00:01:10 _8david: I think you expected something differently than what I just posted 00:01:42 less than a second 00:01:45 want the code? 00:02:05 -!- araujo [n=araujo@gentoo/developer/araujo] has quit [Read error: 148 (No route to host)] 00:02:19 fwiw, i don't care about no licences, a simple mention of authorship suffices 00:03:58 tcr: got it. 00:04:11 -!- Mazingaro [n=tetsuja@host74-228-dynamic.22-79-r.retail.telecomitalia.it] has quit [Remote closed the connection] 00:04:30 weirdo pasted "clean-room readtable-contents" at http://paste.lisp.org/display/72690 00:04:34 ... I was just about to ask "what's signal 11", when I realized... That's sigsegv, isn't it? 00:05:58 tcr: should there be something new on gmane? 00:06:53 oh wait. empty dispatch-chars won't get printed 00:07:34 weirdo annotated #72690 with "now print empty dispatch-chars" at http://paste.lisp.org/display/72690#1 00:08:01 I'm getting to intercept signals now. This is almost kindof fun. 00:09:57 Jabberwo_ [n=jens@dslb-082-083-119-215.pools.arcor-ip.net] has joined #lisp 00:10:43 danderson [n=dave@atlas.natulte.net] has joined #lisp 00:10:45 -!- davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)] 00:10:50 hi. 00:11:15 hi 00:12:51 tcr, or should the iterator be lazy, series style? 00:14:19 weirdo: The iterator should do what the documentation says it should do 00:14:43 danderson: can we help you? 00:15:50 yeah, I was just pastebinning what I need to show :) 00:15:57 (paste.lisp.org appears to be unwell) 00:16:37 http://pastebin.com/d3eacef9e 00:17:45 When I attempt to execute the lambda returned by output-register, my lisp complains that I have an "illegal function call". When I replace ,val with something constant, it works, so I'm assuming I'm incorrectly escaping/unescaping val 00:17:55 but I don't know how I should be doing it right :( 00:18:25 araujo [n=araujo@gentoo/developer/araujo] has joined #lisp 00:19:42 could you provide the full error message? 00:19:56 ((output-register 'data 16 (byte 3 9)) "varname") 00:20:04 <_3b> (funcall (output-register ...) ...) ? 00:20:22 danderson: Common Lisp is not Scheme. The CAR of a form bust be a symbol (or as a special exception (LAMBDA ...)) 00:20:30 s/bust/must/ 00:21:20 aah. 00:21:36 the error was http://pastebin.com/mde31625 btw. 00:21:47 But that explanation makes sense. I must be confusing my lisps. 00:22:32 Thanks! 00:23:01 tcr: I don't think of that as a special exception, but then again you're the one who has studied the spec :) 00:24:07 hefner: It came into the standard late for ISLISP compatibility, just like DEFINE-SYMBOL-MACRO. 00:24:16 merAch [n=none@c-71-199-20-205.hsd1.co.comcast.net] has joined #lisp 00:24:34 *nyef* loves global symbol-macros. 00:25:10 yeah, but I stick to the idea that lambda names a function, and everything seems consistent 00:25:32 -!- HET3 [n=diman@chello084114161225.3.15.vie.surfer.at] has quit ["This computer has gone to sleep"] 00:25:49 except for SATISFIES! 00:26:05 so, in CL, the car has to be either explicitely a symbol, or explicitely a lambda form, but a function returning a lambda isn't allowed? 00:26:20 I'm mildly confused as to why that is consistent 00:26:23 danderson: Yes. The CAR of a form is not evaluated 00:26:34 <_3b> hefner: (setf foo) names a function too though, doesn't it? 00:26:38 oh, that's funny. naughty SATISFIES. 00:26:54 tcr: aah. Except for a lambda, which you said is a special case. 00:26:58 that does make sense. 00:27:25 danderson: Right, and I think it is not even explicitly specified what is evaluated first, the arguments or the lambda. 00:27:31 _3b: heh. I have no idea how/when/why that works, or what the rules are. I abandon my claim that CL makes any sense. 00:27:40 the arguments are specified to be evaluated from left to right, though 00:27:49 CHJA [n=christia@90-231-255-161-no145.tbcn.telia.com] has joined #lisp 00:27:57 kenyon_ [n=kenyon_@xdsl-81-173-148-255.netcologne.de] has joined #lisp 00:28:24 <_3b> that reminds me, does CL specify whether (setf aref) is a function or not? 00:28:38 _3b: No, I don't believe it does. 00:28:43 tcr: beautiful, thanks. Another step towards a deeper understanding of CL. 00:28:50 dash_ [n=dash@dslb-084-057-013-127.pools.arcor-ip.net] has joined #lisp 00:29:00 _3b: It's not, but you can use (setf (apply 'aref ...)) or some such 00:29:13 Is there a way to do a loop inside a loop easily with the loop macro? Something like (loop for blah in foo do (something) for herbert in blah do (something else)) 00:29:16 Yeah, there's special rules for APPLY as a place. 00:29:32 <_3b> so no way to get a function object that sets an array element? 00:29:34 kzar: Yes. It's called "using another loop". 00:29:48 clhs 5.1.2.5 00:29:48 http://www.lispworks.com/reference/HyperSpec/Body/05_abe.htm 00:30:02 <_3b> aside from wrapping a call by hand i guess 00:30:10 nyef: Yea ovbiously I know I could just use another loop, I just was checking you can't do it all from the one loop call 00:30:12 _3b: Not portably, no. 00:30:40 kzar: Sorry, nope. You could try to be clever, but that's usually impenetrable when you come back to it two weeks later. 00:31:46 nyef: Well if the kind of thing I was describing was possible it would be fairly clear 00:32:11 -!- ecraven [n=nex@140.78.42.103] has quit ["bbl"] 00:32:27 -!- kenyon [n=kenyon_@xdsl-81-173-237-104.netcologne.de] has quit [Read error: 60 (Operation timed out)] 00:35:01 S11001001 [n=sirian@74-137-146-187.dhcp.insightbb.com] has joined #lisp 00:35:35 There's nothing quite like realizing that what you thought was a simple utility function in your program is really the main event interface. 00:35:55 -!- beach [n=user@203.210.248.212] has left #lisp 00:36:04 -!- mathrick [n=mathrick@195.116.35.13] has quit [Read error: 110 (Connection timed out)] 00:39:53 -!- dash__ [n=dash@dslb-084-057-051-107.pools.arcor-ip.net] has quit [Read error: 110 (Connection timed out)] 00:40:49 -!- blbrown [n=Berlin@c-71-236-25-127.hsd1.ga.comcast.net] has quit ["Ex-Chat"] 00:48:20 ManateeLazyCat [n=user@222.212.142.244] has joined #lisp 00:48:28 if I get a type-error, how do I know which function threw it in SLDB? 0: (SB-IMPL::OPTIMIZED-DATA-VECTOR-SET # # #) isn't too informative. Should I increase the debug level? 00:48:52 -!- puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has quit [Remote closed the connection] 00:51:14 z0d: How recent is your Slime? 00:51:33 tcr: CVS 00:51:41 HEAD 00:53:07 z0d: Go to a frame belonging to a function of your package 00:53:08 -!- ehu` [n=chatzill@82-170-33-173.ip.telfort.nl] has quit [Read error: 110 (Connection timed out)] 00:53:23 z0d: Then press C-u C-c C-c, this recompiles the function behind that frame with maximum debug settings 00:53:29 z0d: then invoke the RETRY restart 00:53:59 it's the same 00:54:36 well bummer, paste the whole backtrace 00:55:02 rottcodd [n=user@ppp59-167-37-203.lns2.cbr1.internode.on.net] has joined #lisp 00:55:45 z0d pasted "backtrace" at http://paste.lisp.org/display/72692 00:55:48 avdi1 [n=avdi@c-71-58-195-219.hsd1.pa.comcast.net] has joined #lisp 00:56:41 -!- Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has quit [Remote closed the connection] 00:59:06 -!- Yuuhi [n=user@p5483F334.dip.t-dialin.net] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 00:59:17 z0d: you're passing -1 to (setf (aref ...)), or a similiar form 00:59:24 in SPIN 00:59:27 Does Kevin Rosenberg hang out here? 00:59:51 He used to. I don't think he does much Lisping these days. 00:59:59 S11001001: He used to. 01:00:41 Okay, I'm just looking to harass him about some patches anyway. 01:01:32 I'm planning to transform some hacking someone did on Weblocks into CLSQL patches as well, and would like some way to get them into mainline. 01:01:49 tcr: exactly in spin or "in spin or in some function I call from spin" 01:01:51 ? 01:02:35 z0d: Should be in SPIN, or a function that is inlined in SPIN (or a macro) 01:03:19 -!- avdi [n=avdi@c-71-58-195-219.hsd1.pa.comcast.net] has quit [Connection timed out] 01:03:19 can you paste its definition? 01:03:20 Hmm. Thanks 01:03:40 -!- avdi1 [n=avdi@c-71-58-195-219.hsd1.pa.comcast.net] has quit [Read error: 60 (Operation timed out)] 01:03:57 Sure. 01:05:03 z0d annotated #72692 with "definitions" at http://paste.lisp.org/display/72692#1 01:05:52 *finger* is just a counter 01:06:49 -!- milanj [n=milan@79.101.137.186] has quit ["Leaving"] 01:07:43 Umm... Hrm. 01:07:53 How on earth did update-process-status return T? 01:07:57 clhs cond 01:07:57 http://www.lispworks.com/reference/HyperSpec/Body/m_cond.htm 01:07:59 Exciting, the semantics described for *default-special-bindings* in bordeaux-threads do not appear to be implemented 01:08:28 Ah, that'd do it. 01:08:43 -!- Nshag [i=user@Mix-Orleans-106-3-249.w193-248.abo.wanadoo.fr] has quit [Read error: 60 (Operation timed out)] 01:10:18 blitz_ [n=julian@pD95D771E.dip.t-dialin.net] has joined #lisp 01:13:39 S11001001: patches are welcome :) 01:14:41 I considered it, but then realized I need something that works when creating threads not using bt:make-thread. 01:14:48 So I'm implementing hash-table fluids instead. 01:14:57 tiento [n=user@pc-161-172-160-190.cm.vtr.net] has joined #lisp 01:15:13 We are still supporting Hunchentoot 0.15.7 so Hunchentoot doesn't use BT yet 01:15:24 -!- tiento [n=user@pc-161-172-160-190.cm.vtr.net] has left #lisp 01:16:14 S11001001: what are hash-table fluids ? 01:16:19 jrockway: You here? 01:17:17 fe[nl]ix: like http://www.bitbucket.org/svg/weblocks-dev-svg/src/tip/support/clsql-connection-pool/connection-pool.lisp (*process-databases*, *process-databases-lock*) but really having nothing to do with databases 01:19:25 *nyef* shakes his head. 01:19:28 Incredible. 01:19:35 -!- kzar [n=kzar@hardwick.demon.co.uk] has quit [Remote closed the connection] 01:19:53 When you wait(2) for a status change, you get back an integer. 01:20:03 (32-bit integer, near as I can tell.) 01:20:57 The macros defined for operating on this integer allow for 7-bits of termination signal number (and require it to be (< 0 signal 127)), and 8 bits of stop-reason signal number. 01:21:41 One of the ptrace options you can set causes some of the times a process stops to be reported as (logior #x80 SIGTRAP). 01:21:51 That much is okay. 01:23:08 But then there are six other options, and they return (logior (ash some-code 8) SIGTRAP), which can't be detected by the usual macros for dealing with the status from wait(2) and friends. 01:23:30 So I'm looking at this, and... WTF? 01:24:41 There's just no documented way to get at these values. You just have to know how they're encoded. 01:25:02 And that's assuming that they're encoded the obvious way. 01:26:46 (And a quick test shows that they are. But still, WTF?) 01:29:16 fooquux: heh, finding a lot of omissions in the manual? 01:30:44 -!- blitz_ [n=julian@pD95D771E.dip.t-dialin.net] has quit ["Lost terminal"] 01:31:03 -!- ManateeLazyCat [n=user@222.212.142.244] has quit [Read error: 110 (Connection timed out)] 01:31:31 ManateeLazyCat [n=user@222.212.142.244] has joined #lisp 01:37:02 -!- yoshinori [n=read_eva@router1.gpy1.ms246.net] has left #lisp 01:42:38 -!- manuel_ [n=manuel@HSI-KBW-091-089-250-148.hsi2.kabel-badenwuerttemberg.de] has quit [] 01:44:11 -!- smolyn is now known as smolyn|away 01:44:23 -!- ashmawy [n=gfh@41.233.148.52] has quit [Read error: 113 (No route to host)] 01:48:41 Rigdern [n=Rigdern@ool-4350abb3.dyn.optonline.net] has joined #lisp 01:49:17 *tcr* wishes his shell had presentations, and I could drag'n'drop the output of `ls' 01:50:03 careful what you wish for, some gnome bastard might hear you 01:50:07 Sometimes Worse really is Worse 01:50:43 fe[nl]ix: Open to using Trac bugs, or should all things go to -devel? 01:51:40 tcr: You know, I think I might know how to make that happen... 01:51:51 I guess the fact that you have Trac bugs turned off should tell me :) 01:52:26 r2q2 [n=user@c-24-7-212-60.hsd1.il.comcast.net] has joined #lisp 01:52:28 Of course, you need a custom terminal application, and a custom version of ls... 01:52:43 nyef: perhaps some LD_PRELOAD tricks? 01:52:45 Hey what does everyone think about the book "How to solve it by computer" by R.G.Dromley. 01:53:12 ansi-style escape codes for embedding presentations? 01:53:29 tcr: No, because you need to change printf() into something that can create presentations, and calls to printf() rarely have such semantic information available. 01:53:31 I like the way you think! 01:53:35 hefner: That'd work, yes. 01:58:31 -!- robyonrails [n=roby@host60-240-dynamic.10-87-r.retail.telecomitalia.it] has quit ["Sto andando via"] 02:03:10 davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has joined #lisp 02:03:23 -!- ace4016 [i=ace4016@76.168.248.118] has quit ["When there's nothing left to burn, you have to set yourself on fire."] 02:06:30 -!- smolyn|away is now known as smolyn 02:08:49 -!- davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 02:09:17 ace4016 [i=ace4016@cpe-76-168-248-118.socal.res.rr.com] has joined #lisp 02:09:33 avdi [n=avdi@c-71-58-195-219.hsd1.pa.comcast.net] has joined #lisp 02:09:59 -!- chris2 [n=chris@p5B16B10D.dip0.t-ipconnect.de] has quit [Read error: 110 (Connection timed out)] 02:10:07 davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has joined #lisp 02:11:58 S11001001: or if you come up with a patch before I go to sleep I'll apply it right away 02:13:46 grr, a game that writes "negative" values to the scroll register. Now I have to make that work. 02:13:58 wedgeV [n=wedge@cm56-238-229.liwest.at] has joined #lisp 02:15:33 minion: memo for fooquux: fixed the defcfun documentation again. Keep those complaints coming! 02:15:33 Remembered. I'll tell fooquux when he/she/it next speaks. 02:19:10 beach [n=user@203.210.248.212] has joined #lisp 02:23:50 -!- dv_ [n=dv@85-127-102-97.dynamic.xdsl-line.inode.at] has quit ["Verlassend"] 02:24:11 If someone is terribly bored, I'm trying to think up a nice design for an emulator generator in CL, and I'm looking for feedback on the overall design, since I'm rather inexperienced in CL systems 02:24:38 What kind of emulators are we talking about here? 02:24:44 http://pastebin.com/m32befaa0 is the overall idea, along with a couple of design questions 02:24:55 emulator for the m68k cpu 02:25:13 Oh, god. I love the m68k, but I've -never- managed to write an emulator for it that worked. 02:25:54 I don't especially like or dislike it, but my latest world domination project requires one 02:25:59 and written in Java, no less :( 02:26:23 so to spare me the pain, I'm toying with the idea of being meta, and generating the java implementation from a CL high level description. 02:26:27 -!- pstickne [n=pstickne@c-71-236-177-238.hsd1.wa.comcast.net] has quit ["Leaving"] 02:26:53 *hefner* thought about doing the same thing for the 6502, when he gets around to tossing out M6502 and its bogus license 02:27:04 hefner: Marat's core? 02:27:04 granted, 68k is a lot more complicated :) 02:27:17 danderson: Why do you need an emulator for the 68k, and why do you need for the implementation to be in Java? 02:27:25 nyef: yeah, although I've fixed a bunch of random bugs in it. 02:27:37 -!- CHJA [n=christia@90-231-255-161-no145.tbcn.telia.com] has quit [] 02:27:37 (which are probably fixed in the current version, but mine is probably from 1999) 02:27:43 hefner: My 6502 core is yours for the asking, you know. Reasonable license terms, too! 02:28:54 beach: because the sega genesis used an m68k as its CPU, and I want a genesis emulator for Android (google's mobile phone OS) 02:28:58 -!- avdi [n=avdi@c-71-58-195-219.hsd1.pa.comcast.net] has quit [Read error: 110 (Connection timed out)] 02:29:03 kpreid [n=kpreid@pool-72-68-32-150.nwrknj.east.verizon.net] has joined #lisp 02:29:07 and the Android SDK at this time only allows development in Java 02:29:09 nyef: I should probably take you up on that. I'm really not interested in spending any more time on this emulator than it takes to SMB3 and a couple other things working reasonably glitch-free. 02:29:33 danderson: If you're just using this for a code-generator, I'd consider looking at it as a data-transformation problem, particularly given the different overlapping ranges for the instruction encoding spaces. 02:29:45 (or, technically, any language producing JVM bytecode, but it's a bit of a hack) 02:29:46 I wonder whether ABCL could be ported to Android's VM? 02:31:01 if it generates JVM bytecode at runtime, no 02:31:12 *nyef* wonders where the hell his 6502 core ended up, and why it's not in src/mods/emu/cpu/ with the 6280 core. 02:31:16 or it'd need a completely new backend to output (not yet documented) Dalvik VM bytecode 02:31:51 nyef: I'm afraid I don't follow. 02:33:00 danderson: there's also linj, which translates CL source to Java source. 02:33:09 ABCL seems to work in IKVM.. so far.. it generates bytecode.. but IKVM reintepretes the .class to .il 02:33:35 in the case of abcl .cls to .il 02:33:51 abcl's .cls files are standard-format java .class files? 02:33:57 yeah 02:34:13 *Phoodus* .oO( why not just call it *.class then? ) 02:34:28 luis: hmm, interesting. Though I'd still have to run a first stage to output the CL code to translate, but taking care of the ugly translation would be nice. 02:34:58 so let me get this straight 02:35:01 you want CL 02:35:05 translated into Java 02:35:17 which is translated to a different bytecode 02:35:20 which runs on a VM 02:35:23 on a _phone_ 02:35:25 its ticket 34 on trac to fix it http://trac.common-lisp.net/armedbear/ticket/34 02:35:26 to run an _emulator_ 02:35:33 for a _gaming_ system? :-P 02:35:44 Phoodus: very meta, isn't it 02:35:46 intially maybe it was becasue .cls wasnt sure if it was going to do wierder stuff 02:36:02 I'm not put off by the meta, but by the potential speeds you're looking at 02:36:04 -!- wedgeV [n=wedge@cm56-238-229.liwest.at] has quit [] 02:36:10 -!- kpreid [n=kpreid@pool-72-68-32-150.nwrknj.east.verizon.net] has quit [] 02:36:27 Phoodus: I'm not arguing 02:36:28 also .cls keeps the classloader from making assumptions 02:36:40 the biggest hit as far as I'm concerned is running one VM inside another 02:36:44 (the system class loader) 02:36:53 dmiles_afk: makes enough sense 02:37:01 but currently, I don't have a choice, the SDK only allows Java running in a VM 02:37:10 In a non-standard VM. 02:37:16 With no docs. 02:37:42 there are some docs, and the source code is there 02:37:42 I would either write or generate tight java code, instead of translating another language into Java, and have to hold all sorts of wrappers for the original language features 02:38:00 (impressive how google manages to create an environment even more retarded than the iphone) 02:38:00 For the byte-code in the Android VM? 02:38:17 but the fact that it's not a standard JVM is considered an implementation detail: the java .class files are translated to a dalvik .dex file automatically 02:38:22 hefner: it takes lots of geniuses for that. 02:38:41 Ah, so normal JVM byte-code would work. 02:39:03 sounds like it'd work in Android VM then 02:39:03 yes, just not bytecode generated on the phone 02:39:18 Because the translator is a pre-deploy step? 02:39:22 rcy [n=rcy@S01060002553240a8.vc.shawcable.net] has joined #lisp 02:39:32 Jasko [n=tjasko@c-98-235-21-22.hsd1.pa.comcast.net] has joined #lisp 02:39:36 the translator runs at compile time, a post-javac step 02:40:07 I suppose you could port the translator to run inside the VM, if you really wanted to... 02:40:08 gotcha. 02:40:22 kinda like gcj/ikvm.. however both this systems has a .class interpretor just-in-case 02:40:45 so, it's my understanding that things like Jython, which compile python and dynamically output JVM bytecode when the .py is loaded, wouldn't work. 02:41:06 -!- r2q2 [n=user@c-24-7-212-60.hsd1.il.comcast.net] has quit [Connection timed out] 02:41:10 Jython would output JVM bytecode on the phone, bytecode that can't be executed by dalvik. 02:41:31 there are plenty of .class interpetors written in .java already i supposed one could compile first 02:41:42 danderson: ABCL has compile-file for sure, so maybe it'd work? 02:42:04 I think I'd just find a 68k core in C under an agreeable license and transliterate it to java. 02:42:06 also abcl works just as well when the compiler is never used.. just much slower 02:42:39 hefner: one of the ones with the best performance I've seen is actually in the source of a genesis emulator. 