00:01:47 antoszka [~antoszka@unaffiliated/antoszka] has joined #lisp 00:06:32 dreish [~dreish@2002:cf8a:2fad:0:21f:5bff:fe35:ae0d] has joined #lisp 00:08:33 pkhuong: try getting the hashtable microbenchmark run with 1G pages :) 00:10:13 p_l: not that many architectures support those. Plus, I already have decent data for out of cache and in-TLB. 00:10:37 -!- potatishandlarn [potatishan@c-b21e760c-74736162.cust.telenor.se] has quit [] 00:12:49 pkhuong: AMD64 on newer opterons supports 1G pages, iirc 00:14:30 pizzledizzle [~pizdets@pool-96-250-215-244.nycmny.fios.verizon.net] has joined #lisp 00:15:42 -!- bombshelter13b [~bombshelt@76-10-149-209.dsl.teksavvy.com] has quit [Remote host closed the connection] 00:19:50 -!- killerboy [~mateusz@smrw-91-193-87-5.smrw.lodz.pl] has quit [Ping timeout: 246 seconds] 00:20:34 -!- az [~az@2a01:4f8:101:4a2::9c00] has quit [Read error: Operation timed out] 00:22:34 fractal_heart [~matomic@mzhang-linux.ucsd.edu] has joined #lisp 00:23:29 Fare [~Fare@c-24-218-127-11.hsd1.ma.comcast.net] has joined #lisp 00:25:11 xyxxyyy [~xyxu@58.41.0.124] has joined #lisp 00:25:59 -!- carlocci [~nes@93.37.217.224] has quit [Quit: eventually IE will rot and die] 00:28:28 SegFaultAX [~SegFaultA@c-98-234-1-162.hsd1.ca.comcast.net] has joined #lisp 00:28:43 -!- Madsy [~madman@fu/coder/madsy] has quit [Ping timeout: 276 seconds] 00:30:42 neoesque [~neoesque@210.59.147.232] has joined #lisp 00:34:48 hi, do anyone know if there is a perfect hash function generator, as gperf, for CL? 00:36:33 -!- mindCrime [~chatzilla@eth0-2.nmd.four.lax.wayport.net] has quit [Ping timeout: 240 seconds] 00:37:12 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 258 seconds] 00:37:18 abugosh [~Adium@pool-74-103-38-231.bltmmd.fios.verizon.net] has joined #lisp 00:38:10 davazp: haven't seen one so far, been thinking of implementing one myself (though I'd probably have to do equivalent of extensible sequences for it) 00:40:11 why extensible sequences? 00:40:57 -!- abugosh [~Adium@pool-74-103-38-231.bltmmd.fios.verizon.net] has quit [Client Quit] 00:42:11 -!- Yuuhi` [benni@p5483AE9D.dip.t-dialin.net] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 00:42:33 Am I understanding the docs correctly in that nconc does a non-consing concatenation: cats my lists together in place? 00:43:01 So I've been working on the defmemoize macro, and I have something that doesn't quite work yet. Can anyone point out what I'm doing wrong? 00:43:08 (defmacro defmemoized (fun args &rest body) 00:43:09 (let ((cache (gensym))) 00:43:11 `(let ((,cache (make-hash-table :test #'equal))) 00:43:12 (defun ,fun ,args 00:43:14 (multiple-value-bind (val win) (gethash ,args ,cache) 00:43:15 (if win 00:43:16 val 00:43:18 (setf (gethash ,args ,cache) 00:43:20 ,@body))))))) 00:44:19 <_3b`> pasting it into the channel is doing something wrong :p 00:44:58 <_3b`> (aside from annoying people, it makes it harder to read/copy to a repl) 00:45:18 Sorry. Where should I post it? 00:45:27 <_3b`> minion: lisppaste 00:45:34 -!- nixeagle [~user@Wikimedia/Nixeagle] has quit [Read error: Connection reset by peer] 00:45:52 <_3b`> lisppaste: url 00:46:30 <_3b`> elliotstern: http://paste.lisp.org/new/lisp 00:46:32 -!- lolsuper_ [~super_@unaffiliated/lolsuper-/x-9881387] has quit [Ping timeout: 245 seconds] 00:48:09 http://paste.lisp.org/display/112448 00:48:12 <_3b`> elliotstern: possibly a PROGN would help... try macroexpanding a use of that macro, and see if it does what you want 00:49:13 <_3b`> also, don't forget to extract docstrings, declarations, etc and reassemble them correctly into the new function 00:49:52 <_3b`> seems like stuff you would memoize would tend to be stuff you would add optimization declarations to as well 00:50:26 Makoryu [~vt920@pool-71-174-191-10.bstnma.fios.verizon.net] has joined #lisp 00:51:09 right now this is what happens when I call macro-expand in clisp 00:51:11 http://paste.lisp.org/display/112449 00:51:13 <_3b`> also, you missed a few gensyms 00:51:36 <_3b`> (use the annotation button on the paste for more pastes about the same thing. rather than making more pastes) 00:51:43 Ok 00:52:28 -!- ikki [~ikki@201.155.75.146] has quit [Quit: Leaving] 00:52:37 What things should I have gensymmed but didn't? 00:52:47 <_3b`> macroexpand-1 is a function, so its arguments are evaluated 00:53:06 <_3b`> so you tried to macroexpand the results of running the macro, not the macro form itself 00:53:08 ugh, apparently trivial-gray-streams doesn't specialize stream-file-position (and its setf companion) on sbcl :X 00:53:25 <_3b`> adeht: sure you have the most recent version? 00:54:39 *_3b`* thought i fixed that at some point 00:54:53 3b: ah, I didn't.. it's cvs :).. usually I check and update darcs/git repos 00:54:59 3b: thanks 00:55:48 <_3b`> elliotstern: well, which binding form did you use in the function body? 00:55:53 mindCrime [~chatzilla@cpe-075-189-213-049.nc.res.rr.com] has joined #lisp 00:56:07 binding form? 00:56:35 Phoodus [foo@174-17-116-185.phnx.qwest.net] has joined #lisp 00:56:39 <_3b`> elliotstern: and in case it wasn't clear, (macroexpand-1 '(def...)) should be more informative than without the ' 00:57:23 <_3b`> form that establishes bindings of variables 00:57:52 <_3b`> (hint: it has 'bind' in the name) 00:58:01 mbohun [~mbohun@202.124.74.39] has joined #lisp 00:59:31 <_3b`> ah, i guess the problem that needs PROGN isn't the one you are hitting yet, macroexpansion should still be informative though 01:00:34 legumbre_ [~leo@r190-135-40-14.dialup.adsl.anteldata.net.uy] has joined #lisp 01:01:58 bzzbzz [~franco@modemcable240.34-83-70.mc.videotron.ca] has joined #lisp 01:02:03 -!- legumbre [~leo@r190-135-64-86.dialup.adsl.anteldata.net.uy] has quit [Ping timeout: 240 seconds] 01:02:52 -!- cmsimon [~cmsimon@unaffiliated/cmsimon] has quit [Quit: Ex-Chat] 01:03:17 potatishandlarn [potatishan@c-b21e163b-74736162.cust.telenor.se] has joined #lisp 01:03:57 manic12 [~andrew@99-100-67-123.lightspeed.sntcca.sbcglobal.net] has joined #lisp 01:06:37 nixeagle [~user@Wikimedia/Nixeagle] has joined #lisp 01:10:29 So I've made a few changes to it, and it seems to work, although it's limited to functions of a single argument at the moment. Any other suggestions? 01:10:32 -!- Komi [Komi@62.32.148.111] has quit [] 01:12:46 http://paste.lisp.org/display/112448#1 01:12:52 ephcon [~ephcon@c-24-34-195-64.hsd1.ma.comcast.net] has joined #lisp 01:13:21 -!- nyef [~nyef@pool-71-255-129-229.cncdnh.east.myfairpoint.net] has quit [Quit: G'night all.] 01:14:03 <_3b`> elliotstern: what happens if you name an argument win or val? 01:14:13 >_< 01:14:15 >_< 01:14:17 >_< 01:15:10 -!- rootzlevel [~hpd@91-66-217-25-dynip.superkabel.de] has quit [Ping timeout: 276 seconds] 01:16:44 clojay [~clojay@75-119-224-175.dsl.teksavvy.com] has joined #lisp 01:16:52 <_3b`> and if you want to use multiple args, possibly LIST would be useful 01:18:38 -!- elliotstern [~chatzilla@c-71-58-90-255.hsd1.pa.comcast.net] has quit [Ping timeout: 252 seconds] 01:21:24 -!- homie [~user@xdsl-78-34-218-138.netcologne.de] has quit [Ping timeout: 260 seconds] 01:21:31 rootzlevel [~hpd@91-66-217-25-dynip.superkabel.de] has joined #lisp 01:23:43 Guys, I need some help. I'm switching from procedural and OO programming to functional programming. In particular I'm interested in Lisp dialects (especially Clojure). I'm looking for books that will help me learn how to think as a functional programmer, and how to write idiomatic Lisp code. I'm a good programmer but I would struggle to write a compiler, a simple computer algebra system, or even an html parser. My budget is somewhat limited ( 01:23:43 but I'd like to get a couple of good books that will blow my mind and help me become a better functional programmer. There are countless books out there and it's hard to pick something within my budget. What are your suggestions? (SICP and PAIP seem the obvious choices but I'd like to hear more opinions) 01:24:42 <_3b`> this channel focuses on common lisp, which doesn't worry about being 'function' too much 01:25:33 If you're interested in learning clojure, I'd get a clojure book 01:25:58 -!- TDT` [~user@173-30-32-53.client.mchsi.com] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 01:26:09 clojay: CL is more of a multi-paradigm language. Itself it doesn't favor FP too much (it's got wonderful OO system), but it allows you to mix&match and macros are your friends, you can "enhance" it towards FP. 01:26:19 TDT [~user@173-30-32-53.client.mchsi.com] has joined #lisp 01:26:33 <_3b`> if you want to learn about common lisp, practical common lisp and common lisp: a gentle introduction to symbolic computation are both available online 01:26:51 as for clojure, there's a #clojure channel that might have better information 01:27:02 p_l: Thank you 01:27:10 http://pragprog.com/titles/shcloj/programming-clojure is fairly cheap, and a well written book 01:27:43 <_3b`> sicp is also available online, as are videos of lectures using it, and probably uses a more functional style than most CL books (it uses a subset of scheme, not CL) 01:28:11 hi p_l, _3b 01:28:19 manic12: hi 01:28:25 <_3b`> 'lo 01:28:44 I guess I'm not popualr enough to be recognized :( 01:28:45 *p_l* found himself considering going quite serious with "start a startup" business 01:28:54 poor TDT ^_- 01:29:08 <_3b`> p_l: if you have a good idea, might as well :) 01:29:13 -!- Makoryu [~vt920@pool-71-174-191-10.bstnma.fios.verizon.net] has quit [] 01:29:16 hi TDT 01:29:30 -!- Euthydemus [~euthydemu@vaxjo7.165.cust.blixtvik.net] has quit [Read error: Connection reset by peer] 01:29:32 TDT: Thank you for your suggestions. 01:29:32 *_3b`* only has mediocre ideas, in a not very good market, but tries anyway 01:29:35 *TDT* waves at manic12 :) 01:29:46 Euthydemus [~euthydemu@vaxjo7.165.cust.blixtvik.net] has joined #lisp 01:30:04 _3b`: I need to check the market availability further, but the thing is, if there are any tools for the idea I'm looking for, they are quite obscure 01:30:07 -!- xyxxyyy [~xyxu@58.41.0.124] has quit [Ping timeout: 276 seconds] 01:30:40 and the idea is quite simple, make a RAD for Web Apps that can be used with little to none programming knowledge 01:31:34 *_3b`* wants one that skips the requirement for design knowledge instead :p 01:32:13 _3b`: would a set of UI components and simple drop-in layouts be enough for you? :) 01:32:20 <_3b`> probably 01:32:43 Looks like you're thinking of something like visual studio for the web 01:32:55 I want to make it possible to do a fully custom app in it, however that would need CL knowledge ^_- 01:33:00 -!- dialtone [~dialtone@unaffiliated/dialtone] has quit [Ping timeout: 246 seconds] 01:33:03 -!- legumbre_ is now known as legumbre 01:33:06 Which isn't a bad idea, for sure. Some programmatic abiltiy may be helpful, but you could write a DSL to compensate perhaps 01:33:12 <_3b`> maybe add in one of those magic 'colors that look good together' picker things 01:33:14 TDT: except one offered on SaaS basis and directed *only* at web apps 01:33:19 visual studio requires little to none programming knowledge? 01:33:40 manic12: For some of the drag and drop stuff, no. For example, database connectivity is really easy if you're using the controls 01:33:46 manic12: no, unless you mean the case where "designers" use the Web App component of VS to create normal websites 01:34:20 which results in websites that require JS to follow links (because they are implemented as synchronous POST calls) 01:36:31 also, IMHO Visual Studio (nor other tools) don't get the feeling that Delphi had 01:37:54 -!- nixeagle [~user@Wikimedia/Nixeagle] has quit [Read error: Connection reset by peer] 01:38:32 gigamonkey [~user@adsl-99-62-73-203.dsl.pltn13.sbcglobal.net] has joined #lisp 01:38:45 In Delphi, you could quite reasonably get a simple DB application "clicked out" with most of the coding being in the style of "double click the event on the event list, fill in the action, click save" 01:44:13 -!- dreish [~dreish@2002:cf8a:2fad:0:21f:5bff:fe35:ae0d] has quit [Quit: dreish] 01:46:20 -!- peterhil [peterhil@YMKDCCLXXII.gprs.sl-laajakaista.fi] has quit [Quit: Must not waste too much time here...] 01:48:05 -!- pizzledizzle [~pizdets@pool-96-250-215-244.nycmny.fios.verizon.net] has quit [Ping timeout: 264 seconds] 01:49:26 udzinari [~user@209.158.broadband13.iol.cz] has joined #lisp 01:54:40 -!- tayloj [~tayloj@cpe-67-248-22-159.nycap.res.rr.com] has quit [Quit: Leaving.] 01:54:50 nixeagle [~user@Wikimedia/Nixeagle] has joined #lisp 01:55:35 Can someone tell me what (test (constantly t)) is doing in this parameter list: (dirname fn &key directories (test (constantly t)) 01:55:43 Dawgmatix [~dman@c-76-124-9-27.hsd1.nj.comcast.net] has joined #lisp 01:56:40 I'd start by looking up constantly ... 01:57:04 -!- jleija [~jleija@user-24-214-122-46.knology.net] has quit [Quit: good night everyone] 01:57:11 -!- curi_ [~curi@adsl-99-114-139-86.dsl.pltn13.sbcglobal.net] has quit [Quit: This computer has gone to sleep] 01:57:14 Zhivago: I did. But I am more curious about what the list (constantly t) does to test. 01:58:06 -!- gonzojive_ [~red@rescomp-09-162682.Stanford.EDU] has quit [Quit: gonzojive_] 01:58:32 gonzojive_ [~red@128.12.169.254] has joined #lisp 02:01:47 It is a default keyword value. 02:02:34 Zhivago: I think I figured it out, thank you. 02:02:42 SegFaultAX: an annotation that the default test is a constant t? 02:03:22 -!- Dawgmatix [~dman@c-76-124-9-27.hsd1.nj.comcast.net] has quit [Read error: Connection reset by peer] 02:03:43 Dawgmatix [~dman@c-76-124-9-27.hsd1.nj.comcast.net] has joined #lisp 02:05:08 -!- metasyntax [~taylor@72.86.89.174] has quit [Quit: Be seeing you.] 02:17:27 p_l: Not quite. The default test is a function that always returns t. 02:17:55 right 02:33:18 -!- ephcon [~ephcon@c-24-34-195-64.hsd1.ma.comcast.net] has quit [Ping timeout: 240 seconds] 02:35:47 -!- ef [~ihk@m83-185-36-172.cust.tele2.se] has quit [Quit: leaving] 02:39:43 vmmenon [~vmmenon@c-67-183-8-193.hsd1.wa.comcast.net] has joined #lisp 02:40:10 JonSmith [~jon@ip24-250-13-137.ri.ri.cox.net] has joined #lisp 02:40:26 -!- asarch [~asarch@187.132.134.6] has quit [Remote host closed the connection] 02:40:38 hey i've been having trouble getting swank-sbcl to run on my friend's computer 02:40:51 he's trying to do it under vitualbox running ubuntu 02:41:07 -!- Phoodus [foo@174-17-116-185.phnx.qwest.net] has quit [Ping timeout: 245 seconds] 02:42:15 pretty much he compiles/loads all of swank, and then it tells him that it can't find /dev/tty, and goes into ldb 02:42:31 if i just do meta-x run-lisp sbcl runs fine 02:42:58 i got him the latest off of the slime cvs and the 1.0.37 off of sbcl repo 02:42:58 -!- dfox [~dfox@e80.dkm.cz] has quit [Read error: Operation timed out] 02:43:15 (he's running x86-64) 02:44:51 <_3b`> how do you compile/load swank when it fails? 02:45:14 i was just doing alt-x slime 02:46:24 <_3b`> so it writes a bunch of stuff to *inferior-lisp* buffer in emacs, ending with stuff about /dev/tty and ldb? 02:46:47 yup 02:47:45 i googled it a bit and made sure that /dev/tty existed and then that the permissions were right 02:47:58 it seemed they were... 02:48:19 bigjust` [~user@adsl-074-232-230-165.sip.asm.bellsouth.net] has joined #lisp 02:48:22 very confusing as it was an easy setup on the 3 pcs i have here running it 02:48:31 only difference is virtualbox as far as i can tell 02:48:53 <_3b`> paste contents of *inferior-lisp* to http://paste.lisp.org/new/lisp 02:50:03 troussan [~user@c-24-245-15-191.hsd1.mn.comcast.net] has joined #lisp 02:50:43 -!- bigjust [~user@adsl-074-232-230-165.sip.asm.bellsouth.net] has quit [Ping timeout: 276 seconds] 02:51:19 okay, i'll see if i can get ahold of him to do it 02:53:34 -!- mattrepl [~mattrepl@pool-72-83-118-99.washdc.fios.verizon.net] has quit [Quit: mattrepl] 02:54:06 Avisch [~Avisch@cpe-24-93-16-141.rochester.res.rr.com] has joined #lisp 02:54:21 -!- Avisch [~Avisch@cpe-24-93-16-141.rochester.res.rr.com] has quit [Client Quit] 02:54:41 Avisch [~Avisch@cpe-24-93-16-141.rochester.res.rr.com] has joined #lisp 02:55:41 -!- bgs100 [~ian@unaffiliated/bgs100] has quit [Quit: So long and thanks for all the fish] 02:56:32 xyxxyyy [~xyxu@58.41.0.124] has joined #lisp 02:56:33 curi_ [~curi@c-67-180-28-250.hsd1.ca.comcast.net] has joined #lisp 02:57:08 pizzledizzle [~pizdets@pool-96-250-215-244.nycmny.fios.verizon.net] has joined #lisp 02:57:50 -!- vmmenon [~vmmenon@c-67-183-8-193.hsd1.wa.comcast.net] has quit [Remote host closed the connection] 02:59:12 *udzinari* has to choose between clojure and python for a new project and can't decide which to pick :/ 02:59:52 -!- Dawgmatix [~dman@c-76-124-9-27.hsd1.nj.comcast.net] has quit [Quit: Ex-Chat] 03:00:54 -!- nixeagle [~user@Wikimedia/Nixeagle] has quit [Read error: Connection reset by peer] 03:01:00 -!- xyxxyyy [~xyxu@58.41.0.124] has quit [Ping timeout: 252 seconds] 03:01:47 -!- rtoym [~chatzilla@24.148.170.2] has quit [Remote host closed the connection] 03:02:17 nixeagle [~user@Wikimedia/Nixeagle] has joined #lisp 03:03:19 -!- SandGorgon [~OmNomNomO@122.162.122.2] has quit [Ping timeout: 265 seconds] 03:06:44 -!- gigamonkey [~user@adsl-99-62-73-203.dsl.pltn13.sbcglobal.net] has quit [Ping timeout: 248 seconds] 03:08:32 rtoym [~chatzilla@24.148.170.2] has joined #lisp 03:13:33 xyxxyyy [~xyxu@58.41.0.124] has joined #lisp 03:16:12 tsuru [~user@c-174-50-217-160.hsd1.tn.comcast.net] has joined #lisp 03:19:11 -!- clojay [~clojay@75-119-224-175.dsl.teksavvy.com] has left #lisp 03:19:21 AqD|Home [~aqd_home@122-116-21-207.HINET-IP.hinet.net] has joined #lisp 03:19:47 -!- blackwol` [~blackwolf@ool-45763541.dyn.optonline.net] has quit [Remote host closed the connection] 03:26:46 Snamich [~Snamich@166.133.254.178] has joined #lisp 03:31:01 in some other languages, such as python, it's possible (and, in fact, sometimes necessary) to include C files in 'installation' process; those C files get compiled and the shared objects are included in the installed python library. 03:31:15 is this feasible to do with asdf? 03:31:55 I have a library that is very poorly packaged, and essentially I will have to bundle it with my lisp source and compile it myself on whatever machine I want to use it on. 03:31:56 -!- pizzledizzle [~pizdets@pool-96-250-215-244.nycmny.fios.verizon.net] has quit [] 03:32:19 I'm hoping that people will want to use the lisp interface I'm making for this library. So I want to know what the best practice here would be. 03:32:22 adu [~ajr@softbank220043138128.bbtec.net] has joined #lisp 03:32:46 yes, it's possible. I don't know if there's a particularly good way to do it; Fare probably does. 03:33:24 -!- udzinari [~user@209.158.broadband13.iol.cz] has quit [Remote host closed the connection] 03:33:35 notably, this does NOT appear to be a job for cffi-grovel; there are no .h files, and the library is not already installed 03:34:47 asarch [~asarch@187.132.134.6] has joined #lisp 03:34:53 Fare: ping 03:35:39 -!- marvel [~marvel@unaffiliated/marvel] has quit [Quit: leaving] 03:36:30 pong 03:36:55 inklesspen, it's possible for some value of possible. 03:37:23 well, do you have some other suggestion for this situation? 03:37:36 CFFI has that CFFI-wrapper thing that allows you to include C code to be compiled into your image 03:37:44 I mean, compiled then dynamically linked 03:37:59 interesting 03:38:13 cffi-grovel I think contains that wrapper-file thingie 03:38:17 googling cffi-wrapper doesn't appear to turn up anything specific to this. 03:38:34 there may or may not already be c-source-file thingamajigs that do the same directly 03:39:01 in your cffi checkout, see grovel/asdf.lisp 03:39:18 see how it's used in e.g. iolib/src/iolib.syscalls.asd 03:39:26 R3cur51v3 [~Recursive@rrcs-97-77-55-38.sw.biz.rr.com] has joined #lisp 03:40:11 problem is: there is currently no good delivery mechanism, much less a "portable" one. 03:40:34 The current mechanism relies on your doing a asdf:load-system 03:40:35 It would be nice if the manual (or something) had an example of this. It was quite easy to do this with mk:defsys. 03:41:38 dumping an image may or may not work, depending on your implementation, and even when it works (e.g. with SBCL), it may rely on the .so file being there unmoved and undeleted in the asdf output path (or might it possibly work with LD_LIBRARY_PATH? I dunno) 03:42:23 help appreciated 03:42:45 Fare: Anyway, cmucl now has 2.004. 03:42:53 hm, I don't see the "ffi-wrappers-unix" file referenced at http://gitorious.org/iolib/iolib/blobs/master/src/iolib.syscalls.asd#line27 anywhere in the iolib repo 03:43:00 rtoym: yay. You're the first CL implementation to have it, too :) 03:43:07 (as may be obvious, I'm still just learning lisp) 03:43:30 Fare: Did you see my short request for asdf for cmucl? 03:43:40 inklesspen, it's in the syscalls/ subdirectory 03:43:49 rtoym, not sure. What/where/when? 03:43:52 oh, i see. 03:44:02 http://gitorious.org/iolib/iolib/blobs/master/src/syscalls/ffi-wrappers-unix.lisp <-- but this is lisp code, not C. 03:44:09 -!- Snamich [~Snamich@166.133.254.178] has quit [Read error: Connection reset by peer] 03:44:16 I probably didn't, or forgot about it, or I would have done it for 2.004 03:44:31 Snamich [~Snamich@166.133.254.178] has joined #lisp 03:44:32 Fare: I put ASDF 2.004 into the ccl trunk today, also. 03:44:39 rme: thanks! 03:44:39 Fare: I asked earlier if you could include #+cmu (ext:file-comment "$Header$") in asdf.lisp. But now I'm not sure about it. 03:45:01 rtoym: if you send me a patch, I'll apply it. 03:45:21 Fare: I need to include basically this C file: http://www.openbsd.org/cgi-bin/cvsweb/src/lib/libc/crypt/bcrypt.c?rev=1.24;content-type=text%2Fplain 03:45:35 Fare: Sure. It's a 1-line patch. Where do I send it? 03:45:36 and compile it into a shared library when the package is installed 03:46:41 rtoym, to my email. I'll immediately commit it in 2.1, but the next 2.0 release may wait for quite a few weeks. 03:47:08 heard back from him with the paste 03:47:10 http://paste.lisp.org/display/112456 03:47:27 interesting part is the very bottom i think 03:47:51 inklesspen, if you modify cffi/grovel/asdf.lisp to add a c-source-file type similar to cffi-wrapper-file but without the lisp preprocessing, that would be fine. 03:48:17 udzinari [~user@209.158.broadband13.iol.cz] has joined #lisp 03:55:13 -!- R3cur51v3 [~Recursive@rrcs-97-77-55-38.sw.biz.rr.com] has quit [Read error: Operation timed out] 03:56:24 <_3b`> JonSmith: interesting bit is the set_pseudo_atomic_atomic stuff, not sure what to do about it though 03:56:58 same error with 1.0.40 or not? 03:57:09 <_3b`> unless there is an sbcl low-level hacker around to help debug it, best i can suggest is try upgrading sbcl 03:57:19 okay 03:57:32 JonSmith: I'd try to delete the fasls, and maybe install the latest version, but I don't remember anything this major in the last couple months. 03:57:57 he's running 1.0.37 off the website on x86-64 through virtual box 03:58:05 i've deleted fasls a few times 03:58:06 <_3b`> pkhuong: there is a similar bug on lp for same version on windows 03:58:51 i will try him on 32 bit 1.0.40 tomorrow though 03:59:04 hadn't thought of it 03:59:25 <_3b`> which version of ubuntu and vbox is it? 04:01:37 ubuntu is 10.04, not sure about virtual box 04:03:37 -!- udzinari [~user@209.158.broadband13.iol.cz] has quit [Ping timeout: 245 seconds] 04:04:21 -!