--- Log opened Wed May 05 00:00:54 2010
20100505 00:08:26-!- deekay [~dk@wesnoth/developer/dragonking] has quit [Ping timeout: 240 seconds]
20100505 00:11:19-!- deekay [~dk@wesnoth/developer/dragonking] has joined #wesnoth-dev
20100505 00:12:24-!- qemqemqem__ [~quassel@dhcp-urwireless-128-151-180-20.wireless.rochester.edu] has joined #wesnoth-dev
20100505 00:13:58-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 246 seconds]
20100505 00:20:48-!- qemqemqem__ [~quassel@dhcp-urwireless-128-151-180-20.wireless.rochester.edu] has quit [Ping timeout: 276 seconds]
20100505 00:23:21-!- _jbx_ [~jbailey@12.190.80.225] has quit [Quit: It was a new day yesterday, but it's an old day now.]
20100505 00:33:54-!- k23z__ [~k23z__@188.26.49.101] has joined #wesnoth-dev
20100505 00:33:56-!- k23z__ [~k23z__@188.26.49.101] has quit [Changing host]
20100505 00:33:56-!- k23z__ [~k23z__@unaffiliated/k23z--/x-2536701] has joined #wesnoth-dev
20100505 00:37:53< CIA-10> jetryl * r42473 /branches/1.8/data/campaigns/Under_the_Burning_Suns/images/units/ (16 files in 2 dirs): Fixed some images that got botched by OptiPNG.
20100505 00:59:14-!- Bob_The_Mighty [~chatzilla@cpc4-brig15-0-0-cust904.3-3.cable.virginmedia.com] has joined #wesnoth-dev
20100505 01:02:02-!- deekay [~dk@wesnoth/developer/dragonking] has quit [Quit: deekay]
20100505 01:34:30-!- kevg [~kosov.eug@91.194.253.47] has quit []
20100505 01:35:39-!- fakedrake [~fakedrake@ppp-94-64-230-186.home.otenet.gr] has joined #wesnoth-dev
20100505 01:35:47< fakedrake> hi all
20100505 01:35:54< Crab_> hi, fakedrake
20100505 01:50:59-!- wesbot changed the topic of #wesnoth-dev to: 100 bugs, 277 feature requests, 15 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100505 01:51:05-!- Shakey [HydraIRC@c-71-201-89-187.hsd1.il.comcast.net] has joined #wesnoth-dev
20100505 02:01:57-!- Bob_The_Mighty [~chatzilla@cpc4-brig15-0-0-cust904.3-3.cable.virginmedia.com] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539]]
20100505 02:09:24-!- shadowmaster is now known as Kuchen
20100505 02:19:38-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100505 02:32:49-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz]
20100505 02:33:26-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20100505 02:39:03-!- Espreon [~espreon@wesnoth/developer/espreon] has quit [Quit: leaving]
20100505 02:40:15-!- Espreon [~espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20100505 02:47:16-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Quit: bbl]
20100505 02:49:07-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100505 02:49:30-!- Kuchen is now known as shadowmaster
20100505 02:52:06-!- Shakey [HydraIRC@c-71-201-89-187.hsd1.il.comcast.net] has quit [Quit:  HydraIRC -> http://www.hydrairc.com <- *I* use it, so it must be good!]
20100505 02:55:13-!- shadowm_laptop [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100505 02:55:44-!- shadowmaster [~ignacio@wesnoth/developer/shadowmaster] has quit [Read error: error:1408F10B:SSL routines:SSL3_GET_RECORD:wrong version number]
20100505 02:56:04-!- shadowmaster [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100505 03:02:00 * Espreon wants to kill whoever made those pngs indexed...
20100505 03:02:45< shadowmaster> Espreon: which ones?
20100505 03:03:35< Espreon> shadowmaster: ALL of the desert elves...
20100505 03:03:59< Espreon> ... and the Naga Guardian.
20100505 03:04:29< shadowmaster> um, okay, I think the oldest wesnoth-pngcrush did that
20100505 03:04:37< Espreon> ...
20100505 03:04:42< shadowmaster> before my time
20100505 03:04:47< shadowmaster> looong before my time
20100505 03:04:52< Espreon> I... hæt this...
20100505 03:05:08< Espreon> wesnoth-optipng + indexed pngs = ugh...
20100505 03:05:25< shadowmaster> if you need to work with them, go ahead and RGBA (ARGB???)'ize them and run wesnoth-optipng on them
20100505 03:05:41< Espreon> That is what I am trying to do...
20100505 03:05:43< shadowmaster> no idea about how to do the first automatically
20100505 03:05:49< Espreon> ... but I am not going to manually convert them.
20100505 03:05:58< shadowmaster> maybe some png tool software out there can do it
20100505 03:06:13< shadowmaster> check your distributor for stuff that deals with PNGs
20100505 03:10:37< Espreon> If only imagemagick actually worked...
20100505 03:10:56< shadowmaster> well, it does work
20100505 03:11:18< Espreon> ...
20100505 03:11:19< shadowmaster> the problem is that it's rather difficult to learn to use. THe most I could make is a simple script for resizing screenshots from frogatto
20100505 03:11:35< shadowmaster> the documentation is mind-boggling
20100505 03:23:35< Espreon> MEIN GOTT!
20100505 03:23:41< Espreon> Fuck this, I'm manually doing it...
20100505 03:26:03< AI0867> Espreon: let me take a look at the imagemagick docs
20100505 03:26:29< Espreon> AI0867: Good luck.
20100505 03:27:31< AI0867> there's images in wesnoth that I created with imagemagick
20100505 03:30:37< shadowmaster> do you realize I wasn't kidding when I called you "Master of tedious tasks"?
20100505 03:30:38< AI0867> all the utbs images or just the elves?
20100505 03:30:40< shadowmaster> ;)
20100505 03:30:54< Espreon> AI0867: Just do all of them.
20100505 03:30:55< shadowmaster> (the tedious task here is reading im's documentation)
20100505 03:35:34-!- Upthorn [~ogmar@adsl-75-26-184-137.dsl.scrm01.sbcglobal.net] has quit [Ping timeout: 246 seconds]
20100505 03:45:06< AI0867> I'm passing all the right options, but nothing is happening...
20100505 03:45:24< AI0867> the only difference is the doubling in size
20100505 03:45:38-!- Vetinari [~lukjad007@unaffiliated/lukjad007] has quit [Ping timeout: 240 seconds]
20100505 03:48:46-!- k23z__ [~k23z__@unaffiliated/k23z--/x-2536701] has quit [Read error: Connection reset by peer]
20100505 03:53:04< AI0867> gah
20100505 03:53:13< Espreon> This sucks...
20100505 03:53:21< AI0867> convert does work, identify just says PaletteMatte for everything
20100505 03:53:27< AI0867> even when it shouldn't
20100505 03:53:43< AI0867> I'm now using the GIMP for identification...
20100505 03:54:01-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100505 03:56:27-!- happygrue_ is now known as happygrue
20100505 04:03:02-!- King_Elendil [~King_Elen@75.143.233.138] has joined #wesnoth-dev
20100505 04:04:37< AI0867> now running optipng
20100505 04:05:27-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100505 04:06:15-!- shadowm_laptop [~ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 260 seconds]
20100505 04:07:19-!- Crab_ [~Crab_@wesnoth/developer/crab] has quit [Quit: Leaving.]
20100505 04:12:41< CIA-10> ai0867 * r42474 /trunk/data/campaigns/Under_the_Burning_Suns/images/units/ (331 files in 6 dirs): Convert UtBS unit images to RGB mode. They have been recompressed with optipng
20100505 04:13:09-!- King_Elendil [~King_Elen@75.143.233.138] has quit [Read error: Connection reset by peer]
20100505 04:15:40< AI0867> Espreon: ^
20100505 04:15:51-!- King_Elendil [~King_Elen@75.143.233.138] has joined #wesnoth-dev
20100505 04:17:22< Espreon> AI0867: What parameters did you use with convert?
20100505 04:24:05< CIA-10> ai0867 * r42475 /trunk/utils/indexed2rgb.sh: Check in indexed2rgb.sh script, so the imagemagick docs don't need to be read again for this task
20100505 04:24:06< AI0867> Espreon: ^
20100505 04:24:28< Espreon> Awesome, thanks!
20100505 04:24:29< AI0867> ugh, I may have just made a big mistake
20100505 04:24:49< Espreon> Uh....
20100505 04:24:52< Espreon> What?
20100505 04:25:01< AI0867> thankfully, not
20100505 04:25:17< AI0867> I previously had to add 'Matte' or transparency disappeared
20100505 04:25:28< Espreon> OK, good.
20100505 04:42:00< Espreon> AI0867: Wait, add "Matte" to what?
20100505 04:42:58-!- Ivanovic_ [~ivanovic@dtmd-4db2aa48.pool.mediaWays.net] has joined #wesnoth-dev
20100505 04:45:59-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 252 seconds]
20100505 04:46:56-!- Ivanovic_ is now known as Ivanovic
20100505 04:47:14< AI0867> Espreon: 'Matte' is imagemagick's name for 'Alpha'
20100505 04:47:56< Espreon> AI0867: Where in the convert line do I put it, and in what form?
20100505 04:49:00< AI0867> read the script
20100505 04:49:22-!- King_Elendil [~King_Elen@75.143.233.138] has quit [Read error: Connection reset by peer]
20100505 04:50:03-!- King_Elendil [~King_Elen@75.143.233.138] has joined #wesnoth-dev
20100505 04:50:35< Espreon> AI0867: But, it is not present.
20100505 04:50:42< Espreon> ... that is why I wanted to know...
20100505 04:50:54< AI0867> TrueColorMatte
20100505 04:54:33< AI0867> gimp and imagemagick wildly disagree on whether the images are in palette mode though...
20100505 04:56:22< CIA-10> ai0867 * r42476 /trunk/utils/indexed2rgb.sh: Change the usage of an option somewhat
20100505 04:57:57-!- Upthorn [~ogmar@75.26.184.137] has joined #wesnoth-dev
20100505 05:00:10< AI0867> the fire ghost is used in both UtBS and DW. Maybe it should be moved into core?
