--- Log opened Sat Mar 28 00:00:01 2009
20090328 00:04:12< Polarina> Ivanovic: I sent you a revised translation.
20090328 00:07:08-!- noy [n=Noy@wesnoth/developer/noy] has quit [Success]
20090328 00:07:23-!- IneQuation is now known as IneQuation|Away
20090328 00:08:15< mordante> night
20090328 00:08:23-!- mordante [n=mordante@wesnoth/developer/mordante] has quit ["Leaving"]
20090328 00:09:34-!- West_ [n=West@hd5e2754c.gavlegardarna.gavle.to] has quit ["b'bye"]
20090328 00:11:36-!- anakayub [n=anakayub@203.82.79.2] has joined #wesnoth-dev
20090328 00:12:54-!- alink_ is now known as alink
20090328 00:17:16-!- ryochan7 [n=ryochan7@173.9.238.161] has quit ["Leaving"]
20090328 00:21:04< boucman> back (though not for long)
20090328 00:21:48-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has quit ["night all"]
20090328 00:27:01-!- AI0867_ [n=ai@wesnoth/developer/ai0867] has joined #wesnoth-dev
20090328 00:28:47< CIA-30> ivanovic * r34198 /trunk/po/ (26 files in 24 dirs): updated Icelandic and Italian translation
20090328 00:28:58< CIA-30> ivanovic * r34199 /branches/1.6/po/ (26 files in 24 dirs): updated Icelandic and Italian translation
20090328 00:34:07< orian_> night all
20090328 00:34:19-!- orian_ [n=orian@abto229.neoplus.adsl.tpnet.pl] has quit [Remote closed the connection]
20090328 00:36:40-!- gothmog77 [i=992ae250@gateway/web/ajax/mibbit.com/x-89109346b415d9e6] has joined #wesnoth-dev
20090328 00:39:24< Ivanovic> *wow*, SoC is really great, it magically increases the amount of patches we get in
20090328 00:39:26< Ivanovic> ^^
20090328 00:39:27< Ivanovic> https://gna.org/patch/?group=wesnoth
20090328 00:39:47< Polarina> Ivanovic: :)
20090328 00:39:49< boucman> hehe 
20090328 00:40:07< Polarina> I might start developing for Wesnoth after I finish this translation.
20090328 00:40:19< Ivanovic> Polarina: so in maybe three years
20090328 00:40:20< Ivanovic> ;)
20090328 00:40:31< Polarina> It'll take me less than that.
20090328 00:40:36< Ivanovic> getting a translation done as "one man show" takes *lots* of time
20090328 00:40:57< Polarina> Just take a look at the progress: http://www.wesnoth.org/gettext/index.lang.php?lang=is&version=branch
20090328 00:41:10< Ivanovic> oh, at the beginning it is always fast
20090328 00:41:22< Ivanovic> but after the first thousand strings it tends to slow down a lot
20090328 00:41:34< Polarina> How comes?
20090328 00:41:43< Ivanovic> that is just the way it is
20090328 00:41:43< Zen_Clark> Attention span?
20090328 00:42:13< Ivanovic> Zen_Clark: not only this
20090328 00:42:23< Ivanovic> it gets more and more work to keep things consistent
20090328 00:42:27< Polarina> My schedule is minimum 100 strings a day.
20090328 00:42:30< Ivanovic> as in "hmm, how was this named there..."
20090328 00:42:50< Zen_Clark> Ouch....
20090328 00:43:01< Zen_Clark> 100 strings a day?
20090328 00:43:14< Ivanovic> because of this we even started a lovely small list in the german part of the wiki with some "default terms"
20090328 00:43:16< Ivanovic> http://www.wesnoth.org/wiki/GermanTranslationUnitNames
20090328 00:43:23< Polarina> Zen_Clark: Yes. Then I can finish the Icelandic translation in approximately 122 days.
20090328 00:43:29< Ivanovic> Zen_Clark: as *average* this is possible, especialyl when doing campaigns
20090328 00:43:30< alink> I just found http://codepad.org/ and am about to add wesnoth there. Do i make pastes private or public ? http://codepad.org/help/private-projects
20090328 00:43:40 * Polarina doesn't translate names.
20090328 00:43:41< Ivanovic> though for unit descriptions this is probably impossible, those are just too lengthy...
20090328 00:43:53< alink> but that's not really important
20090328 00:44:35< Zen_Clark> I guess adding a translation for lojban might be out of my time league...
20090328 00:44:53< Ivanovic> Polarina: hmm, at least "unit types" need translations
20090328 00:44:56< alink> mmh public then, it's an open source game after all :-)
20090328 00:45:00< Ivanovic> otherwise everything just feels incomplete
20090328 00:45:02< Polarina> Because I do not want to translate "Elvish Fighter" into "Álfskur Bardagamaður" or something crappy like that.
20090328 00:45:13< Polarina> Ivanovic: I think I've already translated the types.
20090328 00:45:32< Ivanovic> uhm, Elvish Fighter is such a "type"
20090328 00:45:34-!- ikarius [n=ross@smtp.gridironsystems.com] has quit []
20090328 00:45:36< Polarina> o.O
20090328 00:45:41 * Polarina won't translate that.
20090328 00:45:50< Ivanovic> what about attack names?
20090328 00:46:00< Polarina> Sure, I can do that.
20090328 00:46:07< Ivanovic> there are plenty of those, too
20090328 00:46:40< Ivanovic> not to mention other "special things" (places and all that stuff, stuff where keeping the english name in the translation does often *not* make sense)
20090328 00:46:54< Polarina> Also, Icelandic has a fairly limited vocabulary when it comes to unit names (Elvish Fighter, Elvish Shaman, etc.).
20090328 00:47:50< Ivanovic> when it comes to "challenging" translations, just go into another cultural area like eg china
20090328 00:47:52< alink> done: http://wesnoth.codepad.org/   http://wesnoth.codepad.org/KGR8v5ZK
20090328 00:48:03< Ivanovic> they in general don't know things like eg gryphons
20090328 00:48:04< Polarina> Hmm...
20090328 00:48:41< alink> loonycyborg: << "Nooooo, not codepad !!!"
20090328 00:48:45< Ivanovic> and in their writing names have a *huge* impact, so in campaigns you explicitly want to translate names
20090328 00:48:58< alink> loonybot << "Nooooo, not codepad !!!"
20090328 00:49:04< loonybot> Nooooo, not codepad !!!
20090328 00:49:15< ilor> why not codepad?
20090328 00:49:28< alink> each time, i make the same mistake :-/
20090328 00:49:31< Polarina> loonybot: << *(NULL)
20090328 00:49:31< loonybot> error: invalid type argument of 'unary *'
20090328 00:50:03< alink> ilor: I just assumed that he doesn't like competition ;p
20090328 00:50:07< loonycyborg> NULL is a macro which expands to 0.
20090328 00:50:20< Polarina> loonycyborg: Evil C++.
20090328 00:50:36< Zen_Clark> Has there been a recent change as to how wesnoth finds added campaigns?
20090328 00:50:36< Polarina> I've also translated all the terrain names. :)
20090328 00:51:01-!- Sirp [n=me@wesnoth/developer/dave] has joined #wesnoth-dev
20090328 00:51:11< Ivanovic> Zen_Clark: depends on what you mean
20090328 00:51:16< Sirp> hi
20090328 00:51:22< boucman> night all
20090328 00:51:24< Ivanovic> if you mean "was the folder structure in trunk currently changed": yes
20090328 00:51:42< alink> gn boucman
20090328 00:51:43-!- boucman [n=rosen@wesnoth/developer/boucman] has quit [Remote closed the connection]
20090328 00:51:46< Ivanovic> if you mean "was the 'algorithm' used to add stuff changed": hmm, different to say
20090328 00:51:57< Zen_Clark> O.k. How was it change?
20090328 00:52:22< Ivanovic> that is: a bug that was reintroduced that forced you to restart the game after installing addons was fixed
20090328 00:52:25< Zen_Clark> Its just not finding my campaign in the .wesnoth1.7 folder
20090328 00:52:27< Ivanovic> so now you don't have to restart
20090328 00:52:55-!- YogiHH [n=chatzill@c174201.adsl.hansenet.de] has quit ["ChatZilla 0.9.84 [Firefox 3.0.7/2009021910]"]
20090328 00:52:57< Ivanovic> yeah, now it has to be in add-ons not campaigns anymore (or was it addons, really not sure?)
20090328 00:53:00< Ivanovic> hi Sirp
20090328 00:53:52< Zen_Clark> O.k.
20090328 00:55:58< ilor> anyone: where is the code for our lobby bot?
20090328 00:56:45< Ivanovic> ilor: probably in utils/
20090328 00:57:02< Ivanovic> mp-lobby-logger.pl
20090328 00:57:08< Ivanovic> that would at least be my guess
20090328 00:57:15< ilor> ouch, it's encrypted :/
20090328 00:57:43< Ivanovic> no, it is perl
20090328 00:57:45< Ivanovic> ^^
20090328 00:57:49< ilor> my point exactly
20090328 00:58:52< Zen_Clark> ilor: At least its not written using the windows api, that would make it much worse.
20090328 00:59:07 * ilor has a vision of a perl program using winapi
20090328 01:00:05 * Zen_Clark does not think it would be called a vision... More like a nightmare.
20090328 01:00:22< Ivanovic> ilor: there is worse than perl...
20090328 01:00:49-!- JosiahSonne [i=992ae250@gateway/web/ajax/mibbit.com/x-4fa22ef059b14b23] has joined #wesnoth-dev
20090328 01:01:00-!- gothmog77 [i=992ae250@gateway/web/ajax/mibbit.com/x-89109346b415d9e6] has quit ["http://www.mibbit.com ajax IRC Client"]
20090328 01:01:01< Zen_Clark> Ivanovic: BASIC?
20090328 01:01:11< Ivanovic> Zen_Clark: nah, haskell
20090328 01:01:30< ilor> ahh, I forgot Ivanovic's favorite ;D
20090328 01:01:43< Ivanovic> though x86 asm ain't too nice either...
20090328 01:01:46< ilor> Ivanovic: what do you think about ocaml?
20090328 01:01:46< Ivanovic> ;)
20090328 01:02:15< loonycyborg> Ivanovic: Perl and Haskell might be hard to read, but for different reasons.
20090328 01:02:25< Ivanovic> ilor: never heard about it, what is it, some arabic food?
20090328 01:02:26< dfranke> Haskell syntax isn't ugly, it's just a sort of linguistic isolate.
20090328 01:02:26< Ivanovic> ;)
20090328 01:02:31-!- JW1 [n=X@c-71-57-85-168.hsd1.il.comcast.net] has joined #wesnoth-dev
20090328 01:03:23< dfranke> a rather beautiful one once you take the time to learn it.
20090328 01:03:34< Ivanovic> one of my mail problems with haskell is debugging
20090328 01:03:38< Ivanovic> s/mail/main
20090328 01:03:52< Ivanovic> plus that it cares so much about whitespaces
20090328 01:03:54< dfranke> eh?  What could be easier to debug than Haskell?
20090328 01:04:11< Ivanovic> hmm, let me think, almost every other lang...
20090328 01:05:05< loonycyborg> Why would one debug Haskell, anyway? :P
20090328 01:05:14< dfranke> it's rare that any bugs make it past the typechecker to begin with, and then it's purely functional, so if you want to see what a piece of code does, just evaluate it, because it does the same thing every time.
20090328 01:06:02< dfranke> every time I've had a Haskell program not work perfectly within a few minutes of its first successful compilation, it's been due to a compiler bug.
20090328 01:07:45-!- molgrum [n=molgrum@c83-249-53-163.bredband.comhem.se] has quit ["Lämnar"]
20090328 01:11:23< ilor> Ivanovic: you do know there's a haskell program in this very channel? ;P
20090328 01:11:49< Ivanovic> ilor: i don't care
20090328 01:12:12< dfranke> there is?
20090328 01:12:49< loonycyborg> dfranke: loonybot is an instance of geordi bot which is written in haskell.
20090328 01:13:12< dfranke> > read "42" :: Int
20090328 01:13:18< dfranke> guess it isn't lambdabot
20090328 01:13:31< loonycyborg> http://www.xs4all.nl/~weegen/eelis/geordi/
20090328 01:15:33< Polarina> loonybot: 2 + 2 * 2
20090328 01:15:33< loonybot> error: expected unqualified-id before numeric constant
20090328 01:15:46-!- Ambushsabre [n=ambushsa@c-75-67-216-121.hsd1.nh.comcast.net] has joined #wesnoth-dev
20090328 01:15:59< Polarina> loonybot: You are an expected unqualified-id before a numeric constant!
20090328 01:16:00< loonybot> error: 'You' does not name a type
20090328 01:16:21< loonycyborg> loonybot: << 2 + 2 * 2
20090328 01:16:23< loonybot> 6
20090328 01:16:34< Polarina> Does it simplify too? :P
20090328 01:16:41< noy_> I'm out people
20090328 01:16:49< noy_> going to see Monsters Vs Aliens!
20090328 01:16:51< noy_> hah
20090328 01:17:03< dfranke> loonybot: << "loonybot: << \"foo\""
20090328 01:17:04< loonybot> loonybot: << "foo"
20090328 01:17:06< Polarina> Monsters == Aliens.
20090328 01:17:12-!- noy_ [n=Noy@S0106001a706b09fb.vc.shawcable.net] has quit ["GO, GET TO THE CHOPPAH!!!"]
20090328 01:17:42< dfranke> if that worked, it'd be fun to give it a quine.
20090328 01:17:54< Polarina> Hehe.
20090328 01:21:37-!- silene [n=plouf@wesnoth/developer/silene] has quit [Read error: 148 (No route to host)]
20090328 01:22:11< Polarina> loonybot: { int crap[3] = { }; cout << crap[3 * 1024]; }
20090328 01:22:16< loonybot> Undefined behavior detected.
20090328 01:22:21< Polarina> :(
20090328 01:22:25 * Polarina goes to bed.
20090328 01:22:44-!- Dragonking [n=dk@wesnoth/developer/dragonking] has quit []
20090328 01:28:20-!- Polarina [n=polarina@wesnoth/translator/Polarina] has quit ["I know you're reading this."]
20090328 01:39:04< fendrin> Ivanovic: In haskell sieve of erathostenes is a single line!
20090328 01:41:19-!- Elvish_Pillager [n=eli@66.189.40.122] 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!"]
20090328 01:45:37-!- West_ [n=West@hd5e2754c.gavlegardarna.gavle.to] has joined #wesnoth-dev
20090328 01:52:26< Ivanovic> time for me to head off to bed, too
20090328 01:52:27< Ivanovic> n8
20090328 01:52:35< Velory> n8 ivanovic
20090328 01:55:03-!- ikarius [n=ross@216.27.182.3] has joined #wesnoth-dev
20090328 01:56:23-!- Sirp [n=me@wesnoth/developer/dave] has quit [Read error: 60 (Operation timed out)]
20090328 02:07:34-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has quit [Remote closed the connection]
20090328 02:07:54-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has joined #wesnoth-dev
20090328 02:08:13-!- ilor [n=user@wesnoth/developer/ilor] has quit []
20090328 02:16:25-!- ryochan7 [n=ryochan7@173.9.238.161] has joined #wesnoth-dev
20090328 02:24:32-!- Zen_Clark [n=user@99-136-80-191.lightspeed.rcsntx.sbcglobal.net] has quit ["Too many bottles bugs on the wall. Too many bottles of bugs."]
20090328 02:26:34< Velory> Well ive finished my healing improvement patch
20090328 02:26:45< Velory> and good night every one :)
20090328 02:26:53-!- [Relic] [n=[Relic]@76.229.202.137] has joined #wesnoth-dev
20090328 02:26:55-!- Velory [n=GProcess@81.213.68.17] has left #wesnoth-dev []
20090328 02:27:49< [Relic]> Hello :)
20090328 02:28:14-!- loonycyborg [n=sergey@wesnoth/developer/loonycyborg] has quit ["Zzzzzzzzzzzzzzzzzzzzzzzzzz"]
20090328 02:28:56-!- loonybot [n=loonybot@wesnoth/bot/loonybot] has quit [Remote closed the connection]
20090328 02:29:37< Crab_> hi, [Relic]
20090328 02:29:46< [Relic]> :)
20090328 02:30:50-!- faryshta [n=faryshta@189.181.149.246] has joined #wesnoth-dev
20090328 02:31:13< faryshta> Hi, I would like to ask information about the Google Summer of Code.
20090328 02:36:16< Soliton> check the link in the topic.
