--- Log opened Sun Apr 11 00:00:39 2010
20100411 00:01:54-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit []
20100411 00:03:39-!- silene [~plouf@wesnoth/developer/silene] has quit [Quit: Leaving.]
20100411 00:07:59-!- Blueblaze [~nick@adsl-99-182-52-75.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100411 00:08:48-!- meric [~Eric@124-170-188-18.dyn.iinet.net.au] has quit [Quit: meric]
20100411 00:10:19-!- mordocai [~mordocai@66.119.9.243] has joined #wesnoth-dev
20100411 00:15:20-!- mordocai [~mordocai@66.119.9.243] has quit [Quit: Leaving]
20100411 00:27:00-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Quit: Leaving.]
20100411 00:31:53< icelus> suppose I add a file to wesnoth
20100411 00:32:01< icelus> i know how to do that with autotools
20100411 00:32:07< icelus> where do i add for your other build systems
20100411 00:32:14< icelus> cmake/scons/visual studio/etc
20100411 00:32:25< icelus> what's the equivalent of Makefile.am
20100411 00:32:52< Crab_> src/CMakeLists.txt
20100411 00:33:04< Crab_> src/SConscript
20100411 00:33:17< Crab_> MSVC will be generated from cmake file
20100411 00:33:49< icelus> Crab_, thanks
20100411 00:34:45< Crab_> also, fix projectfiles/CodeBlocks and projectfiles/CodeBlocks-SCons - it is easy and will save mcshark some work.
20100411 00:35:10< Crab_> ( ignore projectfiles/VC9 )
20100411 00:37:07< icelus> the only danger is i don't know to test that stuff
20100411 00:37:12< icelus> i'll edit and be careful
20100411 00:37:23< icelus> but i don't have codeblocks to test i did it right
20100411 00:37:26< Crab_> well, it's rather easy if you just add files, not dependencies
20100411 00:37:39< icelus> so basically i added a unit_helper.hpp/cpp
20100411 00:37:40< Crab_> just review the diff to see if all is sane
20100411 00:37:44< icelus> pair of files
20100411 00:37:50< icelus> ok
20100411 00:38:06< icelus> the other thing is i would need to tie up my machine for 6 hrs to test all these options =p
20100411 00:38:14< Crab_> oh
20100411 00:38:14< icelus> i tested automake but that took the last few hours
20100411 00:38:26< icelus> i think i added scons/cmake correctly
20100411 00:38:37< icelus> hopefully i can upload this patch and it can be tested
20100411 00:38:55< Crab_> yes, chances are that the patch reviewer will use one of these :)
20100411 00:39:48< icelus> ok cool =)
20100411 00:44:04< Crab_> (just started a new game on old dark forecast with 1st version of your patch)
20100411 00:45:55< Crab_> ok, there's still OOSes there, but the 'level mismatch because of unstore_unit advance=true' is fixed.
20100411 00:46:33< Crab_> icelus: so, congratulations, your patch fixes this particular issue even for the original bug.
20100411 00:47:23< icelus> Crab_, cool, that's good news
20100411 00:47:31< icelus> i'm just about to upload a revised patch
20100411 00:47:35< Crab_> icelus: want to fix the second oos on DF ?
20100411 00:47:41< icelus> unfortunately that version didn't fix all possible OOSes
20100411 00:47:58< icelus> because you could have evilly specified an pre-advance or post-advance action
20100411 00:48:09< icelus> which would have still broken sync.
20100411 00:48:18< icelus> i fixed that though by keeping track of the game phase properly
20100411 00:48:22< icelus> not just relying on the event name
20100411 00:48:34< Crab_> yes, events can call other events...
20100411 00:48:36< icelus> so the revised patch should work no matter what horrible nested actions you throw at prestart
20100411 00:48:58-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20100411 00:50:40< Crab_> icelus: good
20100411 00:51:19< Crab_> icelus: ftp://ftp.terraninfo.net/wesnoth/2p_Dark_Forecast_oos.cfg is the DF modified to show another OOS happening, this time - in unit 'moves left'. it might be due to bad coding, or due to another engine bug.
20100411 00:51:46< Crab_> icelus: just start, recruit a unit with good movement (vision range), and do nothing for several turns
20100411 00:51:47< icelus> Crab_, i didn't ever get DF to replicate the original bug
20100411 00:51:54< icelus> hm ok
20100411 00:52:04< Crab_> icelus: that one is slightly modified to make the bug appear faster
20100411 00:52:12< Crab_> icelus: you'd be able to replicate the original on it
20100411 00:52:19-!- gabba [~gabba@70.35.163.161] has left #wesnoth-dev []
20100411 00:52:28< icelus> Crab_, ok i'll take a look
20100411 00:52:34< icelus> Crab_, there's another sync bug too
20100411 00:52:46< icelus> but this one was keeping me awake at night
20100411 00:52:54< icelus> i just want to kill it for now =p
20100411 00:53:47< Crab_> yes, that's the right thing to do :)
20100411 00:57:42-!- Zarel| [~Zarel@warzone2100/developer/Zarel] has quit [Quit: Leaving]
20100411 01:04:03< icelus> Crab_, so i just wanted to look at this quickly and see if it was obvious now i know more about the sync stuff
20100411 01:04:12< icelus> Crab_, what do you think causes the OOS?
20100411 01:04:16< icelus> is it an event or what?
20100411 01:04:47< icelus> (the other sync bug in DF that you mentioned)
20100411 01:05:23< Crab_> I think that it's advancement event which is processed somewhat differently - e.g. the extra movepoints seem related to the difference of move points between advanced and non-advanced unit
20100411 01:06:55< icelus> really ,an advancement event
20100411 01:07:02< icelus> now i'm interested again =p
20100411 01:07:11< icelus> ok i'll download it and see if i can replicate =)
20100411 01:15:21-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100411 01:16:14-!- ilor_ [~ilor@wesnoth/developer/ilor] has quit [Ping timeout: 240 seconds]
20100411 01:16:32< icelus> Crab_, what sync error am i looking for
20100411 01:16:51< Crab_> when the ai spawns its units, look at movement points left
20100411 01:17:00< Crab_> there are some mismatches, for some units
20100411 01:17:09< Crab_> =0 on one side, and !=0 on another
20100411 01:17:36< icelus> so i see ai enemies
20100411 01:17:39< icelus> they moved toward me
20100411 01:17:42< icelus> i didn't move at al yet
20100411 01:17:44< YogiHH> night everyone
20100411 01:17:46< icelus> just passed every turn
20100411 01:17:49< Crab_> on the first turn when they spawn
20100411 01:18:01< icelus> well i'm under srhoud
20100411 01:18:03< Crab_> its turn 2 or 3, don't remember exactly
20100411 01:18:04< icelus> shroud sorry
20100411 01:18:16< icelus> yeah i'm on turn 4 now 
20100411 01:18:19< Crab_> icelus: that's why I've said 'recruit a unit with good vision range"
20100411 01:18:21< icelus> this was the first turn i saw them
20100411 01:18:22< icelus> oh
20100411 01:18:23< icelus> right
20100411 01:18:24< icelus> god
20100411 01:18:25< icelus> ok
20100411 01:18:34-!- YogiHH [YogiHH@wesnoth/developer/yogihh] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539]]
20100411 01:22:11< icelus> Crab_, ok I see the bug
20100411 01:22:22< icelus> i'll try to figure out why it happens
20100411 01:22:46< Crab_> icelus: good. note that this map is known for ugly WML.
20100411 01:23:03< Crab_> icelus: but, this might be traced down to engine bug, as well
20100411 01:23:11< icelus> it has to be an engine bug on some level
20100411 01:23:22< icelus> really hoping for anyone but software engineers to write synchronized code
20100411 01:23:27< icelus> is unrealistic
20100411 01:23:33< icelus> the engine should never go out of sync really
20100411 01:24:13< Crab_> icelus: yes, but I think it needs to be rewritten from scratch to support this
20100411 01:24:28< Espreon> What's this?
20100411 01:24:35< Crab_> icelus: so, in practice, there's documented cases of 'this is unsafe in MP'
20100411 01:24:48< icelus> but mostly you would expect to be able to solve i think...
20100411 01:24:53< icelus> at least i havent' seen anything yet
20100411 01:24:56< icelus> that tells me it's impossible
20100411 01:25:04< icelus> it just requires a lot of care
20100411 01:25:17< icelus> the trouble with rewrites
20100411 01:25:27< icelus> is you solve all the bugs you've got today
20100411 01:25:36< icelus> and introduce a whole new set you don't know as much about =p
20100411 01:25:57< Crab_> icelus: yes, but there's no 'formal approach' in current code which might be able to ensure that things work within certain limits wrt sync
20100411 01:26:25< icelus> Crab_, that's definitely true
20100411 01:26:26< Crab_> icelus: e.g., your introduced classification is not a first-class citizen in the code.
20100411 01:26:33< icelus> sure
20100411 01:26:39< icelus> but it can be made one probably
20100411 01:26:44-!- Blueblaze [~nick@adsl-99-182-52-75.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 240 seconds]
20100411 01:26:46< icelus> the real difficulty isn't fixing what's here
20100411 01:26:49< icelus> there's a finite amount of it
20100411 01:26:59< icelus> the real problem is making sure the next wml tag that's added
20100411 01:27:06< Crab_> icelus: exactly
20100411 01:27:06< icelus> is MP safe from the start
20100411 01:27:32< Crab_> and there's no automated regression testing for that
20100411 01:27:57< Crab_> (that might be able to help somewhat, by spotting some classes of bugs earlier)
20100411 01:28:10< icelus> yeah regression testing would definitely be the way to go
20100411 01:28:20< icelus> maybe you can get some enthusiast gsoc student to do it
20100411 01:28:21< Crab_> icelus: and, btw, note http://wiki.wesnoth.org/NotSoEasyCoding#add_a_mode_to_improve_OOS_detection_in_MP idea
20100411 01:28:55< icelus> i see that, but i don't think that's the way forward myself
20100411 01:29:01< Crab_> icelus: why ?
20100411 01:29:02< icelus> it's not a bad idea
20100411 01:29:19< icelus> i think it would be very hard to do
20100411 01:29:26< Crab_> icelus: why hard ?
20100411 01:29:26< icelus> so it didn't add exponential code overhead
20100411 01:29:38< icelus> well you need to know EVERYTHING about the game state that wml could possibly influence
20100411 01:29:42< icelus> you need to set up a diff system for it
20100411 01:29:51< icelus> you need to establish safe reliable transport
20100411 01:29:56< icelus> preferably out of band
20100411 01:30:02< icelus> so really open another port i would guess
20100411 01:30:03< Crab_> icelus: well, there's already code which needs to know (almost) everything about the game state
20100411 01:30:14< Crab_> icelus: it's "save game" code
20100411 01:30:28< icelus> sure but does save game contain everything that could go OOS
20100411 01:30:36< icelus> my suspicion is the real things that break
20100411 01:30:41< icelus> are the ones that wouldn't be in a save game
20100411 01:30:57< icelus> if you think about it
20100411 01:31:04< icelus> it's pretty remarkable that with so little code support
20100411 01:31:10< icelus> you can almost always get a reliable MP game
20100411 01:31:25< icelus> the execution paths that are followed have to be identical
20100411 01:31:32< icelus> and you don't have much help for ensuring that
20100411 01:31:40< icelus> but
20100411 01:31:45< icelus> it's also the case that this already works
20100411 01:31:57< icelus> OOS is a rarity and i think with a little bit of care for the edge cases
20100411 01:32:04< icelus> some vicious regression tests
20100411 01:32:10< icelus> you could get it working nicely
20100411 01:32:20< icelus> imo the safest way forward is a regression battery
20100411 01:32:49< icelus> one with a lot of the stuff you want to interact with stubbed
20100411 01:32:53< icelus> to be as unpleasant as possible
20100411 01:32:59< icelus> so you could have rngs 
20100411 01:33:09< icelus> that test whether you're allowed to call them at the moment
20100411 01:33:30< icelus> and any user input could be stubbed and the current context checked
20100411 01:37:10< Crab_> about 'the real things that break  are the ones that wouldn't be in a save game' : to detect an hard OOS, you need a situation where one client can do an action, but another cannot. but, if you load a same save game, the actions allowed will be the same. So, I think that most things which are not in the save will not cause an OOS even if they mismatch.
20100411 01:39:01< icelus> Crab_, well here's what i was thinking
20100411 01:39:15< icelus> let's say you have free reign to make the choice you wish
20100411 01:39:23< icelus> it's going to be put in the replay for you
20100411 01:39:35< icelus> in a moment of madness you call the synced rng to make your choice
20100411 01:39:44< icelus> this only shows up
20100411 01:39:51< icelus> on the NEXT thing that calls the synced rng
20100411 01:39:55< icelus> it might not show up at all
20100411 01:40:00< icelus> or for multiple turns
20100411 01:40:13< icelus> a regression test catches that immediately
20100411 01:40:20< icelus> because you are not allowed to call the synced rng
20100411 01:40:24< icelus> unless everyone is going to too
20100411 01:40:39< icelus> but a game wide diff tells you something happened
20100411 01:40:41< icelus> somewhere
20100411 01:40:43< icelus> you have no idea where
20100411 01:41:32< Crab_> yes, but if it's going to be close enough to the source, it can be traced down more easier
20100411 01:41:58< icelus> the thing about localised regression tests is they tell you before it breaks
20100411 01:42:10< icelus> and they tell you a very small section of code
20100411 01:42:14< icelus> which is implicated
20100411 01:42:16< Crab_> yes, regression is good, too
20100411 01:42:43< icelus> i don't say a game wide state diff is a bad idea
20100411 01:42:44< Crab_> but they need to have a 'cover' a particular small section of code
20100411 01:42:49< icelus> i just think it would be hard to do
20100411 01:43:06< icelus> and i think regression would in my opinion give more bang for the buck
20100411 01:43:13< icelus> but that's just my opinion
20100411 01:43:17< icelus> i don't have anything to back that up with
20100411 01:43:19< Crab_> well, those are different approaches, and they compliment each other
20100411 01:43:21< icelus> beyond the ideas i just said
20100411 01:43:54< Crab_> basically, either we trace down the 'effects' (test particular things) or trace down the 'results' (gamestate diff)
20100411 01:44:13< icelus> well a game state diff is a nice safety net
20100411 01:44:19< icelus> i just suspect that when you got to writing it
20100411 01:44:25< icelus> there would be unpleasantness in there
20100411 01:44:43< icelus> i don't know precisely what; if i knew what I would feel better about it somehow
20100411 01:44:51< icelus> but if this was my job and my boss just told me that plan
20100411 01:44:56< icelus> my heart would be sinking
20100411 01:46:30< Crab_> a hacky way to do that: after each move/attack/recruit/etc, do a a 'save game to memory, checksum it, send checksum as a special replay message, if checksums don't match, panic and OOS (plus, exchange game infos to make a diff).
20100411 01:47:36< Crab_> of course we'll probably need to filter the save, and there'll be other small things
20100411 01:48:14< icelus> Crab_, that could work
20100411 01:48:20< Crab_> but, do you know what the big argument against that idea was when I discussed it some time ago ? 
20100411 01:48:26< icelus> what was it?