20100505 05:02:28-!- King_Elendil [~King_Elen@75.143.233.138] has quit [Read error: Connection reset by peer]
20100505 05:03:08-!- King_Elendil [~King_Elen@75.143.233.138] has joined #wesnoth-dev
20100505 05:03:09< AI0867> and we have an explanation:
20100505 05:03:09< AI0867> The PNG developer says:
20100505 05:03:10< AI0867> I think we mentioned already that "identify" doesn't tell them anything about format that PNG uses. If they only have black pixels, "identify" will tell them it's bilevel no matter how they stored it:
20100505 05:06:51< Espreon> AI0867: I guess... But if the idea is nixed, I'll have to remind myself to have the DW version also use the elemental trait.
20100505 05:21:24-!- King_Elendil [~King_Elen@75.143.233.138] has quit [Read error: Connection reset by peer]
20100505 05:22:02-!- FAAB [~huajie@219.142.242.206] has joined #wesnoth-dev
20100505 05:22:06-!- King_Elendil [~King_Elen@75.143.233.138] has joined #wesnoth-dev
20100505 05:39:23-!- Gambit [~Gambit@pa-67-234-73-7.dhcp.embarqhsd.net] has quit [Read error: Connection reset by peer]
20100505 05:41:14-!- King_Elendil [~King_Elen@75.143.233.138] has quit [Read error: Connection reset by peer]
20100505 05:42:32-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100505 05:52:32-!- dtiger [~dtiger@dynamic-vpdn-93-125-68-175.telecom.by] has joined #wesnoth-dev
20100505 06:09:35-!- dtiger [~dtiger@dynamic-vpdn-93-125-68-175.telecom.by] has quit [Remote host closed the connection]
20100505 06:11:35-!- Blarumyrran [~Blarumyrr@84-50-143-71-dsl.rkv.estpak.ee] has joined #wesnoth-dev
20100505 06:23:01-!- Blueblaze [~nick@adsl-99-158-45-140.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 245 seconds]
20100505 06:41:01-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100505 06:48:28-!- ancestral [~ancestral@97-116-122-90.mpls.qwest.net] has joined #wesnoth-dev
20100505 06:51:34-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100505 07:20:15< FAAB> trunk doesn't compile anymore. The source file portrait.cpp is missing in src/Makefile.am
20100505 07:34:07-!- fendrin [~fabi@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20100505 07:38:23< FAAB> shikadibot: dir trunk
20100505 07:38:23< shikadibot> Web interface URL to directory trunk: http://svn.gna.org/viewcvs/wesnoth/trunk
20100505 08:02:33-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20100505 08:03:02-!- ilor__ [~ilor@aut175.neoplus.adsl.tpnet.pl] has joined #wesnoth-dev
20100505 08:07:19-!- ancestral [~ancestral@97-116-122-90.mpls.qwest.net] has left #wesnoth-dev []
20100505 08:27:38-!- fakedrake [~fakedrake@ppp-94-64-230-186.home.otenet.gr] has quit [Ping timeout: 240 seconds]
20100505 08:30:54-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20100505 08:31:06< mordante> servus
20100505 08:31:55-!- ilor [~ilor@wesnoth/developer/ilor] has joined #wesnoth-dev
20100505 08:33:54-!- ilor__ [~ilor@aut175.neoplus.adsl.tpnet.pl] has quit [Ping timeout: 260 seconds]
20100505 08:34:05< FAAB> anyone working on AI around ?
20100505 08:35:12-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100505 08:35:37< FAAB> I wonder what is the impact of rev 42310 on custom AI [http://svn.gna.org/viewcvs/wesnoth?rev=42310&view=rev]
20100505 08:38:26< FAAB> I got a formula AI working but it gets broken by that revision, it keeps raising "UNABLE TO FALLBACK, fallback ai is NULL"
20100505 08:38:45< FAAB> the error is raised from src/ai/testing/stage_fallback.cpp
20100505 08:39:13< mordante> best ask Crab_
20100505 08:42:33< mordante> YogiHH grid cells are `invisible' the widget needs to provide the background
20100505 08:43:30< FAAB> discard my comment on src/Makefile.am, it was a conflict with a local version
20100505 08:56:11-!- ancestral [~ancestral@97-116-122-90.mpls.qwest.net] has joined #wesnoth-dev
20100505 09:02:24-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Quit: crimson_penguin]
20100505 09:06:02-!- euschn [~eugen@wesnoth/developer/euschn] has joined #wesnoth-dev
20100505 09:43:39-!- Crab_ [~Crab@wesnoth/developer/crab] has joined #wesnoth-dev
20100505 09:43:44< Crab_> hi, FAAB
20100505 09:44:48< Crab_> FAAB: the impact should be 'old default ai no longer available, old code and saves  work almost the same'. if there's an error, it might be a bug, please copypaste or link to your ai config
20100505 09:46:43< FAAB> hi Crab_
20100505 09:47:14< FAAB> what should I do with my fallback stage. It was refering to default AI (see http://wesnoth.pastebin.com/6yjUtUWd)
20100505 09:48:28< Crab_> FAAB: can  you show your entire ai config ?
20100505 09:48:33< FAAB> I have seen you have remove the fallback stage on the poison formula AI, but if I do that, the AI doesn't do anything
20100505 09:48:34< Crab_> not just that stage
20100505 09:49:34< FAAB> ok, it is here http://wesnoth.pastebin.com/qAQdYx2S
20100505 09:52:06-!- Ivanovic [~ivanovic@dtmd-4db2aa48.pool.mediaWays.net] has quit [Changing host]
20100505 09:52:06-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20100505 09:53:24< Ivanovic> moin
20100505 09:53:45< Crab_> FAAB: a quick (but slightly wrong) fix would be copypaste the main loop from data/src/ai/dev/formula_ai.cfg
20100505 09:53:46< mordante> hi Ivanovic 
20100505 09:54:09< Crab_> FAAB: also, you can just remove the [ai] section from testing_ai_default::fallback stage and see if it works.
20100505 09:54:59< FAAB> you mean the code I first pasted (last stage of the AI) ?
20100505 09:55:21< Crab_> yes, from that last stage
20100505 09:56:02< FAAB> ok I will try that, is that a correct fix ? ;)
20100505 09:56:25< Crab_> it's no more incorrect than your initial code, so, it's ok :)
20100505 09:57:21< Crab_> you see, your code has three stages
20100505 09:57:31< Ivanovic> AI0867: will you run the script mean to fix the images on branches/1.8, too?
20100505 09:57:43< Ivanovic> to make sure that images don't end up broken again in there either
20100505 09:58:03< Crab_> FAAB: so, firstly, 'jinx attack' and 'cursed movement' are done. and then, the turn is played as normal
20100505 09:59:09< Crab_> FAAB: the problem with that is, that, if after several moves/attacks, a new 'interesting' jinx/cursed move appears,  it won't be done as the first 2 stages are  history at this point (until the next turn, that is)
20100505 09:59:39< Crab_> it is usually better to 'inject' things into main ai turn sequence at the correct places (priorities)
20100505 10:00:49< FAAB> yeah, that is how I would like to do it, but I didn't figure out how to do it :)
20100505 10:00:54< Crab_> (your code will work, too; just note that feature)
20100505 10:02:58< FAAB> but anyway, I might try to improve the AI when I switch to lua AI as you told me formula AI will not be updated anymore
20100505 10:04:00< FAAB> I just hope I can still use the formula AI until the switch is possible
20100505 10:05:03< Crab_> FAAB: see http://wesnoth.pastebin.com/LriqjgyU
20100505 10:06:24< Crab_> caveat 1: your candidate actions must return a score related to scores of other candidate actions in main_loop.
20100505 10:06:49< Crab_> caveat 2: if you needed to include {core/macros/ai_candidate_actions.cfg}, you still need.
20100505 10:07:21< Crab_> caveat 3: set max_score to the ceil (max) score you can return from your evaluation function, to speed things up.
20100505 10:07:49< Crab_> caveat 4: make sure your filter and evaluation are fast
20100505 10:08:04< FAAB> that rocks !
20100505 10:08:48< Crab_> (the code should work, if it does not, file a bug, i'll fix it today if there's any glitch)
20100505 10:09:42< FAAB> do you mean it should fix the problem I have with rev 42310 ?
20100505 10:10:04< Crab_> yes, of course, at it uses the new rca ai instead of fallback
20100505 10:10:14< Crab_> use --log-debug=ai/config --log-debug=ai/mod  and :inspect to see how it handles the config
20100505 10:10:31< FAAB> ok, I will try now
20100505 10:10:41< Crab_> use --log-debug=ai/stage/rca to debug its behavior at runtime
20100505 10:11:41< Crab_> also, you should assign id= attribute to your candidate actions. for prettier log output, and to allow you to delete them later if needed.
20100505 10:12:21< FAAB> ok I will
20100505 10:13:39< Crab_> also note that you can throw things out of the 'base' ai by using "[modify_ai] action=delete path=stage[main_loop].candidate_action[id]"
20100505 10:13:51< Crab_> where id is the id of the candidate action
20100505 10:14:26< Crab_> this can be used to alter the ai mid-scenario
20100505 10:14:43< Crab_> as you can add/delete things on the fly
20100505 10:15:41< Crab_> FAAB: oops, and I've got a typo in there
20100505 10:15:48< Crab_> instead of path=side['main_loop'].candidate_action use path=stage['main_loop'].candidate_action
20100505 10:16:11< Crab_> or, even more correct, path=stage[main_loop].candidate_action     (no quotes)
20100505 10:16:19< Crab_> sorry :)
20100505 10:16:41< FAAB> alright
20100505 10:17:50< FAAB> lot of info to digest, so it might take me some time :)
20100505 10:18:06< FAAB> thanks for the help, though !
20100505 10:29:59-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20100505 10:30:41-!- stikonas [~and@bcm-131-111-247-104.girton.cam.ac.uk] has joined #wesnoth-dev
20100505 10:30:41-!- stikonas [~and@bcm-131-111-247-104.girton.cam.ac.uk] has quit [Changing host]
20100505 10:30:41-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100505 10:38:42-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: GO, GET TO THE CHOPPAH!!!]
20100505 10:39:10-!- ancestral [~ancestral@97-116-122-90.mpls.qwest.net] has quit [Quit: And that’s the end of THAT chapter.]
20100505 10:45:04-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100505 10:46:13-!- Netsplit *.net <-> *.split quits: shadowmaster, Rhonda
20100505 10:47:31-!- Netsplit over, joins: shadowmaster, Rhonda
20100505 10:47:34-!- shadowmaster [~ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 246 seconds]
20100505 10:47:42-!- shadowmaster [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100505 10:50:04-!- deekay [~dk@wesnoth/developer/dragonking] has joined #wesnoth-dev
20100505 10:51:02< FAAB> Crab_: it is working great !
