--- Log opened Wed Aug 12 00:00:32 2009
--- Day changed Wed Aug 12 2009
20090812 00:00:32< fendrin> thespaceinvader: Yes, zookeeper is doing them. Is there any special filepath they belong in? (Maybe one that suggests the usage with the rouser image)
20090812 00:00:42< Crab_> deekay: old code looked like this: 	if (formula_ai_->do_recruitment()) { LOG_AI << "Recruitment done by formula_ai\n"; return true;	}}
20090812 00:00:55< Crab_> note the code always returning true
20090812 00:01:22< deekay> Not only when do_recruitment() returned true?
20090812 00:01:37-!- Valkier [n=karasawa@c-71-225-240-156.hsd1.pa.comcast.net] has quit []
20090812 00:01:38< Crab_> deekay: yes
20090812 00:01:47< Crab_> deekay: so, old fallback to fai worked like this
20090812 00:02:20< thespaceinvader> fendrin: i would guess that they go in sounds with all the other sounds
20090812 00:02:23< Crab_> deekay: note that there's a code path in ai_default which checks the return value of do_recruitment
20090812 00:02:36< thespaceinvader> bearing in mind that once they're in the game, they could be used by anyone, not necessarily just your goblin
20090812 00:03:00< Crab_> deekay: any ideas about 'how to easily check for fai recruitment' ?
20090812 00:03:16< Crab_> deekay: I can check fai 'made move'.
20090812 00:03:48< deekay> do_recruitmet from formula/ai.cpp should only return true if valid action was done by fai
20090812 00:04:45< Crab_> deekay: ok. in formula_ai::evaluate, I see :  'if (  !var.is_empty() ) { return "Made move: " + var.to_debug_string();   }' - is this a good enough check ?
20090812 00:06:16< deekay> Crab_: if execute_variant returns empty vector, then no valid move was done
20090812 00:06:26< Crab_> deekay: ok.
20090812 00:06:32< fendrin> thespaceinvader: That is true. I will find a place for them :-)
20090812 00:06:45< Crab_> deekay: another question - is stage_side_formulas good enough for your recruit formula ?
20090812 00:07:13< deekay> Does it work like old move= - as work in "loop" ?
20090812 00:07:34< Crab_> deekay: (if I make default_ai call it when it needs to recruit...)
20090812 00:07:45< Crab_> deekay: it has  "  while( !fai_.make_action(move_formula_,callable).is_empty() ) { } "
20090812 00:07:55< deekay> Sure, it's perfect. :)
20090812 00:09:05< Crab_> deekay: ok. so, it will work like this: I will allow to define a stage which will be called by ai_default when it needs to recruit
20090812 00:09:14-!- wesbot changed the topic of #wesnoth-dev to: 70 bugs, 239 feature requests, 12 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090812 00:09:23< deekay> Crab_: Sounds perfect. :)
20090812 00:09:27< Crab_> deekay: so, if you make it a  'stage_side_formulas'
20090812 00:09:41-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 60 (Operation timed out)]
20090812 00:10:01< Crab_> deekay: btw, *any* stage can be used in this manner (as long as it properly returns true only if gamestate has changed)
20090812 00:10:22-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090812 00:10:41< Crab_> deekay: also, it will be quite easy to extract another aspect of ai_default's behaviour in the same manner
20090812 00:11:11< Crab_> deekay: so, if you think about some interesting way to abuse it, tell me about it
20090812 00:11:33< deekay> Crab_: Sounds neat, I'll keep that in mind. :)
20090812 00:12:02< Crab_> deekay: for ai_default's do_recruitment will look literally like:   bool ai_default::do_recruitment() {  return get_recruitment()->play_stage(); }  
20090812 00:12:34< deekay> I do like this level of abstraction that you made there. :)
20090812 00:12:48< Crab_> any by using ai batch testing framework, we can test the impact of any 'replacement' on ai-vs-ai efficiecy
20090812 00:13:02< deekay> Great
20090812 00:13:23< deekay> I doubt that impact of my recruitment will be really big, but I wonder how it will look like with it
20090812 00:13:34< Crab_> deekay: well, we'll see :)
20090812 00:13:59< deekay> Crab_: I'm going to add function to formula AI that calculates what part of map AI is supposed to own based on movement points of a possible recruits
20090812 00:14:09-!- Chusslove [n=Chusslov@adsl-ppp-3181.yubc.net] has quit [Read error: 54 (Connection reset by peer)]
20090812 00:14:18< Crab_> deekay: yes, I've seen those your maps.
20090812 00:14:47< deekay> I'm not exactly sure how it will turn out.... but I tihnk it is good to give it a try
20090812 00:14:48< Crab_> deekay: btw, there's a strange bug which was reported by euschn
20090812 00:14:53< deekay> Yes?
20090812 00:15:09< Crab_> deekay: I even managed to reproduce it with his save: http://wesnoth.pastebin.com/m40454da0
20090812 00:15:45< Crab_> deekay: it is a 'random segfault with no how-to-reproduce'
20090812 00:16:09< deekay> Yea... random segfaults are the best...
20090812 00:16:14< Crab_> deekay: it is caused by events::generic_event::notify_observers dereferencing an invalid non-null pointer
20090812 00:16:39< Crab_> deekay: that should be the pointer to one of the observers
20090812 00:17:19< deekay> hm
20090812 00:17:52< Crab_> deekay: note that each observer should register during construction, and unregister during destruction
20090812 00:18:08< deekay> Yes
20090812 00:18:23< Crab_> so, if any of observers goes out of scope, it should be connected with a call to 'unregister' method
20090812 00:18:24< deekay> I noticed it when adding the observer for formula_ai
20090812 00:18:29< Crab_> and the big question is: WTF?
20090812 00:19:05< deekay> Is there possibility that it is multithreading issue?
20090812 00:19:19< Crab_> deekay: that part of wesnoth is singlethreaded
20090812 00:19:30< deekay> wtf indeed :s
20090812 00:20:41< Crab_> deekay: I know of a way to debug it (basically, by including a 'name' of an observer or printing its address, we can find out the exact observer which causes this)
20090812 00:20:41< deekay> Crab_: You sure all observers unregister in their destuctors?
20090812 00:22:31< deekay> And you said it happens randomly, yes?
20090812 00:22:57< Crab_> deekay: http://wesnoth.pastebin.com/m72d85c6
20090812 00:23:56< Crab_> (plus playsingle_controller.cpp, which is fine, too)
20090812 00:24:02< Crab_> yes, it happens randomly
20090812 00:24:21< Crab_> see  irclog for 11 august
20090812 00:24:56< Crab_> (12:37:46 PM) euschn: Crab_: weird, i still get segfaults. still havent found a better way to reproduce than just play, you could try my start of scenario save: http://stud3.tuwien.ac.at/~e0204097/wesnoth/EI-Weldyn_under_Attack
20090812 00:25:23< Crab_> (10:57:53 AM) euschn: Crab_: when playing EI - Weldyn under attack, im getting "error ai/actions: Return value of AI ACTION was not checked. This may cause bugs!" during ai turns, followed by a segfault. I havent managed to reproduce it yet, but it happens fairly often. any ideas what might cause this?
20090812 00:25:42< Crab_> I reproduced once from his save
20090812 00:25:47< Crab_> around turn 9
20090812 00:25:47< deekay> Hm... did he cnounter it while plaing from the beginning or after loading a save only? Dunno if it makes any difference
20090812 00:26:04< Crab_> I played for 20 turns from :cl, no bug was found
20090812 00:29:35< deekay> wierd indeed... code looks good
20090812 00:30:29< Crab_> deekay: so, I'm just making sure that you know about it. I'll add some debug later, to see if some new info resurfaces, 
20090812 00:30:44< deekay> ok
20090812 00:30:49< deekay> Thanks for the info
20090812 00:38:58< Crab_> deekay: here's how the 'recruitment' config snippet will look like: http://wesnoth.pastebin.com/m3d7a3686
20090812 00:40:25< Crab_> deekay: with some small changes. for example, you will probably want it to not be invalidated on 'new turn' event (to avoid repeating calculations)
20090812 00:42:05< deekay> Crab_: What do you mean by invalidating on a 'new turn' ?
20090812 00:42:20< Crab_> deekay: aspects have defined invalidation rules
20090812 00:42:35< deekay> Are ai variables lost between turns?
20090812 00:42:36< Crab_> deekay: for example, the list of attacks is invalidated after each move/recruit 
20090812 00:43:09< Crab_> deekay: ai variables are part of the formula ai 
20090812 00:43:21< Crab_> deekay: formula_ai is wrapped by fai engine
20090812 00:43:57< Crab_> deekay: so, there is exactly 1 formula_ai per side, per game. it is created on first access to fai engine, and destroyed on game end
20090812 00:44:05< deekay> ok
20090812 00:44:08< deekay> Sounds good
20090812 00:44:13< Crab_> deekay: so, if you store state inside fai variables, they will persist
20090812 00:44:23< deekay> Good, that's what I need :)
20090812 00:45:21< Crab_> just be sure to not hit the set_var-in-a-row limit :)
20090812 00:45:40-!- ABCD_ is now known as ABCD
20090812 00:45:43< deekay> hehe
20090812 00:45:53< deekay> No worries about that :)
20090812 00:50:26-!- kitty_ [n=kitty@wesnoth/artist/kitty] has quit ["good night"]
20090812 00:50:38< shadowmaster> everything. is. timing. out. on. me. :(
20090812 00:50:47-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has quit ["night all"]
20090812 00:50:50< shadowmaster> I hate this wifi network
20090812 00:53:34< Crab_> deekay: ok, I'll finish tomorrow in the morning (All I have to do is to extract ai_default_recruitment_stage to a separate file, plus test & document)
20090812 00:53:56< shadowmaster> zookeeper yes I did
20090812 00:54:25-!- Crab_ [i=crab@wesnoth/developer/crab] has quit ["Leaving."]
20090812 00:54:41< shadowmaster> but I guess fendrin already handled it.
20090812 00:55:55-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has quit [Remote closed the connection]
20090812 01:02:52< shadowmaster> he's a wizard because you can throw any WML question at him, he'll know the answer provided that he's ina good mood and you use proper indentation.
20090812 01:05:38-!- shikadibot is now known as CaptainTimeout
20090812 01:07:24-!- CaptainTimeout is now known as shikadibot
20090812 01:16:02< shadowmaster> AI0867: ^ that was my response to something in #wesnoth. A SSH lag tricked me into writing it to the wrong channel
20090812 01:16:11< shadowmaster> ouch, Aethaeryn . That wasn't mean for AI
20090812 01:16:46< Aethaeryn> ?
20090812 01:17:01< shadowmaster> huh, never mind.
20090812 01:22:45-!- boucman [i=43fa2d5d@wesnoth/developer/boucman] has joined #wesnoth-dev
20090812 01:22:53< boucman> hey all
20090812 01:22:57< deekay> hi boucman 
20090812 01:22:59< shadowmaster> hi boucman 
20090812 01:24:44-!- Zarel [i=Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20090812 01:24:47< Zarel> Hey, guys.
20090812 01:25:04< deekay> hi
20090812 01:25:07< Zarel> I just noticed http://www.wesnoth.org/forum/viewtopic.php?t=9500
20090812 01:25:36< Zarel> But it's in Dev (restricted), so I can't reply to it, so I figure I might as well reply in this channel.
20090812 01:26:16< Zarel> The protocol described in the second and third posts are vulnerable to hash collision attacks.
20090812 01:26:54< shadowmaster> that topic is pretty old, year 2006.
20090812 01:27:24< Zarel> I think I found it as a link from wiki/NotSoEasyCoding. :P
20090812 01:27:40< Zarel> It appears it was never implemented.
20090812 01:27:53< Soliton> there should be something about that in ilor's gsoc wiki page.
20090812 01:28:48< Zarel> Don't see it. http://www.wesnoth.org/wiki/User:Ilor
20090812 01:29:09< Zarel> Oh, nevermind
20090812 01:31:12-!- BenUrban [n=benurban@c-68-50-54-86.hsd1.md.comcast.net] has joined #wesnoth-dev
20090812 01:35:18< Sirp> Zarel: sure, but lots of secure protocols are vulnerable to varying degrees to hash collision attacks. Generally hash collision attacks are pretty hard to implement though, due to the large amount of processing power needed.
20090812 01:35:57< Zarel> Ah, well, it's not a big deal. Ilor's GSoC page just uses a TTP solution, which is probably best.
