--- Log opened Thu Mar 19 00:00:04 2009
--- Day changed Thu Mar 19 2009
20090319 00:00:04< CIA-70> loonycyborg * r33846 /trunk/packaging/windows/Wesnoth.nsi.in: Bump version number to 1.6 in the name of userdata directory used by windows installer.
20090319 00:00:14< esr_> I see.
20090319 00:04:57< Velory> Ivanovic: i generally explained  %)Practical considerations questions before when im explaining other questions so should i write them again ?
20090319 00:05:41< YogiHH> fendrin: ping
20090319 00:07:45< CIA-70> esr * r33847 /trunk/data/ (3 files in 3 dirs): Remove a dead comment annd update one that's still twitching.
20090319 00:08:20< mordante> Velory, not sure what you mean, can you also add a link to your page at the bottom of this one? http://www.wesnoth.org/wiki/SummerOfCodeIdeas
20090319 00:09:15-!- Zen_Clark` [n=user@99-136-80-191.lightspeed.rcsntx.sbcglobal.net] has quit ["ERC Version 5.3 (IRC client for Emacs)"]
20090319 00:09:17< Velory> im talking about
20090319 00:09:30< Velory> http://www.wesnoth.org/wiki/SoC_Information_for_Google
20090319 00:09:54< Velory> im creating my page and answering this questions
20090319 00:11:04-!- Blueblaze [n=Blueblaz@c-98-199-143-139.hsd1.tx.comcast.net] has quit []
20090319 00:11:06< Velory> and one thing should i write my ideas about improving AI to my page but i dont know this ideas worthy i wrote here before if u read
20090319 00:12:15< mordante> you should write your ideas to your AI page, once you did that we'll discuss them and you can fine tune them
20090319 00:12:26< Velory> hmm okey :)
20090319 00:12:34< mordante> your AI ideas to your page *
20090319 00:12:59< fendrin> YogiHH: pong
20090319 00:13:15< mordante> it's always nice to read something first and then discuss about how to improve it
20090319 00:13:17< Velory> how can i write with bold charaters ? and 20 punto maybe
20090319 00:13:20< YogiHH> fendrin: do we still have a scouting problem with the AI?
20090319 00:13:47< Velory> no options for that in editing page
20090319 00:13:56< mordante> ?
20090319 00:13:58< Velory> can i use <b> </b> ? :D
20090319 00:14:06< Velory> creating page
20090319 00:14:09< fendrin> YogiHH: yes.
20090319 00:14:24< YogiHH> fendrin: Can you try something about this, then?
20090319 00:14:35< fendrin> YogiHH: we do even still have the shuffling problem.
20090319 00:14:48< YogiHH> fendrin: you compile yourself, right?
20090319 00:14:58< Velory> ok i found
20090319 00:14:59< Velory> info
20090319 00:15:03< mordante> when editing the wiki page there's a link to the help, which explains which markup you can use
20090319 00:15:28< fendrin> YogiHH: No, it's usually done by the cpu, I just call it. :-)
20090319 00:15:37< YogiHH> :-P
20090319 00:15:56< Crab1> Velory: there is also a preview button to see how it will look like.
20090319 00:16:25-!- Corvvs [n=michaelm@pool-96-231-13-174.washdc.east.verizon.net] has joined #wesnoth-dev
20090319 00:16:44< Crab1> mordante: Hello. I'm also writing a GSoC student page (also about AI improvement :) )
20090319 00:16:50< YogiHH> fendrin: please try this: comment ai_village.cpp line 352 (setting the threat_multiplier) and remove the threat_multiplier from the subsequent line as well (leaving the rest as is).
20090319 00:17:01< mordante> seems the AI is popular again ;-)
20090319 00:17:08< Velory> :)
20090319 00:17:14< mordante> but welcome Crab1 
20090319 00:17:25< Velory> i saw that one :)
20090319 00:17:34< mordante> ?
20090319 00:17:37< YogiHH> fendrin: then see if it helps with scouting
20090319 00:17:56< fendrin> YogiHH: give me a moment please.
20090319 00:18:09< Crab1> mordante: well, if we can manage to make the AI better than it is now, maybe next year the AI topic will not be so popular )
20090319 00:18:18< mordante> :-)
20090319 00:18:31< mordante> last year it also was the hottest topic
20090319 00:18:58< mordante> and for some reason, no matter how good the AI is, people will love the idea to tinker with it
20090319 00:19:15< mordante> Crab1, did you play Wesnoth
20090319 00:19:19< Crab1> yes
20090319 00:19:42< mordante> Crab1, I assume you saw this page http://www.wesnoth.org/wiki/SummerOfCodeIdeas
20090319 00:19:45< Crab1> I've started last year, played all the SP campaigns in 1.4.2
20090319 00:19:47< Crab1> yes, I did
20090319 00:19:55< mordante> nice and nice
20090319 00:19:59< Crab1> i'm writing a page about myself at the moment
20090319 00:20:15< Crab1> mordante: and what did the students managed to do about the AI last year ? 
20090319 00:20:50< Crab1> mordante: for here http://www.wesnoth.org/forum/viewtopic.php?f=10&t=24511   Dave says that 'The state of the AI in 1.6 is somewhat embarrassing'
20090319 00:21:17< mordante> last year Sirp added a formula AI and Dragonking expanded on that quite a bit
20090319 00:21:51< mordante> with the formula AI we didn't change the bigger picture only the implementation
20090319 00:21:58< Crab1> mordante: was that formula AI included in 1.5.8 ?
20090319 00:22:36< mordante> not sure which release it was in first, but it's in 1.6 which has been finished today
20090319 00:26:34< Crab1> mordante: what is the naming convention for GSoC student wikipages ? is 'SummerOfCodeProposal_AI_Crab' good ?
20090319 00:26:41< CIA-70> mordante * r33848 /trunk/src/gui/widgets/window.cpp: 
20090319 00:26:41< CIA-70> Handle layout problems more gracefully.
20090319 00:26:41< CIA-70> Instead of asserting show an error to the user, with this change the
20090319 00:26:41< CIA-70> asserts should really indicate an engine error.
20090319 00:28:03< mordante> Crab1, we've no naming convention, but yours sounds good
20090319 00:28:09< Dragonking> Crab1: SoC is not enough to make good AI, so I think next year it still will be hot topic ;)
20090319 00:28:59< mordante> Crab1, please add a the summer of code Category to your page as well
20090319 00:30:10< Crab1> mordante: ok
20090319 00:30:53< Dragonking> Crab1:  and formula AI aim was to make it a good languate to implement AI for scenario designers, so no implementation of default AI was written in to for wesnoth
20090319 00:32:15< Crab1> Dragonkind: hello ) Yes, I've seen that formula AI is some kind of domain specific language to make AI development easier, but I'm yet to see its source.
20090319 00:33:49-!- ilor_ [n=user@wesnoth/developer/ilor] has quit [No route to host]
20090319 00:37:45< Velory> mordante: what is Summer of Code Category u mentioned?
20090319 00:38:02< Crab1> Velory, it's a tag to be added to a wiki page
20090319 00:38:22< Velory> how can i add this ?
20090319 00:38:26< Crab1> Velory: in a form [[Category:NAME]] 
20090319 00:38:34-!- alink [n=alink@78.129.18.215] has quit [Remote closed the connection]
20090319 00:38:45< YogiHH> good night, everyone
20090319 00:39:15< Velory> so what is our category ? :)
20090319 00:39:26-!- Corvvs [n=michaelm@pool-96-231-13-174.washdc.east.verizon.net] has quit [Read error: 110 (Connection timed out)]
20090319 00:39:26< Velory> i dont understand well :D
20090319 00:39:52< AI0867> [[Category:SummerOfCode]] ?
20090319 00:40:04< Crab1> Velory: look at any page in this category. it's [[Category:Summer of Code]]
20090319 00:40:31< Velory> ok
20090319 00:44:00-!- YogiHH [i=YogiHH@wesnoth/developer/yogihh] has left #wesnoth-dev ["Bye folks"]
20090319 00:47:06-!- Elvish_Pillager [n=eli@66-189-40-122.dhcp.oxfr.ma.charter.com] has quit ["Hi! I'm a quit message virus vaccine. If you see a quit message virus, don't replace your quit message with it!"]
20090319 00:48:50-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090319 00:54:39-!- Espreon [n=espreon@wesnoth/developer/espreon] has quit [Remote closed the connection]
20090319 00:55:16-!- ikarius [n=ross@smtp.gridironsystems.com] has quit []
20090319 00:58:20< mordante> night
20090319 00:58:57-!- mordante [n=mordante@wesnoth/developer/mordante] has quit ["Leaving"]
20090319 01:04:25< silene> Ivanovic: the AI of the 1.6 release is flooding stdout; someone forgot to remove his debug dumps...
20090319 01:08:44< silene> Ivanovic: please have it fixed before the release on sunday
20090319 01:10:12-!- BenUrban [n=benurban@c-68-49-10-243.hsd1.md.comcast.net] has joined #wesnoth-dev
20090319 01:11:09< Ivanovic> silene: it is not flooding anything here
20090319 01:11:17< Ivanovic> at least not in campaign play that i tested
20090319 01:12:00< Ivanovic> okay, in mp gameplay i get it
20090319 01:12:13< Ivanovic> that is: when playing mp with "formula ai based" opponents
20090319 01:12:20< Ivanovic> so either Dragonking or Soliton left something in
20090319 01:12:48< Dragonking> ah yes :<
20090319 01:12:55< Ivanovic> silene: and personally i got to say that i don't think this is critical enough to retag
20090319 01:14:07< Dragonking> Ivanovic: So shall I remove this debug info?
20090319 01:14:35< Ivanovic> yes, you should
20090319 01:14:40< Dragonking> ok
20090319 01:14:46< Ivanovic> though "rereleasing" with this patch is annoying
20090319 01:15:02< Ivanovic> i will better "just" mail this patch to all known packagers and that's it
20090319 01:15:15< silene> Ivanovic: it depends, i'm sure windows user will be happy to have a "stdout" file filling their harddrive when playing an mp game with an AI :-)
20090319 01:15:15< Soliton> Dragonking: print it to info level.
20090319 01:16:15-!- Blueblaze_ [n=Blueblaz@c-98-199-143-139.hsd1.tx.comcast.net] has joined #wesnoth-dev
20090319 01:29:19< Dragonking> Ivanovic: Fixed, sorry for that :<
20090319 01:29:20< CIA-70> dragonking * r33849 /trunk/src/formula_function.cpp: Made debug_print function print to info level, instead to stdout
20090319 01:30:25< Dragonking> I just used it to check if everything is ok till the last minute, to make sure that 1.6 will be ok, and forgot that it has to be removed :/
20090319 01:33:35-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has quit [Remote closed the connection]
20090319 01:39:44< Velory> ok at last  i created my wiki page :)
20090319 01:41:16< Velory> but can be some grammar mistakes or somethings that  i forgot cause i'm sleepy today :)
20090319 01:41:25< Velory> ivanovic are u there ?
20090319 01:41:59< Velory> hmm seems not :(
20090319 01:46:51< Velory> ok goodnite every1
20090319 01:51:03-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has quit ["night all"]
20090319 01:52:34-!- ABCD_ [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090319 01:52:35-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Client Quit]
20090319 01:57:16-!- silene [n=plouf@wesnoth/developer/silene] has quit [Read error: 60 (Operation timed out)]
20090319 02:02:19-!- loonycyborg [n=sergey@wesnoth/developer/loonycyborg] has quit ["Zzzzzzzzzzzzzzzzzzzzzzzzz"]
20090319 02:05:39< dfranke> aww, so silene beat me out for finding the first new bug against 1.6.
20090319 02:05:56< Velory> hmm so i will edit my page its seemed so weak to me after i looked up 2008 proposals:)
20090319 02:12:01-!- esr_ [n=chatzill@static-71-162-243-5.phlapa.fios.verizon.net] has quit [Remote closed the connection]
20090319 02:13:32< Dragonking> night
20090319 02:13:35-!- Dragonking [n=dk@wesnoth/developer/dragonking] has quit []
20090319 02:14:38-!- Velory [n=GProcess@81.213.229.223] has left #wesnoth-dev []
20090319 02:15:27-!- happygrue [n=George@wesnoth/developer/wintermute] has quit [Connection timed out]
20090319 02:18:43-!- ABCD_ is now known as ABCD
20090319 02:22:54-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090319 02:26:44-!- Corvvs [n=michaelm@c-69-139-145-2.hsd1.md.comcast.net] has joined #wesnoth-dev
20090319 02:32:53-!- PK [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has joined #wesnoth-dev
20090319 02:33:10-!- PK [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has quit [Client Quit]
20090319 02:47:45-!- Sirp [n=me@wesnoth/developer/dave] has joined #wesnoth-dev
20090319 02:48:22< Sirp> hi everyone.
20090319 02:50:56-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090319 02:52:16< Shadow_Master> 9hi
20090319 02:52:33< Sirp> hi Shadow_Master 
20090319 02:52:44< Shadow_Master> accepted into GSoC? Nice! :)
20090319 02:52:47< Sirp> so I guess we are not yet flooded with SoC hopefuls? :)
20090319 02:52:50< Sirp> Shadow_Master: yeah
20090319 02:53:17-!- esr_ [n=chatzill@static-71-162-243-5.phlapa.fios.verizon.net] has joined #wesnoth-dev
20090319 02:54:37< Crab1> Sirp: hello. I am a student that wants to apply for a GSoC - improving the AI in Wesnoth. I am currently writing my wikipage and looking through the source of FormulaAI to learn 'the current state of things'. Can you point me towards some small bug or enhancement that I may try to fix\add in the AI, to better familiarize myself with the code ?
20090319 02:55:25-!- Zen_Clark [n=user@99-136-80-191.lightspeed.rcsntx.sbcglobal.net] has joined #wesnoth-dev
20090319 02:57:53-!- Shadow_Master_ [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090319 02:58:26-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has quit [Nick collision from services.]
20090319 02:58:38< Sirp> Crab1: well to begin with, is there a particular reason you are interested in improving the AI vs some other area you could work on? The AI is much more complicated than any other area we are encouraging people to work on for SoC, and so we'd only consider students who have some especially strong background in AI development.
20090319 02:58:42-!- Corvvs [n=michaelm@c-69-139-145-2.hsd1.md.comcast.net] has left #wesnoth-dev []
20090319 02:59:16-!- Shadow_Master_ is now known as shadowmaster
20090319 03:00:14< Sirp> however, in any case, to begin with on the AI I'd recommend trying to develop a small AI of your own that tries to play using a particular strategy or works well on a specific map, and show it to us.
20090319 03:01:13-!- newt [n=newt@c-98-212-201-198.hsd1.il.comcast.net] has joined #wesnoth-dev
20090319 03:01:14< Crab1> Sirp: Well, I looked through other 'areas of interest' in the wiki, and I this objective looked most interesting to me, because it is hard and it requires creative thinking.
20090319 03:01:57-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090319 03:02:01< newt> you guys getting much interest from SoC?
20090319 03:02:23< Sirp> Crab1: Fair enough. :) Do you have any background or experience in writing AI?
20090319 03:02:25< Sirp> newt: a little so far.
20090319 03:03:25< Crab1> Sirp: ok, should I try to develop a small AI based on conventional c++ (as in src/ai.cpp ) or should I try use your domain specific language (formulaAI) ?
20090319 03:05:40< Sirp> Crab1: a good sign that you know enough to ask this question. :) You could really go either way depending on what you try to accomplish (I'm making this open ended, since as you said, an AI developer needs to show they can be creative) ... however you could try developing it using the formula AI and then make any improvements to the formula AI system that you need to in C++ to make your AI work.
20090319 03:07:54-!- Espreon [n=espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20090319 03:08:37< Crab1> Sirp: regarding experience in writing AIs - none practical experience, but I had some university courses in which the algorithms which can be used for writing AIs were discussed (for example, genetic algorithms, which can be used to quickly get the 'good enough' next move in various turn-based games as a result of optimization of a function that analyzes the position on the 'board' and possible moves)
20090319 03:11:30< Crab1> Sipr: ok, I'll try to play with the AI a bit, and I'll see what I can do. Is there any documentation on FormulaAI (beyond the concepts in the wiki) ?
