--- Log opened Tue Mar 23 00:00:01 2010
20100323 00:02:07< Ivanovic> schumi: updated that list
20100323 00:02:10< Ivanovic> off to bed now
20100323 00:05:11< fendrin> Crab_: Yeah, then let's keep it in the plugin for now.
20100323 00:05:18< Crab_> ok
20100323 00:06:39< Crab_> schumi: done
20100323 00:10:55< Crab_> fendrin: what are your code style conventions for java ? same as for wesnoth's c++ ?
20100323 00:11:10< Crab_> (except the naming, of course) 
20100323 00:11:22< fendrin> Crab_: CTRL + I 
20100323 00:11:44< Crab_> 'eclipse defaults', that is?
20100323 00:11:50-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Quit: This computer has gone to sleep]
20100323 00:11:51< fendrin> right
20100323 00:12:11< fendrin> but we can change that if you like.
20100323 00:12:25< Crab_> np with that
20100323 00:14:46< CIA-80> caslav_ilic * r41694 /trunk/data/ (4 files in 4 dirs): Updated localized images for Hungarian.
20100323 00:18:35-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100323 00:18:58-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has quit [Read error: Connection reset by peer]
20100323 00:26:03< Soliton> fendrin, Crab_: any reason why you put your plugin or whatever it is in the resources branch instead of just under utils/?
20100323 00:27:16< Crab_> Soliton: no specific reason, except that for now the code is not tied to specific wesnoth release. noone seemed to object when we were discussing where to put it.
20100323 00:27:35< Crab_> Soliton: we can relocate it, np for me
20100323 00:29:03< Soliton> it doesn't seem to fit into the resource branch at all from what i can tell. and trunk isn't specific to a wesnoth release unlike some branches. ;-)
20100323 00:30:47< Crab_> so, you're suggesting to put it into trunk/utils/eclipse_plugin or trunk/utils/java/eclipse_plugin ?
20100323 00:30:54< Crab_> fendrin: what do you think about it?
20100323 00:31:16< Soliton> don't mind either way.
20100323 00:33:33-!- thespaceinvader [~chatzilla@wesnoth/artist/thespaceinvader] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6/20100115144158]]
20100323 00:33:42-!- LucaMoller [~a1184ae9@gateway/web/freenode/x-bqyifverzbriobed] has joined #wesnoth-dev
20100323 00:34:45< CIA-80> ai0867 * r41696 /trunk/data/schema.cfg: Some use of the new [description] tag in schemas. category attribute is still unused
20100323 00:34:50< CIA-80> ai0867 * r41695 /trunk/data/tools/wesnoth/wmlgrammar.py: Add some rudimentary support for documentation embedded in a schema
20100323 00:35:46< fendrin> Crab_: I am fine with moving to trunk. But after the release. 
20100323 00:35:53< Crab_> ok
20100323 00:36:03< fendrin> Crab_: I don't dare to messup trunk right before a planned release.
20100323 00:36:21< Crab_> there's no reason to hurry
20100323 00:41:55< Soliton> how in the world would you do that?
20100323 00:43:09< Crab_> Soliton: svn move ?
20100323 00:43:15< Crab_> Soliton: from url1 to url2
20100323 00:43:28< Crab_> or you're about the trunk messup ?
20100323 00:44:18< Soliton> the latter.
20100323 00:44:38-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100323 00:47:42-!- ilor [~ilor@wesnoth/developer/ilor] has quit [Ping timeout: 252 seconds]
20100323 00:48:50-!- General_Fou_Fou_ [~Sarah@rrcs-69-193-74-3.nys.biz.rr.com] has joined #wesnoth-dev
20100323 00:49:52-!- General_Fou_Fou_ [~Sarah@rrcs-69-193-74-3.nys.biz.rr.com] has quit [Client Quit]
20100323 00:51:58-!- General_Fou_Fou [~Sarah@rrcs-69-193-74-3.nys.biz.rr.com] has quit [Ping timeout: 264 seconds]
20100323 00:59:34-!- UnknownDevice [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100323 00:59:44-!- Appleman1234 [~Appleman1@131.181.47.1] has quit [Ping timeout: 246 seconds]
20100323 01:01:13< UnknownDevice> Hello
20100323 01:01:20< fendrin> Soliton: I don't think that the move would hurt wesnoth in any way. But I don't see the point to commit a nonworking piece of software into a stable release. I guess Ivanovic would complain.
20100323 01:02:16< Crab_> yeah, 'finally, we are trying to show some respect to the feature freeze' :)))
20100323 01:02:40< Crab_> UnknownDevice: hello
20100323 01:03:13< UnknownDevice> I just came here to introduce myself as I'm interested in applying for the google summer of code.
20100323 01:03:22< Soliton> there is no freeze on utils/ and i don't see the point in developing something where it clearly doesn't belong.
20100323 01:03:32-!- migge_ [~marc@ip-95-223-247-31.unitymediagroup.de] has quit [Read error: Operation timed out]
20100323 01:04:46< Soliton> it'd still be interesting to know why you thought it fits there to prevent future mix-ups.
20100323 01:05:01-!- migge [~marc@ip-95-223-247-31.unitymediagroup.de] has joined #wesnoth-dev
20100323 01:05:07< UnknownDevice> So hello :P
20100323 01:05:38< Soliton> hi UnknownDevice
20100323 01:06:17< Crab_> Soliton: I personally thought that resources is for various data which doesn't fit anywhere else, like unused or old images or files.
20100323 01:06:24< UnknownDevice> So does everyone here work on the game? Would anyone like to know anything about me? 
20100323 01:07:01< Soliton> Crab_: a plugin isn't data, no?
20100323 01:07:53< Crab_> Soliton: well, everything is data, if no better classification can be found
20100323 01:08:16-!- haoyu [~bhy@cm26.delta25.maxonline.com.sg] has joined #wesnoth-dev
20100323 01:12:32< Crab_> UnknownDevice: yes, at least some of the people there work on the game
20100323 01:12:40< UnknownDevice> Ah excellent
20100323 01:13:18< Crab_> UnknownDevice: well, if you want, you can tell us about you. it's actually better to write this in the wiki, since things written in the irc tend to get lost from time to time :)
20100323 01:13:58< Crab_> UnknownDevice: have you picked a project which interests you most ? or, maybe, you'd like to propose your own ?
20100323 01:14:01< UnknownDevice> Alrighty, I'll do that tomorrow then because it's 12am now
20100323 01:14:09< Crab_> 2:14 here, heh :)
20100323 01:14:28< Crab_> UnknownDevice: ok, so thanks for saying hello :)
20100323 01:14:46< UnknownDevice> Hmm well I think it's better to pick one of the suggested ones because the people who suggested them probably know what more important then me.
20100323 01:15:48< Crab_> it depends :) so, what do you think about suggested projects ? or, you'll ready through them later ?
20100323 01:15:54< Crab_> s/ready/read
20100323 01:15:54-!- UnknownDevice__ [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100323 01:16:19< UnknownDevice__> Out of the suggested ones I don't know which one I'm going to pick yet. I'm pretty confident I could any of them in the time frame.
20100323 01:16:36< UnknownDevice__> So I really have to look at them more before I officially pick one.
20100323 01:16:42< Crab_> and quite confident that you can do them better than the other candidates :) ?
20100323 01:16:52< UnknownDevice__> Oh no! XD
20100323 01:17:04< UnknownDevice__> I've no idea what everyone else is like
20100323 01:17:23< Crab_> ok, so good luck picking the project for you :))
20100323 01:17:32< UnknownDevice__> Just from reading the suggested list there's nothing there I wouldn't have some idea where to start.
20100323 01:17:42< Crab_> having ideas is nice :)
20100323 01:17:50< UnknownDevice__> haha yes it is 
20100323 01:18:28< UnknownDevice__> Anything from the list that you think is more important then the rest? 
20100323 01:19:14-!- UnknownDevice [~UnknownDe@188.141.93.69] has quit [Ping timeout: 248 seconds]
20100323 01:19:20< Crab_> quite hard to say
20100323 01:19:30< Crab_> each task is somehow important
20100323 01:20:19< UnknownDevice__> Actually I think the only one I have no idea how to do is the network one
20100323 01:20:25< UnknownDevice__> I don't really know much about that area.
20100323 01:20:33< Crab_> yes, it's highly specialized
20100323 01:20:54< UnknownDevice__> In my opinion I think probably the most important ones are supporting sprite sheets or the input framework
20100323 01:21:27< UnknownDevice__> You really notice the touch controls don't work too well on the iPhone when you play the game. 
20100323 01:21:34< Crab_> then try to catch those developers who know about them, online
20100323 01:21:39< UnknownDevice__> Or atleast that's the feeling I got from playing it.
20100323 01:21:48< UnknownDevice__> You don't develop yourself/ 
20100323 01:23:27-!- Appleman1234 [~Appleman1@131.181.103.236] has joined #wesnoth-dev
20100323 01:24:03-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100323 01:25:06-!- qemqemqem_ [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has quit [Ping timeout: 252 seconds]
20100323 01:25:40< Gambit> I thought the game already supported spritesheets with ~CROP
20100323 01:26:03< Crab_> Gambit: yes, there was a discussion about that on the ML
20100323 01:26:04< CIA-80> crab * r41697 /branches/resources/eclipse_plugin/src/org/ (17 files in 6 dirs): prototype of interface for WMl schema
20100323 01:26:10< Crab_> fendrin: ^
20100323 01:26:16< UnknownDevice__> Hmm, it's on the suggest ideas. So is that incorrect? :(
20100323 01:26:35< Crab_> UnknownDevice__: see the mailing list. it is more than just '~CROP'
20100323 01:26:52< Gambit> Though it would be tedious naming coordinates.
20100323 01:27:04< fendrin> Crab_: Thank you :-)
20100323 01:27:32< Gambit> So a system that automatically cropped sheets into 72x72 pieces and then allowed them to reference "frame 1" would be better.
20100323 01:27:37< Crab_> ah, https://mail.gna.org/public/wesnoth-dev/2010-03/msg00034.html
20100323 01:28:03< Crab_> fendrin: SchemaFactory.getSchemaForTag("campaign") should return a dummy object for you.
20100323 01:28:08< UnknownDevice__> Thanks Crab_ :)
20100323 01:28:24< fendrin> Crab_: very nice.
20100323 01:29:07< UnknownDevice__> Haha
20100323 01:29:16< UnknownDevice__> That's a pretty interesting way of doing it
20100323 01:29:23< UnknownDevice__> I would love to do something like this! :)
20100323 01:31:01-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Quit: Leaving]
20100323 01:31:10-!- UnknownDevice___ [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100323 01:31:30< UnknownDevice___> <UnknownDevice__> Thanks Crab_ for linking me to this, gives me a way better idea of what's expected.
20100323 01:31:37< UnknownDevice___> Just incase you didn't get that
20100323 01:31:48< UnknownDevice___> Sorry my connection is just being really buggy tonight.
20100323 01:31:49< Crab_> (yes, I understood)
20100323 01:32:09< Crab_> np with connection problems, it happens sometimes
20100323 01:32:38< UnknownDevice___> I think they must be doing some kind of work on the line because it's be happening every night this week around 12am
20100323 01:34:25-!- UnknownDevice__ [~UnknownDe@188.141.93.69] has quit [Ping timeout: 246 seconds]
20100323 01:35:32< UnknownDevice___> Hey Crab_ I was just wondering, from the link to the forum post does that mean they're working to a degree, but they want someone to build on that? 
20100323 01:35:40< Crab_> fendrin: in a day or two, I'll add a generation of the schema from AI0867's schema.cfg (either via a proper WML parsing or via 'regexps+state machine' hacks)
20100323 01:35:52< UnknownDevice___> Sorry for asking what might be a stupid question, it's just that I don't want to pick it up wrong.
20100323 01:37:13< Crab_> UnknownDevice___: yes, you're right, the support is there but is not used to its full potential.
20100323 01:37:39< UnknownDevice___> Ah excellent, that makes it more clear for me then, thanks again.
20100323 01:38:19< UnknownDevice___> Sorry again if it seems like I'm asking stupid questions, I just misread something and have a completely wrong idea of what's going on :P
20100323 01:38:30< UnknownDevice___> *I just don't want to
20100323 01:41:16< CIA-80> caslav_ilic * r41698 /trunk/ (26 files in 13 dirs): Updated localized images for Serbian.
20100323 01:45:00-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100323 01:46:57< UnknownDevice___> I've gota now anyway, thanks for all the help Crab_!
20100323 01:47:06< Crab_> :)
20100323 01:47:27< UnknownDevice___> See you tomorrow hopefully if you're around.
20100323 01:47:37-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz]
20100323 01:47:41-!- UnknownDevice___ [~UnknownDe@188.141.93.69] has quit [Quit: Leaving]
20100323 01:48:10-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20100323 01:55:36< CIA-80> crab * r41699 /branches/resources/eclipse_plugin/src/org/wesnoth/wml/ (4 files in 3 dirs): updated WML schema to account for the fact that attributes refer to types, not to individual attributes.
20100323 01:55:41< Crab_> fendrin: ^ sorry, I haven't got this right the first time, a slightly updated version.
20100323 01:57:17< Crab_> e.g. in the schema, it's not ' abbrev="required abbrev"   ', but  '  abbrev="required identifier"  '
20100323 01:58:24-!- General_Fou_Fou [~Sarah@cpe-76-179-189-81.maine.res.rr.com] has joined #wesnoth-dev
20100323 02:00:27< CIA-80> crab * r41700 /branches/resources/eclipse_plugin/src/org/wesnoth/wml/ (core/SchemaFactory.java schema/SchemaChild.java): a minor readability change in the eclipse plugin
20100323 02:11:22< Espreon> Meh, you children and your schemata...
20100323 02:11:57< shadowmaster> what the hell is up with these people who make posts that don't contain anything but a smiley? At this rate the forum will turn into an IRC replica
20100323 02:12:10< Espreon> Indeed.
20100323 02:18:55-!- quaternary [~karol@87-205-197-38.adsl.inetia.pl] has joined #wesnoth-dev
20100323 02:24:53-!- Blueblaze [~nick@adsl-76-202-22-67.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100323 02:26:39< fendrin> Crab_: Ah , okay
20100323 02:26:45< Crab_> ok
20100323 02:30:10< quaternary> Hi
20100323 02:30:17< fendrin> hi
20100323 02:30:22< Crab_> hi, quaternary
20100323 02:31:50< quaternary> I would like to participate in GSoC and your game seems to be quite interesting. I would like to know what projects you would propose and what are your priorities.
20100323 02:32:00< fendrin> Crab_: Don't you like to call interfaces with a name that begins with "I" =
20100323 02:32:02< fendrin> ?
20100323 02:33:04< Crab_> fendrin: no, not much - I prefer the simplest possible unique name for the interface, as I usually work with them
20100323 02:33:22< quaternary> I know C, C++, java and ruby and Im particularly interested in AI.
20100323 02:33:46< Crab_> quaternary: thanks for your interest. the list of projects is located in our wiki, http://wiki.wesnoth.org/SummerOfCodeIdeas
20100323 02:34:22< Crab_> quaternary: there's one AI-related project there (yes, AI projects usually attract somewhat more interest)
20100323 02:34:57< fendrin> quaternary: Are you familiar with eclipse plugin development?
20100323 02:35:05< Crab_> quaternary: the project is extending lua bindings for wesnoth ai and using those bindings to rewrite several areas of current wesnoth ai, which are obsolete
20100323 02:35:21< Crab_> quaternary: and there's java-related work, too.
20100323 02:45:03< Crab_> night
20100323 02:45:16< quaternary> goodnight
20100323 02:45:20< gabm> Crab_: night
20100323 02:45:33-!- Crab_ [~Crab_@wesnoth/developer/crab] has left #wesnoth-dev []
20100323 02:48:52< quaternary> fendrin: No, Im afraid Im not.
20100323 02:50:29< quaternary> Sorry for the delay. I was reading about the lua project.
20100323 02:52:15-!- mpavel [~Pavel@cpc1-dund3-0-0-cust363.sgyl.cable.virginmedia.com] has left #wesnoth-dev []
20100323 02:53:26< quaternary> goodnight
20100323 02:53:28-!- quaternary [~karol@87-205-197-38.adsl.inetia.pl] has quit [Quit: leaving]
20100323 02:53:50-!- Silenthunter [~Silenthun@97-81-86-140.dhcp.athn.ga.charter.com] has left #wesnoth-dev ["Leaving"]
20100323 03:04:07-!- gabm [~gabm@205.240-ppp.3menatwork.com] has left #wesnoth-dev []
20100323 03:07:24-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100323 03:22:41-!- ancestral [~ancestral@mobile-166-137-136-102.mycingular.net] has joined #wesnoth-dev
20100323 03:25:27-!- ryaether [~ryan@173-26-127-85.client.mchsi.com] has joined #wesnoth-dev
20100323 03:25:45-!- Gambit [~Gambit@pa-67-234-116-244.dhcp.embarqhsd.net] has quit [Quit: Leaving.]
20100323 03:27:39-!- ancestral [~ancestral@mobile-166-137-136-102.mycingular.net] has quit [Read error: Connection reset by peer]
20100323 03:27:42-!- Appleman1234 [~Appleman1@131.181.103.236] has quit [Ping timeout: 258 seconds]
20100323 03:27:56-!- ancestral [~ancestral@mobile-166-137-136-102.mycingular.net] has joined #wesnoth-dev
20100323 03:35:11-!- Gambit [~Gambit@pa-67-234-116-244.dhcp.embarqhsd.net] has joined #wesnoth-dev
20100323 03:45:39-!- Appleman1234 [~Appleman1@131.181.47.9] has joined #wesnoth-dev
20100323 03:56:11-!- mrEPIC [~mrEPIC@adsl-70-242-0-245.dsl.hstntx.swbell.net] has joined #wesnoth-dev
20100323 03:58:06< mrEPIC> Good evening everyone! I have a quick question: I am in the process of translating my multiplayer campaign, The Haunted Woods. I will first translate it into (very basic) French, and then I will translate it to Chinese. My question is: for other people to be able to view the translation, do they need to have the add-on? In other words, is the translation sent over multiplayer?
20100323 03:58:32< mrEPIC> I also apologize if I should have asked this in the #wesnoth channel instead
20100323 03:59:27< mrEPIC> anyone?
20100323 04:03:04-!- ancestral [~ancestral@mobile-166-137-136-102.mycingular.net] has quit [Quit: Colloquy for iPhone - http://colloquy.mobi]
20100323 04:03:24< Appleman1234> mrEPIC, depends how you package the addon
20100323 04:04:38< mrEPIC> well, I looked at another addon for 1.6 (although my translated version is currently for 1.8), and they used the directory structure of having a translations folder, inside which are folders for the different languages (e.g. fr for french) which each contain a folder named "LC_MESSAGES"
20100323 04:04:49< mrEPIC> inside of this folder is the .mo file which has the translation
20100323 04:05:26< mrEPIC> My own translation is based off of this structure as well
20100323 04:05:32-!- schumi [~a11848c0@gateway/web/freenode/x-lkucxumntfxcywrx] has quit []
20100323 04:07:02< mrEPIC> specifically, my french translation is at {add-ons/The_Haunted_Woods/translations/fr/LC_MESSAGES/wesnoth-thw.mo
20100323 04:07:43-!- Appleman1234 [~Appleman1@131.181.47.9] has quit [Ping timeout: 246 seconds]
20100323 04:07:45-!- wesbot changed the topic of #wesnoth-dev to: 1.8 commits allowed till March, 24th, 12:00 GMT, afterwards "testfreeze" | string/feature freeze active! | 68 bugs, 254 feature requests, 14 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100323 04:07:54< mrEPIC> I am unsure if this is correct because it is my first time translating something for Wesnoth, but it is recognized by the game, so I would assume that it is at least partially correct
20100323 04:07:56-!- ancestral [~ancestral@97-116-113-184.mpls.qwest.net] has joined #wesnoth-dev
20100323 04:08:06< Espreon> mrEPIC: Uh, I hope that your textdomain doesn't reflect that mo's name.
