2011-03-14T00:00:13 and the engine just spawns food in the one part 2011-03-14T00:00:13 they must come up with some horrible hacks with game mechanics for actual retail games 2011-03-14T00:00:28 thinking it is spawning food for everyone, that can be unfair 2011-03-14T00:00:33 <_flag> McLeopold: Then spawn it in all parts? 2011-03-14T00:00:46 yes :) 2011-03-14T00:01:17 <_flag> Either I'm really confused, or you guys are over complicating things 2011-03-14T00:01:32 antimatroid: so, spawn one food for each section, pick one subsection from each section not touching, pick one free land space in each subsection 2011-03-14T00:01:59 layer appropriately for best results 2011-03-14T00:02:31 for each subsection choice, make sure each one gets an even distribution 2011-03-14T00:02:47 so there isn't one sub-section that got 5 hits and one that got 1 2011-03-14T00:03:00 keep a max diff of 1 or 2 2011-03-14T00:04:02 or, we could say don't spawn food within X of 2 different ants, but that would be expensive to compute 2011-03-14T00:12:26 *** closedbracket has joined #aichallenge 2011-03-14T00:23:50 *** amstan has joined #aichallenge 2011-03-14T00:23:50 *** ChanServ sets mode: +o amstan 2011-03-14T00:24:26 *** Prillicy has joined #aichallenge 2011-03-14T00:26:15 McLeopold: i was thinking like follows 2011-03-14T00:27:50 find the set of squares that never translate to each other at the start (this is easier than it sounds), then randomly put them into a list, then each turn go to your selected square and increment the number of food squares needing to be spawned there, if that was currently 0, add the position to the queue of squares to spawn food at 2011-03-14T00:29:18 sigh: ^^ does that make sense, and go along with your interpretation of things? 2011-03-14T00:30:08 that makes sense 2011-03-14T00:31:27 i don't think that'd be too hard to implement 2011-03-14T00:36:44 I doesn't make sense yet 2011-03-14T00:37:15 McLeopold: do you understand that for any location, translating it by rtrans,ctrans n times will get you back to the same location? 2011-03-14T00:37:32 no 2011-03-14T00:37:36 I need pictures :) 2011-03-14T00:38:05 are rtrans and ctrans variables, or a property of the map 2011-03-14T00:38:44 the map generator picks them, but you can get them from the different between row,col of any two enemy ants starting positions 2011-03-14T00:39:12 so say you have loc1 and loc2, rtrans = abs(loc1.row-loc2.row), similarly for ctrans 2011-03-14T00:40:04 then translating loc1 is just loc = ((loc1.row+rtrans)%rows, ...) 2011-03-14T00:40:14 so, on a 4x4 map, with rtrans=2 and ctrans=1, n would be 4? 2011-03-14T00:40:30 yes, but i wouldn't allow that map 2011-03-14T00:40:33 and there would be 8 sets 2011-03-14T00:40:50 you would have 2 squares to order for placing food 2011-03-14T00:41:04 where do you get 2? 2011-03-14T00:41:19 wait, maybe 4 2011-03-14T00:41:25 yeah, 4 set 2011-03-14T00:41:27 s 2011-03-14T00:41:58 you just take one of those sets, order them randomly, then place the food in each of the equivalent squares selected each turn 2011-03-14T00:42:21 or, order the sets you mean 2011-03-14T00:42:28 yeah 2011-03-14T00:42:33 okay 2011-03-14T00:42:51 my map generator enforces no two enemy ants start in the same row or column, as it looks better and makes it easier to calculate rtrans and ctrans 2011-03-14T00:43:45 still need pictures? they'd be quite messy 2011-03-14T00:44:31 what paint program do i want for ubuntu? 2011-03-14T00:46:01 no, I don't need pictures 2011-03-14T00:46:33 what types of symmetry are you doing? 2011-03-14T00:47:24 that is it 2011-03-14T00:47:29 i can't think of any other way to do symmetry 2011-03-14T00:47:56 well, for 4 players in a square, you could do mirror or radial 2011-03-14T00:48:05 or just tellelate 2011-03-14T00:48:05 yeah, but not arbitrarily for n players 2011-03-14T00:48:10 uh tesselate 2011-03-14T00:48:21 my method atm is just tesselating 2011-03-14T00:48:30 no, each player count would have it's own set of symmetry options 2011-03-14T00:48:49 anything divisible by 2 can be mirror 2011-03-14T00:48:53 true 2011-03-14T00:49:32 that would mean people can come across each other on symmetric terms 2011-03-14T00:49:44 I'm installing ubuntu on my other laptop right now 2011-03-14T00:49:54 in mine, at the start, if you meet an enemy, you both took different starting paths and are facing each other asymmetrically 2011-03-14T00:50:04 i really like it so far 2011-03-14T00:50:27 i am going to have a go at setting up that ssh key stuff with git in a bit 2011-03-14T00:50:40 i don'tunderstand what the hell they mean when they ask for a passphrase 2011-03-14T00:50:44 oh, I've done that on both win and linux 2011-03-14T00:51:00 I've just used a blank passphrase 2011-03-14T00:51:11 haha, i'll try that after 2011-03-14T00:51:15 gotta finish actually cooking food now 2011-03-14T00:51:22 You're ssh key can be used in lots of places, so it's like security your key 2011-03-14T00:51:52 plus, you can copy your generated key to multiple machines 2011-03-14T00:52:58 antimatroid: a passphrase is just a password for your key... otherwise anyone who got access to your key could use it 2011-03-14T00:53:29 or you could generate a different key at each computer, and let github know about all of them 2011-03-14T00:55:38 so i can just come up with a random piece of text? 2011-03-14T00:55:42 can it contain spaces? 2011-03-14T00:55:53 <_flag> Has the food spawning bug been fixed yet? 2011-03-14T00:55:55 yes and yes 2011-03-14T00:55:59 @antimatroid 2011-03-14T00:56:00 sigh: You have no gotten any error messages recently, so here's a random one just to let you know that we care. 2011-03-14T00:56:00 okay cool 2011-03-14T00:56:06 cheers :) 2011-03-14T00:56:34 _flag: suggest me a paint program 2011-03-14T00:56:37 make sure you remember it 2011-03-14T00:56:42 yeah i will 2011-03-14T00:56:46 <_flag> antimatroid: gimp? 2011-03-14T00:56:49 gimp 2011-03-14T00:57:15 if you were on xp, I'd say use visio 2011-03-14T00:57:31 I haven't found a good linux equivalent yet. 2011-03-14T00:57:44 _flag: hasn't been fixed in the main repo, no 2011-03-14T00:57:44 i don't mind learning something new 2011-03-14T00:58:13 McLeopold: what do you think of dia 2011-03-14T00:58:18 ? 2011-03-14T00:58:32 denver airport? 