:11 aha, I think deadlines need o_nonblock on every fd that's accessed 16:47:37 would opening with o_nonblock obviously break something? 16:48:24 mega1: it would break all code that assumes that read/write blocks 16:49:21 sure, I'm only referring to fd-streams in sbcl. 16:50:03 hmm... O_NOBLOCK on stuff other than FIFOs and block devices is unspecified 16:50:10 unless someone's using serve-event it should behave the same, right? 16:50:21 hmm, sbcl apparently works on sbcl, i meant on vms 16:50:39 it doesn't 16:50:45 stassats: not yet... but who knows about future, especially if we get VMS working on lameulator :D 16:51:08 p_l: which will be working on sbcl 16:51:15 stassats: Thus closing the circle! 16:51:22 then my question will be correct 16:51:23 -!- phadthai [i=mmondor@ginseng.pulsar-zone.net] has quit [Read error: 104 (Connection reset by peer)] 16:51:58 However, in order to make SBCL on VMS really interesting, we would need IA-64 code generation 16:52:24 p_l: at least char devices should be supported as well 16:52:43 mega1: only those that support non-blocking ops 16:52:55 "When opening a block special or character special file that supports non-blocking opens:" 16:53:38 now I have to look where I lost _linux_ manual pages, cause `man 3 open` gives me POSIX docs 16:53:59 So I asked the wrong question at the beginning, it's the file type not the file system type. 16:55:25 "When attempting to read a file (other than a pipe or FIFO) that supports non-blocking reads and has no data currently available:" " If O_NONBLOCK is set, read() shall return -1 and set errno to [EAGAIN]." 16:55:38 -!- joshe` is now known as joshe 16:56:07 -!- envi^home [n=envi@220.121.234.156] has quit [Read error: 104 (Connection reset by peer)] 16:56:31 mega1: http://linux.derkeiler.com/Mailing-Lists/Kernel/2007-08/msg06662.html and http://lists.lysator.liu.se/pipermail/lsh-bugs/2003q4/000155.html 16:57:19 yes, I saw the first one. 16:58:09 -!- user___ [n=user@p57A7C95F.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 16:58:30 phadthai [i=mmondor@ginseng.pulsar-zone.net] has joined #lisp 16:58:36 -!- silenius [n=jl@yian-ho03.nir.cronon.net] has quit [] 17:01:37 ehu` [n=chatzill@82-170-33-173.ip.telfort.nl] has joined #lisp 17:06:46 anyone know of an article about "when to evaluate macro arguments?" I am not sure whether to evaluate arguments in a macro for defining "web packages": http://paste.lisp.org/display/79142 17:06:51 is there a good free debugger/stepper that will let me look inside closures? 17:07:10 hi, is there an ide for lisp written in lisp, that ist open source? 17:07:36 krumholt, I think Emacs qualifies under that description 17:08:03 scratch_: debuggers come with the implementation 17:08:12 clhs break 17:08:13 http://www.lispworks.com/reference/HyperSpec/Body/f_break.htm 17:08:25 It seems like arguments to def* in most of lisp are not evaluated, so I am tempted to not evaluate any of the arguments. On the other hand, it seems to add more flexibility if the root-function-name argument is evaluated 17:08:46 -!- addled [n=adlirc@mail.andrewlawson.org] has quit [Remote closed the connection] 17:09:23 krumholt: ist is a typo for is or isn't? 17:09:39 yes 17:10:00 the wasn't helpful 17:10:04 s/the/that/ 17:11:18 anyway, already mentioned slime is written in elisp and is quite good, there is also climacs + clim listener, written in CL and which is not as good yet 17:11:42 there is also portable hemlock, CCL's IDE on OS X 17:12:07 they are all pretty much emcas like 17:12:18 there is also CUSP for eclipse 17:12:52 and simple editor with repl, able 17:12:55 clhs able 17:12:56 Sorry, I couldn't find anything for able. 17:13:02 minion: able 17:13:02 Sorry, I couldn't find anything in the database for ``able''. 17:14:43 i am looking for something like the acl IDE just open source. 17:15:00 stassats: See ABLE: "ABLE is a Common Lisp editor bundled with a compiler and a selection of open source libraries." http://phil.nullable.eu/ 17:15:58 drewc: if you will be a new minion, then you should be a little faster 17:16:00 -!- deech [n=deech@71-10-166-226.dhcp.stls.mo.charter.com] has quit [Read error: 104 (Connection reset by peer)] 17:16:21 deech [n=deech@71-10-166-226.dhcp.stls.mo.charter.com] has joined #lisp 17:16:36 stassats: i've only got the one CPU, and it's 30 years old. 17:17:38 time to upgrade! 17:19:36 drewc: brain is not a one cpu 17:20:58 and there is also Dandelion => http://sourceforge.net/projects/dandelion-ecl 17:21:46 p_l: it's not a single CORE, no, but my motherboard only has space for a single chip. 17:22:11 the beeblebrox upgrade is tempting ;) 17:22:21 -!- matley [n=matley@193.204.39.77] has quit [Read error: 110 (Connection timed out)] 17:23:14 thanks for all the sugestions. i don't really like any of those. The one i like best is emacs but i don't like to remember all the shortcuts. i wish i could just right click and choose evaluate 17:23:47 drewc: you are mixing IO backplane connector with computing cluster unit ;-) 17:24:01 that's counterproductive! 17:24:09 slyrus_ [n=slyrus@dsl092-019-253.sfo1.dsl.speakeasy.net] has joined #lisp 17:24:21 krumholt, the complicated shortcuts actually trigger commands like slime-indent-and-complete-symbol and stuff .. you can bind those to easier to use shortcut keys 17:24:36 p_l: i find this conversation quite dull. It was a throwaway joke, i don't care how you personally map human parts to computer components :) 17:24:41 krumholt, (define-key slime-mode-map [f1] 'slime-hyperspec-lookup) for instance 17:25:22 drewc: sorry, bias from having to do neuropsychology ;-) 17:25:46 -!- dkcl [n=dkcl@unaffiliated/dkcl] has quit ["leaving"] 17:26:03 krumholt, (there are actually two shortcut mappings in a slime context; slime-mode-map and slime-repl-mode-map .. one for code-buffers and one for the repl) 17:26:16 krumholt: it's trivial to make emacs do what you want, but by the time you've learned enough emacs to do so, you'll find you didn't want it after all. 17:26:25 dkcl [n=dkcl@unaffiliated/dkcl] has joined #lisp 17:26:49 lnostdal: don't forget sldb, xref, etc. 17:27:18 puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has joined #lisp 17:27:19 lnostdal, i know i can rebind all shortcuts. it still lacks lots of feature i like to have in an ide like a deploy button or something. or a package management 17:27:21 ah, there are mappings for those too .. i've only added a couple of keys for code and repl buffers :) 17:27:22 hmm, is anyone actually using the able thing? 17:28:27 krumholt: that's what external tools are for (you can write emacs integration for them if you want) 17:28:28 rsynnott: i played with it. 17:28:53 i've tried it .. it is very very basic, rsynnott .. you get CCL with a simple editor + repl and some common libraries out-of-the-box, rsynnott 17:28:53 krumholt: a deploy button? how do you propose that work? 17:29:16 sounds like a real time strategy game 17:29:17 krumholt: and package management .. you mean like ASDF integration? Slime has this. 17:29:32 something like the terrifying ruby capistrano thing? 17:29:37 krumholt: If deployment was as easy as clicking a button, it wouldn't be hard to add such button... when it requires an AI... 17:30:02 *drewc* wants a deploy button too now! 17:30:55 drewc: do you want an AI with that? :D 17:30:56 *stassats* wants a "do everything" button 17:30:56 i know. it's just i have started using lisp a year ago. i was disapointet with other languages and i really liked it. but i'm missing a tool like eclipse. 17:31:07 simply click, and it with upload part of the project, break your webserver and give an unintelligible error! 17:31:14 -!- deliana [n=deliana@147.210.246.189] has quit [Remote closed the connection] 17:31:25 krumholt: if you are missing eclipse, you are doing lisp wrong! 17:31:36 krumholt: but why not just use eclipse? 17:31:36 krumholt: well, there is cusp... 17:31:50 krumholt: if the allegro ide is closest to what you like, why don't you just use it? 17:31:55 *tmh* is missing the structured editor that does not exist. 17:32:05 because allegro is no free. 17:32:05 drewc: but .. you don't refactor? 17:32:09 danlei: i sense a 'because it's not free' bs 17:32:12 it has a heap size limit of 60mb 17:32:22 krumholt: mine does not. 17:32:39 throw money at it, krumholt .. :P 17:32:41 krumholt: because i paid for it ;) 17:32:42 Fare [n=Fare@ita4fw1.itasoftware.com] has joined #lisp 17:32:47 or lispworks, maybe? 17:32:50 i don't have the money 17:32:53 (same objection, fo course) 17:33:11 manuel__ [n=manuel@port-92-205-126-235.dynamic.qsc.de] has joined #lisp 17:33:12 cusp is actual eclipse integration, using slime 17:33:21 drewc, one reason in particular i don't use eclipse. if i tell someone how much i like lisp and that it is great and he wants to try it. i have to tell him he has to install java first... 17:33:21 for Eclipse there is als Dandelion, don't forget that 17:33:22 though it tends to be a bit quirky; slime is better 17:33:24 krumholt: I doubt you really need stuff like "deploy button" till you want to earn money with it :) 17:33:27 (using swank, rather0 17:33:27 and lispworks's ide is too emacsey 17:33:57 lispworks is pretty good too but has the same heap size limitation 17:34:02 free edition 17:34:09 mcl? :) 17:34:11 krumholt: you also need to install libc 17:34:16 (assuming you have a really old mac) 17:34:43 yeah, lisp is fast wrt. development .. but it's not so fast the bottleneck is lack of a "deployment button" .. heh :) 17:35:19 deploy - caching! .. deploy - caching! 17:35:32 /me has recently been working with both a lisp webapp and a ruby-on-rails webapp 17:35:54 despite the lack of magic deploy thingy, the lisp one is MUCH easier to deploy 17:36:24 I guess it's partially because Lisp is much rarer and more self-contained in form of apps :) 17:37:40 (in fairness, this is largely due to hunchentoot being much saner than the methods used to run ruby webapps) 17:38:37 ah, that's too. I guess that there are smaller problems wrt. to speed too :) 17:39:49 oh, indeed. Ruby, where you avoid more than a certain number of elements per page because of the processor cost of filling in HTML templates :S 17:39:57 -!- trebor_dki [n=user@mail.dki.tu-darmstadt.de] has left #lisp 17:39:59 xan [n=xan@cs78225040.pp.htv.fi] has joined #lisp 17:40:04 -!- jao [n=jao@74.Red-80-24-4.staticIP.rima-tde.net] has quit [Read error: 110 (Connection timed out)] 17:40:12 bombshelter13p [n=bombshel@24.114.234.33] has joined #lisp 17:40:31 -!- bombshelter13p [n=bombshel@24.114.234.33] has quit [Client Quit] 17:41:49 bombshelter13p [n=bombshel@24.114.234.32] has joined #lisp 17:42:23 -!- yvdriess [n=yvdriess@progpc35.vub.ac.be] has quit [] 17:43:40 krumholt: personally, i don't let other people's opinions choose which tools i use. I don't care if your buddy doesn't like java, and i don't see what that has to do with lisp. 17:44:29 -!- bombshelter13p [n=bombshel@24.114.234.32] has quit [Client Quit] 17:44:46 bombshelter13p [n=bombshel@24.114.234.32] has joined #lisp 17:45:31 *drewc* is feeling particularly combative today. 17:45:39 tsk 17:45:44 apologies in advance if i take it too far. 17:46:25 rsynnott: it's always amazed me that interpreted web languages are so big on 'templates'... 17:46:56 rsynnott: i mean, isn't that just interpreting another language, in an interpreter, and damn near every page hit? 17:47:02 drewc, it is not about that he doesn't like java. i want to convince people using lisp. but i have i hard time doing that if i have to tell them that the ide for lisp is written injava. they will ask me why they shouldn't learn java instead 17:47:38 drewc: generally, they are implemented by turning the templates into python/ruby code once and compiling where possible 17:47:42 so it isn't that bad 17:47:48 krumholt: the IDE for lisp is emacs. Get used to it! 17:47:51 krumholt: so you don't want one written in java, and apparently you don't want any of the ones written in lisp. 17:47:57 -!