20100323 04:08:25< mrEPIC> err well I have #textdomain wesnoth-thw
20100323 04:08:34< Espreon> IIRC, the textdomain name has to match the addon's directory name.
20100323 04:08:42< mrEPIC> ah ok
20100323 04:09:02< mrEPIC> it was still recognized by wesnoth and I can see my translation in game though
20100323 04:09:26< mrEPIC> but i'll change this immediately if it may cause unforeseen issues in the future
20100323 04:09:54< mrEPIC> if this is the case, what should the mo file be named?
20100323 04:09:54< Espreon> Meh, I guess that is just to have it work for Wescamp's services.
20100323 04:10:11 * Espreon shrugs
20100323 04:10:19< mrEPIC> I'm new to IRC as well
20100323 04:10:46< Espreon> Uh, what does that have to do with this?
20100323 04:11:01< mrEPIC> I just saw 'espreon shrugs'
20100323 04:11:06< Espreon> Oh.
20100323 04:11:08< mrEPIC> I'm not sure how to respond to that xD
20100323 04:11:18< Espreon> If you want to be able to do that, use /me
20100323 04:11:23 * mrEPIC test
20100323 04:11:25< mrEPIC> ah thanks
20100323 04:11:32< Espreon> You are welcome.
20100323 04:12:40< ancestral> Espreon has bestowed mrEPIC the power of the emote. Use it wisely.
20100323 04:12:47< Espreon> ...
20100323 04:13:01< ancestral> With great power comes great… just kidding
20100323 04:13:23-!- mrEPIC [~mrEPIC@adsl-70-242-0-245.dsl.hstntx.swbell.net] has quit [Read error: Connection reset by peer]
20100323 04:16:56-!- Ivanovic_ [~ivanovic@dtmd-4db2f9f7.pool.mediaWays.net] has joined #wesnoth-dev
20100323 04:18:54-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 252 seconds]
20100323 04:18:54-!- Ivanovic_ is now known as Ivanovic
20100323 04:19:50-!- Appleman1234 [~Appleman1@131.181.47.6] has joined #wesnoth-dev
20100323 04:29:58-!- Zarel [~Serei@c-75-72-160-179.hsd1.mn.comcast.net] has joined #wesnoth-dev
20100323 04:34:36-!- isaac__ is now known as isaac
20100323 04:38:10-!- Gambit [~Gambit@pa-67-234-116-244.dhcp.embarqhsd.net] has quit [Quit: Leaving.]
20100323 04:51:19-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100323 05:14:55-!- Mythological [Mythologic@77.28.113.253] has joined #wesnoth-dev
20100323 05:21:47-!- Mythological [Mythologic@77.28.113.253] has quit [Ping timeout: 260 seconds]
20100323 05:21:53-!- Mythological [Mythologic@77.28.113.253] has joined #wesnoth-dev
20100323 05:32:54< ancestral> Hmm, yeah what is attack depth?
20100323 05:41:59< Zarel> Hmm.
20100323 05:42:03< Zarel> Espreon: You there?
20100323 05:42:22< Zarel> Actually, different idea. noy: You there?
20100323 05:45:19-!- Espreon [~espreon@wesnoth/developer/espreon] has quit [Ping timeout: 258 seconds]
20100323 05:47:09-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Quit: crimson_penguin]
20100323 06:00:00< fendrin> ancestral: The maximal number of units that attack a single enemy unit. So it's highest meaningfull value would be 6.
20100323 06:06:53< ancestral> Hmm
20100323 06:09:56< noy> Zarel: yes?
20100323 06:10:14< Zarel> Might want to split the thread from here: http://www.wesnoth.org/forum/viewtopic.php?p=416811#p416811
20100323 06:12:04< Zarel> Oh, hmm, phpBB doesn't have a very good topic split mechanism. It might be difficult to split after so many posts. :/
20100323 06:12:13< noy> should I?
20100323 06:12:39< Zarel> Oh, wait, there's a "split topic from selected post onwards" option.
20100323 06:12:57< Zarel> Your call. If you defer to me, I'd say it's probably better to split.
20100323 06:13:36< Zarel> Into "Ideas". Call it "Proposal: Allow recalls of units that have died in campaign"
20100323 06:15:02< ancestral> Anyone ever played Planescape?
20100323 06:15:14< ancestral> Or a game where you get to re-summon when you die?
20100323 06:15:19< ancestral> Even SimAnt
20100323 06:15:33< ancestral> Could make for a pretty interesting Wesnoth scenario
20100323 06:26:05-!- mjs-de [~mjs-de@vpw.wh.Uni-Dortmund.DE] has joined #wesnoth-dev
20100323 06:28:17-!- FAAB [~huajie@219.142.251.73] has joined #wesnoth-dev
20100323 06:28:55-!- FAAB [~huajie@219.142.251.73] has left #wesnoth-dev []
20100323 06:30:17-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Quit: Leaving]
20100323 06:56:50-!- LucaMoller [~a1184ae9@gateway/web/freenode/x-bqyifverzbriobed] has quit [Quit: Page closed]
20100323 07:07:22-!- General_Fou_Fou [~Sarah@cpe-76-179-189-81.maine.res.rr.com] has quit [Quit: General_Fou_Fou]
20100323 07:15:37-!- Appleman1234 [~Appleman1@131.181.47.6] has quit [Ping timeout: 276 seconds]
20100323 07:25:20-!- ryaether [~ryan@173-26-127-85.client.mchsi.com] has left #wesnoth-dev []
20100323 07:33:52-!- ettin [~jorda@wesnoth/developer/ettin] has quit [Read error: Connection reset by peer]
20100323 07:37:57-!- ettin [~jorda@wesnoth/developer/ettin] has joined #wesnoth-dev
20100323 07:56:32-!- fendrin [~fabi@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20100323 07:56:57-!- fendrin [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20100323 08:00:00-!- ilor [~ilor@wesnoth/developer/ilor] has joined #wesnoth-dev
20100323 08:02:33-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20100323 08:32:18-!- ancestral [~ancestral@97-116-113-184.mpls.qwest.net] has quit [Quit: And that's the end of THAT chapter]
20100323 08:34:19-!- ancestral [~ancestral@97-116-113-184.mpls.qwest.net] has joined #wesnoth-dev
20100323 09:10:33-!- ancestral [~ancestral@97-116-113-184.mpls.qwest.net] has quit [Quit: And that’s the end of THAT chapter.]
20100323 09:14:35-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: GO, GET TO THE CHOPPAH!!!]
20100323 09:15:41-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100323 09:15:41-!- noy [~Noy@wesnoth/developer/noy] has quit [Client Quit]
20100323 09:19:06< ilor> does anyone recall the approximate dates when Wesnoth started using cvs and then svn?
20100323 09:23:28< Rhonda> ilor: svn log -r1?
20100323 09:23:45< Rhonda> :P
20100323 09:24:50< ilor> :P that gives me september 2003 for svn, but what about cvs?
20100323 09:25:22< ilor> anyway, need to run, would appreciate an answer in the logs ;)
20100323 09:31:39-!- Darkas [~quassel@ppp-93-104-139-121.dynamic.mnet-online.de] has joined #wesnoth-dev
20100323 09:34:24-!- Crab__ [~Crab@195.238.92.115] has joined #wesnoth-dev
20100323 09:34:27-!- stikonas [~and@2002:4f84:b38b:b:226:9eff:fe0a:3e55] has joined #wesnoth-dev
20100323 09:34:27-!- stikonas [~and@2002:4f84:b38b:b:226:9eff:fe0a:3e55] has quit [Changing host]
20100323 09:34:27-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100323 09:40:20-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20100323 09:40:31-!- stikonas [~and@2002:4f84:b38b:b:226:9eff:fe0a:3e55] has joined #wesnoth-dev
20100323 09:40:31-!- stikonas [~and@2002:4f84:b38b:b:226:9eff:fe0a:3e55] has quit [Changing host]
20100323 09:40:31-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100323 09:44:10-!- mpavel [~Pavel@cpc1-dund3-0-0-cust363.sgyl.cable.virginmedia.com] has joined #wesnoth-dev
20100323 09:44:42-!- acieroid [~acieroid@ks23738.kimsufi.com] has joined #wesnoth-dev
20100323 09:44:58-!- mpavel [~Pavel@cpc1-dund3-0-0-cust363.sgyl.cable.virginmedia.com] has left #wesnoth-dev []
20100323 09:47:38-!- Blueblaze [~nick@adsl-76-202-22-67.dsl.hstntx.sbcglobal.net] has quit [Read error: Connection reset by peer]
20100323 10:02:15-!- Zarel [~Serei@c-75-72-160-179.hsd1.mn.comcast.net] has quit [Ping timeout: 260 seconds]
20100323 10:10:21-!- Ivanovic [~ivanovic@dtmd-4db2f9f7.pool.mediaWays.net] has quit [Changing host]
20100323 10:10:21-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20100323 10:15:01< Ivanovic> moin
20100323 10:17:08< CIA-80> ivanovic * r41701 /website/start/1.8/ (index.ja.html po/ja.po): updated Japanese announcement translation
20100323 10:17:45< fendrin> Crab__: Any idea about http://www.wesnoth.org/forum/viewtopic.php?f=4&t=29265 ?
20100323 10:18:09< CIA-80> ivanovic * r41702 /trunk/po/wesnoth-units/lv.po: updated Latvian translation
20100323 10:19:49-!- loonybot [~loonybot@ppp79-139-136-187.pppoe.spdop.ru] has joined #wesnoth-dev
20100323 10:19:49-!- loonybot [~loonybot@ppp79-139-136-187.pppoe.spdop.ru] has quit [Changing host]
20100323 10:19:49-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20100323 10:20:38-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100323 10:23:20-!- acieroid [~acieroid@ks23738.kimsufi.com] has left #wesnoth-dev []
20100323 10:27:03-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 276 seconds]
20100323 10:27:39< Crab__> fendrin: will take a look
20100323 10:28:01-!- Mythological [Mythologic@77.28.113.253] has quit []
20100323 10:30:03< fendrin> Crab__: The wml does store and kill the units of the bandits, and creates new ones without the ai_special=guardian attribute. 
20100323 10:30:36< fendrin> Crab__: That is horrible wml coding but I can't see why it does slow down the ai that much.
20100323 10:36:12< Crab__> fendrin: actually it should be easy to fix, but atm I'm at work so it has to wait till evening
20100323 10:36:25< Crab__> fendrin: can you reproduce the issue ?
20100323 10:36:54< Darkas> Crab_: hi
20100323 10:37:03< Crab__> hi, Darkas
20100323 10:37:35< Darkas> Crab_: I currently can't find the functions I have to implement for evaluate and execute
20100323 10:38:57< Crab__> Darkas: see src/ai/composite/engine_lua.cpp lua_candidate_action_wrapper::evaluate and lua_candidate_action_wrapper::execute
20100323 10:39:27< Crab__> they call evaluation_action_handler_->handle and execution_action_handler_->handle
20100323 10:39:49< Crab__> and that's in src/scripting/lua.cpp
20100323 10:39:55< Darkas> Crab_: ok, thanks
20100323 10:40:07< Crab__> lua_ai_action_handler::handle(config &/*cfg*/)
20100323 10:40:19< Crab__> as you see, the config argument is currently not used
20100323 10:41:05< Darkas> Crab_: so it's just one handle function that has to be called with different configs?
20100323 10:41:21< Crab__> in stage, config is not used at all
20100323 10:41:58< Crab__> in candidate action evaluation, config is 'out parameter' - we must set it by getting a config from lua code we execute
20100323 10:42:19-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100323 10:42:40< Crab__> in candidate action execution, config is 'in parameter' - we must pass it to lua code
20100323 10:43:21< Crab__> Darkas: you can add a parameter to handle() to allow the callers (in engine_lua.cpp) to specify the wanted behavior, or you can make three different handle() functions
20100323 10:43:37< Darkas> Crab_: ok, got it
20100323 10:43:45< Darkas> and that's it?
20100323 10:44:03< Darkas> so just returning or pushing the config?
20100323 10:44:27< Crab__> yes. don't forget that you wanted candidate action evaluation to return 1 or 2 results, e.g. "42, {optional config} " or just "10.5"
20100323 10:44:46< Crab__> so, it's most likely about LUA_MULTRET
20100323 10:44:51< Darkas> Crab_: of course
20100323 10:45:42< Crab__> and note that candidate action evaluation in engine_lua.cpp expects the score to be inside the config ( serialized_evaluation_state_["score"] ) - you should make sure that it finds it there, or change the expectation.
20100323 10:45:43< Darkas> Crab_: is this handle function already used somewhere?
20100323 10:46:02< Crab__> it is expected that engine_lua.cpp is (and will be) the only user
20100323 10:46:30< Crab__> and it is used three times - in stage execution, in ca evaluation, in ca execution
20100323 10:46:52< Darkas> ok, so a boolean parameter won't work
20100323 10:48:17< Crab__> Darkas: well, you'd need at least two boolean parameters to distinquish between three options. but, you can note that there's no point in optimizing the initialization of execution of a stage, because stages are called only a few times per turn.
20100323 10:48:40< Crab__> so, you can pass a empty config to the lua code which is called from do_play_stage
20100323 10:49:13< Crab__> and have only two options 1) candidate action evaluation - return config+score from lua  2) execution (both stage and ca) - pass config to lua
20100323 10:49:45< Darkas> ok, so I'll do it with a boolean
20100323 10:49:50< Crab__> yes, it'll work.
20100323 10:50:07< Darkas> should I do that by #define ?
20100323 10:50:28< Darkas> something like #define EVALUATION true
20100323 10:50:36< fendrin> Crab__: Yes, I can reproduce it. But my version is not uptodate.
20100323 10:51:02< Crab__> Darkas: no, no need to. just use a optional boolean
20100323 10:51:08< stikonas> esr: hi, if you are still thinking about version control systems, you may want to look at how Xonotic handles large amounts of data, they've just imported Nexuiz into git, and they also have lots of binary data
20100323 10:51:11< Darkas> Crab_: ok
20100323 10:51:42< Crab__> Darkas: in fact, you can use an enum with three values (STAGE_EXECUTION, CA_EVALUATION, CA_EXECUTION ), as well , instead of boolean.
20100323 10:52:04< Crab__> Darkas: but it's better to keep the number of #define directives low.
20100323 10:52:09< stikonas> esr: http://git.xonotic.org/
20100323 10:52:15< esr> stikonas: Noted.  Our issue with moving to git is really that moving off gna would br a huge pain, and gna doesn't support git.
20100323 10:52:43< Darkas> Crab_: ok, but I think I rather use the boolean
20100323 10:52:43< Crab__> fendrin: reproduce from a fresh start or from a save ?
20100323 10:52:53< Crab__> Darkas: ok, that's good.
20100323 10:52:55< fendrin> Crab__: From the save.
20100323 10:53:03< Crab__> fendrin: ok. I'll check it out today
20100323 10:53:19< fendrin> Crab__: Do you want me to reproduce it from a fresh start?
20100323 10:53:49< Crab__> fendrin: well, no, I don't think that there's a need to do a 'remote debugging session' :)
20100323 10:54:18< Crab__> fendrin: 'avoid' aspect, and 'caution>0', are a usual source of long turn time
20100323 10:55:14< fendrin> Crab__: Not that it takes over a minute. But it doesn't eat 100% cpu during the time it takes. What means that there is a io problem?
20100323 10:55:54< Darkas> Crab_: what is that config object?
20100323 10:55:57< Crab__> fendrin: well, there's always a possibility of 'memory io'
20100323 10:56:27< Crab__> Darkas: just a  config object. lua code can store anything in there in evaluate(), and then read it back in execute()
20100323 10:56:43< Darkas> Crab_: where is the definition of it?
20100323 10:56:49< Crab__> src/config.hpp
20100323 10:57:14< Darkas> thanks
20100323 10:57:19< Crab__> it's an object which is used very often in wesnoth code. it represents a WML snippet
20100323 10:59:07< Darkas> Crab_: is there any good example of a creation of a config object?
20100323 10:59:53< fendrin> Crab__: It's not reproducible from debug mode that easy. 
20100323 11:00:22< Crab__> fendrin: ok.
20100323 11:00:36< Crab__> Darkas: it has a no-arg constructor :)
20100323 11:01:16< Crab__> Darkas: and attributes (strings) can be set easily: " config cfg;   cfg["score"] = str_cast(score);"
20100323 11:01:18< Darkas> Crab_: I've seen that, but I have to get the values from the returned lua table in it, right?
20100323 11:01:37< Crab__> Darkas: there's already code for the conversion, thanks to silene
20100323 11:01:57< Darkas> Crab_: ok, where do I find that?
20100323 11:02:45< Crab__> see luaW_pushvconfig and luaW_toconfig
20100323 11:03:07< Darkas> Crab_: ok, thanks
20100323 11:03:15< Crab__> (vconfig is variable-expanding config, we don't need those in this place)
20100323 11:04:23< Crab__> also see table_of_wml_config
20100323 11:19:50< Darkas> Crab_: how should I return the score? as a part of the config?
20100323 11:23:45-!- Mythological [Mythologic@77.28.123.167] has joined #wesnoth-dev
20100323 11:29:54< Crab__> Darkas: you said yesterday that you preferred a separate return, e.g. 42, {config=optional}
20100323 11:30:12< Crab__> you can then put the str_cast(score) in the config in the c++ code, nothing wrong with that
20100323 11:30:14< Darkas> Crab_: yeah, that's right, but how does handle return it
20100323 11:30:23< Darkas> ok
20100323 11:30:24< Crab__> either ^ or an out parameters
20100323 11:38:34-!- Crab__ [~Crab@195.238.92.115] has quit [Read error: Connection reset by peer]
20100323 11:43:10-!- Crab__ [~Crab@195.238.92.115] has joined #wesnoth-dev
20100323 11:43:57-!- ConfusedFly [~user@2001:da8:203:3:21c:26ff:fe64:26c1] has joined #wesnoth-dev
20100323 11:44:01< Darkas> Crab_: should I now use luaW_pushvconfig?
20100323 11:45:36< Crab__> Darkas: yes, I think you can. then the lua code would be able to get access to data in this config, if it'll need it. or you can change the config to a lua table.
20100323 11:46:55< Darkas> Crab_: what is the way to change it to a table?
20100323 11:49:23< Darkas> I don't really see what luaW_pushvconfig does, if not just pushing a table that contains all the data in the config
20100323 11:51:12< Darkas> Crab__: ah, I see, you're now with 2 underscores :)
20100323 11:52:20< Crab__> there's table_of_wml_config
20100323 11:52:40< Crab__> luaW_pushvconfig pushes config as light userdata, not as a table
20100323 11:52:48< Darkas> Crab__: ok, understood
20100323 11:53:00< Crab__> :)
20100323 11:58:10< Darkas> Crab__: when I call table_of_wml_config(L, cfg); at the beginning, I get "PANIC: unprotected error in call to Lua API (attempt to index a nil value)"
20100323 11:58:43< Crab__> have you created a table before that ?
20100323 11:58:56< Crab__> see how  table_of_wml_config is used
20100323 11:59:02< Darkas> ok
20100323 11:59:25< Darkas> Crab__: ok, I see
20100323 11:59:39< Darkas> I thought it would create the table
20100323 12:00:55< Crab__> 'without creating tables' allows to append several configs into a single table
20100323 12:01:38< Darkas> Crab__: yeah, that's nice
20100323 12:01:45< Darkas> I seem to be finished
20100323 12:02:10< Darkas> I get lots of ingame messages and terminal messages now :)
20100323 12:02:35< Ivanovic> ilor: lovely to be able to have music in the lobby turned on again
20100323 12:03:48< Crab__> Darkas: that's good. can you pastebin all the terminal messages for me ?
20100323 12:03:57< Darkas> Crab__: of course
20100323 12:05:04-!- Tesafilmchen [~quassel@p5B27489C.dip.t-dialin.net] has joined #wesnoth-dev
20100323 12:05:45< Darkas> Crab__: http://pastebin.ca/1850170
20100323 12:07:14< Crab__> yes, looks good. the next step will be to try to pass information from evaluate to execute
20100323 12:07:46< Crab__> e.g., pass "222222" from evaluate to execute in the first ca, and "33333" from evaluate to execute in the second ca.
