2011-05-28T00:36:04 *** Frontier has quit IRC (Quit: Leaving.) 2011-05-28T00:37:43 *** smiley1983 has joined #aichallenge 2011-05-28T01:06:30 *** chris__0076 has quit IRC (Ping timeout: 260 seconds) 2011-05-28T01:19:45 *** chris__0076 has joined #aichallenge 2011-05-28T01:23:20 Frontier: I like the fact that you're showing combat lines in the replay, it makes the action much clearer. 2011-05-28T01:26:18 contestbot: later tell Frontier the combat lines in the visualizer are great. 2011-05-28T01:26:18 smiley1983: OK 2011-05-28T01:28:16 *** chris___0076 has joined #aichallenge 2011-05-28T01:31:23 *** chris__0076 has quit IRC (Ping timeout: 276 seconds) 2011-05-28T01:31:24 *** Palmik has joined #aichallenge 2011-05-28T01:40:38 *** chris__0076 has joined #aichallenge 2011-05-28T01:43:27 *** chris___0076 has quit IRC (Ping timeout: 252 seconds) 2011-05-28T02:04:26 *** chris__0076 has quit IRC (Ping timeout: 240 seconds) 2011-05-28T02:16:57 *** chris__0076 has joined #aichallenge 2011-05-28T02:17:44 "Right -- because array indexing is actually an operation in JS, as well as most other sensible languages. As far as I can tell, PHP treats it as some sort of bizarre mutant special case of variable naming." 2011-05-28T02:17:55 \o/ 2011-05-28T02:26:18 aichallenge: janzert epsilon * ra15612c / setup/worker_setup.py : copy debs from base os to chroot os to cut down on redownloaded content - http://bit.ly/izQ9E3 2011-05-28T02:26:47 *** znutar has left #aichallenge 2011-05-28T03:08:18 ls 2011-05-28T03:12:48 *** _0x47 <_0x47!~slash@p5B12F6EF.dip.t-dialin.net> has joined #aichallenge 2011-05-28T03:22:52 *** chris__0076 has quit IRC (Ping timeout: 240 seconds) 2011-05-28T03:36:01 *** chris__0076 has joined #aichallenge 2011-05-28T04:31:05 *** mcstar has joined #aichallenge 2011-05-28T04:33:22 *** smellyhippy has joined #aichallenge 2011-05-28T04:48:37 *** smiley1983 has quit IRC (Ping timeout: 252 seconds) 2011-05-28T04:50:47 *** smiley1983 has joined #aichallenge 2011-05-28T04:51:33 *** mcstar has quit IRC (Quit: WeeChat 0.3.5) 2011-05-28T04:54:06 *** Eruonen has joined #aichallenge 2011-05-28T05:00:26 *** needsch has joined #aichallenge 2011-05-28T05:11:02 *** amstan has quit IRC (Remote host closed the connection) 2011-05-28T05:11:37 *** tomh has joined #aichallenge 2011-05-28T05:22:26 *** sigh has joined #aichallenge 2011-05-28T05:53:06 *** Redgis has joined #aichallenge 2011-05-28T05:58:51 *** Naktibalda has joined #aichallenge 2011-05-28T06:01:26 *** mcstar has joined #aichallenge 2011-05-28T06:01:50 *** Kingpin13 has joined #aichallenge 2011-05-28T06:09:50 *** sigh has quit IRC (Ping timeout: 260 seconds) 2011-05-28T06:18:38 *** mceier has joined #aichallenge 2011-05-28T06:25:58 *** FireFly has joined #aichallenge 2011-05-28T06:27:08 *** Cyndre_ has joined #aichallenge 2011-05-28T06:27:21 *** Cyndre has quit IRC (Read error: Connection reset by peer) 2011-05-28T06:31:14 *** smiley1983 has quit IRC (Quit: leaving) 2011-05-28T07:03:21 *** delt0r___ has joined #aichallenge 2011-05-28T07:03:28 *** delt0r_ has quit IRC (Read error: Operation timed out) 2011-05-28T07:06:01 *** Naktibalda has quit IRC (Quit: ChatZilla 0.9.86.1 [Firefox 4.0.1/20110504060726]) 2011-05-28T07:54:39 *** orivej has quit IRC (Ping timeout: 255 seconds) 2011-05-28T08:03:51 anyone know what kind of AI systems they use in football games? 2011-05-28T08:15:12 I saw this: http://aigamedev.com/open/articles/football-ai-cheating-machine-learning/ but the document linked is 404'ing, anyone knows which one it is? 2011-05-28T08:32:28 *** Redgis has quit IRC (Quit: ... mains libres) 2011-05-28T08:46:17 as far as i know FSM 2011-05-28T08:46:27 tomh 2011-05-28T08:46:59 ok 2011-05-28T09:54:30 *** smellyhippy has quit IRC (Ping timeout: 240 seconds) 2011-05-28T09:54:35 *** GreenTea has joined #aichallenge 2011-05-28T09:54:50 Hi 2011-05-28T09:54:53 where is http://ai-contest.com/ ? 