--- Log opened Tue Jun 15 00:00:39 2010
20100615 00:02:01-!- phlaem [~a@e178071236.adsl.alicedsl.de] has quit [Quit: Leaving]
20100615 00:21:20-!- _jbx_ [~jbailey@12.190.80.225] has quit [Quit: Dig that hole, forget the sun.]
20100615 00:35:40-!- Gallaecio [~Gallaecio@232.158.60.213.dynamic.mundo-r.com] has quit [Remote host closed the connection]
20100615 00:37:05-!- YogiHH [YogiHH@wesnoth/developer/yogihh] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539]]
20100615 00:39:18< AI0867> boucman: what has TERRAIN_BASE_PROB been replaced with?
20100615 00:43:53< AI0867> 22:43 < gabba> loonycyborg: yeah. Actually, if being GPL v2 hampers introduction of useful code, I think now would be a good time to do a code-wide search and replace :P <-- that doesn't actually need to be done. You can have both GPLv2+ and GPLv3 code in the same project, it's just that the package as a whole becomes GPLv3
20100615 00:45:12< gabba> AI0867: good to know
20100615 00:46:12< AI0867> we previously had some GPLv2-only code
20100615 00:46:24< AI0867> some mailer that wasn't used, so we got rid of it
20100615 00:59:28< CIA-86> gabba * r43443 /trunk/src/ (3 files in 2 dirs): Whiteboard: refactoring methods that create the temporary move arrow.
20100615 00:59:34< CIA-86> gabba * r43444 /trunk/src/ (arrow.cpp arrow.hpp): Arrow: Little change to path validation
20100615 00:59:39< CIA-86> gabba * r43445 /trunk/src/ (mouse_events.cpp whiteboard/manager.cpp): Whiteboard: stop displaying a temp arrow if unit is deselected
20100615 00:59:46< CIA-86> gabba * r43446 /trunk/src/ (unit.cpp unit.hpp unit_animation.cpp): Added "ghosted" animation to be used with whiteboard
20100615 00:59:52< CIA-86> gabba * r43447 /trunk/src/ (5 files): Arrows: removed notification of deletion, since arrows can outlive display and try to notify a dead pointer.
20100615 01:00:00< CIA-86> gabba * r43448 /trunk/src/whiteboard/ (5 files): Whiteboard: use fake units to show destination position.
20100615 01:21:59-!- Shakey [HydraIRC@c-71-201-89-187.hsd1.il.comcast.net] has joined #wesnoth-dev
20100615 01:33:33-!- Upthorn [ogmar@adsl-75-26-202-12.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100615 01:33:33-!- Upth [ogmar@adsl-75-26-170-167.dsl.scrm01.sbcglobal.net] has quit [Read error: Connection reset by peer]
20100615 01:41:00-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100615 01:47:25-!- Shakey [HydraIRC@c-71-201-89-187.hsd1.il.comcast.net] has quit [Quit:  HydraIRC -> http://www.hydrairc.com <- Chicks dig it]
20100615 02:11:09-!- shadowmaster [~ignacio@wesnoth/developer/shadowmaster] has quit [Disconnected by services]
20100615 02:11:20-!- shadowmaster [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100615 02:13:33-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz]
20100615 02:15:13-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20100615 02:16:15-!- Bob_The_Mighty [~chatzilla@cpc4-brig15-0-0-cust904.3-3.cable.virginmedia.com] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539]]
20100615 02:21:34-!- mjs-de [~mjs-de@vpw.wh.Uni-Dortmund.DE] has quit [Remote host closed the connection]
20100615 02:23:20< CIA-86> eleazar * r43449 /trunk/data/core/terrain.cfg: adjusted the default base for some winter/fall combo hill/forest terrains to the brown hills.
