00:00:00 plage [n=user@123.19.43.20] has joined #lisp 00:00:29 Good morning! 00:00:41 -!- Buganini [n=buganini@security-hole.info] has quit ["test new settings..."] 00:01:09 -!- Ogedei [n=user@e178220086.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)] 00:01:11 <_3b> it could also use a better API on top of it for creating .swf files from code. I've done a bit of that, but the names are pretty random if i remember right, and not sure how much i like the way various parts of it worked (haven't actually used it much recently since i'm focusing on compiler, so don't remember the dtails) 00:01:58 _3b: well, that's a start. btw, did you write them for needs in another project? 00:02:37 <_3b> the flash compiler is so i can write flash stuff without using actionscript (probably not the most efficient choice of tools :) 00:02:42 <_3b> the .swf stuff is mainly to support the compiler 00:03:00 -!- Taggnostr [n=x@dyn57-11.yok.fi] has quit ["Configuring christmas present..."] 00:03:09 Buganini [n=buganini@security-hole.info] has joined #lisp 00:05:04 *_3b* 's plan is to write small ad supported flash games in lisp 00:06:34 -!- carlocci [n=nes@93.37.202.139] has quit ["eventually IE will rot and die"] 00:07:47 _3b: so how can i use it, does it work like a plugin? is there a way to use lisp inside the flash program itself, i mean can i write lisp code bound to flash objects directly inside flash using the libraries or should i use it externally, 00:08:48 <_3b> currently it is a crosscompiler, load stuff into host lisp, and compile a .swf, then run that in flash 00:09:06 hi plage 00:09:38 <_3b> at some point, i'd like to add some sort of hooks to send code to a running flash instance as it is compiled 00:09:54 <_3b> and longer term, make it self-hosting and capable of running slime directly for interactive dev :) 00:10:51 <_3b> though i might still leave it as requiring the cross compiler for deployment 00:12:13 _3b: that would be huge :) i may help in documenting them if i achieve to understand it well enough (but it seems a bit long term) too. i guess :) 00:12:38 kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 00:14:26 _3b: anyway, i should go for now, thanks for great work 00:14:40 <_3b> ok, let me know if you have questions about it 00:15:10 eno_ [n=eno@adsl-70-137-149-81.dsl.snfc21.sbcglobal.net] has joined #lisp 00:15:14 _3b: ok, i will :) 00:15:44 -!- merl15__ [n=merl@80-121-51-197.adsl.highway.telekom.at] has quit [Read error: 110 (Connection timed out)] 00:16:31 fe[nl]ix: How are things? 00:16:55 nice and quiet 00:17:04 Did it snow? 00:17:22 some 15 minutes 00:18:32 -!- eno [n=eno@nslu2-linux/eno] has quit [Read error: 104 (Connection reset by peer)] 00:23:22 davazp [n=user@187.Red-83-57-39.dynamicIP.rima-tde.net] has joined #lisp 00:23:54 -!- OmniMancer [n=OmniManc@122-57-20-218.jetstream.xtra.co.nz] has quit [Read error: 104 (Connection reset by peer)] 00:27:28 daniel_ [i=daniel@unaffiliated/daniel] has joined #lisp 00:27:58 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 00:28:30 -!- teiresias [n=user@archlinux/trusteduser/teiresias] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 00:29:08 -!- eno_ is now known as eno 00:32:34 -!- jmr1 [n=jmr@88.238.180.148] has left #lisp 00:32:40 -!- pr [n=pr@unaffiliated/pr] has quit ["Get MacIrssi - http://www.sysctl.co.uk/projects/macirssi/"] 00:33:13 -!- xan [n=xan@155.99.117.91.static.mundo-r.com] has quit [Read error: 110 (Connection timed out)] 00:33:57 -!- HET3 [n=diman@cpc1-cdif12-2-0-cust125.5-1.cable.virginmedia.com] has quit ["Leaving"] 00:35:59 -!- dandersen [n=dkcl@metabug/dandersen] has quit ["leaving"] 00:36:11 dandersen [n=dkcl@metabug/dandersen] has joined #lisp 00:39:07 OmniMancer [n=OmniManc@122-57-20-218.jetstream.xtra.co.nz] has joined #lisp 00:43:41 -!- daniel [i=daniel@unaffiliated/daniel] has quit [Read error: 110 (Connection timed out)] 00:44:18 -!- huangjs`` [n=user@watchdog.msi.co.jp] has quit [Remote closed the connection] 00:47:51 spradnyesh [n=pradyus@122.167.100.247] has joined #lisp 00:48:45 -!- dandersen [n=dkcl@metabug/dandersen] has quit ["leaving"] 00:52:54 Draggor [n=Draggor@216-80-120-145.alc-bsr1.chi-alc.il.static.cable.rcn.com] has joined #lisp 00:56:35 dandersen [n=dkcl@metabug/dandersen] has joined #lisp 00:56:45 xan [n=xan@13.Red-83-36-170.dynamicIP.rima-tde.net] has joined #lisp 00:58:10 <_deepfire> Any slime folks around? 01:00:11 <_deepfire> When I call a function from repl, and then get to the debugger, the function's frame is missing. This doesn't appear to depend on optimisation level. 01:01:29 _deepfire: what do you mean by "the function's frame"? 01:02:01 <_deepfire> I get a (SB-INT:SIMPLE-EVAL-IN-LEXENV #), and then whatever is down the call chain, but no frame for the function. 01:02:17 proun [n=proun@cpe-76-180-84-142.buffalo.res.rr.com] has joined #lisp 01:02:19 <_deepfire> pkhuong, the function's call frame 01:03:01 <_deepfire> pkhuong, were there any other possible interpretations I missed, though? 01:03:12 does anyone know if it's a good practice to run multiple webapps on 1 hunchentoot instance? 01:03:30 _deepfire: is it evaluating parameters? 01:03:33 -!- grouzen [n=grouzen@91.214.124.2] has quit [Read error: 110 (Connection timed out)] 01:03:57 _deepfire: calling the outermost function in the form is the LAST thing an EVAL call does 01:04:15 <_deepfire> rahul, well, there is a single self-evaluating argument 01:04:30 djanderson [n=dja@hltncable.pioneerbroadband.net] has joined #lisp 01:05:13 _deepfire: paste a backtrace and what frame you'd expect (where you'd expect it). 01:05:26 <_deepfire> rahul, and yes, I think I'm well aware of the general scheme of how EVAL works :-) 01:05:30 <_deepfire> pkhuong, sec 01:05:50 davazp` [n=user@220.Red-83-57-37.dynamicIP.rima-tde.net] has joined #lisp 01:06:14 stassats` [n=stassats@wikipedia/stassats] has joined #lisp 01:07:21 Taggnostr [n=x@dyn57-487.yok.fi] has joined #lisp 01:08:04 -!- davazp` [n=user@220.Red-83-57-37.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 01:10:21 manuel_ [n=manuel@pD9E6C9E8.dip.t-dialin.net] has joined #lisp 01:10:49 <_deepfire> pkhuong, folks, sorry for the noise, I added the (declare (optimize debug)) to the wrong function. When I added it to the right function, its call frame appeared as expected. 01:12:33 -!- jmbr [n=jmbr@9.32.220.87.dynamic.jazztel.es] has quit [Read error: 104 (Connection reset by peer)] 01:12:43 jmbr [n=jmbr@9.32.220.87.dynamic.jazztel.es] has joined #lisp 01:15:02 -!- OmniMancer [n=OmniManc@122-57-20-218.jetstream.xtra.co.nz] has quit ["Leaving."] 01:17:52 -!- eno [n=eno@nslu2-linux/eno] has quit [Read error: 60 (Operation timed out)] 01:18:55 redblue [i=star@ppp018.108-253-207.mtl.mt.videotron.ca] has joined #lisp 01:20:56 quek [n=read_eva@117.102.187.225.static.zoot.jp] has joined #lisp 01:20:58 -!- davazp [n=user@187.Red-83-57-39.dynamicIP.rima-tde.net] has quit [Read error: 110 (Connection timed out)] 01:22:02 eno [n=eno@nslu2-linux/eno] has joined #lisp 01:23:43 jordyd [n=jordyd@99-177-65-75.lightspeed.wepbfl.sbcglobal.net] has joined #lisp 01:23:57 -!- stassats [n=stassats@wikipedia/stassats] has quit [Read error: 110 (Connection timed out)] 01:28:37 -!- timor [n=martin@port-87-234-97-138.dynamic.qsc.de] has quit [Remote closed the connection] 01:29:04 QinGW [n=wangqing@203.86.89.226] has joined #lisp 01:32:32 -!- redline6561 [n=redline@adsl-074-232-249-027.sip.asm.bellsouth.net] has quit [Read error: 60 (Operation timed out)] 01:32:52 hello all :) 01:33:24 I'm trying to do a 'nohup ./clbuild lisp' but that does not work. a './clbuild lisp' works perfectly fine though 01:33:31 any idea why this might be happening? 01:33:35 -!- djanderson [n=dja@hltncable.pioneerbroadband.net] has quit [Read error: 110 (Connection timed out)] 01:34:42 -!- Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has quit [] 01:35:02 -!- Dawgmatix [n=dawgmati@c-76-124-8-39.hsd1.nj.comcast.net] has quit [] 01:35:11 -!- Yuuhi [i=benni@p5483F95F.dip.t-dialin.net] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 01:40:35 -!- plage [n=user@123.19.43.20] has quit [Read error: 60 (Operation timed out)] 01:40:50 -!- redblue [i=star@ppp018.108-253-207.mtl.mt.videotron.ca] has quit [Read error: 110 (Connection timed out)] 01:45:13 -!- fiveop [n=fiveop@g229067139.adsl.alicedsl.de] has quit ["humhum"] 01:48:56 rukubites [n=user@114.78.131.136] has joined #lisp 01:49:39 Are the standard releases of sbcl (e.g. Ubuntu 9.10) SMP compatible, or do I have to roll my own? 01:50:12 I remember when I last tried (approx 2 years ago), I had to fiddle with init parameters. 01:50:49 Edward_ [n=Ed@AAubervilliers-154-1-45-127.w90-3.abo.wanadoo.fr] has joined #lisp 01:51:00 -!- ruediger [n=the-rued@p508B6C2B.dip.t-dialin.net] has quit ["This computer has gone to sleep"] 01:51:44 you'll probably have to rebuild. 01:53:11 redblue [i=star@ppp180.108-253-207.mtl.mt.videotron.ca] has joined #lisp 01:54:38 demmeln [n=Adium@dslb-094-216-033-228.pools.arcor-ip.net] has joined #lisp 01:54:54 Oh well, I will then. My problem is simple to partition, are there any nice lisp examples of spawning e.g. *max-jobs* at a time from a list of jobs? 01:55:45 ruediger [n=the-rued@p508B6C2B.dip.t-dialin.net] has joined #lisp 01:56:43 Looks topical. The last git commit was: 1.0.33.26: enabled threads by default on x86[-64] Linux 01:57:08 rukubites: it's no different in lisp than any other language 01:57:17 rukubites: thread pool and a work queue? That's 30 LOC. 01:58:21 You're right. I can cope, just seeing if anyone remembered any easy examples to save me a little time. 01:58:36 You could use nikodemus's sb-queue. 01:58:59 I wouldn't use such a method if I really cared about getting performance 01:59:11 but for small systems it'll work ok 01:59:19 That'll give you lock-free queues. Since you won't have too many threads, busy-waiting shouldn't be too much of an issue. 02:00:02 rahul: it's an easy way to get nice speed ups on SMP systems. i.e., effective in most of the real world. 02:00:03 Probably maximum of four. It's running on linode. 02:00:38 My VM tells me that I have 4 cpus. 02:00:46 4 will probably be ok 02:01:42 plage [n=user@123.19.70.63] has joined #lisp 02:02:40 we've made it scale to hundreds of nodes on multiple lans too, with some tuning. 02:03:18 pkhuong: I guess you'd want to use a tree of distribution for something like that 02:03:55 branch 4 ways, taking 8 or 16 tasks at a time, and then branch 4 ways again 02:03:57 no, ~10 minute (at least) work units. 02:04:11 ok, yeah, then there's little contention 02:04:52 and when the work takes predictably long, you're hard-pressed to find something more efficient 02:04:55 At that scale, worker failures are probably the most important issue. 02:05:02 yeah 02:06:18 By worker failure, are you meaning in the backend, or bugs in the dispatched code? 02:07:05 rukubites: dispatched *code*?! Anyway, the root cause doesn't matter. Sometimes, work units take too much time to return. 02:07:19 -!- kpreid [n=kpreid@cpe-72-228-78-176.twcny.res.rr.com] has quit [] 02:07:39 kpreid [n=kpreid@cpe-72-228-78-176.twcny.res.rr.com] has joined #lisp 02:07:52 bombshelter13b [n=bombshel@76-10-149-209.dsl.teksavvy.com] has joined #lisp 02:08:05 sorry, my terminology is often lacking for #lisp, it seems. :~( ... 02:10:20 rukubites: the work units should all be basically the same thing applied to different data 02:11:16 rahul: Understood, but perhaps some data contains cases which the dispatched work units don't handle correctly. 02:11:54 rukubites: well, if that's the case, then your whole program is buggy 02:11:58 rukubites: or the client randomly failed, or the network hiccuped, ... 02:12:17 rahul: most programs are. 02:12:19 pkhuong: Yes I understand, I should be prepared for a worker unit failure. 02:12:20 the random failures are the ones you need to cope with 02:12:26 pkhuong: fair enough :P 02:12:57 rukubites: but if you can't compute some results, you need to think carefully about what that means to the overall result 02:12:59 is there any specific things that are not good to do in a cffi callback? I get a memory fault when calling a method, it executes but somewhere else it fails, it only contains a print and a call to glfw:close-window, when i call glfw:close-window by itself there is no memory fault 02:12:59 rahul: A program might not be "buggy" but it may have made assumptions about the data which are wrong for this subset. 02:13:05 rukubites: not on an SMP system. 02:13:16 if it's a scatter-gather kind of computation, you're going to be unable to compute a final result 02:13:32 rukubites: that's an even harder class of bug to deal with 02:13:34 it also seems to not execute any other code after 02:13:47 -!- xan [n=xan@13.Red-83-36-170.dynamicIP.rima-tde.net] has quit [Read error: 110 (Connection timed out)] 02:14:25 rahul: Well you just test and debug when the thing breaks. 02:14:39 rukubites: and yeah, if you're dealing with all processors on the same system, then there aren't going to be communication or hardware failures that will be too recoverable 02:15:13 -!- EwS [n=ews@76.102.249.234] has quit [Remote closed the connection] 02:15:15 rukubites: and you want to be able to debug that asynchronously while the rest of the computation continues? 02:15:30 rahul: No, that probably won't be happening. 02:15:38 -!- redblue [i=star@ppp180.108-253-207.mtl.mt.videotron.ca] has quit [Connection timed out] 02:15:41 rukubites: well, then I don't see why you need to worry about it 02:15:42 Well it could, as I am planning on storing the results to postgres. 02:16:06 rukubites: ok, the worker could check the DB to see if the result is partially calculated 02:16:12 rukubites: if you're going to hit the DB, you might as well put the work queue in there too. 02:16:16 rukubites: also, you could store the work queue itself in the DB 02:16:51 I was just worried that pkhuong may have been implying that the sbcl SMP randomly breaks. But you all say that it is fine, and pkhuong said that by worker failure he was meaning e.g. network outages as well as bugs. Understood now. :-) 02:17:09 and then you can easily distribute over the network. keep pending tasks in a separate queue, and when the main queue is exhausted, look at the pending queue and try those computations again 02:17:42 rukubites: well, it did randomly break under heavy load at ITA 02:18:02 but heavy is somewhere in the dozens of threads all chugging away at once 02:18:02 rahul: pkhuong: yes, that is roughly the plan. With this particular set of computations, I am keeping the design possible to be scaled broadly. 02:18:04 rahul: linux? what kernel version? 02:18:28 pkhuong: yeah, 2.6. umm... 20+? 02:18:29 ISTR a kernel bug ~1-2 years ago affecting threading. 02:18:38 it was around that time frame 02:18:50 CCL didn't pass the tests either 02:19:04 Is a 5 minute build for sbcl good? great? average? 02:19:13 rukubites: good to great. 02:19:20 That was on linode. 02:19:44 Building latest git pull using ubuntu 9.10 32 bit. 02:20:44 hmm, what's the right way to define a CLIM command that pops up a little non-modal editor window? 02:21:00 basically, I want to make a Define Class command 02:22:41 where's the section on non-application frames? 02:23:06 ones that are not a separate application 02:25:26 ah, those are sheets 02:26:07 rahul, pkhuong: thanks for your help. Really the main bit of what I needed was the term "thread pool". :) 02:26:39 -!- Intensity [i=[P1DFcMG@unaffiliated/intensity] has quit [Read error: 104 (Connection reset by peer)] 02:26:56 And FYI - I just checked, the ubuntu 9.10 maintained sbcl has :sb-thread enabled. 02:28:12 doug lea's JSR 166(?) most probably does it right, if you want inspiration. 02:28:56 rahul: re threads, what would your feelings be on assigning 2 FDs for synchronisation purposes to each thread? 02:30:49 -!- spradnyesh [n=pradyus@122.167.100.247] has left #lisp 02:31:02 pkhuong: just 2? 02:31:18 I guess if all communication is only with the master 02:31:57 it might be fine... I'd be worried if there are a lot of threads and other apps. the app will be hogging a lot of FDs. 02:32:16 rahul: yeah, me too. Seems like OS X has a limit of 4k FDs/process for instance. 02:32:20 should work fine for small stuff tho 02:32:35 -!- plage [n=user@123.19.70.63] has quit [Read error: 60 (Operation timed out)] 02:32:42 although it may result in more context switches than needed 02:33:06 so I'd avoid it in the case where there are frequent messages being sent 02:33:09 It gives me something like OS-supported thread parking. 02:33:15 yeah 02:33:44 lockless queues would be better if there's heavy message traffic 02:34:26 it's like Newtonian physics compared to Relativistic. 02:34:53 simpler to do and accurate for the simpler cases 02:35:14 rahul: there's no either/or here. I can have lockless data structures that use the OS instead of busy-waiting for too long. 02:35:58 pkhuong: you're just pushing the problem to the next turtle 02:36:06 the OS needs to implement the lock 02:36:55 -!- ruediger [n=the-rued@p508B6C2B.dip.t-dialin.net] has quit ["This computer has gone to sleep"] 02:37:19 rahul: yes, that's the whole point. 02:37:53 right, so the performance problems will be similar to using heavily contended lock-guarded queues 02:38:16 only when the lock-free alternative would waste tons of time busy waiting. 02:38:48 true 02:39:03 but I'd use a tree structure for communication if there are a lot of threads 02:39:16 gruseom [n=daniel@S0106001217057777.cg.shawcable.net] has joined #lisp 02:39:39 -!- gruseom [n=daniel@S0106001217057777.cg.shawcable.net] has quit [Remote closed the connection] 02:39:42 or something else if you can match the communication structure of the actual hardware 02:40:07 gruseom [n=daniel@S0106001217057777.cg.shawcable.net] has joined #lisp 02:40:40 rahul: that's what busy looping does. Not so useful when you have tons more threads than actual hardware. 02:40:53 Do you guys recommend using something like PCall or "Eager Future"? 02:41:13 rukubites: just roll exactly what you need. 02:41:17 rukubites: do you need that? 02:41:49 pkhuong: a tree structure for communcations? 