20100323 12:08:09< Darkas> Crab__: ok
20100323 12:08:10< Crab__> just return a config representation from the evaluation, and read it in the execution and say hello containing this data
20100323 12:09:13< Crab__> after it's ready and works as expected, it's time to submit the patch. You can read the http://www.wesnoth.org/forum/viewtopic.php?p=136866#p136866 and post a patch at patches.wesnoth.org
20100323 12:10:36< Darkas> Crab__: what name does the table has?
20100323 12:11:34< Darkas> Crab__: ah, sorry, I just add a parameter and this parameter is the table then, right?
20100323 12:13:36-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has joined #wesnoth-dev
20100323 12:15:55< Crab__> Darkas: see LuaWML  wiki page
20100323 12:16:19< Crab__> http://wiki.wesnoth.org/LuaWML#Encoding_WML_objects_into_Lua_tables
20100323 12:16:31< Crab__> the config itself is unnamed
20100323 12:16:43< Crab__> it's contents (tags/attributes) have names
20100323 12:18:22< Darkas> but how do I access those values then? if my config table has for example a value test?
20100323 12:19:57-!- ConfusedFly [~user@2001:da8:203:3:21c:26ff:fe64:26c1] has quit [Remote host closed the connection]
20100323 12:22:23< Crab__> Darkas: see the same wiki page
20100323 12:22:42< Crab__> there are examples there
20100323 12:23:17< Darkas> Crab__: ah, so it's ...?
20100323 12:23:33< Crab__> ... allows you to reference all the arguments
20100323 12:23:45< Crab__> if you have two arguments, then you can get both from ...
20100323 12:24:27< Crab__> something like " local ai, cfg = ... "
20100323 12:25:24< Darkas> Crab__: ok
20100323 12:25:33< Crab__> or you can have your c++ code add that cfg to the ai table, under a predefined name
20100323 12:25:50< Crab__> then you'll be able to access that cfg from my_ai functions
20100323 12:25:50< Ivanovic> Crab__, fendrin: https://gna.org/bugs/index.php?15680
20100323 12:26:12< Ivanovic> Crab__, fendrin: https://gna.org/bugs/index.php?15679
20100323 12:26:15 * Crab__ adds a note to self 'before 1.8, playtest entire low'
20100323 12:26:45< Ivanovic> especially the 2nd sounds *really* problematic and should probably be fixed
20100323 12:26:53< shadowmaster> Crab__: good luck with that ;)
20100323 12:27:09< Crab__> heh
20100323 12:27:41< Darkas> Crab__: I'm also currently having another problem
20100323 12:27:44< fendrin> shadowmaster: :-P
20100323 12:28:33< Darkas> if I change return 42 to return 42, {test=123}, then luaL_checkint(L, 0); doesn't work anymore
20100323 12:29:46< Crab__> have you changed your call to expect two return values or multiple return values ?
20100323 12:30:06< Crab__> if 'multiple', have you used a function to get the actual number of returned values ?
20100323 12:30:20< Darkas> Crab__: it's mutiple
20100323 12:30:43< Darkas> and I used gettop to check how many elements are there on the stack
20100323 12:31:40< Crab__> and how many you have ?
20100323 12:33:02< Darkas> 2, as expected
20100323 12:33:13< Crab__> what are their types ?
20100323 12:33:32< Crab__> (you can add debug code to print it out, e.g. something like http://lua-users.org/wiki/CppLuaDataPassing )
20100323 12:35:05< CIA-80> fendrin * r41703 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/19_Costly_Revenge.cfg: LoW 19: Made druid and sorceress no longer recruitable.
20100323 12:35:06< fendrin> Ivanovic: Thank you :-)
20100323 12:36:17< Crab__> Darkas: have you tried using lua_tonumber instead ?
20100323 12:36:26< Darkas> Crab__: not yet
20100323 12:36:37< Crab__> as 'what you have' is not a function argument, but a value returned by function
20100323 12:37:00-!- mordante [~mordante@87.215.201.26] has joined #wesnoth-dev
20100323 12:37:07< mordante> servus
20100323 12:37:16< Crab__> Darkas: : and  luaL_check* are for checking function arguments
20100323 12:37:34< Ivanovic> hi mordante
20100323 12:37:51< Crab__> hi, mordante
20100323 12:37:52< Darkas> Crab__: so I should expect that they write correct code?
20100323 12:37:53< mordante> Ivanovic, I just read the announcement and we break MP compatibility before 1.7.3, didn't that change to 1.7.13 instead
20100323 12:37:57< mordante> hi Crab__ 
20100323 12:38:09< Crab__> Darkas: well, you can check the number of returned values
20100323 12:38:20< Ivanovic> mordante: yeah, it should
20100323 12:38:22< Ivanovic> classical typo
20100323 12:38:37< Darkas> Crab__: yeah, that's what I'm already doing :)
20100323 12:38:52< Ivanovic> mordante: got a link?
20100323 12:38:55< Darkas> ok, it says that it returned a table and a unknown value
20100323 12:39:08< Crab__> Darkas: and you can check the result for nil, too
20100323 12:39:33< mordante> Ivanovic, what kind of link?
20100323 12:39:51< Ivanovic> ah, forget the aprt about "link"
20100323 12:40:59< CIA-80> ivanovic * r41704 /trunk/utils/pofix.py: another fix for the announcement...
20100323 12:41:43< Crab__> Darkas:you can also try some different 'stack printer', such as http://lua-users.org/lists/lua-l/2006-03/msg00335.html , to find out what appears on a stack
20100323 12:42:44< CIA-80> ivanovic * r41705 /website/start/1.8/ (61 files in 2 dirs): compatibility is only available down to 1.7.13, not 1.7.3...
20100323 12:42:55< Crab__> Darkas: you can also pastebin your code, if you want me to take a look at it
20100323 12:43:03< mordante> off again
20100323 12:43:04< Darkas> Crab__: ok
20100323 12:43:05< Crab__> Darkas: or you can try to debug this yourself
20100323 12:43:55-!- mordante [~mordante@87.215.201.26] has quit [Quit: Leaving]
20100323 12:45:23< Darkas> Crab__: yeah, I'll try a bit more
20100323 12:49:49< CIA-80> caslav_ilic * r41706 /trunk/doc/manual/images/ (14 files in 2 dirs): Updated localized images for Serbian.
20100323 12:50:05-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100323 12:50:59-!- stikonas [~and@ctv-79-132-179-139.vinita.lt] has joined #wesnoth-dev
20100323 12:50:59-!- stikonas [~and@ctv-79-132-179-139.vinita.lt] has quit [Changing host]
20100323 12:50:59-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100323 12:51:00< CIA-80> fendrin * r41707 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/17_Breaking_the_siege.cfg: 
20100323 12:51:00< CIA-80> Low 17: Fixed bug #15679
20100323 12:51:00< CIA-80> Uredradia was missing at scenario start.
20100323 12:55:03< CIA-80> caslav_ilic * r41708 /trunk/po/wesnoth-manual/ (sr@ijekavian.po sr@ijekavianlatin.po): Updated translations for Serbian.
20100323 12:56:26< CIA-80> ivanovic * r41709 /website/start/1.8/index.en.html: make sure that the english version of the announcement (that is shown) is up to date, too...
20100323 12:57:44< Darkas> Crab__: it seems working now
20100323 12:57:54< CIA-80> caslav_ilic * r41710 /branches/resources/image-localization/ (4 files in 2 dirs): Updated source localized images for Serbian.
20100323 12:58:09< Crab__> Darkas: good. what was the error ?
20100323 12:58:42< Darkas> Crab__: wrong stack indexes
20100323 12:59:17< Darkas> Crab__: but now, I'm still confused in getting the values from the config table back
20100323 12:59:39< Ivanovic> fendrin: could you please add those two LoW fixes at least to the normal changelog?
20100323 12:59:56< fendrin> Ivanovic: yes
20100323 13:00:05< Ivanovic> thanks
20100323 13:00:23< fendrin> Ivanovic: Is "Fixed bug #xyz" enough?
20100323 13:00:25< Ivanovic> they might even make sense in the players_changelog (since this stuff is clearly user visible) ;)
20100323 13:00:47< Ivanovic> use the format as for the other fixed bugs
20100323 13:01:19< Ivanovic> example: Fixed bug #13882: Map which is invalid aborts map selection.
20100323 13:01:59< Darkas> Crab__: I can use ... only outside the functions, right?
20100323 13:03:27< Crab__> Darkas: only inside functions, which deal with variable number of arguments
20100323 13:03:55< CIA-80> fendrin * r41711 /trunk/ (changelog players_changelog): Added LoW bugfixed to the changelogs.
20100323 13:06:52< Darkas> Crab__: so how would I use it in my_ai:candidate_action_execution_hello() for example?
20100323 13:07:52< Crab__> either inject that cfg into my_ai, then use self. to refer to it
20100323 13:07:56< Crab__> or use my_ai:candidate_action_execution_hello(cfg)
20100323 13:09:25< CIA-80> ivanovic * r41712 /trunk/ (changelog players_changelog): move LoW entries into a "Campaigns" section
20100323 13:11:30< Darkas> Crab__: if I only add a cfg parameter, I get '[string "..."]:20: attempt to index local 'cfg' (a nil value)' as error message
20100323 13:12:47< Crab__> show me your scenario config file, please
20100323 13:14:24< Darkas> Crab__: http://pastebin.ca/1850196
20100323 13:15:45-!- shadowm_laptop [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100323 13:16:23< Crab__> Darkas: well, your function expects a parameter, but you use      execution="(...):candidate_action_execution_hello()"
20100323 13:16:46< Crab__> maybe something like "local ai,cfg = ...        ai:candidate_action_execution_hello(cfg)"   ?
20100323 13:17:17< Crab__> Darkas: so, now you're basically passing a nil parameter to candidate_action_execution_hello, exactly as it says
20100323 13:19:25< Darkas> Crab__: so I have to change execution="(...):candidate_action_execution_hello()" to execution="(...):candidate_action_execution_hello(...)"?
20100323 13:21:05< Darkas> no, that doesn't work too, but this time test is unknown
20100323 13:21:55< Crab__> Darkas: so, ... is a pair of arguments
20100323 13:22:09< Darkas> Crab__: ok
20100323 13:22:18< Darkas> so I have to add ai and cfg as parameters?
20100323 13:22:18< Crab__> Darkas: extract ... to a pair of local variables, then do ai:candidate_action_execution_hello(cfg)
20100323 13:22:38< Darkas> can I do that in execution?
20100323 13:23:40-!- Gambit [~Gambit@pa-67-234-116-244.dhcp.embarqhsd.net] has joined #wesnoth-dev
20100323 13:24:35< Crab__> well, you handle() function has to pass two parameters to lua code
20100323 13:24:56< Crab__> one of them is the ai context, the other is the config representation
20100323 13:25:03< Crab__> is this correct ?
20100323 13:25:49< Crab__> Darkas: ^
20100323 13:26:12< Darkas> Crab__: yes, I think so
20100323 13:26:51< Crab__> then, if from within your function, you want to access those arguments, you can use ... , right ?
20100323 13:26:57< Crab__> "local arg1, arg2 = ..."
20100323 13:28:13< Crab__> if the first of those arguments is the ai context 'ai', and the second is config 'cfg', then the method call is  :   ai:candidate_action_execution_hello(cfg)
20100323 13:28:24< Crab__> before, it was (...):candidate_action_execution_hello()
20100323 13:28:37< Crab__> that was the same as   local ai = ...         ai:candidate_action_execution_hello()
20100323 13:28:48< Crab__> now you have one additional parameter
20100323 13:29:48< Gambit> Hypotheticly; is this possible? http://www.wesnoth.org/forum/viewtopic.php?p=417111#p417111
20100323 13:29:57< Gambit> I'm 99% sure the user is joking.
20100323 13:30:42< Darkas> Crab__: ok, I think I understand
20100323 13:31:19< Crab__> Gambit: well, if we add a 'add LuaAI functions to influence the RNG' part to LuaAI gsoc project, mess up with the server code a little, then it should be possible :)
20100323 13:33:47< Gambit> Crab__: A user suspects that the single player campaign AI cheats by messing with RNG.
20100323 13:34:04-!- Ken_Oh [~briang@static-71-178-174-219.washdc.fios.verizon.net] has joined #wesnoth-dev
20100323 13:35:04< fendrin> Gambit: Even if the ai would cheat the player would see the new percentage.
20100323 13:35:15< Crab__> fendrin: wrong :)
20100323 13:35:31< Gambit> fendrin: Oh no. That AI is a sneaky one ;)
20100323 13:35:48< Crab__> fendrin: if we're talking about SP, it's pretty easy to hijack the rng to get more/less favourable outcomes.
20100323 13:36:12< fendrin> Crab__: How?
20100323 13:36:25< Crab__> fendrin: but, actually, we need to do it only if the ai's attack would lead to a critical unit dying and scenario lost for player :)
20100323 13:39:42< Crab__> fendrin: for example, "apply a modification to our weapon/unit or to defender before the attack,  remove it just after the attack"
20100323 13:40:37< fendrin> hmmmm
20100323 13:41:03< fendrin> Gambit: Honestly, there is no such cheating in mainline campaigns.
20100323 13:41:22< Gambit> I know that, and you know that.
20100323 13:42:36< shadowmaster> more and more threads of the same
20100323 13:42:40< Crab__> fendrin: another way is to store a rng seed, simulate a attack, if the result isn't good, repeat with another rng number, then, store 'best' rng seed found, then do the actual attack :)
20100323 13:42:45< shadowmaster> newbies complaining about the RNG or whatever
20100323 13:43:51< fendrin> It's basically that commercial games are designed to be played without reading a handbook or learning anything new since they are only clones of something the player already able to manage in sleep
20100323 13:43:57-!- FAAB [~huajie@219.142.243.194] has joined #wesnoth-dev
20100323 13:44:46< FAAB> I read in some code that "fai engine is temporary disabled", anyone who know how to enable it back locally ?
20100323 13:44:58< Crab__> fendrin: should be reenabled atm
20100323 13:45:04< Crab__> FAAB: ^
20100323 13:45:08< Crab__> fendrin: sorry
20100323 13:45:14< fendrin> No problem.
20100323 13:45:56< FAAB> Crab__: does it mean in trunk ? I am running  v1.7.14-1.8beta7
20100323 13:46:08< Ivanovic> ahh, beta7 is soooo outdated!
20100323 13:46:15< fendrin> The commercial games player is used to win at normal or hard even on the first try at least for the half of a campaign.
20100323 13:46:41< shadowmaster> fendrin: what games are you talking about?
20100323 13:46:49< Crab__> FAAB: afair, fai engine was disabled for a short time somewhen in autumn 2009
20100323 13:47:06< Crab__> FAAB: is there a particular thing that doesn't work for you ?
20100323 13:47:09< fendrin> shadowmaster: Warcraft3
20100323 13:47:12< fendrin> for example
20100323 13:47:27< shadowmaster> ahh, well, never played those
20100323 13:47:44< FAAB> Crab__: I am trying the poisonner test case in the AI Arena
20100323 13:48:05< FAAB> I will upgrade to 1.7.15 then
20100323 13:48:13< Crab__> FAAB: it's another matter there
20100323 13:48:18< Crab__> FAAB: no need to upgrade
20100323 13:49:07< FAAB> Crab__: ok, should I do anything to make it work ?
20100323 13:51:17< Crab__> FAAB: how exactly it doesn't work ?
20100323 13:51:54< Crab__> e.g. what were you doing, what was the expected behavior, and what actually happened ?
20100323 13:52:22< FAAB> Crab: well I studied the code before running it and I saw taht comment in it so I came for explanation.
20100323 13:52:31< FAAB> I haven't run it yet
20100323 13:52:53< Ivanovic> FAAB: it might be a stray comment
20100323 13:52:59< Ivanovic> please point to the source file and line
20100323 13:53:17< FAAB> I assume that the expected behavior is that the assassin doesn't attack the skeleton on village with poison :/
20100323 13:54:02< Crab__> if he's got another, more interesting targets, then yes
20100323 13:55:09< Crab__> FAAB: I cannot test atm, but it should be pretty easy for me to take a look at it in +6 hours
20100323 13:55:13< FAAB> and what about both assassins hitting the same slasher ? ;)
20100323 13:55:45< Crab__> FAAB: if the second assasin has got a unpoisoned target available, he should had went for it.
20100323 13:56:03< FAAB> Crab__: thanks a lot but I can have a try on my own for now, I just need to be sure the code is actually called
20100323 13:57:31< FAAB> Crab__: yeah, they are surrounded by 6 slashers but still he went for the same (hence already poisonned) one :(
20100323 13:57:53< Crab__> FAAB: the first thing is to verify that the correct ai got loaded (via :inspect)
20100323 13:58:10< Darkas> Crab__: I'm having a very strange problem here
20100323 13:58:11< Crab__> FAAB: then, you can add debug_print statements to formula ai code and you can use --log-debug=ai/stage/rca to see what's going on there.
20100323 13:58:14< Crab__> Darkas: which one ?
20100323 13:58:35< Ivanovic> Crab__: IIRC you only activated RCA with rc1 as default
20100323 13:58:37< Darkas> when I run the first action, it should return 42 and {test=123}, so a number and a table
20100323 13:58:46< Darkas> but it returns a number and a function
20100323 13:59:32< Darkas> (the number is 42)
20100323 13:59:42< Ivanovic> afk for a while
20100323 14:00:09-!- shadowm_laptop [~ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 240 seconds]
20100323 14:00:09< FAAB> Ivanovic: the file that says the formula AI engine is disabled is data/ai/dev/formula_ai_poisoning.cfg
20100323 14:00:09< Crab__> Ivanovic: ai arena is a special case
20100323 14:00:45< FAAB> Crab__: ok, I will try that
20100323 14:00:53< FAAB> thanks a bunch !
20100323 14:02:24< Darkas> Crab__: I think I know what could be the problem, but I got no idea why
20100323 14:02:48< Darkas> I think the top element on the stack is missing, because the second action returns only the function
20100323 14:04:06< Crab__> FAAB: i can help more in the evening
20100323 14:07:04-!- Blarumyrran [~Blarumyrr@81-20-159-197.levira.ee] has joined #wesnoth-dev
20100323 14:07:55-!- shadowm_laptop [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100323 14:08:16< Darkas> Crab__: it really seems like that, because if I add more values to return, I do now get a table and the 99
20100323 14:12:39-!- shadowm_laptop [~ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 260 seconds]
20100323 14:13:54< Crab__> Darkas: that's good that you found out the problem
20100323 14:14:48< Crab__> (Darkas: without code/compiler at hand, it's hard to help you atm)
20100323 14:15:03< Darkas> Crab__: I think I'm getting closer :)
20100323 14:15:22< Crab__> Darkas: I can suggest dumping the stack just before the call, and just after the call
20100323 14:15:34< Darkas> Crab__: I'm doing
20100323 14:15:48< Darkas> and after the call, the topmost value is missing
20100323 14:15:59< Darkas> before the call everything looks alright
20100323 14:18:04< Crab__> can you show me the logs about stack ?
20100323 14:19:00< Darkas> Crab__: http://pastebin.ca/1850235
20100323 14:19:04< Darkas> this is the whole output
20100323 14:19:38< Darkas> this one looks good now, because I've added return values
20100323 14:20:00< Darkas> so, for example the second is actually: return 99, 3
20100323 14:21:12< Crab__> ah, yes :)
20100323 14:21:22< Crab__> it should be 'return', of course
20100323 14:23:44< Darkas> Crab__: where?