20100411 01:48:54< Crab_> it was "we might need to make each wesnoth client know only the part of the full game info, the part which that client sees, to prevent cheating in MP"
20100411 01:49:30< Crab_> and that is somewhat incompatible with 'checksum entire game state or most interesting parts of it after each move'
20100411 01:49:48< icelus> i don't understand the issues well enough i don't think
20100411 01:50:00< icelus> with what i understand so far, that objection whilst valid principle
20100411 01:50:05< icelus> is not realistic in practice
20100411 01:50:25< icelus> because it is in effect a lock-step system
20100411 01:50:36< icelus> you would need a different network protocol i think to do that kind of trick
20100411 01:50:55< icelus> the way it works now is nice but it's symmetric
20100411 01:50:58-!- wesbot changed the topic of #wesnoth-dev to: GSoC: Thanks to all GSoC candidates for submitting proposals to Google. Accepted student applications announced on April 26. | 87 bugs, 264 feature requests, 16 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100411 01:51:01< Crab_> yes.
20100411 01:51:34-!- norbert_ [~norbert@82-171-70-54.ip.telfort.nl] has quit [Quit: Leaving]
20100411 01:51:45< Crab_> another note - currently wesnoth has no 'running' integration testing in place, so any work towards that will be a very good thing
20100411 01:52:10< Crab_> so, that 'create a regression suite for mp sync issues' would be a step in the right direction
20100411 01:52:11< icelus> how do you mean running integration?
20100411 01:52:15< icelus> oh i see
20100411 01:52:27< icelus> i wasn't volunteering to do a regression framework =)
20100411 01:52:41< icelus> too close to the horrors of my day job i think
20100411 01:53:00< icelus> i write optimization software
20100411 01:53:04< icelus> some multithreaded
20100411 01:53:04< Crab_> icelus: no, I'm about 'regression testing of my sync issues is a good thing, because it'll be a good step towards regression testing of wesnoth'
20100411 01:53:19< icelus> right ok
20100411 01:53:20< Crab_> icelus: I'm not about 'someone particular should do it'
20100411 01:53:25-!- schumi [~a11849c4@gateway/web/freenode/x-bnqzxikrpnnzpitb] has joined #wesnoth-dev
20100411 01:53:30< icelus> regression testing in general is a wonderful thing
20100411 01:53:38< icelus> you feel like you waste a vast amount of time setting it up
20100411 01:53:46< icelus> but then afterwards you can't imagine how you lived without it
20100411 01:53:50< icelus> at least that has been my experience
20100411 01:53:51< Crab_> icelus: yes :)
20100411 01:54:08< Crab_> icelus: the same. if you get it running, it'll become a part of life :)
20100411 01:54:30< icelus> but also debugging is so much easier
20100411 01:54:37< icelus> because you KNOW all the components you're working with
20100411 01:54:38< icelus> work
20100411 01:54:47< icelus> the bug is localised immediately to whatever you just changed
20100411 01:54:59< icelus> it also forces the idea of an interface spec onto the stuff you're working with
20100411 01:54:59< Crab_> icelus: not when there are upstream changes ;)
20100411 01:55:11< icelus> well if they pass your regression tests
20100411 01:55:14< icelus> they work
20100411 01:55:18< icelus> that's what i mean
20100411 01:55:22< icelus> so if you develop a new piece of code
20100411 01:55:32< icelus> and you only relied on behaviour that's in the "interface"
20100411 01:55:36< icelus> by which i mean is regression tested
20100411 01:55:38< icelus> then if it doesn't work
20100411 01:55:41< icelus> it's your fault
20100411 01:55:56< icelus> there's no question of hunting through header files or 20 other source files
20100411 01:56:11< Crab_> icelus: yes, that's good if you're developing something sufficiently independent from other code
20100411 01:56:14< icelus> to find out if everything you're interacting with is well-behaved
20100411 01:56:40< icelus> the harder you try for that kind of independence
20100411 01:56:45< icelus> the better it works out i think
20100411 01:56:52< icelus> i haven't written every kind of software
20100411 01:56:55< icelus> but i've written a few kinds
20100411 01:57:12< icelus> and always when i really tried to centre things around small cohesive units
20100411 01:57:13< icelus> life improved
20100411 01:57:24< icelus> once you let things tangle together
20100411 01:57:29< icelus> it takes 100x as much effort to untangle them
20100411 01:57:38< icelus> sometimes you get to a codebase
20100411 01:57:45< icelus> and they've been cheerfully tangling it up for years
20100411 01:57:54< icelus> and it's unrealistic to separate out the parts now
20100411 01:58:08< icelus> but then that kind of place
20100411 01:58:13< icelus> tends not to have regression tests either
20100411 01:58:17< icelus> "it's too big to test"
20100411 01:58:22< icelus> is a maxim i've heard before
20100411 01:59:46< Crab_> well, sometimes there are cases when you can untangle the code, but you're not it's 'owner' and cannot  (for political reasons) make such drastic changes, and so you are stick to making small changes on your own, and trying to make sure that *your* code works.
20100411 02:00:01< icelus> yeah i've been there before too Crab_ =)
20100411 02:00:20< icelus> but i found that when i was asked for new features
20100411 02:00:35< icelus> i could try to separate them out into parts that were cohesive
20100411 02:00:40< icelus> i often ended up with extra code
20100411 02:00:49< icelus> a sort of interface or buffer area
20100411 02:01:15< icelus> which dealt with mixing all my clean logic into the dozen different places it had to be used in the codebase
20100411 02:01:41< Crab_> yes, a good solution :)
20100411 02:02:29-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has quit [Remote host closed the connection]
20100411 02:13:12< Upth> Hrm. I just noticed a bug in 1.9.0-svn it is present in revision 42099 and I have verified that it is not due to a change I made
20100411 02:13:24< Crab_> hi, Upth
20100411 02:13:33< Crab_> what's the bug ?
20100411 02:13:35-!- Upthorn is now known as orn
20100411 02:13:35-!- Upth is now known as Upthorn
20100411 02:13:37-!- orn is now known as Upth
20100411 02:14:35< Upthorn> Well I have a savegame from the first scenario in heir to the throne, where one of my allies has a shaman sitting on a village
20100411 02:15:08< Crab_> ok. and what's wrong with it?
20100411 02:15:13< Upthorn> When that ally's turn is reached, the village healing is not displayed, but the shaman's HP updates.
20100411 02:15:30< Crab_> so, healing animation is absent ?
20100411 02:15:33< Upthorn> yes.
20100411 02:15:42< Crab_> boucman should know more about it
20100411 02:15:55< Crab_> as he's the person to ask about animation issues
20100411 02:16:06< Crab_> (if you want to fix it yourself, which is a good thing)
20100411 02:16:21< Crab_> (otherwise, submit a bug to gna)
20100411 02:18:54-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz]
20100411 02:19:41-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20100411 02:25:03-!- schumi [~a11849c4@gateway/web/freenode/x-bnqzxikrpnnzpitb] has quit [Quit: Page closed]
20100411 02:25:04-!- Chusslove [~Chusslove@brsg-d9bee23a.pool.mediaWays.net] has quit [Read error: Connection reset by peer]
20100411 02:26:23-!- schumi [~a11849c4@gateway/web/freenode/x-suanfymatbkgixzs] has joined #wesnoth-dev
20100411 02:27:10-!- Chusslove [~Chusslove@brsg-d9bee23a.pool.mediaWays.net] has joined #wesnoth-dev
20100411 02:30:57-!- Darkas [~quassel@ppp-88-217-107-190.dynamic.mnet-online.de] has quit [Remote host closed the connection]
20100411 02:32:26-!- pokhbocee [~pokhbocee@ws40.cs.drexel.edu] has quit [Quit: Ex-Chat]
20100411 02:34:23-!- deekay [~dk@wesnoth/developer/dragonking] has quit [Quit: deekay]
20100411 02:47:41-!- pokhbocee [~pokhbocee@ws60.cs.drexel.edu] has joined #wesnoth-dev
20100411 02:52:11-!- Appleman1234 [~Appleman1@CPE-124-191-176-143.oxqn1.cha.bigpond.net.au] has joined #wesnoth-dev
20100411 02:53:26-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: GO, GET TO THE CHOPPAH!!!]
20100411 03:00:28-!- fkhodkov [~fedor76@ppp-78-24-25-45-bras0.istra.ru] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)]
20100411 03:12:50-!- chains [~Rylar@adsl-75-37-46-168.dsl.pltn13.sbcglobal.net] has joined #wesnoth-dev
20100411 03:16:44-!- apoi [~andi@85-126-180-242.volume.xdsl-line.inode.at] has quit [Ping timeout: 240 seconds]
20100411 03:18:38-!- apoi [~andi@85-126-180-242.volume.xdsl-line.inode.at] has joined #wesnoth-dev
20100411 03:18:47-!- schumi [~a11849c4@gateway/web/freenode/x-suanfymatbkgixzs] has quit [Quit: Page closed]
20100411 03:22:57< Crab_> Achilles, Dexter,  aranair ,  awilkins ,  billynux ,  Darkas ,  endercoaster ,  fakedrake ,  gabba ,  Greywhind ,  JustasJ ,  kahowell,  lfernando ,  LucaMoller ,  orfest ,  pokhbocee ,  Sannya ,  schumi ,  Shuger ,  timotei, timotei21 ,  Truongan ,  Upthorn, Upth : I've commented on the gsoc proposals, naming the most important points to improve. good luck.
20100411 03:24:25-!- apoi [~andi@85-126-180-242.volume.xdsl-line.inode.at] has quit [Ping timeout: 252 seconds]
20100411 03:24:31< nagbot> /me good night, Crab_
20100411 03:24:33 * Crab_ thanks nagbot
20100411 03:24:38-!- Crab_ [~Crab_@wesnoth/developer/crab] has quit [Quit: Leaving.]
20100411 03:25:02-!- apoi [~andi@85-126-180-242.volume.xdsl-line.inode.at] has joined #wesnoth-dev
20100411 03:25:36< Espreon> Meh, you children and your nagbot.
20100411 03:26:34< Upthorn> I believe that M should be an H :p
20100411 03:26:54< Espreon> Nope.
20100411 03:33:21< Espreon> ... if anything the 'th' in your name should be a 'þ'... LOL.
20100411 03:33:54< Upthorn> if IRC would support it, it would be.
20100411 03:34:09< Espreon> No, it's just your client.
20100411 03:34:12< Espreon> Oh.
20100411 03:34:30< Espreon> Yeah, I wish that thorn could be in names...
20100411 03:34:44< Espreon> This sux...
20100411 03:35:46< Upthorn> Gah. animation "bug" I encountered was a result of breakpoints screwing up the time.
20100411 03:38:27< Upthorn> a breakpoint I had in calculate_healing was causing the animations to become added with outdated start_tick_ values
20100411 03:39:34< Upthorn> start_tick_ values outdated by more than the length of the total animation, in fact.
20100411 03:44:20-!- lfernando [~luiz@201-43-134-54.dsl.telesp.net.br] has joined #wesnoth-dev
20100411 04:06:04-!- Appleman1234 [~Appleman1@CPE-124-191-176-143.oxqn1.cha.bigpond.net.au] has quit [Quit: Leaving]
20100411 04:18:44-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100411 04:23:06-!- Blueblaze [~nick@adsl-99-182-52-75.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100411 04:30:43-!- pokhbocee [~pokhbocee@ws60.cs.drexel.edu] has quit [Remote host closed the connection]
20100411 04:44:34-!- LucaMoller [~a1184af9@gateway/web/freenode/x-izkpwqyzwfuczbez] has joined #wesnoth-dev
20100411 04:48:25-!- LucaMoller [~a1184af9@gateway/web/freenode/x-izkpwqyzwfuczbez] has quit [Client Quit]
20100411 04:51:43-!- lfernando [~luiz@201-43-134-54.dsl.telesp.net.br] has quit [Quit: Ex-Chat]
20100411 04:59:28< AI0867> icelus: another issue with checksumming a save: there's plenty of WML that can be changed without causing OOS (UI stuff), you'd have to strip that out before doing the checksum
20100411 05:01:13< icelus> AI0867, i don't think i personally am in any danger of implementing such a thing
20100411 05:01:34< icelus> i can believe it is possible and would be useful
20100411 05:01:43< icelus> but like i said in the log i think it'd be hard
20100411 05:01:49< icelus> because of exactly those kinds of edge cases
20100411 05:02:03< icelus> it sets off warning bells in my head that it'd be a big project to take on
20100411 05:02:11< icelus> and an entirely unenjoyable one
20100411 05:02:14< icelus> for me at least =)
20100411 05:02:27-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Quit: asdf]
20100411 05:03:34< AI0867> I was going to direct that comment at crab, but my tab-completer said he left
20100411 05:04:14< Upthorn> icelus: there is a point of confusion that I have which perhaps you could clear up for me. Earlier I saw boucman say that you have a WML persistence proposal, but I cannot find one on the wiki. Was his statement incorrect, or did you not make a wiki page from the proposal template?
20100411 05:04:26< AI0867> it gets worse in that actionWML *can* access this "supposedly only for UI" stuff..
20100411 05:04:52< icelus> Upthorn, i'm not totally sure what boucman meant by a wml persistence mechanism
20100411 05:04:59< icelus> i definitely haven't made a wiki page
20100411 05:05:17< AI0867> I'm not sure if it has even been defined yet
20100411 05:05:42< icelus> the stuff i'm working on is in this patch: https://gna.org/patch/?1610 and the related bug: https://gna.org/bugs/?15560
20100411 05:05:44< AI0867> other than "some way to carry data from one campaign/mp scenario to a different one"
20100411 05:05:55< AI0867> or between multiple instances
20100411 05:06:08< icelus> so i'm trying to make the wml synchronization model sane
20100411 05:06:11< Upthorn> icelus: so you are not a prospective Summer of Code student.
20100411 05:06:17< icelus> Upthorn, no =)
20100411 05:06:42< Upthorn> Okay. Boucman's statement indicated that you were competing with me on http://wiki.wesnoth.org/SoC_Ideas_Persistent_Gameworld
20100411 05:06:49< Upthorn> I am glad that this is not the case.
20100411 05:07:06< icelus> competing? =p
20100411 05:07:14< icelus> i do this for fun
20100411 05:07:21< icelus> as a break from linear optimization codes
20100411 05:09:39< Upthorn> heh. I normally treat this kind of thing as a hobby, too
20100411 05:12:10< Upthorn> But no more than one student will be accepted for each project to google's paid summer internship program, and I wish to participate. So I am glad that you are not also hoping to participate on my same project. :)
20100411 05:12:31< Upthorn> (Since that would likely mean that I would not get to)
20100411 05:16:48< icelus> i'm sure that's not true
20100411 05:17:02< icelus> but anyway good luck
20100411 05:17:28-!- Gambit [~Gambit@pa-67-234-73-7.dhcp.embarqhsd.net] has quit [Quit: Leaving.]
20100411 05:25:52-!- ancestral [~ancestral@97-116-115-135.mpls.qwest.net] has joined #wesnoth-dev
20100411 05:42:35-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Quit: crimson_penguin]
20100411 05:44:19-!- icelus [~ed@cpc9-sgyl11-0-0-cust29.sgyl.cable.virginmedia.com] has quit [Quit: Leaving]
20100411 06:34:13-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100411 06:40:46-!- ancestral [~ancestral@97-116-115-135.mpls.qwest.net] has quit [Quit: And that’s the end of THAT chapter.]
20100411 07:04:34-!- faryshta [~faryshta@201.124.54.8] has joined #wesnoth-dev
20100411 07:08:00-!- ancestral [~ancestral@97-116-115-135.mpls.qwest.net] has joined #wesnoth-dev
20100411 07:08:58-!- fendrin [~fabi@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20100411 07:13:03-!- ancestral is now known as ancestral_
20100411 07:13:23-!- ancestral_ is now known as ancestral
20100411 07:25:11-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20100411 07:25:21< mordante> servus
20100411 07:25:33< Espreon> Ic grete þe.
