--- Log opened Mon Mar 22 00:00:10 2010
--- Day changed Mon Mar 22 2010
20100322 00:00:10< YogiHH> good night everybody
20100322 00:00:21-!- YogiHH [YogiHH@wesnoth/developer/yogihh] has left #wesnoth-dev []
20100322 00:02:32-!- schumi [~a11848c0@gateway/web/freenode/x-pzxgqdhhhnbckyzb] has quit []
20100322 00:07:25-!- AnMaster [~AnMaster@unaffiliated/anmaster] has quit [Ping timeout: 256 seconds]
20100322 00:09:26-!- AnMaster [~AnMaster@unaffiliated/anmaster] has joined #wesnoth-dev
20100322 00:25:30< happygrue> YogiHH:  I see I missed you, but the bug is fixed, so maybe it doesn't matter?
20100322 00:34:27-!- Blueblaze [~nick@adsl-76-202-22-67.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 260 seconds]
20100322 00:36:27-!- Darkas [~quassel@ppp-93-104-154-111.dynamic.mnet-online.de] has quit [Remote host closed the connection]
20100322 00:36:38< ancestral> :(
20100322 00:36:45< Ivanovic> re
20100322 00:36:49< ancestral> The first time I've logged into multiplayer in a while
20100322 00:36:50< shadowmaster> fwd
20100322 00:37:03-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has quit [Remote host closed the connection]
20100322 00:37:16< ancestral> So many different panes and no lines or boxes or anything separating them
20100322 00:37:27< ancestral> Buttons with no padding or margins
20100322 00:37:45< ancestral> Or maybe that's not a button…
20100322 00:39:03< ancestral> Should I be able to scroll the games window with my scroll wheel?
20100322 00:39:38< shadowmaster> ilor might know
20100322 00:39:49< shadowmaster> I guess that you should.
20100322 00:40:11-!- ptn777 [~ptn@201.230.126.89] has joined #wesnoth-dev
20100322 00:40:30-!- ptn777 [~ptn@201.230.126.89] has quit [Client Quit]
20100322 00:40:39< ancestral> I can't :(
20100322 00:40:55< ancestral> Oh
20100322 00:40:57< ancestral> I see
20100322 00:41:23< ancestral> I have to click in the window to scroll. I assumed the mouse cursor in the window was sufficient.
20100322 00:41:51< shadowmaster> okay, that sounds like it usually works on Windows
20100322 00:42:02< shadowmaster> *like the way
20100322 00:42:16< shadowmaster> not as comfortable as in Gtk2 and Qt4
20100322 00:43:33< ancestral> Hmm
20100322 00:43:48< ancestral> Music should be playing while in the lobby?
20100322 00:43:54< ancestral> That changed iirc
20100322 00:44:14< ancestral> Conscious decision?
20100322 00:44:27< CIA-80> ivanovic * r41663 /trunk/po/wesnoth-units/de.po: updated German translation
20100322 00:44:40< ilor> ancestral: probably an oversight
20100322 00:45:18< ancestral> I'm not big with the music and the lobby myself *shrugs*
20100322 00:51:02< Ivanovic> i am off to bed now, n8
20100322 00:51:35< Ivanovic> ilor: have you read todays logs?
20100322 00:51:49< Ivanovic> ilor: regarding the "wrong id stuff" that boucman mentioned
20100322 00:52:04< Ivanovic> i think mordante left a highlight for you, so grepping the logs should work
20100322 00:52:07< ilor> Ivanovic: some of it, boucman's issue was likely among the ones I fixed after rx1
20100322 00:52:18< Ivanovic> the issue was in trunk head
20100322 00:52:20< Ivanovic> so, no
20100322 00:52:33< ilor> well he later said he was using rc1
20100322 00:52:45< Ivanovic> <boucman> [18:45:09] I have tons of messages like
20100322 00:52:48< Ivanovic> <boucman> [18:45:12] 20100321 16:36:51 warning engine: User KenRage has unknown game_id: 2482
20100322 00:52:59-!- qemqemqem_ [~quassel@urwireless-dhcp-128-151-28-248.wireless.rochester.edu] has quit [Ping timeout: 245 seconds]
20100322 00:53:34< ilor> regarding those it's mostly harmless and I've answered that
20100322 00:53:35< Ivanovic> [00:32:29] <mordante> [18:47:51] regarding the warnings, I've no clue best ask ilor about them
20100322 00:53:42< Ivanovic> okay
20100322 00:54:01< Ivanovic> really off to bed now, n8
20100322 00:54:05< ilor> also, off to bed, catch you tomorrow
20100322 00:58:23-!- DesertPanther_ [~Khalid@41.234.234.37] has quit [Quit: Leaving]
20100322 01:14:36-!- UnknownDevice [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100322 01:16:26-!- UnknownDevice [~UnknownDe@188.141.93.69] has quit [Client Quit]
20100322 01:17:00-!- UnknownDevice [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100322 01:18:24< ancestral> So when do I lose the game?
20100322 01:18:38< ancestral> I played multiplayer against three AI's
20100322 01:18:52< ancestral> Free-for-all
20100322 01:19:13-!- UnknownDevice [~UnknownDe@188.141.93.69] has quit [Client Quit]
20100322 01:19:41< ancestral> I wasn't defeated until two AI's were dead
20100322 01:19:50< ancestral> Even though I lost my leader pretty early on
20100322 01:20:01-!- UnknownDevice [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100322 01:20:01< ancestral> When I was "defeated" I still had units left
20100322 01:20:52< Gambit> :s It should end when all the leaders on a team are dead.
20100322 01:21:09< ancestral> This was free-for-all
20100322 01:21:16< Gambit> Weird.
20100322 01:21:20< ancestral> So I should have died way back when, right?
20100322 01:21:23< ancestral> When my leader died
20100322 01:21:33< Gambit> Is the scenario FFA by default?
20100322 01:21:40-!- UnknownDevice [~UnknownDe@188.141.93.69] has quit [Client Quit]
20100322 01:22:04< Gambit> That shouldn't matter but maybe it was going by the "map settings" for teams defeat.
20100322 01:22:04-!- UnknownDevice [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100322 01:22:09-!- Dun1031[AFK] [~chatzilla@24.42.243.79] has quit [Ping timeout: 245 seconds]
20100322 01:22:28< Gambit> Though if that were the case someone surely should have caught it by now because it'd be everywhere.
20100322 01:22:56< ancestral> I suppose I'll post up my save
20100322 01:24:03-!- UnknownDevice_ [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100322 01:24:10< ancestral> In a reply, can I go back a turn?
20100322 01:24:35< Gambit> So you lost right when one of the other AI leaders died?
20100322 01:25:21< ancestral> Yes
20100322 01:25:29< Gambit> What scenario and which AI?
20100322 01:25:32< ancestral> After two of them died and there was one remaining still
20100322 01:25:35< ancestral> Lagoon
20100322 01:25:38< ancestral> Default AI
20100322 01:25:44< Gambit> I mean which AI team?
20100322 01:25:48< Gambit> Which side number?
20100322 01:26:24< Gambit> nevermind on my stupid theory. It is FFA by default.
20100322 01:27:13< ancestral> Hmm
20100322 01:27:24-!- UnknownDevice [~UnknownDe@188.141.93.69] has quit [Ping timeout: 252 seconds]
20100322 01:27:25< ancestral> I'll post it on the forums if you want to download it
20100322 01:27:30< ancestral> But the blue player, I think ai3
20100322 01:27:34-!- UnknownDevice_ [~UnknownDe@188.141.93.69] has left #wesnoth-dev []
20100322 01:27:37< ancestral> or maybe it's 2
20100322 01:27:41< ancestral> I was number 1
20100322 01:27:50< Gambit> blue is 2 by default.
20100322 01:28:59< ancestral> So just to be clear
20100322 01:29:09< ancestral> Should I lose when I lose my leader, or when I lose all my guys?
20100322 01:29:14-!- Dun1031[AFK] [~chatzilla@24.42.243.79] has joined #wesnoth-dev
20100322 01:29:17< ancestral> in a FFA
20100322 01:29:58< Gambit> Your leader.
20100322 01:31:01< Gambit> Wait wait
20100322 01:31:05< Gambit> You said there was one AI left?
20100322 01:31:07< CIA-80> espreon * r41664 /trunk/ (2 files in 2 dirs): Updated localized images for the Spanish translation.
20100322 01:31:11< Gambit> You lost because he won :P
20100322 01:31:30< Gambit> I think you do get to play to the end.
20100322 01:31:38< Gambit> in MP
20100322 01:31:58< Gambit> As a local game with AI should follow the same rules as an online game with humans.
20100322 01:32:14< Gambit> And it wouldn't make sense for an FFA to end in multiplayer just because side one lost his leader.
20100322 01:32:28< Gambit> So actually that behaviour wasn't weird at all.
20100322 01:32:52< Gambit> You should quickly ninja your thread
20100322 01:33:28< ancestral> Ah
20100322 01:33:29< ancestral> Okay
20100322 01:33:32< ancestral> Hmm
20100322 01:33:36< ancestral> Although
20100322 01:33:38< ancestral> Does it matter?
20100322 01:33:42< ancestral> When there are 3 ai's?
20100322 01:33:50< ancestral> Why not just end it?
20100322 01:33:58< ancestral> Or better
20100322 01:34:04< Gambit> Ah. so rework your thread into an idea proposal ;)
20100322 01:34:09< ancestral> Why not display a "you've been defeated" when you are defeated
20100322 01:34:12< Gambit> But no I don't think it should just end it.
20100322 01:34:14< ancestral> Gambit: Good call
20100322 01:34:24< Gambit> Some people with no friends test their new eras and factions with AI
20100322 01:34:30< Gambit> and will want to keep watching even when they lost.
20100322 01:35:24< ancestral> True
20100322 01:35:33< Gambit> Also in multiplayer "Why not display a "you've been defeated" when you are defeated" is because you can still help your ally with your left over units
20100322 01:35:38< Gambit> or in the case of FFA take revenge.
20100322 01:36:23< Gambit> It's not unheard of to one to be killed by a leaderless team in MP FFA's.
20100322 01:38:10< ancestral> How's this? http://forums.wesnoth.org/viewtopic.php?f=12&t=29275
20100322 01:38:21< ancestral> You can still play, of course
20100322 01:38:39< CIA-80> espreon * r41665 /branches/resources/image-localization/es/logo.xcf: Updated the xcf for the Spanish BfW logo.
20100322 01:40:11< ancestral> Does anyone know where the image is for the border on the main screen?
20100322 01:40:22-!- thespaceinvader [~chatzilla@wesnoth/artist/thespaceinvader] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6/20100115144158]]
20100322 01:40:54< Gambit> I've not been able to locate some pieces of the GUI
20100322 01:41:00< Gambit> like the loading bar.
20100322 01:41:44< ancestral> It wouldn't surprise me if some of that is drawn
20100322 01:41:59< ancestral> But that border I'd be surprised
20100322 01:42:01-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz]
20100322 01:42:40-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20100322 01:43:29< Gambit> install-dir\images\misc\
20100322 01:43:35< Gambit> ^^^ has a lot of borders.
20100322 01:45:55< ancestral> It does have borders
20100322 01:46:17< ancestral> and images/dialogs
20100322 01:46:54< ancestral> I don't see it though :(
20100322 01:54:29-!- UnknownDevice [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100322 01:57:10-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100322 01:57:27-!- UnknownDevice_ [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100322 01:58:29-!- UnknownDevice_ [~UnknownDe@188.141.93.69] has left #wesnoth-dev []
20100322 01:58:55-!- UnknownDevice__ [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100322 01:59:34-!- UnknownDevice [~UnknownDe@188.141.93.69] has quit [Ping timeout: 258 seconds]
20100322 01:59:57-!- UnknownDevice__ [~UnknownDe@188.141.93.69] has quit [Client Quit]
20100322 02:02:03-!- Appleman1234 [~Appleman1@131.181.18.237] has quit [Ping timeout: 252 seconds]
20100322 02:02:27-!- UnknownDevice [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100322 02:06:45-!- UnknownDevice__ [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100322 02:07:24-!- Dun1031[AFK]_ [~chatzilla@24.42.243.79] has joined #wesnoth-dev
20100322 02:07:51-!- Dun1031[AFK]_ [~chatzilla@24.42.243.79] has quit [Read error: Connection reset by peer]
20100322 02:08:28-!- Appleman1234 [~Appleman1@131.181.18.237] has joined #wesnoth-dev
20100322 02:08:41-!- Dun1031[AFK]_ [~chatzilla@24.42.243.79] has joined #wesnoth-dev
20100322 02:09:09-!- Dun1031[AFK] [~chatzilla@24.42.243.79] has quit [Ping timeout: 240 seconds]
20100322 02:09:14-!- Dun1031[AFK]_ is now known as Dun1031[AFK]
20100322 02:09:52-!- schumi [~a11848c0@gateway/web/freenode/x-nnpnbkzqwnvoquhs] has joined #wesnoth-dev
20100322 02:10:06-!- UnknownDevice [~UnknownDe@188.141.93.69] has quit [Ping timeout: 268 seconds]
20100322 02:10:14-!- UnknownDevice__ is now known as UnknownDevice
20100322 02:12:53< CIA-80> espreon * r41666 /trunk/ (data/core/images/help/l10n/es/tooltip.png l10n-track): Updated localized images for the Spanish translation.
20100322 02:13:13< UnknownDevice> Hello
20100322 02:13:48< Espreon> Hello.
20100322 02:15:02< UnknownDevice> I saw that Battle for Wesnoth went up as part of the Google Summer of Code, so I came in to introduce myself since I'm interested in applying.
20100322 02:15:30< schumi> hello
20100322 02:15:39< schumi> me too!
20100322 02:15:40< Espreon> UnknownDevice: OK. Out of curiosity, what did you want to do for GSoC?
20100322 02:16:11-!- UnknownDevice__ [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100322 02:16:32< UnknownDevice__> Sorry about that, my connection is acting a bit dodgy tonight.
20100322 02:16:45< Espreon> OK.
20100322 02:17:06< Espreon> Out of curiosity, what were you planning to do for GSoC?
20100322 02:17:50< UnknownDevice__> Being honest I'm not really too sure yet, I've just been looking at the list of ideas for projects. 
20100322 02:17:59< UnknownDevice__> On the wiki that is
20100322 02:18:09< schumi> i'm still thinking of it
20100322 02:18:23< UnknownDevice__> You're looking to apply too schumi? 
20100322 02:18:29< schumi> yes
20100322 02:18:32< UnknownDevice__> Excellent :)
20100322 02:18:38< Espreon> Hmmm... I see...
20100322 02:18:47< schumi> i want to learn more about the projects on the wiki
20100322 02:19:09-!- UnknownDevice [~UnknownDe@188.141.93.69] has quit [Ping timeout: 240 seconds]
20100322 02:19:36< UnknownDevice__> At the moment my thoughts are that it might be better to try and use one of the projects from the list since they're things people want.
20100322 02:19:54< UnknownDevice__> Rather then coming up with someone for the sake of it, and that being just less important
20100322 02:20:01< UnknownDevice__> *something
20100322 02:20:04-!- UnknownDevice__ is now known as UnknownDevice
20100322 02:20:16< UnknownDevice> Still not really sure though
20100322 02:20:27-!- Dun1031[AFK] [~chatzilla@24.42.243.79] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6/20100115144158]]
20100322 02:21:18< UnknownDevice> My background is mainly in C++ for games.
20100322 02:22:59< Zarel> Guys.
20100322 02:23:01< Zarel> http://www.wesnoth.org/forum/viewtopic.php?p=416811#p416811
20100322 02:23:31 * Espreon rolls his eyes
20100322 02:24:20< shadowmaster> bah
20100322 02:24:26< ancestral> In D&D there's a house rule where if someone dies they re-roll a new character one level lower
20100322 02:25:14< ancestral> Thing is
20100322 02:25:24< Zarel> Hmm, we could do the same thing here. Subtract a certain amount of experience from the "resurrected" unit.
20100322 02:25:38< Zarel> Perhaps not an entire level, what with a level cap of 3-4 compared to D&D's level cap of 20.
20100322 02:28:03< ancestral> Allowing people to waste their money on dead units is allowing people to run out of money
20100322 02:28:08< Zarel> We could simply reset experience to 0. Would prevent any advantage gained by resurrecting a unit close to leveling.
20100322 02:28:34< Zarel> People will run out of money if they play poorly regardless.
20100322 02:28:54< Zarel> This is mostly for people who get maybe one or two dead high-level units they want to resurrect.
20100322 02:29:05-!- Appleman1234 [~Appleman1@131.181.18.237] has quit [Ping timeout: 256 seconds]
20100322 02:29:09 * shadowmaster wonders who are "we" ;)
20100322 02:29:18< Zarel> Maybe make it only apply to high-level (L2+) units, at that, so they don't overpay for L1 units.
20100322 02:29:22< Gambit> Zarel and Zarel's mouse.
20100322 02:29:44< Zarel> "we" i.e. "the Wesnoth project". Fun thing about open-source is that everyone who wants to be involved can be involved, eh? ;)
20100322 02:30:22< Gambit> But I don't want "advantage gained by resurrecting a unit close to leveling" removed :(
20100322 02:30:43< shadowmaster> Zarel: yeah. Whether it makes it to mainline or not doesn't matter.
20100322 02:31:13< ancestral> Can it be done with WML?
20100322 02:32:53< shadowmaster> yes, it can
20100322 02:33:19< shadowmaster> although changing a unit's type tends to be affected by interesting new every development cycle
20100322 02:33:20< Zarel> ancestral: Probably.
20100322 02:33:41< shadowmaster> the rest is perfectly doable
20100322 02:34:26< ancestral> You'd have to be sure it only affects when they die due to negligence
20100322 02:34:41< ancestral> In some scenarios it may be where they are supposed to die, or they just don't make it to the next scenario
20100322 02:34:59< ancestral> Storyline plot
20100322 02:35:42-!- UnknownDevice__ [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100322 02:37:45< ancestral> Zarel: How about http://www.wesnoth.org/forum/viewtopic.php?f=12&t=29275
20100322 02:38:41-!- esr [~chatzilla@static-71-162-243-5.phlapa.fios.verizon.net] has joined #wesnoth-dev
20100322 02:39:18-!- UnknownDevice [~UnknownDe@188.141.93.69] has quit [Ping timeout: 276 seconds]
20100322 02:39:23-!- UnknownDevice__ is now known as UnknownDevice
20100322 02:39:29-!- esr [~chatzilla@static-71-162-243-5.phlapa.fios.verizon.net] has quit [Changing host]
20100322 02:39:29-!- esr [~chatzilla@wesnoth/developer/esr] has joined #wesnoth-dev
20100322 02:41:48-!- Gambit [~Gambit@pa-67-234-116-244.dhcp.embarqhsd.net] has quit [Quit: Leaving.]
20100322 02:45:37-!- Appleman1234 [~Appleman1@131.181.47.4] has joined #wesnoth-dev
20100322 02:49:04-!- UnknownDevice__ [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100322 02:51:58-!- shenmage [~dbartkow8@128.151.220.142] has quit [Ping timeout: 276 seconds]
20100322 02:52:01< CIA-80> espreon * r41667 /trunk/ (changelog players_changelog): Updated the changelogs.