20100505 10:51:16< Crab_> cool :)
20100505 10:51:42< FAAB> Crab_: I assume the code is not compatible with 1.8
20100505 10:51:55< Crab_> let me check...
20100505 10:52:10< Crab_> the latest change is 'allow modify_ai to be used without side= in [side]'
20100505 10:52:20< Crab_> let me check when it was done
20100505 10:53:36< Crab_> wesbot: log 42103
20100505 10:53:37< wesbot> crab * r42103 : implement espreon's FR about making modify_ai in [side] work without side= attribute
20100505 10:53:40< wesbot> URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=42103
20100505 10:53:52< Crab_> apr 11, it is
20100505 10:53:59< Crab_> so yes, for that to work in 1.8, you need side= in [modify_ai]
20100505 10:55:37< FAAB> well I changed that part, my code is now: http://wesnoth.pastebin.com/f5WzUyvT
20100505 10:56:33< Crab_> if you inline the ai config, you can do that : http://wesnoth.pastebin.com/N6D6AGa9
20100505 10:56:37< Crab_> that is 1.8 compatible, btw
20100505 10:57:14< Crab_> (there's no need to use modify_ai if you control the entire ai config)
20100505 10:57:31< FAAB> that is cleaner indeed :)
20100505 10:57:37< Crab_> just add all those candidate_actions (order doesn't matter) to a one testing_ai_default::candidate_action_evaluation_loop stage
20100505 10:59:35< Crab_> FAAB: btw, ping me when/if your candidate actions and formula ai code become relatively stable - I'll try porting them to lua, to check what things are needed there.
20100505 10:59:57< Crab_> there's always need for a good use cases :)
20100505 11:00:31< Crab_> FAAB: and add max_score= attribute, if you can
20100505 11:00:47< Crab_> imagine a situation: you set 20 gotos for your ai.
20100505 11:01:33< Crab_> if, on your formula_ai candidate actions, you have a max_score=SOMETHING_LOWER_THAN_GOTO_SCORE , then there will be no need to evaluate your formula_ai candidate actions till all the gotos are dealt with
20100505 11:03:06< Crab_> i.e. if goto evaluates to true, then, what's the point of evaluating that formula_ai code if the maximum that that code can return is lower than the current 'best' score of the goto ?'
20100505 11:03:56< FAAB> got it, and add an id= too ...
20100505 11:04:02< mordante> FAAB, I have a problem (gcc-4.5) with addon_management.cpp:808 you try to convert a boolean to pointer, did you intent two NULL's before the false?
20100505 11:04:21< mordante> (the last false(
20100505 11:04:26< mordante> )*
20100505 11:05:40< FAAB> mordante: let me check
20100505 11:06:36< FAAB> I am ashamed to say it is a copy paste of the code you find line 1001 :/
20100505 11:07:59< mordante> FAAB, it's not, count the number of parameters
20100505 11:09:27< FAAB> right
20100505 11:10:05< FAAB> so there should be 2 null pointers there
20100505 11:11:03< FAAB> I will send a patch
20100505 11:12:37< mordante> FAAB, ok, I can also do it right no
20100505 11:12:39< mordante> now*
20100505 11:13:07< FAAB> thanks, sorry for the mess
20100505 11:13:46< mordante> no problem
20100505 11:20:26< CIA-10> mordante * r42477 /trunk/src/addon_management.cpp: 
20100505 11:20:27< CIA-10> Add two NULL parameters.
20100505 11:20:27< CIA-10> Problem discovered by gcc 4.5.
20100505 11:20:57< mordante> FAAB, ^
20100505 11:22:01< Ivanovic> mordante: there is just a simple problem with moving the scholarship into the "post gsoc time": when shall we release a new stable series?
20100505 11:22:16< Ivanovic> that is: which "really new and bigger projects" would be possible then?
20100505 11:22:27< Ivanovic> most work would have to happen in a branch which is not this nice IMO
20100505 11:22:49< mordante> Ivanovic, of course that's a disadvantage
20100505 11:22:59< mordante> but IMO we need somebody to mentor a student
20100505 11:24:29< Ivanovic> i am not really sure about the border between mentorship and scholarship
20100505 11:24:47< Ivanovic> that is: even in gsoc with a really strong student things are more like a scholarship
20100505 11:24:57< Ivanovic> there is some basic guidance but that's often it
20100505 11:25:16< Ivanovic> the mentor is basically someone who is around in case problems arise
20100505 11:25:57< Ivanovic> so if everything goes smooth and perfectly it is a rather relaxed job for the mentor talking to the student regulary to make sure that there are no real problems and to assist in some bigger design decisions
20100505 11:26:19< Ivanovic> the real intensive work from a mentor is only required if some bad problems come up
20100505 11:26:54< Ivanovic> but this is one of the things that are impossible to forsee
20100505 11:27:02< Ivanovic> you don't know if problems will come up or not
20100505 11:29:17< mordante> exactly that's why we need a mentor for the scholarship student
20100505 11:29:42< mordante> and if we get a really strong student the mentoring can be done by a gsoc mentor
20100505 11:29:55< mordante> but we can't be sure the student is strong
20100505 11:36:00< Ivanovic> we could try to only select a really strong one
20100505 11:36:14< Ivanovic> though what shall we do if no "really strong one" does propose
20100505 11:36:36< Ivanovic> should we then say "sorry, we won't take anyone since none of you was strong enough"?
20100505 11:36:40< Ivanovic> sounds bad to me, too
20100505 11:37:22< mordante> exactly my problem as well, if we run the program we need to go through with it regardless who applies 
20100505 11:37:57< mordante> and I think such strong students are rare
20100505 11:38:21< Ivanovic> yeah
20100505 11:38:39< Ivanovic> though the scholarship itself is *NOT* limited at students only
20100505 11:38:46< Ivanovic> but those are the ones most likely to apply
20100505 11:38:47< mordante> (and it's not really unrealistically to imagine such a student is already working for gsoc)
20100505 11:39:05< Crab_> then the projects will need to be flexible enough to allow to scale it down to match the student
20100505 11:39:31< mordante> I know it's not limited to students, but I either expect a student or somebody between jobs
20100505 11:39:51< Ivanovic> yes, sounds reasonable
20100505 11:39:55< Ivanovic> afk for some hours
20100505 11:40:37< mordante> Crab_, but how much do we want to scale it down
20100505 11:46:36< Crab_> mordante: to match the student's abilities and 'available time'
20100505 11:47:56< Crab_> as we don't know these 'in advance', we have two choices : make some upfront requirements and screen for them (which we don't want to do as this can lead to 'no students accepted because no students match requirements') or make the projects scalable in both directions.
20100505 11:48:53-!- Appleman1234 [~Appleman1@CPE-124-191-176-143.oxqn1.cha.bigpond.net.au] has joined #wesnoth-dev
20100505 11:49:57< mordante> Crab_, well since we also offer a sum of money I don't want to scale down too far
20100505 11:50:09< Crab_> the money is scalable variable, as well
20100505 11:50:11< mordante> and I think a lesser student can succeed if properly mentored
20100505 11:50:24< mordante> Crab_, ok
20100505 11:51:10< Crab_> yes, 'properly mentoring' is also a form of 'flexibility' - e.g., if a certain project can be made much easier by doing some upfront work by mentor and devoting more time to mentor the student - it's good.
20100505 11:54:04< mordante> Crab_, I don't have time for that. Do you?
20100505 11:54:54< Crab_> I have two free days per week, one of them is usually devoted to d&d. so, there's time left.
20100505 11:55:55-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has quit [Ping timeout: 248 seconds]
20100505 11:58:09-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20100505 11:58:56< mordante> well if you have time and are willing to mentor that student, it would make my concern void
20100505 11:59:06-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100505 12:01:25< Crab_> mordante: it depends on the project
20100505 12:08:03< mordante> ok
20100505 12:12:05-!- Vetinari [~lukjad007@unaffiliated/lukjad007] has joined #wesnoth-dev
20100505 12:20:48-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Remote host closed the connection]
20100505 12:21:17-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100505 12:23:50-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Remote host closed the connection]
20100505 12:26:11-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100505 12:43:05-!- timotei [timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20100505 12:45:13-!- fendrin [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20100505 13:04:15-!- timotei [timotei@wesnoth/developer/timotei] has quit [Quit: Leaving]
20100505 13:29:28-!- Blarumyrran [~Blarumyrr@84-50-143-71-dsl.rkv.estpak.ee] has quit [Ping timeout: 276 seconds]
20100505 13:45:25-!- Appleman1234 [~Appleman1@CPE-124-191-176-143.oxqn1.cha.bigpond.net.au] has quit [Quit: Leaving]
20100505 13:50:21-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20100505 13:56:47< Ivanovic> re
20100505 14:03:52-!- fakedrake [~fakedrake@athedsl-419513.home.otenet.gr] has joined #wesnoth-dev
20100505 14:10:06-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100505 14:12:31< fendrin> Crab_: Hello
20100505 14:12:42< fendrin> Crab_: http://www.wesnoth.org/forum/viewtopic.php?p=426456#p426456
20100505 14:13:21< fendrin> Crab_: I removed the assassins from the southern orc leader in 1.8.1 two days ago :-)
20100505 14:15:07< fendrin> Crab_: Are you sure the new uprooting works like expected? Faster deployment of the orcs (wolf riders) , many assassins...
20100505 14:15:47< Crab_> fendrin: I haven't touched 1.8.1, I've touched only trunk
20100505 14:16:02< fendrin> Crab_: That may make the scenario more interesting but still a first scenario shouldn't scare the players away.
20100505 14:16:30< Crab_> fendrin: it's very easy on easy, ok on normal, and quite beatable on 1st try on hard
20100505 14:16:53< Crab_> at least for me. I've lowered the AI initial gold
20100505 14:17:22< Crab_> have you played the new version ? any feedback ?
20100505 14:17:56< fendrin> Crab_: No, haven't touched it. My plan was to balance 1.8.x and port all the changes to 1.9 afterwards.
20100505 14:18:07-!- Ken_Oh [~briang@static-71-178-174-220.washdc.fios.verizon.net] has joined #wesnoth-dev
20100505 14:18:33< fendrin> I have also reduced the gold for every ai in 14. That seems to be much to hard at least on HARD.