20100615 02:52:04< AI0867> boucman: I've replaced TERRAIN_BASE_PROB with TERRAIN_BASE_P, TERRAIN_BASE_PROP_OV w/ OVERLAY_P and TERRAIN_BASE_OV w/ OVERLAY
20100615 02:53:14< AI0867> now I'm running into TERRAIN_ADJACENT_OV and TERRAIN_ADJACENT_PREC (a custom macro based on several other ADJACENT macros)
20100615 02:53:39< AI0867> it seems that every single macro you've removed was used in IftU and(not or) other UMC in wesnoth-umc-dev
20100615 02:54:38-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20100615 02:58:59-!- noy [~Noy@70.70.254.199] has joined #wesnoth-dev
20100615 02:59:09-!- noy [~Noy@70.70.254.199] has quit [Changing host]
20100615 02:59:09-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100615 03:01:46< AI0867> the walls-with-torches seem to have a slightly broken internal transition
20100615 03:02:31< AI0867> build a horseshoe with opening to the south in the editor and you'll see
20100615 03:14:15-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit []
20100615 03:24:03-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has quit [Remote host closed the connection]
20100615 03:43:28-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100615 03:53:43-!- gabba [~gabba@wesnoth/developer/gabba] has left #wesnoth-dev []
20100615 04:02:31-!- Ivanovic_ [~ivanovic@dtmd-4db2e070.pool.mediaWays.net] has joined #wesnoth-dev
20100615 04:06:11-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 258 seconds]
20100615 04:06:25-!- Sangel [~sangel@adsl-99-140-187-157.dsl.chcgil.sbcglobal.net] has joined #wesnoth-dev
20100615 04:06:29-!- Ivanovic_ is now known as Ivanovic
20100615 04:13:54-!- Appleman1234 [~Appleman1@CPE-60-226-180-71.qld.bigpond.net.au] has joined #wesnoth-dev
20100615 04:59:32-!- Ivanovic_ [~ivanovic@dtmd-4db2293b.pool.mediaWays.net] has joined #wesnoth-dev
20100615 04:59:45-!- Sangel [~sangel@adsl-99-140-187-157.dsl.chcgil.sbcglobal.net] has quit [Quit: Sangel]
20100615 05:02:16-!- noy_ [~Noy@70.70.255.54] has joined #wesnoth-dev
20100615 05:02:16-!- noy_ [~Noy@70.70.255.54] has quit [Changing host]
20100615 05:02:16-!- noy_ [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100615 05:02:55-!- Ivanovic [~ivanovic@dtmd-4db2e070.pool.mediaWays.net] has quit [Ping timeout: 258 seconds]
20100615 05:02:58-!- King_Elendil [~King_Elen@75.143.233.138] has joined #wesnoth-dev
20100615 05:03:31-!- Ivanovic_ is now known as Ivanovic
20100615 05:04:29< shadowmaster> wesbot: seen Crab_
20100615 05:04:29< wesbot> shadowmaster: The person with the nick Crab_ last spoke 1d 5h ago.  6h 48m ago they left with the message: Quit: Leaving.
20100615 05:05:09-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 252 seconds]
20100615 05:05:09-!- noy_ is now known as noy
20100615 05:05:14-!- King_Elendil [~King_Elen@75.143.233.138] has quit [Read error: Connection reset by peer]
20100615 05:19:35< CIA-86> espreon * r43450 /trunk/data/core/images/portraits/undead/transparent/ghoul.png: Ran umcpropfix.
20100615 05:23:50< CIA-86> espreon * r43451 /trunk/data/core/images/portraits/undead/ (ghoul.png transparent/ghoul.png): 
20100615 05:23:50< CIA-86> Ran wesnoth-optipng:
20100615 05:23:50< CIA-86> Overall statistics (only for files with a smaller recompressed size):
20100615 05:23:50< CIA-86> Original size: 269 KiB on 2 files
20100615 05:23:50< CIA-86> Optimized size: 251 KiB
20100615 05:23:51< CIA-86> Total saving: 17 KiB = 6% decrease
20100615 05:29:18-!- Greywhind [~Greywhind@pool-96-238-43-241.prvdri.fios.verizon.net] has joined #wesnoth-dev
20100615 05:32:18-!- Greywhind [~Greywhind@pool-96-238-43-241.prvdri.fios.verizon.net] has quit [Client Quit]
20100615 05:33:04-!- Greywhind [~Greywhind@pool-96-238-43-241.prvdri.fios.verizon.net] has joined #wesnoth-dev
20100615 05:35:54-!- Greywhind [~Greywhind@pool-96-238-43-241.prvdri.fios.verizon.net] has quit [Client Quit]
20100615 05:38:16-!- Greywhind [~Greywhind@pool-96-238-43-241.prvdri.fios.verizon.net] has joined #wesnoth-dev
20100615 05:59:03-!- Elvish_Pillager [~eli@71-10-224-192.dhcp.oxfr.ma.charter.com] has quit [Quit: Hi! I'm a quit message virus vaccine. If you see a quit message virus, don't replace your quit message with it!]