- AqD|Home [~aqd_home@122-116-21-207.HINET-IP.hinet.net] has quit [Remote host closed the connection] 04:10:15 well i think he went to bed 04:10:22 thanks for trying 04:10:55 i'll try .40 tomorrow and see what happens 04:13:20 -!- troussan [~user@c-24-245-15-191.hsd1.mn.comcast.net] has quit [Remote host closed the connection] 04:13:24 R3cur51v3 [~Recursive@rrcs-97-77-55-50.sw.biz.rr.com] has joined #lisp 04:15:16 -!- R3cur51v3 [~Recursive@rrcs-97-77-55-50.sw.biz.rr.com] has quit [Excess Flood] 04:17:33 R3cur51v3 [~Recursive@rrcs-97-77-55-50.sw.biz.rr.com] has joined #lisp 04:23:08 -!- davazp [~user@83.55.180.105] has quit [Remote host closed the connection] 04:29:27 -!- Avisch [~Avisch@cpe-24-93-16-141.rochester.res.rr.com] has quit [Quit: This computer has gone to sleep] 04:30:32 Avisch [~Avisch@cpe-24-93-16-141.rochester.res.rr.com] has joined #lisp 04:31:28 -!- redline6561 [~redline@c-66-56-55-169.hsd1.ga.comcast.net] has left #lisp 04:31:33 redline6561 [~redline@c-66-56-55-169.hsd1.ga.comcast.net] has joined #lisp 04:33:48 -!- Avisch [~Avisch@cpe-24-93-16-141.rochester.res.rr.com] has quit [Client Quit] 04:34:02 xyxxyyy1 [~xyxu@58.41.2.149] has joined #lisp 04:34:33 -!- xyxxyyy [~xyxu@58.41.0.124] has quit [Ping timeout: 240 seconds] 04:37:09 Phoodus [foo@174-17-116-185.phnx.qwest.net] has joined #lisp 04:37:15 -!- dstatyvka [ejabberd@pepelaz.jabber.od.ua] has left #lisp 04:43:18 Bobrobyn [~rsmith05@CPE0015e9d40d4f-CM001ac30e9df0.cpe.net.cable.rogers.com] has joined #lisp 04:45:41 -!- nixeagle [~user@Wikimedia/Nixeagle] has quit [Read error: Connection reset by peer] 04:48:03 Good morning everyone! 04:48:29 morning 04:48:30 -!- silenius [~silenius@c-24-130-172-194.hsd1.ca.comcast.net] has quit [Quit: Leaving] 04:50:04 nixeagle [~user@Wikimedia/Nixeagle] has joined #lisp 04:54:08 dialtone [~dialtone@c-67-180-50-184.hsd1.ca.comcast.net] has joined #lisp 04:54:08 -!- dialtone [~dialtone@c-67-180-50-184.hsd1.ca.comcast.net] has quit [Changing host] 04:54:08 dialtone [~dialtone@unaffiliated/dialtone] has joined #lisp 04:56:35 galaxywatcher [~galaxywat@208.20.164.2] has joined #lisp 04:59:09 -!- nixeagle [~user@Wikimedia/Nixeagle] has quit [Read error: Connection reset by peer] 04:59:37 beach: 'morning 05:01:26 nixeagle [~user@Wikimedia/Nixeagle] has joined #lisp 05:11:45 -!- red1ynx [~Dzmitry@91.149.140.201] has quit [Ping timeout: 246 seconds] 05:13:55 OsamaBinWOG [~OsamaBinW@24-177-10-101.dhcp.nwtn.ct.charter.com] has joined #lisp 05:14:54 SandGorgon [~OmNomNomO@122.160.41.129] has joined #lisp 05:15:48 -!- WOG [~OsamaBinW@24-177-10-101.dhcp.nwtn.ct.charter.com] has quit [Ping timeout: 240 seconds] 05:16:10 -!- asarch [~asarch@187.132.134.6] has quit [Quit: Leaving] 05:27:19 -!- lat [~lat@125.167.140.159] has quit [Ping timeout: 252 seconds] 05:28:28 Avisch [~Avisch@cpe-24-93-16-141.rochester.res.rr.com] has joined #lisp 05:29:06 spradnyesh [~pradyus@nat/yahoo/x-arjrswpjpgfngzmf] has joined #lisp 05:29:19 lat [~lat@125.167.140.159] has joined #lisp 05:33:52 Anarch [~olaf@c-67-171-37-107.hsd1.wa.comcast.net] has joined #lisp 05:36:55 -!- nixeagle [~user@Wikimedia/Nixeagle] has quit [Read error: Connection reset by peer] 05:38:25 nixeagle [~user@Wikimedia/Nixeagle] has joined #lisp 05:38:42 Kolyan [~nartamono@93-80-65-136.broadband.corbina.ru] has joined #lisp 05:40:51 rtra [~rtra@unaffiliated/rtra] has joined #lisp 05:46:44 -!- emma [~em@unaffiliated/emma] has quit [Remote host closed the connection] 05:51:10 can someone recommend a good cPRNG, preferably asdf-installable? It seems that common lisp itself doesn't have such a thing. 05:52:19 inklesspen: what about CL:RANDOM ? Doesn't your implementation do what you want? 05:53:05 -!- TR2N [email@89.180.153.197] has left #lisp 05:53:30 milanj [~milanj_@178.223.147.206] has joined #lisp 05:53:30 the hyperspec doesn't say that it's a cryptographically secure PRNG, so I think I must assume that it is not. 05:55:00 Right. 05:55:46 and I need a cryptographically secure one 05:56:07 I thought surely there must be some simple lisp interface over /dev/urandom 05:56:12 but I haven't found it, if there is one 05:56:55 (with-opne-file (rand "/dev/urandom" :element-type '(unsigned-byte 8)) (read-byte rand)) 05:57:00 It should work to open /dev/urandom and just read bytes. 05:57:47 didi [~user@unaffiliated/didi/x-1022147] has joined #lisp 05:57:50 well, I'll use that in the immediate term; it should be portable to anywhere were /dev/urandom exists 05:58:19 inklesspen: you still need to convert the random bytes into your random ranges correctly... 05:58:34 pjb: no, I want random bytes. 05:58:38 ok. 05:58:42 it's for feeding into a C function 06:01:08 <_3b`> beach: don't suppose sicl has code to pull keyword args out of a list of raw args yet? 06:01:37 `/dev/urandom'. Fun stuff. Didn't know. 06:03:37 emma [~em@unaffiliated/emma] has joined #lisp 06:04:07 inklesspen, crypto is hard. 06:04:18 Fare: yes. 06:04:28 are you sure you want to do crypto in lisp? 06:04:37 I am quite sure that I do not. 06:04:38 if so, ironclad needs your help 06:04:46 however, I need a bcrypt library for lisp. 06:04:52 or a FFI to whatever 06:05:03 -!- milanj [~milanj_@178.223.147.206] has quit [Ping timeout: 265 seconds] 06:05:23 (bcrypt is NOT the blowfish cypher; it's a password hashing algorithm based on blowfish, used by OpenBSD and others) 06:05:30 _3b`: you mean like destructuring-bind? 06:05:56 <_3b`> beach: possibly, i haven't quite worked out if that would be directly usable or not 06:06:06 as far as I can tell, nobody has written one yet. I'd happily pay some money for someone to write one, if I know who to ask. but in the meantime, I'm going to try it myself. 06:06:13 _3b`: Either way, I haven't written that yet. 06:06:21 <_3b`> ok 06:06:34 inklesspen: depending on the random number source, making bcrypt with ironclad shouldn't be too hard 06:07:01 <_3b`> once i get keyword args working, i think i can try adding the new cons code from sicl to my flash stuff 06:07:19 p_l: again, bcrypt is not blowfish. I do not feel competent to reimplement bcrypt from primitives; I'm building a FFI to an existing implementation. 06:07:28 last I checked, ironclad does not offer bcrypt 06:07:40 and I didn't see a cPRNG in the list of offerings either 06:07:43 _3b`: Excellent. That module is close to complete. 06:08:55 _3b`: It needs more tests though. 06:09:11 inklesspen: I once read up bcrypt source code, at the heart of each round was a call to blowfish cipher, with some extra operations done at the beginning and at every round 06:09:30 p_l: yes. I don't feel competent to get into it at that level. 06:09:41 <_3b`> beach: probably has more than my code :p 06:09:51 _3b`: Possibly :) 06:11:14 -!- rme [~rme@pool-68-238-7-51.chi.dsl-w.verizon.net] has quit [Quit: rme] 06:12:10 inklesspen: didn't seem that complex... definitelyeasier than trying to implement Skein when you don't understand the notation and try to reverse engineer Java code... 06:12:18 (I gave up with that one) 06:12:32 i don't care; it's not something I'm willing to attempt. 06:13:07 madmuppet [~alexander@203-211-97-90.ue.woosh.co.nz] has joined #lisp 06:13:49 -!- curi_ [~curi@c-67-180-28-250.hsd1.ca.comcast.net] has quit [Quit: This computer has gone to sleep] 06:14:03 WOG [~OsamaBinW@24-177-10-101.dhcp.nwtn.ct.charter.com] has joined #lisp 06:15:48 inklesspen: do you care if its native? 06:15:55 no. 06:16:20 I just care that it's relatively easy to use. 06:17:08 -!- OsamaBinWOG [~OsamaBinW@24-177-10-101.dhcp.nwtn.ct.charter.com] has quit [Ping timeout: 248 seconds] 06:17:39 inklesspen: maybe you would like to talk with nfroyd, the author of ironclad 06:18:02 lhz [~shrekz@c-dba672d5.021-158-73746f34.cust.bredbandsbolaget.se] has joined #lisp 06:18:09 perhaps. 06:18:53 Davse_Bamse [~davse@82.103.143.161] has joined #lisp 06:24:10 -!- SegFaultAX [~SegFaultA@c-98-234-1-162.hsd1.ca.comcast.net] has quit [Quit: Lost terminal] 06:24:14 hmmm... getting it running doesn't look hard 06:26:31 -!- tsuru [~user@c-174-50-217-160.hsd1.tn.comcast.net] has quit [Ping timeout: 276 seconds] 06:27:47 one day job (if not few hours) with cffi 06:28:35 kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has joined #lisp 06:30:47 -!- bozhidar [~user@93-152-185-88.ddns.onlinedirect.bg] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 06:32:12 leo2007 [~leo@cpc1-cmbg13-0-0-cust596.cmbg.cable.ntl.com] has joined #lisp 06:32:56 p_l: what code are you looking at? I've so far found two different C implementations of bcrypt, neither of which is a well-packaged library. 06:33:18 inklesspen: Openwall's 06:33:28 yeah, that's what I'm working with. 06:33:34 this is my first CFFI experience 06:33:52 and the openwall code could stand to be better-documented 06:34:42 -!- neoesque [~neoesque@210.59.147.232] has quit [Ping timeout: 240 seconds] 06:35:06 neoesque [~neoesque@210.59.147.232] has joined #lisp 06:36:32 pdelgallego [~pdelgalle@1503031474.dhcp.dbnet.dk] has joined #lisp 06:36:38 relcomp [~chatzilla@pc92-e03-3.igf.Uni-Osnabrueck.DE] has joined #lisp 06:39:13 -!- WOG [~OsamaBinW@24-177-10-101.dhcp.nwtn.ct.charter.com] has quit [Ping timeout: 276 seconds] 06:39:18 <_3b`> hmm, binding &key vars properly is hard 06:39:51 You mean preserving evaluation order? 06:40:24 <_3b`> evaluation order, binding scope, checking for unknown keywords (or :allow-other-keys), ... 06:40:34 Yeah. 06:40:52 -!- nixeagle [~user@Wikimedia/Nixeagle] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 06:41:11 <_3b`> oh yeah, and duplicate keywords 06:44:16 -!- R3cur51v3 [~Recursive@rrcs-97-77-55-50.sw.biz.rr.com] has quit [Read error: Operation timed out] 06:44:39 That too. 06:44:53 evaluation order is especially troubling with macros 06:45:12 -!- dejones [~dejones@cpe-70-124-77-66.austin.res.rr.com] has quit [Read error: Operation timed out] 06:45:39 dejones [~dejones@cpe-70-124-77-66.austin.res.rr.com] has joined #lisp 06:47:16 -!- jhalogen [~jake@cpe-98-154-250-117.socal.res.rr.com] has quit [Quit: jhalogen] 06:48:32 you need a &rest or &whole parameter to determine the intended order 06:50:17 Athas [~athas@82.211.209.6] has joined #lisp 06:50:27 xavieran [~xavieran@ppp118-209-157-229.lns20.mel6.internode.on.net] has joined #lisp 06:50:34 mrSpec [~Spec@unaffiliated/mrspec] has joined #lisp 06:53:32 -!- adu [~ajr@softbank220043138128.bbtec.net] has quit [Quit: adu] 06:53:40 -!- cmm [~cmm@109.64.200.182] has quit [Ping timeout: 260 seconds] 06:53:43 hi guys Im currently going through common lisp a gentle introduction and I am having a problem understanding exactly what one of the questions is asking for .. the question is how to write the expression (and x y z w) using cond instead of and .. 06:54:08 Well, what does (and x y z w) mean? 06:54:34 Zhivago: and asks if all statements are true 06:54:39 cmm [~cmm@109.64.200.182] has joined #lisp 06:54:46 Now think about it iteratively ... 06:54:52 madmuppet: there is no statement in Lisp. 06:54:55 What happens if x is false? 06:54:59 Stattrav [~Stattrav@117.192.142.26] has joined #lisp 06:55:21 Zhivago: nil is returned 06:55:33 How would you encode that logic using cond? 06:56:33 ((cond (equal x nil) nil) 06:56:48 Why not use the null predicate? 06:56:53 (null x) 06:56:58 ah sorry 06:57:08 Do you understand the problem now? 06:57:55 yes thanks .. I was off by a little bit you have cleared it up somewhat thanks 06:57:56 *_3b`* would use NOT there 06:58:10 Yeah, not is probably a better choice. 06:59:18 wait a min though if x is true cond returns t and exits what happens to the other conditions? 06:59:39 madmuppet: Yes, what? 07:00:13 madmupper: Wrong. 07:00:38 madmuppet: cond only stops if a guard evaluates true. 07:00:38 -!- xyxxyyy1 [~xyxu@58.41.2.149] has quit [Ping timeout: 265 seconds] 07:00:57 Zhivago: yeah brain freeze .. 07:01:10 Zhivago: ... but it never enters an ininite loop ;). 07:01:35 Zhivago: you said "stops". 07:01:43 write it so only last condition is returned as true 07:01:49 relcomp: And? 07:02:01 gonna try that ..:) 07:02:06 -!- kleppari [~spa@bitbucket.is] has quit [Read error: Connection reset by peer] 07:03:19 red1ynx [~Dzmitry@ramonak.bas-net.by] has joined #lisp 07:03:53 p_l: okay, I have a working FFI, except for the lack of a cPRNG 07:03:57 I'll tackle that tomorrow 07:04:20 madmuppet: notice that (and t t t (values 1 2 3)) --> 1; 2; 3. 07:04:25 -!- gz` [gz@clozure-93943513.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has quit [Ping timeout] 07:05:18 -!- gz` [~gz@209-6-40-245.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has quit [Ping timeout: 240 seconds] 07:05:20 xyxxyyy [~xyxu@58.41.3.58] has joined #lisp 07:06:18 kleppari [~spa@bitbucket.is] has joined #lisp 07:08:11 thanks again problem solved .. now same problem using if statements 07:09:03 (macroexpand '(cond ...)) 07:12:14 HG` [~HG@xdslex247.osnanet.de] has joined #lisp 07:13:33 hm, are asdf:oos and asdf:load-op just weird function names, or is single : also some way to interact with packages? 07:13:59 the latter 07:14:08 OliverUv: : is indeed a syntax to qualify the symbols with their package name. 07:14:15 asdf:load-op means 'load-op from package asdf' 07:14:44 (defun asdf-load (&rest systems) (mapcar (lambda (system) (asdf:operate 'asdf:load-op system)) systems)) 07:15:02 (defun asdf-install (&rest systems) (mapcar (lambda (system) (asdf-install:install system)) systems)) 07:17:03 but when I did (defun a) in package X 07:17:20 then (defun a (X:a)) in package Y, it ddn't work 07:17:32 I had to use (X::a) 07:17:34 Sure the syntax is incorrect. 07:17:45 (defun a ()) at least. 07:17:49 <_3b`> : is for symbols exported from the package 07:17:53 Otherwise, X:A is for exported symbols. 07:18:35 well yeah, it was something like (defun almavis (&optional arsalmanacka) (lotsofthings ...)) 07:18:49 ok. 07:18:49 ah, ok, thanks _3b` and pjb 07:19:34 pjb: there's already asdf:load-system 07:20:10 OliverUv: otherwise, once you solved the exported/not exported status of your symbols, defun expressions aren't evaluated in a package (they are _read_ in the current package *package*, but once read, any special relationship with packages is inexistent). 07:20:20 Fare: That's new? 07:23:26 ah, ok 07:23:41 stis [~stis@1-1-1-39a.veo.vs.bostream.se] has joined #lisp 07:23:59 pjb: for some value of "new" 07:24:07 it's been there for quite a while 07:24:18 and it's officially supported in ASDF 2 07:24:28 Great! Thanks. 07:24:49 (hard to give version numbers for ASDF 1) 07:25:16 -!- gonzojive_ [~red@128.12.169.254] has quit [Quit: gonzojive_] 07:26:41 -!- red1ynx [~Dzmitry@ramonak.bas-net.by] has quit [Ping timeout: 240 seconds] 07:29:28 Fare: I think I just found a use case for cl-launch in one of my projects. 07:30:31 bozhidar [~user@212.50.14.187] has joined #lisp 07:30:37 mega1 [~quassel@catv4E5CABA2.pool.t-online.hu] has joined #lisp 07:31:46 antifuchs, woot 07:31:55 Fare: how does it handle *terminal-io* output in sbcl? I was launching it from popen before, and on some errors, sbcl sometimes spams the terminal with backtraces (and then more backtraces, because of more io errors) 07:32:03 my use case was usually testing the same program under multiple implementations. 07:32:09 I was hoping to avoid that. 07:32:18 exactly what I want to do (: 07:32:24 antifuchs, I don't know much about *terminal-io* 07:32:31 I suppose you can #+sbcl something 07:32:49 I think that's to do with --disable-debugger 07:32:59 maybe using that is not such a great idea anyway (: 07:33:03 evening 07:33:18 sellout [~greg@212.234.38.214] has joined #lisp 07:34:53 well, I guess this might just work. yay (: 07:37:42 -!- Fare [~Fare@c-24-218-127-11.hsd1.ma.comcast.net] has quit [Quit: Leaving] 07:40:33 -!- grouzen [~grouzen@195.49.201.58] has quit [Ping timeout: 240 seconds] 07:41:36 hefner [~hefner@ppp-58-9-117-237.revip2.asianet.co.th] has joined #lisp 07:41:44 -!- Dodek [dodek@jest.pro] has quit [Ping timeout: 260 seconds] 07:42:23 grouzen [~grouzen@195.49.201.58] has joined #lisp 07:43:10 -!- didi [~user@unaffiliated/didi/x-1022147] has quit [Ping timeout: 265 seconds] 07:46:25 delYsid [~user@chello084115136207.3.graz.surfer.at] has joined #lisp 07:47:47 xyxxyyy1 [~xyxu@58.41.3.58] has joined #lisp 07:47:51 -!- xyxxyyy [~xyxu@58.41.3.58] has quit [Ping timeout: 246 seconds] 07:49:12 xyxxyyy [~xyxu@58.41.3.58] has joined #lisp 07:52:21 -!- xyxxyyy1 [~xyxu@58.41.3.58] has quit [Ping timeout: 265 seconds] 07:53:24 Taggnostr [~x@dyn57-487.yok.fi] has joined #lisp 07:55:28 rkm1000 [62659c44@gateway/web/freenode/ip.98.101.156.68] has joined #lisp 07:56:18 Dodek [dodek@jest.pro] has joined #lisp 07:56:58 gz [~gz@209-6-40-245.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has joined #lisp 07:57:08 gz_ [~gz@209-6-40-245.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has joined #lisp 07:58:38 -!- Stattrav [~Stattrav@117.192.142.26] has quit [Ping timeout: 265 seconds] 07:58:42 nha [~prefect@imamac13.epfl.ch] has joined #lisp 07:59:47 timor [~timor@port-92-195-135-30.dynamic.qsc.de] has joined #lisp 07:59:55 -!- rkm1000 [62659c44@gateway/web/freenode/ip.98.101.156.68] has quit [Client Quit] 08:05:49 -!- hankhero [~Adium@c-b21fb7c4-74736162.cust.telenor.se] has quit [Read error: Connection reset by peer] 08:05:56 -!- kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has quit [Ping timeout: 248 seconds] 08:05:56 -!- madmuppet [~alexander@203-211-97-90.ue.woosh.co.nz] has quit [Remote host closed the connection] 08:07:57 hankhero [~Adium@c-b21fb7c4-74736162.cust.telenor.se] has joined #lisp 08:15:51 blandest [~user@85.204.33.242] has joined #lisp 08:16:14 -!- xinming [~hyy@115.221.14.235] has quit [Quit: leaving] 08:16:51 Joreji [~thomas@86-129.eduroam.RWTH-Aachen.DE] has joined #lisp 08:22:55 stassats [~stassats@wikipedia/stassats] has joined #lisp 08:24:52 -!- leo2007 [~leo@cpc1-cmbg13-0-0-cust596.cmbg.cable.ntl.com] has quit [Quit: rcirc on GNU Emacs 23.2.10] 08:26:29 kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has joined #lisp 08:28:07 -!- cmm [~cmm@109.64.200.182] has quit [Ping timeout: 265 seconds] 08:28:42 cmm [~cmm@109.64.200.182] has joined #lisp 08:36:25 -!- Avisch [~Avisch@cpe-24-93-16-141.rochester.res.rr.com] has quit [Quit: This computer has gone to sleep] 08:43:32 -!- dmiles_afk [~dmiles@c-67-161-117-56.hsd1.wa.comcast.net] has quit [Read error: Connection reset by peer] 08:44:29 dmiles_afk [~dmiles@c-67-161-117-56.hsd1.wa.comcast.net] has joined #lisp 08:44:33 -!- hugod [~hugod@bas1-montreal50-1279441126.dsl.bell.ca] has quit [Ping timeout: 240 seconds] 08:47:38 -!- dmiles_afk [~dmiles@c-67-161-117-56.hsd1.wa.comcast.net] has quit [Read error: Connection reset by peer] 08:47:43 dmiles_afk [~dmiles@c-67-161-117-56.hsd1.wa.comcast.net] has joined #lisp 08:49:08 -!- kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has quit [Ping timeout: 248 seconds] 08:50:09 hugod [~hugod@bas1-montreal50-1279441534.dsl.bell.ca] has joined #lisp 08:50:48 -!- billitch [~billitch@dslb-088-075-177-102.pools.arcor-ip.net] has quit [Ping timeout: 260 seconds] 08:54:58 attila_lendvai [~attila_le@catv-89-133-171-82.catv.broadband.hu] has joined #lisp 08:55:53 -!- blandest [~user@85.204.33.242] has quit [Ping timeout: 264 seconds] 08:56:07 -!- dmiles_afk [~dmiles@c-67-161-117-56.hsd1.wa.comcast.net] has quit [Ping timeout: 245 seconds] 08:58:56 dmiles_afk [~dmiles@c-67-161-117-56.hsd1.wa.comcast.net] has joined #lisp 09:02:31 gz` [~gz@209-6-40-245.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has joined #lisp 09:02:55 -!- gravicappa [~gravicapp@ppp85-140-117-52.pppoe.mtu-net.ru] has quit [Ping timeout: 265 seconds] 09:04:57 -!- Phoodus [foo@174-17-116-185.phnx.qwest.net] has quit [Read error: Connection reset by peer] 09:06:33 -!- timor [~timor@port-92-195-135-30.dynamic.qsc.de] has quit [Ping timeout: 240 seconds] 09:10:13 killerboy [~mateusz@smrw-91-193-87-5.smrw.lodz.pl] has joined #lisp 09:13:09 Windows newbie , why does the CLISP prompt return nil for probe-file #p"C:\\Users\\Name\\AppData" 09:13:29 kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has joined #lisp 09:13:39 hello lispers 09:14:08 hankhero: CL is not sure if directories are files or not. 09:14:40 No, it works with c:\\Users\\Name 09:14:41 xinming [~hyy@115.221.3.202] has joined #lisp 09:15:08 hankhero: try a \\ at the end. and Zhivago is correct 9: 09:15:29 the cl file system model predates the VFSes of currently popular operating systems 09:16:15 It makes it tricky to make these things work over multiple CL implementations unless you keep the distinction clear. 09:22:22 -!- stis [~stis@1-1-1-39a.veo.vs.bostream.se] has quit [Read error: Connection reset by peer] 09:23:13 timor [~timor@port-92-195-21-179.dynamic.qsc.de] has joined #lisp 09:24:42 billitch [~billitch@e179145115.adsl.alicedsl.de] has joined #lisp 09:28:20 red1ynx [~Dzmitry@ramonak.bas-net.by] has joined #lisp 09:28:24 clisp can probe a file hello.txt in c:\\users but not c:\\users\\name, must be a security thing 09:32:17 -!- Athas [~athas@82.211.209.6] has quit [Remote host closed the connection] 09:32:20 -!- sanjoyd [~sanjoyd@unaffiliated/sanjoyd] has quit [Ping timeout: 258 seconds] 09:32:49 moocow [~new@poco208-2.fredcanhelp.com] has joined #lisp 09:33:51 -!- holycow [~new@poco208-2.fredcanhelp.com] has quit [Ping timeout: 265 seconds] 09:37:30 -!- mbohun [~mbohun@202.124.74.39] has quit [Read error: Connection reset by peer] 09:37:31 ignas [~ignas@78.60.242.102] has joined #lisp 09:38:13 mbohun [~mbohun@202.124.75.27] has joined #lisp 09:39:47 sanjoyd [~sanjoyd@unaffiliated/sanjoyd] has joined #lisp 09:43:53 -!- mbohun [~mbohun@202.124.75.27] has quit [Read error: Connection reset by peer] 09:44:54 mstevens [~mstevens@fsf/member/pdpc.active.mstevens] has joined #lisp 09:47:32 mbohun [~mbohun@202.124.75.27] has joined #lisp 09:50:10 xyxxyyy1 [~xyxu@58.41.3.58] has joined #lisp 09:50:58 -!- red1ynx [~Dzmitry@ramonak.bas-net.by] has quit [Read error: Connection timed out] 09:52:04 -!- xyxxyyy [~xyxu@58.41.3.58] has quit [Ping timeout: 248 seconds] 09:59:41 -!- sanjoyd [~sanjoyd@unaffiliated/sanjoyd] has quit [Read error: Connection reset by peer] 10:04:18 -!- lat [~lat@125.167.140.159] has quit [Ping timeout: 265 seconds] 10:06:45 pjb` [~t@60.Red-80-34-44.staticIP.rima-tde.net] has joined #lisp 10:10:03 -!- jmcphers [~jmcphers@218.185.108.156] has quit [Remote host closed the connection] 10:12:23 yakov [~yakov@mail.stc-spb.ru] has joined #lisp 10:12:29 hey 10:12:48 hi 10:13:14 -!- peddie____ [~peddie@adsl-76-254-67-6.dsl.pltn13.sbcglobal.net] has quit [Quit: peace!] 10:13:21 -!- Joreji [~thomas@86-129.eduroam.RWTH-Aachen.DE] has quit [Ping timeout: 240 seconds] 10:13:27 I own ACL 8.2. Is there a library to talk windows COM (Automation)? 10:13:45 I've googled a bit without result. 10:14:28 -!- kephas [~pierre@AStrasbourg-551-1-17-210.w86-213.abo.wanadoo.fr] has quit [Ping timeout: 248 seconds] 10:14:42 HET2 [~diman@cpc1-cdif12-2-0-cust125.5-1.cable.virginmedia.com] has joined #lisp 10:15:07 yakov: franz support might be able to tell you 10:16:50 -!- DrForr [~drforr@pool-173-58-135-135.lsanca.fios.verizon.net] has quit [Read error: Operation timed out] 10:17:15 sure, I thought maybe we have some open source library. I dont want to go down to C and foreign functions.. 10:19:15 I want to move my python+pywin32 to CL. 10:19:29 DrForr [~drforr@pool-173-58-135-135.lsanca.fios.verizon.net] has joined #lisp 10:19:34 yakov: you might want to look at http://www.weitz.de/rdnzl/ 10:21:02 Athas [~athas@shop3.diku.dk] has joined #lisp 10:21:34 cinch [~cinch@unaffiliated/cinch] has joined #lisp 10:22:01 lat [~lat@125.167.140.159] has joined #lisp 10:23:24 xan_ [~xan@host86-189-10-226.range86-189.btcentralplus.com] has joined #lisp 10:26:25 udzinari [~user@nat/ibm/x-estmzmopzqakblpj] has joined #lisp 10:27:15 -!