02:42:05 You might have guessed, this is the first time I've tried to do thread-based multi-processing. I don't really know what I need. :-) 02:42:20 rahul: map to the hardware. 02:42:29 -!- Guthur [n=Michael@host86-136-49-51.range86-136.btcentralplus.com] has quit ["Computer says no"] 02:43:39 pkhuong: well, if you're doing that, then you can't tell which node you'll be scheduled on, so there's not much you can improve 02:44:07 pkhuong: just do a tree that branches 4 or 8 or maybe even 16 ways at each level, I guess 02:47:36 why do you want to have such a large number of threads? 02:47:48 c|mell: because it's easier to express some programs that way 02:48:11 c|mell: because rahul doesn't believe in cooperative multitasking. 02:48:16 crink [n=crink@unaffiliated/crink] has joined #lisp 02:49:15 it's not that I don't, it's that a cooperatively multitasked system has basically peaked in performance now 02:49:35 instead of getting exponentially faster, it'll probably be logarithmic now 02:49:48 maybe linear 02:50:09 it's easier to express most programs as a single thread of execution -- the only reason for parallelism is performance, so why mess it up with 1000s of context switches on individual execution cores 02:50:45 c|mell: the point is that most of them will be sleeping, so we don't want the context switching 02:51:01 sounds like you want a workqueue :) 02:51:09 heh 02:51:24 if you CPS transform and put the continuations on the queue, that kind of works 02:51:33 *c|mell* does that 02:52:22 i don't have much work to do but a lot of threads so the massive overhead of cl-cont is acceptable 02:52:31 but you have to determine multiple continuations at once 02:52:34 It's the CPS transform part that I don't like, especially since it's hard to represent arbitrary continuations efficiently. 02:53:10 -!- bombshelter13b [n=bombshel@76-10-149-209.dsl.teksavvy.com] has quit ["If only your veins were filled with oil, the world would rush to your rescue!"] 02:53:18 -!- dlowe [n=dlowe@c-24-91-154-83.hsd1.ma.comcast.net] has quit ["Leaving."] 02:53:31 pkhuong, why is it hard? are you thinking of something in particular? 02:53:50 c|mell: it's hard because continuations are arbitrary; our use cases aren't. 02:54:08 c|mell: continuations can be placed anywhere, for one 02:54:26 and most of them will be in places where you don't need them 02:54:31 they contain arbitrarily nested function calls. 02:55:02 and you need to figure out when a continuation is "callable" in a non-linear way 02:55:30 rahul, yes to return back to the land of the concrete, the problem with cl-cont is that it cps transforms stuff that has no chance of call/cc'ing at all, so needn't be transformed 02:55:36 just paralellizing (+ (f) (g)) is non-trivial 02:55:39 schoppenhauer [n=christop@unaffiliated/schoppenhauer] has joined #lisp 02:55:53 c|mell: no, you NEED that for this problem 02:56:23 if you want to do automatic parallelisation without user hints you are screwed 02:56:34 c|mell: how many research papers have you seen on how to implement state machines efficiently. Compare that to the implementation of continuations and their various restrictions. 02:56:52 caoliver [n=oliver@75-134-208-20.dhcp.trcy.mi.charter.com] has joined #lisp 02:56:58 at the end of F, you should not call a continuation that calls G if you want parallelism 02:57:08 you first call F and G in parallel 02:57:29 then F and G both need to call + as a continuation, but only once both results are there 02:57:49 and they could get there in either order, so you can't express that as a sequence of continuations 02:57:53 rahul: continuations and parallelism don't go together 02:58:07 -!- Stattrav [n=Stattrav@202.3.77.132] has quit [Read error: 110 (Connection timed out)] 02:58:12 pkhuong: that's my point :) 02:58:24 the very point of continuations is to make evaluation order *extremely* explicit 02:58:32 cmeow [i=cmeow@207.192.71.173] has joined #lisp 02:58:32 yeah 02:58:34 rahul, you're talking about pcall not about continuations :) 02:58:34 rahul: have fun with your strawman. 02:58:42 -!- manuel_ [n=manuel@pD9E6C9E8.dip.t-dialin.net] has quit [] 02:58:49 pkhuong: c|mell was proposing using continuations 02:59:13 rahul: and parallelism across completely independent continuations. 02:59:43 i use continuations along with non-blocking io and an event loop to avoid busy waiting or blocking a thread 02:59:45 pkhuong: that's called threading :) 03:00:10 c|mell: back in the 80s, they called that cooperative multitasking... 03:00:47 -!- schoppenhauer [n=christop@unaffiliated/schoppenhauer] has quit [] 03:01:00 -!- ASau [n=user@83.69.227.32] has quit [Remote closed the connection] 03:01:03 rahul: m:n can work very well for performance. 03:01:10 rahul: sure, doesn't have to be OS threads. 03:01:10 pkhuong: sure 03:02:19 dys` [n=andreas@krlh-5f72419c.pool.mediaWays.net] has joined #lisp 03:03:32 timchen119 [i=tim@163.16.211.21] has joined #lisp 03:03:42 -!- timchen119 is now known as nasloc__ 03:04:17 with a big lisp system, you'd want to have multiple processes on the same machine i guess, to avoid the gc buildup, so m:p:n or something 03:04:19 schoppenhauer [n=senjak@unaffiliated/schoppenhauer] has joined #lisp 03:04:40 GC buildup? 03:04:50 what definition of "process" are you using? 03:04:55 -!- schoppenhauer [n=senjak@unaffiliated/schoppenhauer] has quit [Client Quit] 03:05:04 independent heaps or independent stacks? 03:05:32 rahul: OS processes. 03:05:35 stacks that use separate heaps can't be shared 03:05:50 erm, stacks that use separate heaps can't share data, I mean 03:06:26 so again, you need to pre-determine the threading 03:07:04 -!- crink [n=crink@unaffiliated/crink] has quit ["Leaving."] 03:07:57 -!- dreish [n=dreish@minus.dreish.org] has quit [] 03:08:36 actually, you can use mmap to share data between processes outside of the lisp heap 03:09:13 c|mell: and risk the OS spilling to disk :\ 03:10:47 snowbeard [n=snowbear@cpe-065-190-191-189.nc.res.rr.com] has joined #lisp 03:11:21 The_Fellow [n=spider1@glida.mooo.com] has joined #lisp 03:11:36 pkhuong: you could use tmpfs 03:11:41 djanderson [n=dja@hltncable.pioneerbroadband.net] has joined #lisp 03:11:48 c|mell: either way, you'd run into threading issues 03:12:19 c|mell: GC could kick in on one process, and the other would be manipulating the data in the mean time 03:12:29 pkhuong, you always risk the os spilling to disk, unless you disable swap . . . 03:13:07 rahul, you can't use the lisp gc on your mmap'd regions 03:14:09 c|mell: can't? 03:14:14 -!- quidnunc [n=user@bas16-montreal02-1242357571.dsl.bell.ca] has quit [Read error: 54 (Connection reset by peer)] 03:14:25 redblue [i=star@ppp178.108-253-207.mtl.mt.videotron.ca] has joined #lisp 03:14:53 pkhuong, okay how would you get sbcl to allocate data in a region you mmap'd from lisp :) 03:15:22 (manardb solves this problem by implementing its own gc system) 03:15:52 c|mell: then what's the use of this? 03:16:07 you've just added loads of non-transparency to your system 03:17:37 -!- dys [n=andreas@krlh-5f72698c.pool.mediaWays.net] has quit [Connection timed out] 03:19:23 the non-transparency can be partially concealed with clos, as manardb does 03:19:34 -!- demmeln [n=Adium@dslb-094-216-033-228.pools.arcor-ip.net] has quit ["Leaving."] 03:19:41 -!- proun [n=proun@cpe-76-180-84-142.buffalo.res.rr.com] has quit ["leaving"] 03:20:09 the use is that you can (1) share massive amounts of data between processes quickly, (2) do so without pressurising the lisp gc 03:20:29 -!- blackened`_ [n=blackene@ip-89-102-22-70.karneval.cz] has quit [] 03:21:27 I'd rather use JBoss 03:21:29 -!- dandersen [n=dkcl@metabug/dandersen] has quit ["leaving"] 03:21:43 ok, probably not, but this is about as convoluted 03:23:26 what is jboss and why do you think it's convoluted? 03:25:14 *c|mell* opines that manardb is a simple solution to a complex problem 03:25:38 ASau [n=user@83.69.227.32] has joined #lisp 03:33:01 in as much as, say, an RDBMS is a simple solution the problem of generating unique identifiers 03:33:12 (a la gensym) 03:38:00 -!- bgs100 [n=ian@unaffiliated/bgs100] has quit ["Leaving"] 03:39:19 -!- kwinz3 [n=kwinz@85.124.207.62] has quit [Read error: 110 (Connection timed out)] 03:39:52 legumbre_ [n=leo@r190-135-0-222.dialup.adsl.anteldata.net.uy] has joined #lisp 03:40:58 -!- Edward_ [n=Ed@AAubervilliers-154-1-45-127.w90-3.abo.wanadoo.fr] has quit [Read error: 110 (Connection timed out)] 03:41:29 flexwheel1 [n=chatzill@ool-182ff1c9.dyn.optonline.net] has joined #lisp 03:41:39 -!- Younder [n=jthing@88.159.202.84.customer.cdi.no] has quit [Read error: 60 (Operation timed out)] 03:41:40 -!- mattrepl [n=mattrepl@pool-71-163-162-204.washdc.fios.verizon.net] has quit [] 03:43:27 if i have a multi-dimensional array such as (make-array '(2 2) :initial-element #\c), can i reference the first row and get an array in response? from all the material i saw, aref is used to get to a particular value, but i need the entire row. 03:44:11 flexwheel1: funny you should mention 03:44:38 i'm reminded of http://tkpapp.blogspot.com/2009/12/tour-of-xarray-part-1.html 03:44:52 flexwheel1: but also you can displace arrays to it 03:49:08 wakeup^ [n=wakeup@koln-5d819ef0.pool.mediaWays.net] has joined #lisp 03:51:21 QinGW1 [n=wangqing@211.151.251.254] has joined #lisp 03:53:19 thank you Xach. so does displacing one array into another the standard way of doing slicing? 03:53:36 by standard i mean without using any outside libraries 03:54:29 snowbeard_ [n=snowbear@cpe-065-190-191-189.nc.res.rr.com] has joined #lisp 03:54:37 -!- legumbre [n=leo@r190-135-32-162.dialup.adsl.anteldata.net.uy] has quit [Read error: 104 (Connection reset by peer)] 03:54:40 -!- snowbeard [n=snowbear@cpe-065-190-191-189.nc.res.rr.com] has quit [Read error: 54 (Connection reset by peer)] 03:54:42 -!- snowbeard_ is now known as snowbeard 03:55:35 spradnyesh [n=pradyus@nat/yahoo/x-ungfqscwmffkflcv] has joined #lisp 03:55:56 flexwheel1: displaced arrays are part of ANSI CL 03:56:22 flexwheel1: you can also use a vector of vectors if that's going to be your primary mode of access 03:58:48 (array displacement will add a layer of indirection to the access) 03:58:58 i might, but i'm just playing around with arrays and trying to find out if there are any straigt-forward way of referencing rows inside the array. 03:59:35 -!- c|mell [n=cmell@115.67.65.93] has quit [Read error: 110 (Connection timed out)] 04:00:42 flexwheel1: yeah, it's not a general array slicing mechanism, tho (it can only take full rows, so you can't say "give me the upper-left quarter" and if you want multiple rows, they'll be linearized) 04:01:17 for the case of a single row or part thereof, it works just fine 04:01:41 you're talking about desplacing? 04:02:19 dys`` [n=andreas@krlh-5f72d5fb.pool.mediaWays.net] has joined #lisp 04:02:53 -!- jleija [n=jleija@adsl-91-0-131.chs.bellsouth.net] has quit ["leaving"] 04:04:59 -!- legumbre_ [n=leo@r190-135-0-222.dialup.adsl.anteldata.net.uy] has quit [Read error: 104 (Connection reset by peer)] 04:05:04 -!- wakeup [n=wakeup@koln-5d81891a.pool.mediaWays.net] has quit [Read error: 110 (Connection timed out)] 04:05:09 Displacement is quite low level: the array is written out (flat) in row-major order, any displacement is defined on that flat piece of memory. 04:05:11 legumbre_ [n=leo@r190-135-0-222.dialup.adsl.anteldata.net.uy] has joined #lisp 04:05:35 flexwheel1: yes 04:05:39 But displacement will definitely solve your problem, flexwheel1 . 04:06:01 -!- QinGW [n=wangqing@203.86.89.226] has quit [Read error: 110 (Connection timed out)] 04:06:26 i'm playing around with it now, seems to be working like i need it to. thank you. 04:06:37 -!- jordyd [n=jordyd@99-177-65-75.lightspeed.wepbfl.sbcglobal.net] has quit ["Leaving"] 04:07:43 QinGW [n=wangqing@203.86.89.226] has joined #lisp 04:09:35 ls 04:09:39 heh ooops 04:12:50 -!- ikki [n=ikki@189.247.80.97] has quit [Read error: 113 (No route to host)] 04:13:31 -!- snowbeard [n=snowbear@cpe-065-190-191-189.nc.res.rr.com] has quit [Remote closed the connection] 04:14:31 snowbeard [n=snowbear@cpe-065-190-191-189.nc.res.rr.com] has joined #lisp 04:14:58 -!- dys` [n=andreas@krlh-5f72419c.pool.mediaWays.net] has quit [Success] 04:15:31 PassingStranger_ [n=Horst@p54BF6355.dip.t-dialin.net] has joined #lisp 04:16:55 -!- djanderson [n=dja@hltncable.pioneerbroadband.net] has quit [Read error: 110 (Connection timed out)] 04:18:55 -!- PassingStranger [n=Horst@p54BF7915.dip.t-dialin.net] has quit [Read error: 60 (Operation timed out)] 04:21:01 Ogedei [n=user@e178202219.adsl.alicedsl.de] has joined #lisp 04:21:36 -!- LiamH [n=nobody@pool-141-156-235-135.res.east.verizon.net] has quit ["Leaving."] 04:24:51 -!- QinGW1 [n=wangqing@211.151.251.254] has quit [Success] 04:26:41 Younder [n=jthing@88.159.202.84.customer.cdi.no] has joined #lisp 04:27:56 mjsor [n=mjsor@71-210-6-200.eugn.qwest.net] has joined #lisp 04:29:26 -!- PassingStranger_ [n=Horst@p54BF6355.dip.t-dialin.net] has quit ["Leaving"] 04:30:38 -!- holycow [n=new@mail.fredcanhelp.com] has quit [Remote closed the connection] 04:30:39 -!- snowbeard [n=snowbear@cpe-065-190-191-189.nc.res.rr.com] has quit [Remote closed the connection] 04:34:10 -!- hugod [n=hugod@bas1-montreal50-1176024051.dsl.bell.ca] has quit [] 04:36:05 galaxywatcher_ [n=galaxywa@ppp-61-90-19-77.revip.asianet.co.th] has joined #lisp 04:39:21 -!- cools [n=user@CPE000f661aca54-CM001692fae248.cpe.net.cable.rogers.com] has quit [Remote closed the connection] 04:42:28 -!- Ogedei [n=user@e178202219.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)] 04:42:33 -!- galaxywatcher [n=galaxywa@ppp-58-8-48-16.revip2.asianet.co.th] has quit [Read error: 110 (Connection timed out)] 04:42:34 -!- galaxywatcher_ is now known as galaxywatcher 04:45:23 dralston [n=dralston@S010600212986cca8.va.shawcable.net] has joined #lisp 04:48:12 snowbeard [n=snowbear@cpe-065-190-191-189.nc.res.rr.com] has joined #lisp 04:49:05 -!- jsoft [n=user@unaffiliated/jsoft] has quit [Remote closed the connection] 04:49:29 slyrus_ [n=slyrus@customer-74.203.304.pdchawaii.com] has joined #lisp 04:51:08 -!- slyrus [n=slyrus@customer-74.203.304.pdchawaii.com] has quit [Read error: 104 (Connection reset by peer)] 04:51:11 -!- slyrus_ is now known as slyrus 05:01:54 sadiquea [n=sadiquea@119.82.102.202] has joined #lisp 05:02:40 -!- galaxywatcher [n=galaxywa@ppp-61-90-19-77.revip.asianet.co.th] has quit [Read error: 60 (Operation timed out)] 05:03:27 -!- slyrus [n=slyrus@customer-74.203.304.pdchawaii.com] has quit [Read error: 104 (Connection reset by peer)] 05:03:34 slyrus [n=slyrus@customer-74.203.304.pdchawaii.com] has joined #lisp 05:07:06 EAlexL [n=EAlexL@83.139.134.70] has joined #lisp 05:08:03 -!- Adlai [n=adlai@unaffiliated/adlai] has quit [Connection timed out] 05:10:08 Stattrav [n=Stattrav@202.3.77.132] has joined #lisp 05:11:35 SandGorgon [n=OmNomNom@122.160.41.129] has joined #lisp 05:12:31 -!- SandGorgon [n=OmNomNom@122.160.41.129] has quit [SendQ exceeded] 05:12:40 shrughes [n=shrughes@c-65-96-172-188.hsd1.ma.comcast.net] has joined #lisp 05:12:46 -!- drwho [n=d@c-98-225-208-183.hsd1.pa.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 05:13:15 SandGorgon [n=OmNomNom@122.160.41.129] has joined #lisp 05:14:48 -!- mjsor [n=mjsor@71-210-6-200.eugn.qwest.net] has left #lisp 05:15:11 -!- EAlexL [n=EAlexL@83.139.134.70] has quit ["Leaving."] 05:15:25 EAlexL [n=EAlexL@83.139.134.70] has joined #lisp 05:19:42 -!- TR2N [i=email@89-180-232-154.net.novis.pt] has left #lisp 05:20:36 holycow [n=new@mail.fredcanhelp.com] has joined #lisp 05:22:53 -!- rukubites [n=user@114.78.131.136] has quit [Remote closed the connection] 05:30:46 -!- dralston [n=dralston@S010600212986cca8.va.shawcable.net] has quit ["Leaving"] 05:37:16 -!- SandGorgon [n=OmNomNom@122.160.41.129] has quit [Connection timed out] 05:38:25 -!- redblue [i=star@ppp178.108-253-207.mtl.mt.videotron.ca] has quit [Read error: 110 (Connection timed out)] 05:39:17 drwho [n=d@c-98-225-208-183.hsd1.pa.comcast.net] has joined #lisp 05:43:09 -!- snowbeard [n=snowbear@cpe-065-190-191-189.nc.res.rr.com] has quit ["Client Quit"] 05:50:04 Kolyan [n=nartamon@95-24-184-87.broadband.corbina.ru] has joined #lisp 05:52:05 -!- flexwheel1 [n=chatzill@ool-182ff1c9.dyn.optonline.net] has quit ["ChatZilla 0.9.86 [Firefox 3.5.6/20091201220228]"] 05:59:45 -!- Oddity [n=Oddity@99.199.136.138] has quit [] 06:02:13 -!- baddog [n=liam@unaffiliated/baddog144] has quit [Read error: 110 (Connection timed out)] 06:06:12 redblue [i=star@ppp040.108-253-207.mtl.mt.videotron.ca] has joined #lisp 06:06:22 -!- ski [n=slj@c-0611e055.1149-1-64736c10.cust.bredbandsbolaget.se] has quit [Read error: 60 (Operation timed out)] 06:11:27 -!- marioxcc [n=user@200.92.86.102] has quit [Remote closed the connection] 06:12:29 c|mell [n=cmell@115.67.191.47] has joined #lisp 06:19:24 SandGorgon [n=OmNomNom@122.160.41.129] has joined #lisp 06:19:52 rahul, are you still there? can you explain a simpler way of transparently sharing data between multiple lisp processes than manardb? 06:20:48 c|mell: you had a problem. you created another process. now you have two problems. 06:21:58 elderK [n=zk@125-238-255-127.jetstream.xtra.co.nz] has joined #lisp 06:22:45 ski [n=slj@c-0611e055.1149-1-64736c10.cust.bredbandsbolaget.se] has joined #lisp 06:23:34 -!- SandGorgon [n=OmNomNom@122.160.41.129] has quit [Client Quit] 06:23:35 -!- ltriant [n=ltriant@lithium.mailguard.com.au] has quit ["leaving"] 06:24:00 SandGorgon [n=OmNomNom@122.160.41.129] has joined #lisp 06:24:50 -!- ski [n=slj@c-0611e055.1149-1-64736c10.cust.bredbandsbolaget.se] has quit [Read error: 60 (Operation timed out)] 06:26:18 rahul, the problem was how to use multiple cores on a lisp that did not support threads 06:26:22 we had no choice but to use processes 06:27:43 in any case, even on a lisp that supports multiple threads, it can often be better to use multiple processes 06:27:46 ski [n=slj@c-0611e055.1149-1-64736c10.cust.bredbandsbolaget.se] has joined #lisp 06:29:28 fork() is a perfectly fine solution for when there's lots of independence between the processes 06:34:12 and with manardb, it's a perfectly fine solution even when the processes have to share a lot of data :) 06:34:56 yes, it's turtles all the way down 06:37:53 -!