20090319 03:12:26< Sirp> I think the documentation in the wiki is all there is.
20090319 03:12:41< Sirp> though I think there is more than 'concept' documentation in the wiki
20090319 03:13:31< Crab1> Sirp: ok )
20090319 03:18:02-!- shadowmaster [n=ignacio@wesnoth/developer/shadowmaster] has quit [Read error: 110 (Connection timed out)]
20090319 03:20:07< newt> re-implementing save games doesn't seem as bad as the wiki page makes it out to be
20090319 03:21:49-!- Blueblaze_ [n=Blueblaz@c-98-199-143-139.hsd1.tx.comcast.net] has quit []
20090319 03:21:52< Sirp> newt: ?
20090319 03:22:24-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has quit [Success]
20090319 03:22:31< newt> it just sounds like you'd need to figure out what save games need, then re-implement slightly hackish code
20090319 03:23:08< Sirp> newt: I don't know what wiki page you're referring to or why you would want to re-implemented saved games. :)
20090319 03:23:46< newt> [edit] Savegame reorganization
20090319 03:23:47< newt> The savegame formats of Wesnoth for single p
20090319 03:23:49< newt> ..
20090319 03:24:09< newt> something about scalability for multiplayer campaigns, whatnot
20090319 03:24:36< crimson_penguin> how can I apply that patch?
20090319 03:24:37< Sirp> newt: to begin with it'd be nice to see a link to the specific wiki page you're talking about. :)
20090319 03:24:49< newt> http://www.wesnoth.org/wiki/SummerOfCodeIdeas
20090319 03:25:08< newt> ^_^;
20090319 03:25:52< Sirp> okay...
20090319 03:27:58< Sirp> and where does the wiki page make it out to be really hard?
20090319 03:28:00< Sirp> I don't see that
20090319 03:28:03< crimson_penguin> Ivanovic: how can I apply that patch?
20090319 03:28:15< newt> err
20090319 03:28:23< Sirp> crimson_penguin: I'm pretty sure Ivanovic is deep in REM sleep right now.
20090319 03:28:25< newt> http://www.wesnoth.org/wiki/SoC_Ideas_Savegame
20090319 03:28:35< crimson_penguin> oh yeah, I guess he's 8 hour ahead
20090319 03:28:47< newt> i dunno.. something about the language used just feels like they're making it out to be difficult
20090319 03:29:02< Sirp> okay.
20090319 03:29:03< Soliton> crimson_penguin: you mean r33849?
20090319 03:29:09< crimson_penguin> Soliton: yeah
20090319 03:29:11< Sirp> FWIW I think it will be difficult.
20090319 03:29:11< newt> might just be reading more into it than i should
20090319 03:29:24< crimson_penguin> I saved it to a file, but alternately, could I get it from svn somehow?
20090319 03:29:31< Sirp> in fact, I didn't know that was one of the ideas, I must admit I'm a little nervous about that idea.
20090319 03:29:39< Soliton> crimson_penguin: svn diff -c r33849 > patch
20090319 03:29:40< crimson_penguin> oh yeah, I guess I could just svn up src/thatfile
20090319 03:29:51< crimson_penguin> ok, thanks
20090319 03:30:46< crimson_penguin> Soliton: uhh, that didn't seem to do much
20090319 03:30:50< Ivanovic> crimson_penguin: this patch is directly from svn
20090319 03:31:03< crimson_penguin> ah ha! he's here!
20090319 03:31:14< Ivanovic> to use it just run this in your "extracted tarball dir": patch < patchname
20090319 03:31:36< crimson_penguin> that's what I tried to do, but it doesn't know what to patch then
20090319 03:32:00< crimson_penguin> "can't find file to patch at input line 3"
20090319 03:32:44< newt> i'm not sure, it seems like it would be more work, not necessarily harder than the other projects
20090319 03:32:52< newt> assuming you find all the save-game related code, that is :X
20090319 03:33:31< Ivanovic> crimson_penguin: in the "file to patch" just tell it to use src/formula_function.cpp
20090319 03:33:58< crimson_penguin> patch: **** malformed patch at line 11: namespace game_logic {
20090319 03:34:25< Ivanovic> beside this, this should work, too: patch -p0 < patchname
20090319 03:34:29< crimson_penguin> I don't suppose I can just up my repository to that?
20090319 03:34:45< crimson_penguin> that gets the same error
20090319 03:34:48< crimson_penguin> anyway, brb
20090319 03:35:25< Soliton> well, you can just get the modified file since nothing else changed.
20090319 03:35:33< Sirp> newt: I'm nervous about it because there is a fairly large potential for bugs.
20090319 03:35:41-!- Lord_Aether [n=castle@207.212.136.10] has joined #wesnoth-dev
20090319 03:36:46< Ivanovic> crimson_penguin: just fetch this file and place it in src/formula_function.cpp and you are done: http://svn.gna.org/viewcvs/*checkout*/wesnoth/trunk/src/formula_function.cpp?rev=33849
20090319 03:36:57< newt> sadly, that's the nature of the beast
20090319 03:37:19< newt> especially with code whose effect spans multiple files
20090319 03:37:31-!- Turuk_ [n=Turuk@FUSE-WS1-208-102-202-188.fuse.net] has joined #wesnoth-dev
20090319 03:37:48< Ivanovic> newt: do not underestimate the effords connected to the savegame mess
20090319 03:38:02< Ivanovic> since it basically comes to "find a nice way that handles *everything*"
20090319 03:38:24< crimson_penguin> Ivanovic: that's a good method :)
20090319 03:38:28< crimson_penguin> and I better go
20090319 03:38:37< Ivanovic> and i am finally off to bed, too
20090319 03:38:39< Ivanovic> n8
20090319 03:39:03< Ivanovic> newt: beside the AI stuff this is probably one of the biggest possible tasks to tackle
20090319 03:44:51-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Success]
20090319 03:49:36-!- Blueblaze_ [n=Blueblaz@c-98-199-143-139.hsd1.tx.comcast.net] has joined #wesnoth-dev
20090319 04:08:17-!- JW1 [n=X@c-71-57-85-168.hsd1.il.comcast.net] has quit []
20090319 04:23:03-!- Turuk_ [n=Turuk@FUSE-WS1-208-102-202-188.fuse.net] has quit [Success]
20090319 04:23:18-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090319 04:27:34-!- Lord_Aether [n=castle@207.212.136.10] has quit []
20090319 04:42:50-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 104 (Connection reset by peer)]
20090319 04:43:05-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090319 05:06:32-!- Zen_Clark [n=user@99-136-80-191.lightspeed.rcsntx.sbcglobal.net] has quit ["There is not point... In anything"]
20090319 05:07:42-!- Turuk_ [n=Turuk@FUSE-WS1-208-102-202-188.fuse.net] has joined #wesnoth-dev
20090319 05:10:17-!- Sirp_ [n=me@c-71-198-3-104.hsd1.ca.comcast.net] has joined #wesnoth-dev
20090319 05:10:58-!- Sirp_ [n=me@c-71-198-3-104.hsd1.ca.comcast.net] has quit [Client Quit]
20090319 05:14:56-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Connection timed out]
20090319 05:52:29-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090319 05:59:44-!- Turuk_ [n=Turuk@FUSE-WS1-208-102-202-188.fuse.net] has quit [Read error: 110 (Connection timed out)]
20090319 06:07:15< crimson_penguin> Ivanovic: Mac package up - hopefully it's all good, and if not, I can upload another on Saturday
20090319 06:11:04-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev
20090319 06:43:09< Crab1> It looks like I've found a bug in rev. 33849 - under specific circumstances (specific map, specific race) the AI goes into a endless loop.
20090319 06:43:09< Crab1> I am able to reproduce it every time (choose specific map, specific race,  do nothing for 2 turns). I've done a backtrace - it loops in formula_ai::play_turn     at src/formula_ai.cpp:1626
20090319 06:43:09< Crab1> 1626	        while(make_move(move_formula_,callable)) { }
20090319 06:43:09< Crab1> How do I proceed further ?
20090319 06:47:24< Sirp> Crab1: well, ideally find out why make_move is returning true all the time.
20090319 06:47:39< Sirp> Crab1: make_move() is meant to actually make a move if it returns true
20090319 06:47:49< Sirp> Crab1: meaning that it would eventually have to return false
20090319 06:47:58< Sirp> Crab1: so find out why it's not doing that.
20090319 06:50:24< Crab1> Sirp: so, it must either  (1) make a move and return true (2) do not move anything and return false  (3) return false for any other reason ?
20090319 06:55:00< Sirp> Crab1: yes (1) or (2)
20090319 06:55:12< Sirp> Crab1: if it doesn't make a move and yet returns true, that would cause a problem.
20090319 07:01:08-!- silene [n=plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20090319 07:07:31-!- ettin [n=jorda@wesnoth/developer/ettin] has joined #wesnoth-dev
20090319 07:09:09-!- Sirp [n=me@wesnoth/developer/dave] has quit ["leaving"]
20090319 07:20:40-!- ettin_ [n=jorda@wesnoth/developer/ettin] has quit [Read error: 110 (Connection timed out)]
20090319 07:26:53-!- silene [n=plouf@wesnoth/developer/silene] has quit [Read error: 110 (Connection timed out)]
20090319 07:44:45-!- cyber_11 [n=muratcan@78.165.103.11] has left #wesnoth-dev []
20090319 07:50:11< Crab1> aha, finally figured it out. in  ai::move_unit ( src/ai.cpp:704 ) , unit will not be moved if the destination square is not empty (which is reasonable), and  ai::move_unit will return 'from' to tell that no move was done. but, formula_ai::execute_variant (src/formula_ai.cpp:1904) ignores that return value and instead checks 'is the destination square now occupied'. Therefore, if the AI persistently tries to move unit from A to occupied 
20090319 07:53:10-!- silene [n=plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20090319 07:55:55-!- zookeeper [n=l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20090319 08:02:04-!- noy [n=Noy@wesnoth/developer/noy] has quit [Read error: 54 (Connection reset by peer)]
20090319 08:02:13-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev
20090319 08:05:56-!- BenUrban_ [n=benurban@c-68-49-10-243.hsd1.md.comcast.net] has joined #wesnoth-dev
20090319 08:10:04-!- turin [n=turin@168.215.249.247] has quit [Read error: 110 (Connection timed out)]
20090319 08:10:25-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit ["This computer has gone to sleep"]
20090319 08:11:32-!- xonev [n=chatzill@59.92.71.167] has joined #wesnoth-dev
20090319 08:22:54-!- BenUrban [n=benurban@unaffiliated/benurban] has quit [Read error: 110 (Connection timed out)]
20090319 08:28:17-!- ikarius [n=ross@216.27.182.3] has joined #wesnoth-dev
20090319 08:35:45-!- xonev [n=chatzill@59.92.71.167] has quit ["ChatZilla 0.9.84 [Firefox 3.0.7/2009030422]"]
20090319 08:43:10-!- noy [n=Noy@wesnoth/developer/noy] has quit [Read error: 104 (Connection reset by peer)]
20090319 08:45:09-!- noy [n=Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090319 08:47:38-!- Dragonking [n=dk@wesnoth/developer/dragonking] has joined #wesnoth-dev
20090319 09:19:34-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090319 09:23:53-!- Blueblaze_ [n=Blueblaz@c-98-199-143-139.hsd1.tx.comcast.net] has quit []
20090319 09:28:41-!- noy [n=Noy@wesnoth/developer/noy] has quit ["GO, GET TO THE CHOPPAH!!!"]
20090319 09:38:23< Crab1> Dragonking, can I talk to you about data/ai/formula/recruitment.fai ?
20090319 09:59:36-!- Crab1 [n=crab@c.218.175.a530.sta.adsl.cyfra.net] has quit ["Leaving."]
20090319 10:03:04< silene> Ivanovic: guess what, there is yet another flood, again caused by the AI: it queries the defense of "off_map" tiles, which it shouldn't since they obviously have none; this fills stderr at an impressive rate
20090319 10:03:09< silene> Dragonking: ?
20090319 10:15:16-!- loonybot [n=loonybot@79.139.247.203] has joined #wesnoth-dev
20090319 10:15:58-!- loonycyborg [n=sergey@79.139.247.203] has joined #wesnoth-dev
20090319 10:19:27-!- grzybacz [i=grzywacz@kolos.math.uni.lodz.pl] has joined #wesnoth-dev
20090319 10:34:14-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090319 10:48:50< Ivanovic> moin
20090319 10:57:10-!- ilor [n=user@wesnoth/developer/ilor] has quit [Read error: 113 (No route to host)]
20090319 11:00:50< Rhonda> http://lists.alioth.debian.org/pipermail/pkg-games-devel/2009-March/010807.html might be an interesting thread to some (Ivanovic prodded me to also show it here. :))
20090319 11:01:26-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20090319 11:01:49< Ivanovic> Rhonda: it was not me, it was fendrin
20090319 11:01:52< Ivanovic> ;)
20090319 11:02:59< Rhonda> Oh, right.
20090319 11:10:43< Ivanovic> zookeeper: https://gna.org/bugs/index.php?13216
20090319 11:11:03< Ivanovic> Dragonking: just assigned you a bug report: https://gna.org/bugs/index.php?13218
20090319 11:11:42< Ivanovic> Dragonking: it might be a good idea to have a look at "which blocking states are possible" since there seem to be several for the AI
20090319 11:13:03-!- busfahrer [n=busfahre@wsl18.informatik.uni-ulm.de] has joined #wesnoth-dev
20090319 11:16:28< Ivanovic> Dragonking: and it might also be good to intorduce a new loglevel for formulaAi-WIP so that you can just get this and don't happen to spam the normal stdout with the messages
20090319 11:16:29< Ivanovic> ;)
20090319 11:16:52< Ivanovic> silene: how common are maps with 'off' terrain in mp?
20090319 11:17:23< Ivanovic> as in: do we make wide usage of it in mainline or is it mainly used in UMCs?
20090319 11:18:01-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Connection reset by peer]
20090319 11:18:06< silene> Ivanovic: it is used by all the mainline maps that have a special shape
20090319 11:18:16-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090319 11:19:10< Ivanovic> grml, agrh
20090319 11:19:38< silene> Ivanovic: at first glance, i would say 18 mainline maps
20090319 11:21:17< Ivanovic> looks to me like we should seperate the AI log level into two (as in C++AI and FORMULAAI) and have both output shown in a general AI loglevel
20090319 11:21:49< Ivanovic> this way it would be easier to just work with a formulaai loglevel when working on this stuff and no need to print it all to normal std::out
20090319 11:22:06< silene> Ivanovic: note that this is not the AI logging, that's the generic engine complaining that the AI is doing something wrong
20090319 11:22:31< Ivanovic> silene: oh
20090319 11:22:37-!- Espreon [n=espreon@wesnoth/developer/espreon] has quit [Read error: 60 (Operation timed out)]
20090319 11:22:53< silene> and you don't want to remove this message, since it is quite useful for wml designers, i guess
20090319 11:23:38-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090319 11:24:30< Ivanovic> do you mean this one? error config: Defense '-1' is '< 0' reset to 0 (100% defense).