- bombshelter13p [n=bombshel@24.114.234.32] has quit [Client Quit] 17:47:58 drewc: I guess that's because of the languages being rather hard to write the output in them, or because of the fact that they need to give the design to someone who should be kept away from code itself... 17:48:00 and template languages are basically necessary once designers get in on the game 17:48:01 what do you want? one written in magic? 17:48:12 bombshelter13p [n=bombshel@24.114.234.32] has joined #lisp 17:48:22 moocow [n=new@burnaby.axiomnetworking.ca] has joined #lisp 17:48:22 -!- bombshelter13p [n=bombshel@24.114.234.32] has quit [Remote closed the connection] 17:48:33 rsynnott: i disagree 100%. I work with designers all the time, and i do not use template languages. 17:48:41 because designers should not be writing code! 17:48:41 -!- tombom [n=tombomp@wikipedia/Tombomp] has quit ["Peace and Protection 4.22.2"] 17:49:07 bombshelter13p [n=bombshel@24.114.234.32] has joined #lisp 17:50:19 -!- legumbre` is now known as legumbre 17:50:29 i want eclipse written in lisp. i know emacs probably has the same functionallity. it is just not as easy accesible 17:50:53 eclimpse! it's eclipse in clim! 17:50:58 (now, go write it) 17:51:01 -!- rullie [n=rullie@bas4-toronto47-1176151245.dsl.bell.ca] has left #lisp 17:51:01 awesome. 17:51:01 hehe 17:51:18 what kind of designers should not be writing code 17:51:24 manic12: most? 17:51:39 manic12: the ones that are good at design? 17:51:47 krumholt: perhaps it would be better to create an emacs that is as clickable as eclipse? 17:51:49 complete B.S. 17:51:50 the ones that are bad at coding 17:51:55 *drewc* doesn't design, and doesn't hire designers to code. 17:51:57 madnificent, something like that 17:52:17 emacs, now with more clicking 17:52:25 -!- rread_ [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has quit [Read error: 60 (Operation timed out)] 17:52:28 -!- holycow [n=new@mail.wjsgroup.com] has quit [Read error: 104 (Connection reset by peer)] 17:52:45 I'm a mechanical design engineer and I write more code than most people with computer science degrees that I know 17:52:50 (is what the editor is written in REALLY such a big deal? I use one written in Python for many things) 17:52:58 manic12: FFS, context man. 17:53:03 manic12: I think he meant something different by 'designer' 17:53:07 i know most of you will have no interest in a product like that because you are allready used to emacs 17:53:13 manic12: "graphic designer" .. is that better? 17:53:19 that's why I asked what kind 17:53:26 as long as my editor has a repl, I don't care too much 17:53:33 -!- JuanDaugherty [n=juan@cpe-72-228-150-44.buffalo.res.rr.com] has quit ["Leaving"] 17:53:46 manic12: ahh.. i thought you meant "What do you mean designers can't code? 17:53:58 blitz_ [n=blitz@cl-76.dus-01.de.sixxs.net] has joined #lisp 17:54:09 manic12: not "What kind of designers are there, and who can't code" :) 17:54:15 I think that even graphics designers should write code if they can and want to 17:54:17 krumholt: I would have interest in it, but only to show it to other people... that way emacs becomes more accessible (now do it with climacs, to keep the real lisp in there 17:54:28 -!- sbahra [n=sbahra@c-98-218-239-172.hsd1.dc.comcast.net] has quit [Read error: 110 (Connection timed out)] 17:54:44 poor emacs lisp! 17:54:46 that's how programs get better, when the users get involved in the software design 17:54:49 holycow [n=new@mail.wjsgroup.com] has joined #lisp 17:54:54 krumholt: It's not that we're not interested, it's that we hear this editor complaint on a regular basis and it gets old. 17:55:00 it's not unreal, it's ajust a bit horrible! 17:55:13 otherwise C.S. people would be living in a vacuum of theory 17:55:15 manic12: sure, and i can wank all day if i like. That doesn't mean we should design programming systems around the fact that some designers can write code, or some programmmers can wank. 17:55:21 rread [n=rread@nat/sun/x-1a45eec256f2cade] has joined #lisp 17:55:48 mattrepl [n=mattrepl@129.174.69.135] has joined #lisp 17:55:55 what about extension languages like script fu for gimp, autolisp for autocad? 17:56:25 i know I caught you out of context, but my point remains 17:56:35 manic12: context man.. if you want a win take it : 17:56:42 *drewc* hands manic12 a winner 17:56:43 thanks 17:56:52 i know. this isn't a complaint about emacs. i am allready getting used to it. it is a complaint that no availability of an editor like i described it is keeping other people from having easy acces to lisp. and by that showing other lisp to other people 17:57:25 krumholt: lisp is a programming system, meant for writing programs, not ease of access or 'showing to people' 17:57:33 *stassats* doesn't want to show lisp to anyone 17:57:44 it's also a really spiffy calculator 17:58:21 *madnificent* taks the winner back. 'overruled by sheer ignorance' 17:58:38 if there's one thing I really miss in python/ruby it's *,**,** 17:58:39 krumholt: Hear that argument all the time as well. This is your choice, learn lisp and don't worry about other people learning it -or- decide that you are worried about other people learning lisp and develop the editor to facilitate that. 17:58:40 stassats: why not? 17:59:00 it's a secret! :) 17:59:08 rsynnott: I chuckle whenever I write things like (* * *) at the repl. 17:59:34 madnificent: nobody has showed me lisp, why should i? 17:59:51 rsynnott: python has _ 18:00:10 *blinks* 18:00:12 so it does! 18:00:28 thanks, I somehow missed that despite working with python every day for two years... 18:00:33 tmh, that is a good point. 18:00:34 Heh, no one showed me lisp, either. And when I told people I was going to start using lisp for my projects, I literally received chuckles and was asked, "Does anyone use lisp anymore?" 18:00:48 It should really be mentioned prominently in the docs; very handy feature 18:01:02 it was in the python tutorial forever ago 18:01:02 i just think it doesn't have to be like this. or it shouldn't 18:01:03 durka42 [n=durka@d81.wireless.swarthmore.edu] has joined #lisp 18:01:04 when I started using it 18:01:11 (that looks sarcastic; I genuinely did not know about '_') 18:01:16 sbahra [n=sbahra@c-98-218-239-172.hsd1.dc.comcast.net] has joined #lisp 18:01:17 krumholt: really, start updating climacs for it. I'd really appreciate it. I'm trying to write stuff in lisp that isn't all too hard to use (I'm terribly bad at it though), but that's going to be the way to get more people into lisp 18:01:28 rsynnott: mind you, someone will think that it's too lispy and will remove it 18:01:35 it basically said "by the way, if you use it as a calculator, here's this thing: _. Only don't use it in an actual program, it's not pythonic" 18:01:39 krumholt: Look at it this way, if the editor situation really bothers you that much, you can develop an editor in lisp and kill 2 birds with 1 stone. 18:01:57 stassats: because even though no one showed you, you may show others... when newton discovered the law of gravity, he showed it too... no one showed it to him either 18:02:10 krumholt: Develop an editor to learn lisp. 18:02:12 stassats: and for your own use: more lispers means more libraries and support :) 18:02:45 stassats: Guido seems to have backed down on that a bit 18:02:56 python 3 still has map and reduce, I think 18:03:05 even though he used to be determined to get rid of them 18:03:07 madnificent: i'm not hiding it either 18:03:15 i actually have to write a thesis for my college. i have 6 months and that is why i started this discussion. i am thinking about developing an ide for lisp that is easier to use for beginners 18:03:19 (in fairness, map is obsolete to an extent in python now) 18:03:36 -!- sbahra [n=sbahra@c-98-218-239-172.hsd1.dc.comcast.net] has quit [Remote closed the connection] 18:03:54 Rsynnot: Elaborate? Obsolete how. 18:03:58 stassats: ah no, just pointing out that it might be good to actively try to motivate people... 18:04:05 scratch_: you can't use it in an actual program. it only works at the interactive interpreter 18:04:12 minion: paste 18:04:12 paste: lisppaste: lisppaste is an IRC bot that runs under the nickname "lisppaste" and can be used (only for the #lisp IRC channel!) at http://paste.lisp.org/new/lisp - or http://paste.lisp.org/ for other destinations 18:04:15 most if not all usage scenarios can be replaces with comprehensions 18:04:20 and they're faster, it seems 18:04:32 my experience shows that they are generally too ignorant to appreciate the features of the language. So your approach may be better anyway 18:04:34 rsynnott: yes, calling a function in python is insanely slow. 18:04:39 -!- HET2 [i=diman@xover.htu.tuwien.ac.at] has quit ["This computer has gone to sleep"] 18:04:44 Faster perfrnance wise? 18:04:48 yep 18:05:00 Hm, curious 18:05:15 calling functions is expensive 18:05:20 -!- tritchey [n=tritchey@c-68-58-88-241.hsd1.in.comcast.net] has quit [] 18:05:24 and apparently map's iteration was actually slow 18:05:26 Sucks since I prefer the map syntax most of the time :/ 18:05:51 ([myfun(x) for x in biglist] would be slower than map(myfun, x) 18:05:56 sorry, faster 18:06:08 tmh pasted "Editor links for krumholt" at http://paste.lisp.org/display/79147 18:06:39 -!- durka42 [n=durka@d81.wireless.swarthmore.edu] has quit [] 18:06:45 *stassats* is too lazy to learn list comprehensions for his seldom uses of python 18:06:56 what's the point of discussing python performance? it's a continuum ranging from "slow" to "damned slow" 18:07:02 tritchey [n=tritchey@c-68-58-88-241.hsd1.in.comcast.net] has joined #lisp 18:07:07 -!- tritchey [n=tritchey@c-68-58-88-241.hsd1.in.comcast.net] has quit [Remote closed the connection] 18:07:13 tritchey [n=tritchey@c-68-58-88-241.hsd1.in.comcast.net] has joined #lisp 18:07:32 -!- xan is now known as xan-afk 18:07:44 thanks, i haven't yet decided if i should do this. i think if it doesn't turn out to be too bad. it might actually be of some use. but i am fairly certain it is almost impossible to do in 6 months 18:07:52 google is allegedly going to make it all nice and fast :) 18:08:13 krumholt: you could start with climacs and spend six months making it more newbie friendly 18:08:23 rsynnott: oh, sure. what was it, 3 to 5 times faster? if they can do that, then do it two or three more times again, they'll be getting somewhere. 18:08:24 krumholt: If you are going to write a lisp IDE, check out those links. 18:08:32 tmh, i will 18:09:04 -!- moocow [n=new@burnaby.axiomnetworking.ca] has quit [Connection timed out] 18:09:23 -!- pstickne [n=pstickne@c-24-21-76-57.hsd1.wa.comcast.net] has quit [Client Quit] 18:09:23 i actually think climacs or the new mcclim drei pane which if i understand correctly are pretty similiar provide a lot help along the way of creating an ide 18:10:15 krumholt: what problem are you trying to solve exactly? 18:11:15 jajcloz [n=jaj@pool-98-118-118-197.bstnma.fios.verizon.net] has joined #lisp 18:12:00 drewc, an environment i can install and start coding lisp without having to think about different cl implementations or slime or emacs command. where most commands are accesible by mouse or icon. 18:12:31 krumholt, Genera? 18:12:33 possibly a nice graphical interface to a debugger 18:12:33 krumholt: You could probably achieve that by customizing an install of emacs. 18:12:40 krumholt: and keep us (me?) posted! ^_^ 18:12:49 I remember patrickc using the Genera emulator to develop Lisp code 18:12:51 krumholt: lispworks! 