20100322 02:52:07-!- UnknownDevice [~UnknownDe@188.141.93.69] has quit [Ping timeout: 260 seconds]
20100322 02:52:13-!- UnknownDevice__ is now known as UnknownDevice
20100322 02:54:37-!- jabagawee [~andrew@ppp-71-129-109-176.dsl.irvnca.pacbell.net] has joined #wesnoth-dev
20100322 02:54:38-!- andrew__ [~andrew@adsl-71-130-125-219.dsl.irvnca.pacbell.net] has quit [Read error: Connection reset by peer]
20100322 02:56:00< Zarel> Well, yes.
20100322 02:56:27-!- shenmage [~dbartkow8@128.151.220.142] has joined #wesnoth-dev
20100322 02:58:32-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Read error: Connection timed out]
20100322 02:58:54-!- Zarel [~Zarel@c-24-118-17-28.hsd1.mn.comcast.net] has joined #wesnoth-dev
20100322 02:58:55-!- Zarel [~Zarel@c-24-118-17-28.hsd1.mn.comcast.net] has quit [Changing host]
20100322 02:58:55-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100322 03:00:46-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has quit [Remote host closed the connection]
20100322 03:01:39-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has joined #wesnoth-dev
20100322 03:02:37< [Relic]> Hello :)
20100322 03:03:31< schumi> hi
20100322 03:04:52-!- UnknownDevice__ [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100322 03:07:30< ancestral> Howdy
20100322 03:08:01-!- UnknownDevice [~UnknownDe@188.141.93.69] has quit [Ping timeout: 265 seconds]
20100322 03:08:20-!- UnknownDevice__ is now known as UnknownDevice
20100322 03:09:38-!- garrik [~garrik@174-143-173-186.static.cloud-ips.com] has joined #wesnoth-dev
20100322 03:12:49< UnknownDevice> What time about is the channel usually most active? :)
20100322 03:13:33< shadowmaster> between 12:00 UTC and 00:00 UTC
20100322 03:14:08< UnknownDevice> Ok, thank! :)
20100322 03:15:05< UnknownDevice> I wanted give a proper intoduction, but it seems I will have to leave it until tomorrow.
20100322 03:15:10< schumi> in workdays it is more active?
20100322 03:15:56< shadowmaster> it varies
20100322 03:16:20< shadowmaster> UnknownDevice: have you read http://wiki.wesnoth.org/SummerOfCodeIdeas btw ?
20100322 03:16:40< UnknownDevice> Yep
20100322 03:17:37< UnknownDevice> I was just reading it, which is why I came here to introduce myself.
20100322 03:17:57-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100322 03:18:04-!- jekintrivedi [~root@116.72.242.56] has joined #wesnoth-dev
20100322 03:20:58-!- UnknownDevice__ [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100322 03:21:32< Zarel> A good idea.
20100322 03:23:42-!- UnknownDevice [~UnknownDe@188.141.93.69] has quit [Ping timeout: 240 seconds]
20100322 03:23:55-!- UnknownDevice__ is now known as UnknownDevice
20100322 03:25:51-!- haoyu [~bhy@cm26.delta25.maxonline.com.sg] has joined #wesnoth-dev
20100322 03:27:45< schumi> bye all
20100322 03:27:53< UnknownDevice> Bye
20100322 03:28:31< schumi> as UnknowmDevice, I will wait for tomorrow to introduce myself
20100322 03:28:51< UnknownDevice> See you tomorrow no doubt :)
20100322 03:29:46< schumi> \quit
20100322 03:29:50< schumi> ops
20100322 03:30:13-!- schumi [~a11848c0@gateway/web/freenode/x-nnpnbkzqwnvoquhs] has quit []
20100322 03:30:33-!- UnknownDevice__ [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100322 03:32:48< UnknownDevice__> Good night everyone
20100322 03:32:57< Espreon> Goodnight.
20100322 03:33:03-!- UnknownDevice__ [~UnknownDe@188.141.93.69] has left #wesnoth-dev []
20100322 03:33:51-!- UnknownDevice [~UnknownDe@188.141.93.69] has quit [Ping timeout: 260 seconds]
20100322 03:34:07-!- esr [~chatzilla@wesnoth/developer/esr] has quit [Remote host closed the connection]
20100322 03:47:52-!- jekintrivedi [~root@116.72.242.56] has quit [Ping timeout: 276 seconds]
20100322 03:47:57-!- qemqemqem_ [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has joined #wesnoth-dev
20100322 03:53:09-!- shenmage [~dbartkow8@128.151.220.142] has quit [Ping timeout: 252 seconds]
20100322 03:55:47-!- shenmage [~dbartkow8@128.151.220.142] has joined #wesnoth-dev
20100322 03:57:10-!- noy [~Noy@wesnoth/developer/noy] has quit [Read error: Connection reset by peer]
20100322 04:03:42-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100322 04:07:45-!- wesbot changed the topic of #wesnoth-dev to: 1.8 commits allowed till March, 24th, 12:00 GMT, afterwards "testfreeze" | string/feature freeze active! | 70 bugs, 255 feature requests, 14 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100322 04:09:37-!- Blueblaze [~nick@adsl-76-202-22-67.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100322 04:31:59-!- esr [~chatzilla@static-71-162-243-5.phlapa.fios.verizon.net] has joined #wesnoth-dev
20100322 04:32:18-!- esr [~chatzilla@static-71-162-243-5.phlapa.fios.verizon.net] has quit [Changing host]
20100322 04:32:18-!- esr [~chatzilla@wesnoth/developer/esr] has joined #wesnoth-dev
20100322 04:34:42-!- ancestral [~ancestral@97-116-105-225.mpls.qwest.net] has quit [Ping timeout: 240 seconds]
20100322 04:39:41-!- ancestral [~ancestral@97-116-105-225.mpls.qwest.net] has joined #wesnoth-dev
20100322 05:01:22-!- ilor [~ilor@wesnoth/developer/ilor] has quit [Ping timeout: 246 seconds]
20100322 05:20:02-!- Appleman1234 [~Appleman1@131.181.47.4] has quit [Ping timeout: 268 seconds]
20100322 05:35:18-!- Espreon [~espreon@wesnoth/developer/espreon] has quit [Quit: WRYYYYYYYYYYYYYYYYYYYY!]
20100322 05:37:49-!- noy [~Noy@wesnoth/developer/noy] has quit [Read error: Connection reset by peer]
20100322 05:39:47< ancestral> #wesnoth-music
20100322 05:49:27-!- qemqemqem_ [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has quit [Ping timeout: 265 seconds]
20100322 05:51:09-!- fkhodkov [~fedor76@ppp-78-24-26-36-bras0.istra.ru] has quit [Ping timeout: 240 seconds]
20100322 05:54:27-!- Appleman1234 [~Appleman1@131.181.47.4] has joined #wesnoth-dev
20100322 05:59:28-!- ancestral [~ancestral@97-116-105-225.mpls.qwest.net] has quit [Quit: And that’s the end of THAT chapter.]
20100322 06:02:34-!- Appleman1234 [~Appleman1@131.181.47.4] has quit [Ping timeout: 245 seconds]
20100322 06:12:01-!- Spidy [~NickFerra@75.139.189.187] has joined #wesnoth-dev
20100322 06:17:15-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Quit: Leaving]
20100322 06:20:14-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100322 06:26:02-!- ancestral [~ancestral@97-116-105-225.mpls.qwest.net] has joined #wesnoth-dev
20100322 06:28:52-!- Blueblaze [~nick@adsl-76-202-22-67.dsl.hstntx.sbcglobal.net] has quit [Remote host closed the connection]
20100322 06:29:05-!- Blueblaze [~nick@adsl-76-202-22-67.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100322 06:35:58-!- ancestral [~ancestral@97-116-105-225.mpls.qwest.net] has quit [Quit: ancestral]
20100322 06:45:58-!- Spidy [~NickFerra@75.139.189.187] has quit [Ping timeout: 276 seconds]
20100322 06:49:08-!- fkhodkov [~fedor76@ppp-188-65-8-66-bras1.istra.ru] has joined #wesnoth-dev
20100322 06:51:47-!- ancestral [~ancestral@97-116-105-225.mpls.qwest.net] has joined #wesnoth-dev
20100322 07:02:23-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100322 07:04:45-!- ilor [~ilor@wesnoth/developer/ilor] has joined #wesnoth-dev
20100322 07:07:04-!- Crab__ [~Crab@195.238.92.115] has joined #wesnoth-dev
20100322 07:14:48-!- Appleman1234 [~Appleman1@131.181.18.32] has joined #wesnoth-dev
20100322 07:28:55-!- fkhodkov [~fedor76@ppp-188-65-8-66-bras1.istra.ru] has quit [Remote host closed the connection]
20100322 07:43:36< fendrin> hi Crab__
20100322 07:44:08< Crab__> hi fendrin
20100322 07:44:33< fendrin> I have know 2 dialogues, one for creating campaigns, one for scenarios.
20100322 07:44:40< fendrin> s/know/now
20100322 07:45:13< fendrin> But I don't get how I can deploy the default file structure to the new campaign project.
20100322 07:46:44< Crab__> fendrin: I suggest using maven archetypes for that
20100322 07:46:59< Crab__> fendrin: it will allow us to host a repository of templates
20100322 07:47:25< fendrin> I have used maven once
20100322 07:47:28< Crab__> and then, anyone with internet access, would be able to use your dialog wizard to generate a new project, downloading the template on first access
20100322 07:48:09< Crab__> or, if we're not interested in having the ability to dynamically update the templates independently of the plugin, we can just bundle the template inside the plugin.
20100322 07:48:57< Crab__> (using any templating system such as http://velocity.apache.org/engine/index.html )
20100322 07:49:21-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Quit: crimson_penguin]
20100322 07:49:28< fendrin> Do you have an idea how Eclipse plugins interact with maven?
20100322 07:49:47< Crab__> there's a maven plugin for eclipse
20100322 07:50:56< Crab__> our 'create a new project from template' will be a 'mvn archetype:generate' command
20100322 07:52:38< Crab__> see http://m2eclipse.sonatype.org/extension-points-and-api.html
20100322 07:53:47-!- CheeseLord [~chatzilla@188-222-208-250.zone13.bethere.co.uk] has joined #wesnoth-dev
20100322 07:54:01-!- CheeseLord [~chatzilla@188-222-208-250.zone13.bethere.co.uk] has quit [Client Quit]
20100322 07:54:28< fendrin> Crab__: Do you have an example how the maven template for a wesnoth campaign would look like?
20100322 07:55:32-!- Spidy [~NickFerra@71.93.0.21] has joined #wesnoth-dev
20100322 07:55:55< Crab__> fendrin: just a normal wesnoth campaign, plus a pom.xml file with meta-information, plus some ${} variables inside files where we'll need to substitute some properties in the template with actual values.
20100322 07:56:22< fendrin> Crab__: We also need to add java, maven and eclipse plugin to the wesnoth gsoc information.
20100322 07:57:03< fendrin> Crab__: I am interested in an example how a maven template file that describes a directory structure looks like.
20100322 07:58:26< fendrin> And I don't know if that es really needed. It's only a _main.cfg, a scenario directory, units, images not much more.
20100322 07:58:53-!- Appleman1234 [~Appleman1@131.181.18.32] has quit [Ping timeout: 256 seconds]
20100322 07:58:56< fendrin> Only for this few informations it's not worth to integrate maven into the Wesnoth Eclipse plugin.
20100322 07:59:23< fendrin> So I guess you plan to have templates for several other information stored in maven templates?
20100322 07:59:45-!- fkhodkov [~fedor76@ppp-188-65-8-66-bras1.istra.ru] has joined #wesnoth-dev
20100322 08:01:03< Crab__> fendrin: well, the main advantages of using maven is the ability to update those templates independently of our plugin. in fact, anyone would be able to create a template, independently of us.
20100322 08:01:10-!- silene [~plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20100322 08:01:31< Crab__> fendrin: and it'll be able to work independently of eclipse, from the command line, as well.
20100322 08:02:00< Crab__> fendrin: plus, we can be able to hijack maven release goals with 'deploy to addon server' stuff.
20100322 08:02:20< Crab__> fendrin: all this is possible using pure eclipse, too.
20100322 08:03:12< fendrin> But still I don't got why it is worth the work only for a directory structure plus one file.
20100322 08:03:32< Crab__> fendrin: then let us don't do this work now, and bundle everything inside our plugin.
20100322 08:03:43< fendrin> which work?
20100322 08:03:55< Crab__> fendrin: 'using mvn as a backend'
20100322 08:04:12< Crab__> fendrin: e.g. we can try to be simple now, and if it works, that's good.
20100322 08:04:32< fendrin> Okay, let's import it into svn
20100322 08:04:47< Crab__> ok
20100322 08:06:33< CIA-80> fendrin * r41668 /branches/resources/eclipse_plugin/: Added directory for the eclipse plugin.
20100322 08:09:22-!- noy_ [~Noy@d99-199-3-111.bchsia.telus.net] has joined #wesnoth-dev
20100322 08:09:22-!- noy_ [~Noy@d99-199-3-111.bchsia.telus.net] has quit [Changing host]
20100322 08:09:22-!- noy_ [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100322 08:10:11< CIA-80> fendrin * r41669 /branches/resources/eclipse_plugin/ (17 files in 7 dirs): Some basic experiments with eclipse plugin Wizards.
20100322 08:11:21< CIA-80> fendrin * r41670 /branches/resources/eclipse_plugin/icons/ (. sample.gif): Icons folder added.
20100322 08:12:28< CIA-80> fendrin * r41671 /branches/resources/eclipse_plugin/build.properties: build.properties file
20100322 08:13:56< CIA-80> fendrin * r41672 /branches/resources/eclipse_plugin/ (4 files in 2 dirs): Hidden eclipse files.
20100322 08:14:09< fendrin> Crab__: Done
20100322 08:14:31< Crab__> that's good :)
20100322 08:15:29-!- Appleman1234 [~Appleman1@131.181.47.3] has joined #wesnoth-dev
20100322 08:25:45-!- ilor [~ilor@wesnoth/developer/ilor] has quit [Ping timeout: 276 seconds]
20100322 08:29:15-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20100322 08:33:08-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100322 08:37:08< Crab__> fendrin: checked it out.  imo, it's better to rename 'Campaign' to 'Wesnoth campaign' in the new project wizard - to allow to quicky select all wesnoth artifacts by typing 'we' in the filter box
20100322 08:37:25< fendrin> Crab__: Agreed.
20100322 08:37:37< fendrin> Don't expect to much from the commited stuff.
20100322 08:37:48< Crab__> ok. still, it's a good start :)
20100322 08:38:52< fendrin> For the scenario creation dialogue it would be useful to have that write/read wml java class.
20100322 08:41:18< Crab__> fendrin: yes, already created a project for that, will commit as soon as some results are available
20100322 08:47:26-!- ryaether [~ryan@173-26-127-85.client.mchsi.com] has left #wesnoth-dev []
20100322 08:47:57< CIA-80> fendrin * r41673 /branches/resources/eclipse_plugin/plugin.xml: Renamed Scenario and Campaign to Wesnoth Scenario and Wesnoth Campaign.
20100322 08:50:31< CIA-80> fendrin * r41674 /branches/resources/eclipse_plugin/ (7 files in 2 dirs): Documentation templates.
20100322 08:52:15< Crab__> fendrin: an idea - it would be quite cool to add a script in there to grab WML documentation from the wiki, preprocess it, and store it as a 'internal help' in eclipse.
20100322 08:52:35< fendrin> Crab__: I have already thought about that.
20100322 08:52:42< fendrin> Crab__: And my plan was different.
20100322 08:52:51< fendrin> Let me smoke a cigarette and explain.
20100322 08:52:54< Crab__> tell me :)
20100322 08:55:53< fendrin> I want that the content of the wiki is part of the wml schema and that the wiki is autogenerated from there.
20100322 08:58:49< Crab__> fendrin: how will you allow our users to add/modify the documentation, add notes/caveats/examples ?
20100322 08:59:21< fendrin> Like it is now.
20100322 08:59:47< Crab__> fendrin: now they can edit the wiki. they'll be able to edit a wml schema from the web interface ?
20100322 09:00:05< fendrin> Crab__: Well users shouldn't do that.
20100322 09:00:33< fendrin> They should be able to add notes and examples but not change the wml syntax description.
20100322 09:02:00< Crab__> fendrin: ok, so, we'll need to set up a wiki structure where each WML tag information comes from two sources - one is autogenerated from 'wml schema' and is readonly for users,  and another is editable for users ?
20100322 09:02:40< Crab__> it's possible to do, for example, SummerOfCodeIdeas has some content readonly and autogenerated
20100322 09:04:13< fendrin> Yes, that is what I had in mind.
20100322 09:04:56< fendrin> And we have everything in one place. The syntax description for the checkers and the description for the coders. That will help to keep the stuff in sync.
20100322 09:05:07< Crab__> yes, that's good
20100322 09:05:27< Crab__> ok. I can do that. that'll require having a wml schema somewhere in svn...
20100322 09:05:49< fendrin> Crab__: AI0867 has commited a schema to trunk.
20100322 09:05:56< Crab__> where it is ?
20100322 09:06:23-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Quit: This computer has gone to sleep]
20100322 09:06:32< fendrin> data/schema.cfg
20100322 09:06:38< Crab__> thanks
20100322 09:06:41< fendrin> and schema-gui.cfg
20100322 09:06:59< fendrin> note that there is no such attribute or subtag that describes the wiki content now.
20100322 09:10:52< Crab__> yes, it would be needed to talk with AI0867 to find out the best way to store the wiki markup for the info
20100322 09:11:42< Crab__> maybe in the external (translatable?) file cross-referenced by some identifiers ?
20100322 09:12:05< fendrin> hmmm
20100322 09:12:43< Crab__> but, once we figure it out, the rest is quite easy - a cronjob to parse that file via a command line parser, generate a list of wiki pages, see which ones changed, submit them to wiki, let wiki templates do the rest.
20100322 09:12:56< fendrin> I would have done it with a subtag called [description]
20100322 09:13:52< Crab__> yes, also possible, whatever suites AI0867 best.
20100322 09:14:01< fendrin> If it is marked translateable we could have different wml references for every language.
20100322 09:14:54< Crab__> if someone will translate them, that is. but I guess at least 'major' active translations will be interested.
20100322 09:14:59-!- Appleman1234 [~Appleman1@131.181.47.3] has quit [Ping timeout: 260 seconds]
20100322 09:16:09< fendrin> wesbot: seen AI0867
20100322 09:16:09< wesbot> fendrin: Queried user last spoke 1d 8h ago.  AI0867 is currently here and on the channel #wesnoth-umc-dev.
20100322 09:16:17-!- fkhodkov [~fedor76@ppp-188-65-8-66-bras1.istra.ru] has quit [Read error: Connection reset by peer]
20100322 09:17:26< fendrin> Crab__: Maybe AI0867 has already some of the stuff we need ready in his python codebase.
20100322 09:17:37< fendrin> We should work with him together more closely.
20100322 09:17:47< Crab__> fendrin: yes
20100322 09:19:15-!- Appleman1234 [~Appleman1@131.181.47.3] has joined #wesnoth-dev
20100322 09:24:14-!- Appleman1234 [~Appleman1@131.181.47.3] has quit [Ping timeout: 245 seconds]
20100322 09:24:47< fendrin> Crab__: My eclipse is crashing very often when using the epc plugin.
