--- Log opened Tue Mar 24 00:00:08 2009
--- Day changed Tue Mar 24 2009
20090324 00:00:08< Velory> Crab:i think we can divide "One Big map" to small arenas then coders can work easily
20090324 00:00:08< Velory> and also we can add comment lines like #add .. candidate moves here#
20090324 00:01:07< Crab_> Velory: your proposition will work with most .fai 's But some .fai, for example recruitment .fai, is meant to look at the entire map. so they will be screwed on such a map and don't work as intended
20090324 00:01:15< Velory> Yes so 
20090324 00:01:20< boucman> Velory, Dragonking, Crab_: do we need non-regression unit test for formulas, or should this scenario only be to allow easy building of test cases that would be droped after that...
20090324 00:01:22< Velory> i told you
20090324 00:01:29< Velory> before
20090324 00:01:40< Velory> And for specific tests users will create their scenarios and we can remove scenarios after we tested maybe ? 
20090324 00:01:41< boucman> my feeling is that we only need non-reg for real AI testing (i.e real inteeligence, not just formulas)
20090324 00:02:06< Dragonking> boucman: I'm talking about 2 scenarios
20090324 00:02:30< Dragonking> One for unit tests, one for formula nice stuff, where we can add nice stuff and people could look there to learn something from it.
20090324 00:02:31< Crab_> boucman: yes, I want to be able to test real ai on 'arena-style' map.
20090324 00:02:39< Crab_> Dragonking: that is good
20090324 00:03:01< Dragonking> I don't understand what we need this 'realoding' for since any "specialized general formula" would not be tested there.
20090324 00:03:22< Dragonking> You can't just make all possible scenarios in one test scenario, you need to develop something else and test on your own.
20090324 00:03:37< Velory> Dragonking:Think so
20090324 00:03:45< Dragonking> So trying to put my recruitment into any 'fixed' scenario is nonsense.
20090324 00:05:12< Crab_> Dragonking: ok ) but what about 'real ai testing' ? if we want to compare several versions of 'full ai', not just 1 formula, in specific pre-created situations on a specific map, to see what variant AI will do better ?
20090324 00:06:01< Velory> Crab_: you mean if we want to test all formulas in one scenario_?
20090324 00:06:15< Dragonking> Thne we need to test it on different scenarios and maps.
20090324 00:06:25< Velory> Ye
20090324 00:06:26< Velory> s
20090324 00:07:24-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has quit ["night all"]
20090324 00:08:35< Crab_> Velory: you have written a formula to improve one aspect of the ai. if you take a bunch of those formulas together and say that 'this is my ai', you will probably want to test it all together.  That's easy - take any map, as Dragonking says, and test it. But what if you want to compare 'your ai' to some other implementation (your earlier version, or some other totally different implementation) ? It would be nice to have a way to put an
20090324 00:08:52< YogiHH> Sapient: it's not possible to change the difficulty within a scenario, right?
20090324 00:08:54< Crab_> Velory: a some sort of 'arena' or a 'gauntlet' for the ai
20090324 00:09:03< Sapient> YogiHH: true
20090324 00:09:24< Crab_> Velory: where ai being tested is not hardcoded in the scenario
20090324 00:09:47< Crab_> Velory: for it's 'one scenario for all ai's, to compare their performance'
20090324 00:10:30< Sapient> YogiHH: you should know that better than me, Master of Replays ;)
20090324 00:10:46< Polarina> Crab's AI vs. Sapient's AI
20090324 00:10:47-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090324 00:10:48< Polarina> :D
20090324 00:11:06 * YogiHH has seen so many things being done with WML that he never thought to be possible ;-)
20090324 00:11:08< Dragonking> "'one scenario for all ai's, to compare their performance'" <- how one scenario can really compare any performance?
20090324 00:11:25< Dragonking> We need many test cases with many scenarios.
20090324 00:11:43< Zen_Clark> What would it take to add some of the replay features to observed multiplayer games? IMO it would be a pretty cool and bragable feature to have DVR like functions.
20090324 00:12:03< Crab_> Dragonking: agreed, "a set of scenarios for all ai's, to compare their performance".
20090324 00:12:07< Dragonking> And simply evaluating things is not a problem - we can change one line of code to simply change whole AI, thne run scenario again.
20090324 00:12:14< Dragonking> YOu just change one {...} in cfg.
20090324 00:12:24< Sapient> Zen_Clark: you might be surprised how many replay features are already in multiplayer games ;)
20090324 00:12:31< Dragonking> Assuming you have set of AI defined in exteran file.
20090324 00:12:39< Sapient> Zen_Clark: it runs right on top of the replay code
20090324 00:13:19< Crab_> Dragonking: I don't like that  '1-line-changing' ) I personally prefer to have the ability to reload ai on the fly
20090324 00:13:36< Dragonking> Crab_: And how would you evaluate?
20090324 00:13:36< Velory> 1-line changing is simple and easy i think
20090324 00:13:50< Zen_Clark> How are they accessed?
20090324 00:13:53< Dragonking> To evaluate enything you need to have a starting pos, then look how it behaves
20090324 00:14:21-!- PK [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has joined #wesnoth-dev
20090324 00:14:21< Crab_> Dragonking: 'loading Ai on the fly' will make it possible to load AI from any position.
20090324 00:14:39< YogiHH> Zen_Clark: the main problem with that is getting the information for that on the server
20090324 00:14:40< Dragonking> Great, but still you need to realod scenario file to this position
20090324 00:14:51< Crab_> Dragonking, for example, imagine yourself playing an sp campaign and then suddenly asking: what the ai will do in this situation ?
20090324 00:15:19< Dragonking> Then I input run AI form file you implmented, that's all.
20090324 00:15:33< Dragonking> Ah no, candidate moves.
20090324 00:15:35< Crab_> Dragonking: I think it would be really nice to just open the console and ask the ai to 'ai #25, play one turn for me, please"
20090324 00:16:10< Crab_> Dragonking, yes, that's I was talking about. we must load a bit more that simple .fai files - we must register those candidate moves, as well
20090324 00:16:25< Crab_> Dragonking, or do any other fancy stuff that ai implementation #25 requires.
20090324 00:16:47-!- noy [n=Noy@70.70.128.133] has joined #wesnoth-dev
20090324 00:16:51< YogiHH> Zen_Clark: You can have a look here for further explanations, although the original topic was slightly different: http://www.wesnoth.org/forum/viewtopic.php?f=15&t=22048
20090324 00:16:53< Dragonking> For such usage we need to load in whole new AI and give control to it.
20090324 00:16:59< Crab_> Dragonking: yes
20090324 00:17:11< Dragonking> That's new feature, independent to the problem of the scenario cfg files.
20090324 00:17:11-!- Homunculi [n=Homuncul@c-67-186-1-202.hsd1.pa.comcast.net] has left #wesnoth-dev []
20090324 00:17:17< Crab_> it is connected.
20090324 00:17:47< Dragonking> Not to what I was talking about. :)
20090324 00:18:28< boucman> night all
20090324 00:18:30< Crab_> if we implement such loading, we can take 'empty' scenario, and load 1) some units into it from setup.fai  2) load ai  3) run ai 
20090324 00:18:37< Crab_> boucman, good night
20090324 00:18:41-!- boucman [n=rosen@wesnoth/developer/boucman] has quit [Remote closed the connection]
20090324 00:18:43< Velory> boucman:good night
20090324 00:19:01< Crab_> Dragonking, so - voila - one map - any situation possible on that map - any ai.
20090324 00:19:09< Velory> i think we can write a specific scenario for test like u say
20090324 00:19:12< Dragonking> And when I was taling about a scenario that shows people varous neat things you can do with ai, it just complicates things for begginers with formula AI
20090324 00:19:38< Crab_> Dragonking: ah, that test scenario is entirely another matter, you are right. It is cool already )
20090324 00:20:02< Dragonking> I don't say your idea is bad - it is _very_ good.  But it was not what I talked about. That's all. :)
20090324 00:20:45< Crab_> Dragonking, ok ) sorry if I inadvertently tried to confuse you
20090324 00:20:53< Dragonking> Nah, no problem. :)
20090324 00:21:22< Dragonking> I did not understood your idea well to begin with.
20090324 00:21:35< Dragonking> But now I know what you meant. :)
20090324 00:21:39< Crab_> Dragonking: ok ) 
20090324 00:22:08< Velory> :):(:S::/
20090324 00:22:57< Crab_> Velory, Dragonking: so, what shall we do about the map to test general purpose formulas like that 'poisoning formula' ?
20090324 00:23:15< Zen_Clark> YogiHH: Would it not be possible to set it up just like a DVR to where the game is recorded locally while the server just sends the same info without ever knowing if it was being recorded or not?
20090324 00:23:30< Dragonking> For something like that you need still scenario designed a bot for it.
20090324 00:23:40< Dragonking> Like with poisoners, undeads, other unit types.
20090324 00:23:47< Dragonking> s/bot/bit
20090324 00:23:57< Crab_> Dragonking: can we keep it empty and populate in with .fai ?
20090324 00:24:03< Dragonking> No.
20090324 00:24:04< Zen_Clark> YogiHH: The server does not need to keep a local replay, just let the observer watch what he has already seen.
20090324 00:24:07< Crab_> why ?
20090324 00:24:24< Velory> .fai cant create units 
20090324 00:24:25< Dragonking> Setting up units is WML-thing.
20090324 00:24:48< Velory> .fai is just for check move and attack
20090324 00:25:00-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090324 00:25:04< Crab_> Velory: and who 'orders' units to move ?
20090324 00:25:19< YogiHH> Zen_Clark: Well, that is an option, although it will be quite a hassle to distinguish between two different "replay modes" (because multiplayer client synchronization is done via replays, too)
20090324 00:25:29< Velory> You in .fai code?
20090324 00:25:31< Crab_> Dragonking: can we populate it from some additional WML-file ?
20090324 00:25:50< YogiHH> Zen_Clark: and you will always have the problem, that you can't run replays backwards
20090324 00:25:51< Crab_> Dragonking: that will be loaded using a console command
20090324 00:25:52< Dragonking> Crab_: Hmm.... that's a question to some WML expert
20090324 00:26:02< Dragonking> zookeaper for example
20090324 00:26:16< YogiHH> Zen_Clark: so you will either need a lot of time or save gamestates regularly
20090324 00:26:17< Crab_> sadly, he isn't around
20090324 00:26:58< Crab_> Dragonking: and why can't we create units from .fai ? is it a design decision ?
20090324 00:27:21< Sapient> too bad there's no wesnoth experts around ;p
20090324 00:27:30< Sapient> too bad there's no wesnoth markup language experts around ;p
20090324 00:27:56< Sapient> 'cause if they were they would likely answer your questions
20090324 00:28:00< Dragonking> Crab_: fai is supposed to be used to create AI
20090324 00:28:04< Polarina> I wish they used XML.
20090324 00:28:06< Dragonking> AI can't put units areound map.
20090324 00:28:11-!- BenUrban [n=benurban@c-68-49-10-243.hsd1.md.comcast.net] has joined #wesnoth-dev
20090324 00:28:15 * Sapient throws sugar packet at Polarina ------------>#
20090324 00:28:21< Crab_> Dragonking: "AI can't put units around map" - agreed.
20090324 00:28:53< Polarina> Sapient: Anything wrong with that?
20090324 00:29:09< Velory> Dragonking: So what about one large divided arenas map for general .fai formula purposes ?
20090324 00:29:15< Sapient> yes, do a forum search on "XML" and never mention that acronym in here again
20090324 00:29:35< Crab_> Velory: maybe let's wait for WML experts)
20090324 00:29:39< Polarina> Sapient: Ok. Extensible Markup Language then.
20090324 00:29:43< Velory> Crab_: OK :)
20090324 00:29:58< Crab_> Velory: if we would be able to populate the map on the fly, it wouldn't be needed to make the map divided
20090324 00:30:35 * Polarina wishes they used Extensible Markup Language instead if designing their own markup language.
20090324 00:30:56< Velory> Crab_: I understand what u mean
20090324 00:30:57< Crab_> Velory: we'll just make a map in which the leaders (if they are required) are jailed somewhere and be happy with using the rest of the map as our playground :)
20090324 00:31:18< Crab_> Velory: and add a way to load the ai from a set of files
20090324 00:31:32 * Sapient draws four poisoned daggers in his fist and looks menacingly at Polarina
20090324 00:31:34< Crab_> Velory: then - load the map, populate it, test what you desire
20090324 00:31:46< Polarina> Sapient: I didn't use that acronym...
20090324 00:32:03< Sapient> you ignore my first instruction to search the forum and read
20090324 00:32:05< Crab_> Velory: when the test ends - just clean up (or restart) and repeat
20090324 00:32:20< Polarina> Sapient: The forums give me nothing of interest as a search result.
20090324 00:32:44< Velory> Crab_: i see thats looking good but do you think we actually need that?
20090324 00:33:14< Velory> Crab_: the only reason is what ai will do here when im playing campaign for that?
20090324 00:33:15< Polarina> Sapient: And there's also nothing wrong with Extensible Markup Language.
20090324 00:33:33< Sapient> Polarina: read this http://www.wesnoth.org/forum/viewtopic.php?p=294681#p294681
20090324 00:34:01< Sapient> and if you want to discuss it further, use the forum
20090324 00:34:01< Crab_> Velory: for example, it will allow you to submit only .fai as patch when you implement some general purpose function
20090324 00:34:05< Crab_> Dragonking: I've heard that you is a multiplayer expert :) maybe you can recommend a specific multiplayer map to base this 'arena' on ?
20090324 00:34:28< Dragonking> If it is going to be small?
20090324 00:34:44< Dragonking> Maybe den of onis. Since it has some intersting mix of terrain
20090324 00:34:51< Crab_> Dragonking: thanks.
20090324 00:36:11< Polarina> Sapient: That explains almost nothing.
20090324 00:36:36< Sapient> Polarina: what part of "if you want to discuss it further, use the forum" do you not understand?
20090324 00:36:43< Sapient> you are very close to a ban
20090324 00:36:57 * Polarina commences developing patches for migration to Extensible Markup Language. :)
20090324 00:37:01 * Polarina hides.
20090324 00:40:27< Velory> Crab_: Ok lets wait wml experts.im with you its simplifying some issues but i dont think this issues are too important and cant solve with current system. 
20090324 00:41:09< Velory> :)
20090324 00:41:18< Crab_> Velory: ok. I'll write a wiki page to explain what we are trying to do, and I'll try to post note a "WML experts wanted" here tomorrow :)
20090324 00:42:18< Velory> Ok :) and call 911 :p
20090324 00:42:20< Crab_> Velory: i think it will be really nice to have a single pre-made map on which you can test your ai without thinking whether your code will conflict with some other formula which is defined on the same map
20090324 00:42:58< Crab_> Velory, a playground of sorts. not for the beginners to see 'what cool things can be done' but for testing ai's and formulas.
20090324 00:44:37< Crab_> (if I manage to hack it out) WML part will be as a 'challenge' - it will create units to form some 'tactical situation', set up the map, and prepare everything for the ai. and then, any of us, as the developer, will run the ai and see how it handles that challenge./
20090324 00:45:28< YogiHH> night everyone
20090324 00:45:36< Crab_> YogiHH: night
20090324 00:45:44< Velory> i understood what you mean
20090324 00:45:46< Crab_> for example, there can be a situation in which ai has a poisoner, has some enemies of different types, and some of them are poisoned. we load this situation and end turn, and see what the AI will do.
20090324 00:45:52-!- YogiHH [i=YogiHH@wesnoth/developer/yogihh] has quit ["Bye folks"]
20090324 00:46:10< Crab_> and the ai will be sort of 'loadable on the fly' - so you would be able to pick any
20090324 00:49:01< Crab_> and the good thing will be that the ai's and the challenges will not depend on each other.
20090324 00:50:01< Crab_> that means, among other things, that they can be created by different people.
20090324 00:50:25< Velory> :)
20090324 00:51:49< Velory> Anyway we can discuss this subject when wml experts here :)
20090324 00:52:17< Velory> Yes its looking good and handy anyway i'm going to logout
20090324 00:52:19< Crab_> Velory, yes they will be able to tell about the way of doing this without hacking code that shouldn't be hacked
20090324 00:52:35< Velory> :)
20090324 00:52:37< Velory> yes
20090324 00:52:42< Crab_> Velory: good night then :)
20090324 00:52:59< Sapient> darn those WML experts for not sticking around when we need them
20090324 00:53:02< Velory> Good night to you too :) 
20090324 00:53:08< Velory> Haha sapient :)
20090324 00:53:12< Crab_> Sapient: we shall find them :)))
20090324 00:53:15 * Sapient shakes his fist at the sky angrily
20090324 00:53:25< Velory> lets be wml experts and always be here :P
20090324 00:53:38< Crab_> 'let the wml experts be with you' :)
20090324 00:53:44< Velory> haha
20090324 00:53:50< Velory> may the force be with you
20090324 00:53:54< Sapient> 'and also with you'
20090324 00:54:55-!- cjhopman_ [n=chris@wesnoth/developer/cjhopman] has joined #wesnoth-dev
20090324 00:54:57< Velory> Anyway Good nites everyone :)
20090324 00:55:05-!- Velory [n=GProcess@78.166.71.46] has left #wesnoth-dev []
20090324 00:55:58< Crab_> Sapient: '..., and beware of the Dark XML Side'
20090324 00:56:18< Sapient> indeed...
20090324 00:56:28 * Sapient shudders
20090324 00:57:48-!- JW1 [n=X@c-71-57-85-168.hsd1.il.comcast.net] has joined #wesnoth-dev
20090324 00:58:38< Sapient> cya later, have fun.
20090324 00:58:38-!- Sapient [n=sapien-x@wesnoth/developer/sapient] has left #wesnoth-dev []
20090324 01:00:02< noy> Dragonking: do you think Den is a good "test arena" map?
20090324 01:00:30< noy> I'd disagree, particularly if you're trying to use it for an AI
20090324 01:01:02< noy> because a small 1v1 map is not what an AI would be useful for... more like a large campaign map. 
20090324 01:03:01< Dragonking> In case of small map it is good since it has different terrain including caves, lava, water etc and we can look for a bugs there
20090324 01:03:21< Dragonking> About designing whole AI, it is better to use bigger maps.
20090324 01:03:36< Dragonking> When we test some particular usages, we can do it on a smaller map.
20090324 01:03:59-!- faktor4u [n=faktor4u@ilm106.internetdsl.tpnet.pl] has joined #wesnoth-dev
20090324 01:06:41-!- Taliesin_ [n=chatzill@93-96-172-175.zone4.bethere.co.uk] has joined #wesnoth-dev
20090324 01:07:22< Taliesin_> Hi
20090324 01:08:52< Crab_> Taliesin_: hi
20090324 01:09:21< Taliesin_> You got any idea why Drake campaigns are so scarce?
20090324 01:09:41-!- cyber_11 [n=muratcan@78.165.101.220] has joined #wesnoth-dev
20090324 01:10:17< Crab_> Taliesin_: no )
20090324 01:10:29< Taliesin_> Hm.
20090324 01:11:21< Taliesin_> Just seems odd. I mean, the faction is actually tolerably original... Eh. Guess I'll get ack to working on the story for mine.
20090324 01:12:36-!- faktor4u [n=faktor4u@ilm106.internetdsl.tpnet.pl] has quit ["BBL..."]
20090324 01:13:38-!- Polarina [n=polarina@freecode-project/mentor/polarina] has left #wesnoth-dev []
20090324 01:20:29-!- cyber_11 [n=muratcan@78.165.101.220] has left #wesnoth-dev []
20090324 01:24:37< Ivanovic> crimson_penguin: https://gna.org/bugs/index.php?13245
20090324 01:25:39< crimson_penguin> Ah, the event must be passed twice...