20100411 07:25:38< mordante> hi Espreon 
20100411 07:30:19-!- ancestral [~ancestral@97-116-115-135.mpls.qwest.net] has quit [Quit: ancestral]
20100411 07:31:21-!- ancestral [~ancestral@97-116-115-135.mpls.qwest.net] has joined #wesnoth-dev
20100411 07:35:07< CIA-32> mordante * r42100 /trunk/src/ (10 files): 
20100411 07:35:07< CIA-32> Readd the tportrait class.
20100411 07:35:07< CIA-32> The code is used in a locally uncommitted branch, reverts r41847.
20100411 07:37:13-!- chains [~Rylar@adsl-75-37-46-168.dsl.pltn13.sbcglobal.net] has quit [Ping timeout: 248 seconds]
20100411 07:44:58-!- happygrue [~George@wesnoth/developer/wintermute] has quit [Read error: Connection reset by peer]
20100411 07:45:06-!- happygrue [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20100411 07:45:09-!- dtiger [~dtiger@dynamic-vpdn-93-125-17-84.telecom.by] has joined #wesnoth-dev
20100411 07:50:58-!- wesbot changed the topic of #wesnoth-dev to: GSoC: Thanks to all GSoC candidates for submitting proposals to Google. Accepted student applications announced on April 26. | 87 bugs, 264 feature requests, 17 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100411 08:14:50-!- stikonas [~and@ctv-79-132-179-139.vinita.lt] has joined #wesnoth-dev
20100411 08:14:50-!- stikonas [~and@ctv-79-132-179-139.vinita.lt] has quit [Changing host]
20100411 08:14:50-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100411 08:22:14-!- FAAB [~huajie@219.142.239.171] has joined #wesnoth-dev
20100411 08:23:36-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 276 seconds]
20100411 08:34:01-!- Espreon [~espreon@wesnoth/developer/espreon] has quit [Quit: WRYYYYYYYYYYYYYYYYYYYY!]
20100411 08:39:44< Upthorn> boucman, crab_: If you read the logs, I have read your comments on my proposal and responded to them.
20100411 08:41:54-!- Appleman1234 [~Appleman1@CPE-124-191-176-143.oxqn1.cha.bigpond.net.au] has joined #wesnoth-dev
20100411 08:43:30-!- FAAB [~huajie@219.142.239.171] has left #wesnoth-dev []
20100411 08:45:05-!- ilor_ [~ilor@wesnoth/developer/ilor] has joined #wesnoth-dev
20100411 08:49:45-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20100411 08:59:55-!- EdB [~EdB@tss37-1-89-84-18-220.dsl.club-internet.fr] has joined #wesnoth-dev
20100411 09:07:20< aranair> soliton: heys, you happen to be around for a review of my proposal?
20100411 09:11:43-!- mordocai [~mordocai@66.119.9.243] has joined #wesnoth-dev
20100411 09:13:52< zookeeper> Upthorn, regarding your allied shaman's healing not showing; are you sure you don't just have the "skip AI moves" option toggled on?
20100411 09:14:22< zookeeper> it'd sound logical that allies' healed animations wouldn't play if it's on
20100411 09:15:13< Upthorn> zookeeper: it wasn't that, but it wasn't a bug.
20100411 09:15:27< Upthorn> I had a breakpoint triggering during calculate_healing
20100411 09:15:38< Upthorn> which caused the animations to start with stale timestamps
20100411 09:15:45< zookeeper> oh, right, i missed that bit in the log
20100411 09:16:03< Upthorn> so when they started to play, their duration had already elapsed and they were skipped
20100411 09:17:17< Upthorn> this is the first bug I've encountered that only happens when the application is being debugged, heh.
20100411 09:18:57< Upthorn> It's so much better than the more common bugs that only go away when the application is being debugged.
20100411 09:21:13< zookeeper> i bet
20100411 09:25:24-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20100411 09:34:37-!- faryshta [~faryshta@201.124.54.8] has quit [Quit: Saliendo]
20100411 09:46:08-!- ancestral [~ancestral@97-116-115-135.mpls.qwest.net] has quit [Quit: And that’s the end of THAT chapter.]
20100411 09:55:45-!- EdB [~EdB@tss37-1-89-84-18-220.dsl.club-internet.fr] has quit [Remote host closed the connection]
20100411 09:58:51-!- qemqemqem [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has quit [Ping timeout: 245 seconds]
20100411 10:01:12-!- alix [alix@ruination.org.uk] has left #wesnoth-dev []
20100411 10:02:32-!- Blueblaze [~nick@adsl-99-182-52-75.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 276 seconds]
20100411 10:05:38-!- happygrue [~George@wesnoth/developer/wintermute] has quit [Read error: Connection reset by peer]
20100411 10:05:51-!- happygrue [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20100411 10:15:04-!- aranair [~boa.homan@cm212.epsilon34.maxonline.com.sg] has quit [Read error: Connection reset by peer]
20100411 10:20:57-!- aranair [~boa.homan@cm212.epsilon34.maxonline.com.sg] has joined #wesnoth-dev
20100411 10:55:17< Soliton> nagbot: aranair?
20100411 10:55:19< nagbot> aranair : http://wiki.wesnoth.org//SummerofCode_Aranair : http://socghop.appspot.com/gsoc/student_proposal/review/google/gsoc2010/aranair/t127067138706
20100411 10:59:28< Soliton> aranair: i read your proposal. it doesn't sound much like standard wesnoth with the different stati and boni and whatnot, more like a special scenario.
20100411 11:04:16-!- mordocai [~mordocai@66.119.9.243] has quit [Quit: Leaving]
20100411 11:04:24< Soliton> i know that's not particularly useful feedback but i'm not that interested in thinking up my own idea of an alliance system right now.
20100411 11:05:35-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Ping timeout: 276 seconds]
20100411 11:09:49-!- loonybot [~loonybot@ppp79-139-136-187.pppoe.spdop.ru] has joined #wesnoth-dev
20100411 11:09:49-!- loonybot [~loonybot@ppp79-139-136-187.pppoe.spdop.ru] has quit [Changing host]
20100411 11:09:49-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20100411 11:10:43-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100411 11:13:47< Ivanovic> moin
20100411 11:14:53< aranair> soliton: mm what do you mean by a special scenario?
20100411 11:16:34< zookeeper> means something used only in a few specific scenarios, not by default
20100411 11:18:27< aranair> zookeeper: I do not understand why it seems to be that way really mm
20100411 11:19:12< aranair> zookeeper: the idea is supposed to cover scenarios, scenarios that the current system covers, as well as new scenarios where new parameters are added
20100411 11:20:31< aranair> zookeeper: if you don't mind  could you do a quick review as well ?(on the wiki version)
20100411 11:21:09< zookeeper> aranair, well, i'm pretty sure changing the default team MP into something that complex is simply not wanted
20100411 11:21:29< zookeeper> for a couple of special scenarios, fine
20100411 11:22:01< zookeeper> but it's not very KISS if the players by default would have access to an alliance system that complicated
20100411 11:22:54< zookeeper> you'd end up spending as much time thinking about the details of your alliances as you would about the actual unit movement and such
20100411 11:23:37< aranair> the original idea of the new alliance system was to have a asymmetric privlege granting with all sides, and i thought it would be even harder to manage
20100411 11:24:00< aranair> as well, it would be possible to emulate the original system even if this system is in place, mm
20100411 11:27:32< aranair> my proposal is, very much, to simplify the number of things that the player has to consider even after you add those new parameters listed by the devs.
20100411 11:30:57-!- deekay [~dk@wesnoth/developer/dragonking] has joined #wesnoth-dev
20100411 11:31:57< zookeeper> well, to me it in fact seems awfully complicated. extra "contributions" from villages due to an alliance, the leader getting more than the rest, all those alliance statuses, healing amounts depending on them..
20100411 11:33:13< zookeeper> sounds like that'd immediately lead to the alliance system becoming a very important mini-game of who manages to optimize their alliance settings to get the most benefits from it
20100411 11:34:44< zookeeper> all in all, frankly i just think it seems very complicated for an alliance system
20100411 11:35:06< aranair> zookeeper: but if the new parameters proposed by the devs were to be added, complexity will be increased no matter what no?
20100411 11:37:26-!- whiz_ [~work@adsl-70-241-66-181.dsl.hstntx.swbell.net] has joined #wesnoth-dev
20100411 11:38:45< zookeeper> aranair, i suppose
20100411 11:40:25< zookeeper> aranair, but note that i
20100411 11:40:44< zookeeper> that i haven't vouched for the idea to begin with, so..
20100411 11:40:49-!- whiz_ [~work@adsl-70-241-66-181.dsl.hstntx.swbell.net] has quit [Remote host closed the connection]
20100411 11:40:51< aranair> zookeeper: just curious though: do you lean towards a system with free control of privileges N-ways for N-sides in a game?
20100411 11:41:06< aranair> ah. you mean for a new alliance system?
20100411 11:41:40< zookeeper> yeah
20100411 11:41:43< aranair> i kinda get where you are coming from now :/ i mean certain levels of complexity will be added if new parameters are added;
20100411 11:41:57< aranair> although i always considered these complexities like game play variety
20100411 11:42:14-!- whiz_ [~work@adsl-70-241-66-181.dsl.hstntx.swbell.net] has joined #wesnoth-dev
20100411 11:42:49< zookeeper> and no, i don't like the ability to set the priviledge details. i think it'd be good to have the functionality so 1) a scenario could set what info/benefits are shared between team members and 2) they could be toggled in the game creation screen (under some advanced tab)
20100411 11:43:31< whiz_> hi all. does anyone have a compiled package for libsdl 1.2-14.3/4 patched which is supposed to handle the mouse click problem 
20100411 11:44:16< whiz_> i am looking for i386 ubuntu package. not sure if I want to patch and compile myself 
20100411 11:44:31< aranair> actually that would be simply how this new system to exposed to wml right? specific toggling of each individual evaluation could be given to the creater mm
20100411 11:44:33< whiz_> and dont want to play wesnoth in full screen mode either. just cant do much of other activity 
20100411 11:44:53< zookeeper> aranair, personally, i've been thinking about a very simple alliance system myself (and actually i've already partly written it in WML): basically just allowing sides to join or leave teams, and for teams to vote on new members or kicking a member, and so on.
20100411 11:46:02< zookeeper> aranair, i don't think anyone would really oppose the optional ability for both a scenario to set which info/benefits (vision, healing, leadership, etc) are shared and also for the host to set those when creating a game
20100411 11:46:57< aranair> ah, would there be additional consequences in your model? like after joining a team for example
20100411 11:47:34< zookeeper> mmh, not really, no
20100411 11:47:43< aranair> nods that is true zookeeper, but i was worried about allowing too much control (too many options)
20100411 11:47:52< aranair> i see
20100411 11:49:02< zookeeper> well, your proposal includes all the same options but they're just available to all players in-game, instead of to the scenario author or the game host :P
20100411 11:50:09< zookeeper> aranair, the system i've been thinking about is basically really simple. i'll write a quick example...
20100411 11:51:12< aranair> mm yeah, in my model, it is either allnewparameters in at one go, or no-newparameters at all in a single game.
20100411 11:51:56< aranair> but i think having a flag for each individual new parameter that allows host to decide which is in, could be coded(im not too sure now ;P)
20100411 11:52:11< aranair> ok im all ears:
20100411 11:53:26< zookeeper> let's say it's a 5 player free-for-all. you're player 1, no alliances yet. if you want to start an alliance, you can send a proposal to side 2 asking if he wants to form an alliance with you. he sees the proposal on his next turn start, and accepts. you'll become allies (either immediately or on your next turn start). either one of you can then disband the alliance, since there's just two members in it.
20100411 11:53:28< zookeeper> then, if you want to invite side 3 into your team, you'll make a "team proposal" to side 3, meaning that you (as side 1) make the proposal, and then side 2 (your ally) also votes on it. if he agrees on the proposal, then it's sent to side 3, who can again either accept or reject.
20100411 11:54:01< zookeeper> if side 4 wants to join your team, then he can send a proposal to your team, and all your team members will get a vote on their turns on whether to accept or not.
20100411 11:55:24< zookeeper> there's no "alliance leader" or such, and all decisions affecting the entire team will be voted on
20100411 11:55:37< aranair> mm nods i get the rough idea. basically there is a team-concept too, but there is no additional consequences
20100411 11:55:37< aranair> nods
20100411 11:55:41< zookeeper> yeah
20100411 11:56:21< zookeeper> frankly i think that's a bit more like what'd suit wesnoth
20100411 11:56:33< aranair> i like the voting idea, i was still seeking for a way (in my model) for members to progress upwards.
20100411 11:56:36< aranair> mmm
20100411 11:56:57< boucman> zookeeper: what I like about my proposal is the disymetry/negociation aspect... it has too many parameters, and we probably will have to regroup some of them to simplify, but the disymetry is the part I like...
20100411 11:56:58< zookeeper> actually i can pastebin you the whole code, in case you want to try it out...
20100411 11:57:37-!- Johannes13 [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20100411 11:58:00< zookeeper> boucman, well, i might like that aspect too, but as said, more like in a special scenario and not something we could have on by default
20100411 11:58:11< boucman> ok...
20100411 11:58:13-!- Johannes13 [~Johannes@unaffiliated/johannes13] has quit [Read error: Connection reset by peer]
20100411 11:58:13< aranair> zookeeper: i haven't had the chance to compile it yet :P
20100411 11:58:13-!- Johannes13_ [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20100411 11:58:35< boucman> aranair: how come ?
20100411 11:58:47-!- Johannes13_ is now known as Johannes13
20100411 11:58:53< aranair> boucman: its actually about 16 days from my exam hehe ;P
20100411 11:59:00< boucman> k
20100411 11:59:21< zookeeper> aranair, well, this should work with 1.8 or any other recent version: http://wesnoth.pastebin.com/0vN0q20J
20100411 11:59:57< aranair> zookeeper: i thought wesnoth wasn't looking for a new default alliance system that is ALWAYS there?
20100411 12:00:00< zookeeper> you can just put it in add-ons/Diplomacy_Era/_main.cfg for example and the era should then appear in the MP era selection
20100411 12:00:17< aranair> zookeeper: if the old system could be simulated by toggling, it would be enough no?
20100411 12:00:22< zookeeper> aranair, well, i'm not saying that "my system" would be on by default either
20100411 12:00:40< zookeeper> aranair, maybe
20100411 12:01:04< zookeeper> anyway, i didn't really intend to remain here talking about this for quite this long, so i gotta be going for at least a while now.. ->
20100411 12:01:13< aranair> zookeeper, nods haha we stand extremely contrasted ;P
20100411 12:01:18< aranair> oopps
20100411 12:01:23< aranair> sorry and thanks anyways (:
20100411 12:01:31< aranair> cya around
20100411 12:01:37< zookeeper> (oh, and the above code has some bugs and so on, so no need to report them, i just figured you might want to look at how it works in-game, that's all)
20100411 12:02:00< aranair> yup it would be of help thanks~
20100411 12:04:22-!- timotei [~timotei@193.34.191.4] has joined #wesnoth-dev
20100411 12:04:54< aranair> boucman: i haven't had the chance to really talk to a MP dev yet heh, soliton did a quick review on it earlier though, mm im not sure if i get him totally but it seems he feels that it is of a special scenario as well heh~
20100411 12:05:37< boucman> aranair: your proposal is a huge change to gameplay, mine (despite the appearence) isn't that big...
20100411 12:05:59< boucman> so impact are huge and the content needs to be particularly clear...