20100322 09:25:13< fendrin> "The error was 'RenderBadPicture (invalid Picture parameter)'."
20100322 09:25:31< fendrin> Crab__: Do you make the same expierience?
20100322 09:26:43-!- fkhodkov [~fedor76@ppp-188-65-8-66-bras1.istra.ru] has joined #wesnoth-dev
20100322 09:27:45< Crab__> fendrin: no, no crashes yet. I've launched only a few times, though.
20100322 09:28:24< fendrin> Crab__: Do you think that it is a good idea to autogenerate the dialogs out of the wml schema?
20100322 09:28:49< Crab__> 'dialogs' ? which dialogs ?
20100322 09:29:01< fendrin> The wizards.
20100322 09:29:29< fendrin> The new campaign wizard would be autogenerated from the campaign tag's schema. 
20100322 09:29:30< Crab__> I think that the list of properties that we need to ask is known only by the 'campaign template'.
20100322 09:29:50< Crab__> so, it's good idea to generate the UI based on the information we get from that campaign template
20100322 09:30:08< fendrin> What do you mean by "campaign template"?
20100322 09:30:36< Crab__> the sample code which 'new wesnoth campaign' should create
20100322 09:31:03< fendrin> And what form should that template have?
20100322 09:31:24< fendrin> maven?
20100322 09:31:25< Crab__> a wml config file, plus template files
20100322 09:31:54< Crab__> i.e., _main.cfg can contain the info
20100322 09:32:10< fendrin> Crab__: I think we are not talking about the same stuff.
20100322 09:32:35< fendrin> I have 2 wizards, one creating a campaign, one creating a scenario.
20100322 09:33:25< fendrin> Those wizards could both call the same code that autogenerates the wizard's gui out of the schema.cfg looking for the [campaign] and [scenario] tags.
20100322 09:34:01< Crab__> I'm thinking about something like this in a template : http://wesnoth.pastebin.com/Rqa72MKW
20100322 09:35:05< Crab__> or, we can generate the contents of the wizard based on schema.cfg, but then we'll need a way to show the most important tags on page 1, and optional tags on other pages.
20100322 09:35:08< fendrin> Crab__: All of that information is already part of schema.cfg, isn't it?
20100322 09:35:44< fendrin> Crab__: The schema.cfg already does specify if a attribute or tag is optional.
20100322 09:35:51-!- Appleman1234 [~Appleman1@131.181.47.6] has joined #wesnoth-dev
20100322 09:36:22< fendrin> That would need to be coded only once. We could over wizards for every wml tag afterwards.
20100322 09:36:27< Crab__> fendrin: might be, or might be not. e.g., if we want our template to include the option "make this a multiplayer campaign" or "make this both a SP and MP campaign", or "make this a SP campaign", how we'll do it ?
20100322 09:36:39< fendrin> And ask the schema.cfg which wizards are callable in the current wml context.
20100322 09:37:12< fendrin> Crab__: We have different wizards for multiplayer campaigns. They are very different.
20100322 09:37:41< Crab__> well, in LoW, we have the same files for both MP and SP
20100322 09:37:47< fendrin> Crab__: And since a multiplayer scenario has [scenario] as toplevel tag and multiplayer scenarios have [multiplayer] as toplevel tag, that isn't a problem.
20100322 09:38:35< fendrin> Crab__: The campaign creation dialogue will have a part that is hardcoded and a part that is autogenerated. 
20100322 09:38:49< fendrin> You have points here but I don't think they are show stoppers.
20100322 09:38:56-!- ancestral [~ancestral@97-116-105-225.mpls.qwest.net] has quit [Quit: ancestral]
20100322 09:39:04< Crab__> ok, that (part that is hardcoded and a part that is autogenerated) will do it.
20100322 09:44:08-!- Appleman1234 [~Appleman1@131.181.47.6] has quit [Ping timeout: 246 seconds]
20100322 09:44:22-!- noy_ [~Noy@wesnoth/developer/noy] has quit [Quit: noy_]
20100322 09:50:45< CIA-80> fendrin * r41675 /branches/resources/eclipse_plugin/ (. src/wesnoth_eclipse_plugin/wizards/CampaignNewWizard.java): Added bin to the svn ignore list.
20100322 10:04:37-!- ancestral [~ancestral@97-116-105-225.mpls.qwest.net] has joined #wesnoth-dev
20100322 10:07:38-!- ancestral [~ancestral@97-116-105-225.mpls.qwest.net] has quit [Client Quit]
20100322 10:09:38-!- Appleman1234 [~Appleman1@131.181.102.66] has joined #wesnoth-dev
20100322 10:11:17< AI0867> 09:17 < fendrin> Crab__: Maybe AI0867 has already some of the stuff we need ready in his python codebase. <-- I don't, there's some doxygen-like comments in the schemas, but they aren't parsed yet. The options are basically converting it to doxygen (I don't know if doxygen is flexible enough for this), writing a separate tool to extract doxygen-like comments (sounds like a bad idea) or integrating it into the structure (the [description] tag you p
20100322 10:12:16< Crab__> AI0867: ... [description] tag you p <line ended>
20100322 10:12:35-!- Spidy [~NickFerra@71.93.0.21] has quit [Quit: To Many Windows!]
20100322 10:12:41< AI0867> proposed)
20100322 10:13:12< Crab__> AI0867: if [description] tag is good for you, we can use it. e.g. [description] text=_"wiki markup here" [/description]
20100322 10:13:34< AI0867> k, I'll take a look at integrating it into wmlgrammar
20100322 10:13:36< Crab__> or,  [description] text=_"wiki markup here" category="scenario" [/description]
20100322 10:13:54< Crab__> to allow us to group wiki tags by category, like as it's done now
20100322 10:14:07< AI0867> sounds good
20100322 10:14:27< Crab__> and we'd probably need to use a separate textdomain for this
20100322 10:14:47< AI0867> so, one function to retrieve all tags of a category, and a method for tags to retrieve their description?
20100322 10:15:05< AI0867> Crab__: wmlparser can handle #textdomain directives
20100322 10:15:11< AI0867> though I'm not sure if they're parsed correctly
20100322 10:15:23< Crab__> AI0867: afk 10 min
20100322 10:26:36-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20100322 10:27:27-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100322 10:27:27< Crab__> AI0867: I was thinking about 'get all description tags for all categories, for each of those tags, generate a wiki page from local template, if it changed, submit it to wiki. (and use the wiki templates to generate pages like ScenarioWML for all tags in category="scenario" )
20100322 10:38:22< Ivanovic> moin
20100322 10:43:40-!- Mkools [~mukul@117.196.216.212] has joined #wesnoth-dev
20100322 10:44:22< Mkools> Hello their, I wanted to become part of Battle of Wesnoth in soc.
20100322 10:44:46-!- Appleman1234 [~Appleman1@131.181.102.66] has quit [Ping timeout: 264 seconds]
20100322 10:46:50< fendrin> Mkools: Great, please make a page for you on our wiki.
20100322 10:47:36< fendrin> Mkools: http://wiki.wesnoth.org/Category:Summer_of_Code_2010_Student_Page
20100322 10:49:39-!- Blueblaze [~nick@adsl-76-202-22-67.dsl.hstntx.sbcglobal.net] has quit [Remote host closed the connection]
20100322 10:49:42< Mkools> fendrin: Why is it necessary?
20100322 10:50:16< Crab__> Mkools: because we want to have a place where we can read about you, and about your proposal
20100322 10:51:12< Crab__> Mkools: and, then, later you'll be able to reference that page in your google application.
20100322 10:51:44-!- Appleman1234 [~Appleman1@131.181.102.66] has joined #wesnoth-dev
20100322 10:52:04< Crab__> Mkools: it's not necessary to do so now, there's still plenty of time left
20100322 10:53:02< Mkools> Crab__: Don't misunderstand me if I wanted to apply to many organizations they also require creating such kind of accounts, It is very difficult for me to cramp all those passwords. Is their another way?
20100322 10:54:29< Crab__> Mkools: well, creating a wiki page takes only a small amount of time compared to the amount of time that is needed to understand the details of a particular projects, check out the source, and compile
20100322 10:55:08< Ivanovic> Mkools: as student you should not forget that writing a good proposal is *work*
20100322 10:55:28< Ivanovic> Mkools: i think it is almost impossible to get more than 2 really good proposals submitted!
20100322 10:55:52< fendrin> Mkools: And we do see that you are really interested if you take the time to do that well.
20100322 10:56:19-!- Appleman1234 [~Appleman1@131.181.102.66] has quit [Ping timeout: 245 seconds]
20100322 10:58:25< Mkools> fendrin: Please give me some time to think it over again.
20100322 10:59:21< Crab__> and yes, that's four accounts to track ('irc, wiki, forum, bugtracker'), but still, you can let just use a plaintext file to store them, or a suitable keystore program.
20100322 10:59:26-!- fkhodkov [~fedor76@ppp-188-65-8-66-bras1.istra.ru] has quit [Read error: Connection reset by peer]
20100322 11:00:31< Mkools> Crab__: Thanks for suggestion. 
20100322 11:01:43< Mkools> Crab__: I will talk to you later after 6 hours.
20100322 11:01:46-!- Mkools [~mukul@117.196.216.212] has left #wesnoth-dev ["Leaving"]
20100322 11:05:00-!- Crab__ [~Crab@195.238.92.115] has quit [Ping timeout: 276 seconds]
20100322 11:05:15-!- Crab__ [~Crab@195.238.92.115] has joined #wesnoth-dev
20100322 11:06:10-!- Appleman1234 [~Appleman1@131.181.102.66] has joined #wesnoth-dev
20100322 11:07:05-!- meric [~Eric@124-171-9-200.dyn.iinet.net.au] has joined #wesnoth-dev
20100322 11:10:08-!- Mythological [Mythologic@77.28.125.85] has joined #wesnoth-dev
20100322 11:15:22-!- mpavel [~Pavel@cpc1-dund3-0-0-cust363.sgyl.cable.virginmedia.com] has joined #wesnoth-dev
20100322 11:16:53< fendrin> Crab__: Will there be a class that is the counterpart of the c++ config class?
20100322 11:18:36< Crab__> fendrin: there's still a problem with macroses... we can expand them and give a read-only version of config objects
20100322 11:19:12< Crab__> fendrin: but, read-write is tricky if macroses are present
20100322 11:20:15< Crab__> the same (but less tricky) situation is with #ifdef statements
20100322 11:21:06< fendrin> I would not expand them. 
20100322 11:21:20< fendrin> At least not for displaying them in the editor.
20100322 11:21:37< Crab__> fendrin: yes, but if we want to read a value...
20100322 11:22:08-!- Appleman1234 [~Appleman1@131.181.102.66] has quit [Ping timeout: 246 seconds]
20100322 11:22:23< Crab__> fendrin: so, answering the original question, there'll be a class that will be a counterpart to the c++ config, but it's functionality will be limited if macroses are present.
20100322 11:23:24< fendrin> Crab__: Okay. Will that class know what toplevel tag it represents and be able to access it's schema informations?
20100322 11:24:09< Crab__> yes. read-only access is not tricky at all.
20100322 11:24:23< Crab__> you'd have to set up a list of preprocessor defines, then you'll be able to get a config
20100322 11:24:39< Crab__> (a list of defines is needed to expand #ifdef's)
20100322 11:24:49< fendrin> I think we talk about different things again.
20100322 11:24:56< fendrin> Let me describe what I want to do.
20100322 11:24:59< Crab__> ok
20100322 11:25:11< fendrin> I have that class WmlTagWizardPage
20100322 11:25:44< fendrin> It will take a Config as constructor argument. 
20100322 11:25:55-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 260 seconds]
20100322 11:26:35< fendrin> In case of campaign that config will not be read from disc but created as an empty one that only knows that it is a scenario tag.
20100322 11:26:54< fendrin> Config scenario = new Config("scenario");
20100322 11:26:56< fendrin> right?
20100322 11:27:22< fendrin> sorry, mixed campaign and scenario. But both will be quite equal.
20100322 11:27:35< Crab__> yes
20100322 11:28:05< fendrin> Now I need to know what possible attributes and subtags that scenario or campaign or whatever wml tag has as valid childs.
20100322 11:28:55< fendrin> so a method like: scenario.getChildsFromSchema() would be very helpful.
20100322 11:29:35< fendrin> Since the Config knows what tag it is it could lookup the schema.cfg (or one of several)
20100322 11:30:01< Crab__> no, IMO, the config itself shouldn't deal with getting anything from files...
20100322 11:30:40-!- dekozo [~dekozo@189.71.50.249] has joined #wesnoth-dev
20100322 11:30:59< Crab__> I'm thinking about something like 'Config schema = WML.readConfigFromFile("data/schema.cfg")'
20100322 11:31:32< fendrin> And then:
20100322 11:31:53< fendrin> Config scenario = new Config(schema) ?
20100322 11:32:20< fendrin> Config scenario = new Config(schema.get_child(scenario)= ?
20100322 11:32:56< Crab__> probably you'd not need a copy..
20100322 11:33:07< Crab__> Config scenario = schema.child("scenario");
20100322 11:34:06< Crab__> if we open schema.cfg now...
20100322 11:34:16< Crab__> it has [campaign]
20100322 11:34:30< Crab__> in it, we have a list of attributes
20100322 11:34:55< Crab__> will that list be enough for your purposes ?
20100322 11:36:52< fendrin> hmmm
20100322 11:37:18< fendrin> We have configs that describe real wmltags
20100322 11:37:40< fendrin> and we have metaconfigs that describe the structure of a wml tag.
20100322 11:38:18< fendrin> Config scenario = new Config(metaconfig)
20100322 11:38:21< Crab__> and you want to generate a example config from a metaconfig ?
20100322 11:38:51< fendrin> Right.
20100322 11:38:56< CIA-80> silene * r41676 /trunk/src/ (unit_types.cpp unit_types.hpp): 
20100322 11:38:56< CIA-80> Avoided rebuilding all the units again and again when the AI recruits units. (Idea by Saphy.)
20100322 11:38:56< CIA-80> Simplified the code along the way.
20100322 11:39:37< Crab__> that's not possible, because the optional attributes are, well, optional.  We can think about something like     ConfigTemplate scenarioTemplate = new ConfigTemplate (metaconfig)
20100322 11:40:35< fendrin> I don't get the problems with optional attributes? What is wrong with them?
20100322 11:40:59< Crab__> the problem is once we create a config, the attribute is either in there or is not in there.
20100322 11:41:18< Crab__> so, we lose the information about a particular attribute being optional
20100322 11:41:52< fendrin> Crab__: No, the config is empty per default. But if you call config.isValid() it will tell you that there are still attributes that are mandatory left to be set.
20100322 11:42:25< Crab__> you want the configs to be 'strongly typed' ? that is not the case even in c++ now.
20100322 11:42:52< fendrin> yes, I want them to be strongly typed no matter that is not the case in c++ now.
20100322 11:43:42< fendrin> Every config created will be created with the help of a schema.
20100322 11:44:27< fendrin> So we have only valid wml code around, or we get exceptions.
20100322 11:46:12< fendrin> We can have a factory that reads the schema.cfg or cfgs and is able to give you the metaconfig for every known wmltag.
20100322 11:46:35< fendrin> A reference of course, not a copy.
20100322 11:46:57< fendrin> Crab__: Not a good ideaß
20100322 11:46:58< fendrin> ?
20100322 11:47:53< Crab__> It can be done, yes...
20100322 11:48:28< fendrin> But you don't like it?
20100322 11:49:12< Crab__> I need more time to think of all the consequences
20100322 11:50:09< Crab__> the code might look like : WmlEngine.setSchemaFile("data/schema.cfg");  WmlSchemaElement scenario = WmlEngine.getSchema().getElement("scenario");
20100322 11:50:20< Crab__> I don't like the fact that we haven't got it in c++ yet.
20100322 11:51:27< fendrin> Crab__: Yes, I am thinking about that since years. I had that in mind when I started to investigate frogatto's wml schemas.
20100322 11:51:45< Crab__> yes, doing it is a good idea.
20100322 11:51:56< Crab__> but I think that it needs to be done in c++ first
20100322 11:52:13< fendrin> Why?
20100322 11:52:33< Crab__> because we need to concentrate our efforts if we want to get the result
20100322 11:52:37< fendrin> Let's fetch experiences in our eclipse plugin before we mess with the game engine.
20100322 11:52:51< Crab__> ok, we can :)
20100322 11:53:03< fendrin> If everything works well we will port it to c++.
20100322 11:53:10< Crab__> just let's don't depend on it working.
20100322 11:53:37< Crab__> e.g. it might work - great!, it might not work , or it might require some big changes of all mainline/umc to work.
20100322 11:53:44< Crab__> we can *try*
20100322 11:54:16< Crab__> fendrin: note that there can be some perfectly valid WML not accounted by schema
20100322 11:54:30< Crab__> fendrin: e.g. lua-based wml tags made by scenario editor for himself
20100322 11:55:36< fendrin> Right, but the coder will have to life with the fact that there is no wizard for that tag right now.
20100322 11:56:05< fendrin> And the c++ engine will call the tag valid because it knows about dynamically added lua tags.
20100322 11:56:55< Crab__> fendrin: it will call the tag valid because it doesn't care if the tag is semantically valid when loading configs
20100322 11:57:46< fendrin> And later, when the engine processes the lua that defines the tag, the schemafactory can be enhanced by the new information.
20100322 12:03:24< Crab__> fendrin: ok, as of now, for start, we only need a small part of wml library, I can code a small prototype to access the WML schema
20100322 12:03:33< Crab__> ~ today's evening
20100322 12:04:43-!- mjs-de [~mjs-de@vpw.wh.Uni-Dortmund.DE] has joined #wesnoth-dev
20100322 12:05:26< fendrin> Crab__: That metaconfig factory?
20100322 12:06:07< Crab__> fendrin: yes, a way to get the schema and to work with it. it will not be a config object.
20100322 12:06:34< Crab__> but it will have a similar interface, but each child will have additional info accociated with it
20100322 12:07:19< fendrin> Crab__: Let's have a class attribute that also holds the type and the description.
20100322 12:07:52< Crab__> yes, it'll have the description, as well
20100322 12:21:03< Crab__> fendrin: what do you think about http://wesnoth.pastebin.com/CAMF6x6B ?
20100322 12:21:57< fendrin> Crab__: Cool
20100322 12:22:09-!- stikonas [~and@2002:4f84:b38b:b:226:9eff:fe0a:3e55] has joined #wesnoth-dev
20100322 12:22:09-!- stikonas [~and@2002:4f84:b38b:b:226:9eff:fe0a:3e55] has quit [Changing host]
20100322 12:22:09-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100322 12:22:18< Crab__> then, use by something like:    SchemaTag campaignSchema = Wml.getSchemaTag("campaign");
20100322 12:22:30< Crab__> (I can s/schema/meta/g if you like it more)
20100322 12:22:52< fendrin> Doesn't matter both is fine.
20100322 12:24:55< Crab__> ok, then, today, I'll commit the source for the basic 'reader' for the schema file that AI0867  has made, which'll populate the internal map of schema elements and allow to use them through this interface.