20100615 06:17:05-!- PK1 [~pk@r74-192-30-57.bcstcmta01.clsttx.tl.dh.suddenlink.net] has quit [Quit: Java user signed off]
20100615 06:35:45-!- hhyloc [71a69f96@gateway/web/freenode/ip.113.166.159.150] has joined #wesnoth-dev
20100615 06:41:38-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Ping timeout: 260 seconds]
20100615 06:42:05-!- hhyloc [71a69f96@gateway/web/freenode/ip.113.166.159.150] has quit [Quit: Page closed]
20100615 06:42:38-!- crimson_penguin [~ben@64.201.60.211] has joined #wesnoth-dev
20100615 06:42:38-!- crimson_penguin [~ben@64.201.60.211] has quit [Changing host]
20100615 06:42:38-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20100615 07:16:13-!- ancestral [~ancestral@97-116-112-18.mpls.qwest.net] has joined #wesnoth-dev
20100615 07:24:52-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: brb]
20100615 07:26:25-!- CIA-86 [cia@208.69.182.149] has quit [Ping timeout: 264 seconds]
20100615 07:26:43-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100615 07:30:53-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Quit: Leaving]
20100615 07:36:46-!- ancestral [~ancestral@97-116-112-18.mpls.qwest.net] has quit [Ping timeout: 265 seconds]
20100615 07:42:05-!- ancestral [~ancestral@97-116-112-18.mpls.qwest.net] has joined #wesnoth-dev
20100615 07:43:05-!- Appleman1234 [~Appleman1@CPE-60-226-180-71.qld.bigpond.net.au] has left #wesnoth-dev ["Leaving"]
20100615 07:46:42-!- ancestral [~ancestral@97-116-112-18.mpls.qwest.net] has quit [Ping timeout: 240 seconds]
20100615 07:51:55-!- ancestral [~ancestral@97-116-112-18.mpls.qwest.net] has joined #wesnoth-dev
20100615 07:54:56-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Quit: crimson_penguin]
20100615 07:55:02-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20100615 08:00:31-!- ancestral_ [~ancestral@97-116-115-34.mpls.qwest.net] has joined #wesnoth-dev
20100615 08:02:15-!- CIA-86 [cia@208.69.182.149] has joined #wesnoth-dev
20100615 08:02:45-!- ancestral [~ancestral@97-116-112-18.mpls.qwest.net] has quit [Ping timeout: 240 seconds]
20100615 08:02:45-!- ancestral_ is now known as ancestral
20100615 08:07:06-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: You are doing it right!]
20100615 08:39:51-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100615 08:53:00-!- Blarumyrran [~Blarumyrr@84-50-143-71-dsl.rkv.estpak.ee] has joined #wesnoth-dev
20100615 09:03:02< Soliton> elias: http://yourpaste.net/5662
20100615 09:06:48-!- Ivanovic [~ivanovic@dtmd-4db2293b.pool.mediaWays.net] has quit [Changing host]
20100615 09:06:48-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20100615 09:08:47< Ivanovic> moin
20100615 09:10:22-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20100615 09:12:46-!- kevg [~kevg@94.232.5.102] has joined #wesnoth-dev
20100615 09:12:51< kevg> hello
20100615 09:28:20-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: system upgrade in progress]
20100615 09:30:29-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100615 09:38:32-!- stikonas [~and@ctv-79-132-162-160.vinita.lt] has joined #wesnoth-dev
20100615 09:38:32-!- stikonas [~and@ctv-79-132-162-160.vinita.lt] has quit [Changing host]
20100615 09:38:32-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100615 09:52:34< Rhonda> ** Changed in: libsdl1.2 (Ubuntu Lucid)                  
20100615 09:52:34< Rhonda>        Status: Fix Committed => Fix Released
20100615 09:53:19< Rhonda> libsdl1.2 | 1.2.14-4ubuntu1.1 | lucid-proposed | source
20100615 09:56:10-!- thespaceinvader [~chatzilla@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20100615 09:57:00-!- ancestral [~ancestral@97-116-115-34.mpls.qwest.net] has quit [Quit: And that’s the end of THAT chapter.]
20100615 09:58:57-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit []
20100615 10:00:03< Ivanovic> hui!!!
20100615 10:02:03-!- euschn [~eugen@wesnoth/developer/euschn] has joined #wesnoth-dev
20100615 10:05:27< CIA-86> billynux * r43452 /trunk/src/ana/ (19 files in 5 dirs): Removed getopt_pp as a requirement of the sample apps, changed to GPLv2 and other minor changes.
20100615 10:15:42-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Quit: ...]
20100615 10:48:01-!- Blueblaze [~nick@adsl-99-171-160-117.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 264 seconds]
20100615 11:34:08-!- fendrin [~fabi@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20100615 11:44:23-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 260 seconds]
20100615 11:44:37-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100615 12:01:30-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100615 12:04:29-!- loonybot [~loonybot@ppp79-139-137-245.pppoe.spdop.ru] has joined #wesnoth-dev
20100615 12:04:29-!- loonybot [~loonybot@ppp79-139-137-245.pppoe.spdop.ru] has quit [Changing host]
20100615 12:04:29-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20100615 12:05:05-!- Elvish_Pillager [~eli@71-10-224-192.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20100615 12:05:21-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100615 12:11:38< elias> Soliton: guess that's the last as last time
20100615 12:12:55< elias> i'm somewhat unsure if i want to look at the old parser... or instead backport the --parse option
20100615 12:20:01< Soliton> still looks like an internal error to me. so that's not possible with the new parser?
20100615 12:22:34-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 264 seconds]
20100615 12:28:11-!- stikonas [~and@ctv-79-132-162-160.vinita.lt] has joined #wesnoth-dev
20100615 12:28:11-!- stikonas [~and@ctv-79-132-162-160.vinita.lt] has quit [Changing host]
20100615 12:28:11-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100615 12:46:03-!- Crab_ [~Crab@wesnoth/developer/crab] has joined #wesnoth-dev
20100615 12:51:02< Crab_> alink: mostly, yes, you are right. but there is a gotcha - there is a possibility that someone might need to implement per-ai visibility of game state, current structure allows to do that more easily.