2011-05-28T09:55:29 *** arakyd has joined #aichallenge 2011-05-28T09:56:20 *** smellyhippy has joined #aichallenge 2011-05-28T10:08:39 *** Accoun has quit IRC (Ping timeout: 240 seconds) 2011-05-28T10:09:24 *** Accoun has joined #aichallenge 2011-05-28T10:12:48 *** Accoun has quit IRC (Read error: Connection reset by peer) 2011-05-28T10:13:12 *** Accoun has joined #aichallenge 2011-05-28T10:39:53 *** hjax has joined #aichallenge 2011-05-28T10:41:50 @later tell amstan ai-contest.com appears to be down 2011-05-28T10:41:50 jbroman: Aye, aye, sir 2011-05-28T10:44:42 why wont my bot work :/ 2011-05-28T10:44:50 * hjax feels like a failure 2011-05-28T10:46:36 can anyone tell me what is wrong with this? 2011-05-28T10:46:38 http://pastebin.com/rJ8uNPcK 2011-05-28T10:52:34 *** McLeopold has joined #aichallenge 2011-05-28T10:54:16 hey McLeopold 2011-05-28T10:54:19 @later tell Frontier If I could watch the broken replay, I might be able to determine the circumstances that cause the issue. 2011-05-28T10:54:19 McLeopold: Ready to serve, my lord. 2011-05-28T10:54:27 j3camero: ping 2011-05-28T10:56:55 grrr my code is broken again :/ 2011-05-28T10:57:33 hjax: did you just upload it? 2011-05-28T10:59:11 ive been trying, apparently my code cant pass the test, even if its syntactically correct 2011-05-28T10:59:27 i dont thing im passing the wrong value to a function again 2011-05-28T10:59:30 think* 2011-05-28T10:59:51 The database has 'null' for errors. Let me see what I can find out. 2011-05-28T11:00:47 cool, i wish it would be more descriptive when it says that my code compiled but failed testing 2011-05-28T11:08:30 *** smiley1983 has joined #aichallenge 2011-05-28T11:10:05 *** berak has joined #aichallenge 2011-05-28T11:10:40 Traceback (most recent call last): 2011-05-28T11:10:40 File "/home/contest/compiled/tmpUlN_v8/bot/ants.py", line 271, in run 2011-05-28T11:10:40 bot.do_turn(ants) 2011-05-28T11:10:41 File "MyBot.py", line 12, in do_turn 2011-05-28T11:10:41 for ant_loc in ants.my_ants: 2011-05-28T11:10:41 TypeError: 'instancemethod' object is not iterable 2011-05-28T11:10:46 hjax ^^ 2011-05-28T11:10:54 hjax: have you tested your bot with test_bot locally? 2011-05-28T11:10:58 *** Mathnerd314 has joined #aichallenge 2011-05-28T11:11:09 from the git? nope 2011-05-28T11:11:20 hjax: for ant_loc in ants.my_ants(): 2011-05-28T11:11:46 You can download the repo as an archive rather than through git 2011-05-28T11:12:15 see it was a stupid mistake >.< 2011-05-28T11:12:33 i forgot to add the () 2011-05-28T11:13:10 I can help you set up local testing if you have some time now 2011-05-28T11:13:56 smiley1983: yes please 2011-05-28T11:14:31 first step is to download the repo, either through git or from the "download" tab on the github page 2011-05-28T11:14:42 Which OS are you using? 2011-05-28T11:15:19 i did that earlier 2011-05-28T11:15:47 im on windows 7, normally i dual boot it with ubuntu, but not atm 2011-05-28T11:15:56 Okay, great 2011-05-28T11:16:13 If it's extracted, go into the aichallenge/ants folder 2011-05-28T11:16:33 are you on command line or GUI folders? 2011-05-28T11:17:09 i can do command line if youd prefer 2011-05-28T11:17:37 probably easiest, so you can see the errors (though I think there's a way to tell the .cmd to keep the window open?) 2011-05-28T11:18:33 i got cmd open now 2011-05-28T11:18:37 Once you've got the command line in the appropriate dir: 2011-05-28T11:18:46 Um . . . wow. http://www.youtube.com/watch?v=L4QmpyfS3fM 2011-05-28T11:19:08 test_bot.cmd "python /path/to/MyBot.py" 2011-05-28T11:19:25 except replace /path/to with the actual path 2011-05-28T11:19:45 see if that works 2011-05-28T11:20:30 lets see 2011-05-28T11:21:34 oops, it should be \ not / 2011-05-28T11:21:36 hmmm python gives a syntax error 2011-05-28T11:21:44 okay, try: 2011-05-28T11:22:16 test_bot.cmd "python submission_test\TestBot.py" 2011-05-28T11:22:42 its complaining about stderr, same error 2011-05-28T11:22:59 can you pastebin it? 2011-05-28T11:23:52 yep 2011-05-28T11:23:53 http://pastebin.com/1vuwZYm1 2011-05-28T11:24:05 just wait... try play_one_game.cmd 2011-05-28T11:24:09 just by itself 2011-05-28T11:25:02 same error 2011-05-28T11:25:16 okay... how recently did you download the source? 