--- Log opened Sat Feb 28 00:00:19 2009
20090228 00:01:04-!- Shadow_Master [n=imorelle@wesnoth/developer/shadowmaster] has quit [Remote closed the connection]
20090228 00:01:04-!- AI0867_ [n=ai@wesnoth/developer/ai0867] has quit [Read error: 104 (Connection reset by peer)]
20090228 00:04:54-!- Sirp_ [n=me@c-71-198-3-104.hsd1.ca.comcast.net] has joined #wesnoth-dev
20090228 00:05:12-!- stikonas [n=stikonas@wesnoth/translator/stikonas] has quit [Read error: 104 (Connection reset by peer)]
20090228 00:10:06-!- stikonas [n=stikonas@wesnoth/translator/stikonas] has joined #wesnoth-dev
20090228 00:14:50-!- zookeeper [n=l@wesnoth/developer/zookeeper] has quit []
20090228 00:21:07< Sirp_> hmm if we could get up to 10,000 visits/day I think that would open up more ads options to us
20090228 00:21:31< Sirp_> a reason for frogatto to be co-hosted on wesnoth.org, perhaps. :)
20090228 00:22:23< voris> Is that your latest project?
20090228 00:31:36-!- loonycyborg [n=sergey@wesnoth/developer/loonycyborg] has quit ["Zzzzzzzzzzzzzzzzzzzzzzzzzz"]
20090228 00:32:19-!- stikonas [n=stikonas@wesnoth/translator/stikonas] has quit [Read error: 110 (Connection timed out)]
20090228 00:36:42-!- sylecn [n=sylecn@resnet-50-205.dorm.utexas.edu] has joined #wesnoth-dev
20090228 00:46:52-!- shadowmaster [n=imorelle@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090228 00:49:49-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090228 00:50:41< shadowmaster> wesbot: seen fabi
20090228 00:50:41< wesbot> shadowmaster: Person, who was using the nick fabi 3h 24m ago, last spoke 3h 54m ago.  is now known as fabi on IRC.
20090228 00:50:46< shadowmaster> fabi: yes I did propose it.
20090228 00:51:16< shadowmaster> Mordante: no, but I'm planning to delve into that crap and fix it
20090228 00:52:45< shadowmaster> fendrin: ^^
20090228 00:52:51< Ivanovic> time for me to head off to bed, n8
20090228 00:53:01< shadowmaster> switching your nicks keeps confusing me.
20090228 00:53:23< Ivanovic> nital: please remind boucman to mark your patch as "done" (since IIRC it was in 1.5.11, correct?)
20090228 00:53:36-!- Sapient [n=sapien-x@wesnoth/developer/sapient] has joined #wesnoth-dev
20090228 00:55:02< shadowmaster> fendrin: Sapient aproved my suggestion of deprecating [set_variable] format= in 1.7.0 and removing it in 1.7.3 or so too
20090228 00:55:23 * Sapient nods
20090228 00:55:53< Sapient> format is a relic of the confusing way things used to be with variable substitution
20090228 00:56:15< Sapient> it is utterly meaningless now
20090228 00:58:09-!- Turuk_ [n=Turuk@FUSE-WS1-208-102-202-188.fuse.net] has quit [Read error: 110 (Connection timed out)]
20090228 01:01:59-!- shadowmaster [n=imorelle@wesnoth/developer/shadowmaster] has quit ["(null)"]
20090228 01:05:41< fendrin> Sapient: What should I use instead when I want to have variable substitution and arithmetic expression to be evaluated?
20090228 01:06:03< Sapient> $(...)
20090228 01:07:01< Sapient> e.g. "$(2*$x + $y)"
20090228 01:07:16< fendrin> value=.... ?
20090228 01:07:34< Sapient> that goes without saying, right?
20090228 01:08:03< Sapient> value="foo"
20090228 01:08:12< fendrin> The wiki says that value isn't able to do the things you mention.
20090228 01:08:35< Sapient> I think you may be reading it incorrectly
20090228 01:08:40-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 60 (Operation timed out)]
20090228 01:08:40< Sapient> or someone messed it up
20090228 01:08:49-!- Espreon [n=espreon@wesnoth/developer/espreon] has quit [Remote closed the connection]
20090228 01:08:58-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090228 01:09:11-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090228 01:10:45< fendrin> "This only interprets dollars signs if it is the very first character, and then the entire value must be a simple variable name."
20090228 01:11:51< Sapient> where are you reading that?
20090228 01:12:01< Sapient> that hasn't been true for many many versions
20090228 01:12:07< fendrin> http://www.wesnoth.org/wiki/InternalActionsWML
20090228 01:12:59< Sapient> "in 1.3.2, has the same effect as format"
20090228 01:13:11< Sapient> so are you using a version older than 1.3.2?
20090228 01:13:47< Sapient> obviously that information is outdated
20090228 01:16:15< fendrin> Sapient: Now I get it. Thank you.
20090228 01:17:30< Sapient> feel free to clarify it
20090228 01:17:36< Sapient> that old stuff can be removed
20090228 01:19:43-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090228 01:23:12-!- noy [n=Noy@wesnoth/developer/noy] has quit [Success]
20090228 01:23:36< fendrin> Sapient: are you editing it as well?
20090228 01:23:59< Sapient> no
20090228 01:25:45< fendrin> Okay, clarified it.
20090228 01:26:20< Sapient> thanks, fendrin
20090228 01:26:29< fendrin> you are welcome
20090228 01:26:51-!- ikarius [n=ross@smtp.gridironsystems.com] has quit []
20090228 01:28:13-!- Elvish_Pillager [n=eli@66-189-40-122.dhcp.oxfr.ma.charter.com] has quit ["Hi! I'm a quit message virus vaccine. If you see a quit message virus, don't replace your quit message with it!"]
20090228 01:37:12-!- Lord_Aether [n=castle@207.212.136.10] has joined #wesnoth-dev
20090228 01:51:02< CIA-70> fendrin * r33131 /trunk/data/campaigns/Under_the_Burning_Suns/scenarios/06a_In_the_Tunnels_of_Trolls.cfg: UtBS 6a: Fixed bug #13068 .
20090228 01:54:43-!- happygrue_ [n=happygru@c-67-176-145-41.hsd1.in.comcast.net] has joined #wesnoth-dev
20090228 01:55:19-!- happygrue [n=George@wesnoth/developer/wintermute] has quit ["http://xkcd.com/91/"]
20090228 01:55:55-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090228 01:56:37-!- happygrue_ [n=happygru@c-67-176-145-41.hsd1.in.comcast.net] has left #wesnoth-dev []
20090228 01:59:01-!- shadowmaster [n=imorelle@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090228 02:01:36-!- shikadibot [n=uid1001@wesnoth/bot/shikadibot] has joined #wesnoth-dev
20090228 02:04:26-!- Corvvs [n=michaelm@pool-96-231-2-122.washdc.east.verizon.net] has joined #wesnoth-dev
20090228 02:05:25-!- shadowmaster is now known as Shadow_Master
20090228 02:30:22-!- Corvvs_ [n=michaelm@pool-96-231-1-183.washdc.east.verizon.net] has joined #wesnoth-dev
20090228 02:36:16-!- PK [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has joined #wesnoth-dev
20090228 02:40:27-!- Corvvs__ [n=michaelm@pool-96-231-4-242.washdc.east.verizon.net] has joined #wesnoth-dev
20090228 02:43:29-!- Corvvs___ [n=michaelm@pool-96-231-8-238.washdc.east.verizon.net] has joined #wesnoth-dev
20090228 02:46:51-!- Corvvs [n=michaelm@pool-96-231-2-122.washdc.east.verizon.net] has quit [Read error: 110 (Connection timed out)]
20090228 02:49:58-!- sylecn [n=sylecn@resnet-50-205.dorm.utexas.edu] has left #wesnoth-dev ["Leaving"]
20090228 02:51:34-!- Corvvs [n=michaelm@c-69-139-145-2.hsd1.md.comcast.net] has joined #wesnoth-dev
20090228 02:54:43-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Connection reset by peer]
20090228 02:54:58-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090228 02:55:42-!- Corvvs_ [n=michaelm@pool-96-231-1-183.washdc.east.verizon.net] has quit [Read error: 110 (Connection timed out)]
20090228 02:57:44-!- Lord_Aether [n=castle@207.212.136.10] has quit []
20090228 02:58:54-!- Corvvs__ [n=michaelm@pool-96-231-4-242.washdc.east.verizon.net] has quit [Read error: 110 (Connection timed out)]
20090228 03:07:08-!- BenUrban [n=benurban@c-68-49-10-243.hsd1.md.comcast.net] has joined #wesnoth-dev
20090228 03:07:47-!- Corvvs___ [n=michaelm@pool-96-231-8-238.washdc.east.verizon.net] has quit [Read error: 110 (Connection timed out)]
20090228 03:18:11< dfranke> I'm going to be logging off for the weekend.  My brain is thrashing and I need to defragment.
20090228 03:18:21< Shadow_Master> good luck!
20090228 03:18:53< dfranke> I'm closing IRC, email, and IM, and I've put all my time-wasting sites in iptables.
20090228 03:19:02< dfranke> That includes wesnoth.org as soon as I push a quick patch.
20090228 03:24:44< CIA-70> dfranke * r33132 /trunk/src/addon_checks.cpp: Disallow tildes in add-on names, for bug #13080
20090228 03:26:27< voris> That reminds me. Is there any point in fixing stuff like colorization of titles in 1.4 server addons or is it good enough to just fix it in 1.6?
20090228 03:26:40< voris> I inherited one that has that.
20090228 03:27:43< Soliton> well, if you can fix it in 1.4 as well. if it's pretty subtle it's not a big deal anyway though.
20090228 03:28:13< dfranke> okay, see you all on Monday.
20090228 03:28:31< voris> Ok. I just wasn't sure if 1.4 was going away or someting so no point in it.
20090228 03:28:35< voris> I'll do it in both.
20090228 03:28:36-!- [Relic] [n=[Relic]@adsl-76-229-202-137.dsl.milwwi.sbcglobal.net] has joined #wesnoth-dev
20090228 03:28:42< voris> Good weekend, dfranke.
20090228 03:28:43< Soliton> curious that there are a lot tildes in addons up to 1.4 but none in 1.5.
20090228 03:29:09< Shadow_Master> probably because most junk authors haven't switched to 1.5 yet.
20090228 03:29:15< Soliton> ;-)
20090228 03:29:17< dfranke> people are using tildes in add-on *names*?
20090228 03:29:21< dfranke> titles are okay.
20090228 03:29:25< voris> You said what I held my tounge on.
20090228 03:29:39< [Relic]> Hello :)
20090228 03:30:01< voris> omg I forgot it's BSG night!
20090228 03:30:21< Sapient> shikadibot: BSG?
20090228 03:30:22< shikadibot> Sorry, I don't know what 'BSG' means...
20090228 03:30:23< Shadow_Master> BSG == Battlestar Galactica? 
20090228 03:30:24< dfranke> maybe I should have made it a permaban rather than just rejecting your submission.
20090228 03:30:27< Soliton> dfranke: you didn't check carefully what the function you changed does then.
20090228 03:30:29< voris> Yes.
20090228 03:30:33< Shadow_Master> voris: oh.
20090228 03:30:35< happygrue> voris: not for those of us who watch from sci-fi.com
20090228 03:30:41< happygrue> then it is one day behind ;)
20090228 03:30:43< dfranke> Soliton: it checks add-on names and file/directory names.
20090228 03:30:48< Soliton> exactly.
20090228 03:30:55< voris> I watch it on Sci-Fi.
20090228 03:31:01< dfranke> but not display titles.
20090228 03:31:01< Soliton> so it's not just titles.
20090228 03:31:04< happygrue> it comes on tonight?
20090228 03:31:05< voris> Clearly you have some mutant version. :)
20090228 03:31:08< happygrue> lol
20090228 03:31:10< voris> Yeah, I'm in Seattle.
20090228 03:31:14< Soliton> well, names or whatever.
20090228 03:31:15< happygrue> well, last season it didn't
20090228 03:31:24< Soliton> i was talking about filenames of course.
20090228 03:31:25< happygrue> oh, on eastern time it updates too late for me I think
20090228 03:31:26< voris> I'm surprised but not entirely that there would be regional variances.
20090228 03:31:33< voris> ahhh
20090228 03:32:14< dfranke> anyway, enjoy your weekends
20090228 03:32:17< Shadow_Master> science fiction servers + wesnoth == add-on bugs?
20090228 03:32:56< Shadow_Master> or was it science fiction servers + bugs == wesnoth add-on? I'm now confused.
20090228 03:33:15< voris> haha
20090228 03:33:34< voris> Forgetting to switch channels when I noticed that mostly. :)
20090228 03:34:38< voris> A science fiction server bugs addon wouldn't be a total loser I don't think though.
20090228 03:35:20-!- Shadow_Master [n=imorelle@wesnoth/developer/shadowmaster] has quit ["(null)"]
20090228 03:36:53-!- shadowmaster [n=imorelle@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090228 03:37:30-!- shadowmaster is now known as Shadow_Master
20090228 03:39:32-!- Corvvs_ [n=michaelm@c-69-139-145-2.hsd1.md.comcast.net] has joined #wesnoth-dev
20090228 03:42:40-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has quit ["ZNC - http://znc.sourceforge.net"]
20090228 03:44:08-!- Lord_Aether [n=castle@207.212.136.10] has joined #wesnoth-dev
20090228 03:49:13-!- PK [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has quit ["Java user signed off"]
20090228 03:55:30-!- Corvvs [n=michaelm@c-69-139-145-2.hsd1.md.comcast.net] has quit [Read error: 110 (Connection timed out)]
20090228 04:14:44< Shadow_Master> Ivanovic: so, why are RACV translation issues posted to the main Wesnoth-Dev ML still?
20090228 04:14:59< Shadow_Master> not even a plain-text message. ugh.
20090228 04:17:12< Soliton> because it is special.
20090228 04:18:17< Soliton> the message i see is also in plain text though.
20090228 04:19:47< Shadow_Master> bah, gmail prefers to show me the rich-text version of a multipart/mixed message
20090228 04:20:27< Soliton> well, if you're in a browser might as well show you the html part?
20090228 04:20:39-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev
20090228 04:22:07< Shadow_Master> indeed.
20090228 04:24:27-!- shikadibot [n=uid1001@wesnoth/bot/shikadibot] has quit ["0"]
20090228 04:25:06-!- Shadow_Master [n=imorelle@wesnoth/developer/shadowmaster] has quit ["int main(int,char**) { return EWUDBNICE; }"]
20090228 04:39:22-!- Ivanovic_ [n=ivanovic@dtmd-4db2a000.pool.einsundeins.de] has joined #wesnoth-dev
20090228 04:41:15-!- boucman [n=rosen@wesnoth/developer/boucman] has quit [Remote closed the connection]
20090228 04:56:44-!- Ivanovic [n=ivanovic@wesnoth/developer/ivanovic] has quit [Read error: 110 (Connection timed out)]
20090228 04:57:21-!- Ivanovic_ is now known as Ivanovic
20090228 05:06:13-!- Lord_Aether [n=castle@207.212.136.10] has quit []
20090228 05:25:59< CIA-70> esr * r33133 /trunk/data/campaigns/Under_the_Burning_Suns/units/Kaleh.cfg: Add indenting for readability. No markup or content changes.
20090228 05:45:12-!- Sapient [n=sapien-x@wesnoth/developer/sapient] has quit []
20090228 05:49:39< CIA-70> esr * r33134 /trunk/data/campaigns/Under_the_Burning_Suns/units/Kaleh.cfg: 
20090228 05:49:39< CIA-70> Another attempt to refactor the Kaleh AMLAs to use only balanced markup.
20090228 05:49:39< CIA-70> Tested by advancing Kaleh.
20090228 06:11:40-!- Corvvs_ [n=michaelm@c-69-139-145-2.hsd1.md.comcast.net] has quit []
20090228 06:21:27-!- happygrue [n=George@wesnoth/developer/wintermute] has quit ["http://xkcd.com/91/"]
20090228 06:30:06-!- Corvvs [n=michaelm@pool-96-231-8-132.washdc.east.verizon.net] has joined #wesnoth-dev
20090228 06:30:19-!- Corvvs [n=michaelm@pool-96-231-8-132.washdc.east.verizon.net] has quit [Client Quit]
20090228 06:31:37< CIA-70> esr * r33135 /trunk/src/play_controller.cpp: 
20090228 06:31:37< CIA-70> I've looked at the code, and "Back to Start" as a poorly thought out feature.
20090228 06:31:37< CIA-70> The fix: perform a featurectomy, and implement mty threaded-save UI in 1.7.
20090228 06:31:37< CIA-70> Fixes bug #12992 ('Back to Start' doesn't do what it says.).