20090812 01:36:01< Sirp> signed messages are also vulnerable to hash collision attacks.
20090812 01:36:22-!- ancestral [n=ancestra@97-116-126-165.mpls.qwest.net] has quit []
20090812 01:37:28< Sirp> I mean I'd be pretty damn surprised if a serious cheater ever ended up using a single hash collision attack to break such a protocol in a Wesnoth game, if we were ever to implement such a system.
20090812 01:42:25< Zarel> I think generatic a generic MD5 collision would only take a few seconds.
20090812 01:43:02< Zarel> But Google is refusing to give me benchmarks. :|
20090812 01:44:40< corn> http://lookit.typepad.com/lookit/2006/03/creating_collis.html
20090812 01:45:04< Zarel> That's from 2006.
20090812 01:45:12< Zarel> MD5 collisions have sped up since then. :P
20090812 01:46:02< Sirp> there are newer algorithms than MD5...
20090812 01:46:14< Sirp> I don't think anyone uses MD5 anymore for resistance against malicious collisions.
20090812 01:50:08< Sirp> SHA-1 is easy to implement and there haven't been any collisions found for it...
20090812 01:50:54< Sirp> and when collisions are found for it, I'm pretty sure that trying to cheat in a game of Wesnoth won't be one of the first applications. :)
20090812 01:51:53-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Connection reset by peer]
20090812 01:52:08-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090812 01:55:05< CIA-62> cornmander * r37688 /website/stats.wesnoth.org/wesstats/ (controllers/killview.py templates/killview.html): 
20090812 01:55:05< CIA-62> Added date and level range selection to the killmaps html template.
20090812 01:55:05< CIA-62> Initial work on filter support for killmaps.
20090812 01:57:00< shadowmaster> I wish phpBB allowed me to view some "most recently registered users" listing
20090812 01:57:35< shadowmaster> SQL-fu would help me if I knew SQL.
20090812 01:57:38< corn> shadowmaster: do you have access to the mysql tables that phpbb runs on? if user registration data is stored
20090812 01:57:41< corn> oh
20090812 01:57:42< corn> I can help you :)
20090812 01:58:03< shadowmaster> I'd need to check the last n entries of the phpbb3_users table
20090812 01:58:24< corn> what is the name of the column where dates are stored?
20090812 01:59:05< shadowmaster> I don't think that matters a lot. Each user gets an user_id. The highest user_id should be the last registered user
20090812 01:59:09< corn> ok
20090812 02:00:18< corn> SELECT * FROM `phpbb3_users` ORDER BY user_id DESC LIMIT 0,100
20090812 02:01:35< corn> change 100 to whatever you want the limit of entries to be
20090812 02:02:59-!- loonycyborg [n=sergey@wesnoth/developer/loonycyborg] has quit ["Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz"]
20090812 02:03:27-!- loonybot [n=loonybot@wesnoth/bot/loonybot] has quit [Remote closed the connection]
20090812 02:04:19-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has quit ["On the road again"]
20090812 02:04:25< shadowmaster> corn: awesome, tanks
20090812 02:04:28< shadowmaster> *thanks
20090812 02:08:07< shadowmaster> I have suddently realized that this is like killing a bug with a nuclear bomb however.
20090812 02:08:34< shadowmaster> I have remembered that it's possible to sort the members list (which is viewable by anyone) to accomplish the same goal from my browser
20090812 02:10:50< Zarel> Sirp: I think there are collision attacks for SHA-1 by now.
20090812 02:11:37< Zarel> Either way, collision resistance is usually the first thing broken in a new hash.
20090812 02:12:09< corn> shadowmaster: ok
20090812 02:12:42< Sirp> Zarel: per wikipedia there have only been algorithms produced which can find collisions, no collisions have been found yet.
20090812 02:13:24< Zarel> Yeah, it's been weakened.
20090812 02:13:29< Zarel> Which means it'll probably be broken soon.
20090812 02:14:02-!- boucman [i=43fa2d5d@wesnoth/developer/boucman] has quit ["Page closed"]
20090812 02:14:42< Sirp> perhaps. But finding some collisions is much different from being able to do so consistently. In any case, it would take a very determined attacker to break Wesnoth, considering Wesnoth is just a fun casual game
20090812 02:15:41< Sirp> many people use SHA-1 for things a lot more important than Wesnoth.....I wouldn't be concerned. If we wanted to use such a method we could easily enough, and simply update our hashing algorithm every few years or whenever one gets broken.
20090812 02:17:05< Zarel> True, but I'm a developer for Warzone 2100, and cheaters are already hacking our game.
20090812 02:17:18< Zarel> And it's not even as popular as Wesnoth. :[
20090812 02:17:47< Zarel> Sure, the cheaters won't be the ones finding SHA-1 attacks, but once one is found, they'll copy+paste it over into a Wesnoth cheat engine, just you wait. <_<
20090812 02:18:09< shadowmaster> cheating script kiddies.
20090812 02:18:19< CIA-62> shadowmaster * r37689 /website/forum.wesnoth.org/phpbb-3.0.5/ (registration_check.patch registration_check.patch.README): Add cycholka's phpBB registration check hack to the repository.
20090812 02:18:43< Sirp> Zarel: ummm we've had multiplayer going for years and we haven't had that many problems with cheaters.
20090812 02:18:57< Sirp> and almost all cheaters use very low tech solutions.
20090812 02:19:01< Zarel> Your protocols are more secure.
20090812 02:19:09< Sirp> huh? our protocols are incredibly insecure
20090812 02:19:24< shadowmaster> hah :P
20090812 02:19:31< Zarel> Sirp: y
20090812 02:19:33< Zarel> I mean
20090812 02:19:41< Zarel> http://forums.wz2100.net/viewtopic.php?p=31520#p31520
20090812 02:19:56< Zarel> They're writing into RAM addresses!
20090812 02:20:12< Zarel> How come _we_ get so many more cheaters than you do? D:
20090812 02:20:15-!- Aethaeryn [n=Michael@wesnoth/umc-dev/developer/aethaeryn] has quit ["Reconnecting"]
20090812 02:20:29< shadowmaster> perhaps it has something to do with the audience
20090812 02:20:30-!- Aethaeryn [n=Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #Wesnoth-dev
20090812 02:20:41< Zarel> Yeah. We get lots more Russians. :P
20090812 02:21:33< Sirp> Zarel: probably our management of the community. We encourage a non-competitive friendly style of play
20090812 02:21:54< Zarel> Who doesn't? o_O
20090812 02:21:54< Sirp> why'd they need to write to memory addresses if the source code is available, though?
20090812 02:22:08< Sirp> most games don't. Most games have official ladders/rankings/etc
20090812 02:22:12< Zarel> Sirp: Don't ask me. :/
20090812 02:22:38< Zarel> Not only is our source code available, but it's a lot easier to cheat just by modding the game data files.
20090812 02:24:47-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090812 02:24:58< CIA-62> shadowmaster * r37690 /website/forum.wesnoth.org/phpbb-3.0.5/ (quota_overrides.patch quota_overrides.patch.README): 
20090812 02:24:58< CIA-62> Add cycholka's hack for overriding PM limits and attachment quotas
20090812 02:24:58< CIA-62> on the Art Director user and hidden group, respectively.
20090812 02:24:58< CIA-62> This patch is in serious need of a cleanup.
20090812 02:25:51< Zarel> Why not just remove the quotas? Are you low on disk space? o_O
20090812 02:26:02< Soliton> cheating by changing game data files is very very easy in wesnoth. but it's also easily detected by the game.
20090812 02:26:34< shadowmaster> Zarel: it may be abused by regular users.
20090812 02:26:41< shadowmaster> we just won't run that risk.
20090812 02:26:49< Soliton> what kind of cheating do you get? are there ranked games or they just do it for fun?
20090812 02:26:57< Zarel> They just do it for fun.
20090812 02:27:03< Zarel> There's no ranking at all...
20090812 02:27:18< Zarel> I mean, sure, each player has a rank.
20090812 02:27:20< Soliton> is mp peer or central server based?
20090812 02:27:40< Zarel> But the rank is a plaintext file. You can just edit it and change your wins and losses to whatever you want it to be.
20090812 02:28:07< Zarel> There's a central server that lists hosted games, but everything else is peer-to-peer, which makes cheating really easy.
20090812 02:28:35< corn> people should be free to do whatever they want for peer-to-peer games
20090812 02:28:41< corn> as in, games not listed through the central server
20090812 02:28:44< Soliton> ah, yeah.
20090812 02:29:11< corn> for games that are listed on a central server, why can't you just ban malicious users?
20090812 02:29:22< Zarel> Ban them by IP?
20090812 02:29:22< Soliton> do people regularly edit those text files with win/loss? :-)
20090812 02:29:50< Zarel> We don't even know who the malicious users are. We just have a few forum threads and bug tracker issues listing usernames of some people who cheat.
20090812 02:29:52-!- xchat_sux [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 60 (Operation timed out)]
20090812 02:29:58< Zarel> Soliton: As far as I know, no.
20090812 02:31:18< Zarel> corn: People should be free to do whatever they want for peer-to-peer games, as far as other players agree to it.
20090812 02:31:27< Zarel> Generally, cheaters don't ask other players "Hey, can I cheat?" first.
20090812 02:31:56< Zarel> Also, when I mean "listed in a central server", I mean all the server does is help users set up peer-to-peer games.
20090812 02:32:21< Zarel> It's basically just a list of IPs and the games hosted on those IPs.
20090812 02:32:38< Soliton> implement a dedicated "Hey, can I cheat?" button? :->
20090812 02:32:57< Soliton> s/can/may/
20090812 02:33:54< Zarel> Soliton: There you go again, inventing your own standards. This is what RFC 3514 is for. ;)
20090812 02:34:02< Soliton> what cheats are those anyway?
20090812 02:34:16< Soliton> i'd guess you get more cheaters since it's a different genre?
20090812 02:35:09< Zarel> afaik, they're mostly money cheats. That's the only way people know they exist - players start games, then their allies suddenly gift them 10x the amount of money they start with.
20090812 02:36:26< Soliton> that should be easy to detect?
20090812 02:36:43< Soliton> the games goes out of sync in such a case in wesnoth.
20090812 02:37:00< Soliton> since the game knows how much money the other pplayers should have.
20090812 02:37:21< Soliton> well, it just goes out of sync if the excess money is actually spent.
20090812 02:37:23< Zarel> In Warzone, they just send a sync request, and the other players just update their data with the new power levels.
20090812 02:37:34< Soliton> ohh, ok.
20090812 02:37:39< Zarel> We never expected to have to authenticate syncs.
20090812 02:37:55< Soliton> that makes it indeed trivial then.
20090812 02:38:25< Sirp> Zarel: in any case I would be profusely surprised if anyone cheated on Wesnoth through exploiting a SHA-1 collision anytime in the next ten yearas.
20090812 02:38:35< Zarel> That's one difference between turn-based and real-time. In turn-based, it's a lot more obvious if the game is out of sync because of cheating.
20090812 02:38:55< Soliton> well, rng cheating can be very hard to detect.
20090812 02:39:12< Soliton> though that's not exactly turn-based specific.
20090812 02:39:27< Zarel> What, even the current engine doesn't have both players agree on a seed beforehand?
20090812 02:39:28< deekay> night everyone
20090812 02:39:39< Sirp> Zarel: depends on how you design your multiplayer model. On another game I develop, Frogatto, it is real time and we're using something similiar to what Wesnoth does for multiplayer.
20090812 02:39:45< Sirp> and if anyone modifies their data file, things will go out of sync.
20090812 02:39:52< Soliton> the attacker generates it afaik.
20090812 02:40:15< Soliton> but ilor is working on a server side rng.
20090812 02:40:47< Zarel> Sirp: Yeah. We're thinking of moving Warzone over to something more cheat-proof, but it requires an overhaul to the netcode that no one wants to do.
20090812 02:41:28< shadowmaster> Sirp: Frogatto will have MP?
20090812 02:41:37< Soliton> s/will have/has/
20090812 02:41:50< deekay> s/?/!/
20090812 02:42:08< shadowmaster> ohhhh. I really need to get opengl support for any of my computers' onboard chipsets ;(
20090812 02:42:12< Zarel> I'll give some idea of how bad it is: AI/UI code is sprinkled with "TurnOffMultiMsg" because otherwise duplicate instructions would be sent.