20100615 12:54:50< Crab_> shadowmaster: you were looking for me ?
20100615 13:28:38-!- kevg [~kevg@94.232.5.102] has left #wesnoth-dev []
20100615 13:45:31-!- Ken_Oh [~briang@static-71-178-174-220.washdc.fios.verizon.net] has joined #wesnoth-dev
20100615 14:01:25-!- King_Elendil [~King_Elen@75.143.233.138] has joined #wesnoth-dev
20100615 14:15:00-!- King_Elendil [~King_Elen@75.143.233.138] has quit [Quit: I hope y'all have a nice day ;)]
20100615 15:15:05-!- fendrin [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20100615 15:20:02< Elvish_Pillager> Hmm. I notice that unit overlays are drawn over standing animations, but not over other animations. Is it possible to create a unit animation (e.g. [extra_anim]) that specifies that orbs, status bars, and overlays are drawn over the unit?
20100615 15:20:29< Elvish_Pillager> (And also, is it possible to create a [standing_anim] that removes them?)
20100615 15:31:03-!- Khezar [~Greywhind@138.16.59.131] has joined #wesnoth-dev
20100615 15:38:55-!- linease [~linease@dslb-092-076-127-224.pools.arcor-ip.net] has quit [Quit: Lost terminal]
20100615 15:40:42< Elvish_Pillager> Ah, I need with_bars=yes in my [animate_unit] tag.
20100615 15:41:41-!- linease [~linease@dslb-092-076-127-224.pools.arcor-ip.net] has joined #wesnoth-dev
20100615 15:42:30-!- FAAB [~huajie@219.142.242.114] has joined #wesnoth-dev
20100615 15:45:44-!- Khezar is now known as Greywhind_
20100615 15:56:59-!- FAAB [~huajie@219.142.242.114] has quit [Quit: Leaving.]
20100615 15:57:24-!- _jbx_ [~jbailey@12.190.80.225] has joined #wesnoth-dev
20100615 16:06:21-!- kevg [~kevg@94.232.5.102] has joined #wesnoth-dev
20100615 16:22:53< kevg> Crab_: hi. What command line argument switches on DBG_AI_TESTING_AI_DEFAULT?
20100615 16:23:11< Crab_> hi, kevg
20100615 16:23:51< Crab_> --log-debug=ai/ca/testing_ai_default
20100615 16:24:04< Crab_> see the constructor argument in static lg::log_domain
20100615 16:24:18< kevg> Crab_: thanks
20100615 16:24:19< Crab_> you can also use --log-debug="ai/ca/*" or --log-debug="ai/*"
20100615 16:24:50-!- DrJones-wesnoth [5479bc31@gateway/web/freenode/ip.84.121.188.49] has joined #wesnoth-dev
20100615 16:24:56< DrJones-wesnoth> Hi!
20100615 16:24:57-!- euschn [~eugen@wesnoth/developer/euschn] has quit [Quit: Leaving.]
20100615 16:25:33< Crab_> hi, DrJones-wesnoth
20100615 16:26:12< DrJones-wesnoth> Is shadowmaster online?
20100615 16:26:59< DrJones-wesnoth> Hello Crab_ !!! Have a lollipop (>")> @---
20100615 16:28:04< DrJones-wesnoth> Is Ivanovic online?
20100615 16:28:15< Ivanovic> no!
20100615 16:28:23< DrJones-wesnoth> D:
20100615 16:28:29< DrJones-wesnoth> I'm out of luck, I guess
20100615 16:28:47< DrJones-wesnoth> :P
20100615 16:29:20< DrJones-wesnoth> Ivanovic, my brother has told me there's something called Wesnoth Summer Art Scholarship
20100615 16:29:36< DrJones-wesnoth> And asked me to register there, but there's only today left to inscribe
20100615 16:29:56< Ivanovic> honestly i don't know how this is supposed to wokr
20100615 16:29:57< DrJones-wesnoth> Unless I read it wrong, and there are no more inscriptions ô_ò?
20100615 16:30:02< Ivanovic> you should talk to noy about it
20100615 16:30:07< DrJones-wesnoth> Well, at least, it says something the wiki
20100615 16:30:11< Ivanovic> (as well as jetrel)
20100615 16:31:23-!- dtiger [~dtiger@dynamic-vpdn-213-184-233-16.telecom.by] has joined #wesnoth-dev
20100615 16:34:22-!- DrJones-wesnoth [5479bc31@gateway/web/freenode/ip.84.121.188.49] has quit [Ping timeout: 252 seconds]
20100615 17:18:17-!- alink [~alink@wesnoth/developer/alink] has joined #wesnoth-dev
20100615 17:18:43< alink> hi
20100615 17:18:53< alink> Crab_: ok
20100615 17:19:02< Crab_> hi, alink
20100615 17:19:38< alink> but when you said per-ai visibility, do you mean fog/invisibility stuff ?