20090328 02:37:02< faryshta> Jajaja RTFA to myself. Thanks Soliton.
20090328 02:37:09-!- happygrue [n=George@wesnoth/developer/wintermute] has quit ["http://xkcd.com/91/"]
20090328 02:43:55-!- res|laptop [n=res@pdpc/supporter/student/res2k] has quit [Read error: 110 (Connection timed out)]
20090328 02:44:37-!- res|laptop [n=res@erft-4d07fc79.pool.mediaWays.net] has joined #wesnoth-dev
20090328 02:56:14-!- West_ [n=West@hd5e2754c.gavlegardarna.gavle.to] has quit ["b'bye"]
20090328 02:59:45-!- BenUrban [n=benurban@c-68-49-10-243.hsd1.md.comcast.net] has joined #wesnoth-dev
20090328 03:13:37-!- Sapient [n=sapien-x@c-71-204-10-166.hsd1.ga.comcast.net] has joined #wesnoth-dev
20090328 03:19:56< Blueblaze> hey Sapient
20090328 03:20:01< Sapient> hello Blueblaze
20090328 03:20:40< Blueblaze> Sapient: How do I use [and] [or] [not] within an [if] statement?
20090328 03:21:22< Sapient> you mean http://www.wesnoth.org/wiki/ConditionalWML ?
20090328 03:21:54< Sapient> here are some examples: http://www.wesnoth.org/wiki/AdvancedConditionalWML
20090328 03:22:19< Blueblaze> ah that's what I wanted, thanks
20090328 03:23:09< Blueblaze> Sapient: It looks like if I wanted to use and, I don't need to even use the statement, just put condition 1 behind condition 2 within [if]
20090328 03:23:41< Sapient> right... there are very few situations where one actually needs [and]
20090328 03:23:46< Blueblaze> That's what I thought, but I was hoping I might be wrong for the sake of debugging :P
20090328 03:24:55-!- faryshta [n=faryshta@189.181.149.246] has quit [Remote closed the connection]
20090328 03:24:56< Sapient> "Useful as a bracket for complex conditions, but not strictly necessary"
20090328 03:25:31< Sapient> the situations where I use [and] usually involve location filters with radius=
20090328 03:26:27< CIA-30> crab * r34200 /trunk/ (changelog src/actions.cpp): Fixed incorrect handling of poisoning attacks when suggesting best attack in user interface
20090328 03:26:30< Sapient> moving parts of the filter into [and] subtag allows you to change the radius= behavior
20090328 03:26:50< Blueblaze> ah, I see
20090328 03:26:54< Shadow_Master> Crab_: you are developer! congratulations!
20090328 03:27:26< Crab_> Shadow_Master: thanks ) yes, from yesterday ) even got myself a nice first commit number )
20090328 03:27:26 * Sapient gives crab a crab-cake
20090328 03:27:47 * Blueblaze gives crab crab coke
20090328 03:27:53 * Crab_ smiles
20090328 03:27:58< Blueblaze> That's what I thought you said originally -_-
20090328 03:28:15< Blueblaze> Sapient: Did you ever upload your map pack to 1.6, or do you not have plans for it there?
20090328 03:28:25< Sapient> mmm... crab flavored coke. sounds better than Coke Black (coffee infustion), though
20090328 03:28:37< Shadow_Master> Crab_: are you using eyeless smileys?  o.O
20090328 03:28:57< Crab_> Shadow_Master: no :)
20090328 03:28:58< Sapient> Blueblaze: I haven't been maintaining that map pack for a long time
20090328 03:29:10< Sapient> Mythological was actually maintaining it for me lately
20090328 03:29:20< Blueblaze> Ahh okay
20090328 03:29:31< Sapient> why did you miss some map?
20090328 03:30:51< Blueblaze> nah lol
20090328 03:30:56< Sapient> I'm kinda disgusted with it after working on it and not getting the maps as good as I wanted them
20090328 03:31:20< Blueblaze> It was always a very popular map pack though
20090328 03:31:35-!- grantwu [n=zenneth@westquad-151193.reshall.umich.edu] has joined #wesnoth-dev
20090328 03:31:44< dfranke> corn: around?
20090328 03:31:47< Sapient> probably part of the popularity was just because I had the top slot ;)
20090328 03:34:48< Blueblaze> heh
20090328 03:35:21< Blueblaze> Sapient: Is there any reason why this wouldn't trigger properly? http://pastebin.com/d3a147a1 Basically the if statements are the same repeated 4 times with [not]s in different places of the conditions
20090328 03:35:26 * Shadow_Master likes popcorn.
20090328 03:35:37< Blueblaze> 4 different posibitilies, yes|yes, yes|no, no|yes, no|no
20090328 03:37:29< CIA-30> crab * r34201 /branches/1.6/ (changelog src/actions.cpp): backport r34200 from trun (Fixed incorrect handling of poisoning attacks when suggesting best attack in user interface)
20090328 03:38:29 * Shadow_Master |= SM_PEDANTIC;
20090328 03:38:42< Shadow_Master> you misspelled trunk. ^^
20090328 03:39:40 * Sapient throws sugar packet at Shikadi ------------>#
20090328 03:40:02< Crab_> Shadow_Master: yes, I know :) 
20090328 03:40:45 * Shadow_Master takes his ion cannon and destroys the sugar packet
20090328 03:40:48< Sapient> Blueblaze: I think that last [if] may be wrong
20090328 03:41:35< Sapient> you probably meant to do [not]a[or]b[/or][/not]
20090328 03:41:37-!- oak_ [i=oakus@c02--112b.kn.vutbr.cz] has quit [Read error: 110 (Connection timed out)]
20090328 03:42:04< Sapient> which could also be expressed as [not]a[/not][not]b[/not]
20090328 03:43:08< Blueblaze> Sapient: so what does [not]ab[/not] do?
20090328 03:43:56< Sapient> that will be true if a,b, or both a+b are untrue
20090328 03:44:30< Sapient> that will be true if a untrue,b untrue, or both a+b are untrue
20090328 03:44:35 * Sapient clarifies
20090328 03:44:55< Blueblaze> ahh I get it
20090328 03:45:13< Blueblaze> So it will trigger off on the no|yes and yes|no conditions with the no|no?
20090328 03:45:45< Sapient> if I translate that into english, then yes
20090328 03:45:58< Blueblaze> heh
20090328 03:49:09-!- JW1 [n=X@c-71-57-85-168.hsd1.il.comcast.net] has quit [Read error: 110 (Connection timed out)]
20090328 03:50:01< Blueblaze> Sapient: Bleh, I tried both what you recommended and it triggers off the no|no command whenever yes|no or no|yes occurs
20090328 03:50:49< Sapient> trust me, I know how conditionalwml works
20090328 03:51:54< Sapient> maybe put in a sanity check to see if you're running the old version somehow, or paste it
20090328 03:52:14< Sapient> but I will be gone to the store for a few mins
20090328 03:52:18 * Sapient afk
20090328 03:52:41< Shadow_Master> cp: target `/tmp/Invasion_from_the_Unknown-1.12.0.fakeroot.dir/idata/metatags/' is not a directory: No such file or directory
20090328 03:52:57< Shadow_Master> AI0867: it seems I cannot write to /tmp for some reason.
20090328 03:53:07< Shadow_Master> wrong channel.
20090328 03:54:22< Soliton> no, it seems you're writing somewhere inexistant.
20090328 03:54:49< Shadow_Master> that's strange. It would work here in my laptop.
20090328 03:56:13< Shadow_Master> no arch/filesystem layout differences besides that machine runs etch and mine runs lenny. um..
20090328 03:56:57< Soliton> i doubt the /tmp dir has the same layout.
20090328 03:58:25< Shadow_Master> the only thing the script assumes is the existence of /tmp. Any subdirs it references are created by it
20090328 03:58:59< Soliton> looks like it failed then.
20090328 03:59:36< Shadow_Master> yup. Just realized that the 'svn export' line isn't working.
20090328 04:04:11< Shadow_Master> pah. it isn't it. Using my no-svn-export method fails the same. I'll go to sleep instead.
20090328 04:24:53-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090328 04:34:08-!- Ivanovic_ [n=ivanovic@dtmd-4db2cd23.pool.einsundeins.de] has joined #wesnoth-dev
20090328 04:38:49-!- elminse [n=elminsei@netblock-208-127-84-182.dslextreme.com] has joined #wesnoth-dev
20090328 04:41:48-!- Crab_ [n=crab_@c.218.175.a530.sta.adsl.cyfra.net] has quit ["Leaving."]
20090328 04:42:13-!- Sirp_ [n=me@69.26.208.78] has joined #wesnoth-dev
20090328 04:42:35< Sirp_> hi
20090328 04:42:56-!- Sirp_ [n=me@69.26.208.78] has quit [Remote closed the connection]
20090328 04:43:52-!- anakayub [n=anakayub@203.82.79.2] has quit []
20090328 04:43:57-!- Sirp_android [n=me@69.26.208.78] has joined #wesnoth-dev
20090328 04:44:07< Sirp_android> hi
20090328 04:48:58< esr> Sirp_android: I may actually have a functional use for one of those titles you occasionally hand out...
20090328 04:50:40-!- Ambushsabre [n=ambushsa@c-75-67-216-121.hsd1.nh.comcast.net] has quit ["Leaving"]
20090328 04:51:08-!- Ivanovic [n=ivanovic@wesnoth/developer/ivanovic] has quit [Read error: 110 (Connection timed out)]
20090328 04:52:06-!- Ivanovic_ is now known as Ivanovic
20090328 04:53:52< Sirp_android> esr: ok but I'm not usually the one who hands them out. :)
20090328 04:54:02< esr> ?
20090328 04:54:07< esr> Who does?
20090328 05:02:21-!- Sirp_android [n=me@69.26.208.78] has quit [Read error: 54 (Connection reset by peer)]
20090328 05:04:06< Sapient> lvanovic
20090328 05:09:35< Blueblaze> oh Sapient, I pasted the last [if] conditionals, it triggers every time regardless if the condition tags are met or not. http://pastebin.com/d3a85a078
20090328 05:15:05< Sapient> I have a hard time believing that
20090328 05:15:15< Sapient> but since you insist, I'll try a test case here
20090328 05:16:14< Sapient> if what you say is true, then something is seriously broken
20090328 05:25:31< Sapient> Blueblaze: I just tested it and it works fine
20090328 05:27:12< Sapient> http://pastebin.com/m74e0db8a
20090328 05:27:57< Sapient> the problem could be with your macro definition or usage
20090328 05:28:46-!- Sirp_android [n=me@69.26.208.78] has joined #wesnoth-dev
20090328 05:29:26< Sirp_android> re-hi
20090328 05:30:49< Blueblaze> Sapient: Hrm, I'll try it without the macro, but I don't see how that could be a problem as the macro is used for the other conditions too without problem
20090328 05:31:25-!- ABCD_ [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090328 05:32:02-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Read error: 104 (Connection reset by peer)]
20090328 05:32:22< Blueblaze> Sapient: So whenever a unit was there, your event didn't trigger off?
20090328 05:32:32< Sapient> right
20090328 05:32:54< Sapient> take that code and put it in the scenario-test.cfg if you want to try it for yourself
20090328 05:33:06< Sapient> those two positions are the silver mage and the alvish avenger
20090328 05:33:12-!- ABCD_ [n=ABCD@wikipedia/ABCD] has quit [Read error: 104 (Connection reset by peer)]
20090328 05:33:48-!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090328 05:34:58< Blueblaze> Sapient: Do you think the fact that I'm using this in a [set_menu_item] cause a problem?
20090328 05:35:13-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Read error: 104 (Connection reset by peer)]
20090328 05:35:51< Sapient> there's nothing wrong with your ConditionalWML except possibly macro usage issues
20090328 05:36:14< Blueblaze> Alright, let me try it without the macro first before anything
20090328 05:37:12-!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090328 05:37:22-!- chains [n=Rylar@adsl-76-206-236-184.dsl.chcgil.sbcglobal.net] has quit [Remote closed the connection]
20090328 05:38:21-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Client Quit]
20090328 05:38:39-!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090328 05:43:53< Blueblaze> okay, so the problem isn't in my macro lol
20090328 05:45:16< Blueblaze> I guess I'll minimalize the scenario and debug it
20090328 05:45:28< Sapient> good luck with that
20090328 06:05:29-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev
20090328 06:08:08-!- ryochan7 [n=ryochan7@173.9.238.161] has quit ["Leaving"]
20090328 06:11:51-!- grantwu [n=zenneth@westquad-151193.reshall.umich.edu] has left #wesnoth-dev []
20090328 06:19:12-!- alink [n=alink@78.129.18.215] has quit [Remote closed the connection]
20090328 06:23:34-!- Sirp_android [n=me@69.26.208.78] has quit [Read error: 110 (Connection timed out)]
20090328 06:26:09-!- csaunders [n=chatzill@ip216-239-77-94.vif.net] has joined #wesnoth-dev
20090328 06:26:36< csaunders> Hello, would anyone be able to help me make sense of some database data dumps?
20090328 06:27:46< Sapient> csaunders: you do realize this is #wesnoth-dev, right?
20090328 06:28:06< csaunders> yes
20090328 06:28:10< csaunders> it is for wesnoth
20090328 06:28:11< csaunders> :P
20090328 06:28:22< csaunders> the summer of code project for the stats server
20090328 06:28:39< csaunders> i am just having a bit of a time trying to get what a few of the CSV columns mean :(
20090328 06:29:27< Sapient> ok, looks like you just missed Sirp unfortunately
20090328 06:30:06< csaunders> shoot
20090328 06:30:14< csaunders> what time is he typically on at?
20090328 06:30:15< Sapient> feel free to just ask your question and maybe somebody lurking will know it or read it in the logs
20090328 06:30:22< csaunders> hmm alright
20090328 06:30:41< csaunders> well essentially, Sirp put up a data dump here -> http://www.wesnoth.org/cgi-bin/stats/dump.pl
20090328 06:31:17< csaunders> sadly, the dump isn't fully documented, and I don't know what each column means for a table description
20090328 06:32:34< Sapient> what particular columns did you have a question about?
20090328 06:32:54< csaunders> for example: game_id|int(11)|NO|PRI||auto_increment is (columnid, type,required(Y/N), key_type, ?, incr_type)
20090328 06:33:29< csaunders> i am just having a tough time trying to make sense of that part.  Once i make sense of that I can get a script to turn all of this into SQL that I can then put in my database
20090328 06:33:42< Sapient> that last column is the "comment" field in mysql
20090328 06:33:59< Sapient> this looks like standard mysql DESCRIBE output
20090328 06:34:53< Sapient> http://dev.mysql.com/doc/refman/5.1/en/describe.html
20090328 06:35:13< csaunders> *facepalm*
20090328 06:35:15< csaunders> thank you :D
20090328 06:36:03< Sapient> no problem
20090328 06:39:39-!- csaunders [n=chatzill@ip216-239-77-94.vif.net] has quit [Remote closed the connection]
20090328 06:40:56-!- csaunders [n=chatzill@ip216-239-77-94.vif.net] has joined #wesnoth-dev
20090328 06:42:34< corn> dfranke: do you know if the upload bug appears under windows?
20090328 06:48:30-!- [Relic] [n=[Relic]@76.229.202.137] has quit ["Leaving"]
20090328 06:52:21-!- Espreon [n=espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20090328 06:55:00< dfranke> corn: no I don't.
20090328 06:55:07< dfranke> corn: don't see why it wouldn't though.
20090328 07:02:59< corn> dfranke: well, tcp timeout is OS specific. if the problem is with an underlying connect() it would depend on the OS
20090328 07:16:41< Soliton> crimson_penguin: there is some guy trying to compile wesnoth on a mac in #wesnoth. maybe you can help him.
20090328 07:18:03-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has quit [Remote closed the connection]
20090328 07:22:16-!- mordante [n=mordante@roadie.xs4all.nl] has joined #wesnoth-dev
20090328 07:22:41< mordante> hi
20090328 07:25:31-!- grantwu [n=Zenneth@westquad-147025.reshall.umich.edu] has joined #wesnoth-dev
20090328 07:28:07< Espreon> Hello mordante.
20090328 07:28:51< Sapient> yo mordante
20090328 07:29:42< Sapient> mordante: is it true that malformed terrain-strings could crash wesnoth?
20090328 07:29:53< mordante> hi Espreon 
20090328 07:29:56< mordante> hi Sapient 
20090328 07:30:13< mordante> Sapient, not sure any reason why you ask?
