2012-10-05T00:00:25 gotta be up for work in 4 hours, see you around 2012-10-05T00:00:35 o/ 2012-10-05T00:09:49 *** amstan has quit IRC (Quit: Konversation terminated!) 2012-10-05T00:35:22 *** mceier has quit IRC (Quit: leaving) 2012-10-05T00:49:49 *** antimatroidl has joined #aichallenge 2012-10-05T00:58:16 *** antimatroidl1 has joined #aichallenge 2012-10-05T00:58:16 *** antimatroidl has quit IRC (Read error: Connection reset by peer) 2012-10-05T01:44:27 *** pairofdice has joined #aichallenge 2012-10-05T01:45:52 *** antimatroidl1 has quit IRC (Quit: Leaving.) 2012-10-05T02:34:50 *** mceier has joined #aichallenge 2012-10-05T03:26:06 *** knyppeld1nan has joined #aichallenge 2012-10-05T03:46:39 *** mlp has joined #aichallenge 2012-10-05T04:22:58 *** epicmonkey has joined #aichallenge 2012-10-05T05:02:07 *** antimatroidl has joined #aichallenge 2012-10-05T05:04:01 *** mceier has quit IRC (Read error: Connection reset by peer) 2012-10-05T05:05:33 *** mceier has joined #aichallenge 2012-10-05T05:37:37 *** wentam has joined #aichallenge 2012-10-05T05:38:46 *** antimatroidl1 has joined #aichallenge 2012-10-05T05:38:47 *** antimatroidl has quit IRC (Read error: Connection reset by peer) 2012-10-05T05:42:03 *** knyppeld1nan has quit IRC (Ping timeout: 255 seconds) 2012-10-05T05:56:03 *** thestinger has joined #aichallenge 2012-10-05T06:08:03 *** antimatroidl1 has quit IRC (Ping timeout: 248 seconds) 2012-10-05T06:32:59 *** mleise has joined #aichallenge 2012-10-05T06:39:55 *** antimatroidl has joined #aichallenge 2012-10-05T06:40:30 *** Eibwen has quit IRC (Ping timeout: 246 seconds) 2012-10-05T07:27:34 *** mcstar has joined #aichallenge 2012-10-05T07:29:25 thestinger: i have a dilemma 2012-10-05T07:30:05 ? 2012-10-05T07:30:15 wasnt sure you were there 2012-10-05T07:30:18 now i know 2012-10-05T07:30:34 thestinger: remember what we talked about yesterday? 2012-10-05T07:30:40 i can do it with dbus 2012-10-05T07:30:52 the problem is, that i have 3 choices 2012-10-05T07:30:59 wait what? asteroids? 2012-10-05T07:31:04 oh 2012-10-05T07:31:10 session management 2012-10-05T07:31:12 right. 2012-10-05T07:31:15 do it in python, do it in haskell or do it in awesome+lua 2012-10-05T07:31:22 ewww, awesome 2012-10-05T07:31:32 thats your choice? 2012-10-05T07:31:56 do it with xmonad :) 2012-10-05T07:32:00 nah 2012-10-05T07:32:03 :( 2012-10-05T07:32:04 xmonad sucks 2012-10-05T07:32:09 it does? 2012-10-05T07:32:12 I have used it :p 2012-10-05T07:32:15 haven't* 2012-10-05T07:32:22 well, it has a crippled mouse interface 2012-10-05T07:32:31 i'd have to set up bluetile with it 2012-10-05T07:32:46 and i already have an awesome setupt that i like 2012-10-05T07:34:08 it'd be nice if i could sript awesome directly 2012-10-05T07:34:31 though lua is not my cup of tea 2012-10-05T07:35:08 awesome has a really crappy lua api 2012-10-05T07:35:23 is that what you read? 2012-10-05T07:35:28 I used to use it 2012-10-05T07:35:31 before i3 2012-10-05T07:35:35 how long? 2012-10-05T07:35:39 2 years? 2012-10-05T07:35:41 and why didnt you tell me before ? 