20100615 17:20:20< Crab_> alink: yes, and stuff related to 'actions with the hypothetical unit map'
20100615 17:20:24< alink> which already work well on the main unit_map
20100615 17:20:35< alink> ah ok "hypothetical unit map"
20100615 17:21:23< Crab_> alink: for example, whiteboard games (which gabba does) play with unit_map to make it display something that is not really there
20100615 17:21:33< alink> then this future ai-dev will have the same messy problems that I mentioned to gabba for his whiteboard
20100615 17:21:37< alink> :)
20100615 17:21:40-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100615 17:21:45< Crab_> yes
20100615 17:22:14< alink> that is many game objects store a reference to the main real unit_map and use it
20100615 17:22:25< Crab_> but it's somewhat easier for the ai...
20100615 17:22:38< alink> and only some functions allow to pass custom unit_map parameter
20100615 17:23:20< Crab_> since all 'things that change the game state' are already encapsulated in the ai in one place, and all thing that check something use the units map from 'ai info'
20100615 17:24:58< alink> "all thing that check something", I suspect that there is hidden uses of the main unit_map. that is ignoring the one of ai info (which is the same for now)
20100615 17:26:43< alink> already found an example : unit abilities 
20100615 17:27:29< alink> Crab_: I would recommend to this future ai dev to use the same trick that I suggested to gabba: use the main unit_map and temporary modify it
20100615 17:28:02< alink> it may even be possible to reuse whiteboard code for such ai task
20100615 17:28:23< Crab_> so, you suggest that we dismantle ai::game_info to avoid duplication ?
20100615 17:28:45< alink> at least remove what is already in resources
20100615 17:29:47< Crab_> yes, that could be done, then. there's a few other gotchas, but they are smaller (e.g., the c++ ai developer would have to know what he can access, and what he cannot access, from resources:: )
20100615 17:30:04< alink> if these refrences are the same, it's a useless redundancy, and if not it will be huge global cleaning to make it work correctly
20100615 17:30:17< alink> you mean to prevent ai cheating ?
20100615 17:30:21< Crab_> as for now, they're the same
20100615 17:30:35< Crab_> not only cheating, but oos'ing, too
20100615 17:31:05< alink> mmh, we maybe need a const resources
20100615 17:31:33< alink> and then be sure that ai only use engine functions to modify sensible data
20100615 17:31:37< Crab_> but it's mainly for developer convenience.. so, if a dev would remember that 'accessing units, teams, map is ok', then there's no problems.
20100615 17:31:45< Crab_> 'ai only use engine functions to modify sensible data'  - already done
20100615 17:31:54< alink> ok good
20100615 17:32:54< alink> but anyway a const resources seems not a bad idea, it's true that we lost const correctness when we replaced const parameters by resources
20100615 17:32:55< Crab_> so yes, you're right, it'll be a good thing to get rid of that part of ai::game_info which is present in resources::, and remove all accociated caches/parameters
20100615 17:34:01< Crab_> but, the 'constness' of resources is not a problem, compared to other issues the current code has
20100615 17:34:21< alink> such as ?
20100615 17:34:41< alink> you mean oos risks
20100615 17:35:20< Crab_> no, the other things.   now, the 'game rules' part is mixed with the UI part. for example, there is no engine function which says "yes, unit X can attack unit Y now, or no, it cannot"
20100615 17:35:56< Crab_> ai does those checks for itself (see src/ai/actions.cpp), ui does it for itself (scattered all over the code), replay code does it for itself
20100615 17:36:21< alink> indeed, no global rules checking
20100615 17:36:35< alink> well some main action pass by actions::
20100615 17:36:49< alink> I mean actions.cpp
20100615 17:36:49< Crab_> yes, something is in src/actions.cpp
20100615 17:37:34< alink> move was unified there recently IIRC
20100615 17:37:44< alink> with your help I think
20100615 17:37:59< Crab_> ai part was cleaned up
20100615 17:38:13< Crab_> ai now delegates some of the main processing to the src/actions.cpp
20100615 17:38:18< alink> replay moves moved there for 1.6
20100615 17:38:38< Crab_> but there's still a lot of things which are scattered and duplicated
20100615 17:39:11< alink> so, one possible plan could be to move all 'game actions' in actions.cpp and add various validity check there
20100615 17:39:39< alink> probably starting with attacks
20100615 17:39:54< Crab_> yes. and extract common code to helper functions to avoid duplication of game rules
20100615 17:40:40< Crab_> for example, the recent bug about segfault on sighted event after move in move+attack...
20100615 17:41:01< alink> indeed, good example
20100615 17:41:26< alink> ok, I see. You are right, for an ai-dev, these problems are bigger than constness
20100615 17:41:32-!- dtiger [~dtiger@dynamic-vpdn-213-184-233-16.telecom.by] has quit [Remote host closed the connection]
20100615 17:42:22< alink> Crab_: about cleaning game_info, the only one left seems to be 'recent_attacks', what is it? Seems not much used
20100615 17:42:58< alink> ah attack_close() reads it
20100615 17:43:29< Crab_> alink: we can remove it from attack_close and unit test the ai, to see if there's negative impact
20100615 17:44:15-!- dtiger [~dtiger@dynamic-vpdn-213-184-233-16.telecom.by] has joined #wesnoth-dev
20100615 17:44:24< Crab_> if there was a clean unified interface to 'do the attack only if it is ok', then no ui bug would cause such a segfault.