20100411 12:07:17< aranair> boucman: i have the impression that the full control might have a similar impact mm
20100411 12:07:50< aranair> boucman: actually it all depends on what is the direction wesnoth is heading to right now :/
20100411 12:08:00< boucman> well, mine is "splitting the current system" into pieces, that's why it's less frightening
20100411 12:08:07< aranair> hehe
20100411 12:09:17< aranair> i think im going to get huge resistance on my proposal :/
20100411 12:10:42< aranair> i was always thinking of more variety and aspects i could ADD into this system mm
20100411 12:16:53< Upthorn> it might be okay to do that, so long as you nest it in the UI, and it doesn't adversely affect performance
20100411 12:17:57< Upthorn> the main problem with adding too many options into the piece-by-piece alliance system is that when a user wants to ally (or something) with another user
20100411 12:18:06< Upthorn> they'll get overwhelmed with options
20100411 12:18:25< boucman> indeed, 
20100411 12:18:26< aranair> mm it is true, that'll boil down to what is the "accepted" number of considerations
20100411 12:18:42< Upthorn> So you want to boil things down to maybe like four settings on the main screen
20100411 12:18:43< boucman> that's the sort of problems where I expect student proposals to differentiate :P
20100411 12:18:54< aranair> oh ;P
20100411 12:18:58< Upthorn> and then for each of those you could maybe have an "advanced" which deals with the pieces
20100411 12:19:14< aranair> actually.
20100411 12:19:38< aranair> that is a good idea upthorn mm
20100411 12:19:39< Upthorn> This is how you can balance functionality with intuitivity.
20100411 12:19:45< aranair> at least it COULD give you the level of details when you want it
20100411 12:19:52< aranair> and still have a default level of detail that is not too overwhelming
20100411 12:19:54 * Upthorn has put a lot of thought into this due to his arguements with apple users
20100411 12:19:59< aranair> LOL.
20100411 12:20:06< boucman> hehe
20100411 12:21:18< aranair> boucman: has there been MP devs who are pretty ok with adding new parameters?
20100411 12:21:55< boucman> well, when I discussed my idea on the ML a couple of weeks ago there were no loud protest, which is usually a good sign :P
20100411 12:22:59< aranair> ok, i think it would be easier to talk to the mp devs who are more ok with this first hehe
20100411 12:23:11< boucman> :P
20100411 12:24:19< aranair> i need an honest opinion of the proposal though, do you think it is too complex of now?
20100411 12:25:21< aranair> it isn't difficult to scale down the detail of consequences affected by the statuses to reduce complexity i think?;P
20100411 12:25:27< boucman> complex is not the main problem, it's too "invasive" of the normal gameplay, and it's hard to explain simply, which means players will have lots of trouble understanding it
20100411 12:25:53< boucman> basically, it will break players habits, and they tend not to like that
20100411 12:25:58< aranair> ah.
20100411 12:27:22-!- Vetinari [~lukjad007@unaffiliated/lukjad007] has quit [Ping timeout: 264 seconds]
20100411 12:27:42-!- Vetinari [~lukjad007@unaffiliated/lukjad007] has joined #wesnoth-dev
20100411 12:27:56< aranair> dang that is one brick there heh
20100411 12:29:46-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100411 12:30:15< aranair> ah, hey noy. 
20100411 12:30:21< noy> hello
20100411 12:30:47< aranair> noy: do you happen to have any time to review a proposal on a new alliance system?
20100411 12:31:08< noy> I'm back from a night of drinking with my rugby team
20100411 12:31:12< noy> sooo
20100411 12:31:17< aranair> oh ;)
20100411 12:31:27< noy> probably not the best of time to ask
20100411 12:31:31< boucman> noy: then this is the perfect moment
20100411 12:31:33< aranair> nah its fine. i'll catch you the next time ;P
20100411 12:31:35< noy> link me
20100411 12:31:35< aranair> LOL
20100411 12:31:37< aranair> http://wiki.wesnoth.org/SummerofCode_Aranair
20100411 12:31:41< boucman> you should play a few ladder games too :P
20100411 12:31:46< aranair> hehe
20100411 12:31:50< noy> my name is not jb
20100411 12:31:52< noy> its noy
20100411 12:33:31-!- whiz_ [~work@adsl-70-241-66-181.dsl.hstntx.swbell.net] has quit [Remote host closed the connection]
20100411 12:34:33< noy> aranair: to be honest I'm really tired and your proposal is long...
20100411 12:34:41< noy> I think I'll do it tomorrow if I have time
20100411 12:34:44< noy> sorry
20100411 12:34:48< aranair> noy: its ok ;)
20100411 12:35:07< aranair> noy: i'll catch you tomorrow then, thanks~
20100411 12:37:54-!- noy [~Noy@wesnoth/developer/noy] has left #wesnoth-dev []
20100411 12:47:16-!- fendrin [~fabi@88-134-14-205-dynip.superkabel.de] has joined #wesnoth-dev
20100411 12:47:16-!- fendrin [~fabi@88-134-14-205-dynip.superkabel.de] has quit [Changing host]
20100411 12:47:16-!- fendrin [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20100411 12:52:33-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100411 12:52:39< CIA-32> timotei * r42101 /trunk/utils/java/ (253 files in 72 dirs): 
20100411 12:52:39< CIA-32> - added the 3rd party library used by the plugin editor (for download see LIBRARIES.txt)
20100411 12:52:39< CIA-32> - commented out the "marking feature"
20100411 12:54:55< timotei> hello fendrin 
20100411 12:55:27< fendrin> hi timotei
20100411 12:55:35< timotei> fendrin, do you have some time to update the plugin just to see that everything works as desired? the plugin starts/compiles
20100411 12:56:02< timotei> fendrin, also, I'm available 2-2 and a half more hours
20100411 12:56:05< fendrin> timotei: Sorry, not now.
20100411 12:56:13< fendrin> Maybe in an hour.
20100411 12:56:15< timotei> ok
20100411 12:56:31< timotei> we'll talk later than
20100411 12:56:35< timotei> s/than/then
20100411 13:01:34-!- Darkas [~quassel@ppp-93-104-159-193.dynamic.mnet-online.de] has joined #wesnoth-dev
20100411 13:09:20-!- Appleman1234 [~Appleman1@CPE-124-191-176-143.oxqn1.cha.bigpond.net.au] has quit [Read error: Connection reset by peer]
20100411 13:14:29< boucman> timotei: what exactly is the external library you just added, is it GPL ? can we include it like that ?
20100411 13:14:46< boucman> (i'm not sure what distribution means you have thought of with fendrin and Crab_)
20100411 13:16:19-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20100411 13:21:23-!- aranair [~boa.homan@cm212.epsilon34.maxonline.com.sg] has quit [Read error: Connection reset by peer]
20100411 13:27:10< timotei> well, yes it's free
20100411 13:27:14< timotei> afair
20100411 13:27:56< timotei> boucman, yes: The Xtext framework is 100% open source and available under the Eclipse Public License. 
20100411 13:28:15< timotei> should I add the LICENSE.txt or something like that?
20100411 13:28:26< timotei> xtext is a plugin through eclipse update manager
20100411 13:28:34-!- aranair [~boa.homan@cm212.epsilon34.maxonline.com.sg] has joined #wesnoth-dev
20100411 13:29:56< timotei> hello Crab_ 
20100411 13:30:05< Crab_> hi, timotei
20100411 13:30:10< boucman> well, tbh we tend not to bundle our dependencies with the game, so having the dependencies with the eclipse plugin ranged a bell in my head
20100411 13:30:30< timotei> oh... well, that project aren't dependencies
20100411 13:30:36< timotei> but adjacent plugin's projects
20100411 13:30:46< boucman> then why did you add them ?
20100411 13:31:08< timotei> well, there's were a part of the eclipse plugin will take part
20100411 13:31:19< timotei> autocompletition, syntax highlithing, all editor things
20100411 13:32:01< timotei> when we'll deploy the plugin, I will try to include everything in one plugin
20100411 13:32:10< timotei> those adjacent projects.
20100411 13:32:15< timotei> isn't that ok?
20100411 13:32:21< Crab_> timotei: this is ok
20100411 13:32:31< Crab_> timotei: but note the difference between 'deploy time' and 'develop time'
20100411 13:32:48< timotei> yes
20100411 13:33:11< timotei> deploy is, when we will supply it to our users, isn't it?
20100411 13:33:13< Crab_> timotei: e.g., if those projects are fully autogenerated, there's no need to include them; if they contain changes from you, then we'll  need to include them in svn
20100411 13:33:15< Crab_> yes
20100411 13:33:36< timotei> well, they *will* contain my changes
20100411 13:33:44< Crab_> ok,then
20100411 13:33:50< Crab_> what's the total size ?
20100411 13:34:18< timotei> of the sources? or the resulting jars?
20100411 13:34:25< Crab_> of the sources
20100411 13:34:36< timotei> 1.6M the org.wesnoth.wml.*
20100411 13:34:50< timotei> 1.8M with the eclipse_plugin project
20100411 13:34:53< Crab_> what about the size of 'everything you put into svn?'
20100411 13:35:20< timotei> 1.8M is the total
20100411 13:35:31< Crab_> ok, I see no problems with that. good.
20100411 13:36:39< timotei> brb 5 mins
20100411 13:44:31< timotei> so, as guideliness: no 'mine-edited' dependencies uploaded on the svn
20100411 13:44:42< timotei> no .exe files
20100411 13:45:00< timotei> what about the size? what's the maximum size of something like that?
20100411 13:50:58-!- wesbot changed the topic of #wesnoth-dev to: GSoC: Thanks to all GSoC candidates for submitting proposals to Google. Accepted student applications announced on April 26. | 88 bugs, 264 feature requests, 17 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100411 13:52:38-!- Vetinari [~lukjad007@unaffiliated/lukjad007] has quit [Ping timeout: 276 seconds]
20100411 14:03:22-!- Tesafilmchen [~quassel@p5B27731F.dip.t-dialin.net] has joined #wesnoth-dev
20100411 14:08:35-!- timotei is now known as timotei_away
20100411 14:13:09-!- timotei_away is now known as timotei
20100411 14:16:24< Crab_> timotei: well, total wesnoth svn checkout is about 300mb
20100411 14:17:10< timotei> well, I was refering to the thing: shouldn't we upload "things" that are bigger than a certain size?
20100411 14:17:28-!- lukjad86 [~lukjadOO7@unaffiliated/lukjad007] has quit [Ping timeout: 258 seconds]
20100411 14:17:35< Crab_> I don't think that there is maximum size
20100411 14:17:53< timotei> ok
20100411 14:18:14< Crab_> e.g., you should upload what you need, if it's reasonably small. e.g. 1 mb is ok, even 10 mb might be ok. but 100 mb is too much ;)
20100411 14:18:20< timotei> yes
20100411 14:18:20< timotei> :D
20100411 14:18:48< Crab_> note that we can use maven or ivy dependency management to avoid bundling jar files with the projects
20100411 14:19:05< Crab_> if the size of dependency jar files will become too bug
20100411 14:19:21< Crab_> e.g., if it becomes something like "2mb of source and 70mb of libs"
20100411 14:20:02< timotei> yes
20100411 14:20:08< timotei> I've used maven before
20100411 14:20:30< timotei> the project I've worked on, the lineage 2 server, used at first ANT
20100411 14:20:34< timotei> but switched then to maven
20100411 14:20:56-!- lukjad86 [~lukjadOO7@unaffiliated/lukjad007] has joined #wesnoth-dev
20100411 14:21:15< Upthorn> oh hey Crab_ did you see the patch I submitted?
20100411 14:21:40< Crab_> yes, I've noticed it
20100411 14:22:40-!- Darkas [~quassel@ppp-93-104-159-193.dynamic.mnet-online.de] has quit [Remote host closed the connection]
20100411 14:22:43< timotei> Crab_, I don't get the difference:
20100411 14:22:48< timotei> state_.classification().campaign_define = "TUTORIAL";
20100411 14:22:50< timotei> cache_.add_define("TUTORIAL");
20100411 14:23:06< timotei> the last one is used with: #ifdef TUTORIAL
20100411 14:23:10< timotei> but the former?
20100411 14:23:17< Crab_> timotei: second is for WML preprocessor, yes
20100411 14:24:01< timotei> is that for wrting [tutorial]  instead of [campaign]?
20100411 14:24:08< timotei> s/wrting/writing
20100411 14:24:18< Crab_> timotei: but the first is for wml preprocessor, too
20100411 14:24:23< Crab_> timotei: I guess that the second is a hack
20100411 14:24:36-!- Darkas [~quassel@ppp-93-104-159-193.dynamic.mnet-online.de] has joined #wesnoth-dev
20100411 14:24:39< Crab_> timotei: see fgrep -Rn campaign_define /wesnoth/src
20100411 14:25:02< Crab_> timotei: note "game_config::scoped_preproc_define campaign_define_def(state_.classification().campaign_define, !state_.classification().campaign_define.empty());" line
20100411 14:25:36< Crab_> timotei: and similar lines
20100411 14:25:51< Crab_> timotei: I guess that the code path for 'playing the tutorial' contains some hacks.
20100411 14:26:17< Crab_> timotei: e.g., if it was the same as with normal scenarios,there would be no need to set campaign_define to TUTORIAL explicitly
20100411 14:27:16< timotei> oh
20100411 14:27:17< timotei> ok
20100411 14:27:30< timotei> cause with TEST is the same thing, but I'll investigate more
20100411 14:27:32< timotei> brb again
20100411 14:27:33-!- timotei is now known as timotei_away
20100411 14:27:40< Crab_> as the tutorial can be considered a campaign and should be eventually changed to be a campaign, as part of 'make mp,sp,tests use more similar code paths'
20100411 14:27:52< Crab_> but, as of now, there are differences in handling
20100411 14:30:58< Upthorn> Hrm. I'm trying to implement a quick prototype of set_global_variable and get_global_variable, and I am having difficulty making [store_unit], variable=hero, ..., [/store_unit], [set_global_variable], from_local=hero, ... [/set_global_variable] to work.
20100411 14:31:19< Crab_> Upthorn: what is the difficulty ?
20100411 14:31:22< Upthorn> the resulting .cfg line just has "hero="yes"
20100411 14:31:37< Crab_> Upthorn: use :inspect
20100411 14:31:52< Crab_> Upthorn: see what's in "hero" variable. is it's value correct ?
20100411 14:32:52< Upthorn> I will try that.
20100411 14:33:09< Upthorn> but I think I am using the wrong function to get hero.
20100411 14:33:22< Crab_> paste the code for store_unit, then
20100411 14:33:37< Upthorn> err, in set_global_variable.
20100411 14:35:19< Upthorn> I think that "cfg[global] = resources::state_of_game->get_variable_cfg(pcfg[local]);" must be wrong
20100411 14:37:20< Crab_> cfg[global] looks not good to me
20100411 14:38:09< Crab_> because it looks like a assignment of boolean value to cfg[global]
20100411 14:38:22< Crab_> and config can be, in fact, considered a boolean in certain contexts
20100411 14:38:23< Upthorn> That would explain the problem.
20100411 14:38:27< Crab_> that's why you get hero="yes"
20100411 14:39:17< Crab_> silene ^: note the fact that the above code silently works wrongly, it might be related to the recent changes in config.
20100411 14:39:47< Upthorn> what is the correct way to duplicate a wml variable in a type-agnostic manner?
20100411 14:41:05< Crab_> well, the wml variable is a config
20100411 14:41:17< Crab_> I think that resources::state_of_game->get_variable_cfg(pcfg[local]) is ok
20100411 14:41:34< Crab_> but, then, you need to add or replace a child in the 'global' config object
20100411 14:41:52< Crab_> it depends on the structure of that object
20100411 14:42:08< Crab_> nagbot: upthorn ?