2011-05-28T11:25:44 an hour ago? 2011-05-28T11:26:08 hm... that's weird... 2011-05-28T11:26:16 how much effort is it to try in ubuntu? 2011-05-28T11:26:51 hjax: Python version? 2011-05-28T11:27:06 2.5 2011-05-28T11:27:36 a lot, id have to setup my partitions for a new dual boot 2011-05-28T11:27:37 I forget which version of Python first allowed using print as a function. I'm pretty sure 2.7 does, but it's possible 2.5 does not. 2011-05-28T11:27:55 hjax: which version of python? 2011-05-28T11:28:03 2.5 2011-05-28T11:28:07 oh, get 2.7.1 for windows 2011-05-28T11:28:07 I think in 2.5 you have to do: "print >>sys.stderr, whatever" instead of "print(whatever, file=sys.stderr)" 2011-05-28T11:28:38 McLeopold: Scrollback much? :P 2011-05-28T11:30:03 I only made it to the pastebin :) 2011-05-28T11:30:51 im installing 2.7 now, i normally use 2.5 anyhow, but replaced with 2.5 for the contest :/ 2011-05-28T11:31:01 i normally use 2.7* 2011-05-28T11:31:22 we need to update the info page then 2011-05-28T11:32:48 it ran but the visualizer failed 2011-05-28T11:33:00 okay, that's progress :) 2011-05-28T11:33:12 failed how? 2011-05-28T11:33:14 http://pastebin.com/ekA3xsfE 2011-05-28T11:35:02 and cmd gave me this 2011-05-28T11:35:04 http://pastebin.com/xsjYDAVn 2011-05-28T11:35:16 strange. Let me update my repo to see if something's changed, mine doesn't reference those files 2011-05-28T11:36:53 Huh. I updated, and now I get the same error 2011-05-28T11:37:03 This isn't your setup, they've broken something :) 2011-05-28T11:37:23 lol its broken on the beta site too 2011-05-28T11:37:36 my bot has been waiting to compile for 15 min 2011-05-28T11:38:06 and hasnt played a game since 30 min ago 2011-05-28T11:38:18 McLeopold: are you seeing this? I updated my repo and now the visualizer is broken :) 2011-05-28T11:38:36 ok 2011-05-28T11:38:48 Frontier isn't here though 2011-05-28T11:40:21 hjax: if you want to install some sort of Windows Git interface, I may be able to walk you through the process of cloning and then rolling back to a non-broken state 2011-05-28T11:40:46 otherwise it'll be necessary to wait until they fix it 2011-05-28T11:40:52 i can wait 2011-05-28T11:40:57 i gtg anyhow 2011-05-28T11:41:03 bbl 2011-05-28T11:41:03 or perhaps you could just make some empty images with the right names in the right locations 2011-05-28T11:41:10 Best of luck with it :) 2011-05-28T11:41:16 thats what i was going to try :P 2011-05-28T11:41:26 bye 2011-05-28T11:43:23 *** Cyndre__ has joined #aichallenge 2011-05-28T11:45:44 *** hjax has quit IRC (Ping timeout: 252 seconds) 2011-05-28T11:46:25 *** Cyndre_ has quit IRC (Ping timeout: 260 seconds) 2011-05-28T11:47:02 contestbot: later tell hjax it's worth installing through git anyway; the code is still under heavy development, so you'll be redownloading a lot of times if you get the whole archive every time, whereas updating through git is fast 2011-05-28T11:47:02 smiley1983: Ready to serve. 2011-05-28T11:50:38 *** UncleVasya has joined #aichallenge 2011-05-28T11:58:23 *** hjax has joined #aichallenge 2011-05-28T11:59:38 Hi. Does anybody writing an F# starter package? 2011-05-28T12:00:23 There's an OCaml package, I believe they're similar... 2011-05-28T12:00:36 I don't think anyone's announced starting an F# package 2011-05-28T12:00:44 Thank you. 2011-05-28T12:00:52 no worries :) 2011-05-28T12:05:45 smiley1983: you make me feel a lot better about my sleeping habbits :d 2011-05-28T12:13:20 antimatroid1: lol, I should go to sleep soon 2011-05-28T12:16:57 *** hjax has quit IRC (Quit: Page closed) 2011-05-28T12:26:17 UncleVasya: I think they're using mono for .net things 2011-05-28T12:32:12 *** sigh has joined #aichallenge 2011-05-28T12:32:42 solifugid is spelling things... 