20100323 14:24:43< Crab__> in evaluation code, which returns a score and a config
20100323 14:25:24< Darkas> Crab__: it is return
20100323 14:25:31< Crab__> Darkas: ok, it seems that i've misunderstood
20100323 14:25:41< Darkas> ok :)
20100323 14:26:54-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has quit [Ping timeout: 265 seconds]
20100323 14:28:00< Crab__> Darkas: well, then I'll let you debug this issue further, have to go a bit afk now. I'll return in the evening, - please submit a patch even if the issue will not be fixed, (in that case, we'll try to fix it in the evening)
20100323 14:28:18< Darkas> Crab__: ok, I'll give my best
20100323 14:28:38< Crab__> thanks
20100323 14:31:18-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has joined #wesnoth-dev
20100323 14:34:29-!- mpavel [~Pavel@cpc1-dund3-0-0-cust363.sgyl.cable.virginmedia.com] has joined #wesnoth-dev
20100323 14:34:53-!- Crab__ [~Crab@195.238.92.115] has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
20100323 14:39:04-!- mpavel1 [~Pavel@cpc1-dund3-0-0-cust363.sgyl.cable.virginmedia.com] has joined #wesnoth-dev
20100323 14:42:51-!- mpavel [~Pavel@cpc1-dund3-0-0-cust363.sgyl.cable.virginmedia.com] has quit [Ping timeout: 265 seconds]
20100323 14:47:27< orfest> hi guys!
20100323 14:48:16< fendrin> hi orfest
20100323 14:59:54-!- migge [~marc@ip-95-223-247-31.unitymediagroup.de] has quit [Quit: leaving]
20100323 15:00:24-!- fkhodkov [~fedor76@ppp-188-65-8-66-bras1.istra.ru] has joined #wesnoth-dev
20100323 15:01:08< Darkas> is there anyone who has experience with luaW_pcall?
20100323 15:01:37-!- shadowm_laptop [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100323 15:01:48< Darkas> because I found a problem with it, but I'm not sure if my solution is clean
20100323 15:09:12-!- shadowm_laptop [~ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 258 seconds]
20100323 15:11:09-!- FAAB [~huajie@219.142.243.194] has left #wesnoth-dev []
20100323 15:21:27-!- Rakkubaby [~chatzilla@122.162.125.80] has joined #wesnoth-dev
20100323 15:22:54-!- fendrin [~fabi@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20100323 15:22:58-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20100323 15:23:40-!- Crab__ [~Crab@195.238.92.115] has joined #wesnoth-dev
20100323 15:24:36< Crab__> Darkas: luaW_pcall is a wesnoth-specific function, silene is familiar with it. I'd say that if it has got a problem, it's most likely in your code, and not in it... but, anything might happen :)
20100323 15:25:10< Darkas> Crab__: luaW_pcall doesn't like LUA_MULTRET ;)
20100323 15:25:29< Darkas> I've already created a fix :)
20100323 15:26:38< Crab__> that's good. but, if it involves modifying existing luaW_pcall code, you should consult with silene first.
20100323 15:26:42-!- fendrin [~fabi@88-134-187-173-dynip.superkabel.de] has joined #wesnoth-dev
20100323 15:26:42-!- fendrin [~fabi@88-134-187-173-dynip.superkabel.de] has quit [Changing host]
20100323 15:26:42-!- fendrin [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20100323 15:27:06< Crab__> (if he won't be here, it'll be better to fork luaW_pcall and use a forked, fixed, version in our code, until silene confirms that the fix is ok)
20100323 15:27:53< Crab__> the problematic line is 'lua_remove(L, -1 - nRets);' ?
20100323 15:28:03< Darkas> Crab__: yes
20100323 15:28:29< Crab__> ok. so, the fixed code fully works ?
20100323 15:29:12< Darkas> Crab__: yes, I just added a check before, if nRets is LUA_MULTRET, and if so, then I called lua_remove(L, 1);
20100323 15:32:59< Crab__> ok, so you've only modified the code path that wasn't working before and wasn't used before ?
20100323 15:33:30< Darkas> Crab__: yes
20100323 15:33:55< Crab__> that's good, then, when you're ready, submit a patch which includes all the changes, changelog entry, and data/core/about.cfg entry in misc. contributors section.
20100323 15:33:56< Darkas> or rather: I added a code path
20100323 15:34:10< Darkas> Crab__: ok
20100323 15:34:24-!- shadowm_laptop [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100323 15:34:42< Crab__> Darkas: what is important is that nothing not-broken-before will break, in any case :)
20100323 15:35:12< Crab__> Darkas: thanks again. I'll review your patch in +5 hours
20100323 15:35:16< Darkas> Crab__: if no one used luaW_pcall with LUA_MULTRET before, then nothing can be broken :)
20100323 15:35:22< Crab__> yes
20100323 15:35:27< Darkas> ok
20100323 15:36:17-!- shadowm_webchat [~929b5303@gateway/web/freenode/x-jckshrgxvoimssfo] has joined #wesnoth-dev
20100323 15:36:32-!- shadowm_webchat [~929b5303@gateway/web/freenode/x-jckshrgxvoimssfo] has quit [Changing host]
20100323 15:36:32-!- shadowm_webchat [~929b5303@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100323 15:36:59-!- Crab__ [~Crab@195.238.92.115] has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
20100323 15:38:08-!- FAAB [~huajie@219.142.243.194] has joined #wesnoth-dev
20100323 15:50:10-!- Rakkubaby [~chatzilla@122.162.125.80] has quit [Ping timeout: 248 seconds]
20100323 15:53:10< shadowmaster> great, a fan from cloudcuckooland
20100323 15:53:32< shadowmaster> either that or he's a smart troll that knows that I absolutely hate fanboyism
20100323 15:53:50< Blarumyrran> Is that a derogative word for India
20100323 15:54:01< shadowmaster> no, I'm pretty sure he isn't from Earth even
20100323 15:54:13< Gambit> So wait I could make you angry(er) by pretending that I worship the very ground you walk upon?
20100323 15:54:24< shadowmaster> http://tvtropes.org/pmwiki/pmwiki.php/Main/Cloudcuckooland
20100323 15:54:37< Blarumyrran> Oh look it's that site again
20100323 15:54:47< shadowmaster> anyway, if India was so eager for my output as he says, my inbox would have EXPLODED.
20100323 15:55:05< shadowmaster> so yeah, not eating that fish, er, tale.
20100323 15:55:29< Blarumyrran> shadowmaster, I can tell him to tell his Indian friends to post fan mail in your inbox if you really want
20100323 15:56:09< Blarumyrran> He thinks you're efficient; he'd do it
20100323 15:56:36< shadowmaster> ...
20100323 15:56:56< shadowmaster> you have a strange affinity with spam
20100323 15:57:14< Gambit> You know with enough indian fans that would almost be a DoS on shadowmasters pm inbox
20100323 15:57:23< shadowmaster> you were suggesting to subscribe someone's email address (rocket16's) to random newspapers and stuff? :)
20100323 15:57:24< Gambit> Though I doubt admins have inbox size limits.
20100323 15:57:37< Blarumyrran> shadowmaster, I didn't just suggest it
20100323 15:57:38< shadowmaster> Gambit: we have.
20100323 15:57:58< Gambit> shadowmaster: Really? :\
20100323 15:58:01< shadowmaster> Gambit: I can break it, of course. But I don't need to.
20100323 15:58:21< shadowmaster> Inbox: Folder is 12% full (235 from 2000 messages stored)
20100323 15:59:07< shadowmaster> I removed most useless messages some time ago when my quota dropped to the Registered Users limit (don't ask), but it was still well below 1000 and above 700 back then
20100323 16:00:04< Blarumyrran> Nobody mails me :( I should make a campaign that claims to be a sequel to whatever is your chronologically latest utbs sequel
20100323 16:00:34< shadowmaster> that would be AtS, which already has a sequel from Espreon even if neither of them are finished
20100323 16:00:36< Gambit> Blarumyrran: Just make anything and then let your thread die
20100323 16:00:44-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20100323 16:00:52< Gambit> You will get tons of PMs for people requesting rediculous features that you've no intention of adding.
20100323 16:01:15< shadowmaster> subscribe to the wesnoth commits ML and you'll get many email messages per day
20100323 16:01:37< shadowmaster> just make sure to keep digests turned off for better effect
20100323 16:01:56< Blarumyrran> Yeah, I had wesnoth's inbox in mind though
20100323 16:04:10< shadowmaster> Ujoin ##shadowm
20100323 16:04:14< shadowmaster> aarg h
20100323 16:05:31< shadowmaster> Gambit: so, yeah, it'd be a banfest.
20100323 16:07:45-!- wesbot changed the topic of #wesnoth-dev to: 1.8 commits allowed till March, 24th, 12:00 GMT, afterwards "testfreeze" | string/feature freeze active! | 68 bugs, 254 feature requests, 15 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100323 16:35:30-!- dtiger [~dtiger@dynamic-vpdn-93-125-116-119.telecom.by] has joined #wesnoth-dev
20100323 16:36:22-!- erl [~erl@vetinte.spelaroll.se] has joined #wesnoth-dev
20100323 16:46:19-!- Crab__ [~Crab@195.238.92.115] has joined #wesnoth-dev
20100323 16:50:23< Crab__> Darkas: thanks for the patch. a few notes: 1) you've submitted some diff for some files not one, but two times (not a big deal)
20100323 16:50:33< Crab__> 2) if(lua_gettop(L) == 1) check can be done better: don't duplicate the "score = lua_tonumber(L, 1);" line, and check that lua_gettop(L)>=2 before fetching the config, to avoid warnings if lua_gettop(L) returns 0
20100323 16:52:02< Crab__> 3) '//stack size is 3 [-1: ai_context -2: ai_action] comments are describing the supposed state of the stack, top to bottom (note the negative numbering). so, instead of "//stack size is 3 [-1: ai_context -2: ai_action -3: table]" you'll need "//stack size is 3 [ -1: table; -2: ai_context -3: ai_action]" comment
20100323 16:53:29-!- Crab__ [~Crab@195.238.92.115] has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
20100323 16:59:47< shadowmaster> (wait, rocket16 hates Microsoft and uses Comic Sans MS? Hmm)
20100323 17:00:56< Blarumyrran> In the pdf?
20100323 17:02:04< Gambit> Maybe he took a stab at Wesnoth being open source; a majority of people here would love linux
20100323 17:02:05-!- stikonas [~and@ctv-79-132-179-139.vinita.lt] has joined #wesnoth-dev
20100323 17:02:05-!- stikonas [~and@ctv-79-132-179-139.vinita.lt] has quit [Changing host]
20100323 17:02:05-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100323 17:02:11< Gambit> And he could endear himself you you by bashing MS.
20100323 17:02:16< Gambit> *to
20100323 17:02:30< loonycyborg> Comic Sans had transcended MS long time ago :P
20100323 17:03:05< Blarumyrran> I wrote an essay for a "social aspects of information technology" wholly in comic sans yesterday
20100323 17:03:25< Blarumyrran> I think it kind of fits social "sciences"
20100323 17:03:43< shadowmaster> loonycyborg: still. They distributed it as a core font of sorts first
20100323 17:06:31< loonycyborg> http://bancomicsans.com/ :P
20100323 17:06:50< Blarumyrran> Oh, I think I signed that once
20100323 17:07:06< Blarumyrran> Online petitions are still ridiculous
20100323 17:13:19< loonycyborg> That site doesn't seem to take itself seriously at all.
20100323 17:14:04< loonycyborg> And I didn't notice any petitions there either..
20100323 17:17:34< CIA-80> ivanovic * r41713 /trunk/data/core/about.cfg: updated Galician translation credit entry
20100323 17:19:23-!- shadowm_webchat [~929b5303@wesnoth/developer/shadowmaster] has quit [Quit: Page closed]
20100323 17:19:36-!- shadowm_laptop [~ignacio@wesnoth/developer/shadowmaster] has quit [Quit: Lost terminal]
20100323 17:24:16-!- bsrk1 [~dce1353c@gateway/web/freenode/x-fkdvtepvqiwqnstl] has joined #wesnoth-dev
20100323 17:25:09-!- shadowm_laptop [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100323 17:28:20-!- bsrk1 [~dce1353c@gateway/web/freenode/x-fkdvtepvqiwqnstl] has left #wesnoth-dev []
20100323 17:28:21-!- Shakey [~jbailey@12.190.80.225] has joined #wesnoth-dev
20100323 17:29:12-!- sidkots [~dce1353c@gateway/web/freenode/x-cwzndywlerttxqzx] has joined #wesnoth-dev
20100323 17:29:23-!- shadowm_laptop [~ignacio@wesnoth/developer/shadowmaster] has quit [Client Quit]
20100323 17:30:22-!- sidkots [~dce1353c@gateway/web/freenode/x-cwzndywlerttxqzx] has left #wesnoth-dev []
20100323 17:31:47-!- sidkots [~dce1353c@gateway/web/freenode/x-nuqhlodqywvfmqxz] has joined #wesnoth-dev
20100323 17:33:55-!- haoyu [~bhy@cm26.delta25.maxonline.com.sg] has quit [Ping timeout: 240 seconds]
20100323 17:34:41< Ivanovic> fendrin: https://gna.org/bugs/index.php?15681
20100323 17:36:29< Blarumyrran> loonycyborg, http://www.petitionspot.com/petitions/bancomicsans
20100323 17:44:16-!- timotei21 [~timotei21@193.34.191.4] has joined #wesnoth-dev
20100323 17:48:00< sidkots> Hello everyone. I am interested to take part in GSOC 2010.
20100323 17:50:25-!- FAAB [~huajie@219.142.243.194] has quit [Quit: Leaving.]
20100323 17:50:32< sidkots> I am thinking to work on using spritesheets(2.7), http://wiki.wesnoth.org/SummerOfCodeIdeas.
20100323 17:54:46-!- sidkots [~dce1353c@gateway/web/freenode/x-nuqhlodqywvfmqxz] has quit [Ping timeout: 252 seconds]
20100323 18:09:05-!- Tesafilmchen_ [~quassel@p5B274945.dip.t-dialin.net] has joined #wesnoth-dev
20100323 18:09:21-!- Tesafilmchen_ [~quassel@p5B274945.dip.t-dialin.net] has quit [Remote host closed the connection]
20100323 18:09:36-!- sidkots [~dce1353c@gateway/web/freenode/x-etdvcxwvtbgcvygc] has joined #wesnoth-dev
20100323 18:09:42-!- sidkots [~dce1353c@gateway/web/freenode/x-etdvcxwvtbgcvygc] has left #wesnoth-dev []
20100323 18:10:10-!- sidkots [~dce1353c@gateway/web/freenode/x-qzgyqulbsiguvkne] has joined #wesnoth-dev
20100323 18:10:14-!- sidkots [~dce1353c@gateway/web/freenode/x-qzgyqulbsiguvkne] has quit [Client Quit]
20100323 18:10:43-!- sidkots [~dce1353c@gateway/web/freenode/x-mjxlgalzyruibkmp] has joined #wesnoth-dev
20100323 18:13:06-!- Tesafilmchen [~quassel@p5B27489C.dip.t-dialin.net] has quit [Ping timeout: 265 seconds]
20100323 18:13:08-!- Zarel [~Serei@c-75-72-160-179.hsd1.mn.comcast.net] has joined #wesnoth-dev
20100323 18:27:47-!- sidkots [~dce1353c@gateway/web/freenode/x-mjxlgalzyruibkmp] has left #wesnoth-dev []
20100323 18:35:13-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: I ATE'NT DEAD]
20100323 18:35:57-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20100323 18:50:49-!- loonybot [~loonybot@ppp79-139-136-187.pppoe.spdop.ru] has joined #wesnoth-dev
20100323 18:50:50-!- loonybot [~loonybot@ppp79-139-136-187.pppoe.spdop.ru] has quit [Changing host]
20100323 18:50:50-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20100323 18:51:37-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100323 19:04:01< CIA-80> loonycyborg * r41714 /trunk/src/font.cpp: 
20100323 19:04:01< CIA-80> Disable fontconfig-specific code if CAIRO_HAS_WIN32_FONT is defined.
20100323 19:04:01< CIA-80> It doesn't seem to be really needed in this case and causes annoying
20100323 19:04:01< CIA-80> WINDOWSTEMPDIR_FONTCONFIG_CACHE directories to be created.
20100323 19:08:03-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has joined #wesnoth-dev
20100323 19:09:04-!- knee [~jakobkang@42.176.250.195.sta.estpak.ee] has joined #wesnoth-dev
20100323 19:09:06< [Relic]> Hello :)
20100323 19:11:19< timotei21> hello Relic
20100323 19:11:38< [Relic]> :)
20100323 19:12:09< knee> hi
20100323 19:12:35< knee> listen, i'm in a bit of a situation here
20100323 19:13:23-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20100323 19:13:55< knee> i need to produce a 4 scenario campaign for wesnoth in 3 hours :)
20100323 19:14:39< Blarumyrran> Or what, you'll be fired?
20100323 19:15:04< knee> no, we will probably get worse grades
20100323 19:15:18< Gambit> Dude you're taking a class that uses wensoth for grading?
20100323 19:15:46< [Relic]> get enough xp and level up, to a class 2 coder  :)
20100323 19:16:02< knee> bit of a disaster with deadlines, its been 3 night sessions of 3-4 people on WML and caffeine
20100323 19:16:26< knee> yeah, we have a class where we have to produce a wesnoth campaign 
20100323 19:16:36< Gambit> It shouldn't be that hard to churn out a little campaign like that.
20100323 19:16:41< knee> well
20100323 19:16:43< Gambit> If each person makes one scenario
20100323 19:16:49< knee> we have 2 scenarios basically done
20100323 19:17:00< Gambit> it takes 3 hours tops to make a scenario
20100323 19:17:03< Gambit> anotehr few to balance it
20100323 19:17:12< knee> one of which is a simple battle, one of which is a collect gold job
20100323 19:17:31< Gambit> Ah so it's not a standard killEmAll campaign?
20100323 19:17:53< knee> now i need to make one with 3 sides, one human, one friendly ai and one enemy ai
20100323 19:18:04< knee> no we try to mix it up :D
20100323 19:18:37< Gambit> Well we can't really help with homework. 
20100323 19:18:41< Gambit> That'd be cheating.
20100323 19:18:48< Gambit> Unless you're supposed to get a mentor.
20100323 19:18:52< knee> were not looking for code
20100323 19:19:10< knee> well, the course is open source management
20100323 19:19:21< Gambit> I want to take that class :o
20100323 19:19:27< knee> the management side is out the window
20100323 19:19:50< knee> i just need to ask questions from time to time when we get stuck coding
20100323 19:19:50< Gambit> I would wow the pants off of the teacher.
20100323 19:20:00< knee> lol :D
20100323 19:20:39< knee> anyway, is a friendly ai side even possible?
20100323 19:20:43< Gambit> Yes
20100323 19:20:51< Gambit> just make sure the team is the same value as the player
20100323 19:20:59< Crab_> Darkas: around ?
20100323 19:21:01< knee> that crashed my game
20100323 19:21:07< knee> first thing i tried
20100323 19:21:12< Darkas> Crab_: yes
20100323 19:21:13-!- General_Fou_Fou [~Sarah@cpe-76-179-189-81.maine.res.rr.com] has joined #wesnoth-dev
20100323 19:21:14< Gambit> This might be better for #wesnoth or #wesnoth-umc-dev though.
20100323 19:21:31< knee> k
20100323 19:21:44< knee> well i'll be moving to another office and meeting some other guys in a bit
20100323 19:21:57< Gambit> team_name and user_team_name should match.
20100323 19:21:59< Crab_> Darkas: in additon to things I've said about the patch earlier, there's an issue that we need to fix in your patch of luaW_pcall
20100323 19:22:10< Gambit> *not* the side number. that should be unique.
20100323 19:22:17< Darkas> Crab_: what is it?
20100323 19:22:17< knee> ok thanks a lot
20100323 19:22:23< knee> little things like this help big time
20100323 19:22:42< Gambit> knee: http://wiki.wesnoth.org/ReferenceWML
20100323 19:22:52< knee> yeah ive read all of that
20100323 19:22:52< Gambit> I hope you stick with the game ever after you pass the class with flying colors :)
20100323 19:22:58< knee> and template WML
20100323 19:22:59< Gambit> :o the whole thing
20100323 19:23:11< Crab_> Darkas: your implementation works only if the stack is 'empty except the function arguments' when we call luaW_pcall
20100323 19:23:24< knee> dude, we have been cranked up on stimulants for 3 nights doing this
20100323 19:23:58< Darkas> Crab_: what can also be on the stack?