20090812 02:43:55< Soliton> try to get into gsoc? :-)
20090812 02:44:04< Zarel> We're planning on it. :P
20090812 02:44:20< Zarel> Not all of us have the publicity Wesnoth does, you know. ;)
20090812 02:44:51< Zarel> And you're lucky you don't have a 3D engine. 90% of our bug reports have to do with bad OpenGL drivers.
20090812 02:45:24< Soliton> yeah, i can imagine that's very annoying.
20090812 02:46:12-!- Netsplit clarke.freenode.net <-> irc.freenode.net quits: wesbot, AnMaster
20090812 02:48:56-!- Netsplit clarke.freenode.net <-> irc.freenode.net quits: CIA-62
20090812 02:49:37-!- Netsplit over, joins: wesbot, AnMaster
20090812 02:49:56-!- Netsplit over, joins: CIA-62
20090812 02:51:34-!- noy [n=Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090812 03:01:20-!- Netsplit clarke.freenode.net <-> irc.freenode.net quits: CIA-62
20090812 03:02:51-!- Netsplit over, joins: CIA-62
20090812 03:11:14< Sirp> Zarel: the other thing is that we would always generate hashes from 64 bit keys. I think it's quite likely -- though I don't know the rigorous mathematical details -- that there *are* no collisions in SHA-1, or perhaps even MD5, if all your keys are 64 bit integers.
20090812 03:11:42< corn> Sirp: there will always be collisions by the pidgeon hole principle
20090812 03:11:59< Zarel> Sirp: Oh, that's true.
20090812 03:12:07< corn> you cannot create a unique hash of all 65 bit integers using 64 bits
20090812 03:12:11< Zarel> There'll be collisions, but they'll be a lot rarer and less exploitable.
20090812 03:13:34< Sirp> corn: but SHA-1 is 160 bits, and we only have 64 bit keys
20090812 03:14:04< corn> Sirp: ah
20090812 03:14:12< Sirp> Zarel: there won't necessarily be collisions. The other side of the pidgeon hole principle is that if you have *more* holes than pidgeons then there isn't necessarily a collision.
20090812 03:14:12< Zarel> Oh, hmm, I think that makes the existence of collisions nearly impossible.
20090812 03:14:22< corn> in that case, it *may* be true that there will be no collisions, but it is dependent on what hashing algorithm you use
20090812 03:14:28< Sirp> corn: yes, correct.
20090812 03:14:35< Sirp> and in the case of SHA-1, I doubt anyone knows for sure.
20090812 03:15:02< Zarel> Sirp: Of course, the birthday paradox means that the existence of at least _one_ collision is pretty likely, given a uniformly random distribution (or the equivalent).
20090812 03:15:11< corn> however... using a hashing algorithm that produces a hash larger than the data you are hashing defeats the purpose
20090812 03:15:22< Sirp> Zarel: actually I'm not sure about that either.
20090812 03:15:27< Zarel> corn: Not if it's for use in an RNG.
20090812 03:15:42< Zarel> Sirp: Lemme do the math.
20090812 03:15:53< Sirp> Zarel: remember there are 365 days in a year, and it takes around 21 (I forget exactly) people to have a 50% chance of a collision.
20090812 03:16:03< Sirp> 365 is only around 18x larger than 21
20090812 03:16:14< Sirp> 2^160 >>> 2^64
20090812 03:16:22-!- karl_m1 [n=karlm@220-253-90-228.QLD.netspace.net.au] has quit ["Leaving."]
20090812 03:17:18< Zarel> (2^160)!/(2^(160*64)(2^160-2^64)!)
20090812 03:17:23< Zarel> You're right, it's negligible.
20090812 03:17:38< Zarel> lol, Google refuses to calculate that number.
20090812 03:17:56< Zarel> Any of you have Mathematica on you? ;)
20090812 03:20:01< corn> you could use wolfram alpha, but it is having trouble with your input verbatim
20090812 03:20:18< Zarel> Sirp: That's assuming it's linear, though.
20090812 03:20:41< shadowmaster> Zarel: I don't think so.
20090812 03:20:53< shadowmaster> unless someone has already figured out how to run computer programs on biological brains.
20090812 03:20:54< Zarel> For the probability of a collision to be around 50%, you need sqrt(hashspace), which is 2^80.
20090812 03:21:49< Sirp> I don't think Wolfram Alpha calculates that number either. It has limits too
20090812 03:22:21< shadowmaster> I do have maple though :P
20090812 03:22:43< Zarel> That, and the way I wrote it had an implicit multiplication.
20090812 03:24:30< Zarel> ((2^160)!)/((2^(160*64))*((2^160)-(2^64))!)
20090812 03:24:39< Zarel> Here it is, written in a way most CASes should understand.
20090812 03:25:32< Zarel> Either way, it's approximately 1 in 2^224.
20090812 03:25:35< Zarel> i.e. negligible.
20090812 03:27:13< Sirp> I think the bigger risk would be that if you sent someone a hash they would be able to do a space search and find the number you used to generate the hash out of all the 64 bit keys. However I think that would be very unlikely with current processing power.
20090812 03:28:46< Zarel> Well, it might be possible if the hash algorithm were weakened.
20090812 03:28:52< Zarel> Such as, say, SHA-1.
20090812 03:28:55 * Zarel coughs
20090812 03:29:15< Zarel> But still, that one require someone to run a crypto attack on Wesnoth specifically, and we're reasonably safe from _that_. ;)
20090812 03:29:50< Zarel> Maybe the focus of my crypto research should be "Attacking Popular Open-Source Turn-Based Strategy Games"
20090812 03:30:10-!- melinath [n=melinath@pool-72-72-193-179.altnpa.east.verizon.net] has joined #wesnoth-dev
20090812 03:34:53-!- noy [n=Noy@wesnoth/developer/noy] has quit []
20090812 03:35:22< Sirp> Zarel: "I think you do it wrongly" <-- I like this FAQ. :)
20090812 03:36:18< Aethaeryn> melinath: just curious... why'd you swap the slasher?
20090812 03:36:35< Aethaeryn> melinath: I could be mistaken, but the old name was more intuitive on the guy who has a giant blade weapon than the line that does not.
20090812 03:37:22< Zarel> Sirp: That's one of the parts I didn't write. :P
20090812 03:37:36< melinath> Aethaeryn: I'm not sure that was originally my idea...
20090812 03:37:47< melinath> Aethaeryn: but I can explain it.
20090812 03:38:00< melinath> Aethaeryn: The problem was that Gladiator didn't fit.
20090812 03:38:04< Aethaeryn> melinath: it just seems counter-intuitive to swap a branch name.
20090812 03:38:21< Aethaeryn> melinath: My two cents would be to swap the names of the lvl 2s or replace "slasher" with something else.
20090812 03:38:33< Aethaeryn> I'd say it'd be better to replace "slasher" if you're content with the others. They actually are nice names.
20090812 03:38:47< Aethaeryn> It'll be confusing to read old guides that say a "slasher" is good for something when the slasher is the opposite.
20090812 03:38:56< melinath> That's true.
20090812 03:39:05< Aethaeryn> Or old forum posts a year from now, newbies would be confused.
20090812 03:39:05< Soliton> and it makes it impossible for an automated upgrade path.
20090812 03:39:06< melinath> Slasher seemed to be the best suggestion at the time.
20090812 03:39:12< Aethaeryn> Well, it's confusing. 
20090812 03:39:22< Aethaeryn> It's easier to think that most/all of the upgraded names are replaced, than one is moved.
20090812 03:39:22< Zarel> Why?
20090812 03:39:27< melinath> If you can think of something better, I have nothing against it being used.
20090812 03:39:34< Aethaeryn> Zarel: Imagine if the lancer and knight names were swamped.
20090812 03:39:38< Aethaeryn> *swapped
20090812 03:39:44< Zarel> Who would do that?
20090812 03:39:57< Zarel> Knight doesn't upgrade to anything, but Lancer upgrades to Grand Knight? :P
20090812 03:40:09< Aethaeryn> Zarel: Grand Lancer. ;)
20090812 03:40:34< Zarel> That would be worth the confusion, just to have a Grand Lancer. :D
20090812 03:41:32< Zarel> Imagine a Grand Lancer doing 37-3 at a favorable time of day. :D
20090812 03:42:01< Zarel> Dark Adepts would be _toast_. :D
20090812 03:42:06< melinath> Aethaeryn: Revision of my last statement: If you can think of something better, I would be happy to see something used that wouldn't cause problems.
20090812 03:42:10< Aethaeryn> melinath: http://thesaurus.reference.com/browse/enforcer
20090812 03:42:13< Aethaeryn> Any of those sound interesting?
20090812 03:42:21< Aethaeryn> I'm disappointed in the thesaurus now... :(
20090812 03:42:41< Zarel> "Enforcer" sound like something "Arbiter" would upgrade to.
20090812 03:42:48< Zarel> They both have "law" connotations.
20090812 03:43:21< Zarel> What was wrong iwth Gladiator?
20090812 03:43:25< melinath> People used thesaurus a lot during the discussions...
20090812 03:43:30< Aethaeryn> melinath: if you want something that rhymes...
20090812 03:43:37< melinath> haha
20090812 03:43:53< melinath> Zarel: define gladiator. That's the problem.
20090812 03:44:03< Zarel> melinath: Someone who fights a lot! :D
20090812 03:44:16< Aethaeryn> melinath: basher, lasher, slasher, smasher, thrasher
20090812 03:44:26< Aethaeryn> or anything else here: http://www.rhymezone.com/r/rhyme.cgi?Word=basher&typeofrhyme=perfect&org1=syl&org2=l
20090812 03:44:38< Aethaeryn> slasher obviously is the one that's already there :P
20090812 03:44:48< Zarel> Blacksmith -> Weaponsmith?
20090812 03:44:51< melinath> Zarel: http://dictionary.reference.com/browse/gladiator
20090812 03:44:58< Zarel> Would make it clearer that those are the ones with the bunch of different weapons, too.
20090812 03:45:26< melinath> Zarel: Also, the fluff about being smiths died and is rightly gone.
20090812 03:45:30< Aethaeryn> basher and smasher sound cool.
20090812 03:45:36< melinath> Zarel: have you read the discussion?
20090812 03:45:47< Zarel> melinath: There are discussions? :P
20090812 03:45:48< Aethaeryn> thrasher also sounds cool but doesn't when you look at the the definitions :P
20090812 03:45:56< Aethaeryn> Drake Farm Machines?
20090812 03:46:00< Zarel> I wasn't even _involved_ with Wesnoth at all until recently.
20090812 03:46:04< melinath> Aethaeryn: I kinda like Thrasher...
20090812 03:46:10< melinath> Zarel: I'll link you.
20090812 03:46:23< Aethaeryn> melinath: it is also my favorite of the rhymes, just the most unfitting. :P
20090812 03:46:24< Zarel> Aethaeryn: lolololol
20090812 03:46:43< Aethaeryn> melinath: could ask esr. He probably named half the units in Thunderstone.
20090812 03:46:45< Zarel> I mean, it's not _that_ bad. I personally wouldn't want to get attacked by a farm machine.
20090812 03:46:49< melinath> Zarel: http://www.wesnoth.org/forum/viewtopic.php?f=32&t=26165
20090812 03:46:56< Zarel> Plus, in Ascension, they could upgrade to Drake Tractors!
20090812 03:47:09< Aethaeryn> I am just wholly (though it doesn't really matter since I'm not with any kind of power over this matter) against slasher being the former gladiator.
20090812 03:47:12< Aethaeryn> just too confusing.
20090812 03:47:15< Aethaeryn> And I'm not a newbie.
20090812 03:47:41< Aethaeryn> Might as well change it before 1.7.3 when it'll stick.
20090812 03:48:06< melinath> Aethaeryn: In the end, the power to change names does not rest with me either. Gladiator didn't fit with the fluff; anything that does is okay.
20090812 03:48:21< Aethaeryn> melinath: I was aware of the thread, btw, I just didn't realize the slasher was swapped until I saw the image with the name. Must not have read it that carefully.
20090812 03:48:38< Aethaeryn> Which is one of the reasons it'll confuse.
20090812 03:48:54< melinath> I think that Thrasher would be all right since it's a less common word.
20090812 03:49:07< melinath> You'd think 'something that thrashes' rather than 'farm  machine'
20090812 03:49:23< melinath> Though it's not perfect either.
20090812 03:49:50< Aethaeryn> melinath: actually, dictionary.com has better definitions
20090812 03:49:51< Aethaeryn> to beat soundly in punishment; flog.