20100505 14:18:34< Crab_> well, my 1.9 changes are less a balance and more a 'rework to make it less simular to httt 1'. I've got a wip on LoW 2 rework, as well (no L2 trolls in there)
20100505 14:18:41< Crab_> ok
20100505 14:19:11< fendrin> No more trolls at lvl2 in LoW2?
20100505 14:19:55< fendrin> Crab_: I have changed the recruit list as well. The player(s) can only recruit fighters/archer in scenario 1-3. 
20100505 14:20:21< fendrin> From 4 on they can recruit scouts and with the freeing of Cleodil they get shamans.
20100505 14:20:25< Crab_> fendrin: those changes are not committed yet ?
20100505 14:20:46< fendrin> No, not yet. And the work is done on 1.8 branch.
20100505 14:21:01< fendrin> That may need a rebalancing as well.
20100505 14:21:23< Crab_> any reason for player not having shamans ?
20100505 14:21:31< fendrin> But it fits the introduction of the recruit list per leader feature.
20100505 14:21:44< Crab_> and any reason for orcs not using many assassins to take advantage of the fact ?
20100505 14:22:25< fendrin> Crab_: I want that in singleplayer Cleodil is the shaman recruiter. Kalenz and Lander can recruit everything else. Olurf can recruit dwarves only.
20100505 14:22:57< Crab_> ok. and what about orcs ? shouldn't they take advantage of the fact that elves don't have shamans and use multiple assasins ?
20100505 14:23:28< fendrin> Crab_: There is no reason. If you changes work I am fine with them. Removing the assassins was an atempt to balance the game. But it is surely possible to balance it in another way as well.
20100505 14:23:56< Crab_> about 'recruit list per leader' - I've got some WIP code on that, seems easy enough to add, but i'm wondering how the UI should display the 'available but not with nearby leaders' units - should it display them 'in red' or not display at all at the recruit list ?
20100505 14:24:27-!- Gambit [~Gambit@pa-67-234-73-7.dhcp.embarqhsd.net] has joined #wesnoth-dev
20100505 14:24:30< fendrin> Crab_: I would like to display them in red :-)
20100505 14:24:42< Crab_> my WIP does it as 'introduce a SUF for leader which filters 'what part of side recruit list they can use'
20100505 14:25:00< Crab_> e.g., Olurf would have race=dwarf  in that SUF
20100505 14:25:48< Crab_> maybe allow both Kalenz and Landar to recruit shamans in LoW 1 ?
20100505 14:26:07< Crab_> but warn in-game that 'after we flee, we'll no longer be able to recruit them'
20100505 14:26:41< fendrin> Crab_: You usually end up with some shamans because Verlorn is recruiting some.
20100505 14:27:00< Crab_> but you don't transfer Velon's units in MP, isn't it ?
20100505 14:27:49< fendrin> Right, that isn't done yet. No idea if I should do it. And the verlon shaman solution is very unpredictable. We could solve the problem by giving some shamans as starting units with them.
20100505 14:27:58< Gambit> Why aren't units transfered in MP by the way?
20100505 14:28:00< Crab_> (I'm ok with recruiting restrictions on Kalenz & Landar, I'm not ok with recruiting restrictions for orcs :) )
20100505 14:28:21< Gambit> Some scenarios can have just as much code as every unit in default.
20100505 14:28:56< fendrin> Gambit: I need code that splits them between the two players. It's not a great deal but not done yet.
20100505 14:29:25< fendrin> Gambit: Your statement about code and units is not understood by me.
20100505 14:29:36< Gambit> Maybe I misunderstood the problem.
20100505 14:29:53< Gambit> [unit_type]'s are not transfered across the internet in multiplayer (AFAIK)
20100505 14:29:56< fendrin> Crab_: That means you do want the orcs to use assassins?
20100505 14:31:15< Crab_> fendrin: In general, I want tactical variety and more challenges for player. assassins using poison is a nice tactical challenge, imo.
20100505 14:32:11< fendrin> Crab_: Yes, but there are too few villages to heal them. And those assassins take long to kill, they are hard to hit. Time is spare in that scenario.
20100505 14:32:32< Crab_> fendrin: (and feel free to balance the 1.8 scenario as you wish, I won't object)
20100505 14:32:48< Crab_> fendrin: well, there isn't time to heal in there anyway :)
20100505 14:33:15< fendrin> Crab_: I will play the 1.9 version if it works well it will replace the 1.8 one.
20100505 14:33:45< Crab_> fendrin: ok
20100505 14:34:28-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 276 seconds]
20100505 14:34:40< Crab_> fendrin: in general, on easy, it's possible to kill any leader, on normal, it is possible to kill either the southern or the eastern one, on hard, only the southern is possible without too much luck.
20100505 14:34:47< fendrin> Crab_: The ai in 14, aldar the loyalist does recruit bad. He should recruit much more horses to use in the open plane. No idea why that fails.
20100505 14:35:13< Crab_> afair, haven't tinkered with that ai yet, will take a look
20100505 14:35:37< Crab_> afair, the smartest thing that that ai can do is to delay the battle till dawn
20100505 14:35:55< fendrin> Crab_: Right, alda should act defensively.
20100505 14:36:21< Crab_> offensively during day, defensively during night
20100505 14:36:34< fendrin> Recruit nonhorses to guard the villages and some horses to do the hack and slaying.
20100505 14:37:56< Crab_> fendrin: when playing LoW1, note that the ai is very 'aggressive', by design (the orcs are careless). so, it's possible to draw initial recruits of the southern orcs away from him, opening the way to Wrulf.
20100505 14:38:22< fendrin> That is a good thing imho.
20100505 14:38:32< Crab_> that's the easiest way to beat it on hard (to avoid the need to hack through most of the 70% defense assasins)
20100505 14:38:46< Crab_> distract them and draw initial recruits somewhat 'off'.
20100505 14:39:19< Crab_> and scouts are a key to do that, since they have very him movement and can easily split the enemy initial recruits from their boss
20100505 14:39:47< Crab_> also note that the eastern orc is a tempting alternate target (but more risky)
20100505 14:40:01< Crab_> s/him/high
20100505 14:41:45< fendrin> hmmm, it's just that most campaigns play with an incomplete recruit list that grows and I always liked that feature.
20100505 14:42:14< Crab_> well, it's ok if the initial list allows for some variety in tactics..
20100505 14:42:54< Crab_> and we can drop the extra recruits in LoW 2
20100505 14:43:02< Crab_> since they're technically 'away from home'
20100505 14:43:30-!- Johannes13 [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20100505 14:43:39< Crab_> but this would still allow to recruit an initial set of units according to player's strategy
20100505 14:43:41< zookeeper> it's practically always a bad idea to remove access to recruits mid-campaign unless you also take away recalls
20100505 14:44:05< Crab_> zookeeper: what about 'making them more expensive' route ?
20100505 14:44:06< fendrin> zookeeper: why?
20100505 14:44:37< Crab_> (if we warn about it beforehand)
20100505 14:44:43< zookeeper> fendrin, because then the silly way of recruiting with all your remaining gold on the last turn becomes a good idea
20100505 14:44:46< fendrin> Crab_: The more expensive route is hacky to code. Have a look at utbs. At least if there is no c++ support for it.
20100505 14:45:00< zookeeper> Crab_, sure, i think that's ok, but as fendrin said, it's annoyingly hacky
20100505 14:45:29< Crab_> is there a reason why c++ doesn't support this ?
20100505 14:45:49< zookeeper> if you take away recruits at scenario 2, then the most important thing in scenario 1 is to spend all your gold recruiting those units so you have a nice recall list of experienceless lvl1 units
20100505 14:45:54< Crab_> e.g. 'does anything known prevent me from adding per-leader-per-scenario recruit costs?'
20100505 14:46:07< fendrin> zookeeper: I see and agree.
20100505 14:46:43< zookeeper> Crab_, no particular reason i guess. i'd imagine it'd still be a somewhat complicated feature.
20100505 14:46:46< fendrin> Crab_: Sounds nice.
20100505 14:46:54< zookeeper> since it'd be per-leader-per-scenario-per-unittype
20100505 14:47:05< fendrin> zookeeper: No, the xp modification framework can be used.
20100505 14:47:34< Gambit> a multiplier.
20100505 14:47:35< zookeeper> sure, if it's good enough to be able to adjust the cost like that, instead of per-unittype
20100505 14:47:56< zookeeper> but sure, a multiplier sounds perfectly fine for most purposes
20100505 14:49:16< fendrin> Crab_: What about [unit] \n canrecruit=yes \n recruit=elvish archer #14, elvish blah #17, ...
20100505 14:50:54< fendrin> where 14 is the cost for an elvish archer recruited by this unit.
20100505 14:51:31< fendrin> Crab_: Note that you would need to choice the cheapest recruiter of a unit if two or more leaders from the same side are in the keeps of the same castle.
20100505 14:51:49< Crab_> fendrin: I already need to find out all leaders which can recruit this unit.
20100505 14:52:21< Crab_> fendrin: so, it isn't too hard to find the cheapest one :)
20100505 14:52:57< fendrin> Yeah, that problem is surely in NP\NP-Hard.
20100505 14:53:30< zookeeper> well, if you wanted something really fancy, you could just allow [recruit_list] [unit] type=Elvish Archer cost=14 number=4 (?) [/unit] [/recruit_list] instead of recruit=
20100505 14:54:10< fendrin> zookeeper: Mighty but not kissy to code.
20100505 14:54:18< zookeeper> sure
20100505 14:54:24-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Remote host closed the connection]
20100505 14:54:27< zookeeper> especially not if there's multiple leaders with different recruit lists
20100505 14:54:30< fendrin> zookeeper: Maybe both syntaxes could coexist?
20100505 14:55:19< Crab_> zookeeper: to keep things simpler, I want a per-team recruit list to stay as it is now
20100505 14:55:36< Crab_> zookeeper: and just add a 'filter' to leaders to make them able to recruit only a subset of that list
20100505 14:56:00< Crab_> zookeeper: that way, dynamically adding/removing recruits will still work.
20100505 14:56:12-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100505 14:56:23< fendrin> Crab_: Clever.
20100505 14:59:13< FAAB> Crab_: do you remember how I wanted to see my custom AI in the list of AI when creating a game ?