20100322 12:25:26< Crab__> I'll add some more methods in there, to allow to get optional/mandatory schemaTag's and schemaAttribute's easier.
20100322 12:37:58-!- meric [~Eric@124-171-9-200.dyn.iinet.net.au] has quit [Quit: meric]
20100322 12:40:02-!- meric [~Eric@124-171-9-200.dyn.iinet.net.au] has joined #wesnoth-dev
20100322 12:42:59-!- meric [~Eric@124-171-9-200.dyn.iinet.net.au] has quit [Client Quit]
20100322 12:45:38-!- Ken_Oh [~briang@static-71-178-174-219.washdc.fios.verizon.net] has joined #wesnoth-dev
20100322 12:54:28-!- mpavel [~Pavel@cpc1-dund3-0-0-cust363.sgyl.cable.virginmedia.com] has quit [Quit: Leaving.]
20100322 12:55:36-!- mpavel [~Pavel@cpc1-dund3-0-0-cust363.sgyl.cable.virginmedia.com] has joined #wesnoth-dev
20100322 13:08:44-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has joined #wesnoth-dev
20100322 13:23:57-!- qemqemqem [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has joined #wesnoth-dev
20100322 13:30:19< Crab__> fendrin: good news and bad news: I was able to generate a eclipse text editor plugin with schema-aware autocompletion and syntax highlighting for WML.  but,  it requires the schema to be inputted in a form which differs from AI0867's schema, so some more work will be required to autogenerate it.
20100322 13:31:04< fendrin> Crab__: What form does it require?
20100322 13:31:51< Crab__> fendrin: something like http://wesnoth.pastebin.com/eLC9LsdM
20100322 13:32:07< Crab__> I think that it can be generated from AI0867's schema
20100322 13:33:59< Crab__> the docs even say that it can do crosslinking, with the right schema (e.g. using available id='...' in autocompletion in places where such id is used)
20100322 13:34:57< fendrin> Sounds good.
20100322 13:36:10< fendrin> Do you plan to parse new files with that syntax or is that handled internal?
20100322 13:36:48< Crab__> sorry, I don't understand the question fully
20100322 13:36:51< Crab__> which 'new files' ?
20100322 13:37:17< fendrin> Convert the schema.cfg file in a file that contains the needed syntax.
20100322 13:37:34< fendrin> Once.
20100322 13:38:00< fendrin> Or do you read the schema.cfg and convert it internal into the needed syntax.
20100322 13:38:05< fendrin> ?
20100322 13:38:56< Crab__> doing it internally at runtime is tricky... it's better to do this beforehand
20100322 13:39:11< Crab__> the reason is that I need to generate the code of the plugin based on the schema
20100322 13:39:28< Crab__> so, it's easier to do that when packaging a plugin
20100322 13:39:47< Crab__> the convertsion procedure can/should be automated, of course.
20100322 13:55:11-!- qemqemqem [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has quit [Ping timeout: 265 seconds]
20100322 13:56:49-!- Master_Chief [~chatzilla@180.149.49.129] has joined #wesnoth-dev
20100322 14:10:10-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has quit [Ping timeout: 265 seconds]
20100322 14:30:30-!- qemqemqem [~quassel@urwireless-dhcp-128-151-186-137.wireless.rochester.edu] has joined #wesnoth-dev
20100322 14:33:27-!- Gambit [~Gambit@pa-67-234-116-244.dhcp.embarqhsd.net] has joined #wesnoth-dev
20100322 14:41:09-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has joined #wesnoth-dev
20100322 14:47:42-!- Crab__ [~Crab@195.238.92.115] has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
20100322 14:51:47-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100322 14:51:49-!- Crab__ [~Crab@195.238.92.115] has joined #wesnoth-dev
20100322 14:57:45-!- jekintrivedi [~root@116.72.242.56] has joined #wesnoth-dev
20100322 15:16:19-!- dekozo [~dekozo@189.71.50.249] has quit [Quit: Ex-Chat]
20100322 15:20:44-!- qemqemqem [~quassel@urwireless-dhcp-128-151-186-137.wireless.rochester.edu] has quit [Ping timeout: 265 seconds]
20100322 15:24:07-!- Netsplit *.net <-> *.split quits: teaser
20100322 15:40:16-!- Crab__ [~Crab@195.238.92.115] has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
20100322 15:44:49-!- Darkas [~quassel@ppp-93-104-137-87.dynamic.mnet-online.de] has joined #wesnoth-dev
20100322 15:51:13-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Quit: This computer has gone to sleep]
20100322 15:53:10-!- Master_Chief [~chatzilla@180.149.49.129] has quit [Remote host closed the connection]
20100322 15:57:39-!- timotei [~timotei@193.34.191.4] has joined #wesnoth-dev
20100322 16:00:23< timotei> hello guys!
20100322 16:00:33< timotei> and girls(if any:P)
20100322 16:12:11< shenmage> timotei: morning
20100322 16:17:07-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100322 16:23:10-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Ping timeout: 264 seconds]
20100322 16:26:09-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20100322 16:27:35-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100322 16:36:02-!- timotei is now known as timotei21
20100322 16:38:40< timotei21> Fix Bug #15656 (OOS errors in LoW from differing starting gold for host vs. other player).
20100322 16:38:47< timotei21> what does it mean "Low"
20100322 16:38:50< timotei21> "Low"*
20100322 16:40:46< zookeeper> legend of wesmere
20100322 16:40:47< Ivanovic> Legend of Wesmere
20100322 16:41:03< Ivanovic> the common short form of all campaigns is using the first letters
20100322 16:41:15< zookeeper> except liberty ;)
20100322 16:41:29< timotei21> ok
20100322 16:42:12< AI0867> zookeeper: yet its textdomain is wesnoth-l
20100322 16:42:45< zookeeper> yeah, but no one knows what you're talking about if you just say l :p
20100322 16:42:57< Ivanovic> off for some hours
20100322 16:48:31-!- timotei21 [~timotei@193.34.191.4] has quit [Ping timeout: 276 seconds]
20100322 17:00:15-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Quit: This computer has gone to sleep]
20100322 17:01:03-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20100322 17:02:39-!- migge_ [~marc@ip-95-223-247-31.unitymediagroup.de] has joined #wesnoth-dev
20100322 17:03:31-!- jekintrivedi [~root@116.72.242.56] has quit [Quit: Leaving.]
20100322 17:04:16< CIA-80> silene * r41677 /trunk/src/hotkeys.cpp: Prevented ctrl+alt from triggering a spurious hotkey.
20100322 17:04:20< CIA-80> silene * r41678 /trunk/src/hotkeys.cpp: Fixed hotkey handling improperly assuming an ascii locale.
20100322 17:05:39-!- migge [~marc@ip-95-223-247-31.unitymediagroup.de] has quit [Ping timeout: 240 seconds]
20100322 17:08:50-!- thespaceinvader [~chatzilla@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20100322 17:10:44< CIA-80> thespaceinvader * r41679 /trunk/ (31 files in 4 dirs): Add and wire new Cuttle Fish base frame and animations by Mefisto. Delete obsolete images. Update changelogs, credits.
20100322 17:12:29< Darkas> Crab_, hi
20100322 17:12:34< Crab_> hi, Darkas
20100322 17:13:18< Darkas> would the berserk ai improvement be a good warming up for the Google Summer of Code?
20100322 17:13:51< Crab_> yes, if done right, it'll be a good thing by itself, and in gsoc context
20100322 17:13:52< Darkas> or do you have better jobs for me to see if the ai project is good for me?
20100322 17:14:05< Crab_> depends on what and how quickly you want to do :)
20100322 17:15:05< Crab_> I suggest that you pick something that you can complete in a week
20100322 17:15:23< Darkas> ok
20100322 17:16:14< Darkas> do you think the berserk ai improvement would be something like this big?
20100322 17:17:02< Crab_> hard to say. the problem with 'berserk ai improvement' is that it's impossible to do without the 'idea'
20100322 17:17:23< Crab_> so, do you have the idea about *how* the ai should play it's berserkers ?
20100322 17:17:43< Darkas> I know, I was told to ask DragonKing
20100322 17:18:00< Crab_> we have the ability to test your idea, e.g. see if dwarven ai becomes better in ai-vs-ai matchup if tested with your changes
20100322 17:18:38< Crab_> 'asking Dragonking' is not the easiest thing nowadays :)
20100322 17:18:43< Darkas> that's good, I think I have a first idea
20100322 17:18:48< Darkas> :)
20100322 17:18:58< Crab_> Darkas:  good :) btw, what do you think about other points at http://wiki.wesnoth.org/SoC_Ideas_Lua_AI ?
20100322 17:19:05-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20100322 17:19:07< Crab_> (..Possible pre-gsoc tasks..)
20100322 17:20:10< Darkas> they sound nice, but I like the berserk thing a bit more :)
20100322 17:20:13< shenmage> Crab_: since I haven't introduced myself yet and I was told you were the person in charge of the AI GSOC, hi :)
20100322 17:20:37< Crab_> hi, shenmage
20100322 17:21:44< boucman> hey all
20100322 17:22:00< Crab_> shenmage: yes. Last year I was a gsoc student for wesnoth, and refactored the ai codebase to make it more modular and add the ability to support multiple languages in a single ai. and now it's time to improve the basic lua support for the ai, which I've added recently with the help of silene.
20100322 17:22:00< Darkas> Crab_: what files are the interesting ones to start with?
20100322 17:22:03< Crab_> hi, boucman
20100322 17:22:20< Crab_> Darkas: ask a question, and I'll tell you what files contain the answers.
20100322 17:22:21-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100322 17:22:37-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100322 17:22:51< shenmage> hey boucman
20100322 17:23:53-!- teaser [~teaser@h-37-106.A254.priv.bahnhof.se] has joined #wesnoth-dev
20100322 17:24:34< Darkas> Crab_: ok, then I'll try to get an overview by myself and then ask you if there are any questions
20100322 17:24:42< Crab_> Darkas: if you want an overview of the AI, I'd start with http://wiki.wesnoth.org/Ai_Module#Source_files , and then, trace the  ai turn sequence, starting with the ai::manager::play_turn in src/playsingle_controller.cpp
20100322 17:25:30< shenmage> Crab_: I was looking at the different pre-gsoc tasks over the weekend, and I was curious which of them would be most helpful for me to work on
20100322 17:25:51< Crab_> shenmage: depends on what is most interesting for you :)
20100322 17:26:12< Crab_> Darkas: it will call src/ai/manager.cpp, which will init the ai contexts (src/ai/contexts.cpp), src/ai/default/contexts.cpp, src/ai/composite/contexts.cpp
20100322 17:26:59< Darkas> Crab_: ok, that sounds good
20100322 17:27:09< Crab_> Darkas: then, it will create the composite AI, src/ai/composite/ai.cpp, which will load the stages, including (in case of RCA ai) candidate action evaluation loop stage, which will load all defined candidate actions, then the candidate action evaluation loop be played.
20100322 17:27:12< Darkas> Crab_: what does the term formula ai mean?
20100322 17:27:30< Crab_> Darkas: it's a domain-specific functional language which can be used for writing AIs.
20100322 17:28:05< Crab_> shenmage: the main point of pre-gsoc tasks is to see first-hand the limitations of current lua ai support
20100322 17:28:07-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100322 17:28:16< Crab_> shenmage: and to code a few improvements...
20100322 17:28:26< Darkas> Crab_: ok, I think I understood
20100322 17:28:51< Crab_> Darkas: see data/ai/formula/poisoner_eval.fai and data/ai/formula/poisoner_attack.fai
20100322 17:29:16< shenmage> Crab_: alright. I'll probably start working on it later tonight, but right now I'm going to finish up some studying for a midterm. I'll keep in touch.
20100322 17:29:24< Crab_> shenmage: ok
20100322 17:30:00< Crab_> Darkas: the main parts that you need to study is to find out what 'stage' and 'candidate action' mean.
20100322 17:31:02< Crab_> each stage is a part of ai turn, each candidate action is a 'if THIS do THAT with priority SCORE'. see more at http://wiki.wesnoth.org/Customizing_AI_in_Wesnoth_1.8
20100322 17:31:24< Darkas> Crab_: ok, I'll do that
20100322 17:31:40< Crab_> Darkas: anything else that you need to know now ?
20100322 17:32:16< Darkas> Crab_: not yet, I'll start studying code/wiki pages now
20100322 17:32:21< Crab_> ok
20100322 17:32:27< boucman> thespaceinvader: around ?
20100322 17:32:41< thespaceinvader> indeed boucman 
20100322 17:33:22< boucman> i'd really like an artist's opnion on the commit worthyness of doofus's different stuff in http://www.wesnoth.org/forum/viewtopic.php?p=416907#p416907
20100322 17:33:29< boucman> esp the underground villages
20100322 17:33:54< boucman> (the sand castle is probably good enough, and the underground castles not ready yet, so the others are the most interesting to me)
20100322 17:34:42< Blarumyrran> the 1. and 3. have giant shading problems
20100322 17:37:43-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100322 17:38:38-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Client Quit]
20100322 17:47:25-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100322 17:49:34-!- haoyu [~bhy@cm26.delta25.maxonline.com.sg] has quit [Ping timeout: 264 seconds]
20100322 17:57:15< thespaceinvader> boucman: the perspective issues are the blocker for me - the villages simply don;t look like they're built right
20100322 17:57:24< boucman> ok, good
20100322 17:57:41< boucman> I don't feel good deciding about art, that's why I ping you
20100322 17:58:51-!- noy [~Noy@d99-199-3-111.bchsia.telus.net] has joined #wesnoth-dev
20100322 17:58:55-!- noy [~Noy@d99-199-3-111.bchsia.telus.net] has quit [Changing host]
20100322 17:58:55-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100322 18:01:35< thespaceinvader> boucman: no worries
20100322 18:01:56< thespaceinvader> if in doubt, i'd generally not commit things, personally
20100322 18:02:05< thespaceinvader> i still tend to check with jet about stuff anyway
20100322 18:03:10< Darkas> Crab_: as far as I can see, part of my job would be writing lua stages, right?
20100322 18:03:11< boucman> k
20100322 18:04:03< Crab_> Darkas: part of the job is to actually improve the ai, and this will involve writing some lua stages. in particular, we now can have multiple leaders per side - so, we need to teach the ai to use this intelligently.
20100322 18:04:35< Crab_> Darkas: and part of the job is to make it possible to write such stages 'easily' - exposing existing c++ things to lua
20100322 18:05:37< Crab_> Darkas: please read the 'exposing game state' section from http://wiki.wesnoth.org/LuaAI#Exposing_game_state
20100322 18:05:56-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100322 18:05:58< Darkas> Crab_: ok
20100322 18:06:03< Crab_> Darkas: some of the info (e.g. unit_map) is already available for the ai
20100322 18:06:37< Crab_> Darkas: and some info is not available (like 'possible moves of a single unit')
20100322 18:07:27< Darkas> Crab_: they should just be exposed to lua?
20100322 18:07:47< Darkas> so they already exist in C++?
20100322 18:07:49< Crab_> yes. and some of them are 'derived info' which can be written in lua, as a 'library of helper functions'
20100322 18:07:56< Crab_> yes
20100322 18:08:18< Crab_> there's a number of caches and parameters, most of them are in readonly_context src/ai/contexts.cpp
20100322 18:09:24< Crab_> we want to get access to them, and to achieve at least a feature parity with formula ai functions.
20100322 18:09:58-!- ilor [~ilor@wesnoth/developer/ilor] has joined #wesnoth-dev
20100322 18:10:05< Darkas> ok
20100322 18:10:46< Darkas> but why do we actually want lua? what makes it interesting if we already have the formula ai?
20100322 18:11:08< Crab_> formula ai is very hard to use (*censored*)
20100322 18:11:50< Darkas> that's a good reason :)
20100322 18:13:03< Crab_> also, lua is also used in wesnoth to interact with wml, so it'll be very easy to make the ai interact with WML code of the scenario
20100322 18:14:04< Darkas> good
20100322 18:14:34< Crab_> and we need to make it easy to use our lua ai for content developers. and using lua  means that they will need to learn only one additional language, not two.
20100322 18:15:30< Darkas> so you want to drop formula ai in the long term?
20100322 18:16:23< Crab_> if no one will be there to maintain it, I guess it would be dropped. but there's no reason to drop working code without even having a replacement.
20100322 18:16:33< Darkas> of course
20100322 18:16:42< Crab_> formula ai things tend to be written in functional style, and that's hard to get right.  e.g.:    me.loc, choose(unit_moves(me.loc),'mloc',-sum(map(simplest_path(mloc,loc(14,9),me.loc), 'path_location', movement_cost( me, path_location ) ))) )
20100322 18:17:10< Darkas> yeah, that's really hard
20100322 18:17:11< Crab_> oops, the full formula example is   "move(me.loc, choose(unit_moves(me.loc),'mloc',-sum(map(simplest_path(mloc,loc(14,9),me.loc), 'path_location', movement_cost( me, path_location ) ))) )  "
20100322 18:17:21< Crab_> and using lua we can do it in pieces
20100322 18:17:42< Crab_> since we can do it procedural-style
20100322 18:17:58< Darkas> yeah
20100322 18:18:21< Darkas> actually, I'm experienced in lua and the C(++) binding
20100322 18:18:41< boucman> what experience ?
20100322 18:18:48< boucman> (too lazy to go to your gsoc page)
20100322 18:19:24< Darkas> I've used it in my own project
20100322 18:19:34< boucman> k
20100322 18:19:49< Darkas> I've been using lua for about a year now
20100322 18:20:24< Crab_> Darkas: so, we want to expose base things to lua code, then code a library of support functions to make simple things really easy for scenario creators.
20100322 18:21:07< Darkas> Crab_: the library should then contain real AI, right?
20100322 18:21:52< Crab_> Darkas: well, the philosophy is "there is only one AI, and it's name is composite AI"
20100322 18:22:06< Crab_> Darkas: so, the library would contain some 'pieces' which can be added to the ai
20100322 18:22:42< Crab_> and there are two kinds of pieces which govern the ai turn sequence - stages and candidate actions
20100322 18:22:48< Darkas> Crab_: something like: put stragegies A and B together to do your job?
20100322 18:23:09< Crab_> Darkas: yes. have you read about candidate action evaluation loop already ?
20100322 18:23:34< Crab_> we want scenario creators to mix-in behavior into default ai
20100322 18:23:42< Darkas> Crab_: yes, but I'm not really sure about the scope of the candidate action
20100322 18:23:58< Crab_> it can be small or large, as desired. see src/ai/testing/ca.cpp
20100322 18:24:03< Darkas> are they for the whole AI or only for special units?
20100322 18:24:20< Crab_> as the candidate action will desire :)
20100322 18:24:38< Crab_> e.g., even now, 'move leader to keep ' candidate action only moves the leader
20100322 18:24:56< Crab_> the candidate action looks at the battlefield and says a score>0 if it wants to play
20100322 18:25:08< Crab_> it might look only at special units, or it can look at all units
20100322 18:25:25< Darkas> I see
20100322 18:25:26< Crab_> and, if it gets to play, it can play the entire turn if it wants to.
20100322 18:26:24< Darkas> so the candidate action with the highest score gets played, and if there are fighters left, then the second candidate action can take them?