- redblue [i=star@ppp040.108-253-207.mtl.mt.videotron.ca] has quit [Success] 06:38:52 do you not understand why separate processes can be preferable to threads? 06:42:59 (gc) (fork) makes some sense when you have a tree of computations that throw away all data after they are complete 06:43:08 but requires very special support from the GC 06:43:38 I'm not sure it's ever really worked in SBCL 06:44:09 The first question that I would ask is "is transparently shared data a good idea?" 06:45:42 especially data to which access is not synchronized or mediated in any way 06:45:57 I used to think that it was a good idea. 06:46:26 But, much like transparent persistence and transparent distribution, I no-longer do. 06:46:36 it's fine if the data is never side-effected. if it's just data upon which a series of computations base their computations on 06:47:24 -!- EAlexL [n=EAlexL@83.139.134.70] has quit ["Leaving."] 06:52:15 franki^ [n=franki@unaffiliated/franki] has joined #lisp 06:57:53 -!- c|mell [n=cmell@115.67.191.47] has quit [Read error: 104 (Connection reset by peer)] 06:58:25 that discussion leads to an other crazy sbcl idea, what if you could make one sbcl proces gc the other? 06:59:25 tcr [n=tcr@host146.natpool.mwn.de] has joined #lisp 07:00:56 That idea is similar to an other one, that you could fire up an new sbcl, connect to a crashed one, move over all objects and select a starting point. 07:01:50 What would be the point? 07:04:11 Zhivago: main reason is exercise :) The latter idea though could have a practical use. 07:04:24 What would that be? 07:05:55 I mean, it's crashed, so what data in the image can we rely upon? 07:07:02 c|mell [n=cmell@111.84.110.1] has joined #lisp 07:10:14 prxq [n=mommer@e179054118.adsl.alicedsl.de] has joined #lisp 07:10:48 grouzen [n=grouzen@91.214.124.2] has joined #lisp 07:11:26 Well if it's a stack-overflow, or heap-overflow (and it's caught in Lisp), you may want to rescue part of the data processed so far. 07:11:50 not that I ever wanted it, but I don't work on precious data :-) 07:12:19 May I emphasize the fun in just doing it? 07:12:26 the only modern common lisp which safely handles stack overflow is lispworks? (or are there others?) 07:12:28 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 07:12:50 c|mell: sbcl 07:13:05 probably ccl and acl, too 07:13:17 actually, maybe not ccl, but I'd expect acl to 07:13:53 lharc: that's a helluva difficult "exercise" 07:14:14 well in my experience sbcl and acl crash quite hard :) 07:14:22 mostly, you'll learn sbcl's memory layout in a bass-acwards way 07:14:33 c|mell: I've managed to recover sbcl a few times 07:14:50 c|mell: it's not 100% reliable 07:15:13 indeed 07:15:18 rahul: :) yes.. I dont have the mental capability for it. It seems all my fun ides like this one, fail on that point. 07:15:53 It's ok to do things that are stupid for fun -- but it's often a good idea to understand if and why they are stupid first. 07:17:44 -!- NNshag [n=shag@lns-bzn-37-82-253-13-222.adsl.proxad.net] has quit [Remote closed the connection] 07:17:55 Zhivago: correct, but I belive that exercise is not stupid because it will take you to wonderful places inside sbcl. 07:18:01 udzinari [n=user@19.221.broadband4.iol.cz] has joined #lisp 07:18:24 Sounds good. 07:18:27 lharc: if anything, working on making ldb more featureful will be orders of magnitude easer and more useful 07:18:33 galaxywatcher [n=galaxywa@ppp-58-8-53-124.revip2.asianet.co.th] has joined #lisp 07:19:59 is source for minion and/or specbot avaliable somewhere? 07:20:34 rahul: good idea, is there more info on what is requested? 07:21:42 ASau` [n=user@77.246.231.86] has joined #lisp 07:21:51 -!- abeaumont [n=abeaumon@84.76.48.250] has quit [Read error: 110 (Connection timed out)] 07:24:04 I think maybe I answered my own question :) .. always just press 'q' when landing in ldb. 07:26:28 lharc: There's a full bug tracker. Easy stuff is tagged `easy' to get one's hand dirty. 07:26:58 anything involving ldb is not likely to be tagged easy :) 07:29:53 there doesn't seem to be particular items about ldb itself 07:30:06 tcr: just a plain search for ldb shows 8 bugs, but noone about ldb. 07:31:01 Yeah, still might make sense to tackle easy stuff first to get a feel for the code base. But I missed the start of the conversation anyway, because you feel more safy in C than in Lisp 07:31:12 +perhaps 07:34:39 baddog [n=liam@unaffiliated/baddog144] has joined #lisp 07:36:32 OmniMancer [n=OmniManc@122-57-20-218.jetstream.xtra.co.nz] has joined #lisp 07:37:37 -!- ia [n=ia@89.169.161.244] has quit ["...may the Source be with you..."] 07:38:28 mrSpec [n=Spec@unaffiliated/mrspec] has joined #lisp 07:41:59 ia [n=ia@89.169.161.244] has joined #lisp 07:42:14 -!- c|mell [n=cmell@111.84.110.1] has quit [Read error: 110 (Connection timed out)] 07:47:11 c|mell [n=cmell@111.84.120.19] has joined #lisp 07:56:27 kejsaren1 [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 08:00:06 mvilleneuve [n=mvillene@LLagny-156-36-4-214.w80-14.abo.wanadoo.fr] has joined #lisp 08:00:27 bombshelter13b [n=bombshel@76-10-149-209.dsl.teksavvy.com] has joined #lisp 08:00:32 good morning 08:00:49 nunb [n=nundan@static-217-133-104-225.clienti.tiscali.it] has joined #lisp 08:03:55 pr [n=pr@p579CAE84.dip.t-dialin.net] has joined #lisp 08:06:42 CBro2007_ [n=manukaul@c-7996e253.026-209-73746f13.cust.bredbandsbolaget.se] has joined #lisp 08:09:16 -!- grouzen [n=grouzen@91.214.124.2] has quit [Read error: 60 (Operation timed out)] 08:11:31 moocow [n=new@mail.fredcanhelp.com] has joined #lisp 08:11:59 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 08:14:21 varjag [n=eugene@122.62-97-226.bkkb.no] has joined #lisp 08:20:11 -!- jmbr [n=jmbr@9.32.220.87.dynamic.jazztel.es] has quit [Read error: 104 (Connection reset by peer)] 08:24:48 -!- pr [n=pr@unaffiliated/pr] has quit ["Get MacIrssi - http://www.sysctl.co.uk/projects/macirssi/"] 08:25:59 -!- dnolen [n=dnolen@ppp-70-249-155-206.dsl.rcsntx.swbell.net] has quit [] 08:26:58 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 08:29:32 blackened` [n=blackene@ip-89-102-22-70.karneval.cz] has joined #lisp 08:29:38 grouzen [n=grouzen@91.214.124.2] has joined #lisp 08:30:08 -!- hohum_ is now known as hohum 08:31:49 -!- kejsaren1 [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 08:32:19 -!- caoliver [n=oliver@75-134-208-20.dhcp.trcy.mi.charter.com] has left #lisp 08:40:42 CyberBlue [n=yong@111.167.0.48] has joined #lisp 08:41:13 -!- baddog [n=liam@unaffiliated/baddog144] has quit [Read error: 110 (Connection timed out)] 08:41:44 nikodemus [n=nikodemu@cs181201111.pp.htv.fi] has joined #lisp 08:41:45 -!- kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 08:41:52 good morning 08:51:58 -!- slyrus [n=slyrus@customer-74.203.304.pdchawaii.com] has quit [Read error: 60 (Operation timed out)] 08:55:53 -!- bombshelter13b [n=bombshel@76-10-149-209.dsl.teksavvy.com] has quit [] 08:56:41 bombshelter13b [n=bombshel@76-10-149-209.dsl.teksavvy.com] has joined #lisp 08:58:20 -!- moocow is now known as holycow_ 08:59:13 -!- bombshelter13b [n=bombshel@76-10-149-209.dsl.teksavvy.com] has quit [Client Quit] 09:04:34 jsoft [n=user@unaffiliated/jsoft] has joined #lisp 09:04:42 kiuma [n=kiuma@85-18-55-37.ip.fastwebnet.it] has joined #lisp 09:05:05 merl15__ [n=merl@80-121-60-5.adsl.highway.telekom.at] has joined #lisp 09:12:14 jdz [n=jdz@85.254.211.133] has joined #lisp 09:14:25 fihi09`` [n=user@pool-71-190-77-62.nycmny.east.verizon.net] has joined #lisp 09:15:52 trebor_dki [n=user@mail.dki.tu-darmstadt.de] has joined #lisp 09:17:04 kejsaren_ [n=kejsaren@wavelan-228.vlan-109.publik.su.se] has joined #lisp 09:17:33 kwinz3 [n=kwinz@85.124.207.62] has joined #lisp 09:17:39 hi nikodemus, have you got a case with the wrong GC time reported in the profile for me? 09:17:47 attila_lendvai [n=ati@catv-89-134-66-143.catv.broadband.hu] has joined #lisp 09:18:29 -!- holycow [n=new@mail.fredcanhelp.com] has quit [Read error: 110 (Connection timed out)] 09:22:41 freiksenet [n=freiksen@hoasnet-fe29dd00-202.dhcp.inet.fi] has joined #lisp 09:23:50 TuxPurple [n=TuxPurpl@unaffiliated/tuxpurple] has joined #lisp 09:23:54 -!- TuxPurple [n=TuxPurpl@unaffiliated/tuxpurple] has quit [Read error: 104 (Connection reset by peer)] 09:25:46 Jafet [n=Jafet@unaffiliated/jafet] has joined #lisp 09:26:11 c|mell: it was (profile "SB-C"), but i don't remember exactly what it was that i compiled 09:27:10 it always matches the (time ...) GC time for me 09:27:54 i'll poke at it later today again 09:28:30 thanks, i don't want this thing to die because of some unreproducible doubts :( 09:28:56 if i can reproduce i'll send the case to the list, if not, i'll look at merging the patch :) 09:29:21 on a related note, what about changing internal-time-units-per-second to 1 000 000? 09:30:00 linux is very keen on the us as a unit 09:30:28 c|mell: can we get us for real time as well? 09:31:11 yes, we can, of course 09:31:21 (still working on my coffee...) 09:32:49 otoh, the ms is quite an ingrained standard for many lisps (e.g. allegro) and overflowing fixnums on 32-bits etc. might be annoying 09:33:42 -!- tcr [n=tcr@host146.natpool.mwn.de] has quit ["Leaving."] 09:33:46 yeah, though there may be some surprising side-effects like increased bignum consing in code that keeps checking the time 09:35:01 Fufie [n=innocent@86.80-203-225.nextgentel.com] has joined #lisp 09:35:06 we *could* do it for 64 bits only... 09:36:15 i was thinking of that, it makes sense to me 09:37:56 in either case it would be nice to provide a high-accuracy function people could explicitly call on 32 bits as well 09:38:11 -!- fihi09` [n=user@pool-96-224-42-9.nycmny.east.verizon.net] has quit [Success] 09:38:29 maybe not get-resource-usage (the differences between windows and unix are too big there), but get-time-of-process or something 09:42:37 -!- lnostdal [n=lnostdal@90.149.113.175] has quit [Read error: 110 (Connection timed out)] 09:43:17 on the other hand, it seems that linux has clock_gettime for doing nanosecond stuff 09:43:26 lnostdal [n=lnostdal@90.149.113.175] has joined #lisp 09:43:42 -!- galaxywatcher [n=galaxywa@ppp-58-8-53-124.revip2.asianet.co.th] has quit [] 09:43:43 and as we're not really interested in the rest of the stuff returned from getrusage in this case, maybe that would be better 09:44:06 -!- fe[nl]ix [n=algidus@ABordeaux-158-1-130-200.w90-60.abo.wanadoo.fr] has quit ["Valete!"] 09:44:09 cmsimon [n=cms@unaffiliated/cmsimon] has joined #lisp 09:44:39 "Some implementations of CLIM may cause the application to "throw" all the way back to run-frame-top-level in order to do this." 09:44:42 GAAAHHH 09:46:22 I want to pop up the interactor pane only when the user's action needs further input 09:52:26 nikodemus, i think that clock_gettime has the very important advantage for the profiling stuff of allowing *perthread* cputime measurement 09:52:41 galaxywatcher [n=galaxywa@ppp-58-8-53-124.revip2.asianet.co.th] has joined #lisp 09:54:22 -!- dmiles_afk [n=dmiles@c-76-104-220-73.hsd1.wa.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 09:55:59 xan [n=xan@13.Red-83-36-170.dynamicIP.rima-tde.net] has joined #lisp 09:56:01 very good point 09:56:03 dmiles_afk [n=dmiles@c-76-104-220-73.hsd1.wa.comcast.net] has joined #lisp 09:58:15 -!- mrsolo [n=mrsolo@adsl-68-126-184-236.dsl.pltn13.pacbell.net] has quit ["Leaving"] 09:59:36 -!- nunb [n=nundan@static-217-133-104-225.clienti.tiscali.it] has quit ["Leaving"] 10:03:45 holycow [n=new@mail.fredcanhelp.com] has joined #lisp 10:04:17 udzinari` [n=user@nat/ibm/x-supnsuibrlxnukll] has joined #lisp 10:05:16 Hun [n=hun@p50993726.dip0.t-ipconnect.de] has joined #lisp 10:08:03 Ogedei [n=user@e178206094.adsl.alicedsl.de] has joined #lisp 10:08:43 -!- blandest [n=blandest@softhouse.is.ew.ro] has quit [Read error: 110 (Connection timed out)] 10:08:46 -!- udzinari` [n=user@nat/ibm/x-supnsuibrlxnukll] has quit [Read error: 104 (Connection reset by peer)] 10:09:33 Reaver [n=Data_Ent@h15a2.n2.ips.mtn.co.ug] has joined #lisp 10:11:40 -!- quek [n=read_eva@117.102.187.225.static.zoot.jp] has left #lisp 10:12:50 -!- nikodemus [n=nikodemu@cs181201111.pp.htv.fi] has quit ["This computer has gone to sleep"] 10:15:38 -!- holycow [n=new@mail.fredcanhelp.com] has quit [Remote closed the connection] 10:15:38 -!- holycow_ [n=new@mail.fredcanhelp.com] has quit [Read error: 104 (Connection reset by peer)] 10:17:44 Edico [n=Edico@unaffiliated/edico] has joined #lisp 10:20:32 nikodemus [n=nikodemu@cs181201111.pp.htv.fi] has joined #lisp 10:24:17 -!- moesenle [n=moesenle@atradig124.informatik.tu-muenchen.de] has quit [Remote closed the connection] 10:24:43 -!- kejsaren_ [n=kejsaren@wavelan-228.vlan-109.publik.su.se] has quit [Read error: 110 (Connection timed out)] 10:24:57 -!- xan [n=xan@13.Red-83-36-170.dynamicIP.rima-tde.net] has quit [Read error: 110 (Connection timed out)] 10:25:20 holycow [n=new@mail.fredcanhelp.com] has joined #lisp 10:27:22 marcoecc [i=me@gateway/gpg-tor/key-0x9C9AAE7F] has joined #lisp 10:27:44 nikodemus: thanks for the cursor key patch, very nice. 10:27:48 moesenle [n=moesenle@atradig124.informatik.tu-muenchen.de] has joined #lisp 10:28:11 Can you tell me how to reproduce the timeout thing? 10:28:40 I have never seen it. Could it be an iolib-on-MacOS thing? 10:30:33 (How well does epoll work on MacOS anyway?) 10:31:14 tcr [n=tcr@host146.natpool.mwn.de] has joined #lisp 10:31:40 oh, actually I don't use epoll, because select worked better. hmm. 10:31:51 -!- dys`` is now known as dys 10:32:24 nikodemus: here? 10:32:24 manuel_ [n=manuel@pD9E6C9E8.dip.t-dialin.net] has joined #lisp 10:33:56 fe[nl]ix [n=algidus@ABordeaux-158-1-130-200.w90-60.abo.wanadoo.fr] has joined #lisp 10:35:06 demmeln [n=Adium@dslb-188-099-119-199.pools.arcor-ip.net] has joined #lisp 10:36:31 -!- QinGW [n=wangqing@203.86.89.226] has quit [Read error: 104 (Connection reset by peer)] 10:37:51 nipra [n=nipra@121.243.225.226] has joined #lisp 10:38:20 mstevens [n=mstevens@zazen.etla.org] has joined #lisp 10:39:52 tcr: for a short while 10:40:09 lichtblau: i'm pretty sure the timeout patch is wrong 10:40:42 nikodemus: I made autodoc to be as fast, if not faster than before. I can move through package-data-list now without any kind of slug. 10:41:29 tcr: that's great news! thanks! 10:41:37 *nikodemus* turns autodoc back on 10:41:44 nikodemus: I'll clean it up and commit it, and I'd like you to update then. There will probably be a bug, or two. 10:41:57 hehe 10:42:04 where bug means getting an error message in the echo area 10:43:26 tcr, could you make the slime compilation note buffer building faster? :) it takes literally minutes sometimes :( 10:43:39 (setq font-lock-verbose nil) 10:43:49 it's an Emacs thing 10:45:29 YuleAthas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has joined #lisp 10:49:00 Is that OSX only? I don't recall seeing it elsewhere. 10:49:40 lichtblau: having had a look at iolib, it seems that the right thing is to call event-dispatch with eg. :min-step of 0.01 10:50:24 -!- ace4016 [i=ace4016@cpe-76-170-134-79.socal.res.rr.com] has quit ["night"] 10:51:42 the thing is that with min-step of zero, if the monotonic time counter is not sufficiently monotonic like on OS X, you end up waiting not at all 10:52:07 assuming i understand this correctly, that is 10:52:37 well, I think min-step > 0 means that timeout gets rounded up to min-step. 10:52:45 xan [n=xan@155.99.117.91.static.mundo-r.com] has joined #lisp 10:53:05 And that results in the hemlock event loop handing for that timeout. 10:53:08 clamped between min-step and max-step 10:53:30 I explicitly had to set the timeouts to 0 to make things not sluggish. 10:54:06 i bet that's because you actually get some waiting with (+ TMP-TIMEOUT (IOLIB.SYSCALLS:%SYS-GET-MONOTONIC-TIME)) 10:54:17 with TMP-TIMEOUT of zero 10:54:50 I think what I was seeing was that it was waiting for the timeout to expire even when there actually was an event or something. 10:55:05 hrr 10:55:21 *attila_lendvai* pokes fe[nl]ix 10:56:22 The other thing to debug is why the epoll backend segfaults in hemlock. I switched to select because of that. 10:56:39 wow 10:57:23 csimon [n=cms@unaffiliated/cmsimon] has joined #lisp 10:57:25 -!- cmsimon [n=cms@unaffiliated/cmsimon] has quit [Read error: 104 (Connection reset by peer)] 10:57:33 -!- csimon [n=cms@unaffiliated/cmsimon] has quit [Remote closed the connection] 10:58:12 bbl 10:58:13 -!- nikodemus [n=nikodemu@cs181201111.pp.htv.fi] has quit ["This computer has gone to sleep"] 10:58:22 <_deepfire> Is there a way to check whether a pathname is a symlink? 10:59:39 lichtblau: the purpose of min-timeout is to balance throughput against latency 11:00:40 tcr, while testing that trick i managed to get emacs to show no notes at all :( 11:00:48 lichtblau: IIRC lnostdal had the problem that with a few thousand connections the loop was run much too often 11:01:56 -!- CBro2007_ [n=manukaul@c-7996e253.026-209-73746f13.cust.bredbandsbolaget.se] has quit [] 11:05:17 <_deepfire> I've a problem with distinguishing dead symlinks from files, as probe-file returns same for an existing file and a symlink. 11:05:28 <_deepfire> Er, a dead symlink. 11:06:10 c|mell: I think luis reported yesterday that that trick worked for him 11:06:12 fe[nl]ix: sounds like an area where careful microbenchmarks are needed as tests. In this case, nikodemus is on MacOS and I'm on Linux, and it looks like there might be differences in behaviour. 11:06:18 <_deepfire> So in the end, I can't seem to be able to tell if a file exists or not. 11:06:24 _deepfire: don't use pathnames :-) 11:06:44 <_deepfire> lichtblau, asdf does... 11:07:29 <_deepfire> lichtblau, I want to modify asdf to signal a proper, distinguishable error /before/ LOAD fails with a useless SIMPLE-FILE-ERROR 11:08:40 _deepfire: (osicat:file-kind #p"/usr/X11R6") 11:08:56 <_deepfire> fe[nl]ix, thanks! 11:08:59 plage [n=user@serveur5.labri.fr] has joined #lisp 11:09:04 Good evening 11:09:15 aintme [n=user@1.37.221.87.dynamic.jazztel.es] has joined #lisp 11:11:48 -!- c|mell [n=cmell@111.84.120.19] has quit ["Leaving"] 11:13:42 nikodemus [n=nikodemu@cs181150041.pp.htv.fi] has joined #lisp 11:15:28 <_deepfire> Oh hell, osicat:file-kind uses CFFI. I'm feeling doomed. 