20090328 07:31:07< Sapient> http://www.wesnoth.org/wiki/TerrainCodesWML states "Note most of these rules are not validated since it would slow down Wesnoth; not following these rules might break Wesnoth."
20090328 07:32:06-!- happygrue [n=George@wesnoth/developer/wintermute] has quit ["http://xkcd.com/91/"]
20090328 07:32:15< Sapient> but our policy is that WML you might encounter while a scenario is running should not crash wesnoth or assert-fail
20090328 07:32:22< mordante> well it's exactly as written there, no idea what happens if somebody feels is bad data
20090328 07:32:55< mordante> the parsing of the terrain-graphics is really slow adding more checks will make it even slower
20090328 07:33:14< Sapient> in that case, all terrain strings encountered by ActionWML or SLF should go through an additional check
20090328 07:33:36< Sapient> it doesn't have to affect terrain graphics in general
20090328 07:35:09< Sapient> a config::error which aborts the scenario is acceptable
20090328 07:36:28< mordante> does it happen in practice? The code has been like that since 1.4 and never heard a problem with that part
20090328 07:36:52< Sapient> no, I was just going by what the wiki states
20090328 07:38:31< mordante> the problem with a lot of those rules I can't check them
20090328 07:38:53< mordante> if an UMC creator uses a terrain without the YyZz things might break
20090328 07:39:21< Sapient> I don't understand exactly what you mean
20090328 07:39:52< Sapient> can't a regular expression be used?
20090328 07:40:30< mordante> no since some things are a 'social' policy which a computer can't check
20090328 07:41:03< Sapient> I still don't understand
20090328 07:41:49< mordante> how can I know whether Gg^Abcd is defined for the main game by a Wesnoth developer or by an UMC creator who thought it looks nice
20090328 07:42:28< mordante> if it's the UMC creator it's not honouring the rules, if it's a Wesnoth dev it's oke
20090328 07:43:33< Sapient> ah, so even if it follows the pattern, it must be a valid terrain string, or wes could break?
20090328 07:44:19< Sapient> my main concern is security
20090328 07:44:25< mordante> if two terrains with the same letter occur there will be undefined behaviour, that was with the old and new engine
20090328 07:45:07-!- turin [n=turin@168.215.250.1] has quit [Read error: 110 (Connection timed out)]
20090328 07:45:58< Sapient> if someone hosts a game with an invalid terrain="^mad^hacks001010010111" and gets it to write in an undesired memory location then that's a security issue
20090328 07:46:08-!- silene [n=plouf@ASte-Genev-Bois-152-1-100-67.w86-203.abo.wanadoo.fr] has joined #wesnoth-dev
20090328 07:46:17< Sapient> that's why I brought it up
20090328 07:46:41< mordante> no there will be no random memory buffers overwritten
20090328 07:46:47< Sapient> ok then
20090328 07:47:38< mordante> I don't want to add security problems ;-)
20090328 07:48:35< mordante> but I think a terrain with a ? will work as well, but we might want to define it later for some special purpose
20090328 07:48:58< mordante> so we define a ? for something special and suddenly the addon might not work anymore
20090328 07:49:38< silene> hi
20090328 07:49:44< Sapient> yo silene
20090328 07:49:49< mordante> in that case it's the problem of the author to fix his/her addon
20090328 07:49:51< mordante> hi silene 
20090328 07:50:22< mordante> and with 7 free to pick letters for addon creators I think they have enough options
20090328 07:51:20< Sapient> right
20090328 07:52:43< Sapient> I think the editor should show a warning the first time you place a UMC terrain
20090328 07:53:26< Sapient> "this terrain may not work properly due to... [] don't show this warning again"
20090328 07:55:44< mordante> in that case it would be easier to do that when parsing the data and show it at the start
20090328 07:55:57-!- KnightR [n=anonymou@bb219-74-177-1.singnet.com.sg] has joined #wesnoth-dev
20090328 07:56:04< mordante> then we don't need to remember the state of which terrain have or have not been used
20090328 07:56:29< Sapient> possibly; I guess ilor would be the one to ask
20090328 07:56:40< Sapient> wesbot: seen ilor?
20090328 07:56:40< wesbot> Sapient: The person with the nick ilor last spoke 6h 45m ago.  5h 48m ago was here and on the channels #wesnoth and #wesnoth-mp with the message:
20090328 07:58:24< CIA-30> soliton * r34202 /trunk/utils/mp-server/send_server_message: typo fix
20090328 07:58:29< CIA-30> soliton * r34203 /trunk/utils/mp-server/update_server: set build flags based on hostname
20090328 07:58:37< CIA-30> soliton * r34204 /trunk/ (changelog src/unit_abilities.cpp src/unit_types.hpp): * Fixed bug #13257: Attack dialog always uses the active name of a weapon special
20090328 08:00:28< mordante> yes
20090328 08:15:53-!- grantwu [n=Zenneth@westquad-147025.reshall.umich.edu] has left #wesnoth-dev []
20090328 08:21:25< Sapient> cya later, have fun.
20090328 08:21:26-!- Sapient [n=sapien-x@wesnoth/developer/sapient] has left #wesnoth-dev []
20090328 08:24:11< noy> dfranke: game?
20090328 08:34:32-!- orian_ [n=orian@adab224.neoplus.adsl.tpnet.pl] has joined #wesnoth-dev
20090328 08:35:13< orian_> hi all:)
20090328 08:36:42< CIA-30> mordante * r34205 /trunk/CMakeLists.txt: Lua is only mandatory for the game.
20090328 08:37:38< mordante> hi orian_ 
20090328 08:54:06-!- csaunders [n=chatzill@ip216-239-77-94.vif.net] has left #wesnoth-dev []
20090328 08:56:37-!- Espreon [n=espreon@wesnoth/developer/espreon] has quit [Remote closed the connection]
20090328 09:04:06-!- ABCD is now known as ABCD_sleep
20090328 09:08:12< mordante> I'm off bye
20090328 09:08:27-!- mordante [n=mordante@wesnoth/developer/mordante] has quit ["Leaving"]
20090328 09:11:11-!- viggy_prabhu [i=d2d40596@gateway/web/ajax/mibbit.com/x-b2178021abb92d84] has quit [Remote closed the connection]
20090328 09:11:11-!- JosiahSonne [i=992ae250@gateway/web/ajax/mibbit.com/x-4fa22ef059b14b23] has quit [Remote closed the connection]
20090328 09:23:51-!- dlr365 [n=doug@h44-199.reznet.ucalgary.ca] has joined #wesnoth-dev
20090328 09:57:01-!- boucman [n=rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20090328 09:59:57-!- oak_ [i=oakus@c02--112b.kn.vutbr.cz] has joined #wesnoth-dev
20090328 10:07:02-!- cristi_ceata [n=morpheus@p22.eregie.pub.ro] has joined #wesnoth-dev
20090328 10:07:21-!- cristi_ceata [n=morpheus@p22.eregie.pub.ro] has left #wesnoth-dev []
20090328 10:07:56-!- cristi_ceata [n=morpheus@p22.eregie.pub.ro] has joined #wesnoth-dev
20090328 10:08:00-!- cristi_ceata [n=morpheus@p22.eregie.pub.ro] has left #wesnoth-dev []
20090328 10:08:40-!- noy [n=Noy@wesnoth/developer/noy] has quit []
20090328 10:22:27-!- loonybot [n=loonybot@79.139.246.190] has joined #wesnoth-dev
20090328 10:23:03-!- loonycyborg [n=sergey@79.139.246.190] has joined #wesnoth-dev
20090328 10:24:47-!- YogiHH [n=chatzill@c174232.adsl.hansenet.de] has joined #wesnoth-dev
20090328 10:27:21< Ivanovic> moin
20090328 10:27:34< boucman> moin moin
20090328 10:33:47-!- euschn [n=chatzill@85-127-104-186.dynamic.xdsl-line.inode.at] has joined #wesnoth-dev
20090328 10:33:51< euschn> good morning
20090328 10:34:56-!- dlr365 [n=doug@h44-199.reznet.ucalgary.ca] has quit ["rawr I'm an emu!"]
20090328 10:35:48-!- YogiHH [n=chatzill@c174232.adsl.hansenet.de] has quit ["ChatZilla 0.9.84 [Firefox 3.0.8/2009032609]"]
20090328 10:37:05< CIA-30> dfranke * r34206 /branches/1.6/src/unit.cpp: Make sure the formula_vars map is initialized before assigning to it. Fixes bug #13279.
20090328 10:38:09-!- zookeeper [n=l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20090328 10:40:52< CIA-30> dfranke * r34207 /trunk/src/unit.cpp: Port r34206 (bug #13279) to trunk
20090328 10:43:04-!- orian_ [n=orian@adab224.neoplus.adsl.tpnet.pl] has quit ["using sirc version 2.211+KSIRC/1.3.12"]
20090328 10:43:06-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090328 10:43:55-!- orian_ [n=orian@adab224.neoplus.adsl.tpnet.pl] has joined #wesnoth-dev
20090328 10:49:06-!- Elvish_Pillager [n=eli@66-189-40-122.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090328 10:50:13< CIA-30> ivanovic * r34208 /trunk/po/ (6 files in 6 dirs): updated Chinese (Simplified) translation
20090328 10:50:19< CIA-30> ivanovic * r34209 /branches/1.6/po/ (6 files in 6 dirs): updated Chinese (Simplified) translation
20090328 10:55:05< boucman> cjhopman: around ?
20090328 11:04:54-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090328 11:10:14-!- Dragonking [n=dk@wesnoth/developer/dragonking] has joined #wesnoth-dev
20090328 11:14:08-!- ABCD_sleep [n=ABCD@wikipedia/ABCD] has quit ["No Ping reply in 30 seconds."]
20090328 11:14:42-!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090328 11:34:39-!- JW1 [n=X@c-71-57-85-168.hsd1.il.comcast.net] has joined #wesnoth-dev
20090328 11:39:32-!- crimson_pingvin [n=ben@S0106000d88bbc764.vc.shawcable.net] has joined #wesnoth-dev
20090328 11:40:49< euschn> I have a question about Easycoding/village renaming: Id like to run the standalone test program in mapgen.cpp, but I failed to build it so far. scons gives me an error: http://wesnoth.pastebin.com/m6d50ebed
20090328 11:41:27< euschn> what is the easiest way to get this to run? should I write my own make file (or sconscript), and whats the easiest way to do this?
20090328 11:42:30-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090328 11:43:12< Ivanovic> uhm, what exactly have you tried to do?
20090328 11:43:31< Ivanovic> you changed mapgen.cpp to include a main function there, correct?
20090328 11:43:56< euschn> Ivanovic: yes, it is already there
20090328 11:44:06< Ivanovic> this will of course not work if you build normally via scons
20090328 11:44:17< euschn> wrapped by a ifdef TEST_MAPGEN
20090328 11:44:47< Ivanovic> okay, and how have you changed the build system to only consider this main function?
20090328 11:44:48< euschn> Ivanovic: yes, I was not really expecting it to work
20090328 11:45:35< euschn> I have not yet at all
20090328 11:45:35< Ivanovic> you could try it with cmake, there it is rather easy to get working: define the set of cpp files (omit the old game.cpp, since this is probably where the "normal" main is in)
20090328 11:46:11< Ivanovic> set an add_definitions() block and you are basically done
20090328 11:46:12< euschn> Ivanovic: thanks, I will try that
20090328 11:46:36-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit [Read error: 110 (Connection timed out)]
20090328 11:49:59< Ivanovic> euschn: should be something like this: http://wesnoth.pastebin.com/m530ebf7d
20090328 11:50:27< Ivanovic> basically you need the set with the source files, create the executable, define the libs to link against and define the extra compile flag only for this target
20090328 11:50:49< Ivanovic> this goes into src/CMakeLists.txt
20090328 11:52:16< euschn> Ivanovic: thanks a lot.
20090328 11:52:55< Ivanovic> keep in mind that this is only a very rude example and i might have missed some stuff, so you will probably at least have to edit the list of libs to link against as well as the set of files
20090328 11:53:20< euschn> sure, I will play around with it
20090328 11:54:30< Ivanovic> for building: just create a new folder somewhere and call: cmake PAT/TO/trunk-checkout
20090328 11:54:49< Ivanovic> afterwards you can do a simple "make mapgen-test" to only build this binary
20090328 11:55:04< euschn> ok
20090328 11:55:35< Ivanovic> (if you run a plain make, it is the same as building all binaries, which might be a little overkill for your situation ;) )
20090328 11:55:48< euschn> hehe yes
20090328 11:56:50< loonycyborg> euschn: Naturally there's an equivalent way of doing it with scons ;)
20090328 11:56:52< Ivanovic> some info about using cmake is also in the INSTALL file as well as on the compiling page in the wiki
20090328 11:57:12< Ivanovic> loonycyborg: i am sure there is, i just don't know it off hand (and for cmake it is easy enough ;) )
20090328 11:58:05-!- RainCT [n=RainCT@ubuntu/member/rainct] has joined #wesnoth-dev
20090328 11:58:15< RainCT> Hi
20090328 11:58:19-!- Velory [n=GProcess@81.213.176.147] has joined #wesnoth-dev
20090328 11:58:20< Ivanovic> hi RainCT
20090328 11:58:26< Velory> Hi everyone
20090328 11:58:41< Velory> boucman: i finished my improvement avout healing 
20090328 11:58:50< Velory> s/avout/about
20090328 11:59:04< loonycyborg> env.WesnothProgram("mapgen-test", [mapgen-test_sources, libwesnoth_core,...])
20090328 11:59:09< loonycyborg> That's it.
20090328 11:59:35< boucman> Velory: good, tested ? submited ?
20090328 11:59:40-!- ABCD [n=ABCD@wikipedia/ABCD] has quit ["No Ping reply in 30 seconds."]
20090328 11:59:44< Velory> tested yesterday
20090328 11:59:47< Velory> now i can submit
20090328 11:59:53< Velory> i created 2 candidate move
20090328 11:59:58< loonycyborg> env.WesnothProgram("mapgen-test", [mapgen-test_sources, libwesnoth_core,...], CPPDEFINES = ["TEST_MAPGEN"])
20090328 12:00:01< Velory> one for healer one for unit that will be healer
20090328 12:00:07-!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090328 12:00:11< Velory> s/will be healer/will be healed
20090328 12:00:41< RainCT> I'm looking into participating with GSoC and the stats.wesnoth.org idea looks appealing :)
20090328 12:00:52< loonycyborg> Ivanovic: I really wish that make hacks like cmake and autotools became history.
20090328 12:01:33< euschn> loonycyborg: thanks, I will try that too
20090328 12:02:06< Ivanovic> loonycyborg: hmm, personally i think it "has to work"
20090328 12:02:10< Ivanovic> this is the case with cmake
20090328 12:02:27< Ivanovic> beside this it offers some "base functionality" that you will most likely always have
20090328 12:02:49< loonycyborg> Ivanovic: The same can be implemented based on scons.
20090328 12:02:55< Ivanovic> yes, it can
20090328 12:03:19< Ivanovic> but as you see several projects failed to do so (at least regarding mr_bones_ as gentoo games packager)
20090328 12:03:23< loonycyborg> The fact that they've decided to invent their own language instead is NIH.
20090328 12:05:27-!- JW1 [n=X@c-71-57-85-168.hsd1.il.comcast.net] has quit []
20090328 12:07:03< Velory> boucman: submitted https://gna.org/patch/index.php?1145 
20090328 12:07:07-!- Crab_ [n=crab_@c.218.175.a530.sta.adsl.cyfra.net] has joined #wesnoth-dev
20090328 12:07:15< Crab_> hi all
20090328 12:08:40< RainCT> So... to whom should I talk about this? :)
20090328 12:09:05< Ivanovic> you should just talk about your ideas about this matter in here
20090328 12:09:08< Ivanovic> someone will react
20090328 12:09:09< Ivanovic> ;)
20090328 12:09:11< Ivanovic> hi Crab_
20090328 12:10:48< Velory> hi Crab_
20090328 12:12:09< RainCT> oh, lots of people already applied for that idea
20090328 12:19:15< Ivanovic> RainCT: there are always many people applying
20090328 12:19:24< loonycyborg> euschn: Actually, if you override CPPDEFINES for some sources this may may cause conflicts with other targets which depend on objects without those overrides.
20090328 12:19:25< Ivanovic> last year we had almost 30 applicants
20090328 12:20:30< ilor> boucman around? I think I have the units-not-under-water issue here as well, maybe I can help debug?
20090328 12:20:42< boucman> yay, great :)
20090328 12:21:05< euschn> loonycyborg: ok
20090328 12:21:07< ilor> a skeleton archer in shallow water should be in said water not on it, I assume?