20100505 14:59:34< Crab_> FAAB: yes
20100505 14:59:47< FAAB> I have take a shot at hit and send a patch, could you check it when you have time ?
20100505 14:59:55< FAAB> at it*
20100505 15:00:00< Crab_> ok, yes
20100505 15:00:15< Crab_> but only if you send it today or tomorrow - or in a 10 days :)
20100505 15:00:16< FAAB> https://gna.org/patch/index.php?1664
20100505 15:00:54< FAAB> are you going on holiday ?
20100505 15:01:16< Crab_> yes
20100505 15:01:20< Crab_> to Finland
20100505 15:02:40< FAAB> wow, that sounds nice :)
20100505 15:05:41< Crab_> FAAB: ok, the patch looks good.
20100505 15:06:33< Crab_> FAAB: I'll commit it later today.
20100505 15:06:59< FAAB> was that the way you saw it done ?
20100505 15:07:11 * FAAB cannot believe it :)
20100505 15:07:13< Crab_> generally, yes
20100505 15:07:28< Crab_> it can have some refactoring, yes. but the approach is very good.
20100505 15:07:33< CIA-10> ai0867 * r42478 /branches/1.8/data/campaigns/Under_the_Burning_Suns/images/units/ (331 files in 6 dirs): Convert UtBS unit images from indexed to RGBA, recompress with optipng
20100505 15:08:42< FAAB> do you see a way for me to make that AI taking over to default one. So it is selected whenever a computer player is selected ?
20100505 15:26:14< Crab_> FAAB: well, you can.
20100505 15:28:18< Crab_> but only for the AI selection list, not for internal value of 'default' ai configuration
20100505 15:33:00< Ivanovic> AI0867: have you run the script on everything of branches/1.8 or explicitly only on the utbs units?
20100505 15:33:02-!- Crab_ [~Crab@wesnoth/developer/crab] has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
20100505 15:34:57< Ivanovic> when running wesnoth-optipng on branches/1.8 i get this message several times:
20100505 15:34:59< Ivanovic> * processing ./data/core/images/halo/teleport-1.png (nice = 19)...
20100505 15:35:00< Ivanovic> convert: tRNS chunk has out-of-range samples for bit_depth `./data/core/images/halo/teleport-1.png.new' @ warning/png.c/PNGWarningHandler/1457.
20100505 15:35:04< Ivanovic> (of course for different files)
20100505 15:35:15< Ivanovic> does anyone know what this might be about and if it can cause problems?
20100505 15:37:00< Ivanovic> here is a complete log of an optipng run on my machine (it is a 1.2MB textfile!):
20100505 15:37:01< Ivanovic> http://files.wesnoth.org/optipng-logs.txt
20100505 15:37:16< Ivanovic> installed packages/versions: app-arch/advancecomp-1.15, media-gfx/optipng-0.6.4 and media-gfx/imagemagick-6.6.1.2
20100505 15:46:49-!- Darkas [~quassel@ppp-93-104-128-7.dynamic.mnet-online.de] has joined #wesnoth-dev
20100505 15:56:37-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Remote host closed the connection]
20100505 15:58:32-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100505 15:58:59-!- _jbx_ [~jbailey@12.190.80.225] has joined #wesnoth-dev
20100505 16:04:41< CIA-10> mordante * r42479 /trunk/src/gui/widgets/ (grid.cpp grid.hpp): 
20100505 16:04:41< CIA-10> Revert r42453.
20100505 16:04:41< CIA-10> Forgot that a vector copies on resize, which causes problems.
20100505 16:19:33-!- fakedrake [~fakedrake@athedsl-419513.home.otenet.gr] has quit [Ping timeout: 265 seconds]
20100505 16:19:55-!- billynux [~billy@wesnoth/developer/billynux] has joined #wesnoth-dev
20100505 16:33:47-!- gabba [~gabba@wesnoth/developer/gabba] has joined #wesnoth-dev
20100505 16:34:13< gabba> hi
20100505 16:34:53< billynux> hi gabba
20100505 16:34:59< gabba> hi billynux
20100505 16:39:52-!- ilor [~ilor@wesnoth/developer/ilor] has quit [Quit: Konversation terminated!]
20100505 16:40:09-!- ilor [~ilor@wesnoth/developer/ilor] has joined #wesnoth-dev
20100505 16:43:16-!- swathanthran [~user@unaffiliated/shyam-k/x-8459115] has joined #wesnoth-dev
20100505 16:43:34-!- swathanthran [~user@unaffiliated/shyam-k/x-8459115] has left #wesnoth-dev ["ERC Version 5.3 (IRC client for Emacs)"]
20100505 17:00:11-!- Johannes13_ [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20100505 17:01:33-!- euschn [~eugen@wesnoth/developer/euschn] has quit [Quit: Leaving.]
20100505 17:03:15-!- Johannes13 [~Johannes@unaffiliated/johannes13] has quit [Ping timeout: 260 seconds]
20100505 17:07:08-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100505 17:09:30-!- stikonas [~and@bcm-131-111-247-104.girton.cam.ac.uk] has joined #wesnoth-dev
20100505 17:09:30-!- stikonas [~and@bcm-131-111-247-104.girton.cam.ac.uk] has quit [Changing host]
20100505 17:09:30-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100505 17:11:14< AI0867> Ivanovic: only utbs units. as all of those were indexed and the rest of utbs wasn't
20100505 17:11:46< Ivanovic> AI0867: it might be a good idea to have it run on all of trunk/branches/1.8 just to be sure
20100505 17:11:51< AI0867> the script doesn't actually check if the image is in indexed mode before converting it yet, and running optipng on *everything* takes quite a bit of time
20100505 17:11:57< Ivanovic> since those convert warnings do sound suspicious
20100505 17:12:12< Ivanovic> yes, it takes more than an hour on my quadcore box
20100505 17:12:22< AI0867> I guess I'll try upgrading the script
20100505 17:12:43< AI0867> imagemagick's identify doesn't actually tell you this though, so I'll need to add a different dependency
20100505 17:12:50< AI0867> pngcheck seems to work
20100505 17:16:08< AI0867> pngcheck says OK: src/wesnoth-git2/data/core/images/halo/teleport-1.png (58x58, 16-bit grayscale+alpha, non-interlaced, 76.8%).
20100505 17:16:30< AI0867> do we want to convert those too?
20100505 17:16:43< AI0867> and what about RGB (no alpha) images?
20100505 17:18:58< Ivanovic> i guess that the best would probably be to have everything as rgb+alpha
20100505 17:19:24< Ivanovic> reason: this way we can simply list this as "save your file as XYZ and it will work!"
20100505 17:20:40-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has joined #wesnoth-dev
20100505 17:36:48< billynux> mordante, around?
20100505 17:38:02-!- billynux [~billy@wesnoth/developer/billynux] has quit [Quit: Leaving]
20100505 17:39:00< gabba> Any git-svn users around? I had problems with "git svn rebase/dcommit" erasing my local commits, and I need help diagnosing what happened.
20100505 17:42:50-!- elias [~elias@allegro/developer/allefant] has joined #wesnoth-dev
20100505 17:43:56< CIA-10> ivanovic * r42480 /trunk/ (7 files in 6 dirs): updated Spanish translation
20100505 17:43:58< CIA-10> ivanovic * r42481 /branches/1.8/ (9 files in 8 dirs): updated Spanish translation
20100505 17:46:20-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100505 17:52:53< CIA-10> gabba * r42482 /trunk/src/ (game_display.cpp game_display.hpp): Small adjustments to multiple fake units code (made an accessor const, optimized a loop, uniformized braces)
20100505 17:53:39-!- Lastmerlin [~Miranda@kalypso.csn.tu-chemnitz.de] has joined #wesnoth-dev
20100505 17:55:32< CIA-10> ivanovic * r42483 /trunk/po/ (467 files in 26 dirs): 
20100505 17:55:32< CIA-10> make sure that svn:eol-style is set to native for all .po and .pot files
20100505 17:55:32< CIA-10> command used (from inside po/):
20100505 17:55:32< CIA-10> for j in wesnoth*; do cd $j; for i in *.po*; do svn propset svn:eol-style native $i@; done; cd ..; done
20100505 17:56:48< Soliton> gabba: well, what happened?
20100505 17:56:50< AI0867> Ivanovic: currently re-optimizing 404 files converted in trunk
20100505 17:57:02< AI0867> most of those seem to be terrain graphics
20100505 17:57:08< AI0867> so it'll be less in 1.7
20100505 17:57:14< AI0867> s/1\.7/1.8
20100505 17:57:32< CIA-10> ivanovic * r42484 /branches/1.8/po/ (414 files in 25 dirs): 
20100505 17:57:32< CIA-10> make sure that svn:eol-style is set to native for all .po and .pot files
20100505 17:57:32< CIA-10> command used (from inside po/):
20100505 17:57:32< CIA-10> for j in wesnoth*; do cd $j; for i in *.po*; do svn propset svn:eol-style native $i@; done; cd ..; done
20100505 17:58:15< Ivanovic> AI0867: looking at the info from my wesnoth-optipng run: there might be several in 1.8, too
20100505 17:58:19< Ivanovic> (that run was on 1.8!)
20100505 17:58:41< gabba> Soliton: gotta go in 2 min unfortunately, but quick summary: I had a local commit, did "git svn rebase", and the local commit just disappeared from the git history (fortunately the files themselves still had my changes, so I was able to commit again)
20100505 17:59:00< AI0867> ugh, I just ran into an issue
20100505 17:59:21< AI0867> the specific convert command I use doesn't convert 16 bit grayscale to RGB
20100505 18:00:02< AI0867> it probably would with 16 bit-per-channel RGB (64bit per pixel in total)
20100505 18:00:10< Soliton> gabba: never happened to me. maybe it was just a conflict in re-applying the commits after the rebase?
20100505 18:00:14< AI0867> more special casing
20100505 18:00:18< gabba> Soliton: earlier, I tried a git svn dcommit, but that was interrupted by an svn commit to the server by someone else. Result: the dcommit was interrupted, and I lost two or three local commits
20100505 18:00:33< AI0867> gabba: that's a known issue with git-svn
20100505 18:01:00< AI0867> if the connection is interrupted after at least one commit went through, the rest remains in the index and working directory, but the commits themselves are lost
20100505 18:01:22< AI0867> shadowmaster is hit by it regularly
20100505 18:01:38< gabba> AI0867: really? that's kind of a letdown, after hearing so many people sing its praises :(
20100505 18:02:01< gabba> Anyways gotta go
20100505 18:02:11-!- gabba [~gabba@wesnoth/developer/gabba] has left #wesnoth-dev []
20100505 18:02:14< Soliton> i'd guess people praise git more than git-svn.