20090319 11:24:49< silene> yes
20090319 11:24:59< Ivanovic> this could be related to some changes by Soliton, too
20090319 11:26:24< silene> yes, that's what i first thought too, but this doesn't seem to be the case here
20090319 11:29:15< silene> (in particular, i didn't notice this flood with the old ai, while i had noticed soliton's bug with it)
20090319 11:35:34< Ivanovic> right, this was about movement costs, not about defense value
20090319 12:07:03-!- velory [n=gprocess@212.175.117.27] has joined #wesnoth-dev
20090319 12:07:14< velory> hi everyone
20090319 12:09:42< Ivanovic> hi velory
20090319 12:11:29< velory> hi  i nearly completed my wiki page :)
20090319 12:11:30< velory> http://www.wesnoth.org/wiki/SummerOfCodeProposal_Velory
20090319 12:11:50 * Ivanovic will now first try to wake up with a cup of tea
20090319 12:13:01< velory> hehe ok
20090319 12:32:23< Ivanovic> silene: i just have been running a test with three ais playing against each other on one of the "lots of off-terrain" maps
20090319 12:32:47< Ivanovic> silene: there are many lines with the defense value stuff around, but this is still bearable and enough to have fixed in 1.6.1
20090319 12:35:31< silene> my favorite mp test is 9p_Merkwuerdigliebe, since it allows to exercize the tactical ai with all the factions at once, and there the flood is hell; but i understand your point
20090319 12:37:10< Ivanovic> sure, in such a testcase it is really annoying especially with the formula stuff spamming ~20 lines per site turn
20090319 12:37:23< Ivanovic> the defense value stuff is not this often per site turn
20090319 12:37:58< Ivanovic> beside this: 1) only the host gets the message in a "real" multiplayer match and 2) those setups are rather seldom since most often human opponents are used
20090319 12:39:06< silene> yes, that's why i said i understand your point
20090319 12:39:35< silene> but i disagree with the "not often"; i have about 300 log lines per turn because of the defense bug
20090319 12:40:12< Ivanovic> hmm, interesting
20090319 12:40:17-!- crimson_penguin [n=ben@S0106000d88bbc764.vc.shawcable.net] has joined #wesnoth-dev
20090319 12:40:35< Ivanovic> after >10 turns with 3 ai players on another map there were about 50 lines
20090319 12:41:33< silene> i guess it scales quadratically: each ai scans all the units, which is already proportional to the number of ai
20090319 12:41:43< Ivanovic> yeah, could b
20090319 12:41:45< Ivanovic> e
20090319 12:44:43< silene> anyway, regardless of the logging issue, i hope the ai gets fixed; because if it takes into account all the tiles, including the ones the units cannot reach, it is bound to take dumb decisions on these special-shape maps
20090319 12:45:37< silene> (for 1.6.1 i mean)
20090319 12:46:11< Ivanovic> yes, would be a good thing to have
20090319 13:02:27< Ivanovic> okay, i currently work on the list of "tags" for wesnoth at http://delicious.com/gsoc2009/
20090319 13:04:16< Ivanovic> hmm, should i add the MySQL tag?
20090319 13:04:29< Ivanovic> since one of our proposals requires it, so it is possible, right?
20090319 13:05:35-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090319 13:07:54< Shadow_Master> hi
20090319 13:09:16< velory> ok my lesson starting goodbye
20090319 13:09:34< velory> and this is my wiki page again
20090319 13:09:35< velory> http://www.wesnoth.org/wiki/SummerOfCodeProposal_Velory
20090319 13:09:49< velory> i will edit again soon
20090319 13:09:53-!- velory [n=gprocess@212.175.117.27] has left #wesnoth-dev []
20090319 13:10:37-!- ilor [n=user@wesnoth/developer/ilor] has quit [Read error: 145 (Connection timed out)]
20090319 13:15:45-!- devvrat [i=d2d43703@gateway/web/ajax/mibbit.com/x-028f1eccdafdd297] has joined #wesnoth-dev
20090319 13:17:54-!- BenUrban_ is now known as BenUrban
20090319 13:19:31-!- devvrat [i=d2d43703@gateway/web/ajax/mibbit.com/x-028f1eccdafdd297] has quit ["mibbit.com: Bbbye"]
20090319 13:28:15< Shadow_Master> sudo invoke-rc.d apache2 stop
20090319 13:28:47< Shadow_Master> g}
20090319 13:29:32< Shadow_Master> psmouse.c: TouchPad at isa0060/serio1/input0 lost synchronization, throwing 4 bytes away. <- ??
20090319 13:32:32-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has quit ["Lost terminal"]
20090319 13:32:49-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090319 13:39:57-!- tushar_hailstone [i=d2d40558@gateway/web/ajax/mibbit.com/x-4754ce53553ad317] has joined #wesnoth-dev
20090319 13:44:56< Shadow_Master> wesbot: seen vladimirslavik
20090319 13:44:56< wesbot> Shadow_Master: The person with the nick VladimirSlavik last spoke 18d 15h ago.  11d 19h ago person left:
20090319 13:54:51-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has quit [Client Quit]
20090319 14:08:34-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Client Quit]
20090319 14:09:02-!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090319 14:16:18-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090319 14:20:49-!- yann [n=dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has quit [Read error: 110 (Connection timed out)]
20090319 14:24:28-!- tushar_hailstone [i=d2d40558@gateway/web/ajax/mibbit.com/x-4754ce53553ad317] has quit ["http://www.mibbit.com ajax IRC Client"]
20090319 14:38:09-!- BenUrban [n=benurban@unaffiliated/benurban] has quit ["Power failu"]
20090319 14:39:42-!- elias [n=elias@allegro/developer/allefant] has joined #wesnoth-dev
20090319 14:39:56< Shadow_Master> all sites but wesnoth.org are timing out for me. how odd.
20090319 14:40:12< Ivanovic> your ISP knows what you really need
20090319 14:40:14< Ivanovic> ;)
20090319 14:41:40-!- tushar_hailstone [i=d2d40558@gateway/web/ajax/mibbit.com/x-d6bf36dca305e87f] has joined #wesnoth-dev
20090319 14:46:08-!- Baufo [n=thomas@wesnoth/developer/baufo] has joined #wesnoth-dev
20090319 14:46:34-!- Dragonking [n=dk@wesnoth/developer/dragonking] has quit [Read error: 110 (Connection timed out)]
20090319 14:49:51< Baufo> Ivanovic, is now a good time to commit the updates for the man pages I told you about some time ago?
20090319 14:50:06< Ivanovic> jupp
20090319 14:50:14< Ivanovic> the perfect time
20090319 14:50:14< Baufo> ok, thanks
20090319 14:53:59< Shadow_Master> argh... I cannot find any clue as to which (of all the syndication mods for phpbb3) mod is the one we use for RSS/Atom feeds
20090319 14:55:17< Shadow_Master> I wish phpbb could use real plugins like e.g. blosxom (perl) does :/
20090319 14:55:44 * Shadow_Master dies of frustration.
20090319 14:55:48-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has quit [Client Quit]
20090319 15:00:07-!- Dragonking [n=dk@wesnoth/developer/dragonking] has joined #wesnoth-dev
20090319 15:06:40< Ivanovic> hi Dragonking
20090319 15:09:17< CIA-70> baufo * r33850 /trunk/doc/man/wesnothd.6: update for the wesnothd man page: added information about the user handler stuff
20090319 15:10:17-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090319 15:18:00-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit ["Leaving"]
20090319 15:18:07-!- crimson_penguin [n=ben@S0106000d88bbc764.vc.shawcable.net] has joined #wesnoth-dev
20090319 15:19:57-!- Crab_ [n=crab_@c.218.175.a530.sta.adsl.cyfra.net] has joined #wesnoth-dev
20090319 15:24:12< Baufo> Ivanovic, out of interest: why is it that we wait a few days between releasing an announcing?
20090319 15:24:40< Ivanovic> Baufo: several reasons
20090319 15:24:43< Ivanovic> 1) packagers need time
20090319 15:24:52< loonycyborg> sf needs time :)
20090319 15:24:53< Ivanovic> 2) translators need time to get the announcement done
20090319 15:25:08< Ivanovic> 3) sf.net needs lots of time to propagate the packages
20090319 15:25:25< Ivanovic> 4) giving us some extra time to find "really, really bad bugs"
20090319 15:25:58-!- Thrawn [n=chatzill@pool-71-126-232-234.bstnma.east.verizon.net] has joined #wesnoth-dev
20090319 15:26:10< Baufo> makes sense :)
20090319 15:26:13< Baufo> thanks for the info
20090319 15:32:42< grzybacz> I need a C++ wizard. ;-)
20090319 15:32:51< grzybacz> Where's Sirp when he's needed? :P
20090319 15:32:59< grzybacz> silene: there?
20090319 15:33:35< loonycyborg> There's a slight chance that I know some wizardly C++ stuff you need :P
20090319 15:34:29< grzybacz> Ah, there.
20090319 15:34:49< grzybacz> what does & on a local reference variable return?
20090319 15:35:28< grzybacz> so... A & ref = somethingReturningRefToA(); cout << &ref;
20090319 15:36:20< loonycyborg> cout << &ref will probably print the pointer to object ref refers to..
20090319 15:36:38< elias> it would be logical - but then, I only know C
20090319 15:37:13< silene> grzybacz: yes
20090319 15:37:22< grzybacz> silene: ^
20090319 15:37:45< loonycyborg> References are just pointers with value semantics.
20090319 15:37:53< grzybacz> Right.
20090319 15:38:03< grzybacz> But you don't really want to test references for NULL.
20090319 15:38:05< grzybacz> Or do you? ;-)
20090319 15:38:06< Ivanovic> could everybody have a short look at the description text and the tags i added for wesnoth here? http://delicious.com/gsoc2009/game_development
20090319 15:38:24< Ivanovic> that is: i added *many* tags even if they are only slightly related to the stuff...
20090319 15:38:36< silene> grzybacz: it returns a pointer pointing to the memory located behind the actual object
20090319 15:40:04< loonycyborg> { int* i = NULL; int& j = *i; cout << j; }
20090319 15:40:19< grzybacz> silene: so, why constructions like if(&ref == 0) are not common and instead most code would just die trying to use a "null reference"?
20090319 15:40:24< loonycyborg> loonybot: { int* i = NULL; int& j = *i; cout << j; }
20090319 15:40:26< loonybot> Undefined behavior detected.
20090319 15:40:43< loonycyborg> So you can have NULL references :)
20090319 15:41:00< elias> you can reference a pointer which is NULL you mean :P
20090319 15:41:15< silene> grzybacz: references are supposed to point to objects, there is no such thing as *NULL reference
20090319 15:41:54< silene> grzybacz: a compiler is allowed to assume your test always fail
20090319 15:41:56< grzybacz> silene: I can do a A & get() { return *((A *) 42); }
20090319 15:42:13< elias> loonycyborg:  ah, no, you actually can reference NULL that way - i always thought you couldn't
20090319 15:42:31< grzybacz> Tested with 2 compilers so far. None does assume anything, it works just like a pointer...
20090319 15:42:37< silene> grzybacz: sure, you can type an arbitrary sequence of character; that doesn't mean it has a valid semantic from a c++ point of view
20090319 15:43:02< grzybacz> that I know
20090319 15:43:29< grzybacz> I've just been puzzled that it's not prohibited by default...
20090319 15:44:57< silene> how could it be prohibited? maybe you know better than the compiler and the memory location 42 actually contains an A object; this is quite improbable and a compiler should warn about it, but it definitely cannot forbid it
20090319 15:45:42< loonycyborg> You're using C-style cast there, so you must know what you're doing :)
20090319 15:46:11< grzybacz> I'm not talking my particular case here. 42 was exeggarated.
20090319 15:46:28< loonycyborg> IIRC true C++ requires reinterpret_cast in this case..
20090319 15:46:39< grzybacz> Buy dereferencing a pointer to return it as a reference and *then* checking for a null reference somewhere in client code...
20090319 15:46:57< grzybacz> s/Buy/But/
20090319 15:47:43< loonycyborg> Check for a null reference *before* you dereference it..
20090319 15:48:05< grzybacz> haha, yes
20090319 15:48:09< grzybacz> that's what I would do
20090319 15:48:42< grzybacz> But I've just run across a piece of code which does that in the client code. A serious wtf when you read it for the first time...
20090319 15:49:18-!- alink [n=alink@wesnoth/developer/alink] has joined #wesnoth-dev
20090319 15:58:12< CIA-70> baufo * r33852 /trunk/src/multiplayer.cpp: remove line that does nothing
20090319 15:58:13< CIA-70> baufo * r33851 /trunk/src/multiplayer.cpp: made error message for unregistered nicks tranlatable
20090319 16:00:22< Soliton> Baufo: looks like you put two spaces into the error message. remove the untranslatable one some languages might do it differently...
20090319 16:01:19< Baufo> Soliton, hmm, why not the translatable?
20090319 16:02:03< silene> Baufo: because there are languages for which spacing does not mean anything
20090319 16:02:19< Baufo> I see
20090319 16:02:22< Ivanovic> " " + _(" This
20090319 16:02:37< silene> for instance, that would break the vertical aligment of chinese characters
20090319 16:02:38< Ivanovic> this part is not good, just make the spaced part normally translateable
20090319 16:02:50< Ivanovic> in fact i don't see a reason for the additional space
20090319 16:03:15-!- busfahrer [n=busfahre@wsl18.informatik.uni-ulm.de] has quit ["leaving"]
20090319 16:03:49< CIA-70> baufo * r33853 /trunk/src/multiplayer.cpp: fix unregistered nick error: remove untranslatable space
20090319 16:04:10< Baufo> Ivanovic, I do neither, it was a silly mistake :)
20090319 16:04:58< Ivanovic> :)
20090319 16:17:49-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 104 (Connection reset by peer)]
20090319 16:18:04-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090319 16:19:57-!- jp0186 [n=jp0186@77-56-56-38.dclient.hispeed.ch] has joined #wesnoth-dev
20090319 16:23:03-!- tushar_hailstone [i=d2d40558@gateway/web/ajax/mibbit.com/x-d6bf36dca305e87f] has quit ["http://www.mibbit.com ajax IRC Client"]
20090319 16:24:26< alink> Ivanovic: about debug output in a release, i was thinking maybe a script to detect obvious ones. It must be improved but something like: 
20090319 16:24:28< alink> svn diff src -r30000 --diff-cmd diff -x '--unified=0' | grep -E '^\+.*std::cout' 
20090319 16:24:43< alink> where the -r30000 is the rev number of the previous release
20090319 16:25:17< alink> probably need to add printf too, I think some dev use it
20090319 16:26:10< Ivanovic> alink: this will often also get some wanted output as in "error messages" that are not normal debug output
20090319 16:26:54< alink> yes but the content of the message may clarify it
20090319 16:27:54< alink> anyway, in doubt,  just drop the result on irc, and the dev who added it may recognize his debug code
20090319 16:28:25< alink> should be simple enough, unless we added ton of such lines
20090319 16:29:06< alink> with -r20000 i have only a dozen of such line and half of them are in comment
20090319 16:29:27< alink> and we are at r33853
20090319 16:30:55< alink> mmh that seems not many, maybe improving the regex will show more
20090319 16:33:56< elias> std::cerr maybe
20090319 16:34:27-!- Velory [n=GProcess@88.254.133.51] has joined #wesnoth-dev
20090319 16:34:33< Velory> hi
20090319 16:34:45< alink> elias: yes that show more change
20090319 16:34:49< alink> hi Velory
20090319 16:35:02< Soliton> why go through the diffs? just grep the source. there aren't many places where std::cout/cerr make sense.
20090319 16:35:28< alink> Soliton: well that was to detect only new introduction
20090319 16:35:59< alink> grep source will always show the same valid std::cout , unless we can mark them somehow
20090319 16:36:28< Soliton> the only valid places are somewhere in game.cpp probably.
20090319 16:36:49< alink> indeed, 81 matches for std::cout but mainly in dfool.cpp
20090319 16:37:51< Crab_> alink: you can mark valid ones with same-line comment ( // some_key_marking_valid), or replace them with a macro.
20090319 16:38:10< alink> if the only valid places are somewhere in game.cpp, maybe just rename that std::cout as main_cout
20090319 16:38:37< alink> or with a macro, indeed Crab_, even if i am not a fan of macro
20090319 16:40:18< alink> anyways, all of that start to be more complicated that my initial svn diff idea :-)   not sure that debug output problem worth the introduction of new convention
20090319 16:41:11< Soliton> s/introduction of new/proper expression of existing/
20090319 16:41:33-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090319 16:41:54< alink> hehe, but coders don't work like that ;-p
20090319 16:42:17-!- ikarius [n=ross@216.27.182.3] has quit []
20090319 16:42:25-!- stikonas [n=stikonas@ctv-79-132-187-171.vinita.lt] has joined #wesnoth-dev
20090319 16:42:51< alink> Ivanovic: btw s I suppose that po tools detect introduction of new translatable strings, or a similar svn diff trick is also usefull ?