2011-03-14T00:58:53 dia is a program which is similar to visio 2011-03-14T00:58:53 <_flag> http://projects.gnome.org/dia/ I think 2011-03-14T00:59:48 looks okay 2011-03-14T00:59:55 looks "old" 2011-03-14T01:00:10 haha, yeah it does a bit 2011-03-14T01:00:14 *** greghaynes has quit IRC (Ping timeout: 250 seconds) 2011-03-14T01:01:38 *** McLeopold is now known as McLeopold_Win7 2011-03-14T01:01:38 <_flag> It uses your gnome theme (I just tried it) so it doesn't look as bad 2011-03-14T01:05:53 *** McLeopold has joined #aichallenge 2011-03-14T01:06:35 *** McLeopold is now known as McLeopold_Ubuntu 2011-03-14T01:10:28 btw, McLeopold, have you done your big merge yet? 2011-03-14T01:10:33 :( 2011-03-14T01:10:49 I just haven't had time to work on it this weekend 2011-03-14T01:10:56 ahh 2011-03-14T01:11:01 stuff happens 2011-03-14T01:11:18 but I'm really close 2011-03-14T01:11:42 and sometimes I get distracted with other things, like os installs 2011-03-14T01:11:47 hehe 2011-03-14T01:12:33 so, when I do merge, there's just a couple of changes to the engine and ants, to support getting json results and a few engine options 2011-03-14T01:12:43 but after that, I'm hoping you'll take over completely 2011-03-14T01:14:39 :P 2011-03-14T01:15:24 haha 2011-03-14T01:16:05 surely the engine is almost done? 2011-03-14T01:18:47 does anyone have any font preferences for coding? 2011-03-14T01:19:36 neg 2011-03-14T01:21:06 *** Prillicy has quit IRC (Quit: I be bangin') 2011-03-14T01:37:33 *** McLeopold_Ubuntu has left #aichallenge 2011-03-14T01:41:14 *** closedbracket has quit IRC (*.net *.split) 2011-03-14T01:41:15 *** jmpespxoreax has quit IRC (*.net *.split) 2011-03-14T01:43:29 *** McLeopold_Ubuntu has joined #aichallenge 2011-03-14T01:43:37 *** McLeopold_Ubuntu has left #aichallenge 2011-03-14T01:47:23 *** antimatroid has quit IRC (Ping timeout: 264 seconds) 2011-03-14T01:57:45 *** delt0r___ has joined #aichallenge 2011-03-14T01:59:06 *** delt0r_ has quit IRC (Ping timeout: 276 seconds) 2011-03-14T02:03:10 *** antimatroid has joined #aichallenge 2011-03-14T02:04:08 i have git fully set up now :D 2011-03-14T02:04:13 *** closedbracket has joined #aichallenge 2011-03-14T02:04:13 *** jmpespxoreax has joined #aichallenge 2011-03-14T02:11:17 http://www.proggyfonts.com/index.php?menu=download 2011-03-14T02:22:22 antimatroid: what's your editor/ide? 2011-03-14T02:22:38 code::blocks 2011-03-14T02:22:43 janzert: thx 2011-03-14T02:22:49 i don't use it for python though 2011-03-14T02:22:55 but I prefer anti-aliased 2011-03-14T02:23:18 vim 2011-03-14T02:23:53 McLeopold: yeah, it's only for c++ 2011-03-14T02:24:15 Yeah, I use vim too. 2011-03-14T02:24:19 and eclipse 2011-03-14T02:25:31 how to i get the terminal to know what python is in ubuntu? 2011-03-14T02:25:47 python should already be installed 2011-03-14T02:25:53 type python 2011-03-14T02:26:10 claims it's not defined :\ 2011-03-14T02:26:19 it does say python is installed 2011-03-14T02:26:33 paste the error 2011-03-14T02:27:05 http://pastebin.com/31HzVrUy 2011-03-14T02:27:18 >>> is a python prompt 2011-03-14T02:27:22 dude, you are in the python repl 2011-03-14T02:27:33 i.e. you're already in python 2011-03-14T02:27:34 lol :P 2011-03-14T02:28:09 works now 2011-03-14T02:28:13 haha 2011-03-14T02:28:49 OMG 2011-03-14T02:32:27 i want frontier to come online :( 2011-03-14T02:32:39 i'm so close to having access to the greek alphabet with my finger tips 2011-03-14T02:37:29 like, for irc? 2011-03-14T02:40:33 i'm hoping it'll compile 2011-03-14T02:40:42 it probably wont 2011-03-14T02:41:28 what will compile 2011-03-14T02:41:38 greek alphabet in me codez 2011-03-14T02:42:23 -_- 2011-03-14T02:42:56 *** McLeopold has joined #aichallenge 2011-03-14T02:42:58 haters gonna hate, i'd love lots more symbols for relations/operations too 2011-03-14T02:43:12 *** McLeopold_Win7 has left #aichallenge 2011-03-14T02:43:19 and god damnit, i want to overload operater precedence! 2011-03-14T02:56:30 :) 2011-03-14T03:06:19 *** closedbracket has quit IRC (Remote host closed the connection) 2011-03-14T03:09:17 *** greghaynes has joined #aichallenge 2011-03-14T03:09:49 *** amstan has quit IRC (Ping timeout: 250 seconds) 2011-03-14T03:19:42 *** McLeopold has left #aichallenge 2011-03-14T03:23:53 *** McLeopold has joined #aichallenge 2011-03-14T03:24:33 *** McLeopold has left #aichallenge 2011-03-14T03:39:35 *** boegel has joined #aichallenge 2011-03-14T03:43:33 *** mega1 has joined #aichallenge 2011-03-14T03:57:06 *** boegel has quit IRC (Quit: *poof!*) 2011-03-14T04:25:37 *** boegel has joined #aichallenge 2011-03-14T04:31:46 *** antimatroid has quit IRC (Quit: Leaving.) 2011-03-14T04:35:40 *** antimatroid has joined #aichallenge 2011-03-14T04:37:00 *** Frontier has joined #aichallenge 2011-03-14T04:45:47 *** kaemo has joined #aichallenge 2011-03-14T04:46:45 *** antimatroid has quit IRC (Quit: Leaving.) 2011-03-14T05:37:59 *** antimatroid has joined #aichallenge 2011-03-14T05:54:52 hmmm i think it'd make it interesting to do the spawning before battle and use a spawn method like battle method 1 2011-03-14T05:55:07 player with the most ants in range gets the spawn, if it's draw the food square dies 2011-03-14T05:55:42 it'd make contests over food much more interesting anyway 2011-03-14T06:17:04 *** kaemo has quit IRC (Ping timeout: 250 seconds) 2011-03-14T06:34:01 *** Naktibalda has joined #aichallenge 2011-03-14T07:03:29 *** Naktibalda has quit IRC (Quit: ChatZilla 0.9.86 [Firefox 3.6.15/20110303024726]) 2011-03-14T07:11:49 *** Naktibalda has joined #aichallenge 2011-03-14T07:18:29 *** stocha has joined #aichallenge 2011-03-14T07:18:46 hi 2011-03-14T07:26:47 *** kaemo has joined #aichallenge 2011-03-14T07:30:30 stocha: hi 2011-03-14T07:54:03 *** kaemo has quit IRC (Ping timeout: 276 seconds) 2011-03-14T07:56:47 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has quit IRC (Quit: Lost terminal) 2011-03-14T08:08:45 *** bss03 has left #aichallenge ("Konversation terminated!") 2011-03-14T08:09:22 *** stocha has quit IRC (Ping timeout: 245 seconds) 2011-03-14T08:14:14 *** bss03 has joined #aichallenge 2011-03-14T08:20:53 *** kaemo has joined #aichallenge 2011-03-14T08:27:22 aichallenge: Marco Leise epsilon * rb9aeb59 / (45 files in 5 dirs): fog of war; removed .class files from repo - http://bit.ly/i5UkBs 2011-03-14T08:32:02 *** UncleVasya has joined #aichallenge 2011-03-14T08:35:56 *** mceier has joined #aichallenge 2011-03-14T08:37:41 Fog of war can now be switched on in the visualizer: http://marco-leise.homedns.org/antgui/v4/visualizer.php?game_id=1#visualizer 2011-03-14T08:44:59 Frontier: I've been waiting for you!!! 2011-03-14T08:45:10 how do i get the greek alphabet? 