20090228 06:33:18< esr> wesbot: topic
20090228 06:33:38 * esr likes the direction tghe bug count is going.
20090228 06:54:10-!- [Relic] [n=[Relic]@adsl-76-229-202-137.dsl.milwwi.sbcglobal.net] has quit ["Leaving"]
20090228 07:13:09-!- silene [n=plouf@AMarseille-251-1-12-171.w83-113.abo.wanadoo.fr] has joined #wesnoth-dev
20090228 07:21:22-!- voris [n=voris@c-24-16-107-149.hsd1.wa.comcast.net] has quit [Remote closed the connection]
20090228 07:30:30-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 104 (Connection reset by peer)]
20090228 07:30:45-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090228 07:41:55-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit ["Leaving"]
20090228 08:12:44< esr> Hi, silene. Got any bug fix plans tonight?
20090228 08:18:52< silene> esr: nothing in particular; if you have noticed a bug that seems in my domain, please do tell
20090228 08:23:28< esr> silene: I'm not certain what your domain is, except you don't seem to mind touching the AI (something I can do but am nervous about).
20090228 08:28:47< silene> i wouldn't touch the ai for adding things, but as long as it's a c++ issue, i can deal with it
20090228 08:48:31-!- zookeeper [n=l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20090228 08:52:02< zookeeper> esr, what's poorly thought about the back to start option?
20090228 08:52:26< zookeeper> it's useful. gets you right back to the start if you need it.
20090228 08:54:12< zookeeper> silene, there's one possible AI fix that would need at least testing out...i'm not sure how complicated it would be to do. sirp said that [side] [ai] [target] might only affect what units the AI tends to like to move towards when it's out of range, and be ignored when deciding which targets are actually attacked out of all that are in range.
20090228 08:55:10< zookeeper> whereas it really IMO should also affect the latter.
20090228 08:57:09< zookeeper> if it's really currently ignoring it when deciding on attack targets, then it's probably the single most annoying and frustrating thing about the AI currently.
20090228 08:58:15< esr> zookeeper: You read dfranke's bug?  I looked at the C++ and saw no clean way to fix it.
20090228 08:59:13< silene> zookeeper: that i won't touch; even if the behavior isn't useful, it is the behavior it was designed for (if i understand you correctly), so changing it would actually amount to implement a new feature in the ai; if needed, i can take a look at it for 1.7, but there is no way i will break the 1.6 ai :-)
20090228 08:59:57< zookeeper> esr, i did. if it's so then fine
20090228 09:00:16< zookeeper> i was just saying that i don't see what's poorly thought out about the feature itself
20090228 09:01:06< zookeeper> silene, ok. although i'm pretty sure 90% of authors think it's going to do what it doesn't currently, and use it accordingly..
20090228 09:15:03-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090228 09:23:44-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has quit ["On the road again"]
20090228 10:01:17-!- Appleman1234 [n=Appleman@131.181.101.7] has joined #wesnoth-dev
20090228 10:03:24-!- loonybot [n=loonybot@79.139.247.203] has joined #wesnoth-dev
20090228 10:04:01-!- loonycyborg [n=sergey@79.139.247.203] has joined #wesnoth-dev
20090228 10:15:29-!- stikonas [n=stikonas@wesnoth/translator/stikonas] has joined #wesnoth-dev
20090228 10:16:53-!- EdB [n=EdB@79.95.12.139] has joined #wesnoth-dev
20090228 10:28:58-!- Sirp_ [n=me@c-71-198-3-104.hsd1.ca.comcast.net] has quit ["leaving"]
20090228 10:35:42-!- Noyga [n=lame-z@wesnoth/developer/noyga] has joined #wesnoth-dev
20090228 10:46:46-!- Appleman1234 [n=Appleman@131.181.101.7] has quit ["Leaving"]
20090228 10:49:43< Ivanovic> moin
20090228 10:55:42-!- noy [n=Noy@wesnoth/developer/noy] has quit []
20090228 11:13:04-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090228 11:14:46-!- mordante [n=mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20090228 11:15:03< mordante> hi
20090228 11:17:06< fendrin> hi mordante
20090228 11:17:16< mordante> hi fendrin 
20090228 11:18:14-!- boucman [n=rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20090228 11:34:37-!- Dragonking [n=dk@wesnoth/developer/dragonking] has joined #wesnoth-dev
20090228 11:37:38-!- mordante_ [n=mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20090228 11:52:57-!- mordante [n=mordante@wesnoth/developer/mordante] has quit [Read error: 110 (Connection timed out)]
20090228 11:57:37< CIA-70> fendrin * r33136 /trunk/data/tools/emacs_mode/ (5 files): Updated the wml emacs mode to the new 1.3.5 upstream version.
20090228 12:01:11-!- Tigge [n=tigge@bacchus.olf.sgsnet.se] has quit [Read error: 131 (Connection reset by peer)]
20090228 12:07:57-!- Elvish_Pillager [n=eli@66-189-40-122.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090228 12:10:25-!- mordante_ is now known as mordante
20090228 12:13:35-!- Tigge [n=tigge@bacchus.olf.sgsnet.se] has joined #wesnoth-dev
20090228 12:35:06< CIA-70> fendrin * r33137 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/ (4 files): LoW 1+2+3+7 applied balancing patches from Santi.
20090228 12:44:10< CIA-70> mordante * r33138 /trunk/src/gui/widgets/ (widget.hpp window.cpp window.hpp): 
20090228 12:44:10< CIA-70> Allow widgets to get the same size.
20090228 12:44:10< CIA-70> This is an initial implementation, it still needs quite some extra work,
20090228 12:44:10< CIA-70> which will be postponed until 1.7.
20090228 12:46:06< esr> Ivanovic: ping?
20090228 12:46:22< Ivanovic> esr: pong
20090228 12:49:02< esr> Ivanovic: A releasable state seems imminent.  There are only two remaining bugs in Security/Important categories and good reason to believe they'll be fixeed in the near term.  I'd say, ewait onbn fixes for those and ship an RC which seems likely to be the final one.
20090228 12:49:03-!- BenUrban [n=benurban@unaffiliated/benurban] has quit [Read error: 54 (Connection reset by peer)]
20090228 12:49:15-!- BenUrban [n=benurban@c-68-49-10-243.hsd1.md.comcast.net] has joined #wesnoth-dev
20090228 12:49:25< esr> s/ewait onbn/wait on/
20090228 12:49:28< Ivanovic> esr: i will follow the original plan
20090228 12:49:32< Ivanovic> that is: rc tomorrow
20090228 12:49:34< mordante> I still like to get a few fixes in today for the listbox sorting
20090228 12:49:41< Ivanovic> and another rc a week later to be really sure that it is fine
20090228 12:49:49< Ivanovic> then the final, if it really is okay
20090228 12:49:55< mordante> :-)
20090228 12:49:56< esr> Right, I'm not suggesting accelerating the final date.
20090228 12:50:22< mordante> I'm against that I still need to do quite some things and don't want to hurry it even more
20090228 12:50:42< mordante> sorry misread
20090228 12:51:00< esr> mordante: I'm in favcor of giving you as much time as you want, actually :-)
20090228 12:51:31< esr> The more proress you make on the graphics bugs, the better.
20090228 12:51:59< mordante> :-)
20090228 12:52:20< mordante> well I still hope two fix campaign dialog bugs
20090228 12:52:43< Elvish_Pillager> Hey guys - does anyone know where the python AI files are supposed to go? I had them in (campaign)/data/ais/ but that's not working in trunk (it says it can't find the files.)
20090228 12:52:51< loonycyborg> esr: What is the difference between RC and beta? :P
20090228 12:53:12< esr> loonycyborg: Our attitude about it.
20090228 12:53:26< esr> (That was a ha ha only serious.)
20090228 12:54:11< esr> loonycyborg: What have you got planned for near future?
20090228 12:54:13< Ivanovic> Elvish_Pillager: into the trashbin!
20090228 12:54:20 * loonycyborg uses his attitudeometer to measure attitude difference
20090228 12:54:20< Ivanovic> python ai was removed with 1.5.11
20090228 12:54:25< Elvish_Pillager> :o
20090228 12:54:42< Elvish_Pillager> I didn't see that in the changelog...
20090228 12:55:08< Elvish_Pillager> so, is there any way to do custom AI now?
20090228 12:55:17< esr> silene: Still around?
20090228 12:55:31< Ivanovic> Elvish_Pillager: basic stuff should be possible via formula ai
20090228 12:55:35< Ivanovic> ask Dragonking for details
20090228 12:55:37< loonycyborg> Elvish_Pillager: Perhaps someone will add Haskell AIs :P
20090228 12:55:52< Elvish_Pillager> and what is the extent of this "basic stuff"?
20090228 12:55:59 * Elvish_Pillager asks with trepidation...
20090228 12:56:06< Ivanovic> Elvish_Pillager: like i said, ask Dragonking
20090228 12:56:21< Dragonking> Elvish_Pillager: Define 'custom AI'
20090228 12:56:28< Dragonking> ;)
20090228 12:56:49< Elvish_Pillager> well, first and most importantly is an AI that will make its units try to flee to the edges of the map
20090228 12:56:54< Dragonking> Elvish_Pillager: YOu may create AI for single unit, for groups of units that meet some condition or for whole side
20090228 12:57:25< Elvish_Pillager> Is there documentation on this somewhere?
20090228 12:57:25< Dragonking> Elvish_Pillager: Possible, just I need to fix first move function :)
20090228 12:57:46-!- wildpenguin [n=chpln@ppp121-45-8-208.lns10.adl2.internode.on.net] has joined #wesnoth-dev
20090228 12:57:47< Dragonking> http://www.wesnoth.org/wiki/FormulaAI
20090228 12:57:57< Elvish_Pillager> I guess I also need to ask _when_ this system will be available? :/
20090228 12:58:15< Dragonking> 1.6
20090228 12:58:16< silene> esr: yes
20090228 12:58:26< Dragonking> And it is in 1.5
20090228 12:58:27< loonycyborg> esr: Regarding wesnoth I have no cogerent plans for near future.
20090228 12:58:57< esr> I meant to ask you last night if you'd have a look at  bug #12839.
20090228 13:02:04< silene> esr: i will take a look
20090228 13:02:17< esr> Thanks.
20090228 13:03:16< fendrin> esr: I have fixed all bugs that are in my campaigns and related to wml. Since I have time now, feel free to assign new wml bugs to me.
20090228 13:04:34< esr> fendrin: Thanks, and I will, and (yay!) I din't think we have any left.  Checking...
20090228 13:06:17< esr> ...nope.  It's till an outside possibiility that  #12983 is WML, but I agree witn you that it's *most likely* an engine problem.
20090228 13:08:47< esr> fendrin: At this point I'd say take a potshot at any of the ramaininfg 43 that you thing you might be able to fix.
20090228 13:09:29< esr> Oh - if you could test and clear any of the "Ready For Test" bugs, that would be great.
20090228 13:13:16-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090228 13:13:24< fendrin> esr: okay
20090228 13:15:57< Elvish_Pillager> Dragonking: I am mostly placated by this Formula AI thing (assuming it will work). I have a couple questions -
20090228 13:16:13< Elvish_Pillager> 1) what _order_ do the unit-specific AIs act in at the beginning of the AI turn?
20090228 13:17:08< Dragonking> Elvish_Pillager: Order is not specified, it;s just how they are defined in unit_map inside the game
20090228 13:17:20< Elvish_Pillager> 2) an example on the FormulaAI page uses a function avg_damage_inflicted, which isn't mentioned on the FormulaAI_Functions page. Is there a way for me to find a list of all functions (somewhere in the source, maybe?)
20090228 13:17:26< Dragonking> Tho you gave me idea of possibility to assign priorities to the unit AIs..
20090228 13:18:22< Elvish_Pillager> it seems to me that individual-unit AIs would be most useful for random minions, making it not too important
20090228 13:18:57< Elvish_Pillager> although the possibility of making Majesty-like scenarios is now intriguing... :P
20090228 13:19:20< Dragonking> Elvish_Pillager: avg_damage_inflicted is not an function
20090228 13:19:26< Dragonking> Is member of 'attacks' object
20090228 13:19:33< Elvish_Pillager> oh, I see
20090228 13:19:42< Dragonking> Stores average damage that we can inflict to the unit
20090228 13:19:45< Dragonking> IIRC
20090228 13:20:02< Dragonking> Elvish_Pillager: Members are not located anywhere..
20090228 13:20:07< Elvish_Pillager> in that case, are there descriptions of the data structures somewhere?
20090228 13:20:12< Dragonking> Nope
20090228 13:20:15< Elvish_Pillager> :(
20090228 13:20:25< Dragonking> Elvish_Pillager: There is function to check members
20090228 13:20:32< Elvish_Pillager> ah
20090228 13:20:33< Elvish_Pillager> heh
20090228 13:20:33< Dragonking> dir(attacks[0]) should work
20090228 13:21:09< Dragonking> Elvish_Pillager: press f in trunk or 1.5.x to get into formula commandline
20090228 13:21:15< Elvish_Pillager> yeah
20090228 13:21:59-!- malumalu [n=malu@hnvr-4dbb5332.pool.einsundeins.de] has joined #wesnoth-dev
20090228 13:24:53-!- ilor [n=user@wesnoth/developer/ilor] has quit [Read error: 60 (Operation timed out)]
20090228 13:25:20< Elvish_Pillager> Dragonking: priority would be nice if it could be given as a function.
20090228 13:26:23< Dragonking> Elvish_Pillager: You can always use candidate moves 
20090228 13:26:40< Elvish_Pillager> candidate moves?
20090228 13:26:43< Dragonking> And specify eval function
20090228 13:26:51< Elvish_Pillager> right
20090228 13:26:59< Dragonking> Ah, it's not documented....
20090228 13:27:01< Elvish_Pillager> not for individual unit AIs though
20090228 13:27:10-!- ilor [n=user@avc112.neoplus.adsl.tpnet.pl] has joined #wesnoth-dev
20090228 13:27:17< Dragonking> Yup
20090228 13:27:23< Dragonking> But you can filter units
20090228 13:27:29< Dragonking> And eval are done for units
20090228 13:31:45< Elvish_Pillager> I see. the FormulaAI page on the wiki is pretty incomplete.
20090228 13:31:54< Dragonking> It is a bit
20090228 13:32:14< Dragonking> I am ony one maintaining it now unfortunately
20090228 13:32:27< Dragonking> 'only
20090228 13:32:32< Elvish_Pillager> if I end up using it successfully, I should help with keeping the page complete
20090228 13:32:44< Dragonking> That would be great
20090228 13:33:04< Dragonking> Elvish_Pillager: There are test scenatios that use formula AI
20090228 13:33:11< Elvish_Pillager> cool
20090228 13:33:13< Dragonking> Run wesnoth with -t formula  
20090228 13:33:37< Elvish_Pillager> does not start
20090228 13:33:47< Dragonking> Is it?
20090228 13:35:05< zookeeper> EP, if i had to make units flee to the edges of the map, i'd do it with WML, not with these silly AI language hacks ;)
20090228 13:36:29< Elvish_Pillager> err, and how would you do that?
20090228 13:36:57< Dragonking> Elvish_Pillager: -t formula -d
20090228 13:38:07< zookeeper> EP, by using gotos?
20090228 13:38:40< Elvish_Pillager> zookeeper, and how would I decide where to put the gotos? I want them to flee in a way that isn't stupid.
20090228 13:39:04< Elvish_Pillager> Leaving the default AI to decide attacks isn't cool either.
20090228 13:39:23< esr> Ivanovic: I introduced new strings with that help fix, but they haven't shown up in the translation stats yet.  Does this mean you need to run a translation update?
20090228 13:39:39< Dragonking> Formula to do something like this is supposed to be quite simple,  and it should be, just there is still one bug that I need to fix linke I said earlier
20090228 13:39:47< Dragonking> I'll try to fix it today since I have some time
20090228 13:39:52< esr> A pot update, I mean?
20090228 13:40:02< wildpenguin> Dragonking: This reminds me: I'm concerned about the amount of information available to players via FormulaAI.
20090228 13:40:27< Elvish_Pillager> Dragonking: yeah, it'll be easier that writing the same thing in Python, which I did already :)
20090228 13:40:34< zookeeper> EP, well, you could use some rough conditionals for figuring out the nearest and least-populated-by-enemies edge.
20090228 13:40:43< Dragonking> wildpenguin: Why?
20090228 13:40:53< Elvish_Pillager> zookeeper: OR I could do it PERFECTLY by using formula AI
20090228 13:41:17< zookeeper> EP, sure, if you can. i just said i wouldn't ;)
20090228 13:41:23< wildpenguin> That is, something as simple as "filter(units, <opposing_side>)" will make render fog useless.
20090228 13:41:42< Elvish_Pillager> Dragonking: does move() leave the unit unable to attack, or just unable to move further?