18:12:55 (sorry) 18:13:01 exactly like lispworks or acl 18:13:05 just it has to be free 18:13:06 Fare: you know very well that Genera is very keyboard-centric. 18:13:25 keyboard and mouse 18:13:48 quick question --- is it possible to run save-lisp-and-die from w/in slime? 18:13:55 sorry! Never mind that last 18:14:12 It was sitting around in a buffer, but was overtaken by events.... 18:14:14 i will take next week for some more research into the topic and then i will decide if i pursue this as my bachelor thesis. 18:14:22 -!- rpg [n=rpg@216.243.156.16.real-time.com] has quit [] 18:14:34 krumholt, also have a look at DrScheme and things like that 18:15:04 krumholt: Second that, the last link is to DivaScheme, you should really look at that. 18:15:16 a nice lisp ide in cl would be quite interesting 18:15:25 *madnificent* seconds rsynnott 18:15:32 but very difficult to make it good enough to be convincing vs emacs 18:15:33 i will have a look at all the links. thanks 18:15:46 and a lot of people (myself included) just don't care for clim 18:16:26 rsynnott: what if there were a gtk/qt theme engine backend? 18:16:47 possibly, if qt and looked/felt reasonably nice 18:16:50 rsynnott, my intention is not so much to get people who use emacs to use my ide. they probably will never be convinced to use something else :) it rather is an effort to try help people new to lisp and not used to emacs to get starte 18:16:56 (I'm on a mac; we don't have gtk here :) ) 18:17:24 those people could use acl/lispworks to get started, no? 18:17:40 by the time they need the heap size and so forth they'll be ready for emacs anyway 18:17:57 *stassats* used clisp's bare repl to get started 18:18:02 yes but i was very dissapointed as a simple programm i wrote analysing web pages allready exceded the heap space 18:18:07 rsynnott: one question is whether to edit and develop within one image, or use separate editor process and something like swank. the former is simpler, more powerful, and prone to explode randomly, while the latter is more work, less integrated, and more reliable. 18:18:10 -!- bombshelter13p [n=bombshel@24.114.234.32] has quit [Read error: 60 (Operation timed out)] 18:18:20 and lispworks will then just exit without warning or something 18:18:33 hefner: I'd go for the second approach 18:18:48 hefner: yep, I'd consider having the editor in the same image as totally unworkable for writing certain types of thing 18:19:03 (if you were integrating a C lib, for instance, it could be terribly annoying) 18:19:05 p_l: me too, but with the feeling of having made a tragic compromise the whole time. 18:19:28 akcom [i=akcom@rrcs-67-79-132-132.se.biz.rr.com] has joined #lisp 18:21:31 unfortunately no Lisp that I know offers apriopriate separation systems to work in one image 18:21:51 now, CL-on-.NET would have such a thing... 18:21:57 some sort of imaginary ideal RPC system would be best 18:22:09 but it can still blowup if you manage to crash the VM itself 18:22:26 imaginary ideal RPC systems, eh? :P 18:22:30 indeed! 18:23:10 rsynnott: what about "sidechannel" instead of current swank? With the control mechanisms being separate from the code in lisp image? :) 18:23:10 that way you'd be isolated from your development image, but it would be just as if you were there :) 18:23:11 CORBA, surely. 18:23:29 nah, SOAP: it's Simple! 18:23:32 a definition of 'ideal' with which I was not previously familiar, there 18:23:50 " no matter how much SOAP they used they couldn't create a clean solution?" 18:24:07 haha 18:24:19 (the correct response is: "a definition of 'simple' with which I was not previously familiar, there" 18:25:00 yep; there are many variations on the 'SOAP stands for simple object action protocol; none of these are correct' joke 18:25:40 but it's in XML, so obviously it's simple 18:25:48 XML makes everything easy 18:26:58 XML can be made easy. It also can be made bloated and impossible to understand. The second one is the more common 18:28:11 It's the same with S-Expressions 18:28:21 rpg [n=rpg@97-127-65-227.mpls.qwest.net] has joined #lisp 18:29:41 The temptation seems to be greater with XML 18:30:19 dlowe: Because there are tools that will change your function/data structure into XML format... 18:31:52 and people are way too lazy to design a format for interchange 18:32:45 it's like "no one was ever fired for using XML!" 18:32:53 pity 18:33:20 p_l: i have not hired someone for choosing XML.. does that count? 18:33:25 djarvelis [n=jarv@dsl-217-155-101-22.zen.co.uk] has joined #lisp 18:33:43 .NET SDK tools for XML integration were quite nice, assuming you started by designing the format first... 18:33:44 drewc, close enough for govt work? 18:34:18 *drewc* has never seen a .NET environment. 18:34:52 hell, i've never used eclipse, unless the window manager in cl counts. 18:35:16 billstclair [n=billstcl@unaffiliated/billstclair] has joined #lisp 18:35:31 I used eclipse for some time with RoR. I gave up quite fast 18:35:34 drewc: for java, eclipse is quite nice. It provides good hinting :) 18:35:39 p 18:35:45 Jacob_H [n=jacob@92.4.155.143] has joined #lisp 18:35:54 p_l: RoR integration with emacs is juist a bazilion times better :P 18:36:07 madnificent: Do you think what I was using later? :D 18:36:15 Well, that and NetBeans 18:36:51 NetBeans for RoR? how was that? 18:37:11 sounda a little masochistic 18:37:13 madnificent: works well, you just have too remember the fact that by default it uses JRuby 18:37:14 -!- spradnyesh [n=pradyus@117.192.14.4] has quit [Read error: 60 (Operation timed out)] 18:37:59 though I find ruby in general unsatisfactory. It has odd anonymous functions 18:38:19 -!- manuel__ [n=manuel@port-92-205-126-235.dynamic.qsc.de] has quit [] 18:38:52 p_l: jikes! (pun intended) 18:39:08 madnificent: ouch! 18:39:17 p_l: I'm trying to move from RoR to lisp though 18:39:25 madnificent: i've avoided having to write more than 100 lines of java in my entire career so far. I don't plan to hit 150 18:39:27 *p_l* is a Sun JDK user, none of that IcedTea crap 18:39:29 -!- Balooga [n=luke@208.87.19.36] has quit ["Balooga has no reason"] 18:39:31 Xof: are you still doing commits today? 18:40:19 drewc: I have school projects (now see me cry!) 18:40:28 madnificent: oh, BTW, i have my own 'migrations' code that i will be integrating with rofl soon 18:40:39 drewc: have you seen mine? 18:40:40 madnificent: thanks for the inspiration! :) 18:40:44 hmm, most of my java code comes from assignments in uni. The rest is a certain web GUI for maxima (Java -> JS, Ruby and maxima) 18:41:06 I was kind of hoping that we wouldn't end up with a bazilion of them, that's why I put it in a separate library:)) 18:41:12 madnificent: i did.. rather than line-by-line critique and patch it, it was easier to roll my own. 18:41:26 madnificent: i didn't like the code :( 18:41:41 my recent hava work has all been duct tape for some clojure stuff 18:41:49 lisppaste: url? 18:41:50 To use the lisppaste bot, visit http://paste.lisp.org/new/lisp and enter your paste. 18:42:05 drewc: albeit true, it still doesn't solve the fact that a bazilion systems will be here soon :) (and yes, the code wasn't built to be pretty, I only thought a bit about the api) 18:43:03 drewc pasted "migrations for rolf (sans defmigration syntax)" at http://paste.lisp.org/display/79149 18:43:14 rofl even 18:43:17 madnificent: ^ 18:43:27 bombshelter13p_ [n=bombshel@24.114.234.33] has joined #lisp 18:43:42 BrianRice-mb [n=briantri@c-98-225-51-246.hsd1.wa.comcast.net] has joined #lisp 18:44:24 -!- bombshelter13p_ [n=bombshel@24.114.234.33] has quit [Client Quit] 18:44:52 drewc: I've added a documentation string to ensure that merges of repositories don't give invalid migrations (and well yes, I'm not a fan of objects where lists are more than sufficient) 18:44:57 jao [n=jao@9.Red-83-36-222.dynamicIP.rima-tde.net] has joined #lisp 18:44:57 tfb [n=tfb@restormel.cley.com] has joined #lisp 18:45:11 -!- scratch_ [n=scratch@adsl-232-6-140.asm.bellsouth.net] has quit [Read error: 110 (Connection timed out)] 18:45:22 allso, how much code do you want to write for a migration ?! 18:46:05 madnificent: you can't possibly be complaining that i have yet to include a macro for syntax, can you? 18:46:27 oh yes I am, that was the only thing I cared about :) 18:46:35 madnificent: jesus f'n c. 18:46:56 *drewc* slaps madnificent with a DEFMACRO 18:47:25 madnificent: your idea of a syntax might be significantly different from what i want, and my idea might change. 18:47:26 drewc: which shouldn't be written by the user 18:47:35 -!- dwave [n=ask@212251218147.customer.cdi.no] has quit ["Be back later"] 18:47:42 madnificent: no, it shouldn't. 18:48:02 madnificent: do you see the title of the paste? 18:48:36 in any case, it is quite trivial and it will probably work. I keep being surprised how much your ideas can differ from mine on essentially the same concepts :) (quite interesting) 18:48:47 madnificent: and, lists are not sufficient.. your migrations code is limited to a single database per image. 18:48:55 drewc: yes, and :) 18:49:07 madnificent: also, you use a bunch of global variables to keep state 18:49:30 drewc: I've been thinking of extending that (but hell, there are way more solutions for that than clos... 18:50:27 madnificent: what, exactly, is wrong with using CLOS? do you not notice that i actually make use of generic functions and specialization to allow for a more robust and flexible solution? 18:50:57 madnificent: do you think the designers of lisp included CLOS so you can go ahead and half-ass an object system for every trivial program>? 18:51:25 drewc: I don't think it is the solution to everything 18:51:28 and you think it is 18:51:36 Adamant [n=Adamant@c-76-29-188-22.hsd1.ga.comcast.net] has joined #lisp 18:51:38 madnificent: and you presume to speak for me now? 18:51:44 there is nothing wrong with that 18:53:08 spradnyesh [n=pradyus@117.192.6.17] has joined #lisp 18:54:11 madnificent: you seem to think macros are the solution to everything, and there is something wrong with that :) 18:54:46 *madnificent* hands drewc the winner he took back a bit earlier 18:57:17 f 18:57:20 oops 18:57:27 -!- chris2 [n=chris@p5B16A7DB.dip0.t-ipconnect.de] has quit ["Leaving"] 19:01:39 lhz [n=shrekz@c-b9aa72d5.021-158-73746f34.cust.bredbandsbolaget.se] has joined #lisp 19:04:02 HET2 [i=diman@xover.htu.tuwien.ac.at] has joined #lisp 19:04:20 -!- mokogobo [n=mokogobo@pcp075595pcs.unl.edu] has quit [Read error: 60 (Operation timed out)] 19:04:58 madnificent: no, really! (let ((version 1)) (upgrade version)) should work in my opinion. i'd also like to FUNCALL it. This, and not the lack of CLOS, is what made me reject your solution. 19:06:13 drewc: you've been handed your win. Please put an x on your geek card and be silent :) 19:06:18 -!- tfb [n=tfb@restormel.cley.com] has quit [] 19:06:31 madnificent: it's not a win for me if i can't use software you write. 19:07:15 drewc: btw: really, I'd 've been glad if I would've gotten some issues reported on it. It works now, and that was all I cared about. 19:07:44 technically, the wright flyer was an airplane. 19:08:16 Fade: tell me more about it ^_^ 19:08:17 madnificent: like i said, it was easier to re-write than to attempt to fix. my implementation took all of an hour. If you want a critisism i can give you one. 19:08:39 hrmn 19:10:50 manuel_ [n=manuel@213.144.1.98] has joined #lisp 19:11:01 -!- Fare [n=Fare@ita4fw1.itasoftware.com] has quit ["Leaving"] 19:11:06 drewc: really, what is your problem today? You seem tense and this isn't helping it I guess 19:11:31 drewc needs a hug. 19:11:47 -!- manuel_ [n=manuel@213.144.1.98] has quit [Client Quit] 19:12:55 -!- krumholt [n=krumholt@port-92-193-18-193.dynamic.qsc.de] has quit [Remote closed the connection] 19:13:52 -!