20090319 16:43:00< alink> s/s//
20090319 16:43:10-!- irmorelle [i=929b362a@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090319 16:43:42< Soliton> i think we are mostly sticking to the convention of using our log macros except in dead code no one cares for like dfool.cpp.
20090319 16:43:45< zookeeper> Velory, please don't make a hundred separate wiki edits each just changing a few words, rather fix more issues at a time to keep the recent changes less cluttered.
20090319 16:44:03< Velory> im not changing anymore
20090319 16:44:06-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit ["Leaving"]
20090319 16:44:12< Ivanovic> alink: yeah, the po tools automatically get the strings when i run a pot-update
20090319 16:44:40< alink> Ivanovic: and then spot the new ones ?
20090319 16:45:00< Ivanovic> alink: internal gettext magic, don't ask me how it works, it just works...
20090319 16:45:02< Ivanovic> ;)
20090319 16:45:20< Ivanovic> probably a case of "just" parsing a list of files given
20090319 16:46:27< irmorelle> Turuk: wanted to talk with me?
20090319 16:46:47< alink> Ivanovic: so, i can't sneak a new one during string freeze without being noticed ;-p
20090319 16:47:27< Ivanovic> alink: correct
20090319 16:47:40< Ivanovic> since i always check if a new one is added in the po files or not
20090319 16:47:55< alink> i knew that this open-source thing was not a good idea <:o)
20090319 16:48:16< Ivanovic> sure, you can always add those untranslated messages to the commandline, but the matter is the stuff that is uservisible and as such normally marked translateable
20090319 16:48:17< Ivanovic> ;)
20090319 16:51:35< alink> hmm does someone has checked tiny-gui recently ?
20090319 16:52:39< alink> (i didn't, just asking)
20090319 16:54:21-!- irmorelle [i=929b362a@wesnoth/developer/shadowmaster] has quit ["http://www.mibbit.com ajax IRC Client"]
20090319 16:57:47< Ivanovic> alink: short version: no
20090319 17:01:16-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090319 17:01:25-!- shikadibot [n=1001@wesnoth/bot/shikadibot] has joined #wesnoth-dev
20090319 17:01:47< Shadow_Master> my uptime!! :(
20090319 17:02:42< Shadow_Master> and my firefox cache! :(
20090319 17:03:36-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 104 (Connection reset by peer)]
20090319 17:03:51-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090319 17:16:58-!- turin [n=turin@168.215.249.247] has joined #wesnoth-dev
20090319 17:18:39-!- ikarius [n=ross@smtp.gridironsystems.com] has joined #wesnoth-dev
20090319 17:33:30 * Ivanovic found a clear case of "won't do": https://gna.org/bugs/index.php?13221
20090319 17:33:40< Ivanovic> anyone who wants to mark it / comment on it?
20090319 17:34:55-!- ettin_ [n=jorda@wesnoth/developer/ettin] has joined #wesnoth-dev
20090319 17:35:31< alink> hehe, that seems to need only a few lines change, if wanted. But probably not wanted
20090319 17:36:49< alink> oh nevermind i misunderstoud it :-/
20090319 17:37:16< Shadow_Master> Sapient or someone else wrote an untested patch to remove randomness in a 1.0.x version of wesnoth IIRC
20090319 17:37:36< alink> I thought he wanted always the same random result after a reload
20090319 17:38:18< alink> "the most expected outcome to happen each time ", so basicaly only use 0% and 100% terrain defense ?
20090319 17:39:29< alink> oh, maybe he means that we must always have the average number of attack 
20090319 17:40:03< Shadow_Master> does it even matter? it's a FPI.
20090319 17:40:58< alink> well, if i don't undersatnd him, i can't say if it's a FPI ;-p
20090319 17:41:19< alink> but my brain is slow these days :-/
20090319 17:41:53< Shadow_Master> he's just asking to disable randomization of fight outcomes.
20090319 17:42:16< Shadow_Master> er, s/disable/add an option to disable/
20090319 17:44:05< alink> yes and he want always the most expected outcome, I assume he is talking about the damage calculations stats
20090319 17:44:34< alink> but looking at that, i see different most expected result for attacker and defender
20090319 17:44:59< Ivanovic> alink: basically it is: make wesnoth deterministic
20090319 17:45:11< alink> i am not sure that both are for the same fight
20090319 17:45:29< Ivanovic> if you are standing on a 70% defense terrain you will get 70% of the max attack value the attacker is able to influct
20090319 17:45:33< Ivanovic> stuff like this
20090319 17:45:52< Ivanovic> s/70% of the max attack value/30% of the max attack value
20090319 17:47:05-!- Blueblaze_ [n=Blueblaz@c-98-199-143-139.hsd1.tx.comcast.net] has joined #wesnoth-dev
20090319 17:48:14-!- ettin [n=jorda@wesnoth/developer/ettin] has quit [Read error: 110 (Connection timed out)]
20090319 17:48:33< alink> mmh I see now, indeed the same FPI as usual
20090319 17:48:58< Shadow_Master> "deterministic". I'll remember that.
20090319 17:50:20< alink> i still want to check how my version of always do the most expected result from damage calculations stats make sense, but that's now only my problem :)
20090319 17:50:28< fendrin> I don't know why this button doesn't exist for a long time. All this debates over randomness bore me.
20090319 17:50:39< Shadow_Master> then ignore those debates.
20090319 17:51:31< AI0867> Ivanovic: I'll do it
20090319 17:52:08< AI0867> done
20090319 17:52:45< Crab_> alink: you can simply warp the [0..1] distribution so it will lean towards the median values. Or, simpler way : for each hit, roll to hit 2N+1 times, declare a hit on N+1 hits. 
20090319 17:52:49 * Shadow_Master hands AI0867 a donut
20090319 17:53:39< Crab_> s/each hit/each attack/
20090319 17:53:49< alink> Crab, i don't undersand your way but i found how make mine works
20090319 17:54:25< alink>  the damage calculations would need to be bi-dimensional to see the most expected result. And we just always do that one. But that would be hard to visualize
20090319 17:54:48< alink> * most expected result for attacker and defender HP
20090319 17:55:36< alink> in fact, IIRC it's like that in memory for the algo
20090319 17:57:02< alink> mmh also need a rule to decide what to do for 50/50 cases
20090319 18:06:03-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has quit ["Lost terminal"]
20090319 18:06:55-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090319 18:09:06-!- sk- [n=sk@pool-68-162-68-65.phil.east.verizon.net] has joined #wesnoth-dev
20090319 18:09:32< sk-> argh, I can't connect to the 1.5 server with 1.6 anymore :(
20090319 18:09:47< Rhonda> why should you?
20090319 18:10:03< sk-> why shouldn't I?
20090319 18:10:49< Shadow_Master> why would you need to do that?
20090319 18:11:26< sk-> because I downloaded 1.6, deleted 1.5 and now I don't want to install/dl 1.5 again.
20090319 18:12:51< Rhonda> Why don't you connect to the 1.6 (stable) server, then?
20090319 18:13:53< sk-> is there such a thing?
20090319 18:14:06< sk-> I just connect to the "official server" within 1.6
20090319 18:14:08< Shadow_Master> multiplayer -> join official server
20090319 18:14:27< sk-> if I do that I get an error saying I can't join the server unless I have 1.5*
20090319 18:14:31< Shadow_Master> well, it isn't working for me. I just reported it in #wesnoth-mp
20090319 18:14:46< sk-> right
20090319 18:15:03< sk-> okay then
20090319 18:15:32< Rhonda> Hmm, does #wesnoth-mp want logging, too?
20090319 18:15:32< sk-> I guess I'll just wait until 1.6 is released.
20090319 18:15:41< Rhonda> Oh, right.
20090319 18:16:27< Rhonda> wesnoth  16949  0.0  0.0   5172  1036 pts/2    S    Mar03   0:00 /bin/sh /home/wesnoth/bin/run_server 1.5
20090319 18:16:30< Rhonda> wesnoth  17842  0.0  0.0   5172  1036 pts/1    S    Mar03   0:00 /bin/sh /home/wesnoth/bin/run_server 1.4
20090319 18:16:33 * Rhonda ponders ...
20090319 18:16:43< Ivanovic> Rhonda: the 1.5 server *is* also the new 1.6 server
20090319 18:16:46< Shadow_Master> I guess the version mask in the server.cfg is wrong
20090319 18:16:51< Shadow_Master> same for server2.wesnoth.org
20090319 18:16:56< Ivanovic> Shadow_Master: yesterday it was "once" correct
20090319 18:17:04< Shadow_Master> Ivanovic: can you try it _today_?
20090319 18:17:04< Ivanovic> ask Soliton to have a look and just wait a little...
20090319 18:17:13< Rhonda> Ivanovic: Then the config isn't adopted yet.
20090319 18:17:30< Ivanovic> looks like the redirect is broken again
20090319 18:17:38< Ivanovic> as in: in the redirect script 1.6 is (again) missing
20090319 18:17:48< Shadow_Master> ah, right, I logged into the 1.5 server with 1.6.0 yesterday
20090319 18:18:07< Rhonda> Ivanovic: But we do have a 1.6/wesnothd.cfg file too
20090319 18:18:26< Ivanovic> Rhonda: really interesting
20090319 18:18:34< Ivanovic> lets just wait for Soliton to clean it up
20090319 18:18:44< sk-> my other complaint, and this is a big one: the gryphon animations... they're cool the first time you see them, but after that they wear on you. I had to eventually turn off animations just to avoid seeing it. and I'm pretty cool with things like the Windmill animation which others abhore.
20090319 18:19:03< Shadow_Master> k
20090319 18:19:11< Rhonda> hmmmm
20090319 18:19:18< Rhonda> versions_accepted="1.5.13*,1.5.14*,1.5.15*,1.6*,test"
20090319 18:19:41< Ivanovic> Rhonda: like i said, the *redirect* server is broken (again)
20090319 18:19:47< Rhonda> ah
20090319 18:19:49< Ivanovic> since by default you connect to the svn server
20090319 18:19:50< Rhonda> which one is that?
20090319 18:19:56< sk-> Yeah, I was able to login yesterday too.
20090319 18:20:03< Ivanovic> and there should a redirect be set for 1.6* to the 1.5.x server
20090319 18:20:19< Ivanovic> lets just wait for Soliton to fix it again and for good
20090319 18:20:32< Rhonda> Ah.
20090319 18:20:34< Rhonda> See it.
20090319 18:20:38< Rhonda> trunk/wesnothd.cfg
20090319 18:21:02< Rhonda> I just don't know how it's started, otherwise I would just edit the file and give it a try.
20090319 18:21:48 * Shadow_Master couldn't even find those directories
20090319 18:22:36-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has quit []
20090319 18:23:02-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev
20090319 18:23:03-!- shikadibot [n=1001@wesnoth/bot/shikadibot] has quit ["Lost terminal"]
20090319 18:29:35< Rhonda> Shadow_Master: Well, that's the difference between a sysadmin and a casual user. ;)
20090319 18:30:21< sk-> shit, Ivanovic, I am getting the AI non-move bug again, except worse now with 1.6
20090319 18:30:31< sk-> Were any changes made to the AI?
20090319 18:30:51< sk-> The bug is different now, last time it was the second wave of AI that froze, now it is the first wave of AI.
20090319 18:30:52< Rhonda> Yes, AI problems were the reasons for the late 1.6
20090319 18:31:38< sk-> I am getting seriously slow performance when it is the AI's turn too.
20090319 18:31:57< sk-> It is also not even attacking when it has enemies within reach which is different.
20090319 18:32:08< sk-> I'd say the situation is worse than the initial bug report I filed.
20090319 18:33:07< Crab_> sk, here is one of the reasons for 'ai freeze' in current trunk - ftp://ftp.terraninfo.net/wesnoth/screen13218.png
20090319 18:33:51< sk-> well, this is in a survival situation without an AI leader.
20090319 18:34:45< alink> sk-: a lot of units on screen ?
20090319 18:35:12< sk-> This is just 5 random units appearing and not attack, moving, or anything. Before they'd collect the nearest village and attack any unit in range... now they don't move, don't collect villages, and don't attack.
20090319 18:35:16< Crab_> sk, it doesn't require a leader. that specific bug just requires to AI to try to move a unit into already occupied square.
20090319 18:35:54< sk->  I can file another bug report with a saved game and a way to duplicate the bug if that would help.
20090319 18:36:06< Rhonda> Crab_: Ouch - a fix for that would sound like it would break compatibility, not?
20090319 18:36:06< silene> sk-: definitely
20090319 18:36:59< Crab_> Rhonda, see https://gna.org/bugs/?13218
20090319 18:36:59< sk-> when i filed the first ai bug ivanovic did something that fixed it completely. something has changed between then and now where it is even worse than before.
20090319 18:37:27< Crab_> sk: post a save )
20090319 18:37:33< sk-> k
20090319 18:37:40< sk-> brb
20090319 18:38:07< Rhonda> Crab_: Yes, but it is a no-fix for 1.6, right? :(
20090319 18:38:50< Rhonda> ... not that I would mind trying to figure it out properly and check for chea^Wabus^Wnifty ways to take benefit of it.
20090319 18:39:13< Crab_> Rhonda: sorry, that question is best asked to someone else ) I'm new here
20090319 18:39:47< Rhonda> It sounds that a fix for that would make AI behave different - and that for sure is a nogo for a stable update I'm afraid. :(
20090319 18:39:54< Rhonda> ... why wasn't that discovered last week!
20090319 18:40:07< alink> Rhonda: I don't think that fixing a AI bug break compatibility
20090319 18:40:19< Soliton> sk-: fixed, sorry.
20090319 18:40:27< alink>  for network, replay or savegame
20090319 18:40:37< silene> Rhonda: because the code was not in last week? (why do you think i send an email to the ml complaining about the amount of changes after rc3?)
20090319 18:40:56< Rhonda> alink: If a unit would move differently on one client than on someone else's?
20090319 18:41:42< Rhonda> silene: ... and yet I get complains to not have pushed rc3 directly into debian unstable. Thanks for confirming that I did the right thing[tm] :)
20090319 18:41:50< alink> Rhonda: IIRC, for the network, we just send AI actions as it is an human player
20090319 18:42:24< Rhonda> alink: Ah, so the AI is only used on one system? Makes sense and sounds a good and save approach, then. :)
20090319 18:42:50< Rhonda> silene: Can you btw. hand me a link to that message? I would like to point the complainer to it.
20090319 18:43:30< alink> Rhonda: Yes, at least some of code that i saw was working like that, and i don't see why do differently
20090319 18:44:28< alink> s/some of code that i saw/all the network code that i saw
20090319 18:44:29< silene> Rhonda: https://mail.gna.org/public/wesnoth-dev/2009-03/msg00064.html but i'm not sure it is helpful to you
20090319 18:45:50-!- Blueblaze_ [n=Blueblaz@c-98-199-143-139.hsd1.tx.comcast.net] has quit [Read error: 60 (Operation timed out)]
20090319 18:46:42< alink> Rhonda: in fact, fixing a AI bug allowing the AI to cheat localy may prevent an OOS, because other client will not accept the invalid action
20090319 18:47:22-!- Sirp [n=me@wesnoth/developer/dave] has joined #wesnoth-dev
20090319 18:47:41< happygrue> Rhonda: personally, I would love logging for #wesnoth-mp
20090319 18:48:18< Rhonda> Now that I have done the logging by wesbot and not me I'm happy to expand the list of channels it is around, no problem with that.
20090319 18:48:26< happygrue> great
20090319 18:48:31-!- Thrawn [n=chatzill@pool-71-126-232-234.bstnma.east.verizon.net] has quit [Read error: 110 (Connection timed out)]
20090319 18:48:34< Sirp> Hello.
20090319 18:48:43< happygrue> hello Sirp 
20090319 18:48:49< Rhonda> I mean, wesnoth-dev-umc is on the list mostly because of that.
20090319 18:49:03< Rhonda> happygrue: Please remind me later, have to move.
