--- Log opened Thu Oct 08 00:00:44 2009
20091008 00:18:53-!- Sirp [n=user@wesnoth/developer/dave] has joined #wesnoth-dev
20091008 00:27:47-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20091008 00:31:24< Espreon> Is it intentional for Wesnoth to crash if a [race] codeblock lacks a 'plural_name' key?
20091008 00:33:24-!- Blueblaze [n=nick@adsl-99-171-161-30.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20091008 00:35:28< Espreon> Eh, never mind..
20091008 00:35:30< Espreon> *...
20091008 00:45:56< shadowmaster> Crab_: is there any (simple) way to stop an AI team from rushing away trying to apparently find a keep?
20091008 00:46:43< Crab_> wait a sec, I'll check...
20091008 00:46:58< Crab_> 'remove all gold' ?
20091008 00:47:06< Crab_> if you're about leader movement
20091008 00:47:42-!- noy [n=Noy@wesnoth/developer/noy] has quit []
20091008 00:47:44< shadowmaster> er, you mean that it'll always look for a keep if it has gold?
20091008 00:48:20< Crab_> well, it'll do it even if it has no gold, if the leader has got nothing else to do
20091008 00:48:34< Crab_> you want the leader to move/fight but not seek keep ?
20091008 00:49:24< shadowmaster> I want it to move/fight but not <reason for rushing to the northwest here>
20091008 00:49:36< shadowmaster> I have an AI side set up like this: http://wesnoth.pastebin.com/d4a8b6b05
20091008 00:50:01< shadowmaster> where {__AI_FLT_HEROES_1} is a SUF for the player's most essential units
20091008 00:51:07< shadowmaster> but when the AI leader is spawned (in  a chamber with no keeps), it doesn't fight with my units until I attack its leader and leave him around 1% HP. Other units of that team do attack mine, however.
20091008 00:52:34< shadowmaster> that is, instead of fighting, it runs away, apparently towards northwest.
20091008 00:53:03< Crab_> has he got 'good' attacks (from your PoV) ?
20091008 00:53:32< shadowmaster> nope.
20091008 00:54:02< shadowmaster> 15-2 melee impact, 23-1 ranged impact (marksman) against an Ancient Lich :P
20091008 00:54:33< Crab_> so, you want him to try anyway ?
20091008 00:54:50< shadowmaster> yeah. That's the idea (and it worked in 1.4 and 1.6 IIRC)
20091008 00:55:01< shadowmaster> wesbot: seen alink
20091008 00:55:01< wesbot> shadowmaster: The person with the nick alink last spoke 1d 2h ago.  23h 22m ago they left with the message: Remote closed the connection
20091008 00:55:55< shadowmaster> thought that setting caution to 0 and aggression to 1 would do it.
20091008 00:56:06< Crab_> an attack which uses leader has aggression -4.0
20091008 00:56:26< Crab_> (I'm running a 'git svn blame' atm to learn 'why')
20091008 00:57:27< Crab_> (there's already workaround - you can try using formula ai to make ai leader suicidal enough, but I want an easier solution)
20091008 00:58:14< shadowmaster> formula aiaiaiai ;(
20091008 01:02:31-!- Nayela [n=Nayela@cpc3-lich6-0-0-cust673.brhm.cable.ntl.com] has quit [Remote closed the connection]
20091008 01:05:00< Crab_> ok, I think that there is two 'easy enough' solutions 1) add a new aspect 'leader_aggression'  with default value -4.0    2) use a side_formulas stage to make the leader attack any enemy units in sight
20091008 01:07:00-!- Bellerophon_true [n=chatzill@BSN-77-172-125.dial-up.dsl.siol.net] has joined #wesnoth-dev
20091008 01:08:42-!- alink [n=alink@host-85-27-114-223.brutele.be] has joined #wesnoth-dev
20091008 01:09:11< Crab_> shadowmaster: so, what we'll choose ? 
20091008 01:09:16< alink> shadowmaster: looking for me ?
20091008 01:09:41< shadowmaster> Crab_: dunno, I'd say 1 because it's simpler for me, the WML author :P
20091008 01:09:48< Crab_> ok
20091008 01:10:12< shadowmaster> alink: is it intentional that the "terrain has a local light bonus" overlay for the ToD pic in the siebar no longer shows?
20091008 01:10:29< CIA-62> crab * r39159 /trunk/data/campaigns/tutorial/scenarios/2_Tutorial.cfg: tutorial 2: get rid of 'ai turn'' event
20091008 01:10:30< shadowmaster> e.g. in maps with illuminated cave tails (Uu^Ii)
20091008 01:10:33< shadowmaster> cave tiles
20091008 01:11:08< alink> i suppose no , but i didn't touched that
20091008 01:11:32< alink> btw i noticed something broken related to this. searching
20091008 01:12:18< shadowmaster> reminds me I wanted to extend that thingy to allow for a negative bonus overlay.