20100323 19:24:10< zookeeper> using the latest windows trunk build, the game is unplayably laggy in IftU's "the heart", and it was the same probably a couple weeks ago already.
20100323 19:24:10< Gambit> Anyway though the column on the right of that page is the most useful thing since some swiss bloke welded a knife and a screwdriver togethor.
20100323 19:24:28< Crab_> Darkas: anything. see example - http://wesnoth.pastebin.com/tx9php5A
20100323 19:24:40< zookeeper> to be precise, there's a major delay whenever hovering over an enemy, so it's probably something to do with the movement range darkening drawing.
20100323 19:25:01< zookeeper> here's a save: http://filebin.ca/jabymy/IftU-1-The_Heart_Turn_12.gz
20100323 19:25:03< Crab_> Darkas: then, luaW_pcall inserts the error handler before arguments, giving us http://wesnoth.pastebin.com/zVAjy3WQ
20100323 19:25:51< Crab_> Darkas: then, if there are 2 returned values, we'll have  http://wesnoth.pastebin.com/GEe8CZbt after the function call
20100323 19:26:23< Darkas> Crab_: I see
20100323 19:26:25< Crab_> Darkas: note that silene's formula  'lua_remove(L, -1 - nRets);' works ok when nRets is known
20100323 19:26:47< Darkas> Crab_: yeah, that's right
20100323 19:27:08< Darkas> but I currently have no idea what we could do otherwise...
20100323 19:27:08< Crab_> Darkas: so, I think that we have to call lua_gettop() one more time before the function call to allow us to find the location of the error handler
20100323 19:27:35< Darkas> Crab_: ok, that would be a solution
20100323 19:27:52< Crab_> after all, *we* insert it, so we have to know where it is :)))
20100323 19:28:00< Darkas> yeah, right ;)
20100323 19:28:24< Crab_> can you do so now ? (in 2-3 nearest hours, also tackling the issues I wrote about a few hours earlier)
20100323 19:28:43< Darkas> Crab_: no problem
20100323 19:28:51< Crab_> that's good! thanks
20100323 19:29:17< Darkas> the issues from before are actually just because I wasn't careful enough ;)
20100323 19:29:31< Crab_> ;)
20100323 19:30:32< Darkas> Crab_: the negative stack indexes are calcualted as lua_gettop(L) - index, right?
20100323 19:30:41< Crab_> yes, -1 is the topmost value on the stack
20100323 19:30:58< Darkas> ok
20100323 19:31:02< Crab_> -2 is the 'value just below top', and so on
20100323 19:31:35< Crab_> the formula should be pretty clear from http://wesnoth.pastebin.com/zVAjy3WQ
20100323 19:32:08< Crab_> and you shouldn't call ' lua_settop(L, 0);//empty stack'
20100323 19:32:25-!- chains [~Rylar@adsl-75-37-45-181.dsl.pltn13.sbcglobal.net] has joined #wesnoth-dev
20100323 19:32:41< Darkas> Crab_: what should I do then?
20100323 19:33:41< Crab_> note the '	int res = lua_pcall(L, nArgs, nRets, -2 - nArgs);  ' line in luaW_pcall
20100323 19:34:04< chains> Everyone has heard about Nexuiz by now right?
20100323 19:34:24< loonycyborg> chains: Probably yes :P
20100323 19:34:49< Crab_> (ah, disregard the note)
20100323 19:35:38< Crab_> Darkas: I'd say  'a simpler solution is to use gettop to find out how many things are in the stack before we call luaW_pcall, and then pop the required number of arguments'
20100323 19:35:54< chains> Total insanity 
20100323 19:36:15< zookeeper> regarding the lag issue i just mentioned: several units in play had a ~PAL function on their base frames, and removing that fixed the issue. seems there's some kind of major inefficiency in that particular function or maybe all base frames with functions cause it.
20100323 19:36:45< Darkas> Crab_: ok
20100323 19:37:05< Darkas> Crab_: what did you mean with the first issue? I didn't really understand that
20100323 19:39:01< Darkas> and something else: we insert the error handler at -2 - nArgs, so why don't we remove it with that formula again?
20100323 19:39:37< Crab_> Darkas: because the args are no more present, they were 'consumed'
20100323 19:39:55< Crab_> Darkas: instead, we have return values , a number of them
20100323 19:40:58< Crab_> Darkas: about (1) , there are three issues
20100323 19:41:22< Crab_> firstly, if(lua_gettop(L) == 1) is not good. lua_gettop(L) can be, say, 13 before the function got executed, and 14 after :)
20100323 19:41:29-!- knee [~jakobkang@42.176.250.195.sta.estpak.ee] has quit [Ping timeout: 246 seconds]
20100323 19:41:51< Darkas> Crab_: yes, I've already changed that
20100323 19:41:54< Crab_> Darkas: instead, you can get the 'initial top' (you'll need it for cleanup anyway), and check for initial_top+X
20100323 19:42:15< Darkas> I just didn't get the first of the issues you were talking about
20100323 19:42:39< Crab_> you could have moved the "score = lua_tonumber(L, 1);//get score" outside the if
20100323 19:42:59< Crab_> (fixing the index, of course)
20100323 19:44:13< Darkas> Crab_: yeah, I've already done that
20100323 19:44:31< Crab_> also, if    " if(lua_gettop(L) == 1) " condition is false, it doesn't mean that you have '2 return values available' - because you can have 0 :)
20100323 19:44:35< Darkas> but you said:  1) you've submitted some diff for some files not one, but two times (not a big deal)
20100323 19:44:48< Crab_> Darkas: ah, that..
20100323 19:44:54< Crab_> Darkas: see your patch at patches.wesnoth.org
20100323 19:45:13< Crab_> https://gna.org/patch/download.php?file_id=8626
20100323 19:45:36< Crab_> and, for example, count the number of times src/scripting/lua.cpp appears in the patch
20100323 19:46:18< Darkas> Crab_: ah, I see
20100323 19:46:50< Darkas> ok, sorry for that, I updated the fix and used >> ;)
20100323 19:47:11< Crab_> as I said, "no big deal" :)
20100323 19:47:27< Darkas> Crab_; so I should make all the indexes I use negative?
20100323 19:47:47< Crab_> either this, or base them on initial_top
20100323 19:48:03< Darkas> ok
20100323 19:51:46< Darkas> Crab_: I clean the stack by with lua_settop(L, oldTop); then?
20100323 19:51:55< Darkas> or is there a better way of doing this?
20100323 19:52:40< Crab_> Darkas: yes, that's perfect. note that you should check the return value of LuaW_pcall to see if you need to do this prematurely without any further processing.
20100323 19:53:34< Crab_> please name it 'initial_top' for consistency with luaW_toconfig
20100323 19:53:43< Darkas> Crab_: ok
20100323 19:55:16< Darkas> Crab_: what does cleaning the stack has to do with the return value of luaW_pcall?
20100323 19:55:51-!- gabba [~gabba@70.35.163.201] has joined #wesnoth-dev
20100323 19:56:11< Crab_> if luaW_pcall returns false, we must assume that the call has failed, and do nothing except cleanup.
20100323 19:56:56< Darkas> ok
20100323 19:58:02< Darkas> Crab_: if the pcall fails, will the arguments be removed from the stack?
20100323 19:58:31< Crab_> that is not defined.
20100323 19:59:35< Crab_> (actually, the behavior is defined, but we shouldn't depend on it)
20100323 19:59:46< gabba> Hi
20100323 19:59:46< Crab_> (we should just cleanup and go)
20100323 19:59:50< Crab_> gabba: hi
20100323 20:00:21< Darkas> Crab_: ok
20100323 20:00:21< gabba> My two proposals are online :D
20100323 20:00:40< Crab_> gabba: yes, I've read them
20100323 20:00:59< gabba> If you or anyone else has comments, they're welcome
20100323 20:01:01< Darkas> Crab_: I've calculated the error handler index as int error_handler_index = lua_gettop(L) - 2 - nArgs;
20100323 20:01:17< Darkas> is that correct?
20100323 20:01:18< Crab_> gabba: yes, I would like to talk about  'Share Undoable Moves with Allies' thing
20100323 20:02:00< gabba> Crab_: whenever you want - maybe you wanna finish your current conversation first?
20100323 20:02:10< Darkas> Crab_: before, it's inserted at -2 -nArgs
20100323 20:02:54< Crab_> Darkas: let me see...
20100323 20:05:36< Crab_> I'd say  lua_gettop(L) - 1 - nArgs;
20100323 20:05:54< Darkas> ok, I'll try out :)
20100323 20:06:06< Crab_> note that the insert 'moves' an element from the top of the stack
20100323 20:06:15< Crab_> that's why we get -2 when inserting
20100323 20:06:40< Darkas> Crab_: ok
20100323 20:06:40< Crab_> if we have:  old;old;old;arg;arg;arg;arg;error_handler;(top)
20100323 20:07:11< Crab_> then, we need to insert error handler before top (-1) before error handler (-1) before args (-nArgs) 
20100323 20:07:16< Darkas> so then I can even remove the if, and always remove with error_handler_index
20100323 20:07:18-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20100323 20:07:28< mordante> servus
20100323 20:07:38< Crab_> Darkas: better don't do that until silene looks at it :) we want to make sure that the old path works ok.
20100323 20:07:56< Darkas> Crab_: ok
20100323 20:08:02< Crab_> and, after error handler is inserted, we get old;old;old;error_handler;arg;arg;arg;arg;(top)
20100323 20:08:19< Crab_> that's why I said about 'lua_gettop(L) - 1 - nArgs;'
20100323 20:08:44< Crab_> Darkas: but you can actually get rid of the if
20100323 20:08:52< Crab_> (in one place)
20100323 20:09:10< Crab_> (if you just calculate the actual nRets)
20100323 20:09:12< Darkas> what if if not the one I added?
20100323 20:09:19< gabba> fendrin: you use eclipse to develop wesnoth, isn't it?
20100323 20:09:40< Crab_> Darkas: ok, then I'll let you fix it and look at the fixed & working code again.
20100323 20:09:59< Darkas> Crab_: it's already done :)
20100323 20:10:09< Crab_> cool :) you're fast :)
20100323 20:10:21< Darkas> I left the if at the end there, because it's simpler calculation
20100323 20:10:45< Darkas> and if silene gives the ok, then we can just delete a few lines :)
20100323 20:10:46< gabba> fendrin: If you make an out-of-tree build, do you still manage to use the eclipse svn integration?
20100323 20:10:50< Crab_> ok, so, ping me when there'll be a new patch available
20100323 20:11:02< Crab_> Darkas: yes
20100323 20:11:06-!- knee [~jakobkang@82.131.116.202.cable.starman.ee] has joined #wesnoth-dev
20100323 20:11:10-!- mjs-de [~mjs-de@vpw.wh.Uni-Dortmund.DE] has quit [Ping timeout: 264 seconds]
20100323 20:11:29< gabba> fendrin: I mean, when you generate your eclipse project with a cmake out-of-tree build
20100323 20:12:04< gabba> Crab_: shall we talk about my proposal?
20100323 20:12:24< Crab_> gabba: I wanted to talk about a thing that is related to your proposal
20100323 20:12:42< Crab_> gabba: about delayed shroud updates
20100323 20:12:58< gabba> Crab_: yup, what about them?
20100323 20:13:43< Darkas> Crab_: the new patch is there
20100323 20:13:49< Crab_> gabba: I'm collecting bugs which are related to them :)
20100323 20:13:49< mordante> Crab_, regarding the eclipse stuff, doesn't the validation depend on the version of Wesnoth? If so trunk would be a better place
20100323 20:13:55< Crab_> wesbot: bug #13256
20100323 20:13:55< wesbot> Bug #13256  Assigned to: Yurii Chernyi <crab>  Status: Postponed  Priority: 5 - Normal
20100323 20:13:58< wesbot> Summary: Assertion failure in  src/actions.cpp, happened when moved (12,8)-> (10,10)
20100323 20:14:01< wesbot> Original submission: Launched a test scenario, moved (12,8)-> (10,10)Happene
20100323 20:14:02< Crab_> wesbot: bug 14269
20100323 20:14:04< wesbot> d once, unable to reproduce so far.==./wesnoth-debug -d -t --log-debug=ai,
20100323 20:14:07< wesbot> URL: https://gna.org/bugs/?13256
20100323 20:14:09< wesbot> Bug #14269  Assigned to: Yurii Chernyi <crab>  Status: Postponed  Priority: 5 - Normal
20100323 20:14:12< wesbot> Summary: name=sighted events not firing on enemy moves and affected by delay_shroud setting
20100323 20:14:15< Crab_> wesbot: bug 15600
20100323 20:14:15< wesbot> Bug #15600  Assigned to: Yurii Chernyi <crab>  Status: Postponed  Priority: 3 - Low
20100323 20:14:18< wesbot> Original submission: This bug seems like it might be a regression (see bug #11
20100323 20:14:21< wesbot> 286 ), and depending on current policy, it may be moot (see the most recent comm
20100323 20:14:24< Crab_> mordante: we'll move
20100323 20:14:25< wesbot> URL: https://gna.org/bugs/?14269
20100323 20:14:27< wesbot> Summary: Shroud updates which are part of apply_shroud_changes interfere with WML modifications to shroud
20100323 20:14:30< wesbot> Original submission: Sometimes it is needed to skip shroud updates which are p
20100323 20:14:33< mordante> Crab_, ok good
20100323 20:14:33< wesbot> art of undo stack unwind.Example: teleport from one 'level' of map to anothe
20100323 20:14:36< wesbot> URL: https://gna.org/bugs/?15600
20100323 20:14:52< gabba> omg, the spam
20100323 20:15:30< Crab_> gabba: yes, a small spamfest to note there's a number of issues (of varying messiness) with delayed shroud updates
20100323 20:16:00< gabba> Crab_: ah - I suppose you already know about my small patch to enable "delay shroud updates" on game start, as well
20100323 20:16:12< Crab_> gabba: yes, I've seen that patch
20100323 20:16:30< Crab_> gabba: so, in 1.9, there's a plan to rewrite the way delayed shroud updates work
20100323 20:16:56< gabba> Crab_: which could conflict/complicate life for my project
20100323 20:17:09< Crab_> gabba: both a conflict and an opportunity
20100323 20:17:11< gabba> Crab_: unless I can integrate it somehow?
20100323 20:17:46< Crab_> gabba: bug https://gna.org/bugs/index.php?13256 is the most easy way to see the breakage of current system
20100323 20:18:02 * gabba goes to look at bug
20100323 20:19:25< Crab_> gabba: the actual segfault is fixed, but the underlying problem remains
20100323 20:19:42< gabba> Crab_: yeah, I was gonna say it looks minor, superficially 
20100323 20:19:57< gabba> Crab_: so what's the underlying issue?
20100323 20:20:57< gabba> Crab_: what I gather, is that it delay shroud updates doesn't play well with the undo stack
20100323 20:20:59< Crab_> gabba: the issue is with situations where we have a bunch of undoable moves
20100323 20:21:00-!- orfest [~kvirc@43.ngu.ac-tel.ru] has quit []
20100323 20:21:40< Crab_> gabba: the issue happens both with and without delayed shroud updates, but with delayed shroud updates it happens a lot more, because more undo moves are available
20100323 20:23:04< Crab_> gabba: if the unit which is moved triggers a wml event, then the event is executed immediately
20100323 20:23:05-!- General_Fou_Fou [~Sarah@cpe-76-179-189-81.maine.res.rr.com] has quit [Quit: General_Fou_Fou]
20100323 20:23:47< Crab_> gabba: and, shroud changes and undo stack unwind happens after that event is fired
20100323 20:24:40< Crab_> gabba: so, if that wml event modified units which were mentioned in undo stack, or if that wml event modified shroud, there can be errors because the undo stack will be processed after the event has fired and changed things
20100323 20:25:20< Crab_> gabba: it all gets funnier once we notice the fact that the applying of shroud changes can trigger 'sighted' events.
20100323 20:26:03< gabba> Crab_: ok, so you could get messy things such as an event repeatedly triggered, or a unit that goes back to a place that doesn't make sense, considering the nature of the WML event
20100323 20:26:40< Crab_> gabba: yes. causality violations happen
20100323 20:27:01< gabba> Crab_: yet the universe didn't implode yet :P
20100323 20:27:17< Crab_> gabba: but it segfaults sometimes :)
20100323 20:27:59< gabba> Crab_: So, do you plan do delay WML events until moves are permanent, or to (gasp) integrate them in the undo stack?
20100323 20:28:27< mordante> ilor, in the lobby I have "games: 0 from 71" shown and I see the games as well
20100323 20:28:45< Crab_> gabba: there is a problem..
20100323 20:28:53< Crab_> gabba: imagine that we have 40 units
20100323 20:28:58< Crab_> we enable DSU
20100323 20:29:04< Crab_> and move them all
20100323 20:29:12< Crab_> we disable DSU
20100323 20:29:17< Crab_> shroud changes are applied
20100323 20:30:10< gabba> and all (potential) 40 events trigger in sequence...
20100323 20:30:16< Crab_> what if we hit a 'sighted' event which will interrupt our move ? what will happen with the rest of our moves ?
20100323 20:30:52< Crab_> another glitch: imagine that there's a task: find a treasury by moving into it.
20100323 20:30:59< Crab_> you have 1 unit, the treasury is near
20100323 20:31:04< Crab_> you enable DSU
20100323 20:31:16< Crab_> move to location A, undo, move to B, undo, ...
20100323 20:31:44< gabba> wait, "sighted" are WML events, i.e. not the same thing as "enemy sighted, press T to continue movement", right?
20100323 20:31:53< Crab_> yes
20100323 20:32:01< Crab_> little buggy wml events
20100323 20:32:14< Crab_> ...repeat until you've safely checked all the nearby locations for treasury (as as soon as you hit it, moveto event will fire which'll abort your move)
20100323 20:32:40-!- EdB [~EdB@tss37-1-89-84-18-220.dsl.club-internet.fr] has joined #wesnoth-dev
20100323 20:32:41< Crab_> the same with ambush - it's possible to find an ambusher hidden in the forest if you move-undo-move-undo-...
20100323 20:33:21< Crab_> Darkas: ok, looking at the patch
20100323 20:33:31< gabba> Crab_: honestly, the only quick-and-clean fix that comes to mind is to empty the undo stack on any WML event
20100323 20:34:12< gabba> Crab_: and I thought ambush already prevented you from undoing moves???
20100323 20:34:40< Crab_> gabba: yes. and that's another problem.
20100323 20:35:03< Crab_> ' empty the undo stack on any WML event' - what will happen if you move a unit to trigger a move event, during emptying the undo stack you'll trigger a sighted event which was meant to  teleport the unit away ?
20100323 20:35:32< gabba> Crab_: or the other, more radical thing that someone proposed on the mailing list: get rid of delay shroud updates altogether (and get lynched by an angry mob)
20100323 20:36:05< Crab_> gabba: well, it's useful :)
20100323 20:36:25< Crab_> gabba: and it's up to you to say how we can make it even more useful while avoiding most of the bugs
20100323 20:37:09< Crab_> gabba: imagine: you have 40 units, you move 30 in temporary locations, thinking "ok, i'll see how it'll be and then i'll undo and do a better thing", then, on unit 31, you hit an ambusher :)
20100323 20:38:05< gabba> Well, for this last one you deserve it, I think
20100323 20:38:19< gabba> It reproduces well the panic of an ambush
20100323 20:38:47< Crab_> gabba: why ? isn't a point of DSU to have a secure way to 'plan ahead, see how the movement of the enemy will be restricted, etc' ?
20100323 20:39:51< Crab_> gabba: so, I was trying to tell about this things to make you think about those issues which surround DSU and shroud-action sharing in more detail.
20100323 20:40:03< gabba> Yeah, but up to a certain point... imagine the other players waiting while you're trying your N^N combinations, where N is your number of units 
20100323 20:40:14< Crab_> gabba: that's what turn timer is for :)
20100323 20:40:22< gabba> true, true
20100323 20:40:25< Ivanovic> re
20100323 20:40:32< Ivanovic> any news regarding world domination?
20100323 20:40:32< Crab_> gabba: and, actually, calculating the combinations by hand will take *more* time
20100323 20:40:45< Crab_> Ivanovic: scheduled for april the first, last time I heard.