2012-10-05T07:35:44 :) 2012-10-05T07:36:17 when I switched from windows -> arch, I started with a silly standalone compiz setup 2012-10-05T07:36:23 quickly realized it sucked and used awesome 2012-10-05T07:36:40 anyway awesome is alright, but it doesn't add much value over dwm 2012-10-05T07:36:59 there's a status bar you can draw on with cairo.... that's about the only good thing I have to say about it 2012-10-05T07:37:38 what does i3 have that awesome doesnt, and its a killer feature? 2012-10-05T07:37:44 tabs? 2012-10-05T07:37:56 or stack, or whatyoumacallit 2012-10-05T07:37:58 tabbing, stacking, managing windows as groups (workspace == tree, so you can manage sub-trees) 2012-10-05T07:38:21 fortunately i dont need that 2012-10-05T07:38:44 for me, its a bit annoying that i3 dynamically adds/removes workspaces 2012-10-05T07:39:51 oh and containers 2012-10-05T07:40:11 also, json ipc API - but it's not very far along yet 2012-10-05T07:45:18 thestinger: do you think i can use the awesome api, or do i need a dbus-lua binding? 2012-10-05T07:45:56 it does have a dbus module but it appears to have no functions 2012-10-05T07:46:01 I think it's just internal stuff 2012-10-05T07:46:07 mcstar: don't use awesome :[ 2012-10-05T07:46:14 mcstar: you can use any wm afaict 2012-10-05T07:46:21 the wm is just another X11 client to restore 2012-10-05T07:47:11 hm? 2012-10-05T07:47:15 nah 2012-10-05T07:47:25 we are not thinking the same thing 2012-10-05T07:47:47 thestinger: i want to register a handler with the session bus, that filters new name requests 2012-10-05T07:47:59 each okular instance does request one 2012-10-05T07:48:17 when that happens, i can quaery the path of the document that okular opened 2012-10-05T07:48:26 now ill store that in my daemon 2012-10-05T07:48:35 and write that to disk 2012-10-05T07:48:57 (not too frequently) 2012-10-05T07:49:17 anyway, this way i can keep track of the open windows/files of okular 2012-10-05T07:49:25 and restore when needed 2012-10-05T07:49:36 and i'd start this daemon when awesome starts 2012-10-05T07:49:56 but if i do this directly in the awesome config, i wouldnt have to bother with a daemon 2012-10-05T07:55:16 *** Scooper has joined #aichallenge 2012-10-05T07:58:37 thestinger: did you look at subtle? 2012-10-05T07:58:47 yes 2012-10-05T07:58:49 and hlwm 2012-10-05T07:58:58 i3 is the best :P 2012-10-05T07:59:18 *** antimatroidl has quit IRC (Read error: No route to host) 2012-10-05T07:59:23 *** antimatroidl1 has joined #aichallenge 2012-10-05T07:59:55 thestinger: hlwm? im not getting the right results 2012-10-05T07:59:59 theres an lwm 2012-10-05T08:00:16 mcstar: herbstluftwm 2012-10-05T08:00:27 you can see why I used an abbreviation... 2012-10-05T08:00:33 lol 2012-10-05T08:00:36 I had to google it to remember the name 2012-10-05T08:00:54 luft is air 2012-10-05T08:00:59 herbst well, herb? 2012-10-05T08:01:05 idk 2012-10-05T08:01:11 I just google herbstuffwm 2012-10-05T08:01:14 googled* 2012-10-05T08:02:25 thestinger: the screenies look good 2012-10-05T08:02:36 just use i3! 2012-10-05T08:03:05 stop that 2012-10-05T08:03:11 im not going to 2012-10-05T08:03:13 no 2012-10-05T08:03:15 never ever 2012-10-05T08:04:06 i dont see a good reason to switch from awesome 2012-10-05T08:12:17 tab 3 arbitrary windows 2012-10-05T08:12:20 mod+a, move them as a group 2012-10-05T08:12:36 fewer bugs + great test suite :P 2012-10-05T08:13:02 ipc interface, so you can write your thing in *any* language 2012-10-05T08:14:20 hide_edge_borders both 2012-10-05T08:14:23 extra row and col of pixels! 