- jao [n=jao@9.Red-83-36-222.dynamicIP.rima-tde.net] has quit [Read error: 110 (Connection timed out)] 19:14:39 athos [n=philipp@92.250.250.68] has joined #lisp 19:14:57 drewc: I hope I'm not stirring it up, but why wouldn't (let ((version 1)) (upgrade version)) work with my code? 19:14:58 madnificent: I spent a lot of time this last week dealing with a horrible POS code base from a young lisper who thought he knew better. I wish i could reach out and tell him "Don't use macros where functions will do, and don't re-invent the wheel", but he knows better, as he is me. 19:15:29 drewc, pont-of-sale? 19:15:37 piece of shit 19:15:47 dysinger [n=tim@cpe-75-80-200-182.hawaii.res.rr.com] has joined #lisp 19:15:50 what fade said. 19:16:06 drewc, on your crusade against macros still, eh? 19:16:13 drewc, so is with-foo really that bad, still? 19:16:18 tic_: no .. but : 19:16:18 a bridge with a shop on it? 19:16:35 http://blog.makezine.com/guitarampstorefront.jpg 19:16:37 _REPLeffect [n=REPLeffe@69.54.115.254] has joined #lisp 19:16:40 please lets be clear on what we are discussing here : 19:16:42 *tmh* just wrote a crapy, throw-away macro, with variable capture, no less :-O 19:16:49 http://github.com/madnificent/database-migrations/blob/bce7daeffb87bf4b03130a757dcf0c6beed10a04/migrations.lisp 19:17:09 madnificent: because you check against the literal number at compile time. 19:17:41 drewc, any specific function/macro? 19:17:50 tic_: upgrade/downgrade 19:18:50 -!- REPLeffect_ [n=REPLeffe@69.54.115.254] has quit [Read error: 113 (No route to host)] 19:19:13 schoppenhauer [n=css@unaffiliated/schoppenhauer] has joined #lisp 19:19:22 drewc: there are better ways of convincing people than to throw dirt around :). Many macro's I write are there to change later on in the process. I've written this to work in the repl (as I assumed that that would be what people would do). I still see no reason as to why the let wouldn't work in the repl :) 19:19:41 drewc: but yes, it could've been cleaner... and no, there really isn't a constructive reason to have that as a macro 19:19:57 Oh, now I see. Yeah, why make that a macro? 19:20:14 It doesn't make any sense whatsoever. 19:20:26 madnificent: i'm not 'throwing dirt', it's nothing personal. 19:20:30 tic_: quick write, no real reason besides "let's not think too much about it" 19:20:56 madnificent, but why? a function should always be the default choice. Is there anything you need to do in compile time w/ the function? 19:20:59 drewc: ah no, you were a bit harsh on your comments, that was the dirt (this macro-thing is quite constructive) 19:21:44 mrsolo [n=mrsolo@adsl-68-126-205-8.dsl.pltn13.pacbell.net] has joined #lisp 19:22:06 tic_: I probably found it easier to write what had to be executed in that case. Nothing constructive 19:22:30 *madnificent* is a bit surprised that he did that 19:22:35 madnificent: it was your code that was dirty.. i don't mean to be harsh, but your code is horrible. 19:22:58 i've written horrible code myself, and probably still do 19:23:50 madnificent: you asked me why i wrote yet-another-system rather than use yours, i iterated the reasons. 19:23:52 drewc: the global variables was something intentional. Rails allows you to set an environment, I wanted to be able to mimic that easily. (that accounts for the database parameter) 19:25:31 The-Kenny [n=moritz@p5087AD31.dip.t-dialin.net] has joined #lisp 19:25:51 hm.. how about passing the db as an optional parameter? 19:25:53 madnificent: my solution allows that as well, sans globals, by using :class allocated slots to store the environment for a certain 'class' of migrations. 19:26:04 tic_: see my paste :) 19:26:05 durka42 [n=durka@d81.wireless.swarthmore.edu] has joined #lisp 19:26:42 Oh, you have a paste? 19:26:56 *looking* 19:26:57 dropping the db environment on a class makes sense to me. 19:27:04 http://paste.lisp.org/display/79149 19:27:10 you could operate on more than one environment at the same time. 19:27:11 tic_: all possible extentions... I wanted to have the functionality I felt needed at the time 19:28:28 tic_: I assumed that people would've used a special binding to allow for that (as it was an implementation of an idea without too much thought (waiting for ideas to drop in) it felt good enough) 19:30:43 i would have used the MOP over the :class allocated slots if i needed a larger environment, or a customizable one. accessing :class slots requires an instance, which is sub-optimal. 19:30:53 mrsolo_ [n=mrsolo@nat/yahoo/x-e3d5e770336ca48b] has joined #lisp 19:31:11 drewc, now I see. I like your solution. It makes sense to put the data associated with the database in, well, the database. 19:31:40 tic_: I think that is the idea of migrations in general? 19:32:17 right. i actually plan to extend the database object somewhat and use it to describe a few other attributes of the DB that my application can use. 19:32:30 'database class' rather 19:32:57 classes are objects too, of course, but not here. 19:33:35 -!- drwhen [n=d@216-67-73-247-rb1.fai.dsl.dynamic.acsalaska.net] has quit [Remote closed the connection] 19:33:48 drwhat [n=d@216-67-73-247-rb1.fai.dsl.dynamic.acsalaska.net] has joined #lisp 19:34:03 drewc: you may still want to add a string in the table with the migrations that have been done. Version numbers may overlap as people are working on the project with multiple people 19:34:55 madnificent: i do 19:34:57 jkennedy_ [n=jkennedy@78-106-157-142.broadband.corbina.ru] has joined #lisp 19:35:10 oh, I didn't see you store in the database, sorry 19:35:10 -!- drwhat is now known as drwhen 19:36:43 Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has joined #lisp 19:36:59 Phoodus [i=foo@ip68-231-38-131.ph.ph.cox.net] has joined #lisp 19:37:13 -!- postamar [n=postamar@x-132-204-253-118.xtpr.umontreal.ca] has left #lisp 19:37:48 madnificent: even if i didn't, you can simply DEFMETHOD either on UPGRADE-DATABASE/DOWNGRADE-DATABASE, or (SETF DATABASE-SCHEMA-VERSION) 19:38:18 madnificent: you'll notice my ASSERT tests at the end do not even use a database. These are the reasons i like to use CLOS. 19:38:36 drewc: I store the several versions/texts in the database, as it may be that joins are made. In that case I don't want the user to have to revert the database back to that version 19:38:50 Davidbrcz [n=david@ANantes-151-1-34-107.w83-195.abo.wanadoo.fr] has joined #lisp 19:39:17 -!- mrsolo [n=mrsolo@adsl-68-126-205-8.dsl.pltn13.pacbell.net] has quit [Read error: 110 (Connection timed out)] 19:39:30 Question: If I'm building an image with save-lisp-and-die, I'm supposed to have my toplevel function either do its biz and exit, or start up the repl. What's the function to start the normal sbcl repl? 19:40:03 -!- jkennedy_ [n=jkennedy@78-106-157-142.broadband.corbina.ru] has left #lisp 19:40:04 rpg: i use (sb-impl::toplevel-init) 19:40:04 19:40:14 but that might not do what you want. 19:40:21 good starting point regardless. 19:40:29 drewc: BTW, Pascal Costanza at ILC gave an example of using class-prototype to access class allocated slots w/o making an instance... 19:40:35 drewc: thanks! 19:40:56 class-prototype is the basis of his singleton metaclass in contextl. 19:41:00 rpg: yeah, i know 19:41:09 you still need MOP for class-prototype though 19:41:20 Right. 19:41:21 if i was going to use MOP, i'd have used a metaclass 19:41:24 I find the code much harder to read, because I need to jump around rather much. But it seems to do it :) 19:42:07 drewc: we were looking at the tradeoff b/w having a slot on a metaclass and having class-allocated slots + class-prototype. I'm still not sure what's the right choice or even what are the parameters of the tradeoff. 19:42:09 and it is more extinsible 19:42:46 rpg: i'd prefer the metaclass simply because i get the accessors and initargs for free 19:43:59 i'd rather see (my-accessor (find-class 'foo)) than (my-accessor (make-instance 'foo)) for slots that are really prt of the class. 19:44:17 i suspect if MOP had have been part of ANSI, we'd not need :class 19:44:41 if you don't have an instance to hand, why aren't you using a special variable? 19:45:01 if you have an instance, then (my-accessor foo) is better than (my-accessor (class-of foo)) 19:46:29 Krystof: i don't have an instance just after defining the class, but i'd like to set the value of the :class allocated slots. 19:46:53 jao` [n=jao@131.Red-79-156-141.staticIP.rima-tde.net] has joined #lisp 19:47:03 -!- scottmaccal [n=scottmac@Sentry3.jay.k12.me.us] has quit ["Gone but not forgotten."] 19:47:05 I see this in some code that a colleague wrote for a test framework. It seems to take the place of java's static methods... 19:47:34 drewc: if you have an :initform, then the value will be set for you when you create the first instance 19:47:49 Krystof: are your commits done for today? and when do you plan to announce a freeze? 19:47:50 if you never read the value without an instance, then everything's cool 19:48:11 kreuter: I am done. I will announce a freeze now, to start in three hours' time (if that suits) 19:48:11 Krystof: will the value be set on the creation of every instance or just once? 19:48:14 drewc: I need to go back and try to grok more deeply the relationship between inheritance and metaclass... 19:48:19 drewc: just once 19:48:30 Krystof: well, it's because i'm an idiot then. 19:48:37 I think 19:48:42 -!- _REPLeffect [n=REPLeffe@69.54.115.254] has quit [Read error: 113 (No route to host)] 19:48:46 there is at least one of :initform and :default-initarg which does what you want to do 19:49:10 araujo [n=araujo@gentoo/developer/araujo] has joined #lisp 19:49:21 Krystof: thanks, i hope you're right! 19:49:45 clhs ensure-class 19:49:46 Sorry, I couldn't find anything for ensure-class. 19:49:49 damnit. 19:49:51 clhs defclass 19:49:52 http://www.lispworks.com/reference/HyperSpec/Body/m_defcla.htm 19:50:20 dwave [n=ask@212251218147.customer.cdi.no] has joined #lisp 19:50:33 Krystof: okay. I just did a 1.0.27.45 commit. I've got one more thing I'd like to commit this month (which bug addressed by the '.45 commit got in the way of). if I can't get it done in the next couple hours, it can wait. 19:50:54 well, I can give you 12 hours no problem 19:51:00 LostMonarch [n=roby@host205-182-dynamic.57-82-r.retail.telecomitalia.it] has joined #lisp 19:51:04 alright, thanks. 19:51:07 it's not like I'll be awake at midnight to frown at late commits or anything 19:51:13 :D 19:51:14 -!- Ragna [i=54a67a2c@gateway/web/ajax/mibbit.com/x-90deb61add58003f] has quit ["http://www.mibbit.com ajax IRC Client"] 19:51:31 -!- dwave [n=ask@212251218147.customer.cdi.no] has quit [Client Quit] 19:52:13 madnificent: what do you mean 'jump around' ... i wrote that code from the top of the file to the bottom, in order, and it's supposed to be read that way! :) 19:55:35 Is there some incantation that's necessary to make sbcl save its own code for meta-dot when building and installing? 19:56:07 lifeh2o [n=lifeh2o@119.152.67.124] has joined #lisp 19:56:26 drewc: trying to understand upgrade-database for instance 19:56:37 drewc: I can't find chapter and verse, but :initform appears to behave how I (and I think you) want it to 19:57:28 Krystof: i beleive it does as well, simply because a slot that is alread initialized won't be re-initialized, and a class slot will be initialized after the first instance is created. 19:57:29 oh, clhs 7.1 mentions it obliquely 19:57:34 yes, quite 19:58:02 rpg: perhaps http://paste.lisp.org/display/72423/ ? 19:58:23 madnificent: what isn't clear? 19:58:59 -!- durka42 [n=durka@d81.wireless.swarthmore.edu] has quit [] 19:59:04 Unit. Testing. Works. 19:59:05 -!- Davidbrcz [n=david@ANantes-151-1-34-107.w83-195.abo.wanadoo.fr] has quit [Remote closed the connection] 19:59:29 fe[nl]ix: hey, that's a neat trick you got there! 