20091008 01:12:30< alink> shadowmaster: you mean the halo thing
20091008 01:13:08< shadowmaster> I don't call it halo. Haloes are a completely different thing here :P
20091008 01:13:12 * alink warns shadowmaster that this ToD stuff is full of ugly hack
20091008 01:13:25< CIA-62> crab * r39160 /trunk/data/campaigns/The_South_Guard/scenarios/05_Choice_In_The_Fog.cfg: TSG 5: get rid of 'ai turn' event
20091008 01:13:33< CIA-62> crab * r39161 /trunk/data/campaigns/Heir_To_The_Throne/scenarios/07_Crossroads.cfg: HttT 7: get rid of 'ai turn' event
20091008 01:13:39< CIA-62> crab * r39162 /trunk/data/campaigns/Eastern_Invasion/scenarios/14.The_Drowned_Plains.cfg: EI 14: get rid of 'ai turn' event
20091008 01:13:44< alink> shadowmaster: i mean the bug is the halo on ToD image is missing
20091008 01:13:44< CIA-62> crab * r39163 /trunk/data/campaigns/Descent_Into_Darkness/scenarios/09_A_Small_Favor3.cfg: DiD 9: get rid of 'ai turn' event
20091008 01:13:49< shadowmaster> I'm talking about that glowy yellow overlay for the ToD picture in the sidebar, which used to be displayed if the selected tile had a lawful bonus.
20091008 01:13:50< CIA-62> crab * r39164 /trunk/data/campaigns/Descent_Into_Darkness/scenarios/11_Descent_into_Darkness.cfg: DiD 11: get rid of 'ai turn' event
20091008 01:13:52< shadowmaster> yes
20091008 01:14:08< alink> shadowmaster: checking and try to see why
20091008 01:14:52< shadowmaster> alink: quicker test: there's one of those tiles in the test scenario under a time_area cave
20091008 01:15:06< shadowmaster> (7,17) (not to mention the lava tiles)
20091008 01:15:23< alink> ah the other thing broken (and related) is [illuminated_time], there is no code support for that. did it work before
20091008 01:15:31-!- loonycyborg [n=sergey@wesnoth/developer/loonycyborg] has quit ["Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz"]
20091008 01:15:37-!- DDR [n=chatzill@66.183.125.196] has joined #wesnoth-dev
20091008 01:15:49< alink> I think it's supposed to allow to replace the ToD image by another one
20091008 01:15:55< shadowmaster> I don't think that worked in 1.4 or 1.6, but the glowy overlay thingy did work
20091008 01:16:05< alink> which look nicer that use the overlay 
20091008 01:16:31< alink> (esp. when you see the nice unused cave image created for this)
20091008 01:16:35< shadowmaster> although I personally preferred the glowy overlay thingy since it was less confusing.
20091008 01:17:23< shadowmaster> in 0.9.6, selecting a first watch tile under the effects of the Illuminate activity would display the dusk ToD pic :P
20091008 01:18:07< alink> there is some [illuminated_time] mainline macros and DM campaign
20091008 01:18:17< alink> and i think it's in the wiki too
20091008 01:18:50< shadowmaster> yeah, it is.
20091008 01:20:53< alink> I confirm the ToD image has problem, even the simple mousehover seems broken
20091008 01:22:13< alink> mmh this remind me another error that i saw there...
20091008 01:23:32< shadowmaster> making a debug build atm... while playing I got a SIGSEGV on a new turn for no reason at alll.
20091008 01:23:41< shadowmaster> Unexpected and unreproduceable :/
20091008 01:24:40-!- Bellerophon_true [n=chatzill@BSN-77-172-125.dial-up.dsl.siol.net] has quit ["ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]"]
20091008 01:25:03< shadowmaster> OTOH... Crab_, do you have any formula suggestions for that issue? :P
20091008 01:25:29< Crab_> shadowmaster: well, I'll commit leader_aggression in half an our
20091008 01:25:31< shadowmaster>  think I'll never be able to learn the language
20091008 01:25:33< shadowmaster> *I
20091008 01:25:34< Crab_> s/our/hour
20091008 01:26:02< shadowmaster> Crab_: um yes, but that won't help 1.7.6 users :P
20091008 01:26:35< Crab_> yes.
20091008 01:27:34< shadowmaster> I should have finished playtesting that scenario before releasing.
20091008 01:29:10-!- gtsteel [n=gtsteel@CPE001346a3fd7f-CM00e06fb8be94.cpe.net.cable.rogers.com] has joined #wesnoth-dev
20091008 01:29:21< shadowmaster> pppppppp 
20091008 01:29:51< shadowmaster> er, right, the entire system lags when linking the debug build. I had forgotten that.
20091008 01:30:04< Crab_> well, there's ai/dev/formula_ai_poisoning.cfg , you can grab the testing_ai_default::candidate_action_evaluation_loop stage from there, and modify it a bit.
20091008 01:30:51< Crab_> and, you should put this stage to be first (in position 0) (via modify_ai )
20091008 01:31:02< alink> YogiHH the ToD area seems not saved in savegame. To see, launch test-scenario, moushover caves, notice the ToD cave image. save/reload. It's now always dawn even in caves
20091008 01:45:05-!- ilor [n=user@wesnoth/developer/ilor] has quit [Read error: 113 (No route to host)]
20091008 01:48:06< shadowmaster> 14.47 MB free. And most of the RAM is an unused cache now.