2011-03-14T08:45:14 i've switched to ubuntu 2011-03-14T08:45:37 γσμ αrε rεαllγ ςrαzγ αbσμτ τhιs :) 2011-03-14T08:45:57 it was the first thing i wanted to do after getting it installed :P 2011-03-14T08:47:10 actually, I don't quite remember what i did 2011-03-14T08:47:22 D: 2011-03-14T08:47:22 i think it was some file in /usr/share 2011-03-14T08:49:32 i'm a big fan so far though 2011-03-14T08:50:46 :) /usr/share/X11/xkb ... somewhere in there 2011-03-14T08:51:23 ah there it is: /usr/share/X11/xkb/symbols 2011-03-14T08:51:34 want to paste bin it for me? :P 2011-03-14T08:52:20 find your layout file there (en) and make a copy. then edit the keys you want to change. My keyboard is german, so you wont like the mix up of y and z :p 2011-03-14T08:52:59 anyway here is my file: http://marco-leise.homedns.org/de_gr 2011-03-14T08:53:05 You can use it for reference 2011-03-14T08:53:12 yeah that might be better 2011-03-14T08:53:23 cheers 2011-03-14T08:54:28 lol, my file for that is 1000 lines long 2011-03-14T08:55:37 oh wait, err 2011-03-14T08:55:53 Have fun ! btw, you'll have to manually set the new layout, as it is unknown to Gnome. open gconf-editor and go to: /desktop/gnome/peripherals/keyboard/kbd/layouts 2011-03-14T08:55:53 Frontier: did you just set if for capslock? 2011-03-14T08:56:16 antimatroid: I set it for AltGr+ combinations 2011-03-14T08:56:27 what is AltGr? 2011-03-14T08:57:01 There is one default layout and one 'l33t' layout that changes some of the latin letters into greek 2011-03-14T08:57:15 AltGr is Ctrl+Alt+ 2011-03-14T08:57:47 Most keyboards have a special key for that. 2011-03-14T08:57:56 altwin maybe? 2011-03-14T08:58:13 no it has nothing to do with the windows key 2011-03-14T08:58:28 it's right Alt on my keyboard 2011-03-14T08:59:20 In /desktop/gnome/peripherals/keyboard/kbd/layouts edit the key 'layouts'. 2011-03-14T09:00:14 I set the list to: [de_gr, de_gr leet] <- note the tab in front of the sub-layout name leet. 2011-03-14T09:00:47 de_gr\tleet 2011-03-14T09:01:33 Τhεη γσμ ςαη sωιτςh γσμr lαγσμτ εαsιlγ ωιτh α hστκεγ σr bγ ςlιςjιηg τhε lαγσμτ ιςση ιη τhε sγsτεm τrαγ. 2011-03-14T09:01:48 *ςlιςκιηg 2011-03-14T09:04:50 ε 2011-03-14T09:05:55 Frontier: so I could really just have leet as the second option in capslock and switch it from normal when i want? 2011-03-14T09:06:47 You can set it up any way you want, yes. 2011-03-14T09:07:12 I just left the original file unchanged so it doesn't get overwritten by updates. 2011-03-14T09:08:29 My basic layout adds greek symbols to the RightAlt and RightAlt+Shift combinations, because I don't need RightAlt (or AltGr) for anything else. 2011-03-14T09:09:32 Τhε lεετ lαγσμτ ωαs jμsτ mεαητ τσ mακε γσμ sωιτςh τσ Lιημχ. 2011-03-14T09:10:23 i get an error when i set that layouts key thing 2011-03-14T09:10:47 does it work when you set my unchanged file? 2011-03-14T09:10:48 nevermind 2011-03-14T09:10:57 kk 2011-03-14T09:12:09 oh wait, i don't think i'm doing this right, hang on 2011-03-14T09:18:32 i can't add files to that folder :\ 2011-03-14T09:18:38 is that a file permissions thing? 2011-03-14T09:21:27 sure, you need to right click the folder and select 'open as administrator# 2011-03-14T09:22:28 In fact a user account should have write access to nothing more than /home/ 2011-03-14T09:22:41 and some tmp directory of course 2011-03-14T09:22:56 open as admin isn't there 8| 2011-03-14T09:23:53 then use either a terminal and the command line to get things done (look for the root-terminal in the gnome menu) or press Alt-F2 2011-03-14T09:24:02 and enter gksu nautilus 2011-03-14T09:25:31 This opens a new file explorer instance as the root user. It will also load any settings from the root account (options, layout, colors, ...) 2011-03-14T09:26:32 well... or you can log into your system as the root user from the login prompt :) 2011-03-14T09:27:09 okaz thatäs done and this is me tzping with it 2011-03-14T09:27:11 Hint: When you select log out... from the menu there is an option to switch the user without actually logging out. 2011-03-14T09:27:27 *** sigh has quit IRC (Remote host closed the connection) 2011-03-14T09:27:34 Haha, it seems to work as expected 2011-03-14T09:27:41 try right alt keys 2011-03-14T09:27:48 η€λλο 2011-03-14T09:27:50 like α or β 2011-03-14T09:27:54 α β 2011-03-14T09:28:26 τηανκσ, ι σηαλλ πλαυ αροθνδ¹ 2011-03-14T09:28:52 first things first, time to see if i can compile and run this 2011-03-14T09:32:33 : 2011-03-14T09:32:35 :( 2011-03-14T09:32:36 no love 2011-03-14T09:34:16 Frontier: "σr bγ ςlιςjιηg τhε lαγσμτ ιςση ιη τhε sγsτεm τrαγ." what layout icon? :P 2011-03-14T09:42:45 meh, that'll do for tonight 2011-03-14T09:43:17 i finally managed to make my first github repository before too :D 2011-03-14T09:47:49 *** kaemo has quit IRC (Ping timeout: 246 seconds) 2011-03-14T09:57:17 *** delt0r___ has quit IRC (Read error: Operation timed out) 2011-03-14T09:57:26 *** delt0r___ has joined #aichallenge 2011-03-14T10:15:07 *** closedbracket has joined #aichallenge 2011-03-14T10:22:05 I have a small area in the tray saying "de_gr". It showed up when I set up more than one layout. 2011-03-14T10:54:41 *** lavalamp has joined #aichallenge 2011-03-14T11:10:38 *** superflit_ has joined #aichallenge 2011-03-14T11:20:33 i just wrote my own rational struct 2011-03-14T11:20:42 it's a little rank, but will do if i ever want to use it 2011-03-14T11:27:10 *** amstan has joined #aichallenge 2011-03-14T11:27:10 *** ChanServ sets mode: +o amstan 2011-03-14T11:48:52 how's it going? 2011-03-14T11:49:00 contestbot: seen j3camero 2011-03-14T11:49:00 amstan: j3camero was last seen in #aichallenge 1 week, 5 days, 10 hours, 55 minutes, and 0 seconds ago: McLeopold: do you have a magic main-server-setup script, or did you manually do a bunch of stuff to create a test server? 2011-03-14T11:49:17 antimatroid: anything new? 2011-03-14T11:49:33 amstan: i'm on ubuntu now 2011-03-14T11:49:39 antimatroid: decent.. decent.. 2011-03-14T11:49:41 and i can make my own git repositories 2011-03-14T11:49:47 lol 2011-03-14T11:49:55 https://github.com/antimatroid/GTL 2011-03-14T11:50:06 still a bit messy and lacking me putting in much :P 2011-03-14T11:50:26 we sort of worked out how to do symmetric food spawning in something more relevant to what you were asking about 2011-03-14T11:51:43 antimatroid: you were asking about jeff, what do you need from him? 2011-03-14T11:52:15 nothing major atm i don't think 2011-03-14T11:52:31 we need to get things running for tests before we really need him again 2011-03-14T11:52:51 antimatroid: tests? i can set that up on hypertriangle 2011-03-14T11:53:18 nah, i mean once the engine gets set up with the visualiser so we can test different parameters/rules 2011-03-14T11:54:19 contestbot: topic 2011-03-14T11:54:19 amstan: Official Google AI Challenge: http://ai-contest.com/ || Channel Logs: http://contestbot.hypertriangle.com/ || Code Repo: http://github.com/aichallenge/aichallenge || Launch Checklist: https://github.com/aichallenge/aichallenge/issues/labels/launch 2011-03-14T11:54:46 antimatroid: do you know anything about mcleopold's starter packages? how complete are they? 2011-03-14T11:55:08 jmcarthur: ping 2011-03-14T11:55:38 antimatroid: also.. how's the visualizer going? 2011-03-14T11:55:45 *** boegel has quit IRC (Ping timeout: 255 seconds) 2011-03-14T11:55:50 Frontier: ^^? 