01:30:53 sellout [n=greg@c-24-128-50-176.hsd1.ma.comcast.net] has joined #ccl 02:12:13 zzz 02:38:45 -!- alms [n=alms@146-115-42-237.c3-0.bkl-ubr1.sbo-bkl.ma.cable.rcn.com] has quit [] 02:49:44 alms [n=alms@146-115-42-237.c3-0.bkl-ubr1.sbo-bkl.ma.cable.rcn.com] has joined #ccl 02:56:09 pem [n=pem@159.226.35.246] has joined #ccl 03:19:55 bfulgham_ [n=brent@adsl-69-234-109-59.dsl.irvnca.pacbell.net] has joined #ccl 04:00:03 -!- rme [n=rme@70.104.116.109] has quit [] 06:17:30 -!- bfulgham_ [n=brent@adsl-69-234-109-59.dsl.irvnca.pacbell.net] has quit [] 07:11:18 -!- billstclair [n=billstcl@unaffiliated/billstclair] has quit [] 08:08:36 billstclair [n=billstcl@unaffiliated/billstclair] has joined #ccl 11:47:32 -!- sellout [n=greg@c-24-128-50-176.hsd1.ma.comcast.net] has quit [] 12:38:44 sellout [n=greg@63.107.91.105] has joined #ccl 14:12:56 hypno [n=hypno@195.43.248.100] has joined #ccl 15:09:03 rme [n=rme@pool-70-104-116-109.chi.dsl-w.verizon.net] has joined #ccl 15:09:53 ouch. looks like there is some sort of loop fd-starvation going on in easygui.asd from trunk? 15:10:21 > Error: Stack overflow on temp stack. 15:10:23 heh 15:11:13 What did you do to elicit that error? 15:11:38 (require :cocoa-application) 15:12:07 I see. You said you're running trunk on Tiger? 15:12:14 yes. 15:13:53 I will start up my old Tiger laptop and take a look. 15:14:19 ; loading system definition from /Users/johnnyzackrisson/Desktop/FOO/ccl/examples/cocoa/easygui/easygui.asd into # 15:14:22 ; loading system definition from /Users/johnnyzackrisson/Desktop/FOO/ccl/examples/cocoa/easygui/easygui.asd into # 15:14:25 ; loading system definition from /Users/johnnyzackrisson/Desktop/FOO/ccl/examples/cocoa/easygui/easygui.asd into # 15:21:29 It worked for me. Do you have stuff in your init file that registers asdf systems? Maybe if you start up with ccl -n (to suppress reading the init file) and then do the (require 'cocoa-application)... 15:23:25 same problem. 15:31:58 What's the output of (lisp-implementation-version), please? 15:36:11 "Version 1.3-dev-r11866M-trunk (DarwinX8632)" 15:36:28 hypno: That's old 15:37:30 so: 15:37:38 Day-to-day development of Clozure CL takes place in an area of the Subversion repository known as the trunk. At most times, the trunk is perfectly usable, but occasionally it can be unstable or totally broken. If you wish to live on the bleeding edge, the following command will fetch a copy of the trunk for Darwin x86 (both 32- and 64-bit versions): 15:37:43 svn co http://svn.clozure.com/publicsvn/openmcl/trunk/darwinx86/ccl 15:37:54 rme: It's been a while ... how do I deal with structs in the Cocoa bridge? eg: frame.size.height = 42 15:37:57 that part in the documentation, http://ccl.clozure.com/ccl-documentation.html, is also wrong or have i missed something? 15:38:06 -!- billstclair [n=billstcl@unaffiliated/billstclair] has quit [Read error: 113 (No route to host)] 15:38:52 hypno: did you rebuild the lisp image? (ccl:rebuild-ccl :full t) 15:39:15 sellout: ah, no. 15:39:55 sellout: there are some convenience macros for NSRect, NSSize, etc. (setf (ns:ns-rect-height frame) 300), for example. 15:40:13 And then I think you need to restart CCL before you require the ide. 15:40:24 rme: Donke. 15:40:30 Danke, ever. 15:40:33 but that didnt work tho. x86-exceptions.c, a lot of thread related defines missing. i guess that has todo with tiger/leopard. 15:40:33 even. 15:40:45 Damn synergy lag. 15:42:01 sellout: otherwise you use pref as normal, e.g., (setf (pref r #>NSRect.size.height) 42.0), but you have to be careful about value types. 15:42:27 hypno: do you have Xcode installed? 15:42:37 rme: yes. 15:44:03 rme: but i guess we can ignore this issue. i have another version of CCL running just great, and my only reason to upgrade was to checkout the new IDE. but i'll wait for screenshots and see if it's something for me. :) 15:45:51 OK. We've made some effort to keep running on Tiger, especially on ppc. I don't have an x86 Tiger system at the moment, though, so I guess it's possible that something got inadvertently hosed. 15:46:03 Upgrade to Leopard already! :-) 15:46:25 heh, yeah. 15:47:14 anyways, wrt the new IDE, has there been any plans to create more of a moving platform? 15:47:37 -!