20091008 01:48:39-!- Chusslove [n=Chusslov@217.190.226.51] has quit [Read error: 145 (Connection timed out)]
20091008 01:48:51< shadowmaster> ah, the side-effects of linking 0.3 GB executables
20091008 01:54:22< CIA-62> alink * r39165 /trunk/src/display.cpp: Fix broken ToD "illuminate" overlay in sidebar (typo in r38613)
20091008 01:54:59< alink> pff what a stupid bug :(
20091008 01:55:41< alink> shadowmaster: fixed ^
20091008 01:58:03-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20091008 02:00:11< CIA-62> alink * r39166 /trunk/src/generate_report.cpp: 
20091008 02:00:11< CIA-62> Fix a very bad missing use of reference.
20091008 02:00:11< CIA-62> (copy the whole gamemap each time we update something in sidebar)
20091008 02:00:28< alink> ^another for my collection of big 1-char optimization  \o/
20091008 02:01:33< CIA-62> crab * r39167 /trunk/ (7 files in 3 dirs): new aspect, leader_aggression, to control the aggression of the leader separately. Default value is -4.0 - that's old implicit default
20091008 02:01:35< Crab_> shadowmaster: ^ try if this one helps. 
20091008 02:01:46-!- Chusslove [n=Chusslov@brsg-d9befd75.pool.mediaWays.net] has joined #wesnoth-dev
20091008 02:01:51< Crab_> as always, more time to compile than to code (
20091008 02:02:37-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit []
20091008 02:51:53< CIA-62> alink * r39168 /trunk/src/game_display.cpp: tiny code cleaning
20091008 02:52:37< alink> shadowmaster: ^ this area is an example of ToD overlay hack I mentioned earlier
20091008 02:53:55< alink> i mean the ugly use of the extra bool
20091008 02:54:52< alink> IIRC introduced during the display/game_display split
20091008 02:57:16< alink> because display doesn't know about ToD I suppose. Anyway that's the kind of annoying things i warned you about, if/when you add more ToD feature there
20091008 03:22:18-!- Appleman1234 [n=Appleman@131.181.102.205] has quit [Read error: 110 (Connection timed out)]
20091008 03:24:32-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has quit [Remote closed the connection]
20091008 03:35:28< Espreon> https://gna.org/bugs/index.php?14472
20091008 03:39:43< shadowmaster> alink: no changelog entry for the overlay fix?
20091008 03:40:23-!- DDR [n=chatzill@66.183.125.196] has quit [Remote closed the connection]
20091008 03:40:47< alink> ah yes, will do since bug was in 1.7.6 I think
20091008 03:41:31< shadowmaster> wtf am I doing, making another debug build?
20091008 03:42:11< shadowmaster> yes, that was in 1.7.6. The release was made around r38889
20091008 03:43:26< shadowmaster> I wonder if there's any way I can allow other programs to run smoothly while the debug build executable is being linked, or the internal object collections archived
20091008 03:43:59< CIA-62> alink * r39169 /trunk/changelog: update changelog
20091008 03:46:09< shadowmaster> ps aux | grep ccdid I
20091008 03:48:37< shadowmaster> well, using -j 4 wasn't exactly a good idea. :p
20091008 03:49:10< Soliton> maybe ionice helps.
20091008 03:50:22< shadowmaster> Espreon: that's a duplicate bug of some other bug I reproted
20091008 03:50:37< Espreon> I see...
20091008 03:50:44< shadowmaster> with a much higher priority since it's a rather important usability issue that was not present in 1.7.6
20091008 03:54:45< shadowmaster> Crab_: doesn't seem to help
20091008 03:54:58< shadowmaster> and I added {AI_SIMPLE_ALWAYS_ASPECT leader_aggression 10.0}
20091008 03:55:04< Crab_> 1.0
20091008 03:55:31< Crab_> can you give/describe me a test case ?
20091008 03:55:33< shadowmaster> higher numbers dowawait f
20091008 03:55:48< shadowmaster> no, it's not necessar. I'm running the wrong build.
20091008 03:56:13< Crab_> ok. ping me if it still doesn't work, i'll try to set up a test case and find out what's wrong
20091008 04:00:02< shadowmaster> Crab_: well, the leader doesn't run away now :P
20091008 04:00:20< shadowmaster> he just moves to an adjacent hex to one of my units and doesn't attack
20091008 04:01:05< Crab_> is there any messages like "Not fighting with leader because recruiting is more preferable" in the logs ?
20091008 04:01:23< shadowmaster> what log level should display that?
20091008 04:01:25< Crab_> if run with --log-debug=ai/general
20091008 04:02:09-!- Appleman1234 [n=Appleman@131.181.102.166] has joined #wesnoth-dev
20091008 04:03:45< shadowmaster> pah.
20091008 04:03:57< shadowmaster> he decided to attack when I loaded the game.
20091008 04:04:49< Crab_> ok. also, for extra effect, set {AI_SIMPLE_ALWAYS_ASPECT number_of_possible_recruits_to_force_recruit 0.0}
20091008 04:05:03< shadowmaster> new turn, he decided not to attack.
20091008 04:05:04< Crab_> it should reduce the tendency to search for keep to recruit something
20091008 04:05:10< shadowmaster> lots of "Not fighting with leader because recruiting is more preferable" indeed
20091008 04:05:38< shadowmaster> okay, I'll test with that.