2012-10-05T08:14:41 thestinger: *any* language? that can be a downside too 2012-10-05T08:15:01 mcstar: why? 2012-10-05T08:15:05 it's all just json over ipc 2012-10-05T08:15:10 what if you want to share your addition with someone? 2012-10-05T08:15:16 so you share it 2012-10-05T08:15:20 yeah 2012-10-05T08:15:26 i3-msg is a sample ipc application 2012-10-05T08:15:32 but possibly they will need another runtime 2012-10-05T08:15:32 so you could just wrap that. 2012-10-05T08:15:42 mcstar: you would just use i3-msg from a shell script 2012-10-05T08:15:54 now thats great 2012-10-05T08:15:58 or the ipc.c/ipc.h from C 2012-10-05T08:16:03 lets customize my wm with shell scripts 2012-10-05T08:16:18 mcstar: so, awesome? 2012-10-05T08:16:31 it has lua 2012-10-05T08:16:31 vicious volume widget -> spawns amixer for every update 2012-10-05T08:16:40 mcstar: yes and lua hardly has any libs 2012-10-05T08:16:43 and it has an awesome-client too, if you want it 2012-10-05T08:16:52 thestinger: i didnt say it was the best choice 2012-10-05T08:17:01 i didnt really care 2012-10-05T08:17:02 the creator of awesome doesn't think it was a good choice :P 2012-10-05T08:17:09 he wrote 2 posts about how much lua sucks 2012-10-05T08:17:09 at the time, i didnt feel too comfortable with haskell 2012-10-05T08:17:26 and he said he wished he had used python so he didn't have to invent his own crappy object system 2012-10-05T08:17:39 hehe, lua tables right? 2012-10-05T08:18:22 i wouldnt have mined python 2012-10-05T08:20:56 mcstar: use i3 2012-10-05T08:21:08 wtf 2012-10-05T08:21:20 do you guys get money for advertising i3? 2012-10-05T08:21:24 XD 2012-10-05T08:21:41 aerique: i expected you to be a heavy stumpwm user :) 2012-10-05T08:22:35 mcstar: i used to :-) went on a tiling wm binge 2 years ago, tried awesome (which i prefer after i3), xmonad, stayed a while with stumpwm, triend some others and finally discovered i3 2012-10-05T08:23:20 aerique: do you have ghc? 2012-10-05T08:24:17 mcstar: is that for haskell? 2012-10-05T08:24:27 yeah, the haskell compiler 2012-10-05T08:24:38 im asking if you have the development envritonment 2012-10-05T08:24:43 mcstar: not on this machine. at a new job and i'm still getting set up 2012-10-05T08:24:50 home? 2012-10-05T08:25:30 aerique: im asking, because i made a little asteroids game in haskell, and you told me, you wanted to get your hands dirty with haskell 2012-10-05T08:26:48 mcstar: no, not at home unfortunately. i wouldn't mind checking it out tonight (if i'm not off to bed early) 2012-10-05T08:27:33 https://github.com/liquid-phynix/asteroids 2012-10-05T08:28:23 meh, boost::optional still has no move constructor 2012-10-05T08:28:28 I'll just make my own 2012-10-05T08:29:18 thestinger: is there a central source of documentation for python packages? 2012-10-05T08:29:27 eggs or whatnot 2012-10-05T08:29:47 mcstar: https://readthedocs.org/ 2012-10-05T08:29:47 like hackage 2012-10-05T08:30:04 mcstar: well there's pypi... 2012-10-05T08:30:19 pretty much every python project uses rst + sphinx 2012-10-05T08:30:28 I kinda like the plaintext docs more than the generated ones 2012-10-05T08:31:17 what they arent hosted in one single place? 