20090324 01:25:53< crimson_penguin> Maybe
20090324 01:25:58< crimson_penguin> That is pretty weird
20090324 01:28:53< Ivanovic> don't ask me for the reasons, it is probably some strange mac thingie...
20090324 01:28:55< Ivanovic> ;)
20090324 01:32:56< Taliesin_> Anybody know where I can find some maps with territory marked out on 'em for wesnoth? I'm idly working on the story for a Drakes campaign, and I'd like the geography to be at least in the ballpark.
20090324 01:36:57< Crab_> Taliesin_, there are some maps there - http://www.wesnoth.org/wiki/WesnothGeography - maybe it can be of some use
20090324 01:38:15< Taliesin_> Hmmm... Ta.
20090324 01:38:37< Taliesin_> So where would Drakes be? I've been assuming the moutnain range just below Lake Vrug, since it's the highest mountain I can see.
20090324 01:38:56-!- res|laptop [n=res@pdpc/supporter/student/res2k] has quit [Nick collision from services.]
20090324 01:38:57-!- res2k3 [n=res@erft-4d07d8b1.pool.mediaWays.net] has joined #wesnoth-dev
20090324 01:39:19-!- res2k3 is now known as res|laptop
20090324 01:39:49< CIA-70> ai0867 * r34074 /trunk/data/multiplayer/ (maps/5p_The_Wilderlands.map scenarios/5p_The_Wilderlands.cfg): Ran set-properties on data/multiplayer.
20090324 01:40:15< Crab_> Taliesin_, see http://www.wesnoth.org/forum/viewtopic.php?f=8&t=9395
20090324 01:40:58< Crab_> Taliesin_, but it will be better to ask on the relevant forum. here, http://www.wesnoth.org/forum/viewforum.php?f=8 , for example
20090324 01:41:26< Taliesin_> Southwest of the continent... Damn. Now I have to work out the campaign route all over again.
20090324 01:42:02-!- ilor_ [n=user@wesnoth/developer/ilor] has quit [Read error: 110 (Connection timed out)]
20090324 01:43:28< Taliesin_> Maybe not actually. Do we know how centralized Drakes are? Maybe there's a few clans scattered around...
20090324 01:43:30< crimson_penguin> Ivanovic: cmd+q gets handled by the Mac menu as well as the Wesnoth menu; the Mac menu sends the same event as closing the window (SDL_QUIT), which quits without asking, but I guess it handles the Wesnoth menu first, giving you the option and then once you've chosen, quitting from the Mac menu
20090324 01:45:18< CIA-70> ai0867 * r34075 /trunk/data/core/ (197 files in 22 dirs): Ran set-properties on data/core.
20090324 01:45:38< Taliesin_> Oh good. Some colonies HAVE spread- perfect. This, I can use.
20090324 01:54:24< CIA-70> ai0867 * r34076 /trunk/images/ (19 files in 2 dirs): 
20090324 01:54:24< CIA-70> Ran optipng on /images
20090324 01:54:24< CIA-70> Original size: 45 KiB on 19 files
20090324 01:54:24< CIA-70> Optimized size: 39 KiB
20090324 01:54:24< CIA-70> Total saving: 5 KiB = 13% decrease
20090324 02:00:40-!- Taliesin_ [n=chatzill@93-96-172-175.zone4.bethere.co.uk] has quit ["ChatZilla 0.9.84 [Firefox 3.0.7/2009021910]"]
20090324 02:03:29-!- alink [n=alink@wesnoth/developer/alink] has joined #wesnoth-dev
20090324 02:05:30-!- ikarius [n=ross@smtp.gridironsystems.com] has quit []
20090324 02:06:11-!- Elvish_Pillage2 [n=eli@66-189-40-122.dhcp.oxfr.ma.charter.com] has quit ["Hi! I'm a quit message virus vaccine. If you see a quit message virus, don't replace your quit message with it!"]
20090324 02:13:02-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090324 02:31:04-!- niadh [n=niadh@78-86-134-181.zone2.bethere.co.uk] has joined #wesnoth-dev
20090324 02:31:51< niadh> Hey, have signed up to GSOC and am looking around for projects, I am limiting myself to applications I like and use a lot, so am just looking at what might be required to work on some of the wesnoth projects
20090324 02:34:09-!- JW1 [n=X@c-71-57-85-168.hsd1.il.comcast.net] has quit []
20090324 02:34:13< Dragonking> hi niadh 
20090324 02:34:23< niadh> hi
20090324 02:34:24< Dragonking> It is best to start here: http://www.wesnoth.org/wiki/SummerOfCodeIdeas
20090324 02:34:48< niadh> I've been looking at them, the two that look interesting are the save games thing and the lobby thing
20090324 02:35:13< niadh> I've not done any real C++ for 2 years and have focused more and more on C# and Python the last 3 years
20090324 02:35:21< niadh> So my C++ skills may be rusty
20090324 02:35:52< Dragonking> You cna start from writing a wiki page about you proposal. Than you might want to look form some easy conding tasks and submit some patches.
20090324 02:36:36< niadh> So I write a proposal based upon one of you guys suggested ideas?
20090324 02:36:53< Crab_> niadh, yes, and read http://www.wesnoth.org/wiki/SoC_People_to_bug_on_IRC for a list of people you'd want to seek on IRC if you need more info
20090324 02:37:18< Dragonking> niadh: You are free to suggest your own idea too.
20090324 02:37:24< niadh> I of course C++ is a common prerequisite in wesnoth?
20090324 02:37:31< niadh> assume*
20090324 02:37:36< Dragonking> Yes.
20090324 02:39:32< niadh> Right ok, I must confess I'm not sure how quick my C++ skills will come back, but as I said, wesnoth is something I use a lot
20090324 02:41:10< Crab_> niadh: 'use' :) btw, some of the projects on the SoC ideas page may require python, not c++ ( for example,  campaign statistics page ). but this being a ( fairly) easy task, there's a bigger competition there.
20090324 02:41:47< niadh> part of my interest in gsoc is to expand my skills
20090324 02:44:22< niadh> what about for this server idea here: http://www.wesnoth.org/wiki/SoC_Ideas_Multiplayer_server
20090324 02:44:36< niadh> Having the ability to search by game creator?
20090324 02:45:00< niadh> I set up my own wesnoth server because the official one was too hard to find the games my friends and I created
20090324 02:45:30-!- noy [n=Noy@wesnoth/developer/noy] has quit [Connection timed out]
20090324 02:46:47-!- PK [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has quit ["Java user signed off"]
20090324 03:05:07-!- cjhopman_ [n=chris@wesnoth/developer/cjhopman] has quit [Remote closed the connection]
20090324 03:11:48-!- cjhopman [n=cjhopman@wesnoth/developer/cjhopman] has joined #wesnoth-dev
20090324 03:14:24-!- niadh [n=niadh@78-86-134-181.zone2.bethere.co.uk] has quit [Remote closed the connection]
20090324 03:14:34-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has joined #wesnoth-dev
20090324 03:18:26< Crab_> Dragonking, I've put a small wikipage about our 'AI testing' discussion -> http://www.wesnoth.org/wiki/AI_Arena
20090324 03:21:02-!- Zen_Clark [n=user@99-136-80-191.lightspeed.rcsntx.sbcglobal.net] has quit ["Zombie raptors are coming to eat my flesh and soil my honor. Must go tame them."]
20090324 03:21:36< Dragonking> Crab_: Ok
20090324 03:25:45< Crab_> Dragonking:  tomorrow, i'm going to ask someone who is a master of WML, about what is needed to implement those parts which require WML.
20090324 03:25:46< Crab_> night
20090324 03:26:04-!- Crab_ [n=crab_@c.218.175.a530.sta.adsl.cyfra.net] has quit ["Leaving."]
20090324 03:28:13-!- Netsplit verne.freenode.net <-> irc.freenode.net quits: Dragonking, happygrue
20090324 03:28:38-!- Netsplit over, joins: happygrue, Dragonking
20090324 03:33:24-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev
20090324 03:38:31-!- Deakolt [n=deakolt@207.224.223.140] has joined #wesnoth-dev
20090324 03:41:03< CIA-70> ai0867 * r34077 /trunk/data/campaigns/ (329 files in 21 dirs): 
20090324 03:41:03< CIA-70> Ran optipng on data/campaigns.
20090324 03:41:03< CIA-70> Original size: 1039 KiB on 329 files
20090324 03:41:03< CIA-70> Optimized size: 997 KiB
20090324 03:41:03< CIA-70> Total saving: 42 KiB = 4% decrease
20090324 03:51:44-!- ikarius [n=ross@216.27.182.3] has joined #wesnoth-dev
20090324 03:53:44-!- Livor [n=chatzill@adsl-67-65-161-98.dsl.hstntx.swbell.net] has joined #wesnoth-dev
20090324 03:53:53-!- Livor [n=chatzill@adsl-67-65-161-98.dsl.hstntx.swbell.net] has quit [Client Quit]
20090324 03:56:21-!- happygrue [n=George@wesnoth/developer/wintermute] has quit ["http://xkcd.com/91/"]
20090324 04:05:38-!- jrmiller [n=jrmiller@71-32-73-200.cdrr.qwest.net] has joined #wesnoth-dev
20090324 04:06:22-!- alink [n=alink@wesnoth/developer/alink] has quit [Remote closed the connection]
20090324 04:08:16-!- PK [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has joined #wesnoth-dev
20090324 04:09:56-!- PK [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has quit [Client Quit]
20090324 04:10:03-!- [Relic] [n=[Relic]@adsl-76-229-202-137.dsl.milwwi.sbcglobal.net] has joined #wesnoth-dev
20090324 04:10:57< [Relic]> Hello :)
20090324 04:12:48-!- Ivanovic_ [n=ivanovic@dtmd-4db26972.pool.einsundeins.de] has joined #wesnoth-dev
20090324 04:29:59-!- Ivanovic [n=ivanovic@wesnoth/developer/ivanovic] has quit [Read error: 110 (Connection timed out)]
20090324 04:30:46-!- Ivanovic_ is now known as Ivanovic
20090324 04:42:04< corn> dfranke: you on?
20090324 04:42:52< corn> dfranke: starting to work on creating a patch, I was wondering if it is neccesary to keep the threading code if I make the function fork off into a seperate process
20090324 04:43:46< dfranke> corn: I don't understand
20090324 04:43:56< dfranke> corn: what do you mean keep the threading code?
20090324 04:44:38< corn> dfranke: check line 77 in upload_log.cpp. my understanding is that a new thread is made currently for uploading logs to server?
20090324 04:45:26< dfranke> corn: one sec, checking on something else atm
20090324 04:45:28< corn> dfranke: line 77 and line 151, where upload_logs is actually called
20090324 04:47:44< dfranke> corn: yeah, the threading would be redundant obviously.  You probably want to fork as the very last thing you do after all other threads terminate.
20090324 04:49:06< corn> dfranke: what do you mean when all other threads terminate? how do I check for that?
20090324 04:55:44< dfranke> corn: don't know that offhand.
20090324 04:58:50< corn> dfranke: any suggestions on how I can induce a slow connection to wesnoth.org to test if my patch works?
20090324 04:59:04< dfranke> corn: iptables?
20090324 04:59:40< corn> dfranke: ok, I'll get a chain set up for it
20090324 04:59:55< dfranke> iptables -A OUTPUT -d www.wesnoth.org -j DROP
20090324 05:00:04< dfranke> then delete that rule after a few seconds.
20090324 05:14:17-!- harikerry [n=harikerr@pool-98-116-170-90.nwrknj.fios.verizon.net] has joined #wesnoth-dev
20090324 05:14:37< dfranke> corn: looking over this some more, I'm not sure if a separate process is actually the best solution.  But anyway, you know the symptom, pick the best cure.
20090324 05:15:39< dfranke> if you could just make the GUI exit cleanly but have the main process persist in the background, that would probably be better.
20090324 05:17:36< corn> yeah, that would be more in line with the rest of the code, since nothing else forks anywhere
20090324 05:18:49< dfranke> yes, and by mixing threads with processes you might run into a messy race condition if there's context switch between the fork() and the exec().
20090324 05:25:51< corn> dfranke: having trouble reproducing the conditions required for this bug. I blocked off access to wesnoth.org, but the client exits cleanly and a logfile is saved under ~/.wesnoth/upload. wouldn't blacklisting the host just cause SDLNet_Resolve Host to fail?
20090324 05:26:27< dfranke> corn: DNS resolution would still work.
20090324 05:26:30< dfranke> corn: connect would hang.
20090324 05:27:34< corn> dfranke: alright, I'll do some more testing
20090324 05:28:25-!- Espreon [n=chatzill@wesnoth/developer/espreon] has joined #wesnoth-dev
20090324 05:34:07< corn> dfranke: I made the client print if it fails to create a socket (SDLNet_TCP_Open), and it correctly fails to make a socket when wesnoth is blacklisted.
20090324 05:34:26< corn> too bad it doesn't help me reproduce the bug
20090324 05:35:34< dfranke> corn: interesting, must be some difference in kernel configuration or something.  If you're behind a router maybe block it there instead?
20090324 05:37:29-!- harikerry [n=harikerr@pool-98-116-170-90.nwrknj.fios.verizon.net] has left #wesnoth-dev []
20090324 05:38:14-!- chains [n=Rylar@adsl-69-209-53-122.dsl.chcgil.ameritech.net] has quit [Remote closed the connection]
20090324 05:38:46-!- tushar_hailstone [i=d2d40558@gateway/web/ajax/mibbit.com/x-e6a81ae393d184db] has joined #wesnoth-dev
20090324 05:38:52-!- tushar_hailstone [i=d2d40558@gateway/web/ajax/mibbit.com/x-e6a81ae393d184db] has quit [Client Quit]
20090324 05:39:15< corn> dfranke: don't think it's possible on my router. is there any way to get iptables to do rate limiting? the same issue should show up if you upload very slowly to the server
20090324 05:40:41< corn> or I'll just put a neverending loop in the function and see what happens :)
20090324 05:43:02-!- [Relic] [n=[Relic]@adsl-76-229-202-137.dsl.milwwi.sbcglobal.net] has quit ["Leaving"]
20090324 05:43:05< dfranke> corn: you can do rate limiting using tc, but have fun with that.
20090324 05:43:14< dfranke> corn: it's... a bit arcane.
20090324 05:43:42< dfranke> +50% damage to undead, easily.
20090324 05:44:06-!- Espreon [n=chatzill@wesnoth/developer/espreon] has quit ["ChatZilla 0.9.84 [Firefox 3.0.7/2009021910]"]
20090324 05:44:24< corn> dfranke: I think I managed something equivalent to the bug by putting in a while(1) {} loop into the code. you want me to make it so that the gui exits cleanly but wesnoth continues running in the background until this thread finishes?
20090324 05:45:04< dfranke> corn: yeah, that's probably close to ideal.
20090324 06:01:29-!- Deakolt [n=deakolt@207.224.223.140] has quit [Read error: 104 (Connection reset by peer)]
20090324 06:05:19-!- Deakolt [n=deakolt@207.224.223.140] has joined #wesnoth-dev
20090324 06:09:27-!- res|laptop [n=res@pdpc/supporter/student/res2k] has quit [Nick collision from services.]
20090324 06:09:28-!- res2k3 [n=res@erft-4d07d3d4.pool.mediaWays.net] has joined #wesnoth-dev
20090324 06:09:40-!- jrmiller [n=jrmiller@71-32-73-200.cdrr.qwest.net] has quit ["Leaving"]
20090324 06:09:46-!- res2k3 is now known as res|laptop
20090324 06:14:34< corn> dfranke: making the gui exit (along with getting rid of the sdl screen) and leaving the uploader running looks like it is pretty hard to do. would it be better to present the user with a window when they exit if the uploader is still running, asking them to wait?
20090324 06:17:48< dfranke> corn: no, I think that would be annoying.  If you can't background it, better to just give it no more than a couple seconds and then kill it.
20090324 06:17:57< dfranke> corn: can always do the upload at a later time.
20090324 06:21:57< noy> dfranke: your turn
20090324 06:33:14< corn> dfranke: SDLNet_TCP_Open already has a timeout. do you know if there is anyway to adjust it?
20090324 06:34:57< corn> ah, sdl mailinglist says no
20090324 06:39:55-!- turin [n=turin@168.215.249.253] has quit [Read error: 110 (Connection timed out)]
20090324 06:42:19< corn> dfranke: I can't think of any way to kill the upload thread without risk of a race condition + resource leaking when SDLNet_TCP_Open is blocking. could it be a better solution to make it so that upload_log is only called when wesnoth exits, and go back to the original idea of spawning a new process? 
20090324 06:44:35< corn> the problem is that if you have a watchdog thread that will check if the upload thread has been running for N seconds, you might kill it when it is _actually_ doing work.
20090324 06:44:37< dfranke> corn: Up to you.  I still think the close-the-GUI-persist-the-process is best though.
20090324 06:45:40< dfranke> Killing while it's doing work is probably okay though, as long as you don't delete the upload file until you're sure it's done.
20090324 06:45:50< dfranke> and the very-occasional duplicate upload is certainly no big deal.
20090324 06:47:45-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090324 06:57:00-!- silene [n=plouf@ASte-Genev-Bois-152-1-33-155.w83-114.abo.wanadoo.fr] has joined #wesnoth-dev
20090324 06:58:46-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has quit [Read error: 110 (Connection timed out)]
20090324 07:07:23-!- cjhopman [n=cjhopman@wesnoth/developer/cjhopman] has quit []
20090324 07:10:16-!- Deakolt [n=deakolt@207.224.223.140] has quit [Read error: 110 (Connection timed out)]
20090324 07:12:28-!- tushar_hailstone [i=d2d40558@gateway/web/ajax/mibbit.com/x-d8b14848473e4d8d] has joined #wesnoth-dev
20090324 07:12:57-!- tushar_hailstone [i=d2d40558@gateway/web/ajax/mibbit.com/x-d8b14848473e4d8d] has left #wesnoth-dev []
20090324 07:17:04-!- cjhopman [n=chris@wesnoth/developer/cjhopman] has joined #wesnoth-dev
20090324 07:18:50-!- ryochan7 [n=ryochan7@173.9.238.161] has quit ["Leaving"]
20090324 07:22:39< corn> ok, should have this done by tomorrow
20090324 07:22:51< corn> this is taking pretty long, but now I have a pretty good idea of how the code works :)
20090324 07:23:02-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090324 07:23:06< dfranke> awesome.  Looking forward to seeing it.
20090324 07:28:30-!- iyonius [n=iyo@a02-0829c.kn.vutbr.cz] has joined #wesnoth-dev
20090324 07:29:17< Sirp> well, slashdot sent about 7000 visits our way, 90% of whom were new visitors to our site.
20090324 07:29:27-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090324 07:29:54< Sirp> what will be truly interesting is if there is a permanent bounce in our visitors going forward. If 1.6 was a good version, there should be.
20090324 07:32:36< silene> what do you mean by "permanent bounce"? people coming back and forth?
20090324 07:33:30-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit ["Leaving"]
20090324 07:34:18< Sirp> silene: no I mean a permanent increase, 'bounce' was probably not a good term.
20090324 07:34:32< Sirp> after previous releases we've seen our average monthly visitors increase permanently.
20090324 07:35:01< silene> i see; what was our average before the release?
20090324 07:35:11< Sirp> around 6000 visits/day
20090324 07:36:14< Sirp> and today we've had over 20,000 visits (final results not in yet). Yesterday we had 11,000
20090324 07:43:10< cjhopman> wow, i get massively higher fps with my mouse offscreen
20090324 07:48:08< cjhopman> or so it says... not sure yet if i believe it
20090324 07:58:50< silene> Sirp: impressive
20090324 07:59:02< silene> cjhopman: x cursor of wesnoth's colored cursor?