20090812 03:49:56< Aethaeryn> is definition #1 for flash.
20090812 03:50:05< Aethaeryn> Perfect for being under the Enforcer.
20090812 03:50:11< Aethaeryn> PERFECT.
20090812 03:51:51< Zarel> "Flasher"?
20090812 03:51:52< Aethaeryn> melinath: http://www.wesnoth.org/forum/viewtopic.php?p=375875#p375875
20090812 03:52:01< Zarel> I'm pretty sure that's not the first thing that comes up when _I_ think of 'flasher'
20090812 03:52:07< melinath> Come back here, you unlawful drake! I'll give you the thrashing you deserve!
20090812 03:52:10< melinath> :-)
20090812 03:52:12< Aethaeryn> Honestly, the definition fits, the name is cool, and it rhymes with clasher, which is the intended thing.
20090812 03:52:29< Aethaeryn> See, too many people use thesauruses when rhyming dictionaries are what they want/need.
20090812 03:52:32< melinath> I'm fine with it.
20090812 03:52:35< Aethaeryn> I <3 writing.
20090812 03:52:39< melinath> haha!
20090812 03:52:46< esr> No good,guys.
20090812 03:52:57< melinath> ah, hi esr.
20090812 03:53:15< esr> "Flashwer" is U.S. slang for a kind of pervert who exposes his genitals in public.
20090812 03:53:30< esr> s/Flashwer/Flasher/
20090812 03:53:31< melinath> That==Zarel's joke.
20090812 03:53:39< melinath> The suggestion was Thrasher.
20090812 03:53:50< esr> Which I think is fine.
20090812 03:54:07< melinath> or possibly his misunderstanding...
20090812 03:54:14< melinath> glad to hear it.
20090812 03:54:24< Zarel> Yeah, it was a joke based on Aethaeryn's typo: <Aethaeryn> is definition #1 for flash.
20090812 03:55:25< Aethaeryn> esr: Thrasher? I edited my post in the Drake thread to be more clear to those not in on the conversation.
20090812 03:55:50< melinath> esr: you still interested in working on a pre-TRoW writing system? I could work on it this weekend.
20090812 03:55:52< Aethaeryn> It sounds good for an underling of an Enforcer, imo, at least if you have the slightly-corrupt-and-forceful connontation.
20090812 03:56:09< esr> melinath: Yes, I'm still interested.
20090812 03:56:26< melinath> esr: glad to hear it. :-)
20090812 03:56:54< Aethaeryn> esr: http://www.wesnoth.org/forum/viewtopic.php?p=375875#p375875 would be the post in question.
20090812 03:57:46-!- Netsplit clarke.freenode.net <-> irc.freenode.net quits: Alesis-Novik_
20090812 03:58:32< esr> I think Thrasher is fine.
20090812 03:58:41< Aethaeryn> Cool.
20090812 03:59:05< Aethaeryn> I know I always call you a Dwarf expert, but you're also kinda a Drake expert too, since you redid their race background info when we redid the wiki.
20090812 04:00:01< Zarel> I think the names should reflect their function more.
20090812 04:00:38-!- Netsplit over, joins: Alesis-Novik_
20090812 04:00:38< Aethaeryn> Zarel: But Thrasher really does fit a function.
20090812 04:00:45< Zarel> Oh, no, Thrasher is fine.
20090812 04:00:50< Aethaeryn> It was an accidental name that actually fits as an underling of the Enforcer when I looked it up.
20090812 04:01:07< Zarel> I think Enforcer should have less of a "law" feel to it; that should be restrained to Arbiter/Warden line.
20090812 04:02:46< Zarel> And there's also that the Thrasher descriptions imply they do more damage, but yet the Arbiter does more base damage.
20090812 04:03:11< Aethaeryn> Zarel: If you think the Thrasher and Enforcer should be swapped...
20090812 04:03:26 * Zarel shrugs
20090812 04:03:43< Zarel> I'm just voicing things I notice. I'm not suggesting anything (at least not yet)
20090812 04:06:47< Aethaeryn> Eh, opinions work.
20090812 04:11:24< melinath> Zarel: all clasher lines are currently cast as lawmen.
20090812 04:11:47< melinath> Zarel: Enforcers are just more... forceful about it.
20090812 04:12:57-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090812 04:14:37< Aethaeryn> melinath: well then, thrasher works more than I thought :)
20090812 04:47:23-!- Ivanovic_ [n=ivanovic@dtmd-4db2a62c.pool.einsundeins.de] has joined #wesnoth-dev
20090812 05:02:31< Aethaeryn> melinath: you seem to be full of ideas
20090812 05:02:38< Aethaeryn> melinath: http://www.wesnoth.org/forum/viewtopic.php?f=12&t=26632
20090812 05:03:30< Aethaeryn> esr: If you have any ideas of the origins of Irdya, this may be your opportunity to elaborate as well.
20090812 05:03:50-!- Ivanovic [n=ivanovic@wesnoth/developer/ivanovic] has quit [Read error: 113 (No route to host)]
20090812 05:03:53< Aethaeryn> You seem to know about an elvish pact of some sort, and that's over 6000 years before the earliest point the rest of us mortals know.
20090812 05:05:21-!- Ivanovic_ is now known as Ivanovic
20090812 05:10:23< shadowmaster> night
20090812 05:11:04< Aethaeryn> 'night
20090812 05:11:08< Aethaeryn> I will do likewise
20090812 05:11:12< CIA-62> soliton * r37691 /trunk/utils/mp-server/tail_log_irc: handle the mod filter differently
20090812 05:11:15-!- Aethaeryn [n=Michael@wesnoth/umc-dev/developer/aethaeryn] has quit ["'night"]
20090812 05:16:14-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit []
20090812 05:24:13-!- ancestral [n=ancestra@97-116-126-165.mpls.qwest.net] has joined #wesnoth-dev
20090812 05:41:19-!- noy [n=Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090812 06:09:03-!- karl_m [n=karlm@220-253-90-228.QLD.netspace.net.au] has joined #wesnoth-dev
20090812 06:13:44< karl_m> anyone on who knows about using lua to call WML events?
20090812 06:16:08-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090812 06:25:11-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 110 (Connection timed out)]
20090812 06:37:38-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has quit [Connection reset by peer]
20090812 06:37:54-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090812 06:49:48-!- [Relic] [n=[Relic]@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit ["Leaving"]
20090812 06:51:48-!- silene [n=plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20090812 06:59:24< silene> karl_m: wesnoth.fire("fire_event", { name="my_event" }) for instance
20090812 07:02:31-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090812 07:04:25-!- Blarumyrran [n=chatzill@82.131.13.45.cable.starman.ee] has joined #wesnoth-dev
20090812 07:05:01< karl_m> silene: cool, I was missing the 'name=' Doh! thank you
20090812 07:09:00-!- Sirp [n=user@wesnoth/developer/dave] has quit ["leaving"]
20090812 07:10:19-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Connection timed out]
20090812 07:22:49-!- allefant [n=allefant@allegro/developer/allefant] has quit [Read error: 60 (Operation timed out)]
20090812 07:32:38-!- allefant [n=allefant@allegro/developer/allefant] has joined #wesnoth-dev
20090812 07:53:12-!- Valkier [n=karasawa@c-71-225-240-156.hsd1.pa.comcast.net] has joined #Wesnoth-dev
20090812 07:56:24-!- Netsplit clarke.freenode.net <-> irc.freenode.net quits: erl
20090812 07:57:28-!- Netsplit over, joins: erl
20090812 08:31:20-!- karl_m [n=karlm@220-253-90-228.QLD.netspace.net.au] has left #wesnoth-dev []
20090812 08:38:33-!- euschn [n=chatzill@wesnoth/developer/euschn] has joined #wesnoth-dev
20090812 08:38:39< euschn> hi
20090812 08:42:24-!- zookeeper [n=l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20090812 08:47:02-!- Kenpachi [n=chatzill@CPE-121-214-224-170.static.qld.bigpond.net.au] has joined #wesnoth-dev
20090812 08:53:39-!- melinath [n=melinath@pool-72-72-193-179.altnpa.east.verizon.net] has quit [Read error: 113 (No route to host)]
20090812 09:11:04-!- YogiHH [i=c3f5f304@wesnoth/developer/yogihh] has joined #wesnoth-dev
20090812 09:11:09< YogiHH> hello
20090812 09:11:22< euschn> trunk crashes when loading the save from https://gna.org/bugs/?14108 , yielding these errors: unit of type Drake Gladiator not found!; game_error: Unknown unit type 'Drake Gladiator'
20090812 09:11:26< euschn> hi YogiHH 
20090812 09:11:52< YogiHH> euschn: how old is that save?
20090812 09:11:56< euschn> 1.7.2
20090812 09:12:18< euschn> in 1.7.2 it works fine
20090812 09:12:29< YogiHH> might be related to renaming of the Drake clasher line, maybe?
20090812 09:13:09< euschn> i see, i wasnt aware that those were renamed
20090812 09:13:42< YogiHH> i am not sure which names were switched exactly, just that there were some renaming recently
20090812 09:14:02-!- Crab_ [i=crab@wesnoth/developer/crab] has joined #wesnoth-dev
20090812 09:17:09< silene> yes, changing the unit name is fine, but changing the id of the unit type was really a dumb move
20090812 09:18:35< YogiHH> silene: well, as long as we don't guarantee version compatibility, i think it is ok
20090812 09:19:10< YogiHH> silene: during development, there will always be changes that break savegames
20090812 09:25:41< silene> sure, but there are two kinds of changes: changes that are necessary and changes that aren't necessary
20090812 09:27:57< Valkier> Yeah like this one time I put on a new pair of pants because I brushed past a table I disliked.
20090812 09:28:00< Valkier> That was neccesary!
20090812 09:29:03< Valkier> But then later I took that same pair of pants off in a Walmart because I felt over dressed. Turns out it was unneccesary, and illegal.
20090812 09:29:24< Valkier> unnecessary*
20090812 09:29:38 * Valkier bows
20090812 09:31:06< zookeeper> anyone notice the ackward silence?
20090812 09:31:26< Valkier> And you went and ruined it.
20090812 09:31:32< zookeeper> oops.
20090812 09:32:38-!- ancestral [n=ancestra@97-116-126-165.mpls.qwest.net] has quit ["And that’s the end of THAT chapter."]
20090812 09:35:19-!- Zarel [i=Zarel@warzone2100/developer/Zarel] has quit [Read error: 110 (Connection timed out)]
20090812 09:35:33-!- stikonas [n=and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20090812 09:36:08-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090812 09:37:45< YogiHH> Valkier: After all, i am not familiar with this change and i can't tell about esr's motives. However, i guess it was due to consistency reasons. If i set up a "can_recruit" attribute and i have to know that the Slasher i want to put in has the id "Gladiator" and vice versa, i'd be pretty confused.
20090812 09:38:41< Valkier> YogiHH: I just do some art. I really don't understand much of what you guys talk about here.
20090812 09:39:12< YogiHH> ah, ok
20090812 10:01:28-!- YogiHH [i=c3f5f304@wesnoth/developer/yogihh] has left #wesnoth-dev []
20090812 10:03:38< silene> YogiHH: you don't have to know it, that's what unit type aliases can be used for
20090812 10:10:41-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090812 10:21:19-!- Netsplit clarke.freenode.net <-> irc.freenode.net quits: erl
20090812 10:22:03-!- Netsplit over, joins: erl
20090812 10:27:11-!- noy [n=Noy@wesnoth/developer/noy] has quit []
20090812 10:36:06-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has quit [Remote closed the connection]
20090812 10:40:15-!- loonybot [n=loonybot@79.139.138.234] has joined #wesnoth-dev
20090812 10:41:00-!- loonycyborg [n=sergey@79.139.138.234] has joined #wesnoth-dev
20090812 10:47:30< Ivanovic> moin
20090812 11:07:11-!- Netsplit clarke.freenode.net <-> irc.freenode.net quits: erl
20090812 11:08:03-!- Netsplit over, joins: erl
20090812 11:13:38< deekay> Morning  everyone
20090812 11:13:49< Crab_> hello deekay
20090812 11:14:21< deekay> Crab_: ai/composite/stage.cpp:116 - forgot about return?