- lat [~lat@125.167.140.159] has quit [Quit: Ex-Chat] 10:29:39 *cinch* is happy to start learning common lisp *it's about time* :) 10:30:28 -!- mstevens [~mstevens@fsf/member/pdpc.active.mstevens] has quit [Quit: leaving] 10:32:05 Yuuhi [benni@p5483BB48.dip.t-dialin.net] has joined #lisp 10:33:50 Komi [Komi@62.32.133.228] has joined #lisp 10:33:51 -!- mbohun [~mbohun@202.124.75.27] has quit [Quit: Leaving] 10:34:26 yakov: http://www.franz.com/support/documentation/current/doc/ole.htm 10:35:39 Joreji [~thomas@86-129.eduroam.RWTH-Aachen.DE] has joined #lisp 10:35:40 also, http://code.google.com/p/cl-win32ole/ -- which I haven't used, but which looks promising 10:38:59 dabd [~dabd@a81-84-79-168.cpe.netcabo.pt] has joined #lisp 10:40:36 mstevens [~mstevens@fsf/member/pdpc.active.mstevens] has joined #lisp 10:40:47 oh, thanks lichtblau! cl-win32ole looks like what I need 10:42:05 if i have , '('(f x) '(g y)) , is there a way to make it => '(f x) '(g y) ? 10:42:28 the original list can vary on length .... 10:42:55 why do you have quotes inside quotes? 10:45:06 stassats, because the macro gives me an error of 'f' or 'g' being a function name... 10:45:28 not being a function name* 10:45:51 Got a program developed on mac/ccl to work on windows/sbcl, including ltk gui :-) Well most of it.. 10:46:08 araujo: looks like you're doing something you don't quite understand 10:46:24 araujo: it is called destructive splicing 10:46:48 destructive? 10:47:04 what does it destroy? 10:47:10 yes because you insert the list in the surrounding one 10:48:06 uh no i'm mixing up destructive splicing is ,. 10:48:49 araujo: you want to discover splicing unquote : ,@ 10:48:55 inside a backquote 10:49:24 billitch, yes, i tried that one .. wondering if there existed other way to do it 10:49:26 does anyone know what araujo is actually trying to do? 10:49:29 `(a b ,@'(c d)) => '(a b c d) 10:51:05 but it depends on what you mean by just '(f x) '(g y), i assume you want to insert them in a surrounding list ? 10:51:18 araujo: paste your code 10:53:05 kephas [~pierre@AStrasbourg-551-1-11-119.w86-213.abo.wanadoo.fr] has joined #lisp 10:55:13 -!- nha [~prefect@imamac13.epfl.ch] has quit [Ping timeout: 260 seconds] 10:59:15 hefner: stimulate discussion 11:00:01 dfox [~dfox@r3ah113.net.upc.cz] has joined #lisp 11:01:06 *Xach* had laser spigot nostalgia today 11:01:56 Xach: did you ever play it? 11:02:59 hefner: No, but the screenshots were lickable 11:03:07 -!- yakov [~yakov@mail.stc-spb.ru] has quit [Ping timeout: 276 seconds] 11:03:38 *hefner* wonders how hard it would be to package up a mac build 11:03:40 marvel [~user@unaffiliated/marvel] has joined #lisp 11:05:39 *Xach* is mac-bound 11:05:59 me too! 11:06:00 hefner: dto's strategy, whatever it is, seems to work nicely. 11:06:29 -!- mstevens [~mstevens@fsf/member/pdpc.active.mstevens] has quit [Quit: leaving] 11:09:13 I need to figure out how to make SDL 1.2 work with ECL (probably easy enough, but SDL 1.3 worked with no effort, so I use it, but it's a unfinished, broken, misguided piece of garbage), figure out how to build everything 32-bit and targetting 10.4 or 10.5, and figure out how to combine everything into a nice application bundle 11:10:04 building windows binaries is much easier! 11:13:25 -!- dabd [~dabd@a81-84-79-168.cpe.netcabo.pt] has quit [Ping timeout: 265 seconds] 11:14:11 my current thinking is that someone should drop a nuke on SDL and most of its users 11:16:19 hefner: why use sdl then ? 11:17:18 stipet [~user@ua.blixtvik.net] has joined #lisp 11:19:08 tsuru [~user@c-174-50-217-160.hsd1.tn.comcast.net] has joined #lisp 11:19:36 -!- tsuru is now known as Guest30350 11:20:19 -!- Guest30350 is now known as tsuru` 11:20:26 -!- tsuru` is now known as tsuru- 11:21:04 that's what i thought.. 11:21:11 oh, fuck off 11:21:21 -!- SandGorgon [~OmNomNomO@122.160.41.129] has quit [Ping timeout: 240 seconds] 11:21:41 actually i'm curious, i never saw sdl with lisp 11:21:41 it's a handy portability layer, provided your app doesn't ask much of the window system 11:22:50 the most important thing about SDL is that there's not a lot of mature 2d graphic library 11:22:53 -!- xyxxyyy1 [~xyxu@58.41.3.58] has quit [Quit: Leaving.] 11:23:17 SDL works on a lot of platform, and handles a lot of tricky stuff 11:23:23 hm i'm not much into it anymore but i remember 2d libraries far more advanced than sdl 11:24:09 Like, say, Postscript? 8) 11:24:10 by 2d graphic I mean, for graphic applications 11:24:27 there's far more advanced drawing library (imlib2 typically) 11:24:57 _KY_ [YKY@unaffiliated/-ky-/x-0649748] has joined #lisp 11:25:01 *Lycurgus* wonders what computing would be like with the more or pointless duplications of effort squeezed out. 11:25:01 oh i was thinking about texture loading, generation and manipulation 11:25:14 I see two legitimate ways to use SDL - as a dumb framebuffer, where it gives me a pointer and I do all the drawing myself, or as a way of opening a window with an OpenGL context, where again I do everything myself through OpenGL 11:25:42 <_KY_> Where can I find the list of functions supported by ABCL? 11:25:43 well you can use a drawing library like imlib2, and use SDL as a graphic driver 11:26:10 there seems to have evolved a bastard middle ground starting from the couple functions in the original SDL to blit a surface into the framebuffer, which were foolishly kludged to also work in OpenGL mode, and it's all downhill from there 11:26:27 you can also achieve that with http://projects.gnome.org/gtkglext/ 11:27:16 that would be more integrated with the window system 11:27:36 but you need gtk which is a far more annoying dependency 11:27:58 i don't know about gtk in cl 11:28:11 sdl is not perfect, but I believe it's the lesser evil for graphic applications (2d games) 11:28:15 billitch: well, another thing about SDL is that it's small, self-contained library I can tote alongside a deployed executable, and in the situations where I'd use it, I don't need another more sophisticated than a fullscreen window to draw in 11:29:16 billitch: which reminds me, I also appreciate it for letting me output sound portably 11:30:11 and input event queues, right 11:30:22 billitch: if I used Gtk, I'd be back in the situation of wanting or needing to use something different on Windows and OS X 11:30:54 gtk is not so well integrated in quartz yet, it sucks 11:31:09 but las time i checked it was pretty close 11:31:40 -!- bigjust` [~user@adsl-074-232-230-165.sip.asm.bellsouth.net] has quit [Remote host closed the connection] 11:31:45 adu [~ajr@softbank220043138128.bbtec.net] has joined #lisp 11:32:28 nha_ [~prefect@imamac13.epfl.ch] has joined #lisp 11:32:48 -!- adu [~ajr@softbank220043138128.bbtec.net] has quit [Client Quit] 11:35:06 -!- nha_ is now known as nha 11:37:55 all these interfaces seem so ad-hoc when you try to imagine portable layers for audio and opengl in win32/osx/linux/bsd 11:38:14 also i always wondered if the GC could not be a problem with 'realtime' audio and video rendering ? 11:39:07 morphling [~stefan@gssn-5f75621c.pool.mediaWays.net] has joined #lisp 11:39:12 i remember it used to be a problem in ocaml 11:40:44 for my ECL hack I mix audio in C (necessary given how SDL's audio API works anyway, and ECL is or was too slow for it anyway), and live with the occasional dropped frame due to the GC (it runs much smoother than I'd have imagined, although I learned too late that ECL doesn't have a generational GC, so the GC-related stutters have gotten more noticeable as my app has grown) 11:42:22 mbohun [~mbohun@ppp115-156.static.internode.on.net] has joined #lisp 11:42:41 -!- Jasko [~tjasko@c-174-59-223-208.hsd1.pa.comcast.net] has quit [Remote host closed the connection] 11:44:07 i dont know much about GC but in D you can disable it when needed, it sounded quite cool 11:45:08 for Shuffletron and random audio hackery using SBCL, I generate audio from lisp, which works fine because the underlying interface there (through both ALSA and libao) is that you push audio at your own pace, and so long as you keep buffered far enough ahead (at least 100ms, more if you want to run in the background glitch-free while other apps stress the CPU), there's no glitches. 11:46:09 yakov [~yakov@mail.stc-spb.ru] has joined #lisp 11:46:30 that said, GCs still introduce enough fear, uncertainty, and doubt in my mind that my new approach for anything action-intensive is just to use C++ instead. 11:46:31 Shuffletron ? =) 11:46:39 minion, shuffletron? 11:46:40 shuffletron: Shuffletron is an MP3 music player running on Linux and Mac OS X systems (and possibly elsewhere). http://www.cliki.net/shuffletron 11:47:07 varjag [~eugene@122.62-97-226.bkkb.no] has joined #lisp 11:48:55 For the last couple days I've been evaluating this strategy by rewriting in C++ a little game that I originally wrote a couple years ago as an exercise to learn Factor. I'll be interested to see how the code comes out, and whether I can keep my sanity using C++. 11:49:25 athlon [~user@94.179.207.86.pool.3g.utel.ua] has joined #lisp 11:49:29 hi 11:49:36 i realize that i totally lost interest into c++... 11:50:18 how to collect atoms from nested list? 11:50:21 why not plain C for the non-gc parts ? 11:50:43 billitch: sorry, I don't follow. non-gc parts? 11:50:53 -!- mega1 [~quassel@catv4E5CABA2.pool.t-online.hu] has quit [Ping timeout: 240 seconds] 11:52:02 mega1 [~quassel@pool-03c9a.externet.hu] has joined #lisp 11:52:43 I have a list like '(1 (2 3)) 11:53:01 I need to get '(1 2 3) 11:53:05 Any idea? 11:53:54 athlon: http://en.wikipedia.org/wiki/Depth-first_search or http://en.wikipedia.org/wiki/Breadth-first_search 11:53:55 i think there's a CL function called flatten or something like that 11:53:55 you need a flatten function 11:54:19 galdor: hi5 11:54:23 :-) 11:54:27 athlon: http://rosettacode.org/wiki/Flatten_a_list#Common_Lisp 11:54:34 :) 11:54:46 athlon: you write a recursive function that walks through the tree and collect the atoms. you could start by writing a function that walks through and prints each atom in the tree, then modify it to collect and return them instead of printing them. 11:54:52 -!- carrl [~carl@61-64-164-206-adsl-tai.STATIC.so-net.net.tw] has quit [Ping timeout: 245 seconds] 11:55:46 carrl [~carl@61-64-164-206-adsl-tai.STATIC.so-net.net.tw] has joined #lisp 11:57:19 that's dfs, when you have a cons you recursively descend into car and cdr, when you have an atom, you push it to your result 11:58:27 hefner: why not have a 'realtime' thread in C and pass messages from CL ? 11:58:47 i guess the proper solution would be to allow to disable GC in specific threads 11:59:04 or at arbitrary points 11:59:21 but i don't know the implications of this 11:59:28 blandest [~user@85.204.33.242] has joined #lisp 11:59:37 -!- hlavaty [~user@77-22-102-45-dynip.superkabel.de] has quit [Remote host closed the connection] 11:59:37 for what purpose? generating audio? 12:00:45 i don't know, but if you think about c++... 12:02:01 xyxxyyy [~xyxu@116.227.164.71] has joined #lisp 12:02:50 well, you say "the proper solution"... so, the proper solution to which problem? 12:03:22 -!- nha [~prefect@imamac13.epfl.ch] has quit [Ping timeout: 258 seconds] 12:06:07 -!- stipet [~user@ua.blixtvik.net] has left #lisp 12:06:21 can i write my_flatten in recursive style only using cons function? 12:06:39 hypno [~hypno@impulse2.gothiaso.com] has joined #lisp 12:07:28 -!- rtra [~rtra@unaffiliated/rtra] has quit [Ping timeout: 276 seconds] 12:07:34 to the problem of GC stopping all threads and inducing latencies 12:08:57 hmm. anyone know if there are some framework or lib (whatever really) that can generate "reports" from a SQL database? say output a pdf file or even text file with metadata and data, etc? 12:09:29 tayloj [~tayloj@cpe-67-248-22-159.nycap.res.rr.com] has joined #lisp 12:09:34 hm, I'm using (load "./otherfile.cl") to include other files, is this correct behaviour? 12:09:48 SBCL seems to have stopped understanding that I created a package 12:10:03 minion: xach-asdf 12:10:04 xach-asdf: Xach's article "Making a small Common Lisp project" can be found at http://xach.livejournal.com/130040.html 12:10:12 thanks! 12:10:17 -!- killerboy [~mateusz@smrw-91-193-87-5.smrw.lodz.pl] has quit [Ping timeout: 245 seconds] 12:10:32 i'm just getting started with packages and such, which weren't covered in the course I went 12:10:35 big thanks 12:10:49 athlon: you have to go through each node of the lists. if you look closely nested lists are actually a tree. 12:10:52 oh lol 12:10:59 my ISP is blocking livejournal 12:11:14 OliverUv: Also read http://weitz.de/packages.html 12:11:21 google cache it is 12:11:37 OliverUv: boo 12:11:49 *Xach* should mirror that post on xach.com (if that's not blocked) 12:11:58 there ought to be a counterpart to that explaining how to do small hacks/scripts with multiple source files or library dependencies, but without using ASDF 12:12:13 Xach: no prob, I've got access through google cache :) 12:12:27 no no, xach.com shouldn't be blocked 12:13:11 i'm living in my gf's parents' old apartment, in the professors' village by a taiwanese uni campus 12:13:21 (which, I suppose, would mostly consist of explaining the difference between read/compile/load/execute times, and how/when to wrap things with eval-when) 12:13:22 aparently they don't want the professors to look at myspace, or porn 12:14:05 athlon: lists and trees are very simple (i mean abstract), recursive structures, so you should definetely write a simple, recursive function 12:14:39 Xach: actually, could you mirror that? For some reason the google cache version has shoved some of the text beyond the left side of the screen 12:14:57 OliverUv: will do. 12:15:00 -!- manic12 [~andrew@99-100-67-123.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 248 seconds] 12:15:09 Stattrav [~Stattrav@117.192.149.38] has joined #lisp 12:15:23 -!- gz_ [~gz@209-6-40-245.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has left #lisp 12:15:45 billitch: actually I would normally recommend to use existing library functions where they are available, since that would give one access to future fixes and performance improvements 12:15:50 mcclim is doing something strange when changing command tables 12:16:08 sometimes it's working, sometimes it's not 12:16:16 stassats: how so? 12:16:20 but i'm guessing athlon is just learning lisp, so then I'd also recommend writing a recursive function 12:16:31 OliverUv: yes but isn't it always a wonderful exercice to parse a tree ? 12:16:40 indeed it is 12:16:46 =D 12:17:04 hefner: i do (setf (frame-command-table frame) 'name), but i not always get additional commands in the interactor 12:17:34 i used to love it, but for some reason lost interest with red black trees.. 12:17:47 adv trees are so confusing at first 12:17:50 lazy me 12:17:58 but when you finally grasp them, it is wonderful 12:18:21 ah! avl tree 12:18:22 sorry 12:18:34 ttp://www.xach.com/lisp/asdf-tutorial/ work ok? 12:18:50 http://www.xach.com/lisp/asdf-tutorial/ - sorry 12:18:58 tea transfer protocol? 12:19:01 http://en.wikipedia.org/wiki/AVL_tree my very heart trembles at its algorithmic optimality 12:19:14 Xach: heh, nope, same problem 12:19:42 stassats: I don't think I've ever tried that, nor considered the possibility that someone would do it. Still, I'd expect it to work provided it's done at the right point in the command loop (while executing a command), as opposed to some random time asynchronous with respect to the command loop, such as a gadget callback. Is that a pattern to when it works, when it doesn't work? 12:19:44 clearly, they're blocking anything with "xach" in it 12:19:52 Xach: getting the same problem in both FF and Chrome 12:20:12 "is *there* a pattern to...", rather 12:20:20 stassats: I got past the blocking long ago, now we're having layout problems 12:20:24 hefner: it's done during value-changed-callback 12:20:37 OliverUv: hang on, let me try something. 12:20:39 AIUI, callbacks happen outside the command loop 12:20:47 so it's not in the right point at all (-; 12:20:56 so, what do i do? 12:20:59 yeah, that won't work until the next command read begins 12:21:30 OliverUv: I ripped out all the stylesheet stuff. does that change the situation at all? 12:21:32 you could emulate a command by throwing to the command loop, I think 12:22:00 Xach: yep, looks good now 12:22:57 ok, at least i understand why it works sometimes and why it doesn't 12:24:37 abugosh [~Adium@pool-74-103-38-231.bltmmd.fios.verizon.net] has joined #lisp 12:25:02 so many good solutions come to mind! "Don't use commands" is one. "Don't use callbacks" the next. 12:25:04 -!- udzinari [~user@nat/ibm/x-estmzmopzqakblpj] has quit [Read error: Connection reset by peer] 12:25:29 I'm a big fan of "don't use commands" 12:25:37 i see "don't use mcclim" somewhere there 12:25:49 oh, I like that one too 12:26:07 what I'd do is invoke the restart that, er, restarts the command loop. it's really there to recover from errors, but you could use it here too 12:27:07 i like commands because i don't like too much buttons or menus 12:27:48 -!- Beetny` [~Beetny@ppp118-208-34-80.lns20.bne1.internode.on.net] has quit [Ping timeout: 260 seconds] 12:28:22 I think antifuchs suggests you could throw a command presentation, provided there's a command to invoke that makes more sense for whatever you're trying to do, but maybe slightly failure prone in the case where the user already entered a command name and it's now accepting arguments, and the argument type happens to be a supertype of command 12:28:46 stassats: I like commands for that reason too, sometimes, but they don't always mix gracefully with buttons and menus 12:29:03 hm, i'm getting too much 'There is no applicable method for the generic function: CLIM:RECTANGLE-EDGES* when called with arguments: (NIL)' 12:29:16 and it goes away after "Clear the output history of the pane and reattempt forceful redisplay" 12:29:17 Jasko [~tjasko@209.74.44.225] has joined #lisp 12:30:23 another thing you might do is put all your commands in one table and try enabling/disabling them from the callback. that might work (I can't say for certain without reading the code). 12:31:34 if that's too kludgey you could maybe keep your current arrangement of command tables, but include them all into one master command table, and then programmatically walk through and disable all the command tables but the one you want commands from, and enable ones you want commands from 12:32:05 i can put my changes inside a display-function and set pane-needs-redisplay to t 12:32:09 -!- abugosh [~Adium@pool-74-103-38-231.bltmmd.fios.verizon.net] has quit [Quit: Leaving.] 12:32:56 can you explain why you want to do this? 12:33:14 redisplay is done inside a command loop? 12:34:38 at the head of the command loop, panes needing redisplay are redisplayed. it goes something like "(re)display, read command, execute command" 12:35:20 I mean, is the value changed callback attached to something like a set of radio buttons that change the mode of the application to do different tasks? 12:35:40 -!- tayloj [~tayloj@cpe-67-248-22-159.nycap.res.rr.com] has quit [Quit: Leaving.] 12:35:46 yeah, it's an option-pane 12:36:32 Xach: you're doing (defpackage #:stumpgrinder ..) 12:36:39 why the # _ 12:36:41 ? 12:36:42 so conceptually that choice could correspond to "select foo mode" / "select bar mode" commands, or "select mode (mode) FOO/BAR" 12:36:57 and i want to have different commands which output to different application panes 12:37:35 hefner: something like that 12:37:37 SandGorgon [~OmNomNomO@122.162.127.184] has joined #lisp 12:38:12 i can do without option-pane, obviously, it isn't critical 12:38:42 i can't even use option-pane with mcclim-freetype, it crashes my X server 12:38:53 boo, hiss 12:38:55 fix your X server. 12:39:12 well, i'm lazy so i disabled mcclim-freetype 12:41:07 clim execute-frame-command 12:41:07 http://bauhh.dyndns.org:8000/clim-spec/28-4.html#_1536 12:41:30 it's still down 12:42:24 (defun my_flatten (l) 12:42:24 (cond ((null l) nil) 12:42:24 ((atom (car l)) (cons (car l) (my_flatten (cdr l)))) 12:42:24 (t (append (my_flatten (car l)) (my_flatten (cdr l)))))) 12:42:37 athlon: don't paste to the channel 12:42:41 sorry 12:42:43 use paste.lisp.org 12:42:50 i know 12:43:05 Xach: thanks a lot for writing all of that up, the wetiz.de site told me what my problem was (at the bottom of the page) and you give a good recipe for a proper system definition, thanks 12:43:18 *Xach* lives to serve 12:43:46 namo amitabha 12:44:26 http://paste.lisp.org/+2ES0 12:44:37 Is it OK function? 12:45:19 athlon: that looks ok 12:45:48 athlon: try: (flatten '(a . b)) and (flatten 'x) 12:46:01 anyway, I wonder if execute-frame-command should restart the command loop if it were called under these circumstances, or if it's strictly intended as a glorified version of APPLY. The spec doesn't address it. 12:46:47 zard1989 [~user@Joe.m4.ntu.edu.tw] has joined #lisp 12:47:05 athlon: we usually separate words with hyphens, not underscores; i.e. my-flatten. As relcomp pointed out, your code doesn't handle atoms; have you even tested it? And, you can improve the runtime complexity of your function a lot. 12:48:05 -!- Komi [Komi@62.32.133.228] has quit [Ping timeout: 264 seconds] 12:48:08 i test it 12:48:22 it imho ok 12:48:29 asarch [~asarch@189.188.141.209] has joined #lisp 12:48:51 athlon: What does (my_flatten '(a . b)) return? 12:48:53 in my idealized notion of how things are supposed to work, stassats' callback should be able to simple use execute-frame-command to switch the application mode. This would be fairly trivial to hack into mcclim too, I think, just modifying how the execute-command-event is handled. 12:50:05 (a.b) 12:50:14 tayloj [~tayloj@cpe-24-92-63-169.nycap.res.rr.com] has joined #lisp 12:50:18 MAYBE execute-frame-command works in mcclim semi-asynchronously 12:50:29 athlon: try (my_flatten '(a . b)) 12:50:41 (with spaces) 12:50:54 athlon: or (flatten 'atom). 12:51:43 ziarkaen [~ziarkaen@87.112.72.23.plusnet.ptn-ag2.dyn.plus.net] has joined #lisp 12:51:49 -!- yakov [~yakov@mail.stc-spb.ru] has left #lisp 12:51:52 rajesh [~rajesh@nylug/member/rajesh] has joined #lisp 12:51:57 i'll go with changing through a command for now 12:53:51 antifuchs: we've done part of the work, queuing up the execute-command-event when you call e-f-c from outside of the application, so that you can remotely control a clim app, so to speak. I'm mistaken about wanting to do this from that event handler, but I think it'd still be an easy thing to make work. 12:54:38 -!- neoesque [~neoesque@210.59.147.232] has quit [Remote host closed the connection] 12:55:16 rrice [~rrice@adsl-99-164-35-250.dsl.akrnoh.sbcglobal.net] has joined #lisp 12:56:30 -!- dfox [~dfox@r3ah113.net.upc.cz] has quit [Ping timeout: 252 seconds] 12:57:01 Odin-` [~sbkhh@157.157.93.22] has joined #lisp 12:58:33 -!- cmm [~cmm@109.64.200.182] has quit [Ping timeout: 240 seconds] 12:59:00 -!- pjb` [~t@60.Red-80-34-44.staticIP.rima-tde.net] has quit [Ping timeout: 246 seconds] 12:59:11 antifuchs: essentially combining that check with what you first suggested, so long as we can ensure the thrown presentation is returned by read-frame-command and not consumed by some nested call to accept 12:59:34 cmm [~cmm@bzq-109-64-200-182.red.bezeqint.net] has joined #lisp 12:59:52 -!