20100322 18:27:06< Crab_> well, after a candidate action plays, the loop runs again. and each (including the first) candidate action gets a chance to play
20100322 18:27:48< Darkas> ok
20100322 18:27:55< Crab_> (we can select the action to play prematurely, if we know upper bounds on remaining candidate action's scores, hence the sorting of candidate actions by desc upper bound of score, where it's known)
20100322 18:28:13< Crab_> see src/ai/testing/stage_rca.cpp
20100322 18:29:52< Crab_> there's, however, a restriction - if candidate action has returned a score>0 and is selected to play, it MUST do at least one move - otherwise it means that it lied to us that It knew how to play, and we'll punish it by disabling it for 1 turn. that is to prevent infinite loops.
20100322 18:30:25< Darkas> ok
20100322 18:30:34< Darkas> how is that score calculated?
20100322 18:31:46< Crab_> the action is free to name any score it wishes.
20100322 18:32:18< Crab_> current actions use static (hardcoded) scores if they can play, 0 (BAD_SCORE) otherwise
20100322 18:32:51< Crab_> user-made actions can use things like 'static score of combat phase+10' which'll mean 'just before combat'
20100322 18:32:57< Darkas> but then every candidate action has to play at least once in a turn?
20100322 18:33:27< Crab_> Darkas: no, not guaranteed. each action changes the game state, so, when higher-score actions have played, it might happen that there's nothing left to move.
20100322 18:34:14< Crab_> Darkas: also note that there is potential to make a 'greater' ai than we have today by making those scores be calculated by a set of external 'position evaluators'
20100322 18:34:33< Crab_> Darkas: the engine supports this, but no such evaluators were made
20100322 18:34:58< Crab_> that way, it'll be possible to directly compare, say, 'defensive move' with an 'attack', instead of current system with hardcoded priorities
20100322 18:35:31< Darkas> yeah, that sounds more logical
20100322 18:36:02< Darkas> so right now, for example "attack" would be always rated higher than "defensive move"?
20100322 18:36:21< Crab_> yes (provided that the attack was considered 'good enough')
20100322 18:36:34< Crab_> and this means that current ai is unable to play defensively.
20100322 18:36:46< AI0867> Darkas: essentially, it's the old sequential AI, converted to the new system
20100322 18:36:56< Crab_> it can play offensively (it's pretty good at killing things) at around 2:1 gold advantage
20100322 18:37:01< AI0867> the next step will be to actually *use* the new system
20100322 18:37:11< Crab_> AI0867: true
20100322 18:37:35< Darkas> wound't that be an even better idea for GSOC?
20100322 18:38:26< Crab_> Darkas: the results are harder to measure. so, it's actually better to do the lua stuff as a 'mandatory program' and include 'ai improvements' as a sideline quest.
20100322 18:39:16< Crab_> Darkas: also, the time-to-develop is faster with lua, as there's no need to do a c++ recompilation
20100322 18:39:35< Crab_> Darkas: so, it's faster to try new ideas in lua, provided that we'll add all the hooks
20100322 18:40:23< Darkas> Crab_: so it would also be possible to use lua as a playground and if it works good, we could implement it in C++?
20100322 18:40:41< Crab_> Darkas: which benefits such reimplementation will bring ?
20100322 18:40:59< Darkas> more speed?
20100322 18:41:09< Darkas> or are they just little parts of code?
20100322 18:41:20< Crab_> Darkas: IMO, speed is mostly determined by an algorithm, not by language chosen
20100322 18:41:55< Darkas> I know, but some guy from #lua told me that C++ code is about 10 times faster
20100322 18:42:47< Crab_> Darkas: well, if we find out that there's a problematic hot-spot in the calculations, we can move it to c++
20100322 18:42:59< Darkas> Crab_: ok
20100322 18:44:00-!- qemqemqem_ [~quassel@urwireless-dhcp-128-151-186-137.wireless.rochester.edu] has joined #wesnoth-dev
20100322 18:44:42< Darkas> Crab_: for the berserk AI improvement, I'd need to modify the candidate actions, right?
20100322 18:45:16< Crab_> Darkas: yes
20100322 18:45:35< Darkas> the ones of attack and move, then check if the unit has a berserk attack and apply special behaviour
20100322 18:45:40< Darkas> ok
20100322 18:46:31< Crab_> Darkas: note 1) that you can use WML config to forbid standard combat phase to process berserkers and deal with the berserkers 'by hand' - by another candidate action
20100322 18:46:47< Crab_> 2) there's no support for lua candidate actions atm - I suppose I'll fix it today.
20100322 18:47:02< Crab_> (there's a small bit that needs to be enabled)
20100322 18:48:06< Crab_> Darkas: but, actually, if you'd got time today, maybe you'll want to fix it for me today ?
20100322 18:48:22< Darkas> Crab_: yes, that's possible
20100322 18:48:41< Darkas> I'll just eat something now and then I'll be back :)
20100322 18:48:52< Crab_> then, after you return, ping me and open src/ai/composite/engine_lua.cpp
20100322 18:49:37< Crab_> (the work will be with c++\lua bindings, which you said you're familiar with - and that's good)
20100322 18:51:53-!- timotei21 [~timotei21@193.34.191.4] has joined #wesnoth-dev
20100322 18:52:56-!- orfest [~kvirc@43.ngu.ac-tel.ru] has quit []
20100322 18:53:55-!- orfest [~kvirc@43.ngu.ac-tel.ru] has joined #wesnoth-dev
20100322 18:54:12< orfest> hi everyone! :)
20100322 18:54:17< Crab_> hi, orfest
20100322 18:54:21-!- Telaron_ [~9442274a@gateway/web/freenode/x-vtlnzqqewvhtzwtm] has joined #wesnoth-dev
20100322 18:54:23< orfest> any news about my patch? :)
20100322 18:54:29< boucman> hello orfest
20100322 18:54:40< boucman> which one was that already ?
20100322 18:54:50< Crab_> orfest: I'll take a look at it today, most likely in ~ 30 minutes
20100322 18:54:57-!- qemqemqem_ [~quassel@urwireless-dhcp-128-151-186-137.wireless.rochester.edu] has quit [Ping timeout: 276 seconds]
20100322 18:54:59< ilor> hi boucman, any more trouble with trunk lobby?
20100322 18:55:01< Crab_> boucman: https://gna.org/patch/index.php?1548
20100322 18:55:22< boucman> ilor: running it now, it crashed yesterday evening, but no crash today so far
20100322 18:55:29< orfest> Crab_: ok :)
20100322 18:55:37< ilor> yesterday evening crash was -trunk?
20100322 18:55:49-!- Telaron_ [~9442274a@gateway/web/freenode/x-vtlnzqqewvhtzwtm] has quit [Client Quit]
20100322 18:55:50< boucman> ilor: yes
20100322 18:55:53< ilor> :/
20100322 18:56:13< ilor> is there anything you can tell me about it?
20100322 18:56:17< boucman> ilor: I stilll have my left-aligned/right-aligned issues in trunk, (if you've seen it)
20100322 18:56:46 * orfest wanna code something ^_^
20100322 18:56:51< ilor> boucman: no, what was it?
20100322 18:57:13< boucman> gimme a sec
20100322 18:57:38< boucman> http://imagebin.org/89749 right aligned 
20100322 18:57:42< boucman> http://imagebin.org/89750 left aligned 
20100322 18:58:01< boucman> ilor: my lobby more or less randomly jump from one aspect to the other
20100322 18:58:26< ilor> boucman: ah, it's not really aligned, as it's jumping between the "expand horizontally and take up space" and "take as much as you need and no more space"
20100322 18:58:31< boucman> with rc1 I knew what would cause one or the other aspect, but in trunk it seems more or less random sofar
20100322 18:59:06< boucman> yes, I figured that out, "left aligned" and "right aligned" was more of an easy naimng convention :)
20100322 18:59:14< ilor> boucman: this is largely a gui2 issue that I don't really know the cause
20100322 18:59:20< boucman> k
20100322 18:59:21< ilor> I'm much more interested in the crash
20100322 18:59:33< boucman> well, no luck so far :°
20100322 18:59:39< ilor> if you can spare the disk space, run with --log-debug=lobby
20100322 19:00:11< Crab_> orfest: then, about the patch, a few minor things: 1) you should add yourself to misc. contributors section of about.cfg , not to 'programmers' (until you get commit access, that is)      2) have you looked at src/Makefile.am - is there anything to fix there ?
20100322 19:00:13< boucman> will it redirect on its own or do I need to do something (i'm running under gdb already
20100322 19:00:22< ilor> it's quite spammy but allowed me to see the cause for crashes that happen later
20100322 19:00:36< ilor> uh, not sure how gdb and stream redirection works, 
20100322 19:00:48< ilor> s/,//
20100322 19:00:54< ilor> argh nvm 
20100322 19:01:54< orfest> Crab_: 1 - ok :) 2 looks that it shouldn't be fixed. grepping boost gives nothing
20100322 19:02:13< orfest> Crab_: 2. which Makefile.am there are many...
20100322 19:02:14< Crab_> orfest: have you tried case-insensitive grep ?
20100322 19:02:21< Crab_> src/Makefile.am
20100322 19:02:58< orfest> Crab_: you are right, it should be fixed. will do in a minute
20100322 19:05:04< Crab_> orfest: 3) your implementation of "static bool good_response(const T& response)"  is too strict - e.g., it will not think that "HTTP/1.1 200 Ok" is a valid response (this doesn't lead to failures now, as you pass it only a "HTTP/1.1 2" part)
20100322 19:06:53< orfest> Crab_: 3) it's just rewritten original implementation. it doesn't lead to failures now, should it still be fixed?
20100322 19:07:05< CIA-80> shadowmaster * r41680 /trunk/misc/fortunes/wesnoth: Cookie
20100322 19:07:22< boucman> ilor: ok, got it to log both...
20100322 19:09:33< Crab_> orfest: yes, I think it's better to fix it now, to ensure that it doesn't lead to someone thinking 'wtf?' later.
20100322 19:09:39< Crab_> orfest: maybe, something like http://wesnoth.pastebin.com/XTdYgk4F
20100322 19:15:59-!- happygrue [~George@wesnoth/developer/wintermute] has quit [Read error: Connection reset by peer]
20100322 19:17:09-!- happygrue [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20100322 19:18:47< Crab_> orfest: also, do you have a stub server against which I can test your patch ? (if no, that's not a problem, I'll see if I can use wesnoth's log handler server script to test)
20100322 19:19:03< Crab_> (server script to handle uploads)
20100322 19:20:04< orfest> Crab_: no,I don't have such a server. During testing I simply uploaded about 100 similar logs to www.wesnoth.org O:)
20100322 19:20:26< Crab_> ok, good enough :))
20100322 19:24:06-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20100322 19:24:15< mordante> servus
20100322 19:24:22< Crab_> hi, mordante
20100322 19:24:36< mordante> hi Crab_ 
20100322 19:25:35< Crab_> mordante: I've seen your comment in the patch for orfest's rewrite of stats uploader, so, a question - what do you think, is it ok to commit it before 1.8 ?
20100322 19:27:01< Crab_> it is a bugfix, fixes bug https://gna.org/bugs/index.php?13094
20100322 19:27:04< mordante> Crab_, no IMO it's too intrusive so short before 1.8
20100322 19:27:55< mordante> I rather see it committed after 1.8 and then we could commit it to 1.8 as well
20100322 19:28:07< mordante> 1.8-branch*
20100322 19:28:39< mordante> and then also ask some Mac and Windows users to test it before 1.8.1 is released
20100322 19:28:51< Crab_> orfest: ^ ok, so, after you fix those little things, we'll treat your patch as 'good and ready for commit', and then commit it after 1.8 branches off.
20100322 19:29:29< timotei21> mordante, you're french?:)
20100322 19:29:31< mordante> Crab_, after of before Ivanovic forks? There well be a time after tagging 1.8 and before forking the branch
20100322 19:29:37< mordante> timotei21, no why?
20100322 19:30:28< timotei21> "servus", this is used also in our country for saluting someone
20100322 19:30:47< Crab_> mordante: after the fork, I think. I don't want it to interfere if some issues with *other* code arise and there'll be a need to retag
20100322 19:30:51< boucman> timotei21: is it ? (i'm french and never heard this except from mordante)
20100322 19:31:07< boucman> Crab_: fyi I've marked all post 1.8 patches as postponed
20100322 19:31:30< Crab_> good
20100322 19:31:37< mordante> Crab_, that time is meant for committing stuff that's needed in trunk and 1.8 ;-) (Like translation fixes)
20100322 19:31:39< timotei21> Yes, it's used very often, but I think is *imported* from French:)
20100322 19:31:52< timotei21> servus/serus
20100322 19:32:28< mordante> it's also used in Germany, but I'm not from German either ;-) I just like the greeting
20100322 19:33:15< timotei21> hehe
20100322 19:33:40< Ivanovic> servus is the greating that is most common in bavaria
20100322 19:34:11< boucman> timotei21: http://en.wikipedia.org/wiki/Servus
20100322 19:34:28< CIA-80> ivanovic * r41681 /trunk/po/wesnoth-low/de.po: updated German translation
20100322 19:34:33< boucman> aparently not that common in french, though the "serviteur" as a greeting could be related to it
20100322 19:36:02< timotei21> oh lol. I don't know why I thought it comes from French. I mean, there are a lot of "frenchness" version of words in our language
20100322 19:36:35< timotei21> anyway, good to know
20100322 19:37:41< apoi> seen Darkas
20100322 19:37:50< apoi> wesbot: seen Darkas
20100322 19:37:50< wesbot> apoi: Queried user last spoke 49m 9s ago.  Darkas is currently in this channel.
20100322 19:39:12< orfest> https://gna.org/patch/index.php?1548 updated
20100322 19:39:22< orfest> Crab_: ok :)
20100322 19:39:26< Crab_> orfest: thanks, I'll take a  look at it now
20100322 19:39:53< Crab_> it's really good that you've reworked the stats uploader quite fast.
20100322 19:40:37 * orfest likes boost::asio
20100322 19:40:47< boucman> orfest: remember to link to the patch in your application page
20100322 19:41:01< orfest> boucman: thanks for the notice
20100322 19:41:02< boucman> when we review them we easily get confused on who did what...
20100322 19:43:21< Crab_> orfest: what do you think about continuing the work and adding http proxy support for the stats uploader (it is not only the matter of using it in stats uploader, it's also about allowing the user to specify it) ? as, later you'll need it anyway for the work on mp server.
20100322 19:44:00< orfest> BTW, could anybody mark my forum account (orfest) as gsoc-student?
20100322 19:44:22< boucman> orfest: right away
20100322 19:44:28< orfest> boucman: thank you
20100322 19:44:49< orfest> Crab_: hm, it'd be interesting.
20100322 19:44:49< boucman> done
20100322 19:45:15< boucman> orfest: it might be a bit complicated if you try to get it from the OS (i.e os dependant code...)
20100322 19:45:33< Crab_> orfest: good :) it's worth discussing here with other developers, how it'll be best to allow the user to specify a proxy
20100322 19:45:37< orfest> Crab_: As I noticed I have two internet connections, the first one is very cheap, it's through proxy and it is *very* slow.
20100322 19:45:59< orfest> Crab_: looks like it'd require some GUI changes
20100322 19:46:34< Crab_> yes, and mordante is the best person to ask about them
20100322 19:46:34< orfest> I won't believe gamers will like to set proxy in plain text files :)
20100322 19:47:02< boucman> orfest: preferences=>MP would be the most logical place
20100322 19:47:03< Crab_> orfest: well, some are *so* desperate to get wesnoth running that they're willing to setup ssh tunnelling by hand :)
20100322 19:47:18< Crab_> orfest: at least there were a few forum threads about it :)
20100322 19:47:59< Crab_> also, note that some proxies are using funny auth schemes
20100322 19:48:49< Crab_> we don't need a '100% good solution', but it'll be interesting to see what can be done and what is not worth adding.
20100322 19:48:51< mordante> I'm off for dinner
20100322 19:48:54< boucman> are there any generic http proxy handling libraries around ?
20100322 19:49:17< Crab_> afair, there is libcurl
20100322 19:49:24< mordante> Crab_, btw regarding the schemes it would be nice if the wiki comment in the gui2 code can be used as basis
20100322 19:50:34< Crab_> but, don't know if it's compatible with boost::asio or not
20100322 19:51:33< boucman> zookeeper: we have a patch bout ANL for 1.8... can I assign to you ?
20100322 19:53:04< orfest> Crab_: googling haven't given me a ready solution for proxies.
20100322 19:56:34< boucman> zookeeper: could you join #wesnoth-mentor for a sec ?
20100322 19:56:51-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100322 19:57:26-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100322 19:59:11< Crab_> orfest: can you investigate this matter in more detail ? e.g. make some kind of table at your proposal page about proxy support - what kind of auth schemes we can support and how much time would it (estimate) take.
20100322 19:59:41< Crab_> orfest: e.g. there are some open-source libs around that can do NLTM auth, like libcurl and http://josefsson.org/libntlm/
20100322 19:59:52< Darkas> Crab_: I'm back
20100322 19:59:53< Crab_> orfest: maybe it'll be possible to reuse one of them
20100322 19:59:59< Crab_> Darkas: good :)
20100322 20:00:04< orfest> Crab_: ok =)
20100322 20:00:17< Crab_> Darkas: then, if you're ready, open src/ai/composite/engine_lua.cpp
20100322 20:00:42< Darkas> Crab_: I'm ready
20100322 20:01:35< Crab_> in that file, you'll see lua_candidate_action_wrapper and lua_stage_wrapper
20100322 20:02:12< Crab_> this lua engine creates ai components backed by lua code,  from config file
20100322 20:02:40< Crab_> Darkas:  lua_candidate_action_wrapper and lua_stage_wrapper both create a lua_ai_action_handler and call it when needed
20100322 20:02:57< Crab_> for stage, nothing else is needed to work
20100322 20:03:12< Crab_> but, for candidate action, we need two lua parts
20100322 20:03:19< Crab_> 'evaluation'  and 'execution'
20100322 20:03:36< Crab_> if evaluation returns a score > 0, then, this candidate action might be selected for execution if it's score is highest
20100322 20:04:03< Crab_> so, we need to return a score (number) from lua code
20100322 20:04:26< Crab_> this is the minimal requirement for lua ai candidate actions to work - evaluation must return a score number to c++
20100322 20:04:26< Darkas> ok, that's no problem
20100322 20:04:48< Crab_> but, there's one more catch - the usual pattern is : "we will return a good score if we find a good specific move to do"
20100322 20:04:56< Crab_> and in, execute, "do that move that we're found"
20100322 20:05:23-!- meric [~Eric@124-171-9-200.dyn.iinet.net.au] has joined #wesnoth-dev
20100322 20:05:32< Crab_> Darkas: please open src/ai/testing/ca.cpp
20100322 20:05:35< Darkas> Crab_: so execute runs a lua function?
20100322 20:05:51< Darkas> done
20100322 20:05:53< Crab_> Darkas: yes, in lua candidate action, both execute and evaluate run lua functions
20100322 20:05:56< Crab_> see the c++ - based goto_phase
20100322 20:06:20< Crab_> note that it finds a 'best goto to do' and stores in in a 'move_ member variable'
20100322 20:06:31< Crab_> and in execute, it simply does that stored move
20100322 20:07:04< Crab_> and, we need to store information in lua 
20100322 20:07:15< Crab_> in context of the specific 'candidate action'
20100322 20:07:50< Darkas> you mean what we've found?