2011-05-28T12:38:34 btw, I can't delete the profile information or change my affiliation - in either case I get an invalid organization ID 2011-05-28T12:39:22 if anyone with server access feels like deleting solifugid's bio (or removing the question from the start) it would be welcome 2011-05-28T12:39:49 also, the bio is obviously now wrong about what the bot does :) 2011-05-28T12:40:39 My main point is not to get .net things, but to try some functional language. Asked about F# because it also was a way to get experience in MSVS. Now I have to choose beetwen Scala, Haskell and OCalm. 2011-05-28T12:41:45 *** sigh has quit IRC (Ping timeout: 255 seconds) 2011-05-28T12:42:23 Ants might be my "hello world" project in a functional world. Is there any difference between what I choose or I can do random click? 2011-05-28T12:43:20 smiley1983: create an issue on github 2011-05-28T12:44:09 I don't know anything about Scala. I think OCaml is a little friendlier to the functional newcomer than Haskell, because it's more friendly to people familiar with procedural and OO styles 2011-05-28T12:44:51 (I wrote the OCaml starter package, blame me for the warnings you'll get in older versions of the compiler) 2011-05-28T12:45:49 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has joined #aichallenge 2011-05-28T12:47:56 So OCalm is my choise. Am I a first user of your package? 2011-05-28T12:49:23 Can I use you as a third teacher (after books and Google)? :) 2011-05-28T12:50:08 Feel free to ask questions, I'll answer them when I can 2011-05-28T12:50:35 I'm not aware of anyone else modifying it, although someone else uploaded it to the server as a test I think 2011-05-28T12:54:36 *** Naktibalda has joined #aichallenge 2011-05-28T13:04:09 *** UncleVasya has quit IRC (Ping timeout: 260 seconds) 2011-05-28T13:07:27 *** smiley1983 has quit IRC (Quit: leaving) 2011-05-28T13:08:35 *** UncleVasya has joined #aichallenge 2011-05-28T13:17:41 *** smiley1983 has joined #aichallenge 2011-05-28T13:19:42 *** Frontier has joined #aichallenge 2011-05-28T13:21:28 McLeopold: Sounds like you would like some visualizer errors to be recoverable. 2011-05-28T13:24:46 *** smiley1983 has quit IRC (Quit: leaving) 2011-05-28T13:32:46 *** amstan has joined #aichallenge 2011-05-28T13:32:46 *** ChanServ sets mode: +o amstan 2011-05-28T13:42:31 Frontier: just enough so that I can see what was happening. maybe a ?debug=true or something 2011-05-28T14:06:08 *** Naktibalda has quit IRC (Quit: ChatZilla 0.9.86.1 [Firefox 4.0.1/20110504060726]) 2011-05-28T14:07:06 janzert: ping 2011-05-28T14:07:19 janzert: did you change anything on ai-contest.com recently? 2011-05-28T14:07:24 all the apache entries are missing 2011-05-28T14:08:16 haven't been even logged into ai-contest.com in a month or so 2011-05-28T14:09:52 yeah.. jeff was playing with it 2011-05-28T14:09:55 *facepalm* 2011-05-28T14:10:09 like.. he doesn't even take this seriously.. 2011-05-28T14:10:11 goody :} 2011-05-28T14:10:29 his bash_history is full of attempts of installing minecraft servers 2011-05-28T14:39:27 yay: copy debs from base os to chroot os to cut down on redownloaded content 2011-05-28T14:39:48 janzert: for testing it's also a good idea to copy the debs back 2011-05-28T14:39:58 but i don't think that should be part of the installation 2011-05-28T14:41:59 McLeopold: ?debug exists, but it only displays the FPS and makes sure the Java Plugin is loaded in a fresh VM every time. 2011-05-28T14:42:42 janzert: any idea how to rebuild the old http file? 2011-05-28T14:42:57 amstan: Installing minecraft is easy on Gentoo. 'emerge minecraft-server' :p 2011-05-28T14:43:06 Frontier: is that the latest? 