2012-10-05T08:31:31 ill just google it 2012-10-05T08:31:36 mcstar: is it a plain asteroids game or something for the ai challenge? 2012-10-05T08:31:55 unrelated to aichallenge 2012-10-05T08:32:11 mcstar: I don't really think so 2012-10-05T08:32:21 they include docs in the source though 2012-10-05T08:32:33 arch packagers often strip that... 2012-10-05T08:33:37 mcstar: https://requests.readthedocs.org/en/latest/ anyway it's all on readthedocs 2012-10-05T08:33:53 aerique: this motivated me http://news.ycombinator.com/item?id=4429407 2012-10-05T08:34:18 "all" :P 2012-10-05T08:36:34 dbus-python is documented on freedesktop.org 2012-10-05T08:38:27 mcstar: yeah cool, i read that too 2012-10-05T08:38:58 aerique: that had levels, and bonuses, and lives, my game cant do that 2012-10-05T08:39:01 has* 2012-10-05T08:39:16 on the other hand, mine looks better(i hope :)) 2012-10-05T09:04:48 thestinger: when they say 'Louise' i always hear they say 'wuss' 2012-10-05T09:05:36 you know, when dexter catches louise snooping around 2012-10-05T09:05:57 oh 2012-10-05T09:32:56 *** mleise1 has joined #aichallenge 2012-10-05T09:32:56 *** mleise has quit IRC (Read error: Connection reset by peer) 2012-10-05T10:29:18 *** MeNistry has joined #aichallenge 2012-10-05T10:30:09 Hi! Is a new AI challenge gonna take place this fall ? 2012-10-05T10:30:27 unlikely 2012-10-05T10:30:46 ok, why? 2012-10-05T10:31:22 no one has decided to step forward and manage a contest 2012-10-05T10:32:17 ok, thanks for your quick answer 2012-10-05T10:32:26 what a pity! 2012-10-05T10:36:51 thestinger: when is __new__ called? 2012-10-05T10:37:10 mcstar: you really don't want to use __new__ and __del__ :P 2012-10-05T10:37:21 why? 2012-10-05T10:37:31 mcstar: it's a static method 2012-10-05T10:37:31 (i dont want to use it directly) 2012-10-05T10:37:43 static how? 2012-10-05T10:37:47 like in c? 2012-10-05T10:37:47 takes class, returns new instance 2012-10-05T10:37:52 mcstar: as in C++/java 2012-10-05T10:37:57 doesn't take an instance as the argument 2012-10-05T10:38:45 thestinger: whats the use case? 2012-10-05T10:38:51 dunno. 2012-10-05T10:38:54 never seen a use case :P 2012-10-05T10:39:03 oh wait 2012-10-05T10:39:08 the dbus binding uses it 2012-10-05T10:39:08 "__new__() is intended mainly to allow subclasses of immutable types (like int, str, or tuple) to customize instance creation. It is also commonly overridden in custom metaclasses in order to customize class creation." 2012-10-05T10:39:21 mcstar: it probably shouldn't 2012-10-05T10:40:19 thestinger: maybe it tries to build a singleton pattern 2012-10-05T10:40:29 but that's evil :( 2012-10-05T10:40:42 modules are singletons. 2012-10-05T10:41:21 there is this class that lets you register a new name with dbus 2012-10-05T10:41:37 and if you request the same name again, it will give back your previous instance 2012-10-05T10:41:43 thats the logic behind it i think 2012-10-05T10:44:24 thestinger: __init__ of that class is just a dumb pass 2012-10-05T10:44:36 thestinger: whats the syntax to call __new__ directly? 