20100505 18:02:29< Soliton> it's always going to be a hack afterall.
20100505 18:10:33-!- Blueblaze [~nick@adsl-99-158-45-140.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100505 18:11:07< Lastmerlin> hello
20100505 18:11:28< FAAB> Crab_: it seems the simplified syntax  you gave me earlier (http://wesnoth.pastebin.com/N6D6AGa9) is not working.
20100505 18:11:35< Lastmerlin> Can anyone tell me, in which wesnoth file the locale change for gettext is done ?
20100505 18:11:46< Lastmerlin> I want to see, what is passed to the setlocale function
20100505 18:12:18< Lastmerlin> my issue is, that the strings that work for linux (for instance en_GB or en_GB.utf8) do nothing for windows
20100505 18:12:42< Lastmerlin> alternatively, if someone can tell me what the problem is directly, I will not complain ;)
20100505 18:12:54< FAAB> Crab_: I filed a bug report with a test case. Check it out https://gna.org/bugs/index.php?15994
20100505 18:14:20-!- Dragonking [~dk@wesnoth/developer/dragonking] has joined #wesnoth-dev
20100505 18:18:43-!- deekay [~dk@wesnoth/developer/dragonking] has quit [Ping timeout: 276 seconds]
20100505 18:23:37-!- FAAB [~huajie@219.142.242.206] has quit [Quit: Leaving.]
20100505 18:30:08-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100505 18:54:25-!- billynux [~c8078d05@wesnoth/developer/billynux] has joined #wesnoth-dev
20100505 18:54:26< AI0867> rgb=0;rgba=0;gray=0;graya=0;other=0;for i in `find . -iname '*.png'`;do result=`pngcheck $i`;if echo $result|grep 'RGB+alpha'>/dev/null;then rgba=$(($rgba+1));elif echo $result|grep 'RGB'>/dev/null;then rgb=$(($rgb+1));elif echo $result|grep 'grayscale+alpha'>/dev/null;then graya=$(($graya+1));elif echo $result|grep 'grayscale'>/dev/null;then grey=$(($grey+1));else other=$(($other+1));echo $result;fi;done;echo "RGBA: $rgba RGB: $rgb GRAY+A: $graya
20100505 18:54:33< AI0867> RGBA: 7807 RGB: 196 GRAY+A: 162 GRAY: 0 OTHER: 9
20100505 18:58:47< CIA-10> eleazar * r42485 /trunk/data/core/ (3 files in 2 dirs): converted the forests to by default use seasonally approprate grass color instead of always using the olive-color, thus also eliminating a glitch when the base was changed. Also deleted unused tile.
20100505 19:01:49-!- timotei [timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20100505 19:13:12-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100505 19:13:53-!- Lastmerlin [~Miranda@kalypso.csn.tu-chemnitz.de] has quit [Ping timeout: 252 seconds]
20100505 19:19:31-!- kpb [~d0dad7a3@gateway/web/freenode/x-uuuhuvvxdrwtpiej] has joined #wesnoth-dev
20100505 19:19:59-!- kpb [~d0dad7a3@gateway/web/freenode/x-uuuhuvvxdrwtpiej] has quit [Client Quit]
20100505 19:23:30< mordante> billynux, I am now
20100505 19:23:39< billynux> hi mordante 
20100505 19:23:47< mordante> hi billynux 
20100505 19:23:54< billynux> mordante: well, ATM I'm pretty busy :(
20100505 19:24:20< billynux> mordante: but, just to update you: I improved the disconnect procedure of clients
20100505 19:24:20< mordante> billynux, ok I will be around for a few hours more
20100505 19:25:18< billynux> mordante: I was looking into allowing generic buffers... I think I'll go with const_buffers for sending data (in the parameters I mean)
20100505 19:25:38-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20100505 19:26:01< mordante> boost::asio's or your own I assume the latter
20100505 19:26:33< billynux> mordante: Well, thats the point, the API is already boost dependent
20100505 19:27:12< billynux>  mordante: And I don't see any issues on using their types. The next thing is providing my own buffer creation function
20100505 19:27:26< mordante> me neither
20100505 19:27:48< billynux> mordante: following their own. I can either duplicate all the methods and just forward...
20100505 19:28:26< billynux> mordante: do a horrible #define to map buffer to boost::asio::buffer (I refuse :) )
20100505 19:28:37< mordante> in that case I would use their version for now
20100505 19:28:55< billynux> mordante: or try to work around it with templates. Although the error messages will be horrible :)
20100505 19:29:04< mordante> I also don't want to see define hacks ;-)
20100505 19:29:42< billynux> mordante: something like template <class P, class R>    R buffer(P& p) { return asio::buffer(p); }
20100505 19:29:49< mordante> yup template errors are horrible, loonycyborg spoke of several MBs for boost spirit...
20100505 19:30:10< billynux> :)
20100505 19:30:18< mordante> great fun to debug
20100505 19:30:35< mordante> boost bind is already horrible if you make a tiny mistake...
20100505 19:31:02< mordante> but your template seems fine to me, would also allow to change the details later on
20100505 19:31:03< billynux> mordante: Copying the method profiles and forward all the calls is not that bad too... just adds some text to the API header
20100505 19:31:05< billynux> yes...bind is terrible :(
20100505 19:31:25< mordante> no bind is great, its error messages are horrible ;-)
20100505 19:31:33< billynux> mordante: true...
20100505 19:32:13< mordante> I leave it to you either use the asio buffers directly or forward to them
20100505 19:32:31< mordante> btw I agree with Crab_ the logo looks like the chaos logo ;-)
20100505 19:32:47< billynux> mordante: ok... but, is it ok? (the log)
20100505 19:32:54< billynux> s/log/logo/
20100505 19:33:37< billynux> mordante: I based it on wikimedia commons' logo :)
20100505 19:35:14< mordante> kind of a hard question, you want to use it later side Wesnoth, personally I think the logo looks to modern for Wesnoth
20100505 19:35:52< billynux> mordante: no, there is no need to use the logo in wesnoth
20100505 19:35:54< mordante> maybe if you change the central circle with the Wesnoth shield and the lines a bit more colours
20100505 19:36:15< mordante> then it would be fitting for Wesnoth
20100505 19:36:39< billynux> mordante: thats the bad thing about coding network stuff, it's background. End users don't "see" the difference :(
20100505 19:37:01< mordante> the same for a lot of the gui code as well...
20100505 19:37:23< mordante> but yes the users only see a small part of the game
20100505 19:38:46< mordante> btw I like the one you posted with ana in it better as the one committed
20100505 19:39:43< billynux> mordante: Yes... thats the one I'll go for... But googlecode is in read-only mode right now
20100505 19:40:01< billynux> mordante: And... BTW, I'll provide zero_copy versions of the send operations
20100505 19:40:48< billynux> mordante: as in zero_copy_send_all , zero_copy_send_if .... and the documentation will provide a sworn statement users will have to sign before calling such a method :)
20100505 19:41:24< billynux> promising to keep the buffer alive and well during the send operation 
20100505 19:41:49-!- timotei [timotei@wesnoth/developer/timotei] has quit [Read error: Connection reset by peer]
20100505 19:41:56-!- timotei [timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20100505 19:42:37< mordante> billynux, didn't we discuss it before and decided to make two functions one which has needed a const buffer and one that is the copying version?
20100505 19:43:03< mordante> and I'm not too fond of the zero_copy prefix
20100505 19:43:55< billynux> mordante: we discussed it, yes. Ok, what prefix/postfix do you propose? (I'll check the logs), and, the fact that the buffer is const doesn't ensure much in this case
20100505 19:44:12-!- Lastmerlin [~Miranda@kalypso.csn.tu-chemnitz.de] has joined #wesnoth-dev
20100505 19:45:23-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has quit [Quit: Leaving]
20100505 19:45:36< mordante> not yet sure what the best name would be
20100505 19:46:11< billynux> mordante: for a Zero Copy operation, the user needs to ensure the buffer will remain valid between the send operation, and the call to the last handler associated with it
20100505 19:46:37< mordante> yes I understand that
20100505 19:47:08< mordante> only the name zero copy sounds awkward
20100505 19:47:43< billynux> mordante: true ... <thinking> ... I can put a flag on the profile, and default it to make a copy operation
20100505 19:47:43< mordante> I like something with static buffer in the name better, if not for the fact that static is a keyword
20100505 19:48:27< mordante> maybe a flag sounds very good
20100505 19:48:33< mordante> s/maybe//
20100505 19:48:43-!- mordante [~mordante@wesnoth/developer/mordante] has left #wesnoth-dev ["Leaving"]
20100505 19:48:52-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20100505 19:48:53< billynux> mordante: server_->send_all( my_buffer, YES_DONT_COPY_IT);  // enum or something (bool?)
20100505 19:49:03< mordante> wrong key :-/
20100505 19:49:17< mordante> a bool
20100505 19:49:20< billynux> :) you missed: mordante: server_->send_all( my_buffer, YES_DONT_COPY_IT);  // enum or something (bool?)
20100505 19:49:33< billynux> mordante: ok, agreed, I'll work on that
20100505 19:49:41-!- Blueblaze [~nick@adsl-99-158-45-140.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 245 seconds]
20100505 19:49:51< mordante> I didn't I like the bool better
20100505 19:49:57< billynux> mordante: so... upcoming: generic buffer support, zero copy option (via parameter)
20100505 19:50:09< billynux> later on: timeouts on send operations
20100505 19:50:30< mordante> I could call the parameter copy_buffer and explain in the documentation what it does
20100505 19:50:59-!- wesbot changed the topic of #wesnoth-dev to: 101 bugs, 277 feature requests, 16 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100505 19:51:13< billynux> mordante: I was considering making my own error_code category (after reading about error_codes) -> Oh... BTW I changed error reporting from bool to boost::system::error_code
20100505 19:52:25< billynux> mordante: Ok, I'll carry on with what I was doing. See you around, and thanks
20100505 19:52:53< mordante> yes I saw you changed it
20100505 19:53:03< mordante> oke see you later
20100505 20:03:09-!- ancestral [~ancestral@97-116-122-90.mpls.qwest.net] has joined #wesnoth-dev
20100505 20:17:44< Lastmerlin> hello, anybode here who can help me with a question to language.cpp and internationalization ?