20090324 07:59:15< silene> or*
20090324 08:03:16< cjhopman> happens with both
20090324 08:04:05< cjhopman> on test scenario, normally i get 40-70 fps... if i move mouse all the way to the left, when it is done scrolling, fps goes to 500-700
20090324 08:04:31< cjhopman> same thing happens in other directions
20090324 08:05:38< cjhopman> down and to the right goes to 1000... (there is only one unit with animation there)
20090324 08:05:45< silene> oh, "higher" meant "10x higher", indeed, that's hard to believe :-)
20090324 08:06:55< cjhopman> i actually think its reasonable... looking at what needs to be done each frame, 500-1000 fps is plausible
20090324 08:11:06< silene> sure, but supposedly, you are not supposed to go much faster than the refresh rate hardcoded in wesnoth, which is 50fps if i understood alink correctly
20090324 08:15:15< silene> hum... now that i look at the code, it seems a bit dubious
20090324 08:18:29< silene> alink, bouncman: in the display::flip function, the volatile things are drawed, then video::flip is called, then the volatile things are undraw; if i'm not mistaken, it means that when using double buffering, the volatile things are not undrawn from the correct buffer
20090324 08:22:19< silene> forget about it, i just checked the code, and wesnoth does not ask for double buffering; which completely defeats the point of calling sdl_flip; tonight i will just remove this call and see if it solves my issue with the memory bandwidth
20090324 08:24:11< cjhopman> i would think it would be better to change it to ask for double buffering
20090324 08:25:48< silene> cjhopman: double buffering is not used if you don't ask for hardware surface along the way, but all wesnoth surfaces are stored inside software surfaces since we are playing lots of games with colors and the like, so it wouldn't work
20090324 08:27:00< silene> (obviously, if the graphical engine was opengl-based and the coloring was sent to the gl engine, then hardware surfaces would make sense, but only in this case)
20090324 08:28:21< silene> as a matter of fact, i just checked and wesnoth explicitly disables hardware surfaces for windows, macosx, and windowed linux
20090324 08:29:20< silene> so the only user of double buffering would be fullscreen linux, which may not worth fighting for
20090324 08:33:52< dfranke> hmm, there's a game in the lobby that has no players joined to it.
20090324 08:36:07< silene> omg, those are ghosts!
20090324 08:36:13< silene> (hum... sorry)
20090324 08:37:05< dfranke> Soliton: ping?
20090324 08:41:30-!- Kamai [n=Kamai@JEFFREY.grinnell.edu] has joined #wesnoth-dev
20090324 08:47:39-!- silene [n=plouf@wesnoth/developer/silene] has quit ["Leaving."]
20090324 08:48:36-!- Sirp [n=me@wesnoth/developer/dave] has quit ["leaving"]
20090324 08:51:08-!- Kamai [n=Kamai@JEFFREY.grinnell.edu] has quit ["Going offline, see ya! (www.adiirc.com)"]
20090324 08:51:34-!- ettin [n=jorda@wesnoth/developer/ettin] has joined #wesnoth-dev
20090324 08:54:04-!- ettin_ [n=jorda@wesnoth/developer/ettin] has quit [Read error: 145 (Connection timed out)]
20090324 08:59:39< dfranke> anyone around who has shell access to the server?
20090324 09:01:55< cjhopman> oh my... no wonder i generally get awful fps... there is a 20 ms delay every frame that isn't scrolling :/
20090324 09:15:17-!- noy [n=Noy@wesnoth/developer/noy] has quit []
20090324 09:38:56-!- wildpenguin [n=chpln@ppp121-45-8-177.lns10.adl2.internode.on.net] has joined #wesnoth-dev
20090324 09:39:14< dfranke> wildpenguin: yo
20090324 09:39:21< wildpenguin> Hello.
20090324 09:39:48< wildpenguin> dfranke: Thinking back, it was Nat who first reported that strange chat behaviour.  Seems likely it is all stemming from that OOS.
20090324 09:40:25< dfranke> wildpenguin: it does seem likely.
20090324 09:43:16< dfranke> in my initial incarnation in that game, it was giving me an illegal command error every time I tried to move or recruit.
20090324 09:43:35< dfranke> which I think is because it still thought I was joined to the other game.
20090324 09:48:16< wildpenguin> dfranke: When reconnecting with another instance of the client, was there a different issue?
20090324 09:49:09< dfranke> wildpenguin: in xivarmy's game I kept getting the same problem even after reconnecting with a new nick.
20090324 09:49:16< dfranke> wildpenguin: I'm writing up a bug report about the whole thing right now.\
20090324 09:50:06< wildpenguin> dfranke: The issues I were having seemed to be resolved after reconnecting to the server with the same instance.  Quite odd.
20090324 10:04:46< dfranke> https://gna.org/bugs/index.php?13252
20090324 10:17:07< Ivanovic> moin
20090324 10:31:03-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has quit [Read error: 110 (Connection timed out)]
20090324 10:32:48-!- zookeeper [n=l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20090324 10:45:52-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has joined #wesnoth-dev
20090324 10:46:14 * dfranke curses AT&T
20090324 11:20:46< dfranke> final attempt before I go to bed: anyone awake who has shell access to our server?
20090324 11:21:31 * Rhonda peeks at dfranke
20090324 11:22:03-!- fendrin [n=fabi@wesnoth/developer/fendrin] has quit [Remote closed the connection]
20090324 11:22:03< Rhonda> dfranke: It would be much more easier if you just would say what you need, otherwise people come around when you sleep and ... don't know what to do.
20090324 11:22:28< dfranke> Rhonda: https://gna.org/bugs/index.php?13252
20090324 11:26:24< Rhonda> Saved the core file
20090324 11:27:00< dfranke> ok, thank you.
20090324 11:27:13< Rhonda> Also saved the binary in case of Soliton doing a rebuild or such.
20090324 11:27:29< dfranke> does it have debug symbols?
20090324 11:27:46< Rhonda> dynamically linked (uses shared libs), for GNU/Linux 2.6.8, not stripped
20090324 11:27:59< Rhonda> Not sure wether it was built with -g
20090324 11:28:18< Rhonda> The core file is 763mb in size, btw.
20090324 11:28:41< dfranke> that seems surprisingly large
20090324 11:29:06< Rhonda> The 1.4 daemon has almost the same size.
20090324 11:29:17< dfranke> ok, guess it's normal then.
20090324 11:29:28< Rhonda> campaignds are half the size
20090324 11:31:44< dfranke> how big is the binary?
20090324 11:32:25< dfranke> my release binary is about 1MB, debug binary is about 16MB.
20090324 11:37:30< dfranke> anyway, off to bed.  Thanks again.
20090324 11:55:43< Ivanovic> Soliton: do we have any way to update the server list at http://wesnoth.yi.org/statistics/ ?
20090324 11:55:58< Ivanovic> would be nice to have stats for 1.6 and 1.4 now, 1.5 can be removed...
20090324 11:56:00< Ivanovic> ;)
20090324 11:56:57-!- loonycyborg [n=sergey@79.139.246.190] has joined #wesnoth-dev
20090324 12:05:14-!- BenUrban [n=benurban@unaffiliated/benurban] has quit []
20090324 12:05:40< CIA-70> ivanovic * r34078 /trunk/ (11 files in 8 dirs): updated Chinese (Simplified) and Slovak translation
20090324 12:05:49< CIA-70> ivanovic * r34079 /branches/1.6/ (11 files in 8 dirs): updated Chinese (Simplified) and Slovak translation
20090324 12:15:49-!- Crab_ [n=crab_@c.218.175.a530.sta.adsl.cyfra.net] has joined #wesnoth-dev
20090324 12:15:56< Crab_> hi
20090324 12:16:29< Ivanovic> hallo Crab_
20090324 12:17:14< CIA-70> loonycyborg * r34080 /trunk/projectfiles/ (4 files in 4 dirs): 
20090324 12:17:14< CIA-70> Committed McShark's projectfiles update
20090324 12:17:14< CIA-70> Added Support for Lua, added folder structure to Dev-Cpp project file.
20090324 12:31:26< ilor> hm, in the 1.6 windows installer most of the stuff is translated, but the "choose where userdata will be stored" bit isn't.
20090324 12:34:35< loonycyborg> That's because it's a custom page.
20090324 12:36:20-!- BenUrban [n=benurban@c-68-49-10-243.hsd1.md.comcast.net] has joined #wesnoth-dev
20090324 12:37:03-!- ilor_ [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090324 12:45:02-!- BenUrban_ [n=benurban@c-68-49-10-243.hsd1.md.comcast.net] has joined #wesnoth-dev
20090324 12:51:58< Ivanovic> loonycyborg: regarind your comment in the gentoo bugtracker:
20090324 12:52:02< Ivanovic> which checks are exactly missing?
20090324 12:52:19< Ivanovic> and which symbols have to be exported for those to work with the code?
20090324 12:52:24-!- ilor [n=user@wesnoth/developer/ilor] has quit [Read error: 110 (Connection timed out)]
20090324 13:02:16-!- BenUrban [n=benurban@unaffiliated/benurban] has quit [Read error: 110 (Connection timed out)]
20090324 13:03:01< loonycyborg> Ivanovic: autoconf works like this: it checks for a symbol and writes #define HAVE_SYMBOL in config.h if found, #undef HAVE_SYMBOL otherwise.
20090324 13:03:37< loonycyborg> Autoconf part is currently fully unimplemented in cmake.
20090324 13:04:16< Ivanovic> loonycyborg: my question is: which files are to be checked for which symbols?
20090324 13:04:55< loonycyborg> Look in SConstruct or configure.ac
20090324 13:05:43 * loonycyborg thinks that implementing it in cmake is empty waste of time since we have scons anyway and it isn't going anywhere :)
20090324 13:06:52< Ivanovic> and even more interesting: which files should be checked for "round"?
20090324 13:06:56< Ivanovic> anything specific?
20090324 13:07:35-!- ilor_ [n=user@wesnoth/developer/ilor] has quit [Read error: 113 (No route to host)]
20090324 13:07:54< loonycyborg> In the standard library.
20090324 13:08:26< loonycyborg> silene added that check since some systems lack round()
20090324 13:12:34-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20090324 13:13:15-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090324 13:19:55< CIA-70> shadowmaster * r34081 /trunk/src/ (7 files): (log message trimmed)
20090324 13:19:55< CIA-70> Partial refactoring of music and sound effects component.
20090324 13:19:55< CIA-70> The music_track struct has been rewritten (and "promoted" to class) and
20090324 13:19:55< CIA-70> music caching behavior in Wesnoth has changed a bit. This should not
20090324 13:19:55< CIA-70> noticeably affect performance or be visible to users in other ways (if
20090324 13:19:58< CIA-70> it is, we've got regressions).
20090324 13:20:00< CIA-70> There's a hidden, disarmed core bomb in this commit. Can you find it?
20090324 13:20:11< CIA-70> shadowmaster * r34082 /trunk/src/intro.cpp: 
20090324 13:20:11< CIA-70> Writing through the lg::wml log domain isn't equivalent to putting stuff
20090324 13:20:11< CIA-70> at the end of the lg::wml_error stringstream... why label a commit like
20090324 13:20:12< CIA-70> part of a different refactoring task if it also has these unrelated
20090324 13:20:16< CIA-70> side-effects?
20090324 13:20:35< CIA-70> shadowmaster * r34083 /trunk/ (changelog src/intro.cpp): 
20090324 13:20:35< CIA-70> [switch] and its children, [else] and [case], may be used in story WML
20090324 13:20:35< CIA-70> ([story]) from now on.
20090324 13:20:35< CIA-70> Fulfills FR / bug #13170.
20090324 13:20:35< CIA-70> Silene or Sapient may want to check the usage of
20090324 13:20:36< CIA-70> vconfig::all_children_iterator in this commit, however.
20090324 13:20:51< CIA-70> shadowmaster * r34084 /trunk/src/image_function.hpp: Clarify C++ documentation of the ~RC() image functor's primitive.
20090324 13:21:17< CIA-70> shadowmaster * r34085 /trunk/src/image.cpp: 
20090324 13:21:17< CIA-70> Be pedantic about image function usage and complain when we find an
20090324 13:21:17< CIA-70> unknown one in a path.
20090324 13:21:17< CIA-70> We are considering Mordante's ~RIGHT() as a special no-op function for
20090324 13:21:18< CIA-70> now. Meh...
20090324 13:22:10< Shadow_Master> wesbot: seen esr
20090324 13:22:10< wesbot> Shadow_Master: The person with the nick esr last spoke 23h 33m ago.  15h 13m ago was here and on the channel #wesnoth-umc-dev with the message: Remote closed the connection
20090324 13:22:15< Shadow_Master> ARGH.
20090324 13:34:34-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090324 13:35:20< Shadow_Master> hi ilor , got some time?
20090324 13:36:04< ilor> Shadow_Master: somewhat, I'm on uni wifi which might drop anytime
20090324 13:36:16< Shadow_Master> ilor: can you check if I didn't screw up your code in r34013 ?
20090324 13:36:42< Shadow_Master> I never got word from you about it on IRC.
20090324 13:37:54< ilor> I must have missed it
20090324 13:38:04< ilor> anyway it looks like the logic's unchanged
20090324 13:38:32< ilor> and I'd like to use this opportunity to point out to gcc that remembering that && goes before || is not a superhuman trait
20090324 13:39:47< Shadow_Master> um, I have never really learned the associativity order with logical operators, thus I always enclose my stuff in extra parenthesis
20090324 13:40:19< Shadow_Master> the gcc warning also sounds like some broken compilers may have slightly different ordering though.
20090324 13:41:22< ilor> I find that highly unlikely. It's been that way all the way since plain C
20090324 13:56:23< Shadow_Master> ARGHÑ.
20090324 13:56:45< Shadow_Master> I PMed Mabuse about using the Edit button and now he started sending me 2 PMs per idea
20090324 13:57:14< Shadow_Master> His future isn't a bright one.
20090324 13:58:02< voris> heh
20090324 13:58:20< Shadow_Master> Mordante: okay, I managed to get something understandable from that kid: http://www.wesnoth.org/forum/viewtopic.php?p=347058#p347058
20090324 13:58:26< Shadow_Master> you should check that assertion failure.
20090324 13:58:35< CIA-70> suokko * r34086 /trunk/ (4 files in 3 dirs): Fixed compilation with -D_GLIBCXX_PARALLEL
20090324 13:58:41< Shadow_Master> lolwut?
20090324 13:59:20-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has quit ["off to uni, cu"]
20090324 14:02:52-!- tushar_hailstone [i=d2d40558@gateway/web/ajax/mibbit.com/x-1b03626a9f4d3ed1] has joined #wesnoth-dev
20090324 14:02:57-!- tushar_hailstone [i=d2d40558@gateway/web/ajax/mibbit.com/x-1b03626a9f4d3ed1] has left #wesnoth-dev []
20090324 14:10:41-!- BenUrban_ [n=benurban@unaffiliated/benurban] has quit ["Power failu"]
20090324 14:16:12-!- Sapient [n=patrickp@wesnoth/developer/sapient] has joined #wesnoth-dev
20090324 14:17:35-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090324 14:20:08< Shadow_Master> one of the C++ experts may want to review suokko's latest patch..
20090324 14:22:42-!- Gauteamus [n=chatzill@c43C4BF51.dhcp.bluecom.no] has joined #wesnoth-dev
20090324 14:26:31< wildpenguin> night.
20090324 14:26:36-!- wildpenguin [n=chpln@ppp121-45-8-177.lns10.adl2.internode.on.net] has quit ["Leaving"]
20090324 14:28:42-!- jmunro [i=james@unaffiliated/jmunro] has joined #wesnoth-dev
20090324 14:29:12< jmunro> hey guys - i'm a gsoc 09 hopeful :)
20090324 14:29:30< iyonius> jmunro: hoj
20090324 14:29:31< Shadow_Master> hi
20090324 14:29:43< iyonius> jmunro: which project?
20090324 14:30:08< Shadow_Master> jmunro: i you haven't done so already, I suggest you check http://www.wesnoth.org/wiki/SummerOfCodeIdeas :)
20090324 14:30:12< Shadow_Master> *if
20090324 14:30:24< Shadow_Master> the first section is essential
20090324 14:30:43< jmunro> i'm currently looking at optimising the WML implementation, but i'm thinking about other areas too
20090324 14:31:00-!- iyonius [n=iyo@a02-0829c.kn.vutbr.cz] has quit [Remote closed the connection]
20090324 14:31:02< jmunro> thanks Shadow_Master 
20090324 14:31:11< CIA-70> suokko * r34087 /branches/1.6/ (4 files in 3 dirs): Fixed compilation with -D_GLIBCXX_PARALLEL
20090324 14:31:47< ilor> a commit by suokko that says he fixed something. I feel a chill.
20090324 14:32:05-!- grzybacz [i=grzywacz@kolos.math.uni.lodz.pl] has joined #wesnoth-dev
20090324 14:32:07-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20090324 14:33:04< Crab_> ilor: looks like he only changed some c-strings to be std::strings, for some exotic compile option to work. no 'game logic' changes, that is.
20090324 14:33:37< Shadow_Master> ilor: yes :-(
20090324 14:33:52< Shadow_Master> ilor: who gave him the link to the 1.6 branch?
20090324 14:33:55< ilor> Crab_: yeah seems harmless enough
20090324 14:34:17-!- alink [n=alink@wesnoth/developer/alink] has joined #wesnoth-dev
20090324 14:34:19< Shadow_Master> jmunro: have you registered a forum account already?
20090324 14:34:22< alink> hi
20090324 14:34:33< Crab_> hi, alink
20090324 14:34:53< alink> hello Crab_
20090324 14:35:06< jmunro> Shadow_Master, just doing that in a few minutes :)
20090324 14:35:06< ilor> lecture over, need to go ;)
20090324 14:35:14-!- ilor [n=user@wesnoth/developer/ilor] has quit []
20090324 14:36:58< alink> cjhopman: "wow, i get massively higher fps with my mouse offscreen", you are probably using --max-fps 1000 and mouse edge scrolling 
20090324 14:38:21< Shadow_Master> --max-fps ? didn't know about it, cool
20090324 14:38:39< alink> with that option, it's a bug with mouse-scrolling, causes high cpu usage
20090324 14:39:33< alink> and when it haapens, I don't think that the high value of fps is true
20090324 14:40:05< zookeeper> btw, i could reproduce this: http://www.wesnoth.org/forum/viewtopic.php?f=4&t=24613
20090324 14:40:19< zookeeper> quite an annoying bug too if you have fog on
20090324 14:40:41< alink> wesbot: bug 13251
20090324 14:40:42< wesbot> Bug #13251  Assigned to: Ali El Gariani <alink>  Status: Confirmed  Priority: 5 - Normal
20090324 14:40:45< wesbot> Summary: Units move on their own when using "next unit"
20090324 14:40:48< wesbot> Original submission: When in game, right click on an empty spot and choose "ne
20090324 14:40:51< wesbot> xt unit", that unit will sometimes move on its own. And if fog is on, sometimes 
20090324 14:40:54< wesbot> URL: https://gna.org/bugs/?13251
20090324 14:40:55< zookeeper> oh, right.
20090324 14:40:56< alink> zookeeper: I agree
20090324 14:42:16< alink> when I cam back from my wesbreak, I found a lot of these false-click bugs. I thought that I fixed all of them, but since i never used the context-menu next unit, I missed that one
20090324 14:45:00< jmunro> Shadow_Master, i've registered a forum account now
20090324 14:45:39< alink> ah found a workaround: click the menu-item with right-click
20090324 14:45:47< Shadow_Master> jmunro: username?