19:59:31 Just wanted to say that. It is painfully tedious, but it really helps work out the bugs. 19:59:32 drewc: jumping around doesn't imply unclear. But you only record the latest migration done? (with the :finally) 19:59:34 <_3b> tmh: if you have a spec meaybe :/ 19:59:45 drewc: what trick ? 20:00:09 fe[nl]ix: the logical-pathname-translations "SYS" trick 20:00:17 fe[nl]ix: Thanks! 20:00:30 sbahra [n=sbahra@pool-71-191-99-52.washdc.east.verizon.net] has joined #lisp 20:00:40 drewc: lol, I'm surprised you didn't know about it 20:00:48 _3b: Test what you can. My unit testing is by no means comprehensive, but I put what I can in place and add when new tests become obvious or necessary. 20:00:54 fe[nl]ix: there's a _lot_ i don't know about :D 20:01:21 <_3b> tmh: yeah, i should test more, just haven't been motivated enough to write a testing lib to test the stuff i want tested :) 20:01:42 mjf [n=mjf@r6y132.net.upc.cz] has joined #lisp 20:02:02 drewc: ah no, I see, you're looping over a list that has (foo . bar) items... (in any case, I found it confusing in my brain as the data is in multiple places (not necessarily a bad thing). 20:02:14 _3b: Well, it is damn tedious. But I've found errors that I would hate to have had to track down based on the use of the library in an application. 20:02:35 madnificent: that's called an «alist» 20:02:46 fe[nl]ix: Do I set those pathnames like that in the build process, or should I put them in my .sbclrc? And if the latter, should I just append them to the existing logical-pathname translations for SYS? 20:02:51 -!- Adamant [n=Adamant@unaffiliated/adamant] has quit [] 20:02:52 <_3b> tmh: i think running a compiler by hand is more than tedious for anything but the most trivial cases :) 20:03:05 madnificent: yes, i only set the schema version at the end. The actual implementation of the method can be more specific 20:03:09 fe[nl]ix: yes, true... 20:03:13 and set whatever you want in the datbase/ 20:03:52 <_3b> tmh: i want something where i can feed it a test case, and let it remember the results, and tell me if it changes 20:03:53 rpg: I have them in the sbclrc, and I don't quite understand what you mean by "appending" here 20:03:59 drewc: you store the fact that you've done a migration for each migration you do, right? 20:04:05 -!- jao` is now known as jao 20:04:08 <_3b> tmh: instead of having to generate the results by hand 20:04:57 madnificent: no, a schema table containg entried for version 1 and version 8 has absolutely gone through 2-7 in my use. 20:05:14 drewc: which you can not know between merges... 20:05:20 madnificent: I used CLOS though, so you are free to DEFMETHOD that to whatever behaviour you choose. 20:05:42 _3b: I'm lucky, if the results are too difficult to generate by hand, I have external tools I trust to generate the results. Why don't I just use the external tools? Well, they work well for generating unit test results, but not for my application. 20:05:46 madnificent: what do you mean? 20:06:04 drewc: say I have built migrations 4 through 9, and pull from someone that built 4 through 15 in the same time. I upgrade and only have his 10 through 15 in my database... isn't that a bit odd? 20:06:05 i'm new to the 'migrations' think.. i just did this with files before. 20:06:13 the SYS thing is even _documented_ 20:06:23 <_3b> tmh: yeah, that would help... do you have a testing lib that can compare trees without getting confused by gensyms? 20:06:58 drewc: I had an argument about it here with antifuchs (I think) 20:07:18 madnificent: i should hope you can communicate with your team do decide on version numbers before checking in to HEAD. 20:07:35 drewc: even rails doesn't do that anymore :) 20:07:37 madnificent: if not, well, it's a DEFMETHOD away innit. 20:07:43 _3b: I do not. I asked a similar question about checking a macro form here a while ago, but with a little help realized that test didn't buy me anything. 20:07:55 madnificent: 'even rails' .. like rails is a good example of how to do things? :) 20:08:08 drewc: it is if it does things better :P 20:08:15 <_3b> tmh: yeah, i think i was the one suggesting hacks to deal with it :) 20:08:20 fe[nl]ix: If I just setf the logical pathname translations like that, I lose the old definition that points at /usr/local/lib/sbcl... 20:08:25 madnificent: so, what is the proper solution in this case then? 20:08:36 <_3b> tmh: more useful in my case, since i'm actually validating the code doing the expansion 20:09:07 _3b: Ah, yeah, it's all a blur. :-) Why don't you just (let ((*gensym-counter* 0)) ...) or similar? 20:09:10 there is a consensus of using the unix system time as a migration number in order to avoid most of it. However, you must still ensure that all numebers have been executed (not only the last ones). I've added the comment string in the database too, just to avoid a bug in which two people created the same migration with the same system time (really, you don't want to search that bug) 20:09:42 rpg: that's the whole point: you either copy the sbcl tree to /usr/local/lib/sbcl or overwrite the pathname translation to point to another path 20:09:46 drewc: for each migration that is ran, you can store the number in the database. If it hasn't been ran yet, you execute it :) 20:09:57 rpg: http://www.sbcl.org/manual/Lisp-Pathnames.html#Lisp-Pathnames 20:09:58 <_3b> tmh: wouldn't be much harder to do it correctly, most of the work would be the infrastructure to make it actually easy to use, which is why i haven't bothered yet 20:10:20 drewc: rails migrations have gone through the same process btw. They initially used what you've created now :) 20:10:36 Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has joined #lisp 20:10:37 madnificent: ok .. add an :after on migration-upgrade-function, done. ;) 20:10:57 drewc: you must do that, not your user ;) 20:11:15 madnificent: why? 20:11:21 _3b: Do you have any experience with LISP-UNIT? 20:11:48 because he shouldn't have to care about these issues... KISS (for) the user 20:11:51 besides that: nvm 20:11:51 blitz__ [n=blitz@2001:6f8:10f6:200:21b:77ff:fe41:11ab] has joined #lisp 20:12:03 madnificent: providing a flexible system that can meet many users needs is way more important to me than making sure i cover every possible use case. 20:12:19 -!- blitz__ [n=blitz@2001:6f8:10f6:200:21b:77ff:fe41:11ab] has quit [Read error: 54 (Connection reset by peer)] 20:12:32 madnificent: keep in mind that i agree your way is a better way to do it. 20:12:33 <_3b> tmh: nope, never really picked up the unit testing frame of mind :( 20:12:40 *madnificent* prefers not having to write things the author should've given me 20:12:47 s/me/him :( 20:12:52 I'm a bit ill 20:12:58 blitz__ [n=blitz@2001:6f8:10f6:200:21b:77ff:fe41:11ab] has joined #lisp 20:13:05 -!- blitz_ [n=blitz@cl-76.dus-01.de.sixxs.net] has quit [Read error: 101 (Network is unreachable)] 20:14:25 Krystof, fe[nl]ix: ah. So you are saying that the only purpose of SYS: is to find the sources. I believe in ACL it's used for other purposes as well, so I have on occasion added to the sys: lpts to add or shadow, but have never just overwritten them. 20:15:02 the only bits currently used are "SRC;**" and "CONTRIB;**" 20:15:12 _3b: I asked because it has ASSERT-EXPANDS, but I never use that form so I'm not sure that it handles GENSYMS properly, although I doubt it does. Give LISP-UNIT a look/try and see if it fits your work flow. If it does, we could add another assertion form to handle GENSYMS. 20:15:14 and only for finding sources 20:15:26 -!- younder [n=jthing@084202013251.customer.alfanett.no] has left #lisp 20:15:40 ryepup [n=ryepup@one.firewall.gnv.acceleration.net] has joined #lisp 20:16:02 -!- deech [n=deech@71-10-166-226.dhcp.stls.mo.charter.com] has quit [Read error: 104 (Connection reset by peer)] 20:16:13 -!- spradnyesh [n=pradyus@117.192.6.17] has quit [Connection timed out] 20:16:17 <_3b> tmh: pretty sure it didn't last i checked, but i have unusual ideas about how workflow should be (unfortunately ideas i don't generally have time to realize) 20:16:44 deech [n=deech@71-10-166-226.dhcp.stls.mo.charter.com] has joined #lisp 20:16:44 we probably reserve the right to do other stuff, but you can replace any existing "SRC" and "CONTRIB" things to point to the system sources 20:16:44 and in practice I would expect any sbcl for the forseeable future to have no other dependence on the SYS host 20:16:51 Krystof: I also see "OUTPUT;**;*.*.*" here 20:17:05 spradnyesh [n=pradyus@117.192.3.9] has joined #lisp 20:17:08 <_3b> tmh: currently i just type in ;;(test form) C-x C-e 20:17:15 <_3b> tmh: i want a testing lib that is that easy to use :) 20:17:40 <_3b> tmh: preferably with dependency tracking, so it can run relevent tests whenever i C-c C-c a function or whatever 20:18:09 <_3b> (though that one might not actually work out in practice, would be annoying for refactoring and such) 20:18:24 Krystof, fe[nl]ix: At any rate, I think it's (possibly over) safe to append (prepend) those two logical pathname translation lists to whatever is the current set of lp translations for SYS.... 20:18:32 _3b: I understand and can relate. After I offered to that, I thought to myself, "Add that to the list of things that need to be updated in the documentation." Now I don't need to. :-) 20:18:34 fe[nl]ix: ah yes, that's new 20:18:41 *_3b* also wants revision control at the level of top-level forms, with automatic checkins at close to C-c C-c level as well 20:18:52 I'll try to update the documentation 20:20:05 -!- mjf [n=mjf@r6y132.net.upc.cz] has quit ["Lost terminal"] 20:23:13 _3b: Those aren't unreasonable features for testing, although I'd want that kind of integration handled by the development environment. So, when you C-c C-c, it compiles the function, runs the test, if the test passes, check it in and ask for a log message, otherwise, prompt you with the error. The only caveat I have with the automagic check-in is that an update may not be confined to one function. 20:24:12 <_3b> tmh: i don't mean at the log message level, more of a premanent undo function :) 20:24:17 Thanks to everybody on the channel, we've got a version of Franz' generate-application that works on SBCL as well as ACL... 20:24:17 A minor oddity, though, when I run it, I see it handling a very large number of SIGHUPs. Does SBCL do something internally with these signals? 20:24:50 rpg: yes, but I think that the point is that you *should* overwrite those L-P-Ts 20:24:58 <_3b> tmh: real checkings with log messages would still be more manual... though i guess they would be reasonable if the # of passing tests increases 20:25:05 madnificent: then, does it make sense at all to number migrations? It seems to me that naming them and having an explicit :depends-on would be a lot cleaner and have less room for error. 20:25:25 fe[nl]ix: If I just prepend, then I shadow them, and any later additions to that lp host will "shine through." 20:25:28 _3b: Ah, I see. Sort of a repository cache? 20:25:46 what else should be made better? (besides the two parameters, about which I want to think longer) http://github.com/madnificent/database-migrations/blob/9492119652d3055006470494571c69f3bb642cd3/migrations.lisp 20:25:49 Repository purgatory? 20:25:57 <_3b> tmh: just part of the super smart editor/ide setup i want :) 20:26:16 drewc: yes it does. The numbering ensures that the order in which you're executing them is valid 20:26:26 <_3b> tmh: forms in git (or whatever) instead of files on filesystem... no manual dependencies anywhere, etc 20:26:29 madnificent: so do explicit dependencies 20:26:59 drewc: you can indeed make it better with a correct depends. But you'll still have to ensure that you don't give nicknames someone else in the team has been using. (which in essense results in the same problem) 20:27:45 that's a different problem... i don't worry about using the same function names that others might be checking in... this is no different 20:27:54 drewc: allso, explicit dependencies are harder to use as a user. Inserting the systemtime is a quick way of working 20:28:18 I use a description + a number, that seems fairly safe to me 20:28:20 madnificent: how is it harder? 