20091008 04:07:12 * shadowmaster notes that custom events + :throw = win
20091008 04:07:32< shadowmaster> (no, not in the end-of-scenario sense)
20091008 04:09:38< shadowmaster> well, sure that guy is effective now. He destroyed my lich :P
20091008 04:09:46< Crab_> hehe ))
20091008 04:10:43-!- Blueblaze [n=nick@adsl-99-171-161-30.dsl.hstntx.sbcglobal.net] has quit [Read error: 104 (Connection reset by peer)]
20091008 04:10:59< shadowmaster> Crab_: thanks
20091008 04:11:21< Crab_> I'm glad I could help
20091008 04:47:10< alink> I now see various bugs with this illuminate ToD overlay. I think I know how to fix them but I just realize now a UI flaw with the mousehover system, you can't read the bonus/malus tooltip of illuminated area :-/
20091008 04:47:53< shadowmaster> alink: but at least now we can see what's the current effect on the unit in the sidebar
20091008 04:47:57-!- ardesh__ [n=ardesh@port-92-206-122-113.dynamic.qsc.de] has quit [Read error: 110 (Connection timed out)]
20091008 04:48:05< shadowmaster> I mean with your recent additions to the sidebar text
20091008 04:48:12< alink> yes indeed
20091008 04:48:32-!- ardesh__ [n=ardesh@port-92-206-22-44.dynamic.qsc.de] has joined #wesnoth-dev
20091008 04:49:02< alink> funnily enough, the invisible change of the content tooltip prevent a update bug caused by the bad code about this brighten bool hack
20091008 04:49:49< alink> (the update check doesn't see that the bool change, but see the tooltip change)
20091008 04:50:35< alink> mmmhn, which maybe means that some non standard WML bonus/malus may trigger this bug
20091008 04:50:56-!- Crab_ [i=crab@wesnoth/developer/crab] has quit ["Leaving."]
20091008 04:51:29< alink> I think it's time that i try killing this ugly bool
20091008 04:53:19< alink> cleaning the implementation may even simply the fix of the other bugs
20091008 04:54:39< alink> which are : (1) don't show "naturally"(not by unit) illuminated areas under fog and (2) allow to detect cave hexes under shroud
20091008 04:55:33< alink> (both are about the sidebar ToD image)
20091008 04:55:43-!- Ivanovic_ [n=ivanovic@dtmd-4db22673.pool.mediaWays.net] has joined #wesnoth-dev
20091008 04:58:24-!- gtsteel [n=gtsteel@CPE001346a3fd7f-CM00e06fb8be94.cpe.net.cable.rogers.com] has quit ["Lost terminal"]
20091008 05:07:25-!- chains [n=Rylar@netblock-72-25-91-59.dslextreme.com] has joined #wesnoth-dev
20091008 05:12:32-!- Ivanovic [n=ivanovic@wesnoth/developer/ivanovic] has quit [Read error: 113 (No route to host)]
20091008 05:13:42-!- Ivanovic_ is now known as Ivanovic
20091008 05:25:15-!- DDR [n=chatzill@66.183.125.196] has joined #wesnoth-dev
20091008 05:32:33-!- Blueblaze [n=nick@adsl-99-171-161-30.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20091008 05:37:35-!- Espreon [n=espreon@wesnoth/developer/espreon] has quit ["WRYYYYYYYYYYYYYYYYYYYY!"]
20091008 05:42:14-!- SonIcco [n=SonIcco@pD95106D2.dip0.t-ipconnect.de] has joined #wesnoth-dev
20091008 05:56:34-!- SonIcco [n=SonIcco@pD95106D2.dip0.t-ipconnect.de] has quit [Remote closed the connection]
20091008 06:10:02-!- alink [n=alink@wesnoth/developer/alink] has quit [Remote closed the connection]
20091008 06:10:46-!- Chusslove [n=Chusslov@brsg-d9befd75.pool.mediaWays.net] has quit [Read error: 110 (Connection timed out)]
20091008 06:17:14-!- Chusslove [n=Chusslov@brsg-d9bee09b.pool.mediaWays.net] has joined #wesnoth-dev
20091008 06:34:39-!- DDR_ [n=chatzill@66.183.125.196] has joined #wesnoth-dev
20091008 06:40:01-!- Sirp [n=user@wesnoth/developer/dave] has quit [Read error: 113 (No route to host)]
20091008 07:41:04-!- DDR_ [n=chatzill@66.183.125.196] has quit [Read error: 110 (Connection timed out)]
20091008 07:53:18-!- DDR [n=chatzill@66.183.125.196] has quit [Read error: 110 (Connection timed out)]
20091008 08:09:13-!- zookeeper [n=l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20091008 09:13:13-!- Blueblaze [n=nick@adsl-99-171-161-30.dsl.hstntx.sbcglobal.net] has quit [Remote closed the connection]
20091008 09:18:40-!- euschn [n=chatzill@wesnoth/developer/euschn] has joined #wesnoth-dev
20091008 09:29:44-!- stikonas [n=and@bcm-131-111-216-119.girton.cam.ac.uk] has joined #wesnoth-dev
20091008 09:41:39-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20091008 09:47:05-!- stikonas [n=and@bcm-131-111-216-119.girton.cam.ac.uk] has joined #wesnoth-dev
20091008 09:47:36-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20091008 09:56:08-!- Appleman1234 [n=Appleman@131.181.102.166] has quit [Read error: 104 (Connection reset by peer)]
20091008 09:59:37-!- [Relic] [n=[Relic]@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit ["Leaving"]
20091008 10:14:19-!- fendrin [n=fabi@wesnoth/developer/fendrin] has quit [Remote closed the connection]
20091008 10:15:15-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20091008 10:17:42< Ivanovic> moin
20091008 10:28:07-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20091008 10:31:06-!- ilor_ [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20091008 10:40:13-!- ilor [n=user@wesnoth/developer/ilor] has quit [Read error: 110 (Connection timed out)]
20091008 11:13:51< CIA-62> ivanovic * r39170 /trunk/po/ (25 files in 25 dirs): msginit created files with the original strings copied to the translated locations, this is obviously *not* what was wanted, going to "completely untranslated" files...