20090324 14:45:54< voris> I'm so prone to sending valid move commands that I didn't mean to I always have delay shroud update on anyway. 
20090324 14:46:04-!- stikonas [n=stikonas@wesnoth/translator/stikonas] has joined #wesnoth-dev
20090324 14:46:07< alink> zookeeper: ^
20090324 14:46:27< jmunro> Shadow_Master, jdmunro
20090324 14:46:35-!- res|laptop [n=res@pdpc/supporter/student/res2k] has quit [Read error: 110 (Connection timed out)]
20090324 14:47:07< Shadow_Master> jmunro: added to the gsoc group
20090324 14:47:26< jmunro> Shadow_Master, thank you!
20090324 14:50:41 * Soliton returns.
20090324 14:51:34< alink> Soliton: welcome back :)
20090324 14:51:40< Soliton> Ivanovic: as usual no access to the wesnothd statistics page unless uso returns or someone sets it up on wesnoth.org.
20090324 14:51:40 * Shadow_Master greets Soliton 
20090324 14:51:51< Soliton> alink: thanks.
20090324 14:52:49< alink> silene: we use software double-buffering (and we must for correct animation). Removing SDL_Flip will completly break the screen drawing
20090324 14:53:05-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has quit ["?"]
20090324 14:53:20< alink> silene: AFAI
20090324 14:53:42< alink> hum AFAIK hardware surface never work on X 
20090324 14:53:45-!- nerwa [n=nerwa@59.78.26.75] has joined #wesnoth-dev
20090324 14:54:10< grzybacz> wesnoth doesn't work with double buffering to begin with
20090324 14:54:14< alink> but execpt maybe some special driver
20090324 14:54:17< grzybacz> and no, no hardware surfaces on X
20090324 14:54:47< alink> grzybacz: you mean hardware double-buffering ?
20090324 14:54:55< grzybacz> alink: I mean any kind of double buffering
20090324 14:55:03< grzybacz> alink: we do partial updates, not redraws
20090324 14:55:16< grzybacz> alink: partial updates with double buffering cause artifacts
20090324 14:55:28< alink> mmh maybe my vocabulary is wrong then
20090324 14:55:41< voris> I can make artifacts show up. :)
20090324 14:56:16< grzybacz> alink: you mean having 2 buffers to draw on?
20090324 14:56:17< alink> we draw all items of the next frame on a second surface, then later switch it that surface with the video surface
20090324 14:56:27< grzybacz> Yes.
20090324 14:56:41< alink> well I assumed the having 2 buffers could be called double-buffering :)
20090324 14:56:42< grzybacz> But currently we don't redraw the whole screen every frame
20090324 14:56:58< grzybacz> We draw only parts that have changed
20090324 14:57:08< grzybacz> So you'd end up with some updates in one buffer, some in the other
20090324 14:57:48< alink> Ah yes this is indeed how we use SDL_UpdateRect
20090324 14:58:04< grzybacz> For double buffering you'd have to redraw the whole screen each frame
20090324 14:58:12< grzybacz> and that is too costly without hardware blitters
20090324 14:58:50< alink> and only when we call SDL_UpdateRect for the whole screen (which do happen) it's equivalent to a sort of double-buffering
20090324 14:58:56< alink> so, my bad then
20090324 15:00:58 * alink was confused because wesnoth call a video::flip() function which call SDL_Flip or just SDL_UpdateRect, depending of the case
20090324 15:01:19< grzybacz> Flip is just a convenience for calling updateRect on the whole screen
20090324 15:01:36< alink> yes for software mode, indeed
20090324 15:03:21-!- nerwa [n=nerwa@59.78.26.75] has quit []
20090324 15:05:27 * alink still thinks that such SDL mechanism could be called "double-buffering with partial update" but the doc disagree :-/
20090324 15:05:34< CIA-70> loonycyborg * r34088 /trunk/scons/lua.py: Clarified message in lua check.
20090324 15:06:51-!- Gauteamus [n=chatzill@c43C4BF51.dhcp.bluecom.no] has quit ["ChatZilla 0.9.84 [Firefox 3.0.7/2009021910]"]
20090324 15:19:10< CIA-70> ivanovic * r34089 /branches/1.6/ (CMakeLists.txt cmake/SearchForStuff.cmake config.h.cmake): 
20090324 15:19:10< CIA-70> add the searching for sendfile/round to the cmake recipe
20090324 15:19:10< CIA-70> now a file for the defines is created automatically...
20090324 15:19:40-!- stikonas [n=stikonas@wesnoth/translator/stikonas] has quit [Read error: 54 (Connection reset by peer)]
20090324 15:26:02-!- stikonas [n=stikonas@wesnoth/translator/stikonas] has joined #wesnoth-dev
20090324 15:28:26-!- [voris] [n=voris@c-24-16-107-149.hsd1.wa.comcast.net] has joined #wesnoth-dev
20090324 15:29:20< CIA-70> ivanovic * r34090 /trunk/ (CMakeLists.txt cmake/SearchForStuff.cmake config.h.cmake): 
20090324 15:29:20< CIA-70> port over r34089 from branches/1.6, commit message was:
20090324 15:29:20< CIA-70> add the searching for sendfile/round to the cmake recipe
20090324 15:29:20< CIA-70> now a file for the defines is created automatically...
20090324 15:31:54-!- voris [n=voris@c-24-16-107-149.hsd1.wa.comcast.net] has quit [Read error: 60 (Operation timed out)]
20090324 15:32:56-!- [voris] [n=voris@c-24-16-107-149.hsd1.wa.comcast.net] has quit [Client Quit]
20090324 15:32:59-!- [voris] [n=voris@c-24-16-107-149.hsd1.wa.comcast.net] has joined #wesnoth-dev
20090324 15:33:12-!- [voris] is now known as voris
20090324 15:34:06-!- voris [n=voris@c-24-16-107-149.hsd1.wa.comcast.net] has quit [Client Quit]
20090324 15:34:46 * alink just used the Moderator Control Panel for the first time (to rename a post).
20090324 15:34:47-!- stikonas [n=stikonas@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090324 15:36:06-!- Cathrak [i=sirokrul@faui08.informatik.uni-erlangen.de] has joined #wesnoth-dev
20090324 15:39:22-!- stikonas [n=stikonas@ctv-79-132-187-171.vinita.lt] has joined #wesnoth-dev
20090324 15:48:41-!- res|laptop [n=res@77.7.243.122] has joined #wesnoth-dev
20090324 15:51:24< CIA-70> zookeeper * r34091 /trunk/data/core/images/terrain/castle/ (4 files in 2 dirs): Made the little keep-to-castle doors show up on all keep tiles north of a castle tile.
20090324 15:54:07< CIA-70> zookeeper * r34092 /trunk/data/core/images/terrain/castle/elven/keep-castle-concave-tr.png: Fixed a very slightly glitchy shadow.
20090324 15:55:24< CIA-70> zookeeper * r34093 /trunk/data/core/images/terrain/castle/ (sunken-ruinkeep1-concave-tl.png sunkenkeep-concave-tl.png): Fixed two off-by-1px tiles.
20090324 16:01:27-!- voris [n=voris@c-24-16-107-149.hsd1.wa.comcast.net] has joined #wesnoth-dev
20090324 16:03:15-!- mysticspiral [n=toni-mai@79.194.40.38] has joined #wesnoth-dev
20090324 16:11:21-!- Ivanovic_ [n=ivanovic@dtmd-4db26ea5.pool.einsundeins.de] has joined #wesnoth-dev
20090324 16:12:24-!- ikarius [n=ross@216.27.182.3] has quit []
20090324 16:13:39-!- esr [n=chatzill@static-71-162-243-5.phlapa.fios.verizon.net] has joined #wesnoth-dev
20090324 16:19:26-!- Ivanovic [n=ivanovic@wesnoth/developer/ivanovic] has quit [Nick collision from services.]
20090324 16:19:29-!- Ivanovic_ is now known as Ivanovic
20090324 16:26:39-!- crimson_penguin [n=ben@S0106000d88bbc764.vc.shawcable.net] has joined #wesnoth-dev
20090324 16:32:14-!- Velory [n=GProcess@88.224.196.170] has joined #wesnoth-dev
20090324 16:32:20< Velory> Hi everyone
20090324 16:35:02-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has joined #wesnoth-dev
20090324 16:35:48-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090324 16:38:48-!- euschn [n=chatzill@85-127-103-245.dynamic.xdsl-line.inode.at] has joined #wesnoth-dev
20090324 16:39:34-!- Shadow_Master is now known as shadowmaster
20090324 16:39:50-!- Shadow_Master [n=shadowm@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090324 16:40:27-!- shadowmaster [n=ignacio@wesnoth/developer/shadowmaster] has quit [Client Quit]
20090324 16:45:57< Shadow_Master> wesbot: seen silene
20090324 16:45:58< wesbot> Shadow_Master: The person with the nick silene last spoke 8h 9m ago.  7h 58m ago they left with the message: "Leaving."
20090324 16:46:10< Shadow_Master> silene: when you get here, please check r34082
20090324 16:46:30< cjhopman> alink, no my fps really was way high
20090324 16:46:35< Shadow_Master> Sapient: can you check my usage of all_children_iterator in r34083 ?
20090324 16:46:42< cjhopman> there is a line in the code that basically says
20090324 16:46:56< cjhopman> if (!scrolling) SDL_Delay(20)
20090324 16:46:59< cjhopman> for every frame
20090324 16:47:14< Sapient> Shadow_Master: unfortunately I am quite busy + afk at the moment
20090324 16:47:21< Sapient> I will review it later definitely
20090324 16:47:50 * Shadow_Master busy too, that's why he is using kvirc to be notified of hilights
20090324 16:48:27-!- ikarius [n=ross@smtp.gridironsystems.com] has joined #wesnoth-dev
20090324 16:49:38-!- euschn is now known as euschn_
20090324 16:51:14< cjhopman> but now though, without that delay... i get 700-1000 fps except when scrolling, then it drops to ~15
20090324 16:51:29< cjhopman> happens when scrolling with mouse or keyboard
20090324 16:51:35< alink> cjhopman: mmh intersting, this line could probably be improved
20090324 16:53:30< zookeeper> shouldn't the game redraw only when something has changed anyway?
20090324 16:53:40< Ivanovic> mordante: once you read the log, please have a look at this forum thread, another assertion in the gui code... http://www.wesnoth.org/forum/viewtopic.php?f=4&t=24596&start=0&st=0&sk=t&sd=a
20090324 16:53:43-!- chains [n=Rylar@76.223.1.90] has joined #wesnoth-dev
20090324 16:54:01< jmunro> is it the mentors that review the gsoc application at the google site too? and is it okay to re use information from my wiki page?
20090324 16:54:03< alink> zookeeper: yes it does that
20090324 16:54:13< zookeeper> alink, ok, nevermind then
20090324 16:54:24< Ivanovic> jmunro: that is exactly what could/should be done
20090324 16:54:35< jmunro> Ivanovic, thanks for that
20090324 16:55:12< Ivanovic> jmunro: that is especially if you are not able to get things done in the charcount allowed, it is better to just have a "my irc nick is ABC and my proposal is in the wesnoth wiki" text
20090324 16:55:13-!- elias [n=elias@allegro/developer/allefant] has joined #wesnoth-dev
20090324 16:55:31< Ivanovic> (of course with a *link* to your proposal)
20090324 16:55:32< Ivanovic> ;)
20090324 16:55:33< cjhopman> but i mean with that delay(20) every frame... at 25 fps you are already spending half your time in SDL_Delay
20090324 16:55:40< alink> zookeeper: but calling "redraw" in code has not a clear meaning, the draw function will ask some things if thet need redraw, and others will be always redrawn.
20090324 16:55:40< jmunro> hehe of course
20090324 16:56:01< jmunro> Ivanovic, which subforum would you suggest is the best for introducing myself to the community, Coder's Corner?
20090324 16:56:21< alink> cjhopman: that is why I meant by "this line could probably be improved"
20090324 16:56:39< Ivanovic> jmunro: introduction is best in here
20090324 16:56:46< jmunro> okay
20090324 16:56:54< Ivanovic> the forums are mainly for "content creation" and players stuff
20090324 16:57:07< Ivanovic> (content as in addons, artwork, music, ...)
20090324 16:57:22< jmunro> noted, thanks
20090324 16:57:26< cjhopman> i'm more interested in why my fps drops so much when scrolling
20090324 16:57:33< Ivanovic> and in general it will be a good idea to talk to us about your idea in here
20090324 16:57:40< jmunro> for now i must hop along to a lecture
20090324 16:57:40< Ivanovic> this serves several purposes:
20090324 16:57:45< Ivanovic> 1) we get to know you
20090324 16:57:51< Ivanovic> 2) we can help you with your idea
20090324 16:57:52< alink> cjhopman: removing that line without using --max-fps seems good
20090324 16:58:07< Ivanovic> (pointing to weaknesses and/or obvious problems)
20090324 16:58:10< alink> (removinh the SDL_Delay(20))
20090324 16:58:29< jmunro> is the author of this post: http://dave.wesnoth.org/?p=9 present in irc?
20090324 16:58:53-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090324 16:59:37-!- turin [n=turin@168.215.249.253] has joined #wesnoth-dev
20090324 16:59:42-!- West_ [n=West@hd5e2754c.gavlegardarna.gavle.to] has joined #wesnoth-dev
20090324 17:00:32< alink> <cjhopman> i'm more interested in why my fps drops so much when scrolling  <- do you mean with the untouched code ?  
20090324 17:03:06-!- l3ftm1n0r [n=plic@122.161.108.139] has joined #wesnoth-dev
20090324 17:03:19< l3ftm1n0r> hi
20090324 17:03:20-!- Sapient [n=patrickp@wesnoth/developer/sapient] has left #wesnoth-dev []
20090324 17:03:33< Shadow_Master> jmunro: it'd be David White a.k.a. Dave a.k.a. Sirp
20090324 17:03:38< Shadow_Master> he isn't here atm, however.
20090324 17:03:42< alink> hi l3ftm1n0r
20090324 17:03:44< jmunro> Shadow_Master, thanks
20090324 17:06:01< cjhopman> alink: yes, it happens regardless of that delay code
20090324 17:06:43< alink> cjhopman: maybe just the cost of copy-paste a big part of the screen and redraw the new stuff then
20090324 17:06:48< l3ftm1n0r> hi
20090324 17:07:02< alink> plus the full screen update
20090324 17:07:07< l3ftm1n0r> alink, hi
20090324 17:07:23< l3ftm1n0r> i was here for the gsoc participation of wesnoth
20090324 17:07:28< alink> cjhopman: wow removing that get_display().delay(20) really seems to help a lot
20090324 17:07:35< l3ftm1n0r> love that game, cool for it to come in soc
20090324 17:08:13< alink> l3ftm1n0r: yes and it was also in SoC last year
20090324 17:08:16< Rhonda> tsk. <server> Removing illegal command from: dfranke. Current player is: (unfound) .
20090324 17:08:24 * Rhonda peeks at dfranke :)
20090324 17:08:26< l3ftm1n0r> oh
20090324 17:08:46< Shadow_Master> bad dfranke
20090324 17:08:48< Rhonda> Soliton: Did you had the time to look into bug #13252 by dfranke?
20090324 17:09:15-!- mordante [n=mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20090324 17:09:19< jmunro> l3ftm1n0r, i'm also a gsoc hopeful, nice to meet you :)
20090324 17:09:27< cjhopman> alink: since that is one of the few SDL_Delays that goes through display::delay(), you can get the same affect with --no-delay
20090324 17:09:30< mordante> hi
20090324 17:09:38< l3ftm1n0r> jmunro, hello
20090324 17:09:40< alink> hi mordante
20090324 17:09:43< Rhonda> Soliton: FWIW, I have the core file in ~rhonda/core/ including the server binary, so feel free to recompile/restart the server.
20090324 17:09:44< mordante> hi alink 
20090324 17:09:48< Shadow_Master> hail mordante
20090324 17:09:56< cjhopman> but yeah, i'm pretty sure that that should be changed
20090324 17:10:02< mordante> alink, I started to use cmake instead of autotools, and you had a question?
20090324 17:10:05< mordante> hi Shadow_Master 
20090324 17:10:38< mordante> alink, and you remember --max-fps, you can use it if you want to test the maximum frame count
20090324 17:10:57< alink> mordante: me no, maybe Pietro in his commit mail for the distribution managed by him
20090324 17:11:15< mordante> I thought it was you :-/
20090324 17:11:53< alink> see the copy paste of his logs yesterday 00h32
20090324 17:12:08< alink> when i was spying #commits for wesnoth commits
20090324 17:13:00< mordante> I was referring to 00:49
20090324 17:13:49< alink> ah yes sorry was not clear, the text between (..) was from his message
20090324 17:14:09-!- Azkar [n=leazkar@rh-116-202.johnstown.resnet.pitt.edu] has quit [Read error: 110 (Connection timed out)]
20090324 17:14:24< euschn_> I have a question about the EasyCoding tasks: Is "more powerful village naming" still free/unsolved? the wiki says ACardBoardRobot is working on it, but the date is 2 years ago
20090324 17:14:45-!- mysticspiral [n=toni-mai@79.194.40.38] has left #wesnoth-dev []
20090324 17:14:46< cjhopman> hm, scrolling without invalidating stuff and basically without actually updating any of the display is still very slow
20090324 17:15:08< Rhonda> (Expert level, 11 scenarios, some under reconstruction)
20090324 17:15:19< Rhonda> Is this still true (the under reconstruction part) for UTBFS?
20090324 17:15:21< Rhonda> -F
20090324 17:15:22< mordante> euschn_, feel free to take it
20090324 17:15:38< euschn_> thanks, ill have a go at it
20090324 17:15:53< mordante> just update the wiki you work on it
20090324 17:15:58< euschn_> ok
20090324 17:16:17< Shadow_Master> Rhonda: well, for 1.6 no.
20090324 17:16:20< alink> cjhopman: did you tried without the line display.cpp:1451 SDL_BlitSurface(screen,&srcrect,screen,&dstrect);
20090324 17:16:24< mordante> Rhonda, I think it unfortunately still true
20090324 17:16:36< Shadow_Master> for 1.7: we are not sure. I can't find anything that needs "reconstruction"
20090324 17:16:36< grzybacz> cjhopman: it's copying the unchanged portion of the screen
20090324 17:16:54< cjhopman> alink: yeah, removing that, still slow
20090324 17:16:57< mordante> Shadow_Master, has everything been polished now ?
20090324 17:17:05< grzybacz> oh
20090324 17:17:21< Shadow_Master> mordante: depends on how you define polished. The campaign wouldn't be in mainline if it wasn't polished when it was added in version 1.1.2
20090324 17:17:38< Rhonda> Shadow_Master: That's from 1.6
20090324 17:17:42< Shadow_Master> it is a bit buggy in some parts, yes, but we know not of any blockers.
20090324 17:17:52< Shadow_Master> but buggy != under reconstruction. The later sounds too violent.
20090324 17:17:52< mordante> and after that some people started to rewrite it and disappear in the process
20090324 17:18:13< Shadow_Master> mordante: well, esr and zookeeper fixed cycholka's mess with the parts that were rewritten
20090324 17:18:24< mordante> well if it was under reconstruction and still buggy I still consider it under reconstruction
20090324 17:18:38< alink> cjhopman: and what do you see when scrolling then ?
20090324 17:18:39-!- mysticspiral [n=toni-mai@p4FC22826.dip.t-dialin.net] has joined #wesnoth-dev
20090324 17:18:48< Shadow_Master> under reconstruction sounds to me like : " do not play or your savegames may become invalid in next release "
20090324 17:19:01< Shadow_Master> that cannot happen in 1.6, obviously.