20100615 17:45:03< Crab_> alink: e.g, the ai, which has such checks, in that situation (defender stolen by wml sighted event) would only compain like 'error ai/actions: Error #1002 (attack_result::E_EMPTY_DEFENDER) in attack by side 1 from location 8,6 to location 7,7 using weapon -1 with aggression 0.4 ' without any funny segfaults.
20100615 17:46:02< alink> well there is still the problem that WML event could invalidate your variables in the middle of local code, so this local code can still crashes
20100615 17:46:53< Crab_> then that becomes a problem in a 'engine layer' of code
20100615 17:47:19< Crab_> e.g., things are separated between the 'ui stuff' and the 'engine stuff', and the latter is easier to get right once.
20100615 17:48:00< alink> yeah, ok it would help to fix/prevent such cases
20100615 17:49:11< Crab_> also, it would help the ai to know more about the game.
20100615 17:49:45< Crab_> e.g., it's now impossible to reliably determine 'how much healing will my unit get if it'll be standing on A next turn'
20100615 17:49:53-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20100615 17:50:03-!- DrJones-wesnoth [5479bc31@gateway/web/freenode/ip.84.121.188.49] has joined #wesnoth-dev
20100615 17:50:06-!- billynux [c8078d05@wesnoth/developer/billynux] has joined #wesnoth-dev
20100615 17:50:06< Crab_> since the actual code which does the healing does all the 'correct' calculations internally
20100615 17:50:08< DrJones-wesnoth> Hi
20100615 17:50:18< alink> hi DrJones-wesnoth
20100615 17:50:21< DrJones-wesnoth> Did I miss something since the last time I wrote?
20100615 17:50:33< DrJones-wesnoth> There was a monsoon or something in my city
20100615 17:50:39< alink> Crab_: WML stuff is also a problem for such 'predictions'
20100615 17:50:40< DrJones-wesnoth> and it seems I disconnected
20100615 17:50:49< DrJones-wesnoth> *got disconnected
20100615 17:51:08< Crab_> alink: yes. but at least if the same code path is reused, it's closer to truth and less code lines.
20100615 17:51:12 * DrJones-wesnoth makes the sad face U_U
20100615 17:51:27< alink> Crab_: agreed
20100615 17:51:35-!- Gambit [~quassel@pa-67-234-73-7.dhcp.embarqhsd.net] has joined #wesnoth-dev
20100615 17:51:58< alink> mmh but back to attack_close(), it seems hard to check, it's deep in a special case.
20100615 17:52:21< alink> Crab_: your ai farm is still playing ?
20100615 17:52:44< zookeeper> DrJones-wesnoth, no, you didn't miss anything. have you registered on the forums?
20100615 17:53:00< Crab_> alink: it's not, but it's online, so we can, say, 'remove it for one of the ais' and set up a test
20100615 17:53:45< Crab_> also, it's not a problem if game_info remains
20100615 17:54:01< Crab_> ai::game_info is rarely passed as a function parameter
20100615 17:54:15< Crab_> since it's usually available via get_info() method which is inherited when you're in ai context
20100615 17:54:15< alink> Crab_:  indeed better keep it for possible future new ai datas
20100615 17:54:54< Crab_> so, there's no price to pay if we keep ai::game_info, we can just get rid of teams/map/units and simplify code all over the ai to use resources::
20100615 17:55:13< alink> I never wanted to kill game_info just remove the resources parts, as I wish to do in other objects
20100615 17:55:28< Crab_> so, the recent_attacks is not a problem
20100615 17:55:39< alink> ok then it's decided
20100615 17:56:03< Crab_> the original purpose of recent_attacks is something like 'if you do stupid things with attacks, do it with a large group of units, maybe it'll work'
20100615 17:57:28< DrJones-wesnoth> zookeeper yes, I have registered under the name DrJones
20100615 17:57:33< alink> aarrgh there is a huge number of get_info()'s use
20100615 17:57:48< alink> ~260
20100615 17:58:32< Crab_> yes, and a large number of uses of the local caches of get_info().units , get_info().map, get_info().teams
20100615 17:59:12< zookeeper> DrJones-wesnoth, ok, well if you're interested in the summer art scholarships then send a forum private message to jetrel and noy, i guess.
20100615 18:00:35< zookeeper> i'm presuming that no one knows who you are or what you can do, so obviously you need to have some convincing prior art to show
20100615 18:02:29-!- Netsplit *.net <-> *.split quits: alink, Elvish_Pillager, Soliton, dtiger
20100615 18:03:40-!- Netsplit over, joins: alink, dtiger
20100615 18:03:53< DrJones-wesnoth> Yes, I guess I'll have to show something
20100615 18:04:50-!- Elvish_Pillager [~eli@71-10-224-192.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20100615 18:04:50-!- Soliton [~Soliton@wesnoth/developer/soliton] has joined #wesnoth-dev
20100615 18:04:51< alink> Crab_: so, I may start this change? Or do you prefer do it yourself?