11:15:37 -!- freiksenet [n=freiksen@hoasnet-fe29dd00-202.dhcp.inet.fi] has quit [Remote closed the connection] 11:15:49 <_deepfire> As if it was unpredictable.. 11:16:41 _deepfire: use FFI of your choise and call stat directly? 11:17:10 <_deepfire> nikodemus, I need to extend ASDF.. 11:17:31 nikodemus pasted "hemlock sb-sprof results from just sitting in the terminal (without the aforementioned timeout or min-step changes)" at http://paste.lisp.org/display/92413 11:18:19 _deepfire: try to find out which implementations already have such a test, for the rest use their own FFI. 11:19:10 _deepfire: it's just a matter of using M-. 11:19:10 _deepfire: you could report that as a bug, i guess. arguably truename of a dead symlink is the symlink itself 11:20:10 <_deepfire> fe[nl]ix, well, the idea is simple, indeed. I'm just having an obvious problem of splicing a #+-ssed FFI-calling monster into asdf. 11:20:47 <_deepfire> fe[nl]ix, it's not that I'm lazy or lacking imagination -- it's about whether such a thing belongs to ASDF or not.. 11:22:47 <_deepfire> nikodemus, well, truename returns the symlink itself, in the dead case here -- but I don't see how this helps me to tell if LOAD will succeed finding its target or not. 11:23:03 without claiming to understand how the event handling is meant to work, my impression is that the API should provide explicitly for three different cases: 11:23:05 1. wait with a non-zero timeout 2. check for events with a zero timeout, i.e. return right away if there are no events and 3. wait without a timeout, i.e. indefinitely 11:23:21 I'm specifying a timeout of zero, but what I mean is (3), not (2). 11:24:26 nikodemus annotated #92413 "macroexpanding the repeat-upon-condition-decreasing-timeout from harvest-events in iolib" at http://paste.lisp.org/display/92413#1 11:25:01 <_deepfire> Maybe I should just handle file-errors within ASDF. OTOH they might be inconclusive, because the loaded .asd might call LOAD or somesuch itself. 11:25:40 <_deepfire> So I won't be able to determine if the file-error was caused by the missing definition /or/ by the definition itself screwing up. 11:26:21 fe[nl]ix: can you look at that annotation. unless i'm terribly misreading something the loop always returns when the select returns for the first time -- not that i think it has direct bearing on this issue, but still doesn't seem quite right... 11:26:43 <_deepfire> I guess I'll go with a 90% solution here. 11:26:53 <_deepfire> God, forgive me. 11:27:35 _deepfire: i hope you're not saying you're installing a reasonably generic handler around loading of a .asd? 11:28:15 <_deepfire> nikodemus, that's the plan, to handle FILE-ERRORs. Do you think that inlining osicat:file-kind is better? 11:28:57 <_deepfire> nikodemus, note that that's strictly around LOAD within ASDF:FIND-SYSTEM. 11:28:59 no, that doesn't sound too good either... just a sec, i may have an idea 11:29:35 G0SUB [n=ghoseb@ubuntu/member/gosub] has joined #lisp 11:30:10 ignas [n=ignas@78-60-73-85.static.zebra.lt] has joined #lisp 11:30:32 (with-open-file (f asd :if-does-not-exist nil) (unless f (error "Could not open ~A" asd)) (load f)) 11:31:08 if you want, you can even check for existing truename after open failed in this case 11:31:27 racy, of course, but i don't think you can really avoid races in this sort of stuff 11:34:39 <_deepfire> The only question is if this works portably.. 11:35:00 <_deepfire> Other than that, all hail nikodemus, the saviour! 11:35:49 the truename bit wont be portable, but LOAD on streams and :IF-DOES-NOT-EXIST NIL is standard -- modulo bugs :) 11:36:07 <_deepfire> Thank you! 11:37:09 lichtblau: iolib seems to provide for case 3, assuming you don't have :max-step or timeouts on handlers themselves 11:39:10 fiveop [n=fiveop@g229064068.adsl.alicedsl.de] has joined #lisp 11:39:12 -!- spradnyesh [n=pradyus@nat/yahoo/x-ungfqscwmffkflcv] has left #lisp 11:43:57 nikodemus annotated #92413 "lichteblau: i *think* this is right" at http://paste.lisp.org/display/92413#2 11:47:41 -!- mstevens [n=mstevens@zazen.etla.org] has quit ["leaving"] 11:48:40 nikodemus: okay, seems to work for me. And it fixes the MacOS busy loop, too? 11:48:40 -!- plage [n=user@serveur5.labri.fr] has quit [Remote closed the connection] 11:49:23 -!- benny [n=benny@i577A3A6E.versanet.de] has quit [Read error: 60 (Operation timed out)] 11:50:46 yes 11:51:57 -!- demmeln [n=Adium@dslb-188-099-119-199.pools.arcor-ip.net] has left #lisp 11:53:17 -!- moesenle [n=moesenle@atradig124.informatik.tu-muenchen.de] has quit [Read error: 60 (Operation timed out)] 11:53:26 -!- G0SUB [n=ghoseb@ubuntu/member/gosub] has quit ["Ex-Chat"] 11:54:58 Sergio` [n=Sergio`@a89-152-191-112.cpe.netcabo.pt] has joined #lisp 11:55:19 lichtblau: was there a way to get a repl in the editor lisp? 11:57:52 longkid [n=lisp@113.22.166.200] has joined #lisp 11:58:47 spiaggia: hello 11:59:04 nikodemus: There is "Start Slave Thread" 12:00:06 (If you really want to debug the hemlock event loop, you can also try "Select Eval Buffer", but it's easy to crash hemlock by making even the smallest typo in there. And it's not a real PREPL.) 12:00:11 dym_ [n=dym@217.20.175.226] has joined #lisp 12:01:00 lichtblau: i'm trying to understand how it all hangs togather, so being able to trace/inspect the live process would be a help 12:01:57 so far i've been writing trace-like output to files in /tmp, which works too but isn't quite as convenient 12:01:58 silenius [n=jl@yian-ho03.nir.cronon.net] has joined #lisp 12:02:18 okay. Start Slave Thread is definitely helpful for that. For more difficult debugging stuff like redisplay errors, M-x Start Swank Server can also be convenient. 12:02:51 TuxPurple [n=TuxPurpl@unaffiliated/tuxpurple] has joined #lisp 12:02:56 ok, thanks 12:03:36 (The slave functionality could in principle be used to connect one editor to another, but I don't think there is a command to expose that at the moment.) 12:05:07 -!- TuxPurple [n=TuxPurpl@unaffiliated/tuxpurple] has quit [SendQ exceeded] 12:05:53 TuxPurple [n=TuxPurpl@unaffiliated/tuxpurple] has joined #lisp 12:06:24 nikodemus: you can do all 3 points with the current API, although perhaps in a contrived way 12:06:56 fe[nl]ix: what should i read to understand the multiplexes api? 12:07:23 nikodemus: the expansion is correct: if EINTR is signalled, the code loops and the select() is retried(with the timeout recalculated) 12:08:38 ah 12:08:50 nikodemus: examples/echo-server.lisp 12:08:56 benny [n=benny@i577A38EE.versanet.de] has joined #lisp 12:08:56 c|mell [n=cmell@115.67.191.177] has joined #lisp 12:09:36 -!- aintme [n=user@1.37.221.87.dynamic.jazztel.es] has quit [""fff""] 12:10:24 -!- nvoorhies [n=nvoorhie@adsl-76-216-21-95.dsl.pltn13.sbcglobal.net] has quit [] 12:12:29 -!- dym [n=dym@217.20.175.226] has quit [Read error: 110 (Connection timed out)] 12:12:33 fe[nl]ix: a few comments there would be nice :) 12:14:29 I know, I suck at documentation 12:15:42 lichtblau: I pushed a fix so that :timeout 0 now works as expected 12:15:45 -!- longkid [n=lisp@113.22.166.200] has left #lisp 12:16:04 Odin- [n=sbkhh@adsl-2-92.du.snerpa.is] has joined #lisp 12:21:30 lichtblau: hah, it turns out my last effort in getting rid of the spinning make the slave thread sluggish... 12:21:48 (having to wait for output ~10 seconds) 12:22:11 nvoorhies [n=nvoorhie@adsl-76-216-21-95.dsl.pltn13.sbcglobal.net] has joined #lisp 12:24:34 nikodemus: that's weird, I have your change and I don't see sluggishness (on Linux). 12:25:01 i hate apple 12:25:29 moesenle [n=moesenle@atradig124.informatik.tu-muenchen.de] has joined #lisp 12:25:32 -!- ia [n=ia@89.169.161.244] has quit [Read error: 60 (Operation timed out)] 12:26:11 -!- nvoorhies [n=nvoorhie@adsl-76-216-21-95.dsl.pltn13.sbcglobal.net] has quit [Client Quit] 12:29:26 -!- The_Fellow [n=spider1@glida.mooo.com] has quit [Read error: 54 (Connection reset by peer)] 12:29:59 The_Fellow [n=spider1@glida.mooo.com] has joined #lisp 12:34:01 -!- The_Fellow [n=spider1@glida.mooo.com] has left #lisp 12:37:25 fe[nl]ix: you need (when max-step (coercef max-step ...)) in EVENT-DISPATCH now 12:40:46 -!- elderK [n=zk@125-238-255-127.jetstream.xtra.co.nz] has left #lisp 12:40:47 thanks :) 12:41:59 -!- jsoft [n=user@unaffiliated/jsoft] has quit [Read error: 110 (Connection timed out)] 12:42:00 ia [n=ia@89.169.161.244] has joined #lisp 12:42:14 -!- Sergio` [n=Sergio`@a89-152-191-112.cpe.netcabo.pt] has quit [Remote closed the connection] 12:44:10 ikki [n=ikki@189.247.80.97] has joined #lisp 12:44:11 Sergio` [n=Sergio`@a89-152-191-112.cpe.netcabo.pt] has joined #lisp 12:44:23 -!- attila_lendvai [n=ati@catv-89-134-66-143.catv.broadband.hu] has quit [Read error: 110 (Connection timed out)] 12:47:53 jsoft [n=user@unaffiliated/jsoft] has joined #lisp 12:48:01 -!- lnostdal [n=lnostdal@90.149.113.175] has quit [Read error: 110 (Connection timed out)] 12:49:04 hugod [n=hugod@bas1-montreal50-1176024051.dsl.bell.ca] has joined #lisp 12:49:07 Geralt [n=Geralt@p5B32FB98.dip.t-dialin.net] has joined #lisp 12:54:42 blandest [n=blandest@softhouse.is.ew.ro] has joined #lisp 12:55:08 -!- Jasko [n=tjasko@174.59.195.12] has quit ["Leaving"] 12:56:51 -!- nowherman [n=pierre@lec67-4-82-235-57-28.fbx.proxad.net] has quit [Remote closed the connection] 12:57:06 Jasko2 [n=tjasko@c-174-59-195-12.hsd1.pa.comcast.net] has joined #lisp 12:57:43 bgs100 [n=ian@unaffiliated/bgs100] has joined #lisp 12:58:06 -!- Jafet [n=Jafet@unaffiliated/jafet] has quit [Read error: 54 (Connection reset by peer)] 12:58:15 Jafet [n=Jafet@unaffiliated/jafet] has joined #lisp 12:58:45 -!- fiveop [n=fiveop@g229064068.adsl.alicedsl.de] has quit ["humhum"] 13:02:52 merl15_ [n=merl@80-121-17-208.adsl.highway.telekom.at] has joined #lisp 13:04:34 -!- kpreid [n=kpreid@cpe-72-228-78-176.twcny.res.rr.com] has quit [] 13:04:54 kpreid [n=kpreid@cpe-72-228-78-176.twcny.res.rr.com] has joined #lisp 13:05:36 -!- Ginei_Morioka [n=irssi_lo@78.112.50.213] has quit [Read error: 110 (Connection timed out)] 13:06:22 -!- CyberBlue [n=yong@111.167.0.48] has quit ["Leaving"] 13:07:29 -!- shrughes [n=shrughes@c-65-96-172-188.hsd1.ma.comcast.net] has quit [Remote closed the connection] 13:08:02 LiamH [n=nobody@pool-141-156-235-135.res.east.verizon.net] has joined #lisp 13:10:41 cmsimon [n=cms@unaffiliated/cmsimon] has joined #lisp 13:12:02 fe[nl]ix: i think i know why darwin is misbehaving 13:12:09 (< iolib.multiplex::+far-into-the-future+ (isys:%sys-get-monotonic-time)) => T ... oops 13:12:35 so harvest-events never longer than :min-step 13:12:48 never waits, even 13:12:57 jmckitrick [n=user@adsl-176-183-44.asm.bellsouth.net] has joined #lisp 13:13:01 quek [n=read_eva@router1.gpy1.ms246.net] has joined #lisp 13:14:03 -!- jmckitrick [n=user@adsl-176-183-44.asm.bellsouth.net] has quit [Client Quit] 13:14:30 nikodemus: where do you see that code ? 13:15:18 in my repl 13:16:34 -!- YuleAthas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has quit [Remote closed the connection] 13:16:38 -!- jdz [n=jdz@85.254.211.133] has quit ["Boot me gently"] 13:16:39 just a sec, let me trace my line of thought again 13:18:45 -!- OmniMancer [n=OmniManc@122-57-20-218.jetstream.xtra.co.nz] has quit ["Leaving."] 13:18:46 Usuario059 [n=Usuario0@76.73.16.26] has joined #lisp 13:18:46 1. i know that i see busy-looping because both (time-to-next-timer timers) and (time-to-next-timer fd-timers) are both 2.0d9 for me (aka +far-into-the-future+): harvest-events substracts the monotonic clock from the min of these 13:19:00 so it's always negative and gets clamped to zero -> busy wait 13:19:14 -!- manuel_ [n=manuel@pD9E6C9E8.dip.t-dialin.net] has quit [] 13:19:29 s/harvest-events/event-dispatch/, sorry 13:19:43 -!- merl15__ [n=merl@80-121-60-5.adsl.highway.telekom.at] has quit [Read error: 110 (Connection timed out)] 13:20:40 -!- Usuario059 is now known as rus 13:21:38 aha! 13:22:22 Edward__ [n=Ed@AAubervilliers-154-1-47-235.w90-3.abo.wanadoo.fr] has joined #lisp 13:22:29 timer-to-next-timer returns +far-into-the-future+ if there is no next timer, not (abs-timeout +far-into-the-future+) 13:27:11 G0SUB [n=ghoseb@ubuntu/member/gosub] has joined #lisp 13:28:08 -!- rus [n=Usuario0@76.73.16.26] has left #lisp 13:29:48 rus_ [n=rusa@203-158-59-49.dyn.iinet.net.au] has joined #lisp 13:31:15 -!- rus_ [n=rusa@203-158-59-49.dyn.iinet.net.au] has quit [Client Quit] 13:32:15 -!- blandest [n=blandest@softhouse.is.ew.ro] has quit [Read error: 110 (Connection timed out)] 13:32:36 nikodemus: try http://common-lisp.net/~sionescu/time-to-next-timer-fix-when-no-timer.patch 13:33:24 thanks 13:33:37 WhiteFlame [i=foo@ip68-231-37-148.ph.ph.cox.net] has joined #lisp 13:33:51 -!- Phoodus [i=foo@ip68-231-37-148.ph.ph.cox.net] has quit [Read error: 104 (Connection reset by peer)] 13:34:24 -!- WhiteFlame [i=foo@ip68-231-37-148.ph.ph.cox.net] has quit [Read error: 54 (Connection reset by peer)] 13:39:28 mstevens [n=mstevens@zazen.etla.org] has joined #lisp 13:39:59 -!- Spaghettini [n=Spaghett@vaxjo6.150.cust.blixtvik.net] has quit [Read error: 110 (Connection timed out)] 13:46:22 dlowe [n=dlowe@ita4fw1.itasoftware.com] has joined #lisp 13:46:24 -!- jsoft [n=user@unaffiliated/jsoft] has quit [Success] 13:49:37 cvandusen [n=user@12.185.80.194] has joined #lisp 13:50:35 a-s [n=user@nat-240.ro.66.com] has joined #lisp 13:51:13 Ginei_Morioka [i=irssi_lo@78.114.143.52] has joined #lisp 13:51:23 Oddity [n=Oddity@99.199.136.138] has joined #lisp 13:52:15 fe[nl]ix: that ends up giving me einval... 13:52:26 (took a while untangle this...) 13:55:59 hmmm 13:56:19 ok, i have a patch 13:57:01 -!- Ginei_Morioka [i=irssi_lo@78.114.143.52] has quit [Nick collision from services.] 13:57:08 Ginei_Morioka [i=irssi_lo@78.114.143.52] has joined #lisp 14:02:58 Spaghettini [n=Spaghett@vaxjo6.150.cust.blixtvik.net] has joined #lisp 14:03:02 nikodemus pasted "this works for me at least..." at http://paste.lisp.org/display/92421 14:06:32 stradivari [i=18542ecc@gateway/web/freenode/x-aqsxnepwgmsqelml] has joined #lisp 14:06:59 -!- Geralt [n=Geralt@unaffiliated/thegeralt] has quit [Read error: 110 (Connection timed out)] 14:07:38 ...and now i can have a responsive slave thread without cpu going through the roof. \o/ 14:08:15 dandersen [n=dkcl@metabug/dandersen] has joined #lisp 14:08:21 -!- kiuma [n=kiuma@85-18-55-37.ip.fastwebnet.it] has quit ["Bye bye ppl"] 14:09:49 blandest [n=blandest@softhouse.is.ew.ro] has joined #lisp 14:10:35 -!- cmsimon [n=cms@unaffiliated/cmsimon] has quit [Read error: 104 (Connection reset by peer)] 14:10:55 cmsimon [n=cms@unaffiliated/cmsimon] has joined #lisp 14:11:09 -!- cmsimon [n=cms@unaffiliated/cmsimon] has quit [Remote closed the connection] 14:13:08 -!- legumbre_ is now known as legumbre 14:13:41 redline6561 [n=redline@c-66-56-16-250.hsd1.ga.comcast.net] has joined #lisp 14:15:13 manuel_ [n=manuel@pD955542E.dip.t-dialin.net] has joined #lisp 14:15:20 nikodemus: could you also try http://common-lisp.net/~sionescu/fix-make-timer.patch ? 14:16:35 -!- SandGorgon [n=OmNomNom@122.160.41.129] has quit [Read error: 60 (Operation timed out)] 14:22:18 bad_alloc [n=marvin@HSI-KBW-078-042-137-224.hsi3.kabel-badenwuerttemberg.de] has joined #lisp 14:23:57 -!- manuel_ [n=manuel@pD955542E.dip.t-dialin.net] has quit [] 14:24:47 hello, i've got a function with an internal loop that is supposed to search an array for consecutive values either backwards or forwards, depending on the arguments it gets passed. so (func 0 30) should search forwards while (func 30 0) should search backwards. how do i implement this in the loop macro without duplicating the loop in an if/else statement? 14:24:49 seems to work fine with cursory wheel-kicking here 14:25:18 nikodemus: does "Fix computation of the event loop timeout in case of no timers." sound good as commit message ? 14:25:38 sounds fine to me :) 14:26:04 -!- sadiquea [n=sadiquea@119.82.102.202] has quit ["Leaving."] 14:26:27 bad_alloc: Pass the start value, the end value, and a step value 14:26:52 to go backwards you'd pass -1 as step value 14:26:58 or use SEARCH, maybe? 14:27:15 nikodemus: thanks. I pushed the changes :) 14:27:21 or FIND 14:27:34 tcr: thanks 14:27:43 oh consecutive values, well :-) 14:27:44 fe[nl]ix: do you have david's other patches in upstream yet? 14:27:47 can search or find find a series of free places? 14:28:05 i have to fight fragmentation here 14:28:27 -!- Stattrav [n=Stattrav@202.3.77.132] has quit [Read error: 110 (Connection timed out)] 14:28:33 what do you mean by a place? 14:28:38 nyef [n=nyef@pool-71-161-71-17.cncdnh.east.myfairpoint.net] has joined #lisp 14:28:47 G'morning all. 14:28:59 nikodemus: not yet. as soon as he sends them to me I will 14:29:02 jmckitrick [n=user@adsl-176-183-44.asm.bellsouth.net] has joined #lisp 14:29:02 hi nyef 14:29:11 Anything interesting happening? 14:29:31 davazp [n=user@89.Red-83-46-6.dynamicIP.rima-tde.net] has joined #lisp 14:29:36 nikodemus: i have a round table and i have to seat as many groups there as possible without having to send out groups who dont fit in the gaps, so i want to put small groups richt and big ones left 14:29:38 hi nyef... long time, no chat 14:29:49 *nikodemus* has been hacking on hemlock.tty today 14:30:25 unfortunately, now's the time for real work 14:30:29 nvoorhies [n=nvoorhie@adsl-76-216-21-95.dsl.pltn13.sbcglobal.net] has joined #lisp 14:30:40 So, it looks like it's possible to determine if a given interrupt context was for an internal-error trap or not. 14:30:46 At least, on x86oid linux. 14:31:58 And I'm thinking that it might be useful to be able to ask ldb to describe the internal error for such contextx. 14:32:14 manuel_ [n=manuel@pD955542E.dip.t-dialin.net] has joined #lisp 14:32:22 (Saves having to build with QSHOW.) 14:32:51 -!