20:28:23 gotta go (dinner) 20:28:34 _3b: I think I understand. 20:29:03 Could my SIGHUP thing be a MacOS issue? 20:29:12 drewc: because you have to search for each dependency you care about :) and you have to remember the name (which means extra scrolling around, which in turn takes relatively much time (its an action you could skip if you wanted to)) 20:29:23 you need to search for the number .. non 20:29:24 ? 20:29:45 if the current schema version is 10, you can't number your migration 5 and expect it to work? 20:29:46 rpg: no, there was no notion of signal in MacOS. 20:30:12 pjb: No notion of signal? I can use kill... 20:30:20 it's just an implicit dependency on the previous version. 20:30:21 rpg: not on MacOS. 20:30:28 Perhaps on MacOSX, but that's a different OS. 20:30:53 MacOS did too have a notion of kill. Force Quit is basically the same thing, and was quite useful. :) 20:30:57 madnificent: since you must know what the previous version is regardless, it doesn't matter if it's a name or a number. 20:31:21 -!- xan-afk is now known as xan 20:31:22 Oh, sorry. I'm a newcomer to the Mac, so I'm not sensitive to that distinction. 20:31:59 Anyway, I'm running sbcl on my Mac, and I'm seeing a lot of SIGHUPs (I have a handler defined for SIGHUP). I'm not sending them, so I wondered what was going on. 20:32:04 The two OSes are quite different. MacOSX is unix-like. 20:32:23 pjb: and its predecessor isn't a real OS ;-) 20:32:26 Force Quit on MacOS was so much better than OSX's force quit. Now it takes about 5 minutes to come up whenever I want it (which is usually because some app is putting my system into swap-hell) 20:32:56 I guess they didn't consider pinning it in memory a good idea. 20:33:34 rpg: unfortunately, it's rather difficult, if not impossible, to determine where signals come from. Perhaps it's related to the memory management? 20:34:15 pjb: that would be a different signal, I guess 20:34:29 Yes, SIGHUP is surprising. 20:35:41 We have an application that wants to throw SIGHUP at our lisp code, and have the lisp code write a status message in response, so it can monitor us being up. 20:36:18 -!- Jasko [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has quit [Read error: 110 (Connection timed out)] 20:37:32 Perhaps there's some library that asks the notification daemon to send it the SIGHUP? http://developer.apple.com/documentation/Darwin/Conceptual/MacOSXNotifcationOv/DarwinNotificationConcepts/DarwinNotificationConcepts.html 20:37:35 drdo [n=psykon@167.111.54.77.rev.vodafone.pt] has joined #lisp 20:38:24 -!- oudeis [n=oudeis@bzq-82-81-239-4.cablep.bezeqint.net] has quit ["This computer has gone to sleep"] 20:38:27 Ah. It's a known Mac OSX bug. Kill sends SIGHUP + signal when sending signal. 20:39:07 http://bugs.mysql.com/bug.php?id=14164 20:39:10 OMG 20:39:16 how comes that Mac OSX is so buggy wrt signals? 20:39:32 because it's not really a Unix kernel 20:39:47 it's a pseudo-unix kernel emulation layer running on top of a Mach-like microkernel 20:39:47 The kernel is not unix, it's Mach. Unix is "simulated" over the microkernel... 20:39:51 *p_l* had just finished reading up on those notifications and wonders who was so braindamaged to make notifyd send SIGHUP 20:40:13 pjb: In OSX they simply ended shoehorning FreeBSD into Mach's kernel space 20:40:45 just look how long it took Apple to fix the "SBCL crashes all the time" problem 20:41:05 Officially it's XNU or something like that and is supposed to be "descendant, bu not exactly Mach" 20:41:20 I could rant for a good long while about this, but I'll go to sleep instead 20:41:24 upon looking through Darwin's sourcecode, I found "OSF/1" :> 20:41:41 there's a fun OS 20:41:45 -!- ryepup [n=ryepup@one.firewall.gnv.acceleration.net] has left #lisp 20:42:11 except that the only real OSF/1 implementation I heard of didn't have so many problems with signals 20:44:48 Fortunately, I develop on SBCL, but deliver on linux, so this goes into the very large pile of things that I can Not Care About. 20:44:55 s/SBCL/MacOSX/ 20:45:24 I don't see that behavior on my version of OSX 20:45:27 -!- tcr [n=tcr@host145.natpool.mwn.de] has quit ["Leaving."] 20:45:37 kill -INT pid only the SIGINT handler gets called 20:46:13 stas@slack: /usr/src/linux-2.6.29.1% grep "OSF/1" -R . | wc -l => 49 20:46:15 10.5.6 20:46:19 Faed [n=quassel@206-248-163-169.dsl.teksavvy.com] has joined #lisp 20:48:38 heh. OSF/1 was probably the only successful Mach-based Unix 20:48:53 and only in one implementation :P 20:49:23 i liked it quite a bit by the time it was tru64. 20:51:04 -!- Faed [n=quassel@206-248-163-169.dsl.teksavvy.com] has quit [Read error: 104 (Connection reset by peer)] 20:51:15 -!- white-rabbit-obj [n=cpc5@cpe-67-241-171-153.buffalo.res.rr.com] has quit ["Leaving."] 20:51:17 bah. graphical irc clients are bad. 20:51:53 -!- cipher [n=cipher@pool-173-48-136-239.bstnma.fios.verizon.net] has quit [Remote closed the connection] 20:54:33 impulse32 [n=impulse@CPE00236916024d-CM001ac3167610.cpe.net.cable.rogers.com] has joined #lisp 20:54:47 pstickne [n=pstickne@69.166.35.201] has joined #lisp 20:54:55 I wonder how far nyef got with his axp emulator. 20:55:37 Fade: He was hitting hw init last time I heard of it 20:55:46 neat 20:55:58 test case was SRM :) 20:56:08 yeah, well. :) 20:56:58 I wonder however if the updated code is available somewhere 20:57:09 i'd be surprised if it was literally nowhere. 20:57:22 I'd like to use as inspiration for my MMIX simulator 20:57:27 there's a lot of legacy gear out there that people would be interested in running. 20:58:27 there was even a project to rebuild Alpha in FPGA 20:58:47 -!- mgr [n=mgr@psychonaut.psychlotron.de] has quit [Read error: 104 (Connection reset by peer)] 20:59:32 -!- spradnyesh [n=pradyus@117.192.3.9] has quit [Connection timed out] 21:00:25 spradnyesh [n=pradyus@117.192.3.253] has joined #lisp 21:01:24 tcr [n=tcr@host145.natpool.mwn.de] has joined #lisp 21:01:54 -!- dlowe [n=dlowe@ita4fw1.itasoftware.com] has quit ["Leaving."] 21:02:32 -!- Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has quit [Read error: 113 (No route to host)] 21:04:02 foom: I'm not sure that's the bug. I do know that I'm seeing a lot of SIGHUPs that I'm not generating. 21:04:28 -!- frank_s_ [n=FrankS@41.145.115.144] has quit [No route to host] 21:04:55 Jasko [n=tjasko@c-98-235-105-148.hsd1.pa.comcast.net] has joined #lisp 21:05:10 -!- anekos is now known as awayekos 21:06:11 thom_logn [n=thom@pool-96-229-99-100.lsanca.dsl-w.verizon.net] has joined #lisp 21:06:32 mgr [n=mgr@psychonaut.psychlotron.de] has joined #lisp 21:06:49 dwave [n=ask@212251218147.customer.cdi.no] has joined #lisp 21:06:58 -!- rpg [n=rpg@97-127-65-227.mpls.qwest.net] has quit [] 21:08:17 -!- dwave [n=ask@212251218147.customer.cdi.no] has quit [Client Quit] 21:09:30 -!- beach` is now known as beach 21:10:00 Good evening. 21:13:54 good evening 21:15:36 -!- LostMonarch [n=roby@host205-182-dynamic.57-82-r.retail.telecomitalia.it] has quit ["raise RuntimeError"] 21:17:41 s0ber_ [n=s0ber@118-168-236-19.dynamic.hinet.net] has joined #lisp 21:23:14 -!- pstickne [n=pstickne@69.166.35.201] has quit [Read error: 110 (Connection timed out)] 21:23:43 -!- xan [n=xan@cs78225040.pp.htv.fi] has quit ["Lost terminal"] 21:24:48 drdo` [n=psykon@167.111.54.77.rev.vodafone.pt] has joined #lisp 21:24:49 Samy [n=sbahra@c-76-26-144-229.hsd1.dc.comcast.net] has joined #lisp 21:25:31 -!- mega1 [n=mega@53d8275f.adsl.enternet.hu] has quit [Read error: 110 (Connection timed out)] 21:28:22 oudeis [n=oudeis@bzq-82-81-239-4.cablep.bezeqint.net] has joined #lisp 21:29:35 -!- s0ber [i=pie@118-168-239-154.dynamic.hinet.net] has quit [Connection timed out] 21:29:45 younder [n=jthing@084202013251.customer.alfanett.no] has joined #lisp 21:31:31 -!- tcr [n=tcr@host145.natpool.mwn.de] has quit ["Leaving."] 21:31:35 kpreid [n=kpreid@216-171-189-59.northland.net] has joined #lisp 21:33:16 drewc: I name them the unix system time, then you don't have to search, you only have to see how the database currently looks (but that is something that you generally know) 21:34:17 -!- LiamH [n=none@common-lisp.net] has quit ["Leaving."] 21:34:47 drewc: if you're going to number manually then that is possible. It still leaves you to thing on what you must base yourself (not really an issue, as it generally doesn't matter that much) 21:34:47 madnificent: you name them the unix system time? like (defmigration 123456789 ...)? 21:35:06 drewc: yup (there is an example in blogworks, I think) 21:35:37 ok, that would work. 21:35:46 drewc: that is what rails currently does too (it is a bit uglier there, as the version and the name of the migration is in the filename there) 21:35:48 ugly though. 21:36:11 tombom [n=tombomp@wikipedia/Tombomp] has joined #lisp 21:36:37 if they are ordered correctly in the file, it's quite okay. You can make it an explicit tree if you want. It could be an interesting tryout 21:36:46 you still have to look up the system time though.. yeah? 21:37:15 so (get-name-of-database-version) vs (get-universal-time) 21:37:26 and implicit dependencies vs explicit. 21:37:31 -!- hkBst [n=hkBst@gentoo/developer/hkbst] has quit [Read error: 104 (Connection reset by peer)] 21:38:06 I use M-1 M-! time +%s and I have the system time pasted in there. My upgrade function simply upgrades everything 21:38:49 I'm trying to use STEP in SBCL+SLIME. I choose the option to step into the function and all it does is finish and return. Is there something I'm missing or is this as good as it gets on SBCL? 21:38:50 what about clock skew between network-distant hosts? 21:39:12 -!- sbahra [n=sbahra@pool-71-191-99-52.washdc.east.verizon.net] has quit [Read error: 110 (Connection timed out)] 21:39:14 -!- drdo [n=psykon@167.111.54.77.rev.vodafone.pt] has quit [Read error: 110 (Connection timed out)] 21:39:14 tmh: compiled with high debug? 21:39:40 stassats: I have no declarations for debug, whatever the default is. 21:39:50 default is 1 21:40:03 Ah, thanks 21:40:16 drewc: again, feel free to try it with a completely defined tree, it may yield interesting results (as in: you could be able to revert only a partial branch etc) 21:40:57 http://www.sbcl.org/manual/Debugger-Policy-Control.html#Debugger-Policy-Control 21:41:07 If debug is greater than all of speed, space and compilation-speed the code will be steppable 21:41:09 -!- ignas [n=ignas@ctv-79-132-160-221.vinita.lt] has quit [Read error: 110 (Connection timed out)] 21:41:13 madnificent: damnit .. you just had me convinced to do it with numbers as i saw no advantage to tags... 21:41:24 then you give me an advantage to tags! 21:42:02 ok, explicit tags it is. seems more lispy to use symbols anyway. 21:42:12 kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 21:42:20 drewc: do signal a warning when a duplicate tag has been used! 21:43:16 -!- tombom [n=tombomp@wikipedia/Tombomp] has quit ["Peace and Protection 4.22.2"] 21:43:28 i wonder if slime could detect such cases and notify about them 21:43:33 -!- Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has quit [Remote closed the connection] 21:44:35 madnificent: oh yes, i will :) 21:45:08 -!- spradnyesh [n=pradyus@117.192.3.253] has quit [Read error: 110 (Connection timed out)] 21:45:56 spradnyesh [n=pradyus@117.192.3.229] has joined #lisp 21:46:19 -!