- Stattrav [~Stattrav@117.192.149.38] has quit [Ping timeout: 245 seconds] 13:00:17 *hefner* now remembers why hacking on mcclim was so fun - the design presented by the spec is only half finished, and we get to figure out all the rest of these details :) 13:00:41 indeed ((: 13:00:49 :C 13:01:04 legumbre_ [~leo@r190-135-66-118.dialup.adsl.anteldata.net.uy] has joined #lisp 13:01:11 maybe that's why it's not so fun to write mcclim applications 13:01:42 it's certainly more fun to hack on mcclim than to hack on a mcclim application (: 13:01:55 for one, you don't always have a spec to look at (: 13:02:16 I always wanted to ask if there is a mcclim tutorial. Seems like there isn't any? 13:02:50 there are some, http://mcclim.cliki.net/Documentation 13:03:19 Stattrav [~Stattrav@117.192.157.179] has joined #lisp 13:03:22 -!- legumbre [~leo@r190-135-40-14.dialup.adsl.anteldata.net.uy] has quit [Ping timeout: 276 seconds] 13:03:22 -!- Stattrav [~Stattrav@117.192.157.179] has quit [Client Quit] 13:04:40 -!- redline6561 [~redline@c-66-56-55-169.hsd1.ga.comcast.net] has quit [Ping timeout: 276 seconds] 13:04:53 OH! Looks nice. Thank you. I got stuck in the spec which is as far from a tutorial as the HS :) 13:05:20 see also Examples directory 13:05:33 OK. 13:06:20 -!- rajesh [~rajesh@nylug/member/rajesh] has quit [Quit: leaving] 13:06:35 also, gilbert's presentation from a few years back is short but great at explaining what makes clim fun 13:06:40 y'all know if my app will work well on AllegroCL's CLIM implementation if I make it work on McClim? 13:07:13 unfortunately, his web server is online only when he is at home 13:07:23 OliverUv: noooot necessarily. 13:07:36 shit 13:07:45 most parts will be similar, but these are different interpretations of an ambiguous spec 13:07:53 mm 13:08:10 why not just run McCLIM on allegro? 13:08:20 OliverUv: *beep* (from censorbot) :) 13:08:24 why not just use allegro clim? (-: 13:08:27 *stassats* dreams of CLIM-3 13:08:35 *antifuchs* has nightmares about clim3 (: 13:08:39 antifuchs: because I don't know how to hack on that =p 13:08:47 hefner: I do! ((-: 13:09:20 aah. this particular distribution of knowledge is actually a very good reason for you not to use allegro clim (: 13:09:24 antifuchs: can't get allegro clim on my computer (i think, you have to pay for that right?) 13:09:32 yeah, it costs money 13:09:46 at the moment I'm telecommuting around the globe (Taiwan-Sweden) 13:10:16 and the latency is too big for me to X11 forward through comfortably 13:10:45 I'm curious, have you tried nx? 13:11:00 (or the open-source reimplementation or whatever it is) 13:11:01 dlowe [~dlowe@ita4fw1.itasoftware.com] has joined #lisp 13:12:38 -!- ziarkaen [~ziarkaen@87.112.72.23.plusnet.ptn-ag2.dyn.plus.net] has quit [Remote host closed the connection] 13:14:21 I think McCLIM suffers from CLIM itself being too ambitious. If you ignored the command-based app framework (a leaky abstraction if ever there was one) and sold McCLIM as a (half-finished) toolkit-style alternative to Gtk or Qt with a nifty CLOS-based graphics canvas supporting presentations as one of its chief features, it'd probably be more successful - much easier to learn, easier to hack on and improve, less likely to baffle or disap 13:15:51 to make this work well you'd have to make a few changes to make presentations more useful and accessible from an event-driven app, and whatever low-hanging fruit there is to make event-driven GUIs less painful to code 13:16:28 Just curious, the commad-based app framework - how is that a leaky abstraction? I haven't used the command-based app framework, but read a bit about the leaky abstraction and didn't realy "get it" from my first read. 13:17:35 mstevens [~mstevens@fsf/member/pdpc.active.mstevens] has joined #lisp 13:18:07 -!- billitch [~billitch@e179145115.adsl.alicedsl.de] has quit [Ping timeout: 265 seconds] 13:18:12 (at this point you'd have a solid base build on simple, proven concepts, and set about reinventing a "CLIM 3.0" app framework as a sort of research project on top of that) 13:18:46 -!- Odin-` [~sbkhh@157.157.93.22] has quit [Quit: Meep.] 13:18:56 billitch [~billitch@e179145115.adsl.alicedsl.de] has joined #lisp 13:19:20 LiamH [~none@pdp8.nrl.navy.mil] has joined #lisp 13:21:20 btbngr [~btbgnr@smorge2.force9.co.uk] has joined #lisp 13:21:30 -!- tayloj [~tayloj@cpe-24-92-63-169.nycap.res.rr.com] has quit [Quit: Leaving.] 13:21:40 -!- kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has quit [Ping timeout: 248 seconds] 13:21:57 Xach: I don't get the reason you have the # in front of package names (eg (asdf:defsystem #:stumpgrinder :depends-on (#:cl-ppcre) :components ((:file "stumpgrinder"))) 13:21:59 TDT: it imposes constraints that all too often don't fit what people actually want to build, and don't integrate well (or at all) with other features in CLIM. it seems to me that it's a special case suited for certain kinds of apps, and shouldn't be what the entire system is organized around. if it isn't "leaky", it's certainly under-engineered. 13:22:42 mattrepl [~mattrepl@pool-72-83-118-99.washdc.fios.verizon.net] has joined #lisp 13:22:56 OliverUv: in those places, the symbol is only used for the string it designates 13:23:12 OliverUv: you could also use "STUMPGRINDER" or :stumpgrinder or stumpgrinder 13:23:15 the way that the climacs application base (I forgot the name) massages the command framework is a very interesting approach, I think 13:23:17 OliverUv: I prefer #:stumpgrinder 13:23:21 make every app into an emacs 13:23:22 ah ok 13:23:25 -!- billitch [~billitch@e179145115.adsl.alicedsl.de] has quit [Ping timeout: 265 seconds] 13:23:31 fe[nl]ix: would you consider making a soft link of static-vectors.asd to the top-level directory so that clbuild will see it? 13:23:31 so # works like "" when there are no spaces? 13:23:31 but I'm not sure this is suited to modern applications anymore 13:23:35 -!- morphling [~stefan@gssn-5f75621c.pool.mediaWays.net] has quit [Remote host closed the connection] 13:23:43 OliverUv: not at all. 13:23:46 or does it eval :stumpgrinder into the symbol STUMPGRINDER 13:23:48 antifuchs: Drei? 13:24:02 TDT: if it were really as all-powerful as it needs to be to justify its position at the top level of your application, I think it should be possible to build graphical gadgets with it, for instance, and it just isn't unless you want to seriously compromise usability for idealogical purity. 13:24:05 OliverUv: #:foo is the syntax for an uninterned symbol with a symbol-name of (probably) "FOO" 13:24:11 I'm only used to seeing it when refering to functions 13:24:14 ah 13:24:15 ok 13:24:17 beach: yes, drei. thanks (: 13:24:23 dabd [~dabd@gtkdbio.inesc-id.pt] has joined #lisp 13:24:29 I didn't know you could manually specify interning of symbols like that 13:24:31 OliverUv: # is used before a second character to perform some standard read macros. 13:24:36 OliverUv: By using #:stumpgrinder you make sure that you are independent of how the reader alters case, and you don't pollute any packages with useless symbols. 13:24:42 beach: ESA, surely. 13:24:48 CL defines #:, #', #A, and some others 13:24:52 hefner: Ah, I don't know much about mcclim to comment any which way on that. 13:24:54 yes, ESA. I /really/ couldn't remember the name (: 13:24:55 very nice, I'll also go for #: then 13:24:58 hefner: Right, all that is in ESA. 13:25:02 Komi [Komi@62.32.147.35] has joined #lisp 13:25:23 OliverUv: but the #: syntax is not related to the #' syntax in any other way 13:25:30 -!- tcr [~tcr@115.132.79.190] has quit [Quit: Leaving.] 13:25:53 ESA is cool, but I imagine its assumptions don't fit every application 13:26:08 hefner: No, only ESAs :) 13:26:12 i was using slime as a GUI, but i figured i could use clim's presentations and better graphics 13:26:13 heh. 13:26:13 yeah, one tells the reader that this is an uninterned symbol, the other tells eval to fetch the function value of a specified symbol 13:26:16 right? 13:26:19 power-user type applications, sure. I'd love to have most of the apps that I spend a significant amount of time in to be emacs-style 13:26:46 beach: Thanks for describing about the difference between #: and :, I didn't know that..and used them kinda randomly. I'm glad there's an actual reason for the difference. 13:26:49 antifuchs: This is, of course, why ESA was written. 13:26:58 i bet i'll get ostrachised when I let you know I'm using vim 13:27:01 :D 13:27:04 TDT: Sure, no problem. 13:27:21 OliverUv: not quite 13:27:36 OliverUv: it's not like you're hurting anyone but yourself 13:27:41 hehe 13:27:45 I kind of wish I was a vim user, just to write a vim-style editor on top of mcclim. 13:27:53 OliverUv: #'foo expands into (function foo), which is a special operator that can get local function bindings too. 13:27:53 do it anyway 13:28:17 OliverUv: no, I need to restrict myself to things I'll actually use. 13:28:18 ah right, I forgot it expanded (like quote) 13:28:38 been about 3 years since I hacked lisp myself, fun to get started with it again (paid, too!) 13:28:38 nus [~nus@unaffiliated/nus] has joined #lisp 13:30:09 *hefner* wanders off 13:30:28 Must be nice, there's no pay opportunities for lis or clojure here. 13:30:53 -!- lusory [~bart@bb220-255-245-158.singnet.com.sg] has quit [Read error: Connection reset by peer] 13:31:40 i'm improving on my uni's lab materials 13:31:57 coding a gui for a calendar app the students finish implementing 13:31:58 beach: indeed, and it works very well for this use case (: 13:32:22 -!- attila_lendvai [~attila_le@catv-89-133-171-82.catv.broadband.hu] has quit [Ping timeout: 245 seconds] 13:33:39 -!- thunk [~user@unaffiliated/thunk] has quit [Ping timeout: 246 seconds] 13:34:29 tayloj [~tayloj@clip-point-02.dynamic2.rpi.edu] has joined #lisp 13:35:41 lusory [~bart@bb119-74-156-187.singnet.com.sg] has joined #lisp 13:37:00 -!- tayloj [~tayloj@clip-point-02.dynamic2.rpi.edu] has quit [Client Quit] 13:37:27 i suggested the project myself, so if you got ties to your alma mater (or current uni) usually the profs are open to suggestions 13:37:32 hopefully 13:37:56 tayloj [~tayloj@clip-point-02.dynamic2.rpi.edu] has joined #lisp 13:40:37 felideon [~user@12.228.15.162] has joined #lisp 13:41:07 kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has joined #lisp 13:42:08 btbngr1 [~btbgnr@smorge2.force9.co.uk] has joined #lisp 13:43:24 -!- btbngr [~btbgnr@smorge2.force9.co.uk] has quit [Ping timeout: 260 seconds] 13:45:15 LiamH: that should be fixed in clbuild 13:45:40 fe[nl]ix: OK thanks 13:46:34 nha [~prefect@imamac13.epfl.ch] has joined #lisp 13:47:51 Odin-` [~sbkhh@157.157.93.22] has joined #lisp 13:47:57 astoon [~astoon@213.141.244.246] has joined #lisp 13:48:15 as in "someone should fix it" or as in "you've fixed it" or as in "the recent git version doesn't have the problem"? 13:50:27 clbuild is the right place to fix the problem of "clbuild doesn't find src/foo.asd" 13:51:17 -!- btbngr1 is now known as btbngr 13:52:22 -!- dialtone [~dialtone@unaffiliated/dialtone] has quit [Ping timeout: 245 seconds] 13:53:02 fe[nl]ix: Oh, I thought you meant clbuild had fixed the problem. 13:54:07 oh, I see the misunderstanding D: 13:54:25 *Xach* mumbles something about how vaporware quicklisp doesn't have that problem 13:54:36 normative vs. descriptive use of "should" 13:54:49 IIANM, clbuild doesn't _have_ static-vectors. 13:55:11 lichtblau: it doesn't, but I wish to add it 13:55:20 at least to my own wnpp-projects 13:55:58 LiamH: aha. in that case, please also add it to link_extra_asds in the clbuild script so that the .asds get taken care of. 13:56:06 *LiamH* eagerly anticipates the condensation of Xach's vaporware 13:56:45 *Xach* is busy adding moons on sticks 13:56:57 :) 13:57:10 -!- Davse_Bamse [~davse@82.103.143.161] has quit [Quit: leaving] 13:57:22 jdavide [~chatzilla@nat/ibm/x-bimslrhqzkzpzish] has joined #lisp 13:57:29 I think I could even release static-vectors 13:57:50 but first I need to compile the latest snapshot of cmucl :) 13:57:57 fe[nl]ix: The development version of GSLL uses it, so that would be nice. 13:57:58 not an easy task 13:58:01 -!- jdavide [~chatzilla@nat/ibm/x-bimslrhqzkzpzish] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.5.10/20100504093643]] 13:58:17 -!- xan_ [~xan@host86-189-10-226.range86-189.btcentralplus.com] has quit [Ping timeout: 264 seconds] 13:59:06 nyef [~nyef@pool-71-255-129-229.cncdnh.east.myfairpoint.net] has joined #lisp 13:59:18 G'morning all. 13:59:19 lichtblau: OK, like this? register_other_asd static-vectors/src 14:00:12 hi nyef 14:00:38 dialtone [~dialtone@c-67-180-50-184.hsd1.ca.comcast.net] has joined #lisp 14:00:39 -!- dialtone [~dialtone@c-67-180-50-184.hsd1.ca.comcast.net] has quit [Changing host] 14:00:39 dialtone [~dialtone@unaffiliated/dialtone] has joined #lisp 14:00:53 -!- legumbre_ is now known as legumbre 14:01:40 LiamH: sounds right 14:02:18 -!- potatishandlarn [potatishan@c-b21e163b-74736162.cust.telenor.se] has quit [Ping timeout: 240 seconds] 14:02:39 lichtblau: I have some other modifications to clbuild as well... should I send diff to mailing list? 14:03:42 LiamH: a pull request would be ideal, i.e. copy your repo to a web server somewhere 14:03:58 -!- UnwashedMeme [~nathan@216.155.97.1] has left #lisp 14:04:19 lichtblau: sorry, not a darcs user (except for darcs pull). I have git repo of it though. 14:04:41 LiamH: btw, you can follow some of the activity on http://github.com/quicklisp 14:05:07 *Xach* wonders why darcshub.com is parked 14:05:22 lichtblau: how about converting clbuild to git ? 14:05:30 thunk [~user@cpe-24-26-199-70.hot.res.rr.com] has joined #lisp 14:05:40 -!- thunk [~user@cpe-24-26-199-70.hot.res.rr.com] has quit [Changing host] 14:05:40 thunk [~user@unaffiliated/thunk] has joined #lisp 14:06:17 So, something is causing the GC to trash my link register... But precisely -what- that something is escapes me. 14:06:36 potatishandlarn [potatishan@c-b21e163b-74736162.cust.telenor.se] has joined #lisp 14:06:55 caoliver [~oliver@75-134-208-20.dhcp.trcy.mi.charter.com] has joined #lisp 14:07:05 I have checked and re-checked each use of the link register against the model, and they -all- seem valid. 14:07:37 Xach: interesting, thanks. Where do we lobby to add projects? 14:08:07 jdz [~jdz@193.206.22.97] has joined #lisp 14:08:13 UnwashedMeme [~nathan@216.155.97.1] has joined #lisp 14:09:10 fe[nl]ix: not for me to say; I wouldn't have chosen darcs in the first place. Ask lukego! 14:09:49 I can push my clbuild repo to repo.or.cz if anyone wants. 14:09:51 LiamH: You could submit an issue if you wanted. Or just tell me. 14:10:01 http://github.com/quicklisp/quicklisp-projects/issues 14:11:11 LiamH: do you have something specific in mind? 14:11:15 Xach: I was thinking of some things that GSLL would like to have, like FSBV and static-vectors. 14:11:38 Xach: I think you got everything else. 14:11:56 where does static-vectors live? 14:12:11 git://gitorious.org/iolib/static-vectors.git 14:12:24 -!- galaxywatcher [~galaxywat@208.20.164.2] has quit [Quit: galaxywatcher] 14:12:24 fe[nl]ix: right? 14:12:41 I think that's the official location. 14:12:55 yes, gitorious is the official repo 14:13:32 LiamH: ok, i've added them to my to-add list 14:13:48 Xach: thanks! 14:14:45 Xach: so what's the target user for quicklisp? Lisp beginners, production users, lisp developers, or all of the above? 14:14:59 -!- HG` [~HG@xdslex247.osnanet.de] has quit [Quit: Leaving.] 14:15:34 Yes. 14:15:37 -!- potatishandlarn [potatishan@c-b21e163b-74736162.cust.telenor.se] has quit [Ping timeout: 258 seconds] 14:15:39 Anyone know if I can make my asdf package load .cl files instead of trying to load .lisp files? 14:16:22 lichtblau: lukego said it's fine for him 14:16:40 LiamH: well, I'm hoping pretty much anyone who wants to rely on a (possibly big) set of third party infrastructure-type libraries. 14:16:51 Xach: that's me! 14:16:59 LiamH: I'd like to take the fear and pain out of traversing the graph of dependencies. 14:17:18 galaxywatcher [~galaxywat@208.20.164.2] has joined #lisp 14:17:23 In the process, hopefully get more people to test libraries, report problems, fix things, etc. 14:17:45 i think clbuild is the right idea but with some aspects that don't work for me personally 14:17:58 Xach: so I could put in installation instructions for GSLL, "load quicklisp" 14:18:05 LiamH: that is my sincere hope 14:18:10 Xach: nice. 14:18:30 LiamH: of course, there's also "load gsl"... 14:18:32 Xach: I agree there are issues with clbuild. 14:18:44 Xach: yes. 14:18:59 redline6561 [~redline@gate-20.spsu.edu] has joined #lisp 14:18:59 But no need to spell out all the different pieces to load. 14:19:01 *Xach* wouldn't mind some kind of quicklisp add-on that knows how to query the debian package system for required foreign libraries 14:19:41 Xach: oooh, *really* nice. One of the things I miss about moving away from Debian lisp is the non-lisp dependency load. 14:19:46 ogamita [~t@81.202.16.46.dyn.user.ono.com] has joined #lisp 14:20:23 rme [~rme@pool-68-238-7-51.chi.dsl-w.verizon.net] has joined #lisp 14:20:24 The more features I think of, the slower it is to finish :) 14:20:52 LiamH: right now cffi doesn't report failed foreign library loads with enough detail to do it. some ambitious hacker could do the work to fix that and submit it to cffi... 14:21:34 muddyferret [~muddyferr@83.39.117.46] has joined #lisp 14:21:38 chrisdone [~user@unaffiliated/chrisdone] has joined #lisp 14:21:44 p_l: where's my trylisp? 14:22:11 slippery slope, considering 42% of quicklisp users are on Mac, with half of those having a strong preference for or against a particular ports system 14:22:12 One thing I wish ASDF had is optional usage of a system; for example, GSLL optionally depends on FSBV (which is probably why you missed it). I like the idea of optional systems because it makes the porting easier. 14:22:14 potatishandlarn [potatishan@c-b21e1588-74736162.cust.telenor.se] has joined #lisp 14:22:23 lichtblau: add-ons for all! 14:22:34 Xach: How about finishing a base set and then have more features on a TODO list that others can see? :p 14:22:37 -!- hankhero [~Adium@c-b21fb7c4-74736162.cust.telenor.se] has quit [Read error: Connection reset by peer] 14:23:03 Xach: don't add features, release it and wait for our bitter complaints about missing features to roll in! 14:23:07 Odin-`: that's my general plan, actually. i'd like to get a reasonable version 0 pushed out asap and then get the ball rolling. 14:23:46 LiamH: i'll have to pull out "Daddy can't understand you when you whine" 14:24:13 Xach: :-) 14:26:07 hankhero [~Adium@c-b21fb7c4-74736162.cust.telenor.se] has joined #lisp 14:26:26 LiamH: That strategy seems to work fine for Apple 14:26:28 -!- hankhero [~Adium@c-b21fb7c4-74736162.cust.telenor.se] has quit [Read error: Connection reset by peer] 14:27:05 *LiamH* hopes Xach places the antennas correctly in quicklisp 14:27:15 and not a knock on apple, have a MBP, ipad, few ipods, etc...but they do well at releasing stuff with minimal features, then people whine, and they add the features later :) 14:27:50 They do tend to do those minimal features quite well. 14:28:10 TDT: Features like working? :) 14:28:36 Odin-`: Well...specifically thinking of the multi tasking recently added to iOS4 14:29:11 I jailbroke my ipad just because I want multitasking. iOS4 won't even come out on the ipad for a good month or so from what I hear. 14:29:20 TDT: yeah; copy-and-paste is another famous example 14:29:30 once you've got that piece of hardware, you're bound to continue using it no matter what. Not so with a free lisp download. First impressions count there. 14:29:37 TDT: yeah; due out in august/september 14:30:02 (in fairness to them, they tend to avoid releasing until they can do whatever well) 14:30:04 Hm, Xach, I followed your packaging guide (now have almavis.asd, package.lisp and almavis.lisp) but when I do (load (compile-file "almavis")) I just get a no such package error 14:30:24 I think it is trying to compile almavis.lisp, hits the (in-package #:almavis) and dies 14:30:39 hankhero [~Adium@c-b21fb7c4-74736162.cust.telenor.se] has joined #lisp 14:30:51 OliverUv: Where is that package defined? 14:30:59 package.lisp 14:31:23 OliverUv: my advice re load/compile-file applies only when everything is in a single file. 14:31:32 ah ok 14:31:39 OliverUv: if there are multiple files, i recommend writing an .asd file 14:31:47 I have an .asd file 14:31:54 then load it with asdf 14:32:03 why are you manually compiling 14:32:24 ah, I got the impression that you only dd that when you were done with it 14:32:35 alright i'll load it into asdf like I did with mcclim then? 14:32:50 (i got mcclim from source, so guessing it'd work the same way) 14:33:25 OliverUv: asdf loading is how you get projects described by an .asd file into the running system. 14:33:47 I should have figured. 14:34:04 Speaking of compiling and all...when having a asdf location of libraries - say I have 4-5 libraries there. Once the libraries are loaded, obviously the fasl files are written, but lets say I wanted to have multiple implementations point at the same asdf-installed location. So those 5 libraries could be shared between the implementations. Would the fasl files cause problems, or will each implmemtnation simply recompile on reload? 14:34:04 Though I did try (require 'almavis) but that didn't work 14:34:17 it just said it didn't know how to require almavis 14:34:33 TDT: it can be a problem. 14:34:37 (using SBCL, btw) 14:34:37 OliverUv: what CL are you using? 14:34:43 it should work with asdf if the asd file is in your asdf:*central-registry* paths 14:34:54 if not you should put/link it from there 14:35:08 OliverUv: what xristos said. one easy way in slime is to use ,cd to go to the path of your project. 14:35:10 TDT: some lisps use the same file ending for fasls. SBCL, CMUCL and SCL name clash. 14:35:18 Xach: Ah, dang, I see. i was hoping to try and avoid hving to have multiple copies of these libraries. 14:35:26 TDT: ASDF2 segregates fasls by default. 14:36:04 asdf2? hmm, is that included by default with sbcl at this point or is it something I need to upgrade to? 14:36:12 xristos: ah, ok, Xach's guide said I could use (require 'almavis) if I was in the same directory, so I assumed there was some error loading my thing normally, thought it had to be evaluated or something first 14:36:27 which was an erronous thought, which got me into the whole compile track 14:36:32 TDT: it's loaded by default. 14:36:46 anyway I'll probably manage it now, big thanks for the help and patience, I appreciate it 14:36:49 lots! 14:37:47 Xach: k, I'll have to see if abcl will play nice with that as well. I really should consider trying to encapsulate all libraries in my project much like clbuild tries to do and just go that route. Kinda like the frozen gems idea with rails. I rely on asdf a lot. 14:37:49 -!