- cvandusen [n=user@12.185.80.194] has quit [Remote closed the connection] 14:33:06 sorry, i mean what do you mean by "place" in the context of lisp. an array element? so a free place would be an array element that doesn't have your person-object there yet, but eg. NIL? sure, you can search for eg. #(nil nil nil) 14:33:51 nyef: you'll hear no objections from me 14:34:10 (it would also be nice if ldb printed out the error name instead of just the code...) 14:34:14 hi jmckitrick 14:34:19 nikodemus: by place i mean a chair at a table. nothing lispy at all 14:34:40 nikodemus: http://repo.or.cz/w/sbcl/nyef.git/commitdiff/df98deecfaf7aa5ab24ea4c4dda740ea19554660 14:34:53 cvandusen [n=user@12.185.80.194] has joined #lisp 14:35:01 i'd post the task, but it is in german and quite long 14:35:24 "can search or find find a series of free places?" -- since neither SEARCH not find knows anything about your problem domain, no 14:35:44 Geralt [n=Geralt@p5B32FB98.dip.t-dialin.net] has joined #lisp 14:35:45 good, i'll do it myself, thanks for the pointer to the steps 14:35:55 but if you represent the problem with an array, sure 14:36:29 nyef: commit! 14:37:10 Okay, one more from the stupid-lowtags-tricks tree for mainline. 14:37:50 -!- ikki [n=ikki@189.247.80.97] has quit ["Leaving"] 14:38:10 -!- blandest [n=blandest@softhouse.is.ew.ro] has quit [Read error: 60 (Operation timed out)] 14:38:36 i'll try to explain: i habe a table whick is supposed to be round with 28 places. groups of random size will enter and take a seat. if no range of free seats of the size of the group is avaliable the group will leave. it is my task to arrange the groups in a way that as few as possible leave without taking a seat. 14:39:17 ... You get to control the group ordering, do you also control the seat allocation policy? 14:39:38 only the seat allocation 14:39:41 wait 14:39:52 Oh, I see. 14:39:54 i control where the groups sit down not thier size 14:40:01 You don't control the ordering, just the allocation policy. 14:40:23 is there a similar problem i could loom up? 14:40:27 *look 14:40:32 Do you know the distribution for the groups? 14:41:05 what distribution do you mean? (my enmglish isn't terribly good, sorry) 14:41:08 Do the groups leave after a while, or do you end up with a full table? 14:41:12 Probability distribution. 14:41:24 blandest [n=blandest@softhouse.is.ew.ro] has joined #lisp 14:41:24 it doesn't say in the task, i guess they leave 14:41:44 Min, Max, Mean, Stddev? 14:41:56 -!- Edward__ [n=Ed@AAubervilliers-154-1-47-235.w90-3.abo.wanadoo.fr] has quit [Read error: 110 (Connection timed out)] 14:42:08 28 > size >= 1 14:42:12 Edward__ [i=Ed@AAubervilliers-154-1-60-144.w90-3.abo.wanadoo.fr] has joined #lisp 14:42:34 well if we give each of them 2 forks to eat their spaghetti with... 14:42:37 i generate them by an (+ 1 (random 27)) 14:42:50 28 philosophers? 14:42:52 If groups don't leave once seated, then it's fairly straightforward. 14:42:55 JonSmith: Heh. 14:43:13 lets assume they do leave 14:43:25 If they do leave then how long do they stay? 14:43:41 dmiles [n=dmiles@c-76-104-220-73.hsd1.wa.comcast.net] has joined #lisp 14:43:51 one iteration maybe 14:43:58 so every time a group enters one leaves 14:44:33 this looks a lot like process scheduling 14:44:41 -!- stradivari [i=18542ecc@gateway/web/freenode/x-aqsxnepwgmsqelml] has quit [Ping timeout: 180 seconds] 14:44:43 what happens when the philosopher from the naboring table grabbs the fork? (MP) 14:44:54 -!- blandest [n=blandest@softhouse.is.ew.ro] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- dandersen [n=dkcl@metabug/dandersen] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- a-s [n=user@nat-240.ro.66.com] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- dlowe [n=dlowe@ita4fw1.itasoftware.com] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- mstevens [n=mstevens@zazen.etla.org] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- quek [n=read_eva@router1.gpy1.ms246.net] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- kpreid [n=kpreid@cpe-72-228-78-176.twcny.res.rr.com] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- dmiles_afk [n=dmiles@c-76-104-220-73.hsd1.wa.comcast.net] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- trebor_dki [n=user@mail.dki.tu-darmstadt.de] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- mvilleneuve [n=mvillene@LLagny-156-36-4-214.w80-14.abo.wanadoo.fr] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- Buganini [n=buganini@security-hole.info] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- saikatc [n=saikatc@c-98-210-192-23.hsd1.ca.comcast.net] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- akamaus [n=maus@94.231.112.177] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- spoofy [n=spoof@78.31.74.25] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- tic [n=tic@c83-249-193-189.bredband.comhem.se] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- mikezor [n=mikael@c-5de570d5.04-404-7570701.cust.bredbandsbolaget.se] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- yahooooo [n=yahooooo@c-67-170-35-27.hsd1.wa.comcast.net] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- j0ni_ [n=joni@192.219.30.200] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- Trystam [n=Tristam@cpe-72-226-127-57.nycap.res.rr.com] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- r0bby [n=wakawaka@guifications/user/r0bby] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- mathrick [n=mathrick@83.1.168.198] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- ecraven [n=nex@octonex.swe.uni-linz.ac.at] has quit [crichton.freenode.net irc.freenode.net] 14:44:54 -!- borisc [n=borisc@borisc2.csbnet.se] has quit [crichton.freenode.net irc.freenode.net] 14:45:02 nunb [n=nundan@static-217-133-104-225.clienti.tiscali.it] has joined #lisp 14:45:59 r0bby [n=wakawaka@guifications/user/r0bby] has joined #lisp 14:46:59 This looks like it could be an interesting problem if it were properly parameterized. 14:47:29 JonSmith: Actually looks more like allocation within a ring buffer. 14:47:59 right 14:48:31 nyef: shall i translaet it for you? 14:48:35 *tanslate 14:48:38 You get to learn something about allocation policy, there are a number of interesting low-level management tricks you can learn in the actual implementation... 14:48:56 i was just looking for the algorithm 14:49:17 bad_alloc: No, thank you. I have my own set of interesting problems to work on today. 14:49:41 kay, just trying 14:49:57 well i'll leave you to it, thanks for the help 14:50:12 -!- bad_alloc [n=marvin@HSI-KBW-078-042-137-224.hsi3.kabel-badenwuerttemberg.de] has quit ["You shoot yourself in somebody else's foot."] 14:51:22 blandest [n=blandest@softhouse.is.ew.ro] has joined #lisp 14:51:22 dandersen [n=dkcl@metabug/dandersen] has joined #lisp 14:51:22 a-s [n=user@nat-240.ro.66.com] has joined #lisp 14:51:22 dlowe [n=dlowe@ita4fw1.itasoftware.com] has joined #lisp 14:51:22 mstevens [n=mstevens@zazen.etla.org] has joined #lisp 14:51:22 kpreid [n=kpreid@cpe-72-228-78-176.twcny.res.rr.com] has joined #lisp 14:51:22 trebor_dki [n=user@mail.dki.tu-darmstadt.de] has joined #lisp 14:51:22 mvilleneuve [n=mvillene@LLagny-156-36-4-214.w80-14.abo.wanadoo.fr] has joined #lisp 14:51:22 Buganini [n=buganini@security-hole.info] has joined #lisp 14:51:22 saikatc [n=saikatc@c-98-210-192-23.hsd1.ca.comcast.net] has joined #lisp 14:51:22 akamaus [n=maus@94.231.112.177] has joined #lisp 14:51:22 spoofy [n=spoof@78.31.74.25] has joined #lisp 14:51:22 tic [n=tic@c83-249-193-189.bredband.comhem.se] has joined #lisp 14:51:22 mikezor [n=mikael@c-5de570d5.04-404-7570701.cust.bredbandsbolaget.se] has joined #lisp 14:51:22 yahooooo [n=yahooooo@c-67-170-35-27.hsd1.wa.comcast.net] has joined #lisp 14:51:22 Trystam [n=Tristam@cpe-72-226-127-57.nycap.res.rr.com] has joined #lisp 14:51:22 j0ni_ [n=joni@192.219.30.200] has joined #lisp 14:51:22 mathrick [n=mathrick@83.1.168.198] has joined #lisp 14:51:22 ecraven [n=nex@octonex.swe.uni-linz.ac.at] has joined #lisp 14:51:22 borisc [n=borisc@borisc2.csbnet.se] has joined #lisp 14:52:01 can't remember the algorithm name, is when you only run the process if it can get all of its pages into memory 14:52:27 working set 14:52:41 Wouldn't surprise me if it's covered in Tenenbaum (sp?). 14:52:59 i think it is 14:53:19 And by that I mean the minix book, not the other one (amoeba?) 14:53:55 Tanenbaum, Operating systems.. Design and implementation 14:54:37 Ooh. It's dying in the compiler, trying to make source info for a pprint-dispatch functio. 14:54:42 s/tio/tion/ 14:55:32 -!- tcr [n=tcr@host146.natpool.mwn.de] has quit ["Leaving."] 14:56:16 -!- jmckitrick [n=user@adsl-176-183-44.asm.bellsouth.net] has quit ["ERC Version 5.2 (IRC client for Emacs)"] 14:56:50 -!- davazp [n=user@89.Red-83-46-6.dynamicIP.rima-tde.net] has quit [Remote closed the connection] 14:58:03 i've got the newest version of it and itis in there 14:58:10 (the tanenbaum book) 14:59:08 ruediger [n=the-rued@p508B380E.dip.t-dialin.net] has joined #lisp 15:04:59 Hey, there's a "maybe delete it towards the end of 2009" comment about unix-gettimeofday in src/code/unix.lisp. 15:06:27 we really need a queue of some kind of for that kind of things 15:07:28 tcr [n=tcr@host146.natpool.mwn.de] has joined #lisp 15:09:02 jmckitrick [n=user@adsl-176-183-44.asm.bellsouth.net] has joined #lisp 15:09:32 I have a quick question about argument lists for functions.... 15:09:49 nowhere_man [n=pierre@lec67-4-82-235-57-28.fbx.proxad.net] has joined #lisp 15:10:10 What would &key ((:foo *global-here*) :whole) mean? 15:10:53 jmckitrick: it means that in the call, use :foo for the keyword, and bind *global-here* in the binding, with a default of :whole if no arg is passed. 15:11:14 so your function would be (froobie :foo 42) and in the body of froobie, *global-here* would be 42. 15:11:18 I thought the defaults went inside the sexp with the keyword... 15:11:20 ok, I see. 15:12:02 i used that a lot in zs3 to dynamically bind the global *CREDENTIALS* from the :credentials keyword arg 15:12:36 So it's sort of like a LET around the body of the functions, which binds a global for the scope of that function.... only it does it in the arglist instead. 15:13:19 nikodemus: Isn't there some sort of "deprecated function" warning thing? 15:13:31 jmckitrick: no. 15:13:56 jmckitrick: it's a way to use a keyword name that differs from the binding name in the body. 15:14:18 jmckitrick: it's not specifically about globals, but that's a convenient way to use it. 15:14:25 fihi09``` [n=user@pool-96-224-165-27.nycmny.east.verizon.net] has joined #lisp 15:14:38 Note that you can also use this to have non-keywordp keyword arguments. 15:14:43 Ah, ok. I've not seen that before, so I'll have to study it. 15:14:54 (Or do I mean non-keywordp argument keywords?) 15:15:03 jmckitrick: it's not very common. i can never remember the syntax. when you need it, though, it's pretty handy. 15:15:25 -!- sellout [n=greg@c-24-128-48-180.hsd1.ma.comcast.net] has quit [] 15:15:37 The syntax gets even more tricky when you destructure &key parameters in macro lambda lists! 15:16:58 leo2007 [n=leo@smaug.linux.pwf.cam.ac.uk] has joined #lisp 15:17:36 tcr: Guess what I'm working on.... LOL 15:17:58 sellout [n=greg@c-24-128-48-180.hsd1.ma.comcast.net] has joined #lisp 15:19:01 jewel [n=jewel@vc-41-28-18-28.umts.vodacom.co.za] has joined #lisp 15:19:06 Will be a cool christmas gift to the sbcl community 15:21:06 kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 15:21:32 dv_ [n=dv@83-64-248-68.inzersdorf.xdsl-line.inode.at] has joined #lisp 15:23:35 tcr: So do we want this to mimic the original browser-based functionality? 15:24:12 mimic in what way? 15:24:19 I.E. Calling the report function in slime opens the popup with the stats and file list? 15:24:54 The start page (cover-index.html) has a nice table there with stats. 15:25:05 Yes 15:25:18 but I would do that at last 15:25:25 first cycling by 15:25:31 `p', `n' should be enough 15:26:36 Ok, so we will show the list of files, p and n will cycle through, and selecting will jump to the buffer with the markup. Correct? 15:27:57 `p', and `n' should also work in the source buffers 15:27:58 *Xach* wonders what's going on 15:28:09 Xach: slime contrib for sb-cover 15:28:29 p and n will jump to each annotated form? 15:28:41 jmckitrick: To the next file in the file list 15:29:00 but I don't know, that are details 15:29:25 first get it working, I'll pester you with detail work when I can try it :-) 15:30:29 Stattrav [n=Stattrav@202.3.77.161] has joined #lisp 15:30:33 cover-in-slime? 15:30:42 wasn't there a nickelodon show about that? 15:31:59 -!- fihi09`` [n=user@pool-71-190-77-62.nycmny.east.verizon.net] has quit [Connection timed out] 15:32:29 -!- TuxPurple [n=TuxPurpl@unaffiliated/tuxpurple] has quit [Connection timed out] 15:35:52 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 15:39:43 *stassats`* once played with displaying results of ccl's coverage in slime 15:40:05 but i was scared by the code i wrote and deleted it 15:40:30 -!- nipra [n=nipra@121.243.225.226] has quit [Read error: 110 (Connection timed out)] 15:40:51 legumbre_ [n=leo@r190-135-11-219.dialup.adsl.anteldata.net.uy] has joined #lisp 15:41:48 You're invited to make slime-cover work for ccl. Shouldn't be too hard once jmckitrick is done 15:42:09 -!- galaxywatcher [n=galaxywa@ppp-58-8-53-124.revip2.asianet.co.th] has quit [Read error: 110 (Connection timed out)] 15:42:12 i even found a screenshot: http://common-lisp.net/~sboukarev/slime-coverage.png 15:42:35 -!- legumbre [n=leo@r190-135-0-222.dialup.adsl.anteldata.net.uy] has quit [Read error: 60 (Operation timed out)] 15:43:02 :-) 15:44:05 jleija [n=jleija@adsl-91-0-131.chs.bellsouth.net] has joined #lisp 15:47:17 tcr: and how to deal with contribs which should use implementation dependant features? putting them in backends breaks optionality of contribs 15:48:40 reader-conditionalization in contrib/swank-cover.lisp? 15:48:55 spradnyesh [n=pradyus@122.167.99.221] has joined #lisp 15:49:12 -!- bipt [i=bpt@cpe-173-095-174-230.nc.res.rr.com] has quit [Read error: 60 (Operation timed out)] 15:49:56 it's an obvious solution, but it's not very nice 15:50:30 Why not? I find it better than having a swank-backends in contrib/ 15:50:45 this way you directly see where and how the stuff is used 15:53:18 -!- ruediger [n=the-rued@p508B380E.dip.t-dialin.net] has quit ["This computer has gone to sleep"] 15:54:03 -!- nunb [n=nundan@static-217-133-104-225.clienti.tiscali.it] has quit ["This computer has gone to sleep"] 15:54:12 now, sbcl doesn't record source locations inside progns, perhaps i should open a ticket 15:55:03 how do you mean? 15:55:29 ruediger [n=the-rued@p508B380E.dip.t-dialin.net] has joined #lisp 15:56:15 -!- Sergio` [n=Sergio`@a89-152-191-112.cpe.netcabo.pt] has quit [Read error: 110 (Connection timed out)] 15:56:16 (progn (defun foo ())) M-. jumps to the progn, not defun 15:58:01 -!- merl15_ is now known as merl15 15:58:36 stassats`: Ah yeah I know 15:58:51 that's easy to fix, I have a branch somewhere with a fix. 16:04:59 nikodemus: I committed my autodoc improvements. Please update. 16:05:45 anyone knows how SBCL optimizes slotaccessing when you change your class at runtime 16:05:50 Nshag [n=shag@lns-bzn-37-82-253-13-222.adsl.proxad.net] has joined #lisp 16:06:56 ikki [n=ikki@201.155.75.146] has joined #lisp 16:07:14 -!- mrSpec is now known as spec[away] 16:09:57 PissedNumlock: as it happens, that is one of things that is well documented in the sbcl internals manual 16:10:15 Isn't it "just" a method call or two? 16:10:30 -!- leo2007 [n=leo@smaug.linux.pwf.cam.ac.uk] has quit ["back-to-work"] 16:11:05 Krystof: do you have a link? I'm looking through the internals wiki 16:11:07 but don't find it 16:11:48 doc/internals in the CVS 16:12:05 http://www.sbcl.org/sbcl-internals/ 16:12:31 Broken link on www.sbcl.org/porting.html points to ww.telent.net/sbcl-internals/ for the cliki. 16:12:44 oh, great :) I was looking here: http://sbcl-internals.cliki.net/index 16:12:52 googling for "sbcl internals manual" should find it 16:13:20 I was looking at http://www.sbcl.org/manual/Internals-Documentation.html#Internals-Documentation 16:13:23 maybe it's worth a link there too? 16:14:37 Oh, neat. My stupid-lowtag-tricks tree actually builds without error for SBCL_ARCH=x86. 16:15:09 HG` [n=HG@xdslfa089.osnanet.de] has joined #lisp 16:15:10 there is a lot wrong with the website 16:15:24 if only there were a web coder willing to take it on! 16:16:00 anyway, PissedNumlock, read chapters 3 and 7 of that 16:16:07 It is in the nature of websites maintained by volunteers that they are ALWAYS out of date :) 16:16:38 rsynnott: hmm, my facebook page is maintained by an army of unpaid volunteers, and i can barely keep up... 16:16:57 *Xach* ponders sbcl-as-social-game trickery 16:17:44 nyef: hurrah 16:17:59 it might be better as some form of wiki :) 16:18:13 (then when people moaned on IRC someone could go and change it to shut them up) 16:18:50 -!- sellout [n=greg@c-24-128-48-180.hsd1.ma.comcast.net] has quit [] 16:19:22 -!- blandest [n=blandest@softhouse.is.ew.ro] has quit ["Leaving."] 16:20:27 *Xach* can send patches against sbcl/doc/manual/intro.texinfo 16:21:27 well, yes, but that's far more trouble :) 16:21:34 -!- grouzen [n=grouzen@91.214.124.2] has quit [Read error: 110 (Connection timed out)] 16:23:16 Eek. 1.0.33.20 is likely to collide my current working tree something awful. :-/ 16:23:49 Mostly for being duplication of changes that are -in- said tree. 16:24:49 Heh. that sucks. I wonder what's a good merging tool? 16:25:34 -!- hypno [n=hypno@impulse2.gothiaso.com] has left #lisp 16:25:51 I have had one isolated case of success os x's builtin FileMerge tool 16:26:29 Hrm... And the next-to-last change in src/compiler/x86-64/move is less than ideal. 16:26:35 I find merging tools completely useless if they don't know how to explain -why- there two conflicting versions, i.e. they need to show the history of each side. 16:26:37 Oh well. 16:26:46 lispm [n=joswig@e177125196.adsl.alicedsl.de] has joined #lisp 16:26:50 (Haven't yet seen a tool that gets it right though.) 16:27:05 lichtblau: that's true. history is more useful than patches. 16:28:21 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 16:28:40 nyef: doh, sorry about that 16:29:05 froydnj: Nah, don't worry about it. You may even have found a place or two that I missed, and it needed doing in -some- form anyway. 16:29:22 oh, bleh, my sbcl git committer email is something long-ago dead 16:29:30 *nyef* has a pile of stuff that's both adding constants and breaking the assumption that n-fixnum-tag-bits = word-shift. 