20090319 18:49:08< happygrue> Rhonda: I see.  I was wondering if there was a reason that was logged but wesnoth-mp was not
20090319 18:49:11< happygrue> sure
20090319 18:49:25< happygrue> Rhonda: you haven't heard the last from me!
20090319 18:49:30< sk-> Soliton, That AI bug is fixed?
20090319 18:49:47< Soliton> no idea.
20090319 18:50:02< sk-> When you said "fixed" what were you referring to?
20090319 18:50:10< Soliton> login to the server.
20090319 18:50:18< sk-> ah k, cool :)
20090319 18:50:57< happygrue> sk-: so it was working under RC3 right?
20090319 18:51:10< happygrue> and you are saying things are more broken with 1.6?
20090319 18:51:18< sk-> Yes.
20090319 18:51:22< happygrue> :(
20090319 18:52:20< happygrue> sk-: you have not posted a bugreport yet, right?
20090319 18:52:36< sk-> no, i'm going to do it within about 10 min or so
20090319 18:52:38< happygrue> go ahead and do that, with how to reproduce it
20090319 18:52:40< happygrue> ok, great
20090319 18:55:29< Ivanovic> grml, damn ai problems, grml
20090319 18:58:26< happygrue> grml == grumble?
20090319 18:58:33< sk-> A really easy way to reproduce it is to just run an Orocia game and watch the AI pile up on the sides, if you approach it, it won't attack.
20090319 18:58:36< sk-> But I'll make a bug report.
20090319 18:58:56< happygrue> sk-: ok.  I don't have 1.6 yet anyway... but I'll look at that
20090319 18:58:57< Ivanovic> make a bug report and explicitly add the addons used
20090319 18:59:09< sk-> k
20090319 18:59:19< Soliton> sk-: post stderr.
20090319 18:59:38< Soliton> it might be recruitment.fai is trying to recruit even though there is no leader.
20090319 19:00:31< Ivanovic> hmm, right
20090319 19:00:53< Ivanovic> though if it does go on, it is not this formula stuff but "something else"(tm)
20090319 19:01:12< Soliton> it won't go on if it doesn't realise there is an error.
20090319 19:01:34< Soliton> or rather it will end turn.
20090319 19:09:19< sk-> alrighty: https://gna.org/bugs/index.php?13222
20090319 19:09:45< Ivanovic> sk-: stderr / stdout output?
20090319 19:09:57< sk-> I'm just noticing that, um... how do I get those?
20090319 19:10:03< Ivanovic> on linux?
20090319 19:10:07< sk-> No, Windows.
20090319 19:10:09< Ivanovic> just start from the terminal
20090319 19:10:22< Ivanovic> on windows it should be created in the binary folder as .txt file
20090319 19:11:18< happygrue> loonycyborg: two very minor packaging questions
20090319 19:12:15< happygrue> 1) is it easy to check where the game is installed, or somehow otherwise detect that it should install to c:\games\Wesnoth, rather than "program files"?
20090319 19:13:03< happygrue> 2) is it wise to also include version number in the start menu shortcut folder?  or are there reasons for not doing so?
20090319 19:13:25< happygrue> I notice, because I install a lot, and I add that bit every time...
20090319 19:13:37< happygrue> perhaps a normal user would not care
20090319 19:13:37< Ivanovic> sk-: ask one of the windows users in here if you need details
20090319 19:13:41< Ivanovic> i am off for a while
20090319 19:13:46< sk-> Okay, I found it.
20090319 19:13:56< sk-> They're attached.
20090319 19:14:20< sk-> there's a ton of errors related to the nearest keep thing.
20090319 19:14:27< happygrue> sk-: just installing 1.6 now, I'll try to confirm it.
20090319 19:16:11< loonycyborg> happygrue: There's a page in the installer that allows you to change install location.
20090319 19:17:26< loonycyborg> happygrue: Probably this is wise. At least for development version.
20090319 19:17:34< happygrue> sure, I do that.  I was just wondering if it is easy to find where wesnoth is intalled, and default to that directory instead
20090319 19:17:41< happygrue> loonycyborg: I agree
20090319 19:17:55< happygrue> since at least most WML folks probably have a few dev versions kicking around at any one time
20090319 19:18:41< sk-> happygrue, try orocia, it will be really easy to notice it.
20090319 19:18:42< loonycyborg> But perhaps stable versions should be handled differently? E.g. uninstall previous version if found.
20090319 19:18:48< sk-> remove fog for easy viewing :)
20090319 19:19:36< happygrue> loonycyborg: I don't know what the best way to handle stable would be.  pros and cons either way I'd say
20090319 19:21:00< happygrue> as long as no one complains, no reason to change it perhaps?  ;)
20090319 19:21:08< Rhonda> happygrue: The reason was that Shadow_Master asked me to do it and you didn't. That simple. :P
20090319 19:21:19< loonycyborg> happygrue: It currently always defaults to program files because each version has its own registry key.
20090319 19:21:23< happygrue> Rhonda: hehe
20090319 19:21:52-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090319 19:22:55< happygrue> loonycyborg: hmmm.  I wonder if most users really put their games in \program files...  Am I a weirdo by not putting it there?
20090319 19:23:05< happygrue> well, it's trivial anyway, I was just curious
20090319 19:23:32< loonycyborg> happygrue: Most users just don't care AFAICT.
20090319 19:24:18< sk-> I found it annoying that by default your userdata is stored in My Documents
20090319 19:24:26< Dragonking> Hi everyone
20090319 19:24:33< Soliton> sk-: remove ai_algorithm=formula_ai from mainline factions and it'll work.
20090319 19:24:33< Dragonking> Ivanovic: Around?
20090319 19:24:35< happygrue> hello Dragonking 
20090319 19:24:36< Dragonking> Crab_: Around?
20090319 19:24:42< Crab_> Dragonking, yes
20090319 19:24:46< Dragonking> Great.
20090319 19:24:46< Soliton> Dragonking: more formula ai issues.
20090319 19:25:03< Dragonking> Soliton: More than this on in bug report?
20090319 19:25:08< Soliton> Dragonking: does recruitment.fai check whether there is a leader?
20090319 19:25:09< Dragonking> one'
20090319 19:25:21< loonycyborg> sk-: Why? Is there a reason why having them at install location is more convenient?
20090319 19:25:44< Dragonking> Soliton: Not really.... I see it should 
20090319 19:25:51< Dragonking> Teamgames. bah
20090319 19:26:13< Soliton> Dragonking: i guess we'll better revert to default ai recruiting.
20090319 19:26:16< Rhonda> uh?
20090319 19:26:27< Dragonking> Soliton: Will be easiest for now.
20090319 19:26:40 * Rhonda blinks at Soliton :P
20090319 19:26:56< Dragonking> Soliton: Can you di it?
20090319 19:26:59< Dragonking> do'
20090319 19:27:04< Soliton> Dragonking: sure.
20090319 19:27:10< Dragonking> Soliton: Thank you
20090319 19:27:27< Rhonda> Soliton: Thanks for fixing the MP server, btw. :)  And don't blink like that.
20090319 19:27:29< happygrue> Rhonda: thanks :)
20090319 19:27:42< Dragonking> Soliton: Let's remove it for 1.6 and make it default for trunk once when we get 1.6 branch 
20090319 19:28:28-!- Turuk_ [n=Turuk@FUSE-WS1-208-102-202-188.fuse.net] has joined #wesnoth-dev
20090319 19:28:39< Soliton> Rhonda: if we wanted logging in #wesnoth-mp lobby would be there but i'll ask around if people mind again.
20090319 19:28:58< Rhonda> Soliton: Hmm, blame happygrue.
20090319 19:29:02 * happygrue hides
20090319 19:29:06< happygrue> Soliton: I asked for it
20090319 19:29:08< Rhonda> I can make wesbot leave again if that's fine ...
20090319 19:29:11< happygrue> I was not aware that it was unwanted
20090319 19:29:28< sk-> how long do you save the logs?
20090319 19:29:34< happygrue> is our secret agenda that important...?  :)
20090319 19:29:39< Rhonda> sk-: Like, forever.
20090319 19:29:45< sk-> ah
20090319 19:29:50< Rhonda> ... or until the next harddisk crash and backup timeout.
20090319 19:30:06< happygrue> which will never happen again!  so forever! ;)
20090319 19:30:17< Rhonda> *fingerscrossed*
20090319 19:30:28< Dragonking> Crab_: Ok, I see yout patch
20090319 19:30:41< Rhonda> Alright, what's the robots.txt syntax again?
20090319 19:31:19< CIA-70> soliton * r33854 /trunk/data/multiplayer/factions/ (12 files): removed formula ai recruiting again; can be added back when there is proper error handling
20090319 19:31:22< Rhonda> User-Agent: *\nDisallow: *?
20090319 19:32:01< Rhonda> Ah, Disallow: /
20090319 19:32:08< Crab_> Rhonda, copy from http://en.wikipedia.org/robots.txt
20090319 19:32:38< Rhonda> Crab_: Definitely not, we don't want indexing of the logs. :)
20090319 19:32:56-!- YogiHH [i=YogiHH@d081190.adsl.hansenet.de] has joined #wesnoth-dev
20090319 19:33:16< Crab_> Rhonda: ))
20090319 19:33:23-!- mordante [n=mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20090319 19:33:28< YogiHH> hello everyone
20090319 19:33:28< Rhonda> There it is, http://irclogs.wesnoth.org/robots.txt
20090319 19:33:38< mordante> hi
20090319 19:34:07< Velory> hi
20090319 19:34:21< mordante> hi Velory 
20090319 19:36:19-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Connection timed out]
20090319 19:37:03-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20090319 19:38:14-!- ettin_ [n=jorda@wesnoth/developer/ettin] has quit [Read error: 110 (Connection timed out)]
20090319 19:39:03-!- Akaps [n=chatzill@cpe-74-65-6-109.rochester.res.rr.com] has joined #wesnoth-dev
20090319 19:39:07< Rhonda> Soliton: Shall I stop it again for the time being until you asked around?
20090319 19:39:16< Sirp> hi YogiHH 
20090319 19:39:22< Sirp> YogiHH: I sent you an email -- did you get it?
20090319 19:39:56< Soliton> Rhonda: nah, leave it there.
20090319 19:40:02< YogiHH> Sirp: yes, did you get my answer, too?
20090319 19:40:57< happygrue> Soliton: what was the rational for not being logged?
20090319 19:41:08< Ivanovic> re
20090319 19:42:00< Soliton> happygrue: talking nasty about you when you're not there.
20090319 19:42:26 * happygrue curses
20090319 19:42:28< happygrue> I knew it!
20090319 19:42:53< Ivanovic> Soliton: come on, you should tell happygrue *everything*
20090319 19:43:03< Ivanovic> even the thing about the scatches about him...
20090319 19:43:04< Ivanovic> ^^
20090319 19:43:38 * happygrue retreats to the dark, where he is loved
20090319 19:44:03< Ivanovic> Dragonking: you pinged me
20090319 19:44:15< Dragonking> Ivanovic: Yes
20090319 19:44:24< Sirp> YogiHH: no.....I didn't.
20090319 19:44:25< Ivanovic> Dragonking: it would be great to have the ai fixes done "really soon" so that i can get out a 1.6a tarball
20090319 19:44:38< Sirp> YogiHH: when did you reply?
20090319 19:44:39< Dragonking> Ivanovic: Working on that
20090319 19:44:44-!- jp0186 [n=jp0186@77-56-56-38.dclient.hispeed.ch] has quit [Connection timed out]
20090319 19:44:45< Ivanovic> good
20090319 19:45:02< YogiHH> Sirp: like 6 hours ago
20090319 19:45:14< Soliton> Ivanovic: all formula related errors have been fixed with my last commit.
20090319 19:45:19< Ivanovic> mordante: any chance you will tackle https://gna.org/bugs/?13217 "soon"?
20090319 19:45:21< Dragonking> Ivanovic: Basically Soliton disabled recruitment formula for now, that sohuld adress MP issue
20090319 19:45:23< Sirp> YogiHH: hmmmmm I thought I checked at home and didn't see it.....could you please re-send?
20090319 19:45:38< Dragonking> Ivanovic: But still there is one bug, but I think that Crab_ fixed it.
20090319 19:45:48 * YogiHH hopes so, it was webmail
20090319 19:45:49< Ivanovic> Soliton: okay, so it is just gna acting up again with the notification
20090319 19:46:04< Dragonking> Crab_: Your patch seems to be good, just lack of whitespaces hurts a bit ;p
20090319 19:46:36< mordante> Ivanovic, will look in a few minutes
20090319 19:46:40< Ivanovic> mordante: okay
20090319 19:47:09< Ivanovic> Dragonking: and please also have a look at the message output silene pointed to this morning
20090319 19:47:13< mordante> zookeeper, regarding the randomness in the terrains, I won't fix it if a new terrain rule gets added the randomness can change 
20090319 19:47:22< Ivanovic> there is something about defense value being -1, some "error engine" message
20090319 19:47:28< mordante> so there's no way to get a terrain determenistic 
20090319 19:47:55< Crab_> Dragonking, that's good. I'll try to be more careful with whitespace )
20090319 19:48:28< Ivanovic> Soliton: your commit was only a single revision?
20090319 19:48:40< Ivanovic> (yes, for 1.6a i will handpick revisions)
20090319 19:49:01< Dragonking> Ivanovic: Hm.. will take a look
20090319 19:49:11< Ivanovic> and, even more important: will the versioning stuff handle a 1.6a release well?
20090319 19:49:25< Ivanovic> Soliton and/or Shadow_Master should know this...
20090319 19:49:42< silene> Dragonking: the ai is querying the engine about "off_map" tiles (used on 17 mainline maps), which induces a flood of error messages since the engine is not quite happy about the query; anyway, regardless of the flood, the ai should never query about the defense of a unit for a tile the unit cannot stand on
20090319 19:50:51< Soliton> Ivanovic: yes..?
20090319 19:50:58< Ivanovic> [19:49:10] <Ivanovic> and, even more important: will the versioning stuff handle a 1.6a release well?
20090319 19:51:19< Ivanovic> Soliton: if i change the version number ingame to 1.6a, will this work nicely with the filtering stuff or will it break anything?
20090319 19:51:24< Soliton> i'm allowing 1.6* on the server if that's what you mean.
20090319 19:51:43< Ivanovic> okay, will check it when the stuff is done...
20090319 19:52:44< Ivanovic> hmm, could this bug "magically" be fixed already by the change YogiHH commited some days ago? https://gna.org/bugs/?13223
20090319 19:52:50< Ivanovic> or is this one still open?
20090319 19:53:28< Dragonking> silene: Please check the latest rev
20090319 19:53:46< Dragonking> silene: Was it MP-only?
20090319 19:54:00-!- Turuk_ [n=Turuk@FUSE-WS1-208-102-202-188.fuse.net] has quit [Connection reset by peer]
20090319 19:54:12< Ivanovic> Dragonking: i had it this morning
20090319 19:54:15-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090319 19:54:18< Ivanovic> that is: when formula ai is active
20090319 19:54:21< Dragonking> Crab_: Do you have time today? As in about 1h?
20090319 19:54:31< Dragonking> Ivanovic: It should be inactive now
20090319 19:54:43< Ivanovic> yes, but the *cause* for it should be fixed, too
20090319 19:54:45< Ivanovic> ;)
20090319 19:54:57< Crab_> Dragonking: yes, I have time at the moment
20090319 19:55:21< silene> Dragonking: no idea; i noticed the bug because my favorite test scenario is an mp scenario
20090319 19:55:43-!- stikonas [n=stikonas@wesnoth/translator/stikonas] has quit [Read error: 54 (Connection reset by peer)]
20090319 19:55:50< Dragonking> Crab_: Great, want to discuss some stuff releated to your AI ideas presented in yout post at the bugtracker, will ping you in about 1h
20090319 19:56:02< Dragonking> Ivanovic: Understood. :)
20090319 19:56:04< Crab_> ok
20090319 19:59:27< zookeeper> mordante, i don't see how that relates to the randomness on a map being different in the editor and when playing
20090319 19:59:47< zookeeper> (but i'm afk so don't expect a speedy reply..)