20100615 18:06:46< CIA-86> alink * r43453 /trunk/src/Makefile.am: 
20100615 18:06:46< CIA-86> Use top_srcdir instead of .. for ../data/tools copy.
20100615 18:06:46< CIA-86> Seems needed to use make in another directory.
20100615 18:06:49-!- ancestral [~ancestral@97-116-115-34.mpls.qwest.net] has joined #wesnoth-dev
20100615 18:09:21< Crab_> alink: yes, I see no problems with it
20100615 18:09:35< Crab_> alink: if you do have time and determination, do it :)
20100615 18:10:20< alink> Crab_: ok, then I fear that I will do it. It's cleaner to fix this before the other changes that I really want
20100615 18:10:51< Crab_> alink: I will modify src/ai/actions.cpp , that one is slightly more tricky than all the others
20100615 18:10:53< kevg> Summer Art Scholarship is something like Summer of Code but for designers?
20100615 18:11:04< Crab_> kevg: for artists, yes
20100615 18:11:08< DrJones-wesnoth> Yes
20100615 18:11:12< alink> Crab_: ok
20100615 18:11:17< DrJones-wesnoth> One thing, I've created my page on the wiki
20100615 18:11:20< kevg> hm, interesting
20100615 18:11:21< DrJones-wesnoth> for the Summer Art
20100615 18:11:35< DrJones-wesnoth> But the link at the bottom of the page shows red even though it exists
20100615 18:11:42< DrJones-wesnoth> http://wiki.wesnoth.org/Wesnoth_Summer_Art_Scholarship
20100615 18:11:51< DrJones-wesnoth> Can someone tell me if they see the last link red or blue?
20100615 18:11:55< DrJones-wesnoth> at the bottom of the page
20100615 18:12:45< thespaceinvader> DrJones-wesnoth: it's a redlink.  But that page is a year out of date, and no longer relevant
20100615 18:13:27< DrJones-wesnoth> Does it mean that I've been writing my application on a page that is no longer used? ô_ò?
20100615 18:13:35< thespaceinvader> see http://www.wesnoth.org/forum/viewtopic.php?f=9&t=30273
20100615 18:13:43< thespaceinvader> yes, it does rather i'm afraid
20100615 18:19:06< alink> zookeeper, Sapient: [event] in [unit_type] (and [unit]) is still considered supported right? Is it still used? because it seems limited and recruit/recall looks more powerful
20100615 18:19:26< alink> asking because the code seems messy and I see possible bugs there
20100615 18:22:32< alink> I don't see any mainline use, but it's hard to search
20100615 18:24:50-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100615 18:32:58< thespaceinvader> DrJones-wesnoth: the best way for us to know whether we can make use of your work is for you to show us some of it ;)
20100615 18:33:34-!- billynux [c8078d05@wesnoth/developer/billynux] has quit [Quit: Page closed]
20100615 18:33:39< DrJones-wesnoth> Yeah, but I don't know what you are looking for, hehehe :P
20100615 18:35:37-!- Crab_ [~Crab@wesnoth/developer/crab] has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
20100615 18:36:17< alink> Crab_: what is this position_callable thing? it does weird things with a private unit_map which it seems not to use
20100615 18:36:25< alink> oh :-/
20100615 18:36:27< DrJones-wesnoth> http://agdiforums.com/forum/download/file.php?id=291 <-- this is the first and last picture I did with Ms Paint. heh
20100615 18:36:57< DrJones-wesnoth> http://losers.of.rokugan.googlepages.com/DrJones-Mascota1.gif <-- and this is a little animation I have around here
20100615 18:37:44< DrJones-wesnoth> Of course, these things aren't pixel art, maybe I should do one and then post it :-/
20100615 18:39:50< alink> Crab_: and worse, it does copy of the main unit_map. If it don't use it, it's a big waste
20100615 18:40:17< DrJones-wesnoth> I also did this desktop wallpaper that won a first prize at some site: http://www.agdinteractive.com/games/qfg2/fans/contest1/QFG2_GrandPrizeWinner1.jpg
20100615 18:41:41< DrJones-wesnoth> However, they all look a bit cartoony, which is why I'm reticent of posting on the boards
20100615 18:42:49 * DrJones-wesnoth needs to work harder! (>")>/~~~~ ( x_x) *tchlack*
20100615 18:50:11-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100615 18:50:29-!- gabba [~gabba@wesnoth/developer/gabba] has joined #wesnoth-dev
20100615 18:50:35< gabba> bonjour
20100615 18:50:44< alink> hi gabba
20100615 18:50:48< gabba> hi alink
20100615 18:51:29< gabba> alink: we need to chat about input
20100615 18:52:10< alink> gabba: ok
20100615 18:53:00< alink> gabba: the commits that I saw seems to do as I would do: a global wb flag modulating the actions done in mouse_events.cpp
20100615 18:54:24< gabba> alink: so, you think that plugging in the various if/else structures that extra wb flag is the way to go? I was kind of looking for a cleaner global approach
20100615 18:55:26< alink> well, at least it's the easier way to go :)
20100615 18:56:18< alink> It depends of which mouse features you'll need, which is hard to tell before having a working prototype
20100615 18:57:18< alink> gabba: but you will still need some of the current mouse features, right? so you can't really replace the whole mouse code
20100615 18:57:50< thespaceinvader> DrJones-wesnoth: i suggest you have a good look round the forums.  There are various threads around there detailing what art is currently needed and what is not.  Most of the art organisation goes on there, rather than on IRC
20100615 18:58:15< gabba> I have a pretty good idea of the features I need: intercept select, intercept pathfinding to define moves, intercept move creation; also, I need to execute programmed moves, for which reusing some methods in mouse_events.cpp might be the best approach.