02:42:45 the compiler converts (car myvar) to myvar.CAR() 02:42:53 danderson: that isn't surprising, really. 02:43:08 or (car myvar) to myvar.symbolValueOrDie().CAR() 02:43:09 but it's implemented as a definition of the 68k assembler dialect, and a C code generator, written in C, that outputs the implementation 02:43:36 it's not exactly pretty 02:43:38 or documented :) 02:43:46 danderson: that's great, you can just bolt on a Java code generator and run that :) 02:43:55 heh 02:44:07 DSL to C to Java to Dalvik 02:44:12 great ;) 02:48:41 -!- crod [n=cmell@cad43e-014.dynamic.tiki.ne.jp] has quit [Read error: 110 (Connection timed out)] 02:49:12 crod [n=cmell@cb8a6f-121.dynamic.tiki.ne.jp] has joined #lisp 02:50:10 -!- beach [n=user@203.210.248.212] has left #lisp 02:52:40 -!- pragma_ [n=pragchao@blackshell.com] has quit [Read error: 60 (Operation timed out)] 02:56:37 envi^home [n=envi@220.121.234.156] has joined #lisp 03:05:18 amnesiac [n=amnesiac@p3m/member/Amnesiac] has joined #lisp 03:06:14 -!- cracki [n=cracki@sglty.kawo2.RWTH-Aachen.DE] has quit ["The funniest things in my life are truth and absurdity."] 03:09:31 -!- bpt [n=bpt@cpe-071-070-209-067.nc.res.rr.com] has quit [Read error: 110 (Connection timed out)] 03:12:01 Okay, I'm signing off, and will be back tomorrow. 03:12:04 -!- nyef [n=nyef@pool-71-255-137-188.cncdnh.east.verizon.net] has quit ["G'night all."] 03:22:42 -!- binarycodes [n=sujoy@59.93.192.83] has quit ["WeeChat 0.2.6"] 03:25:07 -!- rvirding [n=chatzill@h40n5c1o1034.bredband.skanova.com] has left #lisp 03:30:44 -!- ManateeLazyCat [n=user@222.212.142.244] has left #lisp 03:34:03 sword [n=user@c-98-222-51-71.hsd1.il.comcast.net] has joined #lisp 03:34:04 -!- |Soulman| [n=kvirc@138.80-202-254.nextgentel.com] has quit [Read error: 110 (Connection timed out)] 03:34:27 -!- tcr [n=tcr@p4FD3E689.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 03:35:30 -!- smolyn is now known as smolyn|away 03:46:07 -!- amnesiac [n=amnesiac@p3m/member/Amnesiac] has quit ["This computer has gone to sleep"] 03:48:08 srp [n=srp@122.172.10.214] has joined #lisp 03:49:42 someone i know just wrote "there's no CPU executing lisp code" 03:56:28 -!- Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has quit [] 04:04:00 Gah, me hates hidden ASD systems.. 04:05:24 I hoped I could avoid the kludge ASDF-INSTALL has to resort to, but the world of "every system is contained in a similarly named filename" is hopelessly simplistic. 04:06:10 asdf begs for a replacement 04:06:27 XCVB, Mudballs... 04:08:06 except it's still a standard 04:08:51 why does it suck again? 04:08:54 any rants to read? 04:10:16 Ahhh, for instance, I can't find a way to build a dependency graph without loading the stuff. 04:10:32 bombshelter13 [n=bombshel@209-161-240-155.dsl.look.ca] has joined #lisp 04:10:51 fisxoj [n=fisxoj@ool-4356ce50.dyn.optonline.net] has joined #lisp 04:11:28 Well, perhaps monitoring the in-memory systems for additions can work. 04:16:28 pstickne [n=pstickne@c-71-236-177-238.hsd1.wa.comcast.net] has joined #lisp 04:17:35 maybe we could reach a consensus on cll or some wiki? 04:17:47 then implement the stuff, then an asdf compat layer 04:17:57 then make everyone happy 04:20:29 fe[nl]ix: still hacking up some tests :) 04:20:37 was away for a couple hours 04:22:59 weirdo: only if you promise to finish it =) 04:23:20 got too much stuff on my plate to promise 04:23:25 didn't want to break it 04:23:52 bit` [n=bit@c-67-171-211-187.hsd1.or.comcast.net] has joined #lisp 04:24:00 everyone's plate is full at all times =) i expect to be with asdf for a long time 04:25:41 S11001001: I'm going to sleep about now. if you send the patch to the mailing list I'll apply it as soon as I can 04:26:29 yoshinori [n=read_eva@router1.gpy1.ms246.net] has joined #lisp 04:28:05 what if we reached a consensus and then someone volunteered? 04:28:11 would the consensus-reaching process then be anything more than waste of everyone's time? 04:28:26 besides, my plate is often empty :-) 04:28:43 i dont feel that it'd be a waste at all 04:29:07 -!- rottcodd [n=user@ppp59-167-37-203.lns2.cbr1.internode.on.net] has quit [Remote closed the connection] 04:31:15 -!- enodran [n=brandon@ip70-181-14-103.ri.ri.cox.net] has quit [] 04:32:50 karlw [n=user@cpe-76-168-206-252.socal.res.rr.com] has joined #lisp 04:34:01 -!- karlw [n=user@cpe-76-168-206-252.socal.res.rr.com] has left #lisp 04:34:30 -!- hugod_ [n=hugo@bas1-montreal50-1279441402.dsl.bell.ca] has quit [] 04:37:08 -!- Jasko [n=tjasko@c-98-235-21-22.hsd1.pa.comcast.net] has quit [Remote closed the connection] 04:37:42 -!- syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has quit ["Leaving..."] 04:40:37 -!- yoshinori [n=read_eva@router1.gpy1.ms246.net] has left #lisp 04:45:38 wanna write the initial post? 04:45:49 i could do that too, but poorly 04:55:39 rtra [n=rtra@unaffiliated/rtra] has joined #lisp 04:57:50 -!- smolyn|away is now known as smolyn 05:02:08 manic12_ [i=user@c-98-222-75-188.hsd1.il.comcast.net] has joined #lisp 05:03:01 -!- breinded [n=nonamme@h-67-100-110-200.nycmny83.dynamic.covad.net] has quit [Read error: 110 (Connection timed out)] 05:03:08 -!- manic12_ [i=user@c-98-222-75-188.hsd1.il.comcast.net] has left #lisp 05:06:01 -!- fooquux [n=fooquux@udp265832uds.hawaiiantel.net] has quit [Read error: 110 (Connection timed out)] 05:13:54 -!- weirdo [i=sthalik@c144-107.icpnet.pl] has quit [Remote closed the connection] 05:18:17 weirdo [i=sthalik@c144-107.icpnet.pl] has joined #lisp 05:18:37 what the heck is xlib cache entry? 05:18:40 invalidated all and stumpwm crashad 05:19:01 bug in xlib, xorg, stumpwm, sdl or quake2? 05:20:21 -!- pstickne [n=pstickne@c-71-236-177-238.hsd1.wa.comcast.net] has quit [Read error: 110 (Connection timed out)] 05:20:55 oh, a "bug" in xorg 05:23:54 heh. if i wrote the new asdf, i'd have to bitch and moan about debian third-rate committers butchering my code :/ 05:24:30 debian basically redefines redundant intrusive patches 05:25:52 -!- stassats [n=stassats@wikipedia/stassats] has quit [Read error: 113 (No route to host)] 05:27:55 stassats [n=stassats@ppp78-37-147-61.pppoe.avangarddsl.ru] has joined #lisp 05:32:56 -!- davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 05:36:40 -!- jso [n=user@151.159.200.8] has quit [Remote closed the connection] 05:37:28 Well, they are on the bugreport front line anyway, so it should converge to normality. 05:38:09 -!- anekos is now known as awayekos 05:39:32 dagnachew [n=dagnache@modemcable207.114-201-24.mc.videotron.ca] has joined #lisp 05:40:27 -!- dagnachew [n=dagnache@modemcable207.114-201-24.mc.videotron.ca] has quit [Remote closed the connection] 05:44:46 dagnachew [n=dagnache@modemcable207.114-201-24.mc.videotron.ca] has joined #lisp 05:47:16 -!- dagnachew [n=dagnache@modemcable207.114-201-24.mc.videotron.ca] has quit [Remote closed the connection] 05:49:43 dagnachew [n=dagnache@modemcable207.114-201-24.mc.videotron.ca] has joined #lisp 05:51:16 (not related to xlib stuff) 05:51:32 but i saw stumpwm's .asd 05:56:32 -!- dagnachew [n=dagnache@modemcable207.114-201-24.mc.videotron.ca] has quit [Remote closed the connection] 05:58:06 dagnachew [n=dagnache@modemcable207.114-201-24.mc.videotron.ca] has joined #lisp 05:58:22 -!- smolyn [n=smolyn@S01060016cbc4b572.vc.shawcable.net] has quit [] 05:58:29 weirdo: I'm not up on the latest stumpwm business or xorg buts, but if you haven't updated your clx from darcs in several months, you should do that 05:58:48 s/buts/bugs/ 05:59:58 hefner, i'm using 0.7.3. should i use nightly, then? 06:03:36 overlordpuppy [n=adrien@pool-71-124-117-108.chi01.dsl-w.verizon.net] has joined #lisp 06:04:27 yeah, that is ancient 06:04:49 darcs get http://common-lisp.net/~crhodes/clx 06:06:06 Is there a gtk library for lisp? 06:06:37 minion: gtk? 06:06:38 gtk: Gtk (GTK+) is a Graphics Toolkit written in C, primarily developed for use with the X Window System. http://www.cliki.net/gtk 06:07:09 stassats: Just found that. . . I should've searched first then asked. sorry. 06:08:11 but the question 'what to use' is good, though i don't know the answer 06:09:42 I like GTK for python, though there it's really easy to use. Anywhere else I've tried it has been a little worse experience. 06:11:39 hefner, thank you 06:14:21 why does :force recompile recursively? the horror, the horror 06:19:39 time to restart the x session. here goes nothing! 06:19:41 -!- weirdo [i=sthalik@c144-107.icpnet.pl] has quit [Remote closed the connection] 06:19:45 nicolas [n=nicolas@aqu33-5-82-245-96-5.fbx.proxad.net] has joined #lisp 06:20:22 weirdo [i=sthalik@c144-107.icpnet.pl] has joined #lisp 06:20:43 yay! it works like a charm 06:22:05 -!- crod [n=cmell@cb8a6f-121.dynamic.tiki.ne.jp] has quit [Read error: 110 (Connection timed out)] 06:22:36 crod [n=cmell@cb8ac8-115.dynamic.tiki.ne.jp] has joined #lisp 06:25:27 morning 06:25:44 ManateeLazyCat [n=user@222.212.142.244] has joined #lisp 06:31:35 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 06:33:36 -!- Rigdern [n=Rigdern@ool-4350abb3.dyn.optonline.net] has quit [] 06:35:35 -!- fisxoj [n=fisxoj@ool-4356ce50.dyn.optonline.net] has quit [Read error: 110 (Connection timed out)] 06:35:39 Intertricity [n=chatzill@68-119-171-225.dhcp.sffl.va.charter.com] has joined #lisp 06:36:44 -!- mattrepl [n=mattrepl@ip68-98-133-128.dc.dc.cox.net] has quit [] 06:38:33 how big is a leap from pattern matching to context-free grammar? 06:45:51 mathrick [n=mathrick@195.116.35.13] has joined #lisp 06:47:18 -!- ManateeLazyCat [n=user@222.212.142.244] has quit [Read error: 131 (Connection reset by peer)] 06:47:27 -!- crod [n=cmell@cb8ac8-115.dynamic.tiki.ne.jp] has quit [Read error: 60 (Operation timed out)] 06:47:38 -!- brianj_otter is now known as brianj_otterZZZ 06:48:00 ManateeLazyCat [n=user@222.212.142.244] has joined #lisp 06:50:25 -!- tc-rucho [n=tc-rucho@190.191.162.100] has quit [Client Quit] 06:55:35 -!- overlordpuppy [n=adrien@pool-71-124-117-108.chi01.dsl-w.verizon.net] has left #lisp 06:59:11 srp_ [n=srp@122.172.2.116] has joined #lisp 06:59:56 crod [n=cmell@cb8a6a-246.dynamic.tiki.ne.jp] has joined #lisp 07:02:55 yoshinori [n=read_eva@router1.gpy1.ms246.net] has joined #lisp 07:03:32 -!- mrsolo [n=mrsolo@adsl-68-126-206-137.dsl.pltn13.pacbell.net] has quit ["This computer has gone to sleep"] 07:04:56 mrsolo [n=mrsolo@adsl-68-126-206-137.dsl.pltn13.pacbell.net] has joined #lisp 07:07:33 -!- srp [n=srp@122.172.10.214] has quit [Connection timed out] 07:11:29 Intertricity_ [n=chatzill@68-119-171-225.dhcp.sffl.va.charter.com] has joined #lisp 07:13:52 hsaliak [n=hsaliak@cm247.sigma228.maxonline.com.sg] has joined #lisp 07:20:05 -!- yoshinori [n=read_eva@router1.gpy1.ms246.net] has left #lisp 07:23:02 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 07:24:55 Beket [n=Beket@athedsl-22493.home.otenet.gr] has joined #lisp 07:29:33 -!- Intertricity [n=chatzill@68-119-171-225.dhcp.sffl.va.charter.com] has quit [Read error: 110 (Connection timed out)] 07:31:42 usman [n=root@221.132.118.3] has joined #lisp 07:31:55 cpc26 [n=cpc26@72.170.156.242] has joined #lisp 07:31:58 -!- usman [n=root@221.132.118.3] has left #lisp 07:32:47 manuel_ [n=manuel@HSI-KBW-091-089-250-148.hsi2.kabel-badenwuerttemberg.de] has joined #lisp 07:38:01 -!- Intertricity_ [n=chatzill@68-119-171-225.dhcp.sffl.va.charter.com] has quit [Read error: 110 (Connection timed out)] 07:38:07 -!- cpc26 [n=cpc26@72.170.156.242] has quit [] 07:41:00 If I add a slot-value-using-class method for the metaclass of a subclass, are readers and accessors defined in superclasses required to go through it? 07:42:51 yoshinori [n=read_eva@router1.gpy1.ms246.net] has joined #lisp 07:44:47 yes 07:44:55 [citation needed] :P 07:45:02 :P 07:45:23 -!- yoshinori [n=read_eva@router1.gpy1.ms246.net] has left #lisp 07:45:29 oh well, taking a break then breaking out AMOP 07:46:47 yoshinori [n=read_eva@router1.gpy1.ms246.net] has joined #lisp 07:48:53 otherwise it wouldn't make much sense 07:49:04 slot-value calls svuc 07:49:19 and it dispatches on the class and metaclass 07:50:43 well, if you know that your class is a standard-class 07:50:58 it's possible to optimize the accessors' code 07:51:34 down to type-check followed by array access 07:51:40 there are lots of little restrictions in the MOP to allow optimization, who has comprehensive knowledge of them all? 07:52:03 -!- crod [n=cmell@cb8a6a-246.dynamic.tiki.ne.jp] has quit [Read error: 60 (Operation timed out)] 07:52:49 in case anyone has been paying attention to my myriad thoughts tonight, I'm working on a threadwise pool/proxy object for CLSQL databases 07:53:34 http://weblocks.lighthouseapp.com/projects/18897/tickets/61 08:05:31 -!- ASau [n=user@193.138.70.52] has quit [Read error: 113 (No route to host)] 08:06:27 -!- araujo [n=araujo@gentoo/developer/araujo] has quit ["Leaving"] 08:07:04 crod [n=cmell@cb8a6a-246.dynamic.tiki.ne.jp] has joined #lisp 08:08:27 kiuma [i=51d06a4b@gateway/web/ajax/mibbit.com/x-801772dd0a86736d] has joined #lisp 08:08:54 hello lispers 08:14:25 good morning kiuma 08:15:38 hello madnificent, have you spent nice holidays ? 08:19:21 -!- dagnachew [n=dagnache@modemcable207.114-201-24.mc.videotron.ca] has quit ["Leaving"] 08:20:01 madnificent: I need a suggestion, are you available ? 08:26:41 bpt [n=bpt@cpe-071-070-209-067.nc.res.rr.com] has joined #lisp 08:30:37 -!- bpt [n=bpt@cpe-071-070-209-067.nc.res.rr.com] has quit [Read error: 60 (Operation timed out)] 08:33:52 -!- drdo [n=psykon@167.111.54.77.rev.vodafone.pt] has quit [Read error: 104 (Connection reset by peer)] 08:39:19 bpt [n=bpt@cpe-071-070-209-067.nc.res.rr.com] has joined #lisp 08:45:56 Aankhen`` [n=Aankhen@122.162.166.64] has joined #lisp 08:52:07 -!- mrSpec[zzzz] is now known as mrSpec 08:52:17 Hey 08:53:32 Heh. I wonder how long the "I updated SLIME and it b0rked!" -> "You need the slime-repl contrib" messages will go on. 08:53:42 kiuma: sure 08:53:47 I was waking up :) 08:54:35 kiuma: the holidays were fine, but in general it's not my favorite time of the year:) 08:54:56 HET2 [n=diman@chello084114161225.3.15.vie.surfer.at] has joined #lisp 08:57:03 pstickne [n=pstickne@c-71-236-177-238.hsd1.wa.comcast.net] has joined #lisp 09:00:35 kpreid [n=kpreid@72-255-42-36.client.stsn.net] has joined #lisp 09:02:04 ALT777 [n=777ALT@99-163-22-30.lightspeed.mssnks.sbcglobal.net] has joined #lisp 09:04:33 madnificent: on priv plz. 09:08:56 tcr [n=tcr@p4FD3E0FB.dip.t-dialin.net] has joined #lisp 09:09:46 nostoi [n=nostoi@156.Red-79-146-64.staticIP.rima-tde.net] has joined #lisp 09:15:23 -!- H4ns1 [n=Hans@p57BBA32B.dip0.t-ipconnect.de] has quit [Connection timed out] 09:16:12 Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has joined #lisp 09:17:41 lhz [n=shrekz@c-3143e455.021-158-73746f34.cust.bredbandsbolaget.se] has joined #lisp 09:21:15 at all call sites (the place we jump FROM ?) we emit an trace-table-entry but not in generate-call-sequence, why ? 09:24:15 -!- wormphlegm [n=george@c-98-234-189-214.hsd1.ca.comcast.net] has quit [] 09:24:42 -!- stassats [n=stassats@wikipedia/stassats] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 09:25:12 -!- envi^home [n=envi@220.121.234.156] has quit [Read error: 104 (Connection reset by peer)] 09:25:15 -!- ia [n=ia@89.169.165.188] has quit [Read error: 60 (Operation timed out)] 09:25:57 If you look at the first without-scheduling in any non-x86* you will see a nop. I guess that nop is pad, because we lack an trace-table-entry. Else the return point would land wrong ? 09:26:11 wedgeV [n=wedge@cm56-238-229.liwest.at] has joined #lisp 09:28:18 -!- yrb_ is now known as yrb 09:29:05 NoorDextor [n=Game450@unaffiliated/noordextor] has joined #lisp 09:31:20 -!- nostoi [n=nostoi@156.Red-79-146-64.staticIP.rima-tde.net] has quit ["Verlassend"] 09:32:11 -!- sohail [n=Sohail@unaffiliated/sohail] has quit [Remote closed the connection] 09:35:27 jso [n=user@host-9-143-107-208.midco.net] has joined #lisp 09:37:32 jestocost [n=cmell@cb8a9b-024.dynamic.tiki.ne.jp] has joined #lisp 09:37:41 ASau [n=user@193.138.70.52] has joined #lisp 09:38:52 -!- crod [n=cmell@cb8a6a-246.dynamic.tiki.ne.jp] has quit [Read error: 110 (Connection timed out)] 09:41:31 ia [n=ia@89.169.165.188] has joined #lisp 09:44:47 -!- yoshinori [n=read_eva@router1.gpy1.ms246.net] has left #lisp 09:44:49 bpt_ [n=bpt@cpe-071-070-209-067.nc.res.rr.com] has joined #lisp 09:45:09 -!- bpt [n=bpt@cpe-071-070-209-067.nc.res.rr.com] has quit [Read error: 104 (Connection reset by peer)] 09:46:35 vasa [n=vasa@mm-178-89-84-93.dynamic.pppoe.mgts.by] has joined #lisp 09:47:28 -!- bpt_ is now known as bpt 09:47:51 -!- nicolas [n=nicolas@aqu33-5-82-245-96-5.fbx.proxad.net] has quit [Read error: 60 (Operation timed out)] 09:48:11 drdo [n=psykon@167.111.54.77.rev.vodafone.pt] has joined #lisp 09:55:08 -!- Jabberwo_ [n=jens@dslb-082-083-119-215.pools.arcor-ip.net] has quit [Remote closed the connection] 09:58:36 -!- NoorDextor [n=Game450@unaffiliated/noordextor] has quit ["Go Canada!"] 09:59:12 -!- manuel_ [n=manuel@HSI-KBW-091-089-250-148.hsi2.kabel-badenwuerttemberg.de] has quit [] 09:59:48 beach [n=user@203.210.248.212] has joined #lisp 09:59:55 Good afternoon. 10:00:04 NoorDextor [n=NoorDext@unaffiliated/noordextor] has joined #lisp 10:01:39 Hi. 10:03:57 drdo` [n=psykon@167.111.54.77.rev.vodafone.pt] has joined #lisp 10:05:14 What is a good way of reading bytes from a file into an array (32 bit, unsigned) with big endianness? 10:05:41 clhs read-sequence 10:05:41 http://www.lispworks.com/reference/HyperSpec/Body/f_rd_seq.htm 10:10:21 How do I handle the byte order with it? 10:11:47 -!- bombshelter13 [n=bombshel@209-161-240-155.dsl.look.ca] has quit [] 10:12:18 Are you saying the byte order is different from the native one? 10:12:37 Yes 10:13:22 Then either your Lisp system must let you open streams that way, or you might be able to use flexi-streams. If neither of those works, you must probably do it "manually". 10:13:25 afaik, you need to take care of that yourself or rely on implementation specific stuff. :/ 10:14:19 -!- jestocost [n=cmell@cb8a9b-024.dynamic.tiki.ne.jp] has quit [Read error: 60 (Operation timed out)] 10:14:43 Thanks 10:15:55 z0d: flexi-streams seem to handle the endianness. 10:16:27 I have a thread that's deadlocking. Is there any way to see what it's waiting for? I think it's hanging on a READ somewhere... 10:16:35 (I'm using SBCL.) 10:17:12 koning_robot: M-x slime-list-threads 10:17:26 you use g in there to update 10:17:42 you can use d to interrupt a thread and look at the backtrace 10:18:41 -!- drdo [n=psykon@167.111.54.77.rev.vodafone.pt] has quit [Connection timed out] 10:19:16 I'm not using slime (*blush*), but I'll figure out how it does it, thanks! 10:19:36 -!- bit` [n=bit@c-67-171-211-187.hsd1.or.comcast.net] has quit [Remote closed the connection] 10:20:42 koning_robot: What are you using? 10:21:06 plain sbcl 10:22:57 Why that? That's quite anemic. 10:24:22 trolltard [i=bizinatc@cpe-66-75-6-226.san.res.rr.com] has joined #lisp 10:25:34 is anyone here? 10:25:35 How can we get nkechismith in contact with a professional so that he does not commit suicide? 10:25:41 http://www.mahalo.com/answers/health-and-fitness/how-can-we-get-nkechismith-in-contact-with-a-professional-so-that-he-does-not-commit-suicide 10:26:01 this guy on this mahalo answers thing.. asked a question about killing himself 10:26:08 now he's not answering my questions 10:27:05 give him the common lisp ansi standard document. it has healing properties! 10:27:25 tcr: I know... No reason really, other than I'm using vim and not emacs, and learning emacs would in the process of learning lisp would probably slow me down even more. But I hear there's a slimey thing for vim too, so I guess now would be a good time to have a look at that. 10:27:39 jestocost [n=cmell@cb8a9b-024.dynamic.tiki.ne.jp] has joined #lisp 10:27:56 koning_robot: I don't think learning Emacs would slow you down. That's a misconception. 10:28:04 pjb [n=t@81-66-196-92.rev.numericable.fr] has joined #lisp 10:28:27 koning_robot: You need to read the Emacs tutorial to know the basics to use Slime. That's half an hour of investement. 10:28:32 koning_robot: there is viper-mode and others that essentially makes emacs a vim. missing out of emacs is really killing your productivity with common lisp. :/ 10:29:37 if any of you has a spare two minutes, you could sign up with mahalo answers and send this guy a direct question. might help save his life 10:29:49 isn't killing productivity an overstatement ? 