20090319 20:00:20< mordante> zookeeper, I expect some other rules being loaded due to different defines in the editor
20090319 20:01:07< mordante> but I don't care about it since there's no guarantee it remains the same, not even in a stable release
20090319 20:01:40-!- Elvish_Pillager [n=eli@66-189-40-122.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090319 20:06:46< Dragonking> Ivanovic: http://www.wesnoth.org/wiki/SoC_People_to_bug_on_IRC    <- needs to be updated
20090319 20:07:27< Dragonking> Last "Last year he participated at GSoC as a student" in section about grzywacz, and maybe other outdated things.
20090319 20:07:48< Ivanovic> right
20090319 20:07:55< Ivanovic> Dragonking: please send a mail about this to the dev ml
20090319 20:08:13< Dragonking> Ivanovic: ok... sometime soon... :)
20090319 20:08:17< Ivanovic> :)
20090319 20:10:56< Ivanovic> that is after a short look mainly the entry from grzywacz is outdated
20090319 20:11:02< Ivanovic> grzywacz: please update it! http://www.wesnoth.org/wiki/SoC_People_to_bug_on_IRC#Karol_Nowak_.28grzywacz.29
20090319 20:13:41< YogiHH> Ivanovic: yes, bug 13223 sounds like it is fixed by me, but i am pretty sure the fix is in 1.5.14 already...
20090319 20:14:19< grzywacz> Ivanovic, k, in a moment. I didn't even know there was something like that ;)
20090319 20:14:36< Ivanovic> YogiHH: would be great if you could take the time and test it
20090319 20:14:58< YogiHH> Ivanovic: will do
20090319 20:15:42< Ivanovic> just as info for everyone: i will try to get 1.6a out tonight
20090319 20:16:23< Ivanovic> this will basically be 1.6 *only* with handpicked commits to fix "really bad issues" like revision 33849 and especially revision 33854
20090319 20:17:17< Ivanovic> YogiHH: please also have a look here: http://www.wesnoth.org/wiki/SoC_People_to_bug_on_IRC#YogiHH_or_Piotr_Cychowski_.28cycholka.29
20090319 20:17:44< Ivanovic> that is: you can remove cycholka from this block and add (a lot) more about your knowledge in the area of replays and savegames
20090319 20:18:03< loonycyborg> Perhaps I should be added there too ;)
20090319 20:18:14 * Dragonking too
20090319 20:18:22< Ivanovic> loonycyborg, Dragonking: yes, you should!
20090319 20:18:42< Dragonking> But I'm bad at writing about myself. ;p
20090319 20:18:49< happygrue> wesbot: bug 13223
20090319 20:18:51< wesbot> Bug #13223  Assigned to: None  Status: None  Priority: 5 - Normal
20090319 20:18:51< wesbot> Summary: AI does nothing when the leader is far from the keep
20090319 20:18:51< wesbot> Original submission: How-to-reproduce: 
20090319 20:18:52-!- Turuk_ [n=Turuk@FUSE-WS1-208-102-202-188.fuse.net] has joined #wesnoth-dev
20090319 20:18:54< wesbot> f the Basilisk
20090319 20:18:57< wesbot> URL: https://gna.org/bugs/?13223
20090319 20:19:10 * loonycyborg too
20090319 20:22:23-!- Netsplit verne.freenode.net <-> irc.freenode.net quits: esr, fendrin
20090319 20:22:38-!- Netsplit over, joins: fendrin, esr
20090319 20:23:14-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20090319 20:24:48< happygrue> I'm pretty sure that that would be a duplicate of bug 13222
20090319 20:25:13< happygrue> and is already fixed.  Soliton care to verify that?
20090319 20:26:01-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Connection timed out]
20090319 20:28:14 * loonycyborg is surprised to discover that Noy and Noyga are different people :P
20090319 20:28:26< noy> heh
20090319 20:30:02< grzywacz> Ivanovic, done
20090319 20:30:03< Soliton> happygrue: yeah, that should be fixed.
20090319 20:31:08< mordante> Ivanovic, it seems the assert causing 13127 is no longer needed or true so can be removed
20090319 20:31:25< mordante> but I'll test a bit more
20090319 20:31:34< Ivanovic> 13127?
20090319 20:31:51< Ivanovic> or do you mean 13217 ?
20090319 20:32:02< Ivanovic> https://gna.org/bugs/index.php?13217
20090319 20:32:03< mordante> yes and typing is an art 
20090319 20:32:08< mordante> ;-)
20090319 20:32:43< Ivanovic> mordante: then do so
20090319 20:32:50< mordante> maybe I should write a book "the art of typing without errors" ;-)
20090319 20:32:56< Ivanovic> that is: if without this assert the game won't get problems"
20090319 20:33:01< mordante> I do only my fingers trick me
20090319 20:33:20-!- ettin [n=jorda@wesnoth/developer/ettin] has joined #wesnoth-dev
20090319 20:33:24< mordante> I thought you referred to my typing ;-)
20090319 20:33:31< Ivanovic> nah
20090319 20:33:40< mordante> like I said I'm testing it
20090319 20:34:05< Ivanovic> if i would have referred to your typing i would have said "you should first practice it yourself before writing a book about how to do it correctly..."
20090319 20:34:06< Ivanovic> ;)
20090319 20:34:27 * loonycyborg decides to edit http://www.wesnoth.org/wiki/SoC_People_to_bug_on_IRC and hopes to not to interfere with someone else doing the same.
20090319 20:34:46< mordante> nah college learned my you don't need to master a subject in order to teach it ;-)
20090319 20:35:07-!- jp0186 [n=jp0186@77-56-56-38.dclient.hispeed.ch] has joined #wesnoth-dev
20090319 20:35:15< Ivanovic> mordante: but knowing nothing about the subject can easily lead to problems
20090319 20:35:16< Ivanovic> ;)
20090319 20:36:37< mordante> luckily for teachers, the students most of the time know less about the subject so can't detect your errors ;-)
20090319 20:36:44< mordante> their*
20090319 20:37:37< Ivanovic> :)
20090319 20:42:25< mordante> I can fix the crash easily, but I it indeed avoided another bug :-)
20090319 20:42:31-!- fabi [n=fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20090319 20:42:57< mordante> that bug is that when you click on the selected text the cursor is misplaced
20090319 20:43:09< mordante> so a minor bug instead of an assert
20090319 20:43:17< happygrue> hey, does anyone know if Wesnoth comes packaged with the Dell computers that ship Ubuntu?
20090319 20:43:26< mordante> still want to look into the bug
20090319 20:43:32< Velory> my game is stucking when im on multiplayer 2p- Caves Of the Basilisk (default ai)
20090319 20:43:33< Velory> when i press end turn 2 times without doing nothing game stucking
20090319 20:43:43< mordante> Dragonking, ^
20090319 20:44:04< Dragonking> Velory: Fixed.
20090319 20:44:09< Velory> hmm
20090319 20:44:26< Velory> should i download from trunk again ?
20090319 20:44:29< Dragonking> Yes.
20090319 20:44:31< Velory> ok
20090319 20:45:06< mordante> Ivanovic, I'll either remove the assert of do a proper fix before you want to tag 1.6a
20090319 20:45:06< loonycyborg> Ivanovic: Edited: http://www.wesnoth.org/wiki/SoC_People_to_bug_on_IRC#loonycyborg
20090319 20:47:36-!- elias [n=elias@allegro/developer/allefant] has quit ["Leaving"]
20090319 20:53:29< Baufo> umm, is this supposed to be this way? http://imagebin.org/41986
20090319 20:53:48< Baufo> the message only seems to be as wide as the map and not the screen
20090319 20:54:34-!- yann [n=dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has joined #wesnoth-dev
20090319 20:55:52< mordante> seems a recent bug :-( 
20090319 20:57:31< CIA-70> dragonking * r33855 /trunk/src/formula_ai.cpp: Patch by Crab - fixed bug #13218
20090319 20:58:59-!- fendrin [n=fabi@wesnoth/developer/fendrin] has quit [Read error: 110 (Connection timed out)]
20090319 20:59:21< mordante> Baufo, which scenario number is it?
20090319 20:59:57< Baufo> mordante, the second one of descent into darkness
20090319 21:00:18< mordante> ok will look later, pondering about another bug at the moment
20090319 21:07:06< Soliton> Velory: you have to restart though reload won't work, i think.
20090319 21:13:59< Velory> what ?
20090319 21:15:25< Ivanovic> Velory: what Soliton meant is that you can't continue a saved game with this broken ai
20090319 21:15:34< Ivanovic> it will still "halt" in the middle
20090319 21:15:48< Ivanovic> but newly started mp games *should* be working nicely
20090319 21:16:12< Velory> but im not trying to continue a saved game :S
20090319 21:16:44< Soliton> good.
20090319 21:16:47< Velory> and its no problem in the game now  for me
20090319 21:16:53< Ivanovic> even better
20090319 21:16:55< Ivanovic> :)
20090319 21:16:57< Velory> i downloaded from svn again
20090319 21:16:59< Velory> and remake
20090319 21:17:08< Velory> so it working well now
20090319 21:17:18< Velory> if u mean the stuck problem
20090319 21:17:19< Ivanovic> mordante: what is the status of your assert change/fix ?
20090319 21:17:26< Ivanovic> yes, this one we mean, Velory
20090319 21:17:36< Ivanovic> i would like to get 1.6a out *now*
20090319 21:17:38< Velory> ok so no problem now thank you
20090319 21:17:50< mordante> your half a minute too early with that question ;-)
20090319 21:18:06< mordante> but I also want to look at the issue Baufo just posted
20090319 21:19:33-!- Andylee [n=kvirc@188-23-11-27.adsl.highway.telekom.at] has joined #wesnoth-dev
20090319 21:19:34< Andylee> hi
20090319 21:19:40-!- Blueblaze_ [n=Blueblaz@c-98-199-143-139.hsd1.tx.comcast.net] has joined #wesnoth-dev
20090319 21:20:25< Ivanovic> hi Andylee
20090319 21:21:07< Andylee> wtf? 1.6 already released? where?
20090319 21:21:12< Andylee> how?
20090319 21:21:14< Andylee> who?
20090319 21:21:16-!- Baufo [n=thomas@wesnoth/developer/baufo] has quit [Remote closed the connection]
20090319 21:21:24< Andylee> how many victims?
20090319 21:21:54< Ivanovic> even more, will get a hotfix release tagged in some minuites...
20090319 21:22:12< mordante> Andylee, released but not announced
20090319 21:23:15< Andylee> ok, will wait for sunday
20090319 21:23:20< Velory> :)
20090319 21:23:55< Andylee> just wanted to congratulate and thank you for your good work and encourage you to further expand this best free game in the net.
20090319 21:24:04< mordante> thanks :-)
20090319 21:24:19-!- esr_ [n=chatzill@static-71-162-243-5.phlapa.fios.verizon.net] has quit [Remote closed the connection]
20090319 21:24:35< ettin> Ivanovic: may I commit a translation update to svn trunk, or I need to wait until 1.6a is tagged?
20090319 21:25:35< Ivanovic> ettin: i am working with the 1.6 tag copied to 1.6a
20090319 21:25:48< Ivanovic> and i will hand apply single revisions there
20090319 21:26:44< ettin> Ivanovic: nevermind then, I will just continue updating the translation and wait for 1.6.1
20090319 21:26:45-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090319 21:26:49< ettin> it is just a minor update anyway
20090319 21:27:17< Ivanovic> sure
20090319 21:27:58< mordante> arrgh ellipses show ugly on some keeps :-( DiD scenario 1
20090319 21:28:42< Ivanovic> clear case of "fix for post 1.6a"
20090319 21:28:47< Ivanovic> mordante: your fix is online?
20090319 21:28:57< Ivanovic> (since gna is damn slow again with notifications atm)
20090319 21:29:08-!- Blueblaze_ [n=Blueblaz@c-98-199-143-139.hsd1.tx.comcast.net] has quit [Read error: 145 (Connection timed out)]
20090319 21:29:39< mordante> Ivanovic, r33857
20090319 21:31:49< Ivanovic> and yeah, i will remove your changelog entry (since i will not raise the ingame version number, lame as i am)
20090319 21:32:49< mordante> Ivanovic, and I think I've a fix for baufo'
20090319 21:32:54< mordante> s issue as well
20090319 21:33:05< Ivanovic> then commit it!
20090319 21:33:40< mordante> testing it now
20090319 21:33:58< mordante> compilation takes a while here :-(
20090319 21:34:56< Ivanovic> and you are sure it does not introduce a new assert?
20090319 21:34:57< Ivanovic> ;)
20090319 21:35:50< mordante> no I didn't add a new one, I've enough left to be triggered ;-)
20090319 21:41:27-!- Akaps [n=chatzill@cpe-74-65-6-109.rochester.res.rr.com] has left #wesnoth-dev []
20090319 21:49:49< mordante> Ivanovic, r33858
20090319 21:51:23-!- Elvish_Pillage2 [n=eli@66-189-40-122.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090319 21:51:58-!- esr [n=chatzill@static-71-162-243-5.phlapa.fios.verizon.net] has quit [Read error: 104 (Connection reset by peer)]
20090319 21:53:51< Ivanovic> okay, merged the stuff for 1.6a into the extra tag, revisions used: 33849, 33854, 33855, 33857 and 33858
20090319 21:55:29< Ivanovic> and now there is also a nice and little patch that can just be applied... http://rafb.net/p/DG3gLZ97.html
20090319 21:57:32-!- Rrenys [n=rrenys@81-20-159-197.levira.ee] has joined #wesnoth-dev
20090319 21:57:49< CIA-70> ivanovic * r33856 /tags/1.6a/: copy over the 1.6 tag content to get started on the hotfix release 1.6a...
20090319 21:57:57< CIA-70> mordante * r33857 /trunk/ (changelog src/text.cpp): 
20090319 21:57:57< CIA-70> Remove a no longer valid assert.
20090319 21:57:57< CIA-70> Due to the ellipses on the right side the assert is no longer valid and
20090319 21:57:57< CIA-70> thus removed. Fixes bug #13217. Also discoved a small cursor glitch with
20090319 21:57:57< CIA-70> the ellipses, but don't know the sanest solution so left a reminder.
20090319 22:00:03< Dragonking> Ivanovic: Can I commit my patch for error msg flood, or you are doing some weird stuff with svn? :)
20090319 22:00:27< Ivanovic> Dragonking: this one won't go in
20090319 22:00:36< Ivanovic> that is: i have everything in that will be
20090319 22:00:53< sk-> where is 1.6a? i want to test that thing.
20090319 22:01:05< Ivanovic> since formulaai is not active by default atm, this is not too relevant
20090319 22:01:19< Dragonking> Ivanovic: I know
20090319 22:01:22< Ivanovic> better make sure that stuff is *well tested* soon, so that we can have things working nicely in 1.6.1
20090319 22:01:32< Ivanovic> sk-: instead of trunk/ us tags/1.6
20090319 22:01:36< Ivanovic> s/us/use
20090319 22:01:39< Dragonking> Ivanovic: Just asking when I can commit it
20090319 22:01:48< Ivanovic> Dragonking: whenever you feel like
20090319 22:02:06< Dragonking> Ivanovic: Soooo... like now too?
20090319 22:02:19< Ivanovic> jupp
20090319 22:02:22< Dragonking> (that was a purpose of my first question :) )
20090319 22:02:23< Dragonking> ok
20090319 22:02:25< Ivanovic> like i said, i won't be using turnk for this release
20090319 22:03:32< Dragonking> ok
20090319 22:04:29< loonycyborg> Ivanovic: So there already exists 1.6 branch or you're making 1.6a by cherry-picking commits into a tag?
20090319 22:04:50< Ivanovic> loonycyborg: i cherry picked commits into an extra tag
20090319 22:05:07< Ivanovic> because using a plain copy of trunk is not possible with the new strings
20090319 22:06:20< mordante> Velory, please add your page here http://www.wesnoth.org/wiki/SummerOfCodeIdeas#GSoC_Student_pages
20090319 22:06:53< mordante> Velory, looking at your page now and unless you love spam I suggest to obfuscate your email address a bit
20090319 22:07:18< YogiHH> Dragonking: ping
20090319 22:07:34< loonycyborg> Aren't spambots smart enough to see through obfuscation? :P
20090319 22:07:43-!- Elvish_Pillager [n=eli@66-189-40-122.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090319 22:08:18< Velory> mordante:i dont understand what u mean? obfuscate ?