20090328 12:21:55< boucman> i'm semi afk, but could you add a printf to see the value of current_data.submerge in unit_frame::redraw ?
20090328 12:22:01< boucman> I'll come back to you a bit later
20090328 12:22:18< euschn> Ivanovic: I sort of got it to run by cmake, but I had the following error initially: http://wesnoth.pastebin.com/m23155318
20090328 12:22:59< loonycyborg> euschn: So you'll probably need to do something like this: mapgen_sources = Object("mapgen.cpp", CPPDEFINES = ["MAPGEN_TEST"])
20090328 12:23:04< euschn> I got rid of it by commenting that line in wesconfig.h, but I cannot keep it like that of course
20090328 12:23:13< Ivanovic> euschn: run svn up
20090328 12:23:19< Ivanovic> this one was fixed yesterday
20090328 12:23:43< loonycyborg> env.WesnothProgram("mapgen-test", [mapgen-sources, libwesnoth_core,...], True, CPPDEFINES = ["TEST_MAPGEN"])
20090328 12:23:55< loonycyborg> env.WesnothProgram("mapgen-test", [mapgen-sources, libwesnoth_core,...], True)
20090328 12:24:03< euschn> Ivanovic: thanks
20090328 12:24:22< loonycyborg> So override the define only for the object.
20090328 12:24:55-!- fendrin [n=fabi@wesnoth/developer/fendrin] has quit [Remote closed the connection]
20090328 12:25:28< loonycyborg> And for it not to conflict with objects without that symbol add a prefix to it: mapgen_sources = Object("mapgen.cpp", CPPDEFINES = ["MAPGEN_TEST"], OBJPREFX = "mapgen_test")
20090328 12:26:08< euschn> loonycyborg: thanks, I will try that
20090328 12:38:04< RainCT> Well, I probably wouldn't learn much with this anyway (I already have quite some web development experience) so I'll just leave it for something else. Perhaps next year I'll know more C and I'll be able to apply for something more challenging related to the game itself :)
20090328 12:39:32< Ivanovic> it is all your choice
20090328 12:39:47< Ivanovic> though wesnoth is not C, it is C++
20090328 12:39:48< Ivanovic> ;)
20090328 12:40:17< RainCT> Well, as I don't know much about them I find C and C++ to be pretty much the same :P
20090328 12:40:32< loonycyborg> euschn: You know, you can also try experimenting with autotools :)
20090328 12:40:56< euschn> loonycyborg: haha, I think Im good with 2 buildtools
20090328 12:42:06< Ivanovic> RainCT: hmm, there is quite some difference in the way you can use the langs
20090328 12:42:16< Ivanovic> and C is missing the lovely stuff STL provides
20090328 12:42:18< Ivanovic> ;)
20090328 12:45:58< ilor> boucman: submerge stays at 0
20090328 12:46:15< boucman> wow that's weird
20090328 12:46:32< boucman> ilor: is this a "clean" checkout (no addons)
20090328 12:46:46< ilor> no addons, yes
20090328 12:46:51< silene> ilor, boucman: have you seen my post?
20090328 12:47:13< RainCT> Hehe. By the way, about the name generation improvement stuff you've on the wiki, perhaps there's something you can use at http://sourceforge.net/projects/libnamegen (it's pretty much dead -I've heard nothing about it for two years- but it has quite some names in different languages -of people, rivers, cities, mountains, etc.-)
20090328 12:47:24< ilor> boucman: thouigh wait a sec I'll doublecheck
20090328 12:47:31< ilor> silene: I have translations off
20090328 12:48:07< silene> ilor: sure, but i would still be interested in what your locale variables are
20090328 12:48:20< ilor> boucman: yeah I removed my userdata and it's the same
20090328 12:48:55< ilor> silene: which ones? this is on windows, remember
20090328 12:49:38< ilor> there's one unit in the test scenario that gets submerge=0.1, it looks like it's the elvish shyde on 17,11 (on grass)
20090328 12:49:46-!- Polarina [n=polarina@wesnoth/translator/Polarina] has joined #wesnoth-dev
20090328 12:49:53< ilor> 0.01 sorry
20090328 12:50:08< silene> ilor: its unix name is LC_NUMERIC; i don't know what the windows equivalent is, but i know there is one
20090328 12:50:26< boucman> ilor: hmm
20090328 12:50:52< boucman> ilor: you do have units on water, of course (preferably move them around in water to make sure they are the ones you see moving
20090328 12:51:16< boucman> I usually print current_data.image too (it's a string) to know what unit printed the submerge)
20090328 12:51:44< boucman> silene: reading it now
20090328 12:52:08< ilor> silene: if the shyde has a fractional submerge wouldn't that suggest they are working?
20090328 12:52:22< boucman> silene: ok, well, we'll continue investigating both ways anyway
20090328 12:52:33< ilor> silene: anyway my locale regardless of the actual name is something along the lines of pl_PL
20090328 12:53:02 * boucman switches locak
20090328 12:53:21< ilor> boucman: can I just << the current_data.image or must it be some member of the image?
20090328 12:53:27< boucman> I can't select polish :(
20090328 12:53:39< boucman> ilor: that would work too I guess
20090328 12:53:53< ilor> boucman: it doesn't :/
20090328 12:54:30< boucman> image_loc.get_filename() 
20090328 12:54:32< boucman> sorry
20090328 12:54:43< ilor> just found it :)
20090328 12:55:10< silene> boucman: if it's a locale issue, i'm not sure you will experience it on linux, otherwise we would already have (at least i, since i play in french)
20090328 12:55:49< boucman> silene: so do I :)
20090328 12:57:03< silene> ilor: the fact that the shyde has a submerge of 0.01 does not look to me like it's working properly; i mean, why would a unit have 1% submerge?
20090328 12:57:36< euschn> loonycyborg: I still have trouble to get scons to work: the line mapgen-test_sources = env.Object("mapgen.cpp", CPPDEFINES = ["TEST_MAPGEN"], OBJPREFX = "mapgen_test") gives a SyntaxErr: can't assign to operator
20090328 12:58:27< loonycyborg> *OBJPREFIX
20090328 12:58:40< loonycyborg> mapgen-test_sources is not a valid python name
20090328 12:58:45< loonycyborg> remove "-"
20090328 12:59:36< ilor> boucman: I can definitely confirm that the unit which is moved from one shallow water tile to another has zero submerge all the time
20090328 12:59:54< boucman> ok, let's go a bit higher in the stack
20090328 13:00:17< loonycyborg> It should be built with "scons target-name" where target-name is name that was passed to env.WesnothProgram
20090328 13:00:58< boucman> ok, I have an idea where the 0.01 comes from, it's a way for the engine to recognise flying units
20090328 13:01:07< boucman> so it's normal, since the shyde is flying
20090328 13:01:30< boucman> please add a printf at unit.cpp:1825
20090328 13:01:39< boucman> check params.submerge
20090328 13:03:15< ilor> boucman: rebuilding...
20090328 13:06:09< ilor> boucman: zero there
20090328 13:08:22< boucman> ouch
20090328 13:08:42< boucman> ok, so the anim engine is given bad parameters from the terrain layer
20090328 13:09:04< boucman> we need to check the terrain WML code, but this looks a lot like what silene described
20090328 13:09:12-!- Elvish_Pillage2 [n=eli@66.189.40.122] has joined #wesnoth-dev
20090328 13:09:46< boucman> ilor: at this point we're out of my expertise, I don't really know where to look, i'll tell you if I find some place to instrument, but you could look too
20090328 13:10:20< ilor> I get it the zero returned is wrong, I'll try to eork from there
20090328 13:10:52< boucman> it's the percentage of the unit that should be made semi-transparent
20090328 13:10:52-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090328 13:11:17< boucman> seems to be set at terrain.cpp line 62
20090328 13:11:40< boucman> unless it's a terrain with overlay in which case it's line 170
20090328 13:18:08< euschn> loonycyborg: sorry for being annoying, but I still fail at this. I now get "File <string>, line 1         SyntaxError: can't assign to operator (<string>, line 1)".
20090328 13:18:20< euschn> i just added these lines to src/SConscript: http://wesnoth.pastebin.com/m21187ad5
20090328 13:20:39< loonycyborg> You can't use "-" in the name too.
20090328 13:21:25< euschn> ack, i shouldve noticed that
20090328 13:21:44< ilor> silene, boucman: looks like a locale issue after all. atof is locale-dependant and converts "0.4" to 0
20090328 13:22:56< boucman> because in your local, the dot is not the decimal symbol...
20090328 13:22:58< boucman> well done
20090328 13:23:11< boucman> is there a local independant equivalent of atof ?
20090328 13:25:26-!- Elvish_Pillager [n=eli@66-189-40-122.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090328 13:27:59< euschn> loonycyborg: finally its working, thanks again
20090328 13:28:07< loonycyborg> np
20090328 13:29:25< loonycyborg> Ask questions to silene if you change your mind about trying autotools :P
20090328 13:29:47-!- IneQuation|Away [i=1286@2001:470:1f09:409:0:0:0:1286] has quit [Remote closed the connection]
20090328 13:29:52< silene> ilor: now that you are finally ready to test locale-related things, could you put some std::cout << std::setlocale(LC_ALL, NULL) << '\n'; in init_locale so that we now know your values are?
20090328 13:33:20< Crab_> boucman: we need more then 'locale independence', if we want, for example, to read configs from old savegames without quirks ( there is a number of atof's in team.cpp ), and old saves can be, of course, locale-dependent.
20090328 13:33:42< boucman> :(
20090328 13:33:45< boucman> indeed
20090328 13:33:50< boucman> we need to clean all atof I guess
20090328 13:34:41< ilor> silene: stderr has 20090328 13:29:58 info general: locale could not be determined; defaulting to system locale
20090328 13:34:41< ilor> 20090328 13:29:59 debug general: set locale to '' result: 'Polish_Poland.1250'
20090328 13:34:41< Crab_> and if you want to make 1.6.1 savegames open ok in 1.6a , it gets even better
20090328 13:36:00< silene> ilor: could you try the other variables too. in particular LC_NUMERIC?
20090328 13:36:20< ilor> silene: the cout in init_locale returned English_United States.1252 though
20090328 13:37:55< silene> Crab_: no, they are not supposed to be locale-dependent; if they are, it's a bug that should be fixed; we shouldn't bother writing an input function that can cope with all the locales of the world
20090328 13:38:22< ilor> silene: same for LC_NUMERIC, its english in init_locale
20090328 13:38:34< Crab_> silene, ok, I agree. but it is  worth checking.
20090328 13:39:56< silene> ilor: could you also add debug dumps in wesnoth_setlocale? thanks
20090328 13:40:22< ilor> silene: it's trying to set the locale to a '' locale
20090328 13:41:11< silene> ilor: the empty locale is the system locale
20090328 13:41:41-!- molgrum [n=molgrum@c83-249-53-163.bredband.comhem.se] has joined #wesnoth-dev
20090328 13:41:48< ilor> silene: yes. so it gets set to Polish in my case which is not what we want
20090328 13:42:44< silene> what is the category?
20090328 13:44:09< ilor> 1729 and then 1, which is LC_MESSAGES and LC_ALL I think
20090328 13:44:53< ilor> no sorry it's LC_MESSAGES and LC_COLLATE from the two calls in set_language
20090328 13:47:27< silene> is lc_numeric still set to english after the calls to wesnoth_setlocale?
20090328 13:50:55< ilor> silene: it's Polish
20090328 13:51:51< silene> was it changed by the function or was it already wrong before the function is called?
20090328 13:52:18< ilor> uhh. there's a bit of code in the win32-specific part of wesnoth_setlocale that makes is set LC_ALL when it's called with LC_MESSAGES
20090328 13:52:33< ilor> silene: language.cpp:222
20090328 13:54:07< silene> the culprit is "cycholka 2008-06-24"
20090328 13:54:46< ilor> apparently there's no LC_MESSAGES in windows
20090328 13:56:02< silene> how come it worked before? native language support wasn't added last year on windows, wesnoth has worked fine for years
20090328 13:58:36< ilor> I've no idea. Maybe it worked but all the locale-specific stuff like number formatting was always in english
20090328 13:59:40< silene> yes, number formatting was purposely in english, in order to avoid the submerge bug, the ai bug, and several similar ones
20090328 14:01:14< silene> ilor: you could try replacing lc_all by lc_ctype; i just hope we are not relying to much on the length of the strings being deterministic
20090328 14:01:34< ilor> silene: I just tried ctype, submerging works
20090328 14:02:07-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090328 14:02:25-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Read error: 104 (Connection reset by peer)]
20090328 14:02:34< ilor> the resulting locale is polish for collate and ctype and english for numeric, time and monetary
20090328 14:02:36< silene> that's what i was expecting; i just worry that it introduces much more vicious bugs, like memory corruption and the like
20090328 14:02:47-!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090328 14:03:00< ilor> why'd it introduce bugs?
20090328 14:04:05< silene> ilor: because the length of strings may be off, depending on how we queried the value; so we may be accessing past its end if we wrongly think it is longer (due tu polish locale) than it really is (due to utf8) for instance
20090328 14:05:15< silene> there is just so many ways it could go wrong; let's hope
20090328 14:07:18-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Client Quit]
20090328 14:07:23-!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090328 14:08:21-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Client Quit]
20090328 14:08:41-!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090328 14:08:55< silene> ilor: i just checked, the old code was simply setting the LANG environment variable on windows
20090328 14:09:15< silene> unless we find better, i suggest we just revert commit 27471 to get back to the old working behavior
20090328 14:10:30-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Read error: 104 (Connection reset by peer)]
20090328 14:10:50-!- elias [n=allefant@allegro/developer/allefant] has joined #wesnoth-dev
20090328 14:11:32< ilor> silene: the pre-27471 seems to have called a setlocale(LC_MESSAGES) which I thing produces a warning on windows
20090328 14:11:34-!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090328 14:11:58< Ivanovic> yeah, there you get some warnings at startup, but those are at least not breaking things...
20090328 14:11:59< Ivanovic> ;)
20090328 14:12:44< silene> ilor: we could just return early when the category is LC_MESSAGES; as a matter of fact, the LANG variable should only be set when the category is LC_MESSAGES
20090328 14:13:25< ilor> so what about if lc_messages return; in that bit of code?
20090328 14:14:15-!- Sparks_ [n=Sparkste@78-105-212-150.zone3.bethere.co.uk] has joined #wesnoth-dev
20090328 14:14:30< Sparks_> Hello guys! :)
20090328 14:14:41< Ivanovic> hi Sparks_
20090328 14:15:02< Velory> Hi sparks
20090328 14:15:55< Sparks_> I haven't got much time to talk right now..  but I have put a substantial amount of info on my wiki for GSoC..  If anyone would like to take a quick look and comment/ask qns  about it I would be very grateful
20090328 14:16:30< Sparks_> If you could put them under the 'COMMENTS' section that would be great.
20090328 14:17:13< Sparks_> I have to to goto work now.. :(
20090328 14:17:47< Sparks_> But next week I should be able to get on here a bit more often :)
20090328 14:18:05< Sparks_> Velory:  I hope your proposal is goin well to :)
20090328 14:18:15< Velory> Sparks_: Yes hope :)
20090328 14:18:30< Velory> Sparks_: Same wishes to you
20090328 14:18:33< Sparks_> :0
20090328 14:18:35< Sparks_> :0
20090328 14:18:38< Sparks_> :) oops
20090328 14:18:40< Velory> :O ? :D
20090328 14:18:55< Sparks_> kept typing wrong..  anyways  see you later
20090328 14:18:58< Sparks_> ..:)
20090328 14:19:18-!- Sparks_ [n=Sparkste@78-105-212-150.zone3.bethere.co.uk] has quit ["Leaving"]
20090328 14:19:23< silene> ilor: do you are talking with respect to the current code or the old code?
20090328 14:19:50< ilor> silene: the current code. it does look a bit cleaner to me than the old one
20090328 14:19:54< silene> ouch, my english gets worse and worse with time
20090328 14:21:21-!- Ambushsabre [n=ambushsa@c-75-67-216-121.hsd1.nh.comcast.net] has joined #wesnoth-dev
20090328 14:21:38< silene> ilor: then also get rid of all the LC_ALL settings; they don't make sense; and please conditionalize setting the LANG variables so that they are only modified for LC_MESSAGES
20090328 14:22:36< silene> and now that i look at the code, the issue also exists with beos and apple, wth?!
20090328 14:25:17< Ivanovic> how lovely...