2011-05-28T14:43:36 Frontier: i just do this: et -qO- http://www.minecraft.net/download.jsp stdout|sed "s/\"/\n/g"|grep server.jar|grep download|sed "s/download/http:\/\/www.minecraft.net\/download/g"|xargs wget -O minecraft_server.jar 2011-05-28T14:43:39 usually the ebuild is updated frequently 2011-05-28T14:43:50 in a file called update.sh 2011-05-28T14:43:53 Oo 2011-05-28T14:44:02 it downloads the newest version of the server jar in the current folder 2011-05-28T14:44:33 of course, it's not gonna have java.. 2011-05-28T14:47:45 The latest ebuild in Gentoo is "20110526". Is that the current one? 2011-05-28T14:47:54 sounds right 2011-05-28T14:47:59 but latest is broken 2011-05-28T14:48:03 lots of bugs 2011-05-28T14:49:08 I only had a bug where the engine would not load a section of the map from the server and I was running in the void. 2011-05-28T14:49:44 *** UncleVasya has quit IRC (Ping timeout: 260 seconds) 2011-05-28T14:50:19 Frontier: see the new bugs lists: http://www.minecraftwiki.net/wiki/Version_history 2011-05-28T14:50:22 But that was months ago. Since all actions are processed on the server I didn't fall down and could mine blocks. 2011-05-28T14:51:00 advanced OpenGL... lol? this is the game with the most retro graphics ever and now it is suddenly advanced? 2011-05-28T14:51:30 *** UncleVasya has joined #aichallenge 2011-05-28T15:02:42 It means it uses hardware acceleration in certain cases, or something like that 2011-05-28T15:03:19 what i think it means is that it makes some chunks invisible 2011-05-28T15:03:30 *** delt0r_ has joined #aichallenge 2011-05-28T15:04:24 it always used hardware acceleration, maybe now it uses the shader model, instead of 'fixed function' 2011-05-28T15:04:40 *** delt0r___ has quit IRC (Ping timeout: 246 seconds) 2011-05-28T15:05:05 like this: http://www.youtube.com/watch?v=17sxwMovEEI&feature=channel_video_title 2011-05-28T15:05:06 7:20 2011-05-28T15:12:14 *** Accoun has quit IRC () 2011-05-28T15:25:31 *** Accoun has joined #aichallenge 2011-05-28T15:34:47 nice sheeps 2011-05-28T15:35:08 *** Palmik has quit IRC (Remote host closed the connection) 2011-05-28T15:37:11 *** berak has quit IRC (Quit: ChatZilla 0.9.84 [SeaMonkey 2.0a3/20090223135443]) 2011-05-28T15:38:28 incredible how much im not familiar with games, i only know fps 2011-05-28T15:38:50 *** qacek has joined #aichallenge 2011-05-28T15:39:12 amstan: That's the effect I meant. There were missing chunks in earlier version, too. Just not in single-player 2011-05-28T15:40:35 *** Redgis has joined #aichallenge 2011-05-28T15:53:19 *** arakyd has quit IRC (Quit: Page closed) 2011-05-28T16:24:49 *** UncleVasya has quit IRC (Ping timeout: 260 seconds) 2011-05-28T17:18:15 *** mcstar has quit IRC (Quit: WeeChat 0.3.5) 2011-05-28T17:22:00 Lol there exists a simple JavaScript test case with an inner loop that runs 500 times slower on the otherwise amazing Chrome JS engine: http://files.myopera.com/emoller/blog/webkit.html 2011-05-28T17:22:58 Compare that in Firefox 4/Opera vs. Chrome/Safari. Chrome/Safari are 100 times slower :p 2011-05-28T17:24:05 Some argue that comes from optimizing to a single benchmark called SunSpider, while real world applications like this unzip algorithm fall behind. 2011-05-28T17:24:34 I'd imagine it's more a matter of all of the JS engines having some weak spots. 2011-05-28T17:24:43 I'm sure the V8 team is working on it. :P 2011-05-28T17:25:08 i got 1800ms 2011-05-28T17:25:36 12 ms in FF4 :P 2011-05-28T17:26:19 Something is probably just messing up V8's attempt to detect and optimize hotspots. 2011-05-28T17:26:39 (My guess is that this loop is falling completely under the radar due to some unusualy aspect in its structure.) 2011-05-28T17:26:50 Interesting that Nitro is affected too, though. 2011-05-28T17:34:01 There is a bug in java starter pack 2011-05-28T17:34:20 start bot crash on some maps.. 2011-05-28T17:34:26 in 1 turn 2011-05-28T18:00:23 GreenTea: Could it be you are testing this on the beta test server? 