20100323 20:41:36< mordante> Ivanovic, yes failed to achieve it today, maybe tomorrow
20100323 20:42:01< Ivanovic> hmm, tomorrow will be difficult since you just have reduced time to achive it then...
20100323 20:42:16< gabba> Crab_: well, you are successfully making me think, but I can already say there's no way to reconcile, say, ambush and keeping the ability to undo
20100323 20:42:20< Ivanovic> the rest of the time you got to test if wesnoth 1.8 is really ready for tagging...
20100323 20:42:24< ilor> mordante: re the games count, should be fixable very quickly
20100323 20:42:42< gabba> Crab_: I may find ways to solve some annoying corner cases, though
20100323 20:42:59< mordante> ilor, I thought/hoped that
20100323 20:43:07< Crab_> gabba: for example, if you add a way to plan moves on a hypothetical unit map, without actually doing the moves.
20100323 20:43:34< mordante> Ivanovic, tagging 1.8 is claiming world domination ;-)
20100323 20:43:49< Ivanovic> nah, 1.8 is claiming europe
20100323 20:43:56< Crab_> gabba: e.g., then, a selected number of players would be able to see/propose/make hypothetical moves, and then, you'll be able to overlay the 'plan' and actual gamemap and do planned moves by clicking
20100323 20:44:00< Ivanovic> hmm, okay, that is probably most of the sane world these days, right?
20100323 20:44:01< Ivanovic> ^^
20100323 20:44:05-!- General_Fou_Fou [~Sarah@cpe-76-179-189-81.maine.res.rr.com] has joined #wesnoth-dev
20100323 20:44:05< Crab_> gabba: or some other funny workaround
20100323 20:44:16< Crab_> gabba: but most important is for this matter to be *settled*
20100323 20:44:34< gabba> Crab_: oh, of course if you have an abstracted map where you can input supposed enemy positions, and what you just said, yes
20100323 20:44:39< Crab_> gabba: e.g., if you say 'I've thought a lot and sighted events must die a horrible death',, that's  also a possible opinion/solution.
20100323 20:44:58< mordante> Ivanovic, that implies there are sane parts left in this world ;-)
20100323 20:45:22< Ivanovic> yes, i think there are some tiny parts eg hidden somewhere deep in germany
20100323 20:45:36< Crab_> gabba: well, nothing prevents the c++ code from making a copy of the gamemap/unit_map and doing some hypothetical moves
20100323 20:46:09< Crab_> gabba: or, you can add a 'red arrow' for each unit which is planned to move/attack, and make a mode to calculate the ZoC based on 'planned' positions
20100323 20:46:27< Crab_> gabba: it would be cool to watch
20100323 20:46:41< gabba> Crab_: but I'm kind of uneasy with this: part of a strategy game is picturing situations in your head, right?
20100323 20:46:56-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20100323 20:47:06< ilor> okay, so anyone /now/ remeber when Wesnoth started using CVS? ;)
20100323 20:47:22< gabba> Crab_: this sounds like interrupting a chess match to take the board in another room and move the pieces around to see what it would look like
20100323 20:48:00< Crab_> gabba: example arrows : http://www.vpltd.com/images/hoi2pic_02.jpg
20100323 20:48:03< gabba> Crab_: so I'm not sure that adding meta-meta systems to help you visualize what-if situations would actually improve the game
20100323 20:48:21< Ivanovic> ilor: the date of the first revision in svn it is
20100323 20:48:22< Crab_> gabba: so, you can keep the 'main map', but just add some visual candy 
20100323 20:48:42< ilor> Ivanovic: wouldn't that be the import date?
20100323 20:48:44< Ivanovic> ilor: since the svn repo was converted using cvs2svn all cvs revisions were imported into the tree
20100323 20:48:48< ilor> ahh
20100323 20:48:53< Ivanovic> no, the import date was some time later
20100323 20:48:59< Ivanovic> IIRC post 1.0, not sure though
20100323 20:50:18 * mordante hopes boucman shows up soon to test a work around for his late night crash
20100323 20:50:22< Blarumyrran> Crab_, what game is that?
20100323 20:50:53< gabba> Crab_: this said, boucman had the idea of planning moves while it's not your turn
20100323 20:50:58< Ivanovic> and the move to svn was in 2005
20100323 20:51:10< Ivanovic> have a look at the mail from isaac from july 25th 2005
20100323 20:51:11< Crab_> Blarumyrran: hearts of iron 2
20100323 20:51:11-!- elias [~elias@allegro/developer/allefant] has joined #wesnoth-dev
20100323 20:51:48< Crab_> gabba: yes, that can be integrated, too. and you can make it possible to allow your allies to 'suggest' moves for you, too
20100323 20:51:48< Ivanovic> the real conversion was at the 20th september 2005
20100323 20:51:50< gabba> Crab_: after all your idea is very similar
20100323 20:52:13< Crab_> gabba: yes. but the purpose is different :)
20100323 20:52:25< Crab_> Darkas: the patch looks good. will try it out now
20100323 20:52:34< Ivanovic> wow, already 4.5 years at gna.org...
20100323 20:52:38< gabba> Crab_: it's starting to sound like a separate GSoC project :P
20100323 20:52:41< Darkas> Crab_: ok, good
20100323 20:53:01< ilor> Ivanovic: thanks
20100323 20:53:20< Crab_> gabba: well, it's for you to formulate your proposal, I'm simply trying to suggest things that are 'near' it and can be included if you want to.
20100323 20:53:22< mordante> ilor, just curious why do you want the info?
20100323 20:53:41< gabba> Crab_: and I appreciate it
20100323 20:54:15< Crab_> ilor: also see http://www.mail-archive.com/wesnoth-cvs-commits@gna.org/mail29.html
20100323 20:54:19< ilor> mordante: university course on software development, I'll be talking about SCMs and want a concrete example of a project that went through cvs to svn and considers git ;)
20100323 20:54:47< ilor> ... and started off as .tgz passed around ;)
20100323 20:54:51< Ivanovic> ilor: for the discussion have a look at the mailinglist thread started by isaac that i pointed out
20100323 20:55:02< gabba> Crab_: what's the position of the MP devs regarding boosting undo capabilities even more?
20100323 20:55:05< ilor> Ivanovic: thanks a lot
20100323 20:55:14< Crab_> gabba: I'm not a MP dev :)
20100323 20:55:31< gabba> Crab_: you're more in AI I think, that's why I ask
20100323 20:55:51< Ivanovic> gabba: about this you should probably talk to Soliton and noy
20100323 20:55:57< Crab_> gabba: well, I want to get rid of bugs which are connected with delayed shroud updates, and I want sighted events to be actually usable
20100323 20:56:10< mordante> ilor, ah ok
20100323 20:56:10< Crab_> gabba: if changing the undo system will fix this, that's very good.
20100323 20:57:06< gabba> Crab_: I've opened a whole can of worms, I think
20100323 20:57:18< Crab_> gabba: you're near it :)
20100323 20:57:41< gabba> Crab_: I'll organize all this in my head and discuss this left and right
20100323 20:58:14< gabba> Crab_: the thing that annoys me though, is that I don't have a "simple" preliminary task anymore
20100323 20:58:28< gabba> Crab_: i.e. no related simple bug to fix
20100323 20:59:33< Crab_> gabba: well, you can make a prototype of  'send undoable moves via network to allies' as 'a simple bug to fix' ;)
20100323 21:00:17< Crab_> gabba: and let the devs play with it, to see if they like it
20100323 21:00:40< mordante> Ivanovic, you want to change the format of the changelog regarding bug fixed?
20100323 21:00:51< Ivanovic> mordante: ?
20100323 21:00:59< gabba> Crab_: ha ha, "simple", ha ha
20100323 21:01:04< Ivanovic> mordante: i want it kept the way i changed it the last days
20100323 21:01:09< gabba> Crab_:  I'm having shivers down my spine
20100323 21:01:30< mordante> Ivanovic, r41693
20100323 21:01:54< Crab_> gabba: well, it's pretty simple if you do it as a hack
20100323 21:02:10< gabba> Crab_: I'll look into it though, maybe it's less bad than it sounds... as a hack as you say
20100323 21:02:27< CIA-80> ivanovic * r41715 /trunk/ (13 files in 11 dirs): updated Finnish and Russian translation
20100323 21:02:29< Crab_> gabba: yes, as a shameless hack. note that you can involve wml events, too
20100323 21:02:45< Ivanovic> mordante: have a look at the format of the changelog like it is for the 1.7.15+svn changes as of revision 41715
20100323 21:02:58< Ivanovic> this is how i would like things to look for players_changelog and changelog
20100323 21:03:08< Ivanovic> common:
20100323 21:03:26< Ivanovic> * use past for stuff that was changed
20100323 21:03:32< Ivanovic> * up to 80 chars per line
20100323 21:03:55< Crab_> gabba: also note that you'll be able to ask the devs to patch their wesnoth clients for that particular game\set of games
20100323 21:04:06< Ivanovic> * if a reported bug was fixed, write "* Fixed bug #ABCDE: short bug description"
20100323 21:04:15< Crab_> gabba: so, try to think of a good way :)
20100323 21:04:42< Ivanovic> * if a requested feature was added: "* Added feature #ZYXWV: short feature description"
20100323 21:04:49< Ivanovic> differences:
20100323 21:04:49< gabba> Crab_: to patch their client? I'm not following you
20100323 21:05:06< mordante> yes but there are still some "foo bar (bug #xxx)" entries
20100323 21:05:18< mordante> which got me confused
20100323 21:05:19< Ivanovic> * make things look more like (short) sentences in the players_changelog (don't forget a '.' at the end)
20100323 21:05:23< Crab_> gabba: e.g., for a demo,you can make everyone who is checking your idea to play a patched version of wesnoth
20100323 21:05:32< gabba> Crab_: ah ok
20100323 21:05:34-!- UnknownDevice [~UnknownDe@188.141.93.69] has joined #wesnoth-dev
20100323 21:05:57< Ivanovic> * between versions there are two free lines in the players changelog and just one in the normal, categories are seperated by a free line in the players changelog, nothing like this in the normal
20100323 21:06:02< Crab_> gabba: you can try to play with invisibility rules: e.g. play without fog/shroud,  make all the units invisible to enemies but visible to allies, and make units visible again on end of turn
20100323 21:06:13< Ivanovic> mordante: feel free to change/fix them
20100323 21:06:23< Crab_> gabba: whatever is fastest to get the result you want
20100323 21:06:44-!- dnk-88 [~dnk-88@217.21.40.13] has joined #wesnoth-dev
20100323 21:06:45< Ivanovic> and: things should be sorted in the categories: alphabetical sorting
20100323 21:07:04-!- mpavel1 [~Pavel@cpc1-dund3-0-0-cust363.sgyl.cable.virginmedia.com] has left #wesnoth-dev []
20100323 21:07:05< mordante> I will, what about when it fixes an upstream bug (like Debian bugs) also at the front or behind between braces?
20100323 21:07:36< gabba> Crab_: thanks for the suggestions, I better get to work then ;)
20100323 21:07:43< Crab_> good luck :)
20100323 21:07:59< Ivanovic> if it is reported upstream only: Fixed Debian bug #12345:
20100323 21:08:05< gabba> Crab_: but let's just say I'm happy to have my other proposal as a backup: that one is less interesting, but is infinitely simpler
20100323 21:08:19< Crab_> gabba: I think that the hacky prototype will be a very good thing to have :)
20100323 21:08:28< mordante> and after the colon in the changelog, capitalize or not?
20100323 21:08:32< Ivanovic> if it is reported both, upstream and in our tracker: Fixed bug #ABCDE (Debian bug #12345):
20100323 21:08:44< Ivanovic> capitalize
20100323 21:09:44-!- dnk-88 [~dnk-88@217.21.40.13] has left #wesnoth-dev []
20100323 21:09:54< Crab_> gabba: I personally think that the fastest way is to patch the clients to store the 'prematurely received moves' from enemy sides until a signal 'please apply them' is received.
20100323 21:10:33-!- qemqemqem [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has joined #wesnoth-dev
20100323 21:10:50< Ivanovic> mordante: you saw my comment in the libsdl clicking bug that i made sure that the info is available upstream (in their tracker)?
20100323 21:11:41< Crab_> gabba: that way, you'll play without fog/shroud and without delayed shroud updates, but your enemies will have the impression that DSU is on for you.
20100323 21:11:59< Crab_> ..and your allies will see your moves immediately..
20100323 21:12:48< Zarel> Hey, can one of you split a thread for me?
20100323 21:12:51< gabba> Crab_: right, I suppose that currently they're only sent after you lost the undo possibility
20100323 21:13:11< Crab_> gabba: I suppose that you can simulate the undo via WML right-click menu.
20100323 21:13:21< gabba> Crab_: but you're confusing me a little bit about how you think it should look for the active player
20100323 21:13:58< Crab_> move units as usual, if need to undo, undo either via normal undo, or, if this is hard to make work, via wml right-click menu.
20100323 21:14:25< Crab_> after all, you can implement a simple undo stack in wml
20100323 21:14:48< gabba> good to know :)
20100323 21:15:02< Crab_> and the non-allied clients will need to store received moves until they get the signal that they should process it. 
20100323 21:15:20< Crab_> then, they should delete the 'undoed' moves, and just replay the rest
20100323 21:16:04< mordante> Ivanovic, jein, I just did
20100323 21:16:39< mordante> fendrin, didn't you fixed some parts of the Debian UtbS bug?
20100323 21:16:45< gabba> Crab_: Or, maybe I could just sync the moves to everybody for now, as if they were allies, and worry about hiding them from enemies later
20100323 21:17:00< gabba> Crab_: would that do as well for a prototype?
20100323 21:17:21< gabba> Crab_: the advantage is that I won't have to throw code away later
20100323 21:17:35< CIA-80> mordante * r41716 /trunk/changelog: Unify the style of the changelog.
20100323 21:17:51< Crab_> gabba: throwing code away is not a problem :) after all, you gain the ability to hack wesnoth in many unusual ways :)
20100323 21:18:18< Crab_> gabba: well, you will still be able to let the devs share their undoable moves when they'll test it out, so, it's possible, and easier.
20100323 21:18:40< gabba> Crab_: so both prototypes would be acceptable?
20100323 21:18:53-!- EdB [~EdB@tss37-1-89-84-18-220.dsl.club-internet.fr] has quit [Remote host closed the connection]
20100323 21:20:16< gabba> Crab_: (just to make sure I understood well... I'd hate to start on an idea, to realize there was a misunderstanding)
20100323 21:20:38< Crab_> gabba: speaking for myself, I would love to see any of the prototypes.
20100323 21:21:30< Crab_> gabba: but I think that it is important for you to interest other devs, as well.
20100323 21:22:44< gabba> O other potentially interested devs, are you listening ?
20100323 21:22:45< mordante> Ivanovic, the announcement po has an interesting string... "</a>" (won't break anything, but looks odd)
20100323 21:23:06< mordante> gabba, ?
20100323 21:24:11< gabba> mordante: heh, I'm a bit at a loss as to how and to whom I should push this prototype idea
20100323 21:24:29< mordante> Ivanovic, "What's New in Wesnoth 1.8" the capitalized N of New
20100323 21:24:29< gabba> mordante: or if I should just implement something and come back
20100323 21:24:48< Ivanovic> mordante: a little late for changes to the announcement!
20100323 21:24:59< gabba> mordante: (if you followed my discussion with Crab_ )
20100323 21:25:02< Ivanovic> yes, there is a reason why it is frozen and yeah, i don't care too much about this tiny one
20100323 21:25:11< mordante> gabba, haven't followed the entire discussion, can you give a short summery
20100323 21:25:22< Darkas> Crab_: have you already tested the patch?
20100323 21:25:44< mordante> Ivanovic, would it be an idea to make the english version a translation as well to fix these typos?
20100323 21:26:11< Ivanovic> mordante: would be possible, you just would have to change some files like the makefile
20100323 21:26:13< gabba> mordante: well, there are three things here:
20100323 21:26:17< Ivanovic> in fact it would be *good* to do so
20100323 21:27:04< gabba> mordante: 1- Crab_ thinks I should probably tackle the undo moves vs WML events (especially 'sighted') as the same time as my "share undoable moves with allies"
20100323 21:27:20< Crab_> Darkas: yes, patch, but some formatting issues :)
20100323 21:27:39< Crab_> Darkas: you sometimes used spaces instead of tabs, lines with only tab, etc
20100323 21:27:40< Darkas> Crab_: what's with the formatting?
20100323 21:27:51< Darkas> Crab_: ok
20100323 21:28:05< Crab_> Darkas: (no need to redo the patch)
20100323 21:28:12< Darkas> then I have to reconfigure kate ;)
20100323 21:28:34< gabba> mordante: 2- He's suggesting I make a prototype of how shared undoable moves would look like, by basically faking some undos through WML
20100323 21:29:20< gabba> mordante: 3- Or (my take on it) I could make a prototype that just syncs all moves for everybody, and we'll worry about hiding them from enemies later
20100323 21:29:24< Crab_> Darkas: are you interesting in coding something else, really small,  today ?
20100323 21:29:35< Darkas> Crab_: yes
20100323 21:30:16< gabba> mordante: so now, the question is, should I just go ahead and code something up, or should I interest more people in this... ...
20100323 21:31:13< Darkas> I'll just be away for a few minutes, but I'll read it
20100323 21:33:00< Crab_> Darkas: in cfun_ai_execute_attack, you'll find a todo. double aggression = 0.5;//TODO: replace with side agression
20100323 21:33:13< gabba> General question: what's the name of the MP dev channel: #wesnoth-mp, right?
20100323 21:33:17< Crab_> Darkas: try to fix it
20100323 21:33:22< mordante> gabba, yes
20100323 21:33:28< gabba> thnx
20100323 21:34:20< Crab_> Darkas: the method to use is in src/ai/contexts.cpp, double readonly_context_impl::get_aggression()
20100323 21:34:22< mordante> the sighted part regarding events is rather tricky
20100323 21:34:50< Crab_> Darkas: the method is available in engine_lua. 
20100323 21:34:56< gabba> mordante: I gathered that much o_o
20100323 21:35:07< Crab_> Darkas: note that the value cannot be stored during ai creation, since it can change during the game
20100323 21:35:28-!- Ken_Oh [~briang@static-71-178-174-219.washdc.fios.verizon.net] has quit [Read error: Connection reset by peer]
20100323 21:35:40-!- Espreon [~espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20100323 21:35:58< mordante> gabba, regarding 3 I like the unix style more, send to nobody unless explicitly allowed
20100323 21:37:00< mordante> regarding what's the best 2 or 3 I'm not sure
20100323 21:37:12< gabba> mordante: I do as well, much better from a security point of view, but for a prototype?
20100323 21:37:58< mordante> the problem is most prototype code becomes production code 
20100323 21:38:17< gabba> mordante: anyways, if you agree that either 2 or 3 would be a good way to bolster my proposal, that's an added security
20100323 21:38:19< mordante> but you can either discuss it further with other dev or code a prototype
20100323 21:38:47< mordante> personally I think a prototype for a gsoc proposal makes your proposal look stronger
20100323 21:38:52< gabba> mordante: if I hack something up, I don't mind starting from scratch later
20100323 21:39:23< mordante> ok, just make sure you mention that in your application
20100323 21:39:41< gabba> mordante: ok. and thanks.
20100323 21:39:47< mordante> you're welcome
20100323 21:40:09< Crab_> Darkas: the point to start would probably be engine_lua::engine_lua - it has access both to readonly_context, which contains your get_aggression method, and it creates  a lua ai context
20100323 21:40:34< Crab_> so, it's a link where you can sneak a pointer/reference to ai::readonly_context to lua code
20100323 21:40:47< Crab_> readonly_context is stateful, 1 per ai
20100323 21:41:25< Crab_> Darkas: make the solution generic,in the future we'll probably need to move a lot of things from ai contexts to lua code
20100323 21:41:25< Darkas> Crab_: where do I find cfun_ai_execute_attack?