20090328 14:26:43< ilor> ugh, those are from a revision around 4000 by sirp...
20090328 14:27:24< ilor> they only set the evniroment vars which probably didn't break it
20090328 14:27:25< Ivanovic> silene: not sure about "only LC_MESSAGES"
20090328 14:27:32< Ivanovic> what about the date in the load dialog?
20090328 14:27:48< Ivanovic> so it would be LC_DATE, too
20090328 14:27:53< ilor> Ivanovic: the date seems to be handled in the dialog by setting LC_DATE there
20090328 14:28:15< Ivanovic> ilor: and windows understands this?
20090328 14:28:16< Ivanovic> ;)
20090328 14:28:40< ilor> apparently it does, that revision by cycholka cleaned up something there
20090328 14:30:45< silene> ilor: it depends, if it works like on windows, it would (the reason why it works on windows is precisely because we set the LANG variable)
20090328 14:31:43< ilor> silene: the apple/beos code doesn't seem to be calling setlocale(LC_ALL), and probably doesn't get the lovely atof issues
20090328 14:34:35< silene> i hope, but just to be on the safe side, i will remove the games on the LC_ALL env var
20090328 14:38:34< ilor> I'll just check if translations still work and I
20090328 14:38:40< ilor> and I'll commit
20090328 14:38:47< ilor> and my enter button is far to close to '
20090328 14:40:22< CIA-30> ilor * r34210 /trunk/src/language.cpp: 
20090328 14:40:22< CIA-30> Do not call setlocale with LC_ALL on windows when trying to set the unavailable
20090328 14:40:22< CIA-30> on win32 LC_MESSAGES, instead just return. Also only set LANG env var when
20090328 14:40:22< CIA-30> setting LC_MESSAGES and never set the LC_ALL var. This fixes the units not
20090328 14:40:22< CIA-30> submerging bug and possibly more issues as floats will now load correctly
20090328 14:40:25< CIA-30> regardless of locale.
20090328 14:40:49< ilor> silene: ^
20090328 14:45:02< silene> ilor: great
20090328 14:45:18< boucman> congratulation for finding and correcting that bug :)
20090328 14:45:25< silene> let's test it for a bit and then commit it to 1.6
20090328 14:45:43< Velory> boucman: do u have a time for looking my improvement :)
20090328 14:45:57< boucman> I already had a look at your healing patch today
20090328 14:46:01< Velory> Oh ok
20090328 14:46:03< boucman> I posted comments there
20090328 14:46:12< Velory> ok i didnt saw
20090328 14:48:24< Velory> i dont understand this one :you should do all moves in a single candidate move to avoid someone "stealing" your wounded before you get them
20090328 14:49:44< boucman> well, you did two formulas, you probably do just one, by moving both the healer and the other one in a single move formula
20090328 14:50:49< Velory> but i can do just one move in one formula right?
20090328 14:51:03< boucman> no, I think you can return a list of moves
20090328 14:52:32< Velory> hmm so i can do 2 or 3 move in one formula ok then
20090328 14:52:33-!- Sirp [n=me@pool-173-74-2-179.dllstx.fios.verizon.net] has joined #wesnoth-dev
20090328 14:54:56< boucman> Velory: test it, I'm not 100% sure, but I think you can
20090328 14:55:50< Velory> boucman: And check will return the FIRST unit that is a potential candidate. however we never check for other units ,, yes but in unit_control formula other units are going to near of healer too so all units that need heal in area going to near of healer
20090328 14:56:30< boucman> yes, but if the first unit is not good, the healer won't move, even if the second unit is good and needs healing
20090328 14:56:59< boucman> you should look for the "best" unit to heal, right now you're looking for the first
20090328 14:57:14< boucman> or the position where the most units can reach you
20090328 14:57:41< CIA-30> silene * r34211 /trunk/src/language.cpp: Removed modification of the LC_ALL variable and conditionalized setting the LANG variable.
20090328 14:57:54< Velory> boucman: this find returns unit if there is unit that need heal so what is the if the first unit is not good?
20090328 14:58:17< boucman> suppose ther are more than one unit that needs healing
20090328 14:58:18< Velory> find returns first units if there is unit 
20090328 14:58:24< Velory> yes
20090328 14:58:27< boucman> and the first in the list regenerates
20090328 14:58:33< Velory> hmm
20090328 14:58:36< Velory> i see ok
20090328 14:58:41< boucman> :)
20090328 14:58:43< Velory> :D
20090328 14:58:56< Velory> so i can check regenerate before that :p
20090328 14:59:01< boucman> no
20090328 14:59:05< Velory> yes i know
20090328 14:59:06< Velory> i cant :D
20090328 14:59:11< Velory> check is returning same uni,t
20090328 14:59:30< boucman> you could filter for not regenerating inside the find
20090328 14:59:41< boucman> but your find would still return the first unit in the list
20090328 14:59:52< boucman> you want to look for the best possible unit, not just the first one
20090328 15:00:14< Velory> i want to look first one in possible units
20090328 15:00:21< Velory> though
20090328 15:00:23< boucman> no you don't :P
20090328 15:00:33< boucman> why do you want the first one ?
20090328 15:00:40< Velory> hmm
20090328 15:00:41< boucman> what's so special about the first one
20090328 15:00:46< Velory> i love him :)
20090328 15:00:56< boucman> well, I think you understand what I mean
20090328 15:01:08< boucman> "first one" is good for easy testing, but not good enough for commit
20090328 15:01:17< Velory> but
20090328 15:01:29< Velory> in unit_control other ones also coming near to healer
20090328 15:02:07< boucman> suppose that the first one is sompletely to the left (because it's moving to a village for healing)
20090328 15:02:19< boucman> and all others are on the right (fighting)
20090328 15:02:31< Velory> hmm
20090328 15:02:35< boucman> you healer will move away from most unit to go and heal the lonely one
20090328 15:02:48< Velory> yes ure right
20090328 15:02:54-!- anakayub [n=anakayub@203.82.79.2] has joined #wesnoth-dev
20090328 15:04:07< Velory> but in the end i must choose one unit
20090328 15:04:57< boucman> yes
20090328 15:05:27< Velory> so u say find best one not first one righT?
20090328 15:05:42< boucman> right
20090328 15:05:52< Velory> hmm i think i can handle it with choose maybe
20090328 15:06:04< happygrue> is it possible to choose by location instead?
20090328 15:06:27< happygrue> I don't want to sound like an idiot, but don't you want the location where most healing occurs?
20090328 15:06:44< boucman> yes, choose is probably the way to go
20090328 15:07:13< boucman> happygrue: it's a bit more complicated, you want the location that most wounded units can choose
20090328 15:07:20< boucman> but yes, that would be the best
20090328 15:07:23< happygrue> I see
20090328 15:10:08-!- Sirp [n=me@pool-173-74-2-179.dllstx.fios.verizon.net] has quit [Read error: 111 (Connection refused)]
20090328 15:12:51-!- happygrue [n=George@wesnoth/developer/wintermute] has quit ["http://xkcd.com/91/"]
20090328 15:13:17< Velory> boucman: No i cant do 2,3 move in one formula
20090328 15:13:36-!- jmunro[a] is now known as jmunro
20090328 15:15:47< boucman> oh :(
20090328 15:18:15< jmunro> afternoon guys
20090328 15:18:27< jmunro> anyone catch the F1 qualifying ;)
20090328 15:18:40-!- anakayub_ [n=anakayub@203.92.154.39] has joined #wesnoth-dev
20090328 15:19:11-!- Sirp [n=me@wesnoth/developer/dave] has joined #wesnoth-dev
20090328 15:19:36-!- YogiHH [n=chatzill@d051064.adsl.hansenet.de] has joined #wesnoth-dev
20090328 15:20:05< YogiHH> hello everyone
20090328 15:20:09< Crab_> hello, YogiHH
20090328 15:20:15< Velory> Hi yogihh
20090328 15:21:45< boucman> hey YogiHH
20090328 15:22:51< Dragonking> hi everyone
20090328 15:22:53< Crab_> YogiHH, I'll post my work-in-progress patches today. Regarding team.cpp -> I wanted my changes (at this point) to not impact savegames at all, so I tried to make AI parameter handling to be bug-for-bug compatible with the current behavior.
20090328 15:22:58< Velory> Hi dragonking
20090328 15:22:58< Crab_> hi Dragonking
20090328 15:23:17< boucman> hey Dragonking
20090328 15:23:29< Dragonking> "Velory boucman: No i cant do 2,3 move in one formula"  <- could you elaborate please what's the problem? I'm a bit out of sync with discussion :)
20090328 15:23:44< YogiHH> Crab_: ok, cool
20090328 15:23:47< Velory> Dragonking: ok
20090328 15:23:48< boucman> we are considering a "healer" formula
20090328 15:24:00< Velory> Dragonking: i created 2 candidate move for healing improvement
20090328 15:24:04< boucman> in that case we want to move both the healer and wounded
20090328 15:24:21< Velory> Dragonking: yes in first formula im moving healer to units
20090328 15:24:33< Velory> and in other formula im moving others to healer
20090328 15:24:39< boucman> but we want to do it in one go, so nobody "steals" our unit between two evaluation rounds
20090328 15:25:07< boucman> so the idea (I thought it worked) was to return more than one move in the move part, so that they all be executed one after the other
20090328 15:25:30< Dragonking> then formula=[ calcualte_healer_move(), calculate_wounded_move() ]  where calcualte_healer_move() returns move(X,Y) for healer, and simmilar for wounded unit.
20090328 15:25:31< Velory> and boucman with 2 candidate move im get out of next unit problem too
20090328 15:25:52< boucman> Dragonking: yes, that's the idea
20090328 15:26:19< Dragonking> And what's the problem with that?
20090328 15:26:47< boucman> Dragonking: according to Velory it doesn't work, we can only return a single move
20090328 15:26:54< boucman> but it would be a good thing :)
20090328 15:26:56< Dragonking> No, it works.
20090328 15:27:04< Velory> hmm :(
20090328 15:27:28< Velory> so im silly person :D
20090328 15:28:12< Velory> so in calculate_healer_move() there wil be one move statement and other one there will be one righT?
20090328 15:28:32-!- Mad_Troll [n=madtroll@77-97-206-48.cable.ubr15.sgyl.blueyonder.co.uk] has quit [Read error: 104 (Connection reset by peer)]
20090328 15:28:42< boucman> Velory: not sure I understand
20090328 15:28:53< Velory> i dont understand quiet formula=[ calcualte_healer_move(), calculate_wounded_move() ] 
20090328 15:29:41< Dragonking> Result would be for example [ move(loc(10,20), loc(10,21) ), move( loc(9,18), loc(11,21) ) ]
20090328 15:29:56< Dragonking> First would be healer movement, second - unit movement
20090328 15:29:56-!- anakayub__ [n=anakayub@203.82.79.2] has joined #wesnoth-dev
20090328 15:30:10< Velory> ok
20090328 15:30:17< Velory> i tried to say same thing:)
20090328 15:30:39< Velory> hmm
20090328 15:30:48< Velory> but in evaluation i will be die :)
20090328 15:32:45< Dragonking> boucman: I've been really busy tis week, many tests, still 3 to go on monday, tuesday, wednesday.... I'll write my proposal today/tomorrow (it will be quite different than other AI ones I think)  and will work on some things after my uni-mess is over...
20090328 15:33:38< Dragonking> Actually I am so busy right now mainly because I ignored some uni-things during the last 2 weeks before 1.6 releasing, whe I've been hunting for bugs here :)
20090328 15:34:14< boucman> silene: could you backport your language fix to 1.6 or is it too dangerous
20090328 15:34:21< YogiHH> Dragonking: Well, then don't worry, that's ok ;-)
20090328 15:34:44< boucman> Dragonking: that's ok, we know about that
20090328 15:34:52< Dragonking> Great :)
20090328 15:36:38< ilor> boucman: I thinks silene wanted to first test the changes for a bit before backproting them
20090328 15:37:12< ilor> s/thinks/think
20090328 15:37:47-!- anakayub [n=anakayub@203.82.79.2] has quit [Read error: 110 (Connection timed out)]
20090328 15:39:37< boucman> ok, fair enough
20090328 15:42:14< CIA-30> boucman * r34212 /trunk/src/unit_frame.cpp: avoid useless search of image in cache when invalidating a unit that fits in the hex
20090328 15:43:16-!- anakayub_ [n=anakayub@203.92.154.39] has quit [Read error: 110 (Connection timed out)]
20090328 15:43:29< Crab_> boucman, btw, do you take into accounts those units which are displayed elevated (when standing in castle, for example), when checking 'if it fits in hex' ?
20090328 15:43:36< Crab_> s/accounts/account
20090328 15:45:50< boucman> yes
20090328 15:46:03< Crab_> thats good )))
20090328 15:46:08< boucman> "fit in hex" is a parameter of the image that tells if the unit fits in the hex shape
20090328 15:46:20< boucman> however we also control that it's correctly positionned when invalidating
20090328 15:47:25< Velory> boucman: so when choosing the best i must look unit's location if they can reach that unit right?
20090328 15:47:47< Velory> where rhey = other units that need healing :)
20090328 15:47:58< Velory> :p
20090328 15:48:18< boucman> yes
20090328 15:48:31< Velory> i started using variables in normal conversations :D
20090328 16:01:56< Crab_> Velory: try using WML in conversations instead :) you'll be able to speak in any language - for it supports gettext internationalization )
20090328 16:03:00< YogiHH> Sirp, you there?
20090328 16:04:16< Sirp> YogiHH: in a manner of speaking; what's up?
20090328 16:04:48< YogiHH> Is there a specific reason why gamestatus (TOD, number of turns) and game_state are separated from each other?
20090328 16:06:24< YogiHH> Sirp: Is there a specific reason why gamestatus (TOD, number of turns) and game_state are separated from each other?
20090328 16:07:37< Sirp> YogiHH: I believe that game_state is data that carries over from one scenario to another, while gamestatus is only relevant during a scenario
20090328 16:08:11< YogiHH> Sirp: ah, ok. Thanks
20090328 16:08:17< Sirp> and so the data is somewhat conceptually different. That's not to say that this couldn't be redesigned.
20090328 16:08:35 * YogiHH already thinks about redesigning it ;-)
20090328 16:09:06< Sirp> well that is what I figured. :)
20090328 16:09:11< Sirp> you could start with some better names
20090328 16:09:29< YogiHH> yup
20090328 16:22:09-!- jmunro is now known as jmunro[a]
20090328 16:23:43-!- Gnutoo [n=gnutoo@host169-162-dynamic.44-79-r.retail.telecomitalia.it] has joined #wesnoth-dev
20090328 16:26:23< Gnutoo> hello,I'm compiling wesnoth-1.6a with the autotools...but I've a config.h problem...just to check what I understood: is config.h generated by the autotools and used to #define things during the compilation? so here's the real question: wesnoth's config.h conflicts with SDL_config.h so should the SDL_config.h be included while compiling wesnoth?
20090328 16:27:44< Polarina> Gnutoo: autotools compiling is obsolete and unsupported (that is, we will not assist you).
20090328 16:27:48< Polarina> Gnutoo: Use scons instead.
20090328 16:28:04< Gnutoo> ouch...can I use cmake?
20090328 16:28:10< YogiHH> Gnutoo: You might want to check yesterdays irc log, there was some issue with config.h
20090328 16:28:12< silene> Gnutoo: yes, it is generated by autoconf; what kind of conflict do you have?
20090328 16:28:17< loonycyborg> Polarina: Actually silene decided to reverse that decision.
20090328 16:28:24< Gnutoo> redefinitions
20090328 16:28:35-!- KnightR [n=anonymou@bb219-74-177-1.singnet.com.sg] has quit []
20090328 16:28:39< Gnutoo> I thought you wouldn't migrate to something else than autotools...
20090328 16:29:20< Gnutoo> http://rafb.net/p/92KIKz59.html
20090328 16:29:39< loonycyborg> Gnutoo: We tried, but failed to reach consensus as to which buildsystem to keep :)
20090328 16:30:12< silene> Gnutoo: which version of sdl is that? it was a bug in an old sdl
20090328 16:30:32< Gnutoo> libsdl-x11_1.2.11.bb
20090328 16:31:08< Polarina> Why not make a custom build-system? :P
20090328 16:31:10< Gnutoo> or maybe I shoud make 2 or 3 recipes? one for each build system?
20090328 16:31:15< Gnutoo> lol
20090328 16:32:43< loonycyborg> Polarina: Preferably scripted in WML :P
20090328 16:32:59< Polarina> loonycyborg: Of course! :D
20090328 16:33:32-!- allefant [n=allefant@allegro/developer/allefant] has joined #wesnoth-dev
20090328 16:33:39< Gnutoo> so both scons and cmake are supported but not the autotools?