20090324 17:19:12< mordante> true
20090324 17:19:19-!- mysticspiral [n=toni-mai@p4FC22826.dip.t-dialin.net] has left #wesnoth-dev []
20090324 17:20:24< Shadow_Master> it *may* happen in 1.7, but unlike fendrin, I'm not that pesimistic about the current state. I don't think rewriting any scenario of this campaign can be done by anyone by the original author without losing flavor
20090324 17:20:54< Shadow_Master> specially since Espreon isn't interested and fendrin and I aren't native english speakers like that original author (quartex). Ah yes, Espreon, fendrin and I are the current "maintainers"
20090324 17:21:45-!- orian [n=opera@aece127.neoplus.adsl.tpnet.pl] has joined #wesnoth-dev
20090324 17:21:56< Shadow_Master> fendrin suggests that I rewrite some WML. But I won't do like cycholka and expose my changes to the users in the form of mangled/unfinished/missing stuff
20090324 17:22:05< Shadow_Master> and of course I won't do it in 1.6
20090324 17:22:36< Shadow_Master> that is: who let cycholka mess UtBS in 1.5.x?
20090324 17:22:42< alink> cjhopman: mmh, in heavy case, just removing get_display().delay(20) helps the framerate but seems less nice for the cpu. Need to find some smart compromise
20090324 17:22:53< cjhopman> alink: removing that and some other stuff and i basically see no update when scrolling... the only thing kept updated are things with animation
20090324 17:22:56< Shadow_Master> had he maintained other campaigns before? I think not.
20090324 17:22:56< mordante> who let suokkos mess in trunk ;-)
20090324 17:23:02< mordante> suokko's
20090324 17:23:03< mordante> *
20090324 17:23:12< orian> hi
20090324 17:23:15< mordante> hi orian 
20090324 17:23:26< cjhopman> alink: removing it but keeping max-fps should be fine, i would think
20090324 17:23:26< Shadow_Master> mordante: suokko is still alive and destroying stuff. news at 11!
20090324 17:23:48< mordante> Shadow_Master, no pun intended ;-) http://xkcd.com/
20090324 17:24:01< happygrue> hehe
20090324 17:24:13< mordante> (I already saw my commit mails)
20090324 17:24:24< Shadow_Master> mordante: yeah, good one.
20090324 17:25:08 * mordante always gets scared when either xkcd or dilbert fulfill their prophecy on the same day
20090324 17:25:15< Shadow_Master> esr: pling pling.
20090324 17:26:25< Soliton> Rhonda: nope, no time. seems like side assignment is messed up for some reason. i'm not sure a server restart will help. does the problem happen in other games too?
20090324 17:26:49-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has quit [Remote closed the connection]
20090324 17:27:56< alink> cjhopman: what i mean is that in a heavy setup with a lot of ghosts, my fps drop to 30fps and 70%cpu. when using no-delay, it's 50fps but 100%cpu (which make sense)
20090324 17:28:52< Ivanovic> off to do some sports, cu
20090324 17:29:10< mordante> bye Ivanovic 
20090324 17:29:12< Shadow_Master> Ivanovic: wait a second! oh, too late
20090324 17:29:19< alink> mmh ghosts are not a good test for that, their continuous alpha change are not standard animation. rechecking
20090324 17:29:21< Shadow_Master> Ivanovic: don't forget to ask Torangan to update wescamp
20090324 17:29:35< cjhopman> i would think that we'd prefer to not have the fps drop...
20090324 17:30:04< mordante> corn, around?
20090324 17:30:32< Ivanovic> Shadow_Master: wescamp *is* updated
20090324 17:30:48< Ivanovic> Shadow_Master: the only thing missing as "content on the addon server" and the "new" addonserver setup correctly
20090324 17:30:49< Shadow_Master> Ivanovic: no.
20090324 17:30:55< orian> alink: sorry to butt in, but maybe you could try get_display().delay(10); ? and show results?
20090324 17:31:00< Shadow_Master> Ivanovic: trunk contains 1.6 UMC instead of 1.6
20090324 17:31:07< alink> cjhopman: depends, there is other things that rendering: AI, WML, etc.. Need to check if it doesn't slow them
20090324 17:31:13< Shadow_Master> http://www.wesnoth.org/gettext/index.php?package=allun&order=trans&version=branch
20090324 17:31:27< Ivanovic> Shadow_Master: yeah, and the "fix" is to make the addon server upload stuff to branches/1.6 instead of trunk/ like it does atm
20090324 17:31:35< Ivanovic> nothing torangan can do much about...
20090324 17:31:40< Shadow_Master> so the old content can't be moved away?
20090324 17:31:51< Shadow_Master> * branched
20090324 17:32:07< alink> orian: well, i can, but as i said but as i said my test case was not "standard". 
20090324 17:32:28< alink> trying..
20090324 17:33:12< Ivanovic> Shadow_Master: we "should" soon fix the upload stuff, then a simple svn mv on all trunk stuff to branches/1.6 does the trick
20090324 17:33:13< orian> alink: ok ;)
20090324 17:33:22< Ivanovic> Shadow_Master: but this can be done by anybody, even by me
20090324 17:33:31< Ivanovic> though atm the other stuff is not setup correctly, IIRC
20090324 17:33:36< Ivanovic> really off now, cu
20090324 17:33:39< Shadow_Master> k. go to your filthy sports now
20090324 17:33:44< cjhopman> alink: but other things shouldn't be affected... they can't do anything during that delay, anyway
20090324 17:33:45< Shadow_Master> ;)
20090324 17:33:47< alink> orian: seems to be somehow in the middle of the previous results 
20090324 17:34:23< alink> cjhopman: ah yes good point :)
20090324 17:34:31< Shadow_Master> I am sure Ivanovic calls "sports" to practicing archery using users as targets
20090324 17:35:43-!- VladimirSlavik [n=chatzill@wesnoth/translator/VladimirSlavik] has joined #wesnoth-dev
20090324 17:35:58< VladimirSlavik> hello, world!
20090324 17:36:10< Shadow_Master> no!
20090324 17:36:18< Shadow_Master> no more java plz
20090324 17:36:33< alink> VladimirSlavik: hello VladimirSlavik, but this just #wesnoth-dev here ;p
20090324 17:36:41< orian> I don't know how start, but I'm here becouse I'm looking information about GSOC. I read a http://www.wesnoth.org/wiki/SummerOfCodeIdeas..
20090324 17:36:42< alink> *this is just
20090324 17:37:26< alink> reading that was a good start ;)
20090324 17:37:31< jmunro> orian, im a gsoc student too, what do you need help with?
20090324 17:37:37< orian> and the refactoring memory usage or wml validation schemes looks quite interesting;)
20090324 17:37:43< VladimirSlavik> alink: with the GSOC crowds around... almost :)
20090324 17:39:06< mordante> orian, did you already create a forum account?
20090324 17:39:11< orian> no,
20090324 17:39:31< mordante> when you did, please ping me and I'll put you in the gsoc group
20090324 17:39:50< orian> ok
20090324 17:40:50< jmunro> orian, when it comes to writing the wiki page, take a look at the others for help on structure, thats what i did
20090324 17:41:03-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev
20090324 17:41:16< orian> jmunro: ok, thx;)
20090324 17:41:27< mordante> good advice ;-)
20090324 17:41:29< jmunro> no probs
20090324 17:42:03< orian> mordante: I created a forum account, nick the same: orian
20090324 17:42:11< mordante> Ivanovic, regarding the round stuff for cmake I added links to the commits which made those changes in the CMakeLists.txt
20090324 17:42:58 * alink wonders if it will make all the page looks the same or if it will push people to try to be more original
20090324 17:43:05< alink> s/page/pages
20090324 17:43:23< mordante> orian, added you to the group
20090324 17:43:46< orian> mordante: thx:)
20090324 17:43:48< mordante> alink, look what others did and do it better ;-)
20090324 17:44:58< Shadow_Master> loonycyborg: around ?
20090324 17:45:35< loonycyborg> Shadow_Master: Yep.
20090324 17:47:24-!- l3ftm1n0r [n=plic@122.161.108.139] has quit [Connection timed out]
20090324 17:47:37-!- l3ftm1n0r [n=plic@122.161.108.139] has joined #wesnoth-dev
20090324 17:48:18< mordante> loonycyborg, against which version of pango do you link on Windows?
20090324 17:51:09< loonycyborg> mordante: 1.20.5 according to .pc files.
20090324 17:51:43< mordante> same as I'm using here :-/
20090324 17:55:12< mordante> orian, can you tell a little about yourself?
20090324 17:55:20< esr> Shadow_Master: You looking for me?
20090324 17:55:41< Shadow_Master> esr: yes. What is the state of that priority 7 FR?
20090324 17:55:55< mordante> alink, cjhopman that 1000 fps mouse of the map "feature" is there for a long time
20090324 17:56:15-!- Sirp [n=me@wesnoth/developer/dave] has joined #wesnoth-dev
20090324 17:56:40< alink> mordante: yes i also think that i noticed it a long time ago
20090324 17:56:49< cjhopman> ~2 years now, i believe
20090324 17:57:07< mrfinch> Hey guys! Any GSoC people lurking?
20090324 17:57:09< cjhopman> that's when the 20ms delay per frame was added
20090324 17:57:10< mordante> yes might be 
20090324 17:57:26< cjhopman> or rather, per non-scrolling frame
20090324 17:57:37< alink> but maybe affected a little by the recent change about the behavior when mouse leave the window
20090324 17:57:40< mordante> yes I think I added that fix since before there was no limit
20090324 17:57:43< CIA-70> ai0867 * r34094 /trunk/data/core/images/ (424 files in 23 dirs): 
20090324 17:57:43< CIA-70> Ran optipng on data/core/images.
20090324 17:57:43< CIA-70> Original size: 4491 KiB on 424 files
20090324 17:57:43< CIA-70> Optimized size: 4313 KiB
20090324 17:57:43< CIA-70> Total saving: 178 KiB = 3% decrease
20090324 17:58:24< Sirp> Hello.
20090324 17:58:30< mordante> hi Sirp 
20090324 17:58:41< l3ftm1n0r> mrfinch, yea 
20090324 17:59:00< cjhopman> I'm pretty sure that delay should go... but it's been here longer than me, so someone else might know better
20090324 17:59:24-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090324 17:59:29< mordante> why?
20090324 17:59:35< grzybacz> 3% increase in file sizes is hardly worth the effort ;)
20090324 17:59:55< grzybacz> *decrease
20090324 18:00:04< cjhopman> because for every frame we have a 20ms delay... i mean at just 25 fps half your time is spent delaying
20090324 18:00:30< mordante> it delays if there's nothing better to do
20090324 18:00:33< cjhopman> no
20090324 18:00:39< mordante> no?
20090324 18:00:39< cjhopman> it delays every frame for 20ms
20090324 18:00:52< mordante> that's not what I implemented
20090324 18:01:00< cjhopman> this is something else
20090324 18:01:15< mordante> is this for dialogs?
20090324 18:02:00< cjhopman> no... controller_base.cpp line 209
20090324 18:02:38< mordante> ah yes not the code I worked on
20090324 18:02:41< cjhopman> every non-scrolling frame we have an SDL_Delay(20) call
20090324 18:03:02-!- mrfinch [n=mrfinch@82-32-33-143.cable.ubr05.aztw.blueyonder.co.uk] has quit ["Leaving"]
20090324 18:03:16< mordante> I was talking about display::draw_wrap
20090324 18:04:21< mordante> never looked at the controller_base coe
20090324 18:04:23< mordante> code*
20090324 18:04:31< cjhopman> no, the controller_base is the big one... my fps goes from ~50-70 to ~500-1000 on my laptop when i disable it
20090324 18:04:55< mordante> the 1000 is the limit imposed by display::draw_wrap
20090324 18:05:50< cjhopman> ah
20090324 18:06:13< cjhopman> alink: and i think i've figured out why scrolling is so slow
20090324 18:06:30 * alink is listening
20090324 18:06:32< esr> Shadow_Master: On my list -- I've reviewed it -- but I'm fighting gpsd firtes just now.  Your requirements were clear, thank you.
20090324 18:07:08< mordante> this limit is normally 20 ms but that delay is only used if we spend less than 20 ms since the last frame
20090324 18:07:09< alink> cjhopman: you have a huuuuuge screen ? ;-p
20090324 18:07:16< cjhopman> lol, no
20090324 18:07:18< mordante> --max-fps overrides that value
20090324 18:07:26-!- shikadibot [n=shikadi@wesnoth/bot/shikadibot] has joined #wesnoth-dev
20090324 18:07:37 * mordante starts to listen to cjhopman as well
20090324 18:07:38< shikadibot> Shadow_Master writes understandable stuff! News at 11!
20090324 18:07:43< cjhopman> yeah, the max-fps stuff works great
20090324 18:08:47< cjhopman> ok, so video.cpp at line 205 we are adding a new invalidated rectangle to a list of them, and we check it against all the others to see if they are contained in each other.
20090324 18:09:11< cjhopman> we do this, because later we call SDL_InvalidateRects() with this list and that call won't check for overlap and such
20090324 18:10:43< cjhopman> the problem is that when scrolling we get ~300-500 of these rects per frame...
20090324 18:10:53-!- West_ [n=West@hd5e2754c.gavlegardarna.gavle.to] has quit ["b'bye"]
20090324 18:10:56-!- l3ftm1n0r [n=plic@122.161.108.139] has quit [Read error: 60 (Operation timed out)]
20090324 18:11:19< cjhopman> its actually faster to just remove that containment check loop and pass them all to SDL
20090324 18:11:39< cjhopman> that improves fps from ~14 to ~25
20090324 18:11:42< mordante> don't we if invalidate too much already invalidate the entire screen?
20090324 18:12:08< alink> this is worse than that and has an other solution video::flip() check if we update >80%
20090324 18:12:35< mordante> but a rather nice improvement 
20090324 18:12:56< alink> so a scrolling probably introduce a big rectangle (>80% of the screen), we may stop rectangle there. 
20090324 18:13:14< alink> mmmh no the big rectangle may be added at the end
20090324 18:13:44< alink> or not, scolling code seems to start by that
20090324 18:14:25-!- Andylee [n=kvirc@188-23-6-2.adsl.highway.telekom.at] has joined #wesnoth-dev
20090324 18:14:26< Andylee> hi
20090324 18:14:32< Andylee> i am confused!
20090324 18:14:35< mordante> hi Andylee 
20090324 18:15:08< Andylee> wtf is wesnoth 1.6a? is this a version higher than 1.6? why is there no announcement made on wesnoth.org?
20090324 18:15:16< alink> cjhopman: or just call one SDL_InvalidateRect(map area) instead of SDL_InvalidateRects(ton of rectangles)
20090324 18:15:18< Andylee> is there a changelog available?
20090324 18:15:19< Shadow_Master> Andylee: assume 1.6a is what was announced.
20090324 18:15:23< loonycyborg> 1.6a is a bugfix release.
20090324 18:15:24< Andylee> ah
20090324 18:15:25< Andylee> ok
20090324 18:15:26< Shadow_Master> 1.6a is 1.6 + some critical bugfixes
20090324 18:15:29< Andylee> ah
20090324 18:15:30< Shadow_Master> emphasis on critical.
20090324 18:15:43< Andylee> and so i will have to DL again?
20090324 18:16:09< Andylee> damn... thats really not funny with my internet connection down here... 200kb/s
20090324 18:16:18< Shadow_Master> Andylee: do you use the source code?
20090324 18:16:23< loonycyborg> Checkout tags from svn :P
20090324 18:16:32< Shadow_Master> loonycyborg: not helpful.
20090324 18:16:38< Andylee> no i don't 
20090324 18:16:51< cjhopman> yeah, we actually do check the coverage of the rectangles... and if it is over a certain threshold we do SDL_Flip insteand of invalidaterects()
20090324 18:16:59< Andylee> i am already reading too many projets source code
20090324 18:17:15< Andylee> i would turn mad if i would have to do the same with wesnoth :)
20090324 18:17:24< Shadow_Master> Andylee: not as in read. I meant "compiling"
20090324 18:17:34< Andylee> no, i don't
20090324 18:17:35< alink> cjhopman: but currently after having checked their overlapping which you said take times
20090324 18:17:37< Shadow_Master> if you did, you'd just have to download a ~3 MB XDelta file to upgrade it
20090324 18:17:46< Andylee> hehe
20090324 18:17:55< Andylee> i cannot on this computer
20090324 18:17:56< cjhopman> alink: exactly
20090324 18:18:11< Andylee> and the other one is some hundred kilometres away :)
20090324 18:18:38-!- grantwu [n=Zenneth@westquad-147025.reshall.umich.edu] has joined #wesnoth-dev
20090324 18:18:55< Andylee> ah, one question: there was a call for participation... do you need german translation help?
20090324 18:19:11< alink> cjhopman: maybe store the total surface currently coverred by thes rectangle and check that value against 80% after each introduction of a new rectangles
20090324 18:19:13< cjhopman> alink: though the check just sums the areas of the rectangles... which could actually all overlap each other
20090324 18:19:15< Shadow_Master> Andylee: the people at #wesnoth-de probably know better than us
20090324 18:19:20< Shadow_Master> chrber is the maintainer
20090324 18:19:29< Andylee> ok, will just switch over there
20090324 18:19:59< cjhopman> I've gotta go to class now though... will definitely look into this some more
20090324 18:20:02< alink> cjhopman: I thought that we precisely prevented overlapping rectangle sin update_rect () ? 
20090324 18:20:06< Andylee> chrber is not online :)
20090324 18:20:21< alink> cjhopman: ok, nice catch and keep me posted
20090324 18:20:25< cjhopman> alink: no, just ones contained in other ones
20090324 18:20:33< cjhopman> entirely contained
20090324 18:20:44< alink> ah ok, that's different indeed
20090324 18:20:58< alink> and with our imbricated hexes that's important
20090324 18:21:22< cjhopman> ok, later 
20090324 18:22:51< Andylee> is there also interest in klingon or quenya translation?
20090324 18:23:10< loonycyborg> klingon would be nice :P
20090324 18:23:24< Andylee> i am serious... robot
20090324 18:23:27< loonycyborg> Never heard about quenya :/
20090324 18:23:35-!- Shadow_Master [n=shadowm@wesnoth/developer/shadowmaster] has quit ["foo"]
20090324 18:23:37< Andylee> it's elvish
20090324 18:24:01< alink> cjhopman: quick idea: watch the numbers of pixels witch coordinates multiples of 20(or so) and when 80% of them are updated, consider that 80% of the screen is updated
20090324 18:24:03< Andylee> a higher elvish language in which many ancient elvis books are written
20090324 18:24:14< alink> s/witch/with
20090324 18:24:17< loonycyborg> There's already latin translations, so wesnoth clearly doesn't have problems with rare languages.
20090324 18:25:23< Andylee> a, yeah alink
20090324 18:25:36-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090324 18:25:36< Andylee> my klingon is better then my english
20090324 18:25:45< Andylee> have to go now... will be back later
20090324 18:25:52< alink> Andylee: ?
20090324 18:25:52-!- Andylee is now known as Dachorganisation
20090324 18:26:11-!- l3ftm1n0r [n=plic@122.161.108.139] has joined #wesnoth-dev
20090324 18:26:18< ilor> mordante around?