20090812 11:14:33< deekay> And hi Crab_ :)
20090812 11:16:38< Crab_> deekay: that snippet of stage.cpp looks like http://pastebin.mozilla.org/666694 atm
20090812 11:17:04-!- YogiHH [i=c3f5f304@wesnoth/developer/yogihh] has joined #wesnoth-dev
20090812 11:17:06< deekay> Crab_: Commit? :)
20090812 11:17:39< Crab_> deekay: I'll commit soon, compiling some changes atm
20090812 11:17:44-!- Netsplit clarke.freenode.net <-> irc.freenode.net quits: erl
20090812 11:17:44< deekay> k
20090812 11:18:48-!- Netsplit over, joins: erl
20090812 11:33:44< silene> Crab_: yes, please commit only things that actually compiles
20090812 11:34:18< Crab_> silene: I have 'strict compilation' off, and I compile before committing.
20090812 11:36:05< loonycyborg> Why not enable it?
20090812 11:36:50< Crab_> loonycyborg: 'scons strict=true' ?
20090812 11:36:56< loonycyborg> Yes.
20090812 11:37:08< Crab_> ok
20090812 11:37:42< loonycyborg> strict compilation should work unless you get warnings from boost headers or something :P
20090812 11:45:50-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20090812 11:47:14-!- Noyga [n=lame-z@wesnoth/developer/noyga] has joined #wesnoth-dev
20090812 12:10:10-!- Valkier [n=karasawa@c-71-225-240-156.hsd1.pa.comcast.net] has quit []
20090812 12:26:27< Crab_> loonycyborg: yes, it worked
20090812 12:35:44-!- fendrin [n=fabi@wesnoth/developer/fendrin] has quit [Remote closed the connection]
20090812 12:37:49-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090812 12:57:41< CIA-62> ivanovic * r37692 /trunk/po/ (wesnoth-ei/de.po wesnoth-thot/de.po): updated German translation
20090812 13:06:05< CIA-62> ivanovic * r37693 /trunk/ (761 files in 24 dirs): 
20090812 13:06:05< CIA-62> pot-update
20090812 13:06:05< CIA-62> regenerated doc files
20090812 13:12:48< CIA-62> ivanovic * r37694 /branches/1.6/ (117 files in 3 dirs): 
20090812 13:12:48< CIA-62> pot-update (no new/changed strings)
20090812 13:12:48< CIA-62> regenerated doc files
20090812 13:15:20< Ivanovic> Crab_: uhm, get a warning...
20090812 13:15:31< Ivanovic> [ 83%] Building CXX object src/CMakeFiles/wesnoth.dir/ai/composite/stage.cpp.o
20090812 13:15:32< Ivanovic> cc1plus: warnings being treated as errors
20090812 13:15:34< Ivanovic> /src/wesnoth/src/ai/composite/stage.cpp: In member function 'ai::stage_ptr ai::ministage::get_stage_ptr(ai::ai_context&)':
20090812 13:15:35< Ivanovic> /src/wesnoth/src/ai/composite/stage.cpp:117: error: control reaches end of non-void function
20090812 13:15:36< CIA-62> crab * r37695 /trunk/src/ai/ (7 files in 2 dirs): added a recruitment ministage (not used yet)
20090812 13:15:41< Crab_> Ivanovic: fixed
20090812 13:15:51< Crab_> deekay: done
20090812 13:16:02< deekay> Crab_: Thank you :)
20090812 13:16:02< Crab_> deekay: see data/ai/dev/formula_ai.cfg
20090812 13:16:08< deekay> ok
20090812 13:17:16< Crab_> deekay: just wait a sec for r37695 to upload..
20090812 13:17:25< Crab_> s/r37595/r37596
20090812 13:18:16< Ivanovic> deekay: so this evening a test game to check if your ai improvements help?
20090812 13:18:44-!- AnMaster [n=AnMaster@unaffiliated/anmaster] has quit ["ZNC - http://znc.sourceforge.net"]
20090812 13:19:29< CIA-62> crab * r37696 /trunk/ (12 files in 5 dirs): recruitment as aspect
20090812 13:19:58< deekay> Ivanovic: Well, I would prefer tomorrow, thate are still some improvements I'd like to do, and addictionaly mine and grzywacz's opponents from 2v2 tournament want to play game this evening too... they couldn't come with better timing
20090812 13:20:01< Crab_> deekay: now it's ok. (some pot-updates sneaked in during a svn rebase :) )
20090812 13:20:50< deekay> Crab_: ok, thanks :)
20090812 13:22:30< deekay> Ivanovic: Or I'll just set up a game for 18GMT+2 and will just observe it
20090812 13:23:02< Crab_> why Drake Burner (Lvl 1) has a same attacks as Drake Glider (Lvl1) ?
20090812 13:23:15< deekay> Crab_: ???
20090812 13:23:32< Crab_> both have 3-3 fire breath marksman,  6-2 impact slam
20090812 13:23:39 * Crab_ suspects something is not right..
20090812 13:25:25-!- Kenpachi_ [n=chatzill@CPE-58-169-231-110.sa.bigpond.net.au] has joined #wesnoth-dev
20090812 13:25:55-!- xchat_sux [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090812 13:26:33< deekay> Crab_: Yes, apparently drake burner was weakened when it got new frames
20090812 13:27:08< Crab_> wesbot: log 37670
20090812 13:27:12< wesbot> thespaceinvader * r37670 : Wire new Drake Burner flight and fire frames.
20090812 13:27:15< wesbot> URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=37670
20090812 13:27:24< Crab_> there ^
20090812 13:27:38< deekay> That's easy to fix
20090812 13:27:41 * deekay will do
20090812 13:27:48< Crab_> thanks
20090812 13:28:09-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090812 13:28:48< Crab_> deekay: also note Sky Drake
20090812 13:29:06< Crab_> deekay: and Hurricane Drake
20090812 13:29:21< deekay> Yes
20090812 13:29:47< Crab_> wesbot: log 37513
20090812 13:29:48< wesbot> thespaceinvader * r37513 : New frames and animations for the Hurricane Drake, delete unused old frames.
20090812 13:29:51< wesbot> URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=37513
20090812 13:30:16-!- xchat_sux is now known as Elvish_Pillager
20090812 13:31:20< Crab_> wesbot: log 37417
20090812 13:31:20< wesbot> thespaceinvader * r37417 : New Sky Drake base frames and animations, remove old frames, minor fix to Glider WML.  Update changelogs.
20090812 13:31:24< wesbot> URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=37417
20090812 13:38:39-!- Kenpachi___ [n=chatzill@CPE-124-182-233-77.sa.bigpond.net.au] has joined #wesnoth-dev
20090812 13:39:22< CIA-62> dragonking * r37697 /trunk/data/core/units/drakes/ (Burner.cfg Hurricane.cfg Sky.cfg): Reverted accidental changes to attacks of burner, sky and hurricane drakes
20090812 13:39:37< deekay> Crab_: seems it is ok now
20090812 13:39:44< Crab_> good :)
20090812 13:42:32-!- Kenpachi [n=chatzill@CPE-121-214-224-170.static.qld.bigpond.net.au] has quit [Read error: 110 (Connection timed out)]
20090812 13:45:42-!- Kenpachi_ [n=chatzill@CPE-58-169-231-110.sa.bigpond.net.au] has quit [Read error: 110 (Connection timed out)]
20090812 13:51:34-!- Kenpachi____ [n=chatzill@CPE-58-166-242-16.sa.bigpond.net.au] has joined #wesnoth-dev
20090812 13:51:35-!- Kenpachi____ is now known as Kenpachi
20090812 13:57:13-!- Kenpachi___ [n=chatzill@CPE-124-182-233-77.sa.bigpond.net.au] has quit [Read error: 110 (Connection timed out)]
20090812 13:58:14-!- YogiHH [i=c3f5f304@wesnoth/developer/yogihh] has left #wesnoth-dev []
20090812 14:09:05-!- euschn [n=chatzill@wesnoth/developer/euschn] has quit ["ChatZilla 0.9.85 [Firefox 3.0.13/2009080315]"]
20090812 14:09:19-!- Kenpachi_ [n=chatzill@121.214.210.39] has joined #wesnoth-dev
20090812 14:22:47-!- Noyga [n=lame-z@wesnoth/developer/noyga] has quit [Remote closed the connection]
20090812 14:25:24-!- Kenpachi__ [n=chatzill@CPE-124-177-185-63.sa.bigpond.net.au] has joined #wesnoth-dev
20090812 14:27:51-!- Kenpachi [n=chatzill@CPE-58-166-242-16.sa.bigpond.net.au] has quit [Read error: 110 (Connection timed out)]
20090812 14:28:01-!- Kenpachi__ is now known as Kenpachi
20090812 14:34:05-!- Kenpachi__ [n=chatzill@CPE-139-168-197-136.sa.bigpond.net.au] has joined #wesnoth-dev
20090812 14:38:31-!- boucman [i=43fa2d5d@wesnoth/developer/boucman] has joined #wesnoth-dev
20090812 14:38:37< boucman> hey all
20090812 14:38:41< Crab_> hi boucman
20090812 14:39:55< boucman> hey
20090812 14:42:55< Crab_> boucman: so, I've added the capability to wrap any stage inside aspect. Look into ai_default::do_recruitment to see the way this can be used.
20090812 14:43:03-!- Kenpachi_ [n=chatzill@121.214.210.39] has quit [Read error: 110 (Connection timed out)]
20090812 14:43:34< boucman> i'll check your commit, direct access to code is tricky from here
20090812 14:44:02< Crab_> http://pastebin.mozilla.org/666725
20090812 14:46:15< boucman> that's great, 
20090812 14:46:40< boucman> i can see how it could be usefull for "special cases"
20090812 14:47:12< Crab_> Yes.  and note that there was not need to create a separate 'fai recruitment stage' - we just reused a side_formulas stage
20090812 14:47:12< boucman> do you intend to use that mechanism to provide DK's error catching mechanism ?
20090812 14:47:40< Crab_> boucman: explain more :)
20090812 14:48:11< Crab_> boucman: afair, DK's error catching mechanism is implemented in formula_ai::execute_variant , and is not related
20090812 14:48:55< boucman> i must admit his error catching idea was a bit confusing for me too
20090812 14:49:14< boucman> I thought it was related to catching up on gameplay errors
20090812 14:50:57-!- Kenpachi [n=chatzill@CPE-124-177-185-63.sa.bigpond.net.au] has quit [Read error: 110 (Connection timed out)]
20090812 14:53:08-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090812 14:53:17< Crab_> boucman: his error catching (i.e. on_fail) is an easier way to write formulas which can fail because of unexpected circumstances
20090812 14:53:43< boucman> yes, that's what i understood
20090812 14:53:59< Crab_> and this my work is just 'executing a stage outside of normal turn flow', e.g. "when we need to recruit, do THIS"
20090812 14:54:33< Crab_> note that 'THIS' can be resolved to different implementations (in different time of days, or on different turns)
20090812 14:57:28< boucman> yes, I got that
20090812 14:57:56< Crab_> boucman: the next 'global' thing to do with the AI is to think out its 'goals\targets' part
20090812 14:57:57< boucman> I was more interested in what case, appart from recruitement, we could use that mechanism...
20090812 14:58:18< boucman> yes, that could be an interesting part
20090812 14:58:51< Crab_> boucman: since we need to provide a good way for scenario creators (or by player using right-click menu) to set AI goals
20090812 14:59:14< Crab_> boucman: also, AI would need to pick some goals for itself, too (ai_default uses that in a number of places)
20090812 14:59:17< boucman> hmm
20090812 14:59:41< Crab_> e.g. "is this village interesting to us?"
20090812 15:00:46-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 110 (Connection timed out)]
20090812 15:00:46< boucman> it seems there are two different thing we call goals here... and we need not to get confused
20090812 15:00:50< Crab_> note that this is connected with DK's recruitment work. "is this terrain\ choke point interesting to us ? if yes, should we recruit units which are good on it? how many ? "
20090812 15:01:27< boucman> on the one hand you have scenario goals, which are set by WML, or default to "kill opposite leader"
20090812 15:01:45< Crab_> yes
20090812 15:01:58< boucman> and on the other hand you have tacticall goals like evaluating a given hex to get tactical advantages
20090812 15:02:17< boucman> scenario goals could probably be a composite aspect
20090812 15:03:04< boucman> with a cpp special engine that would parse a wml description and/or a fai engine that would return a new type of variant to describe an objective
20090812 15:03:37< boucman> tactical goals are purely AI internal, and probably don't need to be accessible in an AI agnostic manner
20090812 15:04:28< Crab_> boucman: what is the difference between scenario goal and tactical goal ?