20091008 11:18:42-!- loonybot [n=loonybot@79.139.139.50] has joined #wesnoth-dev
20091008 11:20:10-!- loonycyborg [n=sergey@79.139.139.50] has joined #wesnoth-dev
20091008 11:41:45-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has quit ["On the road again"]
20091008 11:45:35< Ivanovic> hi loonycyborg
20091008 11:45:49< Ivanovic> loonybot: have you done anything regarding setting language as fallback?
20091008 11:45:51< loonybot> error: 'have' does not name a type
20091008 11:46:13< Ivanovic> since yeah, i think it might make sense to just set the language= block if the locale is not available
20091008 11:55:47< loonycyborg> Not yet. Should it be activated with a command-line option at first?
20091008 11:56:23< AI0867> po/wesnoth/en@shaw.po:8: some header fields still have the initial default value
20091008 11:58:07-!- freim [n=hogne@wesnoth/artist/freim] has joined #wesnoth-dev
20091008 11:59:40< Ivanovic> AI0867: of course they do, because the files are brand new!
20091008 12:00:44< AI0867> that's the only file it gave that message on though
20091008 12:01:01< Chusslove> They shouldn't have initial default values after POT->PO.
20091008 12:01:05< Ivanovic> yes, gettext is sometimes strange...
20091008 12:01:15< Ivanovic> Chusslove: after msginit they *do* have the initial values
20091008 12:01:27< Chusslove> That's a problem.
20091008 12:01:29 * Chusslove checks.
20091008 12:01:45< Ivanovic> i originally created the files using msginit
20091008 12:02:01< Ivanovic> i just saw an hour ago that msginit directly copied over the original strings
20091008 12:02:20< AI0867> that particular file contains stuff like PACKAGE VERSION and FULL NAME <EMAIL@ADDRESS>
20091008 12:02:41< Ivanovic> so i opened all "old" files in kate and copied over all .pot files over the new files just to replace the headers right afterwards with the ones from msginit
20091008 12:03:06< Ivanovic> (yes, some damn work by hand, but now these are empty po files with the original msginit stuff set)
20091008 12:04:19< Chusslove> A guess: msginit thought it's like en_GB, and hence copied translations over, because that's what en_GB translator would normally do.
20091008 12:05:07< Chusslove> But not Shavian translator, no. (I wonders how *that* will turn out... :)
20091008 12:05:33< Chusslove> Auto transliteration should not be possible...
20091008 12:05:57< Ivanovic> Chusslove: arc is in #wesnoth
20091008 12:06:08< Ivanovic> he thinks that auto transliteration might work for most stuff
20091008 12:10:15-!- fendrin [n=fabi@88-134-143-95-dynip.superkabel.de] has joined #wesnoth-dev
20091008 12:10:47< Chusslove> Ivanovic: Yep, msginit copies over original to translation if target language is en<something>.
20091008 12:11:06< freim> what is the default build system these days? make, cmake, scons?
20091008 12:11:10< Ivanovic> Chusslove: that is why i changed this by hand
20091008 12:11:25< Ivanovic> freim: all three should work, though in releases i only test if cmake works (for the dev tree)
20091008 12:15:54 * loonycyborg does everything with scons, including windows releases
20091008 12:16:38< Ivanovic> loonycyborg: not sure if we should do the language stuff via commandline option
20091008 12:16:57< Ivanovic> since in theory a commandline option would require an added string to the manpages and we are in string freeze...
20091008 12:17:23< Ivanovic> so i'd guess it should be the fallback case for all those cases that currently are greyed out in the selection
20091008 12:18:12< Ivanovic> (as in: instead of greying and not doing anything, grey out (to show that support might not be as perfect as with the rest (date format!)) but use language= to set it when selected instead of not setting)
20091008 12:18:33< Ivanovic> okay, off for breakfast, cu
20091008 12:19:40< loonycyborg> Not sure whether greying out but not disabling is possible with that scrolling list widget..
20091008 12:21:54< freim> scons seemed to be pretty straightforward
20091008 12:24:24< Chusslove> Ivanovic: Here's one way to clear translations, in case needed again: msgfilter -i foo.po -o foo.po --keep-header sed -e 'd'
20091008 12:31:58< Soliton> fendrin: data/campaigns/Legend_of_Wesmere/utils/ai_controller.deprecated  can be removed?