10:30:22 holycow [n=rtaylor@S01060016b6b53675.vf.shawcable.net] has joined #lisp 10:30:25 trolltard, go away 10:30:36 trolltard: This is off topic, and looks like scam. 10:31:02 its not a scam, I came here because I thought freenode and lisp in particular would be full of benevolent people 10:31:27 lhz: well, my point was that if you are fucking around with with vim and try to play the lisp game in the same way you play the C game, you /are/ not doing it very efficient, now are you? 10:31:27 lhz: I think Slime provides a 10 fold increase of productivity, if not more, compared to using copy&pasting into the SBCL repl. 10:31:29 galdor [n=galdor@bur91-2-82-231-160-213.fbx.proxad.net] has joined #lisp 10:31:40 the guy exchanged like three messages with me, mahalo answers wont let me send more questions to him, i just dont want a guy to actually die 10:32:16 so you are on irc? 10:32:20 fuck off and all 911 10:32:36 can someone kick/ban this dick ... 10:33:03 i dont know who it is 10:33:42 hypno, tcr: you both point out how to update code in your running image. I do that by using asdf. But yes, emacs do that in an smaller circle. One thing. If you have several sources in emacs and update here an there, how do you send all changes to your image ? 10:34:08 lhz: I send my changes as I do them. 10:34:16 dito for me. 10:34:37 lhz: But it's not just that. The Slime debugger, and Slime inspector are very value pieces to the puzzle, too. 10:34:54 lhz: And automatic display of a function's arglist, etc etc 10:36:13 and, of course, all the other features that emacs brings to the table. 10:36:43 lhz: I think you're really missing something out. Especially if you never tried it. 10:39:58 -!- trolltard [i=bizinatc@cpe-66-75-6-226.san.res.rr.com] has quit [] 10:40:14 Yes I do miss out, but I just can't get confortable to use it. On my long-time running image I instead have an telnet-interface where I do updates. 10:41:35 For what it's worth, I think humans are capable of totally miscalculating their long-term gains in productivity from reasons of short-term discomfort. 10:42:08 I see this daily, when my students waste more than half their time just because they don't want to look up a command or two. 10:42:26 <_8david> it's probably as much work to learn a new development environment or editor as it is to learn a new programming language 10:42:34 <_8david> (in the end, neither is really hard though, so just get it over with) 10:42:43 _8david: That sounds about right. 10:42:44 hacker vs engineer 10:42:53 hacker skills are great when the tools aren't around 10:43:00 beach: some even realise it and feel very bad about it whilst not looking the data up (I might be talking about myself) 10:43:06 but when the tools are there, continuing to hack half-baked solutions is suboptimal :) 10:43:33 it is very strange that I sometimes can't find the energy to do that, it feels like I'm sleeping whilst I have that 10:44:01 madnificent: yes, it seems to be a very strong psychological phenomenon. 10:44:41 I wonder why we do that, it can't possibly be that we evolved into something that stupid 10:45:04 "stupid" is not global in that sense 10:45:13 madnificent: It's actually very wise. 10:45:16 It seems people don't "naturally" strive for productivity, but instead settle for an approximation such as "familiarity". 10:45:47 if we develop a trait that helps in X situation, yet new Y situation comes up where the trait is a detriment, initially developing the trait wasn't stupid 10:45:51 madnificent: If you behave like you behaved in the past, you know that you survided last time doing it, so there's a good chance that you'll survive in future, too. 10:46:09 Perhaps familiarity is a good approximation in most cases, but it fails spectacularly when it comes to using programming tools. 10:46:24 familiarity is the enemy of progress 10:46:56 _8david: I don't think learning a new development enviornment is as much work as learning a new programming language. 10:47:12 tcr: point taken, yet you could assume that people searching for better ways to survive (whilst not being haunted) had a bigger chance of survival. Yet most of us seem to drop down hald-dead when the work is (allmost) done. 10:47:37 -!- ALT777 [n=777ALT@99-163-22-30.lightspeed.mssnks.sbcglobal.net] has left #lisp 10:47:41 tcr, _8david: depends on the language and the environment 10:47:44 ;) 10:47:48 tcr: idunno, it's a different challenge. You commit a lot of you environment to reflexes and such 10:48:31 _8david: Learning a new programming language is pretty much a one, or two year investement. Learning a new editor is a much more incremental experience. 10:48:59 Anyway, as I point out to my students, it's their duty to overcome this psychological force, do the calculation to see what is more productive, and then just go ahead with the investment. 10:49:29 robyonrails [n=roby@host60-240-dynamic.10-87-r.retail.telecomitalia.it] has joined #lisp 10:49:47 beach: and how many interpret that? It seems to me that it would perfectly fit in the story, to just ignore that advise as well :) 10:49:48 beach: The problem is that you see yourself possibly doing a wrong investment. 10:50:28 in the IT-world just about everything is hyped as an improvement tho. i think most of us here are rather conservative in the sense that we block out a lot of the new bullshit that gets around only to be "replaced" a year later. :) 10:50:30 tcr: Yes, sometimes we have no choice but trying something that might fail. 10:50:38 tcr: that too can be calculated, though. 10:51:13 http://www.labri.fr/perso/strandh/Teaching/Langages-Enchasses/Common/Strandh-Tutorial/psychology.html Relevant? 10:51:16 madnificent: Sure, just the way everything we try to communicate is ignored by some 85% of the students anyway. 10:51:53 hypno: XML! XML! X-M-L! rah rah! ;) 10:52:10 except some of it doesn't get replaced :( 10:52:23 beach: that is fair. Reaching 15% of the students by some short information that might raise their overal productivity by say 20% is very good 10:52:26 sykopomp: there should be a shorter URL for that http://dept-info.labri.fr/~strandh/Essays/psychology.html or something like that. 10:52:33 sykopomp: and yes, it is very relevant. 10:52:40 hypno: broken by design -_- 10:53:19 madnificent: I know it is good enough. I did the calculation when I got depressed because I wasn't convinced I served any great purpose. Now I am convinced. :) 10:53:43 I mean, I think it's also worth noting that regardless of the effort to be put into learning something that's apparently 'better', it's also very hard to get out of one's comfort zone and just learn something properly. 10:54:07 -!- benny [n=benny@i577A20FA.versanet.de] has quit [Read error: 110 (Connection timed out)] 10:54:17 For example: I've had a horrible time learning anything other than Lisp in the past few months. My eyes glaze over, I get very irritated, and I keep wanting to go back and hack something up in Lisp. 10:54:28 CHJA [n=christia@90-231-255-161-no145.tbcn.telia.com] has joined #lisp 10:55:21 It bothers me, since I -do- want to learn these other languages (right now, C, python, asm), but I'm getting impatient and annoyed if I even read through a tutorial. The most I managed to get myself to do was write a couple of functions in python (half of them without a body) 10:56:23 sykopomp: It is also entirely possible that Lisp is one of the better languages, in which case no such investment will represent a gain in productivity, and so any investement is bound to be a net loss. 10:56:50 sykopomp: that feeling, I recognise, however it is twofold. I want to be able to create the 'right' abstractions in the program that is being written. As I see the abstraction, I want to map it to the language at hand. However, many languages (say Java) don't support the constructs to define those abstractions. Somehow my brain gets stuck in an endless loop, trying to figure out how it can be done. That lasts untill someone hacks 10:56:50 something very ugly up for me, and I iterate over it some times to make it more better (then at least I don't feel like I'm destroying the program) 10:56:54 sykopomp: but then you have to invoke the other reason to learn something new, namely to have arguments in favor of what you can now compare. 10:57:14 <_8david> tcr: let's discuss docstrings, not IDEs! 10:57:28 beach: Lisp might be technically and ideally better, but there's a lot of gains to be had from learning other languages. Lisp isn't quite that nice for scripting *nix, for example, it's also not as integrated as C. There's also a lot of work out there if you know C, or even python, and there's actually good tools for those languages. 10:57:58 Lisp isn't perfect, even though I love it more than anything else, so it makes me feel like I'm falling into the same trap I accuse everyone else of falling into. 10:58:00 madnificent: You have to consider the possibility that your idea of what is the right abstraction is determined by the languages you know, and that there might be better ones in other languages. 10:58:47 sykopomp: I see your point. 10:58:50 beach: exactly. You should strive to use the language as it was intended to be used, to see where its strengths lie 10:59:09 -!- Quadrescence [n=quad@c-71-193-68-220.hsd1.mn.comcast.net] has quit [Remote closed the connection] 10:59:20 beach: yes, I was pointing at the writing of code in other languages, not the learning of new languages. Learning languages still seems interesting to me. 10:59:25 Quadrescence [n=quad@c-71-193-68-220.hsd1.mn.comcast.net] has joined #lisp 10:59:30 malumalu [n=malu@hnvr-4dbb4474.pool.einsundeins.de] has joined #lisp 10:59:41 madnificent: I thought about that, which is why I included python in the example. I learned python before, I wrote pretty long scripts in it... before I started learning lisp. Now I'm all rusty. The thing about python is that it has a big enough toolkit that a lot of these 'how do I do this' questions get answered pretty fast. 10:59:42 _8david: I already pulled out. I saw your posting, but haven't read through it yet. 11:00:03 sykopomp: but that is only a problem beacuse you are trying to forcefit Lisp into the world of unix. (it's the same with the idiots who try to force fit Erlang semantics upon CL yet remain Common Lisp). it is also rather evident amongst most FFI-hacked up interfaces to lisp. these things are not bad because of lisp, but because they are never lispified and so remain to function at a retarded C level. 11:01:02 hypno: I'm not trying to forcefit it, but CL doesn't have as much easy integration with the system. This is not so much a flaw with CL as it is a natural advantage of C-like languages. 11:01:29 sykopomp: or a disadvantage of Unix. 11:01:40 I mean, it's like trying to fit C code into Genera or something. 11:02:01 sykopomp: But there are two ways of fixing that: replace the langauge, or replace the OS. 11:02:52 beach: which is why I would love a LispOS. But since there's not one right now that I could simply switch to, I don't mind using languages like python or *sh for system scripting. 11:03:00 sykopomp: I for one think that "modern" OSes still use abstractions from 40 years ago that should be gotten rid of as soon as possible. 11:03:22 sykopomp: faire enough. 11:03:24 *fair 11:03:30 OSs seem to naturally be slow-evolving beasts, too. 11:03:38 Indeed. 11:04:11 And they're also not exactly easy projects :< 11:04:31 People are so used to the "Save" menu entry. Just like stick shifts on cars. Both are flaws of the underlying technology that people get so used to they actually want them. 11:05:17 sykopomp: I don't think it's *that* hard to write an OS. 11:05:17 I think stick shifts are part of the --omg-optimized world that C++ people and the like seem to live in. 11:06:23 Nshag [i=user@Mix-Orleans-106-3-192.w193-248.abo.wanadoo.fr] has joined #lisp 11:06:56 sykopomp: yeah, give me paddle shifters any day :) 11:07:03 writing a complete OS might be to stretch it tho. why is it not enought to encapsulate the stuff that matters on existing unix kernels? 11:07:29 because not everybody uses Unix kernels? :-P 11:07:35 yeahh... 11:07:36 the only reason it is seen as impractical in CL is because it has never been integrated properly. integrating these things are way easier than a new kernel. 11:07:57 and I mean, if you're writing a lispOS, it's much nicer to integrate it completely. 11:08:05 hypno: you mean standards, right? 11:08:27 yoshinori [n=read_eva@router1.gpy1.ms246.net] has joined #lisp 11:08:40 hypno: not just a new kernel, a completely different operating system. One big fat lisp image (or several images to switch between!). 11:09:08 personally, the idea of a full-fledged lispOS makes me drool :) 11:09:28 sbcl-os should get some support from people that actually know how to write an OS. I'd be glad to help out by writing some inferior code though. But I lock useable knowledge of the underlying hardware. 11:09:36 s/lock/lack 11:09:53 Phoodus: hmm. well, in part perhaps. i'm mainly saying that these things could be working as nicely as the rest of Cl with proper abstractions. that part is rarely done, and so people settle to fuck around with low level sockets and whatever nonsense istead of proper closified frameworks which abstracts the functionality away from the underlying nonsense. 11:09:55 *sykopomp* needs to get the dinosaur book, too. 11:10:00 hypno: that is very hard to do, given the semantics of Unix when it comes to the difference of primary and secondary storage. 11:10:42 hypno: yeah, I see that a lot. When writing apps, they build it around unixisms/win32isms/etc instead of around their application abstractions 11:11:52 beach: you lost me. what specificly do you mean? 11:11:55 kzar [n=kzar@hardwick.demon.co.uk] has joined #lisp 11:12:27 hypno: Unix has files. I see a Lisp OS as having a single, persistent memory. 11:12:42 chris2 [n=chris@p5B16C75E.dip0.t-ipconnect.de] has joined #lisp 11:13:21 beach: delicious in-memory trees of CLOS objects. 11:13:35 beach: the problem with that is the whole notion of "commit". When editing something in a text editor, you have your "in progress" stuff, while the latest "committed" stuff is on disk 11:13:41 Geez. An extensible CLOS-based filesystem would be amazing. 11:13:48 hypno: You can write a Lisp OS as a Unix process, but you have to make sure that when you pull the plug, your objects are still around when you come back up. 11:14:03 beach: editing everything live is dangerous unless you set up another edit/commit cycle somehow 11:14:17 Phoodus: we should have unrestricted undo instead of the binary semantics of Unix. 11:14:48 or state branches or something 11:15:01 Phoodus: I think you got it wrong. What is dangerous is to never be able to go back once you say "OK". 11:15:38 sykopomp: why would you want files? 11:15:39 how do you say "OK" with a single memory image? 11:16:03 if you need to change 5 separate things in a live asynchronous system, that's not easy 11:16:05 beach: ah, heh, yes indeed. in theory of course, i would prefer a real lispified os (with a 3d-gui) to boot. i think it is less work to get "almost there" with existing kernels tho. otherwise all the time will be eaten up with driver devleopment. :( 11:16:09 Phoodus: every keystroke (or nearly so) can be undone so it is a "Maybe". 11:16:43 For an example of a single level store, have a look at EROS, and in particular: http://www.eros-os.org/papers/storedesign2002.pdf 11:16:45 beach: I didn't quite mean files in the same way they are in UNIX, but more in the sense of "I have this thing here that holds X data, which I can locate in a number of ways, possibly by name" 11:17:02 I really don't know what to call something like that if not a 'file' 11:17:08 -!- kiuma [i=51d06a4b@gateway/web/ajax/mibbit.com/x-801772dd0a86736d] has quit ["http://www.mibbit.com ajax IRC Client"] 11:17:17 Phoodus: If you need that kind of functionality, it is easy to accomplish if you have unlimited undo. The other way around is much harder. 11:17:38 kiuma [i=51d06a4b@gateway/web/ajax/mibbit.com/x-ad9d4ac08de8564f] has joined #lisp 11:17:38 I don't know what "undo" exactly implies in the case of highly async systems like OSes 11:17:42 i for one welcome our new 1980s-office-metaphor-busting overlords 11:17:49 beach: I can certainly see a middle ground here where anything you edit can be revision-controlled. Tagged, reverted, patched, etc. 11:18:00 but still, everything discussed here was solved and implemented in mainframes decades back 11:18:01 hypno: I disagree, actually. I think you can get very far if you do it as a Unix process, and use a battery to avoid data loss. 11:18:29 sykopomp: Sounds better. I would prefer tagging, sort of like gmail. 11:18:40 sykopomp: call them "objects". 11:18:51 pjb: yes, EROS is nice. 11:18:54 beach: worksforme 11:19:07 pjb: But I have come to understand that they are solving the wrong problem. 11:19:25 beach: sounds like we are in agreement. for a practial approach, use existing kernel and leverage the support they bring. a lispos for production use is very close to a pipe dream. 11:19:57 Phoodus: not much is new in effect. 11:20:12 beach: it's worth entertaining the idea that since a system like this would run on top of CLOS, we could conceivably make it so flexible that you could simply write a new class/method and change the behavior of the system 'revision-control'. It would be awesome to leave that open and let people mess with it. :) 11:20:25 Well, single level storage is only part of it. The main purpose of EROS is to implement a capability based OS. In coyotos, I think they even dropped the SLS. 11:20:37 sykopomp: exactly. 11:20:56 pjb: right you are. 11:21:02 sykopomp: sounds more like the realm of an expert system to me 11:21:08 *tcr* just rm'd a file by mistake 11:21:13 gosh I hate unix 11:21:58 binarycodes [n=sujoy@59.93.242.192] has joined #lisp 11:22:10 pjb: A Lisp OS would be "naturally" capability based though. Just (or perhaps with some variation) use the pointer to the object as the capability. 11:22:28 tcr: alias rm='mv $1 ~/.trash'? (I don't think that works?) 11:22:30 Yes. And SLS could be implemented over linux processes. journaled file system already do the hard part. 11:22:39 tcr: you may be able to resurrect said file with debugfs. But, it may be more trouble than its worth. 11:22:57 sykopomp: better, use your own rm binary. 11:23:27 sykopomp: That'd be awfully slow if the file is on another partition 11:23:47 tcr: indeed 11:24:02 joachifm [n=joachim@ti132110a340-2798.bb.online.no] has joined #lisp 11:24:03 tcr: you could also alias it to something that would ask for confirmation... 11:24:15 alias rm='rm -i' 11:24:25 pjb: I know understand that persistence and checkpointing are very different animals. 11:25:21 Yes, checkpointing is a technique used to implement persistence in EROS. 11:25:22 pjb: Persistence can be had easily with just battery backup. Checkpointing is useful for a number of other things. 11:25:47 pjb: yes, which I know think is the wrong way of doing it. 11:26:37 sykopomp: I'd just type `y' out of habit all the time 11:27:03 tcr: the solution is obviously to join or start an effort to write a proper lispOS :) 11:27:05 beach: that's 11:27:05 tcr: Interaction designers agree with you. 11:27:18 *tcr* wishes debufs supported some kind of "show recently removed blobs" 11:27:55 beach: that's funny that no vendor addd batteries to their computers... 11:28:09 pjb: I agree. It seems so obvious. 11:28:13 beach: The first principle of UI design is that _everything_ should be undoable. 11:28:23 tcr: indeed. 11:28:35 pjb: and UPSes only last for a very short time. 11:29:02 tcr: http://e2undel.sourceforge.net/ may be of assistance 11:29:23 pjb: would you like to write an OS? :) 11:29:30 binarycodes_ [n=sujoy@59.93.255.215] has joined #lisp 11:29:46 I toyed with the idea for a long time. :-) 11:30:27 jso: only for ext2 11:31:35 pjb: Following my conviction that it is best done top-down, the next step would be to have a persistent Unix process with a tags-based object store. 11:31:58 tcr: does lsdel and undelete do what you need? 11:31:59 pjb: One has to give some thought to protection and perhaps multi-user issues. 