20090228 13:41:52< Dragonking> unable to move IIRC
20090228 13:42:01< Elvish_Pillager> (sounds like the right idea)
20090228 13:42:41< Dragonking> wildpenguin: You are not supposed to use formula AI in mp, and using it in sp is just like looking at savegame 
20090228 13:43:18< Dragonking> Use formul AI in mp as: formula AI commandline
20090228 13:43:22< wildpenguin> Not being supposed to is not much of an assurance.  It provides full scenario information to players all the same.
20090228 13:43:22< Dragonking> Not genrally
20090228 13:43:55< Dragonking> wildpenguin: How will you acess this info if you can't use formula commandline?
20090228 13:44:12< loonycyborg> Dragonking:  IMO formula AI is still too low level. It would be nice if there were high level constructs.
20090228 13:44:21< wildpenguin> As it is now, I believe this is still available to players in MP?
20090228 13:44:22< Dragonking> You caoul "prepare" scenatio that prints out some info, but you need to host it
20090228 13:44:34< loonycyborg> Which don't deal with individual units, but with unit groups.
20090228 13:44:40< Dragonking> wildpenguin: It shouldn't be, if it is, then we have a bug.
20090228 13:44:59< Elvish_Pillager> Dragonking: (running the test scenario) ...it's very slow? :/
20090228 13:45:04< wildpenguin> I've tested this with 1.5.11.  It still works.
20090228 13:45:10< Dragonking> Elvish_Pillager: Why slow?
20090228 13:45:17< Dragonking> wildpenguin: ok, thanks for info
20090228 13:45:41< Elvish_Pillager> I take it it's not supposed to stall after the orc archer and the walking corpse move?
20090228 13:45:47< Elvish_Pillager> (it is doing so)
20090228 13:45:56< Dragonking> it is not supposed
20090228 13:46:08< Elvish_Pillager> oh it's giving errors
20090228 13:46:27< Elvish_Pillager> and not responding to ctrl-C
20090228 13:46:36< Dragonking> Interesting
20090228 13:47:16< Dragonking> Works here just fine
20090228 13:47:24 * Elvish_Pillager updates to latest SVN
20090228 13:47:57< Dragonking> Elvish_Pillager: If it still gives errors, please paste them somewhere and gimme link, ok?
20090228 13:48:09< Elvish_Pillager> okay
20090228 13:52:13< Dragonking> loonycyborg: What do you mean exactly by "deal with unit groups" ?
20090228 13:53:48< Dragonking> Elvish_Pillager: formula AI is still a bit buggy... if you encounter problems please talk to me :)
20090228 13:55:11< loonycyborg> E.g. a high level variant of move() which tells a group units to move to a location while using appropriate terrain and defending itself.
20090228 13:55:46< Dragonking> loonycyborg: Anyone is free to write such high level function using existing ones
20090228 13:56:36< loonycyborg> Most likely they won't bother to write it :)
20090228 13:56:55< Elvish_Pillager> I would definitely write it if I had an occasion to use it
20090228 13:57:18< Dragonking> loonycyborg: Notice, we do have library of build-in functions that are low-level, but we plan to make big library of functions defined in formula language that are more 'high level'
20090228 13:57:19< loonycyborg> Since they're interesting in writing their campaigns, not hacking AIs.
20090228 13:57:31< Elvish_Pillager> I'd even make it dynamically choose scout units and send them slightly ahead so that it can decide whether to use defensive terrains or just go with greatest speed
20090228 13:57:32< loonycyborg> Dragonking: Great.
20090228 14:02:22< Elvish_Pillager> Dragonking: If I'm interpreting this candidate_move thing correctly, couldn't you do the same thing quite easily without it?
20090228 14:03:15< Dragonking> Elvish_Pillager: What kind of 'same thing' ?
20090228 14:03:31< CIA-70> mordante * r33139 /trunk/src/config.cpp: Added constructor to initialize all members.
20090228 14:03:33< CIA-70> mordante * r33140 /trunk/src/gui/widgets/settings.hpp: Initialize all members.
20090228 14:03:36< CIA-70> mordante * r33141 /trunk/src/playmp_controller.cpp: Initialize all members.
20090228 14:03:40< Elvish_Pillager> giving moves priorities, then choosing a move with the highest priority
20090228 14:03:45< CIA-70> mordante * r33142 /trunk/ (4 files in 2 dirs): 
20090228 14:03:45< CIA-70> Listbox columns have the same width again.
20090228 14:03:45< CIA-70> The header row still has a different width, which will be fixed later.
20090228 14:04:35< mordante> the changelog for the current version seems quite empty and IIRC quite some bugs are fixed since the last release...
20090228 14:04:54< Dragonking> Elvish_Pillager: You would need to write custom filter, then chose one with top score, then do some code regarding that
20090228 14:05:23< Dragonking> Elvish_Pillager: Candidate moves are great, since you can evaluate normal candidate move, for unit, of attack candidate move, for unit-enemy pair
20090228 14:05:44< Dragonking> It's much simpler to use it than to write something simillar in general formula
20090228 14:05:53< Elvish_Pillager> can you not have a function which takes a move or a move-and-attack interchangably?
20090228 14:06:00-!- Tigge [n=tigge@bacchus.olf.sgsnet.se] has quit [Remote closed the connection]
20090228 14:06:46< Dragonking> Elvish_Pillager: I don't understand what you mean
20090228 14:08:25-!- isaac [n=isaac@173.Red-79-154-223.dynamicIP.rima-tde.net] has quit [Read error: 104 (Connection reset by peer)]
20090228 14:09:06-!- isaac [n=isaac@debian/developer/isaac] has joined #wesnoth-dev
20090228 14:09:09< Elvish_Pillager> well, why wouldn't it work to just say move=choose(generate_moves_function, eval_function) ?
20090228 14:10:24< Dragonking> it would, just then write generate_moves_function
20090228 14:10:48< Dragonking> It would be far more complex to do same thing properly in general fomrula
20090228 14:10:55< Dragonking> Thank it is now with candadate moves
20090228 14:11:35< Elvish_Pillager> what, can I not have a data structure that's a list of moves and attacks, and combine the things?
20090228 14:12:01< Dragonking> You have list of attacks
20090228 14:12:05< Dragonking> 'attacks'
20090228 14:12:13< Dragonking> Along with dst where to move to attack
20090228 14:12:35< Elvish_Pillager> I want a list that includes both moves and attacks
20090228 14:12:58< Dragonking> You have another one 'possible_moves'
20090228 14:13:07< Dragonking> Or something like that
20090228 14:13:12< Dragonking> It is function I tihnk
20090228 14:13:18< Dragonking> Specified for one unit
20090228 14:13:48< Dragonking> Elvish_Pillager: Candidate moves are used for 'specialized' things, general formulas are general
20090228 14:14:55< Dragonking> If you want few candidate moves, you just write them, and you have quite clear scenario file, writing general formula to act like candadate moves that takes care of different things would be dirty
20090228 14:15:23< Dragonking> Need to filter out thinkg, do some choosing etcd - it is done already by candidate moves itself
20090228 14:19:00-!- isaac_ [n=isaac@14.Red-88-26-177.staticIP.rima-tde.net] has joined #wesnoth-dev
20090228 14:19:28-!- wildpenguin [n=chpln@ppp121-45-8-208.lns10.adl2.internode.on.net] has quit [Remote closed the connection]
20090228 14:23:36< Ivanovic> esr: yes, for strings to appear i have to run a pot-update
20090228 14:24:00< Ivanovic> but i don't want to run one after each commit, so i tend to wait for "some days" so that string changes can sum up a little
20090228 14:24:12< esr> Ivanovic: OK.
20090228 14:26:20< Ivanovic> running a pot-update now
20090228 14:26:41< Ivanovic> or are there any outstanding "small" bugfixes that will require stringchanges?
20090228 14:26:58-!- Baufo [n=thomas@wesnoth/developer/baufo] has joined #wesnoth-dev
20090228 14:27:11-!- isaac__ [n=isaac@173.Red-79-154-223.dynamicIP.rima-tde.net] has joined #wesnoth-dev
20090228 14:27:14< esr> Ivanovic: Not that I know of.
20090228 14:27:14< Elvish_Pillager> Dragonking: err, how does a formula AI end its turn?
20090228 14:28:07< Ivanovic> ettin: around?
20090228 14:28:33< Ivanovic> ettin: i saw you commited some stuff for the new announcement, can the text esr created so far already be added?
20090228 14:30:06-!- isaac [n=isaac@debian/developer/isaac] has quit [Nick collision from services.]
20090228 14:30:08-!- isaac__ is now known as isaac
20090228 14:30:13< Dragonking> When all candidate moves evals return number <0 or when in general formula there is nothing more to do
20090228 14:30:51< Elvish_Pillager> what does the general formula return for "nothing more to do"?
20090228 14:31:18-!- fabi [n=fabi@78.52.73.6] has joined #wesnoth-dev
20090228 14:32:36< Dragonking> Elvish_Pillager: That it does not evaluate to a move
20090228 14:32:45< Elvish_Pillager> okay.
20090228 14:32:51< Dragonking> Elvish_Pillager: You can also write 'end_turn' 
20090228 14:33:03< Elvish_Pillager> cool
20090228 14:34:55< Dragonking> Elvish_Pillager: Generally, formula evaluates to something, it if is move, an attack, settign some variable, recruit, it is evaluated again, otherwise it ends turn
20090228 14:36:49< CIA-70> ivanovic * r33143 /trunk/ (244 files in 8 dirs): 
20090228 14:36:49< CIA-70> pot-update (several string changes!!!)
20090228 14:36:49< CIA-70> regenerated doc files
20090228 14:41:21-!- isaac_ [n=isaac@14.Red-88-26-177.staticIP.rima-tde.net] has quit [Read error: 113 (No route to host)]
20090228 14:46:49< Elvish_Pillager> Dragonking: the test scenario seems to work now, although it still outputs this error message several times: http://pastebin.com/m27d086d
20090228 14:47:17< Elvish_Pillager> aaaaand it apparently crashes if you pick "back to turn 1" from the menu.
20090228 14:47:26< Elvish_Pillager> but that probably has nothing to do with the AI.
20090228 14:47:39-!- fendrin [n=fabi@wesnoth/developer/fendrin] has quit [Read error: 110 (Connection timed out)]
20090228 14:47:47< Elvish_Pillager> oops wrong link
20090228 14:48:20< Elvish_Pillager> http://pastebin.com/m494c3768
20090228 14:49:30< Dragonking> I'll look at it later
20090228 14:50:02 * Dragonking enqueues things to do with formula AI
20090228 14:50:05< Elvish_Pillager> hang on, got the same error as before
20090228 14:50:22< Elvish_Pillager> if you move the elvish archer to the castle hex southeast of your leader
20090228 14:50:33< Elvish_Pillager> and then recruit four dwarves and end turn
20090228 14:51:20< Dragonking> Elvish_Pillager: I see, thanks
20090228 14:51:31< Elvish_Pillager> it just stalls and repeatedly says: http://pastebin.com/md8dcd44
20090228 14:51:34< Dragonking> Looks like a broken formula
20090228 14:52:55< zookeeper> esr, "these specis are lesser"? specis?
20090228 14:53:14< Elvish_Pillager> formula AI can't recall? heh :P
20090228 14:53:19< esr> Ooops.  Typo.
20090228 14:53:37< Dragonking> Elvish_Pillager: You think it should? ;)
20090228 14:53:38< zookeeper> also i don't think the word "species" is used a lot at all. shouldn't you just use "race"?
20090228 14:53:54< Elvish_Pillager> Dragonking: it should be able to do anything a human player can
20090228 14:53:57< fabi> zookeeper: species != race
20090228 14:54:12< zookeeper> yeah, and race is used for everything, not species.
20090228 14:54:15< Elvish_Pillager> well, both terms are vague enough!
20090228 14:54:16< Dragonking> Elvish_Pillager: Maybe in the future.. ;)
20090228 14:54:29< Elvish_Pillager> Dragonking: but I want to make an AI to win campaigns for me ;)
20090228 14:54:29< esr> fabi is right, though it's a distinction Wesnoth has tended to flout.
20090228 14:55:10< Elvish_Pillager> I mean, making an AI that meticulously levels up its units just wouldn't be cool without the ability to recall them. :P
20090228 14:55:10< loonycyborg> Elvish_Pillager: I'm getting those errors on stdout too, but it doesn't stall for me.
20090228 14:56:03< Dragonking> It does for me.
20090228 14:56:06< Elvish_Pillager> you made the same set of moves on turn one, and are getting exactly the same error messages, but the AI moves?
20090228 14:56:36< loonycyborg> Yes. It keeps moving after have moved archer and wlking corpse.
20090228 14:56:47< zookeeper> ers, no one disagrees with fabi on that.
20090228 14:56:58< loonycyborg> *after having moved archer and walking corpse
20090228 14:57:20< Dragonking> What kind of move is next?
20090228 14:57:38< loonycyborg> IIRC ghost
20090228 14:58:10< Ivanovic> esr: another stringchange imminent?
20090228 14:59:01< esr> Ivanovic: I'll put it in pofix.
20090228 14:59:10< Ivanovic> just fix it and i'll directly update
20090228 14:59:37< loonycyborg> Wait.. I've moved the elf to southermost castle tile.
20090228 14:59:39< Ivanovic> that is *better* than having everything in pofix unless pofix eventually gets too many "uhm, you will have to handfix this due with the comments"
20090228 15:00:34< esr> Ivanovic: So you want me not to put it in pofix, just fix the original string?
20090228 15:00:45< Ivanovic> yes
20090228 15:00:54< Ivanovic> and make clear if it should be species or races
20090228 15:01:35< loonycyborg> OK. After properly screened the elf with dwarves I've got the freeze too.
20090228 15:01:47< CIA-70> esr * r33144 /trunk/data/core/help.cfg: Typo fix.
20090228 15:02:10< esr> Ivanovic: If you say "lesser races" in English you raise some ugly ghosys from history.
20090228 15:02:23< Ivanovic> ahh, i see what the string is
20090228 15:02:26< esr> s/ghosys/ghosts/
20090228 15:02:43< esr> I'd rather stick with "species" for this use.
20090228 15:02:56< Ivanovic> yeah, i got to agree with this one
20090228 15:04:04-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090228 15:05:55< CIA-70> ivanovic * r33145 /trunk/po/wesnoth/ (48 files): just another pot-update (fix to one of the newly introduced strings)
20090228 15:12:23< Ivanovic> wow, this is really a perfect report... https://gna.org/bugs/index.php?13083
20090228 15:13:07< Elvish_Pillager> lol
20090228 15:13:51< mordante> yes I love it :-/
20090228 15:14:25< mordante> I hoped the original submitter would add some more info, but seems it was not an error to give no more info
20090228 15:15:34-!- Turuk_ [n=Turuk@FUSE-WS1-208-102-202-188.fuse.net] has joined #wesnoth-dev
20090228 15:19:43< Elvish_Pillager> yes, the submitter's message appears to be truncated
20090228 15:19:59< mordante> :-)
20090228 15:20:46< Ivanovic> lets say it this way: it is minimalistic...
20090228 15:20:53< esr> I was wondering if you'd martk that one "Need Info".  I almost did myself.
20090228 15:21:02< mordante> I just did
20090228 15:21:12< esr> Ys, I see.
20090228 15:21:30< mordante> and the bug has also been posted anonymously... so I don't expect a reaction
20090228 15:21:35-!- BenUrban [n=benurban@unaffiliated/benurban] has quit ["Power failu"]
20090228 15:21:38< esr> Nor dpo I.
20090228 15:21:57< mordante> which means we close the report
20090228 15:23:43-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 110 (Connection timed out)]
20090228 15:24:17< silene> mordante: not yet, give it a few dayx, so that the submitter can notice your post
20090228 15:24:42< mordante> silene, yes I normally give them a few weeks
20090228 15:25:14< mordante> also that part of the gui will be rewritten post 1.6, which should fix all truncating issues
20090228 15:28:13< Ivanovic> mordante: so which "must be fixed before 1.6" items are still open on your list?
20090228 15:28:23< silene> it has been a long time, i have forgotten, and a cursory look doesn't bring anything relevant in the documentation: how do you jump to a given scenario in a campaign?
20090228 15:28:26< Ivanovic> are you done with the listbox for the moment, or is there more to come?
20090228 15:28:54< ettin> Ivanovic: yes, feel free to edit template.html
20090228 15:29:14< mordante> silene, :cl in debug mode
20090228 15:29:52< Ivanovic> ettin: beside this: any idea when you will find the time to create a "clear" list of commands for generating the po4a related docs?
20090228 15:30:25< Ivanovic> yes, i know that i am currently annoying you with this, i would just like to have documented in a human readable way what *has* to be done
20090228 15:31:27< Ivanovic> esr: where do you have your announcement text?