20100322 20:07:57< Crab_> yes
20100322 20:08:00< Darkas> ok
20100322 20:08:15< Darkas> that could be little more tricky
20100322 20:08:20< Crab_> to avoid the situation "huh, we know that we've just found a good move. now, what the move was?"
20100322 20:08:30< Darkas> yeah, right
20100322 20:08:46< Crab_> there's two ways to do that - either pass it back to c++ in evaluate() (as a serialized config object) and pass it to lua again in execute(), or find a way to store it in evaluate() and retrieve it back in execute() 
20100322 20:08:50-!- gabm [~gabm@205.240-ppp.3menatwork.com] has joined #wesnoth-dev
20100322 20:09:03< Crab_> this data is transient, we don't care if it's lost on save/load/end turn/etc
20100322 20:09:32< Darkas> is it executed on the same lua state?
20100322 20:09:40< Crab_> yes, there's only one lua state in wesnoth
20100322 20:09:44< Crab_> see src/scripting/lua.cpp
20100322 20:09:47< Darkas> ok, that's good
20100322 20:10:04< Darkas> the rest are coroutines then?
20100322 20:10:39-!- mjs-de [~mjs-de@vpw.wh.Uni-Dortmund.DE] has quit [Ping timeout: 240 seconds]
20100322 20:10:39< Crab_> Darkas: no, we're singlethreaded atm
20100322 20:11:05-!- General_Fou_Fou [~Sarah@cpe-76-179-189-81.maine.res.rr.com] has joined #wesnoth-dev
20100322 20:11:10< Darkas> Crab_: ok, that makes things easier
20100322 20:11:50< Crab_> Darkas: either approach to fix it is good enough, you can choose. currently, there's half-done approach 1 ( pass info back to c++ in evaluate() as a serialized config object and pass it to lua again in execute() )
20100322 20:12:06< Crab_> it's done in src/ai/composite/engine_lua.cpp
20100322 20:12:33< Darkas> I think that's the better one
20100322 20:12:54< Crab_> and it's not done in src/scripting/lua.cpp ( see lua_ai_action_handler::handle(config &/*cfg*/) )
20100322 20:13:53< Crab_> we can add something like 'if we're a candidate action then pass ai_context+config to lua code and get a config back from lua code else pass just ai_context to lua'
20100322 20:15:06< Crab_> basically, when we init the lua engine we retrieve a lua table from user code which'll contain the 'user ai', and then, on each call to lua stage/candidate action execute/candidate action evaluate, we pass that 'user ai' as an argument'
20100322 20:15:25< Crab_> see data/ai/scenarios/scenario-lua-ai.cfg for example usage
20100322 20:15:34< Darkas> why not put the ai_context and the config in a single table?
20100322 20:15:54< Crab_> Darkas: yes, you can put that config in ai_context
20100322 20:16:04< Crab_> Darkas: and get it back from there
20100322 20:16:06< gabm> boucman, happygrue: I'm looking into the "delay shroud update on new game" option that was mentioned the other day on the mailing list
20100322 20:16:20< Crab_> Darkas: just note that the ai context is shared by all stages/candidate actions of a single ai
20100322 20:16:42< Darkas> Crab_: ok, then it's a bit more difficult
20100322 20:17:02< Crab_> Darkas: yes, either passing the config around from/to c++ , or some tricks with unique keyings
20100322 20:17:03< Darkas> I think then it's better to put it in another table
20100322 20:17:51< boucman> gabm: yes, and how is it going ?
20100322 20:17:52< Crab_> Darkas: ok. feel free to suggest a solution, taking into account the fact that 12:00 GMT March 24th is a testfreeze, and I want this to be committed before that
20100322 20:18:00< Darkas> I think simply adding a parameter with default to NULL would to the trick
20100322 20:18:45< Crab_> Darkas: I personally think that passing a config to/from c++ code will be good enough for now, and is easier to do than other solutions
20100322 20:18:46< timotei21> hey Crab_, what it means the "testfreeze"
20100322 20:19:09< timotei21> I've never encountered this term before
20100322 20:19:13< gabm> boucman: When you add a new option to preferences, how do you make sure it's initialized?
20100322 20:19:30< Darkas> Crab_: me too
20100322 20:19:36< boucman> I usually copy/paste existing code :P
20100322 20:19:49< Darkas> Crab_: is there any good way I can test my code?
20100322 20:20:00< boucman> IIRC you just use a default value when reading the WML
20100322 20:20:20< gabm> boucman: :P
20100322 20:21:04< Crab_> Darkas: add a candidate action evaluation loop and a lua candidate action to data/ai/scenarios/scenario-lua-ai.cfg , launch "wesnoth -t lua_ai -d"
20100322 20:21:04< gabm> boucman: I'll experiment with what I found then
20100322 20:21:35< boucman> gabm: if you look into closed patches, I think there was a couple that added options
20100322 20:21:43< boucman> though they should be pretty old by now...
20100322 20:21:47< Crab_> timotei21: testfreeze = "no commits except where approved by release manager", testing to prepare for 1.8 release.
20100322 20:21:55< gabm> boucman: ok
20100322 20:22:34< Crab_> Darkas: I can code a sample test for you to modify
20100322 20:22:53< boucman> Crab_: except in that particular case it seems rc1 will become 1.8 and the freeze is more of a pre-branch commit time (IICRC, Ivanovic could tell us the details)
20100322 20:22:57< Darkas> Crab_: that would be nice, thanks
20100322 20:24:08< gabm> boucman: another question: delay_shroud_updates_ is a boolean member of team...
20100322 20:24:35< gabm> boucman: my problem, then, is how to go from preferences, which is local to a client, to team
20100322 20:24:57< boucman> hmm
20100322 20:25:11< gabm> boucman: how to tell which team(s) are linked to the local client?
20100322 20:26:16< boucman> gabm: remember it's only a "default on new game" so my advice would be simply to default the value in "team" from the one in config, but not back it to config if it's changed in the team
20100322 20:26:25< boucman> only if it's changed in the preference menu
20100322 20:27:27< Ivanovic> boucman: it is exactly what Crab_ said
20100322 20:27:41< boucman> oops, my bad
20100322 20:27:45< Ivanovic> it is the time where svn is basically frozen so that everyone can do some last minuite testing
20100322 20:27:55< Ivanovic> to see if the stuff currently in trunk is fine
20100322 20:28:05< Ivanovic> if problems are found, talk to me before commiting
20100322 20:28:32< Ivanovic> that is: i will already bump the version number to 1.8.0 when the commit freeze starts
20100322 20:28:44< Ivanovic> about 24h later i will (if things do work!) tag 1.8.0
20100322 20:28:50< boucman> ilor: still no crash :(
20100322 20:29:14< Ivanovic> another 24h after tagging i will branch trunk off into 1.8 and trunk is open for the 1.9.x commits
20100322 20:29:24< Ivanovic> those extra 24h are for spelling fixes and the likes
20100322 20:31:32< Crab_> Darkas: what do you think is better/easier for UMC creators ? a) "return {score=10, from_x=3, from_y=5, to_x=5, to_y=8}" or  b)           "return 10, {from_x=3, from_y=5, to_x=5, to_y=8}"
20100322 20:31:54< gabm> boucman: I don't know much about the client-server structure, but is "team" used server-side? or always initialized client-side?
20100322 20:32:27< Crab_> e.g., a) return a config representation containing the score  b) make score first return value, and optional config representation - optional second return value
20100322 20:32:48< gabm> boucman: If each client initializes its own teams then I can use values from "preferences in the constructor, otherwise it doesn't make sense.
20100322 20:32:58< boucman> gabm: i don't know, yogi might know more...
20100322 20:32:59< Crab_> (see http://wiki.wesnoth.org/LuaWML#Encoding_WML_objects_into_Lua_tables for 'config representation' )
20100322 20:33:12< Darkas> Crab_: I think b) is better
20100322 20:33:16< boucman> probably not "server side" per se, but more communicated from client to client
20100322 20:33:34< Darkas> because it differs between the score and the other things
20100322 20:33:41< boucman> not sure if the delay shroud update is used over the ntework... my guess would be no, but it's just a guess
20100322 20:33:46< Crab_> Darkas: yes, and all other things are optional.
20100322 20:34:10< Darkas> yeah, that too
20100322 20:34:45< gabm> wesbot: seen YogiHH
20100322 20:34:46< wesbot> gabm: The person with the nick YogiHH last spoke 20h 34m ago.  20h 34m ago person left:
20100322 20:35:00-!- teaser [~teaser@h-37-106.A254.priv.bahnhof.se] has quit [Remote host closed the connection]
20100322 20:35:02< Gambit> Is there anyway to turn off that annoying rooster crow in WML?
20100322 20:35:51< noy> I need more rooster in my version
20100322 20:35:51-!- teaser [~teaser@h-37-106.A254.priv.bahnhof.se] has joined #wesnoth-dev
20100322 20:35:56< gabm> YogiHH: I'd appreciate if you could take a look at my conversation with boucman, above
20100322 20:37:38< Crab_> gamb: see the CONTROLLER controller; field of team
20100322 20:38:00< gabm> boucman: thanks for the tips
20100322 20:39:26< Ivanovic> zookeeper: https://gna.org/patch/?1549
20100322 20:39:31< Ivanovic> zookeeper: could you look at this patch?
20100322 20:39:40< gabm> Crab_: ah. So HUMAN basically means local
20100322 20:39:59< Crab_> gabm: yes
20100322 20:40:06< zookeeper> Ivanovic, yeah, give me a minute..
20100322 20:40:21< Ivanovic> ilor: you saw those two reports? https://gna.org/bugs/index.php?15668 and https://gna.org/bugs/index.php?15669 ?
20100322 20:40:46< Ivanovic> ilor: and yes, i think especially for the mp lobby it would be great to use the default playlist instead of playing the main menu music all the time
20100322 20:41:02< Crab_> Darkas: see http://wesnoth.pastebin.com/t7xJnn0K
20100322 20:41:12-!- mjs-de [~mjs-de@wh.uni-dortmund.de] has joined #wesnoth-dev
20100322 20:41:26< gabm> Crab_: thanks
20100322 20:41:33< ilor> Ivanovic: I think I can manage than this evening
20100322 20:41:41< Ivanovic> ilor: would be really great
20100322 20:41:56< Darkas> Crab_: thanks
20100322 20:42:02< Crab_> Darkas: ./wesnoth -t lua_ai -d --log-debug='ai/stage/rca'
20100322 20:42:06< Darkas> is this a complete scenario?
20100322 20:42:25< ilor> I didn't notice that because I develop with --no-music everywhere... the first 10 seconds of the title music get old fast
20100322 20:42:35< Crab_> yes, replace the code in data/ai/scenarios/scenario-lua-ai.cfg with it
20100322 20:42:51< zookeeper> Ivanovic, urgh, looks like a major rewrite
20100322 20:43:06< Ivanovic> zookeeper: outsch
20100322 20:43:28-!- qemqemqem_ [~quassel@urwireless-dhcp-128-151-22-157.wireless.rochester.edu] has joined #wesnoth-dev
20100322 20:44:09< Crab_> Darkas: or see the updated version - http://wesnoth.pastebin.com/Bf2DVJc3
20100322 20:44:16< Crab_> Darkas: i've added 'name=' to both candidate actions
20100322 20:44:24< Crab_> this is to make the debug output slightly pretties
20100322 20:44:27< Crab_> *prettier
20100322 20:44:32< Darkas> Crab_: in what file should I put that?
20100322 20:44:47< Crab_> download it as file and replace your data/ai/scenarios/scenario-lua-ai.cfg with it
20100322 20:44:59< Crab_> then launch with ./wesnoth -t lua_ai -d --log-debug='ai/stage/rca'
20100322 20:45:10< zookeeper> Ivanovic, so in any case it'll take time to just look through that and test if everything works
20100322 20:48:31< Darkas> Crab_: should I try playing that scenario?
20100322 20:48:42< mordante> orfest, what kind of gui changes do you need?
20100322 20:49:27< Crab_> Darkas: no, it's only purpose is to test if lua ai candidate actions work as expected
20100322 20:49:48< Crab_> Darkas: that's a testscase for a first part of the task - making them work with no data returned except the score
20100322 20:49:50< Darkas> Crab_: so I just end my turn and see what happens?
20100322 20:49:53< Crab_> yes
20100322 20:49:57< Darkas> ok
20100322 20:50:15< Crab_> and see the console output 
20100322 20:50:39-!- qemqemqem_ [~quassel@urwireless-dhcp-128-151-22-157.wireless.rochester.edu] has quit [Ping timeout: 241 seconds]
20100322 20:51:50< Darkas> Crab_: no output, shouldn't that be logged to the file?
20100322 20:52:15< Crab_> no in-game output, or no console output, or no both ?
20100322 20:53:25< Darkas> Crab_: only the message that tells me that it's round 2
20100322 20:53:38< Darkas> err, round 1, computer side
20100322 20:53:41-!- meric [~Eric@124-171-9-200.dyn.iinet.net.au] has quit [Quit: meric]
20100322 20:53:46< Crab_> hmm.. that's the old code
20100322 20:53:50< Darkas> and no console output
20100322 20:54:04< Darkas> maybe it wasn't overwritten properly
20100322 20:54:51-!- General_Fou_Fou [~Sarah@cpe-76-179-189-81.maine.res.rr.com] has quit [Quit: General_Fou_Fou]
20100322 20:58:04< Crab_> Darkas: maybe. the new version should say 'hello from candidate action evaluation!' 'hello from second candidate action evaluation!' 'hello from stage!'
20100322 20:58:16< Darkas> Crab_: ok
20100322 20:59:06< Crab_> Darkas: the old version said something about current turn
20100322 20:59:21< Crab_> be sure to launch with -d, and with that --log-debug statement
20100322 20:59:35< Darkas> Crab_: looks good now
20100322 21:00:30< Crab_> good. and the console log output ?
20100322 21:01:26< Darkas> looks like there's happening something
20100322 21:02:00-!- Ken_Oh [~briang@static-71-178-174-219.washdc.fios.verizon.net] has quit [Read error: Connection reset by peer]
20100322 21:02:38< Darkas> ok, so if my code works, the candidate actions do what?
20100322 21:03:03< Crab_> they will return correct values (42 and 99), and the one which returned 99 will be executed
20100322 21:03:41< Crab_> then, some funny things will happen :) and, sooner or later, both actions will be blacklisted and the turn will end with a hello from stage.
20100322 21:03:50< Darkas> ok :)
20100322 21:04:03< Darkas> I think I'll get it working tomorrow ;)
20100322 21:04:48< Crab_> note: if it won't work tomorrow in the evening, I'll need to do it myself - I need to commit it before 1.8. but I trust you to make it happen :)
20100322 21:06:53< Crab_> note that you'll need to mess with the code in src/ai/composite/engine_lua.cpp to check the correct return value instead of cfg["score"] - our put that score in the config before returning the cfg from src/scripting/lua.cpp
20100322 21:07:05< Crab_> s/our/or
20100322 21:08:42< Darkas> Crab_: ok
20100322 21:08:54< Darkas> I'll try my best, but I have lots of time today :)
20100322 21:09:02< Crab_> good :)
20100322 21:09:06< Darkas> tomorrow, not today
20100322 21:17:01-!- mjs-de [~mjs-de@wh.uni-dortmund.de] has quit [Ping timeout: 258 seconds]
20100322 21:18:06-!- Silenthunter [~Silenthun@97-81-86-140.dhcp.athn.ga.charter.com] has joined #wesnoth-dev
20100322 21:19:04< Darkas> Crab_: will you be here tomorrow, if I have any questions?
20100322 21:19:32< Crab_> Darkas: yes, although only partially during the 'day' (as of GMT)
20100322 21:20:01< Darkas> Crab_: that's ok
20100322 21:20:19< Crab_> anything else that you need ?
20100322 21:20:34-!- mjs-de [~mjs-de@wh.uni-dortmund.de] has joined #wesnoth-dev
20100322 21:22:19< Darkas> maybe a bit, I'm not really sure what lua_ai_action_handler::handle does with the aisKey
20100322 21:22:42< CIA-80> crab * r41682 /trunk/ (changelog data/ai/formula/new_recruitment.fai): Fix bug #14247: make formula_ai behave correctly if the side has only 1 recruit
20100322 21:26:32< Crab_> Darkas: in ai code, we need to initialize a lua table for each ai, and we need to compile a lua function for each handler that we have. to keep this compiled code within reach, we store a table  (search for 'Create the ai elements table' in lua.cpp ) and store a index of the element of that table which contains our function
20100322 21:26:53< CIA-80> ilor * r41683 /trunk/ (data/multiplayer/_main.cfg src/multiplayer.cpp): make the lobby play music defined in a [lobby_music] wml tag and set it to default_playlist + main menu theme. Fixes bug #15668.
20100322 21:27:19< Crab_> Darkas: so, lua_ai_action_handler::handle gets that 'global table which contains various ai objects' first, and then gets a specific element from this table which contains our current function
20100322 21:31:07< boucman> Mythological: what era are you using on the 1.8 server, I can't find it...
20100322 21:31:58< Darkas> Crab_: ok, I think I understood
20100322 21:32:22< Darkas> I just have to refresh my knowledge about stack manipulation and light userdata a bit
20100322 21:32:44< Crab_> see the lua engine initialization in data/scenarios/scenario-lua-ai.cfg
20100322 21:33:05< Crab_> we store the returned value of the function as the 'ai context'
20100322 21:33:13< Crab_> we also store it in that global table
20100322 21:37:28-!- schumi [~a11848c0@gateway/web/freenode/x-tqcpuhoatbwxzugi] has joined #wesnoth-dev
20100322 21:38:02< CIA-80> ilor * r41684 /trunk/ (data/core/editor/_main.cfg src/game.cpp): 
20100322 21:38:02< CIA-80> Make the titlescreen play more than just the mian menu song, fixing fr / bug #15669
20100322 21:38:02< CIA-80> The playlist is taken from a new [titlescreen_music] tag, which has been set to default_playlist + main menu song
20100322 21:38:02< CIA-80> Note that music will always restart with the main_menu theme when going back into the titlescreen e.g. from the editor, it'd be difficult otherwise to make it always start with main_menu.ogg on startup.
20100322 21:38:13< ilor> Ivanovic: ^
20100322 21:39:08< Darkas> Crab_: where is the lua engine initialized in that file?
20100322 21:39:26< Darkas> I can just find a few functions
20100322 21:39:41< Crab_> [side]side=2[engine] name=lua code= << HERE >>
20100322 21:41:43-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: GO, GET TO THE CHOPPAH!!!]
20100322 21:42:13-!- silene1 [~plouf@AMarseille-251-1-65-254.w83-205.abo.wanadoo.fr] has joined #wesnoth-dev
20100322 21:42:35< Darkas> Crab_: ah, now I understood
20100322 21:42:44< Darkas> in this case, it's empty of course
20100322 21:42:54< Crab_> why empty ?
20100322 21:42:59-!- schumi [~a11848c0@gateway/web/freenode/x-tqcpuhoatbwxzugi] has left #wesnoth-dev []
20100322 21:43:13-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100322 21:43:16< Darkas> my_ai is just empty isn't it?