11:33:02 Capabilities seem nice for that. 11:33:03 jso: lsdel is part of e2undel? 11:33:24 tcr: nah, debugfs 11:33:27 -!- mrSpec [n=SomeOne@88.208.105.1] has quit ["c Ya!"] 11:33:31 pjb: One idea I had was to use a few (8?) bits of a 64-bit pointer to indicate access rights. Each object type would interpret those bits as it sees fit. 11:33:35 tcr: have you typed 'help' ? 11:34:12 jso: Oh nice. 11:34:14 pjb: so there could be several pointers with different access rights to the same object. 11:34:50 That's basically what capabilities are, pointers with access rights. 11:34:56 I know, yes. 11:35:07 jso: It doesn't find find my file though 11:35:12 pjb: I was just trying to figure out an efficient way of implementing them in the context of Lisp. 11:36:01 decafbad [n=mehmet@88.232.66.140] has joined #lisp 11:36:08 Make it work first, make it efficient then. 11:36:19 tcr: I'm playing with it a little, its been a while since I used debugfs. And when I did it was for an OS class project... so, I really didn't worry about recovering data. 11:36:25 pjb: before accessing memory, these additional bits would be masked out, which is pretty fast. :before methods on some types would check the access bits. 11:37:07 In MacOS we had a bad expercience with using bits in addresses for something else than address... 11:37:32 pjb: bah, we do that for type information already. 11:37:38 Transition to full 32-bit pointers was painful. 11:38:16 pjb: I am betting it is going to take a while before we need more than 56 address bits. 11:38:29 Well, it's true that in the case of lisp, we're speaking of a virtual machine, so we can consider we have 56 bit addresses, yes. 11:38:39 -!- wedgeV [n=wedge@cm56-238-229.liwest.at] has quit [] 11:39:19 jso: I wonder why it can't find it, the file was big. 11:39:40 well, if there hasn't been much activity, it should still be there. 11:40:06 jso: It was on the root system, but all my interesting directories are on seperate lvm volumes (/home, /tmp, /var) 11:40:27 -!- vasa [n=vasa@mm-178-89-84-93.dynamic.pppoe.mgts.by] has quit ["I am not vasya, i am vasa"] 11:42:22 Hm. A MUD is an OS. Mind = Blown. 11:42:42 beach: http://www.ai.mit.edu/projects/aries/course/notes/tagged_paper.pdf mentions M-machines that has "guarded pointers". 11:42:51 -!- binarycodes [n=sujoy@59.93.242.192] has quit [Read error: 110 (Connection timed out)] 11:43:03 Yuuhi [n=user@p5483F991.dip.t-dialin.net] has joined #lisp 11:45:01 sykopomp: ? 11:45:03 konqueror 11:45:05 bah 11:45:40 madnificent: It occurs to me it would be pretty fantastic to think of an entire MUD engine as just an operating system. Answers a lot of design questions by default :) 11:46:07 Mazingaro [n=tetsuja@host74-228-dynamic.22-79-r.retail.telecomitalia.it] has joined #lisp 11:46:41 if you're going to call a MUD an OS, then a Lisp machine is even moreso an OS and you're done :-P 11:47:09 and the abbreviation of mud would be what in this context? 11:47:20 madnificent: multi user dungeon. A game. 11:47:29 oh, allright 11:47:36 sykopomp: tron style ;) 11:48:09 madnificent: MUDs are text-based online games :) 11:48:10 pjb: thanks. Intersting. 11:48:14 think WoW without a GUI. 11:48:27 -!- holycow [n=rtaylor@S01060016b6b53675.vf.shawcable.net] has quit [Read error: 104 (Connection reset by peer)] 11:48:48 holycow [n=rtaylor@S01060016b6b53675.vf.shawcable.net] has joined #lisp 11:49:25 sykopomp: wow, that's one hell of a suggestion. I'd say, "Think Zork, but with more people." 11:49:32 sykopomp: besides racing games, graphics are actualy one of the only things I care about in games (perhaps I'm some sort of anti-gamer) 11:50:12 madnificent: you and i agree on this point 11:50:14 madnificent: There are things you can't possibly express through graphics. I'd like to make those things shine. 11:51:13 nowaitfiles [n=lookiden@c-76-109-123-17.hsd1.fl.comcast.net] has joined #lisp 11:51:32 or you could describe a MUD as IRC with single-room presence, bots and stats 11:51:56 madnificent: i also like when attention i paid to gameplay (visually) and control 11:52:20 tcr: I've found a decent start. dump the directory that held the file out to a file 11:53:13 -!- nowaitfiles [n=lookiden@c-76-109-123-17.hsd1.fl.comcast.net] has left #lisp 11:53:19 Phoodus knows all my irc notification triggers I think: MUD, Cyc and ABCL 11:53:22 tcr: that file will contain the inode table of the directory. We can get the inode from that. You should be able to find the name of your old file stored in there. 11:53:27 heh 11:53:36 Dinner, I might be back later. 11:53:40 -!- beach [n=user@203.210.248.212] has left #lisp 11:54:14 -!- O_4 [n=souchan@125-236-190-161.broadband-telecom.global-gateway.net.nz] has quit ["Come alive!"] 11:55:20 Phoodus: why one channel? An IRC-server with one channel per location would be far cooler. That way you only hear things for the room you're currently in. The backend would then send you from room to room (where indeed, bots give you credits and whatnot) 11:56:00 Phoodus: It could even be straight-forward to implement (and with only an irc-client needed to play the game, it is playable everywhere) 11:56:03 "single-room presence" == you can only be in 1 room at a time 11:56:05 madnificent: I think that's what he meant. 11:56:19 Phoodus: oh, and fights could be in private chats :P 11:56:27 Phoodus: nice of you! 11:56:31 sykopomp: could be... 11:56:35 madnificent: also, if you develop a flexible enough engine, you can actually link up just about any client to it (that's what I'm doing. MVC ftw) 11:56:38 sykopomp: perhaps something for in your tutorial? 11:57:29 jso: You see my privmsg? 11:57:39 madnificent: pffft :P 11:57:51 lol 11:57:55 so about that tut ... 11:57:57 *cough* 11:57:58 hehehe 11:58:08 ;_; 11:58:23 sykopomp: what, is there something wrong with it? 11:58:37 sykopomp: Make me laugh 11:58:37 madnificent: nah, I just keep putting it off. 11:59:18 you have 3 days to new years 11:59:21 get cracking :) 11:59:21 Quadrescence: (cl-ppcre:regex-replace-all "ss|s" "What kinds of things make you laugh?" "th") 11:59:33 sykopomp: how about this: you publish it, you add a link to some channel on freenode and you allow people to rewrite/add what they want. You review it and choose what you want in there... That way we can link to the tutorial and say 'it isn't perfect yet' and you don't have to take the blame :) 11:59:34 holycow: I'll get to it next year or something. 12:00:03 sykopomp: har har har 12:00:10 ThomasI [n=thomas@unaffiliated/thomasi] has joined #lisp 12:00:15 madnificent: I mean, sure, I'll do that once I get another page written. It's mostly that I keep having other stuff come up (like being home with family, finishing a semester, etc) 12:00:19 lol 12:00:31 ironChicken [n=richard@79-75-104-125.dynamic.dsl.as9105.com] has joined #lisp 12:00:47 it's not so much that it's not perfect, but that I've been procrastinating :3 12:00:56 sykopomp: point taken, days ore not long enough 12:01:16 sleep is annoying. Major design flaw, imho. 12:01:35 younder [i=jpthing@062016236162.customer.alfanett.no] has joined #lisp 12:01:36 "Sleep is only for those who don't deny their need for it." - me 12:01:38 i wouldn't mind if someone would write two shot'ish tuts 12:01:41 ID would be more convenient if little-man-on-cloud had a Trac or something. 12:01:47 -!- bpt [n=bpt@cpe-071-070-209-067.nc.res.rr.com] has quit ["Leaving"] 12:01:53 one on 'intro to lisp- what every noob should know' 12:01:56 cl of course 12:02:09 Lesson 1 - The Meta-Object Protocol 12:02:16 hahaha 12:02:19 the other would be just a bunch of excercises onhow to get data into and out of various data stores 12:02:24 schaueho [n=schauer@dslb-088-066-025-243.pools.arcor-ip.net] has joined #lisp 12:02:36 whatever you call arrays and such 12:02:39 Phoodus: I would think Lesson-1: Macros would be more ridiculous, but it's already been done (Casting SPELs) 12:03:09 The most obscure area is the protocol for modifying the reader 12:03:27 reader macros for everyone! 12:04:13 madnificent: what I'm more likely to do is publish a version of the tutorial with no images, or only some images, and put a request out for people to help fill in the blanks, edit images, etc. 12:05:00 jli [n=jli@adsl-074-229-201-181.sip.gnv.bellsouth.net] has joined #lisp 12:05:09 sykopomp: what kind of tut have y ou been thinking about doing anyway? 12:05:32 holycow: I've been planning out what to do for the third page (introducing lists) 12:06:15 Lesson 3.1 - caaddaaaddddddaaddr 12:06:32 Phoodus: haha. YES. 12:06:47 or alternatively, mapcan 12:06:48 isn't there a macro for making setf-like functions? 12:06:57 Yes We MapCan 12:07:39 jii: yes 12:07:43 jli: (defun (setf reader) (new-value target) ...) 12:08:06 sykopomp, that's not what I was thinking of 12:08:23 sykopomp: no defmethod (setf .. 12:08:35 jli: define-setf-expander 12:08:49 younder: the name of the function is (setf reader) 12:09:01 or something. It's a function. Unless you want to be able to extend it. 12:09:30 -!- kzar [n=kzar@hardwick.demon.co.uk] has quit [Remote closed the connection] 12:09:59 kzar [n=kzar@hardwick.demon.co.uk] has joined #lisp 12:10:29 jli: sykopomp showed you the right way to create setf functions. 12:10:43 jli: define-setf-expander does not define a setf function, but a setf expander. 12:11:03 tcr: I think he meant something other than setf functions... 12:11:32 I didn't mean setf functions. I meant a macro for defining macros that have side-effects, I think 12:11:36 jli: In general setf functions are preferred for the same reason that functions are preferred over macros. 12:11:48 -!- awayekos is now known as anekos 12:11:51 jli: you're perhaps thinking of define-modify-macro 12:12:09 -!- jestocost [n=cmell@cb8a9b-024.dynamic.tiki.ne.jp] has quit [Read error: 60 (Operation timed out)] 12:12:21 that sounds kind of right 12:12:50 sykopomp: You almost always see it used in that context 12:13:36 yes, I think that's it. thanks tcr (and sykopom, younder, locklace) 12:13:43 setf was after all introduced to allow easy access to CLOS slots 12:14:00 younder: uhm, no it wasn't 12:14:21 It later found uses in other areas 12:14:38 setf existed before CLOS 12:15:14 mulligan [n=user@e178031222.adsl.alicedsl.de] has joined #lisp 12:15:14 tcr: CLOS existed before LISP 12:15:29 wow, define-modify-macro is kind of magical 12:15:40 no, set, followed by setq followed by setf 12:15:48 clhs define-modify-macro 12:15:48 http://www.lispworks.com/reference/HyperSpec/Body/m_defi_2.htm 12:16:18 jli: It's just a wrapper around GET-SETF-EXPANSION. 12:16:36 and setf.. 12:16:58 jli: Which you may need some time to wrap you head around 12:17:31 Peter Norvig has a pretty good explanation in PAIP 12:17:55 pchrist|univ [n=spirit@gateway.hpc.cs.teiath.gr] has joined #lisp 12:18:08 In general, you don't really need to wrap your head around. :) Just use CALLF! (http://common-lisp.net/~trittweiler/darcs/alexandria/modify-macros.lisp) 12:18:55 shmho [n=user@58.142.15.103] has joined #lisp 12:20:22 -!- schaueho [n=schauer@dslb-088-066-025-243.pools.arcor-ip.net] has quit ["Bye bye"] 12:21:57 Does anyone here use the Pretty Printer to format data? 12:22:25 younder: you mean FORMAT? 12:22:37 with ~W 12:23:32 Multiline formats witch indent according to context. 12:23:43 Is there a clever way to concatenate a list of strings? I made a macro to do it but it doesn't work if the list is made dynamically 12:24:41 kzar: (let ((*print-pretty* nil)) (format nil "~{~A~}" list)) 12:25:20 tcr: Oh yea I didn't think of using format 12:26:12 is *print-pretty* t by default in SBCL? 12:26:49 younder: Yea, that's why he needed to turn it off 12:27:01 works well, cheers tcr 12:28:14 ah. same as CMUCL.. Not all compilers have it enabled. it severly slows down output to files. 12:28:44 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 12:28:45 kzar: what's wrong with (apply #'concatenate 'string list-of-strings) ? 12:29:44 that's `(conc @,list-of-strings) for cl-who users :-P 12:29:54 *fusss* loves jumping in the middle of conversations 12:30:03 ,@ even 12:30:24 -!- ManateeLazyCat [n=user@222.212.142.244] has quit [Read error: 131 (Connection reset by peer)] 12:30:33 Hmm pretty snazzy 12:30:44 locklace: call-arguments-limit 12:31:07 -!- Phoodus [i=foo@ip68-231-38-131.ph.ph.cox.net] has quit [] 12:31:16 tcr: fine, (reduce #'(lambda (a b) (concatenate 'string a b)) list-of-strings) 12:31:19 ManateeLazyCat [n=user@222.212.142.244] has joined #lisp 12:31:43 #' considered superflous 12:31:53 nyef [n=nyef@pool-71-255-137-188.cncdnh.east.verizon.net] has joined #lisp 12:31:56 G'morning all. 12:32:11 concatenate is a bit heavy.. 12:32:22 kzar: use the reduce form if list-of-strings is large 12:32:36 younder: i'd have thought format was heavier 12:32:38 try with-output-to-string 12:33:33 fusss, is it? I thought it was good style 12:34:20 do all implementations have lambda defined as a macro? 12:34:22 fusss_ [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 12:34:26 -!- binarycodes_ [n=sujoy@59.93.255.215] has quit [Client Quit] 12:34:44 That just adds the string to the end 12:34:49 binarycodes [n=sujoy@59.93.255.215] has joined #lisp 12:34:50 jli: a recent post at planet.lisp.org goes through all the gory details, for your amusement 12:35:01 -!- Beket [n=Beket@athedsl-22493.home.otenet.gr] has quit [] 12:35:34 locklace: Still not good enough. 12:35:47 locklace, looking 12:35:58 locklace: you need (reduce #'(lambda (result new) (concatenate 'string result new)) (rest list) :initial-value (first list)) 12:36:13 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 12:36:19 -!- fusss_ is now known as fusss 12:37:13 <_8david> tcr: that results nil rather than "" for the empty list, which is the right result either, is it? 12:37:25 <_8david> but anyway, the REDUCE suggestion is the worst approach so far 12:37:42 <_8david> portable, but portably bad 12:37:50 or (with-output-to-string (dolist (s list) (princ s stream))) 12:38:30 couldn't you write something with dolist and side-effect? 12:38:56 there should be a (stream) after with-output-to-stream 12:39:21 _8david: why worst? 12:39:33 <_8david> It's also marking a new low for this channel that younder offers a better solution than the rest of the channel. 12:39:57 <_8david> (Not taking into account tcr's very cute FORMAT suggestion, but it went downhill from that.) 12:42:28 _8david: prepare for many "low's" ;) 12:43:52 _8david: ummm, younder is a _pretty_ good lisper :-) 12:44:28 Not according to xach 12:44:52 I am in fact the idiot of the chanel 12:45:05 whatever 12:45:09 -!- Yuuhi [n=user@p5483F991.dip.t-dialin.net] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 12:45:22 younder: how often have you been frequenting here? 12:45:39 -!- ThomasI [n=thomas@unaffiliated/thomasi] has quit ["Bye Bye!"] 12:45:44 just glad the /whois showed up a familiar name, from cll 12:45:49 since the last time I was 'banned' abot a week 12:46:03 mrSpec [n=SomeOne@88.208.105.1] has joined #lisp 12:46:14 he tends to drool in confusion 12:47:12 ok, i really can't get into personal histories and chan politics 12:47:20 Let's let personal differences aside and focus on Lisp rather than a pointless word war (I never drool) 12:48:08 vicious 12:52:38 oh well, good thing I have a Herzog film lying around in case there isn't much here :-) 12:53:13 kenyon [n=kenyon_@xdsl-81-173-184-180.netcologne.de] has joined #lisp 12:54:10 hugod [n=hugo@bas1-montreal50-1279441402.dsl.bell.ca] has joined #lisp 12:55:28 Anyhow back to the Pretty Printer A.I Memo No. 1102a might be a souce of inspiration 12:56:00 jgracin [n=jgracin@82.193.208.195] has joined #lisp 12:59:20 ruediger [n=the-rued@p508B6F49.dip.t-dialin.net] has joined #lisp 12:59:44 ejs [n=eugen@92-49-229-119.dynamic.peoplenet.ua] has joined #lisp 13:02:07 -!- kenyon_ [n=kenyon_@xdsl-81-173-148-255.netcologne.de] has quit [Read error: 113 (No route to host)] 13:03:59 madnificT [n=user@83.101.62.132] has joined #lisp 13:04:00 fwiw, the only reasons *i* come here and no longer read cll are: 1) whinnage about lisp popularity, 2) whinage about lack of libraries or difficulty of use, and 3) whining people who want to write MFC style apps in Lisp, for free, and still demand their tools remain buzzword compliant with the latest and greatest of reddit tags. 13:04:26 -!- m4dnificent [n=user@83.101.62.132] has quit [Read error: 113 (No route to host)] 13:04:39 -!- madnificent [n=user@83.101.62.132] has quit [Read error: 113 (No route to host)] 13:05:09 -!- holycow [n=rtaylor@S01060016b6b53675.vf.shawcable.net] has quit [Remote closed the connection] 13:05:27 right.. LISP is for people that DO 13:06:33 string-downcase ;-) 13:06:40 yhara [n=yhara@7.193.12.221.megaegg.ne.jp] has joined #lisp 13:06:51 I guess Edi Weitz is the best example 13:07:07 yeah. use what you can, share what you can. and cheers to you. simple :-) 13:07:13 fusss: oh, it's like, you dont approciate John Harrops excellent Ocaml/F# articles or Eranns continous doomsday preaching? :) 13:07:22 -!- pstickne [n=pstickne@c-71-236-177-238.hsd1.wa.comcast.net] has quit ["Leaving"] 13:07:25 fusss: MFC? 13:07:49 Harrop has a clear agenda. some people need to trash the competition to sell their services. 13:07:59 the whole GUI/interop thing? 13:08:09 holycow [n=rtaylor@S01060016b6b53675.vf.shawcable.net] has joined #lisp 13:08:21 amos_ [n=amos_@host86-145-162-199.range86-145.btcentralplus.com] has joined #lisp 13:08:31 -!- holycow [n=rtaylor@S01060016b6b53675.vf.shawcable.net] has quit [Remote closed the connection] 13:08:46 holycow [n=rtaylor@S01060016b6b53675.vf.shawcable.net] has joined #lisp 13:08:49 fusss: (format nil "~:(~a~)" "LISP") would be more accurate imnsho 13:09:08 Ron is a good hacker, he just needs to come back to his sense and stop bugging people. You can demand features from your vendor (or hire the boys at sbcl-dev) but you can't demand a community of your choice :-) 13:09:54 younder: battyani seems to be into some really cool things too. i wish i had come up with that fpga thing. ubercool. 13:10:04 *sykopomp* doesn't really like all the doomsday talk about lisp. 13:12:02 if someone can't find the good stuff, to hell with him. can't go on click and search? oh, too bad. 13:13:07 s/click/cliki/ 13:13:13 hypno: WTF are you talking about? 13:13:47 Any McCLIM hackers about? 13:15:40 m4dnificent [n=user@83.101.62.132] has joined #lisp 13:16:06 -!- madnificT is now known as madnificent 13:16:53 hypno: what GUI interop thing ar you talking about, because i have a surprise for you :-) 13:17:11 younder: you said lisp was for people that /do/. then you continued and mentioned edi weitz as an example. i chipped in with Marc Battyani as another example. then you have no fucking clue what i am talking about. well, nevermind. 13:18:13 hypno: those are example of people who do things, and share it with other people. usually big tangible things. 13:19:28 IRL, very small solutions tend to have sufficient commercial value to pay the rent, but they're themselves of no use to most people 13:20:13 -!- jli [n=jli@adsl-074-229-201-181.sip.gnv.bellsouth.net] has left #lisp 13:20:32 you have idiots like me who basically use CL like they would use perl :-) 13:21:17 -!- jso [n=user@host-9-143-107-208.midco.net] has quit ["reboot"] 13:21:27 hypno: sorry, I really have no idea what that fpga thing is 13:22:07 tcr: I recommend . Might not have been helpful in your specific case but it sure comes handy in a lot of situations. 13:22:09 perhaps you would care to enlighten me? 13:22:29 younder: indeed, http://www.hpcplatform.com/ 13:23:14 vy [n=user@213.139.194.186] has joined #lisp 13:23:22 younder: you seem to be drawing lots of apparently random flak. My condolences :) 13:24:47 -!- ManateeLazyCat [n=user@222.212.142.244] has quit [Remote closed the connection] 13:24:48 smaglo [i=4b77e9e9@gateway/web/ajax/mibbit.com/x-fc7f0795963210af] has joined #lisp 13:26:01 -!- smaglo [i=4b77e9e9@gateway/web/ajax/mibbit.com/x-fc7f0795963210af] has left #lisp 13:27:19 luis: in my particular case it was a movie, a bit too big to sync online 13:28:57 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 13:29:13 fusss_ [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 13:29:13 -!- fusss_ is now known as fusss 13:30:43 tcr: fwiw, if quality doesn't matter but size does, you can rip it to flv (flash video) with ffmpeg 13:34:38 fusss: isn't FLV just a container format? 