2012-10-05T10:44:45 I don't think you can. 2012-10-05T10:44:50 i dont mean .__new__() 2012-10-05T10:44:55 mcstar: new returns you the object 2012-10-05T10:45:04 foo = Foo() is calling __new__ 2012-10-05T10:45:46 thestinger: why doesnt that call init on Foo? 2012-10-05T10:45:50 init is passed the instance new creates 2012-10-05T10:45:56 as the first param 2012-10-05T10:46:50 def __new__(cls 2012-10-05T10:46:54 ok, i get it 2012-10-05T10:47:01 thats why its first param is called cls 2012-10-05T10:53:03 mcstar: related to your session management stuff, look at 'checkpoint' 2012-10-05T10:53:09 *** mceier has quit IRC (Quit: leaving) 2012-10-05T10:53:15 mcstar: https://ckpt.wiki.kernel.org/index.php/Main_Page 2012-10-05T10:53:43 yeah 2012-10-05T10:53:45 seen it 2012-10-05T10:54:00 theres some language implementation that uses it 2012-10-05T10:54:05 i forgot which 2012-10-05T10:57:29 *** MeNistry has quit IRC (Quit: Page closed) 2012-10-05T11:03:18 *** amstan has joined #aichallenge 2012-10-05T11:03:18 *** ChanServ sets mode: +o amstan 2012-10-05T11:08:53 *** epicmonkey has quit IRC (Ping timeout: 246 seconds) 2012-10-05T11:22:12 *** Areks|2 has left #aichallenge ("Once you know what it is you want to be true, instinct is a very useful device for enabling you to know that it is") 2012-10-05T11:28:33 *** flowenol has joined #aichallenge 2012-10-05T11:28:48 sooooo 2012-10-05T11:28:57 is the contest dead 2012-10-05T11:39:25 no 2012-10-05T11:39:30 just smells funny 2012-10-05T11:44:39 *** flowenol has quit IRC (Quit: Page closed) 2012-10-05T12:03:52 *** mlp has quit IRC (Quit: Page closed) 2012-10-05T12:22:37 *** Kingpin13 has joined #aichallenge 2012-10-05T12:28:06 *** mceier has joined #aichallenge 2012-10-05T12:32:36 *** Accoun has quit IRC (Ping timeout: 246 seconds) 2012-10-05T12:39:13 *** mpstyler has joined #aichallenge 2012-10-05T12:39:56 probably th most popular question here, will there be an AI challenge 2012?? 2012-10-05T12:51:34 No. 2012-10-05T12:51:48 we could run one in 2013 and call it 2012 2012-10-05T12:52:48 the previous ones were very exciting. its a pity that there wont be one this year 2012-10-05T12:58:05 *** Accoun has joined #aichallenge 2012-10-05T13:01:20 so why there will be no 2012 challenge? 2012-10-05T13:12:16 no one's been working on it? 2012-10-05T13:12:36 and i have no idea what's been going on org-wise, in terms of money etc 2012-10-05T13:17:44 if we could print money, i think we could still organize a contest 2012-10-05T13:18:09 from federal prison? 2012-10-05T13:18:21 does canada have prisons? 2012-10-05T13:20:57 :) 2012-10-05T13:21:06 thestinger doesnt even know 2012-10-05T13:21:10 typical canadian 2012-10-05T13:37:34 *** Garf has joined #aichallenge 2012-10-05T13:38:01 *** Apophis_ has quit IRC (Read error: Connection reset by peer) 2012-10-05T13:42:05 thestinger: how do you think i should store the my dict? 2012-10-05T13:42:08 on disk 2012-10-05T13:42:15 what kind of dict? 2012-10-05T13:42:18 how big? 2012-10-05T13:42:21 string to string 2012-10-05T13:42:25 not long 2012-10-05T13:42:32 so why on disk? :P 2012-10-05T13:42:41 i needs to be saved 2012-10-05T13:42:51 mcstar: just use json.dump then 2012-10-05T13:42:51 to be available across sessions, remember? 