20100505 20:17:53< Lastmerlin> its for the windows part specifically
20100505 20:19:39< timotei> Lastmerlin, just post here your question, and if it somebody who knows the answer will post the answer
20100505 20:19:48< timotei> s/if it/if it is
20100505 20:19:52< Lastmerlin> ok
20100505 20:20:09< Lastmerlin> I am trying so set environment variablen to change gettext locale
20100505 20:20:18< Lastmerlin> works perfect for linux, does not work for windows
20100505 20:20:35< Lastmerlin> I copied almost exactly the wesnoth code and cant see my mistake
20100505 20:20:59< mordante> Lastmerlin, maybe post your code on a pastebin
20100505 20:21:13< loonycyborg> AFAIK there's some caveat in the way i18n support on windows works.
20100505 20:22:08< Lastmerlin> http://wesnoth.pastebin.com/ZfXxcSmU
20100505 20:22:39< loonycyborg> Look at language_win32.ii . I don't see en_GB etc there
20100505 20:23:02< Lastmerlin> but en
20100505 20:23:16< Lastmerlin> and I take only the first two characters of the locale
20100505 20:23:21< Lastmerlin> and then go into the two ifs
20100505 20:23:33< Lastmerlin> which is, if I am not mistaken, is exactly the same wesnoth does
20100505 20:24:57< loonycyborg> You're using GetEnvironmentVariable. Those don't seem to affect gettext at all IME
20100505 20:25:17< Lastmerlin> hmm i just copied that from language.cpp
20100505 20:25:23< Lastmerlin> and hoped to fix it
20100505 20:26:39< Lastmerlin> http://wesnoth.pastebin.com/r1hjBUib
20100505 20:26:45< Lastmerlin> here is some output of the program
20100505 20:27:11< Lastmerlin> note, that neither getenv nor getEnvironmentVariable return useful thing for the first call
20100505 20:27:43< Lastmerlin> gettext itself works - i have system default == german and it translated english strings to german
20100505 20:28:07< Lastmerlin> what does not work, is to change language to english so that english strings are NOT translated
20100505 20:29:01< CIA-10> ai0867 * r42486 /trunk/utils/indexed2rgb.sh: Make indexed2rgb.sh check if the image isn't already in RGBA mode. Add an option for automatically running optipng afterwards.
20100505 20:29:25< CIA-10> ai0867 * r42487 /trunk/utils/indexed2rgb.sh: Check, then doublecheck if the image really has been converted.
20100505 20:31:33< CIA-10> ai0867 * r42488 /trunk/ (418 files in 58 dirs): Convert all trunk images to RGBA format
20100505 20:35:03-!- EdB [~EdB@tss37-1-89-84-18-220.dsl.club-internet.fr] has joined #wesnoth-dev
20100505 20:35:19< Lastmerlin> by the way, it works, if I change language for windows globally
20100505 20:35:47< Lastmerlin> but it does not work to change it ingame
20100505 20:37:25< loonycyborg> Lastmerlin: Does it work if you change those environment variables in terminal before running?
20100505 20:38:01< Lastmerlin> hmm what are the commands for windows to get and set env vars ?
20100505 20:38:18< loonycyborg> IIRC set var=value
20100505 20:39:00< loonycyborg> And just 'set' will print them all
20100505 20:39:46< Lastmerlin> well there is not variablen LANG defined so far o_O
20100505 20:39:54< Lastmerlin> what is the name for windows
20100505 20:40:26< timotei> weird, git svn rebase doesn't work, but git svn fetch -rHEAD works....
20100505 20:40:52< Lastmerlin> looking at the variables I can not see anything that looks like the locale variable :O
20100505 20:41:06< Lastmerlin> so what should I set ?
20100505 20:41:14< Lastmerlin> set LANG=English ?
20100505 20:41:27< loonycyborg> Yes.
20100505 20:42:28< loonycyborg> Or set LANG=en
20100505 20:42:54< Lastmerlin> no influence, its still german, if I choose german in the control panel
20100505 20:42:55< mordante> timotei, what does git tell you?
20100505 20:42:59< Lastmerlin> in both cases
20100505 20:43:20< timotei> mordante, nothing. it just stays like is waiting for something
20100505 20:43:35< timotei> (i've been staying with a git svn rebase console for about an hour and nothing)
20100505 20:43:44< timotei> but np, as long as I can update it, is ok:D
20100505 20:43:56< loonycyborg> Lastmerlin: You could also try LC_MESSAGES or LC_ALL 
20100505 20:45:43< Lastmerlin> tried all possible combinations, no success
20100505 20:45:58< mordante> timotei, odd and git status?
20100505 20:48:45< timotei> mordante, just some untracked items
20100505 20:49:07-!- Sirp_ [~97c1dc1d@wesnoth/developer/dave] has joined #wesnoth-dev
20100505 20:49:16-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20100505 20:49:18< Lastmerlin> well, thank you anyway, get to go now
20100505 20:49:34< Lastmerlin> will be back in a few hours and ask annoying questions again :D
20100505 20:50:02< timotei> mordante, do you know what it may cause that problem? maybe cause there are many files?
20100505 20:50:07< mordante> I won't be around to answer then in a few hours :-P
20100505 20:50:35< timotei> even though, I can update the "fetch" way, it's ok. the next thing is to get that working on linux
20100505 20:50:39< mordante> timotei, if an untracked file has the name of a newly added file git complains that it can't detach the head
20100505 20:50:57< mordante> the solution is to remove that untracked file
20100505 20:51:12-!- billynux [~c8078d05@wesnoth/developer/billynux] has quit [Quit: Page closed]
20100505 20:56:03-!- ancestral [~ancestral@97-116-122-90.mpls.qwest.net] has quit [Quit: ancestral]
20100505 20:58:05-!- Skystriker [~croselius@pool-70-111-210-233.nwrk.east.verizon.net] has joined #wesnoth-dev
20100505 21:00:19-!- Upthorn [~ogmar@75.26.184.137] has quit [Ping timeout: 246 seconds]
20100505 21:01:07-!- ancestral [~ancestral@97-116-122-90.mpls.qwest.net] has joined #wesnoth-dev
20100505 21:01:21-!- ancestral [~ancestral@97-116-122-90.mpls.qwest.net] has quit [Client Quit]
20100505 21:06:03< mordante> I'm off bye
20100505 21:07:32-!- yann_ [~dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has joined #wesnoth-dev
20100505 21:07:45< Skystriker> night mordante
20100505 21:08:27< timotei> bye mordante 
20100505 21:13:46-!- noy_ [~noy@74.198.148.79] has joined #wesnoth-dev
20100505 21:13:46-!- noy_ [~noy@74.198.148.79] has quit [Changing host]
20100505 21:13:46-!- noy_ [~noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100505 21:17:23-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 260 seconds]
20100505 21:17:26-!- noy_ is now known as noy
20100505 21:18:14-!- kevg [~kosov.eug@91.194.253.47] has joined #wesnoth-dev
20100505 21:18:24< kevg> Hi to all.
20100505 21:18:31< Crab_> hi, kevg
20100505 21:20:33-!- noy [~noy@wesnoth/developer/noy] has quit [Quit: noy]
20100505 21:20:55< kevg> mordante, i have a question about https://gna.org/bugs/?7470. I want to try to implement it.
20100505 21:21:37< kevg> Where can i find this fonts in game? Are them in lobby room?
20100505 21:22:22< Crab_> kevg: note - 15 minutes ago mordante said "I'm off bye", you might have missed him by a few minutes.
20100505 21:23:08< kevg> ;) Next time i would read logs better.
20100505 21:26:05< mordante> kevg, I'm still around but only for a short time
20100505 21:26:49< mordante> but that bug is regarding the old gui, and I don't know too much about that part
20100505 21:27:31< Crab_> mordante: the new gui doesn't have that problem ?
20100505 21:27:32< kevg> Ok, any another simple programming task?
20100505 21:28:03< kevg> I`m a novice C++ programmer.
20100505 21:28:22< mordante> Crab_, no idea, but it was only in the in-game chat and not in the lobby
20100505 21:28:29< Crab_> mordante: ok
20100505 21:28:52< Crab_> kevg: what do you think about https://gna.org/bugs/?15921 ?
20100505 21:29:25< Crab_> see ./src/play_controller.cpp +574
20100505 21:30:22< Crab_> sometimes, the WML author thinks that it is a bad thing to do that gui_->scroll_to_leader , for a given side
20100505 21:30:28< kevg> Suitable. I'll see.
20100505 21:30:54< kevg> WML is some kind of internal language-game config?
20100505 21:30:59< Crab_> yes
20100505 21:31:26< Crab_> so, the reporter of that feature request suggested we add a new WML attribute to [side] tag, to allow to disable that  gui_->scroll_to_leader  for that side.
20100505 21:32:08< mordante> really off now, night
20100505 21:32:17< Crab_> mordante: night
20100505 21:32:24< kevg> bye
20100505 21:32:29-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20100505 21:32:47< Crab_> kevg: in that function in play_controller, 'current_team' is the current team object. see src/team.cpp for it.
20100505 21:32:56< kevg> What [side] is? One of opponents?
20100505 21:33:08< Crab_> [side] is a configuration of a particular team
20100505 21:33:22< Crab_> that scroll happens at the start of turn of a particular team
20100505 21:33:54< Crab_> take a look at any campaign scenario .cfg file
20100505 21:34:17< Crab_> it has [scenario] [side] config of side 1 [/side] [side] config of side 2 [/side] ... [side] config of side N [/side]  other stuff [/scenario]
20100505 21:34:22-!- timotei [timotei@wesnoth/developer/timotei] has quit [Quit: Leaving]
20100505 21:34:30< Crab_> from each  [side] config of side N [/side] , we construct a team object
20100505 21:34:47< Crab_> see team::team_info::team_info(const config& cfg) in src/team.cpp 
20100505 21:34:59< Crab_> 'config' is a c++ representation of the WML tag
20100505 21:37:19< Crab_> so, a implementation plan might look like this: add a new boolean field to team to store that option; read the value of that field from config at team construction; save the value of that field to config at team serialization; add an accessor for that field; check for that boolean field in current_team in play_controller and conditionally skip scrolling part. add a changelog entry; add entry to about.cfg; fix wiki documentation; submit a patch;
20100505 21:38:35< Crab_> ask questions, if any.