20090319 22:08:22< mordante> loonycyborg, depends on how lazy the programmer was ;-)
20090319 22:08:42< mordante> my_addressATfooDOTcom or something like that
20090319 22:08:48< Andylee> hey!
20090319 22:09:00< loonycyborg> Truely lazy programmer probably will reuse someone else's smart spambot :)
20090319 22:09:02< Velory> oh okey
20090319 22:09:11< Andylee> you said, 1.6 is not released by now.... but look at this: http://www.chip.de/downloads/The-Battle-for-Wesnoth_17625051.html
20090319 22:09:22< Andylee> it is already out at the software-portals!
20090319 22:09:34< YogiHH> Andylee: it's not officially released by now ;-)
20090319 22:09:41< Velory> mordante: u want me to write my link below == GSoC Student pages == right?
20090319 22:09:45< mordante> Andylee, as I said it's released but not announced ;-)
20090319 22:09:51< mordante> Velory, yes
20090319 22:09:54< Velory> okey
20090319 22:10:01< Dragonking> YogiHH: Yes?
20090319 22:10:14< loonycyborg> Velory: Probably they're monitoring SF.
20090319 22:10:21< mordante> we want to test some more and give packagers time to upload the binary to SF and let SF mirror it a bit
20090319 22:10:29< YogiHH> Dragonking: why am i using formula AI stuff if i chose default AI?
20090319 22:10:31< loonycyborg> s/ Velory:/Andylee: 
20090319 22:10:49 * mordante goes to read the review ;-)
20090319 22:10:56< Dragonking> YogiHH: You do not in latest trunk and in 1.6a :)
20090319 22:11:10< Andylee> ok... then I misunderstood you
20090319 22:11:12< YogiHH> Dragonking: ah, ok :-)
20090319 22:11:15< Dragonking> YogiHH: It was attempt to make AI recruit better, but still a bit too buggy
20090319 22:11:18< Andylee> my english reading arts... ;)
20090319 22:11:23-!- zookeeper [n=l@wesnoth/developer/zookeeper] has quit []
20090319 22:11:33< YogiHH> Dragonking: yes, and with an endless loop ;-)
20090319 22:12:19< sk-> I lost the link to get to trunk/ which won't help me to find tags/1.6 to get 1.6a. :S .. what is the link again?
20090319 22:12:21< Dragonking> YogiHH: Details... lo
20090319 22:12:24< alink> pfff, they don't even mention Linux and Mac in OS supported :-(
20090319 22:12:27< Dragonking> s/lo/;p/
20090319 22:12:43< mordante> <sarcasm>great all packagers are ready including BeOs</sarcasm>
20090319 22:13:08< Ivanovic> sk-: have a look at svn.wesnoth.org
20090319 22:13:25< Dragonking> Ivanovic: Should I write this mail regarding SoC people to talk to site?
20090319 22:13:34< Ivanovic> Dragonking: yes
20090319 22:13:36< Dragonking> ok
20090319 22:13:42< Dragonking> Will do it now
20090319 22:14:25< Ivanovic> good
20090319 22:17:28< sk-> oh.. svn... I'll wait until there's an exe or installer.
20090319 22:18:24-!- Turuk_ [n=Turuk@FUSE-WS1-208-102-202-188.fuse.net] has quit [Read error: 104 (Connection reset by peer)]
20090319 22:18:39-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090319 22:19:17< mordante> Velory, did you also read this page http://www.wesnoth.org/wiki/WhyWritingAWesnothAIIsHard
20090319 22:19:28< Velory> yes
20090319 22:19:32< mordante> ok good
20090319 22:19:43< Velory> i read http://www.wesnoth.org/wiki/WritingYourOwnAI
20090319 22:19:45< Velory> too
20090319 22:20:01< Velory> and most pages about AI
20090319 22:20:14< mordante> well the problem is most people underestimate how hard writing an AI is
20090319 22:21:09< Velory> hmm i cant :)
20090319 22:21:10< loonycyborg> I personally don't underestimate it..
20090319 22:21:34< Dragonking> Ivanovic: Done
20090319 22:21:43< sk-> interesting article
20090319 22:21:53< loonycyborg> Though for some reason that area leaves me uninterested. At least for now..
20090319 22:22:27< Ivanovic> Dragonking: thanks
20090319 22:22:47< Dragonking> No problem
20090319 22:24:05< sk-> I always thought that the AI problem of so many maps and units would be helped by allowing map makers the power to control very detailed aspects of the AI's behavior for their maps.
20090319 22:25:20-!- Andylee [n=kvirc@188-23-11-27.adsl.highway.telekom.at] has quit ["KVIrc 3.4.2 Shiny http://www.kvirc.net/"]
20090319 22:25:33< sk-> anyway, I await the 1.6 ...... later guys
20090319 22:25:40< Velory> mordante:btw i forgot to say which tools im using-so im using eclipse in linux and qt4 but for c++ codes im using vim text editor...On windows im using both visual studio and eclipse
20090319 22:25:41< Velory> for database development:im using mysql
20090319 22:25:44-!- sk- [n=sk@pool-68-162-68-65.phil.east.verizon.net] has quit ["Leaving"]
20090319 22:26:29< Crab_> Dragonking: You've said that you wanted to discuss the rest of the bugfix with me. Will you do so today ? (I'm going to be here today for about an hour)
20090319 22:26:37< CIA-70> mordante * r33858 /trunk/ (changelog src/gui/widgets/window.cpp): 
20090319 22:26:37< CIA-70> The dialog now always uses the full map area width.
20090319 22:26:37< CIA-70> Baufo discovered if the map is smaller as the available map area the
20090319 22:26:37< CIA-70> dialogs will use the map width instead of the full area available for
20090319 22:26:37< CIA-70> the map. Now always use the available width.
20090319 22:26:48< Dragonking> Crab_: Sorry, was a bit busy till now
20090319 22:26:55< Dragonking> Crab_: Gimme about 5 min
20090319 22:26:58< CIA-70> ivanovic * r33859 /tags/1.6a/ (16 files in 3 dirs): 
20090319 22:26:58< CIA-70> merge revisions 33849, 33854, 33855, 33857 and 33858 into 1.6a to fix a crashes
20090319 22:26:58< CIA-70> (due to an assertion failed) and problems with the AI (fixes to the formula
20090319 22:26:58< CIA-70> stuff as well as deactivating the usage of the formula recruitment by default)
20090319 22:26:59< Crab_> Dragonking: all ok )
20090319 22:27:30< mordante> Velory, do you also have experience with web development?
20090319 22:27:42< Velory> yes i do
20090319 22:28:03< Ivanovic> created the tarball, uploading to files.wesnoth.org
20090319 22:28:31-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Client Quit]
20090319 22:28:32< Velory> i dont have my domain but i know javascript,PHP and html and i know xml etc.
20090319 22:28:42< Velory> my own domain i mean*
20090319 22:29:02-!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090319 22:29:18< mordante> what do you think of this idea http://www.wesnoth.org/wiki/SummerOfCodeIdeas#Implement_campaign_statistics_reports_on_stats.wesnoth.org
20090319 22:29:20< nital> Is it a safe assumption that no files within wesnoth directory contain spaces in their filenames?
20090319 22:29:39< Velory> and im familiar with 3d design i used cinema 4d for desining some object,i also know actionscript for animations
20090319 22:29:49< mordante> assumptions are the mother of all *****ups ;-)
20090319 22:29:52< CIA-70> dragonking * r33860 /trunk/src/formula_ai.cpp: Fixed defense_on function causing errors while trying to get defense value on unpassable terrain
20090319 22:30:17< Velory> hmm i think i can be in 3 different ideas AI idea,savegame organization idea and statistics reports idea but i choose the best i wanted :)
20090319 22:30:43< CIA-70> ivanovic * r33861 /tags/1.6a/configure.ac: set tarball name to 1.6a, the ingame version number will be identical for 1.6a
20090319 22:30:44< Crab_> nital, "/wesnoth%find ./ -name '* *' -print " gives no results
20090319 22:31:02< nital> thanks Crab_
20090319 22:32:15< mordante> nital, we only have no control of your preferences directory, regarding addons and maps
20090319 22:32:42< alink> or savegames
20090319 22:34:23< Crab_> nital: in prefs directory, cache entries contain spaces (for example - /cache/cache-v1.6 (33853M)-3c73140f332c8bc2b70048cc09e68a1f2ff812ed.gz )
20090319 22:34:40< nital> well, since I'm creating a patch utility which will work on fresh install of Wesnoth
20090319 22:34:51< nital> there will be no cache or userdata, so it doesn't matter
20090319 22:35:03< mordante> Velory, of course it just that I expect a lot of AI ideas so would be nice to know whether you're able and willing to do another project as well
20090319 22:35:39< nital> I will probably add spaces support later, but now I have a working example asap ;)
20090319 22:35:47< nital> s/have/want to have/
20090319 22:36:35< Velory> ok i can also edit my wiki page for other ideas if u want i just dont want to seem like "i want all ideas :D "
20090319 22:37:29< mordante> I understand, but since you said you could do mysql I was wondering
20090319 22:37:41< mordante> no need to change your proposal 
20090319 22:37:57< Velory> Ok
20090319 22:38:18< mordante> I'll keep it in mind
20090319 22:38:48-!- iyo [n=iyo@a02-0829c.kn.vutbr.cz] has joined #wesnoth-dev
20090319 22:38:51< Velory> good for me :)
20090319 22:38:57-!- iyo is now known as iyonius
20090319 22:39:07-!- iyonius is now known as iyo
20090319 22:39:12< Rhonda> Oh? 1.6a?
20090319 22:39:23< Rhonda> ..brown paper 1.6 version?
20090319 22:39:49< Velory> will we edit and develop AI or will we create all new framework for AI with new approaches ?
20090319 22:39:50< mordante> Rhonda, brown paper bags
20090319 22:39:53-!- iyo [n=iyo@a02-0829c.kn.vutbr.cz] has quit ["Leaving."]
20090319 22:39:56< Rhonda> I take it that that one does include the AI debug disabling thing that was mentioned.
20090319 22:40:22< mordante> btw I read the debian-games posts and hope to have some time to reply tonight
20090319 22:40:54< Sirp> mordante: debian-games posts?
20090319 22:41:05-!- iyonius [n=iyo@a02-0829c.kn.vutbr.cz] has joined #wesnoth-dev
20090319 22:41:09< Rhonda> Sirp: http://lists.alioth.debian.org/pipermail/pkg-games-devel/2009-March/010807.html thread
20090319 22:41:23< mordante> Velory, not sure yet, but a lot will depend on the strength of the candidate and what we do with the AI 
20090319 22:41:30< mordante> Sirp's idea is quite new
20090319 22:41:37 * Rhonda . o O ( and yes, I know in the meantime that my mentioned problem was between .12 and .13 and not .13 and .14 )
20090319 22:42:46< Velory> Ok
20090319 22:42:47< Dragonking> Crab_: ok I'm here
20090319 22:42:52< Crab_> aha
20090319 22:43:04< loonycyborg> Ivanovic: So 1.6a tag should be used to produce another installer?
20090319 22:43:58< Dragonking> Crab_: Ok, regarding this recruting formula and your proposals
20090319 22:44:11-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 104 (Connection reset by peer)]
20090319 22:44:13< Crab_> and what do you think ?
20090319 22:44:25< Dragonking> Crab_: a) and b) sound perfectly ok, I wonder about c) Keep available within reach of a leader but is occupied by a enemy: attack the keep with the leader
20090319 22:44:26-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090319 22:44:49< Dragonking> What is point of attacking with a leader?
20090319 22:44:54< Crab_> what do you propose ? it seems *reasonably enough* and quites simple
20090319 22:45:13-!- crimson_penguin [n=ben@S0106001cf0705af9.cc.shawcable.net] has joined #wesnoth-dev
20090319 22:45:56< Dragonking> Crab_: c is not really reasonable imho :)
20090319 22:46:46< Crab_> another solution is to ask the AI gang up other troops (if the ai can think it can kill the occupying creature with a good % of success)
20090319 22:47:13< Dragonking> Yes, that's a bit better. 
20090319 22:47:16< Dragonking> Imho.
20090319 22:48:42< Dragonking> Crab_: Did you take a look (generally) how current recruitment works?
20090319 22:49:34< Crab_> a bit of. it tries to find 'most effective troops' by using , for example, map  terrain information
20090319 22:49:54< Dragonking> Good, so generally it does sometihng like that
20090319 22:50:10< Dragonking> make sure unit is in keep,  evluate, recruit, fallback to default AI
20090319 22:50:19< Dragonking> It's aim is to make recruitment better
20090319 22:50:55< Velory> Dragonking: can i ask u question about recruitment?
20090319 22:51:18< Dragonking> Velory: Just ask, will try to answer that too
20090319 22:52:39< Dragonking> Crab_: Assuming this formula works like intended... can you see some drawback of it?
20090319 22:52:41-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit ["Leaving"]
20090319 22:53:22< Crab_> Dragonking: not yet
20090319 22:54:05< Ivanovic> loonycyborg: yes
20090319 22:54:12< Velory> Dragonking:Okey so can AI look how many enemy unit and how many allied unit that have ?
20090319 22:54:25< Ivanovic> loonycyborg: that is it is ready now
20090319 22:54:33< Ivanovic> but you have to copy over your windows installer changes
20090319 22:54:53< Ivanovic> i have not seen them fitting as "really important" for copying things over
20090319 22:55:09< Crab_> Dragonking: yes, i think it can. the list of units on the map is available and can be looked through
20090319 22:55:49< Velory> u said me or Dragonking ? :)
20090319 22:56:07< Crab_> Dragonking: some functions there are doing just that ( evaluate_attacker_against_opponents , for example)
20090319 22:56:14< loonycyborg> Ivanovic: You mean 1.6 version bump? It'll probably just stay there once I'll do svn switch.
20090319 22:56:30< Ivanovic> loonycyborg: exactöy
20090319 22:56:31< loonycyborg> Since I use the same working copy for all releases :)
20090319 22:56:35< Ivanovic> s/ö/l
20090319 22:56:38< Dragonking> Velory: AI has info about *everything*
20090319 22:57:08-!- loonycyborg [n=sergey@wesnoth/developer/loonycyborg] has quit ["KVIrc 3.4.2 Shiny http://www.kvirc.net/"]
20090319 22:57:18< Dragonking> Crab_: Was that answer for my or Velory's question, cause I'm a bit confused. :)
20090319 22:57:26< Ivanovic> the stuff is in tags/1.6a, cherrypicked the revisions that are important
20090319 22:57:26< Crab_> s/Dragonking/Velory
20090319 22:57:33< Dragonking> Thought so :)
20090319 22:57:44-!- loonybot [n=loonybot@wesnoth/bot/loonybot] has quit [Remote closed the connection]
20090319 22:58:05< Dragonking> Ok, so for now I leave question of what is drawback of current recruitment implementation open :)
20090319 22:58:17< Crab_> Dragonking: ok
20090319 22:58:53< Dragonking> Crab_: Your idea is quite good, what I would do in c) scenario, is just fallback to default AI
20090319 22:59:05< Crab_> ok
20090319 22:59:52-!- boucman [n=rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20090319 23:00:12-!- loonycyborgw32 [i=kvirc@79.139.247.203] has joined #wesnoth-dev
20090319 23:00:33< Crab_> I thought something like: (c):
20090319 23:00:33< Crab_> 1) check if units nearby (without leader) can (with good % of success) kill that enemy unit in keep.