2012-10-05T13:42:58 it* 2012-10-05T13:43:17 *** mviel has quit IRC (Quit: Leaving) 2012-10-05T13:43:30 oh 2012-10-05T13:43:37 i dont need that actually 2012-10-05T13:43:45 i just need to save the filenames 2012-10-05T13:44:37 thestinger: how about saving a list of strings directly? 2012-10-05T13:44:50 the string representation can reconstruct the object 2012-10-05T13:44:57 just json.dump it 2012-10-05T13:45:00 its similar to haskell read/show 2012-10-05T13:45:05 it's faster than pickle and human readable 2012-10-05T13:45:11 mcstar: repr/ast.literal_eval 2012-10-05T13:45:14 but don't bother 2012-10-05T13:45:18 why? 2012-10-05T13:45:20 json is faster 2012-10-05T13:45:25 ok 2012-10-05T13:45:28 and a standard, easy to work with format 2012-10-05T13:45:38 i saw json 2012-10-05T13:45:48 just that python has alraedy support for what i want 2012-10-05T13:46:00 with a nice simple document api 2012-10-05T13:46:32 mcstar: there's also pickle, but it's binary and slower than json 2012-10-05T13:46:42 and you can pickle code... 2012-10-05T13:47:59 i used pickle before 2012-10-05T13:48:08 saved whole classes with massive data 2012-10-05T13:48:17 once it couldnt load it back though 2012-10-05T13:48:29 thestinger: how can pickle be slower? 2012-10-05T13:48:46 because the json module is a fast piece of C 2012-10-05T13:48:48 json is simple 2012-10-05T13:48:51 easy to parse, simple format 2012-10-05T13:48:59 pickle is complex - has code, all kinds of escape things, etc. 2012-10-05T13:49:01 yeah, but pickle is binary 2012-10-05T13:49:06 but it's a bad binary format 2012-10-05T13:49:08 anyway, ill use json 2012-10-05T13:49:18 msgpack/protobuf are the sane binary serialization formats 2012-10-05T13:49:22 but text is fine 2012-10-05T13:49:48 mcstar: pickle is basically bytecode 2012-10-05T13:49:59 code, rather than data 2012-10-05T13:55:09 thestinger: about 'with', if i want to open a file with it, and lets say that file doesnt exist, the body of with wont be executed, and i cant branch on that , can i? 2012-10-05T13:55:42 mcstar: you'll still get an exception you can catch 2012-10-05T13:55:57 try: around with? 2012-10-05T13:56:07 yes but then you might as well use try/except/finally 2012-10-05T13:56:11 with close in finally 2012-10-05T13:56:11 yeah 2012-10-05T13:56:18 thats what i was thinking 2012-10-05T13:56:22 with is just a wrapper for the common case 2012-10-05T13:56:27 it's nice for stuff like db transactions 2012-10-05T13:56:40 with will be good though now, i just wanted to know 2012-10-05T14:08:29 thestinger: there are a lot of os.spawn* variants 2012-10-05T14:08:39 mcstar: don't use any 2012-10-05T14:08:48 hehe 2012-10-05T14:08:51 subprocess.check_output 2012-10-05T14:09:03 why use subprocess if theres os? 2012-10-05T14:09:04 returns stdout, throws exception on non-zero return 2012-10-05T14:09:11 mcstar: os is deprecated for that stuff 2012-10-05T14:09:12 i dont want to communicate with their stdin/stdout 2012-10-05T14:09:33 so subprocess.check_call(["foo", "-a"]) 2012-10-05T14:09:38 and a non-list works too 2012-10-05T14:09:46 does it now? 2012-10-05T14:10:00 check_call("exit 1") 2012-10-05T14:10:02 i had to split the parameters 2012-10-05T14:10:36 oh it might only work when shell is on (which you don't want) 2012-10-05T14:10:53 anyway the basic one is .call(["foo", "bar"]) 2012-10-05T14:11:01 the check ones throw on non-zero 2012-10-05T14:12:16 thestinger: but .call is a blocking call 2012-10-05T14:12:28 obvously i want to spawn a thread, and dont want to block it 2012-10-05T14:12:35 should u just append &? 