20100323 21:41:34< Crab_> in src/scripting/lua.cpp
20100323 21:42:54< Crab_> my preferred solution: right now, cfun_ai_execute_attack is a closure on side. make it a closure on a userdata containing a pointer to engine_lua / readonly_context.
20100323 21:43:25< Crab_> (that is, cfun_ai_execute_attack gets side from lua_upvalueindex(1) )
20100323 21:43:45< gabba> see you all later
20100323 21:44:09< Crab_> Darkas: see the start of LuaKernel::create_ai_context where such c-closures are created
20100323 21:44:16< mordante> bye gabba 
20100323 21:44:30-!- gabba [~gabba@70.35.163.201] has left #wesnoth-dev []
20100323 21:44:52< Crab_> we can pass a pointer to readonly_context (or, better, to engine_lua to which we can add an accessor to context) , to  LuaKernel::create_ai_context 
20100323 21:45:13< Crab_> then, make those functions use that pointer as an upvalue
20100323 21:45:34< Crab_> then, retrieve, get the pointer, dereference, call get_aggression(), voila
20100323 21:46:32< Crab_> this way, those 'attack/move/all other future' functions will have access to the ai context - object which'll contain everything they need to know
20100323 21:46:38< Soliton> gabba: the more undoability the better. nothing wrong with letting the computer help you plan.
20100323 21:48:02< Darkas> Crab_: where does the pointer to readonly_context gets passed to LuaKernel::create_ai_context?
20100323 21:48:24-!- qemqemqem [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has quit [Ping timeout: 245 seconds]
20100323 21:48:28< Crab_> Darkas: nowhere. but the caller can do it.
20100323 21:48:40-!- qemqemqem [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has joined #wesnoth-dev
20100323 21:48:42< Darkas> Crab_: ah, so I have to add it
20100323 21:48:43< Crab_> the caller is engine_lua::engine_lua 
20100323 21:50:36< Crab_> note: please work on files which don't include your changes
20100323 21:50:45< Crab_> (from the previous task)
20100323 21:50:52< Darkas> Crab_: ok
20100323 21:51:29< Darkas> Crab_: how do I get back to a clean repository in svn?
20100323 21:51:42< Crab_> svn revert
20100323 21:56:56< zookeeper> esr, do you have time to add a couple of image renaming rules to wmllint?
20100323 21:57:17< esr> zookeeper: Yes.
20100323 21:57:20< zookeeper> i'm getting rid of the old-style orcish warlord portraits
20100323 21:57:28< esr> Good thing.
20100323 21:58:12 * esr gently dubs zookeeper with the Anti-Fish of Encouragement
20100323 21:58:24< zookeeper> oh, great
20100323 21:58:35< zookeeper> here's how i figured they ought to be converted: http://wesnoth.pastebin.com/P8fx7ek3
20100323 21:58:47 * zookeeper will commit the image changes in a minute
20100323 21:58:55< esr> I need to come up with the opposite of a fish.  Perhaps a kitten?
20100323 21:59:11< zookeeper> i'd prefer a pie
20100323 21:59:47< esr> That's easily done.
20100323 21:59:49< Darkas> Crab_: then we don't need the side parameter in create_ai_context anymore, right?
20100323 22:00:08< Crab_> yes, context has a get_side() method
20100323 22:01:14< Zarel> zookeeper: Can you split a thread for me?
20100323 22:01:32< zookeeper> Zarel, sure
20100323 22:01:43< Zarel> All posts starting from here: http://www.wesnoth.org/forum/viewtopic.php?p=416811#p416811
20100323 22:01:46< CIA-80> zookeeper * r41717 /trunk/data/core/images/portraits/orcs/ (8 files in 2 dirs): Removed the old-style orcish warlord portraits, changed the current warrior portrait into a warlord portrait and added Lordbob's latest as the new warrior portrait.
20100323 22:01:52< Zarel> Split it into Ideas. Whatever title you want; I can edit it.
20100323 22:03:23< zookeeper> Zarel, done
20100323 22:03:43< Zarel> Thanks
20100323 22:04:02< mordante> fendrin, in case you did UtbS, could you also update the changelog?
20100323 22:04:44< esr> zookeeper: Shall I apply those changes to mainline now?
20100323 22:04:44< Espreon> What's this about UtBS?
20100323 22:04:52< mordante> Espreon, ?
20100323 22:05:04< zookeeper> esr, yeah, that'd be nice
20100323 22:05:17< esr> zookeeper: Working...
20100323 22:05:42< Espreon> mordante: What may have been done to it by fendrin?
20100323 22:06:45< mordante> Espreon, some parts of http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=561580
20100323 22:07:15< Espreon> Ah, I see.
20100323 22:07:46-!- wesbot changed the topic of #wesnoth-dev to: 1.8 commits allowed till March, 24th, 12:00 GMT, afterwards "testfreeze" | string/feature freeze active! | 69 bugs, 254 feature requests, 15 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100323 22:10:18< CIA-80> crab * r41718 /trunk/ (6 files in 5 dirs): Add support for AI candidate actions written in lua. patch LuaW_pcall to support LUA_MULTRET. Patch #1558 by Darkas.
20100323 22:10:24< Crab_> Darkas: so, it's in ^
20100323 22:10:45< Darkas> that's great :)
20100323 22:10:50< Crab_> yes :)
20100323 22:11:14< Crab_> that's very good that we've included it in 1.8
20100323 22:11:32< esr> zookeeper: Conversioin done, now running sanity check...
20100323 22:11:42< CIA-80> mordante * r41719 /trunk/ (4 files in 3 dirs): (log message trimmed)
20100323 22:11:42< CIA-80> Worked around a assertion failure in the lobby.
20100323 22:11:42< CIA-80> The bug seems rather rare and only boucman has reported it, since I
20100323 22:11:42< CIA-80> haven't been able to reproduce it I'm not conviced this fix really
20100323 22:11:42< CIA-80> works. But since the current code causes an assertion failure, there's
20100323 22:11:42< CIA-80> no risk of regressions.
20100323 22:11:43< CIA-80> Boucman when possible please test and look whether you get the error
20100323 22:12:05< mordante> boucman the log contains more info for you to read^
20100323 22:12:22< mordante> boucman the commit log, not the chan log
20100323 22:17:13< Darkas> Crab_: should the userdata for the closure be a engine_lua?
20100323 22:17:28< Darkas> or rather: does engine_lua has a readonly_context?
20100323 22:17:39< Crab_> yes, it has it initially
20100323 22:17:44< Crab_> and it gains a bigger context later
20100323 22:18:05-!- Darkas [~quassel@ppp-93-104-139-121.dynamic.mnet-online.de] has quit [Read error: Connection reset by peer]
20100323 22:18:10-!- Darkas [~quassel@ppp-93-104-139-121.dynamic.mnet-online.de] has joined #wesnoth-dev
20100323 22:18:40< zookeeper> mordante, btw, [unit_type] [portrait] isn't documented in the wiki
20100323 22:18:47< Darkas> Crab_: and where do I get the engine_lua from?
20100323 22:19:29< CIA-80> zookeeper * r41720 /trunk/data/core/units/orcs/Warlord.cfg: Added the [portrait] definition for orcish warlord.
20100323 22:20:13< Crab_> Darkas: well, if you create the lua ai context from engine_lua::engine_lua  ...
20100323 22:20:29< Crab_> Darkas: then, 'this' in there will point to engine_lua
20100323 22:20:46< zookeeper> ...1.8 has "hundreds of new and revised unit portraits"? :P
20100323 22:20:48< Darkas> Crab_: ah, so I pass this as a parameter too
20100323 22:21:04< mordante> zookeeper, true the syntax is not really fixed, wanted to do that during 1.8 but somehow never got around it
20100323 22:23:03< Crab_> Darkas: make sure that src/scripting/lua.hpp doesn't get any new #Include directives
20100323 22:23:22< timotei21> hey guys, I have a question... 
20100323 22:23:27< Darkas> Crab_: how?
20100323 22:23:35< CIA-80> esr * r41721 /trunk/data/ (19 files in 8 dirs): Oc portrait reference fixes. Goes with r41717.
20100323 22:23:45< Crab_> Darkas: well, if you need to pass pointers, you can just declare the 'class foo;
20100323 22:24:18< Crab_> like it's done for struct lua_State; in there
20100323 22:24:33< timotei21> I'm currently searching of what project to do as part of GSOC, and one question I have: if there is somebody else "wanting" to work on a feature, should I not pick it?
20100323 22:25:28< Ivanovic> timotei21: there will always be several people interested to work on all the features
20100323 22:25:42< Ivanovic> timotei21: so expect other candidates for the projects!
20100323 22:26:09-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100323 22:26:30< Ivanovic> last year we had >30 applications and could only accept 6 students
20100323 22:26:38< zookeeper> esr, thanks, i'll check it..
20100323 22:26:43< Ivanovic> this year we even have less mentors, so we can accept less projects, too
20100323 22:27:19< esr> zookeeper: I think there may be an error on line 7 of Warlord.cfg.
20100323 22:27:44-!- dtiger [~dtiger@dynamic-vpdn-93-125-116-119.telecom.by] has quit [Remote host closed the connection]
20100323 22:28:08< zookeeper> esr, err, actually no, but that'll still interfere with the new renaming rule...
20100323 22:28:09< Ivanovic> looking at the list of existing project proposals that we do provide this would lead to some 4 to 5 students proposing each idea if we get some 30 applications this year
20100323 22:28:32< esr> zookeeper: Yes, it will.  What can we do about that?
20100323 22:28:38< esr> Rename it?
20100323 22:29:06< zookeeper> esr, can you make wmllint not rename it if it's in a [unit_type]?
20100323 22:29:24< zookeeper> it ought to get renamed if it's for a single unit profile, but not if it's for a unit type profile
20100323 22:29:33< esr> zookeeper: Not easily.  That would require new code.
20100323 22:29:55-!- YogiHH [YogiHH@d040082.adsl.hansenet.de] has joined #wesnoth-dev
20100323 22:30:02-!- Sapient [~patrickp@wesnoth/developer/sapient] has joined #wesnoth-dev
20100323 22:30:12-!- YogiHH [YogiHH@d040082.adsl.hansenet.de] has quit [Changing host]
20100323 22:30:12-!- YogiHH [YogiHH@wesnoth/developer/yogihh] has joined #wesnoth-dev
20100323 22:30:22< zookeeper> esr, right...
20100323 22:30:25< zookeeper> how annoying
20100323 22:30:28< Sapient> Gambit: thou hast summoned me?
20100323 22:30:39< Gambit> Erm yes
20100323 22:30:48< Gambit> I noticed something weird and I was told to ask you if it was intentional
20100323 22:31:01< Gambit> [set_menu_item]'s transfer between scenarios in SP campaigns?
20100323 22:31:11< zookeeper> esr, how about detecting whether the .cfg being processed happens to be somewhere within a units/ directory?
20100323 22:31:22< zookeeper> (since unit type .cfg's are always in such a location)
20100323 22:31:22< Ivanovic> fendrin: you saw this bug? (assigned it to you) https://gna.org/bugs/index.php?15681
20100323 22:31:59< Sapient> Gambit: when I first wrote the feature I was more concerned about MP than SP
20100323 22:32:20< Sapient> so I'm not sure whether it was intended or not; but it seems reasonable enough
20100323 22:32:35-!- qemqemqem__ [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has joined #wesnoth-dev
20100323 22:32:39< Gambit> Well there is an easy enough work around I guess.
20100323 22:32:59-!- qemqemqem [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has quit [Ping timeout: 245 seconds]
20100323 22:33:24< Crab_> Gambit: I wondered about it some time ago, and, afair, I was told that it was intended.
20100323 22:33:56-!- mrEPIC [~mrEPIC@ppp-70-242-9-60.dsl.hstntx.swbell.net] has joined #wesnoth-dev
20100323 22:34:17< Sapient> for that matter, it would be nice to make persistent [event]s that persist across scenarios
20100323 22:34:30< timotei21> ok, thank you
20100323 22:34:32< Crab_> Gambit: in one of 1.7.x betas, I introduced a bug which allowed people to command ai leader's to move to any map location :))
20100323 22:34:52< Crab_> Gambit: by forgetting about [set_menu_item] persistence across scenarios
20100323 22:34:59< Darkas> Crab_: there's a problem
20100323 22:35:01< Gambit> Sapient can't that be done with a util included in _main.cfg
20100323 22:35:06< Crab_> Darkas: which  ?
20100323 22:35:16< Darkas> I can't just define readonly_context in lua.hpp
20100323 22:35:22-!- ancestral [~ancestral@12.145.225.25] has joined #wesnoth-dev
20100323 22:35:29< Sapient> Gambit: can it? I didn't know about such a util
20100323 22:35:30< Crab_> Darkas: why ? you pass a pointer to it?
20100323 22:35:37< Darkas> because in lua.cpp, I use readonly_context::get_side()
20100323 22:35:46< Crab_> Darkas: then include it in lua.cpp :)
20100323 22:35:55< Darkas> Crab_: ok ;)
20100323 22:35:58< Gambit> Sapient: You can include cfg files in other cfg files
20100323 22:36:01< Darkas> why not in lua.hpp?
20100323 22:36:14< Sapient> Gambit: of course. I know that 
20100323 22:36:28< Crab_> Darkas: if you include in lua.hpp, then all the files which include lua.hpp will have to be recompiled each time something in the ai changed.
20100323 22:36:37< Crab_> Darkas: and this is bad, and unnecessary.
20100323 22:36:49< zookeeper> esr, or maybe you can just remove that renaming rule and simply output something along the lines of "make sure you're referencing the right portrait on file.cfg:x" to the user?
20100323 22:36:49< Gambit> I think you could put events in one.
20100323 22:36:51< Darkas> Crab_: ok
20100323 22:36:58< Crab_> Darkas: it's better to keep the header dependencies to a minimum
20100323 22:37:06< Gambit> No I gues they wouldn't be in between the [scenario] tags then
20100323 22:37:11< Sapient> the point of a persistent event is that you don't need to edit static files by hand. You can dynamically create it
20100323 22:37:13-!- knee [~jakobkang@82.131.116.202.cable.starman.ee] has quit [Ping timeout: 264 seconds]
20100323 22:37:36< timotei21> and also, when choosing a project. We should start working on it starting right now, or instead working on some EasyCoding/NotSoEasyCoding or patches before?
20100323 22:37:48< zookeeper> Sapient, yeah, it'd be somewhat nice to not need to include events in every scenario of a campaign
20100323 22:38:09< zookeeper> but that's really only a minor inconvenience, it's just one line in every scenario after all.
20100323 22:38:16< Crab_> Darkas: e.g., there's files like src/formula_fwd.hpp, which are only there to reduce dependencies via recoupling formula interfaces from implementation
20100323 22:38:32< Crab_> s/recoupling/decoupling
20100323 22:39:07< Darkas> Crab_: is there any way I can access readonly_context from engine_lua?
20100323 22:39:34< Ivanovic> timotei21: those easy coding projects are there to give you an easier way into our codebase
20100323 22:39:36< Sapient> zookeeper: well I can't take credit for the idea. esr was the one who proposed it several months ago
20100323 22:39:43< Ivanovic> which is not a tiny and easy to grasp thing
20100323 22:39:54< Crab_> Darkas: this
20100323 22:40:08< timotei21> ok
20100323 22:40:10< Darkas> Crab_: this?
20100323 22:40:26< zookeeper> Sapient, i'm pretty sure it's been suggested years ago already ;)
20100323 22:40:36< Gambit> Sapient: If _main.cfg could have #on_scenario_load
20100323 22:41:01< Crab_> Darkas: ah, no
20100323 22:41:21< Crab_> there's ai_ 
20100323 22:41:46< Crab_> (which is from src/ai/composite/engine.hpp )
20100323 22:42:02< Sapient> Gambit: as long as we're day-dreaming about features, let's not make it ... ugly (*shudder*)
20100323 22:42:29< Gambit> :(
20100323 22:42:31< Sapient> [event] persistent=yes should be sufficient
20100323 22:42:58< esr> zookeeper: Wouldn't it be easier just to rename the portrait to something the renaming rule won't match?  After all, by hypothesis, you're only going to use it in that one spot.
20100323 22:43:00< Gambit> Yeah but you could do more than events with #on_scenario_load
20100323 22:43:04< Gambit> you could do side definitions.
20100323 22:43:08< Gambit> Or the schedule.
20100323 22:43:39< Sapient> I think you want  a templating mechanism for scenarios
20100323 22:44:00-!- Tesafilmchen [~quassel@p5B274386.dip.t-dialin.net] has joined #wesnoth-dev
20100323 22:44:13< Gambit> Would kill both birds.
20100323 22:44:27< Zarel> ...hrm.
20100323 22:44:32< Sapient> what you seem to be overlooking is that events can be spawned based on dynamic conditions
20100323 22:44:33-!- piva [~piva@linux.ime.usp.br] has joined #wesnoth-dev
20100323 22:44:46< Zarel> zookeeper: Hmm, you only split the first few posts... wasn't there a really long discussion about it?
20100323 22:44:47< zookeeper> esr, well, sure that'd work, i just don't like renaming things in a confusing way
20100323 22:44:53< piva> hello
20100323 22:45:08< Darkas> Crab_: ok
20100323 22:45:29< Gambit> Sapient: How would it be any different than including another cfg at the very begining of a scenario?
20100323 22:45:34< zookeeper> Zarel, err, didn't look like it. i assumed the ones i split were the last ones of the thread.
20100323 22:45:40< Crab_> hi, piva
20100323 22:45:43< Zarel> zookeeper: There's a "Split from this post onward" option in phpBB.
20100323 22:45:55< esr> zookeeper: Better than adding bug-prone code and a warning message that requires hand interrvention by the person running wmllint.
20100323 22:46:02< Sapient> Gambit: one is conditional, the other is unconditional
20100323 22:46:07< Zarel> The last two pages (starting here: http://www.wesnoth.org/forum/viewtopic.php?p=416920#p416920 ) are also exclusively discussing the proposal.
20100323 22:46:18< Zarel> Split again and merge? Sorry for all the trouble. :/
20100323 22:46:36< Sapient> Gambit: one is dynamic, the other is static
20100323 22:46:58< piva> Hey Crab_, I have some questions about a project suggested for Google Summer of Code
20100323 22:47:09< Crab_> you're welcome to ask them :)
20100323 22:47:24< zookeeper> Zarel, ok, so split from that post onwards, over 2 pages? just making sure here
20100323 22:47:30< Zarel> zookeeper: Yes.
20100323 22:47:36< Sapient> consider that the entire event could be dynamically constructed from user-supplied variables and then spawned at mid-scenario
20100323 22:47:52< Sapient> how would you do that with # syntax ? 
20100323 22:47:56< Zarel> And then merge the split topic with http://www.wesnoth.org/forum/viewtopic.php?f=12&t=29298
20100323 22:47:57< piva> I was interested in the project to extend Wesnoth LUA AI support
20100323 22:48:12< Crab_> that's good :)
20100323 22:48:24< piva> I'm a fairly decent programmer in C++, but I'm not very familiar with LUA
20100323 22:48:26< Gambit> Sapient: Okay.
20100323 22:48:41< zookeeper> Zarel, ok, is it right now?
20100323 22:48:41< piva> would that be a big problem, or is there space (and time) to learn it?
20100323 22:48:59< zookeeper> esr, i guess
20100323 22:49:24< Crab_> piva: well, lua-cpp bindings are tricky, but quite possible to learn fast. and lua is an easy language to learn. so, 'not very familiar with LUA' is not a blocker.
20100323 22:50:10< piva> Also, I like AI field very much. Is there any expected algorithmic background?
20100323 22:50:21< CIA-80> ilor * r41722 /trunk/src/gui/dialogs/lobby_main.cpp: fix lobby "Games: showing 0 out of N" issue
20100323 22:50:35< ilor> mordante, Ivanovic: ^
20100323 22:50:40< Zarel> zookeeper: Yes.
20100323 22:50:47< mordante> thanks ilor 
20100323 22:51:25< mordante> I'm off night
20100323 22:51:27< Darkas> Crab_: there's one more problem
20100323 22:51:29< Crab_> piva: during the project, tools for ai developers will be made, which will be used to improve some areas of the AI which currently need improvement. and, there's much room for various experiments. 