20090319 23:00:33< Crab_> 2) if so, do just that
20090319 23:00:33< Crab_> 3) else, forget about this keep
20090319 23:00:51< Dragonking> Crab_: But writing function (in formula language) that looks for all keeps in unit's range may be a bit painul now
20090319 23:01:16-!- stikonas [n=stikonas@wesnoth/translator/stikonas] has joined #wesnoth-dev
20090319 23:02:42< Crab_> Dragonking:  why ? the number of keeps on the map is quite limited. we can write a c++ function that iterates through them and sees what keeps are reachable. and use this function as a  formula language element
20090319 23:03:01-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Client Quit]
20090319 23:03:05< Dragonking> Crab_: *(in formula language)*
20090319 23:03:19< Dragonking> That is in case we want to make recruitment 1.6 compatible
20090319 23:03:36-!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090319 23:04:01< Dragonking> I don't want to have new C++ functions added in 1.6 branch to keep all formula usages compatible with *every* 1.6.X
20090319 23:04:28< Crab_> Dragonking: aha ) I was thinking about extending it. if that is not possible at the moment due to 1.6 compatability, then I'll guest it will be better to fallback then...
20090319 23:04:54< Dragonking> Crab_: But your prposal is really good one for 1.7+
20090319 23:05:51< Crab_> Dragonking: I've got some more proposals about AI codebase at my SoC wikipage (  http://www.wesnoth.org/wiki/SummerOfCodeProposal_AI_Improvement_Crab )
20090319 23:06:06 * Dragonking goes to read
20090319 23:06:07-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Client Quit]
20090319 23:06:37-!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090319 23:07:20< Velory> Dragonking:OK so i think comp. Player can look how many unit that comp. player have and look how many unit that other player have and if comp. player stronger than (have more unit) enemy i think there is no need for recruitment so AI can pass recruitment phase after that AI can recruit gooder units,Or can recruit a lot of units at same time.(I dont know if AI already doing that just i think AI dont have to enter recruitment phase every turn and thats 
20090319 23:08:39-!- Blueblaze_ [n=Blueblaz@c-98-199-143-139.hsd1.tx.comcast.net] has joined #wesnoth-dev
20090319 23:09:12< Dragonking> Crab_: "June 11- July 10: Improve the AI and make sure it is better that current AI."  Seems you're really optimistic. :)
20090319 23:09:35-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Client Quit]
20090319 23:09:46< mordante> Crab_, can you also add your page to this section http://www.wesnoth.org/wiki/SummerOfCodeIdeas#GSoC_Student_pages 
20090319 23:09:49< Crab_> Dragonking: yes, I am. But at least I think that it will have less subtle bugs which can spoil te fun
20090319 23:10:02< mordante> Crab_, and I also don't know how fond you are of spam 
20090319 23:10:13-!- Blueblaze_ [n=Blueblaz@c-98-199-143-139.hsd1.tx.comcast.net] has quit [Client Quit]
20090319 23:10:29< Crab_> mordante: quite fond of ) I ran a catch-all on @terraninfo.net and diligently read all the spam that comes to me :)
20090319 23:10:57-!- Blueblaze_ [n=Blueblaz@c-98-199-143-139.hsd1.tx.comcast.net] has joined #wesnoth-dev
20090319 23:11:27< mordante> and how much viagra and credit cards did you get offers for ;-)
20090319 23:12:09< Crab_> mordante: not much ) Icedove's Spam filter and spamassasin spoil all the fun )
20090319 23:12:25< Dragonking> Velory: I don't think skipping recruitment just because we are stronger makes much sense
20090319 23:12:39< Dragonking> Each unit on battlefield counts, especially in game like wesnoth.
20090319 23:13:17< Crab_> Dragonking: done with wikilink
20090319 23:13:20< mordante> hmm they should prohibit these nasty programs, you should be about to get those important messages
20090319 23:15:45-!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090319 23:16:10< Crab_> mordante, all right, I've obfuscated it a bit :)
20090319 23:16:44< mordante> ok
20090319 23:17:00< Crab_> used cyrillic letters AT for additional security )
20090319 23:17:22< Dragonking> Crab_: ok.. you write 2) the ability for the AI to intelligently form/recruit tactical squads to achieve those goals. A squad is a small group of units which will work towards a single goal. Squad composition may be overridden by scenario creators using FormulaAI.
20090319 23:17:28< Dragonking> Can you temm something more about it?
20090319 23:17:41< Dragonking> What kind of "goals" you have in mind?
20090319 23:17:53< Dragonking> How yould you split units into groups?
20090319 23:18:05< Ivanovic> okay, the new 1.6a tarball is now at sf.net, too
20090319 23:18:12< Ivanovic> the old 1.6 tarball is gone
20090319 23:18:21< boucman> Ivanovic: still announcing on staruday ?
20090319 23:18:29< Ivanovic> boucman: still announcing sunday
20090319 23:18:33< boucman> ok
20090319 23:18:45< Dragonking> Crab_:  Would there be no single units, but only managable groups?
20090319 23:20:02< Crab_> Dragonking: "What kind of "goals" you have in mind?" - one of the most important goals, IMO, is to kill a single enemy or to cut it's escape or to protect weak (wounded) units while they heal. it is usually not possible to do with single unit - multiple units must cooperate.
20090319 23:21:08< Crab_> Dragonking: and I think It will be fun to watch ai attempt such tactics
20090319 23:21:44< Dragonking> Can group objective change during the single turn?
20090319 23:21:49< Dragonking> Or only between turns?
20090319 23:21:56< Crab_> Dragonking: "How would you split units into groups?" - on need-to-do-something basic (or on the whim of scenario creators, of course)
20090319 23:22:06-!- ABCD_ [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090319 23:22:13< Crab_> Dragonking: yes, i personally think that objectives can change dynamically
20090319 23:22:33< Dragonking> Crab_: Well, aim of the AI is to have good default AI too, so we can't let only scenario designers decide.
20090319 23:22:34< Crab_> for example, if a wizard hits all 4 of his 70% attacks and kills the enemy faster than expected )
20090319 23:22:53< Dragonking> Or if wizard missess all 4
20090319 23:22:54< Dragonking> :)
20090319 23:22:56< Crab_> aha
20090319 23:23:01-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Connection reset by peer]
20090319 23:23:21< Dragonking> Crab_: So good explanation how to form grups would be good too
20090319 23:23:46< Dragonking> And does the groups can change members in a dynamic way too?
20090319 23:23:48< Crab_> i'm an optimist ) and, it's easier to hit 4 70% [~25%] than to miss them all [0.81%] :)
20090319 23:24:17< Dragonking> Statistics have nothing to do with wesnoth, believe me. ;)
20090319 23:24:30-!- alink [n=alink@wesnoth/developer/alink] has quit [Remote closed the connection]
20090319 23:24:46< Crab_> Dragonking: yes, i think that groups shall built purely for tactical purposes and change on the whim of the AI
20090319 23:24:55< Crab_> s/built/be built
20090319 23:25:12< Dragonking> What this "whim" could be?
20090319 23:26:13< mordante> no regarding a lot of users mages just keep missing all 4 hits at 70% chance to hit ;-)
20090319 23:26:43< Crab_> mordante: yes, players notice those a lot more :) it's a mystery of human perception )
20090319 23:26:44-!- Espreon [n=espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20090319 23:26:56< Crab_> Dragonking: one of the variants is to combine a statical position evaluator with some optimization routine (for example, using genetic algorithms)
20090319 23:27:20< Dragonking> ok
20090319 23:27:24< mordante> I know, it's just odd to hear how unfair the game is time after time
20090319 23:27:33-!- loonycyborgw32 [i=kvirc@79.139.247.203] has quit ["KVIrc 3.4.0 Virgo http://www.kvirc.net/"]
20090319 23:27:38< mordante> but nobody able to come up with hard facts
20090319 23:27:42< Dragonking> Crab_: What about my questions regarding single units?
20090319 23:27:53< Dragonking> Would there be groups only?
20090319 23:27:54-!- stikonas [n=stikonas@wesnoth/translator/stikonas] has quit [Read error: 110 (Connection timed out)]
20090319 23:28:15< Crab_> Dragonking: a single-unit group is a legimate group in my opinion, and it will be easier to make the AI think in 'groups'
20090319 23:28:20< mordante> but another fact is that it's late here, so I'm off to bed; night
20090319 23:28:30-!- mordante [n=mordante@wesnoth/developer/mordante] has quit ["Leaving"]
20090319 23:28:34< Crab_> so, 'groups only', to make it simpler
20090319 23:28:40< boucman> Crab_: yeah, particularly if that single is very special (healer)
20090319 23:29:01< Crab_> boucman: no more special than the group of 3 healers :)
20090319 23:30:21< Dragonking> Crab_: Whay do you tihnk so?
20090319 23:30:22< Crab_> boucman: but of course you are right ) some groups will have special capabilities (such as - ability to heal, ability to lead, ability to level-up after next combat, ability to recruit), and the AI has to take these into account, at least a bit )
20090319 23:31:58< Crab_> Dragonking: because I just want to simplify things a bit. It is fair to make the ai favor making 1-unit-groups for certain unit types, of course.
20090319 23:32:59< Dragonking> How 3-healer group would be simpler than 1-unit one?
20090319 23:33:11< Crab_> Dragonking: "Crab_: Well, aim of the AI is to have good default AI too, so we can't let only scenario designers decide." - regarding this, I personally think that it will be better to make a core AI in C++ and make formulaAI a convenient way to tune coreAI parameters for scenario creators
20090319 23:33:18< Dragonking> I'm just asking about your point of view. :)
20090319 23:33:54< Crab_> Dragonking: having to deal with 'groups' is easier that having two separate code branches for 'groups' and 'single units'. less duplication.
20090319 23:34:20< Dragonking> Crab_: Ok, just I wonder about mentioned usage
20090319 23:34:26< Dragonking> 3 healers vs 1 healer group
20090319 23:34:47< Dragonking> How would it "control" 3 healers, how 1 healer, and how it could differ?
20090319 23:35:22< Crab_> so i am not saying that '3-healer group would be simpler than 1-unit one', i am saying that the difficulty to control a group of three healers (which stick fairly together on the map) is not conceptually different from controlling one healer'.
20090319 23:36:12< Dragonking> Ok, but usually in a game we have different types of units moxed
20090319 23:36:14< Dragonking> mixed'
20090319 23:36:29< Dragonking> So like one healer supports mage and fighters/archers
20090319 23:36:30< Crab_> Dragonking: yes, i know. 
20090319 23:36:36< Dragonking> Especially in MP-type of game
20090319 23:36:40-!- loonybot [n=loonybot@79.139.247.203] has joined #wesnoth-dev
20090319 23:37:10< Crab_> Dragonking: it will be nice for the ai to utilize them as a group - so the healer actually 'supports' and other troops cover it
20090319 23:37:13-!- loonycyborg [n=sergey@79.139.247.203] has joined #wesnoth-dev
20090319 23:37:30< Dragonking> How would you decide up on unit's usage?
20090319 23:37:50< Dragonking> For example there are units that may be use as supporters, but also as these that attack as first
20090319 23:37:51< Crab_> AI wound need to calculate the advantages and disadvantages a healer will bring to him (again, in terms of position evaluation)
20090319 23:38:00< Crab_> s/wound/would
20090319 23:38:50< Crab_> Dragonking: yes, so they give multiple benefits
20090319 23:39:47< Dragonking> How would you evaluate position?
20090319 23:40:23< Crab_> Dragonking: my first goal is to make the ai system easy to debug\manage\extend. afterwards it will be easier to experiment without risking breaking too much 
20090319 23:40:53< Crab_> Dragonking: about position evaluation - i think that there is already some code that does that in the codebase. Am I wrong ?
20090319 23:41:09< Dragonking> Well, there are some.
20090319 23:41:55< Dragonking> Tho I would like to know to what factors/things you pay attention to.
20090319 23:44:07-!- grzybacz [i=grzywacz@kolos.math.uni.lodz.pl] has quit ["leaving"]
20090319 23:44:09< nital> how do you create tables in wesnoth wiki?
20090319 23:44:53< nital> or maybe somebody could point me to one
20090319 23:44:54< Crab_> nital, see http://www.wesnoth.org/wiki/Help:Tables
20090319 23:45:27< nital> thanks
20090319 23:45:53< boucman> Crab_: you're fast :)
20090319 23:46:12-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 60 (Operation timed out)]
20090319 23:47:23< Soliton> nital: what kind of patch utility are you making?
20090319 23:47:25-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090319 23:47:36< Crab_> Dragonking: we can use that existing code, and then try to bring some math to  it.
20090319 23:47:37< Crab_> I personally would consider:
20090319 23:47:37< Crab_>  - forces present at position
20090319 23:47:37< Crab_>  - my income and finance
20090319 23:47:37< Crab_>  - enemy income and finance
20090319 23:47:37< Crab_>  - what damage (in terms of units and gold) will I cause and get if I attack and they attack (now or after several turns)
20090319 23:47:40< Crab_>  - what damage (in terms of units and gold) will I cause and get if I attack and they defend(now or after several turns)
20090319 23:47:43< Crab_>  - what damage (in terms of units and gold) will I cause and get if I defend and they attack(now or after several turns)
20090319 23:47:46< Crab_>  - what damage (in terms of units and gold) will I cause and get if I defend and they defend(now or after several turns)
20090319 23:47:49< Crab_> (basically I think that the AI needs to take day/night cycle into account in order to defend during the day and attack during the night as chaotic)
20090319 23:47:52< Crab_>  - possible future changes due to reinforcement and new units arriving
20090319 23:49:12< Crab_> Dragonking: basically, i want the computer to predict the flow of the situation for a few turns ahead (on basic level: am I going to win here? am I going to lose here?) and act aggressively or defensively accordingly.
20090319 23:49:49< Crab_> and the position is good if it 'looks like I'm going to win there'.
20090319 23:49:53< Crab_> with some numbers attached
20090319 23:50:29< Crab_> also, it is important to take cashflow and unit flow (reinforcements by both sides) into account
20090319 23:50:47< Dragonking> "(now or after several turns)" <- You want to calculate what kind of damage you tkae after several turns?
20090319 23:50:57-!- loonybot [n=loonybot@wesnoth/bot/loonybot] has quit [Read error: 104 (Connection reset by peer)]
20090319 23:51:43< Crab_> Dragonking: no, i want to determine just a basic 'my perspectives looks good here, I should attack' or 'i will soon be outnumbered and overwhelmed, i must fall back'
20090319 23:51:44< noy> AI talk?
20090319 23:51:47< Crab_> aha
20090319 23:52:59< Dragonking> Crab_: What about fog of war? Would your AI consider it?
20090319 23:53:01< Crab_> Dragonking: or : ' i get more gold than he, so let him come and attack me, i'm not in hurry'
20090319 23:53:10< Soliton> trying to guess where the enemy will send reinforcements sounds pretty impossible.
20090319 23:53:11< Crab_> Dragonking: i prefer AI to cheat the FoW :)
20090319 23:53:27< Crab_> Soliton: how humans do that ?
20090319 23:54:03< Soliton> well, depends heavily on the situation. if there is just one front then it's trivial of course.
20090319 23:54:12< Crab_> And if there's two ?
20090319 23:54:34< Crab_> or something like 'caves of the basilisk' ?
20090319 23:54:36< fabi> I see the need for telling the ai something about it's victory conditions.
20090319 23:54:37< Soliton> but as soon as there are more... well, that's complicated.
20090319 23:55:16< Soliton> you can guess based on what you think would be a good idea, i guess.
20090319 23:55:51< fabi> For example could the orc leader in the middle of the siege of elensefar be told that his only goal is to survive and hold the town, he doesn't need to kill every enemy unit.
20090319 23:55:52< Soliton> i'd just not worry about it.
20090319 23:56:02< Crab_> fabi: in my opinion, coreAI doesn't need to know anything about that. but it must be modular and allow configuration from scenario creators, using FormulaAI
20090319 23:57:40< Crab_> fabi: because any restrictions lead to suboptimal behavior. it is better to keep those restrictions external,in my opinion.
20090319 23:58:40-!- BenUrban [n=benurban@c-68-49-10-243.hsd1.md.comcast.net] has joined #wesnoth-dev
20090319 23:59:01< Crab_> fabi: for example, the designer may need to script movement of some units for a few turns. It's better to let him use FormulaAI, not C++, to do this
20090319 23:59:21-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has quit [Remote closed the connection]
--- Log closed Fri Mar 20 00:00:50 2009