2012-10-05T14:12:38 u->i 2012-10-05T14:12:50 oh, you don't want blocking? 2012-10-05T14:12:54 ofc not 2012-10-05T14:12:55 mcstar: no it doesn't run in a shell by default 2012-10-05T14:12:59 it's like popen in C 2012-10-05T14:13:32 mcstar: if you don't want that, then sure use spawn :P 2012-10-05T14:13:38 haha, ok 2012-10-05T14:13:41 wait actually no 2012-10-05T14:13:44 that's deprecated too 2012-10-05T14:13:56 how do you know that btw? 2012-10-05T14:13:59 just Popen(["foo", "bar"]) 2012-10-05T14:14:30 http://www.python.org/dev/peps/pep-0324/ 2012-10-05T14:14:40 oh 2012-10-05T14:15:32 Popen(["foo", "-a"]) just returns a pid object 2012-10-05T14:15:45 the functions are just wrappers for common cases 2012-10-05T14:15:58 os is a pretty bad module now 2012-10-05T14:16:20 it's a mess, there's shutil and subprocess which replace parts 2012-10-05T14:20:06 *** foRei has joined #aichallenge 2012-10-05T14:27:06 thestinger: great, its working 2012-10-05T14:27:28 68 lines of awesomeness 2012-10-05T14:27:49 this python is a useful language 2012-10-05T14:30:54 thestinger: http://sprunge.us/IiIc?python 2012-10-05T14:32:34 mcstar: nitpick - shouldn't use except: 2012-10-05T14:32:40 it catches even SyntaxErrors 2012-10-05T14:33:13 what does that mean? 2012-10-05T14:33:19 how can syntax error happen there? 2012-10-05T14:33:37 anyway, i can put a FileERror or something 2012-10-05T14:33:45 yeah, do that :P 2012-10-05T14:34:06 thestinger: but explain what did you mean 2012-10-05T14:34:46 mcstar: http://sprunge.us/RJIJ it won't throw any error on that, since you're catching it 2012-10-05T14:34:55 anyway it could happen in any function you call 2012-10-05T14:34:59 better to just catch specific stuff 2012-10-05T14:35:01 oh 2012-10-05T14:35:08 you mean it hides a syntrax error 2012-10-05T14:35:11 yeah 2012-10-05T14:35:22 or undefined vars, etc. 2012-10-05T14:35:33 ok 2012-10-05T14:36:21 mcstar: oh and a.__str__() can be str(a) 2012-10-05T14:36:37 i could actually get rid of that i think 2012-10-05T14:37:10 im not sure if json behaves differently on python unicode strings or on that dbus.String 2012-10-05T14:37:19 that the functions actually return 2012-10-05T14:37:29 they had a different __repr__ though 2012-10-05T14:37:37 mcstar: it will convert both to json strings, which are utf-8 2012-10-05T14:37:57 ok, if it looks the same in the file, then i can get rid of that 2012-10-05T14:38:35 mcstar: well, unless it doesn't know how to serialize those 2012-10-05T14:38:46 well see 2012-10-05T14:39:15 thestinger: its the same 2012-10-05T14:39:19 \o/ 2012-10-05T14:39:35 now i will try it out with awesome 2012-10-05T14:39:42 *** mcstar has quit IRC (Quit: mcstar) 2012-10-05T14:40:57 *** jacob_strauss has joined #aichallenge 2012-10-05T14:53:24 *** mcstar has joined #aichallenge 2012-10-05T14:53:55 thestinger: it worked 2012-10-05T14:54:00