20090324 18:26:21< alink> mmh with 20 that still a fine grid, maybe use a bugger value since we use 72x72 tiles
20090324 18:26:30< mordante> ilor, yes
20090324 18:26:32< alink> s/bugger/bigger
20090324 18:26:53< ilor> mordante: up for some assisted winxp gui2 bug hunting in around 1.5hrs? ;)
20090324 18:27:23< mordante> Dachorganisation, you should talk to elias he also wanted to add klingon ;-)
20090324 18:28:06< mordante> ilor, not sure whether I've much time tonight but we can try
20090324 18:28:16< Crab_> mordante, are there any more suggestions regarding my run_file patch ( https://gna.org/patch/?1137 ) ?
20090324 18:28:52< elias> mordante: i said that just in an attempt to troll that royal-catalan-whatever mailinglist spam :P
20090324 18:28:53< mordante> Crab_, not yet catching up with all wesnoth stuff, will do so later
20090324 18:29:02< Crab_> mordante: ok
20090324 18:29:33< mordante> elias, and now you see what comes from trolling ;-)
20090324 18:29:54< elias> heh
20090324 18:32:07-!- Polarina [n=polarina@freecode-project/mentor/polarina] has joined #wesnoth-dev
20090324 18:34:23-!- cjhopman [n=chris@wesnoth/developer/cjhopman] has quit ["Konversation terminated!"]
20090324 18:37:55-!- csaunders [n=christop@137.122.202.12] has joined #wesnoth-dev
20090324 18:41:06-!- Rrenys [n=rrenys@81-20-159-197.levira.ee] has joined #wesnoth-dev
20090324 18:42:04< mordante> Dragonking, can you have a look at https://gna.org/patch/?1137 to see whether the formula part is ok?
20090324 18:42:06-!- shikadibot [n=shikadi@wesnoth/bot/shikadibot] has quit [Read error: 110 (Connection timed out)]
20090324 18:43:13< mordante> Crab_, patch looks good to me, only the doxygen comment should have a @params part, but I can do that
20090324 18:43:39< mordante> Crab_, I let DK have a look since he's more familiar with that code and then I'll commit it
20090324 18:43:57< Crab_> mordante: ok. thanks. 
20090324 18:44:07< mordante> no problem, thank you for the patch
20090324 18:45:30< csaunders> mordante: hello, I am a student who would like to participate in the summer of code.  Could you mark that on my account (csaunders)
20090324 18:49:09< mordante> csaunders, done
20090324 18:49:31< mordante> csaunders, haven't seen you here before did you read the wiki page?
20090324 18:50:26< csaunders> mordante, yes.  I've been looking at it.  Started this morning.
20090324 18:50:59< mordante> good did you already picked a project you want to work on?
20090324 18:51:57< csaunders> yes, going for a somewhat popular one.  Game saves
20090324 18:52:35< mordante> most seem popular ;-)
20090324 18:52:58-!- stikonas [n=stikonas@wesnoth/translator/stikonas] has quit [Read error: 54 (Connection reset by peer)]
20090324 18:53:01< mordante> can you tell a bit about yourself?
20090324 18:53:08< csaunders> Sure!
20090324 18:54:56< csaunders> I'm a 4th year CS student (with co-op option) from the University of Ottawa.  I've done quite a bit of work with data in rather raw forms, such as interpreting 3rd party CSV and XML data files.  I've also done a few projects for school using various languages and tools.  Most being in java.
20090324 18:55:38< Sirp> hi csaunders, welcome.
20090324 18:55:53< csaunders> I've done some work with C++ and my most recent work with it has been for graphics using the OpenGL API.
20090324 18:55:54-!- Azkar [n=leazkar@rh-116-202.johnstown.resnet.pitt.edu] has joined #wesnoth-dev
20090324 18:56:19< csaunders> Sirp: hello
20090324 18:56:39< Sirp> csaunders: is there any chance at all I could interest you in doing some work on our stats database and making a website to run reports on it?
20090324 18:57:06-!- ryochan7 [n=ryochan7@173.9.238.161] has joined #wesnoth-dev
20090324 18:57:16< Sirp> personally I think that is a much easier project than the saved games one, though of course you should do what interests you the most.
20090324 18:57:39< csaunders> Sirp: I could take a look at it!
20090324 18:58:02< csaunders> The saved game one really caught my eye.  But I can understand it being rather difficult from what I've read on the wiki
20090324 19:00:13-!- Andrai [n=Andrai@150.165.63.86] has joined #wesnoth-dev
20090324 19:00:51< csaunders> Sirp: I like the option of using a framework of ones choice for the stats server.
20090324 19:01:03-!- ayona [n=ayona@151.57.11.24] has joined #wesnoth-dev
20090324 19:01:06< Sirp> csaunders: yes, I strongly believe in giving developers choices where possible.
20090324 19:01:53< Sirp> csaunders: have you done any web development in the past?
20090324 19:03:07-!- silene [n=plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20090324 19:03:07< csaunders> Sirp: My last two work terms were doing web development with Tomcat.  And I will be working for another web dev company using Rails.
20090324 19:03:32< silene> hi
20090324 19:03:42< Sirp> hi silene 
20090324 19:03:47< Crab_> hi silene
20090324 19:03:49< csaunders> Sirp: I am more into the backend logic of things, but I am always open for getting better at CSS and such.
20090324 19:03:51< Sirp> csaunders: I see. And do you have experience with MySQL?
20090324 19:04:22< Sirp> csaunders: actually I don't think you need to worry about the web site being "pretty". If you make it work and it is nice I'm sure someone else will play with all that CSS stuff.
20090324 19:04:52< Sirp> csaunders: as you probably know, we just released Wesnoth 1.6 a few days ago....
20090324 19:04:53< csaunders> Sirp, yes.  I've used mysql during my workterms, as well as for my own services at home such as MediaTomb.
20090324 19:05:25< mordante> hi silene 
20090324 19:05:48< Sirp> csaunders: so I hacked up a script to read our stats db and show us this: http://www.wesnoth.org/cgi-bin/stats/usage-1.6.pl
20090324 19:05:57< Sirp> it shows how many Wesnoth scenarios have been completed each day since release.
20090324 19:06:05< csaunders> Sirp: I've only used the Ubuntu repository versions of Westnoth.  So I am not too sure at what version I am at.  Though, I have downloaded the trunk code.  Just need to compile it
20090324 19:06:40< Sirp> csaunders: however what we really want is a *framework* which can easily show us all sorts of graphs and charts like this. e.g. what if someone wanted to know how many people had played a specific campaign or scenario, or if users of different operating systems etc had played...
20090324 19:07:16< silene> loonycyborg: as a matter of fact, i didn't add the check in the code, it was already there; i added the autoconf test so that the code would be optimized on my computer; for people not using autoconf, the code would have stayed as bad as it was before
20090324 19:07:34< Sirp> so that is where it gets both more complicated and interesting.....having something nice and generic yet easy to use which allows diverse reporting.
20090324 19:07:41< mordante> silene, regarding your lua code, could you add a new log domain and write your log message there instead of on std::cerr?
20090324 19:07:54< mordante> s/on/to/
20090324 19:08:35< silene> mordante: sure, i thought i had done it, i must have forgotten or lost the commit
20090324 19:08:37< Sirp> csaunders: anyhow was just throwing that out as a potentially interesting alternative for you to consider.
20090324 19:08:41< csaunders> Sirp: that is understandable.
20090324 19:09:01 * Sirp hopes Yogi doesn't notice and try to get revenge on him for trying to 'steal' students off of the save game project. :)
20090324 19:09:08-!- ayona [n=ayona@151.57.11.24] has left #wesnoth-dev []
20090324 19:09:20< mordante> thanks
20090324 19:09:43-!- alink [n=alink@wesnoth/developer/alink] has quit [Remote closed the connection]
20090324 19:10:48-!- ayona [n=ayona@151.57.11.24] has joined #wesnoth-dev
20090324 19:10:58-!- ayona [n=ayona@151.57.11.24] has left #wesnoth-dev []
20090324 19:11:30< csaunders> Sirp: I am quite interested in it!  I've never really made a framework, so it should be rather challenging!
20090324 19:11:50-!- KnightR [n=anonymou@bb219-74-177-1.singnet.com.sg] has joined #wesnoth-dev
20090324 19:13:39< csaunders> I've got to go, but I will be back later!
20090324 19:13:44-!- csaunders [n=christop@137.122.202.12] has left #wesnoth-dev []
20090324 19:13:50-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has quit [Read error: 110 (Connection timed out)]
20090324 19:14:24-!- ryochan7 [n=ryochan7@173.9.238.161] has quit ["Leaving"]
20090324 19:16:33< silene> Shadow_Master: when i designed the logging system of wesnoth, the goal was that the user could control the logging level; i didn't know the policy had changed since then, and that the user was no longer allower to prevent a message from being displayed
20090324 19:16:38-!- Andrai [n=Andrai@150.165.63.86] has quit ["Leaving"]
20090324 19:17:56-!- l3ftm1n0r [n=plic@122.161.108.139] has quit [Read error: 110 (Connection timed out)]
20090324 19:18:21-!- l3ftm1n0r [n=plic@122.161.108.139] has joined #wesnoth-dev
20090324 19:19:14-!- Baufo [n=thomas@wesnoth/developer/baufo] has joined #wesnoth-dev
20090324 19:22:04< silene> alink: when there is no double buffering; there is no point in calling sdl_flip; sdl_updaterects is always sufficient and often faster
20090324 19:24:21< loonycyborg> silene: Is there an easy way to test whether lua support works?
20090324 19:24:34< silene> Shadow_Master: oh, sorry, i understand now; who the hell decided to put this stringstream in the lg namespace, that's confusing to the possible
20090324 19:25:16< silene> loonycyborg: would downloading the tutorial and running it be okay?
20090324 19:26:04< loonycyborg> Probably. Is it somewhere on wiki?
20090324 19:26:22< loonycyborg> I'd appreciate if lua was used somewhere in test scenatio.
20090324 19:26:42< silene> there was the link in my mail, there is the link on the wiki, and both point to the file stored on gna
20090324 19:27:11< silene> loonycyborg: yes, i intended to at first, but that was before i discovered the content of the test scenario, then i fled in terror :-)
20090324 19:28:11< silene> i will try to come up with something
20090324 19:33:34< loonycyborg> silene: To make use of of the lua tutorial you overwrite 1_Tutorial.cfg in data/campaigns/tutorial/scenarios, right?
20090324 19:33:36< AI0867> is there any reason for ~data/maps to exist?
20090324 19:33:49< silene> loonycyborg: right
20090324 19:33:51-!- csaunders [n=christop@137.122.205.167] has joined #wesnoth-dev
20090324 19:34:03< silene> loonycyborg: your installed location at least
20090324 19:34:43-!- elias [n=elias@allegro/developer/allefant] has quit ["Leaving"]
20090324 19:36:05< silene> mordante: while translating the release notes, i understood you intended to have a user interface controlable from wml (?); is there a document describing it somewhere? i would be interested to see if it is possible to have lua call it too somehow
20090324 19:36:41-!- lu_zero [n=lu_zero@gentoo/developer/lu-zero] has joined #wesnoth-dev
20090324 19:36:44< lu_zero> hi
20090324 19:37:35< loonycyborg> silene: Is there something special in lua tutorial that differs from purely wml tutorial that you can notice when playing it?
20090324 19:37:42< lu_zero> mr_bones_ told me that the autotools build system got deprecated because nobody was available to keep it up
20090324 19:37:44< mordante> hi lu_zero 
20090324 19:38:12< lu_zero> if is possible I'd volunteer to keep it up to date
20090324 19:38:31< mordante> silene, http://wesnoth.org/wiki/GUIToolkit
20090324 19:39:00< mordante> the end result should make it easier to integrate with WML, but this hasn't been done yet
20090324 19:39:43< silene> loonycyborg: unfortunately no (in fact yes, but it is hardly noticible); what i do in practice is that i insert a syntax error in the lua code and look at the console :-)
20090324 19:39:44< mordante> lu_zero, thanks for the offer, not sure whether it's needed silene still uses autotools
20090324 19:39:49< loonycyborg> lu_zero: Since then silene reappeared. He uses autotools. Perhaps he would appreciate help.
20090324 19:40:34< loonycyborg> s/help/your help
20090324 19:40:41< mordante> and the change build system project ended in a fiasco so we're stuck with three systems instead of one :-/
20090324 19:40:48< silene> lu_zero: i would indeed appreciate; as far as i know it is up-to-date, but if you notice anything, just tell
20090324 19:41:29< loonycyborg> mordante: Enjoy the choice :P
20090324 19:42:19< mordante> loonycyborg, I'm not complaining ;-) But the project was a fiasco 
20090324 19:42:54< silene> mordante: is the wiki representative of the current code? or are there some wishlists (?) i should skip?
20090324 19:43:33< lu_zero> silene I'll have a look and pester mr_bones about updating the ebuild accordinly =P
20090324 19:43:33< mordante> the page is up to date only I should add some info about the future goals
20090324 19:44:07< silene> great
20090324 19:45:33< mordante> I also might change the formula part to lua... but that depends whether or not lua is mandatory
20090324 19:47:08< silene> i intend to push it to be mandatory; it was not my goal to have it optional, especially as it fixes some things that are currently not feasible in pure wml, for instance dynamic objectives
20090324 19:48:20< mordante> that's what I thought as well
20090324 19:49:33< loonycyborg> silene: Be sure to tell that to sapient.
20090324 19:49:45-!- esr [n=chatzill@static-71-162-243-5.phlapa.fios.verizon.net] has quit ["ChatZilla 0.9.84 [Firefox 3.0.7/2009030423]"]
20090324 19:54:29< loonycyborg> lu_zero: IIRC there where some wesnoth autotools buildsystem bugs in Gentoo that may help you to find something to work on.
20090324 19:55:46-!- esr [n=chatzill@static-71-162-243-5.phlapa.fios.verizon.net] has joined #wesnoth-dev
20090324 19:59:16-!- euschn_ [n=chatzill@85-127-103-245.dynamic.xdsl-line.inode.at] has quit ["ChatZilla 0.9.84 [Firefox 3.0.7/2009030423]"]
20090324 20:21:01-!- KnightR [n=anonymou@bb219-74-177-1.singnet.com.sg] has quit []
20090324 20:21:05-!- VladimirSlavik_ [n=chatzill@88.103.147.178] has joined #wesnoth-dev
20090324 20:22:53< loonycyborg> Man. It looks like my cross-compiled lua dll actually works! Everything that was required is two tweaks to lua's Makefiles..
20090324 20:23:06< mordante> :-)
20090324 20:23:22< mordante> btw I saw sapient made it optional 
20090324 20:26:42< loonycyborg> That means that we probably won't have to ship msvcr80.dll required by dll from lua's binary package.
20090324 20:28:05< mordante> :-)
20090324 20:28:16< Polarina> Wait.. Lua in Wesnoth, is that the plan?
20090324 20:30:18-!- jmunro is now known as jmunro[a]
20090324 20:31:04< loonycyborg> Polarina: Yes. Unless someone manages to come up with acceptable python solution.
20090324 20:31:39< Polarina> What's the issue with using Python?
20090324 20:32:09-!- noy [n=Noy@wesnoth/developer/noy] has quit []
20090324 20:33:18< loonycyborg> Because python is not sandboxed making addons require user to trust addon's author to execute arbitrary code.
20090324 20:36:10< CIA-70> silene * r34095 /trunk/data/scenario-test.cfg: Added Lua check to test scenario.
20090324 20:39:35-!- VladimirSlavik [n=chatzill@wesnoth/translator/VladimirSlavik] has quit [Read error: 110 (Connection timed out)]
20090324 20:42:16< Ivanovic> re (somehow...)
20090324 20:42:21< Ivanovic> have i missed much?
20090324 20:43:13-!- stikonas [n=stikonas@ctv-79-132-187-171.vinita.lt] has joined #wesnoth-dev
20090324 20:43:21< mordante> not really
20090324 20:45:47< VladimirSlavik_> oh and Ivanovic, thanks for fixing these links
20090324 20:46:12-!- l3ftm1n0r [n=plic@122.161.108.139] has quit [Read error: 60 (Operation timed out)]
20090324 20:46:49< Ivanovic> VladimirSlavik_: you're welcome
20090324 20:47:12< Ivanovic> VladimirSlavik_: like i said in my mail to the list: i wanted to make sure that they are correct in the versions that were translated and thus directly checked things
20090324 20:47:23< Ivanovic> time to prepare me something for lunch
20090324 20:47:35< Ivanovic> (bah, exam tomorrow, who wants to go there in my place?)
20090324 20:47:45< VladimirSlavik_> what is it?
20090324 20:49:11< mordante> oh Ivanovic I saw you already added the poll round etc detection to cmake :-)
20090324 20:49:16< Ivanovic> "current research in embedded system software: operating system technics"
20090324 20:49:48< Ivanovic> basically stuff about creating product lines of embedded operating systems and how to do it in a "nice way"
20090324 20:50:02< mordante> I think some of use might want to go for you, the question is which mark do you hope to get ;-)
20090324 20:50:23< Ivanovic> allowing the same base being used for various final systems while still keeping it usable for embedded systems
20090324 20:50:30< Ivanovic> mordante: sure, that is always a question
20090324 20:50:45< Ivanovic> mordante: it is the lecture with the lovely ifdef-hell i showed you in october...
20090324 20:51:02< mordante> I already thought that ;-)
20090324 20:51:27< lu_zero> silene sdl could be inferred by pkgconfig
20090324 20:51:44< silene> Ivanovic: that's quite specific for a course topic; what grade (?) is it?
20090324 20:51:57< Ivanovic> silene: the so called "diplom"
20090324 20:52:04< Ivanovic> basically equivalent to a master
20090324 20:52:27< Ivanovic> and yeah, those are the "special lectures" where you specialize into an area (and i specilize into embedded systems)
20090324 20:52:32 * lu_zero is digging a bit in the configure.ac
20090324 20:52:57< Ivanovic> those of you understanding german might be able to make some use of the official page: http://ess.cs.uni-dortmund.de/DE/Teaching/WS2008/BST/
20090324 20:53:41< silene> lu_zero: you mean trying again with pkg-config if sdl-config failed? or is it the other way around?
20090324 20:53:49< Ivanovic> basically it is about all the lovely stuff from design to implementation (including aspect and feature orientation as well as syntecising objects via template meta programming)
20090324 20:54:06< mordante> lu_zero, if you have changes http://patches.wesnoth.org is the best place to post them
20090324 20:56:02< lu_zero> silene the other way around
20090324 20:56:08< lu_zero> first pkgconfig
20090324 20:56:11< orian> i'm looking for info about http://www.wesnoth.org/wiki/SoC_Ideas_Schemes, what exactly is "schemes", how the work or anything
20090324 20:56:21< lu_zero> (should be quick) then other means
20090324 20:56:28< silene> Ivanovic: aspect-oriented programming; sorry, i couldn't resist laughing at the sight of it :-)
20090324 20:56:41< mordante> orian, you're familiar with xml
20090324 20:56:44-!- csaunders [n=christop@137.122.205.167] has quit [Read error: 110 (Connection timed out)]
20090324 20:56:52< lu_zero> I notice that there are few buildchecks that could be avoided if pkgconfig is available
20090324 20:57:03 * mordante puts his armour
20090324 20:57:09< lu_zero> also there is a peculiar way to use libtool
20090324 20:57:38< lu_zero> maybe LT_foo stuff could replace it (depending on what you want to support in this regard
20090324 20:57:56< Ivanovic> silene: there are some nice things possible with it
20090324 20:58:18< Ivanovic> and the head is not smoking as badly as it is with too much template meta programming
20090324 20:58:20< Ivanovic> ;)
20090324 20:59:10< orian> mordante: ok, I think I understand xml, but i'm looking for some detail or docs, can you give me any clue?