13:35:21 indeed 13:35:27 bloody hell it's 13:35:27 i should know better 13:35:36 i actually hacked on the format! 13:35:38 IIRC the video is H.263 13:35:48 or a variant thereof 13:36:04 you're right S11001001 13:36:17 pkhuong: Is it okay to include string-case.lisp wholesale in a project? 13:38:17 Aankhen``: private msg? 13:38:43 Aankhen``: He was about to license it under a MIT-style license (iirc) for _8david. 13:39:03 fusss: Not sure I understand. 13:39:30 tcr: Ah, I see. 13:39:51 -!- cods [n=cods@tuxee.net] has quit ["moving the server"] 13:40:11 -!- ASau [n=user@193.138.70.52] has quit [Remote closed the connection] 13:40:11 Aankhen``: i would make a request like that in private :-) 13:40:55 ASau [n=user@193.138.70.52] has joined #lisp 13:41:41 jso [n=user@host-9-143-107-208.midco.net] has joined #lisp 13:41:55 fusss: Then you'd miss out on what tcr could have told you about it. :-) 13:42:24 *Aankhen``* wonders if an MIT-style licence is compatible with a BSD-style licence. 13:42:25 _8david: There's a typo in the annotate-documentation of ANNOTATE-DOCUMENTATION, in :argument name (s/for/form/) 13:42:28 <--- clueless about the legalities. 13:42:29 -!- joachifm [n=joachim@ti132110a340-2798.bb.online.no] has quit [Read error: 110 (Connection timed out)] 13:44:23 |Soulman| [n=kvirc@138.80-202-254.nextgentel.com] has joined #lisp 13:44:37 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 13:45:05 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 13:45:12 a few days ago i asked if i can rename the win32 clisp binary to something else, because i use it as a win32 long-running daemon and didn't want the client to close it accidentally. 13:45:19 Anyone know the status of the OpenGL backend to McCLIM? I've never seen it work, but I'm not exactly a McCLIM expert. 13:45:22 schme_ [n=marcus@c83-249-80-232.bredband.comhem.se] has joined #lisp 13:45:49 didn't get a clear answer. deployed it as "clisp.exe" and gave the users a ten minute lecture :-) 13:46:47 it scares me that most of is (including myself) are willing to share code, but have no clue whatsoever what legal 13:46:58 it scares me even more, that we have the same issue with about everything we do 13:47:38 jso: I know the Cairo backend works. you might have a better luck leaving a memo for beach or hefner 13:47:55 stassats [n=stassats@wikipedia/stassats] has joined #lisp 13:48:43 m4dnificent: hmm? what is scary about it? you pick a license according to goals and believes and there isnt much else to it? 13:48:55 m4dnificent: why should that scare you and WGS what is "legal" 13:49:14 Aankhen``: yes. 13:49:51 hypno: not knowing what is combinable for instance 13:50:15 fusss: Yeah. I'm aware of that one working. But, I'm beginning development of a multitouch device and I'd like to interact with OpenGL. Instead of building a UI from the ground up, it seems that jumping on the work McCLIM has is a good idea. 13:50:44 ASau: Cool, thank you. 13:50:47 fusss: getting sued over things you have no knowlege about (or just not even knowing the rules of the society we live in) 13:51:05 jso: ping beach and he will give you the low-down. or you can try the mcclim mailing list 13:51:19 jso: I'm not sure. I know it seems to be getting worked on. You should write the mcclim mailing list and see what's up. I know they at least want to get it :) 13:52:01 I'll do that later today when I'm more awake. Thanks for the advice. :D 13:52:14 :) 13:52:26 m4dnificent: oh, well, there's the BSD/MIT/X11/PD-camp and there is the GPL-camp, basically. for "fun" (as in depressing) license issues, checkout why CLISP ended up beeing GPL. :) 13:52:26 jso: you can have minion drop beach a line, actually 13:52:36 jso: he does memos :) 13:52:56 hypno: will do that some day 13:53:00 help memo 13:54:00 sykopomp: usually individuals actively involved know what is working or not :-) many times I was able to fix things and make dow with them, even when they're considered "broken" (usually just means undocumented) 13:54:00 minion: help memo 13:54:24 hmm 13:54:24 To send a memo, say something like ``minion: memo for nick: the memo''. I'll remember the memo for any nick which is the same as the given nick, +/- differences in punctuation, and any nick which is an alias for it, and give it to them when they next speak. 13:54:55 malumalu_ [n=malu@hnvr-4dbbf76e.pool.einsundeins.de] has joined #lisp 13:57:29 m4dnificent: I studied the law a bit 13:58:13 -!- jgracin [n=jgracin@82.193.208.195] has quit [Remote closed the connection] 13:58:22 jgracin [n=jgracin@82.193.208.195] has joined #lisp 13:58:33 envi^home [n=envi@220.121.234.156] has joined #lisp 14:00:45 m4dnificent: there's a concept of imputability, which is accepted in European law systems. 14:00:59 m4dnificent: at least to my knowledge. 14:01:28 m4dnificent: in short, when you reach some age, it is assumed 14:01:33 -!- schme [n=marcus@sxemacs/devel/schme] has quit [Read error: 110 (Connection timed out)] 14:02:05 m4dnificent: that you are to understand responsibilities 14:02:10 usocket, bordeaux-threads, clsql and cl-ppcre; someone tell Garret they exist, and accepting patches for his OS 9 :-) 14:02:33 m4dnificent: and that not knowing the law doesn't relieve. 14:02:37 -!- drdo` is now known as drdo 14:03:57 m4dnificent: if you are not in the state of imputability 14:05:46 caliostro [n=caliostr@host-84-222-134-28.cust-adsl.tiscali.it] has joined #lisp 14:05:57 fusss: are you sure? 14:06:07 What Lisps run on OS 9? MCL I guess? 14:06:54 We could possibly torture SBCL into it. 14:07:16 why bother 14:07:20 Exactly! 14:07:24 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 14:07:25 masochism 14:07:31 -!- ASau [n=user@193.138.70.52] has quit [Remote closed the connection] 14:07:42 But at least it's "just" a new OS interface and not a whole new backend. 14:07:43 anyway, my fluid databases thing is finished I think 14:08:05 now to get Kevin Rosenberg to integrate it... 14:08:24 ASau [n=user@193.138.70.52] has joined #lisp 14:08:51 S11001001: what is that? 14:09:08 http://weblocks.lighthouseapp.com/projects/18897-weblocks/tickets/61 14:09:14 silenius [n=jl@yian-ho03.nir.cronon.net] has joined #lisp 14:10:19 magical thread safety for database API and standard-db-objects 14:10:41 niiice 14:12:01 -!- malumalu [n=malu@hnvr-4dbb4474.pool.einsundeins.de] has quit [Read error: 110 (Connection timed out)] 14:12:33 m4dnificent: e.g. mad sick, you're exempt from some parts of law, 14:12:34 m4dnificent: but you're simply treated another way (e.g. isolated into psychiatric hospital). 14:13:21 -!- robyonrails [n=roby@host60-240-dynamic.10-87-r.retail.telecomitalia.it] has quit ["Sto andando via"] 14:13:23 m4dnificent: that's about your "getting sued over things you have no knowlege about". 14:13:32 This system is under too heavy load. :( 14:14:02 <_8david> S11001001: my recommendation: nuke clsql connection pools from orbit. it's the only way to be sure. 14:14:35 I don't even know what that means but thanks 14:15:03 -!- kpreid [n=kpreid@72-255-42-36.client.stsn.net] has quit [] 14:15:18 _8david: Did you try that genesis modification yet, or is that something to try later? 14:15:37 i am new to emacs and slime too, how do i associated a source file into slime, so that i can compile and run through C-c C-c? 14:16:16 store your sources in .lisp files 14:16:43 ASau: but still, I could get sued for bundling two software-pieces together whereas both of them had a 'free' license. If I didn't know both were incomptible, I'd be breaking the law, altough it wouldn't be obvious for me... 14:16:46 <_8david> something to try later 14:17:22 m4dnificent: how can you prove that you know or don't know something? 14:17:38 stassats: fine! It's working! thanks 14:17:40 Okay. I'd be curious to know how it turns out, if/whenever you do try it. 14:18:31 ASau: another example: apparantly in belgium, we are not allowed to sell stuff without warranty that it will work. Therefore, the GPL is not really completely applicable in this country. I didn't know that at first (and I still don't really know what the consequences are). 14:18:42 stassats: it simple scares me, I'm not saying that it's unjust :) 14:19:01 -!- yoshinori [n=read_eva@router1.gpy1.ms246.net] has left #lisp 14:19:29 m4dnificent: Not allowed to sell stuff without warranty doesn't scare me so much as not being allowed to give stuff away without warranty. 14:20:07 nyef: see, I wonder about that. I don't know how it is stated. And I have no clue whatsoever as to how I could find out (on a student's budget, that is) 14:20:26 Can anyone summarize what dynamic-wind does in one or two sentences? 14:20:54 If it's a legal precedent thing, you're kindof screwed. If it's legislated, the text is possibly online. 14:21:49 I know that things like the Massachusetts General Laws are online, but that's only half the story. 14:22:20 -!- vy [n=user@213.139.194.186] has quit [Read error: 104 (Connection reset by peer)] 14:22:34 By the same token, AIUI, in some juristictions, it's illegal to "practice law" without having passed some exams... and that means even for your own benefit. 14:23:10 nyef: I have followed a legal course especially to get a grip on these things, but I'm just not really comfortable with it. I just tend to find it rather strange, that's all I guess 14:23:31 There's probably some way of looking at it that makes it all at least comprehensible. 14:23:50 yes 14:27:49 Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has joined #lisp 14:28:52 m4dnificent: sure, there're several conflicts that has to be ruled out. 14:29:02 # ; that always gives me a warm, fuzzy feeling inside. 14:29:03 m4dnificent: that's why lawyers get paid. 14:29:47 -!- amos_ [n=amos_@host86-145-162-199.range86-145.btcentralplus.com] has quit [Read error: 110 (Connection timed out)] 14:31:18 <_8david> btw, anyone at 25c3 this year? 14:31:20 m4dnificent: prior to some damn IP changes here, GPL was effectively equal to BSD here. 14:31:34 m4dnificent: now GPL is even worse than commercial. 14:32:11 m4dnificent: some companies prefer to buy Linux-based distribution at least 1 copy 14:32:34 m4dnificent: just to get rid of some pecularities of tax law. 14:32:43 -!- ejs [n=eugen@92-49-229-119.dynamic.peoplenet.ua] has quit [Read error: 110 (Connection timed out)] 14:33:15 ejs [n=eugen@92-49-225-247.dynamic.peoplenet.ua] has joined #lisp 14:35:31 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 54 (Connection reset by peer)] 14:36:12 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 14:36:37 caliostro: you can also do something like (add-to-list 'auto-mode-alist '("\\.foo'" . slime-mode)) in your .emacs, where ".foo" is the file extension you want to interpret as lisp code. Franz uses .cl btw. 14:37:02 -!- mrSpec is now known as spec[afk] 14:37:03 not recommended 14:38:37 cods [n=cods@tuxee.net] has joined #lisp 14:38:38 i only need to associated a file with the slime buffer 14:39:20 and compile that file with C-c C-c key binding in emacs, but i am quite relative new to emacs, slime and lisp too..Thank u for suggestions 14:39:33 m4dnificent: as for "no warranty" conflict, the consequences are that you have to pay damages, if your product doesn't work and you sell it. 14:39:45 m4dnificent: but you're to consult your local lawyer. 14:40:50 _8david: esden will be there 14:40:51 caliostro: do you have anterograde amnesia? You just asked that, got a response, and said it was working 14:41:19 caliostro: open the buffer and do M-x slime-mode 14:41:55 S11001001: it worked once 14:42:14 ecraven [n=nex@140.78.42.103] has joined #lisp 14:42:19 S11001001: maybe he wants something temporarily? I do that when i'm writing long text that needs some lisp in between. switch between modes to suit each type of writing. 14:42:46 ok i am in slime-mode now 14:44:55 Hun [n=Hun@dial18-65.RZ.FH-Augsburg.DE] has joined #lisp 14:45:33 caliostro: the emacs cheat-sheet might come handy in the near future :-) 14:46:02 ok i may say..it's definitively working right now 14:47:19 <_8david> tcr: thanks 14:47:56 <_8david> tcr: I'm a little overwhelmed by your documentation macro. Do you think our macros could be merged into one? 14:48:47 caliostro: when i was learning emacs, i found it helpful to use it for everything :-) editing, mail, cvs, shell, web, calendar, calculator, notes and memos. etc. it comes handy when you're stuck in a unix console and need to be productive. 14:48:57 -!- Quadrescence [n=quad@c-71-193-68-220.hsd1.mn.comcast.net] has quit [Remote closed the connection] 14:49:24 fusss: thats excellent advice 14:49:25 Quadrescence [n=quad@c-71-193-68-220.hsd1.mn.comcast.net] has joined #lisp 14:49:28 thats how i'm learning emacs now 14:49:31 Or stuck in a X terminal and need to be productive. 14:49:42 Or stuck in a Aqua console and need to be productive. 14:49:53 Not mentionning what MS-Windows uses... 14:50:10 o'reilly's "unix power tools" is a massive tome of text, written for the sole purpose of creating an slow, bug ridden, and incomplete environment half of emacs :-) 14:50:11 _8david: That was just an experiment. I think I since moved on to like Scribe syntax more for prosa documentation (i.e. @foo{...} that's also used in TexInfo.) 14:50:58 My main gripe with TexInfo is that its macros are so crippled. 14:51:09 *tcr* (bbl) 14:51:13 hey tcr 14:51:23 are you also at the congress? 14:51:45 Hun: No, I'll drive back to munich tomorrow. 14:51:49 *tcr* afk. 14:53:39 -!- holycow [n=rtaylor@S01060016b6b53675.vf.shawcable.net] has quit [Remote closed the connection] 14:54:22 -!- ASau [n=user@193.138.70.52] has quit [Remote closed the connection] 14:55:24 ASau [n=user@193.138.70.52] has joined #lisp 14:57:22 topo [n=topo@240.pool85-58-128.dynamic.orange.es] has joined #lisp 14:58:17 amos_ [n=amos_@host86-145-162-199.range86-145.btcentralplus.com] has joined #lisp 14:59:16 Yeah, Texinfo's macros suck. 14:59:39 -!- HET2 [n=diman@chello084114161225.3.15.vie.surfer.at] has quit ["This computer has gone to sleep"] 15:00:13 -!- CHJA [n=christia@90-231-255-161-no145.tbcn.telia.com] has quit [] 15:01:37 -!- carbocalm [i=carbocal@75-119-233-233.dsl.teksavvy.com] has quit [] 15:03:29 milanj [n=milan@77.46.225.124] has joined #lisp 15:07:41 -!- caliostro [n=caliostr@host-84-222-134-28.cust-adsl.tiscali.it] has left #lisp 15:09:44 -!- Hun [n=Hun@dial18-65.RZ.FH-Augsburg.DE] has quit [Read error: 60 (Operation timed out)] 15:11:48 -!- ecraven [n=nex@140.78.42.103] has quit ["bbl"] 15:18:35 xan_ [n=xan@cm136195.red83-165.mundo-r.com] has joined #lisp 15:19:42 -!- spec[afk] is now known as mrSpec 15:21:14 -!- binarycodes [n=sujoy@59.93.255.215] has quit [Read error: 145 (Connection timed out)] 15:22:19 Hun [n=Hun@Hunlappy2.visitor.congress.ccc.de] has joined #lisp 15:23:12 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 15:23:24 fusss_ [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 15:23:24 -!- fusss_ is now known as fusss 15:23:25 -!- Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has quit [] 15:25:21 later 15:25:21 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Client Quit] 15:25:35 -!- jlouis [n=jlouis@port1394.ds1-vby.adsl.cybercity.dk] has quit [Remote closed the connection] 15:25:48 is deftype opaque? can user-defined types be portably expanded? 15:26:05 manuel_ [n=manuel@91-67-169-12-dynip.superkabel.de] has joined #lisp 15:27:10 -!- manuel_ [n=manuel@91-67-169-12-dynip.superkabel.de] has quit [Client Quit] 15:28:25 weirdo: yes. 15:31:13 yes to former, no to latter? 15:31:32 yes to both. 15:32:26 -!- hypno [n=hypno@impulse2.gothiaso.com] has quit [Remote closed the connection] 15:32:31 hypno [n=hypno@impulse2.gothiaso.com] has joined #lisp 15:32:50 pjb, then how does one go around expanding user-defined types without shadowing deftype? 15:33:22 (deftype a () '(integer 0 10)) (deftype b () '(or a (integer 20 30))) 15:33:39 -!- sword [n=user@c-98-222-51-71.hsd1.il.comcast.net] has quit [Nick collision from services.] 15:34:41 I don't think that's what he meant by "expansion". 15:35:20 Ah, if you mean a -> (integer 0 10), then indeed, it's totally opaque and you have to shadow deftype. 15:38:13 blbrown [n=Berlin@c-71-236-25-127.hsd1.ga.comcast.net] has joined #lisp 15:40:59 not like it warrants a CDR or anything 15:43:14 -!- ASau [n=user@193.138.70.52] has quit [Remote closed the connection] 15:44:32 LiamH [n=nobody@pool-72-75-101-81.washdc.east.verizon.net] has joined #lisp 15:45:16 ASau [n=user@193.138.70.52] has joined #lisp 15:46:55 -!- ASau [n=user@193.138.70.52] has quit [Remote closed the connection] 15:47:05 ASau [n=user@193.138.70.52] has joined #lisp 15:47:53 kpreid [n=kpreid@pool-72-68-32-150.nwrknj.east.verizon.net] has joined #lisp 15:49:59 Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has joined #lisp 15:51:15 vy [n=user@213.139.194.186] has joined #lisp 15:51:45 Hi all. 15:52:00 What is the "lightest" lisp for ARM use, if I want a REPL? 15:52:04 -!- mulligan [n=user@e178031222.adsl.alicedsl.de] has quit [Read error: 104 (Connection reset by peer)] 15:52:17 Possibly clisp? 15:52:19 Hi! There was a discussion on tbnl-devel about how to limit upload file size in hunchentoot? Does anybody have the pointer to the related threads in the archives? 15:52:46 SBCL will likely never be the "lightest" lisp on ARM, even after the port is usable. 15:52:48 nyef: it is marked as broken on ARM in pkgsrc. 15:53:06 nyef: someone did it, thus probably there's a reason. 15:53:15 Fair enough. I have -no- idea. 15:53:33 I have no means to check. 15:53:37 What about ECL? 15:53:55 I have no means to check. 15:53:57 Well... 15:54:14 It involves setting emulated environment... 15:55:04 Would it be possible to get a lisp running on an ARM without linux? 15:55:37 You mean, a "bare metal" ARM? 15:55:59 Yes. 15:56:19 Hm. 15:56:27 It would be -possible-. 15:56:31 I don't know about easy, though. 15:57:13 Because I sort of want to write a small OS for the ARM 15:57:15 I guess "possible" implies rolling your own environment. 15:57:29 There are some existence-proofs of lisp running on bare-metal x86 PCs, after all. 15:57:38 I know. 15:57:38 But those are seriously customized environments. 15:58:05 (Movitz is a complete from-scratch environment, and SBCL-os is a major hackjob.) 15:58:08 I don't know the cost of customizing existent Lisp implementation to standalone environment. 15:58:29 And for either of those you'd be looking at a new compiler backend anyway... 15:58:57 I did that for Forth five years ago. 15:59:01 No idea about Lisp. 15:59:13 Now, if you don't mind losing CL semantics and just want -a- lisp, it's not hard to write a simple evaluator or cross-compiler. 15:59:23 Forth is easy to bootstrap anyway. 15:59:32 My problem is that I'd really like to have CLOS 16:01:04 I'm willing to do an extensive amount of work to make that happen (get some existing CL implementation onto a bare-metal arm) but basically unsure where to start. 16:02:06 Well, step one is to figure out which implementation to use. 16:02:36 I've only used SBCL and ECL. 16:02:57 You'll need to worry about device drivers, interrupt handling, memory management, filesystems, etc. 16:03:00 -!- younder [i=jpthing@062016236162.customer.alfanett.no] has left #lisp 16:03:06 -!- drdo [n=psykon@167.111.54.77.rev.vodafone.pt] has quit [Remote closed the connection] 16:03:21 That is pretty much exactly what I would lke to be worrying about. 16:03:32 My POV is you should use ready kernel. 16:03:45 NetBSD, Linux, whatever. 16:04:07 I'm partial to Linux, even with the complete botchjob which is the BKPT instruction. 16:04:23 Then again, I've never tried running NetBSD on my ARM. 16:04:26 syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has joined #lisp 16:05:21 I want to write my own micro-kernel 16:05:40 -!- amos_ [n=amos_@host86-145-162-199.range86-145.btcentralplus.com] has quit ["Leaving"] 16:05:49 (I have no illusions about this being something that anyone would want to ever use, I just need to teach myself os programming) 16:06:13 So... why on an ARM and not on a PC? 16:06:20 And I want to involve CL at the lowest possible language, because I have the easiest time programming in CL. 16:06:27 Because I know ARM asm and not x86 asm? 16:06:37 -!- vy [n=user@213.139.194.186] has quit [Remote closed the connection] 16:06:52 Also I've worked with embedded devices in C and asm before and written a couple device drivers. 16:06:56 Umm... You realize that very little ASM is actually involved, right? 