20100505 21:39:29< kevg> Now have no questions. I need to examine information. I work very slow.
20100505 21:40:16< Crab_> kevg: it's ok. you can always ask them later. the usual policy is 'just ask' and anyone who knows the answer might answer.
20100505 21:41:06< Crab_> good luck :)
20100505 21:41:11< kevg> Also i`m worry about my time. Most possible i'll complete this. But it might occur very late.
20100505 21:41:16< kevg> Thank you.
20100505 21:41:39< kevg> My english is ugly? :)
20100505 21:41:51< kevg> I had no practice before.
20100505 21:42:11< Crab_> kevg: it's good enough to understand you.
20100505 21:42:17< Crab_> kevg: so, it's all ok :)
20100505 21:42:30< kevg> Sounds nice.
20100505 21:43:10< shadowmaster> I've seen worse.
20100505 21:43:12< Crab_> kevg: and the feature request itself is quite easy to implement, no hidden bugs in there :)
20100505 21:43:27< shadowmaster> from native speakers, even
20100505 21:44:23< CIA-10> ai0867 * r42489 /branches/1.8/ (386 files in 54 dirs): Convert all 1.8 images into RGBA format
20100505 21:45:07< shadowmaster> AI0867: 386? that sounds like an awful amount
20100505 21:46:40< shadowmaster> what's the size difference, though?
20100505 21:46:42< AI0867> there were only a few indexed ones
20100505 21:46:53< AI0867> most were RGB or grayscale with or without alpha
20100505 21:46:58< AI0867> not much
20100505 21:47:02< AI0867> a few percent increase for most
20100505 21:47:05< AI0867> some only a few bytes
20100505 21:48:38-!- shadowm_laptop [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100505 21:51:00< AI0867> second check reveals that some increased quite a bit more than that though =/
20100505 21:52:19< CIA-10> ai0867 * r42490 /trunk/utils/png_format_check.sh: Add png_format_check.sh, to check the format of PNGs
20100505 21:52:26< shadowmaster> uh huh
20100505 21:55:37-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100505 21:56:09< AI0867> this was png_format_check's report: RGBA: 7130 RGB: 197 Gray+A: 148 Gray: 0 Index+A: 25 Index: 9 Other: 0
20100505 21:56:44< CIA-10> crab * r42491 /trunk/ (4 files in 3 dirs): Patch #1664 by faab: Display the AIs defined by selected era in the list of available AIs, when creating a game
20100505 21:57:57-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100505 22:19:43< shadowm_laptop> fendrin: I see you've not committed the music update to trunk yet?
20100505 22:20:04-!- Johannes13__ [~Johannes@pD9501555.dip0.t-ipconnect.de] has joined #wesnoth-dev
20100505 22:23:39-!- Johannes13_ [~Johannes@unaffiliated/johannes13] has quit [Ping timeout: 260 seconds]
20100505 22:27:15-!- Ken_Oh [~briang@static-71-178-174-220.washdc.fios.verizon.net] has quit [Read error: Connection reset by peer]
20100505 22:29:14-!- shadowm_laptop [~ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 240 seconds]
20100505 22:30:32-!- EdB [~EdB@tss37-1-89-84-18-220.dsl.club-internet.fr] has quit [Remote host closed the connection]
20100505 22:34:54-!- yann_ [~dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has quit [Remote host closed the connection]
20100505 22:39:13-!- happygrue_ [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20100505 22:39:21-!- happygrue [~George@wesnoth/developer/wintermute] has quit [Remote host closed the connection]
20100505 22:45:49< CIA-10> crab * r42492 /trunk/src/ai/composite/engine_fai.cpp: fix bug #15994: formula_ai in candidate actions specified in [side][ai] doesn't work, but works in candidate actions added later via modify_ai
20100505 22:45:53< CIA-10> crab * r42493 /trunk/src/ai/ (formula/ai.cpp formula/ai.hpp testing/stage_rca.cpp): improve logging output of ai candidate actions framework
20100505 22:48:14< CIA-10> crab * r42494 /trunk/changelog: changelog entry for the bfix of bug #15994
20100505 22:55:28< Lastmerlin> Hello folks
20100505 22:55:48< Lastmerlin> I am still wondering, how to set the language locale used by gettext under windows
20100505 22:56:22< Lastmerlin> things already tried: setenv with LANG, SetEnvironmentVariable with LANG, setlocale
20100505 22:56:35< Lastmerlin> each one with values "en", "en_GB" and "English"
20100505 22:57:01< Lastmerlin> also tried without success: setting environment variable LANG or LC_ALL on cmd before starting
20100505 22:57:10< Lastmerlin> changing windows global language works
20100505 22:57:19< Lastmerlin> any ideas what to try next ?
20100505 23:05:08< AI0867> checking...
20100505 23:05:22< Lastmerlin> checking what ?
20100505 23:06:22< AI0867> oh, windows
20100505 23:06:27< AI0867> I don't know there
20100505 23:06:37< Lastmerlin> for linux setlocale works fine
20100505 23:06:53< Lastmerlin> for windows it does not
20100505 23:06:54< AI0867> under linux, LANGUAGE="nl" wesnoth works
20100505 23:07:14< Lastmerlin> yes, for linux there are a lot of options
20100505 23:07:15< AI0867> try the environmentvariable LANGUAGE
20100505 23:07:19< AI0867> rather than LANG
20100505 23:07:26< Lastmerlin> setting lang env variable should also work
20100505 23:07:33< Lastmerlin> for windows
20100505 23:07:34< AI0867> it didn't here
20100505 23:07:35< Lastmerlin> ok,
20100505 23:07:40< Lastmerlin> i give it a try
20100505 23:11:45< Lastmerlin> It works (!)
20100505 23:11:48-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit []
20100505 23:11:50< Lastmerlin> for english ^^
20100505 23:12:01< Lastmerlin> LANGUAGE=German does not
20100505 23:12:07< AI0867> try de
20100505 23:12:13< AI0867> LANGUAGE="de"
20100505 23:12:59< Lastmerlin> perfect :)
20100505 23:13:11< Lastmerlin> thank you very much :)
20100505 23:15:27-!- elias [~elias@allegro/developer/allefant] has quit [Quit: Leaving]
20100505 23:21:43< Lastmerlin> args x(
20100505 23:21:51< Ivanovic> Lastmerlin: uhm, you know, the easiest way to switch to a different lang is to start the game and jsut select the language you want in the "Languages" menu...
20100505 23:21:51< Lastmerlin> i tried to set the value in the app now
20100505 23:21:54< Ivanovic> ;)
20100505 23:21:59< Lastmerlin> and what I got is
20100505 23:22:05< Lastmerlin> it works for the first set of the language
20100505 23:22:08< Lastmerlin> but not for the second
20100505 23:22:18< Ivanovic> yes, restart the program to have the changes take effect
20100505 23:22:20< Lastmerlin> but you are right, I will accept this behavior now
20100505 23:22:25< Ivanovic> windows sucks regarding multilang support!
20100505 23:22:44< Lastmerlin> i think I have noticed that now
20100505 23:23:05< Ivanovic> (okay, windows does suck in really many areas if you are used to linux systems
20100505 23:23:59< Lastmerlin> Lets say everything that is not multimedia and games ?
20100505 23:24:08-!- Johannes13__ [~Johannes@pD9501555.dip0.t-ipconnect.de] has quit [Ping timeout: 252 seconds]
20100505 23:24:47< Ivanovic> even in multimedia windows does suck
20100505 23:24:55< Ivanovic> and games like wesnoth are faster on *nix, too
20100505 23:25:13< Lastmerlin> yes but count the number of commersial quality games for linux
20100505 23:25:23< Ivanovic> (yeah, in multimedia the player are a lot more responsive and they "just work", no need for installing systemwide codec packs and the likes)
20100505 23:25:24< Lastmerlin> thats the main reason, why windows still has >90% market share
20100505 23:25:42< Ivanovic> yes, the amount of AAA games on windows is a lot larger
20100505 23:25:59< Lastmerlin> as soon as most games *just work* and multimedia *just works* windows is screwed
20100505 23:25:59< Ivanovic> but honestly, that's about it
20100505 23:26:23< Ivanovic> and if the rumors about steam support on linux holds true some time soon even the area of games might change
20100505 23:26:29< Lastmerlin> I am quite optimistic that I will see the day, this happens
20100505 23:26:34< Lastmerlin> ok not steam
20100505 23:26:38< Ivanovic> (at least it is likely that valve would port their source engine based games)
20100505 23:26:44< Lastmerlin> good game support for linux in general
20100505 23:27:00< Ivanovic> Lastmerlin: ehm, there *is* some linux support of steam (GUI)
20100505 23:27:03< Ivanovic> though nothing official
20100505 23:27:15< Ivanovic> cf http://www.phoronix.com/scan.php?page=news_item&px=ODIwNQ
20100505 23:27:43< Ivanovic> and since they *do* port their source engine stuff to macosx it is likely that they directly do linux, too (since it is not much additional work)
20100505 23:28:02< Lastmerlin> damn I have somehow screwed my windows localization now
20100505 23:28:13< Lastmerlin> no env variable i can see is still set
20100505 23:28:17< Lastmerlin> deleted them all
20100505 23:28:23< Ivanovic> Lastmerlin: it was screwed right from the start when MS created windows...
20100505 23:28:24< Ivanovic> ;)
20100505 23:28:29< Lastmerlin> and windows control panel say german localization
20100505 23:28:44< Lastmerlin> but my keyboard setup is fucked up
20100505 23:29:07< Lastmerlin> grml I hate gill bates x(
20100505 23:33:10-!- Sirp_ [~97c1dc1d@wesnoth/developer/dave] has quit [Ping timeout: 252 seconds]
20100505 23:35:17-!- Blueblaze [~nick@adsl-99-158-45-140.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100505 23:35:24-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100505 23:36:48-!- Darkas [~quassel@ppp-93-104-128-7.dynamic.mnet-online.de] has quit [Remote host closed the connection]
20100505 23:41:44-!- _jbx_ [~jbailey@12.190.80.225] has quit [Quit: It was a new day yesterday, but it's an old day now.]
--- Log closed Thu May 06 00:00:00 2010