2011-05-28T18:01:26 Yes 2011-05-28T18:03:04 for example http://aichallengebeta.hypertriangle.com/visualizer.php?game=4501 2011-05-28T18:03:06 *** mceier has quit IRC (Quit: leaving) 2011-05-28T18:03:19 why it crashed, I don't know.. 2011-05-28T18:03:20 It probably happens because it uses virtualization software that practically makes very little memory available to Java 2011-05-28T18:04:52 Java typically allocates 1GB of virtual memory, which is fine without that virtualization. But on the test server it is like allocating 100% of available RAM. 2011-05-28T18:06:44 hope developers will fix this.. 2011-05-28T18:06:49 no 2011-05-28T18:07:03 why? 2011-05-28T18:07:20 The fix is to use other virtualization software or none at all :p 2011-05-28T18:08:23 on previous constest there where no such problem. so I think it is possible to fix) 2011-05-28T18:08:54 It is not useful to force Java into 100mb memory max, which is the only way to 'fix' it for now by a developer. 2011-05-28T18:09:25 On the real constes bots will be written that may well use several 100mb of RAM. 2011-05-28T18:09:34 *contest 2011-05-28T18:10:28 I'd just wait until the contest starts and the real servers can handle Java's insane demand on virtual memory. 2011-05-28T18:11:57 To be fair the development team only found out about the problem with Java on the test server (donated by demerzel) when it was already in use and we tried to compile some Java on it. 2011-05-28T18:12:11 I see all java bots are crushed somtimes.. 2011-05-28T18:12:28 It's a pity that it is no possible to see error lot 2011-05-28T18:12:50 The error would be "out of memory" I'm sure. 2011-05-28T18:15:16 McLeopold: I looked at my source code for error handling and it is all very generic. I'd rather not bloat it. That would easily double my error handling code at this point :p. Would it be ok for you to download the defective replay and just fix the length of the invalid command string? 2011-05-28T18:15:45 maybe set "100mb memory max" for test server.. as workaround 2011-05-28T18:18:04 But that would be pretty much Roibos with vanilla, no more pure GreenTea... :( 2011-05-28T18:20:37 I also find the following in the server logs: java.lang.ArrayIndexOutOfBoundsException: 18 | at Ants.getPointType(Ants.java:303) | at MyBot.move(MyBot.java:104) | MyBot.doTurn(MyBot.java:164) | at Ants.run(Ants.java:347) 2011-05-28T18:21:00 I'm not sure if that is for the starter bot. 2011-05-28T18:21:34 when it happen, now? 2011-05-28T18:28:30 actually yes, an hour ago or so 2011-05-28T18:29:00 oh that was your bot, haha 2011-05-28T18:32:02 By the way.. java desctop java visualizer is not ready yet? 2011-05-28T18:32:49 and game engine.. 2011-05-28T18:32:58 not find them in starter package 2011-05-28T18:33:20 No they are not there. We use the engine from the git repository. 2011-05-28T18:34:15 But you can download the http://aichallengebeta.hypertriangle.com/visualizer/visualizer.jar 2011-05-28T18:34:51 You run it as "java -jar visualizer.jar xxx.replay" 2011-05-28T18:35:14 playgame.py ? 2011-05-28T18:35:58 o, thank you! 2011-05-28T18:35:59 that's the executable script yes. it depends on some more files in the directories 2011-05-28T18:36:51 You can also try play_one_game.sh or play_one_game.cmd. It will open a replay in your browser 2011-05-28T18:38:27 The only reason the desktop visualizer is ready is that I didn't code it from ground up, but just run the JavaScript inside Java using Rhino from Mozilla. That is why it is very slow and you may curse the visualizer.jar one day :p 2011-05-28T18:39:48 =) 2011-05-28T18:43:10 *** needsch has quit IRC (Quit: Leaving.) 