20100323 22:51:33< Darkas> the ai_ member is protected
20100323 22:51:33< Crab_> Darkas: which ?
20100323 22:51:42< Darkas> should I write a get method for it?
20100323 22:51:50< piva> I see, thanks Crab_
20100323 22:51:53< Crab_> Darkas: yes, create a get_readonly_context() for it
20100323 22:52:21< piva> So, if I were to be more interested in developing the AI rather than working on the tools and framework, would there be space for that?
20100323 22:52:27< Darkas> Crab_: in engine_lua or in engine?
20100323 22:52:30< zookeeper> esr, if you want to rename it somehow then go ahead, but really i wouldn't expect false positives in any UMC.
20100323 22:52:44< Crab_> Darkas: a good question. in engine, I think.
20100323 22:52:46-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20100323 22:52:53< Darkas> Crab_: ok
20100323 22:53:09< esr> zookeeper: Then I could just comment out that renaming rule?
20100323 22:53:11< zookeeper> also, i gotta be going, i'll finish checking the conversion tomorrow before the deadline and also commit the assassin portrait if tsi hasn't done it at that point
20100323 22:53:45< Crab_> piva: yes, there would be space. but the mandatory program is 'make lua ai feature-complete', and the rest is extra. the reason for it is that is harder to check concrete results in 'pure ai development'
20100323 22:53:47< zookeeper> esr, well, no, since pretty surely UMC uses the old pic for single unit profiles
20100323 22:53:59< zookeeper> but pretty surely not for unit type profiles
20100323 22:54:18< Crab_> piva: but, if you code the requried things fast, you'll have room for experiments and improvements
20100323 22:54:44< zookeeper> anyway, i'll be off now, i'll read the logs in the morning...
20100323 22:54:47-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit []
20100323 22:55:14< piva> Crab_: Thanks very much
20100323 22:55:26< Crab_> piva: anything else that you want to know ?
20100323 22:56:09< Crab_> fendrin: I've investigated the utbs ai slowness issue.
20100323 22:56:23< piva> Crab_: Are projects 'per student' or do projects get more than one studen assigned?
20100323 22:56:53< Crab_> piva: it is possible for one project to get two students (with slightly different proposals). it happened in the past for wesnoth. 
20100323 22:57:19< Crab_> piva: but,in general, 'students accepted < project ideas available'
20100323 22:57:32< piva> I see
20100323 22:58:06< piva> Crab_: Thanks for the answers, I'll surely be back if I come up with some ideas.
20100323 22:58:19< Crab_> piva: you're welcome
20100323 22:58:35< Crab_> piva: note : prototypes are welcome more than ideas :)
20100323 22:58:57< piva> Crab_: ok =D
20100323 22:59:19< piva> Crab_: Just one more thing, is there a place where I can check previous years student's accepted proposals?
20100323 22:59:50< Crab_> piva: crosslink the history of the summer of code ideas page versus google information for previous years
20100323 22:59:51< piva> Crab_: (And perhaps developed code during the project)
20100323 23:00:12< Crab_> piva: there's a page at google for that, but, most of it is integrated in wesnoth
20100323 23:00:24< timotei21> piva: here is the list: http://socghop.appspot.com/gsoc/program/list_projects/google/gsoc2009
20100323 23:00:41< piva> timotei21: Great, thanks!
20100323 23:00:44< CIA-80> ivanovic * r41723 /trunk/ (data/core/about.cfg po/wesnoth-low/de.po): updated German translation
20100323 23:00:48< timotei21> you're welcome:D
20100323 23:00:54< piva> Later!
20100323 23:01:01< Crab_> fendrin: cause of UtBS issue is some stupid WML (basically, they ask the ai to protect hundreds of locations at the same time, instead of asking to protect a single location with a radius around it) which hasn't worked before ([protect_location] was somewhat broken), but now it works as designed, so it's slow.
20100323 23:01:29< CIA-80> esr * r41724 /trunk/data/core/units/orcs/Warlord.cfg: Prevent wmllint from trying to convert an image it shouldn't
20100323 23:01:32-!- piva [~piva@linux.ime.usp.br] has quit [Quit: Saindo]
20100323 23:01:47-!- ancestral [~ancestral@12.145.225.25] has quit [Quit: Colloquy for iPhone - http://colloquy.mobi]
20100323 23:03:28< Crab_> fendrin: UtBS should be fixed, to protect not ~1500 locations but a few with a bigger protection radius.
20100323 23:04:19< Crab_> I'll review the mainline usage of protect_location tag
20100323 23:05:20-!- stikonas [~and@2002:4f84:b38b:b:226:9eff:fe0a:3e55] has joined #wesnoth-dev
20100323 23:05:21-!- stikonas [~and@2002:4f84:b38b:b:226:9eff:fe0a:3e55] has quit [Changing host]
20100323 23:05:21-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100323 23:05:43< Crab_> yes, only Under_the_Burning_Suns/scenarios/02_Across_the_Harsh_Sands.cfg misuses that old tag
20100323 23:07:00< Ivanovic> Crab_: and you got a good fix at hand?
20100323 23:07:02< Ivanovic> ;)
20100323 23:07:09-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has quit [Read error: Connection reset by peer]
20100323 23:07:26< timotei21> ok guys, I'm out
20100323 23:07:28< timotei21> good night
20100323 23:07:31-!- timotei21 [~timotei21@193.34.191.4] has quit [Quit: bye bye]
20100323 23:07:35-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has joined #wesnoth-dev
20100323 23:07:58< Ivanovic> esr: what is the status of the Dead Water rework?
20100323 23:08:00< Crab_> Ivanovic: well, it's quite simple, just something like 'protect each location in x=1-39 y=1-35' with 'protect x=18 y=17 with 'radius ~ 25'
20100323 23:08:13< Ivanovic> esr: is it ready for inclusion in trunk right after the 1.8 branch is created?
20100323 23:08:25< Crab_> Ivanovic: (the exact numbers have to be determined by looking at the map)
20100323 23:08:33< esr> Ivanovic: We believe it is, yes.
20100323 23:08:39< Ivanovic> esr: great
20100323 23:08:51< Ivanovic> i'd love to see this done this weekend
20100323 23:09:14< fendrin> Crab_: Thank you, I will have a look at the wml coding.
20100323 23:09:17< Ivanovic> currently i plan to open trunk for 1.9 development on friday
20100323 23:10:09< esr> Ivanovic: The author did a *remarkably* good job on DW.  In terms of both WML and prose quality it was the most ready-for-prime-time UMC I have ever seen.
20100323 23:10:14< Crab_> fendrin: old [protect_location] tag has a radius attribute, which should be used instead of specifying a big group of locations
20100323 23:10:28< Ivanovic> esr: cool!
20100323 23:10:43< esr> Ivanovic: Do you actually want it grafted into 1,8 before release, is that why you're asking?
20100323 23:10:52< Ivanovic> no!
20100323 23:11:02< Ivanovic> i just want to make sure that it is added to the dev tree as early as possible
20100323 23:11:52< Ivanovic> you know, "adding big, new features as early in the dev process as possible and the likes"
20100323 23:11:53< Ivanovic> ;)
20100323 23:13:23< esr> OK.
20100323 23:13:54< esr> What sorts of string changes *should* fo into 1,8 before 1.9 forks, if any?
20100323 23:14:33< Darkas> Crab_: is there any way I can test my changes?
20100323 23:15:12< stikonas> adding an entirely new campaign on the last day is ludicrous :)
20100323 23:15:23< Crab_> the easiest way is to add a std::cerr << "Hey, I'm using aggression "<<aggression_value<<std::endl;
20100323 23:15:26< Ivanovic> uhm, those that were delayed due to the stringfreeze
20100323 23:15:42< Crab_> Darkas: default value is 0.4
20100323 23:15:59< Crab_> Darkas: campaigns will see different things.
20100323 23:16:00< Ivanovic> there is lots of stuff i think that AI0867 and fendrin wanted to have commited that currently is commited in a, uhm, "strange" form including the _ in the string so that the strings are not translateable
20100323 23:16:17< Darkas> Crab_: so I just have to run a testgame?
20100323 23:16:18< Ivanovic> in general 1.8.x should be bugfix only
20100323 23:16:18< Crab_> Darkas: but, you can load it in the lua test scenario
20100323 23:16:26< Darkas> Crab_: ok
20100323 23:16:29< Ivanovic> so only string changes that are required for fixing bugs *should* go in
20100323 23:16:45< Ivanovic> esr: IIRC you had some bugs that would be fixed in a nicer way with some additional string, right?
20100323 23:17:01< Crab_> Darkas: ai includes attacks with 'default aggression',  and '0.9' aggression
20100323 23:17:09< Crab_> Darkas: in  my_ai:do_moves(), that is
20100323 23:17:19< esr> Ivanovic: Two of them, yes.  That's why I was wondering.
20100323 23:17:24< Crab_> post a patch (without about.cfg entry this time :) )
20100323 23:17:34< Ivanovic> esr: those are perfectly fine for 1.8 once tagged
20100323 23:17:34< Crab_> (after you verify that it works :) )
20100323 23:17:53< Ivanovic> (and yes, for those there is the merge window
20100323 23:18:01< esr> Ivanovic: Noted.
20100323 23:18:20-!- mpavel [~Pavel@cpc1-dund3-0-0-cust363.sgyl.cable.virginmedia.com] has joined #wesnoth-dev
20100323 23:18:33< Ivanovic> esr: so yes, it would be great if you already started to gather a list of those strings so that they can be easily added once i tagged on thursday
20100323 23:19:45< esr> Ivanovic: Related issue: Should we clear the list of ad-hoc string fixes in pofix.py at the 1.8 boundary? It should be cleaned out sometime, otherwise obsolescent ones will pile up forever. 
20100323 23:20:03< Ivanovic> yes, this should be done for trunk once it is branched off
20100323 23:20:21< Ivanovic> planned to do so myself anyway
20100323 23:20:30< Darkas> Crab_: in the loop, I have to call ai::engine_lua** data = (ai::engine_lua**)lua_newuserdata(L, sizeof(ai::engine_lua*)); *data = engine; right?
20100323 23:20:52< Darkas> instead of pushing the value for the side
20100323 23:21:10< esr> Ivanovic: OK, I'll leave that to you then.
20100323 23:21:43< Crab_> Darkas: the idea is correct
20100323 23:22:05< Darkas> Crab_: good
20100323 23:22:13< Ivanovic> esr: okay
20100323 23:22:23< Darkas> what is with the side pushed on the stack before the loop?
20100323 23:22:24< Crab_> Darkas: but, you only need to store a pointer..
20100323 23:22:46< Crab_> Darkas: http://pgl.yoyo.org/luai/i/lua_pushlightuserdata
20100323 23:22:58< Darkas> Crab_: so only light userdata?
20100323 23:23:13< Crab_> it should be enough
20100323 23:23:58< Crab_> since we need it here only for c++ code ,for now
20100323 23:24:11< esr> Ivanovic: One thing you can reassure your translators about, BTW, is that there will be less mainline churn in the future.  My long task of bring the mainlne prose up to what I consider decent standards is nearly done.  Unut descriptions still need some work, though.
20100323 23:24:26< Ivanovic> esr: and i think i will (in the 1.9.x series) add a comment behind each string stating when it was added
20100323 23:24:43< esr> Ivanovic: Good idea,
20100323 23:24:47< Ivanovic> (in pofix.py)
20100323 23:25:04< esr> (I understood :-) )
20100323 23:25:08< Ivanovic> not sure yet how i will do so, either the version (1.9.0+svn, ...) or the date
20100323 23:25:11-!- Mythological_ [Mythologic@77.28.112.101] has joined #wesnoth-dev
20100323 23:25:23< Ivanovic> since revisions would suck if we switched the version control system so something else
20100323 23:25:24< Ivanovic> ;)
20100323 23:25:35< Darkas> Crab_: how do I get my pointer back then?
20100323 23:25:44< esr> Ivanovic: Yes, I've been dealing with that very issue recently.
20100323 23:25:51-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Quit: Leaving]
20100323 23:26:29< Crab_> 'what is with the side pushed on the stack before the loop?' - push it (take from get_side()
20100323 23:26:34-!- Mythological [Mythologic@77.28.123.167] has quit [Ping timeout: 246 seconds]
20100323 23:26:49< Crab_> Darkas: it's for ai.side - if lua code wishes to now the side
20100323 23:27:32< Darkas> Crab_: ok
20100323 23:27:53< Ivanovic> AI0867, Crab_, esr, fendrin, ilor, loonycyborg, Sapient, shadowmaster, Soliton, YogiHH, everyone else who cares: please remember that you have only a little more than 13 hours left for commits that are meant to go into 1.8.0
20100323 23:28:20< Darkas> Crab_: and how do I get my pointer back from the light userdata?
20100323 23:28:31< Ivanovic> from then on trunk is frozen and you are only allowed to commit after asking me until i either revoke this rule (due to some really serious issue) or till i tagged 1.8.0
20100323 23:28:46< esr> Ivanovic: I have no pending work.  I'm mainly watching for easy-to-shoot-down WML bugs and wmllint glitches at this point.
20100323 23:28:51-!- mrEPIC [~mrEPIC@ppp-70-242-9-60.dsl.hstntx.swbell.net] has quit [Quit: Leaving]
20100323 23:29:03< YogiHH> Ivanovic: I am fine, too
20100323 23:29:06-!- Shakey [~jbailey@12.190.80.225] has quit [Quit: It was a new day yesterday, but it's an old day now.]
20100323 23:29:13< Ivanovic> esr: is this maybe an easy to fix bug? https://gna.org/bugs/index.php?15681
20100323 23:29:17< Ivanovic> got *no* idea...
20100323 23:29:18< Espreon> esr: O RLLY?
20100323 23:29:43< Sapient> Ivanovic: if I was going to change something before 1.8 was released, it would be this -- http://www.wesnoth.org/forum/viewtopic.php?f=4&t=29167
20100323 23:29:47< esr> Ivanovic: Looking...
20100323 23:30:01< Sapient> but, I don't plan to change anything
20100323 23:30:15< Darkas> Crab_: I already found out
20100323 23:30:22< Crab_> Darkas: good. how ?
20100323 23:30:26< Ivanovic> that is: once i bumped the version to 1.8.0 (which i will do tomorrow right when the freeze starts!) everyone should test trunk as excessivly as possible
20100323 23:30:26< Crab_> Darkas: cast it ?
20100323 23:30:32< Darkas> lua_touserdata
20100323 23:30:35< Crab_> ah
20100323 23:30:40< esr> Ivanovic: It may not be a bug at all.  fendrin is the one to take that.
20100323 23:30:43< Darkas> and then just cast it to ai::engine_lua
20100323 23:30:50< Darkas> code seems working
20100323 23:30:55< Ivanovic> esr: okay, so i assigned it correctly
20100323 23:31:21< Crab_> Darkas: then (if it works in a test), post a patch 
20100323 23:31:29< Darkas> I got lots of 'Hey, I'm using aggression 0.4' messages now :)
20100323 23:31:34< Crab_> cool :)
20100323 23:33:26< Darkas> Crab_: I just defined readonly_context& get_readonly_context(); in engine.hpp, should I add any virtual keyword or something else?
20100323 23:33:28< CIA-80> thespaceinvader * r41725 /trunk/ (18 files in 4 dirs): Add and wire new Giant Rat images by Stern to DiD. Remove obsolete image. Updates changelogs, campaign credits.
20100323 23:33:47< Crab_> Darkas: no, not needed now
20100323 23:33:58< Darkas> Crab_: ok
20100323 23:34:20< esr> Ivanovic: Just as general info, my planned projects for 1.9 are (1) Wings of Victory (which is blocked now as fendrin is busy with more orgent things) and (2) game-threaded interface for the load-game dialog (which is blocked on mordante converting that dialog to GUI2).
20100323 23:36:36< fendrin> Ivanovic: I believe 15681 is in fact a feature.
20100323 23:38:08< esr> fendrin: Leacve a comment so noting, then, and mark it Invalid.
20100323 23:39:24-!- Remiel_ [~Remiel@c-76-16-143-181.hsd1.il.comcast.net] has joined #wesnoth-dev
20100323 23:39:53< Darkas> Crab_: the patch is now there
20100323 23:39:57< Crab_> good
20100323 23:40:44-!- Appleman1234 [~Appleman1@CPE-124-191-176-143.oxqn1.cha.bigpond.net.au] has joined #wesnoth-dev
20100323 23:41:27< Darkas> have fun with it ;)
20100323 23:41:33< Crab_> :)
20100323 23:41:36-!- YogiHH [YogiHH@wesnoth/developer/yogihh] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6/20100115144158]]
20100323 23:42:43< Remiel_> Greetings everyone, potential Gsoc applicant here 
20100323 23:42:48< Crab_> hi, Remiel_
20100323 23:43:03< Remiel_> Ah, just the person I wished to chat with
20100323 23:43:24< Remiel_> I'm currently checking over the things requried to port the AI to lua
20100323 23:43:33< Crab_> that's good :)
20100323 23:43:48< Crab_> any questions ?
20100323 23:43:52< Darkas> Crab_: I'll leave now
20100323 23:43:57< Darkas> bye
20100323 23:44:00< Crab_> Darkas: bye :)
20100323 23:44:12< Remiel_> at the moment, no, but I'm trying to teach myself the basics of lua first so...
20100323 23:44:35-!- Darkas [~quassel@ppp-93-104-139-121.dynamic.mnet-online.de] has quit [Remote host closed the connection]
20100323 23:44:36< Crab_> Darkas: the patch looks good (but still a few minor quirks with formatting)
20100323 23:46:13< shadowmaster> Ivanovic: okay, good. I don't think I've got anything to help with atm, and I'll be busy for the following days
20100323 23:47:07< Crab_> Remiel_: well, lua itself should be pretty easy to learn :)
20100323 23:47:27< Crab_> Remiel_: lua c++ bindings are trickier, however.
20100323 23:47:48< Remiel_> Yes, lua seems similar to Scheme and Lisp
20100323 23:48:10< Remiel_> minus the parenthesis
20100323 23:49:18< Remiel_> and perhaps a dash of python
20100323 23:50:41< Crab_> Remiel_: in the past (~ a year ago), wesnoth had support for python scripting, as well. but it was removed due to security reasons
20100323 23:51:00< esr> Remiel_: The designer himself describes lua as very Scheme-like.
20100323 23:51:12< Remiel_> Gotcha
20100323 23:52:12< Remiel_> Is the assumption that I (or someone else) would use a library like tolua++ or that we would write the binding ourselves?
20100323 23:52:25< Crab_> Remiel_: see src/scripting/lua.cpp
20100323 23:52:39< Crab_> Remiel_: we use lua c++ api directly
20100323 23:52:40-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20100323 23:53:42-!- Gambit [~Gambit@pa-67-234-116-244.dhcp.embarqhsd.net] has quit [Ping timeout: 276 seconds]
20100323 23:53:51< Remiel_> Gotcha
20100323 23:55:28< Crab_> 20th March 07:14 <haoyu> eh, just curious, why you have a lua-binding written from scratch instead of using something like luabind? 
20100323 23:55:44< Crab_> 20th March 13:12 <loonycyborg> I asked silene why he didn't use luabind some time ago. iirc he said that it won't be necessary because he considered wesnoth's use of it simple enough to stick with C api. 
20100323 23:57:25-!- Gambit [~Gambit@pa-67-234-116-244.dhcp.embarqhsd.net] has joined #wesnoth-dev
20100323 23:57:36< Gambit> website is down again?
20100323 23:57:56< Gambit> http://downforeveryoneorjustme.com/wesnoth.org
20100323 23:58:27< shadowmaster> yeah, I can't log into the server
20100323 23:58:36< shadowmaster> oh wait, i cn
20100323 23:58:47< shadowmaster> server lag is horrible, though
20100323 23:58:48< Crab_> I just logged in, after a pause
20100323 23:59:19< Crab_> looks like network issues
20100323 23:59:33< shadowmaster> yeah, CPU usage is fine
--- Log closed Wed Mar 24 00:00:11 2010