20090328 16:34:00< Ivanovic> autotools will work as long as they work, but that's it
20090328 16:34:14< boucman> and they willwork in 1.6 of course
20090328 16:34:25 * Polarina bets Gentoo has USE flags on which build system to use.
20090328 16:34:26< Ivanovic> where cmake as well as scons are in the area of what you could call "well maintained"
20090328 16:34:33< Ivanovic> as in problems coming up will be dealt with
20090328 16:34:39< Ivanovic> Polarina: no, it has not
20090328 16:34:44< Polarina> Ivanovic: :(
20090328 16:35:16< Ivanovic> the offical maintainer of the games rep has had bad experiences with scons and thus switched to cmake for the "offial recipe"
20090328 16:35:24< Gnutoo> Polarina, gentoo uses cmake
20090328 16:35:36< Ivanovic> though in our rep loonycyborg uploaded a recipe for using portage with scons
20090328 16:36:04< loonycyborg> Ivanovic: Hopefully he'll switch to autotools so people could actually have working wesnoth :)
20090328 16:36:34< Gnutoo> i'll go for cmake then but I don't know it well...if there are problems I don't know if I will be able to resolve them without help...
20090328 16:37:17< loonycyborg> Gnutoo: You can stay with autotools too since silene decided to start maintaining it.
20090328 16:37:27< Gnutoo> ah ok thanks
20090328 16:37:30-!- elias [n=allefant@allegro/developer/allefant] has quit [Read error: 110 (Connection timed out)]
20090328 16:37:53-!- jmunro[a] is now known as jmunro
20090328 16:38:30< silene> Gnutoo: that being said, while i maintain it, i can hardly make it work with a broken sdl; so you would better use scons or cmake to work around the bug of your sdl
20090328 16:38:44< Gnutoo> ok thanks a lot
20090328 16:38:55< loonycyborg> scons and cmake generate config.h too :)
20090328 16:39:08< loonycyborg> Better fix sdl.
20090328 16:39:09< Crab_> boucman, YogiHH - I've posted my work-in-progress in ai redeployment here -  https://gna.org/patch/index.php?1146 - It touches a bit of ai.cpp, ai.hpp, ai_interface.hpp, formula_ai.cpp, menu_events.cpp, multiplayer_connect.cpp, playsingle_controller.cpp, team.cpp, team.hpp,  and I've added two new files (ai_manager.cpp ai_manager.hpp, 862 lines in new files). Basically, it is now  in the state "doesn't break anything, doesn't do anythi
20090328 16:39:25< Gnutoo> ok thanks...what's the minimum libsdl version that I should use?
20090328 16:39:46< YogiHH> Crab_: thanks
20090328 16:39:56< boucman> that's a nice subset :P
20090328 16:40:03< Gnutoo> gentoo says 1.2.7 but I have 1.2.11
20090328 16:40:11< Gnutoo> on openembedded
20090328 16:40:30-!- allefant is now known as elias
20090328 16:40:37< silene> Gnutoo: not sure, mine is 1.2.13, but i don't know if the bug is fixed in sdl or if was fixed in the debian package
20090328 16:40:38< Crab_> boucman, YogiHH: mostly minor tweaks in lookup code for AIs and AI parameters. team.cpp suffered most 
20090328 16:40:58< Gnutoo> silene, ok maybe I should try cmake
20090328 16:41:28< Crab_> there is, basically, std::stack<AI> for each AI. and it should work the same when there is only one AI in a stack.
20090328 16:42:02< Crab_> ai_manager.hpp is the interface for the AI manager. I commented it, to make everything in that interface clear.
20090328 16:42:18< loonycyborg> Gnutoo: You could as well try scons and tell me why gentoo packager had bad experiences with it.
20090328 16:43:02< Gnutoo> loonycyborg, I use gentoo on my laptop,desktop,and router, but I'm not making a package for gentoo but for openembedded
20090328 16:43:39< Gnutoo> loonycyborg, so maybe It will just work on openembedded...
20090328 16:43:43< silene> Crab_: all the ai_manager::instance() are a bit ugly; can't you make the public functions static instead?
20090328 16:43:55< Crab_> silene, yes, of course.
20090328 16:46:29< silene> why are there local variables with a _ suffix? do you intend to move them to class scope?
20090328 16:47:08< Crab_> silene, in which file ?
20090328 16:49:05< silene> team.cpp, the declarations marked by "this is a temporary variable" (which is quite obvious since they are at local scope, or did you mean something else?)
20090328 16:50:09< Crab_> silene, this is a work-in-progress. I intend to clean out that part of code (//START OF MESSY CODE -> //END OF MESSY CODE), and after that I'll rename those vars. those vars were previously class variables.
20090328 16:53:35-!- crimson_pingvin is now known as crimson_penguin
20090328 16:59:58< YogiHH> Crab_: does formula_ai comply to ai_interface?
20090328 17:00:35< Crab_> YogiHH: yes.
20090328 17:01:03< silene> Crab_: i don't understand the code of ai_manager::add_ai_for_team_from_param; what's the point of pushing an object on a stack that is getting destroyed?
20090328 17:02:57< Crab_> silene. thanks. I forgot to place some ref's there. will fix.
20090328 17:03:34-!- crimson_pingvin [n=ben@S0106000d88bbc764.vc.shawcable.net] has joined #wesnoth-dev
20090328 17:03:38< silene> Crab_: what happens if someone modifies the objects returned by ai_manager::get_active_ai_parameters_for_team?
20090328 17:05:12< silene> no, forget it
20090328 17:06:05< Crab_> btw, the old version has come comments on "should we allow the user to modify the config directly or not?"
20090328 17:07:40-!- elminse [n=elminsei@netblock-208-127-84-182.dslextreme.com] has quit [Read error: 110 (Connection timed out)]
20090328 17:09:48-!- grantwu [n=Zenneth@westquad-147025.reshall.umich.edu] has joined #wesnoth-dev
20090328 17:09:52< silene> Crab_: some style issues: please remove the tabulations at the start of the lines of ai_manager.cpp (your lines are already way long, no need to make it worse); also i would prefer if your multi-line doxygen comments put the description on the second line (the current comment of get_fallback_ai looks a bit dumb for instance), if you were avoiding using tabs inside the comments (just a single space is fine), and if you were to use @note for you
20090328 17:10:25-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit [Read error: 110 (Connection timed out)]
20090328 17:10:48< silene> for your notes instead of NOTE
20090328 17:11:19-!- elminse [n=elminsei@netblock-208-127-84-182.dslextreme.com] has joined #wesnoth-dev
20090328 17:12:54< Crab_> silene: thanks. I'll do that. Regarding the 'description on the second line' - like this ? http://wesnoth.pastebin.com/m66599e83
20090328 17:16:08< silene> yes, and you don't even need to put two spaces between the star and the text (that way you won't have spurious tabs there)
20090328 17:16:23< Crab_> ok
20090328 17:17:01< Velory> afk about 2 hours
20090328 17:17:10< silene> Crab_: remove_ai_for_team seems dubious; and for remove_all_ais_for_team, you could just empty the stack
20090328 17:17:29< silene> oups, sorry, scrap the first one
20090328 17:18:16< silene> (they were too similar, i didn't notice the difference)
20090328 17:19:45< Crab_> thanks for the second one 
20090328 17:20:12< Crab_> copypasting is evil )
20090328 17:22:00< silene> Crab_: please rewrite create_transient_ai the way you suggest it in the first note; it will indeed look much better
20090328 17:22:52< Crab_> 'the way you suggest it in the first note' ?
20090328 17:23:08< Crab_> aha
20090328 17:23:15< Crab_> got it
20090328 17:24:23-!- torghul [n=chatzill@78.172.155.22] has joined #wesnoth-dev
20090328 17:27:36< silene> Crab_: the point of std::make_pair is to avoid passing template parameters; so in get_or_create_ai_stack_for_team, just remove the "make_" part :-)
20090328 17:28:22< Crab_> silene, ok.  btw, note that get_available_ais() will be later rewritten to 'look for all .cfg files in a specific dir, and store the list of AIs constructed from those"
20090328 17:29:14< silene> ok, i was wondering what it was supposed to return; so you say they are not C++ AIs but user-defined AIs?
20090328 17:30:08< Crab_> silene, AI = 'core C++ part' plus 'config part' plus 'external .fai or lua or ... files'
20090328 17:30:23< silene> by the way, what is the function used for? user interface?
20090328 17:30:27< Crab_> yes
20090328 17:30:41< Crab_> in multiplayer_connect.cpp
20090328 17:31:20< Crab_> I want the user to be able to 'drop a AI config file into specific folder, launch wesnoth, and see this AI in a list of AIs to play against'
20090328 17:32:17< silene> it doesn't have to be done right away, but later it may be worth expanding a it the type, in particular with a t_string that would contain a description of the ai
20090328 17:33:41< Crab_> silene, it will need at least two strings: AI_file_location and AI_name_to_show_to_user. so, taking into account :author-description-etc-...,  I guess a helper config object or c++ class will be needed )
20090328 17:35:11< Crab_> basically, since the AI description will be parsed from WML (exactly as it does it now as part of scenario loading), it will be easy to get to it.
20090328 17:35:27< silene> yes, that's why i'm mentioning it now; even if you don't propagate the data up to multiplayer_connect.cpp, you can already think about the object/config
20090328 17:35:46< Crab_> silene: yes, I'm already thinking about it.
20090328 17:36:26-!- Mellar [n=brebbesv@ti400720a080-3037.bb.online.no] has joined #wesnoth-dev
20090328 17:37:58< silene> what's the difference between a param, an effective param, and a memory? at first glance, it seems like there is one too many
20090328 17:39:04< Crab_> basically, memory is a space for the AI to put its notes, which will be persisted during save/load
20090328 17:39:35< Crab_> ai parameters -is a stuff that someone (scenario editor, for example) told to the AI.
20090328 17:39:57< Crab_> there can be many different configs in ai parameters
20090328 17:40:03< Crab_> one for day, one for night, for example
20090328 17:40:11< Crab_> so it is a vector of configs
20090328 17:40:34< silene> sure, but it will have to be persistent too (?), so what makes it any different from the memory?
20090328 17:40:58< Crab_> then, in team::set_time_of_day, effective parameters are determined
20090328 17:41:38< silene> hum... i see
20090328 17:41:44< torghul> disco
20090328 17:41:53-!- torghul [n=chatzill@78.172.155.22] has quit ["ChatZilla 0.9.84 [Firefox 3.0.8/2009032609]"]
20090328 17:42:33< Crab_> silene: 1) 'what makes it any different from the memory?' thats how current implementation works. I'm not really sure that the current AI even uses it.   2) config is supposed to come 'from the outside' of the AI, memory is supposed to come from 'inside of the AI'.
20090328 17:45:15< silene> yes, that's the distinction i'm making too; but i feel like once designers start to script (part of) ais directly inside scenarios, the frontier between "inside" and "outside" will get quite thin
20090328 17:46:12< Crab_> the question is, IMO: is AI allowed to change its own config ?
20090328 17:46:30< Crab_> if it is allowed, then there's no effective distinction between config and memory
20090328 17:46:54-!- RainCT [n=RainCT@ubuntu/member/rainct] has quit ["bbl"]
20090328 17:47:38< Crab_> also (another solution), the memory can be put  inside that config vector, as an additional element.
20090328 17:48:29< silene> i wouldn't mind the ai changing its own config; for instance, once its leader is dead, the ai (or at least the thing controlling it) may want to adopt a whole new behavior (e.g. no need to grab villages)
20090328 17:48:50< Crab_> and there is another problem that I see - when effective config is determined, there is a lots of copying from config to effective config.
20090328 17:49:18< silene> anyway, it's getting late, i have to be afk for a while
20090328 17:49:35< Crab_> if the ai will try to store a large number of values inside its config (use it as a persistent cache, for example), that will be slow.
20090328 17:49:36< Crab_> ok
20090328 17:51:47< silene> that may just mean that the "effective" part is over-engineered: the ai itself may be the best place to decide which parameters to use (really afk now)
20090328 17:57:09< molgrum> in what files do i find the save game code?
20090328 17:57:24-!- cjhopman [n=chris@wesnoth/developer/cjhopman] has quit [Read error: 110 (Connection timed out)]
20090328 17:57:54< molgrum> i am mostly interested in the savegame reorganization idea
20090328 17:58:23< molgrum> so if i can take a look at that maybe i can see what i can do :)
20090328 18:02:56< Crab_> molgrum, loading game - start, for example, from : multiplayer_connect.cpp:1520        saving game - start, for example, from playcampaign.cpp:565
20090328 18:10:40-!- benetnash [n=benetnas@c137-122.icpnet.pl] has joined #wesnoth-dev
20090328 18:12:12-!- cristi_ceata [n=morpheus@p22.eregie.pub.ro] has joined #wesnoth-dev
20090328 18:12:21-!- cristi_ceata [n=morpheus@p22.eregie.pub.ro] has left #wesnoth-dev []
20090328 18:13:14< benetnash> Hi! Is anybody out there? I've already wrote it on forum, but I'd like to ask again: After dropping support of python scripts is there any possibility of writing custom multi-purpose AIs for wesnoth?
20090328 18:13:34-!- penguin_crimson [n=ben@S0106000d88bbc764.vc.shawcable.net] has joined #wesnoth-dev
20090328 18:14:01< boucman> benetnash: you can write it in C++ :P
20090328 18:14:15< Crab_> benetnash, or try http://www.wesnoth.org/wiki/FormulaAI
20090328 18:15:04-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev
20090328 18:15:45< benetnash> Crab_: I already searched wiki and documentation but I didn't find how to write and run stand alone Formula AI script
20090328 18:15:48-!- crimson_pingvin [n=ben@S0106000d88bbc764.vc.shawcable.net] has quit [Read error: 60 (Operation timed out)]
20090328 18:17:57< Crab_> benetnash: have you seen scenario-formula.cfg (it is in wesnoth data dir) ?
20090328 18:18:52< benetnash> for my master degree work I have to evolve AI using coevolutionh. Now when I'm in half way and python support was dropped I'm starting to regret, that I didn't chose something else (like ORTS)
20090328 18:19:39< Crab_> benetnash: can't you, then, just use the old wesnoth version for your master degree work?
20090328 18:20:22< benetnash> Crab_: yes, of course but then I cannot share my work with other players ;)
20090328 18:22:35< Crab_> benetnash. I understand that. but, there was a major security problem with python - there was a possibility to write exploits using python AI code. So, It was decided to drop python support in order to encourage AI/scenario sharing between players.
20090328 18:23:54< benetnash> I understand that. Is there any way to use FormulaAI script for custom multiplayer scenario? If so I'll switch to Formula AI.
20090328 18:23:59< boucman> there might be some possibilities using LUA, but the focus is more on WML extensions than AI programming
20090328 18:24:25-!- Zen_Clark [i=4b0beeb9@gateway/web/ajax/mibbit.com/x-7f3f043535c7d091] has joined #wesnoth-dev
20090328 18:25:05-!- torghul [n=chatzill@88.242.141.52] has joined #wesnoth-dev
20090328 18:25:23< boucman> there are various ways to use FAI code from custom MP scenarios (i.e setting a [side] to  use a custom formula, but FAI is a completely different lanquage from python, i'm not sure it would be possible to port your work...
20090328 18:29:35< benetnash> boucman: I evolve set of fuzzy rules controlling AI and the AI script is generetad from evolved genotype. Changeing phenotypic script will be quite easy. I know about [side] tags, but is there metod for choosing script using command line (with --algorithm<number>=value)?
20090328 18:30:11< boucman> nope, I don't think you can easily from the command line i'm afraid
20090328 18:30:18< benetnash> using command line would make running several instances of wesnoth parallel easier
20090328 18:30:36< boucman> yes, I understand why you would want to do that...
20090328 18:30:52< boucman> Crab_ might know a little more about that, he's in that area right now
20090328 18:31:08-!- Elvish_Pillager [n=eli@66-189-40-122.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090328 18:31:14< Crab_> benetnash, work to add such support is in progress (I'm currently working on making the ai hot-deployable and I'll going to add ai-testing command-line support)
20090328 18:31:58-!- turin [n=turin@168.215.250.1] has joined #wesnoth-dev
20090328 18:33:07< Crab_> benetnash: I expect it to be finished in 2-3 weeks. there will be possibility to write a stand-alone .cfg file containing the AI details, and there will be possibility to load the ai for the specific side from that file.