- dialtone [~dialtone@unaffiliated/dialtone] has quit [Read error: Operation timed out] 14:38:03 OliverUv: i have a directory ~/code/lisp with all my lisp projects, then i have another directory ~/code/lisp/asdf where i link all the asd files, then i can do (require :project) or (asdf:oos 'asdf:load-op :project) 14:38:16 Xach: Well global asdf-installed libraries, I should say. 14:38:22 mm 14:39:24 OliverUv: also something like (push "/Users/xristos/code/lisp/asdf/" asdf:*central-registry*) in my site-init.lisp which get's loaded from .sbclrc 14:41:26 rdd [~user@c83-250-52-182.bredband.comhem.se] has joined #lisp 14:41:28 yup, then you can create a symbolic link to where ever you are developing the package and it will load no matter where your repl's current working directory is. 14:42:55 dfox [~dfox@r3ah113.net.upc.cz] has joined #lisp 14:44:51 Current theory: Something is loading the link register from reg_LIP... when reg_LIP is improperly based (not placed at a useful offset from a boxed register). 14:47:11 xyxxyyy1 [~xyxu@116.227.164.71] has joined #lisp 14:48:16 -!- nus [~nus@unaffiliated/nus] has quit [Ping timeout: 260 seconds] 14:48:18 -!- xyxxyyy [~xyxu@116.227.164.71] has quit [Ping timeout: 240 seconds] 14:48:25 Fullma [~fullma@81.192.102.2] has joined #lisp 14:49:04 Actually, I know of a handful of locations where that's still done... And examining the program counter would be a good start. 14:49:48 lexa_ [~lexa_@seonet.ru] has joined #lisp 14:50:16 -!- lexa_ is now known as Guest85644 14:50:42 -!- ignas [~ignas@78.60.242.102] has quit [Ping timeout: 245 seconds] 14:50:57 -!- varjag [~eugene@122.62-97-226.bkkb.no] has quit [Remote host closed the connection] 14:53:38 ah, getting some errors because CL and MCCLIM both define a describe symbol, which should shadow which? 14:54:52 neither 14:54:55 Technically, neither: The CLIM spec calls for a separate CLIM-LISP package. 14:54:59 OliverUv: there's the CLIM-LISP package that your package can use instead of CL 14:55:06 To be used instead of the COMMON-LISP package. 14:55:11 ah ok thanks! 14:56:08 Now, that said, the "good" CLIM implementations are sufficiently tied to the host lisp that the packages are identical, giving some CLIM benefits to non-CLIM programs and libraries. 14:57:02 by that you mean MCL and the lisp machine based ones? (: 14:57:10 chrisdone: been working on my startup (after trying to fix some personal issues) 14:57:35 Bwah? The -program counter- is also in oldspace? 14:58:01 Okay, now I know my model is broken. 14:58:16 it is all working now, hooray! 14:58:18 p_l: a lisp startup? 14:58:44 hypno: yes 14:58:49 cool 14:58:58 assuming it gets beyond the idea stage 14:59:05 antifuchs: I mean that the degree of host integration is a measure of the quality of the CLIM implementation. 14:59:08 -!- rahul [~rjain@66-234-32-150.nyc.cable.nyct.net] has quit [Ping timeout: 265 seconds] 14:59:10 also, I'm starting to seriously consider dropping out of school :/ 14:59:16 Even though CLIM -is- horribly broken as specified. 14:59:17 p_l: so what is it? 14:59:18 Oh? 14:59:25 from talking about lisp with my project supervisors on gmail I've started getting ads for lisp jobs 14:59:30 which is pretty nice 15:00:14 davazp [~user@83.55.180.105] has joined #lisp 15:00:54 rahul [~rjain@66-234-32-150.nyc.cable.nyct.net] has joined #lisp 15:01:56 az [~az@2a01:4f8:101:4a2::9c00] has joined #lisp 15:02:09 hypno: simple, non-programmer friendly system to build sort of "web apps" 15:04:07 p_l: ah. sounds like a good idea. like a generic version of "build your own webshop"-site then? :) 15:05:53 -!- kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has quit [Ping timeout: 265 seconds] 15:06:15 hypno: yeah, with graded complexity, so I can lure people who actually want to do more by allowing access to more powerful tools. 15:06:39 p_l: while a cool project, i think it hasnt been done because it sounds very hard. 15:06:41 Plus integrated hosting etc., selling this in SaaS model 15:07:19 hypno: actually, bits and pieces are laying around and similar projects exist, but for different environment 15:07:27 (like OpenBlocks for Java) 15:08:06 -!- johs [~johs@hawk.netfonds.no] has quit [Read error: Operation timed out] 15:08:09 ie, making html/css/js and all the other crap that makes up the web to something coherent and idiot friendly requires tons of hard work. :) 15:08:10 johs [~johs@hawk.netfonds.no] has joined #lisp 15:08:38 -!- Fullma [~fullma@81.192.102.2] has quit [Quit: Fullma] 15:08:54 -!- rahul [~rjain@66-234-32-150.nyc.cable.nyct.net] has quit [Ping timeout: 258 seconds] 15:09:42 -!- galaxywatcher [~galaxywat@208.20.164.2] has quit [Quit: galaxywatcher] 15:10:27 lichtblau: could you create a project on gitorious for clbuild ? 15:10:49 kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has joined #lisp 15:10:58 rahul [~rjain@66-234-32-150.nyc.cable.nyct.net] has joined #lisp 15:11:02 hypno: sure, but I'm actually going for certain limitations to decrease the issue 15:11:27 at worst case I can put the code into flash to avoid performance issues on web browsers 15:11:57 Heh. For the changes I'm working with now, I have my SBCL build-and-test cycle bottleneck down to the test part of the cycle. 15:13:34 ikki [~ikki@201.155.75.146] has joined #lisp 15:13:54 DanLentz [~danlentz@c-68-32-54-29.hsd1.nj.comcast.net] has joined #lisp 15:14:01 billitch [~billitch@dslb-088-075-177-102.pools.arcor-ip.net] has joined #lisp 15:14:35 HAH! I think I pretty much finished the SICL module that implements the conses dictionary. A lot more tests are needed, though, but if anyone wants to send me additional tests (preferably some that indicate defects) that would be great! 15:15:23 beach: Congratulations. 15:15:25 What would be the quickest way to teach Lisp to a professional programmer? 15:15:36 nyef: Thanks! 15:15:41 pjb: Hire them, and throw them in at the deep end. 15:15:44 zomgbie [~jesus@chello062178135013.2.14.vie.surfer.at] has joined #lisp 15:15:47 p_l: ah, well good luck with that. we're all thursty for another lisp webapp success story. :) anyway, what do you study and why do you wanna drop out? 15:16:00 nyef: that's what I'm thinking too, but if they need some help? 15:16:15 pjb: What is the context? Are you paid to do this? Is it a friend? Does the person *want* to be taught Lisp, or was it imposed? 15:16:38 I'm teaching Lisp to a colleague to develop a new project. 15:17:04 pjb: I hear the French Foreign Legion have a great system for learning French fluently and quick. I bet it would work for CL too. Tell them: "You learn, or you get your head kicked in." ;) 15:17:08 hypno: CS & AI at university of Aberdeen, and I had a really bad year right now, trying to somehow pass (unclear yet If I will) 15:17:09 stis [~stis@1-1-1-39a.veo.vs.bostream.se] has joined #lisp 15:17:18 -!- blandest [~user@85.204.33.242] has quit [Ping timeout: 240 seconds] 15:17:18 He is not hostile, but a little submerged with all there is to learn at once, emacs, slime, lisp, libraries... 15:17:20 pjb: Provide a mentor? 15:17:23 barcon332 [~barcon332@208.89.210.254] has joined #lisp 15:17:23 pjb: These days, I pretty much always start with the evaluation model (REPL, semantics as a sequence of interactions). 15:17:38 p_l: master or phd or? 15:17:56 sanjoyd [~sanjoyd@unaffiliated/sanjoyd] has joined #lisp 15:18:21 We've covered that already. 15:18:24 I guess we just need more time to digest everything... 15:18:51 hypno: the french foreign legion are a fucking bunch of racists 15:19:07 hypno: undergrad that adds an extra bit (UK specific) on top of BSc that makes it relevant to jumping straight to PhD 15:19:16 -!- astoon [~astoon@213.141.244.246] has quit [Quit: astoon] 15:19:46 not that i like this kind of gross generality =P 15:19:46 billitch: In the movies "Le Randonneurs", the capitain is black and talks with the same racists terms (which are really only code words) than his colonnel. :-) 15:19:49 Ugh. Always fails in the same place, but only with the full context of the rest of the file. :-/ 15:19:53 pjb: I spend a lot of time on higher-order functions just to illustrate the concept of first-class and anonymous functions. 15:19:57 p_l: i see. less CL/irc/Y.C/reddit for you! more study! the job market for half-assed academics is bad. ;) 15:20:24 -!- carrl [~carl@61-64-164-206-adsl-tai.STATIC.so-net.net.tw] has quit [Ping timeout: 258 seconds] 15:20:33 pjb: After that, I do generic functions, classes, method combinations. 15:20:36 pjb: i know of people who were detached with them 15:20:43 but hollywood gotta be right 15:20:45 =) 15:20:56 hypno: I'd say I've got fucked enough by being homeless for three months 15:21:01 carrl [~carl@61-64-164-206-adsl-tai.STATIC.so-net.net.tw] has joined #lisp 15:21:21 -!- Guest85644 [~lexa_@seonet.ru] has left #lisp 15:21:27 Jabberwockey [~Jens@port-5104.pppoe.wtnet.de] has joined #lisp 15:21:52 -!- redline6561 [~redline@gate-20.spsu.edu] has quit [Ping timeout: 260 seconds] 15:21:53 beach: we spend a couple of hours on that (and the equivalence closure<->classes), but indeed, I could insist a little more on the higher order functions, and functional abstraction. 15:22:47 p_l: homeless? why? there are no student apartments, etc over there? or you miss state funding due to bad performance or? 15:22:51 billitch: oops, it wasn't in "Les Randonneurs". but in "Le Raid" (2002). Very funny movie. 15:24:30 beach: I was hoping for a magic bullet^W booklet that would teach them the whole CL and enlightment in 20 pages :-) 15:26:08 talking about the red pill.. "[neo's eyes flickering] ... - I know Lisp." 15:26:28 Yes :-) Something like that. 15:26:47 hypno: no state funding, bad events related to work during summer leaving me penniless etc. 15:27:02 hypno: now add to that me falling into depression 15:27:29 pjb: I don't think that would be possible, unless the person already knows similar languages before. 15:28:29 NIR[f] [~root@78.24.45.0] has joined #lisp 15:28:48 That's not the case. 15:29:26 milanj [~milanj_@178.223.147.206] has joined #lisp 15:29:51 -!- fatblueduck [~user@rrcs-67-52-188-98.west.biz.rr.com] has quit [Remote host closed the connection] 15:30:27 -!- bozhidar [~user@212.50.14.187] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 15:32:20 p_l: ah. i've been in similiar situation this summer (living beneath minimal income), hustling every god damn day for survival... but it could be worse and things will eventually change for the better. :) 15:32:40 -!- rahul [~rjain@66-234-32-150.nyc.cable.nyct.net] has quit [Ping timeout: 240 seconds] 15:32:42 bgs100 [~ian@unaffiliated/bgs100] has joined #lisp 15:33:22 In the US? 15:33:28 pjb: I've had success with PCL and some little CL projects I've written at work. 3 interns and 2 full-timers. Hand them a well-defined and smallish project, run a few lessons, be generally available. 15:34:01 jpanest: They were all learning at once? 15:34:09 pjb: were/are 15:34:14 rahul [~rjain@66-234-32-150.nyc.cable.nyct.net] has joined #lisp 15:34:43 This might be more motivating to learn in a group however. 15:34:48 jpanest: That helped, though. I think. Some have some recent academic scheme (but very little). 15:35:18 lat [~lat@125.167.140.159] has joined #lisp 15:38:12 -!- nha [~prefect@imamac13.epfl.ch] has quit [Ping timeout: 248 seconds] 15:38:33 -!- billitch [~billitch@dslb-088-075-177-102.pools.arcor-ip.net] has quit [Ping timeout: 240 seconds] 15:39:16 I'm talking to myself. 15:39:26 pjb: yeah, i think that was a factor 15:39:55 pjb: i dont think you can teach cl to someone who doesnt want to learn it 15:40:01 -!- Xach [~xach@cpe-72-227-90-1.maine.res.rr.com] has quit [Ping timeout: 276 seconds] 15:40:25 all you get is one of those "yeah, i tried lisp a few years ago. what a mess! it should stay in university." 15:40:37 This is ok. The problem is that there's a lot to learn and he feels a little overwhelmed. 15:42:47 -!- Athas [~athas@shop3.diku.dk] has quit [Remote host closed the connection] 15:42:58 bbl 15:43:01 well, that and that you actually have to learn the language by the spec and books and not so much by trial and error. it's hard work and not newbie friendly. 15:45:35 Davse_Bamse [~davse@82.103.143.161] has joined #lisp 15:47:19 "Yeah, I tried C++ a few years ago. What a mess! It should stay in whatever research lab spawned it." 15:47:34 Well it does take awhile to really get CL, but quite frankly, that's one thing I enjoy about the language 15:48:44 heh, lkcl seems to have a knack for writing long rants. 15:48:51 When I picked up PHP, I was making database connections within a week and writing crappy but functional code pretty fast. After a year, PHP didn't have much more to offer my learning...been using it 5 years now, and haven't really learned anything new for at least 2-3 of those years. CL is going on..2 years now, ish or so...and I still feel like a newbie at the language. In 5 years I think I'll still be learning a lot 15:50:31 foom: hmm. ref? 15:50:51 also on COM-ish crap -> http://lists.debian.org/debian-devel/2010/07/msg00179.html 15:50:52 http://norvig.com/21-days.html -- sums up it well :) 15:51:42 funnily enough the changes to mozilla were from > 6 months ago, and yet he hasn't actually fixed python-xpcom yet. 15:52:30 nyef: =) 15:52:51 Ugh. Okay, found at least one bug. 15:53:03 I forgot that the PC for an 15:53:22 I forgot that the PC for an LRA is the LRA, plus n-word-bytes, minus other-pointer-lowtag. 15:53:40 Which turns out to be (surprise!) negative three. 15:54:13 Instant hole, one instruction wide, on all function returns. 15:54:34 Also applies to the link register. 15:54:42 Xach [~xach@cpe-72-227-90-1.maine.res.rr.com] has joined #lisp 15:55:29 -!- Snamich [~Snamich@166.133.254.178] has quit [Quit: Snamich] 15:55:29 -!- jdz [~jdz@193.206.22.97] has quit [Read error: Operation timed out] 15:55:35 nowhere_man [~pierre@AStrasbourg-551-1-22-10.w86-213.abo.wanadoo.fr] has joined #lisp 15:55:42 -!- kephas [~pierre@AStrasbourg-551-1-11-119.w86-213.abo.wanadoo.fr] has quit [Ping timeout: 245 seconds] 15:56:41 ASau [~user@83.69.227.32] has joined #lisp 15:57:24 -!- mega1 [~quassel@pool-03c9a.externet.hu] has quit [Ping timeout: 248 seconds] 15:57:43 mega1 [~quassel@pool-061e4.externet.hu] has joined #lisp 16:00:10 homie [~user@xdsl-87-79-190-34.netcologne.de] has joined #lisp 16:00:36 -!- gzip4 [~xxx@78.108.73.250] has quit [Ping timeout: 260 seconds] 16:00:49 -!- Draggor [~Draggor@adsl-99-148-31-40.dsl.emhril.sbcglobal.net] has quit [Read error: Operation timed out] 16:01:51 TDT: there's plenty to learn with PHP too, specially since they keep screwing up^W^Wupdating the language. 16:03:18 slyrus: site down? 16:03:51 TDT: and there _are_ little details that will bite you in the ass several years later. I think a big difference is that CL gets you so involved in the details of the language. The language is specified, so you can't just hammer something out at the repl and expect the same behavior in a different implementation (or even in the same one a couple versions down). There's also the fact that you need to understand something pretty well in 16:03:51 order to program it. It's not so much that PHP is simple (it isn't), it's that it can't do a lot of things CL can. 16:04:43 -!- spradnyesh [~pradyus@nat/yahoo/x-arjrswpjpgfngzmf] has left #lisp 16:07:43 jhalogen [~jake@cpe-98-154-250-117.socal.res.rr.com] has joined #lisp 16:08:42 -!- abend [~alx@delta.muted.org] has quit [Ping timeout: 258 seconds] 16:10:48 -!- benny [~user@i577A7669.versanet.de] has quit [Ping timeout: 240 seconds] 16:11:36 xyxxyyy [~xyxu@116.227.164.71] has joined #lisp 16:11:46 curi_ [~curi@adsl-99-114-139-86.dsl.pltn13.sbcglobal.net] has joined #lisp 16:11:57 hjpark [~user@116.40.135.21] has joined #lisp 16:13:13 benny [~user@i577A8785.versanet.de] has joined #lisp 16:14:20 -!- xyxxyyy1 [~xyxu@116.227.164.71] has quit [Ping timeout: 240 seconds] 16:15:36 Draggor [~Draggor@adsl-99-141-191-116.dsl.emhril.sbcglobal.net] has joined #lisp 16:17:34 -!- chrisdone [~user@unaffiliated/chrisdone] has quit [Remote host closed the connection] 16:17:48 sykopomp: personally I feel PHP is fairly simple, but you're right - there are little details, such as casting issues and the == vs === that will bite even someone who's done it years...but the barrier of entry to getting even the most basic thing done is very low in PHP 16:18:05 -!- athlon [~user@94.179.207.86.pool.3g.utel.ua] has quit [Remote host closed the connection] 16:18:35 -!- relcomp [~chatzilla@pc92-e03-3.igf.Uni-Osnabrueck.DE] has left #lisp 16:20:29 abeaumont [~abeaumont@84.76.48.250] has joined #lisp 16:20:30 -!- kiuma [~kiuma@85-18-55-37.ip.fastwebnet.it] has quit [Quit: Bye bye ppl] 16:21:34 TDT: basic from which perspective? 16:22:03 bozhidar [~user@93-152-185-88.ddns.onlinedirect.bg] has joined #lisp 16:22:35 I'll answer my own question here. Basic from the perspective that you can write a bit of code and see a result in a familiar context 16:22:41 -!- abeaumont [~abeaumont@84.76.48.250] has quit [Client Quit] 16:22:59 The familiar context is the killer feature and one that the REPL lacks, I think 16:23:05 Who is hyped for the Boston Lisp Meeting?! 16:23:06 *Xach* is hyped 16:23:18 I'd be more hyped if I could ever make it to one :p 16:23:33 abeaumont [~abeaumont@84.76.48.250] has joined #lisp 16:23:33 I'd be hyped if I lived nearby :( 16:23:49 Something has come up for me during it for every month except the first two 16:24:01 dlowe: What about this month? 16:24:08 Taking a trip to Sydney 16:24:09 dlowe: I should clarify the comment. A very common program in these "learn X in 21 days" is a shopping cart application for more web-friendly languages. I can see creating a shopping cart application being far easier in PHP than in Lisp for the newbie looking at the shelf keeping the database connectivity stuff in mind. This is what I mean by the barrier of entry as well. 16:24:22 Xach: Only if I lived in Boston, or near Boston :) 16:24:31 dialtone [~dialtone@unaffiliated/dialtone] has joined #lisp 16:25:34 TDT: you keep coming up with problems. no lisp companies in iowa! no boston lisp meeting in iowa! blah blah blah! just make it happen! 16:25:58 if you start biking now, you will have no trouble! 16:26:06 Xach: lol 16:26:10 TDT: I'd say it takes about far more effort to parse through PHP's database interface documentation as it does to install a db interface in a lisp implementation 16:26:12 -!- mstevens [~mstevens@fsf/member/pdpc.active.mstevens] has quit [Quit: leaving] 16:26:35 dlowe: I find clsql considerably more difficult to use than PEAR:DB, personally 16:26:43 but I *like* clsql much better though :) 16:26:57 dlowe: I don't have a similar impression. 16:28:05 postmodern's interface is sexier 16:28:17 *Xach* likes postmodern a lot 16:28:36 *Xach* mumbles something about lowering bars with quicklisp 16:28:54 postmodern is excellent. 16:28:57 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 240 seconds] 16:37:04 thanks xach 16:37:24 *slyrus* needs to upgrade freebsd to something from the last couple years... 16:37:45 Clozure (the company) now writes Obj-C? 16:39:01 hefner: yeah, I have a graph package called epigraph. not sure if anyone else uses it. 16:40:04 felideon: For a long time. 16:40:13 slyrus: Does it have a description online? I've never heard of it before. 16:41:08 Xach: thx 16:41:24 peterhil [peterhil@YKXVI.gprs.sl-laajakaista.fi] has joined #lisp 16:41:25 mega1_ [~quassel@pool-01bb8.externet.hu] has joined #lisp 16:41:57 -!- mega1 [~quassel@pool-061e4.externet.hu] has quit [Ping timeout: 245 seconds] 16:42:15 felideon: Clozure still develops Clozure CL and does lisp consulting, of course. 16:43:38 redline6561 [~redline@c-66-56-55-169.hsd1.ga.comcast.net] has joined #lisp 16:43:53 Is there some command reference for Clozure CL:s Listener? 16:44:33 I mean, I want to know how to get input from history. And tab-completion would be nice also. 16:45:44 nus [~nus@81.24.80.50] has joined #lisp 16:45:44 -!- nus [~nus@81.24.80.50] has quit [Client Quit] 16:46:02 nus [~nus@unaffiliated/nus] has joined #lisp 16:46:58 -!- xyxxyyy [~xyxu@116.227.164.71] has quit [Quit: Leaving.] 16:47:30 peterhil: In the IDE, meta-p and meta-n go through input history. meta-/ works for completion. 16:47:49 Blkt [~user@dynamic-adsl-78-13-252-73.clienti.tiscali.it] has joined #lisp 16:48:23 If you mean just a plain listener in a terminal, you need to use Slime or something. 16:49:07 HET3 [~diman@w220.engin.cf.ac.uk] has joined #lisp 16:50:14 fisxoj [~fisxoj@HSI-KBW-095-208-111-226.hsi5.kabel-badenwuerttemberg.de] has joined #lisp 16:50:47 jleija [~jleija@user-24-214-122-46.knology.net] has joined #lisp 16:51:52 chairos [~jon@inklesspen.com] has joined #lisp 16:53:05 faux [~user@1-1-4-21a.gkp.gbg.bostream.se] has joined #lisp 16:54:46 abend [~alx@67.136.131.11] has joined #lisp 16:54:53 -!- Draggor [~Draggor@adsl-99-141-191-116.dsl.emhril.sbcglobal.net] has quit [Ping timeout: 240 seconds] 16:55:17 -!- hjpark [~user@116.40.135.21] has quit [Ping timeout: 245 seconds] 16:55:40 Xach: no, I guess it needs a blog entry, at least... 16:55:42 http://git.cyrusharmon.org/cgi-bin/gitweb.cgi?p=epigraph.git;a=summary 16:56:11 it's used by chemicl, my still-as-yet-unborn chemistry informatics library 16:57:37 nixeagle [~user@Wikimedia/Nixeagle] has joined #lisp 17:03:07 danlen-1 [~danlentz@mobile-166-137-139-117.mycingular.net] has joined #lisp 17:03:45 fatblueduck [~user@rrcs-67-52-188-98.west.biz.rr.com] has joined #lisp 17:06:57 -!- JonSmith [~jon@ip24-250-13-137.ri.ri.cox.net] has quit [Ping timeout: 245 seconds] 17:07:12 -!- DanLentz [~danlentz@c-68-32-54-29.hsd1.nj.comcast.net] has quit [Ping timeout: 265 seconds] 17:09:00 -!- mega1_ [~quassel@pool-01bb8.externet.hu] has quit [Read error: Operation timed out] 17:10:49 *Xach* ran into a bug that looked something like this: (defgeneric foo (bar baz) (:method (baz bar) ...)) 17:10:51 hjpark [~user@116.40.135.21] has joined #lisp 17:11:12 of course, the args weren't metasyntactic variables, so it wasn't as easy to realize one set was out of order. 17:11:33 surely sbcl could issue a style-warning for that... 17:11:33 -!- NIR[f] [~root@78.24.45.0] has quit [Ping timeout: 265 seconds] 17:12:00 lvillani [~lvillani@fedora/lvillani] has joined #lisp 17:12:43 -!- ogamita [~t@81.202.16.46.dyn.user.ono.com] has quit [Ping timeout: 258 seconds] 17:15:30 lispm [~lispm@d226022.adsl.hansenet.de] has joined #lisp 17:17:30 lispm_ [~lispm@d226022.adsl.hansenet.de] has joined #lisp 17:17:30 -!- lispm [~lispm@d226022.adsl.hansenet.de] has quit [Read error: Connection reset by peer] 17:17:31 -!- lispm_ is now known as lispm 17:17:53 mega1 [~quassel@catv4E5CABA2.pool.t-online.hu] has joined #lisp 17:19:55 varjag [~eugene@84.118.202.84.customer.cdi.no] has joined #lisp 17:21:02 silenius [~silenius@c-24-130-172-194.hsd1.ca.comcast.net] has joined #lisp 17:22:08 -!- antoszka [~antoszka@unaffiliated/antoszka] has quit [Quit: +++ killed by SIGSEGV +++] 17:23:08 -!- lispm [~lispm@d226022.adsl.hansenet.de] has quit [Remote host closed the connection] 17:28:01 Snamich [~Snamich@161.210.164.242] has joined #lisp 17:28:02 Makoryu [~vt920@pool-71-174-191-10.bstnma.fios.verizon.net] has joined #lisp 17:28:13 -!- zomgbie [~jesus@chello062178135013.2.14.vie.surfer.at] has quit [Ping timeout: 276 seconds] 17:28:48 -!- nus [~nus@unaffiliated/nus] has quit [Ping timeout: 260 seconds] 17:29:22 rme: Yes, that's exactly what I wanted. Thanks! 17:30:40 -!- TDT [~user@173-30-32-53.client.mchsi.com] has quit [Ping timeout: 260 seconds] 17:31:17 -!- zeroish [~zeroish@135.207.174.50] has quit [Read error: Operation timed out] 17:31:53 -!- dabd [~dabd@gtkdbio.inesc-id.pt] has quit [Ping timeout: 260 seconds] 17:32:30 slyrus: is it ready for public consumption? 17:33:34 pkhuong:
instead of a 
? that is a little odd.