- oudeis [n=oudeis@bzq-82-81-239-4.cablep.bezeqint.net] has quit ["This computer has gone to sleep"] 21:46:52 frank_s [n=FrankS@41.145.115.144] has joined #lisp 21:47:43 -!- impulse32 [n=impulse@CPE00236916024d-CM001ac3167610.cpe.net.cable.rogers.com] has quit [Read error: 113 (No route to host)] 21:49:34 white-rabbit-obj [n=cpc5@cpe-67-241-171-153.buffalo.res.rr.com] has joined #lisp 21:50:06 -!- frank_s [n=FrankS@41.145.115.144] has quit [Client Quit] 21:50:12 frank_s [n=FrankS@41.145.115.144] has joined #lisp 21:51:11 -!- sellout [n=greg@guest-fw.dc4.itasoftware.com] has quit [] 21:51:42 I didn't notice until I needed a slightly altered reader that backquoted expression get read in an implementation-defined way. eg (first (read-from-string "`(a ,b)")) => SB-IMPL::BACKQ-LIST 21:52:07 dreish [n=dreish@minus.dreish.org] has joined #lisp 21:52:24 You know it's time for a break when there is a glaring obvious error in your code and you resort to (debug 3), STEP and BREAK to find it. 21:54:38 karvus [n=thomas@193.213.35.168] has joined #lisp 21:56:04 And I knew this about an hour ago when I first encountered the error. 21:57:36 -!- hugod [n=hugod@modemcable086.138-70-69.static.videotron.ca] has quit [] 21:57:55 tmh: ouch. 22:01:26 -!- HG` [n=wells@91.111.13.16] has quit [Client Quit] 22:03:47 metasyntax [n=taylor@pool-71-127-85-87.aubnin.fios.verizon.net] has joined #lisp 22:04:13 plutonas [n=plutonas@c-83-233-152-13.cust.bredband2.com] has joined #lisp 22:04:55 -!- kib2 [n=kib2@bd137-1-82-228-159-28.fbx.proxad.net] has quit ["When two people dream the same dream, it ceases to be an illusion. KVIrc 3.4.2 Shiny http://www.kvirc.net"] 22:10:24 nvoorhies_ [n=nvoorhie@adsl-75-36-207-191.dsl.pltn13.sbcglobal.net] has joined #lisp 22:10:44 -!- nvoorhies [n=nvoorhie@adsl-75-36-207-191.dsl.pltn13.sbcglobal.net] has quit [Read error: 54 (Connection reset by peer)] 22:10:48 -!- Fufie [n=innocent@86.80-203-225.nextgentel.com] has quit ["Leaving"] 22:10:52 alec [n=alec@pool-96-233-16-99.bstnma.east.verizon.net] has joined #lisp 22:10:54 JAS415 [n=jon@c-24-34-16-25.hsd1.ma.comcast.net] has joined #lisp 22:11:09 ejs [n=eugen@196-194-124-91.pool.ukrtel.net] has joined #lisp 22:11:59 octoberdan [n=user@c-65-96-169-141.hsd1.ma.comcast.net] has joined #lisp 22:12:04 Hello everyone :-) 22:12:21 tomoyuki28jp [n=tomoyuki@w221062.ppp.asahi-net.or.jp] has joined #lisp 22:12:47 Is there a way to write this in a shorter way? (defmacro mac (&optional x) (if x `(fn ,x) `(fn))) 22:12:53 . 22:13:17 (defmacro m (&optional x) (if x `(fn ,x) `(fn))) 22:13:26 That'll be two characters, please. 22:14:19 drewc: are you two still debating the merits of numbers vs. date-named db migrations? 22:14:29 Is there a cleaner way of saying... (if (> a b) a b) 22:14:31 (defmacro mac (&optional x) `(fn ,@(mklist x))) if you have MKLIST defined in the usual way. 22:14:52 octoberdan: (max a b) 22:14:58 For the win! :-) 22:15:11 pjb: Thank you. I appreciate your help 22:15:18 drewc: my reason for using the date kind is that after a merge from a co-worker's tree, I don't have to renumber an unknown number of migrations; instead, I just need to test whether the current order still works. 22:16:01 tomoyuki28jp: (defmacro mac (&optional x) `(fn . ,x)) 22:16:37 fe[nl]ix: eh? (mac 10) => (fn . 10) 22:16:50 s/&optional/&rest/ 22:16:51 &rest instead of option? 22:16:58 damn 22:17:12 (defmacro mac (&rest x) `(fn ,@x)) 22:17:26 tee hee 22:17:30 antifuchs: oh yeah, it seems like the best. 22:17:53 thanks you, guys. 22:17:59 s/thanks/thank/ 22:18:02 but then you can just use (fn x) or (fn) anyway... 22:18:43 -!- schoppenhauer [n=css@unaffiliated/schoppenhauer] has quit ["leaving"] 22:19:28 antifuchs: I don't want the mac returns (fn nil) 22:20:15 ooh, ok, that makes sense 22:20:42 -!- athos [n=philipp@92.250.250.68] has quit [Remote closed the connection] 22:20:50 tomoyuki28jp: you mean you want (mac nil) => (fn) 22:21:02 as well as (mac) => (fn) and (mac 10) => (fn 10) 22:21:57 oh, good point. In this case, yes > (mac nil) => (fn) 22:22:10 Nowis there a way to simplify (append (max a b) (min a b) rest)? 22:22:19 Then my version with MKLIST does it. Where mklist is (defun mklist (x) (if (listp x) x (list x))) 22:22:40 then I think it could return `(apply #'fn ',x) with x being a rest arg 22:22:54 gigamonkey: or alexandria:ensure-list 22:23:04 octoberdan: append needs lists, and max returns integers 22:23:05 (append (sort (list a b) #'>) rest) 22:23:08 fe[nl]ix: yeah, because that's a lot shorter. ;-) 22:23:18 antifuchs: fn might be also a macro. 22:23:42 tomoyuki28jp: hmmm, so I think your first version with the IF was pretty much optimal 22:25:31 except, it should read `(if ,x (fn ,x) `(fn)) or better, (let ((#1=#:VALUE ,x)) `(if ,#1# (fn ,#1#) (fn))) ; (substitute a gensym for #1=:VALUE) 22:25:46 antifuchs: that much is not in debate.. now the argument was to bother numbering them at all, or to just name them and have explicit :depends-on links 22:25:51 antifuchs: In my case, actually this one works just fine. (defmacro use-mac (&rest args) `(mac ,@args)) 22:25:52 antifuchs: does that even work? (mac z) turns into (apply #'fn '(z)) 22:25:57 antifuchs: I believe I follow the logic, but I'm a bit confused by the #'>. Is that passing in a function for sort to use to compare? 22:26:19 -!- jajcloz [n=jaj@pool-98-118-118-197.bstnma.fios.verizon.net] has quit [] 22:26:20 I'm probably borderlining on being kicked with these lame questions 22:26:22 Say you have (let ((z 10)) (mac z)) 22:26:25 octoberdan: yes 22:26:43 Which results in applying #'fn to the symbol Z not the value 10. 22:26:44 octoberdan: that's a good question, don't worry 22:27:08 oudeis [n=oudeis@77.124.71.204] has joined #lisp 22:28:13 So #' followed by a symbol converts the symbol into a function? 22:28:37 hmm, I think my vocab is off 22:28:44 octoberdan: itdepends on the criteria. If you want to simplify the source, append+sort might be good. If you want to simplify the generated code, (if (< a b) (list* b a rest) (list* a b rest)) might be better. 22:28:59 -!- Jacob_H [n=jacob@92.4.155.143] has quit ["Leaving"] 22:29:02 -!- ikki [n=ikki@201.155.75.146] has quit [Read error: 113 (No route to host)] 22:29:35 -!- willb [n=wibenton@wireless74.cs.wisc.edu] has quit ["Leaving"] 22:29:43 octoberdan: yes 22:29:58 octoberdan: #' is a dispatching reader macro. #'x reads as (function x). function is a special operator that returns the function whose name is given as argument, like quote is a special operator that return its argument. 22:30:00 #'name is like writing (function name) 22:30:12 manuel___ [n=manuel@HSI-KBW-078-043-184-124.hsi4.kabel-badenwuerttemberg.de] has joined #lisp 22:30:14 Ah! Okay 22:30:27 -!- fiveop [n=fiveop@pD9E6C1DA.dip.t-dialin.net] has quit ["humhum"] 22:31:02 drewc: oh, I missed that part of the discussion. 22:33:23 -!- The-Kenny [n=moritz@p5087AD31.dip.t-dialin.net] has quit ["I am Locutus of Borg! You will assist us!"] 22:33:24 drewc: sounds like one way to get an ordering, but I'm not sure this is really necessary 22:35:02 you're very probably not going to do them in parallel, and very often I want one definite order in which these things get executed 22:35:25 -!- mattrepl [n=mattrepl@129.174.69.135] has quit [] 22:35:50 -!- lhz [n=shrekz@c-b9aa72d5.021-158-73746f34.cust.bredbandsbolaget.se] has quit ["Leaving"] 22:36:50 what I'm saying is, I can't think of a case where this is what I'd want to do (: 22:37:07 ("this" ... "write depends-on lists") 22:37:39 vs. having things happen in the order they are defined (or explicitly numbered, I don't know) 22:37:45 -!- phadthai [i=mmondor@ginseng.pulsar-zone.net] has quit [Read error: 113 (No route to host)] 22:40:55 -!- manic12 [n=manic12@c-76-29-88-103.hsd1.il.comcast.net] has quit [Read error: 110 (Connection timed out)] 22:42:29 sylar: maybe unstable or experimental or testing branch 22:42:36 uuups 22:44:23 -!- jao [n=jao@131.Red-79-156-141.staticIP.rima-tde.net] has quit [Connection timed out] 22:45:37 "order they are defined" .. i considered that and threw it away outright .. i can't remember quite why now. 22:48:03 antifuchs: any good reason why the order of execution can't be the order of definition? Is there a hairy case when merging that i can't think of right now? 22:48:43 ok, I can only speak for the rails approach here (which has worked well for me in the past, so bear with me) 22:48:50 they use separate files, one per migration 22:49:04 and that causes zero merge conflicts (: 22:49:17 how are the files ordered? 22:49:27 is there an asdf-alike that handles builds? 22:49:29 by file name, the prefix is the iso date 22:49:40 well, there's rake, which is a make alike thing 22:49:53 (it's a bad hack, but it does the job) 22:50:12 -!- dcrawford [n=dcrawfor@ita4fw1.itasoftware.com] has quit [Read error: 54 (Connection reset by peer)] 22:50:38 ok, so what if i have ASDF handle the ordering. 22:50:44 Adamant [n=Adamant@c-76-29-188-22.hsd1.ga.comcast.net] has joined #lisp 22:50:50 -!- sepult [n=buggarag@xdsl-87-78-121-209.netcologne.de] has quit ["leaving"] 22:51:05 I'd ask why 22:51:18 you can achieve the same effect from the output of DIRECTORY, and ordering by :name 22:51:37 not if i don't include the date in the name of the file.. 22:51:47 with /far/ less maintenance overhead from writing :depends-on clauses and doing manual file merges on conflicts 22:52:16 chessguy_work [n=chessguy@67-130-43-2.dia.static.qwest.net] has joined #lisp 22:52:27 there is little overhead in the depends on .. it's just the _last_ version.. i should have called it :supercedes 22:52:35 -!- chessguy_work is now known as chessguy 22:52:51 what's the order for migrations that supersede the same migration? 22:53:10 unspecified. 22:53:21 oh well 22:53:47 I suppose you should try how it works for you... I suppose I have my migrations setup already (it's all of three functions or so (-;) 22:54:09 how does having a date fix that same problem? 22:54:39 davazp` [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has joined #lisp 22:54:44 it provides an ordering that will not cause unspecified collisions in an overwhelmingly large amount of cases 22:54:56 if i am developing on top of version 01, and we both check in migrations from that version (dated), it is exactly he same issue. 22:55:05 it's not version 01 22:55:13 whatver, version 2009-01-01\ 22:55:15 it's version 20090422 22:55:22 actually, no, not even that. 22:55:25 it's version 20090422220203 22:55:31 right.. 22:55:57 so you want to check in a new migration.. at the same time as somebody else. 22:56:08 right. 22:56:16 "overwhelmingly large amount of cases" 22:56:22 this is not a problem in practice 22:56:22 -!- davazp [n=user@56.Red-79-153-148.dynamicIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)] 22:56:28 you both worked from (so supersede) 20090422220203 22:56:43 antifuchs: of course it's not, that's my point. 22:57:01 same for an explicit :depends-on :previous-version-name 22:57:08 oh yes, there it is 22:57:14 -!- gemelen [n=shelta@shpd-78-36-166-34.static.vologda.ru] has quit ["I wish the toaster to be happy, too."] 22:57:27 it's easy for two people to create a migration that have the same parent and then merge them 22:57:57 hugod [n=hugod@bas1-montreal50-1279442133.dsl.bell.ca] has joined #lisp 22:58:00 regardless of the method used to name and order them. 22:58:02 it's very much not easy for two people to accidentally create a migration at the same time and then not test whether their code still works after a merge 22:58:25 indeed. 22:58:47 the date-based approach has the advantage that it suggests an ordering that will probably work by default, and if it doesn't, you can easily fix it 22:59:27 no, i don't see the difference between changing a date, and changing a :depends-on label. 22:59:32 -!