20090324 20:59:39< orian> mordante: i'm thinking about schemes
20090324 20:59:41< mordante> orian, also familiar with xml scheme validation (xsd files)
20090324 21:00:57< Polarina> loonycyborg: Can't you simply disable/modify offending modules?
20090324 21:01:20< CIA-70> silene * r34096 /trunk/src/ (log.cpp log.hpp scripting/lua.cpp): Added a "lua" log domain.
20090324 21:01:30-!- Dachorganisation is now known as Andylee
20090324 21:01:54< orian> mordante: thx, i will make little search ;)
20090324 21:02:21< loonycyborg> Polarina: Nope. There always be ways of getting to them through other modules.
20090324 21:02:29< mordante> orian, no need I can explain, just wanted to know whether you were familiar with xsd
20090324 21:02:43< Polarina> loonycyborg: Other modules?
20090324 21:02:52< mordante> just if you know xsd my explaination can be much shorter
20090324 21:03:13< Polarina> loonycyborg: If modules are the only issue, one could simply create his own set of modules.
20090324 21:03:49< loonycyborg> Polarina: But rich standard library is whole point of python :)
20090324 21:03:50< happygrue> wesbot: seen Noyga
20090324 21:03:50< wesbot> happygrue: The person with the nick Noyga 2d 10h ago they were seen quitting on the channel #wesnoth-mp with the message: Client Quit
20090324 21:04:17< ilor> how is the lua test in the test scenario supposed to work?
20090324 21:04:29< mordante> orian, schemes are a description file of which elements in WML are required optional what values are allowed etc
20090324 21:04:31< Polarina> loonycyborg: Then just create a wrapper around the wesnoth accessable modules and the Python modules implementing non-abusable functions and classes.
20090324 21:05:11< mordante> orian, have a look here http://www.wesnoth.org/devdocs/canvas_8cpp-source.html at line 98
20090324 21:05:31< loonycyborg> Polarina: Perhaps python implementation in parrot will produce really sandboxed python in future.
20090324 21:05:33< mordante> this is a kind of scheme for a line
20090324 21:05:40< dfranke> Polarina: that's what was originally tried.  See https://gna.org/bugs/index.php?13048
20090324 21:06:13< Polarina> Hmm...
20090324 21:06:16< loonycyborg> http://en.wikipedia.org/wiki/Parrot_virtual_machine
20090324 21:06:46< loonycyborg> Currently parrot's python implementation is rather incomplete..
20090324 21:07:05< mordante> orian, it defines all key value pairs for a line and whether or not they're optional and what their default value is
20090324 21:07:12< Polarina> But are there any other issues than modules?
20090324 21:08:17< mordante> orian, this is in human readable form in the source for the gsoc project we want the scheme files to stand on their own or be generated from the sources
20090324 21:08:45< mordante> these files will then be used to validate whether or not the loaded WML is valid and if not give a sane diagnostic
20090324 21:09:04< loonycyborg> Polarina: The other issue was that Python AIs sucked. They were too low-level and implemented using raw python/C api while Boost.Python would be more appropriate.
20090324 21:09:30< Polarina> I see.
20090324 21:09:30< mordante> this way the user can easily fix the problem and the game engine can be sure it will only get valid syntax 
20090324 21:10:16< Polarina> If there's going to be Lua, I would prefer some functions, such as bigint support (like long on Python)
20090324 21:12:13-!- cjhopman [i=cjhopman@wesnoth/developer/cjhopman] has joined #wesnoth-dev
20090324 21:12:16< loonycyborg> ilor: It will print a message on one of hexes: "lua says.."
20090324 21:13:23-!- natasiel [n=chatzill@bas3-sherbrooke40-1279576088.dsl.bell.ca] has joined #wesnoth-dev
20090324 21:19:46-!- Elvish_Pillager [n=eli@66-189-40-122.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090324 21:20:25< orian> mordante: I understand generating from source, but what do you mean by "stand on their own"?
20090324 21:20:35< natasiel> "https://gna.org/bugs/?13252" I can't reproduce but can tell what approximatively happened, letting 2 zombies game in lobby
20090324 21:21:12< natasiel> where should i write my deposition?
20090324 21:21:25< mordante> natasiel, add the info to the bug report
20090324 21:21:32< natasiel> thank you mordante 
20090324 21:22:20< mordante> orian, stand on their own; that they're plain files so you don't need to have the source around to generate them
20090324 21:25:25< orian> mordante: but after changing source, you will need to change them too, am I right?
20090324 21:25:46< silene> mordante: it isn't obvious to me while reading the documentation; can a wml event create an arbitrary window? get information from it? interact with it?
20090324 21:26:15< mordante> orian,  yes, but if you generate them from the source you'll need to write a generator that extracts them
20090324 21:26:40< mordante> and this could be added to the build system to automatically generate them
20090324 21:27:02< mordante> silene, no not yet it can only show a message dialog
20090324 21:27:29< mordante> what kind of window do you have in mind
20090324 21:28:09< silene> mordante: forinstance, i know designers have wanted for a long time to be able to have rpg-like inventory window
20090324 21:28:22< orian> mordante: schemas are not present in wesnoth ?
20090324 21:28:30< silene> if we can somehow achieve it, it would be great
20090324 21:28:31< mordante> orian, no
20090324 21:28:41-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20090324 21:28:49< Crab_> mordante: 'what kind of window do you have in mind'. I am interested in that too. For example - what about a dialog box to ask the user to type a string, or a dlg box to select from a (possibly, large) list.
20090324 21:29:10< orian> grzywacz: sounds like polish nick :);>
20090324 21:29:24< zookeeper> did someone just break some trunk WML? i'm getting some errors...
20090324 21:29:30< mordante> Crab_, type a string can already be done with the message dialog and selecting from a list as well
20090324 21:29:32< grzywacz> orian, yes, why?
20090324 21:29:37< Crab_> mordante: thanks.
20090324 21:29:43< orian> grzywacz: tak, ja te¿ ;P
20090324 21:29:49< grzywacz> ah, good luck then ;)
20090324 21:29:51< CIA-70> ai0867 * r34097 /trunk/ (data/core/units.cfg src/filesystem.cpp): Remove the ~units, ~multiplayer and ~ais directories.
20090324 21:30:00-!- grzybacz [i=grzywacz@kolos.math.uni.lodz.pl] has quit ["leaving"]
20090324 21:30:10< loonycyborg> zookeeper: What errors?
20090324 21:30:27< zookeeper> loonycyborg, WML errors naturally?
20090324 21:31:09< mordante> Crab_, start the test scenario and move the units around 18,10 on the texts
20090324 21:31:18< loonycyborg> Naturally. What are error messages? Are they on stdout or in message box?
20090324 21:31:25< mordante> that will give you an idea of the current possibilities
20090324 21:32:19< mordante> silene, yes it should be possible to do, but not implemented
20090324 21:32:24< zookeeper> loonycyborg, message box on startup
20090324 21:32:37< zookeeper> stderr too
20090324 21:32:54-!- kitty_ [n=kitty@wesnoth/artist/kitty] has joined #wesnoth-dev
20090324 21:32:59< loonycyborg> Hm.. Perhaps I need to update it..
20090324 21:33:38< Crab_> mordante, can additional WML parts be loaded in this way ? for example 'user types "ai/formula/test.cfg"  and that file is parsed as a WML and added to current scenario.' use case.
20090324 21:34:49< silene> mordante: i'm sure it should be possible; what i'm interested in is if it's part of your plans to provide this ability to scripts? :-)
20090324 21:35:00< mordante> Crab_, no but the problem is not a GUI issue and also not sure whether that functionallity is really wanted
20090324 21:36:53< Crab_> mordante, ok. btw, played with the test scenario and got it to segfault :)
20090324 21:37:45< zookeeper> esr, btw, i have a fair list of things i'd like to see done in 1.7...one of them would be writing some nice and short narrator story messages on defeat for most campaign scenarios in the vein of AOI.
20090324 21:37:55< mordante> silene, this idea was not yet a real idea but I see the use of it and I think lua makes it much easier to add the interface
20090324 21:38:28< mordante> I wanted to give WML more access to the GUI code to solve some of these issues
20090324 21:38:35< mordante> but mainly in the panels
20090324 21:38:57< mordante> but I see a lot of usage for dialogs as well
20090324 21:39:51< mordante> Crab_, that sounds bad, can you post a bug report? (Please also mention your screensize)
20090324 21:40:06< mordante> s/screensize/wensoth window size/
20090324 21:41:04< silene> okay then, we should discuss it some day (not tonight, i had a meeting the whole day long, and i'm starting to pay its toll)
20090324 21:41:24< loonycyborg> zookeeper: Updated.
20090324 21:41:53< mordante> ok
20090324 21:45:08< Crab_> mordante, s/segfault/assertion failure. it looked like a race condition. see https://gna.org/bugs/index.php?13256
20090324 21:45:16-!- Andylee [n=kvirc@188-23-6-2.adsl.highway.telekom.at] has quit [Read error: 104 (Connection reset by peer)]
20090324 21:45:27< CIA-70> zookeeper * r34098 /trunk/data/campaigns/Heir_To_The_Throne/scenarios/02_Blackwater_Port.cfg: Fixed bug #13255. Also reordered the events a bit and changed the facing of the reinforcements.
20090324 21:48:02< Ivanovic> crimson_penguin: would be nice if you could have a lok at this thread: http://www.wesnoth.org/forum/viewtopic.php?f=4&t=24587&start=0&st=0&sk=t&sd=a
20090324 21:49:17< CIA-70> zookeeper * r34099 /branches/1.6/data/campaigns/Heir_To_The_Throne/scenarios/02_Blackwater_Port.cfg: Ported r34098 to 1.6.
20090324 21:49:58-!- busfahrer is now known as duschfahrer
20090324 21:51:29< crimson_penguin> Ivanovic: that second person didn't install the game; they're running it from the dmg
20090324 21:51:38< crimson_penguin> but it's still really weird
20090324 21:51:40< Ivanovic> then post this as reply
20090324 21:52:10< crimson_penguin> I will
20090324 21:52:23< esr> zookeeper: Reasonable.  I could do that.
20090324 21:56:10< orian> mordante: big thx for i have few more questions, is any tool described on ideas page ready? is there any demand on programing language?
20090324 21:56:26< orian> mordante: big thx for clear explanation ;)
20090324 21:57:02< mordante> Crab_, thanks for the report, you really found the best message for it ;-)
20090324 21:57:12< Crab_> :)
20090324 21:57:23< CIA-70> ivanovic * r34100 /branches/1.6/cmake/SearchForStuff.cmake: typo fixes, thanks stikonas
20090324 21:57:33< CIA-70> ivanovic * r34101 /trunk/cmake/SearchForStuff.cmake: typo fixes, thanks stikonas
20090324 21:57:41-!- faktor4u [n=faktor@chello089077167144.chello.pl] has joined #wesnoth-dev
20090324 21:57:49< mordante> orian, there are no tools to extract the schema files from the source since we haven't defined how they exactly should look
20090324 21:58:27< mordante> orian, there is a tool wiki_grabber.py that extracts the WIKI comment (that page I pointed you at before)
20090324 21:58:54< mordante> that tool generates wiki pages like http://wesnoth.org/wiki/GUICanvasWML
20090324 21:59:16< orian> mordante: i found wiki tool, again big thx mordande;) I think I will be creating my wiki page slowly ;)
20090324 21:59:31< mordante> :-) 
20090324 21:59:37< mordante> I look forward to that
20090324 22:00:44< orian> mordante: so if i good understand the first point will be to define how it would look like, and so on.. but what about programming languages and utils?
20090324 22:01:14< mordante> well the part that does the validation should be inside Wesnoth and thus C++
20090324 22:01:25-!- silene [n=plouf@wesnoth/developer/silene] has quit [Read error: 113 (No route to host)]
20090324 22:02:35< mordante> for the extractor you can pick something else not sure what languages you're comfortable with
20090324 22:03:16< orian> mordante: i think that C++ isn't a problem ;) i was little scared about python because i'm rather begginer in it ;)
20090324 22:03:38< mordante> me too ;-)
20090324 22:04:06< mordante> you're more familiar with another language which can easily extract text?
20090324 22:04:07 * loonycyborg was a python beginner some time ago too
20090324 22:05:19 * esr was a Python beginner, before he learned to level mountains with the laser beams from his eyes...
20090324 22:05:51< mordante> I'm still a beginner, but I can write simple extractor tools, which suits my needs
20090324 22:06:07< mordante> and python code is still readable after a month ;-)
20090324 22:06:23< esr> ...unlike Perl code.
20090324 22:06:47< mordante> yes perl convinced me bitrot really exists ;-)
20090324 22:07:25< orian> mordante: I think I can learn bit more python ;)
20090324 22:07:41< mordante> gsoc is about learning
20090324 22:07:43< mordante> ;-)
20090324 22:08:17< Ivanovic> *life* is about learning...
20090324 22:08:17< Ivanovic> ;)
20090324 22:08:19< orian> since now I was mostly using C/C++, since year java and C# for short period,
20090324 22:08:26< orian> :)
20090324 22:10:31< Ivanovic> mordante: https://gna.org/bugs/index.php?13256
20090324 22:10:35 * Polarina does crazy stuff.
20090324 22:11:49 * Ivanovic is feeling really crazy today
20090324 22:11:54< Ivanovic> i am doing *normal* stuff
20090324 22:11:56< Ivanovic> ^^
20090324 22:11:58 * mordante was just wondering how long it would take before somebody wants to give that bug to me :-(
20090324 22:12:44< Polarina> Ivanovic: Would you consider this (http://simnet.is/gabrielp/improved-fan.jpg) normal stuff or crazy stuff? :P
20090324 22:12:45< Ivanovic> sorry that it took me almost half an hour
20090324 22:12:47< Ivanovic> ;)
20090324 22:13:12< Crab_> mordante, 25 minutes 15 seconds, that is )
20090324 22:13:33< Ivanovic> Polarina: i would consider it impractical since airflow is broken (airflow via the outlet in the powersupply)
20090324 22:13:33< mordante> Crab_, the error is not in the GUI code so I'll pass
20090324 22:13:57< Crab_> mordante: ok
20090324 22:14:13< mordante> if it was I would have assigned it to myself
20090324 22:14:33< mordante> I'm off night
20090324 22:14:42< Crab_> mordante: good night
20090324 22:14:54< Ivanovic> okay, then it is probably either for Dragonking or for Sirp
20090324 22:15:08< orian> mordante: good night;)
20090324 22:15:34< mordante> Ivanovic, not sure it is in some undo code
20090324 22:16:08< VladimirSlavik_> good night mordante
20090324 22:16:33-!- mordante [n=mordante@wesnoth/developer/mordante] has quit ["Leaving"]
20090324 22:17:00< VladimirSlavik_> Polarina: is that toilet paper?
20090324 22:17:24-!- Pk_JoA [n=pkjoa@190.183.107.26] has joined #wesnoth-dev
20090324 22:17:28< Polarina> VladimirSlavik_: No, it's a normal paper.
20090324 22:18:43< Pk_JoA> Good afternoon everyone :D
20090324 22:18:47< Polarina> VladimirSlavik_: What makes you think it could be a toilet paper?
20090324 22:18:52< Polarina> Pk_JoA: Good afternoon.
20090324 22:20:02< Pk_JoA> Can someone tell me somethin... I have transalated an scenario from Up From Slavery. How do I install it to see if it working? The translation, i mean. I have the .po file but I don´t know where to put it
20090324 22:20:13< Pk_JoA> (sorry, I am VERY new at this)
20090324 22:20:39< VladimirSlavik_> Polarina: well the way it is tucked int othe box makes it seem "thick inside"... I can't epress that well
20090324 22:20:48< VladimirSlavik_> Pk_JoA: you use poedit?
20090324 22:21:55< Polarina> Pk_JoA: /usr/share/locale/<locale>/
20090324 22:22:11< CIA-70> loonycyborg * r34102 /trunk/projectfiles/ (4 files in 4 dirs): Committed McShark's projectfiles update.
20090324 22:22:59< Pk_JoA> I didn´t use poedit. Just the text editor of Ubuntu but highlited for gettext
20090324 22:23:33< VladimirSlavik_> ok, you need to compile the .po file with... msgfmt I think,to .mo (or .gmo?)
20090324 22:23:42< loonycyborg> VladimirSlavik_: .mo
20090324 22:24:00< VladimirSlavik_> then put it where it belongs, which is where I could help you only if you used Win
20090324 22:24:00< Pk_JoA> thanks
20090324 22:24:06< Pk_JoA> I will do it right now
20090324 22:24:08< Pk_JoA> lets see...
20090324 22:24:56-!- Baufo [n=thomas@wesnoth/developer/baufo] has quit ["Leaving"]
20090324 22:25:12< CIA-70> ivanovic * r34103 /trunk/ (5 files in 4 dirs): updated Russian translation
20090324 22:25:24< CIA-70> ivanovic * r34104 /branches/1.6/ (5 files in 4 dirs): updated Russian translation
20090324 22:25:33< Pk_JoA> Ready. I have now an .mo file called messages
20090324 22:26:05-!- faktor4u [n=faktor@chello089077167144.chello.pl] has quit ["..."]
20090324 22:28:50< Pk_JoA> Where should I move it? 
20090324 22:29:54-!- duschfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090324 22:34:04< zookeeper> loonycyborg, right, using the new build solved the errors...i guess someone changed something in somewhere.
20090324 22:35:33-!- boucman [n=rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20090324 22:35:53< loonycyborg> Yes. There were some source changes accompanied by WML changes.
20090324 22:36:12< VladimirSlavik_> Pk_Joa: sorry I'm no Windows, I have no clue... but do you know where are installed your addons?
20090324 22:36:33< Pk_JoA> No windows either here. I am on Ubuntu 8.10
20090324 22:36:43< Pk_JoA> Nevertheless... yes, I know where is the addon
20090324 22:36:53< Pk_JoA> /home/pkjoa/.wesnoth1.6/data/campaigns/Up_From_Slavery
20090324 22:37:00< loonycyborg> Pk_JoA: Look at some translated addon, e.g. TDH and follow its example.
20090324 22:37:10-!- Sapient [n=sapien-x@c-71-204-10-166.hsd1.ga.comcast.net] has joined #wesnoth-dev
20090324 22:38:26< VladimirSlavik_> Pk_JoA: so in there translations/<your_language_code>/LC_MESSAGES/<textdomain>.mo
20090324 22:38:48< Pk_JoA> Ok...
20090324 22:39:06< VladimirSlavik_> code is for English en, German (Deutsch) de and so on
20090324 22:39:39< boucman> hey all
20090324 22:39:40< Polarina> Is there a specific channel for translations?
20090324 22:39:43< boucman> esr: around ?
20090324 22:39:49< Pk_JoA> Got it... but it is not working. Maybe I screw the code
20090324 22:39:51< Crab_> hi boucman
20090324 22:39:56< boucman> hey Crab_
20090324 22:40:02-!- cjhopman [i=cjhopman@wesnoth/developer/cjhopman] has quit [Read error: 110 (Connection timed out)]
20090324 22:40:04< esr> boucman: Yes, what's up?
20090324 22:40:42< VladimirSlavik_> Pk_Joa for textdomain look into some of the addon's cfg files, it has #textdomain something, so your file name must match that
20090324 22:40:52< boucman> esr: two days ago, you removed {{DevFeature}} tags from FormulaAIFunctions, EventWML and InterfaceActionsWML
20090324 22:40:57< zookeeper> Soliton, Sapient, it seems that a weapon special like [damage] ... active_on=defense [filter_opponent] defense=50 [/filter_opponent] [/damage] is active even on offense if the [filter_opponent] matches.