16:30:29 Okay, that's the internal-error descriptions in runtime thing committed. 16:31:15 SandGorgon [n=OmNomNom@122.162.51.84] has joined #lisp 16:32:15 hah. i can't seem to untrace a macro-character function. 16:32:52 how did you manage to trace it in the first place? 16:32:52 *Xach* re-evaluates the set-macro-character call 16:32:55 Xach: Untrace #.(character-macro-function) ? 16:33:16 Or whatever the magic is? 16:33:26 rahul: trace @-reader function, set-macro-character #\@ '@-reader, i think 16:33:32 Xach: sounds about right 16:33:35 Xach: I realized the other day that my trees library does the exact eager coercion of symbols to functions that you were grousing about the other day =/ 16:33:45 Xach: so yeah, you'd need to set again after untracing 16:33:50 -!- mstevens [n=mstevens@zazen.etla.org] has quit ["leaving"] 16:33:50 yay for grey areas in the spec 16:33:51 *Xach* was more ptarmiganing about it 16:36:05 Xach: that is taking verb creation a step too far 16:36:42 tcr: Does the 'n' and 'p' stuff require creation of a minor mode? 16:36:51 -!- jewel [n=jewel@vc-41-28-18-28.umts.vodacom.co.za] has quit [Read error: 110 (Connection timed out)] 16:38:14 jmckitrick: Yes 16:38:28 jmckitrick: I'm not sure if it requires it, but you should do it. 16:38:45 jmckitrick: look at the macroexpansion stuff in slime.el for inspiration 16:38:59 jmckitrick: or xref stuff 16:39:17 tcr: I'll grab my 'Writing GNU Emacs Extensions' book off the shelf for this. 16:39:28 Now I'll get some use out of it. 16:39:38 kejsaren1 [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 16:40:28 reprore [n=reprore@ntkngw356150.kngw.nt.ftth.ppp.infoweb.ne.jp] has joined #lisp 16:41:07 jewel [n=jewel@vc-41-30-129-104.umts.vodacom.co.za] has joined #lisp 16:41:46 fe[nl]ix: is there a reason why iolib syscall-error doesn't include the name of the syscall that failed in the condition? 16:42:31 since DEFSYSCALL inlines the call, it can be quite had to figure just what is going wrong where... 16:43:10 grouzen [n=grouzen@91.214.124.2] has joined #lisp 16:43:19 -!- kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 16:46:06 -!- HG` [n=HG@xdslfa089.osnanet.de] has quit [Client Quit] 16:47:32 froydnj, nyef: one of the reasons to switch to git properly is to help in these situations. If either of you know whether it would actually have helped, could you explain how to me? 16:49:19 rpg: here? 16:49:43 marioxcc [n=user@200.92.185.171] has joined #lisp 16:53:21 -!- manuel_ [n=manuel@pD955542E.dip.t-dialin.net] has quit [] 16:54:14 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 16:55:05 Krystof: I don't think it would have helped in this situation, though it might help nyef do his rebase-for-final-commit. 16:55:16 Krystof: Actually, I'm not sure it matters. Either way, it's got to be merged somehow. 16:55:55 it might have avoided the situation, if nyef and froydnj had been aware of the work being duplicated 16:56:10 which is a pretty big if, i admit :) 16:56:17 I was just thinking about rewriting my branch history and rebasing when I found it, though. 16:56:41 rread [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has joined #lisp 16:56:48 mstevens [n=mstevens@zazen.etla.org] has joined #lisp 16:59:03 -!- rread [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has quit [Remote closed the connection] 16:59:09 rread [n=rread@nat/sun/x-xqgpjdxmsjrkgwlh] has joined #lisp 17:00:46 Sergio` [n=Sergio`@a89-152-191-112.cpe.netcabo.pt] has joined #lisp 17:01:40 -!- xan [n=xan@155.99.117.91.static.mundo-r.com] has quit [Read error: 110 (Connection timed out)] 17:02:40 nikodemus: Okay, the commit you requested has shown up in git. 17:03:46 sellout [n=greg@c-24-128-48-180.hsd1.ma.comcast.net] has joined #lisp 17:05:13 kiuma [n=kiuma@93-35-225-51.ip57.fastwebnet.it] has joined #lisp 17:06:31 -!- SandGorgon [n=OmNomNom@122.162.51.84] has quit [Read error: 110 (Connection timed out)] 17:08:30 -!- mvilleneuve [n=mvillene@LLagny-156-36-4-214.w80-14.abo.wanadoo.fr] has quit ["Lost terminal"] 17:08:31 -!- spradnyesh [n=pradyus@122.167.99.221] has left #lisp 17:08:35 nikodemus: that's a pity about the fork problem on MacOS. I had some of that code in C originally, and was glad to find that a Lisp version worked as well. 17:09:17 Question for all the people here who toyed around with fork() stuff in Lisp recently: Did you happen to make that work on MacOS, and if so, how? 17:10:41 ... fork() stuff? 17:11:15 xan [n=xan@155.99.117.91.static.mundo-r.com] has joined #lisp 17:11:47 nyef: ISTR a fork discussion in the context of xcvb, BICBW. 17:11:52 Ah. 17:12:23 tcr: I have a list of cons cells mapping short filenames to full pathnames. In CL, I'd use string= or something like that to look up the long ones. Do you recommend something similar in elisp, or a different approach to storing the data? 17:12:24 nikodemus: what is the failure mode, BTW? 17:12:55 So, nothing like fork()ing a threaded lisp and hoping that the other threads either survive or are properly cleaned-up? 17:13:32 (I know how to go about making either happen, but am not even remotely convinced that it's a good idea.) 17:14:43 -!- jewel [n=jewel@vc-41-30-129-104.umts.vodacom.co.za] has quit [Read error: 110 (Connection timed out)] 17:16:38 nyef: well, all I want to do is to call exec() right away, but according to nikodemus even that fails because that mach exception handling thread. 17:16:48 Oh, ouch. 17:17:10 ... slyrus is the person I'd ask about that. 17:17:36 lichtblau: lose("Unhandled memory fault. Exiting."); 17:17:38 I guess there's a reason SBCL has 'spawn' implemented in C. :-) 17:17:54 in src/runtime/bsd-os.c 17:18:00 Heh. 17:18:19 nikodemus: hmmm. Perhaps it needs a #+sbcl (unprotect-the-entire-heap)? GC is disabled at that point anyway. 17:18:21 tcr: just back... 17:18:47 Wait, wait... -What's- your failure mode? 17:18:50 authentic [n=authenti@85-127-38-99.dynamic.xdsl-line.inode.at] has joined #lisp 17:18:55 Is this for accessing the strings you pass to exec()? 17:19:14 And if so, how about allocating them as alien strings prior to forking? 17:19:29 -!- redline6561 [n=redline@c-66-56-16-250.hsd1.ga.comcast.net] has quit [Read error: 110 (Connection timed out)] 17:19:33 -!- cvandusen [n=user@12.185.80.194] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 17:19:34 lichtblau: i don't see the gc being disabled... 17:19:56 jewel [n=jewel@vc-41-27-46-220.umts.vodacom.co.za] has joined #lisp 17:20:54 -!- nikodemus [n=nikodemu@cs181150041.pp.htv.fi] has quit ["Leaving"] 17:22:38 snowbeard [n=snowbear@cpe-065-190-191-189.nc.res.rr.com] has joined #lisp 17:23:06 wonder if we can do without that hack on 10.6 17:23:28 lnostdal [n=lnostdal@90.149.113.175] has joined #lisp 17:24:44 milanj [n=milan@93.87.142.53] has joined #lisp 17:25:31 cvandusen [n=user@12.185.80.194] has joined #lisp 17:26:47 -!- reprore [n=reprore@ntkngw356150.kngw.nt.ftth.ppp.infoweb.ne.jp] has quit [Remote closed the connection] 17:27:32 nikodemus [n=nikodemu@cs181150041.pp.htv.fi] has joined #lisp 17:29:50 freiksenet [n=freiksen@hoasnet-fe29dd00-202.dhcp.inet.fi] has joined #lisp 17:30:26 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 17:32:32 pkhuong: might be worth seeing if the posix stuff is up to snuff even in 10.5 17:32:48 -!- Patzy [n=somethin@coa29-1-88-174-11-170.fbx.proxad.net] has quit [Read error: 104 (Connection reset by peer)] 17:32:54 i think the last time slyrus looked at it and decided that it's not good enough was around 10.4 17:32:57 Patzy [n=somethin@coa29-1-88-174-11-170.fbx.proxad.net] has joined #lisp 17:34:22 So, there's this comment in src/code/debug-int at around line 1606 about how genesis can't handle a #.(find-package ...) form. Does anyone know if this is actually -true-? 17:36:48 ... Guess it's irrelevant: the xc won't even -dump- such a thing. 17:38:15 Makes sense, though, as the package objects aren't even created until partway through cold-init. 17:38:39 Might be possible to dump something that would cause a usable l-t-v to appear, though. 17:38:54 a package-cell :) 17:39:02 I don't get it. SBCL is claiming that defmethod defines a new type for args named "stream" 17:39:05 -!- mstevens [n=mstevens@zazen.etla.org] has quit ["leaving"] 17:39:12 -!- Madsy [n=madsy@fu/coder/madsy] has quit ["leaving"] 17:39:16 Lock on package COMMON-LISP violated when declaring the type of STREAM. 17:39:38 madsy [n=madsy@ti0050a380-2204.bb.online.no] has joined #lisp 17:39:39 lisppaste the code? 17:39:42 rahul: You specializing on that arg? 17:39:50 nyef: no 17:39:56 Hrm... 17:39:57 nikodemus: it's mcclim 17:39:59 ... dunno. 17:40:07 just a print-object method with arg name stream 17:40:07 -!- madsy is now known as Madsy 17:40:23 also happens for accept presentation methods, fwiw 17:40:44 sbcl 1.0.31 17:41:35 rahul: i can take a quick look if you tell me the file and method to look at 17:42:56 And if you can produce a minimal test case to reproduce it, bisection! 17:43:43 kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 17:44:29 presentations.lisp line 47 17:44:56 -!- kejsaren1 [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 17:46:04 how do you trigger the error? 17:46:16 odd. minimal examples don't have the problem 17:46:26 let me see if clearing the fasls helps 17:46:51 -!- PeakerWork [n=PeakerWo@nat/ibm/x-ywmrljmjcydwohqa] has left #lisp 17:47:38 nyef pasted "Proposed debug-int change, comments?" at http://paste.lisp.org/display/92433 17:48:12 -!- silenius [n=jl@yian-ho03.nir.cronon.net] has quit [] 17:48:32 It's less KLUDGEy, but it still feels a bit like a KLUDGE. 17:49:05 looks on to me 17:49:18 ok, even 17:50:34 In the comment being removed there, it seems that there's a desire to reduce the amount of stuff in the cold-core. 17:50:39 nikodemus: hmm, you're right. I thought hi::invoke-without-interrupts did that, but it's really only for CCL at the moment. 17:50:51 nyef: looks good. 17:51:01 -!- Oddity [n=Oddity@99.199.136.138] has quit [] 17:51:43 Is less stuff in the cold-core still a desireable thing, and if so why? 17:52:26 -!- Reaver [n=Data_Ent@h15a2.n2.ips.mtn.co.ug] has quit [Connection timed out] 17:52:53 -!- varjag [n=eugene@122.62-97-226.bkkb.no] has quit ["Ex-Chat"] 17:55:24 hmm looks like clearing fasls did the trick 17:56:31 reprore [n=reprore@ntkngw356150.kngw.nt.ftth.ppp.infoweb.ne.jp] has joined #lisp 17:57:55 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 17:59:20 -!- Sergio` [n=Sergio`@a89-152-191-112.cpe.netcabo.pt] has quit [Remote closed the connection] 17:59:48 hm, i could get Grep occasionally working earlier, but now all i get is EINVAL 17:59:58 attila_lendvai [n=ati@catv-89-134-66-143.catv.broadband.hu] has joined #lisp 18:00:13 oops, ENOENT 18:00:25 manuel_ [n=manuel@pD9E6D8CC.dip.t-dialin.net] has joined #lisp 18:00:57 ... Actually, how kludgey is it to use l-t-v find-package instead of #. find-package ? 18:01:35 more right to me. 18:02:31 You'd rather use l-t-v? 18:02:35 yes. 18:02:36 dumping a package to fasl seems odd 18:02:44 if that's what you're talking about 18:02:49 (I think I would as well, #. usually feels like a kludge to me.) 18:03:18 nikodemus: Missing directory perhaps. What does M-: (hemlock::default-directory) RET say? 18:03:33 rahul: Yeah, dumping a literal package reference. It's legal, as they have print-read consistency, but it doesn't work in the xc because the package structures aren't set up by that point. 18:04:23 "" 18:04:34 nyef: yep, that would be an even worse place to do that :) 18:05:38 I still think this needs a comment, though. Just not sure -what-. 18:05:42 clim frame-layout 18:05:43 Sorry, I couldn't find anything for frame-layout. 18:05:50 clim layout-frame 18:05:51 http://bauhh.dyndns.org:8000/clim-spec/28-3.html#_1512 18:05:55 clim frame-current-layout 18:05:56 http://bauhh.dyndns.org:8000/clim-spec/28-3.html#_1509 18:06:16 Sergio` [n=Sergio`@a89-152-191-112.cpe.netcabo.pt] has joined #lisp 18:07:10 ok, when frame-current-layout is changed, McCLIM does not re-layout the frame 18:07:26 -!- marcoecc [i=me@gateway/gpg-tor/key-0x9C9AAE7F] has quit [Remote closed the connection] 18:07:34 I think it just needs to call note-space-requirements-changed 18:07:59 -!- skeptomai|away is now known as skeptomai 18:08:32 ... I remember running into that years ago. 18:08:40 demmeln [n=Adium@dslb-188-099-119-199.pools.arcor-ip.net] has joined #lisp 18:08:54 jmckitrick: still here? 18:09:32 I think what I ended up doing was setting the layout I wanted before starting the toplevel, and abandoning the project before getting to the point where I'd have to make it work at runtime. 18:09:43 dnolen [n=dnolen@ppp-70-249-155-206.dsl.rcsntx.swbell.net] has joined #lisp 18:10:27 "In the case that layout-frame is invoked, it isn't necessary to call note-space-requirements-changed since a complete re-layout of the frame will be executed." 18:10:37 nyef: resizing the window works for me 18:11:12 Yes, but the point is that I wanted it to happen entirely under program control, not with a user interaction step. 18:11:44 -!- jewel [n=jewel@vc-41-27-46-220.umts.vodacom.co.za] has quit [Read error: 110 (Connection timed out)] 18:12:22 -!- demmeln [n=Adium@dslb-188-099-119-199.pools.arcor-ip.net] has left #lisp 18:12:54 "Note: CHANGE-SPACE-REQUIREMENTS also mentions an :resize-frame option which is missing here. [edit]-- Gilbert Baumann 2003-03-16 14:48Z" cute 18:13:08 nyef: sure, same here :) 18:13:46 I have got replace but I want http://en.wikipedia.org/wiki/Comparison_of_programming_languages_(string_functions)#replace 18:13:50 what to do? 18:14:43 -!- trebor_dki [n=user@mail.dki.tu-darmstadt.de] has quit [Read error: 113 (No route to host)] 18:15:20 tcr: Ja. 18:15:33 tcr: when you get finished with jmckitrick .... 18:15:45 lichtblau: hacking to default-directory to never to return "" gets me a bit farther... but still no joy 18:16:04 "ignoring error: A non-existent ERRNO-VALUES syscall error has been signaled: UNKNOWN, 45." 18:16:09 manuel__ [n=manuel@pD9E6C512.dip.t-dialin.net] has joined #lisp 18:16:54 rpg: I'll send comments about the patch per mail 18:17:22 jmckitrick: Could you give an example of your data? 18:17:34 tcr: the docs patch, right? 18:18:16 tcr: I'm past that point, actually, and now I'm passing location data (start end state) to the markup function. Hold on.... 18:18:17 hm, that's ENOTSUP 18:18:24 nikodemus: if Grep succeeded at some point in the past, I take it that fork&exec happens to work occasionally? 18:18:25 -!- amnesiac__ [n=amnesiac@charanda.sandino.net] has quit [Client Quit] 18:18:32 amnesiac [n=amnesiac@p3m/member/Amnesiac] has joined #lisp 18:18:32 posix isatty 18:18:33 http://www.opengroup.org/onlinepubs/007908799/xsh/isatty.html 18:18:37 rpg: yeah, there's lots of whitespace issues 18:18:45 rpg: you introduce lots of trailing whitespace 18:18:46 yes 18:18:52 error in process filter: Wrong type argument: integer-or-marker-p, [[102 130 2] [219 253 1] [132 189 1] [200 266 1]] 18:18:52 Oh, lovely. 18:19:02 tcr: thanks. I wonder why emacs didn't catch that for me.... 18:19:03 it worked occasionally -- failing pretty randomly 18:19:23 tcr: THat's where I am now. Thought I could pass a list to this function, not sure why not yet. I just dove into this part, so it might be obvious 18:19:27 now i can't get it to work at all, but at least the errors are more consistent :) 18:19:37 jmckitrick: Uh why do you have a vector of vectors? I'd rather use vector of lists so you can use destructuring-bind on each datum 18:19:48 tcr: My edits were pretty local --- I wonder if texinfo mode, which recalculates the various nodes, etc., introduces trailing whitespace... 18:19:59 rpg: That may be the case 18:20:05 tcr: Emacs converted that when I pasted. It's actually a list 18:20:25 I have slime bindings active now, so ( becomes [ ;-) 18:20:55 tcr: I will upload a new patch. 18:21:06 jewel [n=jewel@vc-41-31-74-43.umts.vodacom.co.za] has joined #lisp 18:21:52 progress: ignoring error: # 18:23:27 the stack traces really need to go into a buffer rather than the tty. 18:23:33 (except during redisplay perhaps...) 18:24:23 in this case the trace comes from the child, so it's not too bad -- it's just useless the as the syscall doesn't appear there... 18:24:27 rpg: wait I have more comments 18:24:28 tcr: Why would an elisp function choke on a list? 18:24:45 jmckitrick: because it does not expect it? Paste the full backtrace to paste.lisp.org 18:24:47 tcr: I've tried with and without interactive 18:25:06 Actually, it might be the process filter of swank... 18:25:20 fun, it's %sys-execvp that's complaining 18:26:38 New plan: Tell MacOS users to run the TTY backend using Qt's event loop rather than iolib. :-) 18:28:01 lichtblau: Qt on os x is sort of painful to build, no? 18:28:28 TuxPurple [n=TuxPurpl@unaffiliated/tuxpurple] has joined #lisp 18:29:35 is it? I wouldn't know, because I don't have a Mac. 18:31:02 jmckitrick: Not necessarily, you get that prefix for every error that happens in the extent of a slime-eval-async 18:31:08 In general, Qt is huge and takes a while to compile, so I prefer binaries over sources if I have a choice. 18:31:24 trebor_dki [n=user@mail.dki.tu-darmstadt.de] has joined #lisp 18:31:34 jmckitrick: Really paste the backtrace, otherwise I can't help I'm afraid :-) 18:32:59 tcr: I'm trying to find one, but it's just an error in the process filter. I need to generate a bt... 18:34:44 -!- Stattrav [n=Stattrav@202.3.77.161] has quit [Remote closed the connection] 18:35:22 -!- manuel_ [n=manuel@pD9E6D8CC.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 18:35:28 mrsolo [n=mrsolo@nat/yahoo/x-rpphlbrguxtwtuov] has joined #lisp 18:35:48 Stattrav [n=Stattrav@202.3.77.161] has joined #lisp 18:35:55 jmckitrick: M-x toggle-debug-on-error 18:36:25 where one can find minion sources 18:36:41 minion: minion? 18:36:42 minion: minion is an IRC robot (who prefers the term "electronically composed.") For online help, try /msg minion help Minion is hosted at common-lisp.net and is usually connected to the #lisp IRC channel. http://www.cliki.net/minion 18:37:05 pkhuong: Must one install qt from macports? 18:38:52 -!