2011-05-28T18:48:14 <_flag> What timezone is the server in? It tells me my most recent game was played tommorow 2011-05-28T19:17:56 *** hjax has joined #aichallenge 2011-05-28T19:17:58 *** McLeopold has quit IRC (Ping timeout: 240 seconds) 2011-05-28T19:18:26 did they fix the visualizer error yet? 2011-05-28T19:30:55 aichallenge: Marco Leise epsilon * r79b35d6 / (5 files in 2 dirs): visualizer: adress field can have &row=y&col=x&turn=t, visualizer will zoom in on that square in that turn - http://bit.ly/limswg 2011-05-28T19:33:05 aichallenge: Marco Leise epsilon * r71dd200 / ants/visualizer/js/Application.js : ... addition to last commit - http://bit.ly/kTiEaP 2011-05-28T19:34:11 This is an example link for this feature: http://marco-leise.homedns.org/antgui/stable/visualizer.php?game=8&row=69&col=27&turn=543 2011-05-28T19:34:27 hjax: What can I fix? 2011-05-28T19:35:03 the visualizer was crashing asking for a few pictures that didnt exist eariler 2011-05-28T19:35:25 oh yes, they are committed now. 2011-05-28T19:37:41 cool 2011-05-28T19:37:48 *** McLeopold has joined #aichallenge 2011-05-28T19:45:34 im annoyed :/ 2011-05-28T19:46:21 i couldnt figure out what was wrong with my bot, so i gave it some code for writing a log, and it cant make a log :/ 2011-05-28T19:47:41 You can use stderr and let the engine log that for you. There is a switch for that. 2011-05-28T19:49:11 where? 2011-05-28T19:58:19 *** sigh has joined #aichallenge 2011-05-28T19:59:29 -E or -e I think 2011-05-28T20:02:12 *** hjax has quit IRC (Ping timeout: 252 seconds) 2011-05-28T20:05:52 *** GreenTea has quit IRC (Ping timeout: 252 seconds) 2011-05-28T20:06:15 aichallenge: Marco Leise epsilon * re3cba1b / (2 files): visualizer: debug option allows loading of broken replay files in some cases - http://bit.ly/jFB3Bu 2011-05-28T20:07:19 @later tell McLeopold: You can now load broken replays in debug mode. It would be good if you could see an ant's id on the board... hmm 2011-05-28T20:07:19 Frontier: Aye, aye, sir 2011-05-28T20:08:01 cool 2011-05-28T20:14:10 McLeopold: some people ask for an engine download. something alongside the starter packages would be great 2011-05-28T20:23:26 Can someone that has worked on the compiler module explain what safeglob is trying to protect against? 2011-05-28T20:24:58 janzert: I think patterns like "../*" and "/usr/lib/…" 2011-05-28T20:26:09 jbroman: from where? the paths are all coming out of os.walk and the glob patterns given to it come from the language definitions within the module itself 2011-05-28T20:27:33 *** FireFly has quit IRC (Quit: swatted to death) 2011-05-28T20:27:36 I suspect it's safe without it. I have to confess that I don't know for sure. 2011-05-28T20:27:50 also the SAFEPATH regex allows ".." anyway 2011-05-28T20:28:03 * janzert shrugs 2011-05-28T20:28:36 I hate trying to modify this when I don't know for sure what it is trying to accomplish 2011-05-28T20:31:25 aichallenge: Marco Leise epsilon * r0f2c656 / (2 files): visualizer: second mode to the 'toggle player letters on ants' is 'display global ant ids' - http://bit.ly/iumvWr 2011-05-28T20:32:09 McLeopold: Now we can see clearly that the error appears when on start ants that survive until the player crashes/times out 2011-05-28T20:34:10 Here is the link for the lazy :p : http://marco-leise.homedns.org/antgui/stable/visualizer.php?game=9&row=86&col=8&turn=270&debug=true 2011-05-28T20:44:01 aichallenge: Marco Leise epsilon * re9ca861 / ants/visualizer/js/Application.js : visualizer: added url option to override config (i.e. ...&config={"label":2}) - http://bit.ly/kVxSmu 2011-05-28T20:45:47 So much for overengineering: http://marco-leise.homedns.org/antgui/stable/visualizer.php?game=9&row=86&col=8&turn=270&debug=true&config={%22label%22:2} 2011-05-28T20:48:18 This loads 9.replay (game=9), ignoring errors in the parsing process (debug=true), zooms in on row=86, col=8 at turn=270 and shows global ids on ants (config={"label":2}) 2011-05-28T21:04:46 Numbers should probably hidden once they're too small to read. But the visualizer is getting very impressive. 2011-05-28T21:05:16 *** Mathnerd314 has quit IRC (Ping timeout: 264 seconds) 2011-05-28T21:05:37 (Also, dragging appears to work in the large map but not the minimap. But again, a minor complaint.) 