2011-03-14T11:56:02 amstan: not sure sorry 2011-03-14T11:56:16 bss03: you were working on matchmaking based on trueskill, is that integrated yet? 2011-03-14T11:56:42 ok.. seems like everyone's afk 2011-03-14T11:57:10 contestbot: later tell Frontier how finished is the visualizer? both of them(canvas+ie). 2011-03-14T11:57:10 amstan: Yep. 2011-03-14T11:57:28 contestbot: later tell jmcarthur how's it going with your haskell starter pack? 2011-03-14T11:57:28 amstan: Ready to serve, my lord. 2011-03-14T11:57:37 amstan: Take a look: http://marco-leise.homedns.org/antgui/v4/visualizer.php?game_id=1#visualizer 2011-03-14T11:59:06 Frontier: worked the first time, but then i got: https://gist.github.com/869360 2011-03-14T12:00:20 :-( 2011-03-14T12:01:20 amstan: Do we have a fixed date yet? That would be helpful to know. 2011-03-14T12:01:31 Frontier: try for 2 weeks 2011-03-14T12:02:01 a figure that has been circulating around 2011-03-14T12:04:14 antimatroid: what about mcleopold's starter packs? 2011-03-14T12:06:17 aichallenge: Daniel Smith epsilon * rcb64553 / (3 files): 2011-03-14T12:06:17 aichallenge: Fixed compile bug introduced in last change. Added more convienient 2011-03-14T12:06:17 aichallenge: image debugging. Added Map.DoInRad for doing something in a circular 2011-03-14T12:06:39 aichallenge: Alexandru Stan epsilon-new-schema * r35f6ce8 / (61 files in 9 dirs): Merge branch 'epsilon' into epsilon-new-schema (+7 more commits...) - http://bit.ly/efKvtX 2011-03-14T12:06:50 amstan: not sure sorry, i think they're fairly up to date 2011-03-14T12:07:29 sigh.. i haven't closed any of these: https://github.com/aichallenge/aichallenge/issues/labels/launch 2011-03-14T12:08:10 antimatroid: have you noticed sigh's branch about sym food spawning? 2011-03-14T12:08:31 nope, how old is it? 2011-03-14T12:08:38 we only just came up with the newest method earlier 2011-03-14T12:08:48 i haven't really been happy with a method till that 2011-03-14T12:09:14 antimatroid: i have no idea, i just know there's a branch in sigh's repo: https://github.com/sigh/aichallenge/commits/sym-food 2011-03-14T12:09:42 yeah i think that's older 2011-03-14T12:10:08 the new method might actually do a pretty good job at keeping it symmetric from starting positions in a "fair" way 2011-03-14T12:10:15 ok 2011-03-14T12:10:27 antimatroid: how's the code separated right now? 2011-03-14T12:10:35 antimatroid: is there anything in manager/ that's ants related? 2011-03-14T12:10:47 i haven't touched that code 2011-03-14T12:11:01 i can tell you exactly how it should work, but i haven't actually looked at it 2011-03-14T12:11:04 aichallenge: Marco Leise epsilon * r0d3d58a / ants/www/visualizer/js/Application.js : fixed double parsing of replays - http://bit.ly/hRRMVn 2011-03-14T12:11:05 aichallenge: Marco Leise epsilon * r9e700f4 / (7 files in 2 dirs): Merge branch 'epsilon' of github.com:aichallenge/aichallenge into epsilon - http://bit.ly/hCuYYf 2011-03-14T12:11:24 well, except a few functions here and there when asked to verify them 2011-03-14T12:11:39 antimatroid: i would just like to keep the ants stuff separated from manager 2011-03-14T12:11:51 so.. no mapgens, tournament engines, food spawners in there 2011-03-14T12:12:33 antimatroid: what's manager/map_analyzer.py? 2011-03-14T12:12:51 it's to help throw out stupid asymmetric maps 2011-03-14T12:13:09 i placed that where mcleopold told me to, you'll need to talk to him about where stuff is meant to be going 2011-03-14T12:13:31 it shouldn't be much of an issue to move it i don't think 2011-03-14T12:13:45 antimatroid: see if something calls it, if not, move it 2011-03-14T12:14:02 antimatroid: also.. this is planet wars, right? https://github.com/aichallenge/aichallenge/blob/epsilon/manager/map_generator.py 2011-03-14T12:14:21 that's planet wars 2011-03-14T12:14:33 ok.. going to delete it if nothing calls it 2011-03-14T12:15:05 i'm going to just leave the analyser there for the moment 2011-03-14T12:15:12 i'm going to bed in a minute, it's after 3am 2011-03-14T12:15:26 i'll mention it to mcleopold later though and get it moved 2011-03-14T12:15:47 antimatroid: ok, cool 2011-03-14T12:15:49 amstan: I wouldn't bother doing a whole lot with the epsilon branch right now 2011-03-14T12:15:59 janzert: why not? 2011-03-14T12:16:15 Mcleopold already did a lot of major cleaning and changing but it's still all sitting in the new schema branch 2011-03-14T12:16:26 so will be cleaned up when he merges back in 2011-03-14T12:16:30 or someone does 2011-03-14T12:16:59 janzert: darn.. ok.. so.. should i just rename epsilon-new-schema? 2011-03-14T12:17:00 aichallenge: Marco Leise epsilon * r4f3cfad / (ants/bots/d/MyBot.d ants/bots/d/ants.d): made D bot conform to the starter pack rules - http://bit.ly/ev5UA1 2011-03-14T12:17:30 * janzert shrugs 2011-03-14T12:17:34 I was ready to last week :P 2011-03-14T12:17:48 well don't just rename it I suppose 2011-03-14T12:17:50 janzert: well.. why did you put it in new-schema? lol 2011-03-14T12:18:06 I haven't touched new-schema 2011-03-14T12:18:20 janzert: oh.. nvm, you said mcleopold 2011-03-14T12:18:48 janzert: ok, i'll have to talk to him then to see what's happening 2011-03-14T12:19:01 any idea when he goes online? 2011-03-14T12:19:20 usually late afternoon or evening 2011-03-14T12:19:28 he's in pacific time I believe 2011-03-14T12:19:33 oh, decent 2011-03-14T12:19:58 contestbot: later tell mcleopold wanna dev tonight? i'm in EST 2011-03-14T12:19:58 amstan: Ready to serve, my lord. 2011-03-14T12:21:23 McLeopold has 9:18 AM right now 2011-03-14T12:22:55 Frontier: if he's pst he should have 10:18, daylight savings was this weekend 2011-03-14T12:23:07 wait.. no, i can't do math 2011-03-14T12:23:09 9 is right 2011-03-14T12:23:32 amstan: Yeah, we have DST, too. 2011-03-14T12:23:53 jbroman: somehow i did 12-3=10 2011-03-14T12:24:25 amstan: Time math is hard. I always read 1530 as 2:30 PM. (For some reason, 1430 is also 2:30 in my mind.) 2011-03-14T12:24:41 lol 2011-03-14T12:34:38 hehe, I just assume Gnome can do the math if I set the right time zone in the clock applet. 2011-03-14T12:34:41 janzert: what about the simple-json stuff? you say that it's a pretty simple process to convert them? will i be able to do that myself? 2011-03-14T12:35:15 should be able to, if mcleopold hasn't already 2011-03-14T12:35:36 I'm pretty sure everywhere it is used currently the import is done as "import simplejson as json" 2011-03-14T12:35:46 in which case the conversion is "import json" 2011-03-14T12:35:50 :) 2011-03-14T12:36:23 and it might only be used in one place, the worker tournament manager 2011-03-14T12:36:27 *** Kingpin13 has joined #aichallenge 2011-03-14T12:36:36 in which case I think mcleopold did a full rewrite 2011-03-14T12:36:44 janzert: you have it in the install script too: https://github.com/aichallenge/aichallenge/blame/epsilon-new-schema/setup/server_setup.py 2011-03-14T12:37:12 ahh, not using it just installing it though right? 