20090324 22:41:17< boucman> these features were 1.7 only IIUC, so I think they shoud be added back, but I wanted to make sure with you beforehand
20090324 22:41:33< Pk_JoA> #textdomain wesnoth-Up_From_Slavery
20090324 22:41:33< Pk_JoA> [textdomain]
20090324 22:41:33< Pk_JoA>     name="wesnoth-Up_From_Slavery"
20090324 22:41:33< Pk_JoA>     path="data/campaigns/Up_From_Slavery/translations"
20090324 22:41:33< Pk_JoA> [/textdomain]
20090324 22:41:37< esr> boucman: Yes, and a few hours later realized I may have done so in error.  If you caught those errrorsd, please fix them.
20090324 22:41:37< Sapient> zookeeper: sounds like a bug, then
20090324 22:41:38< Pk_JoA> it matches in everything
20090324 22:41:48< boucman> esr: ok, will do
20090324 22:42:11< Pk_JoA> Well, I dont bother you anymore. I will just ask Shadow_Master in a PM
20090324 22:42:32< Pk_JoA> thanks everyone aniway :)
20090324 22:42:34< Pk_JoA> see you
20090324 22:42:35-!- Pk_JoA [n=pkjoa@190.183.107.26] has left #wesnoth-dev []
20090324 22:43:37< boucman> esr: done
20090324 22:44:49< crimson_penguin> Ivanovic: by the way, remind me about that stats thing later - I'm currently visiting my uncle's family 3 time zones away, and am rather sick
20090324 22:45:26< Ivanovic> crimson_penguin: sure
20090324 22:46:32< Velory> boucman:i implemented new global function to fai which is named distance_list_between so we can take a list which is between one unit and all other units
20090324 22:46:53< Velory> boucman:i need this global function for healing improvement
20090324 22:47:17< boucman> Velory: if you think it's needed...
20090324 22:47:27< Ivanovic> boucman: saw your mail
20090324 22:47:36< Ivanovic> an i*what* should i create?
20090324 22:47:48< Velory> for example if we write distance_list_between(my_leader.loc,map(my_units,loc))
20090324 22:47:49< Ivanovic> sorry, but i got no idea what you are talking about with this
20090324 22:47:56< boucman> ical, that's the standard format to exchange agenda entries
20090324 22:48:11< Velory> we can take list lie [0,6,8,9,5]
20090324 22:48:15< Ivanovic> my standard format would be a plain page in the wiki
20090324 22:48:21< boucman> (if you use google calandar, that's what it can export, but most calandar can import/export it)
20090324 22:48:29< Ivanovic> listing stuff everybody wants to work on and time needed for it
20090324 22:51:13< Polarina> Do I have to do anything else than sending a mail to a translation maintainer to become part of a translation team?
20090324 22:51:27< Ivanovic> since we won't have any exact dates for anything anyway, this will work as well
20090324 22:51:46< Ivanovic> (or do you mean stuff like "I won't be around in this week, will be busy in that one, ..."
20090324 22:52:10< boucman> Ivanovic: my point was to have a planning apearing on my agenda, so I keep it in mind :P
20090324 22:52:35< Ivanovic> my planning is only a "feels correct this way"
20090324 22:52:36< Ivanovic> ;)
20090324 22:52:48< VladimirSlavik_> Polarina: probably nothing more afaict
20090324 22:53:10< Polarina> VladimirSlavik_: Ok. I'll simply wait.
20090324 22:53:14< Ivanovic> Polarina: this is perfectly fine
20090324 22:53:37-!- zookeeper [n=l@wesnoth/developer/zookeeper] has quit []
20090324 22:54:03< Ivanovic> the maintainer will/should handle the rest needed (as in looking that everything is fine and sending me or torangan the stuff (also saying how you shall appear in the credits)
20090324 22:54:58< Polarina> Ivanovic: Ok. Thank you.
20090324 22:57:34< Velory> boucman:should i upload that as patch? or i dont know
20090324 22:57:44< Polarina> Ivanovic: How long should I wait for a reply?
20090324 22:57:54< CIA-70> ivanovic * r34105 /trunk/po/ (5 files in 5 dirs): updated Italian translation
20090324 22:57:57< CIA-70> ivanovic * r34106 /branches/1.6/po/ (5 files in 5 dirs): updated Italian translation
20090324 22:58:08< Ivanovic> Polarina: which translation are you working on?
20090324 22:58:20< Polarina> Ivanovic: I want to work on the Esperanto translation.
20090324 22:58:21< Ivanovic> (sorry for not reading everything)
20090324 22:58:26< Ivanovic> hmm, esperanto
20090324 22:58:44< VladimirSlavik_> oh this will make Viliam happy
20090324 22:58:56< Ivanovic> the original maintainer for it was not speaking english and basically using some "trick" to generate translation files from slovak files
20090324 22:59:03< boucman> Velory: I don't really see the point, it's simple map(my_units, distance_betwee,(my_leader.loc,loc)
20090324 22:59:08< Ivanovic> the work was originally going via viliam
20090324 22:59:14< Ivanovic> that is: you should ping him about it
20090324 22:59:23< Polarina> Ivanovic: What's his IRC name?
20090324 22:59:26< boucman> I don't see the point of a new C++ function for that
20090324 22:59:32< Ivanovic> he is not in irc, only in the forums
20090324 22:59:38< Polarina> Ivanovic: Ok.
20090324 22:59:39< Velory> Oh
20090324 22:59:53< boucman> Velory: or did I misunderstand what you did ?
20090324 23:00:00-!- Taliesin_ [n=chatzill@93-96-172-175.zone4.bethere.co.uk] has joined #wesnoth-dev
20090324 23:00:15< Polarina> Ivanovic: What's his forum name?
20090324 23:00:22< Ivanovic> viliam
20090324 23:00:33< Ivanovic> that is: search for this nick, there are not many with a comparable nick
20090324 23:00:34< Ivanovic> ;)
20090324 23:00:59< Polarina> :)
20090324 23:01:25< Velory> boucman:no you are right but i didnt know we can use distance_between(my_leader.loc,loc) :) so you are right anyway i improve my skills about writing global functions :)
20090324 23:01:52< boucman> Velory: I think we can, but I wrote that from memory
20090324 23:02:07< Velory> Yes 
20090324 23:02:16< Velory> its returning same value i think
20090324 23:05:03< Velory> boucman:so can we write location values ,which is include list of locations, in distance_between function ?
20090324 23:05:13< Taliesin_> >	We know anything about Drake battlecries? The kind of stuff they'd swear on (gods names and such)?
20090324 23:05:34< CIA-70> ai0867 * r34107 /trunk/data/ (80 files in 11 dirs): s/{@/{/g on mainline WML.
20090324 23:05:49< boucman> Velory: no, but you can put a distance_between() call inside a map() to obtain a similar result
20090324 23:06:20< boucman> Taliesin_: not that I know of, drake culture hasn't been studied much yet
20090324 23:06:26< boucman> esr might have some ideas, though
20090324 23:06:29< Polarina> Ivanovic: I could also do Icelandic translation, but I would prefer Esperanto. :)
20090324 23:07:14< Ivanovic> :)
20090324 23:07:25< Taliesin_> Hmm. Fair enough, only I'm working on the story for a 7-level campaign, and I have a few battlecries worked out, but I figured I'd ask if there were any already known.
20090324 23:07:34< Polarina> Whichever there is more demand for, I can do. :)
20090324 23:09:13-!- grantwu [n=Zenneth@westquad-147025.reshall.umich.edu] has quit [Nick collision from services.]
20090324 23:09:26< CIA-70> sapient * r34108 /trunk/src/scripting/lua.cpp: make use of lua logger instead of std:cerr
20090324 23:10:06-!- grantwu [n=Zenneth@westquad-147025.reshall.umich.edu] has joined #wesnoth-dev
20090324 23:10:31< Polarina> Ivanovic: I can also do both. I got plenty of free time to spare. :)
20090324 23:10:43< Ivanovic> Polarina: better do one first in a good way
20090324 23:10:49< Ivanovic> you will see that it is *lots* of stuff
20090324 23:11:01< Polarina> Ivanovic: So I figured.
20090324 23:11:23-!- grantwu [n=Zenneth@westquad-147025.reshall.umich.edu] has quit [Nick collision from services.]
20090324 23:11:44-!- grantwu [n=Zenneth@westquad-147025.reshall.umich.edu] has joined #wesnoth-dev
20090324 23:11:49< Polarina> Icelandic is my native language.
20090324 23:12:42< CIA-70> sapient * r34109 /trunk/src/scripting/lua.cpp: typo: 'err' logstream
20090324 23:13:16< Sapient> wml=_"my native language"
20090324 23:13:19< Sapient> ;)
20090324 23:13:40< Polarina> :O
20090324 23:14:04< Polarina> <you>are weird</you>
20090324 23:14:07< Polarina> *grin*
20090324 23:16:14< Sapient> no gettext support in your joke, tsk tsk :p
20090324 23:16:49< loonycyborg> Sapient: So, you've managed to compile lua after all?
20090324 23:17:14-!- YogiHH [n=chatzill@c219113.adsl.hansenet.de] has joined #wesnoth-dev
20090324 23:17:33< Sapient> loonycyborg: no but I have managed to compile DISABLE_LUA
20090324 23:17:34< Crab_> boucman: I've been thinking about ways to test the AI performance, and after some discussions (with Velory and Dragonking ) I've put up a wiki page about the ideas - http://www.wesnoth.org/wiki/AI_Arena       basically, two things are to be done to make it work like a charm - "create one arena map with some WML events which are to be used to setup and teardown the testing", and hack up a way to hot-redeploy the entire AI from a WML fil
20090324 23:17:47< Polarina> Ivanovic: What would I have to do to start an Icelandic translation?
20090324 23:18:01< YogiHH> hello everyone
20090324 23:18:05< boucman> Crab_: yes, I saw it and skimmed through...
20090324 23:18:09< Sapient> yo Yogi
20090324 23:18:32< Ivanovic> Polarina: first provide me with some "spare time"
20090324 23:18:34< Ivanovic> ;)
20090324 23:18:37< loonycyborg> Sapient: There's a mingw compiled lua dll on files.wesnoth.org/hidden/win if you need it.
20090324 23:18:43< Crab_> boucman: what do you think ? i think it will benefit the entire AI team, since the implementation will be AI-independent.
20090324 23:18:45< boucman> Ivanovic: there are two patches assigned to you (translation stuff) btw
20090324 23:18:50< Ivanovic> Polarina: sorry, i won't be of any help now, will relax a little because of an exam tomorrow
20090324 23:19:02< Polarina> Ivanovic: Ah. Good luck with that.
20090324 23:19:08< Ivanovic> we can talk tomorrow in the afternoon, fine with you?
20090324 23:19:14< Polarina> Sure.
20090324 23:19:21< Ivanovic> boucman: wrong
20090324 23:19:31< Ivanovic> boucman: only one is "open"
20090324 23:19:33< Ivanovic> ;)
20090324 23:19:35< boucman> Crab_: I am not convinced that live-reloading of AI is that usefull, (though I could change my mind) but overall, yes, we definitely need that sort of framewor,k
20090324 23:19:45< Ivanovic> argh, no, the cjk one
20090324 23:19:49< Ivanovic> this one is most likely not for me
20090324 23:19:54< boucman> Ivanovic: :P
20090324 23:19:56< boucman> oh
20090324 23:19:57< Sapient> loonycyborg: is that 4.2.1-dw2 ?
20090324 23:20:11< Taliesin_> How vengeful are Elves, once roused to war? I get that they're (supposedly) hard to get involved in a war, but once they're involved, how much do they hold a grudge?
20090324 23:20:19< Ivanovic> should be done by someone who knows src/help.cpp
20090324 23:20:23< loonycyborg> No. 4.3.2, cross-compiled.
20090324 23:20:42< loonycyborg> dw2 is irrelevant since lua is C.
20090324 23:21:28< Sapient> ah, in that case I may be able to use the official dev libs
20090324 23:21:44< Crab_> boucman: live-reloading will also allow you to do things like "AI #25, play 2 turns for me, please", or "AI #5, suggest candidate moves for my unit at (13,12)" So you could, at any moment of SP game, yield some control to the AI and see how it turns out. You could even save beforehand and try different AI's :)
20090324 23:22:11< loonycyborg> Sapient: They seem to depend on msvcr80.dll
20090324 23:22:21< Sapient> yuck
20090324 23:22:32< boucman> Crab_: well, then AI reloading is not specific to the test case and should be considered a different project
20090324 23:23:08< Ivanovic> afk now and then off to bed
20090324 23:24:23< Taliesin_> How vengeful are Elves, once roused to war? I get that they're (supposedly) hard to get involved in a war, but once they're involved, how much do they hold a grudge? I'm wondering what it would take to make them leave their forests to seek vengeance.
20090324 23:25:32< Sapient> AI0867: why did you remove the ~units, ~multiplayer and ~ais directories?
20090324 23:25:57< boucman> Taliesin_: the answer would be "as much as you need them to be for your scenario"
20090324 23:26:01-!- grantwu1 [n=grant@67-194-9-219.wireless.umnet.umich.edu] has joined #wesnoth-dev
20090324 23:26:13< Taliesin_> Ah yes. The Power of Plot.
20090324 23:27:29< esr> Taliesin_: I'd say it depends.  Are you chopping down trees?  Yes.  Have you injured a major Elven lineage?  Yes.  As a general thing, maybe not.  You don't hegenerally bother being bitterly vengeful against beings you think are inferior.
20090324 23:29:23< Taliesin_> esr_: How about threatening their forests, leading (even if accidentally, but they don't know that) a major undead force to their doorstep, and then fleeing from said Undead, making an escape rout with swords and fire?
20090324 23:29:24-!- Rrenys [n=rrenys@81-20-159-197.levira.ee] has quit [Read error: 104 (Connection reset by peer)]
20090324 23:29:36< Taliesin_> route*
20090324 23:30:07< Taliesin_> The elves in question would be the residents of Glyn's Forest, if that helps.
20090324 23:30:27< AI0867> Sapient: because they're completely unused?
20090324 23:30:46< esr> Yeah, that'd piss them off.  Burning trees, definitely a no-no.
20090324 23:31:11< Crab_> boucman: yes, AI reloading is a separate project (quite small project, I must say, for it could be hacked pretty easily, since all AI's conform to the same interface and only need some parameters to create, and syntax for those parameters is already known - http://www.wesnoth.org/wiki/AiWML - nothing to invent). But it is needed to make testing work using just a single ai-independent map which will contain all 'test challenges'. For, a
20090324 23:31:30< Sapient> AI0867: the reason they are there is so the users know where to put such things
20090324 23:31:39< Sapient> AI0867: otherwise how are they to know?
20090324 23:32:26< AI0867> ~ais serves no purpose and neither does ~units
20090324 23:32:54< Sapient> by '~' do you mean user directory or game directory?
20090324 23:33:12< AI0867> ~multiplayer can be useful for maps, so that maybe that's worth putting back
20090324 23:33:21-!- kitty_ [n=kitty@wesnoth/artist/kitty] has quit ["bye"]
20090324 23:33:26< AI0867> by ~... I mean userdata
20090324 23:33:38< Sapient> then yes, they are used
20090324 23:33:50< Sapient> *when* the user puts their custom ai into it
20090324 23:34:22< AI0867> 'ais' does not work, mainline has long changed to ai/{formula,python}
20090324 23:34:35< Sapient> ok, then it needs to be updated, not removed
20090324 23:34:58< AI0867> except that python has been removed and formulaAI cannot be used without WML to begin with
20090324 23:35:01< Sapient> we cannot assume that users will have write permission to the game install directory
20090324 23:35:30< AI0867> so you *need* an add-on to even use it
20090324 23:35:56< Sapient> or a hand written multiplayer scenario
20090324 23:36:11< Sapient> ... in ~multiplayer
20090324 23:36:48< AI0867> I can remember quite clearly my frustration with moving everything around *again* when I figured out I had to move everything to ~data/campaigns in order to distribute
20090324 23:37:33< Sapient> having an empty directories is far less burdensome than explaining where to put stuff over and over
20090324 23:37:50< AI0867> but if you think it's a bad idea, I'll revert it and we can discuss it first
20090324 23:37:55< Sapient> it is not uncommon to download single MP scenarios from the forum
20090324 23:38:05< Sapient> add-ons are not the end-all-be-all
20090324 23:38:44< Sapient> see here: http://www.wesnoth.org/forum/viewtopic.php?p=125704#p125704
20090324 23:39:27< Sapient> looks like my post is a bit out-dated / wrong
20090324 23:39:44< Sapient> I see it was edited by Turuk
20090324 23:40:42< Sapient> what the heck.... he removed all the lines that said "In Linux, save to "
20090324 23:44:11< AI0867> Sapient: according to your post, everything should go into userdata/editor or userdata/data/maps
20090324 23:44:16< VladimirSlavik_> bye...
20090324 23:44:18-!- VladimirSlavik_ [n=chatzill@88.103.147.178] has left #wesnoth-dev []
20090324 23:44:35< Sapient> AI0867: my post has been highly edited by persons other than myself
20090324 23:44:47< Sapient> it's probably wrong
20090324 23:46:06< Sapient> also, people have extensively changed the directory structure of wesnoth since that time
20090324 23:46:45< Sapient> data\scenarios\multiplayer was changed to data\multiplayer\scenarios, for example
20090324 23:47:14< AI0867> but I don't see where ~data/multiplayer is included from
20090324 23:47:53< Sapient> lines such as map_data="{multiplayer/maps/2p_Cynsaun_Battlefield.map}"
20090324 23:48:29< Sapient> and probably elsewhere...
20090324 23:48:35< AI0867> /editor/maps seems to be the only thing included from the C++
20090324 23:48:47< AI0867> and the WML doesn't include *anything* since I stripped out {~units}
20090324 23:49:02< AI0867> and that's just does plain standard maps
20090324 23:49:08< AI0867> s/'s//
20090324 23:49:45< Sapient> here's the main point
20090324 23:50:08< Sapient> if I download a cfg and a .map from the forum, there should be a logical place to put both
20090324 23:50:36< AI0867> I'm trying to find atm if there *is* a place to put the .cfg
20090324 23:50:40< AI0867> because there doesn't seem to be
20090324 23:50:50< Sapient> so that everyone's map_data="{~multiplayer/maps/...  will work
20090324 23:51:21< boucman> night all
20090324 23:51:26-!- boucman [n=rosen@wesnoth/developer/boucman] has quit [Remote closed the connection]
20090324 23:52:18< Sapient> likewise if I download a custom ai frfom the forum
20090324 23:52:41< Sapient> there should be an obvious directory to drop it in that doesn't require admin rights
20090324 23:52:52< AI0867> well, it doesn't seem to work
20090324 23:53:07< Sapient> it used to...
20090324 23:53:24< Sapient> I guess our file structure gods have abandoned us
20090324 23:53:43< AI0867> did it work in 1.4?
20090324 23:54:03< Sapient> did 1.4 have the core/ ?
20090324 23:54:12< Sapient> if so, then no
20090324 23:54:12< AI0867> I believe we did
20090324 23:54:46< AI0867> http://svn.gna.org/viewcvs/wesnoth/tags/1.4/data/multiplayer/_main.cfg?rev=24221&view=auto
20090324 23:54:51< AI0867> {~scenarios/multiplayer/}
20090324 23:54:55< AI0867> we lost that line
20090324 23:55:47 * Sapient subtly points at that guy whose name starts with E and S ;)
20090324 23:55:53-!- BenUrban [n=benurban@c-68-49-10-243.hsd1.md.comcast.net] has joined #wesnoth-dev
20090324 23:58:34< Sapient> AI0867: maybe add it to easycoding page? I dunno
--- Log closed Wed Mar 25 00:00:34 2009