20090228 15:31:43< silene> mordante: thanks a lot (and i'm quite sure there was no interface back in the early days, nice)
20090228 15:31:51< mordante> Ivanovic, I'll update the wiki later today, I hope the finish some more items today
20090228 15:32:13< esr> Ivanovic: http://rafb.net/p/bxHRyu33.html
20090228 15:32:13< mordante> silene, no before you could only jump to the next, but IIRC alink added that feature
20090228 15:32:16< Ivanovic> that is: the template file for the announcement is in website/start/1.6 (instead of branches/ or trunk/ )
20090228 15:32:40 * mordante reads the announcement
20090228 15:33:31< ettin> Ivanovic: it isn't that difficult, actually (specially now that I updated the Makefile)
20090228 15:33:33< Ivanovic> esr: will you transform it into the html or shall i take care of it?
20090228 15:33:40< ettin> Ivanovic: but I'm not feeling very well today
20090228 15:33:51< Ivanovic> ettin: i am talking about the manual and the manpages stuff
20090228 15:33:55< Ivanovic> not the website thingie
20090228 15:34:03< Ivanovic> sure, no need to hurry at all
20090228 15:34:06< esr> Ivanovic: Please do it.  I'm going to have my hands full until this coming Wednesday.
20090228 15:35:44< mordante> do we want to keep the double spaces after a point?
20090228 15:36:11-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090228 15:36:17< ettin> Ivanovic: the release announcement also uses po4a
20090228 15:36:24< esr> mordante: They won't show in HTML.
20090228 15:36:36< mordante> "There are two new units (Dread Bat and Royal Warrior), and one new terrain (cave path)." the comma before and is that correct
20090228 15:36:40< mordante> good point
20090228 15:37:26< esr> mordante: The comma is correct but optional; there's a bitof a style war among grammarians about that.
20090228 15:37:58< mordante> ok, was not 100% sure
20090228 15:38:41< esr> mordante: Neither are most native speeakers :-)  See http://en.wikipedia.org/wiki/Serial_comma
20090228 15:40:25< silene> just wondering, i'm looking at the code of THoT 5, and i see both "{NO_SCOUTS}" and "recruitment_pattern=scout" in the ai parameters; is there a case where it makes sense?
20090228 15:40:59 * esr thinks that if all anyone questions is his comma placement he has done OK :-)
20090228 15:41:30 * mordante has more ;-)
20090228 15:41:40< esr> silene: YThat may be an error, I'll look at it.
20090228 15:42:00< mordante> "and more campaigns now include references to events in that overall history" s/that/the/ 
20090228 15:42:50< esr> Either form would be correct; "that" implies a back-reference to a specific earlier mention of the history, "the" does not.
20090228 15:42:58< mordante> I know ;-)
20090228 15:42:59< zookeeper> silene, if the side can only recruit scout units, then no
20090228 15:43:13< zookeeper> {NO_SCOUTS} only makes the side not recruit any extra scouts for village grabbing
20090228 15:43:32< zookeeper> it won't prevent it from picking them if they're part of its recruitment_pattern
20090228 15:43:34< mordante> to me "that" sounds like somebody made up some version of the history and "the" as if it's the real version of the history
20090228 15:43:44< silene> zookeeper: i see, thanks
20090228 15:43:47< mordante> and of course Wesnoth is real ;-)
20090228 15:44:00< esr> Ah, in that case {NO_SCOUTS} is correcty here, no worries.
20090228 15:44:37< Elvish_Pillager> Nobody's arguing that Wesnoth's not real. I happen to be friends with an Orcish Archer myself.
20090228 15:45:51-!- Turuk_ [n=Turuk@FUSE-WS1-208-102-202-188.fuse.net] has quit [Read error: 110 (Connection timed out)]
20090228 15:46:03< CIA-70> ivanovic * r33146 /website/start/1.6/template.html: 
20090228 15:46:03< CIA-70> add the preliminary version by esr, will need some work to really have it fit in
20090228 15:46:03< CIA-70> add some TODO blocks marking where stuff has to be altered
20090228 15:46:06< mordante> whow we even have friendship among elves and orcs ;-)
20090228 15:46:22< Elvish_Pillager> @#$%ing racism
20090228 15:46:42 * Elvish_Pillager vows to never again write a racist campaign
20090228 15:47:07< mordante> no the history tells us they normally only fight with eachother
20090228 15:47:24< mordante> seems we made some progress in Wesnoth
20090228 15:47:35< Elvish_Pillager> well, at least as much as it borders on Earth
20090228 15:47:40< Ivanovic> esr: to see how the announcement currently looks: http://svn.gna.org/viewcvs/*checkout*/wesnoth/website/start/1.6/template.html?rev=33146
20090228 15:47:41< Elvish_Pillager> Earth is more progressive than Wesnoth
20090228 15:47:50 * esr is busy because he's writing an analysis of the scope of the patents in the TomTom lawsuit.
20090228 15:48:01< Ivanovic> regarding the style: this is the 1.4 one: http://svn.gna.org/viewcvs/*checkout*/wesnoth/website/start/1.4/template.html?rev=33146
20090228 15:48:04< Ivanovic> esr: good luck
20090228 15:48:07< mordante> esr, line 42 ands
20090228 15:49:20< mordante> Ivanovic, do we really remove the gamma slider?
20090228 15:49:26< Ivanovic> yes, we do
20090228 15:50:30< mordante> ok, that means the dialog should be fixed accordingly, will look at it later today
20090228 15:51:01< Ivanovic> mordante: if you point me to where this is, i can have a look at moving the elements around
20090228 15:51:13< Ivanovic> so that you can concentrate on the more important stuff
20090228 15:51:42< mordante> it would also be nice to mention some of the gui stuff in the release notes 
20090228 15:51:59< Ivanovic> mordante: it would be nice to mention it "more clearly"
20090228 15:51:59< mordante> especially that it will allow users to fully define the looks of the gui
20090228 15:52:15< Ivanovic> that is: the dialog stuff is mentioned, but not as clearly as it could be
20090228 15:52:38< mordante> well I hope to get some people to start to look at their own ideas of widgets
20090228 15:54:40< mordante> Ivanovic, game_preferences_display.cpp
20090228 16:01:16< zookeeper> now take a look at these..
20090228 16:01:17< Ivanovic> hmm, lets see if i edited the correct place and it does work as expected
20090228 16:01:29< CIA-70> zookeeper * r33147 /trunk/data/core/ (103 files in 3 dirs): Jetryl's new pine, deciduous and mixed forests. Some variation tiles are still missing, but none of the old deciduous (non-winter) trees remain now.
20090228 16:02:04 * esr applauds zookeeper
20090228 16:04:10< CIA-70> ivanovic * r33148 /trunk/src/game_preferences_display.cpp: 
20090228 16:04:10< CIA-70> remove the space occupied by the not displayed gamma tab (this fixes bug #13050 )
20090228 16:04:10< CIA-70> at the moment the lines are just commented out, clean removal of all gamma slider related stuff has to follow later
20090228 16:05:00 * esr applauds Ivanovic, too.
20090228 16:06:00< Ivanovic> hmm, to cleanup the leftover gamma stuff will be "some more" work
20090228 16:06:54< mordante> I can do that while waiting for my compilation finishes
20090228 16:07:12< esr> Ivanovic: I'd suggest leaving as much of its possible in, but conditoned out, with a comment that it can be re-renabled when SDL fixes the upstream problem.
20090228 16:07:23< Elvish_Pillager> Dragonking: does formula AI have any form of "and" or "or" for conditionals?
20090228 16:07:36< Ivanovic> esr: the matter is that it is spread all over the place
20090228 16:08:07< Ivanovic> with other words: the leftovers clutter *many* places, just grep for 'gamma' in src/ and get what i mean
20090228 16:08:43< esr> Still, they'll fix the upstream problem someday.  Pribably.
20090228 16:08:44< Dragonking> Elvish_Pillager: Sure, try f and '0 and 1'
20090228 16:09:13< Dragonking> Elvish_Pillager: There is also 'not'
20090228 16:09:29< Elvish_Pillager> good. thank you for being my living documentation :P
20090228 16:09:41< Dragonking> You welcome. :)
20090228 16:10:00< mordante> esr, Ivanovic I prefer to get rid of the crud before 1.6
20090228 16:10:06< mordante> should I look at it?
20090228 16:10:25< esr> You'd be the logical person...
20090228 16:10:33< Ivanovic> mordante: i can look at it, though i might miss some places
20090228 16:10:47< Ivanovic> though i think in all items "gamma" is used, so i should be able to wipe it out
20090228 16:10:54 * mordante thinks I really got the all gui is yours hat
20090228 16:11:20 * esr thinks Mordante asked for this pain...
20090228 16:11:37< mordante> Ivanovic, I would do it the same way ;-) just want to avoid two people working on the same issue
20090228 16:12:02< mordante> esr, no I started to rewrite it since I didn't want to work on the old stuff :-/
20090228 16:12:34< esr> (intones) That's always how it begins...
20090228 16:13:31< mordante> yes, guess I should get rid of the old stuff asap
20090228 16:13:56< Ivanovic> mordante: concentrate on the stuff you wanted to work on and i will takle the display slider
20090228 16:14:10< Ivanovic> i just stepped over a problem with the fribidi check in cmake
20090228 16:14:39< mordante> Ivanovic, ok guess I can remove that point from my todo list then :-)
20090228 16:17:29-!- voris [n=voris@c-24-16-107-149.hsd1.wa.comcast.net] has joined #wesnoth-dev
20090228 16:25:19-!- EdB [n=EdB@79.95.12.139] has quit ["Konversation terminated!"]
20090228 16:25:26< CIA-70> esr * r33149 /trunk/po/wesnoth/en_GB.po: en_GB.po update for new help strings.
20090228 16:28:56< esr> AI0867: ping?
20090228 16:29:19< Elvish_Pillager> Dragonking: does shortest_path() take into account enemy units?
20090228 16:29:49< Dragonking> Elvish_Pillager: I don't remember
20090228 16:30:57< Dragonking> Elvish_Pillager: Yes, it does
20090228 16:31:13< Elvish_Pillager> including ZoC?
20090228 16:31:17< Dragonking> Yes
20090228 16:31:43< Elvish_Pillager> hmm.,
20090228 16:31:51< Dragonking> Elvish_Pillager: IN test scenario f 'shortest_path( my_leader.loc, loc(6,7) )
20090228 16:32:23< Dragonking> You can see, result path avoids archer's ZoC
20090228 16:32:33-!- Tigge [n=tigge@bacchus.olf.sgsnet.se] has joined #wesnoth-dev
20090228 16:33:18< Dragonking> Elvish_Pillager: Btw, if you type dir(self) you may find many intersting things. :)
20090228 16:35:34< Elvish_Pillager> I see. it's good for what it does.
20090228 16:39:07 * Dragonking found another formula bug -_-
20090228 16:43:02< Ivanovic> no idea why the check does give out an "okay, i found the function" though i don't find it in any header...
20090228 16:48:48< silene> esr: i don't think i can fix bug 12839 in a safe way, the ai just wasn't designed to cope with real-time modifications to the environment
20090228 16:49:19< mordante> Ivanovic, ?
20090228 16:50:22< esr> silene: Then please explain that in detail in a comment and mark it "Won't Fix".  I'll add a comment about this limitation to THoT WML.
20090228 16:50:37< Ivanovic> mordante: cmake probs
20090228 16:51:09< Ivanovic> mordante: that is: the fribidi detection is borked for me, as it was already added at the top of the CMakeLists.txt file
20090228 16:51:25< Ivanovic> it should detect that i don't have the "old" version of fribidi, but fails miserably
20090228 16:52:15-!- cjhopman [n=chris@wesnoth/developer/cjhopman] has quit [Read error: 110 (Connection timed out)]
20090228 16:53:40< mordante> I had more problems with some parts of cmake, that's why I want to move to 2.6 and get rid of some of the cruft in the current implementation
20090228 16:53:47< Elvish_Pillager> Dragonking: what's the structure of "map"? there's no doc and printing it is truncated to the screen size.
20090228 16:54:01< mordante> not sure why it's there and whether or not it's needed for 2.4
20090228 16:56:17< Dragonking> Elvish_Pillager: map as gamemap?
20090228 16:56:33< CIA-70> esr * r33150 /trunk/data/campaigns/The_Hammer_of_Thursagan/scenarios/05_Invaders.cfg: Document the limitation discovered in bug #12839.
20090228 16:56:50< Elvish_Pillager> never mind, got the information I needed from the commandline
20090228 16:56:56< Dragonking> it has 3 fields:  terrain, w, h   :  w is width, h i heigth
20090228 16:57:03< Dragonking> Elvish_Pillager: ok :)
20090228 16:57:41< CIA-70> mordante * r33151 /trunk/data/gui/default/window/mp_server_list.cfg: 
20090228 16:57:41< CIA-70> Cleanups and improvements.
20090228 16:57:41< CIA-70> This prepares the listbox to have it's header the same size as it's
20090228 16:57:41< CIA-70> content.
20090228 16:57:55< CIA-70> mordante * r33152 /trunk/src/gui/widgets/listbox.cpp: 
20090228 16:57:55< CIA-70> Make sure the new header and footer keep their id.
20090228 16:57:55< CIA-70> This is needed to make listboxes work properly for getting the proper
20090228 16:57:55< CIA-70> width.
20090228 16:58:00< CIA-70> mordante * r33153 /trunk/src/gui/widgets/listbox.cpp: 
20090228 16:58:00< CIA-70> Header now has the same size as the content.
20090228 16:58:01< CIA-70> The implementation still needs some polishing, but that can be done
20090228 16:58:03< CIA-70> later. Want to get it out soons so 1.6 can be ready in time.
20090228 16:58:05< CIA-70> mordante * r33154 /trunk/src/gui/widgets/window_builder.cpp: Update some comment.
20090228 16:58:09< CIA-70> mordante * r33155 /trunk/data/gui/default/window/addon_list.cfg: 
20090228 16:58:11< CIA-70> Cleanups and improvements.
20090228 16:58:13< CIA-70> The listbox now also has proper columns also for the header.
20090228 17:02:48-!- crimson_penguin [n=ben@64.201.60.212] has joined #wesnoth-dev
20090228 17:03:59< CIA-70> mordante * r33156 /website/start/1.6/template.html: Fix a typo.
20090228 17:13:02< CIA-70> ivanovic * r33157 /trunk/src/ (7 files): "real" removal of the gamma slider leftovers (slightly related to the fix for bug #33148 since the mess is getting cleaned up now...)
20090228 17:13:58< Ivanovic> ah, and you will love me for having to rebuild basically everything due to an edit of preferences.hpp
20090228 17:14:30< mordante> Ivanovic, we love you :-P
20090228 17:14:39< Ivanovic> i know
20090228 17:14:40< Ivanovic> ^^
20090228 17:15:28< Dragonking> Ivanovic: Who maintains cmake?
20090228 17:15:41< Ivanovic> Dragonking: that is probably basically me and now mordante
20090228 17:15:54< Elvish_Pillager> Dragonking: using the 'f' prompt allowed me to order an impossible move... :/
20090228 17:16:16< Dragonking> Ivanovic: I wonder, build result of cmake goes under src, why not directly under root directory?
20090228 17:16:26< Dragonking> Elvish_Pillager: Elaborate please.
20090228 17:16:54< Ivanovic> Dragonking: because it is set this way atm
20090228 17:17:04< Ivanovic> i think there is no specific reason for this
20090228 17:17:16< Dragonking> Ivanovic: Can it be changed?
20090228 17:17:20< Elvish_Pillager> my leader was sitting on 7,4 with no moves left. I entered attack(my_leader.loc, loc(8,4), loc(8,5)), and it moved and attacked.
20090228 17:17:23< Ivanovic> that is: should be simple edit in src/CMakeLists.txt to append some ../
20090228 17:18:18< Dragonking> Elvish_Pillager: hmmmmm I'll look into this
20090228 17:18:24< Dragonking> I have bad feeling about it...
20090228 17:18:33< mordante> you can configure the output directory, mine ends in bin
20090228 17:18:52< Elvish_Pillager> ordering another move-and-attack made it move, but not attack (presumably since it had no attacks left.)
20090228 17:19:06< mordante> and I will have a look at the entire cmake system post 1.6 and give it some love
20090228 17:19:23< Dragonking> mordante: How you can specify it?
20090228 17:20:43< mordante> I set BINDIR to bin
20090228 17:20:48< boucman> hey all
20090228 17:20:58< mordante> looked at the cmake files to do that
20090228 17:21:00< mordante> hi boucman 
20090228 17:21:23< boucman> so, what's up ?
20090228 17:21:23< Dragonking> I have bindir set to bin
20090228 17:21:28< boucman> release today ?