20090812 15:05:10< boucman> scenario goals are "what you need to do to win the map"
20090812 15:05:24< Crab_> boucman: for example, both scenario creator and 'ai internal evaluator' can say "get this village as quick as you can"
20090812 15:05:28< boucman> tactical goals are "what you want to achieve this turn"
20090812 15:05:48< Crab_> boucman: ok, understood
20090812 15:06:37< Crab_> then, my example about 'get this village as quick as you can' will be a scenario goal, yes ?
20090812 15:06:46< boucman> not that this does not contradict with what you sqid
20090812 15:07:08-!- lizard_r [n=Miranda@wesnoth/umc-dev/developer/lizard] has joined #wesnoth-dev
20090812 15:07:49 * boucman is not used to qwerty keyboards yet
20090812 15:08:12< Crab_> boucman: ( what keyboard do you usually use ? )
20090812 15:09:11< Crab_> boucman: so, we'll need a bunch of persistent 'scenario goals', which can be set by scenario creator / ai / player, and we'll need a set of 'tactical goals' which will be recalculated each turn. right ?
20090812 15:10:50< boucman> azerty, i'm french
20090812 15:11:08< boucman> Crab_: yes and no
20090812 15:11:15< boucman> scenario goals, yes. 
20090812 15:11:22-!- allefant [n=allefant@allegro/developer/allefant] has quit [Read error: 60 (Operation timed out)]
20090812 15:11:39< boucman> and in an AI independant way, that's why I thought aspect would be the way to go
20090812 15:12:15< boucman> tactical goals could be considered purely AI internal, some AI will use that, and other (default AI) wouldn't
20090812 15:12:29< Crab_> so, two aspects, of 'scenario goals' and 'tactical targets' ?
20090812 15:12:45< Crab_> ai_default already uses tactical targets :)
20090812 15:13:02< Crab_> (and note the proposed terminology): 'goals' vs 'targets'
20090812 15:14:15< boucman> Crab_: one aspect for scenario goals, I'm not sure if tactical goals should be an aspect (I'm not sure letting scenario designers tweak with tactical goals is a good idea)
20090812 15:14:47< boucman> i like the tactical vs scenario goal terminology if that's fine with you
20090812 15:15:09< Crab_> boucman: just because something is an aspect doesn't mean we would encourage scenario creators to mess with it ;)
20090812 15:15:22< Crab_> boucman: ok
20090812 15:15:55< Crab_> boucman: well, tactical goals would also be shared between some ai parts (for example, ai_default_recruitment_stage uses the same get_targets algorithm that ai_default::move_to_targets )
20090812 15:16:11-!- BenUrban [n=benurban@unaffiliated/benurban] has quit ["Power failu"]
20090812 15:16:23< boucman> Crab_: indeed, but what do we gain in making it a mandatory part of the engine ?
20090812 15:16:38< boucman> hmm
20090812 15:17:14< boucman> so make it an aspect that is limited to some ai, not mandatory for all ai
20090812 15:17:18< Crab_> boucman: we will gain the ability to share those 'tactical goals' between all those ai parts which want to share it.
20090812 15:17:45< Crab_> boucman: yes. we just will not make the promise 'this aspect is respected by all ai, you can use it'
20090812 15:18:01< boucman> ok, so to write down what we agree on
20090812 15:18:02< Crab_> boucman: we can even state 'this aspect is purposely not documented' in the documentation
20090812 15:18:28< boucman> Crab_: sounds good
20090812 15:18:54< boucman> 1) tactical and scenario goals are two different things that needs to be handled differently
20090812 15:19:02-!- crimson_p [n=irchon@64.201.60.211] has joined #wesnoth-dev
20090812 15:19:31-!- Aethaeryn [n=Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #Wesnoth-dev
20090812 15:19:34< boucman> 2) tactical goals are purely internal to AI, will probably be implemented via aspects, and should not be user modifiable
20090812 15:20:03-!- crimson_p [n=irchon@64.201.60.211] has quit [Client Quit]
20090812 15:20:07< boucman> afk ~15' 
20090812 15:20:28< Crab_> boucman: but tactical goals generation algorithm should be developer-modificable (there's a bunch of magic parameters in it that we can tinker with)
20090812 15:26:32-!- allefant [n=allefant@allegro/developer/allefant] has joined #wesnoth-dev
20090812 15:26:38-!- stikonas_ [n=and@ctv-213-164-121-103.vinita.lt] has joined #wesnoth-dev
20090812 15:27:25-!- crimson_p [n=irchon@64.201.60.211] has joined #wesnoth-dev
20090812 15:28:50-!- crimson_p [n=irchon@64.201.60.211] has quit [Remote closed the connection]
20090812 15:32:18-!- Netsplit clarke.freenode.net <-> irc.freenode.net quits: nital, loonycyborg
20090812 15:32:57-!- Netsplit over, joins: loonycyborg, nital
20090812 15:39:52-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 110 (Connection timed out)]
20090812 15:40:32-!- Kenpachi__ is now known as Kenpachi
20090812 15:40:52< Soliton> Crab_: 20090812 03:53:24 error ai/actions: Error #-1 in recruitment by side 4 of unit type [Saurian Skirmisher] on any suitable location
20090812 15:40:56< Soliton> wesnoth: src/ai/../unit_map.hpp:113: typename iter_types::pointer_type unit_map::iterator_base<iter_policy, iter_types>::operator->() const [with iter_policy = unit_map::unit_policy, iter_types = unit_map::const_iter_types]: Assertion `policy_.valid(i_, map_) && "Attempt to dereference invalid iterator"' failed.
20090812 15:41:20< Soliton> Crab_: should be reproducable in a game with the pick-your-recruits era and an ai side.
20090812 15:41:35< Crab_> Soliton: ok, I'll try
20090812 15:43:02< boucman> Crab_: agreed, but that enters into the "ai specific aspect" category for me
20090812 15:43:15< boucman> i.e each ai can have its own syntax for it
20090812 15:43:20< Crab_> boucman: yes
20090812 15:44:27< boucman> ok, then we agree on everything :)
20090812 15:46:07-!- Kenpachi_ [n=chatzill@CPE-58-170-82-51.sa.bigpond.net.au] has joined #wesnoth-dev
20090812 15:52:22< Crab_> Soliton: found out the reason (my error in post-recruit check combined with unusual handling of recruits in that era). I'll test & commit a fix.
20090812 15:55:08< Soliton> cool, thanks.
20090812 15:56:11< Crab_> Soliton: note that ' Error #-1 in recruitment by side 4 of unit type [Saurian Skirmisher] on any suitable location' will not go away after my fix.
20090812 15:56:44< Soliton> sounds like it should be a warning then?
20090812 15:57:10< Crab_> Soliton: from the ai POV, it's an error. It tried a legal move: "recruit a unit", but it failed.
20090812 15:57:28< Soliton> yeah, ok.
20090812 15:57:42< Crab_> Soliton: properly fixing this requires adding a way for the WML to tell the ai 'take it easy, it's all ok' 
20090812 15:58:19< Soliton> if the ai doesn't freak out after that error it's fine.
20090812 15:58:39< Crab_> Soliton: or, another way is to add an [ai][/ai] snippet to era config to instruct the ai to do turn-1 recruiting differently
20090812 16:00:00< Soliton> that'd be best to actually make the ai work well in that era. though i think the random recruitment currently is fine so far.
20090812 16:00:08< Crab_> deekay: ^
20090812 16:00:24< deekay> huh?
20090812 16:00:30< Crab_> Soliton: ai engine support for 'special recruiting on turn 1' is already present
20090812 16:00:55< Crab_> deekay: there's an 'pick-your-recruits' era in Addons, which requires special turn-1 recruiting to be effective.
20090812 16:00:59-!- stikonas_ is now known as stikonas
20090812 16:01:27< deekay> What's so special about turn 1 that current solution fails?
20090812 16:01:33< Soliton> ok, i suppose it'd be good if you could tell the AI to recruit a fighter, archer, scout etc to make a somewhat well rounded faction.
20090812 16:01:47< Crab_> deekay: in that era, on turn 1 you pick your recruits
20090812 16:02:01< deekay> ok
20090812 16:02:04< Crab_> deekay: so, you can recruit any 1st level recruits
20090812 16:02:06< Soliton> deekay: on turn 1 you recruit the units you will be able to recruit throughout the rest of the game.
20090812 16:02:17< zookeeper> hmh?
20090812 16:02:21< Crab_> deekay: and you need to pick 5 unit types.they do not actually appear on the map
20090812 16:02:25< deekay> In case of AI, who choses recruits?
20090812 16:02:29< Crab_> deekay: AI
20090812 16:02:41< Crab_> deekay: picking is done by 'trying to recruit'
20090812 16:02:51< deekay> ah, simple
20090812 16:02:56< Crab_> deekay: and, from turn 2 onwards, you stick with these recruit types you've chosen
20090812 16:03:14-!- Kenpachi [n=chatzill@CPE-139-168-197-136.sa.bigpond.net.au] has quit [Read error: 110 (Connection timed out)]
20090812 16:03:17< deekay> Well.. we could write formula to just pick randomly units
20090812 16:03:30< deekay> Stor in the variable and tread as possible recruits
20090812 16:03:39< deekay> treat'
20090812 16:04:15< deekay> For that we need to provide AI with list of units to choose from
20090812 16:04:22< zookeeper> a formula to pick one fighter, one archer, etc would probably work about right. although i think the old default AI has actually worked pretty well out-of-the-box.
20090812 16:04:22< Crab_> deekay: well, I'm sure that there are better ways than 'pure random choice' there :)
20090812 16:04:40< deekay> Yes
20090812 16:05:29< Crab_> deekay: so, I was wondering if your recruiting work can be used there.. note that the recruits chosen need to be unpredictable enough
20090812 16:05:47< zookeeper> one fighter, one scout, one archer, and after that random but so that you won't get more than 2 units of the same usage?
20090812 16:07:02< deekay> Crab_: I guess my current recruitment work can't be used, but it shouldn't be too hard to write one for it 
20090812 16:07:34< deekay> Crab_: What would be returned by my_recruits for the 1st turn in this scenario?
20090812 16:07:42< Crab_> deekay: ok. There's no hurry, I just wanted you to consider that situation - where are many recruit types present at turn 1...
20090812 16:07:49< Crab_> deekay: all lvl 1 unit types
20090812 16:08:04< deekay> ok
20090812 16:08:22< deekay> Crab_: And later it changes?
20090812 16:08:28< Crab_> deekay: yes
20090812 16:08:35< Crab_> on turn 2
20090812 16:08:55< deekay> Crab_: Ok, so we need special random formula for 1st turn, and then just play normally
20090812 16:09:20< deekay> I'll think about it later
20090812 16:09:29< Crab_> deekay: yes. note that we can already do this - one formula for turn 1, another for all other turns.
20090812 16:09:34< deekay> I'm in a middle of cleaning big mess in my files
20090812 16:09:39< deekay> Crab_: Sure :)
20090812 16:09:41< Crab_> ok
20090812 16:14:01-!- crimson_penguin [n=ben@64.201.60.211] has joined #wesnoth-dev
20090812 16:17:47< zookeeper> eh, i think we need to upgrade our loading screens a bit: http://www.wesnoth.org/forum/viewtopic.php?f=6&t=26635
20090812 16:21:24< Soliton> yeah, that's pretty nice.
20090812 16:25:42-!- dk_ [n=dk@chello089077095238.chello.pl] has joined #wesnoth-dev
20090812 16:26:30-!- deekay [n=dk@wesnoth/developer/dragonking] has quit [Nick collision from services.]
20090812 16:26:36-!- dk_ is now known as deekay
20090812 16:29:42< Kenpachi_> that is as nice as blizzard's (tm) World of Warcraft (tm) loading screens
20090812 16:30:47-!- Kenpachi_ is now known as Kenpachi
20090812 16:36:56< boucman> zookeeper: indeed
20090812 16:42:43-!- busfahrer` [n=busfahre@ulmg-5f70cb90.pool.einsundeins.de] has joined #wesnoth-dev
20090812 16:44:38-!- busfahrer` [n=busfahre@ulmg-5f70cb90.pool.einsundeins.de] has quit [Client Quit]
20090812 16:45:07-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090812 16:54:41-!- melinath [n=melinath@pool-72-72-193-179.altnpa.east.verizon.net] has joined #wesnoth-dev
20090812 16:57:45< deekay> Ivanovic: I most likely won't make it today
20090812 16:58:17< deekay> Ivanovic: I need to help my family to bring some new furniture in their house.. meh
20090812 17:04:28< Aethaeryn> zookeeper: yeah, I like the looks of it
20090812 17:05:10-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 60 (Operation timed out)]
20090812 17:17:30-!- Crab_ [i=crab@wesnoth/developer/crab] has quit [Read error: 110 (Connection timed out)]
20090812 17:19:07< Ivanovic> deekay: sure
20090812 17:19:18< Ivanovic> deekay: especially since i will be watching soccer at 6
20090812 17:19:39< Ivanovic> (germany playing vs "unknown country")
20090812 17:21:44< Aethaeryn> Ivanovic: unknown country?