17:33:48 -!- cmm [~cmm@bzq-109-64-200-182.red.bezeqint.net] has quit [Ping timeout: 265 seconds]
17:34:34 cmm [~cmm@109.64.200.182] has joined #lisp
17:35:09 Dawgmatix [~dman@c-76-124-9-27.hsd1.nj.comcast.net] has joined #lisp
17:35:37  Xach: yeah. It's all in the CSS... generated by a third-party tool coded by someone who really believes in presentation in css.
17:35:59 moah [~gnu@188.109.158.189] has joined #lisp
17:36:02 SegFaultAX [~SegFaultA@c-98-234-1-162.hsd1.ca.comcast.net] has joined #lisp
17:36:29  really fucks up the ol' syndicators
17:36:41 syamajala [~syamajala@68-116-188-249.dhcp.oxfr.ma.charter.com] has joined #lisp
17:36:55 -!- danlen-1 [~danlentz@mobile-166-137-139-117.mycingular.net] has quit [Quit: Colloquy for iPhone - http://colloquy.mobi]
17:36:56  Xach: no documentation, the API changes occasionally. it's certainly ready for someone else to take a look at it though :)
17:37:00  Yup. I'll frob the output a bit soon.
17:37:02 NIR[f] [~root@pppoe-45-231.prtcom.ru] has joined #lisp
17:37:53  slyrus: i'm always looking to pad my quicklisp project count, even with marginal projects. anything to stay ahead of the competition.
17:38:00  the marginaler the better.
17:38:05 -!- potatishandlarn [potatishan@c-b21e1588-74736162.cust.telenor.se] has quit []
17:38:18  CLAN?
17:38:28  What's CLAN?
17:38:59  C*AN = Comprehensive [insert programming language here] Archive Network :p
17:39:20  damn, never heard of them! how many projects do they have?
17:39:25 *Xach* will scramble to catch up
17:39:31  Xach: the lisp one? near nothing
17:39:40  Sure sounds comprehensive to me
17:39:42  oh, cclan?
17:39:43  There is a decent Lisp one? :p
17:40:01  Odin-`: where did you see "decent"? :D
17:40:09  p_l: 8)
17:40:32 *Odin-`* suffers from having been introduced to Perl's CPAN at about the same time as programming.
17:40:47  "What do you mean I have to fetch a library manually?"
17:41:15 *lichtblau* checks
17:41:16  cxml-rpc is missing!
17:41:23 -!- Jasko [~tjasko@209.74.44.225] has quit [Quit: Leaving]
17:42:05 *Xach* adds to the queue
17:43:24 austinh [~austinh@c-24-21-81-46.hsd1.or.comcast.net] has joined #lisp
17:43:57  the marginaler the better?
17:43:58  soundex and levenshtein (see cliki) are absolute toy projects someone did, which I enjoyed recently.  A rare moment of "oh, look, someone wrote those 10 lines of CL code for me already".
17:45:11 *Xach* queues and queues
17:46:01  Odin-`: once quicklisp launches, that'll probably be more like "what do you mean I have to use an external application to install a library? Why can't I just require it?"
17:46:04  \o/
17:46:15  ha ha ha
17:46:37 *Xach* has pure-CL GPG signature verification nearly working
17:46:39  Presentation in CSS is fine, but you're supposed to still have the semantics in the html. :)
17:46:51  Xach: \o/
17:47:08  sykopomp: a require hook that automatically provided "INSTALL" as a restart would be pleasant :)
17:47:22  dlowe: indeed!
17:47:32  I might do that anyway and have it hook into asdf-install
17:47:34  how do require hooks work, anyway? Is there a way to set them up portably?
17:47:41  sykopomp: not portably.
17:47:45  they're not portable
17:47:49  aw :\
17:47:51  big oversight, really
17:48:02  well, require is deprecated, no?
17:48:05  hah
17:48:11  sykopomp: yeah, next version of CL and it's gone!
17:48:13  It's kind of nice how everyone can finally agree that Clojure is a nice, practical lisp for everyone to standardize on and write their code in
17:48:30  Makoryu: Yes, absolutely everyone.
17:48:35  At the same time it's reassuring to know that CL is slightly healthier than D
17:48:35  Makoryu: I can't tell if you're trying to be funny or not
17:48:42  Xach: along with all those pesky -if-nots!
17:48:47  dlowe: That's the way I like it
17:48:52  I can't wait for the -if-nots to finally be gone!
17:48:54  lichtblau: got my message ?
17:49:03  D is going to be included into the next Fedora version.
17:49:11  the new LLVM-based implementation
17:49:33  dlowe: no one would say that seriously, so he's obviously joking.
17:49:54  sykopomp: you're an optimist, I see
17:49:56 lexa_ [~lexa_@seonet.ru] has joined #lisp
17:50:24 -!- lexa_ is now known as Guest75255
17:50:29  dlowe: no, I'm just completely willing to dismiss Clojure and its advocates.
17:50:31  sykopomp: Or am I
17:50:42  fe[nl]ix: ah, wouldn't have seen it if you hadn't pointed it out.
17:50:45  fe[nl]ix: will do.
17:50:55  Makoryu: like I said, you can't be serious, so I know you're not.
17:50:57  sykopomp: no, about removing the if-nots.
17:51:03  sykopomp: Or do you
17:51:05  sykopomp: they're the pessimist's functions.
17:51:20 -!- Guest75255 [~lexa_@seonet.ru] has left #lisp
17:52:04  Xach: I'm told that when you're trying to communicate with kids (and often, adults), you need to state things in a positive way because the 'not' in your sentences tend to be removed by others' brains.
17:52:56  oh no! my requires!
17:53:07 -!- austinh [~austinh@c-24-21-81-46.hsd1.or.comcast.net] has quit [Quit: leaving]
17:56:58 -!- davazp [~user@83.55.180.105] has quit [Read error: Connection reset by peer]
17:58:39 Stattrav [~Stattrav@117.192.143.77] has joined #lisp
18:02:01 -!- NIR[f] [~root@pppoe-45-231.prtcom.ru] has quit [Quit: Leaving.]
18:02:10 -!- sanjoyd [~sanjoyd@unaffiliated/sanjoyd] has quit [Ping timeout: 258 seconds]
18:02:10 -!- abend [~alx@67.136.131.11] has quit [Quit: Ex-Chat]
18:04:13 -!- minion [~minion@common-lisp.net] has quit [Ping timeout: 240 seconds]
18:06:45  lichtblau: the repo is @ git://common-lisp.net/users/sionescu/clbuild.git
18:07:28 -!- tayloj [~tayloj@clip-point-02.dynamic2.rpi.edu] has quit [Ping timeout: 260 seconds]
18:08:26  lichtblau: then you can cd /project/clbuild/public_html/git/ ; make-git-mirror git://gitorious.org/clbuild/clbuild.git
18:08:37 -!- tsuru- [~user@c-174-50-217-160.hsd1.tn.comcast.net] has quit [Read error: Operation timed out]
18:08:45  OliverUv: Tell my friends hello from me: Arne Jönsson, Erik Sandewall, Anders Haraldson, etc.
18:09:25 -!- nixeagle [~user@Wikimedia/Nixeagle] has quit [Remote host closed the connection]
18:09:28 tsuru- [~user@c-174-50-217-160.hsd1.tn.comcast.net] has joined #lisp
18:09:46 sanjoyd [~sanjoyd@unaffiliated/sanjoyd] has joined #lisp
18:14:22 Fare [~Fare@ita4fw1.itasoftware.com] has joined #lisp
18:16:04 mengu [~mengu@unaffiliated/mengu] has joined #lisp
18:16:07 -!- mengu [~mengu@unaffiliated/mengu] has left #lisp
18:18:06 tfb [~tfb@restormel.cley.com] has joined #lisp
18:18:58 tayloj [~tayloj@graciliraptor-89.dynamic2.rpi.edu] has joined #lisp
18:22:34 nixeagle [~user@Wikimedia/Nixeagle] has joined #lisp
18:23:20  fe[nl]ix: wait, what's the plan here?
18:23:52  upload your repo (which is on common-lisp.net) to gitorious, then set up common-lisp.net so that it mirrors from gitorious?
18:25:00  beach: haha, Anders Haraldsson is my prof!
18:25:01  If the master is on gitorious, why mirror to common-lisp.net?  The main advantage of having a repo directly on common-lisp.net is that existing contributors keep their write access, and mirroring negates that advantage.
18:25:19  he is so great!
18:25:24  OliverUv: what about Harald Andersson?
18:25:38  Sorry, he is probably not involved.
18:25:43  Never heard about him
18:25:57 -!- lvillani [~lvillani@fedora/lvillani] has left #lisp
18:26:11  Anders Haraldsson held all the lectures for our beginner courses in programming
18:26:22  taught us all the lisp we know1
18:26:24 gigamonkey [~user@adsl-99-2-149-140.dsl.pltn13.sbcglobal.net] has joined #lisp
18:26:32 -!- mega1 [~quassel@catv4E5CABA2.pool.t-online.hu] has quit [Quit: No Ping reply in 180 seconds.]
18:26:44 mega1 [~quassel@catv4E5CABA2.pool.t-online.hu] has joined #lisp
18:26:55 -!- dialtone [~dialtone@unaffiliated/dialtone] has quit [Read error: Connection reset by peer]
18:27:11  we love him like everybody loves Sussman
18:27:43  he also started two of the computer science programs in Sweden, mine is one of them
18:28:29 -!- Xach [~xach@cpe-72-227-90-1.maine.res.rr.com] has left #lisp
18:28:49  time to sleep
18:28:53  g'night
18:29:13  anyone into OpenGL programming? is cl-opengl stable and usable more or less like the C libarary?
18:30:12 dabd [~dabd@a81-84-79-168.cpe.netcabo.pt] has joined #lisp
18:31:52 -!- Makoryu [~vt920@pool-71-174-191-10.bstnma.fios.verizon.net] has quit []
18:32:33 dialtone [~dialtone@70.36.244.244] has joined #lisp
18:32:33 -!- dialtone [~dialtone@70.36.244.244] has quit [Changing host]
18:32:33 dialtone [~dialtone@unaffiliated/dialtone] has joined #lisp
18:32:40  z0d: I believe that you will find the #lispgames crowd to be a more... accurately self-selected audience for such a question.
18:33:11  (I also hear that Mac OSX likes to be actively annoying when it comes to OpenGL and lisp.)
18:36:22 -!- syamajala [~syamajala@68-116-188-249.dhcp.oxfr.ma.charter.com] has quit [Remote host closed the connection]
18:37:12  z0d: cl-opengl is actively developed, and stable enough to develop on.
18:37:21 stassats [~stassats@wikipedia/stassats] has joined #lisp
18:37:46  ok, thanks guys
18:37:48 -!- jhalogen [~jake@cpe-98-154-250-117.socal.res.rr.com] has quit [Ping timeout: 260 seconds]
18:37:49  z0d: as far as OSX goes, there are issues involving using GLUT, because of the way threading on Clozure CL works, but there are workarounds.
18:37:50 daniel_ [~daniel@p5082E4AA.dip.t-dialin.net] has joined #lisp
18:37:51  I'll give it a try
18:38:03  I'm using SBCL on Linux
18:38:03 -!- hankhero [~Adium@c-b21fb7c4-74736162.cust.telenor.se] has quit [Read error: Connection reset by peer]
18:38:05  z0d: #lispgames, like nyef said, is a very good place for conversations about this :)
18:38:29  z0d: SBCL on Linux is more than okay, imo.
18:38:59 potatishandlarn [potatishan@c-b21e593f-74736162.cust.telenor.se] has joined #lisp
18:39:03 -!- hefner [~hefner@ppp-58-9-117-237.revip2.asianet.co.th] has quit [Quit: Arrrr!]
18:39:03 -!- gigamonkey [~user@adsl-99-2-149-140.dsl.pltn13.sbcglobal.net] has quit [Ping timeout: 265 seconds]
18:39:13 -!- daniel__ [~daniel@p5082C4B9.dip.t-dialin.net] has quit [Ping timeout: 240 seconds]
18:39:24 -!- Stattrav [~Stattrav@117.192.143.77] has quit [Remote host closed the connection]
18:39:46  So, I was taking a bit of a break from banging my head against the wall of this GC bug I've been trying to nail for the past few days, and I realized that I'd actually written the truth down in my documentation and then proceeded to attempt to be clever with it instead of doing the right thing.
18:40:18  There are -five- possible interior-pointer registers.
18:40:43 -!- mega1 [~quassel@catv4E5CABA2.pool.t-online.hu] has quit [Remote host closed the connection]
18:40:43  Being pc, npc, link-register, counter-register, and reg_LIP.
18:41:00  And they should all be treated the same way.
18:41:55  No cleverness. No deciding that one of them can only pair certain registers, just all of them dealt with the same way.
18:42:04 -!- caoliver [~oliver@75-134-208-20.dhcp.trcy.mi.charter.com] has left #lisp
18:42:48 -!- barcon332 [~barcon332@208.89.210.254] has quit [Ping timeout: 240 seconds]
18:43:15  one ring to rule them all
18:44:08  ... And still didn't fix it. :-/
18:45:57 eugu [~Miranda@212.1.246.237] has joined #lisp
18:48:13 hankhero [~Adium@c-b21fb7c4-74736162.cust.telenor.se] has joined #lisp
18:52:22 syamajala [~syamajala@68-116-188-249.dhcp.oxfr.ma.charter.com] has joined #lisp
18:54:06 -!- syamajala [~syamajala@68-116-188-249.dhcp.oxfr.ma.charter.com] has quit [Remote host closed the connection]
18:58:32  On the upside, the GC part of this fiasco is now obviously correct.
19:00:15  ... Which leaves me with only one place I know to be ignoring the invariants.
19:01:44 -!- _KY_ [YKY@unaffiliated/-ky-/x-0649748] has left #lisp
19:03:40  nyef: and that place is?
19:04:53 -!- konr [~user@201.82.129.247] has quit [Ping timeout: 264 seconds]
19:05:37  Anywhere in src/assembly/ppc/arith.lisp that calls a static-function.
19:07:17  What's the invariant?
19:08:25  nyef: you've left everyone at the edge of their seats.
19:08:30  When reg_LIP or the program counter point to within an object that can be moved, one of the boxed registers must also point to the same object.
19:08:48  (Actually applies to any interior pointer.)
19:09:58 -!- cmm [~cmm@109.64.200.182] has quit [Ping timeout: 276 seconds]
19:10:05 holycow [~new@poco208-2.fredcanhelp.com] has joined #lisp
19:10:34 lispm [~lispm@d226022.adsl.hansenet.de] has joined #lisp
19:10:40 cmm [~cmm@bzq-109-64-200-182.red.bezeqint.net] has joined #lisp
19:10:57 -!- Bobrobyn [~rsmith05@CPE0015e9d40d4f-CM001ac30e9df0.cpe.net.cable.rogers.com] has quit [Ping timeout: 265 seconds]
19:10:57 -!- REPLeffect [~REPLeffec@69.54.115.254] has quit [Ping timeout: 265 seconds]
19:12:22  Doesn't quite fit with the failure mode, though, for some reason. :-/
19:14:11 -!- lispm [~lispm@d226022.adsl.hansenet.de] has quit [Remote host closed the connection]
19:14:13 ziarkaen [~ziarkaen@87.112.72.23.plusnet.ptn-ag2.dyn.plus.net] has joined #lisp
19:14:19  Ah, ok.   I see that cmucl has a comment that the register pair for reg_LIP is sometimes -1, which means no register pair.  Currently, the code decides not to change reg_LIP if so.  I guess that's wrong.
19:14:20 -!- moocow [~new@poco208-2.fredcanhelp.com] has quit [Ping timeout: 265 seconds]
19:16:20  No, that's right.
19:16:36  The problem comes when reg_LIP should -have- a register pair, but doesn't.
19:16:47  Or any other interior pointer, such as the program counter.
19:17:03 ephcon [~ephcon@c-24-34-195-64.hsd1.ma.comcast.net] has joined #lisp
19:19:48  Isn't the PC always paired with reg_CODE?
19:19:53  No.
19:20:15 -!- tsuru- is now known as tsuru`
19:20:28  Well, I guess that's a huge problem then.
19:20:28  (Trivially demonstrated: You don't update reg_CODE in the same instruction as you jump to a new function during funcall.)
19:21:49  Hmm.  That assumption is used in Cheney GC too, and I haven' touched that code.
19:22:04  It also turns out to be correct -most- of the time.
19:22:23  Consider how wide the window is for a bogus reg_CODE during funcall.
19:22:32  It's three instructions wide -at worst-.
19:22:47  Are you really going to take an asynchronous interrupt in that window?
19:23:02  And then run a GC while the handler is active?
19:23:04 roygbiv [~None@pdpc/supporter/active/roygbiv] has joined #lisp
19:23:23  Yeah, very unlikely.  But I have had some weird crashes with sparc, but rarely, never repeatable.
19:23:27  The First Law of GC: it always happens at the worst possible time.
19:23:32  (Yes, this is two parts. You have to take an interrupt in a vulnerable window, and then you have to GC in the handler.)
19:23:34 Bobrobyn [~rsmith05@CPE0015e9d40d4f-CM001ac30e9df0.cpe.net.cable.rogers.com] has joined #lisp
19:23:36 -!- HET3 [~diman@w220.engin.cf.ac.uk] has quit [Quit: Leaving]
19:23:45  You also have a wider window during return.
19:24:04  What's the vop for a funcall?  (Can't remember.)
19:24:07  I suspect that the whole reason I'm seeing it so repeatably is that I have multiple threads.
19:24:31  Ah, CALL, CALL-NAMED, TAIL-CALL, TAIL-CALL-NAMED, CALL-MULTIPLE, TAIL-CALL-MULTIPLE, CALL-VARIABLE, NAMED-CALL-VARIABLE... I think.
19:24:55  nyef: PPC ?
19:25:01  Have a look for define-full-call in call.lisp, for whatever backend it is.
19:25:08  nyef: care to tell you tale at the BLM?
19:25:25  Fare: Yeah, I'm on PPC. x86oids get a free pass on a whole -lot- of crap.
19:25:52  Umm... Maybe?
19:26:01  You looking for a lightning-talk sort of thing?
19:26:10  nyef: On sparc, it looks like code-tn is set in the branch delay slot of the jump.
19:26:23  rtoym: sparc has an NPC register, doesn't it?
19:26:29  Yes.
19:26:47  NPC register is subject to the same constraints as the program counter, isn't it?
19:27:19  Suddenly, reg_CODE needs to have two different values at the same time.
19:27:21  Should be, I guess.  It's always supposed to point one word past the PC.
19:28:00  It doesn't point such in a branch-delay situation, though.
19:28:22 WOG [~OsamaBinW@24-177-10-101.dhcp.nwtn.ct.charter.com] has joined #lisp
19:28:41  All of this is moot, by the way, if fake_foreign_function_call() or whatever it is in cmucl detects and cleans up such situations.
19:28:45 ehu [~ehuels@ip118-64-212-87.adsl2.static.versatel.nl] has joined #lisp
19:30:27  minion: Paste 112474?
19:30:40  Damn it, minion!
19:30:56  http://paste.lisp.org/display/112474
19:31:15  My current interrupt-context scavenging.
19:31:43 -!- SandGorgon [~OmNomNomO@122.162.127.184] has quit [Ping timeout: 265 seconds]
19:32:58  I'm not really sure where nPC points to in a branch/delay case.
19:33:22  MIPS has one too, I believe.
19:33:38  I don't think cmucl's fake_foreign_function_call does anything to handle the situations.
19:33:55  On the branch instruction it's PC, PC+4. On the delay slot, it's PC+4, dest. On the branch target, it's dest, dest+4.
19:34:42  And how they're used when resuming from an interrupt is to jump to the PC, then in the branch delay slot of that jump, jump to the next-pc.
19:34:54 -!- homie [~user@xdsl-87-79-190-34.netcologne.de] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)]
19:35:41  It's a little mind-twisting.
19:36:25  Can an interrupt happen between the jump and the delay slot?
19:36:57  Depends on the architecture, and possibly on how long it takes for the interrupt enable to take effect.
19:37:30  MIPS has some sort of "write hazard" thing where a write to the interrupt control doesn't actually "happen" for a few instructions.
19:37:41 mstevens [~mstevens@fsf/member/pdpc.active.mstevens] has joined #lisp
19:38:03 syamajala [~syamajala@68-116-188-249.dhcp.oxfr.ma.charter.com] has joined #lisp
19:38:59  Oh! The other factor with interior pointers is that you should only base them on boxed registers whose value is a pointer.
19:39:20  It's a missed test in the stock SBCL versions.
19:39:35  But the argument for why it's necessary is also trivial (commentary in my paste above).
19:40:40  Your next potential problem is FDEFINITION objects and raw-addrs.
19:40:53  Don't know if that's how cmucl does things.
19:41:21  But SBCL has a nasty tendency to just load the raw-addr out of an fdefn into reg_LIP and branch.
19:41:24 -!- Komi [Komi@62.32.147.35] has quit [Ping timeout: 248 seconds]
19:42:10  Which is fine... if you're calling a closure or a funcallable-instance, due to the closure_tramp.
19:43:39 -!- tayloj [~tayloj@graciliraptor-89.dynamic2.rpi.edu] has quit [Quit: Leaving.]
19:45:23 rajesh [~rajesh@nylug/member/rajesh] has joined #lisp
19:45:33  Fare: When is BLM this month, anyway? The 26th?
19:46:42 TeMPOraL [~user@188.147.196.168.nat.umts.dynamic.eranet.pl] has joined #lisp
19:46:50 -!- syamajala [~syamajala@68-116-188-249.dhcp.oxfr.ma.charter.com] has quit [Remote host closed the connection]
19:47:23  nyef: yup
19:48:34 -!- hjpark [~user@116.40.135.21] has quit [Remote host closed the connection]
19:48:51  Okay, a GC occurs, this time through generation 7.  After resuming, an illegal instruction trap happens. The backtrace generated errors out on the escaped frame due to the PC-OFFSET not being within the code object...
19:49:12  ... and the PC and "computed return" (link register) have the same value.
19:49:42 Stattrav [~Stattrav@117.192.130.245] has joined #lisp
19:49:52  So, the code object moved out from under.
19:50:17  cmucl does pretty much what your paste does for reg_LIP.  reg_LR and reg_CTR are handled this way, though.