20090228 17:21:40< Dragonking> Better not looking at formula AI
20090228 17:21:43< AI0867> esr: pong?
20090228 17:22:21< boucman> Dragonking: it's that bad ?
20090228 17:23:36< Dragonking> boucman: I have whole queue of things to fix
20090228 17:23:58< boucman> Dragonking: hmm
20090228 17:24:08< boucman> poison handling will be 1.7 I guess...
20090228 17:24:20< Dragonking> Nah
20090228 17:24:28< Dragonking> Don't be so negative. ;p
20090228 17:26:00 * Dragonking goes to the shop, bbl
20090228 17:27:06-!- cjhopman [n=chris@wesnoth/developer/cjhopman] has joined #wesnoth-dev
20090228 17:28:41-!- malumalu_ [n=malu@hnvr-4dbb5332.pool.einsundeins.de] has joined #wesnoth-dev
20090228 17:30:31< Ivanovic> Dragonking: bindir is only for installing
20090228 17:30:40< Ivanovic> Dragonking: use EXECUTABLE_OUTPUT_DIR
20090228 17:40:08< Ivanovic> Dragonking: working on the changes to make the root the default output dir
20090228 17:40:16< Ivanovic> and the correct var to set is EXECUTABLE_OUTPUT_PATH
20090228 17:42:41< Dragonking> Ivanovic: Thanks
20090228 17:43:00< boucman> Turuk: stupid idea about thread resurection...
20090228 17:43:07< Turuk> What?
20090228 17:43:27< boucman> how hard would it be to have a forum mode that would pop-up an extra warning when replying to a thread that is older than a given date...
20090228 17:43:55< boucman> I am certain most thread res are done by people skimming the forum and forgetting that they are reading old threads
20090228 17:44:05-!- Jimm_ [n=jim@c-24-14-111-0.hsd1.il.comcast.net] has joined #wesnoth-dev
20090228 17:44:10< Turuk> Hmm, I'm not sure, though I do *think* that there is a mod that will lock threads after a certain time passes
20090228 17:44:16-!- malumalu [n=malu@hnvr-4dbb5332.pool.einsundeins.de] has quit [Read error: 110 (Connection timed out)]
20090228 17:44:46< Turuk> That would be a good thing to have, a warning "You are replying to a thread that has been inactive for over a month.  Thread resurrection is frowned upon, so think if what you have to say is really worth it."
20090228 17:44:51< CIA-70> ivanovic * r33158 /trunk/CMakeLists.txt: 
20090228 17:44:51< CIA-70> output binary files after building in "dir where cmake was run" and no longer in "dir where cmake was run"/src (as requested by Dragonking)
20090228 17:44:51< CIA-70> turn fribidi off by default since the detaction for "you got fribidi 2?" seems to not work as expected
20090228 17:45:44< Turuk> I will ask Shadow_Master as well, he might know more about that or be able to figure out how to do so
20090228 17:45:56< Dragonking> boucman: Any clue how to block formula commandline in MP games?
20090228 17:46:32< boucman> not really
20090228 17:46:42< boucman> should we ?
20090228 17:46:48< Dragonking> Definiately.
20090228 17:47:01< boucman> or should we rather block whatever info the player shouldn't acces (fogged unit)
20090228 17:47:45< Dragonking> How you would like to do that?
20090228 17:47:59< boucman> no idea, it's an open question
20090228 17:48:23< Dragonking> Blocking formula commandline should do the trick.
20090228 17:48:39< boucman> hmm
20090228 17:48:51< Dragonking> I can swar it was implemented long time ago...
20090228 17:48:52< boucman> there are already quite a few thing we block in MP (debug comes to mind)
20090228 17:49:09< boucman> you might find out how to know if we're mp in the debug activation code
20090228 17:49:32< Dragonking> ok
20090228 17:49:35< Dragonking> Thanks
20090228 17:53:44< Elvish_Pillager> Dragonking: it looks like id and type are mixed up in the formula AI. what's id in WML is type in fai and vice versa.
20090228 17:53:46< Ivanovic> Dragonking: could you check if my changes work nicely for you?
20090228 17:54:00< Dragonking> Ivanovic: I'll
20090228 17:54:10-!- jorda [n=jorda@6.207.219.87.dynamic.jazztel.es] has joined #wesnoth-dev
20090228 17:54:15< Dragonking> Elvish_Pillager: Are you sure?
20090228 17:54:23< Elvish_Pillager> e.g. id -> 'Dwarvish Steelclad', type -> 'side_1_leader'
20090228 17:54:26< Ivanovic> okay
20090228 17:54:38< Elvish_Pillager> that's what I get when I enter my_leader
20090228 17:55:23< Dragonking> And ID in WML is name of the unit?
20090228 17:56:09< Elvish_Pillager> well, not exactly, ID in WML is internal.
20090228 17:57:10< Elvish_Pillager> but in campaigns, yes, it's usually the non-translatable name of named units.
20090228 17:57:22< Dragonking> Same in formula
20090228 17:57:39< Elvish_Pillager> while type is the unit type.
20090228 17:57:50< Dragonking> hm
20090228 17:57:55< Dragonking> ok
20090228 17:58:00< Dragonking> I get it I think
20090228 17:59:05< Dragonking> Elvish_Pillager: I can't see 'type' in WML besides in [attack]
20090228 17:59:45< Elvish_Pillager> in single [unit]
20090228 17:59:54< Elvish_Pillager> unless it's changed recently
20090228 18:00:06< Dragonking> I look under UnitWML site
20090228 18:00:16< Elvish_Pillager> SingleUnitWML
20090228 18:00:25-!- alink [n=alink@wesnoth/developer/alink] has joined #wesnoth-dev
20090228 18:00:31< alink> hi
20090228 18:00:52< Dragonking> Elvish_Pillager: formula AI id is like id form UnitWML
20090228 18:01:01< mordante> hi alink 
20090228 18:01:02< Dragonking> type is like name
20090228 18:01:16< Dragonking> Unfortunate name, prolly should be changed
20090228 18:01:20< alink> hello mordante
20090228 18:01:26< Elvish_Pillager> yeah, and that's confusing, because id from UnitWML is the same as type from SingleUnitWML
20090228 18:01:35< Elvish_Pillager> and SingleUnitWML is what I'd expect it to reflect
20090228 18:03:29< Dragonking> You opened can with worms
20090228 18:03:53< Elvish_Pillager> no, the guys who made two different variables with the same name opened it.
20090228 18:04:18< CIA-70> mordante * r33159 /trunk/data/gui/default/widget/button_default.cfg: 
20090228 18:04:18< CIA-70> Refactor the button widget.
20090228 18:04:18< CIA-70> This way multiple button variations can be defined in the macro.
20090228 18:04:21< Dragonking> Well, it's been done according to UnitWML
20090228 18:04:28< Dragonking> With the ID stuff
20090228 18:04:48< Elvish_Pillager> yes, and IMO that's wrong, because you're displaying stats of a single unit, not of a unit type.
20090228 18:05:02< Elvish_Pillager> It's pretty clear-cut.
20090228 18:05:56-!- ettin [n=jorda@228.207.219.87.dynamic.jazztel.es] has quit [Read error: 110 (Connection timed out)]
20090228 18:06:18-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has left #wesnoth-dev []
20090228 18:07:21< Dragonking> So ID should display some internal ID, type should be unit non translatable name and 'name' key for translatable name
20090228 18:08:12< Elvish_Pillager> for formula AI?
20090228 18:08:24< Dragonking> Yes, for unit in formula AI
20090228 18:08:32< Dragonking> Cause there exists something like unit type as well
20090228 18:08:40< Dragonking> IN the formula AI
20090228 18:08:57< Elvish_Pillager> oh gosh
20090228 18:08:59 * zookeeper wonders why it just doesn't allow a regular SUF
20090228 18:09:00< Elvish_Pillager> let me look at that
20090228 18:09:32< Dragonking> Elvish_Pillager: See for example at possible recruits
20090228 18:09:47< Dragonking> s/See/Look/
20090228 18:10:15-!- Elvish_Pillage2 [n=eli@66-189-40-122.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090228 18:10:35< Elvish_Pillage2> yeah, FAI unit-type structure should reflect WML unit-type structure, and FAI single-unit structure should reflect WML single-unit structure.
20090228 18:11:06< Dragonking> ok
20090228 18:11:35 * Dragonking enqueues this too...
20090228 18:11:46< Elvish_Pillage2> which means WML single-unit 'id' (which is an internal ID) becomes FAI 'id', WML single-unit 'type' (which is the non-translatable name of the unit _type_) becomes FAI 'type'
20090228 18:11:59< Elvish_Pillage2> not sure if the type-name is given in WML single unit structure
20090228 18:13:04< zookeeper> single units have everything.
20090228 18:13:29-!- Jimm_ [n=jim@c-24-14-111-0.hsd1.il.comcast.net] has quit ["Leaving."]
20090228 18:13:59< zookeeper> not sure if that's relevant.
20090228 18:14:04-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090228 18:14:34< Elvish_Pillage2> yeah, WML single-unit 'language_name' is the translatable unit type name.
20090228 18:15:45< Dragonking> I see single unit has 'name' too
20090228 18:16:03< CIA-70> mordante * r33160 /trunk/ (6 files in 3 dirs): 
20090228 18:16:03< CIA-70> New transparant button in transparent dialog.
20090228 18:16:03< CIA-70> The art might need some more polishing, but rather leave that to an artist.
20090228 18:16:53< Elvish_Pillage2> yeah, 'name' is the _individual_ unit's translatable name.
20090228 18:17:04< Dragonking> Ivanovic: After building there's no wesnoth in root dir, it's still under src/
20090228 18:18:15-!- Jimm_ [n=jim@c-24-14-111-0.hsd1.il.comcast.net] has joined #wesnoth-dev
20090228 18:21:16< Elvish_Pillage2> Dragonking: if I have a source, destination, and which attack to use, is there a way to get information on the attack results? I can't figure out exactly what is in 'attacks'
20090228 18:22:25< Dragonking> Elvish_Pillage2: dir(attacks[0])
20090228 18:22:39< Dragonking> There is chance to kill, avg dmg taken etc
20090228 18:22:56< Dragonking> Elvish_Pillage2: attacks returns a vector of possible attacks
20090228 18:23:13< Elvish_Pillage2> how do I pick out which attack corresponds to which weapon?
20090228 18:23:56< Dragonking> Elvish_Pillage2: hmm... I *think* it corresponds to weapon that can inflict most damage
20090228 18:24:10< Elvish_Pillage2> So it _doesn't_ actually contain all possible attacks?
20090228 18:24:23< Dragonking> With all weapons - no
20090228 18:24:32< Dragonking> But I'll check to make sure
20090228 18:24:35< Elvish_Pillage2> the information it gives is a little haphazard, too
20090228 18:24:49< Dragonking> Well, I did not implement it :)
20090228 18:24:56< Elvish_Pillage2> average damage inflicted and taken is good. then it has chance to kill, and omits chance to die?
20090228 18:25:10< Dragonking> Apparently
20090228 18:26:13-!- Elvish_Pillager [n=eli@66-189-40-122.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090228 18:32:40-!- Sirp_ [n=me@c-71-198-3-104.hsd1.ca.comcast.net] has joined #wesnoth-dev
20090228 18:34:48< CIA-70> mordante * r33161 /trunk/ (17 files in 2 dirs): 
20090228 18:34:48< CIA-70> Add a new set of transparent navigation buttons.
20090228 18:34:48< CIA-70> These will be used in the transparent wml message dialog.
20090228 18:35:36-!- shadowmaster [n=imorelle@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090228 18:37:28< shadowmaster> boucman, Turuk I don't want to know about forum mods.
20090228 18:39:09< shadowmaster> esr: why did you rewrite Kaleh.cfg again?
20090228 18:39:17< shadowmaster> and how extensive was your testing?
20090228 18:40:40< shadowmaster> that is: did you ever realize that your first rewrite caused bugs that were hard to detect in-game?
20090228 18:41:16< CIA-70> dragonking * r33162 /trunk/src/formula_ai.cpp: Fix shortest_path function bug
20090228 18:42:43-!- Espreon [n=espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20090228 18:44:40< esr> shadowmaster: All I did was change some macros to do balanced markup, and I tested by advancing Kaleh and seeing that the results were correct.  I did it to shut up a wmliterator complaint.
20090228 18:45:26< shadowmaster> esr: did you advance hi to level 3?
20090228 18:45:33< CIA-70> mordante * r33163 /trunk/data/gui/default/widget/vertical_scrollbar_default.cfg: Made the scrollbar code more generic.
20090228 18:45:46< shadowmaster> Kaleh started having problems at level 2 in your last rewrite
20090228 18:45:46< esr> No. Is that where the old bugs showed up?
20090228 18:45:54< shadowmaster> esr: L2 or L3, yes
20090228 18:46:06< esr> Oh.  I'll run that test.
20090228 18:52:41< CIA-70> mordante * r33164 /trunk/images/buttons/ (12 files): Add the transparent groove images.
20090228 18:54:38< AI0867> esr: you pinged me?
20090228 18:55:33< Ivanovic> re
20090228 18:55:44< Ivanovic> Dragonking: how do you build, out of source or in the checkout?
20090228 18:55:53< fabi> The new 1.6 version requires a forum acount, right?
20090228 18:56:01< shadowmaster> no.
20090228 18:56:20< Dragonking> Ivanovic: I did svn up, make clean, autogen, configure, make
20090228 18:56:27< Ivanovic> ARGH
20090228 18:56:27< shadowmaster> you only need to use a forum account username and password if you want to
20090228 18:56:40< Ivanovic> Dragonking: you are not using cmake to build when using autogen+configure
20090228 18:56:52< Ivanovic> and i won't change the behaviour of the autotools build system
20090228 18:56:54< Dragonking> Ivanovic: ah, stry, i did after configure
20090228 18:56:56< Ivanovic> Dragonking: try cmake
20090228 18:56:57< stikonas> Ivanovic: https://gna.org/bugs/?13086 should probably be assigned to you.
20090228 18:57:04< Dragonking> Ivanovic: I used ccmake
20090228 18:57:20< Dragonking> after autogen
20090228 18:57:29< Dragonking> By 'configure' I meant 'ccmake .' :)
20090228 18:57:34< Ivanovic> you do not need autogen *at all*
20090228 18:57:39< Dragonking> ok
20090228 18:57:49< Ivanovic> and ccmake . is a bad choice since it only works on "already configured"
20090228 18:58:05< Dragonking> ok
20090228 18:58:24< Ivanovic> so please create an extra dir like wesnoth_cmake, cd into it and run: cmake PATH/TO/WESNOTH_CHECKOUT
20090228 18:58:28< fabi> shadowmaster: If I want to ... what?
20090228 18:58:48< Ivanovic> afterwards you can run "ccmake ." in this folder to alter values
20090228 18:58:48< boucman> fabi: any forum account name is "reserved" you need a forum passwd to use it
20090228 18:58:49< shadowmaster> you only need to use a forum account username and password if you want to use such
20090228 18:59:01< boucman> however any non-forum name can be used without pwd
20090228 18:59:16< fabi> Ah, I see.
20090228 18:59:19< Ivanovic> (they can also be used when running cmake PATH, but i am to lazy to list whatever arguments you want changed)
20090228 18:59:19< esr> AI0867: Yes, hold on...
20090228 18:59:29< Ivanovic> after altering settings, you have to hit 'c' and then 'g'
20090228 19:00:54-!- shikadibot [n=uid1001@wesnoth/bot/shikadibot] has joined #wesnoth-dev
20090228 19:01:35< esr> AI0867: I was wondering about bug #12428.  It appeasrs possible this was fixed by the removal of Python AIs.
20090228 19:01:39< Ivanovic> if you really want to run the buildtool stuff in your checkout dir, just call "cmake ." and afterwards you can already use make
20090228 19:01:45< Ivanovic> (or ccmake . to alter settings)
20090228 19:02:33< Dragonking> Ivanovic: ok, I though that if I ccmake .   then c then g    it confures and generates everything like it should
20090228 19:02:50< Ivanovic> the reason that nothing was changed for you is, that the var where i changed the default is one of the "standard variables" which will not alter anything if you have already configured it
20090228 19:03:13-!- shadowmaster is now known as Shadow_Master
20090228 19:03:20< Ivanovic> that is: it is automatically set to "empty" when cmake is run the first time
20090228 19:03:38< Ivanovic> and it will not alter a var "set" (set as in "it does exist, even if the value is empty)
20090228 19:03:47< Ivanovic> stikonas: replied in your report
20090228 19:03:59< mordante> Ivanovic, regarding bug 13086 I prefer to have the code in cmake
20090228 19:04:10< AI0867> esr: it still dies
20090228 19:04:19< AI0867> but the VALIDATE may not be as specific as it could be
20090228 19:04:34< Ivanovic> mordante: it depends on the steps needed
20090228 19:04:36< Elvish_Pillage2> Dragonking: It would be nice if 'attacks' had one entry for each weapon. It's barely useful as-is.