20100411 14:42:09< nagbot> upthorn : http://wiki.wesnoth.org//SummerOfCodeProposal_JodyNorthup : http://socghop.appspot.com/gsoc/student_proposal/review/google/gsoc2010/upthorn/t127026899399
20100411 14:42:42< Crab_> Upthorn: what's the structure of your [variables] tag ?
20100411 14:43:57< Upthorn> [variables] valid variables [/variables]
20100411 14:44:01< Crab_> if it's the same as in savegame, then see how lua/savegame code does it - basically, there's two code paths depending on whether the variable is a scalar or container
20100411 14:44:29< Upthorn> yes, it is intended to be the same as in savegame.
20100411 14:44:31< Crab_> Upthorn: check this http://wesnoth.pastebin.com/yySd6CzQ
20100411 14:45:34< Crab_> Upthorn: but what do you think about http://wesnoth.pastebin.com/CGUtraL2 ?
20100411 14:45:53< Crab_> Upthorn: e.g., the savegame structure is bad for expanding, if you'll ever need to add additional properties to variables
20100411 14:46:03< Crab_> what about using a more formal approach ?
20100411 14:48:07< Upthorn> I haven't considered this deeply yet. I'm not sure about using [variable], because people looking at the cfg files might get it confused with the [variable] in http://wiki.wesnoth.org/ConditionalActionsWML#True_Condition_Tags
20100411 14:48:23< Crab_> use [global_variable] :)
20100411 14:48:46< Crab_> then it's quite easy, just find-child-by-id on top-level config, if found, replace, if not found, add, then save.
20100411 14:49:33< Upthorn> If we use a different format than savegame currently uses, I'd also want to update savegame's format to the improved one.
20100411 14:49:40< Crab_> why ?
20100411 14:50:08< Crab_> to be more specific 'why fix something that is not broken?'
20100411 14:50:46< Crab_> e.g., the format in the savegame exists and works for local variables, I see no reason to change it. if you see such reasons, tell me.
20100411 14:50:53-!- Gambit [~Gambit@pa-67-234-73-7.dhcp.embarqhsd.net] has joined #wesnoth-dev
20100411 14:51:12< Crab_> but the format for global vars doesn't exist yet, so we can pick the one we like without worrying for backward-compatibility
20100411 14:52:05< Crab_> for example, if you will later want to add a 'this variable is SP-only' flag to the variable, then the savegame format will not suit you.
20100411 14:52:26< Crab_> Upthorn: (^ just a fictional example)
20100411 14:52:55< Upthorn> I just feel like persistence .cfg files and savegames should look the same. I guess there is no real reason.
20100411 14:53:14< Upthorn> I see what you mean, though.
20100411 14:53:38< Crab_> then I'd say that you can add an optional part of your project 'make  persistence .cfg files and savegames should look the same', but let's consider it only if+after all other things are done.
20100411 14:54:17< Crab_> note that the gamemaster's persistent variables might end up stored in a transactional SQL database, so there'll not be a 'the same format' for all types of things :)
20100411 14:54:20< Upthorn> by the way
20100411 14:54:34< Crab_> (or in key-value database)
20100411 14:55:35< Upthorn> does [set_variable] to_variable=foo [/set_variable] work if foo is a container?
20100411 14:56:08-!- Darkas [~quassel@ppp-93-104-159-193.dynamic.mnet-online.de] has quit [Remote host closed the connection]
20100411 14:56:28< Upthorn> from the code it looks like it only works if foo is a scalar.
20100411 14:56:37< Crab_> Upthorn: haven't tried this :)
20100411 14:58:25< Crab_> Upthorn: zookeeper should know. or we can just try and see.
20100411 14:58:26-!- Darkas [~quassel@ppp-93-104-159-193.dynamic.mnet-online.de] has joined #wesnoth-dev
20100411 15:03:57< fendrin> timotei_away: still around?
20100411 15:12:39-!- elias [~elias@allegro/developer/allefant] has joined #wesnoth-dev
20100411 15:14:42-!- timotei_away [~timotei@193.34.191.4] has quit [Read error: Connection timed out]
20100411 15:19:14-!- FAAB [~huajie@219.142.232.61] has joined #wesnoth-dev
20100411 15:20:15< FAAB> hi Crab_, there ?
20100411 15:21:02< FAAB> Crab_: could you check the bug report I filed there : https://gna.org/bugs/?15837
20100411 15:21:25< Upthorn> indeed, [set_variable] variable=somename, to_variable=foo [/set_variable] does not work if foo is a container.
20100411 15:21:38< Crab_> hi, FAAB
20100411 15:22:24< Crab_> FAAB: thanks, I'll do so now
20100411 15:22:27< Upthorn> but set_variables looks like it should.
20100411 15:23:33< FAAB> Crab_: ok, I will stay on line if you have feedback. I am going AFK now.
20100411 15:23:55< Crab_> ok
20100411 15:28:04< Upthorn> I conclude [set_global_variables] and [get_global_variables] tags are missing from the suggested syntax.
20100411 15:30:24< Crab_> Upthorn: yes, were missing. in fact, you can add them to the project..
20100411 15:31:10< Crab_> but, it's more important to get the basic stuff right, since set_global_variables can be emulated by set_global_variable and set_variables.
20100411 15:38:32< Crab_> FAAB: reproduced
20100411 15:41:34-!- Tesafilmchen [~quassel@p5B27731F.dip.t-dialin.net] has quit [Read error: Connection reset by peer]
20100411 15:44:19< Crab_> FAAB: ok, launched in formula ai debugger, found out the reason (all formula ai code is currently somewhat broken in trunk, me.x and me.y are both 0). will investigate and fix.
20100411 15:47:54< Crab_> question: is there anyone with stock wesnoth 1.8 who can run a simple test for me ?
20100411 15:49:45< Crab_> launch wesnoth 1.8, enter any campaign, enter debug mode with :debug, type "f" to get to formula ai console, type "[my_leader.id , my_leader.loc]" and post the result
20100411 15:52:59< FAAB> Crab_: I can help if tell me what campaign you are talking about
20100411 15:53:03< Crab_> any
20100411 15:53:06< Crab_> but in 1.8
20100411 15:53:24< FAAB> single player campaign like httt ?
20100411 15:53:26< Crab_> yes
20100411 15:53:32< FAAB> will do
20100411 15:53:34< Crab_> thanks
20100411 15:53:50< Crab_> most likely, it'll return not-null value for the location
20100411 15:53:53< Crab_> but I want to be sure
20100411 15:54:26< Crab_> I've found the bug in trunk, it's most likely related to the big unit_map changes which were done in trunk since 1.8 is out
20100411 15:56:03< FAAB> In httt, I get ['Konrad', {x->20,y->23}]
20100411 15:56:04-!- iwaim [~iwaim@2001:2c0:40e::1] has joined #wesnoth-dev
20100411 15:56:10< Crab_> good
20100411 15:56:20< Crab_> and it trunk, it's konrad, null
20100411 15:56:29< Crab_> or 0,0 , whatever.
20100411 15:56:34< Crab_> ok, I'll fix it.
20100411 15:56:47< FAAB> good :)
20100411 15:57:33< FAAB> get the bug or my forum thread updated so that I know what I can try again, will you ?
20100411 15:57:42< Crab_> the bug will get updated
20100411 15:58:05< FAAB> thanks a bunch !
20100411 16:00:02< Crab_> FAAB: you can try http://wesnoth.pastebin.com/5qSsUvCT if you want something fast.
20100411 16:00:23< Crab_> 90% that it'll fix the bug.
20100411 16:02:20-!- teaser [~teaser@h-37-106.A254.priv.bahnhof.se] has quit [Remote host closed the connection]
20100411 16:02:37-!- teaser [~teaser@h-37-106.A254.priv.bahnhof.se] has joined #wesnoth-dev
20100411 16:03:14< FAAB> Crab_: testing ...
20100411 16:05:57< FAAB> Crab_: it is doing the trick on the AI arena scenario :)
20100411 16:06:03< Crab_> FAAB: good.
20100411 16:07:08< FAAB> Crab_: BTW, was that useful that I spent time on that scenario or could I have just file a bug that it would have been enough ?
20100411 16:07:41< Crab_> FAAB: yes, definitely useful, as it allowed me to reproduce very quickly
20100411 16:08:08< Crab_> and, after reproducing, I was able to use the formula ai debugger to find out a simpler testcase, which is my_leader.loc
20100411 16:08:38< Crab_> and that lead me to the place in the code which was accidentally broken.
20100411 16:12:11-!- lukjad86 [~lukjadOO7@unaffiliated/lukjad007] has quit [Ping timeout: 246 seconds]
20100411 16:12:17< FAAB> Crab_: ok, its good to know that was useful ;)
20100411 16:12:37< Crab_> FAAB: very useful. fixed two bugs at the same time, btw :)
20100411 16:12:48< Crab_> and it was broken in trunk only, since sun Mar 28
20100411 16:13:25< FAAB> that was fast indeed. I let you do your stuff now. See you around.
20100411 16:13:25< wesbot> FAAB: Sometimes we are fast
20100411 16:13:48-!- lukjad86 [~lukjadOO7@unaffiliated/lukjad007] has joined #wesnoth-dev
20100411 16:13:53< FAAB> wesbot: what do you mean ?
20100411 16:14:29< nagbot> we, bots, sometimes listen to your conversations
20100411 16:14:30< FAAB> ok, that is just a robot :S
20100411 16:14:43< Crab_> yes, one of our bots :)
20100411 16:15:26-!- FAAB [~huajie@219.142.232.61] has left #wesnoth-dev []
20100411 16:23:27< CIA-32> crab * r42102 /trunk/src/callable_objects.hpp: Fix bug #15829. Fix bug #15837. Formula AI in trunk was broken (unit locations were always null) since Mar 28. Thanks to FAAB for a test case.
20100411 16:56:14< Crab_> wesbot: seen espreon
20100411 16:56:14< wesbot> Crab_: The person with the nick Espreon last spoke 9h 30m ago.  8h 22m ago was here and on the channel #wesnoth-umc-dev with the message: Quit: WRYYYYYYYYYYYYYYYYYYYY!
20100411 16:56:54< CIA-32> crab * r42103 /trunk/src/ai/manager.cpp: implement espreon's FR about making modify_ai in [side] work without side= attribute
20100411 16:57:54-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20100411 17:02:25-!- teaser [~teaser@h-37-106.A254.priv.bahnhof.se] has quit [Ping timeout: 264 seconds]
20100411 17:12:10-!- lukjad86 [~lukjadOO7@unaffiliated/lukjad007] has quit [Ping timeout: 260 seconds]
20100411 17:15:09-!- lukjad007 [~lukjadOO7@unaffiliated/lukjad007] has joined #wesnoth-dev
20100411 17:20:49-!- Blarumyrran [~Blarumyrr@81-20-159-197.levira.ee] has joined #wesnoth-dev
20100411 17:37:43-!- Darkas [~quassel@ppp-93-104-159-193.dynamic.mnet-online.de] has quit [Remote host closed the connection]
20100411 17:44:58< Soliton> AI0867: http://forum.wesnoth.org/viewtopic.php?f=6&t=29417 might there be some problem with the advanced preference?
20100411 17:47:00-!- knotwork [~markm@142.177.235.201] has joined #wesnoth-dev
20100411 17:48:21< Gambit> Oh my gosh. Someone just called Dave a dumbass.
20100411 17:48:31< Gambit> Smite!
20100411 17:50:19< Blarumyrran> He didn't
20100411 17:50:55< Blarumyrran> He was parodizing(?) Dave
20100411 17:51:00< Gambit> Hmm. He's accusing Dave of thinking people who complain about luck are dumbasses?
20100411 17:51:06< Blarumyrran> Yes
20100411 17:51:16< Gambit> Well... they are. :P
20100411 17:51:26< Gambit> jk. Nevermind.
20100411 17:51:32 * Gambit goes away.
20100411 17:55:39-!- ilor_ [~ilor@wesnoth/developer/ilor] has quit [Ping timeout: 260 seconds]
20100411 17:58:50-!- chains [~Rylar@adsl-75-37-46-245.dsl.pltn13.sbcglobal.net] has joined #wesnoth-dev
20100411 18:24:53-!- icelus [~ed@cpc9-sgyl11-0-0-cust29.sgyl.cable.virginmedia.com] has joined #wesnoth-dev
20100411 18:31:11-!- lfernando [~luiz@201-43-134-54.dsl.telesp.net.br] has joined #wesnoth-dev
20100411 18:33:36< lfernando> hello
20100411 18:33:43< Crab_> hi, lfernando
20100411 18:34:14< Crab_> the bug in LoW 2 which you've reported, is fixed
20100411 18:34:26< lfernando> hi crab_ =)
20100411 18:35:12< lfernando> nice
20100411 18:35:20< lfernando> I saw your fix
20100411 18:36:12< Crab_> there are two constructors in there, previously different constructor was called. but the caller code got simplified, and accidentally the caller started to call 1-arg constructor, which used null for location.
20100411 18:36:44< Crab_> my fix is somewhat hacky (loc_ can be removed from that code, entirely), but works.
20100411 18:39:08-!- Espreon [~espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20100411 18:39:14< Espreon> Crab_: Jeß?
20100411 18:39:26< Crab_> Espreon: [modify_ai] in [side] should now work without side=
20100411 18:39:47< Espreon> Cool, thanks.
20100411 18:39:53< icelus> Crab_, by the way i looked at that other DF bug you mentioned
20100411 18:40:08< Crab_> icelus: great. is it DF-specific thing or an engine bug?
20100411 18:40:19< icelus> it's definitely an engine bug, but I don't understand yet
20100411 18:40:22< icelus> quite how it works
20100411 18:40:27< icelus> there are two things that are definitely wrong
20100411 18:40:27< Espreon> Ivanovic: Around?
20100411 18:40:33< Ivanovic> depends
20100411 18:40:46< icelus> one thing that's definitely wrong is set_movement(0) is called on one host but not the other
20100411 18:40:47< Crab_> icelus: yes, the end result is definitely bad
20100411 18:40:59< Espreon> Ivanovic: Do our policies forbid adding new images to stable versions?
20100411 18:41:07< Ivanovic> yes
20100411 18:41:20< Espreon> Thanks.
20100411 18:41:20< icelus> the other thing that's even more definitely wrong is that a unit unstore has different parameters passed to it
20100411 18:41:27< icelus> so the wml variables are certainly out of sync
20100411 18:41:34< icelus> that's the cause of the bug (at the very most basic level)
20100411 18:41:45< icelus> one host unstores a unit with 2 movement
20100411 18:41:49< icelus> one unstores a unit with 0 movement
20100411 18:42:04< icelus> there is a preceding set_movement(0) but i'm not clear if they're connected
20100411 18:42:13< icelus> in general the execution traces are _wildly_ different
20100411 18:42:24< icelus> like adding/deleting/extracting units
20100411 18:42:31< icelus> happens 3x as often on one host than the other
20100411 18:42:38< icelus> i still don't understand why though =p
20100411 18:43:29< icelus> so far all i've done is instrument every single way you could modify a unit's movement
20100411 18:43:55< icelus> there's some weirdness with AI here...