20090328 18:33:37< Crab_> the format of that .cfg file will correspond to current format of [side]
20090328 18:34:00< Crab_> ( http://www.wesnoth.org/wiki/AiWML )
20090328 18:36:39< Crab_> benetnash, but, you can achieve it even now by just some shell-scripting work.
20090328 18:36:47< benetnash> Ok, thx. I think I can still stick to AI fixed in custom map, I just have to change my framework.
20090328 18:38:33-!- Elvish_Pillage2 [n=eli@66.189.40.122] has quit [Read error: 145 (Connection timed out)]
20090328 18:38:46< Crab_> benetnash, you can make a custom scenario, which will be regenerated by your framework. And then - start wesnoth, telling it to load this map via a command-line switch (or via using the wesnoth unit tests, but those will need some C++ hacking)
20090328 18:43:55-!- jmunro is now known as jmunro[a]
20090328 18:44:16-!- lu_zero [n=lu_zero@gentoo/developer/lu-zero] has quit [Read error: 110 (Connection timed out)]
20090328 18:47:53-!- anakayub [n=anakayub@203.92.154.37] has joined #wesnoth-dev
20090328 18:50:01< euschn> I have a question on Easycoding/village renaming: where can I find the config file that stores the village_naming attribute?
20090328 18:50:43< benetnash> yes, I'll create scenario for each pair of algorithms I want to test. Now I have to read whole documentation of Formula AI to learn it ;)
20090328 18:50:51< benetnash> Thank you very much!
20090328 18:51:35< Crab_> benetnash; as FormulaAI is a work-in-progress, do not hesitate to return here and suggest improvements )
20090328 18:58:35< boucman> euschn: data/core/macros/names.cfg
20090328 18:58:51< euschn> boucman: thanks
20090328 18:59:43< euschn> I will also have to modify the string_table variable at some point (to add name patterns with more than one sylabus), is there any best practise on where and how to modify it?
20090328 19:00:02< boucman> not that I know of
20090328 19:00:27< euschn> ok
20090328 19:02:54-!- anakayub [n=anakayub@203.92.154.37] has quit [Read error: 145 (Connection timed out)]
20090328 19:06:15< grantwu> can someone point me in the direction of the files for the AI formula improvements listed in EasyCoding?
20090328 19:06:24-!- anakayub__ [n=anakayub@203.82.79.2] has quit [Read error: 110 (Connection timed out)]
20090328 19:08:25< grantwu> is it all contained within formula_ai.cpp?
20090328 19:09:03< Crab_> grantwu, ../data/scenario-formula.cfg will be a good start - there are examples of using a formula to make the AI consider moves.
20090328 19:09:27< grantwu> Crab: All right thanks a lot
20090328 19:09:51-!- noy [n=Noy@wesnoth/developer/noy] has quit []
20090328 19:10:09< Crab_>  formula_ai::make_candidate_moves() is also worth seeing, especially if you don't trust the wiki documentation :)
20090328 19:10:32< grantwu> lol okay
20090328 19:13:59< Crab_> grantwu, and also note that the 'Poisoning Improvement' is already done (by some extent), by Velory (and he's working on healing improvements at the moment)
20090328 19:14:14< grantwu> ruh roh, okay, I guess I'll think of something else
20090328 19:15:19< grantwu> Crab_: do you know if anybody is working on the Berserker?
20090328 19:15:36< boucman> grantwu: nobody
20090328 19:15:46< boucman> (afaik that is)
20090328 19:16:07< Crab_> grantwu: AFAIK, no one. You can also pick Illuminate - it's not even in that list, but is very similar to leadership :)
20090328 19:16:53< grantwu> Crab_: I'm guessing Illuminate is a support ability?
20090328 19:17:20< Crab_> yes. see the in-game description.
20090328 19:17:46< grantwu> all righty, thanks for the help everyone, i'll take a look at those
20090328 19:17:51< Crab_> for example, debug-create a human mage of light
20090328 19:18:14< euschn> boucman: names.cfg does contain the current village names, but I cannot add another config node by editing it, which is what I would like to do
20090328 19:18:23< boucman> well, illuminate/leadership/healing are quite tricky, berserker is probably easier to start with
20090328 19:18:32< euschn> i.e., supply a second set of sylables for village names
20090328 19:19:11< euschn> do you know how to do that?
20090328 19:19:35< Crab_> boucman: it's also tricky - using ZOC to protect the berserker, for example (in a canonical situation where berserker is used to counter dark adepts, and there's lost of skeletons around)
20090328 19:20:11< boucman> well, yes, but still less than the others, you only have one unit to take into account
20090328 19:20:21-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090328 19:20:48-!- Zen_Clark [i=4b0beeb9@gateway/web/ajax/mibbit.com/x-7f3f043535c7d091] has quit ["http://www.mibbit.com ajax IRC Client"]
20090328 19:21:18< Crab_> boucman: yes, you are right. it is easier. but, still, it is not so easy...
20090328 19:21:23< silene> note that the current ai is already quite good with berserkers, so it may be hard to make something exciting
20090328 19:21:58< grantwu> Oh man, all right I'll keep all of this in mind
20090328 19:21:59< boucman> silene: except that it's not able to "not attack" with a unit, which is dangerous with berserkers
20090328 19:22:00< Crab_> silene: thats because the current AI knows everything about the chance to kill :)
20090328 19:22:48< silene> Crab_: yes, and it also knows about zoc :-)
20090328 19:22:53< silene> boucman: what do you mean?
20090328 19:22:59< boucman> euschn: data/multiplayer/scenarios/Random_Scenario_Desert.cfg  calls the macro I pointed you to (and other scenarios) please look in wiki to see how it works exactly
20090328 19:23:28< boucman> well, it tends to attack with allunits that can attack, not keeping some units in the back for later
20090328 19:23:31< euschn> boucman: thanks again
20090328 19:23:50< Crab_> silene: not much ) it doesn't cheat zoc and it knows what are its effects. but, is it able to create 'safe zones' for its berserker ?
20090328 19:24:47< boucman> Crab_: my "pragmatic" approch wrt berserkers would be to put the berserker in a safe place when you find one, not "build it"
20090328 19:25:01< silene> Crab_: depends what you mean by "create"; it tries to pin enemy units, and it recognizes safe zones; but the ai won't create a line of units to protect a berserker
20090328 19:25:30< boucman> Crab_: and there is no interaction between FAI and C++ atm, so it's tricky at best
20090328 19:25:50< Crab_> boucman: I intend to make that 'integration part' pluggable
20090328 19:26:50< Crab_> boucman: I'm about formula_ai::make_candidate_moves(). For example, there may be implementation which tries candidate moves with a pair of units...
20090328 19:26:54< boucman> Crab_: that would be cool
20090328 19:27:28< Crab_> for backstab, pinning, etc - require 'considering more that one unit at a time' to get right
20090328 19:28:24< Crab_> and 'healing' requires 'considering a number of low-hp units and potential healer at a time' to get right
20090328 19:29:39< Gnutoo> it nearly compiled...it stopped at...97%...I will need to fix boost with BOOST_SP_USE_PTHREADS
20090328 19:29:43-!- stikonas [n=stikonas@ctv-79-132-187-171.vinita.lt] has joined #wesnoth-dev
20090328 19:29:48< boucman> Crab_: hmm
20090328 19:29:49< Crab_> also, things like 'swap positions for two units' requires two units and three moves...
20090328 19:29:55< Gnutoo> (boost-1.36.0)
20090328 19:30:07-!- stikonas [n=stikonas@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090328 19:30:08< boucman> we currently have two types "attack" and "move"
20090328 19:30:38< Crab_> so, if you evaluate candidate moves one-unit-a-time, it will be very hard to get right. That's what Velory already run into.
20090328 19:30:39< boucman> we could do a generic "three part" formula
20090328 19:31:05< boucman> a first formual which would return a list of combinations for eval
20090328 19:31:10< boucman> a second which would eval
20090328 19:31:14< boucman> and a third to do the move
20090328 19:31:57< boucman> so we could easily do formulas that run for all pair of (healer, safe hex)
20090328 19:33:24< Crab_> I'm thinking of a slightly different approach - something of the lines of  '  callable.add("me", action_unit_callable); callable.add("me2", action_unit_callable2);  ' in c++ before calling a formula
20090328 19:33:47< Dragonking> There is something like that - for "attack"
20090328 19:33:51< Crab_> yes
20090328 19:33:54< Dragonking> It adds "me" and "target"
20090328 19:34:04< Dragonking> I was thinking about simmilar
20090328 19:34:16< Crab_> then that is probably a good idea )
20090328 19:34:19< Dragonking> "first" and "second" for 2 units of my own side.
20090328 19:34:30< boucman> so you want a new type of candidate move for evey pair of friendly unit ?
20090328 19:34:35< Dragonking> Yes.\
20090328 19:34:42< boucman> how about doing it my way, that's more generic :)
20090328 19:34:49< Crab_> maybe only those who are 'close enough' 
20090328 19:35:35< Dragonking> boucman: Seems like a good approach.
20090328 19:35:37< boucman> a first formula would return [{me=>unit1,me2=>unit2},{me=>unit1,me2=>unit3}...]
20090328 19:35:44< Dragonking> Cause current one is noe optimal one.
20090328 19:35:50< boucman> which would be used for params to the eval one
20090328 19:35:55< Dragonking> For example if we look for units with poison.
20090328 19:36:19< Dragonking> We still need to evaluate all non-poisoners agains all enemy units even if eval returns all the time -1 for them.
20090328 19:36:56< boucman> Dragonking: that's a bit moving the problem around, since we would have to reevaluate the list every turn, which would have a similar cost
20090328 19:37:17< boucman> however, having a list like I described is more versatile and easier to adapt to every need
20090328 19:37:38< Dragonking> boucman: I tihnk it would be more optimal.
20090328 19:37:56< Dragonking> YOu eveluate for example once a unit - it has no poison - not include into list.
20090328 19:38:10< boucman> Dragonking: yes, but not much... the versatility is the selling point for me
20090328 19:38:11< Dragonking> Currently you evaluate it n times, where n is number of enemies.
20090328 19:38:29< Dragonking> That's another benefit. :)
20090328 19:38:38< boucman> true
20090328 19:46:53-!- euschn [n=chatzill@85-127-104-186.dynamic.xdsl-line.inode.at] has quit ["ChatZilla 0.9.84 [Firefox 3.0.7/2009030423]"]
20090328 19:49:46< boucman> Crab_: your proposal is a good start, but you should try to detail what you want a little more
20090328 19:50:15< boucman> apparently you have a pretty good idea of what you want to do, (listening to you on the chan) however your proposal is very vague and general
20090328 19:50:43< boucman> you could add pointers to the other wiki pages you wrote (like AI testing) right now, it's a bit empty...
20090328 19:50:45< orian_> \me is going to watch a movie, cu all ;)
20090328 19:51:07 * orian_ is ashamed
20090328 19:51:20-!- orian_ [n=orian@adab224.neoplus.adsl.tpnet.pl] has quit ["using sirc version 2.211+KSIRC/1.3.12"]
20090328 19:51:41< boucman> guest here, will be back in a couple of hours
20090328 19:51:47< Crab_> boucman: I'll do a better description, but a bit later. now I am focusing my efforts towards the ai testing (and I've updated my wiki page taking our earlier conversation about SoC priorities into account)
20090328 19:56:12-!- torghul [n=chatzill@88.242.141.52] has quit ["ChatZilla 0.9.84 [Firefox 3.0.8/2009032609]"]
20090328 19:56:51-!- CeeGe [n=contact@r74-192-217-160.bcstcmta02.clsttx.tl.dh.suddenlink.net] has joined #wesnoth-dev
20090328 19:57:09< CeeGe> Ivanovic/Sirp?
20090328 20:00:04-!- iyonius [n=iyo@a02-0829c.kn.vutbr.cz] has quit ["Leaving."]
20090328 20:00:32< CeeGe> Ivanovic. Sirp: Good day! It seems you are busy
20090328 20:00:55-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20090328 20:01:02< CeeGe> Ivanovic, Sirp: anyway, i just wanted to let you know that I have submitted my application for GSoC 2009
20090328 20:01:08< CeeGe> and have filled out all requirements
20090328 20:01:15< CeeGe> Hoping to hear from you guys soon
20090328 20:01:29< CeeGe> Ivanovic. Sirp: Thanks you...Bye
20090328 20:01:49-!- CeeGe [n=contact@r74-192-217-160.bcstcmta02.clsttx.tl.dh.suddenlink.net] has quit [Client Quit]
20090328 20:07:14< Velory> back
20090328 20:07:18< Velory> Hi everyone
20090328 20:07:20< Crab_> hi
20090328 20:09:03-!- grantwu [n=Zenneth@westquad-147025.reshall.umich.edu] has left #wesnoth-dev []
20090328 20:11:02-!- elias [n=allefant@allegro/developer/allefant] has quit ["Leaving"]
20090328 20:12:38-!- grantwu [n=zenneth@westquad-149018.reshall.umich.edu] has joined #wesnoth-dev
20090328 20:15:30-!- iyonius [n=iyo@a02-0829c.kn.vutbr.cz] has joined #wesnoth-dev
20090328 20:19:01< Soliton> hmm, we also need a require_scenario key or so since even if a scenario uses a known era it can spawn whatever units it likes.
20090328 20:20:53-!- fendrin [n=fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20090328 20:22:54< zookeeper> hmh?
20090328 20:23:38< Soliton> i just tried to observe a game with known era and unknown scenario and didn't have a needed unit.
20090328 20:24:03< zookeeper> umm, how can that happen?
20090328 20:24:27-!- ettin [n=jorda@wesnoth/developer/ettin] has quit [Read error: 110 (Connection timed out)]
20090328 20:24:42< Soliton> alternatively it could also be a require_era key and it would take a list of needed eras. but then you can still have scenario specific units that aren't really in any era.
20090328 20:25:10< benetnash> I'm now fighting with Formula AI. I'm totally newbee, so please tell me how to run this: I placed move="{ai/test.fai}" in [ai]. In test.fai is only debug message. When I run wesnoth and set ai controller it falls back to default formula AI. What is wrong? Is there something I don't know?
20090328 20:25:30< Soliton> zookeeper: the era defines what units you can recruit not really all units that are in a scenario.
20090328 20:25:30< zookeeper> ah, right. yes, makes sense.
20090328 20:26:04< benetnash> of course [ai] are in [side], I modified  2p_Den_of_Onis.cfg
20090328 20:26:08< zookeeper> maybe it and require_era= could be replaced with a common require_addon= though
20090328 20:26:16< Soliton> benetnash: set ai_algorithm in yur scenario instead.
20090328 20:26:49< benetnash> yes, in [side] I placed ai_algorithm=formula_ai
20090328 20:27:54< Soliton> so, what happens if you don't choose formula ai in the menu?
20090328 20:31:01< benetnash> I run game: wesnoth --log-info=formula_ai  -m --scenario=multiplayer_test --controller2=ai . In menu, when I starting single scenario in multiplayer game I have only default AI to choose.
20090328 20:35:05-!- happygrue [n=George@wesnoth/developer/wintermute] has quit ["http://xkcd.com/91/"]
20090328 20:38:04< benetnash> ok, I replaced move="{ai/test.fai}" with formula="{ai/test.fai}" and it works. Is there any way to prohibit default AI take control at the end of my script and move units with mp left?
20090328 20:39:13< Soliton> if you have no fallback() calls it shouldn't afaik.
20090328 20:42:49-!- penguin_crimson is now known as crimson_penguin
20090328 20:42:52< Dragonking> benetnash: Paste somewhere both of your files: cfg and fai
20090328 20:43:40< benetnash> here is cfg with map and .fai http://devourer.no-ip.org/~benetnash/ai/
20090328 20:44:10< benetnash> at the end of turn, when enemy leader cannot make move default AI moves him instead
20090328 20:44:24-!- Andrai [n=Andrai@150.165.63.86] has joined #wesnoth-dev
20090328 20:46:51-!- Zen_Clark [n=user@99-136-80-191.lightspeed.rcsntx.sbcglobal.net] has joined #wesnoth-dev
20090328 20:46:51< Dragonking> benetnash: I honestly doubt that it is working
20090328 20:47:39< Dragonking> formula= is used for unit formulas, for side one, you should use move= inside [ai] tag.
20090328 20:48:21< benetnash> so change it to move= and see, that it doesn't work at all
20090328 20:48:35-!- [Relic] [n=[Relic]@76.229.202.137] has joined #wesnoth-dev
20090328 20:49:08< Dragonking> Well, formula={..} is not even read from cfg file from what I can see in wesnoth's code.