20091008 12:36:41-!- euschn [n=chatzill@wesnoth/developer/euschn] has quit ["ChatZilla 0.9.85 [Firefox 3.0.14/2009090216]"]
20091008 13:53:09< loonycyborg> Ivanovic: http://wesnoth.pastebin.com/m49181159
20091008 13:54:45< loonycyborg> That's a preliminary LANGUAGE fallback patch. Not commit-ready yet since it un-grey-outs all languages and uses setenv which is unavailable on windows.
20091008 13:57:55-!- ardesh_ [n=ardesh@port-92-206-48-100.dynamic.qsc.de] has joined #wesnoth-dev
20091008 14:14:33-!- ardesh__ [n=ardesh@port-92-206-22-44.dynamic.qsc.de] has quit [Read error: 110 (Connection timed out)]
20091008 14:35:46-!- stikonas [n=and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20091008 15:34:04< fendrin> Soliton: right
20091008 15:34:12-!- fendrin [n=fabi@wesnoth/developer/fendrin] has quit [Remote closed the connection]
20091008 15:42:52-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20091008 15:45:44< CIA-62> soliton * r39171 /trunk/src/server/server.cpp: refactored some simple_wml handling to be more readable
20091008 15:46:06< CIA-62> soliton * r39172 /trunk/data/campaigns/Legend_of_Wesmere/utils/ai_controller.deprecated: removed an obsolete file
20091008 16:00:51-!- ilor_ [n=user@wesnoth/developer/ilor] has quit [Read error: 110 (Connection timed out)]
20091008 16:05:19-!- alink [n=alink@wesnoth/developer/alink] has joined #wesnoth-dev
20091008 16:15:14-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20091008 16:42:58-!- ilor_ [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20091008 16:45:49-!- ardesh_ [n=ardesh@port-92-206-48-100.dynamic.qsc.de] has quit ["Quis custodiet ipsos custodes"]
20091008 16:46:15-!- SonIcco [n=SonIcco@pD95106D2.dip0.t-ipconnect.de] has joined #wesnoth-dev
20091008 16:47:05-!- Blueblaze [n=nick@adsl-99-171-161-30.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20091008 16:47:26-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20091008 16:50:08-!- ilor [n=user@wesnoth/developer/ilor] has quit [Read error: 60 (Operation timed out)]
20091008 16:50:55-!- stikonas [n=and@bcm-131-111-216-119.girton.cam.ac.uk] has joined #wesnoth-dev
20091008 17:06:56< thespaceinvader> wesbot seen boucman
20091008 17:06:57< wesbot> thespaceinvader: The person with the nick boucman last spoke 19h 14m ago.  17h 50m ago they left with the message: "Leaving."
20091008 17:17:55< CIA-62> ivanovic * r39173 /trunk/ (4 files in 3 dirs): add files missing for Shavian translation
20091008 17:20:45< Soliton> wesnothd still produces bad alloc exceptions. :-/
20091008 17:22:40< Soliton> probably this time because the kernel doesn't like to give one process more than 2.5GB address space. even though we turned overcommitting to 1 which means always allow it.
20091008 17:24:38< Soliton> well, the 2.5GB is probably more a limitation of our server being 32bit.
20091008 17:25:55-!- SonIcco_ [n=SonIcco@pD9510AA9.dip0.t-ipconnect.de] has joined #wesnoth-dev
20091008 17:43:08-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20091008 17:43:22-!- SonIcco [n=SonIcco@pD95106D2.dip0.t-ipconnect.de] has quit [Read error: 110 (Connection timed out)]
20091008 17:54:23< alink> Soltion: what kind of data do we put in these 2.5GB ?
20091008 17:55:48< alink> it can't be just WML stuff from the running games, or is it just wasted dirty memory ?
20091008 17:56:02 * alink doesn't know anything about wesnothd
20091008 17:57:05< alink> s/Soltion/Soliton
20091008 17:59:44< Soliton> alink: i'm talking about the virtual address space. really using was wesnothd about 1.3GB. no idea how much is used for what but most is probably needed for scenario/replay WML of each game.
20091008 18:00:51-!- ilor_ [n=user@wesnoth/developer/ilor] has quit [Read error: 110 (Connection timed out)]
20091008 18:01:14< alink> ah so keeping replay info compact is not only useful for bandwith and user's disk space but also for our sever's memory usage
20091008 18:01:31< Soliton> seems quite likely that there are also some memory leaks since it runs fine for a while and the amount of games/users doesn't change much each day.
20091008 18:02:44< alink> is there a low hour when we can reset it each day ?
20091008 18:03:12-!- Blueblaze [n=nick@adsl-99-171-161-30.dsl.hstntx.sbcglobal.net] has quit [Read error: 104 (Connection reset by peer)]
20091008 18:03:32< Soliton> low hour still means ~100 users but sure we could do that.
20091008 18:04:03< alink> mmh minimum is ~100 wow, then no seems a bit too much people
20091008 18:04:50< Soliton> we can restart the server gracefully which means people can finish there games and the server only terminates once all games finished.
20091008 18:05:02< Soliton> s/there/their/
20091008 18:05:26< alink> and forbiding new games during that wait period ?
20091008 18:05:43< Soliton> yes.