20100411 18:43:59< icelus> does the AI run on one host only
20100411 18:44:01< icelus> or on both
20100411 18:44:07< icelus> because only one host is doing A*
20100411 18:44:40< icelus> i think it's ok because the AI runs on one host and injects its decisions into the replay
20100411 18:44:56< icelus> if so the most likely cause of the bug is that the AI has side effects that aren't injected into the replay
20100411 18:44:59< icelus> that's my current guess
20100411 18:45:08< icelus> but perhaps someone else would know if that's plausible
20100411 18:45:25< Crab_> AI runs only on one side
20100411 18:45:28< Crab_> by default, on host
20100411 18:45:38< icelus> ok that makes sense
20100411 18:45:48< Crab_> ' if so the most likely cause of the bug is that the AI has side effects that aren't injected into the replay' - yes, I've seen it before
20100411 18:45:49< icelus> how is it enforced that everything the ai does
20100411 18:45:51< icelus> ends up in the replay
20100411 18:45:55< icelus> ok
20100411 18:45:57< icelus> ok that's good
20100411 18:46:05< icelus> that fits with what i was working on
20100411 18:46:17< Crab_> please check unit::remove_movement_ai
20100411 18:46:19< icelus> is there an enforcement model for the AI Crab_ ?
20100411 18:46:21< icelus> yeah i saw that
20100411 18:46:28< Crab_> it is the only 'unenforced' place
20100411 18:46:35< icelus> ok cool
20100411 18:46:40< Crab_> because the ai needs a way to say 'ok, I'm done with that unit'
20100411 18:47:00< icelus> so perhaps what's going wrong is
20100411 18:47:08< icelus> the AI makes its moves on the host
20100411 18:47:10< Crab_> src/ai/actions.cpp is the only place where the ai is allowed to change the game state
20100411 18:47:20< icelus> then zeroes out its movements
20100411 18:47:27< Crab_> I want to get rid of stopunit action, but, unfortunately, some things in ai algorithms depend on it
20100411 18:47:32< icelus> but that zeroing doesn't happen remotely
20100411 18:47:50< icelus> it's also a safe OOS right
20100411 18:47:54< icelus> if that is the issue
20100411 18:47:56< icelus> it's not a big deal
20100411 18:48:02< icelus> we can fix it, i'm not suggesting we don't
20100411 18:48:06< Crab_> yes, if it's so,it's safe.
20100411 18:48:09< icelus> but the AI isn't going to move those units
20100411 18:48:09< Crab_> but best to fix
20100411 18:48:12< icelus> yeah
20100411 18:48:12< Crab_> yes
20100411 18:48:22< Crab_> note that it's not 'fully' safe
20100411 18:48:22< icelus> the only way it could make a difference is resting effects i guess
20100411 18:48:31< Crab_> resting is accounted for
20100411 18:48:31< icelus> yeah
20100411 18:48:35< icelus> oh ok
20100411 18:48:44< Crab_> (it caused oos, which I fixed last year)
20100411 18:48:55< Crab_> the problem is that WML can filter on movement left
20100411 18:49:06< icelus> Crab_, ok well we need to fix
20100411 18:49:10< icelus> i will try to submit a patch
20100411 18:49:12< Crab_> the better way to fix it is to remove the 'stop unit' action from the ai
20100411 18:49:28< Crab_> because the human cannot stop a unit
20100411 18:49:28< icelus> i'll look at options when i understand it better...
20100411 18:49:34< icelus> unless you want to fix it?
20100411 18:49:39< Crab_> (another hacky way is to transmit stopunit via a replay'
20100411 18:49:45< Crab_> I want, but too gsoc-busy atm
20100411 18:49:46< icelus> we could do that
20100411 18:49:55< icelus> ok i'll look at it
20100411 18:50:38< icelus> my idea at the moment is whenever the ai changes game state
20100411 18:50:43< icelus> we need to put that in the replay
20100411 18:50:52< icelus> an even better solution is the AI only changes game state in safe ways
20100411 18:50:57< icelus> it's a question of which is harder
20100411 18:51:09< icelus> often AIs do things which aren't safe...
20100411 18:51:16< icelus> and it's not always practical to fix that
20100411 18:51:34< Crab_> ' often AIs do things which aren't safe...' - where ?
20100411 18:51:42< icelus> i just mean in general in games
20100411 18:51:46< icelus> AI is not really like a player
20100411 18:51:48< Crab_> afair, the 'stop unit by removing its movement and/or attacks' is the only hack left.
20100411 18:52:00< icelus> it's much harder if the AI has to behave like player strictly
20100411 18:52:02< Crab_> I got rid of others last year
20100411 18:52:06< icelus> ok
20100411 18:52:14< icelus> well maybe i should look at removing that hack then
20100411 18:52:21< icelus> that seems to be the solution you favour?
20100411 18:52:24< Crab_> yes
20100411 18:52:28< icelus> ok
20100411 18:52:48< Crab_> it should be easier to remove this hack for new ai
20100411 18:53:18< Crab_> note that 'full movement' action also uses the hack , because 'full move'=move+stop after move
20100411 18:53:44< icelus> i'm tied up for the moment doing my dayjob
20100411 18:53:50< icelus> but i'll look more closely later this evening
20100411 18:53:58< Crab_> ok. user_end_turn flag should be used instead of setting MP to 0
20100411 18:54:43< Crab_> and unit::remove_attacks_ai should be just removed
20100411 18:55:10< Crab_> (so, we can go without fixing code in src/ai/default)
20100411 18:55:52< icelus> Crab_, ok i'll look at how it works
20100411 18:56:03< Crab_> (most of the code there is the old default ai which'll most likely gets removed from source, if no big problems in 1.8 will arise
20100411 18:56:18< Crab_> be sure to ask any questions you'll need to ask
20100411 18:56:25< icelus> Crab_, will do
20100411 18:56:44< Crab_> e.g. I know the code in src/ai too well, having spent a lot of time on it last year.
20100411 18:57:21< icelus> =)
20100411 18:57:32< Crab_> see the http://wiki.wesnoth.org/Ai_Module for the file descriptions
20100411 18:57:43-!- lukjad007 [~lukjadOO7@unaffiliated/lukjad007] has quit [Ping timeout: 260 seconds]
20100411 18:57:50< CIA-32> ivanovic * r42104 /trunk/ (31 files in 25 dirs): updated Galician and Slovak translation
20100411 18:57:51< CIA-32> ivanovic * r42105 /branches/1.8/ (31 files in 25 dirs): updated Galician and Slovak translation
20100411 19:00:03-!- Noyga [~noyga@wesnoth/developer/noyga] has joined #wesnoth-dev
20100411 19:00:21-!- lukjad007 [~lukjadOO7@unaffiliated/lukjad007] has joined #wesnoth-dev
20100411 19:08:32-!- lukjad007 is now known as lukjad86
20100411 19:09:24-!- teaser [~teaser@h-37-106.A254.priv.bahnhof.se] has joined #wesnoth-dev
20100411 19:12:06-!- ancestral [~ancestral@97-116-115-135.mpls.qwest.net] has joined #wesnoth-dev
20100411 19:13:53-!- lukjad86 [~lukjadOO7@unaffiliated/lukjad007] has quit [Ping timeout: 276 seconds]
20100411 19:17:08-!- teaser [~teaser@h-37-106.A254.priv.bahnhof.se] has quit [Ping timeout: 276 seconds]
20100411 19:24:33-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100411 19:25:06-!- icelus [~ed@cpc9-sgyl11-0-0-cust29.sgyl.cable.virginmedia.com] has quit [Ping timeout: 245 seconds]
20100411 19:26:52-!- icelus [~ed@cpc9-sgyl11-0-0-cust29.sgyl.cable.virginmedia.com] has joined #wesnoth-dev
20100411 19:27:20-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20100411 19:30:14< Crab_> wesbot: seen Darkas
20100411 19:30:15< wesbot> Crab_: The person with the nick Darkas 1h 52m ago they left with the message: Remote host closed the connection
20100411 19:36:16< CIA-32> noyga * r42106 /branches/1.8/po/ (6 files in 6 dirs): French translation update
20100411 19:36:38-!- Blueblaze [~nick@99.182.52.75] has joined #wesnoth-dev
20100411 19:43:37< CIA-32> noyga * r42107 /trunk/po/ (6 files in 6 dirs): French translation update
20100411 19:48:42-!- Espreon [~espreon@wesnoth/developer/espreon] has quit [Quit: WRYYYYYYYYYYYYYYYYYYYY!]
20100411 19:50:58-!- wesbot changed the topic of #wesnoth-dev to: GSoC: Thanks to all GSoC candidates for submitting proposals to Google. Accepted student applications announced on April 26. | 86 bugs, 264 feature requests, 16 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100411 19:55:58-!- pokhbocee [~pokhbocee@ws40.cs.drexel.edu] has joined #wesnoth-dev
20100411 19:56:06< pokhbocee> hey everybody
20100411 19:56:12< Crab_> hi, pokhbocee
20100411 20:07:50-!- Vetinari [~lukjad007@unaffiliated/lukjad007] has joined #wesnoth-dev
20100411 20:09:05-!- lukjad007 [~lukjadOO7@unaffiliated/lukjad007] has joined #wesnoth-dev
20100411 20:11:36< happygrue> aranair: I read through the proposal and I see you have already gotten some feedback
20100411 20:11:37-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has joined #wesnoth-dev
20100411 20:12:22< happygrue> I would largely echo that.  It looks like quite a large change to gameplay.
20100411 20:12:51-!- ilor_ [~ilor@wesnoth/developer/ilor] has joined #wesnoth-dev
20100411 20:12:55< happygrue> could be interesting in a scenario setting, but not something that would be a good default setup for mainline play I think.
20100411 20:15:01< happygrue> some specifics:  there are some clearly defined game types that are common, simple and well liked by core groups of players
20100411 20:15:59< happygrue> for example, 2v2, 1v1, 3v3, free-for-all, co-op/survival
20100411 20:16:02< pokhbocee> Crab_: wesnoth make all initial letter of class names lowercase,
20100411 20:16:20< pokhbocee> i used to do them upper case to prevent confusion
20100411 20:16:27< pokhbocee> is it problem or do u want me to make them lowercase?
20100411 20:18:16< happygrue> aranair: anything that would change or mess around with those types of games is going to produce strong resistance.  Adding new types (without change to the others) would probably be OK. (for example, boucman's system would allow anyone wanting to play their 'normal' games to continue but adds some flexibilty)
20100411 20:18:45< pokhbocee> Crab_: and also do you want me to use memory efficiently ? or just do the ai work, cuz my current design is use little bit too much memory. 
20100411 20:18:56< happygrue> aranair: that is my take on it anyway, feel free to ask any questions that you have.
20100411 20:18:59< Crab_> pokhbocee: it's not a problem now, but, for any code to be included in wesnoth, it has to respect our coding conventions
20100411 20:19:22< Crab_> pokhbocee: about memory, my notebook has 2GB. will I be able to test your prototype ?
20100411 20:19:31-!- lukjad007 is now known as lukjad86
20100411 20:19:45< Crab_> (it's ok if a 'little bit too much' is used, since it can be dealt with later)
20100411 20:20:08< pokhbocee> I believe so, but not for high amount of units
20100411 20:20:12< Crab_> ok
20100411 20:20:51-!- orfest [~chatzilla@16.ngu.ac-tel.ru] has joined #wesnoth-dev
20100411 20:21:04< pokhbocee> ill do this one throw-away prototype. to finish it asap.
20100411 20:21:13< Crab_> very good !
20100411 20:21:15< orfest> Crab_: updated https://gna.org/patch/index.php?1597
20100411 20:21:24< Crab_> orfest: thanks
20100411 20:21:42-!- ^Noyga^ [~noyga@AVelizy-151-1-65-25.w82-124.abo.wanadoo.fr] has joined #wesnoth-dev
20100411 20:21:54< Crab_> orfest: what do you plan to do next ?
20100411 20:22:10-!- Noyga [~noyga@wesnoth/developer/noyga] has quit [Disconnected by services]
20100411 20:22:13-!- ^Noyga^ is now known as Noyga
20100411 20:22:20< orfest> Crab_: as we discussed before, the design proposal
20100411 20:22:27-!- Noyga [~noyga@AVelizy-151-1-65-25.w82-124.abo.wanadoo.fr] has quit [Changing host]
20100411 20:22:27-!- Noyga [~noyga@wesnoth/developer/noyga] has joined #wesnoth-dev
20100411 20:22:39< Crab_> orfest: when we can expect it ?
20100411 20:22:47< orfest> I'll submit it on Tuesday
20100411 20:22:50< orfest> is it ok?
20100411 20:22:56< Crab_> orfest: yes, it's good
20100411 20:23:22< Noyga> i did import the existing translations to the 1.8 branch for several addons on wescamp
20100411 20:24:14< pokhbocee> Crab_: btw i believe my other patches(1577 and 1596) are working atm. i will only modify 1577 to improve filtering after i finish the ai prototype. should it wait in progress??
20100411 20:24:36< Crab_> please finish the prototype first
20100411 20:24:52< pokhbocee> ok :D
20100411 20:24:57< Crab_> we will make sure the patches are ok, later. right now the c++ ai prototype has more priority
20100411 20:25:07< Crab_> thanks!
20100411 20:25:19< pokhbocee> no, thank you.
20100411 20:38:42< Zarel> LOL
20100411 20:38:50< Zarel> Someone edited my post and replaced "dick" with "ass"
20100411 20:39:13< Crab_> hehe. ministry of truth in action :)
20100411 20:39:25< Zarel> Maybe it was the wordfilter.
20100411 20:39:40< Zarel> Although I don't know of any wordfilter that would replace a mild swear with a stronger swear.
20100411 20:40:03< Zarel> Whoever edited it must have different standards of what counts as a swear than I do.
20100411 20:40:26< Zarel> Or maybe I was really out of it and typed "ass" when I meant to type "dick"? Seems unlikely; I pretty much never type "ass"
20100411 20:41:02-!- Elsewise [~Dmitry@80.233.231.12] has joined #wesnoth-dev
20100411 20:42:26< Blarumyrran> I think ass is milder
20100411 20:43:46< pokhbocee> Crab_: which attributes of unit consist the information about attack?
20100411 20:44:27< Crab_> there is attacks() vector in there
20100411 20:44:46< Crab_> const std::vector<attack_type>& attacks()
20100411 20:45:01< pokhbocee> i find it. ty
20100411 20:45:17-!- iwaim [~iwaim@2001:2c0:40e::1] has quit [Ping timeout: 268 seconds]
20100411 20:45:23< lfernando> If I have a config object
20100411 20:45:24< Crab_> src/unit.hpp:187
20100411 20:45:32< lfernando> and I want to add an attribute to it
20100411 20:45:53< Crab_> config cfg;  cfg["attribute_name"] = value;
20100411 20:45:55< lfernando> how should I do it
20100411 20:46:03< Crab_> note the difference between child and attribute
20100411 20:46:11< lfernando> ah
20100411 20:46:12< lfernando> ty
20100411 20:46:26< lfernando> for childs there are methods, right?
20100411 20:46:42< lfernando> add_child(), add_child_at()
20100411 20:46:49< Crab_> yes, add_child("child_name",child_config);
20100411 20:47:07< Elsewise> Hello... I am a wandering developer walking around looking for something interesting. From one conversation I understood  that I can contribute to this project by implementing ideas. The question is: can I implement what i want or if i join the community, I will be given a task?
20100411 20:47:21< Crab_> Elsewise: hello
20100411 20:47:37< Crab_> yes, you can implement what you want, if no one objects :)
20100411 20:47:42< lfernando> welcome elsewise
20100411 20:48:21< Crab_> Elsewise: e.g., there are some ideas which are 'not wanted', to keep the 'spirit of the game'. but, apart from that, you can pick the area which interests you.
20100411 20:48:26< Elsewise> Crab_, and it will be added to the actual project, so that I can show it to my employers later? 
20100411 20:48:50< Crab_> Elsewise: yes, if it adds value to the game or fixes a bug, it will be added in.