2011-05-28T21:07:34 *** tomh has quit IRC (Quit: tomh) 2011-05-28T21:09:54 jbroman: The point of showing the numbers even when they are too small is to show the user that they are active. If you zoom out and toggle them you would get zero feedback. 2011-05-28T21:10:28 Mmmmmm . . . okay, fine. 2011-05-28T21:10:36 I understand now. 2011-05-28T21:19:53 *** Mathnerd314 has joined #aichallenge 2011-05-28T21:24:26 *** Bhasker has joined #aichallenge 2011-05-28T21:24:38 What's up 2011-05-28T21:29:06 *** Bhasker has quit IRC (Ping timeout: 252 seconds) 2011-05-28T21:30:42 j3camero: ping 2011-05-28T21:43:03 *** hjax has joined #aichallenge 2011-05-28T21:43:48 Anyone got a good way to debug a bot? 2011-05-28T21:44:18 I just want to know what's wrong with mine :/' 2011-05-28T21:48:41 Anyone here? 2011-05-28T21:49:43 McLeopold: your wisdom is requested 2011-05-28T21:52:11 Anyone? -__- 2011-05-28T21:55:16 *** Eruonen has quit IRC () 2011-05-28T21:55:59 I'm really bored now :( 2011-05-28T22:00:26 hjax: I just print to stderr and use the -E flag with playgame.py 2011-05-28T22:01:43 ? I tried having my program write to a file, it never wrote the first entree, which should have been right after it made the file :/ 2011-05-28T22:02:20 So my program keeps timing out and i have no clue why 2011-05-28T22:03:03 aichallenge: Marco Leise epsilon * rd5c7d54 / ants/visualizer/js/Application.js : visualizer: minimap can be dragged (idea by jbroman); when row and col are given in the url, change the 'center map' button to these coordinates - http://bit.ly/k0QmPC 2011-05-28T22:04:25 qacek: do you know python? 2011-05-28T22:04:40 hjax: IO is usually buffered. You should check that you flush these buffers when you want a line to appear in your file. 2011-05-28T22:04:49 *** Kingpin13 has quit IRC (Quit: quit) 2011-05-28T22:05:29 Frontier: I've never had problems writing to files before :/ 2011-05-28T22:05:30 This is done automatically when you close a file, but not when your program is terminated by the engine. 2011-05-28T22:05:43 Moan :/ 2011-05-28T22:06:20 In C++ you would write: out << "my entry" << endl; 2011-05-28T22:06:32 hjax: I would just use stderr: print >> sys.stderr, 'Text' 2011-05-28T22:06:38 endl in this case adds a line-break and also makes sure the buffer is flushed 2011-05-28T22:07:31 Here's my code if anyone wants to read it: http://pastebin.com/m2GjWaMJ 2011-05-28T22:08:22 hjax: Either print to stderr, or flush the file buffer. You should be ok then. 2011-05-28T22:09:30 Hmmm either way I can't try it until tomorrow :/ 2011-05-28T22:09:41 It is only understandable that not every byte is written to a file at once. The computer can process large blocks of several kilobytes much more efficiently, so the bytes are kept in a buffer until there is enough data to write. 2011-05-28T22:09:41 Thanks though :D 2011-05-28T22:10:07 Frontier: makes sense, I'll go with printing then 2011-05-28T22:10:59 I should become a teacher 2011-05-28T22:11:13 Lol 2011-05-28T22:12:34 I just realized a flaw in my code as well that would make it ineffective even if it did work :/ 2011-05-28T22:13:20 For each food I'm finding the closest ant, but there may be foods closer to the ant that I haven't looped through yet 2011-05-28T22:13:47 is it possible to submit a pull request for a submodule on github? 2011-05-28T22:15:54 *** hjax has quit IRC (Quit: Rooms • iPhone IRC Client • http://www.roomsapp.mobi) 2011-05-28T22:19:51 amstan: I don't know how to submit a pull request without forking, could you pull my haskell bot, I made some changes to improve readability and speed 2011-05-28T22:20:09 qacek: link? 2011-05-28T22:20:32 https://github.com/qacek/ants-haskell 2011-05-28T22:20:49 it should be in the .gitmodules file though right? 2011-05-28T22:21:13 qacek: yeah, i can handle that 2011-05-28T22:21:16 thanks 2011-05-28T22:21:33