20090228 19:04:43< Ivanovic> i think it can be *strange* to have this stuff in cmake
20090228 19:04:51< CIA-70> mordante * r33165 /trunk/images/buttons/ (24 files): Add the transparent grip images.
20090228 19:04:53< loonycyborg> mordante: And I'd prefer to have it in scons(and I do)
20090228 19:05:08< Dragonking> Elvish_Pillage2: Well, it is useful for simple attacking
20090228 19:05:17< Dragonking> But I agree that it could be somehow extended
20090228 19:05:22< Ivanovic> [19:02:32] <Dragonking> Ivanovic: ok, I though that if I ccmake .   then c then g    it confures and generates everything like it should
20090228 19:05:26< Ivanovic> and this is 100% correct
20090228 19:05:29< esr> AI0867: Any prospect of a fix before 1.6?  Who knows this part of the code well?
20090228 19:05:39< Dragonking> Ivanovic: Then I did that.
20090228 19:05:54< mordante> loonycyborg, cool :-) I prefer not to have everything in shell scripts, otherwise there's no reason at all to move away from autotools ;-)
20090228 19:06:00< Ivanovic> but it uses the settings you entered in the ccmake interface, which is "empty" for the EXECUTABLE_OUTPUT_PATH (was so before my change for the default)
20090228 19:06:08< Ivanovic> afk, time for lunch
20090228 19:06:22< Dragonking> ok
20090228 19:06:24< Elvish_Pillage2> Dragonking: well, consider it like 'my_moves'. that contains every possible way you can move any of your units. It would make sense for 'attacks' to contain every possible way you can attack with your units.
20090228 19:06:24< loonycyborg> mordante: Even autotools is better than shell scripts :)
20090228 19:06:40< esr> ...though not by much :-)
20090228 19:07:28< mordante> well autotools tends to use shell scripts for harder things ;-)
20090228 19:08:51< loonycyborg> mordante: Really? Wesnoth's autotools buildsystem doesn't seem to be using shellscripts anywhere..
20090228 19:09:22< Shadow_Master> ?
20090228 19:09:35< Shadow_Master> the configure script is a huge generated shell script, for starters
20090228 19:09:41< CIA-70> mordante * r33166 /trunk/data/gui/default/ (4 files in 3 dirs): Add a new transparent scrolllabel for wml_message.
20090228 19:09:42-!- JW1 [n=X@c-71-57-85-168.hsd1.il.comcast.net] has joined #wesnoth-dev
20090228 19:09:47< mordante> sure, I thought it did at some points
20090228 19:09:54< loonycyborg> Ah. I forgot about that one.
20090228 19:09:55< Shadow_Master> talk about embedded shell scripting in the m4 sources
20090228 19:10:22< Shadow_Master> so, eh, try to remember how autotools works next time you criticize it ;-)
20090228 19:10:24< loonycyborg> But wesnoth's pot-update and manual generation stuff seems to be fully in make.
20090228 19:10:49< Shadow_Master> probably, I'm just correcting your hastily generalized statement
20090228 19:11:11< CIA-70> mordante * r33167 /trunk/data/gui/default/widget/vertical_scrollbar_default.cfg: Add a file forgotten in my last commit.
20090228 19:11:41< loonycyborg> Shadow_Master: I was referring to wesnoth's buildsystem, so it's not a generalization..
20090228 19:11:54< JW1> zookeeper: around?
20090228 19:12:11< fabi> fucking hell, what happened to the forests?
20090228 19:12:11< Shadow_Master> loonycyborg: so configure.ac doesn't count as part of the buildsystem? :-(
20090228 19:13:01< loonycyborg> Shadow_Master: configure.ac isn't a shell script, but a template processed by autoconf :)
20090228 19:13:35< Shadow_Master> loonycyborg: but it has embedded shell scripting
20090228 19:13:58< Shadow_Master> if test x$foo = xbar ...
20090228 19:16:56< loonycyborg> And there are very interesting reasons why you need x$foo = xbar and not $foo = bar :P
20090228 19:17:31< Shadow_Master> yes. interoperatibility with different implementations of the Bourne shell and avoiding m4 syntax clashing, IIRC
20090228 19:19:04< mordante> btw which build system do you use Shadow_Master ?
20090228 19:19:10< Shadow_Master> scons
20090228 19:19:20< loonycyborg> Yes. So don't even think about using unportable bash-specific stuff in it..
20090228 19:19:55< Shadow_Master> it's easier to manage -O3, debug, gprof'd and glibcxx debug builds in the same tree with scons
20090228 19:20:41< CIA-70> mordante * r33168 /trunk/data/gui/default/widget/horizontal_scrollbar_default.cfg: Made the scrollbar code more generic.
20090228 19:21:30-!- Lord_Aether [n=castle@207.212.136.10] has joined #wesnoth-dev
20090228 19:22:04< mordante> well I do that with cmake ;-)
20090228 19:22:11< Elvish_Pillage2> heh
20090228 19:22:20< mordante> haven't tried whether it's even possible with autotools
20090228 19:22:22< Elvish_Pillage2> my Python code is about five times as long as my formula AI code to do about the same thing
20090228 19:22:57< Shadow_Master> but cmake uses UGLY_UPPERCASE_DEFINES
20090228 19:23:02< CIA-70> dragonking * r33169 /trunk/src/menu_events.cpp: Disable formula AI in multiplayer
20090228 19:23:33< Elvish_Pillage2> (correction: it has about five times as many lines. Its lines are about half as long on average.)
20090228 19:23:37< CIA-70> mordante * r33170 /trunk/data/gui/default/widget/horizontal_scrollbar_default.cfg: Add the transparent horizontal scrollbar.
20090228 19:23:44< Shadow_Master> it --almost=feels like passing.the --parameters directly,to,the,compiler
20090228 19:25:02< loonycyborg> mordante: I find that it's very annoying that you need to use two utils to use(cmake/autogen + make) instead of one (scons)
20090228 19:25:42< mordante> I just really prefer Makefiles, which I can manually modify if needed
20090228 19:26:02< mordante> I often turn on -ggdb3 for a few files
20090228 19:26:04< loonycyborg> IMO the entire idea of makefile generators is screwy.
20090228 19:26:36< loonycyborg> mordante: I prefer to manually modify sconstruct files :P
20090228 19:26:59< boucman> Elvish_Pillage2: what are you working on ?
20090228 19:27:26-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev
20090228 19:27:34< mordante> well I'm quite sure we can't convince the other which system is better ;-)
20090228 19:27:56-!- Sirp_ [n=me@c-71-198-3-104.hsd1.ca.comcast.net] has quit ["leaving"]
20090228 19:28:21< loonycyborg> mordante: I'm not trying to convince you. I'm just explaining why I personally prefer scons.
20090228 19:35:20< mordante> and I why I like cmake
20090228 19:35:39< Shadow_Master> and why I don't like any of the three but prefer scons
20090228 19:35:42-!- EdB [n=EdB@239.12.95-79.rev.gaoland.net] has joined #wesnoth-dev
20090228 19:36:46< Elvish_Pillage2> boucman, a campaign
20090228 19:37:32< Elvish_Pillage2> right now I'm trying to get it back to the state it was in when Python AI existed and formula AI didn't :P
20090228 19:37:34< CIA-70> silene * r33171 /trunk/data/campaigns/An_Orcish_Incursion/scenarios/6_A_Detour_through_the_Swamp.cfg: Fixed typo.
20090228 19:44:25< mordante> Shadow_Master, I think the world can use another build system ;-)
20090228 19:44:50< loonycyborg> mordante: It can and it already does :P
20090228 19:45:10< loonycyborg> It uses an insane amount of build systems.
20090228 19:45:20< mordante> yep I know
20090228 19:45:59< loonycyborg> Unfortunately, many of them are based on ancient make utility.
20090228 19:46:16< loonycyborg> This is bad. It's time to move on.
20090228 19:47:12< Shadow_Master> there is a "new" make somewhere
20090228 19:47:38< loonycyborg> There are tons of new makes.
20090228 19:47:39< Shadow_Master> "cook". I don't know more.
20090228 19:48:08< loonycyborg> Shadow_Master: Perhaps you'll like this: http://makepp.sourceforge.net/
20090228 19:48:41< ilor> loonycyborg:wouldn't manually modifying the sconstruct file trigger a full rebuild?
20090228 19:49:04< Shadow_Master> it does AFAICT.
20090228 19:49:18< loonycyborg> ilor: That depends on what you modify.
20090228 19:50:34< loonycyborg> E.g. if you change cflags for particular files, then they(and only they) will be recompiled.
20090228 19:51:24< ilor> ah, ok then.
20090228 19:52:12< Ivanovic> re
20090228 19:53:38< Shadow_Master> ahhh, we need a framed [img] bbcode ;-)
20090228 19:53:51< CIA-70> mordante * r33172 /trunk/images/misc/ (9 files): Add transparent listbox selection images.
20090228 19:57:50< CIA-70> mordante * r33173 /trunk/data/gui/default/widget/toggle_panel_wml_message.cfg: Add a transparent toggle panel.
20090228 19:58:49< CIA-70> mordante * r33174 /trunk/data/gui/default/ (widget/listbox_wml_message.cfg window/wml_message.cfg): Add a transparent listbox to the wml_message.
20090228 20:01:04< CIA-70> mordante * r33175 /trunk/data/gui/default/window/wml_message.cfg: Add the transparent button to the right side.
20090228 20:05:29-!- fabi is now known as fendrin
20090228 20:06:18< mordante> freim around?
20090228 20:10:43< AI0867> esr: removing the VALIDATE results in a segfault in certain cases and I'm not very familiar with the code
20090228 20:11:42< mordante> Dragonking, regarding r33169 shouldn't the new message be translatable?
20090228 20:11:58< Dragonking> mordante: um
20090228 20:12:07< Dragonking> mordante: You're right
20090228 20:12:48-!- noy [n=Noy@wesnoth/developer/noy] has quit []
20090228 20:13:10< Dragonking> Ivanovic: What is policy now regarding new translatable messages?
20090228 20:13:23< Ivanovic> Dragonking: simple: don't add them!!!
20090228 20:13:28< Shadow_Master> a grue will eat you
20090228 20:13:28< Dragonking> Ivanovic: ok
20090228 20:13:38< Dragonking> mordante: I'll just remove msg then
20090228 20:13:38< Ivanovic> what do you want to do?
20090228 20:14:04 * happygrue makes the noise grues make
20090228 20:14:27< Dragonking> Ivanovic: I wanted to write in game that you can't open up formula commandline in MP game
20090228 20:14:39< Dragonking> But I may as well just not display anything
20090228 20:14:43< Ivanovic> jupp
20090228 20:14:55< Dragonking> So remove msg. ?
20090228 20:15:24< Shadow_Master> I don't have problems with translating just one...
20090228 20:15:26< Shadow_Master> ;-)
20090228 20:17:27< CIA-70> dragonking * r33176 /trunk/src/menu_events.cpp: Remove info msg
20090228 20:17:33< Dragonking> Ivanovic: Done.
20090228 20:17:43< Dragonking> mordante: Thx for noticing that.
20090228 20:17:50< mordante> no problem
20090228 20:19:07< happygrue> loonycyborg: what is the reason for userdata being stored elsewhere than in the Wesnoth directory in recent builds?
20090228 20:19:11< Shadow_Master> zookeeper: have you got around to ask Cernunnos about how he wants to appear in the credits?
20090228 20:19:22< Shadow_Master> happygrue: Vista
20090228 20:19:46< zookeeper> Shadow_Master, oww, good reminder. i'll send him a PM now.
20090228 20:19:46< happygrue> it seems that it is all in "wesnoth 1.5", which would be a problem if I had multiple versions installed... which I do...
20090228 20:19:47< happygrue> I see
20090228 20:19:51< Shadow_Master> although in general you'd also have problems if you installed wesnoth in windows xp under an admin account and attempted to use it as a limited account
20090228 20:20:12< happygrue> I see
20090228 20:20:13< Shadow_Master> the difference is that Vista tries harder to convince users to not use admin accounts for the Internetz
20090228 20:20:21< freim> mordante: yeah
20090228 20:20:33< happygrue> yes, I noticed that when my wife just got a new (vista) laptop
20090228 20:21:12< happygrue> is it as it seems that it is all in /my docs/my games/wesnoth 1.5?  Because that seems like a bad idea to me
20090228 20:21:13< Soliton> happygrue: you can use --config-dir to specify a different dir under "My Documents\My Games\".
20090228 20:21:45< happygrue> Soliton: which I do, but it seems that the default should use the full version name for the directory?
20090228 20:21:51< mordante> freim, for the new gui I added some new widgets in the transparent dialogs
20090228 20:22:04< Soliton> happygrue: versions are compatible now.
20090228 20:22:10< mordante> could you have a look at them and make them look beautiful ?
20090228 20:22:11< happygrue> I see
20090228 20:22:13< Shadow_Master> in Unix, the default is to use a single dir for all versions too.
20090228 20:22:16< happygrue> and will be for 1.6 I guess
20090228 20:22:33< mordante> and I'm not really sure what the best background for the listbox would be
20090228 20:22:43< mordante> (when not selected)
20090228 20:22:44< Soliton> yes. we should use the full version for 1.7 though, i agree.
20090228 20:23:30< Shadow_Master> fendrin: were you complaining about the new tree graphics?
20090228 20:24:08< loonycyborg> happygrue: There's a custom page in the installer to disable placing userdata in My Documents.
20090228 20:24:17< loonycyborg> Since last release.
20090228 20:24:45< happygrue> yes, I noticed that -which is good.
20090228 20:25:11< loonycyborg> You're not the first person who was annoyed by that :P
20090228 20:25:24< happygrue> heh
20090228 20:25:50-!- EdB [n=EdB@239.12.95-79.rev.gaoland.net] has quit ["Konversation terminated!"]
20090228 20:25:54< Shadow_Master> zookeeper: I'll let you write the relevant changelog entries for the "Terrain" section regarding the new tree graphics for tomorrow btw
20090228 20:26:09< Shadow_Master> if there's anything WML auhtors should now about the new soil, please mention it in there too
20090228 20:26:59< freim> mordante: I can look at them tomorrow
20090228 20:27:12< mordante> great :-)
20090228 20:28:10< Shadow_Master> mordante: can I reassign https://gna.org/bugs/?13053 to me, or are you actually working on it?
20090228 20:28:37< Shadow_Master> the problem was introduced when it started mentioning the name of the add-on there btw
20090228 20:28:58< mordante> yes I assumed that was the cause, only not sure who added it
20090228 20:29:21< mordante> feel free to fix it, I've some other things to work on
20090228 20:30:18-!- Shadow_Master [n=imorelle@wesnoth/developer/shadowmaster] has quit ["bbs"]
20090228 20:32:18-!- cjhopman [n=chris@wesnoth/developer/cjhopman] has quit [Read error: 60 (Operation timed out)]
20090228 20:36:28-!- ikarius [n=ross@216.27.182.3] has joined #wesnoth-dev
20090228 20:39:04-!- shadowmaster [n=imorelle@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090228 20:40:01-!- shadowmaster is now known as Shadow_Master
20090228 20:40:55-!- shikadibot [n=uid1001@wesnoth/bot/shikadibot] has quit [Read error: 104 (Connection reset by peer)]
20090228 20:41:12-!- shikadibot [n=uid1001@wesnoth/bot/shikadibot] has joined #wesnoth-dev
20090228 20:42:14-!- shikadibot [n=uid1001@wesnoth/bot/shikadibot] has quit [Client Quit]
20090228 20:42:20-!- Shadow_Master [n=imorelle@wesnoth/developer/shadowmaster] has quit [Client Quit]
20090228 20:43:07< CIA-70> mordante * r33177 /trunk/ (7 files in 2 dirs): 
20090228 20:43:07< CIA-70> Listbox rows no longer have a fixed height.
20090228 20:43:07< CIA-70> Remove the old part of the code which allowed that.
20090228 20:44:10-!- shadowmaster [n=imorelle@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090228 20:44:39-!- shikadibot [n=uid1001@wesnoth/bot/shikadibot] has joined #wesnoth-dev
20090228 20:44:50< fendrin> shadowmaster: yes
20090228 20:45:20< shadowmaster> PM Jetryl about it
20090228 20:45:26-!- shadowmaster is now known as Shadow_Master
20090228 20:45:49< fendrin> Shadow_Master: But I don't want to complain. They look great.
20090228 20:46:12< Shadow_Master> fendrin: why did you say "yes" then? ;-)
20090228 20:46:33< fendrin> Shadow_Master: Because I was refering to them but not complaining.