20100411 20:49:08< Crab_> Elsewise: generally, after you submit two non-trivial patches you'll be offered commit access
20100411 20:49:46< Crab_> Elsewise: then, you'll be able to work, directly committing to the game source, and you'll be marked as a 'programmer' in the credits.
20100411 20:50:32-!- Darkas [~quassel@ppp-88-217-112-100.dynamic.mnet-online.de] has joined #wesnoth-dev
20100411 20:50:36< Crab_> Elsewise: is there a particular thing which interests you ?
20100411 20:50:37< Crab_> hi, Darkas
20100411 20:52:04-!- timotei [~timotei@193.34.191.4] has joined #wesnoth-dev
20100411 20:52:09< Elsewise> Crab_, I am generally interested in two things - physics and AI. Since the game has no actual physics engine, I would probably like to work on the AI
20100411 20:52:25< timotei> hello everyone
20100411 20:52:39< Crab_> lfernando: also note that it *might* be easier to use a different format for variables than it's used in the save game. e.g. [global_variables][global_variable]id=myid [value] ... [/value] [/global_variable] [global_variable]id=myid23 [value] ... [/value] [[/global_variable] [/global_variables]
20100411 20:52:47< Crab_> hi, timotei
20100411 20:53:04< Crab_> timotei: is there a particular reason for committing compiled .class files to svn :) ?
20100411 20:53:23< timotei> oh, well I commited everything
20100411 20:53:29< timotei> and as long as I think
20100411 20:53:35< timotei> the compiled classes
20100411 20:53:48< Crab_> Elsewise: that's good. you're welcome to make wesnoth's AI better :)
20100411 20:53:51< timotei> are defining actual behaviour of the editor
20100411 20:54:09< timotei> or you're telling about... the compiled one from the plugin?
20100411 20:54:11< timotei> let me check
20100411 20:54:50< Elsewise> Crab_, can you, please, give me a waypoint on where to start from?
20100411 20:55:14< Crab_> Elsewise: have you checked out the source ? on what operating system are you ?
20100411 20:55:45< timotei> Crab_, you're right
20100411 20:55:58< Crab_> Elsewise:      http://wiki.wesnoth.org/WesnothSVN - for the info about getting the source code                  http://wiki.wesnoth.org/CompilingWesnoth - for the info about compiling
20100411 20:56:04< Elsewise> Crab_, well, I can use both linux and windows, and haven't seen the source yet
20100411 20:56:24< timotei> the fact was, I was a bit dizzy trying to add the java project to the svn from eclipse :P and used the tortoise and added *everything* to be uploaded
20100411 20:56:31< timotei> is ok if I delete them now?
20100411 20:56:48< Crab_> timotei: of course. and add bin to svn:ignore afterwards
20100411 20:57:01< Crab_> Elsewise: you can work both on linux and on windows, compiling currently has about the same complexity. most developers use linux.
20100411 20:57:40< Crab_> the first step is to checkout the source code of 'trunk' version of wesnoth, and compile it. you can use svn or git to work with wesnoth.
20100411 20:58:06< Crab_> if you prefer git, say so and we'll point you to the location of the tarball which can be used to seed the git-svn repository.
20100411 20:58:34< Crab_> if you prefer svn, just use the "svn co http://svn.gna.org/svn/wesnoth/trunk wesnoth"  command or do it from a GUI.
20100411 20:58:47< Crab_> about 300mb (there's pictures and music in there)
20100411 20:59:18-!- YogiHH [YogiHH@d127214.adsl.hansenet.de] has joined #wesnoth-dev
20100411 20:59:33-!- YogiHH [YogiHH@d127214.adsl.hansenet.de] has quit [Changing host]
20100411 20:59:33-!- YogiHH [YogiHH@wesnoth/developer/yogihh] has joined #wesnoth-dev
20100411 20:59:53< Crab_> Elsewise: the, see http://wiki.wesnoth.org/AI_Module for the descriptions of AI-related files. you can ask me any questions you like about the files in src/ai
20100411 21:00:07< Elsewise> So then I just start hacking the code, and when I am done, I write something similar to a forum post, where people discuss my solution?
20100411 21:00:28< Crab_> if it's a bugfix or a clear improvement, you can submit a patch to patches.wesnoth.org
20100411 21:00:45< Crab_> if it's an idea, it's better to discuss it in the 'Ideas' forum first
20100411 21:00:56< icelus> Elsewise, i just started on the same path myself a few days ago
20100411 21:01:24< Crab_> Elsewise: there's some coding conventions (http://wiki.wesnoth.org/CodingStandards )
20100411 21:01:56< icelus> i recommend pulling the code and looking at stuff like http://wiki.wesnoth.org/EasyCoding http://wiki.wesnoth.org/NotSoEasyCoding http://wiki.wesnoth.org/NotSoEasyCoding https://gna.org/bugs/index.php?group=wesnoth&set=custom&report_id=101&status_id=1&chunksz=150&boxoptionwanted=1
20100411 21:02:02< icelus> while you figure out the codebase
20100411 21:02:12< icelus> it's pretty nice code, it's quite easy to hack on
20100411 21:02:16< icelus> at least that's my finding so far
20100411 21:02:24< pokhbocee> Crab_: is it possible to copy game_info and apply the action on it separate from the original?
20100411 21:02:25< icelus> ~110k lines say, so mid size
20100411 21:03:13< Crab_> AI is about 17k
20100411 21:03:18< CIA-32> timotei * r42108 /trunk/utils/java/ (6 files in 6 dirs): - delete compiled classes
20100411 21:04:23< timotei> good, removed them:P
20100411 21:04:29< timotei> thanks for pointing it out
20100411 21:04:29< Crab_> timotei: great :))
20100411 21:04:31-!- Noyga [~noyga@wesnoth/developer/noyga] has left #wesnoth-dev ["Quitte"]
20100411 21:04:32< Crab_> np :)
20100411 21:04:51< Elsewise> java? I though the project was C++ based
20100411 21:04:59< timotei> did you checked them out? to see if it works?
20100411 21:05:11< Crab_> Elsewise: it's a side project, an eclipse plugin prototype
20100411 21:05:22< Elsewise> oh, got me scared :)
20100411 21:05:42< Crab_> Elsewise: so, it's in java. but the engine of the game is c++ , and scenarios are in WML (custom markup language) and lua
20100411 21:05:52< Crab_> AI can be written in lua, too
20100411 21:06:23< Crab_> pokhbocee: yes, ai uses src/ai/game_info.hpp to get the game info from. you can swap/replace it.
20100411 21:07:08< Crab_> Elsewise: there's some perl and python side utilities, too
20100411 21:08:08< pokhbocee> Crab_: how do i apply an action on the game_info and also how do i undo a move that i do?
20100411 21:08:47< Elsewise> Crab_, I got so excited about the idea that I even forgot that I haven't played the game yet
20100411 21:08:57< Crab_> pokhbocee: you can store current game info, and then restore it after your actions.
20100411 21:09:06< Elsewise> well, I have, but only for like 10 minutes to understand the gameplay
20100411 21:09:13< Crab_> Elsewise: yes, playing the game is the important part of the development process :))))
20100411 21:09:30< Crab_> Elsewise: especially for the AI-related work.
20100411 21:09:57< icelus> Crab_, by the way about boost...
20100411 21:10:04< Crab_> Darkas: around ?
20100411 21:10:08< icelus> i commented on this in the forum a while back
20100411 21:10:09< pokhbocee> Crab_: i just want to see how actions affect the game_info. according to that ill create my own method
20100411 21:10:14< icelus> i think you were only person who responded
20100411 21:10:15< Darkas> Crab_: yeah
20100411 21:10:29< icelus> boost 1.34 is sufficient providing you have boost_serialization
20100411 21:10:35< icelus> is there any chance of getting this rolled back
20100411 21:10:51< pokhbocee> Crab_: all of my units will belong to single team right? not separate teams
20100411 21:10:52< icelus> i still only have 1.34 =p
20100411 21:10:57< Crab_> Darkas: I wanted to to ask you, how the project is going for you.
20100411 21:11:20< Elsewise> Crab_, (offtop) are you russian? (read your name @ the site)
20100411 21:11:28< Crab_> Elsewise: ukrainian
20100411 21:11:53< Elsewise> oh, well, you speak russian do you?
20100411 21:11:56< Crab_> icelus: well, if you add  boost foreach to wesnoth's source, you can make wesnoth work with 1.34. but, as soon as boost::asio is added, you'll have to use 1.35
20100411 21:12:04< Crab_> Elsewise: yes, it's one of my native languages
20100411 21:12:07< Darkas> Crab_: well, right now I'm still thinking about several things, how it could be implemented the best way, etc.
20100411 21:12:15< icelus> Crab_, my system runs just fine right now with 1.34 only
20100411 21:12:24-!- teaser [~teaser@h-37-106.A254.priv.bahnhof.se] has joined #wesnoth-dev
20100411 21:12:25< icelus> Crab_, going to 1.35 would be annoying
20100411 21:12:42< icelus> Crab_, i literally just edited configure.ac, no other changes
20100411 21:12:44< Crab_> icelus: cool. but there are definite plans to use boost::asio for network stack
20100411 21:12:55< icelus> Crab_, this is unfortunate =p
20100411 21:13:48< Crab_> Darkas: ok. have you read the mentor's comments on the GSoC site ?
20100411 21:13:49< icelus> why are we moving to asynchronous io, is there a page or thread i should read on this
20100411 21:14:33< Darkas> Crab_: not yet
20100411 21:14:57< Crab_> icelus: there were some talks on IRC, if you grep the logs, it should be possible to find it.
20100411 21:15:15< Elsewise> Crab_, by the way, if I contribute to the project now, will I be allowed to apply for GSoC next year? 
20100411 21:15:40< Crab_> Elsewise: yes, during the previous years, some of the GSoCers were already developers :)
20100411 21:15:57< Elsewise> That's totally great
20100411 21:16:10< Crab_> icelus: for me, the biggest benefit is that it would be possible to code a client which will be able to play multiple games at the same time, which is important for 'massive multiplayer' type of games.
20100411 21:17:07< icelus> Crab_, what do you see asio adding that allows you to do that
20100411 21:17:42< Crab_> icelus: the architecture rework associated with the rewrite. it should be possible to do that rework while using sdl.net , too.
20100411 21:18:30< icelus> Crab_, i'm sure that's the case
20100411 21:18:52< icelus> i'm always reluctant to involve more libraries
20100411 21:19:19< icelus> i guess from my perspective a library is a straight trade of
20100411 21:19:31< Crab_> also, there were issues with timeout values for sockets, with sdl.net
20100411 21:19:37< icelus> getting something more cheaply now versus paying way more for it later
20100411 21:20:23< icelus> well the SDL stuff in general is a pretty good example of that i think
20100411 21:20:48< icelus> SDL_mixer works out of the box but when you come to balancing different versions there are terrible bugs in many standard versions
20100411 21:21:06< icelus> SDL_ttf has incompatible behaviours on mac and linux if you try to do fine control of layout
20100411 21:21:32< icelus> i lost count of the hours i poured into trying to manage sdl_ttf bugs for silvertree
20100411 21:23:53< timotei> Crab_ is ok if I start working on the C++ codebase a little? just to "show off" my C++ skills too?
20100411 21:23:53< timotei> not only java one's
20100411 21:24:16< Crab_> timotei: it's ok! just post a patch instead of committing, for the first few times.
20100411 21:24:40< Crab_> timotei: you'd be able to commit them yourself, but, for the first times, only after a short review.
20100411 21:25:46< timotei> yes
20100411 21:26:15< Crab_> good :)
20100411 21:26:32-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has quit [Ping timeout: 265 seconds]
20100411 21:27:56-!- timotei is now known as timotei_away
20100411 21:34:43-!- dtiger [~dtiger@dynamic-vpdn-93-125-17-84.telecom.by] has quit [Remote host closed the connection]
20100411 21:38:26< boucman> wow, busy day in here
20100411 21:38:28< boucman> hello all
20100411 21:38:42< Crab_> hi, boucman
20100411 21:40:02< CIA-32> ivanovic * r42109 /trunk/po/ (wesnoth/gl.po wesnoth-anl/gl.po): updated Galician translation
20100411 21:40:06< CIA-32> ivanovic * r42110 /branches/1.8/po/ (wesnoth/gl.po wesnoth-anl/gl.po): updated Galician translation
20100411 21:41:00-!- Gambit [~Gambit@pa-67-234-73-7.dhcp.embarqhsd.net] has quit [Quit: Leaving.]
20100411 21:41:25< mordante> icelus, we want to trade SDL_Net for boost asio
20100411 21:41:44< mordante> from what I heard getting good timeouts in SDL_Net is impossible
20100411 21:42:00< mordante> maybe boucman remembers more details about the problems with SDL_Net
20100411 21:44:38< mordante> Elsewise, if you become a developer now, you will have an advantage the next GSoC we ask our own developers first
20100411 21:45:20-!- Elsewise [~Dmitry@80.233.231.12] has quit [Ping timeout: 276 seconds]
20100411 21:47:15< boucman> icelus: yes, one of the problem of sdl_net is that it's API is "incorrect" in some cases. IIRC (not sure, would have to check) some calls to read don't have a timeout parameter, and a reading thread can stay stuck indefinitely if the other end disapears without closing the socket (which means a hard freeze at the other end, but on a project as large as wesnoth it does happen)
20100411 21:47:51< icelus> boucman, yeah that's unfortunate
20100411 21:48:02< icelus> i stopped arguing against this when i realised you're using sdl_net at the moment
20100411 21:48:08< icelus> my experience of the SDL libraries is anything but positive
20100411 21:48:25< icelus> and trading them for something more professional has usually been desirable in my experience
20100411 21:48:40< boucman> so we need to get rid of SDL_net, and redeveloping from the socket level seems tedious and not worth it. since we already use chunks of boost, moving to asio seemed a sane choice
20100411 21:48:53< icelus> yeah i can't argue with that
20100411 21:49:12< icelus> it's unfortunate in that my distribution seems to have peaked at 1.34 for boost
20100411 21:49:23< icelus> but i was getting sick of it anyway so if the time comes i'll upgrade i guess
20100411 21:49:25< mordante> which distribution is that?
20100411 21:49:35< icelus> i'm still running vanilla ubuntu 8.10 i think
20100411 21:49:59< icelus> it's quite hard to get antyhing to work on this machine
20100411 21:50:05< mordante> ah ok, I use Debian Stable which has both 1.34 and 1.35
20100411 21:50:07< icelus> so i get reluctant to try for upgrades
20100411 21:50:27< mordante> I can imagine
20100411 21:50:38-!- Espreon [~espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20100411 21:50:39< icelus> the real issue is the thunderbird doesn't have some of the sse type instructions
20100411 21:50:51< icelus> so binaries are a problem
20100411 21:50:57< icelus> like opengl drivers and such
20100411 21:51:38< boucman> ok...
20100411 21:53:46< mordante> night
20100411 21:53:54< icelus> night
20100411 21:53:56-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20100411 21:57:24-!- gabba [~gabba@70.35.163.161] has joined #wesnoth-dev
20100411 21:59:24< gabba> Ivanovic, Crab_: thanks for your comments on my proposal
20100411 21:59:30< gabba> (s)
20100411 22:00:16< Crab_> gabba: ok :)
20100411 22:00:30< Crab_> thanks for submitting them :)
20100411 22:01:02-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has joined #wesnoth-dev
20100411 22:03:38< pokhbocee> Crab_: i couldnt find the method which commits a move or applies it on game_info
20100411 22:03:38< Soliton> shadowmaster: http://forum.wesnoth.org/viewtopic.php?p=420919#p420919  maybe you have an idea about how that could have happened.