2011-03-14T12:37:22 it can just be dropped from the list of install packages 2011-03-14T12:38:09 but I'm mostly waiting for the grand merge to take place before doing much more on the setup scripts since a lot of things have changed 2011-03-14T12:38:45 yeah, i think we should merge, there's no point to not do it, neither branches work anyway 2011-03-14T12:38:57 yep 2011-03-14T12:39:15 and some people started to apply unrelated changes in the new-schema branch 2011-03-14T12:39:24 the only way to have those in the normal branch is to merge 2011-03-14T12:39:42 the new-schema branch has much more than the name implies yes 2011-03-14T12:39:52 that shouldn't be the case 2011-03-14T12:40:17 contestbot: later tell bss03 Please look over this: https://github.com/aichallenge/aichallenge/issues/#issue/7/comment/869773 2011-03-14T12:40:17 amstan: I think that worked... 2011-03-14T12:40:32 it's really the de facto main line of development for the last couple weeks 2011-03-14T12:40:57 janzert: i've been trying to merge starter packages and Frontier has been working on the visualizer in the epsilon branch 2011-03-14T12:41:21 ok.. oppening an issue 2011-03-14T12:41:21 yeah, then compare that too all the changes going on in new-schema 2011-03-14T12:41:30 :) 2011-03-14T12:41:49 who have i caused trouble ? :) 2011-03-14T12:42:33 ahh, manager/ec2-check.py uses simplejson as well at the momement 2011-03-14T12:42:51 conversion there should be a simple import line change as well 2011-03-14T12:44:40 Frontier: no trouble 2011-03-14T12:44:42 amstan: You once said that a Makefile can compile the Java applet and combine JavaScript files. I know little about Makefiles, but I would give it a try as it has to do with the visualizer. Now my question is, is there already a Makefile or where would I link it in? 2011-03-14T12:45:35 Frontier: feel free to make a new one; we don't have a way to connect the makefiles yet. ask me for help if you need any; i like makefiles. 2011-03-14T12:47:16 Frontier: my schedule for the day: class from 3:00 to 4:00, then driving home for about 25 min, other than that free 2011-03-14T12:47:18 *** mega1 has quit IRC (Read error: Operation timed out) 2011-03-14T12:47:21 Is there a keyword in use, like 'install' that copies the compiled files from the source tree into the destination directory? 2011-03-14T12:47:36 Frontier: sure 2011-03-14T12:47:59 janzert: feel free to comment on the merge issue: https://github.com/aichallenge/aichallenge/issues/24 2011-03-14T12:55:16 amstan: There is no Makefile in epsilon that installs anything. Or maybe I'm blind. 2011-03-14T12:55:33 I believe all the makefiles were removed 2011-03-14T12:55:36 Frontier: nope, they were all out of date, i deleted them, there's one makefile in the starter package folder 2011-03-14T12:56:48 amstan: thanks for merging my stuff. I still plan at looking at the worker setup script to update the go installation process. I think I may be able to switch it to using a premade package now (I saw on the go mailing list that someone is now making those) instead of building the source. 2011-03-14T12:57:14 lavalamp: decent 2011-03-14T12:58:38 *** readncode has joined #aichallenge 2011-03-14T12:58:57 awww, my awesome makefile that made gifs is gone: https://github.com/aichallenge/aichallenge/commit/56f606b750854e560c8d3879ed901fb4123a7c06 2011-03-14T12:59:18 i made that for jeff's simpleants prototype 2011-03-14T13:00:06 *** closedbracket has quit IRC (Ping timeout: 255 seconds) 2011-03-14T13:01:26 oh man.. as far as i can see, only the first few commits in new-schema are actually about the new schema 2011-03-14T13:01:52 but it looks pretty mergeable 2011-03-14T13:03:44 darn, i hate cherry picks for downstream changes 2011-03-14T13:03:45 https://github.com/aerique/aichallenge/commit/594e977897b0ef03dee06a1992fb44910c40c27a 2011-03-14T13:03:54 janzert: can you help me locate where he cherry picked that from? 2011-03-14T13:05:34 Frontier: darn.. correction on my schedule, my class is from 2-3 2011-03-14T13:05:45 https://github.com/aerique/aichallenge/commit/594e977897b0ef03dee06a1992fb44910c40c27a 2011-03-14T13:06:04 janzert: nope, that's the cherry-pick, where's the original 2011-03-14T13:06:07 heh oops 2011-03-14T13:06:18 just noticed 2011-03-14T13:06:29 janzert: i'm just interested in the branch it's from 2011-03-14T13:06:45 epsilon 2011-03-14T13:06:46 https://github.com/aichallenge/aichallenge/commit/5f867e18eb58f9cb1865255572ef1c4651619d7f 2011-03-14T13:07:01 gah 2011-03-14T13:07:13 https://github.com/aichallenge/aichallenge/commit/dc10af08be58e150b4f6df0e51449ec121d26ff7 2011-03-14T13:07:26 there 2011-03-14T13:07:38 janzert: darn, why did he cherry-pick instead of pull? 2011-03-14T13:08:03 no idea 2011-03-14T13:08:26 maybe it doesn't matter actually 2011-03-14T13:08:58 amstan: I'll wait with the Makefile until you created some of them. When I merge the JavaScript files I have to replace either the main js file or visualizer.php. Without an output or destination directory that's no good :p 2011-03-14T13:09:35 Frontier: well.. i have some time, i can work on this 2011-03-14T13:09:52 why not just make a script that takes the output location as an argument? 2011-03-14T13:10:29 Frontier: what do you mean by replace at merge? why? 2011-03-14T13:10:50 *** bhasker has joined #aichallenge 2011-03-14T13:11:17 Frontier: also, is there a way to not include jquery in the repo? this makes javascript be the most used language: https://github.com/aichallenge/aichallenge/graphs/languages 2011-03-14T13:12:02 Frontier: we could link it from google's servers: http://code.google.com/apis/libraries/devguide.html#jquery 2011-03-14T13:12:03 either i create a 'visualizer-min.js' and have to update the reference in visualizer.php or I replace the main js file 'visualizer.js' with the combined version 2011-03-14T13:12:20 *** readncode has quit IRC (Remote host closed the connection) 2011-03-14T13:12:39 amstan: I do not use jquery. Is it in the repo? 2011-03-14T13:12:56 Frontier: yes, i saw it for the visualizer somewhere 2011-03-14T13:13:06 or viewer: https://github.com/aichallenge/aichallenge/tree/epsilon/ants/viewer 2011-03-14T13:13:09 that's mcleopold then 2011-03-14T13:13:29 To answer your question in PlanetWars the file was requested from google and if not found there, loaded locally. 