20100322 21:43:20-!- noy [~Noy@wesnoth/developer/noy] has quit [Remote host closed the connection]
20100322 21:43:31< Crab_> a few functions are in there :)
20100322 21:43:34-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has joined #wesnoth-dev
20100322 21:43:51< Crab_> e.g., my_ai:stage_hello() adds a function to my_ai table.
20100322 21:43:55< Darkas> ah, right, I totally oversaw
20100322 21:44:00-!- silene [~plouf@wesnoth/developer/silene] has quit [Ping timeout: 252 seconds]
20100322 21:44:27< Crab_> Darkas: that returned value (my_ai) is stored and then is passed as first argument to stage and candidate action handlers
20100322 21:44:31-!- Espreon [~espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20100322 21:44:34< [Relic]> Hello :)
20100322 21:44:53< Espreon> timotei21: http://en.wiktionary.org/wiki/servus#Etymology
20100322 21:44:56< Crab_> Darkas:  that (...):stage_hello() is actually "    local my_ai = ...                 my_ai:stage_hello()   "
20100322 21:46:07< Darkas> I see
20100322 21:46:15< Crab_> Darkas: there's even a proposal (a good proposal) from silene https://gna.org/patch/?1432#comment5 to make it lighter - just allow to specify a method name in stages and candidate actions
20100322 21:48:06< Espreon> Also, ic gretaþ ge.
20100322 21:48:35< Espreon> No...
20100322 21:48:37< Espreon> I fail...
20100322 21:48:41< Crab_> Darkas: see prototype: http://wesnoth.pastebin.com/yyBgjNVy
20100322 21:49:06< Crab_> Darkas: that way it should be simplier for WML authors...
20100322 21:49:07< Espreon> *grete
20100322 21:49:29< Darkas> Crab_: yeah, that looks good
20100322 21:49:45< Crab_> Darkas: that's another good thing to do after you finish this one.
20100322 21:50:12< Darkas> Crab_: right, that should be no problem :)
20100322 21:50:45< Crab_> it'll hide those "local ai = ... local my_ai = { }   local ai_stdlib = wesnoth.require('ai/lua/stdlib.lua'); ai_stdlib.init(ai)"
20100322 21:51:05< Crab_> ( ai/lua/stdlib.lua is the file to which lua-based standard functions are to be added)
20100322 21:52:35< Darkas> Crab_: ok
20100322 21:53:17< timotei21> Esperon: thanks for the link
20100322 21:53:25< timotei21> Espreon*
20100322 21:53:32< Espreon> You are welcome.
20100322 21:55:19-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 240 seconds]
20100322 21:56:38< Ivanovic> ilor: cool, thanks
20100322 21:56:49-!- qemqemqem_ [~quassel@128.151.29.128] has joined #wesnoth-dev
20100322 21:57:24-!- silene1 [~plouf@AMarseille-251-1-65-254.w83-205.abo.wanadoo.fr] has quit [Quit: Leaving.]
20100322 21:58:34-!- YogiHH [YogiHH@d081167.adsl.hansenet.de] has joined #wesnoth-dev
20100322 21:58:45-!- YogiHH [YogiHH@d081167.adsl.hansenet.de] has quit [Changing host]
20100322 21:58:45-!- YogiHH [YogiHH@wesnoth/developer/yogihh] has joined #wesnoth-dev
20100322 21:59:35-!- General_Fou_Fou [~Sarah@cpe-76-179-189-81.maine.res.rr.com] has joined #wesnoth-dev
20100322 22:00:48< fendrin> Crab_: I fixed the crashing eclipse. It was a bugy mozilla version that is used by eclipse to display javadoc.
20100322 22:01:06< Crab_> that's good to hear :)
20100322 22:01:35-!- General_Fou_Fou [~Sarah@cpe-76-179-189-81.maine.res.rr.com] has quit [Client Quit]
20100322 22:02:18< fendrin> Crab_: Yeah, it was really annoying and slowed me down *alot*
20100322 22:03:07< Zarel> YogiHH: Are you Yogibear on the forums?
20100322 22:03:28< YogiHH> Zarel: yep
20100322 22:03:51< Zarel> YogiHH: Cool! Was your comment directed at me or Velensk? It was kinda hard to tell.
20100322 22:04:08< YogiHH> you mean "gameplay trumps realism"?
20100322 22:05:06< Darkas> Crab_: I'll go into bed now
20100322 22:05:09< Darkas> bye
20100322 22:05:10< fendrin> Crab_: I have thought about the wml editor highlighting. Have you considered to change the autogenerated code to read AI0815's syntax?
20100322 22:05:33< Zarel> YogiHH: Yeah.
20100322 22:05:37< CIA-80> ivanovic * r41685 /website/start/1.8/ (13 files in 2 dirs): updated German announcement translation
20100322 22:06:06< YogiHH> Zarel: It was directed to anyone taking part in a more or less pointless discussion as the thread is about people not having fun playing campaigns. So everything that is giving them more fun (and i think your proposal does) is a good thing if it doesn't drive others away (which i think your proposal does not)
20100322 22:06:21< Zarel> Okay.
20100322 22:06:31-!- Darkas [~quassel@ppp-93-104-137-87.dynamic.mnet-online.de] has quit [Read error: Connection reset by peer]
20100322 22:06:35-!- Darkas [~quassel@ppp-93-104-137-87.dynamic.mnet-online.de] has joined #wesnoth-dev
20100322 22:06:36< Crab_> fendrin: when the code is already autogenerated, it's too late to read the schema - since autogenerated code already contains it.
20100322 22:06:52< Crab_> fendrin: as the code is autogenerated on the grammar which represents the language schema
20100322 22:06:54-!- Darkas [~quassel@ppp-93-104-137-87.dynamic.mnet-online.de] has quit [Remote host closed the connection]
20100322 22:07:12< fendrin> Crab_: Ah, so we would need to recode the autogenerator.
20100322 22:07:36< Crab_> fendrin: we would need to add a preprocessing step just before autogenerator
20100322 22:07:45-!- wesbot changed the topic of #wesnoth-dev to: 1.8 commits allowed till March, 24th, 12:00 GMT, afterwards "testfreeze" | string/feature freeze active! | 70 bugs, 254 feature requests, 14 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100322 22:08:03< Crab_> fendrin: to autogenerate the grammar which will be used to autogenerate the eclipse editor plugin.
20100322 22:08:12< Ivanovic> ilor: one thing missing in your commit: the changelog entry!
20100322 22:08:21< ilor> Ivanovic: ow
20100322 22:08:41< ilor> Ivanovic: commits, two of them ;)
20100322 22:08:42< fendrin> Crab_: Is the generated editor mighty enough to justify the work?
20100322 22:08:53< CIA-80> ivanovic * r41686 /trunk/ (17 files in 15 dirs): updated Finnish and Latvian translation
20100322 22:09:31< Ivanovic> wesbot: topic
20100322 22:09:33< wesbot> Ivanovic: Don't be impatient, wait until something changes...  *tsk*
20100322 22:09:52< YogiHH> lol
20100322 22:09:58< Crab_> Ivanovic: wesbots should be more friendly towards release managers, isn't it :) ?
20100322 22:10:11< Ivanovic> can anyone reproduce this or is it a case of broken linking of libjpeg? https://gna.org/bugs/index.php?15676
20100322 22:10:20< Crab_> fendrin: well, syntax highlight, context-sensitive autocompletion for tags, and other interesting things not related to editor. I'd say yes, it's worth it
20100322 22:11:00< CIA-80> ivanovic * r41687 /trunk/data/core/images/units/monsters/ (20 files): 
20100322 22:11:00< CIA-80> ran utils/wesnoth-optipng on the new cuttlefish
20100322 22:11:00< CIA-80> Overall statistics (only for files with a smaller recompressed size):
20100322 22:11:00< CIA-80> Original size: 68 KiB on 20 files
20100322 22:11:00< CIA-80> Optimized size: 29 KiB
20100322 22:11:00< CIA-80> Total saving: 39 KiB = 57% decrease
20100322 22:11:13< Crab_> error reporting (highlight errors in wml), outline, etc.
20100322 22:11:27< gabm> boucman, happygrue: here you go, there's your option, tested in multiplayer and everything: https://gna.org/patch/index.php?1556 . Please review!
20100322 22:11:50< fendrin> Crab_: That sounds good.
20100322 22:11:59< Crab_> basically the things you get in a java editor. if we add cross-linking capabilities to the schema, then (as the docs say) it'll be possible to add a way to jump around the code like you can jump on java functions in java editors
20100322 22:12:15< Crab_> e.g. jump to other scenario from a scenario id
20100322 22:12:22< YogiHH> argh, Ivanovic :-P
20100322 22:12:42< YogiHH> I updated 2 minutes ago :-P
20100322 22:12:47< gabm> YogiHH: you can ignore my request way above in the log
20100322 22:12:53< fendrin> Crab_: Or jumping to where a variable is defined?
20100322 22:13:05< Ivanovic> YogiHH: those are 70kb
20100322 22:13:12< Ivanovic> YogiHH: your line should be able to manage those
20100322 22:13:14< Ivanovic> ;)
20100322 22:13:24< Crab_> fendrin: I think yes, if it's via set_variable tag
20100322 22:13:39< YogiHH> Ivanovic: Maybe release managers commiting all the time are the reason why people don't edit the changelog :-P
20100322 22:13:58< boucman> gabm: nothing to say, i've postponed until 1.8 split
20100322 22:14:10 * boucman is going to have A LOT to do once 1.8 is branched
20100322 22:14:22< Ivanovic> hey, i only edit the changelog when i either *fix* it (because some developers do not like the 80 chars per line limit) or when putting updated translations in
20100322 22:14:38< Ivanovic> okay, and i update when it is time to bump the version
20100322 22:14:44< gabm> boucman: alright, let me know if I did something wrong
20100322 22:15:07< YogiHH> just kidding :)
20100322 22:15:29< Ivanovic> YogiHH: though at least the OOS bugfix belongs into players_changelog, too!
20100322 22:15:30< Ivanovic> ;)
20100322 22:15:32< YogiHH> gabm: Not sure what you are referring to, but i will simply ignore it like you said ;)
20100322 22:15:45< gabm> YogiHH: :P
20100322 22:16:07< CIA-80> ilor * r41688 /trunk/changelog: changelog entry for bug #15668
20100322 22:16:15< CIA-80> ilor * r41689 /trunk/changelog: changelog entry for bug #15669
20100322 22:16:22< Ivanovic> ilor: yes, the music is player visible, don't forget players_changelog
20100322 22:16:40< YogiHH> no, no, no, don't edit the changelog again O-O
20100322 22:16:45< ilor> Ivanovic: ... *now* you remind me ;)
20100322 22:16:55< Ivanovic> ilor: of course i do
20100322 22:17:25< Ivanovic> taking a shower now, looking at the changelog afterwards
20100322 22:17:26< Ivanovic> ;)
20100322 22:17:33< YogiHH> pah!
20100322 22:17:35< gabm> YogiHH: You don't do something like this to catch up on stuff?: wget -O - http://wesnoth.debian.net/%23wesnoth-dev-2010-03-22.log | grep YogiHH
20100322 22:17:42< ilor> Ivanovic: is it really worth putting in players_changelog?
20100322 22:17:58-!- mjs-de [~mjs-de@wh.uni-dortmund.de] has quit [Ping timeout: 258 seconds]
20100322 22:18:19< Ivanovic> ilor: yes
20100322 22:19:09< YogiHH> gabm: I am not only on windows, i also don't like command lines, so i use this fantastic tool TortoiseSVN, which does all the magic for me without typing a word :) . Except preventing hypermotivated devs from editing the changelog, that is :-P
20100322 22:19:12< Ivanovic> most players tend to have music on unless they are too annoyed by the main menu theme in the lobby
20100322 22:19:33< Ivanovic> so it is an important change for them and tells them "yeah, i can turn music back on"
20100322 22:20:25< CIA-80> jhinrichs * r41690 /trunk/ (changelog src/multiplayer_create.cpp): Fix bug #13882: Map which is invalid aborts map selection
20100322 22:20:31< Espreon> YogiHH: But... typing things is fun and more efficient sometimes...
20100322 22:20:49< gabm> YogiHH: oh, I see. Actually the command above is for filtering the irc log for a certain day and see only the lines with your nick, so TortoiseSVN won't help with that (even though it's a nice tool ;) )
20100322 22:20:52-!- Crab_ [~Crab_@wesnoth/developer/crab] has quit [Quit: Leaving.]
20100322 22:20:56< YogiHH> Espreon: That is clearly a matter of taste, i can assure you :)
20100322 22:21:02-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20100322 22:21:37< gabm> YogiHH: I guess you can always open the log in your browser and Ctrl+F for you nick then... this way you won't miss my message next time :P
20100322 22:22:29< CIA-80> ilor * r41691 /trunk/ (changelog players_changelog): changelog edit and players_changelog update wrt bugs 1566[89]
20100322 22:22:45 * ilor is a good dev today ... ^
20100322 22:22:46< ilor> ;)
20100322 22:22:47< YogiHH> gabm: Oops, the chatlog o-o. Yes, i use the browser for that. But i am lazy, atm it's more a game of chance to keep messages for me in the log :)
20100322 22:22:52-!- Appleman1234 [~Appleman1@131.181.102.66] has joined #wesnoth-dev
20100322 22:22:59< ilor> Ivanovic: ^
20100322 22:24:23< boucman> YogiHH: your last commit, did you use the patch for it, or did you rewrite it ?
20100322 22:25:12< YogiHH> boucman: Errm, no, i rewrote it. Actually i wasn't aware there is a patch...
20100322 22:25:39< boucman> darn :(
20100322 22:26:15< boucman> YogiHH: how similar was it to the patch ?
20100322 22:26:21 * YogiHH checks
20100322 22:28:52< esr> Ivanovic: Is it still policy to declare UMC-only WML bugs invalid and boot them off the main tracker?  See  bug #15673.
20100322 22:30:49-!- mjs-de [~mjs-de@wh.uni-dortmund.de] has joined #wesnoth-dev
20100322 22:30:49-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100322 22:31:17< YogiHH> boucman: It's quite different in that i prevented the "launch_game" button to be available if the map is not valid instead of catching the exception afterwards.
20100322 22:31:41< YogiHH> boucman: I can revert it, if you want, so you can put in the patch
20100322 22:33:11< boucman> YogiHH: hmm, not sure what's best...
20100322 22:33:20< CIA-80> jhinrichs * r41692 /trunk/players_changelog: Updating the players_changelog with relevant bug fixes.
20100322 22:33:58< YogiHH> boucman: You can have a look at the fix and i leave it up to you, ok?
20100322 22:34:46< boucman> ok, i'll comment on the patch and leave your commit
20100322 22:36:49< YogiHH> night everyone
20100322 22:37:05< Crab_> night, YogiHH
20100322 22:37:46-!- YogiHH [YogiHH@wesnoth/developer/yogihh] has left #wesnoth-dev []
20100322 22:38:10-!- stikonas [~and@2002:4f84:b38b:b:226:9eff:fe0a:3e55] has joined #wesnoth-dev
20100322 22:38:10-!- stikonas [~and@2002:4f84:b38b:b:226:9eff:fe0a:3e55] has quit [Changing host]
20100322 22:38:10-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100322 22:39:22-!- General_Fou_Fou [~Sarah@rrcs-69-193-74-3.nys.biz.rr.com] has joined #wesnoth-dev
20100322 22:43:12-!- zielony_kac [~zielony_k@chello089077167251.chello.pl] has joined #wesnoth-dev
20100322 22:43:19< timotei21> hey, has anyone time to look over a 1st draft of my wiki page?:D
20100322 22:43:20< timotei21> http://wiki.wesnoth.org/SummerofCode_Timotei21
20100322 22:43:25< timotei21> for GSOC?
20100322 22:43:36< zielony_kac> hello.
20100322 22:43:52< boucman> timotei21: doing..
20100322 22:43:52< timotei21> is not complete, yet
20100322 22:43:56< boucman> hello zielony_kac
20100322 22:43:58< timotei21> hello zielony_kac
20100322 22:44:29< Crab_> timotei21: I'll take a look, too
20100322 22:44:38< timotei21> ok, thank you
20100322 22:44:38< fendrin> Is the world of wesnoth a round planet?
20100322 22:45:02< boucman> fendrin: probably, though I don't think it has ever been formally decided
20100322 22:46:31< fendrin> boucman: Okay, so assume that as fact from now on.
20100322 22:46:38< boucman> :)
20100322 22:46:49< Espreon> fendrin: We just call it Irdya.
20100322 22:47:01< fendrin> Espreon: Is it a round planet?
20100322 22:47:18 * Espreon shrugs
20100322 22:47:30< fendrin> I thought I had read that Irdya is a plain.
20100322 22:47:45< Crab_> timotei21: ok, read through your wiki page page. Looks good. but, it would be interesting to see when you'd put information about the project that you'll propose.
20100322 22:47:54< Espreon> fendrin: Hwy do you want to know?
20100322 22:48:27< fendrin> Espreon: I have just read about how the second sun was created and what happened after that.
20100322 22:48:36< timotei21> yeah, I'm gonna decide on that tomorrow. I've been very busy lately, but shoud be more free next weeks
20100322 22:48:37< Espreon> Ah, I see...
20100322 22:48:42< fendrin> But it doesn't make quite sense.
20100322 22:48:52< boucman> ilor: I got it to crash
20100322 22:49:19< fendrin> The story explains the creation of the long dark with one of the suns touching the earth and getting out of track.
20100322 22:49:21< ilor> boucman: what's the crash message?
20100322 22:49:46< boucman> 20100322 22:45:06 info lobby: Clean game in listbox 3066 (row 6)
20100322 22:49:46< boucman> 20100322 22:45:06 info lobby: Clean game in listbox 3068 (row 7)
20100322 22:49:46< boucman> 20100322 22:45:06 info lobby: Clean game in listbox 3069 (row 8)
20100322 22:49:46< boucman> 20100322 22:45:06 info lobby: Adding game to listbox 3070
20100322 22:49:46< boucman> 20100322 22:45:06 debug lobby: lobby_info::sync_games_display_status20100322 22:45:06 debug lobby: games_by_id_ size: 1020100322 22:45:06 debug lobby:  -> 10
20100322 22:49:47< boucman> wesnoth-debug: src/gui/widgets/grid.cpp:538: virtual void gui2::tgrid::place(const gui2::tpoint&, const gui2::tpoint&): Assertion `false' failed.
20100322 22:50:02< fendrin> That explains why the world is no longer all day like it was before and what was the whole reason to create the second sun.
20100322 22:50:24< fendrin> But it doesn't explain why the long night is double as long as the old night.
20100322 22:50:42-!- schumi [~a11848c0@gateway/web/freenode/x-lyeonlygpddvhlku] has joined #wesnoth-dev
20100322 22:51:00< boucman> ilor: backtrace at http://wesnoth.pastebin.com/nyxDj2XK
20100322 22:51:07< fendrin> So I thought it would be a good explanation if the sun touching irdya has also slowed the planets rotation down.
20100322 22:52:05< fendrin> esr: ^Is that your resort?
20100322 22:52:14< Crab_> fendrin: well, 'the world has moved on' is also a good explanation :)
20100322 22:52:29< zookeeper> fendrin, or the second sun just slows the first one down. gravity or whatever.