20090812 17:21:56< Aethaeryn> Someone chose Random? ;)
20090812 17:22:16< Ivanovic> not exactly
20090812 17:22:55< Ivanovic> but would you be able to find Azerbaijan on a map without using wikipedia, google or any other information sources?
20090812 17:23:11< Aethaeryn> It's a former Soviet Republic.
20090812 17:23:18< Aethaeryn> I'd guess it would be east of Georgia
20090812 17:23:25< Aethaeryn> in the southeast of the caucasus regions
20090812 17:23:47< Aethaeryn> Time to look it up and see...
20090812 17:23:49< Ivanovic> http://en.wikipedia.org/wiki/File:Europe_location_AZE2.png
20090812 17:23:56< Aethaeryn> YES. \o/
20090812 17:24:00< Aethaeryn> It was a 50/50 guess.
20090812 17:24:32< Ivanovic> a clear case of "NO" since the caucasus is east
20090812 17:24:40< Aethaeryn> Actually, no.
20090812 17:24:43< Aethaeryn> That's exactly what I meant.
20090812 17:24:52< Aethaeryn> The southeast of the former caucasus former soviet republics.
20090812 17:25:09< Aethaeryn> I know my European geography, the caucasus region area is a bit fuzzier
20090812 17:25:16< Ivanovic> :)
20090812 17:25:25< Ivanovic> it is a clear case of "very fuzzy"
20090812 17:25:37-!- Crab_ [i=crab@wesnoth/developer/crab] has joined #wesnoth-dev
20090812 17:25:41< Aethaeryn> It's the state where I *meant*
20090812 17:25:43< Ivanovic> (at least for me)
20090812 17:25:52< Aethaeryn> probably should've said west of the Caspian sea
20090812 17:26:15< Aethaeryn> wanted to get the location in fast though so you know I didn't look.
20090812 17:26:48< Aethaeryn> Yep, it is Georgia, Armenia, and Azerbaijan in that area, clockwise.
20090812 17:27:13< Aethaeryn> I can name Europe, North Africa, and the Middle East without looking.
20090812 17:27:30< Aethaeryn> That's more than can be said for the continent I'm on (North America), I always screw up the central american countries.
20090812 17:27:50< Ivanovic> in school i had to learn africa, asia and america
20090812 17:28:01< Ivanovic> europe (where i live...) was implicated to be known
20090812 17:28:30< Ivanovic> with other words: screw the former sovjet states, there are too many of those!
20090812 17:28:55-!- Kenpachi [n=chatzill@CPE-58-170-82-51.sa.bigpond.net.au] has quit ["ChatZilla 0.9.85 [Firefox 3.0.13/2009073022]"]
20090812 17:29:13< Aethaeryn> Ivanovic: In the US they make you memorize the states... in multiple grades.
20090812 17:29:33< Aethaeryn> I had to learn Africa, but I think that's when I was living overseas in middle school. I don't recall doing it here in the States.
20090812 17:29:35< Ivanovic> brrr
20090812 17:29:58< Ivanovic> in 5th and 6th grade it was really *ugh* here
20090812 17:29:59< Aethaeryn> Of course, the only places I know flawlessly (and thus the only places I'll count) are the U.S. states, European countries, and the Middle East (counting North Africa)
20090812 17:30:13< Aethaeryn> I guess I accepted a B in the other regions ;)
20090812 17:30:33< Ivanovic> we had to be able to fill in blank maps (where just the borders and place of capitals were marked) with the respective state names and the names of the capitals
20090812 17:30:42< Aethaeryn> I know most of Asia, but the former Soviet republics in central area really screw me up.
20090812 17:31:00< Aethaeryn> In other words, they would've been better off staying with Russia ;)
20090812 17:31:21< Ivanovic> for most country names i at least have a rough idea where to look on a map, this is by far above average and enough for me
20090812 17:31:27< Aethaeryn> Hmm, yeah, I don't remember the capitals anymore.
20090812 17:31:37< Aethaeryn> I'm sure I could relearn 'em
20090812 17:31:47< Aethaeryn> I'm good enough at geography that if I had to memorize the whole world, I could probably do it.
20090812 17:31:49< Ivanovic> when i hear the names i know roughly where on the world to place them
20090812 17:32:12 * Ivanovic really sucks at learning things by heart
20090812 17:32:22< Ivanovic> i prefer understanding stuff and seeing connections
20090812 17:32:26< Aethaeryn> Oh, I know them because I love looking at maps to supplement history/etc.
20090812 17:32:45 * loonycyborg once knew all countries in the world but by now probably already forgotten a few obscure ones
20090812 17:32:47< Aethaeryn> I didn't really memorize most of the places I know, I can just logic it out.
20090812 17:33:11< Ivanovic> okay, to make it interesting: where is lichtenstein located?
20090812 17:33:27< Aethaeryn> the dot between austria and switzerland
20090812 17:33:30< Aethaeryn> :P
20090812 17:33:36< Ivanovic> :)
20090812 17:33:57< Aethaeryn> Where's San Marino?
20090812 17:34:26< Ivanovic> italian border
20090812 17:34:30-!- melinath [n=melinath@pool-72-72-193-179.altnpa.east.verizon.net] has quit [Read error: 113 (No route to host)]
20090812 17:34:34< Ivanovic> bhutan
20090812 17:34:34< Aethaeryn> not quite border, Ivanovic 
20090812 17:34:44< Aethaeryn> it's in the east of Italy, but is landlocked and not on a border...
20090812 17:35:00< Aethaeryn> they sheltered the Italian nationalists or something along those lines, so they were allowed to keep their state.
20090812 17:35:14< Aethaeryn> Ivanovic: north of India?
20090812 17:35:16< Ivanovic> in fact it is some border
20090812 17:35:21< Ivanovic> north east
20090812 17:35:27< Ivanovic> north of bangladesh
20090812 17:35:50< Ivanovic> ah, i always loved this one: where is "Chad"?
20090812 17:35:56< Aethaeryn> Ivanovic: west of Sudan
20090812 17:36:07< Aethaeryn> south of libya
20090812 17:36:14< Ivanovic> jupp
20090812 17:36:17< Aethaeryn> they fought with Libya over a disputed territory in the north for a while
20090812 17:36:29< Ivanovic> just dessert anyway...
20090812 17:36:29< Aethaeryn> (I like randomly reading up about things on Wikipedia, which helps with geography)
20090812 17:36:53< Aethaeryn> Only some, though. That's how I know about San Marino
20090812 17:36:55< Aethaeryn> Ivanovic: http://en.wikipedia.org/wiki/File:SAN_MARINO-en.png
20090812 17:37:00< Ivanovic> hmm, the whole collection of whateverSTAN is too easy (as in former sowjet, south of rusland)
20090812 17:37:04< Aethaeryn> btw, San Marino is on the border with Italy and... San Marino
20090812 17:37:08< Ivanovic> s/rusland/russi
20090812 17:37:11< Aethaeryn> It's surrounded by italy, like Vatican City
20090812 17:37:20< Ivanovic> like i said, italian border
20090812 17:37:25< Ivanovic> completely italian border
20090812 17:37:41< Aethaeryn> :P
20090812 17:38:00< Aethaeryn> Ivanovic: the collection of STANs are easy, but only if you memorize them
20090812 17:38:11< Aethaeryn> I've memorized and forgot them at least twice, remembering only Kazakhstan
20090812 17:38:16< Ivanovic> it is easy enough to roughly say where they are
20090812 17:38:19< Aethaeryn> They're just so insignificant.
20090812 17:38:47< Aethaeryn> The only thing important to know about them is that they're continuing the Stalinistic policy of draining lakes to fertilize the desert...
20090812 17:38:56< Aethaeryn> Which only helps destroy the lakes.
20090812 17:38:56< Ivanovic> hmm, okay, lets make it difficult (at least for US people): where is germany?
20090812 17:39:12< Ivanovic> and no, it does not share a border with the iran or russia
20090812 17:39:14< Aethaeryn> XFD
20090812 17:39:15< Ivanovic> ;)
20090812 17:39:20< Aethaeryn> Germany is between France and Poland
20090812 17:39:24< Aethaeryn> And south of Denmark
20090812 17:39:27< Aethaeryn> and north of Switzerland
20090812 17:39:34< Aethaeryn> It's very easy to know, for anyone who knows about either world war
20090812 17:40:03< Ivanovic> yes, it really is
20090812 17:40:08< Aethaeryn> Germany is between East Germany and France (old map) ;)
20090812 17:40:29< Ivanovic> the matter is that some *large* us news network had a slightly wrong map in some news
20090812 17:40:53< Ivanovic> placing germany with a border to iraq
20090812 17:40:56< Aethaeryn> ...
20090812 17:40:58< Aethaeryn> News...
20090812 17:40:59< Ivanovic> (or was it iran?)
20090812 17:41:04< Aethaeryn> People trust American TV news?
20090812 17:41:12< Aethaeryn> It's more entertainment than actual news.
20090812 17:41:13< Ivanovic> mr bush did!
20090812 17:41:14< loonycyborg> I wonder how did they manage to do that :P
20090812 17:41:15< Ivanovic> ^^
20090812 17:41:29< Aethaeryn> if something big happens (healthcare, healthcare, healthcare).... they'll cover only it
20090812 17:41:46< Aethaeryn> And ignore anything else of significance, except maybe a 30 minute summary at the top of the hour
20090812 17:41:51< Ivanovic> Aethaeryn: if you take the right "old" map your categorisation of "between east germany and france" is completely wrong
20090812 17:42:00< Aethaeryn> And they use buzz words, this week's is "town hall meeting"
20090812 17:42:02< Ivanovic> okay, the poland and france one, too, at that time
20090812 17:42:13< Aethaeryn> andway, brb
20090812 17:42:19< Aethaeryn> *anyway
20090812 17:43:01< Ivanovic> wait, found the problematic image: http://funfire.de/bilder/399/irak-oder-deutschland.jpg
20090812 17:43:03< loonycyborg> right "old" map categorisation is between west germany and poland :P
20090812 17:43:06< Ivanovic> instead of germany it was iraq
20090812 17:43:29< Ivanovic> since there are many us troops in germany and iraq it is somehow valid, isn't it?
20090812 17:43:55< loonycyborg> They just renamed Germany to Iraq.
20090812 17:44:05< Ivanovic> bad enough
20090812 17:44:34< boucman> hehe
20090812 17:45:06< Ivanovic> at that time i was already thinking "hmm, will the US troops find the correct targets to bomb, or are they using the same maps?"
20090812 17:45:08< Ivanovic> ;)
20090812 17:49:36< CIA-62> crab * r37698 /trunk/src/ (ai/actions.cpp ai/manager.cpp ai/manager.hpp game_events.cpp): wip on ai aspect modification via [modify_side], plus fix of one of the assertion failures after recruitment on choose-your-recruits era
20090812 17:49:44< Crab_> Soliton: ^ Ok, I fixed 1st segfault on choose-your-recruits map, but there's another one, in replay code, http://wesnoth.pastebin.com/m151efe4b
20090812 17:49:46< Aethaeryn> loonycyborg: but East Germany doesn't exist anymore
20090812 17:49:53< Aethaeryn> It just disappeared when the communists went away
20090812 17:50:01< Aethaeryn> Now there's a big void in the map of Europe ;)
20090812 17:50:32< Crab_> Soliton: strangely, it is only present in debug mode :)
20090812 17:50:55< Aethaeryn> Ivanovic: Yes, the US bombed and is occupying the Czech Republic...
20090812 17:51:08< Aethaeryn> Unfortunately, no one has been alerted to the error.
20090812 17:51:59-!- Aethaeryn [n=Michael@wesnoth/umc-dev/developer/aethaeryn] has quit ["bbl"]
20090812 17:53:35< loonycyborg> No. They're using Czech Republic as staging ground in their assault on Iraq :P
20090812 17:54:06< Soliton> Crab_: yeah, that function only gets called with mp_debug set. it's very untested i guess.