16:07:01 I guess I'm just more comfortable with the platform? 16:07:37 That's fair. 16:07:43 -!- topo [n=topo@240.pool85-58-128.dynamic.orange.es] has left #lisp 16:08:13 No, I really don't. Everything that I've read has a lot of theory but very little implementation stuff, beyond isolated components. 16:08:19 THe OS class I just took was awful. 16:08:20 You could take frodef's approach, and build your entire lisp from scratch. 16:08:35 Hang on, let me see what I have handy as an example... 16:09:37 So he had to implement all of CL himself? on top of x86 asm? 16:10:12 ISTR that he used Closette, from AMOP, as the basis for his CLOS implementation, but beyond that he even wrote his own assembler. 16:10:36 Wow. 16:11:09 Okay, you might find http://www.lisphacker.com/temp/lispos/os-0.9.14-m3.patch and http://www.lisphacker.com/projects/sbcl-os/how-it-works-2008-04-19.txt of interest. 16:11:28 Hun` [n=Hun@Hunlappy2.visitor.congress.ccc.de] has joined #lisp 16:11:31 i think that is a rather smart approach if you want to do it in CL. a lisp-assembler and then perhaps a proto-lisp (similiar to the capabilities of C) and then work your way through the necessities. 16:11:53 Note that SBCL-os is -very- rough and incomplete. 16:11:53 i sure wish i had gone that route instead of hacking C and assembler for my kernel. :/ 16:12:08 Heh. I tried the C-and-assembler route briefly. 16:12:17 For the duration of a "hello-world" kernel. 16:12:32 And then I wrote an ITC Forth in ASM. 16:12:36 And used that instead. 16:12:51 so, asm->forth->lisp? 16:12:51 i went a bit further: pmode, simple interrupt driven scheduler and mm/paging system. :) 16:13:44 hypno: I switched to pmode directly from my bootloader, and wrote the code to set up my IDT from within the running Forth environment. 16:13:55 (Yes, I had working keyboard and disk drivers by that point.) 16:14:13 HET2 [n=diman@chello084114161225.3.15.vie.surfer.at] has joined #lisp 16:14:14 ah, i see. 16:14:46 A basic pmode switch is, what, three instructions and a GDT? 16:15:38 Basically LGDT, LMSW, and JMP FAR. So that's four instructions, as you can't LMSW from anything but a register. 16:15:46 yeah, i guess. a20-fix, gdt, set up the segs and let the magic roll. 16:16:03 Don't even need to fix a20 if you load low. 16:16:09 davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has joined #lisp 16:16:35 mogunus: Err... Yeah, I used my forth system to bootstrap the lisp environment. But it doesn't run after starting lisp. 16:17:03 cracki [n=cracki@sglty.kawo2.RWTH-Aachen.DE] has joined #lisp 16:17:06 i set out to write a simple "network operating system" beeing frustrated with Unix. :) incidentally, that's sort of my way to lisp: discussing emacs lisp problems while hacking kernels Common Lisp was revealed to me. :)) 16:17:35 Ugh. I need to figure out exactly what sort of environment SBCL needs to run arbitrary lisp. 16:17:51 mogunus: Yeah, if you go the SBCL route. 16:18:01 And SBCL doesn't have an ARM backend yet, either. 16:18:33 *weirdo* is sleep-deprived and has an idea about clos abstraction 16:19:35 And the other thing about SBCL-os is the need to write a garbage collector, as it completely ignores the whole C runtime thing. 16:19:37 -!- kpreid [n=kpreid@pool-72-68-32-150.nwrknj.east.verizon.net] has quit [] 16:19:37 it might be useful to add slots to the base class for user subclasses 16:20:04 weirdo: Why can't the subclasses just add their own slots? 16:20:25 my solution: declare a list of forms evaluating to successive base-classes, inheriting from the previous one, a'la REDUCE 16:20:33 finally, call #'(setf find-class) on it 16:20:46 Hm. I could generate one. otoh, I'm trying to figure out if CLISP has an arm baskend. 16:20:56 CLISP is bytecoded anyway. 16:21:27 minion: arm port log? 16:21:27 arm port log: http://www.lisphacker.com/projects/sbcl-arm/port-log.txt 16:21:46 I'm hoping to get back to that at some point next month. 16:21:49 nyef, i'd add a "memoization" mixin to speed up context-free grammar and above (as per chomsky), so i could do stuff like ((list :memoize t) a b 42) 16:22:07 nevermind the lack of context-freeness in the example 16:22:07 mattrepl [n=mattrepl@ip68-98-133-128.dc.dc.cox.net] has joined #lisp 16:22:28 weirdo: And why can't the mixin just add its own slots? 16:23:24 nyef, because different matcher-keywords are written by different users 16:23:47 and all that can be said about them is they satisfy some protocols and at best use a common base-class... 16:24:57 Sounds like that might be a job for some MOPpery or MACROtica. 16:25:17 though it could be done by reducing slots, options, inheritance on APPEND 16:26:02 instead of reducing on inheriting anonymous classes only to assign itto a global identifier 16:26:11 nyef: Clisp added native JIT compiling recently, I believe. 16:26:25 tcr: Via some other GNU project, wasn't it? 16:26:32 gnu lightning 16:26:35 srp [n=srp@122.172.2.116] has joined #lisp 16:26:46 -!- yhara [n=yhara@7.193.12.221.megaegg.ne.jp] has quit [Read error: 113 (No route to host)] 16:26:49 so many ideas and don't feel like coding... /me tries to pull oneself together 16:27:47 -!- Hun [n=Hun@Hunlappy2.visitor.congress.ccc.de] has quit [Connection timed out] 16:28:10 weirdo: Is there something you'd rather be doing than coding? 16:29:32 nyef: this port-log is really informative 16:29:43 mogunus: Glad you like it. 16:30:08 It tells me that I really need to learn more about SBCL internals. Is there a good jumping-off point for that? 16:30:31 The sbcl-internals cliki, some of the internals documentation in the source tree. 16:31:15 http://www.lisphacker.com/codex/001-sbcl-vops.txt and http://www.lisphacker.com/codex/005-sbcl-nlx.rev2.txt 16:31:49 Probably jsnell's weblog, places like that. 16:32:29 -!- Hun` [n=Hun@Hunlappy2.visitor.congress.ccc.de] has quit [Read error: 110 (Connection timed out)] 16:33:21 amos_ [n=amos_@host86-145-162-199.range86-145.btcentralplus.com] has joined #lisp 16:33:43 nyef, having social life. but it's TMI, so i'll stop 16:34:10 nyef: thanks 16:34:31 mogunus: And as this used to be #sbcl-implementors, there's lots of help to be had here. 16:35:26 mogunus: Trying to fix bugs is a nice way to get acquainted with the source base. 16:35:50 Also, there is a lot of SBCL that just isn't documented, or is only documented in source comments. 16:36:07 Oh, another thing to look for is the "encycmuclopedia". 16:36:08 tcr: the last round of sbcl bugs that I remember being discussed here were in (format), which is terrifying to me. 16:36:30 mogunus: https://bugs.launchpad.net/sbcl 16:36:31 A lot of low-level stuff from there still applies to some extent. 16:36:31 but that's false, it depends on a subjective false idealized perception of different humans. in real-world conditions, social activity doesn't seem more satisfying than hanging out on irc 16:36:56 weirdo: FWIW, I'm hanging out on IRC and getting some hacking done. 16:37:26 tcr: oh, thanks. is that the "canonical" bug repo? 16:37:33 i sleep every 3 days and smoke 60 cigarettes a day or so 16:37:42 that's why i need to pull myself together 16:37:57 weirdo: That must cost a fortune in tobacco. 16:38:02 it does. 16:38:03 mogunus: Not yet. It has chances to become the official bug tracking mechanism for sbcl. 16:38:21 i once quit, for a week or so 16:38:35 drank some vodka and smoked again 16:38:44 weirdo: Do some sports, and hang out with friends. 16:38:53 tcr: cool. Hmm. I guess I should sign up for sbcl-devel. 16:38:58 Doing sports is really important. Mens sana in corpore sano. 16:39:21 tried the same method, nicotine gum, but this time video game's rubberband ai made me furious to the point of screaming 16:39:28 -!- syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has quit [Read error: 60 (Operation timed out)] 16:39:59 at least i was smart enough not to harm the hardware 16:40:05 though that happened once 16:41:01 Do people mind reading over documentation generated for SBCL? I learn fairly well by documenting things, but of course there's the probability that I generate something very wrong-headed. 16:41:43 sbcl's texinfo stuff? the format's nice, how unlike tinaa, edi's stuff et al 16:41:50 mogunus: Documentation for SBCL tends to get reviewed, yes. Particularly about some aspect of the internals. 16:41:56 -!- srp_ [n=srp@122.172.2.116] has quit [Read error: 110 (Connection timed out)] 16:42:06 great, that's what I'd be interested in writing for 16:43:13 [tmi] wikipedia says sleep deprivation is a coping mechanism for clinical depression 16:43:36 Really? I can actually believe that. 16:44:23 allegedly until resting, then some enzyme or other chemical coursing through the brain gets re-set 16:44:48 syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has joined #lisp 16:45:55 -!- xan_ [n=xan@cm136195.red83-165.mundo-r.com] has quit ["Leaving"] 16:46:07 good evening. 16:46:12 Hello tic. 16:46:15 hi 16:46:31 Any interesting Lisp-related news lately? 16:47:23 Umm... We have one potential new SBCL hacker in mogunus. Does that count? 16:47:35 There doesn't seem to be much on internals in the texinfo manual. 16:47:43 Jabberwockey [n=jens@dslb-082-083-080-185.pools.arcor-ip.net] has joined #lisp 16:47:46 Would contributions to that be welcome? 16:47:47 nyef, yay! 16:48:17 mogunus: sure, or the wiki if that's more your style. 16:48:27 mogunus: Have a look at doc/internals and doc/internals-notes in the sbcl source distro if you haven't seen them already. 16:50:09 (Basically, there are separate user and internals manuals, both texinfo.) 16:50:25 ahh okay 16:52:04 -!- petro [i=Peter@p4FE97473.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 16:52:12 pet-ro [i=Peter@p4FE955D0.dip.t-dialin.net] has joined #lisp 16:53:35 -!- Jabberwockey [n=jens@dslb-082-083-080-185.pools.arcor-ip.net] has quit [Remote closed the connection] 16:54:26 Jabberwockey [n=jens@dslb-082-083-080-185.pools.arcor-ip.net] has joined #lisp 16:54:38 hmm. might there lure a secret lisp success story in the Tellme Inc. deal? 16:56:06 Hun [n=Hun@vpn18-231.RZ.FH-Augsburg.DE] has joined #lisp 16:56:45 -!- mattrepl [n=mattrepl@ip68-98-133-128.dc.dc.cox.net] has quit [] 16:57:15 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 16:57:28 What do most sbcl people use for vcs? 16:57:52 git these days 16:57:57 i like mercurial. it's faster to type than git 16:58:15 git, I think. There's a cvs->git gateway, a repo thingy, antifuchs wrote a wonderful document on cvs<->git, and it does branches neatly. 16:58:20 I use CVS for SBCL itself and darcs for a couple of my projects, but I'm planning to reconsider my options at some point in the mid-term future. 16:58:21 (not a sbcl hacker, but git.) 16:59:13 Ah, nice. git is my second-favorite vcs. 16:59:51 (use darcs for most everything) 16:59:56 RCS here 17:00:00 -!- jsoft [n=Administ@unaffiliated/jsoft] has quit [Read error: 145 (Connection timed out)] 17:03:13 araujo [n=araujo@gentoo/developer/araujo] has joined #lisp 17:06:19 josemanuel [n=josemanu@150.1.222.87.dynamic.jazztel.es] has joined #lisp 17:07:23 <_8david> I wonder whether any SBCL committer is left who's actually using CVS rather than git-cvsexportcommit. 17:07:56 If I'm still a committer, then I probably am. 17:08:00 _8david: nyef (: 17:09:22 I still have those debugger cleanups to consider committing, don't I? 17:09:54 <_8david> nyef is probably also running a hand-written CVS on a Posix implementation for an old Lisp machine dialect powered by an emulator written in Forth 17:10:12 -!- syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has quit [Read error: 60 (Operation timed out)] 17:10:16 I'm not, actually, but that'd be pretty cool. 17:12:12 So, should I commit http://www.lisphacker.com/temp/random-sbcl-patches/reunite-sub-access-and-set-debug-var-slot.diff soon, or leave it a while longer? 17:12:52 To my amazement, while it reduces the file size a bit, it actually makes the conditionalization -worse-. 17:14:12 _8david: Do you have any plans to make use of slot documentation in future versions of Atdoc? 17:15:43 -!- davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)] 17:15:54 davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has joined #lisp 17:15:55 tcr: how do you feel about an OO reader? (Mostly) backward compatible syntax but gives us structures to frob instead of lists and symbols? 17:16:06 syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has joined #lisp 17:17:28 <_8david> Aankhen``: have you already tried the :include-slot-definitions-p keyword argument? 17:17:59 binarin [n=user@epoint.ru] has joined #lisp 17:18:12 -!- binarin [n=user@epoint.ru] has quit [Client Quit] 17:18:22 <_8david> ... I'm not quite sure how that will work in the new world of parse-docstrings though, because the annotation macro can only annotate things named by symbols at this point, not other objects like slot definitions. 17:18:58 _8david: Er, sorry, I misread the source. I kept failing to see the (documentation slot-def t). 17:19:19 pkhuong: I thought about extending what a list it: a) to introduce first class immutable, proper lists, and b) to have syntax-objects that result from reading. 17:20:11 pkhuong: My implementator's hat isn't big enough yet to judge how much that would have a negative impact on performance. 17:22:09 -!- phytovor [n=doxtor@unaffiliated/mitja] has quit [Remote closed the connection] 17:25:28 holycow [n=rtaylor@S01060016b6b53675.vf.shawcable.net] has joined #lisp 17:28:08 tcr: I don't see how moving to a more structured representation for CL':READ data could slow things down (except if the reader itself is that much slower or consier) 17:29:20 -!- envi^home [n=envi@220.121.234.156] has quit [Read error: 104 (Connection reset by peer)] 17:29:24 extra dispatching on all list operations 17:29:43 tcr: that would be for immutable lists? 17:29:59 yeah 17:30:19 njsg [n=njsg@unaffiliated/njsg] has joined #lisp 17:30:26 Yeha, that would be hairy. 17:31:01 sohail [n=Sohail@unaffiliated/sohail] has joined #lisp 17:31:12 Hun` [n=Hun@Hunlappy2.visitor.congress.ccc.de] has joined #lisp 17:31:16 <_8david> couldn't you just put immutable conses on their own memory pages with write protection? 17:31:17 Maybe some BIBOPish scheme for mutable/immutable data (which could also help the GC) 17:31:46 Wow, SBCL SIGSEGVs a lot when firing up the compiler. 17:32:11 _8david: That sounds nice because it'd also work for other things, like string, wouldn't it? 17:33:00 <_8david> yes. would be gencgc only though. 17:33:38 that's the copying one? 17:33:41 _8david: cheney could be extended, by using 2 sets of alloc/fix-up pointers. 17:34:10 tcr: both are copying, but the non-generational only handles oldspace and newspace. 17:34:30 <_8david> allocating them in the first place could be tricky, since you wouldn't want to trigger write protection all the time while filling up the first page, and wouldn't want write protection only after the first copy 17:34:31 -!- davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)] 17:35:04 _8david: So double-map the pages for immutable data, writing to one copy and returning an address to the other? 17:35:15 Wouldn't work on win32, though. 17:35:20 nyef: is that portable across posix? 17:35:33 <_8david> would it even work on Linux? 17:35:45 -!- Hun [n=Hun@vpn18-231.RZ.FH-Augsburg.DE] has quit [Read error: 145 (Connection timed out)] 17:35:46 I know it'd work on linux, but it might have to be an SHM segment there. 17:36:08 (One of the concurrent GC algorithms I looked at back when did it.) 17:36:59 davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has joined #lisp 17:39:11 extend the type lattice with mutable/immutable for the appropriate cases, and there shouldn't be too much of a slowdown with appropriately updated code. 17:40:40 SBCL already plays sillybuggers with lowtags for lists, though. Wouldn't this interfere with that? 17:40:47 Basically I want literals expand to immutable data, so their modification is properly caught 17:40:57 -!- syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has quit [Read error: 60 (Operation timed out)] 17:41:01 when you're defining a view class with clsql, what is the :initarg for? 17:42:16 -!- oudeis [n=oudeis@bzq-79-177-106-178.red.bezeqint.net] has quit ["This computer has gone to sleep"] 17:42:58 <_8david> what nyef says. conses doen't have a header, so we're talking (a) twice as much memory consumption for conses (one new header word, plus one for alignment). (b) perhaps trouble making the layout of nil compatible with both cons and symbol. 17:44:07 nyef: not if you stick some dat ain the page. 17:44:21 JuanDaugherty [n=juan@cpe-72-228-150-44.buffalo.res.rr.com] has joined #lisp 17:45:16 *some data in the pages (BIBOP-style). 17:46:08 <_8david> or would you rename cons to mutable-cons and define cons as (or mutable-cons immutable-cons) with the latter having an entirely different layout? that would require type checking at places where you can currently optimize it away completely, wouldn't it? 17:47:07 <_8david> oh wait, bibop. well, if gencgc can handle that without a rewrite...? 17:47:09 syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has joined #lisp 17:47:35 _8david: for backward compatibility, I think that's almost the only way. But that's pretty much orthogonal to how the mutableness is represented. 17:48:21 Doesn't gencgc need a rewrite anyway, or have the memory exhaustion problems since been fixed? 17:49:16 _8david: nikodemus's recent commit of code-only pages for gencgc might be extendable to accomodate RO pages. 17:50:48 -!- ASau [n=user@193.138.70.52] has quit [Read error: 104 (Connection reset by peer)] 17:51:03 nyef: I guess (and no). Unfortunately, writing a GC wasn't considered complex enough for a master's course. 17:52:17 ASau [n=user@193.138.70.52] has joined #lisp 17:52:38 oudeis [n=oudeis@89-139-68-237.bb.netvision.net.il] has joined #lisp 17:53:22 If that's not complex enough, I'm not sure I want to know what the hell -is-. 17:55:26 Ah, well, I'll be trying to hack SBCL's GC next year. Feel free to shoot ideas and suggestions. 17:55:28 nyef: I might have been able to bargain something like GC + peephole + regalloc, or maybe two of these. I went with a native code compiler in the end. 17:55:50 nyef: maybe being something completely useless is a requirement 17:55:53 luis: please generate most of the code from the primitive object definitions. 17:56:20 hefner: hacking on something that can be integrated in gambit (so useful) was a big implicit plus ;) 17:56:30 pkhuong: Is now the time to mention that I might be building a proof-of-concept for some of that for an unrelated purpose? 17:57:01 nyef: Link time! 17:57:18 pkhuong: for scavenging? 17:57:33 http://www.lisphacker.com/temp/article-drafts/object-memory-dumper.txt 17:58:10 pkhuong: To make it work properly, it has to know which parts of an object are boxed and which unboxed. 17:58:35 overlordpuppy [n=adrien@pool-71-124-46-5.chi01.dsl-w.verizon.net] has joined #lisp 17:59:10 nyef: (: I want to use mine to send data (and closures or even functions) across a cluster. 17:59:19 How can I get (* 100 (/ 5 (+ 5 6 8))) to return in decimal form? Or at least whole numbers and fractions. . . 17:59:26 how can I remove the most recently commit changeset from git? 17:59:30 comitted 17:59:40 Oh, my. 17:59:50 pkhuong: Thank you! 17:59:56 *nyef* bookmarks. 18:00:53 overlordpuppy: truncate, or convert to a floating point number. 18:00:54 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit [Read error: 104 (Connection reset by peer)] 18:01:04 Now, if I tie that into the build process for a cold-core to produce a suitable type system definition... Yeah, there's definate possibility here. 18:01:06 tcr git reset? 18:01:59 nyef: I plan on getting it mostly working over the break. 18:02:10 tcr: if you haven't pushed it to some other repository, you can erase the commit with git reset --hard HEAD^ 18:03:28 fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has joined #lisp 18:03:48 alek_b [n=alek_b@93.86.176.112] has joined #lisp 18:04:06 tcr: re cleaning a development branch up, I think I repeatedly git merge --squash. 18:05:12 nyef: just a little remark: iirc it is not permissible to ignore parameters that the method is specialized on 18:05:47 adeht: SBCL didn't complain, therefor it is permissible for my target platform. 18:07:03 -!