2011-03-14T13:13:55 So the local jquery copy was meant as a fallback 2011-03-14T13:14:18 Frontier: ic, idk if it should be in the repo though, maybe make the installer script take care of downloading it 2011-03-14T13:14:48 As I said, I don't use it so... yeah 2011-03-14T13:14:56 Frontier: anyway, so what were you saying about the -min? that you have to change the referrences when you make one? 2011-03-14T13:16:40 Frontier: how about visualizer.full.js(your original from the repo), visualizer.min.js(compiled) visualizer.js(softlink to either of them, git will have it pointed to full) 2011-03-14T13:16:43 I have a script tag in the php file saying src="visualizer/js/visualizer.js". If I store the combined JavaScript in a different file I have to overwrite the php to make the change. Otherwise I have to overwrite the js to keep the php file as is 2011-03-14T13:16:57 Frontier: and add gitignore for min and the softlink 2011-03-14T13:17:12 ok.. i g2g for noms 2011-03-14T13:17:27 kk 2011-03-14T13:22:05 *** amstan has quit IRC (Ping timeout: 248 seconds) 2011-03-14T13:24:45 @later tell amstan: For the starter bots there has to be some install command anyway to copy them into the target www directory. And there is also the Java applet which needs to be compiled. I have the sources for that right next to the JavaScript sources in the www directory (bad idea?). I'm not sure what 'www' should actually contain. php, images and css only? unminimized JavaScript? Java sources? Java .class files? 2011-03-14T13:24:45 Frontier: As you wish. 2011-03-14T13:31:02 *** McLeopold has joined #aichallenge 2011-03-14T13:31:48 amstan: you're willing to dev? 2011-03-14T13:36:35 Frontier: want to organize your visualizer? 2011-03-14T13:38:21 McLeopold: Yes, I have split the JS files and a few Java classes. 2011-03-14T13:38:57 did you push lately? 2011-03-14T13:39:16 just today. I added fog of war 2011-03-14T13:39:38 You're working on epsilon, right? 2011-03-14T13:39:58 yep 2011-03-14T13:42:23 I'm reading your readme... 2011-03-14T13:42:36 oh man ... that one is old ^^ 2011-03-14T13:42:49 did you ever use the visualize_locally.py file from the last contest? 2011-03-14T13:44:20 no, but if that uses the html visualizer through a python script that might be what people meant when they were using the 'html visualizer locally' 2011-03-14T13:45:57 that was part of it 2011-03-14T13:46:12 it would create a temp html file and write the replay output directly into the html 2011-03-14T13:46:18 I'd like to avoid that this time 2011-03-14T13:46:29 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has joined #aichallenge 2011-03-14T13:46:39 it also had a neat feature that would launch your default browser with the page 2011-03-14T13:46:57 I would like to avoid writing the replay into the html through the php script as well ^^ 2011-03-14T13:47:14 I think your paste option is nice. 2011-03-14T13:47:22 We can add a file open option as well 2011-03-14T13:48:36 That's right. I'm still dreaming of using Rhino, the Java JavaScript interpreter/compiler to get an executable .jar out of the JavaScript. 2011-03-14T13:48:46 And I'm hoping we can implement the engine to launch the browser automatically after a test game as well 2011-03-14T13:49:41 what would that get you? 2011-03-14T13:50:24 McLeopold: Imagine having the visualizer in a Java app, just the way the IE version works now. That would mean a HTML/JavaScript version, an applet and an application all from the same code base :) 2011-03-14T13:51:07 cool 2011-03-14T13:51:18 so, locally I would launch a java app? 2011-03-14T13:51:38 Yes, that's the point :) I never took the time to dive deeper into Rhino to see if that was possible. 2011-03-14T13:51:55 how does ie version work now? 2011-03-14T13:52:22 You can define 'ScriptableObjects' or something like that, that are visible to the script engine like the 'window' object in web implementations of JavaScript 2011-03-14T13:53:05 so, you created an object that acts like the canvas, but the javascript is still outside of the java applet? 2011-03-14T13:53:07 Currently the IE version uses an applet that communicates with JavaScript through LiveConnect. It really is a wrapper around the canvas api 2011-03-14T13:54:27 With Rhino I could move the JavaScript engine into the applet as well. That may be faster as well. LiveConnect works horribly slow. 2011-03-14T13:54:29 so, is the java running the javascript, or the other way around? 2011-03-14T13:54:54 The current applet is only a drawing surface 2011-03-14T13:55:53 But it integrates transparently into my drawing routines. context.fillRect(0,0,10,10) means the same for the html canvas and the applet 2011-03-14T13:56:21 is the context object the piece that is different? 2011-03-14T13:57:13 You can put it that way, yes. The canvas version and the applet version share the same interface when it comes to the rendering context, but implement it differently. 2011-03-14T13:57:45 of course I only had to write the Java implementation ^^ 2011-03-14T13:58:48 I could reuse that Java implementation if I used Rhino. In the end it just means moving the JS engine out of the browser into the applet as well. 2011-03-14T13:59:01 can you point me to the section that does the detect of the browser, and switches the objects? 2011-03-14T14:00:33 The canvas detection happens here: https://github.com/aichallenge/aichallenge/blob/epsilon/ants/www/visualizer/js/Application.js#L146 2011-03-14T14:01:19 and the objects are created here: https://github.com/aichallenge/aichallenge/blob/epsilon/ants/www/visualizer/js/Application.js#L373 2011-03-14T14:03:59 The applet takes some time to initialize and I must not access it during that time, so it has to call back after it is initialized. But this call back again can break the single-threadedness of JavaScript on some browsers. It just stops the current execution there and executes the callback. To avoid any race conditions I wrap every Java-to-JS call in a window.setTimeout() call. These timeouts work as expected in all browsers. T 2011-03-14T14:04:15 Frontier: do you have more than 1 canvas? 2011-03-14T14:04:44 I have loads of them. They are ideal offscreen buffers for stuff that never changes, like the map 2011-03-14T14:04:50 and, shouldn't you only have to worry about IE browsers? 2011-03-14T14:06:39 Sure, but I develop under Linux and had to test the applet there before I reboot into windows. 2011-03-14T14:07:18 Anyway I wanted to go sure that my code is clean and just works with all browsers. 2011-03-14T14:07:35 It's Java, it has to. 2011-03-14T14:07:40 ture 2011-03-14T14:07:41 true 2011-03-14T14:07:52 so, for organization, I would just remove the www dir 2011-03-14T14:08:15 a make file can combine the js, compile the java and move both over to the website dir 2011-03-14T14:08:36 remove the www dir? Literally everything is in there 2011-03-14T14:08:47 oh wait.. you mean the ants www dir 2011-03-14T14:08:51 yes 2011-03-14T14:09:01 1 extra layer too many 2011-03-14T14:09:23 and, hopefully, we can get rid of that php file 2011-03-14T14:09:27 what about visualizer.js. It is really a typical www candidate 2011-03-14T14:09:35 hmm 2011-03-14T14:09:48 the makefile will know where to put it 2011-03-14T14:11:21 if we concat all your files, will it still work? 2011-03-14T14:12:54 no that is too simple though. The rule would be: Take visualizer.js and replace this file by the content of all files referenced by "import('xyz')" lines, but remove the import lines and don't concatenate any duplicate imports. 