20090812 17:54:42< Soliton> that's a special debug mode you only get with -d.
20090812 17:55:23< Crab_> Soliton: should I submit a bug report ? whom to bug ?
20090812 17:55:39< Soliton> Xan - MIA
20090812 17:55:56< Soliton> a bug report would be good though, i suppose.
20090812 17:57:27< Soliton> probably just needs a check whether the unit is still there before doing the checksum.
20090812 17:57:56< Crab_> Soliton: well, note that any other unit can be there at that moment :)
20090812 17:58:04< Crab_> Soliton: the WML can, say, replace a unit with another
20090812 18:13:33< boucman> bye
20090812 18:13:42-!- boucman [i=43fa2d5d@wesnoth/developer/boucman] has quit ["Page closed"]
20090812 18:23:14-!- YogiHH [n=chatzill@d083114.adsl.hansenet.de] has joined #wesnoth-dev
20090812 18:36:20-!- stikonas [n=and@ctv-213-164-121-103.vinita.lt] has joined #wesnoth-dev
20090812 18:40:42-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20090812 18:47:45-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090812 18:48:51-!- Elvish_Pillager [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090812 18:51:00-!- EdB [n=edb@79.95.12.150] has joined #wesnoth-dev
20090812 18:54:57-!- giusef [n=giusef@unaffiliated/giusef] has joined #wesnoth-dev
20090812 18:58:42-!- cib0 [n=cib@p5DC43377.dip.t-dialin.net] has joined #wesnoth-dev
20090812 19:22:08-!- mordante [n=mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20090812 19:22:21< mordante> servus
20090812 19:22:46-!- noy [n=Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090812 19:27:11< mordante> what's the status of the savegame breakage?
20090812 19:27:23< mordante> (due to the unit rename action
20090812 19:27:24< mordante> )
20090812 19:28:40< Soliton> nothing new afaik. what's the problem with it?
20090812 19:28:49-!- cjhopman [n=chris@wesnoth/developer/cjhopman] has joined #wesnoth-dev
20090812 19:29:18< mordante> IMO it's a bad idea to break savegames, since it can hinder fixing bugs
20090812 19:29:37< mordante> so it should only be done if there's a good reason
20090812 19:29:57< Soliton> well, how can you not break them with a unit rename?
20090812 19:30:03< mordante> hi cjhopman long time no see
20090812 19:30:59< mordante> of course it's hard to do when you rename a unit, but why was the unit rename needed in the first place?
20090812 19:31:20< mordante> and of course you can hardcode map the old name to the new one in the code for a certain time
20090812 19:31:50< Soliton> the old name didn't fit in with the role of the unit.
20090812 19:32:04< Soliton> there's a thread on the forum about it.
20090812 19:32:26< Soliton> http://www.wesnoth.org/forum/viewtopic.php?f=32&t=26165
20090812 19:38:02< Soliton> if the renaming wouldn't involve giving an existing name to another unit savegame fixing could even easily be done by wmllint. maybe that will be rectified soon.
20090812 19:39:48< mordante> I glimpsed over the thread and too long to read entirely, but I assume there are good reasons for it
20090812 19:40:01< mordante> but we can do better as breaking the savegames 
20090812 19:40:20< mordante> and I prefer an engine work around for two releases, as with our usual policy
20090812 19:42:58< Soliton> well, as i just mentioned that's not really possible.
20090812 19:43:33-!- Valkier [n=karasawa@c-71-225-240-156.hsd1.pa.comcast.net] has joined #wesnoth-dev
20090812 19:43:39-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20090812 19:43:47< mordante> why not, did another unit get the old name?
20090812 20:07:38-!- Crab_ [i=crab@wesnoth/developer/crab] has left #wesnoth-dev []
20090812 20:21:54-!- noy [n=Noy@wesnoth/developer/noy] has quit []
20090812 20:40:22-!- giusef [n=giusef@unaffiliated/giusef] has quit ["exit (-1);"]
20090812 20:51:57-!- EdB [n=edb@79.95.12.150] has quit [Remote closed the connection]
20090812 20:58:43-!- Noyga [n=lame-z@wesnoth/developer/noyga] has joined #wesnoth-dev
20090812 20:59:53-!- Noyga [n=lame-z@wesnoth/developer/noyga] has left #wesnoth-dev ["Quitte"]
20090812 21:10:09-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has joined #wesnoth-dev
20090812 21:18:01< CIA-62> thespaceinvader * r37699 /trunk/ (47 files in 3 dirs): Add and wire new Drake Fighter frames. Remove old unused frames. Update changelogs.
20090812 21:21:58-!- lizard_r [n=Miranda@wesnoth/umc-dev/developer/lizard] has quit ["Saurian Augur - I'll heal you by 4 hp if you leave next to me"]
20090812 21:23:57-!- YogiHH_ [n=chatzill@d052095.adsl.hansenet.de] has joined #wesnoth-dev
20090812 21:28:20-!- YogiHH [n=chatzill@d083114.adsl.hansenet.de] has quit [Read error: 60 (Operation timed out)]
20090812 21:28:24-!- YogiHH_ is now known as YogiHH
20090812 21:32:05-!- kitty_ [n=kitty@wesnoth/artist/kitty] has joined #wesnoth-dev
20090812 21:33:11< thespaceinvader> hey kitty_ 
20090812 21:33:20< kitty_> hi
20090812 21:42:58-!- Zarel [i=Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20090812 21:58:12-!- Crab_ [i=crab@wesnoth/developer/crab] has joined #wesnoth-dev
20090812 21:58:18-!- Crab_ [i=crab@wesnoth/developer/crab] has left #wesnoth-dev []
20090812 22:04:50-!- stikonas_ [n=and@ctv-213-164-121-103.vinita.lt] has joined #wesnoth-dev
20090812 22:15:57-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 110 (Connection timed out)]
20090812 22:18:40-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 104 (Connection reset by peer)]
20090812 22:18:55-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090812 22:24:43-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Connection timed out]
20090812 22:25:08-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090812 22:30:19-!- Crab_ [i=crab@wesnoth/developer/crab] has joined #wesnoth-dev
20090812 22:30:58< Valkier> Hey Kitty, Nice asp!
20090812 22:31:05 * Valkier falls over laughing
20090812 22:34:57< Zarel> ...I don't get it. :(
20090812 22:37:11< Valkier> Asp is a snake that as a spelling remarkably similair to ass..
20090812 22:37:19< Valkier> and she's drawing a naga.
20090812 22:37:30 * Zarel falls over laughing
20090812 22:37:35< Valkier> I know!!!
20090812 22:37:47< Zarel> I can't see, because forum is overloaded. :(
20090812 22:38:14< Valkier> Oh noes!
20090812 22:38:33< Zarel> Oh, hey, it works agai. :D
20090812 22:38:39< Valkier> ZOMG!
20090812 22:38:55< Valkier> Guess I'm gonna finish up the revenant.
20090812 22:39:17< kitty_> hmpf - ha!
20090812 22:39:55< Valkier> That's about the reaction I was expecting.
20090812 22:40:36< Valkier> I'm surprised you're letting me get away with the blood smears, Kitty.
20090812 22:40:57< kitty_> dry blood is ok - just no gore.
20090812 22:41:08< Zarel> Gore, on a skeleton?
20090812 22:41:20< Valkier> Who'd have thunk it?!
20090812 22:41:27< Zarel> That's like... like legs on a naga!
20090812 22:41:38< kitty_> there are many depictions with remains of rotten flesh and the like out theere...
20090812 22:41:42< Valkier> Well.. that does make sense...
20090812 22:41:43< Zarel> True.
20090812 22:42:08< Zarel> But blood smears are easily plasible-denied. "It's red paint!"
20090812 22:42:30< kitty_> :P
20090812 22:42:34< Zarel> Which is technically true.
20090812 22:42:54< Zarel> I mean, you could use the paintbrush tool in Photoshop and set it to "red"
20090812 22:43:01< Valkier> The Draug is going to have rat holding someones eye up for him to inspect
20090812 22:43:03< Valkier> Is this cool?
20090812 22:43:16< kitty_> what? you joking?
20090812 22:43:30< Valkier> I wish I wasn't...
20090812 22:43:46< Valkier> But I know you guys would never let me get away with that.
20090812 22:43:57< kitty_> it sounds awesome in a way but also terribly silly... show a sketch!
20090812 22:44:17< Valkier> Seriously? Gonna let me do something like that if I can make it look good?
20090812 22:44:24< Valkier> Cause I'll do it 0_0
20090812 22:44:46< kitty_> not if you make it good - if you make it at least somewhat serious...
20090812 22:45:12< Valkier> I will draw the serious thing well!
20090812 22:45:35< kitty_> i look forward to seeing if you can pull that off!
20090812 22:45:58< Valkier> We should get a voice chat channel or something for Wesnoth.
20090812 22:46:09< Valkier> I keep switching from outlining to typing :-P
20090812 22:46:38< kitty_> oh no - my pronouciation is worse than my spelling... and i'm outlining now, too...
20090812 22:46:56< Valkier> Pfft. I've been speaking to your Geutschlanders since I was 10.
20090812 22:47:02< Valkier> Deutschlanders*
20090812 22:47:17< Valkier> Not necessarily in the correct language, but you get the point.
20090812 23:05:19-!- allefant [n=allefant@allegro/developer/allefant] has quit [Read error: 104 (Connection reset by peer)]
20090812 23:05:46-!- silene [n=plouf@wesnoth/developer/silene] has quit [Read error: 110 (Connection timed out)]
20090812 23:06:39-!- silene [n=plouf@AMarseille-251-1-64-135.w83-205.abo.wanadoo.fr] has joined #wesnoth-dev
20090812 23:06:41-!- fendrin [n=fabi@g228014154.adsl.alicedsl.de] has joined #wesnoth-dev
20090812 23:06:46-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has quit [Remote closed the connection]
20090812 23:09:27-!- stikonas_ is now known as stikonas
20090812 23:21:01-!- allefant [n=allefant@allegro/developer/allefant] has joined #wesnoth-dev
20090812 23:23:17< Valkier> bah! I hate doing outlines >_<
20090812 23:25:16-!- zookeeper [n=l@wesnoth/developer/zookeeper] has quit []
20090812 23:26:04< Crab_> silene: hello. a small question. Is it ok to add "#define reverse_foreach BOOST_REVERSE_FOREACH " to src/foreach.hpp ?
20090812 23:26:45< mordante> Crab_, reverse_foreach is 1.35, isn't it?
20090812 23:26:51< silene> Crab_: it is, but please check from which version of boost it comes from
20090812 23:27:32< silene> (that doesn't ring a bell, so it wasn't there during the review of the original boost foreach)
20090812 23:27:42< CIA-62> mordante * r37700 /trunk/src/ai/manager.cpp: Fix a compiler warning.
20090812 23:27:46< CIA-62> mordante * r37701 /trunk/src/formula_debugger.cpp: Fix a compiler warning.
20090812 23:27:48< mordante> I'm sure it was added later, I think 1.35
20090812 23:28:32< Crab_> unfortunately, it seems that it's from 1.36.   ok, will just work around it.
20090812 23:30:42< CIA-62> loonycyborg * r37702 /trunk/src/server/server.cpp: Fixed a compile error on mingw.
20090812 23:33:03-!- melinath [n=melinath@pool-72-72-193-179.altnpa.east.verizon.net] has joined #wesnoth-dev
20090812 23:41:35-!- silene [n=plouf@AMarseille-251-1-64-135.w83-205.abo.wanadoo.fr] has quit ["Leaving."]
20090812 23:53:08-!- xchat_sux [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090812 23:53:58-!- Appleman1234 [n=Appleman@131.181.102.98] has joined #wesnoth-dev
20090812 23:55:04< Valkier> Hurray! Done with the revenant!
20090812 23:55:51< kitty_> Valkier: looks good :) but there are a couple of smears around his legs/arm/axe - could you clean those up?
20090812 23:56:01< Valkier> Yeah. Just noticed that
20090812 23:56:06< Valkier> And the answer is NO! mwahahha!
20090812 23:56:14< Valkier> But I will.. I guess..
20090812 23:56:14< kitty_> sigh
20090812 23:58:35< Valkier> No clue where all these smudges came from weird.
--- Log closed Thu Aug 13 00:00:44 2009