19:50:35  Are or aren't?
19:50:45  Oops.  Aren't.
19:51:11  Yeah, I hadn't been, but then I realized that an interior pointer is an interior pointer, and trying to be clever about it only invites subtle bugs.
19:51:21  As if GC bugs weren't subtle enough already.
19:52:07  I'll need to look through a sparc book to see if interrupts can happen between the jump and the delay slot.
19:52:27  Oh, hell. There's a race in call_into_c.
19:52:38  I'm not going to worry about LR and CTR because, unfortunately, for all intents cmucl/ppc is dead because I don't have access to one anymore.
19:52:52  That's fair enough, I guess.
19:54:21  Hrm. Race in call_into_c is on the outbound side, and would modify the C function being called randomly, but is otherwise not horrible.
19:54:27  Han Hubner used to give me access, but it's no longer available.
19:54:35  And certainly couldn't cause this effect.
19:54:42  What's the issue with raw-addr slots?
19:55:02  The typical approach is to load them directly into reg_LIP and use them as a branch address.
19:55:27  Without loading the function object first.
19:55:59  Don't know if you've got that in cmucl, but here it's a clear violation of the invariants for interior pointers.
19:57:32  In the full call stuff, the raw-addr is loaded in to a descriptor and jumped to.
19:57:42  Yeah, you're on sparc.
19:57:46  Forgot about that.
19:57:52  Try mips or ppc.
19:59:57 AQUAF|RE [~AQUAF_RE@119.155.115.95] has joined #lisp
20:00:03  Hello !
20:00:11  !Rules
20:00:16  I need some help
20:00:41  so is the solution to somehow make function return/calls specially pseudo-atomic?
20:00:41 zomgbie [~jesus@mk092248000189.a1.net] has joined #lisp
20:01:07  that's one solution...
20:01:08  sorry dont know about that
20:01:12  ok fine
20:01:16  Fare: That's one possible solution. Safepoints are another. And following the rules by making sure the function is loaded first is a third.
20:01:57  hello ?
20:02:02  need some help plz
20:02:04 *nyef* went with the third, as it was minimally invasive... and embedding PA during a call sequence is irritating.
20:02:27  AQUAF|RE: Yes, you said that already.
20:05:26  nyef:  AFAICT, on sparc, the lip is only used in 3 places:  call-out, set-fdefn-function, the print vop.  There's a comment about it in lisp-jump, but it doesn't appear to be used.
20:05:41 Xach [~xach@cpe-72-227-90-1.maine.res.rr.com] has joined #lisp
20:05:57 *rtoym* makes me wonder how the register pair is ever -1.  It must have happened at one point.
20:06:31  p_l: hey, you around?
20:06:35  If the value of reg_LIP is numerically less than the value of all of the boxed registers.
20:06:39  Hi Fare!
20:06:47  (I'm inklesspen, but at work, so different nick.)
20:06:51  or, yeah, Fare
20:07:07  Fare: FWIW, the version of asdf at http://common-lisp.net/project/asdf/asdf.lisp says "VERSION:2.003"
20:07:14  I wrote up a blog post with my bcrypt experiences: http://www.letsyouandhimfight.com/2010/07/14/cl-bcrypt-a-first-attempt/
20:07:21  Okay, only places that the link register is used from lisp code are when it points to an assembly-routine (thus will not be paired during scavenge and will not move), in the LISP-RETURN macro, which does things properly, and for calling assembly-routines with a call-style of :raw or :none.
20:07:32 -!- nixeagle [~user@Wikimedia/Nixeagle] has quit [Read error: Connection reset by peer]
20:07:41  Xach: the weird styling issue should be fixed.. Nothing like inserting more python in the pipeline ;)
20:07:44 Sikander [~soemraws@5356ECA7.cable.casema.nl] has joined #lisp
20:07:55  hi Sikander
20:08:00  Hi
20:08:10  rtoym: AIUI, sparc has a jump instruction that takes a register and an immediate offset?
20:08:12  Does anyone know if cl-ncurses is still being maintained?
20:08:26  LiamH: How's life?
20:08:31  pkhuong: sweet
20:08:55  thanks for putting up with the madness.
20:08:57  I have some problems getting the tests of cl-ncurses to run
20:09:04  Sikander: good, busy.  New grid coming soon.
20:09:18  Xach, oops. Fixed.
20:09:20 nixeagle [~user@Wikimedia/Nixeagle] has joined #lisp
20:09:28  LiamH: I've been using grid with slicing quite extensively lately. It works great!
20:09:49  whats grid?
20:09:58 -!- zomgbie [~jesus@mk092248000189.a1.net] has quit [Ping timeout: 260 seconds]
20:10:05  Sikander: check out the "convert" branches in gsd and gsll, if you're feeling adventurous.
20:10:27 tayloj [~tayloj@clip-point-02.dynamic2.rpi.edu] has joined #lisp
20:10:31  Revolve: generalized arrays, particularly for numerical applications.
20:10:38  When calling a :raw or :none style assembly-routine, the link register ends up with reg_CODE (or whatever the current pair for the program counter is) as a pair, so should be moved appropriately.
20:10:39  LiamH: I've noticed that, at least on sbcl, slicing basically returns a copy of the slice. Is this always true, or can I count on it always just copying?
20:10:48  LiamH: Hmmm, I actually might.
20:10:58  These calls are -never- generated without reg_CODE being valid.
20:11:00 -!- Snamich [~Snamich@161.210.164.242] has quit [Quit: Snamich]
20:11:09  nyef: You going to this month's BLM?
20:11:15  So no one knows anything about cl-curses or has used it successfully?
20:11:25  And reg_CODE is mentioned in precisely one assembly-routine, and that's UNWIND.
20:11:26  Sikander: Yes, it will always copy.
20:11:38  nyef:  Do you know why set-fdefn-function needs to make the lip the closure_tramp address?  (There's a potential issue with the current sparc implementation of that.)
20:11:40  Xach: I plan to. We'll see what happens on the day, though.
20:11:58  LiamH: great, thanks.
20:12:03 *Xach* also plans to for sure
20:12:12 *Xach* nearly has the train ticket(s) in hand
20:12:23  Sikander: I've revamped all of foreign-array and GSLL arrays.  All the basics are in foreign-array now, and GSLL just sticks its needed metadata into a slot.
20:12:32  I need a family tree script for my site
20:12:44  could i get soem help in that ?
20:12:45  This makes it easy(ier) to call different foreign libraries.
20:13:01  I hope.
20:13:11  AQUAF|RE: If you want one, I think you will have to be the first to write it.
20:13:35  rtoym: On SBCL there are three different kinds of function objects. SIMPLE-FUNs, CLOSUREs, and FUNCALLABLE-INSTANCEs. The first slot in all of these objects points to a SIMPLE-FUN which is the actual function to run (or the funcallable-instance tramp).
20:13:48  LiamH: you've really put together quite an awesome lisp library around GSL... Do you use GSL that often?
20:13:53  Sikander: the foreign arrays are now built on static-vectors, so there's no pinning needed.
20:13:53  where to write it ?
20:14:00  I am poor in programming
20:14:02  rtoym: The "named" call VOPs don't do this indirection, though, they just grab the raw-addr and run with it.
20:14:05  AQUAF|RE: in a text editor, most likely.
20:14:27  i cant my fellow being !
20:14:34  i am poor in programming
20:14:42  or u may say noob :)
20:14:44  Sikander: No, but partly that's because GSLL is relatively new.  I could frequently make use of various parts of GSL, hence GSLL.
20:14:45  AQUAF|RE: This channel is for people who want to get better at Common Lisp. Do you want to get better?
20:14:56  So if a fdefinition gets pointed to a non-simple-fun, it must set the raw-addr to the closure_tramp to load the function object, do the indirection, and execute the correct function.
20:15:07  mmmm,,,,, C is too tough
20:15:26  LiamH: Well, just to let you know, I've been using GSLL for reading my experimental data, processing it (filtering, extracting parts, transforming and fitting).
20:15:29  AQUAF|RE, ask your teacher for help
20:15:56  lol i have no teacher of programming , i am a commercial pilot by profession
20:15:59  Sikander: and as I think I've explained, my aim is to have interfaces (or at least make it easy to build an interface) to all kinds of numerical libraries.
20:16:06  Sikander: good to hear that.
20:16:15  AQUAF|RE: If you don't want to learn Common Lisp, I don't think you should stay in this channel.
20:16:34  AQUAF|RE, maybe get one? If you suck at programming, you might want to try HtDP (ask around #racket).
20:16:58  LiamH: That would be great. I was looking a bit at if fftw could be wrapped in as well because it has a much faster fft than gsl
20:17:03  Sikander: I'm curious to know, what parts are you using?  I mean other than grid.
20:17:34  Sikander: Yes, I've thought fftw would be a nice next candidate.
20:17:35  LiamH: fitting, statistics and fft
20:17:54  nyef:  Yes, sparc has a jump instruction where the source is the sum of 2 regs or a reg and an offset.
20:18:23  Mmm. And I could almost look forward to doing call VOPs for ARM, where the program counter is an addressable GPR.
20:18:28  LiamH: I'm just reading plain old ascii data tables as matrices, and process the columns as needed. Also, I processed some CCD images and did 2D fitting.
20:18:38  so cant i get scripts ?
20:18:40  (or would that be 3d fitting, I never know)
20:19:15  AQUAF|RE: This is not a place to get scripts for websites.
20:19:23  oh sorry
20:19:28  i am so sorry for that !
20:19:31  LiamH: Actually, I also used ode before.
20:19:33  No problem! Now you know.
20:19:39  yes !
20:19:43  Sikander: Good.  One area I'd like to see built up is data input/output.  I have some very preliminary work: http://repo.or.cz/w/gsd.git/blob/c8cb7aec259e7e6bf9e5aafb2b36f8d9073dcdf2:/grid/read-write/read.lisp
20:19:53  but its not good
20:20:07  if someone has something so he must help others :)
20:20:10  isn't it ?
20:20:24 killerboy [~mateusz@smrw-91-193-87-5.smrw.lodz.pl] has joined #lisp
20:20:51  Sikander: I remember, you went from not knowing of GSLL to working ODE solution in 5 minutes.
20:21:02  wow
20:21:15  nyef:  On sparc the "named" call loads the raw-addr into a descriptor and later on jumps to that reg.  So I guess that's different from ppc, and should be safe.
20:21:28  LiamH: Nice. I have some very quick'n'dirty reading/writing functions I'm trying to clean up.
20:21:32  LiamH: Yep, that's true
20:21:38 tritchey [~tritchey@205.233.9.181] has joined #lisp
20:22:05  LiamH: My GSLL solution was faster, shorter and immediately gave correct results, as opposed to my colleague's python/numpy code. I was hooked!
20:22:36  rtoym: Yeah, sparc is probably safe as far as function-calling due to that quirk... Modulo the whole program-counter-as-interior-pointer thing.
20:22:38  Sikander: are you using SBCL?  If so, it's reasonable to pull the convert branches now.  For imps that static-vectors doesn't support, it would be a bad idea.
20:23:03 HET3 [~diman@cpc1-cdif12-2-0-cust125.5-1.cable.virginmedia.com] has joined #lisp
20:23:05  LiamH: oh, yeah. I sometimes feel like I waste half of my coding time massaging shapeless data into useful tables.
20:23:16  LiamH: Yes, SBCL. Ok, I'll  try it
20:23:23 -!- Davse_Bamse [~davse@82.103.143.161] has quit [Quit: leaving]
20:24:06  pkhuong: Yup, I (and I bet a lot of us) have a lot of ad-hoc converters, readers, etc., and I'd like to pull the ideas together into the grid system.
20:24:19  LiamH: Is there somewhere I can submit some reading/writing functions for you to look over? Don't know if they are "fast" though...
20:24:50 -!- mstevens [~mstevens@fsf/member/pdpc.active.mstevens] has quit [Quit: leaving]
20:24:54 Beetny [~Beetny@ppp118-208-34-80.lns20.bne1.internode.on.net] has joined #lisp
20:24:57  Sikander: be forewarned, some things have changed (no more maref, vector-double-float etc. are in grid package)
20:25:09 -!- Beetny [~Beetny@ppp118-208-34-80.lns20.bne1.internode.on.net] has quit [Read error: Connection reset by peer]
20:25:34  My problem might be weirder because I also have to handle a very large number of test parameters.
20:25:36 -!- HET2 [~diman@cpc1-cdif12-2-0-cust125.5-1.cable.virginmedia.com] has quit [Ping timeout: 246 seconds]
20:25:38  Sikander: yes, absolutely, don't care about fast yet.  Post to GSLL mailing list or lhealy  -at- common-lisp.net
20:26:35 -!- Kolyan [~nartamono@93-80-65-136.broadband.corbina.ru] has quit []
20:26:35  pkhuong: not sure what you mean, but I've thought about being able to import/export structs, arrays of structs, and SQL tables.  No code to show yet though.
20:26:41  Is it possible to write threaded applications intended for multicore machines with common lisp?
20:26:51  Revolve: yes.
20:26:54  woot!
20:26:55  minion: tell Revolve about pcall
20:27:00  thanks
20:27:04  I'll just google it LiamH
20:27:19  Revolve: yup, minion's asleep I guess
20:27:26  Revolve: some people even use it on clusters or bluegene.
20:27:34  wow
20:27:47  see, I was thinking of using it in a very powerful environment, but nothing like a supercomputer
20:28:14  I take it there a well paying engineering jobs for lisp programmers then?
20:28:22 zomgbie [~jesus@h081217133138.dyn.cm.kabsi.at] has joined #lisp
20:28:43  Sikander: so how goes your day job?
20:28:53 -!- HET3 [~diman@cpc1-cdif12-2-0-cust125.5-1.cable.virginmedia.com] has quit [Ping timeout: 264 seconds]
20:29:03  I doubt you will find much Lisp in a HPC setting. I've yet see anything really in the CFD world.
20:29:06 -!- killerboy [~mateusz@smrw-91-193-87-5.smrw.lodz.pl] has quit [Quit: bye]
20:29:47 Beetny [~Beetny@ppp118-208-34-80.lns20.bne1.internode.on.net] has joined #lisp
20:30:12 -!- Stattrav [~Stattrav@117.192.130.245] has quit [Ping timeout: 265 seconds]
20:30:36  HPC ? CFD ?
20:30:56  HPC = High Performance Computing, CFD = Computational Fluid Dynamics.
20:30:56 syamajala [~syamajala@68-116-188-249.dhcp.oxfr.ma.charter.com] has joined #lisp
20:31:12  HPC, you should have seen the previous BLM.
20:31:13  LiamH: Eh, problematic. my experiment won't work properly because I have equipment that's older than I am.
20:31:31  Fare: the FPGA stuff? :)
20:31:46  Sikander: :-(
20:31:51  CFD, I don't know, but I saw people use expert systems to super-optimize models of fluid dynamics for nuclear reactors.
20:31:54  LiamH: Hmm, in gsll, when I git pull origin convert, I get two conflicts: array-structs.lisp and init.lisp.
20:32:17  Fare: Sounds very cool. What project was that?
20:32:20  Sikander: git checkout convert
20:32:50 Blkt` [~user@dynamic-adsl-78-13-247-126.clienti.tiscali.it] has joined #lisp
20:33:08  I'm not sure why there'd be a conflict, unless you had uncommitted mods to those files.
20:33:12  LiamH: ah right. Dammit, git is just too damn complicated.
20:33:23  Nope, I didn't.
20:33:48  Currently, my solution to fix problems is to just delete the whole damn tree and get it fresh. :(
20:33:52  Sikander: Lisp is more complicated than git.  It's just a matter of using it often enough that you remember what to do.
20:34:05  hypno: something at EDF, years ago.
20:34:11  Sikander: not a bad way to go.
20:34:13  Fare: Most people that deals with this are using Fortran or C/C++. I guess CL could work too, but you need to know a lot about the code, such a cache locality, inlining, etc. I guess numeric math DSL or somesuch could do.
20:34:17 HET2 [~diman@cpc1-cdif12-2-0-cust125.5-1.cable.virginmedia.com] has joined #lisp
20:34:24  not sure whether they used Lisp or rolled their own expert system in C/C++
20:34:32  LiamH: wait, git checkout convert would only work if git branch mentions the existence of that branch
20:34:46  LiamH: but I don't have that branch.
20:34:52 -!- tritchey [~tritchey@205.233.9.181] has quit [Quit: tritchey]
20:35:05  hypno, also, stalin was written for high performance image processing, I believe.
20:35:12  Sikander: Right.  git branch -b convert -t origin/convert, I think
20:35:38  no -b
20:35:40 -!- syamajala [~syamajala@68-116-188-249.dhcp.oxfr.ma.charter.com] has quit [Remote host closed the connection]
20:36:08  arrgh.  git checkout -b convert -t origin/convert
20:36:16  nyef:  Don't quite follow you on the set-fdefn-function vop.  On sparc, it uses the lip register.  But set-fdefn-function should just set the raw-addr appropriately.  It should work just as well if any descriptor reg was used instead of lip, right?
20:36:20 -!- Blkt [~user@dynamic-adsl-78-13-252-73.clienti.tiscali.it] has quit [Ping timeout: 248 seconds]
20:37:03  Fare: I've yet to take a look at Stalin (it is Scheme after all). I think a CL solution should use a DSL that goes down to newest and greates CPU instructions, etc tho.
20:37:51  On sparc, yes, because the closure_tramp is boxed like any other function.
20:37:57  On any other platform, no.
20:38:04 Avisch [~NONE@cpe-24-93-16-141.rochester.res.rr.com] has joined #lisp
20:39:22 -!- AntiSpamMeta [~MetaBot@unaffiliated/afterdeath/bot/antispambot] has quit [Excess Flood]
20:40:11 -!- lhz [~shrekz@c-dba672d5.021-158-73746f34.cust.bredbandsbolaget.se] has quit [Quit: Leaving]
20:41:27 AntiSpamMeta [~MetaBot@unaffiliated/afterdeath/bot/antispambot] has joined #lisp
20:41:45  LiamH: ok, that worked. Where do I get #:foreign-array from?
20:41:51  hypno: probably. With recursive blocking and everything, I'm hoping that it'll be enough to only handle very large basic blocks in the DSL (or very restricted extensions like hyperblocks).
20:42:06  Sikander: convert branch of gsd
20:42:19 -!- peterhil [peterhil@YKXVI.gprs.sl-laajakaista.fi] has quit [Ping timeout: 240 seconds]
20:42:28  LiamH: Hmmm, I have it... perhaps it's time to remove some fasls...
20:42:51  Sikander: yeah, I usually wipe fasls and start over for big changes.
20:43:09  pkhuong: Hmm. You are working on a DSL for HPC?
20:44:35  Sikander: re code contributions: I've added you as a committer to the gsd repo.  Make your own branch and add to gsd/read-write if you like.
20:44:40  LiamH: Ok, it's compiling. What things should I look out for? Are there any changes in the interface, or things you'd like me to keep an eye out for?
20:45:07  LiamH: Oh, right, thanks. I'll put something there in the next days.
20:45:40  Yes, changes in the interface.  Use grid:gref instead of gsl:maref, that's the big one.
20:46:39  hypno: I wrote a compiler for one in Feeley's compiler course. The idea was to take a large expression and perform CSE on it, reorder array accesses to make sure accesses are streaming, aggressive constant propagation, regalloc-aware (MAXLIVE, actually) instruction selection and spilling/regalloc with an algorithm specialised for basic blocks.
20:46:44  Ok, I'll be working with it the coming days, thanks
20:46:54  Also to make a foreign-array (aka GSL array) use make-foreign-array or make-grid.  #m still works too, but you can't enter initial data with a flat list anymore, you need the embedded row lists for matrices.
20:47:38  pkhuong: heh, very cool.
20:47:39  LiamH: So make-marray is no more eh
20:47:49  nyef:  Ok.  Thanks.   That is probably the source of the occasional lip_register_pair = -1 issue I saw long ago.  Note also that I changed cmucl/ppc to make closure_tramp and undefined_tramp to be boxed functions.  To be more like sparc.  Turns out this was probably not a good idea because of the way it interacts with linkage tables.
20:47:51  Sikander: OK very interested to see how you like it.  I do intend to write docs so that people know why this is better.
20:47:52  The only thing I couldn't get to work was CSE, because I was too aggressively performing algebraic transformations to simplify the code, and recognize identical or ordered expressions (the latter is weird, but being able to tell that 2*stride+4 > stride+10 is essential to reorder vector accesses)
20:48:00  Sikander: right, marray is gone
20:48:31  LiamH: Ok, thanks. I'll figure it out, or bother you some more later on ;)
20:48:53  And then I started worrying about I$, and wrote a loop inference pass. That gave . Once I fixed the movsd into movaps, I had performance identical to gcc's on this admittedly trivial problem (dot product for vectors of length 64)
20:48:57  Sikander: you know, free software, documentation is the last thing to happen.
20:49:01  So, no one here knows if cl-ncurses is maintained/working etc?
20:49:42  rtoym: It interacts with linkage-tables? What?
20:49:46  LiamH: Yeah, I know. And if I remember correctly, I should still submit the GSL FFT tests for GSLL :(
20:50:03  Sikander: I believe I've heard that people have used it successfully.
20:50:09  Oh, god...
20:50:27  Sikander: well I was going to port the GSL tests.  I've ported a bunch of other tests, but not the FFT tests yet.
20:50:30  Hrm, no, that goes through the CTR...
20:50:51  reb: Ah... I'd like to understand why the tests don't even run for me, so I'd need to talk to someone with first-hand experience. Thanks, though.
20:50:59  It'd be at risk of rebasing, if that were the case, but not to the extent that I'm seeing.
20:51:35  LiamH: Ah, ok. And have you found a good documentation tool yet? I'm still looking for something for myself as well...
20:51:59 ejs [~eugen@92.49.220.56] has joined #lisp
20:52:00  nyef: The raw-addr slot is loaded with the real closure_tramp (which is treated as data).  If the C code is compiled so  that closure_tramp moves, the core still has functions pointing to the old closure_tramp.  This causes some warnings to be printed during GC or building or something.  It's been a while since I've seen this.
20:52:05  LiamH: Something to extract comments from functions and make a nice html/TeX manual?
20:52:14  Sikander: yes, I think.  http://orgmode.org/
20:52:31  hypno:  for what I had by the end of the semester, with a couple reports in french.
20:52:39 -!- Jabberwockey [~Jens@port-5104.pppoe.wtnet.de] has quit [Quit: Verlassend]
20:52:40  Sikander: but I've only used it for notetaking, not for publication of doc
20:52:56  rtoym: when using cross-build-world.sh, how do I achieve the equivalent of "-u Don't build CLX, CLM, or Hemlock", which is a an option of build.sh ?
20:53:19  rtoym: Oh. Right, cmucl doesn't re-build the world after rebuilding the C code.
20:53:36  fe[nl]ix: Yes.  And pcl isn't built either.
20:53:45  LiamH: Argh, and I still don't use emacs :(
20:53:46  In which case, why aren't the closure_tramp and such in read-only space instead of the runtime?
20:54:20  nyef:  I think I changed the script to (optionally?) rebuild the world after rebuilding the C code.
20:54:46  Sikander: well I think for these changes I'll just update the existing texinfo doc, I'm not ready to sink the time into converting yet
20:54:55  pkhuong: ah, heh. thanks.
20:55:33  Mmm. I am drawing a complete blank on how the link register could be getting corrupted while it's still live.
20:55:40  fe[nl]ix: Oops.  I mean cross-build-world doesn't compile any of those.  You get just Lisp and the compiler.  No addons, not even pcl.  So you need to do a regular build afterwards to get pcl.
20:55:48  LiamH: You have quite some documentation in documentation strings of each function. Do you have a tool to extract that to texinfo or did you do that by hand, or...?
20:55:57 -!- jajcloz [~jaj@pool-98-110-225-6.bstnma.fios.verizon.net] has quit [Quit: jajcloz]
20:56:27  nyef:  Why would closure_tramp be in read-only?  It never gets moved because it's not in any of the normal spaces.
20:56:34 -!- NNshag [user@lns-bzn-27-82-248-27-130.adsl.proxad.net] has quit [Ping timeout: 276 seconds]
20:56:51  Sikander: sadly nothing yet.  There are a number of tools out there but I haven't tried them, I don't know how good they are (the names Albert and TINAA come to mind, don't know if those are right).
20:57:00  So that it also doesn't get moved when you rebuild the runtime?
20:57:10  LiamH: So you copy/paste by hand, then?
20:57:13 -!- HET2 [~diman@cpc1-cdif12-2-0-cust125.5-1.cable.virginmedia.com] has quit [Ping timeout: 260 seconds]
20:57:34  Sikander: yes, but I haven't done much of that.
20:57:38  rtoym: http://paste.lisp.org/+2EOX/2
20:57:55  nyef:  Oh.  I misinterpreted your comment.  Yeah, read-only would be better.  Don't know how to do that  though.
20:58:27  I do.
20:58:51  Pray tell, please!
20:58:58  Step one is to recreate it as an assembly-routine.
20:59:16  LiamH: You want read.lisp to read generic objects, right? I mean, a grid is not necessarily an array or table of numbers?
20:59:23  Step two is to change any fixups that refer to it to pick up the assembly-routine version.
21:00:08  Umm... I don't -think- closure_tramp is referred to from genesis...
21:00:31  Oh. It is.
21:00:46  Sikander: yes, I think it's reading in a list now, because it served my immediate purpose.  However, it should read using (setf grid:gref) or pass that list as an :initial-contents.