- kiuma [n=kiuma@93-35-225-51.ip57.fastwebnet.it] has quit ["Bye bye ppl"] 18:39:35 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 18:39:37 ahh, sorry, should look at cliki first 18:39:57 i justa have a need for similar jabber bot 18:40:23 Guthur [n=Michael@host81-159-209-251.range81-159.btcentralplus.com] has joined #lisp 18:40:24 supose it's bsd/mit or something like that 18:41:27 milanj: Note that the version in the cl-irc repository may not be exactly what's running in the channel, but if you're just looking for a sample bot to work from then it should suffice. 18:43:53 -!- lukjad007 is now known as lukjad100 18:44:15 -!- lukjad100 is now known as lukjad007 18:44:38 kejsaren1 [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 18:45:07 Intensity [i=[MApGjQS@unaffiliated/intensity] has joined #lisp 18:45:11 -!- spec[away] is now known as mrSpec 18:45:26 manuel_ [n=manuel@pD9E6FF16.dip.t-dialin.net] has joined #lisp 18:45:59 rpg: sent 18:46:39 yes, example should be enough, thanks nyef 18:46:43 jmckitrick: did that help? 18:47:03 kejsaren2 [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 18:47:57 rpg: You should be able to confirm bugs (everyone can), and also change Importance (only admin people can, but you should hopefully be part of it) 18:48:09 fe[nl]ix: Is RPG a member of the asdf hackers group? 18:49:07 tcr: Thanks. Just updated now. 18:49:20 Jabberwockey [n=jens@port-93973.pppoe.wtnet.de] has joined #lisp 18:49:28 your email bits are taking a while to cross the Atlantic! 18:50:21 tcr: Yes, thanks. It seems elisp is wrapping my args in lists when I did not expect it. Or esles something is happening I don't understand completely. I'll get it, though. 18:51:03 jmckitrick: elisp does not behave differently than CL in that regard, at least not what I'd know of 18:51:11 -!- trebor_dki [n=user@mail.dki.tu-darmstadt.de] has quit [Read error: 113 (No route to host)] 18:51:38 tcr: How can I get the text from find-file-noselect without losing the buffer I'm currenly in? 18:51:57 -!- xan [n=xan@155.99.117.91.static.mundo-r.com] has quit [Read error: 60 (Operation timed out)] 18:52:35 -!- legumbre_ is now known as legumbre 18:53:25 jmckitrick: (with-current-buffer (find-file-noselect "/etc/passwd") (buffer-substring-no-properties (point-min) (point-max))) 18:53:27 -!- kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 18:53:37 tcr: I'll write another function... ah, okay. 18:53:43 I missed the WITH- part 18:55:20 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Connection timed out] 18:55:25 -!- Jafet [n=Jafet@unaffiliated/jafet] has quit [Connection timed out] 18:55:59 -!- manuel__ [n=manuel@pD9E6C512.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 18:56:11 tcr: Now we're getting somewhere! 18:56:24 jmckitrick: insert-file-contents 18:56:26 is better 18:58:11 -!- Taggnostr [n=x@dyn57-487.yok.fi] has quit [Connection timed out] 18:58:40 tcr: ok, that's good. Question: will I need a separate face for each of the states? THere is no way to consolidate them? 18:58:50 I've not done an emacs face before. 18:59:02 Taggnostr [n=x@dyn57-487.yok.fi] has joined #lisp 18:59:23 -!- Intensity [i=[MApGjQS@unaffiliated/intensity] has quit [Remote closed the connection] 18:59:26 what kind of consolidation? 19:00:50 Rather than an 8-line face definition for each of the half-dozen or so states. 19:01:23 -!- attila_lendvai [n=ati@catv-89-134-66-143.catv.broadband.hu] has quit [Read error: 60 (Operation timed out)] 19:01:24 8line? 19:02:09 I wouldn't bother to distinguish between light and dark background 19:02:37 ok, np 19:02:58 make sure to also create an appropriate defgroup 19:03:00 -!- kejsaren2 [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 19:04:27 -!- snowbeard [n=snowbear@cpe-065-190-191-189.nc.res.rr.com] has quit ["Client Quit"] 19:05:18 -!- ignas [n=ignas@78-60-73-85.static.zebra.lt] has quit [Connection timed out] 19:07:30 -!- Kolyan [n=nartamon@95-24-184-87.broadband.corbina.ru] has quit [] 19:09:25 rpg: Below the "Affects ... Status ..." table, there's a (+) Also affects project 19:09:39 rpg: that can be used to tracks bugs across different projects 19:09:53 rpg: launchpad is also able to track across different bug trackers (buzilla, and trac) 19:11:56 -!- manuel_ [n=manuel@pD9E6FF16.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 19:15:36 tcr: I'm at good stopping point to send you what I have. It's obviously very rough, but you'll get an idea of what I've done so far. 19:16:06 -!- dboswell [n=user@64.55.42.130] has quit [Remote closed the connection] 19:17:25 attila_lendvai [n=ati@catv-89-134-66-143.catv.broadband.hu] has joined #lisp 19:17:38 -!- authentic [n=authenti@unaffiliated/authentic] has quit ["Lost terminal"] 19:17:56 -!- legumbre [n=leo@r190-135-11-219.dialup.adsl.anteldata.net.uy] has quit [Read error: 104 (Connection reset by peer)] 19:18:02 authentic [n=authenti@85-127-38-99.dynamic.xdsl-line.inode.at] has joined #lisp 19:18:06 legumbre [n=leo@r190-135-11-219.dialup.adsl.anteldata.net.uy] has joined #lisp 19:18:51 jmckitrick: that or upload it to some place 19:19:29 https://bugs.launchpad.net/launchpad-registry/+bug/208238 -- the following is for gmane support of bug notifications. I'd like you to press on "This bug also affects me" (which can be found below the bug's title) 19:19:51 -!- jewel [n=jewel@vc-41-31-74-43.umts.vodacom.co.za] has quit [Read error: 110 (Connection timed out)] 19:21:27 jewel [n=jewel@vc-41-29-248-228.umts.vodacom.co.za] has joined #lisp 19:22:52 ravster [n=user@CPE000c41a8878f-CM00195efb5296.cpe.net.cable.rogers.com] has joined #lisp 19:22:57 Hello all 19:24:13 -!- nikodemus [n=nikodemu@cs181150041.pp.htv.fi] has quit ["This computer has gone to sleep"] 19:24:22 I am trying to re-create the example in the TK tutorial found at http://www.tkdocs.com/tutorial/firstexample.html , and I would like to know how to read a value that has been entered into the entry text area. Could someone provide me this information? I tried looking in the ltk-docs, but have been unsuccessful. Thank you. 19:25:14 aintme [n=user@1.37.221.87.dynamic.jazztel.es] has joined #lisp 19:26:49 Pepe_ [n=ppjet@mna75-4-82-225-76-148.fbx.proxad.net] has joined #lisp 19:27:23 -!- Pepe__ [n=ppjet@mna75-4-82-225-76-148.fbx.proxad.net] has quit [No route to host] 19:28:58 tcr: There are obviously slime and swank modules, plus changes to sb-cover 19:29:13 Pepe__ [n=ppjet@mna75-4-82-225-76-148.fbx.proxad.net] has joined #lisp 19:29:48 -!- Pepe_ [n=ppjet@mna75-4-82-225-76-148.fbx.proxad.net] has quit [Read error: 113 (No route to host)] 19:31:32 Lycurgus [n=Ren@cpe-72-228-150-44.buffalo.res.rr.com] has joined #lisp 19:33:01 MrRandom [n=chatzill@nrw216.fn.pl] has joined #lisp 19:33:05 http://www.wolframalpha.com/input/?i=IBM+Apple+MICROSOFT+GOOGLE I would have guessed it was like that but a little jarring to see it. 19:33:34 how is that related to lisp? 19:34:01 SandGorgon [n=OmNomNom@122.162.51.84] has joined #lisp 19:37:48 it doesn't but there was silence for 11 minutes. Carry on. 19:38:05 *isn't 19:38:20 -!- daniel_ is now known as daniel 19:38:23 silence doesn't mean you can send whatever you want 19:38:29 stassats`: Are you on slime head? 19:38:37 tcr: i think i am 19:38:57 -!- jewel [n=jewel@vc-41-29-248-228.umts.vodacom.co.za] has quit [Read error: 60 (Operation timed out)] 19:38:58 yes 19:39:03 -!- Pepe__ is now known as Pepe_ 19:39:44 Good, I need people testing my latest changes :-) 19:40:17 no that's right, there's something else that covers that 19:40:50 -!- Lycurgus [n=Ren@cpe-72-228-150-44.buffalo.res.rr.com] has left #lisp 19:41:33 -!- attila_lendvai [n=ati@catv-89-134-66-143.catv.broadband.hu] has quit [Read error: 110 (Connection timed out)] 19:42:32 tcr: Where would you like it posted? 19:45:41 -!- Edward__ [i=Ed@AAubervilliers-154-1-60-144.w90-3.abo.wanadoo.fr] has quit [Read error: 54 (Connection reset by peer)] 19:46:53 tcr: Is it possible to record the relationship between two bugs? 19:48:32 [stoop] [n=stoop@c-68-34-110-14.hsd1.dc.comcast.net] has joined #lisp 19:49:13 attila_lendvai [n=ati@catv-89-134-66-143.catv.broadband.hu] has joined #lisp 19:49:42 snearch_ [n=olaf@e179142058.adsl.alicedsl.de] has joined #lisp 19:49:51 jewel [n=jewel@vc-41-28-38-171.umts.vodacom.co.za] has joined #lisp 19:51:54 ignas [n=ignas@ctv-79-132-160-221.vinita.lt] has joined #lisp 19:53:16 -!- Nshag [n=shag@lns-bzn-37-82-253-13-222.adsl.proxad.net] has quit [Read error: 110 (Connection timed out)] 19:53:21 jmckitrick: Either put it up somewhere, or send it to me 19:53:39 Nshag [n=shag@lns-bzn-22-82-249-125-68.adsl.proxad.net] has joined #lisp 19:53:41 rpg: I don't know 19:53:54 Is there any chance that a lisper would /not/ know what a fasl file is? E.g., if you were to learn lisp entirely on some implementation that did not use this term? 19:54:18 rpg: a common lisper, no. 19:54:21 I am not sufficiently familiar with, e.g., clisp, ecl,lispworks... 19:55:05 I was just wondering if "where does ASDF put its fasl files" was guaranteed to be clear... 19:55:38 rpg: Practical Common Lisp explains the term 19:57:00 -!- reprore [n=reprore@ntkngw356150.kngw.nt.ftth.ppp.infoweb.ne.jp] has quit [Read error: 54 (Connection reset by peer)] 19:57:48 -!- attila_lendvai [n=ati@catv-89-134-66-143.catv.broadband.hu] has quit ["..."] 19:57:55 jbjohns [n=user@52-45.3-213.fix.bluewin.ch] has joined #lisp 19:57:56 -!- stoop [n=stoop@unaffiliated/stoop] has quit [Read error: 110 (Connection timed out)] 19:58:11 -!- kwinz3 [n=kwinz@85.124.207.62] has quit [Read error: 110 (Connection timed out)] 19:58:17 l 19:58:22 thanks. 19:58:25 hi 20:00:25 rus [n=rusa@124-171-39-234.dyn.iinet.net.au] has joined #lisp 20:00:57 http://paste.lisp.org/display/92436 20:01:27 In that code there, I use labels to introduce functions that users of the macro can use, but one of them didn't work as I expected 20:01:37 attila_lendvai [n=ati@catv-89-134-66-143.catv.broadband.hu] has joined #lisp 20:01:53 -!- Xantoz [n=hejhej@c-8cb7e253.01-157-73746f30.cust.bredbandsbolaget.se] has quit [Read error: 104 (Connection reset by peer)] 20:01:59 I have another package called #:net-impl that has a function called #:accept, which is imported into #:net 20:02:05 kwinz3 [n=kwinz@85.124.207.62] has joined #lisp 20:03:14 but when I try to use #:accept-stream in client code it fails because it says #:accept excects 0 arguments. But I explicitly said that I meant #:net-impl:accept, why is it picking the lables one when the package is explicitly specified? 20:03:27 nikodemus [n=nikodemu@cs181201111.pp.htv.fi] has joined #lisp 20:03:30 -!- rus [n=rusa@124-171-39-234.dyn.iinet.net.au] has quit [Client Quit] 20:04:41 jbjohns: net-impl:accept and accept are the same symbol 20:04:58 inside the labels inside the net package? (using sbcl btw) 20:04:59 jbjohns: if you're importing it, they are eq. 20:05:08 Xach: no, one's ACCEPT, the other ACCEPTJ ;) 20:05:19 *Xach* crys 20:05:31 yea, I did that in my code to verify that it works and forgot to remove it for the paste 20:05:57 Xach: ok, that makes sense, so labels is overwriting the accept from net-impl? 20:06:18 lexically shadowing. 20:06:18 jbjohns: i think you would call that "shadowing". 20:06:29 same as (let ((x ...)) ... (let ((x ...)) ...)) 20:06:52 but once accept is shadowed I have no way to refer to the shadowed one? 20:07:05 jbjohns: use flet. 20:07:10 er, the base one, don't know the terminology, but the original 20:07:32 I thought about that, but accept-stream is defined in terms of accept, are later flets able to see previous ones? 20:07:44 -!- MrRandom [n=chatzill@nrw216.fn.pl] has quit ["ChatZilla 0.9.86 [Firefox 3.5.6/20091201220228]"] 20:09:22 pr [n=pr@p579CAF0C.dip.t-dialin.net] has joined #lisp 20:09:46 jbjohns: (symbol-function 'accept) will access the global function slot 20:10:07 Ok, just verified it, later flet bindings can't see earlier ones so I would have to nest flets. I guess it doesn't mater since it's inside a macro 20:10:08 (funcall 'accept ...) then. 20:10:28 schoppenhauer [n=christop@unaffiliated/schoppenhauer] has joined #lisp 20:10:30 Xantoz [n=hejhej@c-8cb7e253.01-157-73746f30.cust.bredbandsbolaget.se] has joined #lisp 20:10:30 -!- Xantoz [n=hejhej@c-8cb7e253.01-157-73746f30.cust.bredbandsbolaget.se] has quit [Remote closed the connection] 20:10:33 Xantoz [n=hejhej@c-8cb7e253.01-157-73746f30.cust.bredbandsbolaget.se] has joined #lisp 20:11:20 jbjohns: I think it's a bad idea what you're doing there 20:11:44 For example, Slime will display (accept socket-handler) in the echo area even when you're in the scope of with-bound-service 20:12:46 tcr: true. But there is no way for slime to properly show the info for that function no matter what I name it 20:13:30 unless I do some trick of registering it with slime, I guess call-next-method must work like this? 20:14:18 You can register it indeed. It's also able to take up local function definitions of direct FLET/LABELS/MACROLET but not those comming implicitly by macroexpansion 20:14:37 pkhuong: that worked, thanks. Thanks Xach too 20:14:48 jbjohns: but tcr is right. 20:15:07 pkhuong: why don't let the user name the socket-handle? 20:15:10 sorry 20:15:18 You really shouldn't change the interface of a function randomly. 20:15:22 jbjohns: why don't you let the user name the socket handle? 20:15:33 jbjohns: then he can just use the global definition of accept 20:15:41 tcr: yea, that's what I mean, these are local functions introduce via macro expansion so there is no way slime can know what they will look like 20:16:07 the user can name the socket handle, that is the socket-handle key word. If not specified it just uses a gensym 20:16:26 Sure it can know :-) the new arglist would in fact make that possible, in principle. I don't think the efforts are worth it though 20:16:39 Overriding the definition with a specialised version of the same is fine, but an incompatible function with the same name isn't very useful. 20:16:41 jbjohns: I'd prefer it to be a required parameter 20:16:51 but having the user actually specify the port is redundant, inside a with-bound-service we would only be talking about the bound service 20:17:13 lichtblau: i've tracked down the execvp failure. you're going to enjoy this... 20:17:17 tcr: ok, I meant "it can't automatically know just from the code" :) 20:17:17 jbjohns: change the name. 20:17:19 jbjohns: then call the function accept-service, or something 20:17:43 varjag [n=eugene@103.80-202-117.nextgentel.com] has joined #lisp 20:18:44 I appreciate your points. I'll give it some thought. But for me personally it makes sense because it's like the normal accept has to have a port because there is no contextual one to use. In this case the "with-bound-service" macro sets the context of the port we're talking about 20:19:08 so the socket is optional... kind of. :) 20:19:56 jbjohns: and what if I want to do something with another service in the body of with-bound-service? 20:20:19 That's an aweful lot of implementation details that a reader of that code must know 20:20:25 lichtblau: in short, execvp fails if the mach exception handling thread is running. if it isn't running, we have to be _extremely_ careful about the code we run or it's lose() time 20:20:45 *nikodemus* goes write some c 20:20:46 pkhuong: fair point 20:21:17 pkhuong: Hey, I'm thinking to do a history-rewrite and possibly a forward-port on the stupid-lowtag-tricks branch. 20:21:36 nyef: sure, I'll rebase everything. 20:21:43 let me see if I have anything fresh 20:21:47 tcr: if the code, overall, behaves that way then it becomes expected 20:22:14 tcr: but I'll consider it. I'll have to give this some thought. Sucks with naming though, as always 20:22:29 -!- jmckitrick [n=user@adsl-176-183-44.asm.bellsouth.net] has quit ["ERC Version 5.2 (IRC client for Emacs)"] 20:23:14 accept is precisely what I mean with the global 'accept function, but inside with-bound-port that is also the precise name, just means something different in that context. But I'll have to think about nested with-bound-port calls 20:23:41 -!- SandGorgon [n=OmNomNom@122.162.51.84] has quit [Read error: 110 (Connection timed out)] 20:23:45 nyef: you have the fix for alloc_code_object? 20:24:06 actually.... :) If I let it take an optional socket object then I think that fixes that issue 20:24:23 carlocci [n=nes@93.37.193.53] has joined #lisp 20:24:47 astalla [n=astalla@93-36-229-181.ip62.fastwebnet.it] has joined #lisp 20:24:54 pkhuong: make_fixnum to explicit WORD_SHIFT ? 20:25:04 For the number of boxed words? 20:25:12 right. 20:25:23 Then yes, I have it. 20:25:50 good then. 20:25:51 nyef: oh how is the fixnum bit salvage going 20:26:01 it was fixnum wasn't it? 20:26:25 Guthur: Not well. It can get to a REPL, but it then tends to chew up all my CPU time on most inputs. 20:26:37 Xantoz_ [n=hejhej@c-8cb7e253.01-157-73746f30.cust.bredbandsbolaget.se] has joined #lisp 20:27:33 ya that doesn't sound good, hopefully there is solution there somewhere 20:27:58 -!- Xantoz [n=hejhej@c-8cb7e253.01-157-73746f30.cust.bredbandsbolaget.se] has quit [Read error: 110 (Connection timed out)] 20:28:35 mattrepl [n=mattrepl@pool-71-163-162-204.washdc.fios.verizon.net] has joined #lisp 20:28:58 thanks everyone. Back to hacking 20:29:09 -!- JonSmith [n=jon@c-71-233-58-7.hsd1.ma.comcast.net] has left #lisp 20:29:12 -!- jbjohns [n=user@52-45.3-213.fix.bluewin.ch] has quit ["lots more code to write"] 20:29:52 great. Nice crash of sleping system just as I connected power :/ 20:30:57 -!- robewald [n=robert@105.81-167-153.customer.lyse.net] has quit [Read error: 110 (Connection timed out)] 20:32:07 -!- jewel [n=jewel@vc-41-28-38-171.umts.vodacom.co.za] has quit [Read error: 110 (Connection timed out)] 20:32:22 jewel [n=jewel@vc-41-30-66-106.umts.vodacom.co.za] has joined #lisp 20:33:38 francogrex [n=user@33.142-64-87.adsl-dyn.isp.belgacom.be] has joined #lisp 20:35:30 hey all 20:35:35 -!- ravster [n=user@CPE000c41a8878f-CM00195efb5296.cpe.net.cable.rogers.com] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 20:37:16 are function parameters considered local bindings 20:37:34 yes, but what's the context? 20:37:49 if they're not special. 