2011-03-14T14:13:54 I don't have any duplicate imports, and they aren't needed at some point, so that rule could be ignored. 2011-03-14T14:14:38 I think a Python script could do that. 2011-03-14T14:14:41 so, you want to be able for it to run split and combined? 2011-03-14T14:16:22 I need the split version while I am working on the files. And for the web it would be better if they were combined and minified with http://developer.yahoo.com/yui/compressor/ 2011-03-14T14:17:07 Closure Compiler might be cool, too. 2011-03-14T14:17:47 jbroman: Isn't that a native compiler? 2011-03-14T14:18:36 Frontier: Hmm? 2011-03-14T14:18:41 Closure Compiler is JS to JS. 2011-03-14T14:18:51 JS to smaller/faster/optimized JS. 2011-03-14T14:18:56 http://code.google.com/closure/compiler/ 2011-03-14T14:19:05 ah ok, I thought that was an executable code compiler 2011-03-14T14:19:55 It will do inlining, dead code removal and other stuff I don't think YUIC does (haven't looked at it lately though). 2011-03-14T14:20:01 Frontier: I'd probably refactor it so you're not doing document.write stuff. Just create a debug html file that has a bunch of script tags. 2011-03-14T14:20:41 Then we can just concat the files at it will mostly be the same 2011-03-14T14:21:27 *** choas has joined #aichallenge 2011-03-14T14:23:36 *** Naktibalda has quit IRC (Quit: ChatZilla 0.9.86 [Firefox 3.6.15/20110303024726]) 2011-03-14T14:27:05 hmm, I debug in both the offline.html and visualizer.php. 2011-03-14T14:29:50 a js minimizer generally does much more than just concat files together 2011-03-14T14:30:09 using a real minimizer of some sort is probably a good idea 2011-03-14T14:32:41 McLeopold: I haz idea: http://www.linuxtopia.org/online_books/linux_tool_guides/the_sed_faq/sedfaq4_025.html 2011-03-14T14:32:55 runs directly from the Makefile I assume 2011-03-14T14:34:43 janzert: my thought was to concat, and then minimize 2011-03-14T14:34:53 we only want minimized for the official site 2011-03-14T14:35:06 as a local tool, we want the full thing 2011-03-14T14:35:16 right, but let something like closure do all of it 2011-03-14T14:36:08 okay, but it's not going to factor out the document.write 2011-03-14T14:36:21 if it does, then great! 2011-03-14T14:37:30 it won't change how the code works, no 2011-03-14T14:42:49 *** bhasker has quit IRC (Ping timeout: 250 seconds) 2011-03-14T14:42:59 *** bhasker has joined #aichallenge 2011-03-14T14:50:19 *** Apophis_ has joined #aichallenge 2011-03-14T14:52:20 *** contestbot has joined #aichallenge 2011-03-14T14:52:20 *** ChanServ sets mode: +o contestbot 2011-03-14T14:52:57 *** Apophis has quit IRC (Ping timeout: 240 seconds) 2011-03-14T14:57:05 *** closedbracket has joined #aichallenge 2011-03-14T15:04:45 *** lavalamp has quit IRC (*.net *.split) 2011-03-14T15:16:40 *** janzert has quit IRC (Read error: Connection reset by peer) 2011-03-14T15:18:13 *** janzert has joined #aichallenge 2011-03-14T15:28:05 *** amstan has joined #aichallenge 2011-03-14T15:28:05 *** ChanServ sets mode: +o amstan 2011-03-14T15:30:06 Frontier: ok so.. 2011-03-14T15:30:08 Frontier: still there? 2011-03-14T15:30:36 McLeopold: yay, you're here too 2011-03-14T15:31:47 * amstan is waiting for either 2011-03-14T15:32:53 *** kaemo has joined #aichallenge 2011-03-14T15:32:53 i have 20 minutes 2011-03-14T15:33:06 Frontier: what www folder are you talking about? 2011-03-14T15:33:09 ants/www? 2011-03-14T15:33:34 yeah 2011-03-14T15:33:40 McLeopold: so.. merge? 2011-03-14T15:33:43 I suggest it was 1 layer 2 much 2011-03-14T15:34:43 McLeopold: what? 2011-03-14T15:35:05 and I only really care about too many ad layers 2011-03-14T15:35:14 what ads? 2011-03-14T15:35:41 sorry, I was trying to be funny. McLeopold wants to move stuff from ants/www into ants 2011-03-14T15:36:13 how about `mv ants/www ants/visualizer` 2011-03-14T15:36:55 then you get ants/visualizer/visualizer/js/visualizer.js :) 2011-03-14T15:37:08 well, more like 'mv ants/www/visualizer ants/visualizer' 2011-03-14T15:37:24 McLeopold: +1 for that 2011-03-14T15:37:47 ok, but then the php file doesn't work any longer 2011-03-14T15:37:51 amstan: so, you really want to get rid of epsilon-new-schema? 2011-03-14T15:38:11 the ants/visualizer/*.php files are not ants specific and should be included in the /website/ folder 2011-03-14T15:38:12 McLeopold: yes 2011-03-14T15:38:14 *** bhasker has quit IRC (Quit: bhasker) 2011-03-14T15:38:35 amstan: okay, give me a few hours, I'm in the middle of some stuff, then we can merge 2011-03-14T15:39:06 Frontier: i was thinking more about the makefiles.. i think we should go with what the others are doing and getting a python install script 2011-03-14T15:39:17 *** bhasker has joined #aichallenge 2011-03-14T15:40:29 allright, I don't know Python so I can relax and let others do the work 2011-03-14T15:40:40 Frontier: yes, but you should tell us what you need to be done 2011-03-14T15:40:58 *** smellyhippy has quit IRC (Read error: Connection reset by peer) 2011-03-14T15:41:36 cool. i can make an issue for that or just explain it here. 2011-03-14T15:42:03 Frontier: makefile or .sh script could work for now, and actually be functional 2011-03-14T15:42:17 in case nobody gets to it 2011-03-14T15:42:54 *** smellyhippy has joined #aichallenge 2011-03-14T15:43:25 In the visualizer/java is the code base for the applet. It has to be compiled into classes. 2011-03-14T15:43:36 so.. javac *.java 2011-03-14T15:44:16 I used com.aicontest.... package names, but you are right 2011-03-14T15:44:36 i'm not that familiar with java packages 2011-03-14T15:45:22 javac visualizer/java/com/aicontest/visualizer/*.java may actually work 2011-03-14T15:45:48 I always use Eclipse for that, so I must check that myself 2011-03-14T15:45:49 Frontier: it should be s/aicontest/aichallenge 2011-03-14T15:46:06 wait.. 2011-03-14T15:46:15 does it have to be a fqdn? 2011-03-14T15:46:54 no, it doesn't have to be anything real, but it is common practice to use the company's web site 2011-03-14T15:47:45 and http://aichallenge.com/ would be utterly wrong ^^ 2011-03-14T15:48:12 yeah 2011-03-14T15:50:14 amstan: the command is slightly more complex. It needs the Java Web Plugin jar on the classpath. On my computer it would be: 2011-03-14T15:50:15 javac java/com/aicontest/visualizer/*.java -classpath /opt/sun-jdk-1.6.0.24/jre/lib/plugin.jar 2011-03-14T15:51:01 ok, i'll it 2011-03-14T15:51:14 on my own though, you don't have to wait for me 2011-03-14T15:51:28 Or if the system uses java-config-2 it is independent from the sdk version and can use: 2011-03-14T15:51:29 javac java/com/aicontest/visualizer/*.java -classpath /etc/java-config-2/current-system-vm/jre/lib/plugin.jar 2011-03-14T15:54:05 Then there is the visualizer/js folder. The would be a good job for a python script. Collect all lines with "import('...');" in *.js files, make a unique list out of it and write a new file from the content of the imports. 