21:00:47 jmcphers [~jmcphers@218.185.108.156] has joined #lisp
21:00:54  Hrm. That's a little broken. Doesn't account for funcallable-instances.
21:00:58 -!- ziarkaen [~ziarkaen@87.112.72.23.plusnet.ptn-ag2.dyn.plus.net] has quit [Remote host closed the connection]
21:01:07 -!- AQUAF|RE [~AQUAF_RE@119.155.115.95] has left #lisp
21:01:31 -!- moah [~gnu@188.109.158.189] has quit [Quit: Leaving]
21:01:37  nyef:  Ah, I once asked Rob why closure tramp was not an assembly routine.  I think he said it just wasn't.
21:01:57 HET2 [~diman@cpc1-cdif12-2-0-cust125.5-1.cable.virginmedia.com] has joined #lisp
21:02:03  Okay, so you also have to fix genesis to load the fasls from assemfile before creating the first fdefinitions.
21:02:29  LiamH: Right. Just checking. I have something similar, but after that map over the list and coerce it to floats (or whatever the flavour is I want) before setting it as the initial content of the (deceased) marray.
21:02:32  And tweak it to pull the assembly-routine version instead of using cold-foreign-symbol-address.
21:02:33  fe[nl]ix: I annotated the paste with the correct invocation of cross-build-world.  Should work better.
21:02:46  And then you can do the same with the undefined_tramp.
21:02:55  (And funcallable-instance trampoline, if you have one of those.)
21:03:09  LiamH: Anyway, I'll have a look if I can contribute in some way. It's time for me to get some sleep now.
21:03:25  Sikander: I need to build-in consistent auto-coercion.  Right now it happens some usage and not in others.
21:03:27  I've done this twice now for SBCL at different times for different reasons.
21:03:32  Sikander: OK, good night.
21:03:33  nyef:  If I still had a reasonably fast sparc, I'd try that.
21:03:49  Goodnight
21:03:55 -!- Sikander [~soemraws@5356ECA7.cable.casema.nl] has quit [Quit: leaving]
21:05:39  rtoym: http://repo.or.cz/w/sbcl/nyef.git/commitdiff/59bda6eac7a44facd5b14fac50113bc642aa8589
21:05:39 btbngr1 [~btbgnr@smorge2.force9.co.uk] has joined #lisp
21:06:10  It's in the context of a more involved project, and has an unrelated change or two in it, but otherwise shows part of how it's done.
21:06:22 -!- Joreji [~thomas@86-129.eduroam.RWTH-Aachen.DE] has quit [Read error: Connection reset by peer]
21:06:26 -!- dlowe [~dlowe@ita4fw1.itasoftware.com] has quit [Quit: Leaving.]
21:07:18  nyef:  Oh, nice.  Thank you very much!
21:07:24  rtoym: and once that's done, do how do I call build.sh ?
21:08:06 -!- stassats [~stassats@wikipedia/stassats] has quit [Read error: Connection reset by peer]
21:08:12  ISTR there being some other changes in support of this at some point, though.
21:08:28  fe[nl]ix: src/tools/build.sh -b linux -C "" -o /lisp/lisp
21:08:33  Probably build-system work I did with the desymlinkification...
21:08:53 -!- btbngr [~btbgnr@smorge2.force9.co.uk] has quit [Ping timeout: 265 seconds]
21:09:01 -!- btbngr1 is now known as btbngr
21:09:08 NNshag [user@lns-bzn-32-82-254-30-23.adsl.proxad.net] has joined #lisp
21:10:22  nyef:  Is allocate-vector still an assembly routine?  Just curious.
21:10:45  I think it depends on the platform. Some of them probably just do an inline allocation.
21:10:56 -!- TeMPOraL [~user@188.147.196.168.nat.umts.dynamic.eranet.pl] has quit [Quit: bye]
21:11:21 Komi [Komi@83.231.86.190] has joined #lisp
21:12:48 galaxywatcher [~galaxywat@pat.harte-hanks.com] has joined #lisp
21:13:00 konr [~user@201.82.129.247] has joined #lisp
21:14:40 -!- ephcon [~ephcon@c-24-34-195-64.hsd1.ma.comcast.net] has quit [Quit: leaving]
21:14:56 Joreji [~thomas@86-129.eduroam.RWTH-Aachen.DE] has joined #lisp
21:15:43 Athas [~athas@shop3.diku.dk] has joined #lisp
21:16:53 badipod [~badipod@unaffiliated/baddog144] has joined #lisp
21:17:21  rtoym: same error
21:17:28 -!- galaxywatcher [~galaxywat@pat.harte-hanks.com] has quit [Client Quit]
21:17:57 [mzm] [~ziggy@92.98.161.154] has joined #lisp
21:18:05  fe[nl]ix: Aargh.  I get an error about it, but it never stops the cross-compile.
21:18:13 -!- MetalDust [~metaldust@75-32-203-61.lightspeed.ftwotx.sbcglobal.net] has quit []
21:20:59 *rtoym* tries it himself.
21:21:50  fe[nl]ix: Did you clean out the fasl before doing the cross-compile?
21:22:11  (Use -c option or run src/tools/clean-target.sh target cross)
21:24:34 MetalDust [~metaldust@75-32-203-61.lightspeed.ftwotx.sbcglobal.net] has joined #lisp
21:24:53 -!- roygbiv [~None@pdpc/supporter/active/roygbiv] has left #lisp
21:25:25  same thing
21:25:57  chairos: yes?
21:26:58  rtoym: btw, could you make build.sh & co. return non-zero when build fails ?
21:27:26  chairos: ah, /last is my friend :)
21:31:36 -!- eugu [~Miranda@212.1.246.237] has quit [Quit: eugu]
21:33:16 Madsy [~madman@94-246-6.65.3p.ntebredband.no] has joined #lisp
21:33:16 -!- Madsy [~madman@94-246-6.65.3p.ntebredband.no] has quit [Changing host]
21:33:16 Madsy [~madman@fu/coder/madsy] has joined #lisp
21:38:09 -!- WOG [~OsamaBinW@24-177-10-101.dhcp.nwtn.ct.charter.com] has quit [Read error: Connection reset by peer]
21:40:41 WOG [~OsamaBinW@24-177-10-101.dhcp.nwtn.ct.charter.com] has joined #lisp
21:43:30 -!- timor [~timor@port-92-195-21-179.dynamic.qsc.de] has quit [Remote host closed the connection]
21:45:02  fe[nl]ix: Bummer.  What is going on?  I'll try it now, but I'm pretty sure it works.  Have you done a cross-compile before?
21:45:07  no
21:45:21  fe[nl]ix: Sure, I'll make build.sh return non-zero.
21:46:39  fe[nl]ix: Hmm. build.sh is already returning 1 if it fails.
21:47:09  not in this case
21:47:27 btbngr1 [~btbgnr@smorge2.force9.co.uk] has joined #lisp
21:47:39  Oh, hmm.  lisp doesn't necessarily set the exit code correctly.  I'll have to took into that.
21:48:16  Do you mean cross-build-world.sh or build.sh?
21:48:53 -!- btbngr [~btbgnr@smorge2.force9.co.uk] has quit [Ping timeout: 240 seconds]
21:49:16  cross-build-world.sh in this case
21:49:57  and ISTR instances where I would send a SIGINT to the lisp and it would exit with code 0
21:50:10  which made the build scripts continue and make a big mess
21:50:29 -!- christoph_debian [christoph@cl-1281.dus-01.de.sixxs.net] has quit [Remote host closed the connection]
21:51:14  p_l: yeah. did you see my link?
21:51:40 -!- LiamH [~none@pdp8.nrl.navy.mil] has quit [Quit: Leaving.]
21:52:51  I have it open
21:53:20  rtoym: how do I get a SAP pointing to a static array's data ?
21:54:23  fe[nl]ix: That might be an issue with -batch.  I usually just use C-z and kill.
21:54:39  fe[nl]ix: Do you mean the static-vector stuff that was added recently?
21:54:42  yes
21:55:25 mrbug [~user@unaffiliated/mrbug] has joined #lisp
21:55:36  chairos: good to see a short & to the point version of your argument (I didn't notice the difference between blowfish variant used by bcrypt)
21:55:53  thanks
21:56:39 -!- rajesh [~rajesh@nylug/member/rajesh] has quit [Quit: leaving]
21:57:10  fe[nl]ix: (sys:vector-sap ) should do it.
21:57:55  chairos: http://github.com/codahale/bcrypt-ruby <--- possible source of alternative C library
21:58:47  p_l: that's the OpenBSD implementation.
21:58:52 -!- Blkt` [~user@dynamic-adsl-78-13-247-126.clienti.tiscali.it] has quit [Quit: Error: do not makunbound t please.]
21:59:05  Though, I grant you, it seems the extraction work has been done.
21:59:33  but it's also out-of-date; the bcrypt.c in OpenBSD source tree is 1.24 and this is 1.22
21:59:36  chairos: If you want something a little nicer than /dev/urandom this might work: http://hcsw.org/blog.pl/34
21:59:44  fe[nl]ix: Are you still cross-compiling with the old lisp using -batch?  Try it without -batch.
21:59:45  and it doesn't solve the library installation issue
21:59:53 timor [~timor@port-92-195-21-179.dynamic.qsc.de] has joined #lisp
21:59:55  thanks for the link, though
21:59:55  what about using SSL cprng routines?
22:00:29  redline6561: yeah, I saw that. I'm a little wary of it because I don't know how much analysis it's received, but it's one thing I'm considering.
22:00:50  chairos: Fair point.
22:01:31  p_l: do you know if there is a good existing SSL library that exposes the CPRNG, or would I need to use CFFI on the raw OpenSSL libs?
22:01:33 bombshelter13b [~bombshelt@76-10-149-209.dsl.teksavvy.com] has joined #lisp
22:01:46  chairos: CFFI it seems
22:02:21 -!- timor [~timor@port-92-195-21-179.dynamic.qsc.de] has quit [Remote host closed the connection]
22:02:35  I think I would prefer to write a small library that can do a variety of approaches and figures out which ones will work on the system at hand. urandom, ECD, openssl, etc etc
22:02:40  but that is of course more work
22:03:18 -!- ehu [~ehuels@ip118-64-212-87.adsl2.static.versatel.nl] has quit [Ping timeout: 240 seconds]
22:03:49  fe[nl]ix: Yeah, that's it.  If I cross-compile with -batch, it stops.  Without it, it continues and finishes.   But I should really fix the issue.
22:04:07  OpenSSL has the bonus that it already accesses system prng/entropy sources on many systems (including windows)
22:04:35  rtoym: I'll try tomorrow. now I'm going to sleep :)
22:04:53 -!- badipod [~badipod@unaffiliated/baddog144] has quit [Remote host closed the connection]
22:04:57  fe[nl]ix: Good night, then.  Ping me tomorrow if it still doesn't work.
22:05:02  p_l: yeah.
22:05:11  rtoym: ok
22:05:29  p_l: unless you're on debian. :P
22:05:48  heh
22:05:54  Is bcrypt blowfish encryption?
22:06:09  no.
22:06:28  it is a password hashing algorithm based on a variant of the setup phase of the blowfish cypher.
22:06:51  unfortunately, the name 'bcrypt' is also used by a symmetric-key-encryption utility that uses blowfish, which causes some confusion.
22:07:45  pity that skein can't really be used for password hashing... it actually kinda specifically defeats bcrypt's purpose
22:08:10  (Skein is a hashing function based on Blowfish/Twofish/Threefish etc.)
22:08:18  heh
22:08:25  chairos: I see.  I just did a google search and the first hit was blowfish file encryption.
22:08:29 antoszka [~antoszka@unaffiliated/antoszka] has joined #lisp
22:08:57  rtoym: yeah. look at http://www.usenix.org/event/usenix99/provos.html
22:09:00 *rtoym* wonders if ironclad supports this.
22:09:13  it does not.
22:09:22  chairos: Skein was made to be as fast as possible while being strong enough to be serious competitor for SHA-3
22:09:52 Jasko [~tjasko@c-174-59-223-208.hsd1.pa.comcast.net] has joined #lisp
22:09:56  something about ~6.1 cpu cycles per bit :)
22:09:56  p_l: yeah. (I read Schneier's blog, among others.)
22:10:12 -!- tayloj [~tayloj@clip-point-02.dynamic2.rpi.edu] has quit [Ping timeout: 260 seconds]
22:10:32  heh. My security-related experiments were in slightly different area than crypto
22:10:43  oh?
22:11:24  chairos: I've been interested in AI-driven automatic information warfare :)
22:11:34  nice
22:12:19  basically combining IDS/IPS systems with "reasoning" module and maybe even leaving a human-unlockable counterstrike functionality :)
22:12:45  heh. what could *possibly* go wrong? ;)
22:13:08  btbngr1: many things
22:13:14 -!- felideon [~user@12.228.15.162] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)]
22:13:19  I never claimed sanity, did I?
22:13:23  hehehe
22:13:45  Not that I'd give AIs weaponry that easily, in difference to for example USA
22:14:07  p_l: oh, you're this guy: http://www.imdb.com/title/tt0382992/
22:14:10  :P
22:15:02  ... it would be better if the movie wasn't of the kind that made me throw up by seeing the poster... <---- inactive (civilian) pilot
22:15:34  hmm. let's say that you have a file that contain lisp data. is there a way to read the file in such a way that you get a new "environment" back, with all defined objects/symbols within the file?
22:17:01  i could, i suppose, define a new package at the top for each file.. but it's a bit, um,  brutal for temporary reads and writes.
22:17:40 lispeval [~lispeval@6.169.16.62.customer.cdi.no] has joined #lisp
22:20:06 syamajala [~syamajala@c-75-68-111-84.hsd1.ma.comcast.net] has joined #lisp
22:20:25  lispeval: (+ 2 2)
22:20:25  4
22:20:30  :-)
22:20:38  andreer: that there is dangerous!
22:20:44  yes, certainly
22:21:05  lispeval: (lisp-implementation-version)
22:21:05  Version 1.5-r13651  (DarwinX8664)
22:21:18  lispeval: (lisp-implementation-type)
22:21:18  Clozure Common Lisp
22:21:32  (exit)
22:21:37  lispeval: (exit)
22:21:38  #
22:21:40  aww
22:21:44  quit works ^^
22:21:51  lispeval: (ccl:quit)
22:21:51 -!- lispeval [~lispeval@6.169.16.62.customer.cdi.no] has quit [Remote host closed the connection]
22:21:55  :)
22:22:15  no such thing as a lisp sandboxing library is there?
22:22:21  andreer: nope.
22:22:25  and it's been tried.
22:22:30  Mmm. Evalbots are dangerous, particularly to the channel signal/noise ratio.
22:22:37  indeed.
22:23:17  gah, I could dust off my notes and write some info on the sandboxing, but my implementation required patching of kernel
22:23:43  p_l: which lisp implementation were you working on this for?
22:23:43 -!- syamajala [~syamajala@c-75-68-111-84.hsd1.ma.comcast.net] has quit [Remote host closed the connection]
22:24:04  sykopomp: if you do it at the OS level, the implementation doesn't really matter.
22:24:26  sykopomp: it was OS level
22:24:34 udzinari [~user@209.158.broadband13.iol.cz] has joined #lisp
22:24:41  pkhuong: well, you still have to stop people from doing (ccl::quit), or other things that will end up killing the bot.
22:24:54 <_3b`> nah, juts start a new one every time
22:24:56  even though you can stop it from hurting your system.
22:25:00  sykopomp: Easy enough, there's this FMAKUNBOUND thing...
22:25:06  _3b`: or make it loop infinitely.
22:25:10  aye, but you could still fork bomb it
22:25:18  or that
22:25:19 <_3b`> resource limits/VMs
22:25:28  sykopomp: to make a good sandbox for something like that you'd need to design the VM for that from scratch, like .NET one
22:25:31  sykopomp: you have to deal with the jailed process randomly dying because of resource limits anyway
22:25:32  oh, it's not insurmountable, just. i'd rather not try :)
22:25:39  pkhuong: true.
22:25:42 <_3b`> last one that was here seemed to handle things pretty well, aside from the signal-noisse stuff
22:25:52 dreish [~dreish@minus.dreish.org] has joined #lisp
22:25:54  it's certainly possible, but I've yet to see anyone pull it off.
22:26:06  and it seems like something people keep trying to do, and just not succeeding.
22:26:30  "possible" would mean "can not crash" or what in this context?
22:26:37 <_3b`> hard to do a general purpose sandbox, but evalbot isn't too bad
22:26:49 camilosebastianl [~camiloseb@186.10.165.42] has joined #lisp
22:27:10   borre acpi=force y apm=power_off de grub.cfg, y cuando apague el pc quedo como al principio que tuve que apagarlo manualmente.
22:27:13  Después utilice esta solución y me dio resultado.
22:27:13  sudo gedit /etc/default/grub (entrar aquí y agregar el acpi=force)
22:27:13  GRUB_CMDLINE_LINUX=""
22:27:13  Quedaría así:
22:27:13  GRUB_CMDLINE_LINUX="acpi=force"
22:27:16  después updatea
22:27:19  sudo update-grub
22:27:19  reestarta el pc
22:27:22  christ
22:27:26  am' sorri
22:27:49  ERC madness? :)
22:28:10  i remember franz used to let you telnet in to a repl, the link is still here: http://www.lisperati.com/casting.html
22:28:47  anyway i'm not gonna run a bot in this channel, might make a new channel for it if i do :)
22:28:55 -!- camilosebastianl [~camiloseb@186.10.165.42] has quit [Quit: Ex-Chat]
22:30:53 -!- Athas [~athas@shop3.diku.dk] has quit [Remote host closed the connection]
22:33:12  andreer: #EVALWAR - tons of eval bots each trying to kill eachother by feeding 'em with obscure lisp hax0r codez. that would be almost kind of funny.
22:33:38  What do you think of newlisp?
22:33:57  newlisp still exists?
22:33:59  konr: it is not much like old lisp.
22:34:18  newlisp is the one that uses fexprs and calls them macros, isn't it?
22:34:21 <_3b`> 'off-topic' is probably the most polite opinion of newlisp around here :p
22:35:17 -!- faux [~user@1-1-4-21a.gkp.gbg.bostream.se] has left #lisp
22:35:22 tayloj [~tayloj@cpe-67-248-22-159.nycap.res.rr.com] has joined #lisp
22:35:24 *Odin-* hadn't realised there was any real on-topic. O_o
22:35:35  Odin-: common lisp is ontopic.
22:36:06 -!- tayloj [~tayloj@cpe-67-248-22-159.nycap.res.rr.com] has quit [Client Quit]
22:36:07  sykopomp: That's a pretty wide-ranging topic, though.
22:36:17  Odin-: righ, but newlisp is not common lisp!
22:36:18  hence...
22:36:36  Although I suppose I would be told off if I started talking about russian spy activities in the U.S.. Dunno.
22:37:22  Odin-: I heard they used a mysterious parenthesized language to communicate the latest information about local PTA activity.
22:37:51  ,     .
22:39:31 -!- nixeagle [~user@Wikimedia/Nixeagle] has quit [Read error: Connection reset by peer]
22:41:13 -!- ejs [~eugen@92.49.220.56] has quit [Ping timeout: 240 seconds]
22:42:27 abend [~alx@delta.muted.org] has joined #lisp
22:44:08  cAr = Armed; cDr = Dangerous. CONS = Armed and Dangerous   :-)
22:44:56  (RPLACA x (cdr x)) - self destruction.
22:44:58 lolsuper_ [~super_@unaffiliated/lolsuper-/x-9881387] has joined #lisp
22:45:41 abugosh [~Adium@pool-74-103-38-231.bltmmd.fios.verizon.net] has joined #lisp
22:52:30 nixeagle [~user@Wikimedia/Nixeagle] has joined #lisp
22:53:18 -!- HET2 [~diman@cpc1-cdif12-2-0-cust125.5-1.cable.virginmedia.com] has quit [Quit: This computer has gone to sleep]
22:54:44  btbngr1: why?
22:55:19 HET2 [~diman@cpc1-cdif12-2-0-cust125.5-1.cable.virginmedia.com] has joined #lisp
22:55:21  beach: yea, i kind of wondered myself after typing that.
23:00:33 -!- stis [~stis@1-1-1-39a.veo.vs.bostream.se] has quit [Read error: Connection reset by peer]
23:01:03 -!- HET2 [~diman@cpc1-cdif12-2-0-cust125.5-1.cable.virginmedia.com] has quit [Quit: This computer has gone to sleep]
23:01:50 -!- tfb [~tfb@restormel.cley.com] has quit [Remote host closed the connection]
23:02:23  btbngr1: (setf *print-circle* nil) (let ((cons (cons 1 nil))) (setf (cdr cons) cons))? :)
23:02:27  is that what you meant?
23:02:28 -!- lemoinem [~swoog@216.252.89.90] has quit [Ping timeout: 265 seconds]
23:02:48  hehe, yes! that would do :)
23:02:52 fusss [~bob@ip70-187-235-200.dc.dc.cox.net] has joined #lisp
23:02:56  hmm
23:03:02  indeed
23:03:14  hey sykopomp
23:03:16 *btbngr1* tries to look like he meant that all along rather than just flippently replying
23:03:32 -!- pdelgallego [~pdelgalle@1503031474.dhcp.dbnet.dk] has quit [Ping timeout: 248 seconds]
23:03:49  fusss: hey fusss
23:03:54  sykopomp: can i bug you with some questions on prototype-based OOPLs? how familiar are you with Self?
23:04:13  fusss: I've never coded in it, but I read some stuff about it.
23:04:34  just trying to clarify my understanding of how data/code objects are implemented
23:04:53  i understand that each object has a few properties {parent, (size?), and slots}
23:04:56  well, implementation-wise, it looks a lot like instances of a class.
23:05:01  yes
23:06:19 -!- varjag [~eugene@84.118.202.84.customer.cdi.no] has quit [Quit: Leaving]
23:06:47  what I am wondering is how would a potential Self programmer access the objects he created in a session without: 1) assigning them to variables, or 2) the objects themselves having a name or other identifiable attribute and put into a hash-table, or 3) limiting instantiation/cloning to a select few "meta-objects" that are accessible by name/ID.
23:06:50  fusss: in Sheeple, objects are #S(mold precedence-list property-values roles)
23:07:26  fusss: which, translated to CLOS speak is #S(class class-precedence-list slot-values method-dispatch-positions)
23:08:13  if objects are not being classified/ID'ed by name, ID, type or any other classfication/clustering mechanism, then how are they distinct? two anonymous object can have identical values
23:08:19  fusss: I think the way Smalltalk-oids deal with this is they have a Lobby object or something like that.
23:08:29  yes, a lobby object
23:08:48  I'm not sure what you're trying to ask.
23:09:00  but lobby shows the top-level objects, and not their internal constituent objects which are pretty much anonymous
23:09:16  well, they're not anonymous. They're in slots.
23:09:24  or they're slots of objects in Lobby's slots, etc.
23:09:47  it looks like Self has an object-access problem that hasn't been address, something that could be begging for XPath or similar graph-traversal DSLs
23:09:48  not sure what the problem is. Objects in a vector are 'anonymous', but you can still refer to them just fine...
23:10:43  is there a prototype language that you would recommend to play with, other than javascript that is.
23:10:56  you could try Io.
23:11:04  Omega was beautiful but I couldn't find a working impl.
23:11:08  Sheeple is like Io+CLOS in some respects.
23:11:33  but Io is pretty simple, and tries to be pretty interactive.
23:11:43  through the command-line, though. No fancy smalltalky IDE.
23:11:59  fusss: http://www.iolanguage.com/
23:12:03  IDEs, don't want
23:12:38 HET2 [~diman@cpc1-cdif12-2-0-cust125.5-1.cable.virginmedia.com] has joined #lisp
23:14:24  fusss: well, right. It has no IDE. It just has what seems to be a fairly nice REPL.
23:15:39 -!- Joreji [~thomas@86-129.eduroam.RWTH-Aachen.DE] has quit [Ping timeout: 240 seconds]
23:16:14 amnesiac [~amnesiac@p3m/member/Amnesiac] has joined #lisp
23:17:28 Draggor [~Draggor@adsl-99-141-191-116.dsl.emhril.sbcglobal.net] has joined #lisp
23:18:44 -!- plediii [~plediii@nat-168-7-232-71.rice.edu] has quit [Remote host closed the connection]
23:19:47 lemoinem [~swoog@216.252.89.150] has joined #lisp
23:22:34  fusss: Dave Ungar asked the most interesting questions at oopsla 2007
23:24:01  Xach: yeah?
23:24:05 *fusss* googles
23:24:19  Xach: cliffnotes?
23:24:36 Athas [~athas@82.211.209.6] has joined #lisp
23:25:10 *sykopomp* is curious too.
23:25:45  Xach: dish! I will be back in a bit
23:26:27 -!- sellout [~greg@212.234.38.214] has quit [Quit: sellout]
23:31:17 -!- Amadiro [~whoppix@ti0021a380-dhcp1747.bb.online.no] has quit [Ping timeout: 264 seconds]
23:35:19 -!- milanj [~milanj_@178.223.147.206] has quit [Ping timeout: 240 seconds]
23:38:30 Amadiro [~whoppix@ti0021a380-dhcp1747.bb.online.no] has joined #lisp
23:40:34 milanj [~milanj_@109.93.104.234] has joined #lisp
23:42:51  fusss: i went to the Dynamic Languages Symposium, which was had a lot of student presentations, and he played the Old Guy Who's Been There role. I wish I could remember specifics.
23:44:39 -!- MetalDust [~metaldust@75-32-203-61.lightspeed.ftwotx.sbcglobal.net] has quit [Ping timeout: 240 seconds]