20091008 18:06:05< Soliton> and you can't reconnect to it since the port is used by the new instance.
20091008 18:06:07-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20091008 18:06:53< alink> or is it possible to start a temporary second wesnothd just for that case (probably a bad idea, but just trowhing it in case someone see a better variant)
20091008 18:07:53< Soliton> hmm?
20091008 18:08:34< alink> no ok seems stupid, and your port remark seems to prevent having 2 wesnoths running in parallel 
20091008 18:08:41< alink> s/wesnoths/wesnoth
20091008 18:08:46< alink> *wesnothd
20091008 18:10:01< Soliton> we do run two wesnothd's in parallel when we do a graceful restart.
20091008 18:10:47< alink> ah so you can't reconnect to the old one but instead can connect with the new one
20091008 18:11:07< Soliton> yes.
20091008 18:11:11< alink> nice
20091008 18:14:24< freim> the INSTALL file seems to be incorrect with regards to how to build the editor. It says "wesnoth_editor" as build target, but that gives "scons: *** Do not know how to make target `wesnoth_editor'.  Stop"
20091008 18:15:52< Soliton> the editor is built-in nowadays.
20091008 18:16:34< freim> ok, then the INSTALL file should be updated to reflect that
20091008 18:18:59-!- SonIcco__ [n=SonIcco@pD9510AA9.dip0.t-ipconnect.de] has joined #wesnoth-dev
20091008 18:21:15< Ivanovic> i am off for the evening, cu
20091008 18:23:56< CIA-62> loonycyborg * r39174 /trunk/INSTALL: Updated INSTALL. wesnoth_editor scons target no longer exists.
20091008 18:25:30< loonycyborg> freim: Updated
20091008 18:27:57< freim> loonycyborg: thanks
20091008 18:28:09-!- Nayela [n=Nayela@cpc3-lich6-0-0-cust673.brhm.cable.ntl.com] has joined #wesnoth-dev
20091008 18:31:45-!- SonIcco_ [n=SonIcco@pD9510AA9.dip0.t-ipconnect.de] has quit [Read error: 113 (No route to host)]
20091008 19:01:18-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20091008 19:09:24< alink> seeing what data there is a replay, i notice a lot of [random] for recruit. There is a anti-OOS for different- localetrick
20091008 19:09:36< alink> *trick
20091008 19:10:04< alink> always generating 12 random for markov chain stuff
20091008 19:10:18< alink> 12 seems a lot for a name, but anyway
20091008 19:10:58< alink> i was wondering if we could replace it by a simple noise function using only 1 (or maybe 2) random ?
20091008 19:11:32< alink> (for generating these 12 randoms)
20091008 19:12:08< alink> mordante: since your are more familiar with this, any opinion ^
20091008 19:13:00< alink> of course, i mean determinist noise function
20091008 19:13:12-!- Aethaeryn [n=Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #Wesnoth-dev
20091008 19:13:59< alink> i also see one useless random for unit_type having only 1 gender. We roll a 1-face die for them
20091008 19:14:15< alink> *dice
20091008 20:24:43-!- DDR [n=chatzill@66.183.125.196] has joined #wesnoth-dev
20091008 20:34:48-!- DDR [n=chatzill@66.183.125.196] has quit [Remote closed the connection]
20091008 20:43:03-!- DDR [n=chatzill@66.183.125.196] has joined #wesnoth-dev
20091008 20:48:07-!- SonIcco__ [n=SonIcco@pD9510AA9.dip0.t-ipconnect.de] has quit [Remote closed the connection]
20091008 21:02:21-!- DDR [n=chatzill@66.183.125.196] has quit [Read error: 60 (Operation timed out)]
20091008 21:06:08-!- DDR [n=chatzill@66.183.125.196] has joined #wesnoth-dev
20091008 21:29:39-!- DDR [n=chatzill@66.183.125.196] has quit [Read error: 60 (Operation timed out)]
20091008 21:38:43< CIA-62> dave * r39175 /trunk/src/server/ (game.cpp simple_wml.cpp simple_wml.hpp): fix to memory leak in wesnothd
20091008 21:42:10-!- silene [n=plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20091008 21:42:18< silene> hi
20091008 21:44:25-!- stikonas [n=and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20091008 21:57:27-!- Sirp [n=user@wesnoth/developer/dave] has joined #wesnoth-dev
20091008 22:01:34< CIA-62> soliton * r39176 /branches/1.6/src/server/ (game.cpp simple_wml.cpp simple_wml.hpp): fix to memory leak in wesnothd
20091008 22:24:09-!- Blueblaze [n=nick@adsl-99-171-161-30.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20091008 22:24:54< shadowmaster> how nice, I have had a svn client process stuck for a whole day
20091008 22:27:16< shadowmaster> how nice, I got an unreproducible SIGSEGV while playing again :/
20091008 22:27:53< shadowmaster> and there seem to be no NULL pointers involved. *sighs*
20091008 22:28:05< Sirp> shadowmaster: backtrace?
20091008 22:28:21-!- Crab_ [i=crab@wesnoth/developer/crab] has joined #wesnoth-dev
20091008 22:28:47< shadowmaster> I only have the first 6 frames. I got frustrated while analyzing it :P
20091008 22:28:56-!- PK3 [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has joined #wesnoth-dev
20091008 22:29:01< Sirp> so you deleted the core?