20100411 22:04:09-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has quit [Read error: Connection reset by peer]
20100411 22:04:12-!- Elsewise [~Dmitry@80.233.231.12] has joined #wesnoth-dev
20100411 22:04:37< Crab_> pokhbocee: they are, atm, not in src/ai
20100411 22:04:48< Crab_> pokhbocee: code in src/ai/actions.cpp asks the game to do a move/attack/etc
20100411 22:05:32< Crab_> pokhbocee: if you want to do a 'real' move/attack, those methods will work. (e.g., check_move_action or execute_move_action from within ai context, or just by creating the action object directly)
20100411 22:06:47< pokhbocee> Crab_: i want to see how a real action is done, but i couldnt understand it from actions.cpp 
20100411 22:07:18< pokhbocee> Crab_: i couldnt find the classes attack_result_ptr and move_result_ptr etc.
20100411 22:07:44< Crab_> pokhbocee: src/ai/actions.?pp
20100411 22:08:09< Crab_> pokhbocee: note the large list of typedefs in src/ai/game_info.hpp
20100411 22:08:49< Crab_> typedef boost::shared_ptr<attack_result> attack_result_ptr;
20100411 22:09:07-!- Darkas [~quassel@ppp-88-217-112-100.dynamic.mnet-online.de] has quit [Remote host closed the connection]
20100411 22:09:09< pokhbocee> oh its typedef thats why i couldnt find the class :D ahah im so stupid
20100411 22:10:11< Crab_> fgrep is your friend :)
20100411 22:10:20< Crab_> pokhbocee: but now you know, so you're smarter now :)
20100411 22:10:45< Crab_> see attack_result::do_execute
20100411 22:11:10< Crab_> it's in src/ai/actions.cpp, it calls attack_unit from src/actions.cpp
20100411 22:12:52< pokhbocee> :D yeah i am :D
20100411 22:15:57< gabba> Ivanovic, mordante: Since I already list some low/medium-priority items in my calendar, would it be ok to expand on that section with more details? Or do you like a lot more the idea of a list of mandatory/optional items like ilor and Crab_ did last year? An example of one or two details that you think are missing would be great, so I have a more concrete idea of what you want.
20100411 22:16:40< Crab_> gabba: ok, will do so now
20100411 22:18:12< pokhbocee> Crab_: sorry i asked too many questions but i have one more. there are graphs that show probabilities of hitpoint results for each unit before attacking. 
20100411 22:18:12< gabba> Crab_: thanks -- even though you weren't the one asking for an even more detailed list :)
20100411 22:18:45< pokhbocee> Crab_: i want to get that info
20100411 22:18:51< Crab_> gabba: yes, but I want to check anyway :)
20100411 22:18:59-!- Espreon [~espreon@wesnoth/developer/espreon] has quit [Quit: WRYYYYYYYYYYYYYYYYYYYY!]
20100411 22:19:51-!- timotei21 [~timotei@188.24.5.157] has joined #wesnoth-dev
20100411 22:19:58< Crab_> pokhbocee: no, it's good that the questions are asked ;) see calculate_outcome_function in src/ai/formula/function_table.cpp
20100411 22:20:38< gabba> Crab_: I'll be afk for some minutes
20100411 22:20:53< Crab_> pokhbocee: it does  things like      battle_context bc(    lots of parameters    );       std::vector<double> hp_dist = bc.get_attacker_combatant().hp_dist;
20100411 22:21:46< Crab_> gabba: I think that 'what would be dropped if things get bad' will be a nice addition
20100411 22:21:51< Crab_> gabba: e.g., 'optional things'
20100411 22:22:47-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100411 22:22:58< Crab_> gabba: you'd want to drop networking ?
20100411 22:23:55-!- timotei_away [~timotei@193.34.191.4] has quit [Ping timeout: 258 seconds]
20100411 22:29:11< Crab_> gabba: but yes, you're right, most things apart from 'expand the list of things which can be dropped if things don't go well' look ok to me.
20100411 22:39:47-!- thespaceinvader [~chatzilla@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20100411 22:40:44< Ivanovic> gabba: what crab said
20100411 22:41:02< Ivanovic> basically making sure that there is a fallback mode if things go down the drain due to whatever
20100411 22:41:21< Ivanovic> so that you got a "real minimum" as well as "would be really nice to have"
20100411 22:41:38< Ivanovic> where in a normal case, without any extremly bad bugs and the likes, both would get done
20100411 22:42:18-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has quit [Remote host closed the connection]
20100411 22:44:21-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Ping timeout: 265 seconds]
20100411 22:44:50< zookeeper> stats.wesnoth.org gives me a "The server encountered an internal error or misconfiguration and was unable to complete your request."
20100411 22:46:12< Ivanovic> there is no working stats.wesnoth.org!
20100411 22:46:23< Ivanovic> in fact it was not working for quite some time by now
20100411 22:48:11< Crab_> zookeeper: what do you wanted to look up at stats.wesnoth.org ?
20100411 22:48:23< Crab_> maybe it'll be possible to fix it, at least a little...
20100411 22:48:54< zookeeper> Crab_, just thought i'd randomly browse the stats
20100411 22:49:16< zookeeper> ok, well, i just didn't know/remember it's not working anymore
20100411 22:49:34< Crab_> ok :)
20100411 22:49:35< zookeeper> is the data still collected?
20100411 22:49:42< Crab_> yes, it should be collected
20100411 22:49:57< Crab_> at least the uploader gets http 200 ok
20100411 22:50:47< gabba> back
20100411 22:52:08< nagbot> gabba, welcome back
20100411 22:52:20< gabba> thanks nagbot
20100411 22:53:24< timotei21> damn it, scons no more compiles
20100411 22:53:35< Crab_> timotei21: what it says ?
20100411 22:53:47< timotei21> well, it doesn't found anymore sdl... but some days ago worked
20100411 22:53:48< timotei21> :-??
20100411 22:53:55< gabba> Ivanovic, Crab_: ok. Networking's already on the "drop if things go bad list", I'll add a few more elements to the calendar as low-priority. I guess I hadn't really integrated some things such as the "alternate display on tab" to the calendar, it was one of the later ideas.
20100411 22:54:22< Crab_> well, there's lots of stuff on 'future things' list, yes. np with that.
20100411 22:54:56-!- Chusslove [~Chusslove@brsg-d9bee23a.pool.mediaWays.net] has quit [Changing host]
20100411 22:54:56-!- Chusslove [~Chusslove@unaffiliated/chusslove] has joined #wesnoth-dev
20100411 22:55:16< gabba> Crab_: ah yes, but the future things are stuff I'm 95% confident I won't have time to implement, whereas I really hope the alternate display when pressing tab gets in
20100411 22:55:44< loonycyborg> timotei21: Try --config=force
20100411 22:55:47< Crab_> yes, 'alternate display' is a really good way to clear things up
20100411 22:55:51< gabba> Crab_: cf the newest mockups I did a few days ago
20100411 22:55:58< timotei21> tried that already
20100411 22:55:58< Crab_> gabba: yes, I've seen them.
20100411 22:56:10< gabba> the contrary would have surprised me :P
20100411 22:56:10< timotei21> guys, at last I see how painful windows it is
20100411 22:56:16< loonycyborg> timotei21: Post build/config.log
20100411 22:56:18< Crab_> :)
20100411 22:56:23< timotei21> I think I'm no more windows/microsoft lover :))
20100411 22:56:40-!- allefant [~elias@allegro/developer/allefant] has joined #wesnoth-dev
20100411 22:56:43< gabba> well, gotta go now, bye all
20100411 22:56:46< Crab_> bye
20100411 22:56:49-!- gabba [~gabba@70.35.163.161] has left #wesnoth-dev []
20100411 22:56:57< timotei21> bye gabba
20100411 22:57:21< Crab_> timotei21: what are your work plans for the week ?
20100411 22:57:31< Crab_> timotei21: e.g., what you'd try to implement this week ?
20100411 22:57:37< Crab_> and how we can help
20100411 22:57:47< timotei21> well, I think I'll implement starting custom scenario/campaign
20100411 22:57:54< timotei21> right now I'm thinking of
20100411 22:58:02-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit []
20100411 22:58:04< timotei21> how we should start it
20100411 22:58:14< timotei21> -tc <campaign_name> <scenario_name>
20100411 22:58:25< timotei21> something like that
20100411 22:58:28< Crab_> ok. how do you want to do this ? extend current hacky logic or ask mordante about a way to hack into a gui2 event loop ?
20100411 22:58:44< timotei21> well
20100411 22:58:48-!- elias [~elias@allegro/developer/allefant] has quit [Ping timeout: 258 seconds]
20100411 22:59:09< timotei21> why go in that gui2 event loop?
20100411 22:59:23< timotei21> I mean, it's just like jumping straight to the scenario
20100411 22:59:45< Crab_> yes, 'jumping straight' will work, too. but it is a hack.
20100411 23:00:06< timotei21> so, we should "load" the titlescreen, and then load the scenario?
20100411 23:00:19< timotei21> or add a "start custom scenario"?
20100411 23:00:34< Crab_> yes, the 'cleanest route'  is to extend the UI to allow a human to start any scenario from start, in debug mode.
20100411 23:00:43< Crab_> and then, add a way to script that UI
20100411 23:00:44< timotei21> we still need some command line-directly start, from eclipse
20100411 23:00:55< timotei21> isn't that?
20100411 23:00:57< Crab_> that's simple -  on game start, put a list of 'things to do' in a global vector, and then, from a game event loop, read them one-by-one and execute them
20100411 23:01:05< timotei21> yes
20100411 23:01:25< Crab_> the hard part is finding a correct place to check 'what I should do next?'
20100411 23:02:07< Crab_> e.g.: launching..   'what I should do next?' - OPEN CAMPAIGNS;  'what I should do next?' SELECT HTTT;  'what I should do next?' SELECT 'debug scenario 2'
20100411 23:02:41< Crab_> the benefit is that the code path is the same
20100411 23:03:25< Crab_> the side benefit is that it'll be possible to script 'check if all scenarios in all campaigns can be launched correctly from command line', which will help with Ivanovic's release procedure a bit.
20100411 23:03:57< Ivanovic> would be lovely
20100411 23:04:01< Crab_> timotei21: if you try to go that way, you should catch mordante to see the best way to hack those checks for  'what I should do next?' in.
20100411 23:04:17< Ivanovic> then i would go the real hardcore way: "open every campaign in every difficulty level"
20100411 23:04:47< Crab_> Ivanovic: 'every scenario in every campaign in every difficulty level' :)
20100411 23:05:03< Ivanovic> nah, that is a little too hardcore and would take too long...
20100411 23:05:05< Ivanovic> ;)
20100411 23:05:09< Ivanovic> afk for a while
20100411 23:05:19< boucman> and then click on the little cross 1024 times :P
20100411 23:05:23< Crab_> Ivanovic: well, you can start a command, then go afk, and return to see the result :)
20100411 23:05:37< Crab_> boucman: no, exiting wesnoth will be read from command-line arg, too.
20100411 23:05:42< Ivanovic> boucman: have you never heard of the command "killall"?
20100411 23:05:50< Crab_> Ivanovic: ah, a brilliant hack :)
20100411 23:05:59< Crab_> boucman: but, actually, I was thinking of gui2 lua scripting.
20100411 23:06:03< boucman> Crab_: and it's read from the command line as you said :P
20100411 23:06:11< timotei21> I think I understand what you are saying
20100411 23:06:36< Crab_> boucman: e.g. ./wesnoth --script start-everything-up-and-check-so-that-release-manager-can-drink-beer.lua
20100411 23:06:56< timotei21> well, I'm so sorry, but I can't stay any longer, I need to go
20100411 23:07:01< Crab_> timotei21: but you should code a proof-of-concept only.
20100411 23:07:02< timotei21> we'll talk tomorrow then, ok?
20100411 23:07:08< timotei21> yes
20100411 23:07:13< timotei21> I'll think about i
20100411 23:07:15< timotei21> it*
20100411 23:07:16< Crab_> timotei21: which'll just get the campaign name and scenario name from command line and launch them
20100411 23:07:18< Crab_> bye :)
20100411 23:07:20< shadowmaster> ...updating an iterator in python isn't as easy as ++i?
20100411 23:07:28< timotei21> good night everyone
20100411 23:07:37< timotei21> Crab_, tomorrow at what time you'll be online?
20100411 23:07:39< timotei21> in UTC time?
20100411 23:07:55< Crab_> timotei21: depends. in +11 h, might be.
20100411 23:08:02< timotei21> ok
20100411 23:08:05< shadowmaster> http://wesnoth.pastebin.com/yefLg65i
20100411 23:08:13< shadowmaster> I get "0: <string blah blah>" on every line :/
20100411 23:09:01< allefant> for i, a in enumerate(ign):
20100411 23:09:06< allefant> that should work
20100411 23:09:10< shadowmaster> allefant: ! you are the person I need right now!
20100411 23:09:12< allefant> or else: i += 1
20100411 23:09:28-!- timotei21 [~timotei@188.24.5.157] has quit [Quit: Leaving]
20100411 23:09:42< shadowmaster> I'll file a bug report first though
20100411 23:11:09< Crab_> shadowmaster: but note that with enumerate you'll get 0-based numbering.
20100411 23:13:33< YogiHH> fendrin, you there?
20100411 23:13:41< fendrin> YogiHH: yes
20100411 23:13:43< shadowmaster> allefant: do you think you can check https://gna.org/bugs/?15846 ?
20100411 23:14:20< YogiHH> fendrin: i submitted the fork to gna and are waiting for acceptance atm. I will tell you once things get going.
20100411 23:14:34< allefant> shadowmaster: hm, let me try recover my gna password :P
20100411 23:15:02< fendrin> YogiHH: :-)
20100411 23:15:04< boucman> hey allefant, long time no see :)
20100411 23:15:33< shadowmaster> allefant: and if you want a test case, download shadowmasters_test_addon from the 1.8 server.
20100411 23:15:36< shadowmaster> (it has the passphrase!)
20100411 23:16:01< allefant> hi boucman :)
20100411 23:28:58-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100411 23:30:11-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20100411 23:31:37-!- YogiHH [YogiHH@wesnoth/developer/yogihh] has left #wesnoth-dev []
20100411 23:35:38< CIA-32> elias * r42111 /trunk/data/tools/wesnoth_addon_manager: possible fix for bug #15846 - _server.ign not quite working
20100411 23:35:46< allefant> shadowmaster: can you re-upload the testcase? my testing overwrote it :P
20100411 23:42:03< CIA-32> ivanovic * r42112 /trunk/po/ (8 files in 8 dirs): updated Slovak translation
20100411 23:42:04< CIA-32> ivanovic * r42113 /branches/1.8/po/ (8 files in 8 dirs): updated Slovak translation
20100411 23:42:23-!- lfernando [~luiz@201-43-134-54.dsl.telesp.net.br] has quit [Quit: Ex-Chat]
20100411 23:46:04-!- Elsewise [~Dmitry@80.233.231.12] has quit [Ping timeout: 276 seconds]
20100411 23:48:51-!- thespaceinvader [~chatzilla@wesnoth/artist/thespaceinvader] has quit [Ping timeout: 245 seconds]
20100411 23:51:06-!- allefant [~elias@allegro/developer/allefant] has quit [Quit: Leaving]
20100411 23:58:49-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Quit: Leaving.]
20100411 23:59:47-!- Polarina [~Polarina@wesnoth/translator/Polarina] has quit [Remote host closed the connection]
--- Log closed Mon Apr 12 00:00:47 2010