- pem [n=pem@159.226.35.246] has quit [verne.freenode.net irc.freenode.net] 15:47:37 -!- gbyers [n=gbyers@c-68-35-15-143.hsd1.nm.comcast.net] has quit [verne.freenode.net irc.freenode.net] 15:47:46 in the sense you can make it independant of CL and use it as a way to deploy user interfaces? 15:47:50 pem [n=pem@159.226.35.246] has joined #ccl 15:47:50 gbyers [n=gbyers@c-68-35-15-143.hsd1.nm.comcast.net] has joined #ccl 15:49:15 that and some modern object and system introspection, possibly remote. that would be rather darn cool. 15:50:40 There's been some talk about a multi-process style development environment, where the IDE would be one Unix process, and the lisp that runs the application would be another. 15:51:08 billstclair [n=billstcl@unaffiliated/billstclair] has joined #ccl 15:52:10 So, in some sense, the IDE would be like emacs + slime.el, and it would talk to some other lisp via IPC (corresponding to the swank protocol). 15:52:34 well, that's a start. i was thinking perhaps more Smalltalkish. i think the java and smalltalk people have conceptually got it very, very right. the ide-environment having an understanding of functions, macros, packages etc and allows for modifications across some sort of link. 15:52:38 So, the IDE wouldn't necessarily be in the application under development. 15:54:00 Right, the idea would be that the IDE tools, like the inspector, listener, backtrace, etc., would all operate on the inferior/slave/tethered lisp's objects via some sort of protocol over an ipc channel. 15:54:58 alright. 15:56:23 I'm not 100% sold on it personally (I like the idea of having everything all in one address space), but the design of Cocoa makes the one-address-space thing hard. (Debugging errors in the event thread is the main problem.) 15:57:03 anRch [n=markmill@12.147.116.243] has joined #ccl 15:59:59 hmm. cocoa is basically one thread-style thing? or why is this an issue? 16:00:39 Cocoa requires nearly all GUI stuff to happen on the main thread. 16:01:35 I think there are situations where the extra robustness of a separate-process style interaction can be very useful. Working with native code is one of those cases. 16:01:45 sure, and isn't it enough to have other threads just do an atomic-push to some global list and have the main thread popping in back and do it's thing? 16:01:57 If I manage to screw up and cause a memory fault somehow in native code, at least my IDE won't go on the fritz as a result. 16:01:57 If there's an error in an event handler, the lisp wants to run a break loop. But if the event thread is sitting break loop, then it can't process events, so the IDE tools won't work. 16:02:45 chandler: that's a strong argument for the two-process style development environment, yes. 16:09:22 rme: I'll still want to be able to get a listener, etc for the IDE process itself ... I like being able to change the IDE as I work :D 16:09:46 I think there are a few obvious reasons why it works better in, say, Smalltalk: it's much less common to call out to unrestricted native code, there's no real equivalent of (safety 0), and it's usually possible to safely interrupt execution. 16:11:20 sellout: Yes, that can be cool. But try modifying the lower levels of editor while you're working on it. "D'oh, I screwed up insert-region again!" (But then again, I'm prone to screwing up early and often.) 16:12:28 All programmers are. The ones that say they aren't are lying. 16:12:39 rme: Yeah, I only tend to play with things like the inspector (which is what I needed struct stuff for ... now the window shrinks to the content. 16:12:40 ) 16:12:53 I need paredit for IRC. 16:13:19 We can start with a good CL IRC client. 16:59:42 -!- anRch [n=markmill@12.147.116.243] has quit [] 17:39:02 anRch [n=markmill@ip-64-7-14-173.bos.megapath.net] has joined #ccl 17:49:10 milanj [n=milan@93.86.114.155] has joined #ccl 18:07:30 -!- anRch [n=markmill@ip-64-7-14-173.bos.megapath.net] has quit [] 18:20:53 anRch [n=markmill@nmd.sbx07258.melroma.wayport.net] has joined #ccl 19:35:32 -!- anRch [n=markmill@nmd.sbx07258.melroma.wayport.net] has quit [] 20:44:57 -!- sellout [n=greg@63.107.91.105] has quit [] 21:21:24 sellout [n=greg@c-24-128-50-176.hsd1.ma.comcast.net] has joined #ccl 21:22:28 -!- milanj [n=milan@93.86.114.155] has quit ["Leaving"]