20091008 22:29:22< shadowmaster> sort of, no coredumps enabled.
20091008 22:29:34< PK3> in soviet russia, core deletes *you*!
20091008 22:29:43< shadowmaster> I was playing with gdb attached to a debug build
20091008 22:29:46< Sirp> ahh
20091008 22:29:58< Sirp> shadowmaster: you only looked at the first 6 frames and got frustrated? :p
20091008 22:30:55< shadowmaster> what really annoys me is that reloading from a saved game doesn't help me reproduce the crash
20091008 22:32:21< shadowmaster> silene: is the inclusion of /data/lua/wml-tags.lua hard-coded somewhere, or is that specified by WML?
20091008 22:33:46< silene> shadowmaster: it's hardcoded in wml: data/core/_main.cfg
20091008 22:44:53< Sirp> Soliton: so how soon will we find out how well the new server is doing? :)
20091008 22:47:07< Soliton> Sirp: i'll restart it in a couple of hours and then we should see different resource usage over time and.. no more crashes.
20091008 22:49:38< Sirp> Soliton: hopefully. :)
20091008 22:51:10-!- giusef [n=giusef@unaffiliated/giusef] has joined #wesnoth-dev
20091008 22:56:49-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20091008 22:58:30-!- stikonas [n=and@bcm-131-111-216-119.girton.cam.ac.uk] has joined #wesnoth-dev
20091008 22:59:07-!- zookeeper [n=l@wesnoth/developer/zookeeper] has quit []
20091008 22:59:49-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20091008 23:02:39-!- stikonas [n=and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20091008 23:08:40-!- Doppp is now known as LFC|Doppp
20091008 23:08:57-!- DDR [n=chatzill@66.183.125.196] has joined #wesnoth-dev
20091008 23:09:04-!- LFC|Doppp is now known as Doppp
20091008 23:24:52-!- Blueblaze [n=nick@adsl-99-171-161-30.dsl.hstntx.sbcglobal.net] has quit [Read error: 104 (Connection reset by peer)]
20091008 23:28:24-!- Espreon [n=espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20091008 23:31:28-!- Appleman1234 [n=Appleman@131.181.102.166] has joined #wesnoth-dev
20091008 23:41:31-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20091008 23:43:05-!- silene [n=plouf@wesnoth/developer/silene] has quit ["Leaving."]
20091008 23:46:42-!- boucman [n=rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20091008 23:47:22< boucman> hey all
20091008 23:47:27< boucman> thespaceinvader: looking for me ?
20091008 23:48:03< thespaceinvader> boucman: yeah - i was trying to work out the drake in-water defence anim, but i couldn't figure out where to put in the base frame to avoid it using the one from the flying animation
20091008 23:48:06< thespaceinvader> any ideas?
20091008 23:49:06< boucman> thespaceinvader: did you have a look at the generic defense animation ?
20091008 23:49:14< thespaceinvader> yes
20091008 23:49:24< thespaceinvader> it doesn't say anything about the standing frame being used anywhere
20091008 23:49:40< thespaceinvader> unless i am misreading it
20091008 23:49:56< boucman> IIRC it does, let me check
20091008 23:50:41< boucman> i'm in animation-utils.cfg line 52
20091008 23:52:17< thespaceinvader> yeah
20091008 23:52:49< thespaceinvader> but BASE_IMAGE in most defence animations is defence-1, REACTION_IMAGE is defence-2
20091008 23:52:55< Aethaeryn> "the drake in-water defence anim" <- aren't you being a tad bit specific?
20091008 23:53:04< thespaceinvader> an image i've not specified at all in the defence macro is playing
20091008 23:53:11< thespaceinvader> Aethaeryn: no, not at all
20091008 23:53:18< thespaceinvader> that's wherew the bug is ;)
20091008 23:53:30< thespaceinvader> for most drakes anyway
20091008 23:53:30< boucman> hmmm
20091008 23:53:50< thespaceinvader> it only occurs when the drake would be fllying over that terrain, and is defending but not hit
20091008 23:53:54< thespaceinvader> as far as i can work out
20091008 23:54:24< boucman> I thought it was that particular image, but in that case it might still be code, i'll look deeper
20091008 23:54:30< thespaceinvader> i think there's a problem with the interaction between a flying standing animation, and a landed base frame
20091008 23:54:48< thespaceinvader> s/base frame/defence anim
20091008 23:55:03< boucman> the unit goes in standing anim between each strike, but normally there are no redraws, so it should be invisible
20091008 23:55:10< thespaceinvader> yeah
20091008 23:55:14< boucman> it seems there is an extra "redraw" somewhere
20091008 23:55:23-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20091008 23:55:28< boucman> i'll tweak the code to not go in standing and see if it solves the problem
20091008 23:55:29< thespaceinvader> i think there must be a redraw when the unit gets missed, or something
20091008 23:56:33-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has quit [Remote closed the connection]
20091008 23:57:44-!- stikonas [n=and@bcm-131-111-216-119.girton.cam.ac.uk] has joined #wesnoth-dev
20091008 23:59:04-!- PK3 [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has quit ["Java user signed off"]
--- Log closed Fri Oct 09 00:00:57 2009