20100322 22:52:32< ilor> boucman: if you have the --log-debug=lobby, can you paste the last few kilobytes of it?
20100322 22:52:49< boucman> timotei21: i'll second what Crab_ said, that's good so far, but the project is one of the most interesting part for us, so not much more to say at this point
20100322 22:52:56< boucman> ilor: sure
20100322 22:53:41< boucman> ilor: http://wesnoth.pastebin.com/4LN3BizU
20100322 22:53:55< fendrin> zookeeper: No, the first one is a real sun like hours. In the center of everything. The artificial sun is only a small rock fired by magic. It is far nearer to earth. Your explanation doesn't work.
20100322 22:54:06< ilor> how sweet, the assert has a "// This shouldn't be possible..." comment right above it
20100322 22:54:19< timotei21> boucman, Crab_: ok, thanks
20100322 22:54:20< ilor> boucman: this looks really more like mordante's issue
20100322 22:54:26< fendrin> Crab_: You suggest that the distance between irdya and the central sun has increased?
20100322 22:55:03< boucman> mordante: around ?
20100322 22:55:24< ilor> it died right after a new game appeared
20100322 22:55:50< ilor> I figure it may be related to the new game appearing and a particural view of the gamelist
20100322 22:56:09-!- Blarumyrran [~Blarumyrr@81-20-159-197.levira.ee] has quit [Quit: Lahkun]
20100322 22:56:11< Crab_> fendrin: no, I was quoting the "Dark Tower"
20100322 22:56:35< fendrin> hmmmm
20100322 22:56:40< fendrin> never heard off
20100322 22:56:44< fendrin> s/off/of
20100322 22:56:50< Ivanovic> esr: the policy is unchanged IIRC
20100322 22:56:58< Gambit> So Naia orbits Irdya?
20100322 22:56:59< Ivanovic> that is: if it is an engine bug: valid
20100322 22:57:25< Ivanovic> if it is a problem due to documentation not covering this extra case: cover it and close it (or improve the engine if it makes sense)
20100322 22:57:53< Ivanovic> if it is a bug due to faulty wml in the addon: ask to post this in the respective forum thread
20100322 22:58:52< ilor> boucman: mordante: I can be sure that given those logs, the number if items in the visibility vector passed to the listbox is correct, so it's a weird gui2 glitch
20100322 22:58:58-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit []
20100322 22:59:55< boucman> mordante: i'll keep the debug session opened until you're around or I go to bed...
20100322 23:00:09< mordante> boucman, not really, finishing my work on the German translation and then I head off to bed
20100322 23:00:13< ilor> boucman: what are the values of best_size and size in frame #3?
20100322 23:00:24< esr> Ivanovic: OK, I'll boot that bug.
20100322 23:00:32< timotei21> OOS errors in LoW
20100322 23:00:35< timotei21> what is OOS?
20100322 23:00:40< boucman> (gdb) p best_size
20100322 23:00:41< boucman> $1 = {x = 200, y = 17}
20100322 23:00:45< Crab_> timotei21: Out-of-Sync
20100322 23:00:52< timotei21> oky
20100322 23:00:54< boucman> $2 = (const gui2::tpoint &) @0x7fffffffb160: {x = 191, y = 17}
20100322 23:01:14< mordante> boucman, but what did you change to get this output "games_by_id_ size: 1020100322 22:45:06 debug lobby:  -> 10"
20100322 23:01:34< mordante> or better said, does it mean what I think ;-)
20100322 23:01:53< boucman> mordante: ??? the only change I have to trunk afaik is the extra asserts you had me add yesterday
20100322 23:02:28< mordante> ilor, I prefer assertion failures over going on in a corrupted state and ending with a crash later on ;-)
20100322 23:02:30< Crab_> timotei21: wesnoth's multiplayer works by sending replays across the network, and the other side receives the replay data and does the same modifications to the game state as the 'our' side. so, they independently modify the game state, but in the same way, 'in sync'.
20100322 23:02:43< Crab_> timotei21:  when something goes wrongly, things can go 'out of sync' - when different clients consider the game state to be different. this is 'out of sync', OOS. and this is a bad thing.
20100322 23:03:08< ilor> mordante: this debug info is the result of me forgetting what happens if you LOG_ twice without a newline
20100322 23:03:29< ilor> mordante: unclobbered it'd be 10 -> 10 which is what should be there
20100322 23:03:38< timotei21> oh, I understand:D
20100322 23:03:59< timotei21> thanks
20100322 23:04:04< mordante> ilor, ah ok 
20100322 23:04:25< mordante> boucman, I thought you added some extra debug statements
20100322 23:05:13< boucman> no, only extra asserts
20100322 23:05:18< mordante> but I won't have time tonight, really about to go to bed only need to fix one more string and then I can email the translation
20100322 23:05:29< mordante> boucman, yeah ilor explained the output
20100322 23:06:01< ilor> maybe grab a core dump? someone who knows gdb would be useful
20100322 23:06:07< CIA-80> ivanovic * r41693 /trunk/ (changelog players_changelog): 
20100322 23:06:07< CIA-80> some changelog fixes/changes:
20100322 23:06:07< CIA-80> 1) use past for "Fixed bug #ABC"
20100322 23:06:07< CIA-80> 2) keep conventions for players_changelog in regards to "end each entry as if it was a sentence"
20100322 23:06:07< CIA-80> 3) sort items in misc secion alphabetically/by bug number
20100322 23:06:11< boucman> mordante: ok, i might have time to try to reproduce on wensday, but it seems to happen only when I am about to go to bed...
20100322 23:06:21< boucman> hmm
20100322 23:06:42< boucman> could an incorrect client joining could cause that ?
20100322 23:07:14< ilor> no. the stack trace suggest a weird glitch in gui2 sizing as far as I can tell
20100322 23:08:06-!- Mythological_ [Mythologic@77.28.116.82] has joined #wesnoth-dev
20100322 23:08:10< ilor> *suggests
20100322 23:08:48< schumi> hello
20100322 23:09:04< schumi> I'm here for the summer of code
20100322 23:09:15< boucman> hello schumi
20100322 23:09:21< Crab_> hi, schumi
20100322 23:10:50< schumi> how can I start the application?
20100322 23:11:03< Crab_> schumi: you've already started, by coming here :)
20100322 23:11:14-!- Mythological [Mythologic@77.28.125.85] has quit [Ping timeout: 268 seconds]
20100322 23:11:15-!- qemqemqem_ [~quassel@128.151.29.128] has quit [Ping timeout: 258 seconds]
20100322 23:11:24< Crab_> schumi: the next steps is to talk about the possible projects, figuring out the thing that you'd want to do
20100322 23:11:42-!- zielony_kac [~zielony_k@chello089077167251.chello.pl] has quit [Ping timeout: 276 seconds]
20100322 23:11:59< boucman> and start a wiki page :)
20100322 23:12:06< timotei21> schumi: here is the step-by-step guide: http://wiki.wesnoth.org/SummerOfCodeIdeas#I_want_to_be_one_of_your_Google_Summer_of_Code_students.2C_what_should_I_do...
20100322 23:12:17< mordante> boucman, and accidentally that's about the same time as I go to bed ;-)
20100322 23:12:19< schumi> Crab_: yes :).  How I can learn more about the projects?
20100322 23:12:20< Crab_> schumi: yes, boucman is right - then, set up a wiki page for yourself (by copying the template), then, starting to work towards your chosen project, compiling wesnoth, submittting patches, and hawing fun
20100322 23:13:04< Crab_> schumi: well, the http://wiki.wesnoth.org/SummerOfCodeIdeas page lists various project ideas (and, you can propose your own project, as well)
20100322 23:13:04< boucman> mordante: my question is : if a user has a tweaked client and logs at the same time every day, that might explain some of the things we see
20100322 23:13:05< schumi> I'm still thinking about which project to choose
20100322 23:13:06< mordante> boucman, ilor I also guess there's a sizing issue left
20100322 23:13:21< Crab_> schumi: yes, it's sometimes hard to choose
20100322 23:13:48< Ivanovic> how to find a project: read through the proposals, play the game, ask questions about the proposals or things you think that the game does lack
20100322 23:13:55< mordante> boucman, pretty far fetched, for now I believe it's a normal bug
20100322 23:14:08< Ivanovic> yes, a mixture of those three things makes most sense!
20100322 23:14:12< boucman> mordante: ok, 
20100322 23:14:42< boucman> mordante: i'll kill my gdb, then
20100322 23:15:29< schumi> Crab_: I've read them
20100322 23:15:34< timotei21> ok guys, had a great time here, but I have to go
20100322 23:15:40< timotei21> see you tomorrow;)
20100322 23:15:45< boucman> see you timotei21
20100322 23:15:45< timotei21> 'night and have fun
20100322 23:15:58< mordante> night timotei21 
20100322 23:16:05< Crab_> timotei21: night
20100322 23:16:14< timotei21> you can call me "timo", it's enough. I don't even know why I'd put a number after my name... but anyways
20100322 23:16:27< Crab_> :)
20100322 23:16:35-!- timotei21 [~timotei21@193.34.191.4] has quit [Quit: bye bye]
20100322 23:16:38< gabm> General question about the Summer of Code applications: can I follow a format like this: http://wiki.wesnoth.org/User:Gabba , as long as I cover all questions from the GSoC questionnaire?
20100322 23:17:17< gabm> copy-pasting the questionnaire and answering inline is ugly and unreadable imho
20100322 23:17:23-!- mjs-de [~mjs-de@wh.uni-dortmund.de] has quit [Ping timeout: 258 seconds]
20100322 23:17:38< Crab_> gabm: my personal opinion: good enough. the main part is your project idea and 'how you work'.
20100322 23:18:13< gabm> Crab_: I'll submit at least two projects, so this page will have only my personal info
20100322 23:18:35< Crab_> good :)
20100322 23:18:44< boucman> k
20100322 23:18:48< Crab_> 'at least' intrigues :)
20100322 23:19:20< gabm> I think it's gonna be just two :)
20100322 23:19:31< gabm> But I'm into idea overflow mode, so you never know :P
20100322 23:19:44-!- schumi [~a11848c0@gateway/web/freenode/x-lyeonlygpddvhlku] has quit [Ping timeout: 252 seconds]
20100322 23:19:48-!- schumi_ [~a11848c0@gateway/web/freenode/x-frfjhwsxxoivqsnr] has joined #wesnoth-dev
20100322 23:20:04< boucman> it's still early in the submission process so that's fine
20100322 23:20:09-!- schumi [~a11848c0@gateway/web/freenode/x-lkucxumntfxcywrx] has joined #wesnoth-dev
20100322 23:20:11-!- schumi_ [~a11848c0@gateway/web/freenode/x-frfjhwsxxoivqsnr] has quit [Client Quit]
20100322 23:20:26< boucman> though my advice would be to focus on one (or at most two) proposals eventually
20100322 23:20:45< boucman> you can ask for advice if you don't know what to decide
20100322 23:20:53< mordante> I'm off night
20100322 23:21:14< Crab_> Gambit: then please use the {{SoC2010}} wiki template for your user page, and 'normal' (what you get for copypasting ' SOC_Template_of_Student_Page ') proposal wiki template for your proposal pages.
20100322 23:21:15< gabm> boucman: yes, in any case I don't think I can manage more than two detailed proposals
20100322 23:21:25< Crab_> Gambit: sorry
20100322 23:21:36< Crab_> gabm: ^ 
20100322 23:21:40< schumi> sorry, I had some connecion problem
20100322 23:21:49< gabm> Crab_ : noted
20100322 23:21:51< Gambit> Crab_: Maybe a few years. :)
20100322 23:22:00-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20100322 23:22:12< Crab_> :)
20100322 23:22:41< schumi> Crab_: Can you talk more about your project (persistent storage of gameworld data)?
20100322 23:23:52-!- zielony_kac [~zielony_k@chello089077167251.chello.pl] has joined #wesnoth-dev
20100322 23:25:26-!- Appleman1234 [~Appleman1@131.181.102.66] has quit [Ping timeout: 258 seconds]
20100322 23:25:52< Crab_> schumi: when a user plays a game, he can save his progress to file. when he progresses to the next scenarios, information about his previous scenarios moves with him. but, there are many interesting things that are impossible\hard to do by just 'moving information forward'
20100322 23:26:18< Ivanovic> gabm: in fact most users only manage one good proposal
20100322 23:26:25< Ivanovic> gabm: and that is all required
20100322 23:26:31< Ivanovic> there is no need for several
20100322 23:26:33< Ivanovic> ;)
20100322 23:26:56< Crab_> schumi: the 'first hard to do \ impossible' thing is moving information 'backward in time'. e.g., when you restart a campaign, and some characters in it know that you've played this campaign before, and know your 'achievements'
20100322 23:27:21< gabm> Ivanovic: my proposal last year was rejected, so...
20100322 23:27:43< gabm> Ivanovic: I'll put most effort on a very good one, but I want to have a backup plan
20100322 23:27:43< Crab_> schumi: e.g., in diablo series, you are be able to restart your progress through the world while keeping your character levels and inventory
20100322 23:28:14-!- Mythological_ [Mythologic@77.28.116.82] has quit []
20100322 23:29:16< Crab_> schumi: if we add such a functionality, to 'store' some information and 'access' it on later plays of the same campaign/scenario to wesnoth, then user-made-content creators would be able to code very interesting things using it.
20100322 23:29:42< schumi> Crab_: so the project is about finding a way to keep this data through wml?
20100322 23:29:53< Crab_> schumi: not necessarily WML.
20100322 23:30:02< Crab_> schumi: whatever method works
20100322 23:30:18< gabm> Ivanovic: this said, if you think a second proposal is just spam, I can instead submit one to a different organisation, but that'll disperse my efforts even more
20100322 23:30:21< Crab_> schumi: for SP,different addons should be able to exchange this information as well
20100322 23:30:43-!- qemqemqem_ [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has joined #wesnoth-dev
20100322 23:31:15< Crab_> gabm: by all means, you can submit two proposals. but, you'll need to actively work on a proposal to have a chance to be accepted - will you be able to manage both ?
20100322 23:31:32< Crab_> schumi: the real fun starts once we get to MP
20100322 23:32:21< gabm> Crab_: I think so. Can you elaborate on what you mean by "actively working on a proposal" ?
20100322 23:33:14< boucman> night all
20100322 23:33:15< Crab_> schumi: for MP, our context, in which we want to save info, is the 'extended multiplayer campaign', or 'massive multiplayer campaign' - a group of games which are related by having a 'same' game world, a same story, and which can exchange information, both before game creation, and during games, as well.
20100322 23:33:17-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Quit: Leaving.]
20100322 23:33:53-!- zielony_kac [~zielony_k@chello089077167251.chello.pl] has left #wesnoth-dev []
20100322 23:34:22< schumi> Crab_: yes, it seens hard, but fun
20100322 23:35:05< Crab_> gabm: example: we asked orfest to submit boost::asio example code, fixing one of the long-time bugs in stats uploader, and, having fixed this bug, he was asked to try to add proxy support to it (a thing that will be needed for his project anyway)
20100322 23:36:04< Crab_> gabm: example: Darkas was asked to fix a bug (that should be fixed no later than tomorrow) that is directly connected with lua ai support
20100322 23:36:56< Crab_> gabm: so, for each a proposal there's work that needs to be done to 'get to the point', 'get to the edge', and actively start writing a proposal having experienced the ins-and-outs of the project
20100322 23:37:57< schumi> Crab_: so it needs to support events in the mmg on real time?
20100322 23:37:59< Crab_> schumi: we were thinking of adding a special player, called 'the gamemaster', which will be present in each and every game of a 'extended multiplayer campaign'
20100322 23:38:00< gabm> Crab_: I see. Boucman did give me a similar task on the mailing list - up to now I've preferred to write patches for small features, to have some concrete work to show
20100322 23:38:24< Crab_> gabm: yes. the more the task is connected to your proposal, the better.
20100322 23:38:29< gabm> Crab_: just in case Boucman's idea proves very long to implement
20100322 23:39:04< gabm> Crab_: but after doing the initial wiki pages I'll get started on his idea
20100322 23:39:23< Crab_> schumi: such gamemaster would be able to hold state for all the games he's currently in, and exchange information between games
20100322 23:39:37< Crab_> schumi: also, he'll be able to create scenarios on-demand
20100322 23:40:06< gabm> Crab_: now I see better what you mean. I probably won't be able to do prototyping code or small related tasks for more than one proposal.
20100322 23:40:24< Crab_> gabm: you might be able to, at start.
20100322 23:41:12< Crab_> gabm: but then you'd probably see that it'll be better for you to drop the proposal that has less potential to succeed.
20100322 23:41:21< gabm> Sure
20100322 23:41:43< Crab_> gabm: so, the fun part is figuring out where you'll be the best :)
20100322 23:42:14< schumi> Crab_: ok :) I will start my wiki and think more about this project
20100322 23:42:20< Crab_> schumi: good luck to you :)
20100322 23:43:25< schumi> Crab_: thanks!
20100322 23:44:01-!- Appleman1234 [~Appleman1@131.181.47.1] has joined #wesnoth-dev
20100322 23:44:38< schumi> Crab_: the site says about implementing some small changes. How can I do that?
20100322 23:45:39< Crab_> find a suitable small change ( look at bugs.wesnoth.org, http://wiki.wesnoth.org/EasyCoding, project-specific page), discuss it, and try to implement it
20100322 23:46:35< fendrin> Crab_: Any progress on the plugin front? A interface would be enough to start coding the wizard.
20100322 23:46:44< Crab_> schumi: btw, see http://www.wesnoth.org/forum/viewtopic.php?p=362242
20100322 23:47:31< Crab_> fendrin: interfaces for what ?
20100322 23:51:38< schumi> Crab_: ok. thanks for the help!
20100322 23:55:09< schumi> Ivanovic: hi, can you mark my account in the forum (schumi) as a GSoC Student?
20100322 23:55:27< Ivanovic> schumi: Crab_ can do it, too
20100322 23:55:29< Ivanovic> :)
20100322 23:55:46< Ivanovic> that is: make sure to name your account on your wiki page, then we will eventually do it
20100322 23:55:58< fendrin> Crab_: the schema 
20100322 23:56:05< Crab_> fendrin: if for wml schema, then those that I've showed you today should work...  ( http://wesnoth.pastebin.com/CAMF6x6B )
20100322 23:56:15< Crab_> fendrin: our you need the 'stub' implementation, as well ?
20100322 23:56:35< fendrin> Crab_: Feel free to commit it if it compiles.
20100322 23:57:04< Crab_> ok, then I'll commit a stub implementation, hopefully in ~2 hours
20100322 23:57:17< fendrin> Crab_: Cool, thank you very much.
20100322 23:57:34< fendrin> Crab_: Can you commit the interface right away?
20100322 23:58:27< schumi> Ivanovic: ok.I asked you because i didn't find Crab in the list. But the name is right
20100322 23:58:28< Crab_> you want it to be in the same project ? or a separate project ? if separate, what buildsystem do you want to use ? just eclipse ? ant ? maven ?
20100322 23:59:05< fendrin> Crab_: Same project for now. Or do you see that being a mistake?
20100322 23:59:34< Crab_> well, it's not a mistake, java code is very easy to refactor, at least until it grows big :)
--- Log closed Tue Mar 23 00:00:01 2010