20090228 20:46:43< Shadow_Master> the old snowy tree tiles bother me now though.
20090228 20:47:28< Shadow_Master> *pine.
20090228 20:47:50< fendrin> Yes, they don't fit. There is more terrain that does not fit the new style of the forest.
20090228 20:49:03< Ivanovic> mordante: have you already started on the rewrite stuff for po things in cmake?
20090228 20:49:11< Ivanovic> or is this a post 1.6 thing for you?
20090228 20:51:22< mordante> I started on it
20090228 20:52:07< loonycyborg> Why do they require a rewrite, anyway?
20090228 20:52:17< Ivanovic> loonycyborg: because it is ugly as hell
20090228 20:52:38< Ivanovic> generates more make targets than needed
20090228 20:53:40< mordante> well the more targets depends on whether you want to keep them ;-)
20090228 20:54:02< mordante> and I removed the shell scripts and move the code to cmake
20090228 20:54:07< zookeeper> fendrin, any specific examples?
20090228 20:55:41< Ivanovic> stikonas: one general comment regarding your make clean problem: don't run it!
20090228 20:56:01< Ivanovic> stikonas: that is: use out of tree builds, there you just remove the whole folder and that's it, no need at all for make clean...
20090228 20:56:03< Ivanovic> ;)
20090228 20:56:42-!- Turuk_ [n=Turuk@FUSE-WS1-208-102-202-188.fuse.net] has joined #wesnoth-dev
20090228 20:56:55< fendrin> zookeeper: no, it's just like if they are from another game. I saw some games which copied wesnoth graphics and merded it with thier own. I had the same expression with the new trees. They use another colour sheme if you want to know my layman thoughts.
20090228 20:56:56-!- Lord_Aether is now known as Lord_Aether0
20090228 20:56:57< loonycyborg> Ah."Don't do it then" approach to bug-fixing. Can fix any bug :P
20090228 20:57:12< fendrin> s/expression/impression
20090228 20:57:51< Ivanovic> loonycyborg: it is the easiest and cleanest thing to do
20090228 20:58:02< Ivanovic> running it right in the checkout dir is a thing you just "should not do"
20090228 20:58:16< Ivanovic> like you should not shoot into your own foot just because you got a gun
20090228 20:58:25< Ivanovic> there are things you "can do" but you should not...
20090228 20:58:27< Ivanovic> ;)
20090228 20:58:41< loonycyborg> Yes. At least because it'll clobber some versioned makefiles.
20090228 20:58:42-!- Lord_Aether0 is now known as Lord_Aether
20090228 20:59:58-!- Shadow_Master is now known as Shadow_Master__
20090228 21:00:12-!- Shadow_Master__ is now known as Shadow_Master
20090228 21:00:15-!- shikadibot [n=uid1001@wesnoth/bot/shikadibot] has left #wesnoth-dev []
20090228 21:00:23-!- Baufo [n=thomas@wesnoth/developer/baufo] has quit [Read error: 110 (Connection timed out)]
20090228 21:00:35-!- shikadibot [n=uid1001@wesnoth/bot/shikadibot] has joined #wesnoth-dev
20090228 21:00:50< zookeeper> fendrin, well, i somewhat agree with the colour theme issue when it comes to the pines and the summer deciduous trees.
20090228 21:01:01-!- Baufo [n=thomas@wesnoth/developer/baufo] has joined #wesnoth-dev
20090228 21:01:16< fendrin> zookeeper: please load the map from scenario 02 of low to see a problem with the transitions between hills and forest tiles.
20090228 21:02:09< zookeeper> looking...where, exactly?
20090228 21:02:49< fendrin> isn't the problem obvious? Or does it only occour in the editor. Let's see.
20090228 21:02:55 * zookeeper notes that he forgot some transition definitions
20090228 21:02:58< stikonas> Ivanovic: I do out of tree build and therefore only today I noticed that out of tree 'make clean' command removes those files in the source tree.
20090228 21:03:04< zookeeper> i'm using the editor, i don't see anything obviously wrong here
20090228 21:03:26< zookeeper> (except the transition glitch which i'm going to fix now)
20090228 21:03:44< fendrin> :-)
20090228 21:04:07-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Connection timed out]
20090228 21:04:19< zookeeper> (but it has nothing to do with forest<->hills)
20090228 21:04:21< Ivanovic> stikonas: oh, this is really bad...
20090228 21:05:02< stikonas> Ivanovic: yeah, I noticed that git svn rebase refuced to work after make clean
20090228 21:05:16< stikonas> s/refuced/refused/
20090228 21:05:23< Shadow_Master> git reset --hard ? (drop all your changes)
20090228 21:06:10< stikonas> yes, it worked. I know git enough to do this :)
20090228 21:06:19< fendrin> zookeeper: sorry, it's forest<->mountains
20090228 21:07:16-!- Shadow_Master [n=imorelle@wesnoth/developer/shadowmaster] has quit ["(null)"]
20090228 21:07:22< zookeeper> fendrin, ah, you mean how some trees are placed on the mountain slopes?
20090228 21:08:23< fendrin> yes
20090228 21:09:03-!- shadowmaster [n=imorelle@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090228 21:09:22< fendrin> If it's intended then some of them need work to make it look right.
20090228 21:09:40< shadowmaster> I don't think it's intended.
20090228 21:09:42-!- shadowmaster [n=imorelle@wesnoth/developer/shadowmaster] has quit [Client Quit]
20090228 21:09:49-!- Shadow_Master [n=imorelle@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090228 21:10:20< zookeeper> how the layering technically works is intended, but not all those particular cases are
20090228 21:10:44< zookeeper> 16,21 looks silly, whereas 21,24 looks just right
20090228 21:11:22-!- Espreon [n=espreon@wesnoth/developer/espreon] has quit [Remote closed the connection]
20090228 21:11:52< fendrin> zookeeper: yep
20090228 21:12:19< zookeeper> i can also see some other glitches with the mountains that i didn't spot when making those changes
20090228 21:12:38< zookeeper> like 28,28 and 7,33
20090228 21:13:23-!- stikonas [n=stikonas@wesnoth/translator/stikonas] has quit [Read error: 104 (Connection reset by peer)]
20090228 21:13:31< zookeeper> oww, and 20,2
20090228 21:13:46< zookeeper> i'll try to sort those out
20090228 21:13:54< fendrin> zookeeper: cool
20090228 21:15:48-!- stikonas [n=stikonas@ctv-213-164-120-106.vinita.lt] has joined #wesnoth-dev
20090228 21:21:19-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090228 21:26:13-!- Lord_Aether [n=castle@207.212.136.10] has quit []
20090228 21:27:01-!- Espreon [n=espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20090228 21:29:18-!- Turuk_ [n=Turuk@FUSE-WS1-208-102-202-188.fuse.net] has quit [Read error: 110 (Connection timed out)]
20090228 21:33:27< Shadow_Master> fendrin: did you file the persistence test case bug?
20090228 21:33:48< Shadow_Master> it would be great that someone could take care of it for 1.6...
20090228 21:34:12< CIA-70> soliton * r33178 /trunk/ (changelog src/campaign_server/campaign_server.cpp): * Disallow uploading of add-ons with no title, type, author, version or description specified and warn about invalid versions.
20090228 21:43:52-!- silene1 [n=plouf@AMarseille-251-1-48-169.w83-113.abo.wanadoo.fr] has joined #wesnoth-dev
20090228 21:44:57< fendrin> Shadow_Master: no, but someone else did it.
20090228 21:45:28< fendrin> Shadow_Master: https://gna.org/bugs/index.php?13016
20090228 21:47:46< Shadow_Master> why is it marked as a FR?
20090228 21:48:13< fendrin> Shadow_Master: no idea, I had no contact to the persion who filled the FR.
20090228 21:48:33< Shadow_Master> does anyone know who the "expert" in this area would be to ask them?
20090228 21:50:05< Shadow_Master> wesbot: topic
20090228 21:50:27< fendrin> Shadow_Master: My testcase campaign still causes an Assertion `false`failed and aborts.
20090228 21:50:56< Shadow_Master> fendrin: fix it.
20090228 21:50:57< fendrin> Shadow_Master: didn't you fix that?
20090228 21:51:04< Shadow_Master> I asked you to do it in WML for now.
20090228 21:51:10< fendrin> okay
20090228 21:52:09< Shadow_Master> wait a minute 
20090228 21:55:12< CIA-70> zookeeper * r33179 /trunk/data/core/terrain-graphics.cfg: Forgot to add the transitions for the forest floor.
20090228 21:56:02-!- voris [n=voris@c-24-16-107-149.hsd1.wa.comcast.net] has quit [Remote closed the connection]
20090228 21:57:03< mordante> Ivanovic, updated http://www.wesnoth.org/wiki/Wesnoth1.6Features#Mordante
20090228 21:58:44< Ivanovic> so atm only the shortcutkey is missing on your side?
20090228 21:59:03< mordante> no also the tooltips and I'm working on them
20090228 21:59:11< Ivanovic> ah, okay
20090228 22:00:59< mordante> still hope to get the campaign dialog fixed as well
20090228 22:01:54< CIA-70> ivanovic * r33180 /trunk/po/ (CMakeLists.txt wesnoth-manual/CMakeLists.txt): 
20090228 22:01:54< CIA-70> some *tiny* improvements to cmake script:
20090228 22:01:54< CIA-70> * reorder stuff for manual to only add targets when those can be built (due to po4a stuff)
20090228 22:01:54< CIA-70> * break textdomain list into "one per line" and add the wesnoth-test domain
20090228 22:05:30-!- silene [n=plouf@wesnoth/developer/silene] has quit [Read error: 110 (Connection timed out)]
20090228 22:07:02-!- fb0teg [n=cra@home-pool-176-130.com2com.ru] has joined #wesnoth-dev
20090228 22:07:04< Shadow_Master> foo
20090228 22:07:13< fendrin> bar
20090228 22:07:30< shikadibot> baz
20090228 22:07:44< Shadow_Master> no seriously, GUI1 doesn't have a way to set a dialog's maximum size? :-/
20090228 22:08:32-!- VladimirSlavik [n=chatzill@wesnoth/translator/VladimirSlavik] has joined #wesnoth-dev
20090228 22:08:37< VladimirSlavik> hi
20090228 22:08:41< mordante> hi VladimirSlavik 
20090228 22:08:44< Shadow_Master> and why is it recompiling everything again? aaoooow.
20090228 22:08:46< mordante> Shadow_Master, I guess no
20090228 22:09:00 * Shadow_Master goes around the house screaming in despair
20090228 22:10:12< zookeeper> uh oh, some weird engine bug AFAICT: http://www.wesnoth.org/forum/viewtopic.php?p=342683#p342683
20090228 22:10:17< zookeeper> i'm trying to pinpoint it
20090228 22:10:56 * Shadow_Master smashes head against the wall
20090228 22:11:31< zookeeper> ahah
20090228 22:11:44< zookeeper> it was my fault after all
20090228 22:11:56< VladimirSlavik> Ivanovic you might want to know that last pot-update included Kaleh specs that have broken indent
20090228 22:12:32< Ivanovic> ARGH
20090228 22:12:34< shikadibot> VladimirSlavik: esr 's rewruite probably
20090228 22:12:45< Ivanovic> esr: what have you misfixed again?!?
20090228 22:12:47< VladimirSlavik> yup, that is it
20090228 22:13:13< Shadow_Master> Ivanovic: he rewrote Kaleh.cfg to fix some wmliterator/wmllint complaints before I could get around to do it myself
20090228 22:13:22< Shadow_Master> s/shikadibot/Shadow_Master/
20090228 22:13:51< Shadow_Master> wesbot: log 33134
20090228 22:13:52< wesbot> esr * r33134 : Another attempt to refactor the Kaleh AMLAs to use only balanced markup.
20090228 22:13:55< wesbot> URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=33134
20090228 22:14:37< shikadibot> wesbot: log 33133
20090228 22:14:38< wesbot> esr * r33133 : Add indenting for readability.  No markup or content changes.
20090228 22:14:41< wesbot> URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=33133
20090228 22:14:56< Ivanovic> +                description= _ "Sustenance:
20090228 22:14:57< Ivanovic> +        This unit's resistances are increased by 20% in daylight, on defense and attack. Vulnerabilities are not affected."
20090228 22:14:59< Shadow_Master> 33133 is the guilty commit actually. He indented text in multiline translatable strings
20090228 22:14:59< Ivanovic> there it is
20090228 22:15:36< zookeeper> "oops" ;)
20090228 22:15:55< Shadow_Master> it seems to be manual indentation
20090228 22:16:15-!- Corvvs [n=michaelm@c-69-139-145-2.hsd1.md.comcast.net] has joined #wesnoth-dev
20090228 22:16:43-!- Espreon [n=espreon@wesnoth/developer/espreon] has quit [Read error: 60 (Operation timed out)]
20090228 22:16:45< Corvvs> hello Espreon, Turuk.
20090228 22:17:06< Shadow_Master> Ivanovic: should I wait for it to be fixed before unfuzzying strings in the .po?
20090228 22:17:13< VladimirSlavik> well the indent is small enough to be still considered fuzzy
20090228 22:17:20< Ivanovic> Shadow_Master: yes
20090228 22:17:24-!- Espreon [n=espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20090228 22:17:31< Shadow_Master> k
20090228 22:18:45< fb0teg> can someone consult me in SDL plz? Just a few stupid questions
20090228 22:19:19< CIA-70> zookeeper * r33181 /trunk/data/campaigns/Son_Of_The_Black_Eye/scenarios/04_The_Siege_Of_Barag_Gor.cfg: Fixed the scout that's supposed to get recalled for the cutscene getting lost entirely instead.
20090228 22:19:20< Shadow_Master> #sdl is the official SDL help channel, just so you know
20090228 22:19:29< fb0teg> oops sorry :)
20090228 22:19:36< Shadow_Master> but if the question is Wesnoth+SDL-specific, then sure this is the correct placee
20090228 22:19:38< mordante> fb0teg, we can only tell if we know the question ;-)
20090228 22:20:20< fb0teg> I'm just wondering how the map tiles are blitted in Wesnoth. Are those hexagonal tiles SDL-kind?
20090228 22:20:41< mordante> we blit rects, sdl can only blit rects
20090228 22:22:47< Ivanovic> VladimirSlavik: three strings affected or are there more?
20090228 22:23:05< VladimirSlavik> iirc Czech had 100% so only hte two
20090228 22:23:17< fb0teg> mordante: And how you achieve hexagonality then? It's just an eye illusion?
20090228 22:23:44< VladimirSlavik> Ivanovic: my last sync brought utbs just with this and the traits in main catalog
20090228 22:24:01< mordante> fb0teg, well it's no eye illusion, wish we could do those tricks ;-)
20090228 22:24:24< mordante> the rect is partly transparent so the part outside the hex has no effect
20090228 22:25:16< mordante> have a look in data/core/images/terrain/ to see how the hexes look
20090228 22:25:59< fb0teg> Oh, I see..
20090228 22:26:23< CIA-70> ivanovic * r33182 /trunk/data/campaigns/Under_the_Burning_Suns/units/Kaleh.cfg: 
20090228 22:26:23< CIA-70> fixed f***ed up indention
20090228 22:26:23< CIA-70> esr: next time please take care of translateable string, whitespaces in those *DO* matter!!! (that is newlines, tabs and spaces)
20090228 22:26:27< Ivanovic> VladimirSlavik: okay, then it is fixed
20090228 22:26:44< Ivanovic> updating the po files now, those fuzzy strings should be gone in a second
20090228 22:26:49< VladimirSlavik> thanks
20090228 22:27:36< fb0teg> Thanks a lot, sorry for bothering
20090228 22:27:46< VladimirSlavik> welllllllll... technically they should stay fuzzy since the value changed again, and since it's kinda stateless like http...
20090228 22:27:50-!- fb0teg [n=cra@home-pool-176-130.com2com.ru] has left #wesnoth-dev []
20090228 22:28:07< VladimirSlavik> or not?
20090228 22:28:29< Shadow_Master> if nobody removes the commented strings in the po files used for fuzzy resolution
20090228 22:28:58< Ivanovic> VladimirSlavik: no, they should not
20090228 22:29:06< Ivanovic> since the old strings are in the commented out area
20090228 22:29:32< VladimirSlavik> no, the ones matched as fuzzy are not kept as old
20090228 22:30:02< Ivanovic> argh
20090228 22:30:11< VladimirSlavik> i see it here
20090228 22:30:11< Ivanovic> okay, then they are still fuzzy unless i do some magic
20090228 22:30:15< Ivanovic> which i will do
20090228 22:30:54 * VladimirSlavik imagines Ivanovic going grrrrrr and opening 50 files in poedit
20090228 22:31:44< stikonas> svn revert is probably faster
20090228 22:31:51< CIA-70> shadowmaster * r33183 /trunk/po/wesnoth/es.po: Spanish translation update.