2011-03-14T15:54:13 *That 2011-03-14T15:55:01 Frontier: sure 2011-03-14T15:55:48 Closure Compiler can be used on the result to minimize the file size. 2011-03-14T15:57:16 just give closure each of the js files separately 2011-03-14T15:58:03 janzert: Why would that be better? A browser waits for each separate .js file as far as I know. 2011-03-14T15:58:11 and rewrite them so they aren't using the dynamic js loading 2011-03-14T15:58:19 it'll compile all the js files into one 2011-03-14T15:59:08 *** boegel has joined #aichallenge 2011-03-14T15:59:17 hmm, ok. if I write one js file containing all the imports (and nothing else) it could work like that. 2011-03-14T15:59:28 * janzert nods 2011-03-14T16:00:20 or just statically put all the separate script tags needed into the html when working with the unminified "development" version 2011-03-14T16:00:49 that's probably what I would do myself, but it probably doesn't really matter 2011-03-14T16:02:56 janzert: I prefer to use the visualizer.php and offline.html as is. I think the solution with one import file is quite a good compromise. 2011-03-14T16:04:07 yeah, I wouldn't use a visualizer.php for testing at all myself. just a plain test_visualizer.html 2011-03-14T16:04:14 I keep the name visualizer.js for this import file and when we install the www files we compile the other js files into a file with the same name. the web files can remain untouched 2011-03-14T16:04:43 yep, that would probably work 2011-03-14T16:05:49 It is nice to have a database with sample replays in the background. A change to the game_id parameter and I can test a different replay. :) 2011-03-14T16:07:04 ok, i'm out for tonight. cu tomorrow 2011-03-14T16:09:20 later 2011-03-14T16:11:07 *** Accoun has quit IRC () 2011-03-14T16:16:11 * amstan is just waiting for McLeopold 2011-03-14T16:20:47 *** amstan has quit IRC (Remote host closed the connection) 2011-03-14T16:21:27 *** amstan has joined #aichallenge 2011-03-14T16:21:27 *** ChanServ sets mode: +o amstan 2011-03-14T16:24:53 *** closedbracket has quit IRC (Read error: Connection reset by peer) 2011-03-14T16:25:07 *** closedbracket has joined #aichallenge 2011-03-14T16:27:54 *** readncode has joined #aichallenge 2011-03-14T16:29:50 *** closedbracket has quit IRC (Ping timeout: 255 seconds) 2011-03-14T16:30:33 *** Accoun has joined #aichallenge 2011-03-14T16:36:54 *** closedbracket has joined #aichallenge 2011-03-14T16:38:09 *** readncode has quit IRC (Read error: Operation timed out) 2011-03-14T16:41:16 *** amstan has quit IRC (Remote host closed the connection) 2011-03-14T16:41:24 *** amstan has joined #aichallenge 2011-03-14T16:41:24 *** ChanServ sets mode: +o amstan 2011-03-14T17:04:56 *** boegel has quit IRC (Quit: Leaving) 2011-03-14T17:24:19 *** UncleVasya has quit IRC () 2011-03-14T17:54:12 *** Naktibalda has joined #aichallenge 2011-03-14T17:58:04 *** delt0r_ has joined #aichallenge 2011-03-14T17:59:29 *** delt0r___ has quit IRC (Ping timeout: 260 seconds) 2011-03-14T18:09:46 *** superflit_ has quit IRC (Quit: superflit_) 2011-03-14T18:17:02 McLeopold: ping 2011-03-14T18:17:39 ping 2011-03-14T18:18:33 amstan 2011-03-14T18:18:43 McLeopold: so.. when are you available to code then? should i make other plans for today? 2011-03-14T18:19:12 not till late for you 2011-03-14T18:19:39 McLeopold: as in 10pm? 2011-03-14T18:20:44 I'll be around then 2011-03-14T18:21:32 ok 2011-03-14T18:21:34 cool 2011-03-14T18:21:46 tomorrow is better, if you want to try and work together 2011-03-14T18:21:50 ok 2011-03-14T18:21:59 tonight, I plan to merge, so that would give you time to go over my changes 2011-03-14T18:24:40 ok, decent 2011-03-14T18:40:44 *** choas has quit IRC (Quit: leaving) 2011-03-14T18:55:55 *** amstan has quit IRC (Ping timeout: 250 seconds) 2011-03-14T18:59:21 *** mceier has quit IRC (Quit: leaving) 2011-03-14T19:07:19 *** bhasker has quit IRC (Ping timeout: 240 seconds) 2011-03-14T19:09:00 *** bhasker has joined #aichallenge 2011-03-14T19:10:26 *** sigh has joined #aichallenge 2011-03-14T19:11:21 *** bhasker has quit IRC (Client Quit) 2011-03-14T19:20:20 *** closedbracket has quit IRC (Remote host closed the connection) 2011-03-14T19:32:48 *** kaemo has quit IRC (Ping timeout: 248 seconds) 2011-03-14T20:01:20 *** Meatkat has joined #aichallenge 2011-03-14T20:13:59 *** Kingpin13 has quit IRC (Quit: You're a kitty!) 2011-03-14T20:21:06 amstan: pong 2011-03-14T20:21:25 contestbot: Thanks, already there. 2011-03-14T20:21:26 bss03: User error, it's not my fault. 2011-03-14T20:36:30 *** nann has joined #aichallenge 2011-03-14T20:40:31 *** Apophis_ has quit IRC (Ping timeout: 255 seconds) 2011-03-14T21:02:03 *** amstan has joined #aichallenge 2011-03-14T21:02:03 *** ChanServ sets mode: +o amstan 2011-03-14T21:15:50 bloody c++ templates 2011-03-14T21:15:54 i want them back, but i don't 2011-03-14T21:23:33 *** Mathnerd314 has joined #aichallenge 2011-03-14T21:28:40 *** amstan has quit IRC (Ping timeout: 255 seconds) 2011-03-14T21:46:33 *** amstan has joined #aichallenge 2011-03-14T21:46:33 *** ChanServ sets mode: +o amstan 2011-03-14T21:47:48 *** amstan_ has joined #aichallenge 2011-03-14T21:47:48 *** ChanServ sets mode: +o amstan_ 2011-03-14T21:50:22 *** Frontier has quit IRC (Quit: Leaving.) 2011-03-14T21:56:43 *** amstan has quit IRC (Read error: Connection reset by peer) 2011-03-14T21:56:43 *** amstan__ has joined #aichallenge 2011-03-14T21:56:43 *** ChanServ sets mode: +o amstan__ 2011-03-14T22:10:47 *** Meatkat has quit IRC (Ping timeout: 252 seconds) 2011-03-14T22:24:07 *** Naktibalda has quit IRC (Quit: ChatZilla 0.9.86 [Firefox 3.6.15/20110303170601]) 2011-03-14T22:31:41 *** closedbracket has joined #aichallenge 2011-03-14T22:52:33 McLeopold: ping 2011-03-14T23:09:41 templates are back 2011-03-14T23:13:04 hmm? 2011-03-14T23:13:06 aww 2011-03-14T23:13:15 *** amstan_ is now known as amstan 2011-03-14T23:17:44 what's with the temaples? 2011-03-14T23:18:29 *** amstan__ has quit IRC (Ping timeout: 255 seconds) 2011-03-14T23:36:40 sigh: for my game theory library 2011-03-14T23:38:46 without: https://github.com/antimatroid/GTL/blob/master/ngame/NGame.h 2011-03-14T23:38:46 with: http://pastebin.com/asZHbsgW 2011-03-14T23:39:02 sure without has all that in the .cc, but it was nice to have it separated