- blbrown [n=Berlin@c-71-236-25-127.hsd1.ga.comcast.net] has quit ["Ex-Chat"] 18:07:20 adeht: And does that mean you have to -use- eql-specialized parameters? 18:07:42 nyef: by "ignore" I actually meant declare ignore 18:09:31 Well, if I don't declare them ignore, I might have to declare them ignorable. 18:09:47 nyef: why is that? 18:09:54 Because I don't use them? 18:10:05 nyef: you do - in specializing the method on them 18:10:16 So? 18:10:27 nyef: so a warning should never be made 18:10:29 adeht: I'm not sure that's specified. You could have fresh bindings for the method. 18:10:31 That's outside the scope of their binding. 18:11:12 that's how I remember it.. 18:11:14 Heck, you probably have to for call-next-method. 18:11:33 Is it possible to create a patch bundle in git like it's possible for darcs? 18:11:52 what's a patch bundle? 18:12:46 diffs+changelog 18:13:32 hmm (coerce (* 100 (/ 5 (+ 5 6 8))) 'float) returns 26.31579. Almost right. But what confuses me is that (round (coerce (* 100 (/ 5 (+ 5 6 8))) 'float)) returns 25, then 0.31579018. Is there some way I can get this down to 2 decimal places? 18:13:57 If I could just round that 5 beside the 1 it would be a correct answer. 18:13:58 I think you want format-patch, which gives you one file per commit. 18:14:22 overlordpuppy: You're confused. 18:14:27 overlordpuppy: use FORMAT. 18:14:49 overlordpuppy: (round foo 1/100) will round to the hundredth. 18:16:19 jlouis [n=jlouis@port1394.ds1-vby.adsl.cybercity.dk] has joined #lisp 18:16:41 pkhuong: Umm... Why do you go to all the trouble of extracting the lowtag and using it as a displacement in real-widetag-of when you could instead just mask off the lowtag bits? And have you considered what happens if you pass NIL to it? 18:17:19 wormphlegm [n=george@c-98-234-189-214.hsd1.ca.comcast.net] has joined #lisp 18:18:19 -!- Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has quit [Remote closed the connection] 18:18:34 overlordpuppy: (format nil "~,2F" 26.31579) => "26.32" 18:18:43 nyef: some debugging noise from when I was trying to understand why widetag-of wasn't always right, I think. 18:19:06 Heh. widetag-of is one of those VOPs that I want to kill anyway. 18:19:19 Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has joined #lisp 18:19:40 hm, how do I make a string out of a list of chars? 18:19:55 nyef, pkhuong: the paragraph beginning with "The expansion of the defmethod macro ``refers to'' each specialized parameter (see the description of ignore within the description of declare)." in the DEFMETHOD entry may be of interest 18:20:38 nyef: NIL is weird. Teh End. 18:20:48 danderson: (coerce list 'string) 18:21:19 luis: cheers. 18:21:32 binarycodes [n=sujoy@59.93.200.172] has joined #lisp 18:22:14 adeht: not applicable in nyef's case, since none of the parameters were specialised. 18:22:29 pkhuong: is that so? 18:22:53 pkhuong: in the document link he pasted, there's a method for READ-WORD 18:23:22 adeht: and indeed there is. 18:24:20 pkhuong: and in this method the first parameter is specialized and then declared as ignored 18:24:35 +upon 18:25:06 -!- NoorDextor [n=NoorDext@unaffiliated/noordextor] has quit [Connection timed out] 18:25:23 luis: Thanks. 18:25:38 ThomasI [n=thomas@unaffiliated/thomasi] has joined #lisp 18:25:56 -!- ejs [n=eugen@92-49-225-247.dynamic.peoplenet.ua] has quit [Read error: 110 (Connection timed out)] 18:26:55 adeht: but in either case, the usedness is purely because of the specification, not because a (sane) implementation would have to do that as a side effect. 18:27:33 pkhuong: that may be so, but I care mostly about the former when I write programs. 18:28:41 adeht: sure. 18:28:48 Are there any nice printed resources for CLIM? 18:30:52 -!- njsg [n=njsg@unaffiliated/njsg] has quit [Connection reset by peer] 18:31:18 njsg [n=njsg@unaffiliated/njsg] has joined #lisp 18:31:52 is :documentation a good way of commenting slots in clsql view classes? or is there something that actually puts comments in the database table definition? 18:32:17 equivalent to comment in mysql 18:32:18 madnificent: ping 18:34:48 -!- overlordpuppy [n=adrien@pool-71-124-46-5.chi01.dsl-w.verizon.net] has left #lisp 18:36:07 ejs [n=eugen@92-49-225-90.dynamic.peoplenet.ua] has joined #lisp 18:39:58 -!- josemanuel [n=josemanu@150.1.222.87.dynamic.jazztel.es] has quit ["Saliendo"] 18:40:33 akkartik [n=4c7e5689@208-75-86-102.slicehost.net] has joined #lisp 18:40:49 -!- akkartik [n=4c7e5689@208-75-86-102.slicehost.net] has left #lisp 18:45:36 syamajal_ [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has joined #lisp 18:49:25 -!- ThomasI [n=thomas@unaffiliated/thomasi] has quit ["Bye Bye!"] 18:49:46 -!- chris2 [n=chris@p5B16C75E.dip0.t-ipconnect.de] has quit ["Leaving"] 18:50:53 ironChicken: :documentation for slots classes and generic functionss as well 18:51:22 doxtor [n=doxtor@unaffiliated/mitja] has joined #lisp 18:51:25 ironChicken: and unit tests too 18:51:51 -!- araujo [n=araujo@gentoo/developer/araujo] has quit [Read error: 113 (No route to host)] 18:52:18 -!- syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has quit [Read error: 110 (Connection timed out)] 18:53:38 I couldn't find how the to tell sbcl, not to print circular lists forever in repl. I just want to see the notation in topic 18:53:52 decafbad: (setf *print-circle* t) 18:53:59 thanks 18:54:30 is there a document about these? 18:54:39 clhs *print-circle* 18:54:39 http://www.lispworks.com/reference/HyperSpec/Body/v_pr_cir.htm 18:55:17 decafbad: it's all standard. 18:56:23 it seems *print-xxx variables controls the repl 18:56:24 kiuma: so will the :documentation, er, argument (?) for a slot end up as a comment in the database table definition? 18:56:25 -!- binarycodes [n=sujoy@59.93.200.172] has quit ["WeeChat 0.2.6"] 18:57:30 decafbad: not only the repl, but printing in general 19:04:13 -!- ejs [n=eugen@92-49-225-90.dynamic.peoplenet.ua] has quit [Read error: 110 (Connection timed out)] 19:04:50 -!- Hun` [n=Hun@Hunlappy2.visitor.congress.ccc.de] has quit [Success] 19:06:12 -!- kzar [n=kzar@hardwick.demon.co.uk] has quit [Remote closed the connection] 19:10:27 -!- syamajal_ [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has quit [Read error: 60 (Operation timed out)] 19:12:52 -!- oudeis [n=oudeis@89-139-68-237.bb.netvision.net.il] has quit ["This computer has gone to sleep"] 19:13:37 -!- amos_ [n=amos_@host86-145-162-199.range86-145.btcentralplus.com] has quit ["Leaving"] 19:16:55 syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has joined #lisp 19:18:50 -!- ruediger [n=the-rued@p508B6F49.dip.t-dialin.net] has quit ["This computer has gone to sleep"] 19:20:09 __jasonherald__ [n=__jasonh@pool-72-66-198-13.ronkva.east.verizon.net] has joined #lisp 19:20:30 -!- anekos is now known as awayekos 19:30:41 Argh. This whole "SLIME gets updated as a dependency and refuses to work as a result" thing is getting really annoying. 19:31:55 -!- brianj_otterZZZ is now known as brianj_otter 19:32:06 dnolen [n=dnolen@ppp-70-242-123-153.dsl.rcsntx.swbell.net] has joined #lisp 19:32:49 ejs [n=eugen@92-49-229-186.dynamic.peoplenet.ua] has joined #lisp 19:37:25 benny [n=benny@i577A0090.versanet.de] has joined #lisp 19:42:26 hkBst [n=hkBst@gentoo/developer/hkbst] has joined #lisp 19:44:25 -!- netaust1n [n=austinsm@adsl-70-130-220-67.dsl.stlsmo.swbell.net] has quit [Read error: 110 (Connection timed out)] 19:45:16 -!- BrianRice [n=briantri@c-98-225-51-246.hsd1.wa.comcast.net] has quit [] 19:45:42 wol [n=wol@c-24-4-220-163.hsd1.ca.comcast.net] has joined #lisp 19:47:33 -!- silenius [n=jl@yian-ho03.nir.cronon.net] has quit [] 19:47:57 rtra_ [n=rtra@unaffiliated/rtra] has joined #lisp 19:48:15 _8david: How does one inclue a literal at sign in a docstring? 19:48:24 @@ doesn't appear to work. 19:50:41 Hun [n=Hun@81-163-30-176.visitor.congress.ccc.de] has joined #lisp 19:50:55 -!- Jabberwockey [n=jens@dslb-082-083-080-185.pools.arcor-ip.net] has quit [Remote closed the connection] 19:51:11 drdo [n=psykon@167.111.54.77.rev.vodafone.pt] has joined #lisp 19:51:56 Jabberwockey [n=jens@dslb-082-083-080-185.pools.arcor-ip.net] has joined #lisp 19:52:58 oudeis [n=oudeis@HFA62-0-190-81.bb.netvision.net.il] has joined #lisp 19:53:24 -!- rtra [n=rtra@unaffiliated/rtra] has quit [Read error: 145 (Connection timed out)] 19:54:36 Davidbrcz [n=david@ANantes-151-1-38-22.w83-195.abo.wanadoo.fr] has joined #lisp 19:55:06 -!- schme_ is now known as schme 19:55:14 at [n=at@nat.garanta.ru] has joined #lisp 19:55:26 lichtblau pasted "bugfix for Aankhen``" at http://paste.lisp.org/display/72712 19:55:31 Well this is great. (find-system :atdoc) works fine at the REPL but dies when it's used within Atdoc. 19:55:58 _8david: Thanks! 19:56:06 slyrus_ [n=slyrus@207.189.195.44] has joined #lisp 19:56:28 <_8david> @@ was meant to work, but I must have broken it 19:56:44 <_8david> not sure what you mean regarding find-system and "within atdoc" 19:56:47 pcc [i=pchapin@dialup1741.wnskvtao.sover.net] has joined #lisp 19:57:44 hello. With 'M-x slime' I get *inferior-lisp* but no CL-USER REPL. How do I get CL-USER REPL? 19:58:35 -!- pcc [i=pchapin@dialup1741.wnskvtao.sover.net] has left #lisp 19:58:45 _8david: (asdf:find-system :atdoc) works fine at the REPL, but when it's called by ATDOC::MAGIC-NAMESTRING, it fails to find the TRUENAME of something to do with SBCL. I'm restarting SLIME to see if maybe it has something to do with *DEFAULT-PATHNAME-DEFAULTS*. 19:58:58 Although that would seem unlikely. 19:59:35 at: Use the slime-fancy contrib 19:59:54 failed to find the TRUENAME of /home/aankhen/lisp/clbuild/target/lib/sbcl/: No such file or directory 20:00:07 That directory does seem to exist, though. 20:00:43 The call is (asdf:find-system :atdoc) in atdoc.lisp:35. 20:01:33 HET3 [n=diman@chello084114161225.3.15.vie.surfer.at] has joined #lisp 20:01:38 Why is the repl a contrib now? Seems like a rather basic feature. 20:02:07 I think so too. 20:02:19 You could argue that it's just sugar around *inferior-lisp* though 20:02:19 *Aankhen``* is guessing that has something to do with Madhu's messages to slime-devel. 20:03:55 Still, whether it lives in a separate file is not important, but it definitely should be loaded by default. 20:04:10 any cms systems in lisp? :) 20:04:13 Hell, slime-fancy should be loaded by default. 20:04:23 -!- tompa [n=thomas@h-224-178.A151.priv.bahnhof.se] has quit [Read error: 110 (Connection timed out)] 20:04:42 luis: I argued for it in the past. Initially, a meta contrib like slime-fancy wasn't even planned 20:04:55 Agreed. It's better to err on the side of excessive features than too spartan a first impression. 20:08:03 -!- dmiles_afk [i=dmiles@c-24-16-246-158.hsd1.wa.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 20:08:50 dmiles_afk [i=dmiles@c-24-16-246-158.hsd1.wa.comcast.net] has joined #lisp 20:09:33 For some reason, ASDF:CONTRIB-SYSDEF-SEARCH is failing when it says (TRUENAME "/home/aankhen/lisp/clbuild/target/lib/sbcl"), though that works fine in the REPL. 20:10:00 You should have voiced your opinion back then. I mostly gave up on slime development in the meanwhile. 20:10:21 -!- kiuma [i=51d06a4b@gateway/web/ajax/mibbit.com/x-ad9d4ac08de8564f] has quit ["http://www.mibbit.com ajax IRC Client"] 20:10:25 -!- HET2 [n=diman@chello084114161225.3.15.vie.surfer.at] has quit [Read error: 110 (Connection timed out)] 20:10:47 Aankhen``: missing / at the end 20:11:03 -!- syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has quit [Read error: 60 (Operation timed out)] 20:11:08 asdf goes crazy with too few / 20:11:14 luis: Sorry, that was my typo. It's correct in the source. 20:12:03 (SB-IMPL::QUERY-FILE-SYSTEM "/home/aankhen/lisp/clbuild/target/lib/sbcl/" :TRUENAME) calls (SB-IMPL::QUERY-FILE-SYSTEM "/home/aankhen/lisp/clbuild/target/lib/sbcl/" :TRUENAME T) which calls (FLET SB-IMPL::FAIL). 20:12:14 And all this works happily from the REPL. 20:13:57 -!- fusss [n=chatzill@ip70-187-234-43.dc.dc.cox.net] has quit ["ChatZilla 0.9.84 [Firefox 3.0.5/2008120122]"] 20:15:06 jgracin_ [n=jgracin@82.193.208.195] has joined #lisp 20:15:15 schme_ [n=marcus@c83-249-80-232.bredband.comhem.se] has joined #lisp 20:15:23 syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has joined #lisp 20:18:56 -!- ejs [n=eugen@92-49-229-186.dynamic.peoplenet.ua] has quit [Read error: 110 (Connection timed out)] 20:21:43 Jasko [n=tjasko@c-98-235-21-22.hsd1.pa.comcast.net] has joined #lisp 20:23:41 -!- schme [n=marcus@sxemacs/devel/schme] has quit [Read error: 145 (Connection timed out)] 20:23:45 _8david: Anyway, Atdoc was choking earlier because my package exports some symbols from the CL package (long story), and it blew up when it tried to dump some of the docstrings. I patched it locally to take a function that filters symbols, which in my case tested that the symbol wasn't being re-exported from :CL. Would you be interested in said patch? 20:24:13 Then again, exporting symbols from CL: would seem to be a rather niche area of interest. 20:24:34 -!- jgracin [n=jgracin@82.193.208.195] has quit [Read error: 110 (Connection timed out)] 20:30:26 *tic* kicks stumpwm gently in the groin. 20:34:29 <_8david> sure, keep the bug reports and/or patches coming 20:36:03 BrianRice [n=briantri@dsl081-166-030.sea1.dsl.speakeasy.net] has joined #lisp 20:38:30 fooquux [n=fooquux@udp265832uds.hawaiiantel.net] has joined #lisp 20:42:11 -!- oudeis [n=oudeis@HFA62-0-190-81.bb.netvision.net.il] has quit ["This computer has gone to sleep"] 20:43:46 Aankhen pasted "Atdoc patch: add SYMBOL-FILTER-FUNCTION" at http://paste.lisp.org/display/72714 20:43:58 _8david: There you go. 20:44:13 Hmm, I wonder how those whitespace changes crept in even though I said --ignore-space-change. 20:44:57 Ah, those lines included other subtle changes too. 20:46:19 hello guys 20:46:42 syamajal_ [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has joined #lisp 20:47:00 i'm trying to add all numbers who are bellow 1000 and who are multiples of 3 or 5 20:47:05 -!- slyrus_ [n=slyrus@207.189.195.44] has quit [Read error: 110 (Connection timed out)] 20:47:16 i'm doing this by recursion 20:47:28 but my stack explode 20:47:41 does 1000 recursiv call are too much for lisp ? 20:50:12 first, are you sure that your code is correct? second, did you compile the function? 20:50:32 adeht: no, i'm a begginer in lisp 20:50:33 jgracin__ [n=jgracin@82.193.208.195] has joined #lisp 20:50:48 so it's more than possible that my code is faulty 20:50:51 are* 20:50:55 is** 20:51:09 -!- syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has quit [Read error: 110 (Connection timed out)] 20:51:10 and i compiledthe function with clisp 20:52:03 -!- brianj_otter is now known as brianj_ottterAFK 20:52:38 type (compile 'function-name) in the repl, then try again 20:53:52 first of all, i want to have the right result with 10 20:55:25 puchacz [n=puchacz@87-194-5-99.bethere.co.uk] has joined #lisp 20:59:26 -!- jgracin_ [n=jgracin@82.193.208.195] has quit [Read error: 110 (Connection timed out)] 21:00:37 -!- Jabberwockey [n=jens@dslb-082-083-080-185.pools.arcor-ip.net] has quit [Remote closed the connection] 21:01:02 Athas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has joined #lisp 21:04:05 Davidbrcz: recursion is not a smart idea, as it might blow the stack if it cannot fold (right word?) it. Ensure you have tail-end recursion if you're trying to do so. An iterative approach is too simple to not give it to you: (loop for x from 0 below 1000 when (or (= 0 (mod x 3)) (= 0 (mod x 5))) sum x) 21:05:49 m4dnificent: i would use an recursive approche as much as possible 21:07:55 because... 21:08:09 the lecturer told so 21:08:45 tcr: implying that I shouldn't paste my solution, right ^_^ 21:09:14 m4dnificent: i've read use recussion is better in most of case in lisp, ocaml, and all of sort of languages 21:09:56 Recursion is not better in most cases than iteration. 21:10:08 Honestly, this sounds like a job for LET, DOTIMES, WHEN, and INCF. 21:10:22 -!- syamajal_ [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has quit [Read error: 60 (Operation timed out)] 21:10:53 Now you read the opposite. The opposite was stated more recently, so please take up on. 21:10:56 -!- Hun [n=Hun@81-163-30-176.visitor.congress.ccc.de] has quit [Read error: 110 (Connection timed out)] 21:11:00 Davidbrcz: sounds like a bunch of rubbish. You must use what is A) the clearest and/or B) the fastest (or anything in between as to what you want to reach). Saying that one 'better' than the other without extra definition is... something bad! 21:11:23 nyef: Not LOOP? 21:11:47 Aankhen``: Not LOOP. Besides, someone already suggested LOOP. 21:11:48 That would seem to be the easiest route. 21:11:53 Aankhen``: loop has been pasted, nyef probably doesn't like loop (he is not the only one) 21:12:02 Oh, I didn't see that. 21:12:27 Bleh, I'm going to bed before I start missing entire pages. 21:12:28 G'night. 21:12:31 -!- Aankhen`` [n=Aankhen@122.162.166.64] has quit [" Samba-- # for adopting gplv3."] 21:12:34 No, I love LOOP. I have one LOOP with six FOR clauses. But this isn't necessarily a job for LOOP and, as I said, LOOP has already been presented. 21:13:02 nyef: I judged too fast, my appologies 21:13:16 blbrown [n=Berlin@c-71-236-25-127.hsd1.ga.comcast.net] has joined #lisp 21:13:41 m4dnificent: No worries. I'm just not one of those vocal ITERATE or SERIES fans. 21:14:03 dlisboa [n=diogo@bd214061.virtua.com.br] has joined #lisp 21:14:44 -!- srp [n=srp@122.172.2.116] has quit [] 21:14:53 I haven't checked it yet, but series does look like a nice and powerfull addition to the constructs that can be made by loop. I tend to like things that offer for greater abstractions (and series seems to have proven it works) 21:15:01 syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has joined #lisp 21:15:41 -!- davazp [n=user@158.Red-88-8-228.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 21:17:34 yeah ! 21:17:40 i've done it ! 21:19:10 Hrm. 21:20:16 -!- malumalu_ [n=malu@hnvr-4dbbf76e.pool.einsundeins.de] has quit ["Verlassend"] 21:21:43 I have a class which manages my ptrace interface. Part of this is a function for checking for events via waitpid. As I want to be able to background this entire thing via a signalfd and serve-event, I can't pass a function handle to the event-checking function. 21:22:03 The event-handling function thus needs to be stored in a slot on the instances. 21:23:16 Now, should I assume that this function will always be a closure or there will only be one instance, or should I pass the ptrace interface instance as an argument? 21:30:11 bakkdoor [n=bakkdoor@bakkdoor-ubuntu.visitor.congress.ccc.de] has joined #lisp 21:30:22 -!- blbrown [n=Berlin@c-71-236-25-127.hsd1.ga.comcast.net] has quit ["Ex-Chat"] 21:31:35 -!- bakkdoor [n=bakkdoor@bakkdoor-ubuntu.visitor.congress.ccc.de] has quit [Client Quit] 21:31:37 netaust1n [n=austinsm@adsl-70-130-228-169.dsl.stlsmo.swbell.net] has joined #lisp 21:34:04 nyef: unless you're following some sort of guide, I think it is fairly safe to assume it is a closure. Users can take care of possible concurrency issues themselfves, no? 21:34:14 bakkdoor [n=bakkdoor@bakkdoor-ubuntu.visitor.congress.ccc.de] has joined #lisp 21:34:19 myrkraverk [n=johann@unaffiliated/myrkraverk] has joined #lisp 21:36:48 blbrown [n=Berlin@c-71-236-25-127.hsd1.ga.comcast.net] has joined #lisp 21:38:17 lispm [n=joswig@f054055037.adsl.alicedsl.de] has joined #lisp 21:38:27 m4dnificent: Concurrency? With this interface?!? God forbid! 21:38:56 (And "some sort of guide"? Can you even -hire- guides for this kind of crap?) 21:39:20 -!- blbrown [n=Berlin@c-71-236-25-127.hsd1.ga.comcast.net] has quit [Client Quit] 21:42:21 nyef: Well, I don't know what you're writing the library for :) With guide I meanth anything as guide (a standard, a request of a client, ...) 21:44:38 pstickne [n=pstickne@c-71-236-177-238.hsd1.wa.comcast.net] has joined #lisp 21:46:21 syamajal_ [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has joined #lisp 21:47:20 petro [n=pet-ro@p4FE955D0.dip.t-dialin.net] has joined #lisp 21:47:22 nyef: pass the instance, or use a singleton-class metaclass if you really want a singleton. 21:47:33 *drewc* just got here and myay have missed important details 21:47:36 may 21:48:31 drewc: can you explain costanza's dynamic-wind endeavor to me? 21:49:02 tcr: I think so... i half-hacked something similar up myself. :) 21:49:12 tcr: how familiar are you with contextl? 