20:37:58 bordeaux-threads 20:38:33 special will still be considered special then 20:39:03 bordeaux-threads might not be the context answer you were looking for 20:39:15 http://trac.common-lisp.net/bordeaux-threads/wiki/ApiDocumentation 20:39:27 Xantoz [n=hejhej@c-8cb7e253.01-157-73746f30.cust.bredbandsbolaget.se] has joined #lisp 20:39:29 specifically the third bullet point in make-thread 20:40:12 -!- Xantoz_ [n=hejhej@c-8cb7e253.01-157-73746f30.cust.bredbandsbolaget.se] has quit [Read error: 104 (Connection reset by peer)] 20:41:06 -!- aintme [n=user@1.37.221.87.dynamic.jazztel.es] has quit ["/* */"] 20:41:16 notice that the paragraph preceding the items list says it's about dynamic variables 20:43:45 the variable in question probably will be special, but it's passed to the method, 20:44:29 so does point 3 apply then 20:44:38 you don't pass variables around. 20:45:48 well the parameter is binding to it, correct? 20:48:43 I'm stagnating 20:49:00 I need new challenges 20:49:18 work on threads on sbcl+win 20:49:28 francogrex: I wondered what the odor was. 20:49:47 xach: what odor? 20:49:51 -!- dv_ [n=dv@83-64-248-68.inzersdorf.xdsl-line.inode.at] has quit ["Verlassend"] 20:50:02 stagnant francogrex 20:50:20 odor of failure 20:54:00 -!- cmm [n=cmm@bzq-79-183-44-124.red.bezeqint.net] has quit [Remote closed the connection] 20:54:11 cmm [n=cmm@bzq-79-183-44-124.red.bezeqint.net] has joined #lisp 20:55:10 redline6561 [n=redline@adsl-074-232-249-027.sip.asm.bellsouth.net] has joined #lisp 20:55:45 i don't suppose there is a way to get the variables global binding from the parameter 20:56:02 besides just passing the symbol 20:56:28 I've mostly been tuning my firewall settings. iptables, psad and fwsnort.. francogrex: when uninspired to the 'housecleaning' tasks you never seem to get around to. 20:56:44 -!- akamaus [n=maus@94.231.112.177] has quit [Read error: 110 (Connection timed out)] 20:58:29 shell 20:59:41 -!- freiksenet [n=freiksen@hoasnet-fe29dd00-202.dhcp.inet.fi] has quit [Remote closed the connection] 20:59:52 Guthur: even the symbol may not suffice 21:00:07 dralston [n=dralston@S010600212986cca8.va.shawcable.net] has joined #lisp 21:01:03 Guthur: to reliably access the global value you have to use something like sb-ext:symbol-global-value 21:01:21 maybe *default-special-bindings*? 21:01:30 actually i need to read it more thoroughly 21:02:10 perhaps nicklevine has written a chapter about threads already? 21:03:04 is he an author? 21:03:16 for none less than oreily 21:03:51 nikodemus: an alternative would be to say #+(and sbcl macos) (run-program) 21:04:00 jgracin [n=jgracin@vipnet213-86.mobile.carnet.hr] has joined #lisp 21:04:37 oh a lisps and threads paper 21:04:39 I was trying to avoid that, because I was hoping that it would be reasonably easy to write a portable "iolib-run-program" library. 21:05:08 (That a Windows version would have to be different in that library seemed obvious, but I didn't expect MacOS issues.) 21:05:14 -!- tcr [n=tcr@host146.natpool.mwn.de] has quit ["Leaving."] 21:06:26 bipt [i=bpt@cpe-173-095-174-230.nc.res.rr.com] has joined #lisp 21:07:31 -!- jewel [n=jewel@vc-41-30-66-106.umts.vodacom.co.za] has quit [Read error: 110 (Connection timed out)] 21:07:36 timor [n=martin@port-87-234-97-138.dynamic.qsc.de] has joined #lisp 21:08:36 lichtblau: i think can do that, but fork-and-exec has to be in c 21:09:02 "i think you can do that", even 21:09:14 actually, i have it almost done, i hope 21:09:17 indeed 21:09:26 nikodemus: you have ? 21:09:28 cool 21:09:38 oh my friend closure could come to the rescue, locking might be easier that way as well 21:11:33 -!- snearch_ [n=olaf@e179142058.adsl.alicedsl.de] has quit ["Ex-Chat"] 21:12:33 fe[nl]ix: well, it's just a 1:1 translation of hemlock's fork-and-exec pretty much 21:13:24 Soulman [n=kvirc@154.80-202-254.nextgentel.com] has joined #lisp 21:13:35 jsoft [n=user@unaffiliated/jsoft] has joined #lisp 21:15:00 demmeln [n=Adium@dslb-094-216-206-061.pools.arcor-ip.net] has joined #lisp 21:15:08 -!- demmeln [n=Adium@dslb-094-216-206-061.pools.arcor-ip.net] has left #lisp 21:17:43 nyef pasted "Any objections to this getting checked in?" at http://paste.lisp.org/display/92443 21:18:48 nyef: wow. ugly. but probably pretty helpful 21:19:01 Hrm... Maybe it should %primitive PRINT about the failed aver as well. 21:19:24 -!- jgracin [n=jgracin@vipnet213-86.mobile.carnet.hr] has quit ["Leaving"] 21:19:38 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 21:19:50 -!- skeptomai is now known as skeptomai|away 21:19:53 -!- merl15 [n=merl@80-121-17-208.adsl.highway.telekom.at] has quit [Read error: 110 (Connection timed out)] 21:20:13 Basically, the problem is that if the package system isn't set up yet then FIND-PACKAGE will lose, so you need to do something else. And when I just added the boundp test it turned out that the standard I/O wasn't set up either. 21:20:42 merl15 [n=merl@80-121-0-189.adsl.highway.telekom.at] has joined #lisp 21:20:48 This -is- a band-aid, though, and I have no idea if there's a window between package initialization and stdio setup that can also cause lossage... 21:21:03 Jasko [n=tjasko@174.59.195.12] has joined #lisp 21:21:45 can't you test that by sticking in an (aver nil) in !cold-init 21:21:48 -!- Jasko2 [n=tjasko@c-174-59-195-12.hsd1.pa.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 21:21:52 I suppose I could. 21:21:59 or straight after whatever it is that sets *package-names* 21:22:06 And probably will, now that it's been mentioned. 21:22:21 baddog [n=liam@unaffiliated/baddog144] has joined #lisp 21:23:13 *prxq* rofls on the 'gargle blaster' 21:23:39 is it necessary the an eventual rewrite of SBCL's I/O system not use CLOS at all ? 21:23:43 -!- [stoop] is now known as stoop 21:23:46 nyef: would (the nil expr) work for the integer cases as well? 21:24:43 nikodemus: It might. Are you really likely to (aver 3) in the code, though? 21:24:43 francogr` [n=user@140.112-65-87.adsl-dyn.isp.belgacom.be] has joined #lisp 21:24:56 fe[nl]ix: that depends whether it happens before or after pcl is built as part of the regular build rather than afterwards 21:25:09 -!- francogr` [n=user@140.112-65-87.adsl-dyn.isp.belgacom.be] has quit [Client Quit] 21:25:42 I think I'd rather the fundametal I/O system not be CLOS-based, as then it has a chance of working even if PCL gets hosed. 21:26:49 does that happen often ? 21:27:04 I don't know. 21:27:12 luis: here? 21:27:31 I view PCL as a large mass of code that I don't understand, and thus am leery of having as a fundamental system dependency. 21:27:39 we need IO to compile PCL... 21:27:54 pkhuong: how so ? 21:28:08 pkhuong: Would that still be a problem if we managed to wedge PCL into the cold-load? 21:28:24 nyef: if (: 21:28:41 Yes, big if. 21:29:23 And directly opposed to the notion of having -less- stuff in the cold-load, which, as I commented earlier today, was apparently considered desireable at some point. 21:29:48 nyef: I think the comments are contradictory. Personally, I think more in the cold-load is aesthetically preferable if potentially harder to debug 21:30:13 in particular, I have cursed several times at not having trace until after CLOS, which seems mad 21:30:33 o wedge PCL into the cold-load? 21:30:33 [16:28] < pkhuong> nyef: if (: 21:30:37 sorry 21:33:46 Hey, what priority should be set for the interactive-stream-p thing on windows? 21:33:51 http://pleac.sourceforge.net/pleac_commonlisp/strings.html  examples here use a (split ...) function. Where do I get that from? 21:34:12 cl-pcre? 21:34:30 *nyef* leaves at undecided. 21:35:16 -!- Ogedei [n=user@e178206094.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)] 21:35:25 clhs remove 21:35:26 http://www.lispworks.com/reference/HyperSpec/Body/f_rm_rm.htm 21:35:53 Krystof: more stuff in cold load? why? 21:36:03 s/p/pp/ 21:36:32 *froydnj* was considering trying to move a lot of the octets/external-format stuff *out* of cold load 21:38:15 because I hate debugging stuff when other stuff isn't working. I have a particular hatred of debugging stuff without a working print 21:39:06 I'm ok with stuff out of cold-load if it's also not in sbcl.core 21:39:25 -!- francogrex [n=user@33.142-64-87.adsl-dyn.isp.belgacom.be] has quit [Read error: 110 (Connection timed out)] 21:44:38 slash_ [n=Unknown@whgeh0250.cip.uni-regensburg.de] has joined #lisp 21:48:14 clhs define-setf-expander 21:48:15 http://www.lispworks.com/reference/HyperSpec/Body/m_defi_3.htm 21:48:48 HET2 [n=diman@cpc1-cdif12-2-0-cust125.5-1.cable.virginmedia.com] has joined #lisp 21:50:40 nikodemus pasted "lichteblau & fe[nl]ix: fork_and_exec" at http://paste.lisp.org/display/92445 21:55:29 -!- redline6561 [n=redline@adsl-074-232-249-027.sip.asm.bellsouth.net] has quit ["Leaving."] 21:56:24 Edward__ [i=Ed@AAubervilliers-154-1-12-148.w86-212.abo.wanadoo.fr] has joined #lisp 21:58:06 -!- nikodemus [n=nikodemu@cs181201111.pp.htv.fi] has quit ["This computer has gone to sleep"] 21:58:13 rus [n=rusa@124-171-39-234.dyn.iinet.net.au] has joined #lisp 21:58:15 rajesh [n=rajesh@nylug/member/rajesh] has joined #lisp 21:59:19 demmeln [n=Adium@dslb-094-216-206-061.pools.arcor-ip.net] has joined #lisp 22:03:58 -!- Edward__ [i=Ed@AAubervilliers-154-1-12-148.w86-212.abo.wanadoo.fr] has quit [Read error: 60 (Operation timed out)] 22:04:16 Edward__ [i=Ed@AAubervilliers-154-1-69-59.w81-249.abo.wanadoo.fr] has joined #lisp 22:04:33 -!- dlowe [n=dlowe@ita4fw1.itasoftware.com] has quit ["Leaving."] 22:04:43 ltriant [n=ltriant@lithium.mailguard.com.au] has joined #lisp 22:05:56 How do I get dpkg to be aware of a new install, instead of the old 'make install'? I know there is a command, but I forget it's name. 22:06:53 more spesifically ImageMagick-6.5.8-8 22:07:56 make sure /usr/lib or whatever appears before /usr/local/lib or whatever in your path. 22:08:21 Younder: you don't 22:08:51 rahul, ? 22:08:52 Younder: you're supposed to make a debian package out of whatever you were doing 22:08:59 or use the package that's already in debian 22:09:14 rahul, I know.. But I forget the command for doing so 22:09:25 #debian 22:09:45 www.debian.org -> user manual 22:09:57 aptitude, mostly 22:10:20 -!- TuxPurple [n=TuxPurpl@unaffiliated/tuxpurple] has quit ["bye..."] 22:12:55 -!- Sergio` [n=Sergio`@a89-152-191-112.cpe.netcabo.pt] has quit [Read error: 60 (Operation timed out)] 22:15:52 Hrm... No nikodemus? 22:16:05 Oh well. 22:16:08 kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 22:16:25 kejsaren2 [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 22:17:35 -!- b4|hraban [n=b4@a83-163-41-120.adsl.xs4all.nl] has quit ["Leaving"] 22:17:49 -!- varjag [n=eugene@103.80-202-117.nextgentel.com] has quit ["Leaving"] 22:19:56 -!- udzinari [n=user@19.221.broadband4.iol.cz] has quit [Read error: 110 (Connection timed out)] 22:23:08 -!- HET2 [n=diman@cpc1-cdif12-2-0-cust125.5-1.cable.virginmedia.com] has quit ["Leaving"] 22:23:30 is there a git mirror of sbcl's cvs somewhere? 22:24:11 lmgtfy 22:24:51 k... 22:25:05 the one at repo.or.cz looks good 22:27:14 -!- cvandusen [n=user@12.185.80.194] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 22:27:39 ... why does src/code/filesys:ensure-directories-exist ignore the return value from unix-mkdir? 22:29:31 -!- Edico [n=Edico@unaffiliated/edico] has quit ["Ex-Chat"] 22:30:17 nicklevinehome [n=chatzill@cpc1-cmbg14-2-0-cust14.5-4.cable.virginmedia.com] has joined #lisp 22:30:49 -!- kejsaren1 [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 22:30:49 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 22:31:02 tcr asked about a threads chapter -- answer is yes, written and posted 22:31:06 http://lisp-book.org/contents/ch15.html 22:32:05 -!- holycow [n=new@mail.fredcanhelp.com] has quit [Read error: 60 (Operation timed out)] 22:32:33 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 22:32:36 kejsaren1 [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 22:32:44 -!- nicklevinehome [n=chatzill@cpc1-cmbg14-2-0-cust14.5-4.cable.virginmedia.com] has quit [Client Quit] 22:33:51 -!- a-s [n=user@nat-240.ro.66.com] has quit [Remote closed the connection] 22:34:25 holycow [n=new@mail.fredcanhelp.com] has joined #lisp 22:38:25 kwertii [n=kwertii@c-67-180-202-175.hsd1.ca.comcast.net] has joined #lisp 22:41:30 -!- merl15 [n=merl@80-121-0-189.adsl.highway.telekom.at] has quit [Read error: 60 (Operation timed out)] 22:41:57 rahul, sorry, that was supposed to go to #ubuntu. (The answer is checkinstall) 22:45:12 -!- alec [n=aberryma@ita4fw1.itasoftware.com] has quit ["Leaving"] 22:45:35 nicklevinehome [n=chatzill@cpc1-cmbg14-2-0-cust14.5-4.cable.virginmedia.com] has joined #lisp 22:47:34 -!- kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 22:47:50 -!- kejsaren2 [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 22:49:06 -!- skeptomai|away is now known as skeptomai 22:49:36 -!- attila_lendvai [n=ati@catv-89-134-66-143.catv.broadband.hu] has quit [Read error: 60 (Operation timed out)] 22:49:37 ace4016 [i=ace4016@cpe-76-170-134-79.socal.res.rr.com] has joined #lisp 22:50:09 -!- astalla [n=astalla@93-36-229-181.ip62.fastwebnet.it] has quit ["Sto andando via"] 22:50:46 -!- nicklevinehome [n=chatzill@cpc1-cmbg14-2-0-cust14.5-4.cable.virginmedia.com] has quit ["ChatZilla 0.9.86 [Firefox 3.5.6/20091201220228]"] 22:56:08 -!- wlr [n=walt@c-65-96-92-150.hsd1.ma.comcast.net] has quit [Remote closed the connection] 22:56:44 anyone knows if and how you can add/remove slots at runtime? 22:56:58 Im looking through the whole MOP but don't find it 22:56:59 kejsaren_ [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 22:57:45 -!- demmeln [n=Adium@dslb-094-216-206-061.pools.arcor-ip.net] has quit ["Leaving."] 22:58:33 -!- rajesh [n=rajesh@nylug/member/rajesh] has quit ["leaving"] 22:58:33 rread_ [n=rread@c-98-234-219-222.hsd1.ca.comcast.net] has joined #lisp 22:58:47 wlr [n=walt@c-65-96-92-150.hsd1.ma.comcast.net] has joined #lisp 22:58:47 PissedNumlock: you can redefine the class at runtime. 22:59:00 PissedNumlock: (eval '(defclass ...)) for example. 22:59:15 PissedNumlock: there's a protocol for how existing instances can be updated to the new definition. 22:59:52 thanks. I actually thought there was a method like (remove-slot class slot) 23:00:09 PissedNumlock: i don't know about that. 23:00:23 I'm probably incorrect :) 23:00:25 thanks 23:01:24 davazp [n=user@9.Red-83-55-179.dynamicIP.rima-tde.net] has joined #lisp 23:03:50 -!- Jabberwockey [n=jens@port-93973.pppoe.wtnet.de] has quit [Read error: 54 (Connection reset by peer)] 23:05:02 you need to go through ensure-class-using-class, at least, to change the slots 23:05:09 the class needs to be refinalized 23:05:50 you could write a remove-slot that reflected on the slots and removed the slot in question, and reconstructed a defclass form or ensure-class call out of it 23:06:32 if you do write that, please consider contributing it to closer-mop 23:11:47 -!- kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has quit [Read error: 110 (Connection timed out)] 23:12:25 -!- dralston [n=dralston@S010600212986cca8.va.shawcable.net] has quit ["Leaving"] 23:14:31 -!- rread [n=rread@nat/sun/x-xqgpjdxmsjrkgwlh] has quit [Read error: 110 (Connection timed out)] 23:14:31 -!- rread_ is now known as rread 23:20:00 lacedaemon [n=algidus@ABordeaux-158-1-37-246.w90-55.abo.wanadoo.fr] has joined #lisp 23:20:59 Phoodus [i=foo@ip68-231-37-148.ph.ph.cox.net] has joined #lisp 23:21:22 tcr: thanks for the mention of nicklevine earlier, very helpful 23:21:31 i see he has post the chapter as well 23:22:10 you don't happen to know which implementations are affected by point 3 in the bordeaux-threads make-thread 23:22:47 i have something working in SBCL, but i'd like to test on an implementation that has the trouble with the bindings 23:24:05 -!- kwinz3 [n=kwinz@85.124.207.62] has quit [Connection timed out] 23:25:29 dreish [n=dreish@minus.dreish.org] has joined #lisp 23:26:50 kwinz3 [n=kwinz@213162066172.public.t-mobile.at] has joined #lisp 23:29:37 Ferrari_308_GTS [n=algidus@ABordeaux-158-1-34-161.w90-55.abo.wanadoo.fr] has joined #lisp 23:32:18 kenanb [n=kenanb@94.54.225.165] has joined #lisp 23:37:38 -!- mathrick [n=mathrick@83.1.168.198] has quit [Read error: 104 (Connection reset by peer)] 23:37:49 -!- mrSpec is now known as spec[away] 23:38:36 -!- spec[away] [n=Spec@unaffiliated/mrspec] has quit [] 23:39:35 mathrick [n=mathrick@83.1.168.198] has joined #lisp 23:40:15 -!- fe[nl]ix [n=algidus@ABordeaux-158-1-130-200.w90-60.abo.wanadoo.fr] has quit [Read error: 110 (Connection timed out)] 23:42:00 snowbeard [n=snowbear@cpe-065-190-191-189.nc.res.rr.com] has joined #lisp 23:44:39 -!- stassats` [n=stassats@wikipedia/stassats] has quit ["ERC Version 5.3 (IRC client for Emacs)"] 23:44:40 -!- slash_ [n=Unknown@whgeh0250.cip.uni-regensburg.de] has quit [Client Quit] 23:45:02 -!- kwinz3 [n=kwinz@213162066172.public.t-mobile.at] has quit [Read error: 60 (Operation timed out)] 23:45:32 YuleAthas [n=athas@0x50a157d6.alb2nxx15.dynamic.dsl.tele.dk] has joined #lisp 23:45:40 dlowe [n=dlowe@c-24-91-154-83.hsd1.ma.comcast.net] has joined #lisp 23:47:13 -!- lacedaemon [n=algidus@ABordeaux-158-1-37-246.w90-55.abo.wanadoo.fr] has quit [Read error: 110 (Connection timed out)] 23:50:06 kejsaren [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 23:50:13 kejsaren2 [n=kejsaren@111.25.95.91.static.ras.siw.siwnet.net] has joined #lisp 23:50:57 djanderson [n=dja@hltncable.pioneerbroadband.net] has joined #lisp 23:51:00 lacedaemon [n=algidus@ABordeaux-158-1-98-120.w86-201.abo.wanadoo.fr] has joined #lisp 23:51:18 -!- dlowe [n=dlowe@c-24-91-154-83.hsd1.ma.comcast.net] has quit ["Leaving."] 23:51:45 -!- Geralt [n=Geralt@unaffiliated/thegeralt] has quit [Remote closed the connection] 23:54:41 -!- bipt [i=bpt@cpe-173-095-174-230.nc.res.rr.com] has quit [Read error: 54 (Connection reset by peer)] 23:54:42 bipt` [i=bpt@cpe-173-095-174-230.nc.res.rr.com] has joined #lisp 23:56:24 -!- amnesiac [n=amnesiac@p3m/member/Amnesiac] has quit [] 23:56:28 -!- kenanb [n=kenanb@94.54.225.165] has quit [Read error: 60 (Operation timed out)] 23:56:41 -!- Ferrari_308_GTS [n=algidus@ABordeaux-158-1-34-161.w90-55.abo.wanadoo.fr] has quit [Read error: 60 (Operation timed out)] 23:58:46 kwinz3 [n=kwinz@213162066161.public.t-mobile.at] has joined #lisp