20090228 22:31:54< Shadow_Master> svn merge -c -NNNN ?
20090228 22:32:10< Shadow_Master> svn revert only reverts changes at your working copy to go back to BASE
20090228 22:32:30< VladimirSlavik> that was my first thought, too - roll back and pot-update
20090228 22:34:33< VladimirSlavik> Anyway, now that I have caused enough chaos and stayed to enjoy it, it's time to disappear again.
20090228 22:34:35< VladimirSlavik> bye!
20090228 22:34:38-!- VladimirSlavik [n=chatzill@wesnoth/translator/VladimirSlavik] has left #wesnoth-dev []
20090228 22:36:59< CIA-70> ivanovic * r33184 /trunk/po/ (240 files in 5 dirs): pot-update (reference update and fixing the 2 strings esr broke in utbs)
20090228 22:38:57-!- shikadibot [n=uid1001@wesnoth/bot/shikadibot] has quit ["assert(false);"]
20090228 22:44:33-!- Noyga [n=lame-z@wesnoth/developer/noyga] has quit [Remote closed the connection]
20090228 22:46:51< mordante> I'm off night
20090228 22:47:06-!- mordante [n=mordante@wesnoth/developer/mordante] has quit ["Leaving"]
20090228 22:48:00< ilor> Ivanovic: how does gettext deal with moving a translatable string between sourcefiles (within the same textdomain)?
20090228 22:49:50< Ivanovic> it does not really care about this
20090228 22:50:08< Shadow_Master> uuuuugh.
20090228 22:50:19< Shadow_Master> the forum pages do not validate in the W3C validator.
20090228 22:50:46< Ivanovic> ilor: that is: gettext is not "place oriented", it just cares about strings and compares strings
20090228 22:51:01< Ivanovic> and for "matching strings" the new "place" is shown
20090228 22:51:10< ilor> Ivanovic: ok
20090228 22:51:38< ilor> Shadow_Master: I value "working" above "validates" tbh
20090228 22:52:27< Shadow_Master> the header says XHTML 1.0 Strict but the code isn't fitting.
20090228 22:52:46< Shadow_Master> <img blabity blah> </a> instead of <img blabity blah /> </a>
20090228 22:54:03< Ivanovic> i am off to bed now, n8
20090228 22:55:07< Shadow_Master> a simple fix killed 6 validation errors at once. :-)
20090228 22:56:13-!- silene1 [n=plouf@AMarseille-251-1-48-169.w83-113.abo.wanadoo.fr] has quit ["Leaving."]
20090228 23:01:18-!- Noyga [n=lame-z@wesnoth/developer/noyga] has joined #wesnoth-dev
20090228 23:06:11< Elvish_Pillage2> dragonking when are you going to fix the move function, battles don't work very well when the enemy doesn't move :P
20090228 23:06:21-!- Derekkk [n=chatzill@host86-165-245-32.range86-165.btcentralplus.com] has joined #wesnoth-dev
20090228 23:06:44< Dragonking> Elvish_Pillage2: Will try tomorrow morning
20090228 23:07:00< Elvish_Pillage2> cool
20090228 23:07:54< Shadow_Master> the other validation problems seem to be caused by misuse of bbcode only
20090228 23:08:13 * Elvish_Pillage2 meanwhile converts the other python AIs to formula
20090228 23:08:29< Dragonking> Elvish_Pillage2: What do you think about formula?
20090228 23:08:48< Elvish_Pillage2> I like it. It could use a few more features, I suppose
20090228 23:09:00< Dragonking> Write down ideas. ;)
20090228 23:09:13< Elvish_Pillage2> but the fact that it took me about half as much code to do the same thing as I did in the more-features Python interface, clearly shows that it works
20090228 23:09:33< Elvish_Pillage2> oh, ideas? That I can supply you with! :D
20090228 23:09:42< Dragonking> Go ahead.
20090228 23:09:54< Dragonking> I'll try to fix something in a meantime.
20090228 23:09:54< Dragonking> :)
20090228 23:10:18< CIA-70> ilor * r33185 /trunk/ (3 files in 3 dirs): work around major issues of bug #12777 and editor's hotkey descriptions/tooltips in a way that does not introduce new strings or mess with the code much
20090228 23:11:23< Shadow_Master> ilor: true, but it's better a page that claims to be XHTML 1.0 Strict doesn't use invalid code
20090228 23:11:38< Shadow_Master> it would break in hypothetically strict browsers
20090228 23:12:30< Derekkk> Dragonking: In formula AI, it would be great if recursion can be done
20090228 23:12:36< Shadow_Master> in particular those that really do interpret it as XML rather than SGML/HTML
20090228 23:12:41< CIA-70> zookeeper * r33186 /trunk/data/core/terrain-graphics/mountains.cfg: Fixed layering problems with some mountain tiles (and trimmed some whitespace).
20090228 23:12:48< zookeeper> that looks like it took care of the mountain glitches, i'll see about the forest<->mountains stuff next..
20090228 23:12:55< Dragonking> Derekkk: For what?
20090228 23:14:08< Shadow_Master> http://www.wesnoth.org/forum/viewtopic.php?p=342698#p342698
20090228 23:14:10< Shadow_Master> pfffffff
20090228 23:14:13< Elvish_Pillage2> Dragonking: first thing I think of is as I mentioned before, adding all weapons to 'attacks' and giving chance to die in 'attacks'
20090228 23:14:25< Elvish_Pillage2> then - first, does it pay any attention to fog right now?
20090228 23:14:33< Dragonking> No.
20090228 23:14:37< Elvish_Pillage2> Oh yeah, recalling would be good too :P
20090228 23:14:52< Elvish_Pillage2> well, it would be good for it to pay attention to fog
20090228 23:15:04< Dragonking> Define 'pay attention'
20090228 23:15:16< Elvish_Pillage2> Not tell the AI where any units are that are in the fog.
20090228 23:15:26< Derekkk> Dragonking: for efficient looping. I think it lacks facilities to loop at the moment
20090228 23:15:47< Elvish_Pillage2> there was another thing I wanted to know - darn, what was it?
20090228 23:16:27< Dragonking> Derekkk: OK.  Jus curious, any idea where to use loop?
20090228 23:16:49< CIA-70> alink * r33187 /trunk/src/mouse_events.cpp: 
20090228 23:16:49< CIA-70> Fix bug #11551 : "if attacker dies after a move+attack we never clear shroud"
20090228 23:16:49< CIA-70> Now clear shroud before starting the attack, partially fixing bug #12077.
20090228 23:16:49< CIA-70> Paradoxically delaying shroud update already dit that, but had doubled or
20090228 23:16:50< CIA-70> incorrect sighted events, plus some possible other bugs. All seems fixed now.
20090228 23:17:18< alink> ^aah glad that I managed to slip this bugfix in the RC (relatively safely)
20090228 23:17:39< Elvish_Pillage2> Dragonking: The shortest_path function to be able to give the shortest path for a unit to move between any two locations, not just from where it's standing.
20090228 23:17:41-!- stikonas [n=stikonas@wesnoth/translator/stikonas] has quit [Read error: 110 (Connection timed out)]
20090228 23:17:51< Elvish_Pillage2> Dragonking: Another function that would be like shortest_path and ignore units.
20090228 23:17:59< alink> that is without creating new bug (if there is one, it was already there when delaying shroud update)
20090228 23:18:20< Dragonking> Elvish_Pillage2: so shortest_path(loc, loc, unit)
20090228 23:18:24< zookeeper> alink, nice
20090228 23:18:36< Shadow_Master> good
20090228 23:18:47< Elvish_Pillage2> Dragonking: probably (unit, loc, loc) for consistency with other functions
20090228 23:19:05< Derekkk> Dragonking: say, I was trying to move a unit through the path returned by the shortest_path function. A recur would be helpful.
20090228 23:19:09< Dragonking> Elvish_Pillage2: Nah, unit would be optional
20090228 23:19:11< alink> zookeeper: feel free to test it with some tricky WML, but normal cases was tested
20090228 23:19:13< Dragonking> So it goes last
20090228 23:19:22< Elvish_Pillage2> Dragonking: Unit can't be optional.
20090228 23:19:30< Dragonking> Elvish_Pillage2: Why?
20090228 23:19:42< Elvish_Pillage2> Because it's the unit that determines the terrain movement costs.
20090228 23:19:48< zookeeper> alink, i don't think i want to start to write test cases for that :P
20090228 23:19:49< Dragonking> And how it works now?
20090228 23:20:03< alink> Dragonking: in code we place temporary unit for that at the wanted location
20090228 23:20:10< Elvish_Pillage2> Gives the shortest path for the unit at one location to move to another
20090228 23:20:18< Dragonking> alink: I know, I just ask Elvish_Pillage2 
20090228 23:20:23< Dragonking> Exactly.
20090228 23:20:28< Dragonking> So it is not needed.
20090228 23:20:29< Dragonking> :)
20090228 23:20:43< Elvish_Pillage2> huh?
20090228 23:20:49< Dragonking> If we get 3rd parameter, we use it, if not, we look for unit at first loc.
20090228 23:20:56< Elvish_Pillage2> Oh, I see :)
20090228 23:20:58< Dragonking> Simple as that.
20090228 23:21:06< Shadow_Master> I haaaate GUI1... 
20090228 23:21:09< Dragonking> Just need to implement
20090228 23:22:14< Shadow_Master> Sapient: do you have any ideas about https://gna.org/bugs/?13053 ?
20090228 23:23:15< Shadow_Master> Sapient: need a smarter test case? in src/dialogs.cpp, go to line 1193 and replace the msg in the second parameter to gui::dialog_frame::dialog_frame() with a very long static string
20090228 23:23:39< Shadow_Master> recompile and try to connect to the add-ons server or MP server. 
20090228 23:26:52< Elvish_Pillage2> Dragonking: here's one it'd be harder to code. A function that takes a move description - src and dst - and returns an ai struct that is as the gamestate would be after that move.
20090228 23:27:24< Dragonking> Elvish_Pillage2: What for?
20090228 23:27:39< Elvish_Pillage2> hmm, you'd want that to take a _list_ of moves, rather.
20090228 23:28:05< Elvish_Pillage2> For dealing with ZoC defenses when moving multiple units
20090228 23:28:29< Dragonking> I don't follow.
20090228 23:28:43< Elvish_Pillage2> since it's quite impractical to calculate which units could move where within the AI
20090228 23:28:58-!- Zen_Clark [n=user@99-136-80-191.lightspeed.rcsntx.sbcglobal.net] has joined #wesnoth-dev
20090228 23:29:46< Derekkk> Elvish_Pillage2: if it is to be coded anyway, why not take it further? instead of looking 1 step ahead, maybe look  even further?
20090228 23:29:51< Dragonking> Elvish_Pillage2: I still don't follow. Gime me an example maybe?
20090228 23:30:30< Elvish_Pillage2> Dragonking: Sure. My AI just killed an enemy unit with several Mages. I also control a bunch of Cavalrymen nearby which are much more expendable. I now want to move the Cavalrymen so that the enemy can't reach the Mages.
20090228 23:31:20< Elvish_Pillage2> Derekkk: It's not deterministic if an attack is involved.
20090228 23:34:29< Derekkk> Elvish_Pillage2: Yes, I am not sure how feasible it is to store multiple game states after a non-deterministic step
20090228 23:35:21< Dragonking> Elvish_Pillage2: Um, then you do it like: calculate move, do it, then calculate it again after the move and decide what to do
20090228 23:35:23< Elvish_Pillage2> Derekkk: one could have the AI specify how the attack turns out, I suppose.
20090228 23:35:40< Elvish_Pillage2> Dragonking: uh, no, because AIs can't undo
20090228 23:35:54< Elvish_Pillage2> Dragonking: Actually, that would be pretty cool. Let the AI undo
20090228 23:35:59< Dragonking> No. :)
20090228 23:36:11< Elvish_Pillage2> heh.
20090228 23:36:11< Dragonking> You do move. then have different state. then move 
20090228 23:36:15-!- Baufo [n=thomas@wesnoth/developer/baufo] has quit [Remote closed the connection]
20090228 23:36:27< Dragonking> What is uese of getting state of move beforehand?
20090228 23:36:31< Dragonking> Just to make move after
20090228 23:36:38< Elvish_Pillage2> but the point is, I don't _know_ vital, but calculable, information about the gamestate after!
20090228 23:36:41< Dragonking> Just do the move, have state, do next one
20090228 23:36:57< Elvish_Pillage2> the situation is that I want to find the optimal places to put my Cavalrymen
20090228 23:37:16< Elvish_Pillage2> I have to figure that out before moving any of them
20090228 23:37:30< Dragonking> You have enemy position etc, so what's a problem?
20090228 23:37:42< Elvish_Pillage2> and where they should end up is closely related to where the enemies will be able to move and attack afterwards,
20090228 23:38:01< Elvish_Pillage2> and the formula AI doesn't provide a good way for _me_ to calculate the effects of terrain and ZoC
20090228 23:38:17< Elvish_Pillage2> it doesn't even give me the terrain movement costs of the enemy units!
20090228 23:38:51< Dragonking> movement_cost function?
20090228 23:38:56< Elvish_Pillage2> nor does it tell me whether my units have ZoC (although that _usually_ corresponds with level)
20090228 23:39:18< Elvish_Pillage2> oh, didn't notice that
20090228 23:39:47< Dragonking> enemy_moves gives you current possible moves of an enemy
20090228 23:40:08< Dragonking> I'm not sure how dificult would be to implement something simillar, just after set of moves
20090228 23:40:11< Elvish_Pillage2> _current_ possible enemy moves are almost useless
20090228 23:40:38< Dragonking> Why?
20090228 23:40:46< Dragonking> YOu can filter out which units can reach some hexes
20090228 23:40:50< Dragonking> And block these
20090228 23:41:09< Elvish_Pillage2> suppose you have other units that are currently blocking some off, but will inevitably move before the end of the turn.
20090228 23:41:19< Elvish_Pillage2> It would be better to have enemy_moves work like ctrl-B.
20090228 23:41:31< Elvish_Pillage2> (Or to have both, for that matter.)
20090228 23:42:03< Dragonking> Ok. I got the point.
20090228 23:43:09< Derekkk> I think the question is: how do you know if a move can block the enemies, without actually doing the move.
20090228 23:43:42-!- malumalu_ [n=malu@hnvr-4dbb5332.pool.einsundeins.de] has quit ["Verlassend"]
20090228 23:43:45< Elvish_Pillage2> hmm. maybe this would cover everything: a function that returns enemy_moves after a certain set of moves, ignoring units of yours that haven't yet moved.
20090228 23:43:46< Dragonking> It's more deep, ther's also case of revealing your units to attack
20090228 23:44:31< Elvish_Pillage2> (i.e. ignoring them for blocking/ZoC purposes.)
20090228 23:46:06< Elvish_Pillage2> actually, hang on. instead of working out exactly what I _would_ want, the function could _specify_ which units were considered.
20090228 23:47:06< Elvish_Pillage2> reworded: Calculate enemy_moves as if you had made a given set of moves already and another given set of your units were removed from the map.
20090228 23:48:34< Elvish_Pillage2> With a function like that, I would be ready to use formula AI to take on the standard AI. >:)
20090228 23:49:41< CIA-70> ai0867 * r33188 /trunk/src/actions.cpp: Remove a VALIDATE, fix the bug that caused the segfault it shielded. (bug #12428)
20090228 23:50:18< Derekkk> Elvish_Pillage2: I actually wrote a procedure as you have specified there in C++ for python AI.... but unfortunately I couldn't test it coz my pc can't compile wesnoth
20090228 23:50:34< Derekkk> it's useless anyway now
20090228 23:50:42< Elvish_Pillage2> useless now, why?
20090228 23:50:59< Derekkk> the procedure was for python AI
20090228 23:51:11< CIA-70> ai0867 * r33189 /trunk/data/tools/wesnoth/wmltools.py: Fix what looks like some typos, caught by pychecker. ESR, please check this.
20090228 23:51:29< Elvish_Pillage2> well, if it can calculate the relevant information, shouldn't it be possible to use some of it?
20090228 23:51:36< Shadow_Master> Ivanovic: I'm going to raise the attachment size limit for generating thumbnails
20090228 23:51:57< Shadow_Master> currently it is 12000 (around 11 KB), I'll make it 512 KB
20090228 23:55:34< Derekkk> Elvish_Pillage2: I am sure you have see the C++ procedure for looking 1 step ahead in Python AI. Mine was just an extension through putting additional moves into an array.
20090228 23:56:02< Elvish_Pillage2> ahhh
20090228 23:56:06< AI0867> wesbot: topic
--- Log closed Sun Mar 01 00:00:28 2009