20090328 20:49:28-!- Munk [n=munkhsoy@CPE0015e91a861d-CM0019474d8044.cpe.net.cable.rogers.com] has joined #wesnoth-dev
20090328 20:49:36< [Relic]> Hello :)
20090328 20:49:39-!- stikonas [n=stikonas@ctv-79-132-187-171.vinita.lt] has joined #wesnoth-dev
20090328 20:49:43< Dragonking> hi [Relic] 
20090328 20:49:54< benetnash> strange, but it works for me. I've got 1.6a
20090328 20:50:10< [Relic]> :)
20090328 20:50:36< Dragonking> benetnash: Does leader move like you order him to?
20090328 20:50:55< benetnash> yes, but only when is use formula= instead of move=
20090328 20:51:01< Dragonking> And once he uses up all moves turn just ends?
20090328 20:51:05< benetnash> *when I
20090328 20:51:49< benetnash> yes, but when he cannot move default AI takes control and moves him to the keep
20090328 20:52:02-!- Munk [n=munkhsoy@CPE0015e91a861d-CM0019474d8044.cpe.net.cable.rogers.com] has quit [Client Quit]
20090328 20:52:31< Dragonking> I'll have a look at it.
20090328 20:52:38< benetnash> ok, thx
20090328 20:53:15< benetnash> It's my first script and Formula AI and I might done something wrong. I used documentation from wiki.
20090328 20:56:25-!- fabi [n=fabi@f051090125.adsl.alicedsl.de] has joined #wesnoth-dev
20090328 20:56:27< Dragonking> benetnash: Where di you put cfg and fai files?
20090328 20:56:59< YogiHH> Ivanovic, if i add a new file to the sourcecode, what is needed to do with regard to the build systems?
20090328 20:57:00< benetnash> in /usr/share/data/ in ai/ and multiplayer/scenarios/
20090328 20:57:33< benetnash> /usr/share/wesnoth/data of course
20090328 20:57:59< Crab_> YogiHH, add it to CMakeLists.txt SConscript Makefile.am
20090328 20:58:19< YogiHH> Crab_: ok, thanks
20090328 21:00:57-!- Pk_JoA [n=pkjoa@190.183.72.84] has joined #wesnoth-dev
20090328 21:01:03-!- Pk_JoA [n=pkjoa@190.183.72.84] has left #wesnoth-dev []
20090328 21:04:43< Dragonking> Interesting...
20090328 21:05:20< Dragonking> benetnash: Apparently wesnoth treats your "formula=" in [side] [ai] like unit formula for a leader...
20090328 21:06:20< benetnash> it moves leader because in fai file the only command is move of leader one hex right.
20090328 21:07:01< Soliton> [side] contains stats for the leader so if a unit takes formula= then that's why.
20090328 21:07:49< Shadow_Master> oops.
20090328 21:08:10< benetnash> so how should I implement and place my AI properly?
20090328 21:09:05< CIA-30> soliton * r34213 /branches/1.6/src/server/server.cpp: made accepted_versions_ a vector so we can choose what version to tell rejected clients by the order in the config file
20090328 21:09:19< Dragonking> Soliton: Hmm.. apparetnly leaving formula setting in factions cfg was quite a bad idea :/
20090328 21:10:11< Crab_> Dragonking: I'm already working to refactor that
20090328 21:10:26< Dragonking> Crab_: Refractor what?
20090328 21:10:42< Crab_> Dragonking: allow loading ai settings from stand-alone files.
20090328 21:11:21< CIA-30> soliton * r34214 /trunk/src/server/server.cpp: made accepted_versions_ a vector so we can choose what version to tell rejected clients by the order in the config file
20090328 21:11:27< Dragonking> It is somehow not related to current problem.
20090328 21:11:54< Crab_> Dragonking: related a bit.
20090328 21:12:02< Dragonking> Apparently when multiplayer scenario is made to use formula_ai now, it takes settings from faction cfgs and block usage of custom formulas.
20090328 21:12:30< Dragonking> It does not affect single player scearios in any way.
20090328 21:12:32-!- fendrin [n=fabi@wesnoth/developer/fendrin] has quit [Read error: 110 (Connection timed out)]
20090328 21:13:06< benetnash> so there is no way (no point?) in developing own AI for wesnoth 1.6?
20090328 21:13:44< Dragonking> Soliton: Is it possible to make multiplayer scenario not load default faction settings?
20090328 21:15:09< Soliton> i'm not sure what that means but sure why not.
20090328 21:15:28< Dragonking> Soliton: How?
20090328 21:16:24< Soliton> by changing the source code..? i'm not sure what you're asking.
20090328 21:16:32< Dragonking> Soliton: I mean, now when someone tries to write formula ai for MP scenario, it takes settings from factions cfg - in which we have specified formula settings
20090328 21:16:56< Dragonking> I'm asking if someone can make multiplayer scenario and make it not load these faction cfg files 
20090328 21:17:07< Dragonking> To move around the problem in 1.6a
20090328 21:17:49< Soliton> if you use default factions you get their config.
20090328 21:19:14< Soliton> so the problem is that [multiplayer_side] [ai] [team_formula] overwrites [scenario] [side] formula or what?
20090328 21:20:09< Dragonking> Problem is that what we have under data/multiplayer/factions overwrites any side formula in custom multiplayer scenario.
20090328 21:20:28< Soliton> that's what i just said?
20090328 21:20:57< Soliton> so is that behaviour wanted to begin with?
20090328 21:21:02< Dragonking> if [scenario] [side]  is equal to [multiplayer] [side] then yes.
20090328 21:21:12< Dragonking> No, it is not wanted.
20090328 21:21:57< Soliton> well, then just changing that behaviour fixes it.
20090328 21:22:22< Dragonking> I'm trying to do find solution that will work in 1.6a
20090328 21:22:34< Dragonking> So we won't have no multipler formulas in 1.6
20090328 21:23:26< benetnash> if so I'll evolve AI for my master degree for wesnoth 1.4 in python.
20090328 21:23:38< Soliton> i guess i still don't get the problem.
20090328 21:27:36< Dragonking> benetnash: I guess that to make formula AI work in 1.6a you need to specify your own era.
20090328 21:28:39< Dragonking> Problem lies now in what is defined in default and AoH eras, that makes formula AI not work in multiplayer scenarios.
20090328 21:29:16-!- ettin [n=jorda@wesnoth/developer/ettin] has joined #wesnoth-dev
20090328 21:29:18< benetnash> Dragonking: IMO that's pointless. Easier for me will be make it in python in 1.4 because I don't have to learn formula AI.
20090328 21:30:33< Crab_> Dragonking: and what about just deleting the unneeded info in data/multiplayer/factions/*.cfg ?
20090328 21:30:51< Dragonking> Crab_: It will work for 1.6.x 
20090328 21:30:59< Soliton> you can just remove the [team_formula] part from default era factions. but sure no one can tell how much work it is for you to transition from python to formula ai.
20090328 21:31:17< Dragonking> Crab_: And I'll do that for sure.
20090328 21:31:21< Dragonking> It's a bug after all.
20090328 21:31:40< Soliton> why is that a bug if they should not override side formulas?
20090328 21:31:57-!- ikarius [n=ross@216.27.182.3] has quit [Read error: 131 (Connection reset by peer)]
20090328 21:32:56< Dragonking> Soliton: Because they do override.
20090328 21:33:11-!- ikarius [n=ross@216.27.182.3] has joined #wesnoth-dev
20090328 21:33:22< Soliton> so fix that then. removing the formulas is not fixing the underlying problem.
20090328 21:33:58< Dragonking> I would not like to mess with 1.6 too much regarding that.
20090328 21:34:20< Dragonking> But guess I'll have to.
20090328 21:34:28< Crab_> Soliton, but,at least, it may quickly fix it for benetnash. no need to do it in the trunk.
20090328 21:34:54< Dragonking> Crab_: But that does not change anything... it still is the same for every other 1.6a user.
20090328 21:35:22< Dragonking> I guess there is no point for benetnash to write formula that only he with modified config can use.
20090328 21:35:45< Crab_> Dragonking: they don't require a solution so quickly. so, it will give us time to fix the underlying problem.
20090328 21:35:50< Soliton> anyone with 1.6.1 will be able to as well if it is properly fixed.
20090328 21:36:15< Dragonking> Yup, Soliton.
20090328 21:36:20< Crab_> yes
20090328 21:36:58< Dragonking> Soliton: Ok, I'm going to look at it tomorrow.
20090328 21:38:18< Crab_> benetnash, so, (if you haven't already ), delete the [team_formula] part from data/multiplayer/factions/*-default.cfg
20090328 21:38:30< Soliton> btw, i'm not saying don't remove the useless team_formula statements though.. just that it does not seem to be the final solution.
20090328 21:38:44< Dragonking> Soliton: And you're right. Completely. :)
20090328 21:38:57< Soliton> ok. :-)
20090328 21:40:35< Dragonking> Crab_: What are you working on now? I have number of issues I want to adress during the upcoming week and I would not want to waste time working on something that you do already.
20090328 21:40:39-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has quit [Remote closed the connection]
20090328 21:41:11 * Crab_ is writing a way to hot-redeploy the AI's, draft patch is available for review
20090328 21:41:32< Crab_> i am going to implement that 'AI arena' thing we talked about
20090328 21:41:36< Dragonking> ok
20090328 21:41:51< Crab_> that will take 1st half of the week.
20090328 21:42:14< Crab_> see https://gna.org/patch/?1146 for my work-in-progress
20090328 21:43:18< Dragonking> ok:)
20090328 21:43:18< benetnash> Ok, when I removed [team_formula] AI works fine. Thx
20090328 21:43:35< Dragonking> benetnash: And remember to change formula= to move=
20090328 21:44:01< benetnash> yes, I already changed that
20090328 21:44:05< Dragonking> :)
20090328 21:46:02-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20090328 21:56:43-!- viggy_prabhu [n=vignesh@210.212.5.150] has joined #wesnoth-dev
20090328 22:03:56-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has joined #wesnoth-dev
20090328 22:21:25-!- Zen_Clark [n=user@99-136-80-191.lightspeed.rcsntx.sbcglobal.net] has quit ["CPU time needed for other things, like cum'pile'in stuff."]
20090328 22:24:19< Gnutoo> hello, what's the exact license of wesnoth-1.6a? GPLv2 or 'GPLv2 or later'?
20090328 22:25:00< Polarina> Obviously the latter.
20090328 22:25:06 * Polarina double checked too.
20090328 22:25:10-!- stikonas [n=stikonas@wesnoth/translator/stikonas] has quit [Read error: 110 (Connection timed out)]
20090328 22:25:25< Gnutoo> Polarina, thanks a lot...that permit me to include code from gcc
20090328 22:25:58< Polarina> Gnutoo: You could, no matter which license Wesnoth was under.
20090328 22:26:10< Gnutoo> ah ok
20090328 22:26:13< Crab_> btw, not all parts of wesnoth are gplv3-compatible
20090328 22:26:20< Gnutoo> ouch...
20090328 22:26:22< Crab_> see src/server/jwsmtp/compat.h
20090328 22:27:06< Gnutoo> thanks a lot
20090328 22:27:37 * Crab_ is not sure whether how that jwSMTP is actually used by Wesnoth, but it is "GPL v2 only"
20090328 22:27:43< Gnutoo> ah no I'm stupid and tired...it has to be included in boost...sorry
20090328 22:27:55< Gnutoo> and boost is BSD
20090328 22:28:00< Gnutoo> or something like that
20090328 22:28:39-!- Zen_Clark [n=user@99-136-80-191.lightspeed.rcsntx.sbcglobal.net] has joined #wesnoth-dev
20090328 22:31:13< benetnash> Gnutoo: boost license isn't viral
20090328 22:31:31< Gnutoo> yes that's good for me
20090328 22:35:04< CIA-30> ivanovic * r34215 /trunk/ (data/core/about.cfg po/wesnoth-sotbe/ru.po): updated Russian translation
20090328 22:35:19< CIA-30> ivanovic * r34216 /branches/1.6/ (data/core/about.cfg po/wesnoth-sotbe/ru.po): updated Russian translation
20090328 22:35:27-!- Ambushsabre [n=ambushsa@c-75-67-216-121.hsd1.nh.comcast.net] has quit [Read error: 110 (Connection timed out)]
20090328 22:35:59-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090328 22:40:04< YogiHH> O-O
20090328 22:40:39< YogiHH> someone sneaked a "save replay" command behind my back in ;-)
20090328 22:41:48< YogiHH> or not...
20090328 22:47:40 * Polarina sneaks code in to disable fog of war on multiplayer maps.
20090328 22:48:41< Soliton> duplicated code is bad.
20090328 22:48:58< Polarina> Disable, so I can see others, while they cannot see me.
20090328 22:55:19-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has quit ["On the road again"]
20090328 22:57:11 * YogiHH is getting old. He has been playing for one year without noticing that new menu command "Save replay"...
20090328 22:58:31< Polarina> YogiHH: Stuff happens. Ever tried to allow infants play at your computer? They slam in various key combinations you thought never existed (and you never figure out which combination it was...)
20090328 22:58:34< Crab_> no, "getting old" is when you figure out that it was actually YOU who had committed that a year ago ^)
20090328 22:59:08< YogiHH> Crab_: I suspected that first but thank god it wasn't true :-D
20090328 22:59:19< Crab_> :-D
20090328 23:00:56< YogiHH> Polarina: I know what you are talking about. I had let my daughter "use" my laptop when she was like 1 year old. I thought "hey, what can happen, she doesn't know how to do things".
20090328 23:01:04< YogiHH> Polarina:  I stopped her after 5 seconds when she had opened the registry editor...
20090328 23:01:21< Polarina> YogiHH: :P
20090328 23:03:46< Polarina> Seriously? The registry editor? How did she do that?
20090328 23:04:15-!- Espreon [n=espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20090328 23:04:35< Crab_> IMO, easiest way is Win+R, then some arrows up, if regedit is in last-run-commands-list )
20090328 23:05:26< YogiHH> I think she started on the desktop and hit Start->Execute, regedit was last called and appeared as default
20090328 23:06:31< YogiHH> I am not sure, though, it went pretty fast :-)
20090328 23:06:36< Crab_> )))
20090328 23:06:36< Polarina> :P
20090328 23:06:48< Polarina> I've also seen an infant shutting down a computer.
20090328 23:07:11< Polarina> Using the keyboard.
20090328 23:08:48-!- abhiSri [i=AB-Alway@123.239.25.118] has joined #wesnoth-dev
20090328 23:33:10-!- kristjan_ [n=kristjan@188-196-n.ipv4.vnet.ee] has joined #wesnoth-dev
20090328 23:33:28< kristjan_> --nomusic and --nosound are only sound-related runtime options?
20090328 23:36:53-!- kristjan_ [n=kristjan@188-196-n.ipv4.vnet.ee] has quit ["Leaving"]
20090328 23:36:58-!- Andrai [n=Andrai@150.165.63.86] has quit [Remote closed the connection]
20090328 23:37:11< zookeeper> esr, i'll leave this to you: http://www.wesnoth.org/forum/viewtopic.php?p=348036#p348036
20090328 23:38:35< esr> zookeeper: Concept good, dialogv a bit bland.  I'll take care of it.
20090328 23:40:56-!- kristjan_ [n=kristjan@188-196-n.ipv4.vnet.ee] has joined #wesnoth-dev
20090328 23:41:11-!- Gnutoo [n=gnutoo@host169-162-dynamic.44-79-r.retail.telecomitalia.it] has quit ["Leaving"]
20090328 23:41:26-!- silene [n=plouf@wesnoth/developer/silene] has quit ["Leaving."]
20090328 23:43:02-!- owls [i=akhilpai@markley-135168.reshall.umich.edu] has joined #wesnoth-dev
20090328 23:45:26-!- owls [i=akhilpai@markley-135168.reshall.umich.edu] has quit [Client Quit]
20090328 23:45:38-!- gozaboru [i=akhilpai@markley-135168.reshall.umich.edu] has joined #wesnoth-dev
20090328 23:46:13-!- kristjan_ [n=kristjan@188-196-n.ipv4.vnet.ee] has quit ["Leaving"]
20090328 23:50:26-!- alink [n=alink@wesnoth/developer/alink] has joined #wesnoth-dev
20090328 23:52:48-!- Elvish_Pillage2 [n=eli@66-189-40-122.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
--- Log closed Sun Mar 29 00:00:10 2009