21:49:29 totally unfamiliar 21:49:44 ok .. well we have to start there. 21:49:46 what does dynamic-wind do? I thought it was scheme's unwind-protect, but it seems to do more 21:50:22 in scheme it's pretty much unwind-protect .. in CL with the presense of dynamic scope, it has to do a bit more. 21:50:30 especially when you mix in call/cc 21:50:54 I read Costanza's first paper about ContextL so I know roughly what it's supposed to provide. I also know the history behind contextl (a means for providing dynamic binding to methods) 21:51:13 not just methods.. but slots as wel 21:51:14 well 21:51:27 and that is really where this extension comes in really handy. 21:52:45 -!- syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has quit [Read error: 110 (Connection timed out)] 21:52:53 the major issue we've all been having is that you couldn't close over a dynamic environment. 21:53:13 so yeah your continuations were broken in that regard 21:53:32 right yeah .. closures were broken really. 21:53:58 you'd have to explicitly bind and rebind the special values you wanted. 21:54:03 no lexical closure are lexical closures, they're not supposed to close over everything. And in most cases it does even make sense 21:54:20 right.. agreed. 21:54:40 but the dynamic environment, otoh, is nice to grab on to. 21:55:15 is to grab on to? or do you mean it would be nice if it was possible to easily grab on to? 21:55:18 slash_ [n=Unknown@p4FF0A7D8.dip.t-dialin.net] has joined #lisp 21:55:24 drewc, when is it nice to grab on to? 21:55:27 hello all 21:55:43 (boundp 'fe[nl]ix) 21:55:57 t 21:56:05 S11001001: watch out you're perhaps looking in the wrong namespace! 21:56:09 err too slow 21:56:24 tcr: in LoL, for example, i have a major need to 'close over' the values of a lot of 'special places' 21:56:35 fe[nl]ix: if you've looked, do you like the patch? 21:56:39 drewc: special places are slots of layered classes? 21:56:55 -!- decafbad [n=mehmet@88.232.66.140] has quit ["Leaving"] 21:57:02 layered slots of layered classes, but yes 21:57:13 yes that's what I meant 21:57:23 BIND ME HARDER 21:57:28 kpreid [n=kpreid@pool-72-68-32-150.nwrknj.east.verizon.net] has joined #lisp 21:57:36 i go a lot further in LoL, and slots actually have different values in different layers 21:57:48 tcr: maybe I should look in *FEATURES* yes? 21:58:20 -!- Davidbrcz [n=david@ANantes-151-1-38-22.w83-195.abo.wanadoo.fr] has quit ["Ex-Chat"] 21:59:26 S11001001: You can be lucky that fe[nl]ix doesn't hide as GO, or CATCH tag! 21:59:36 :D 21:59:50 tcr: in contextL , special slots are really just a gensym, and slot-value pulls the symbol-value of that (special) gensym 22:00:34 DLETF does a PROGV on the symbols, and voila, dynamic slots. 22:00:34 drewc: Yeah, to get the extra level of indirection 22:01:11 so it can be very useful to capture those values as part of a greater 'context' 22:01:30 Yes that's what I supposed. (I wished for dynamic rebinding of methods on occasion.) 22:01:33 daniel_ [i=daniel@unaffiliated/daniel] has joined #lisp 22:01:36 S11001001: I think it's a little overengineered 22:01:48 tcr: you can do that in clojure :) 22:02:10 -!- brianj_ottterAFK is now known as brianj_otter 22:02:13 fe[nl]ix: What should I take out? 22:02:15 also, actually, it's easy to emulate with contextl 22:02:39 drewc: the rejected GENERIC-FLET stuff provided this, too 22:02:56 indeed .. i always wondered why we lost that one. 22:03:09 wait .. generic-flet was lexical though .. non? 22:03:19 It was not obvious to most everyone else RPG what it's supposed to do 22:03:29 no, generic-flet was dynamic, with-added-methods was lexical 22:03:34 or the other way around 22:03:39 they both sound painful to implement to me 22:03:52 ahh yes, that makes sense. 22:04:09 S11001001: I think I'd prefer wrapping make-thread in a PROGV in order to emulate thread-local specials 22:04:20 well, layered methods provide identical functionality really. 22:04:43 fe[nl]ix: I thought about that, but it won't work for threads that weren't created with bt:make-thread 22:05:16 in what sense ? 22:05:26 only you don't inline the method body, it's deflayer, define-layered-method and in-layer 22:06:24 and actually ... using an anonymous layer and ensure-layered-method you could easily do with-added-methods (if that's the dynamic one). 22:06:45 fe[nl]ix: Well, I guess that's mainly talking about if you put it in bt:make-thread itself 22:07:01 I'm speaking from Weblocks's position here. 22:07:06 Jabberwockey [n=jens@dslb-082-083-080-185.pools.arcor-ip.net] has joined #lisp 22:07:30 Our threads are created by Hunchentoot; we don't have an opportunity to bind at their creation 22:08:01 it's easy to change make-thread 22:08:10 i do something very similar in lol. slot-value actually funcalls an anonymous layered method, and that method can of course be specialized in different contexts, so slot values differ in different layers. 22:08:15 except that hunchentoot 0.15.7 doesn't use bt:make-thread 22:08:47 tcr: to capture all that reliably, i need Pascal's latest genius. 22:08:54 drewc: I think, yes, with-added-methods is the dynamic one, makes more sense 22:08:58 rvirding [n=chatzill@h40n5c1o1034.bredband.skanova.com] has joined #lisp 22:09:06 drewc: But what does dynamic-wind do? 22:09:35 He's not talking about saving the whole dynamic environment, including all special variables, is he? 22:09:47 only what's relevant to contextl? 22:10:45 no, i wouldn't think it _implicitly_ support that. 22:11:00 only the contextl environment is how i read it. 22:11:22 then again ... 22:12:57 i'm wrong... 22:13:02 S11001001: why not just patch hunchentoot ? 22:13:09 dynamic-wind? scheme or sumthin'? 22:13:16 -!- Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has quit [] 22:13:39 oh wow i see what he did there. 22:13:56 S11001001: redefine hunchentoot:process-run-function in terms of BT 22:13:58 dys [n=andreas@p5B316887.dip.t-dialin.net] has joined #lisp 22:14:19 fe[nl]ix: I tried patching it once 22:14:30 it broke ? 22:14:32 Edi told me to hack the svn version instead 22:14:48 I fixed a deadlock on SBCL in 0.15.7 22:15:50 I'll be more comfortable with making Weblocks users use the bknr svn version instead of 0.15.7 when Edi links to it on the official Hunchentoot page 22:16:30 -!- lhz [n=shrekz@c-3143e455.021-158-73746f34.cust.bredbandsbolaget.se] has quit ["Leaving"] 22:16:32 -!- jgracin__ [n=jgracin@82.193.208.195] has quit [Remote closed the connection] 22:17:41 syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has joined #lisp 22:18:22 -!- daniel [i=daniel@unaffiliated/daniel] has quit [Read error: 110 (Connection timed out)] 22:18:31 tcr: it doesn't capture the lisp dynamic environement.. I _think_ it only captures any dynamic bindings introduced within the DYNAMIC-WIND form. 22:19:11 yes exactly. 22:19:41 alinrus [n=root@89.123.225.38] has joined #lisp 22:20:21 DYNAMIC-WIND creates a first class dynamic environment which you can capture. 22:20:34 and compose 22:21:36 he talks about the possibility of making dynamic-wind a function 22:21:39 -!- syamajal_ [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has quit [Read error: 110 (Connection timed out)] 22:21:49 I'm not sure how he can capture anything from within a function 22:22:14 *tcr* gtg for a walk. (bbl) 22:22:18 he doesn't have to .. he's not actually capturing anything. 22:22:21 blbrown [n=Berlin@c-71-236-25-127.hsd1.ga.comcast.net] has joined #lisp 22:22:36 maybe he's off his rocker, and the whole thing is in fact the huge kludge it appears to be from a distance 22:22:57 he runs the code that establishes the bindings again. 22:23:28 hefner: genius insanity 22:24:20 dys` [n=andreas@p5B316887.dip.t-dialin.net] has joined #lisp 22:24:24 i know exactly how to use this, right now, to eliminate tons of redundant code and clean up the design of my systems significantly. 22:24:58 that the existence of the svn version of Hunchentoot is a mystery unless you go looking for it says something 22:25:11 I wouldn't be surprised if it's leaky in some way 22:25:14 clbuild installs 0.15.7 from luis's Darcs repo instead of checking out from bknr 22:25:28 tcr: a functional DYNAMIC-WIND would simply have you pass the continuation to 22:25:41 s/to// 22:26:01 -!- CrEddy [n=CrazyEdd@wrongplanet/CrazyEddy] has quit [Read error: 110 (Connection timed out)] 22:26:41 drewc: You're right, thank you. Passing the process object to the event-function is indeed the correct approach. 22:27:38 wow .. DYNAMIC-WIND FTW! 22:27:47 the more this sinks in the happier i am with it. 22:28:03 (This would have been obvious had I twigged to the incredible code duplication involved in handling the possibility of a NIL wait-function earlier. One helper-function later and I realized that it would an extensibility hook if it was a generic function...) 22:28:16 I think I can see at least one problem the way he specified it. 22:28:25 but I rally gotta go now :) 22:28:25 tcr: do tell 22:28:32 Heh. 22:28:32 ok, come back soon :) 22:29:17 CrEddy [n=CrazyEdd@220-253-60-178.VIC.netspace.net.au] has joined #lisp 22:29:31 I think I'm close to being able to write some version of strace(1). 22:29:32 this fixation on dynamic scope makes me nervous, but it's been a couple years since I read the contextl paper 22:29:57 hefner: contextl makes dynamic scope cool again. 22:30:07 -!- dys [n=andreas@p5B316887.dip.t-dialin.net] has quit [Connection timed out] 22:31:02 O_4 [n=souchan@125-236-190-161.broadband-telecom.global-gateway.net.nz] has joined #lisp 22:31:03 slyrus_ [n=slyrus@207.189.195.44] has joined #lisp 22:33:53 aszarsha [n=Aszarsha@dsl-67-55-12-235.acanac.net] has joined #lisp 22:34:28 yhara [n=yhara@7.193.12.221.megaegg.ne.jp] has joined #lisp 22:34:56 dys`` [n=andreas@p5B316887.dip.t-dialin.net] has joined #lisp 22:35:19 Yuuhi [i=benni@p5483F991.dip.t-dialin.net] has joined #lisp 22:37:05 -!- Krystof [n=csr21@84-51-132-95.christ977.adsl.metronet.co.uk] has quit [Read error: 113 (No route to host)] 22:38:25 -!- dys`` is now known as dys 22:40:11 -!- blbrown [n=Berlin@c-71-236-25-127.hsd1.ga.comcast.net] has quit [Connection timed out] 22:40:13 -!- jlouis [n=jlouis@port1394.ds1-vby.adsl.cybercity.dk] has quit [Read error: 110 (Connection timed out)] 22:40:35 -!- dys` [n=andreas@p5B316887.dip.t-dialin.net] has quit [Connection timed out] 22:41:08 blbrown [n=Berlin@c-71-236-25-127.hsd1.ga.comcast.net] has joined #lisp 22:46:06 syamajal_ [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has joined #lisp 22:48:13 jlouis [n=jlouis@port1394.ds1-vby.adsl.cybercity.dk] has joined #lisp 22:52:49 -!- syamajala [n=syamajal@c-75-68-104-47.hsd1.ma.comcast.net] has quit [Read error: 110 (Connection timed out)] 22:52:54 oh noes! 22:53:04 *drewc* panics as his heater runs out of fuel. 22:53:29 Consing keeps you warm. 22:53:29 fooquux, memo from luis: fixed the defcfun documentation again. Keep those complaints coming! 22:54:40 -!- rtra_ is now known as rtra 22:55:14 luis, are you here? 22:55:21 I have another CFFI bug. 22:55:28 Heh. 22:55:36 This one's minor. 22:55:39 Documentation bug? 22:55:44 A little of both. 22:55:56 The docs for DEFCVAR specify a documentation string. 22:56:17 But DESCRIBE on a variable created with DEFCVAR doesn't show the docstring. 22:56:28 At least on SBCL, not sure about others. 22:56:56 indeed, the documentation string is ignored. 22:57:07 that should be easy to fix 22:57:12 Thanks! 22:57:22 I <3 CFFI by the way. 22:58:05 beach [n=user@203.210.248.212] has joined #lisp 22:58:10 Good morning. 22:58:16 hi beach 23:00:44 mulligan [n=user@e178011196.adsl.alicedsl.de] has joined #lisp 23:00:45 oudeis [n=oudeis@bzq-79-177-106-178.red.bezeqint.net] has joined #lisp 23:01:14 chandler: I tried to poke my sysadmin to bring the Alpha Krystof used some time ago, but i think the box's definitely dead. 23:01:27 *bring back 23:03:34 hey beach 23:03:40 -!- __jasonherald__ [n=__jasonh@pool-72-66-198-13.ronkva.east.verizon.net] has quit [Read error: 110 (Connection timed out)] 23:04:03 __jasonherald__ [n=__jasonh@pool-71-171-38-188.nwrknj.east.verizon.net] has joined #lisp 23:05:20 yhara_ [n=yhara@7.193.12.221.megaegg.ne.jp] has joined #lisp 23:05:32 drewc: still here? 23:05:39 t 23:06:05 drewc: do you understand the mark thingie, and why it doesn't for deactivation? 23:06:36 -!- yhara [n=yhara@7.193.12.221.megaegg.ne.jp] has quit [Read error: 104 (Connection reset by peer)] 23:07:52 sledge [n=chris@pdpc/supporter/student/sledge] has joined #lisp 23:08:13 yes. the reason we need dynamic wind is so we can reliably capture the current 'context' and re-execute code within that context. 23:08:53 the issue is that any given context is a set of layers that have been explicitly activated or de-activeated. 23:09:17 -!- bakkdoor [n=bakkdoor@bakkdoor-ubuntu.visitor.congress.ccc.de] has quit [Read error: 110 (Connection timed out)] 23:09:37 but .. deactivations are simply removing a layer object from the list of active contexts .. 23:09:38 what are partial continuations? 23:09:46 delimited continuations 23:10:21 -!- ironChicken [n=richard@79-75-104-125.dynamic.dsl.as9105.com] has left #lisp 23:10:49 like we have in most lisp call/cc libs.. an explicit (with-call/cc ...) delimits the environment in which continuations can be captured/executed 23:11:06 bakkdoor [n=bakkdoor@81-163-62-234.visitor.congress.ccc.de] has joined #lisp 23:11:20 I have a slime question. 23:11:28 I am working on a curses program. 23:11:56 Is there a way that my lisp image can talk via slime and still run in a terminal? 23:12:04 Pointers to docs gladly accepted. 23:12:21 tcr: it's indeed leaky in that it doesn't really capture the dynamic environment .. it captures the code that establishes that environment. 23:13:19 so side-effecting within a dynamic-wind form, outside of PROCEED (which should be called CONTINUE imo).. 23:13:26 -!- lemoinem [n=swoog@modemcable125.83-81-70.mc.videotron.ca] has quit [Nick collision from services.] 23:13:33 is going to cause issues. 23:13:38 lemoinem [n=swoog@modemcable125.83-81-70.mc.videotron.ca] has joined #lisp 23:13:56 -!- lispm [n=joswig@f054055037.adsl.alicedsl.de] has quit [] 23:13:56 pkhuong_ [n=pkhuong@modemcable125.83-81-70.mc.videotron.ca] has joined #lisp 23:14:03 but, for the uses we require (using contextl with call/cc), it is exactly what we need. 23:14:05 fooquux: I don't know what your question is supposed to mean. Could you elaborate what you want? 23:14:29 fooquux: Load swank, then use it to create a server, then use M-x slime-connect from emacs? 23:14:32 drewc: I'm currently trying to wrap my head around wiki's entry for delimeted continuations :) 23:14:49 -!- skv [n=user@076-076-146-016.pdx.net] has quit [Read error: 110 (Connection timed out)] 23:14:53 tcr: i've written call/cc and i don't understand most of that page. 23:15:04 I think what I'm imagining will require threads. 23:15:50 I want a single lisp image that has terminal i/o in a curses environment, but also a slime connection too. 23:16:02 -!- kpreid [n=kpreid@pool-72-68-32-150.nwrknj.east.verizon.net] has quit [] 23:16:05 drewc: If I get it the right the delimeted continuations evaluates everything between reset and shift 23:16:17 is that right, pkhuong? 23:16:20 tcr: http://blog.moertel.com/articles/2008/05/05/olegs-great-way-of-explaining-delimited-continuations 23:16:22 (`delimited') 23:17:10 tcr: that is essentially it 23:17:58 the code between resent and shift is the continuation, but it's not a full continuation.. it knows nothing outside of reset 23:18:22 the scope of the continuation is delimited by the reset form. 23:18:54 this surely looks like magic. Let me digest that blog's comparison. 23:19:28 in CL we really can only have partial or delimited continuations... full reified continuations would require compiler support, and fix that pesky UNWIND-PROTECT issue. 23:19:50 -!- slyrus_ [n=slyrus@207.189.195.44] has quit [Read error: 60 (Operation timed out)] 23:21:43 fooquux: pushed a patch, should be working now. 23:22:46 -!- hkBst [n=hkBst@gentoo/developer/hkbst] has quit [Remote closed the connection] 23:23:57 -!- BrianRice [n=briantri@dsl081-166-030.sea1.dsl.speakeasy.net] has quit [] 23:25:01 -!- pkhuong [n=pkhuong@modemcable125.83-81-70.mc.videotron.ca] has quit [Read error: 110 (Connection timed out)] 23:26:20 -!- alinrus [n=root@89.123.225.38] has quit [Read error: 110 (Connection timed out)] 23:27:53 Can you have two different SHIFTs within the scope of the same RESET? 23:28:43 can you have a shift within a shift that both refer to the same reset? 23:29:14 -!- bakkdoor [n=bakkdoor@81-163-62-234.visitor.congress.ccc.de] has quit [Read error: 110 (Connection timed out)] 23:29:35 sepisultrun [n=user@ks35219.kimsufi.com] has joined #lisp 23:29:55 nyef: You're working on some direct syscall invocation machinery? 23:29:58 -!- sepisultrun [n=user@ks35219.kimsufi.com] has left #lisp 23:30:32 -!- yhara_ [n=yhara@7.193.12.221.megaegg.ne.jp] has quit [Read error: 113 (No route to host)] 23:33:08 tcr: I was toying with it a couple days ago, why? 23:33:16 You want my working notes? 23:33:38 -!- petro [n=pet-ro@p4FE955D0.dip.t-dialin.net] has quit ["Verlassend"] 23:33:47 nyef: I just scribbled (syscall 'read ()) on my scratchpad! 23:33:52 -!- beach [n=user@203.210.248.212] has left #lisp 23:34:27 tcr: not as i understand it.. the function would never return .. 23:34:34 or wait.. hold on. 23:35:27 tcr: I have it worked out for x86 and amd64, as far as knowing what the conventions and entry points are, but haven't actually written the code yet. 23:35:52 On how many multiple projects are you simultaneously working on? 23:36:06 Not sure, really. 23:36:20 The syscall thing was a bit of a diversion. 23:36:51 why syscalls? escaping libc? 23:36:53 I mean, sure, I wanted to do it, but if it hadn't been for me typing "pthread" instead of "ptrace" in a critical spot it would have been later. 23:36:57 sepisultrun [n=user@ks35219.kimsufi.com] has joined #lisp 23:37:13 how can I check if a symbol has been bound allready? 23:37:16 lucca: That's one option, yes. 23:37:19 sepisultrun: BOUNDP. 23:37:20 sepisultrun: boundp 23:37:47 oh, that simple, thanks 23:37:49 lucca: I'm also looking at an... unconventional model for a LispOS. 23:38:22 alinrus [n=root@89.123.225.38] has joined #lisp 23:38:40 But that's a subject for another time. 23:39:37 http://paste.lisp.org/display/72719 is this the idiomatic way to do something onece in common lisp? 23:40:52 unless quote 23:41:21 hmm paste.lisp.org ain't working for me 23:41:27 ah not it is 23:42:04 sepisultrun: No it's not very idiomatic. 23:42:19 tcr: ok, how would you do it? 23:42:41 S11001001: wouldn't it be easier to tell your users to use Hunchentoot from bknr, possibly showing how to do that on the weblocks wiki ? 23:43:14 incidentally I implemented *default-special-bindings*, writing up a test now 23:44:00 I'd have to see if it was okay with Leslie, because Weblocks isn't ported to bknr Hunchentoot yet 23:44:17 -!- Nshag [i=user@Mix-Orleans-106-3-192.w193-248.abo.wanadoo.fr] has quit ["Quitte"] 23:45:44 and it would also be nice if clbuild would also make the same choice, since many people getting into weblocks are using it to install 23:45:44 -!- mrSpec [n=SomeOne@88.208.105.1] has quit ["Bye"] 23:46:49 sepisultrun: (defun maybe-start-swank-server () (unless (swank::default-connection) (swank-server))) 23:49:21 is bknr hunchentoot single threaded ? 23:50:10 -!- alinrus [n=root@89.123.225.38] has quit [Remote closed the connection] 23:50:11 tcr: ahh, thank you. I like this maybe-... naming convention. 23:51:42 no 23:52:17 phf [n=phf@c-98-231-140-203.hsd1.md.comcast.net] has joined #lisp 23:52:22 -!- milanj [n=milan@77.46.225.124] has quit ["Leaving"] 23:53:00 but I'm going to move ht fluids into clsql because it's easier to make fluid-database thread-safe without removing unboundedness than hacking the *default-special-bindings* every time I make a fluid-database 23:53:26 *default-special-bindings* is great if you only need one named fluid, not so much for any number 23:54:31 tcr: ok, you can indeed have many shifts within on reset. 23:54:35 one reset 23:54:48 as i know understand it, revised from earlier :) 23:57:05 i think the inner shift is delimited by the outer shift when nesting. 23:58:33 What does (reset (* 2 (shift k (+ (k 3) 1)) 5)) return?