- davazp` is now known as davazp 22:59:53 i think the only thing a date buys is the implicit ordering... 22:59:56 the supersede approach needs manual intervention at every merge step that involves two migrations with the same parent, and has possibly-disastrous consequences if lazy developers rely on their unspecified order to be the one on the production system 23:00:14 does it? 23:00:38 i'm willing to be that most merges from the same parent don't rely on eachother, and won't interfere with eachother. 23:00:45 i can specify an order if you like.. 23:00:53 I've seen too many :depends-on problems in asdf to ever again believe it will solve anything related to deterministic orders (: 23:01:01 alphabetical from symbol name :) 23:01:06 -!- Yuuhi [i=benni@p5483EAC3.dip.t-dialin.net] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 23:01:16 that's hy :depends-on is the wrong name 23:01:22 it's a single link :) 23:01:25 then you don't need the whole depends-on thing at all ((: 23:01:47 as I would then name my migrations by -descriptive-name (-: 23:02:02 what if i'm inb japan and you're in europe? what if i'm on VMS and you're on windows? 23:02:38 anyway. I'm sure there are different strategies for this kind of thing that can work in parallel. 23:03:19 I suggest you try it and we'll see if it works well enough in practice. I can only offer judgements based on my experience with rails migrations, and my limited experience with my copy of them in cl 23:03:22 i'm just trying to figure out what the arguments for the various strategies are.. i'm new to this and have previously just 'done it manually' 23:03:34 The-Kenny [n=moritz@p5087AD31.dip.t-dialin.net] has joined #lisp 23:03:40 drewc: unix timestamp is defined afaik as seconds since 1970-01-01 00:00 in UTC, so it would work, except that naming would have to be slightly corrected :D 23:04:15 -!- JAS415 [n=jon@c-24-34-16-25.hsd1.ma.comcast.net] has left #lisp 23:04:17 my experience was that one file per migration, prefixed by iso timestamp is a pretty good and simple solution that requires very little manual intervention even in the face of complicated merges 23:04:43 and iso timestamps are easier to read than unix ones ;D 23:05:03 yeah 23:06:11 Errh... "This function is not defined: APPEND." Isn't that standard in common lisp? 23:06:32 octoberdan: it is. you probably in-packaged to somewhere that doesn't import common-lisp 23:06:44 (cl:in-package :cl-user) ; should get you back into known territory 23:07:28 I will blame it on slime ;-). Thanks for the help. That would explain why it /stopped/ working after working before 23:08:01 -!- ehu` [n=chatzill@82-170-33-173.ip.telfort.nl] has quit [Read error: 110 (Connection timed out)] 23:08:09 yeah, in some rare cases it loses track of where you are... are you using a very old sbcl and a very old slime, by chance? 23:08:30 (it used to be that it could happen very easily if you had an error in a defpackage form, but that changed a year ago or so) 23:10:19 According apt-cache show... 1:20080223-2 on slime, 1:1.0.18.0-2 on sbcl 23:10:24 Not sure why they are prefixed with 1: 23:10:46 hmm, you may be better off getting clbuild and installing a newer sbcl through that. 23:10:48 minion: clbuild 23:10:49 clbuild: clbuild [common-lisp.net] is a shell script helping with the download, compilation, and invocation of Common Lisp applications. http://www.cliki.net/clbuild 23:11:12 cool 23:11:36 the current version is 1.0.27, and there was one release per month (-: 23:11:51 octoberdan: they're prefixed with 1: because at some point either sbcl or debian screwed up the monotonicity of version numbers 23:11:57 soooo... half a year, not bad, but no baby-smooth skin anymore, either 23:12:16 <_3b> does the 1.0.18 they ship still have broken TIME? 23:12:26 antifuchs: you'd shudder to see the tech.coop webserver.. 0.8.14 iirc 23:12:34 _3b: How would I find out? 23:12:47 <_3b> octoberdan: try (time (sleep 2)) 23:13:13 hmm, still no append 23:13:18 I must be doing something else wrong 23:14:03 Yeah, isolated there's an issue with append. Something in my code is throwing it out of whack 23:15:08 sellout [n=greg@c-24-128-50-176.hsd1.ma.comcast.net] has joined #lisp 23:15:36 _3b: Jerk :-p 23:15:53 _3b: Flooded with 0s... is that what it's supposed to do? 23:15:57 nyef [n=nyef@24.220.182.229] has joined #lisp 23:16:02 <_3b> octoberdan: ah, not fixed then :( 23:16:04 Hello all. 23:16:04 -!- deech [n=deech@71-10-166-226.dhcp.stls.mo.charter.com] has quit [Read error: 104 (Connection reset by peer)] 23:16:38 <_3b> octoberdan: sorry about that, should have warned you 23:16:45 deech [n=deech@71-10-166-226.dhcp.stls.mo.charter.com] has joined #lisp 23:16:47 pstickne [n=pstickne@c-24-21-76-57.hsd1.wa.comcast.net] has joined #lisp 23:17:21 <_3b> octoberdan: C-c C-o will clear last command's output from repl if that helps :) 23:17:27 <_3b> (in slime that is) 23:17:30 -!- octoberdan [n=user@c-65-96-169-141.hsd1.ma.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 23:17:54 octoberdan [n=user@c-65-96-169-141.hsd1.ma.comcast.net] has joined #lisp 23:18:18 _3b: Argh it crashed emacs and therefor my irc client 23:18:30 -!- spacebat [n=akhasha@ppp121-45-11-174.lns10.adl2.internode.on.net] has quit [Read error: 60 (Operation timed out)] 23:18:34 _3b: Though eventually I regained control. So, is that what you were talking about? 23:18:35 spacebat [n=akhasha@ppp121-45-40-38.lns10.adl2.internode.on.net] has joined #lisp 23:18:39 I just had a build failure with SBCL 1.0.27.45 from 1.0.11, x86-64 linux both sides. An internal error #26 (why isn't there a lookup table for this in the runtime?) "SC: 15, Offset: 0 $1= 0x1000001895: other pointer" too early in init during make-target-2. 23:18:53 <_3b> octoberdan: yeah, didn't expect it to cause that much problem :( 23:19:04 Under file-write-date under !debug-info-cold-init. 23:19:15 _3b: Things just locked down for awhile as the buffer was flodded. No problem. I got a kick out of it, hehe 23:19:16 <_3b> nyef: was just seeing that in win32 23:20:00 Okay, so it's not entirely news. That's good. 23:20:05 <_3b> octoberdan: C-c C-c interrupts slime repl commands, C-c M-o clears repl, might be useful for that sort of situation 23:20:58 -!- The-Kenny [n=moritz@p5087AD31.dip.t-dialin.net] has quit ["I am Locutus of Borg! You will assist us!"] 23:21:18 Hi nyef 23:21:38 p_l: Hello. 23:21:55 *_3b* thinks git needs to be able to bisect with base greater than 2 :p 23:22:07 <_3b> (or sbcl needs to learn to do multicore compiles) 23:23:09 git doesn't need to do bisect at all 23:23:14 -!- lifeh2o [n=lifeh2o@119.152.67.124] has left #lisp 23:23:15 it's a trivial script to write 23:23:16 nyef: is there a place where current code of your emulator exists? 23:23:38 p_l: Yes. My hard drive. 23:23:44 <_3b> foom: or someone that knows enough scripting to make it trivial should write a script for me :) 23:23:58 p_l: Looking to do some Alpha emulator hacking yourself, then? 23:24:09 nyef: Looking at inspiration for my MMIX project 23:24:18 And maybe some testing, maybe even coding :) 23:24:29 don't expect much, though 23:25:24 Fair enough. I'll tarball what I have in a few minutes. 23:25:37 -!- karvus [n=thomas@193.213.35.168] has quit ["Leaving."] 23:26:34 -!- puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has quit [Remote closed the connection] 23:27:08 -!- sellout [n=greg@c-24-128-50-176.hsd1.ma.comcast.net] has quit [] 23:27:37 -!- jewel [n=jewel@41.242.183.106] has quit [Read error: 113 (No route to host)] 23:27:49 and that comment about MKR was after backtracking into containing directory after googling "lameulator.lisp" :) 23:28:25 _3b: Thanks for the tips :-) 23:28:35 <_3b> well, 1.0.27.22 looks ok 23:28:55 <_3b> 27.34 not ok 23:28:58 _3b: I might try to submit an updated package 23:30:15 The-Kenny [n=moritz@p5087AD31.dip.t-dialin.net] has joined #lisp 23:30:54 p_l: Ah. 23:32:20 *_3b* wonders if picking test points based on change in committer name for series of patches would speed up bisecting over purely binary search 23:34:41 <_3b> .31 not ok 23:34:46 -!- oudeis [n=oudeis@77.124.71.204] has quit ["This computer has gone to sleep"] 23:35:13 p_l: http://www.lisphacker.com/temp/axp-tools.tgz 23:35:40 p_l: You can safely ignore ecoff.lisp, and the files will load cleanly in -some- order but I haven't written that order down anywhere yet. 23:35:55 <_3b> ok, problem seems to be 1.0.37.31 23:36:03 nyef: thanks 23:38:05 -!- Samy [n=sbahra@c-76-26-144-229.hsd1.dc.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 23:38:30 Samy [n=sbahra@c-76-26-144-229.hsd1.dc.comcast.net] has joined #lisp 23:38:48 p_l: file-utils, disassembler, alcor-2, lameulator, cpu-instructions. Or you can move alcor-2 earlier, I think that's actually the latest it can be defined. 23:40:09 -!- spradnyesh [n=pradyus@117.192.3.229] has quit [Read error: 110 (Connection timed out)] 23:40:22 heh :) 23:40:33 fileutils is certainly an useful file, though :) 23:40:37 rolly1975 [n=rory@5ac959a9.bb.sky.com] has joined #lisp 23:41:12 Yeah, fileutils is an elaboration of the file of the same name I used for the mkr stuff. 23:41:48 btw, is there any "state of the art" binary format library? 23:41:48 The emulator still needs a little hand-holding to get all the way to the point where it tries to exit PALmode, mostly just skipping a couple troublesome instructions, and I still haven't settled in to defining the ITLB and ICache yet. 23:42:07 It probably depends on the state and the art. 23:42:10 spradnyesh [n=pradyus@117.192.6.193] has joined #lisp 23:42:15 :) 23:42:24 *_3b* needs a better one of those too :) 23:42:36 <_3b> my attempt tends to kill sbcl 23:42:56 _3b: reminds me, I could make the sbcl build archive public 23:43:19 <_3b> antifuchs: do you build for win32 also? :) 23:43:28 Well, my main concern is to get an emulator that can execute NNIX OS usermode code with MMIX 1.1 model 23:43:29 nope 23:43:40 but I suppose it would be a win on linux/x86-64 and x86 23:43:51 <_3b> yeah, could be 23:44:11 of course, if you want to donate a machine's spare cpu cycles to the autobench effort... 23:44:25 -!- lyte [n=lyte@unaffiliated/neerolyte] has quit ["Ex-Chat"] 23:44:37 <_3b> if i had some to spare :/ 23:45:17 right (: 23:45:36 I have a very interesting offer here from a company that would like to donate 20 virtual servers to open source software projects 23:45:48 <_3b> is it building all versions automatically now? 23:46:05 antifuchs: wow 23:46:31 Someone make a buildbot for Lisp libraries, k thx. 23:46:33 it builds every revision on unthreaded x86 and x86_64, and every release and every fourth non-release version in the threaded config on both arches 23:46:58 <_3b> has it tried to build anything since .27.31? 23:47:06 <_3b> and if so, did it work? 23:47:30 jup. everything up to and including .40 was built and benchmarked correctly today. 23:48:12 <_3b> hmm, strange, wonder if i'm doing something odd 23:48:37 which platform are you testing? 23:48:52 <_3b> nyef: you said you were getting that failure on x86_64 linux, right? 23:48:56 Yes. 23:49:06 Just starting bisection now. 23:49:08 <_3b> antifuchs: win32, but nyef had same problem 23:49:14 <_3b> nyef: try .30 and .31 23:49:26 ah, well, it's all linux here. 23:49:27 <_3b> .30 builds for me, but not .31 23:49:50 It might also be dependent on host lisp. 23:50:00 <_3b> i couldn't build head from .30 23:50:37 <_3b> and was using 1.0.25.something before that 23:54:34 apo_ [n=apo@pD9E7E4B2.dip.t-dialin.net] has joined #lisp 23:55:21 oudeis [n=oudeis@92.61.225.131] has joined #lisp 23:56:09 -!- nvoorhies_ [n=nvoorhie@adsl-75-36-207-191.dsl.pltn13.sbcglobal.net] has quit [] 23:57:45 zbrown [n=rufius@unaffiliated/zbrown] has joined #lisp 23:59:06 -!- alec [n=alec@pool-96-233-16-99.bstnma.east.verizon.net] has quit ["leaving"]