20100615 18:58:38< DrJones-wesnoth> Okey, thespaceinvader, thank you!
20100615 18:58:43< gabba> Well I also need to intercept recruit and attack creation eventually
20100615 18:58:56< thespaceinvader> no worries DrJones-wesnoth 
20100615 18:59:37< gabba> alink: I can't really replace all the mouse code... that would amount to copy-pasting a lot of it in my own class
20100615 19:00:15< alink> gabba: and what kind of 'interception' have you in mind ?
20100615 19:00:55< gabba> alink: basically what I do now: if we're about to move a unit, if whiteboard is active, create a planned move instead... and so on
20100615 19:01:20< alink> using if(wb) else
20100615 19:01:26< gabba> exact
20100615 19:01:59< alink> I don't see a simple OO 'cleaner' global way to do this
20100615 19:02:17< alink> but you may split the functions becoming too big
20100615 19:02:48< gabba> I thought of maybe subclassing the mouse handler and forwarding to it only the mouse events I don't handle, but that'll cause a lot of code duplication too
20100615 19:03:28< alink> gabba: yeah but you will need to handle most of the main mouse events
20100615 19:04:19< alink> and you can't let the old code run normally, so there will be if(!wb) there anyway
20100615 19:05:21< alink> IMO the most tricky one will be mouse motion
20100615 19:06:05< gabba> well, it would have been the same relationship as mouse handler vs base handler: for instance mouse handler calls wb handler, and if returns false does its own stuff... but I agree it's rather similar to "if (wb) else"
20100615 19:07:28< alink> gabba: if you really like such design... but I always find this not very flexible
20100615 19:07:54< alink> mainly because you can't use only parts of functions
20100615 19:07:56< gabba> I don't really like it... just thought I'd mention it in case it rung a bell :)
20100615 19:09:25< gabba> I can keep working as I do now; another thing I wonder about is how hard it will be to call such a function from my whiteboard class: "::move_unit(NULL, steps, &recorder, resources::undo_stack, true, &next_unit_, false, check_shroud);"
20100615 19:09:42< alink> maybe start by if(wb) else, see how they spread and when there is a clear pattern, transform it in a (more complex to write) subclass system
20100615 19:10:13< gabba> make it work and then refactor, gotcha
20100615 19:10:22< alink> ::move_unit() will execute a real move, is it what you want?
20100615 19:10:33< gabba> alink: that would be the idea
20100615 19:10:52< alink> triggering by a mouse event ?
20100615 19:12:29-!- Blueblaze [~nick@99.171.160.117] has joined #wesnoth-dev
20100615 19:12:30< alink> I'd say it's not hard to call if your wb class have all the info :)
20100615 19:12:40< gabba> alink: probably triggered by a left click on the planned move destination, or with a hotkey ("execute move #1")
20100615 19:13:53< alink> this is not an input problem, right ?
20100615 19:14:04< gabba> precisely, is that info relatively easy to obtain, i.e. not too context-dependent on the input class&
20100615 19:14:17< gabba> s/&/?
20100615 19:15:20< alink> maybe next_unit is not cleanly separated 
20100615 19:15:46< alink> I am not sure about replay
20100615 19:16:16< zookeeper> alink, still got questions about unit type events?
20100615 19:16:32< alink> gabba: but the ai and replay manage to call it, so it should not be too linked to input
20100615 19:17:01< gabba> alink: ok. Thanks, I think you answered my main questions.
20100615 19:17:03< alink> zookeeper: just: is it really used ?
20100615 19:17:15< zookeeper> alink, yeah, feeding uses it and a lot of UMC too AFAIK
20100615 19:17:24< zookeeper> but it's a bit limited, yes
20100615 19:17:47< alink> zookeeper: no too much bug complains ?
20100615 19:18:09< alink> zookeeper: because I see various potential ones
20100615 19:19:29< zookeeper> alink, well, i'm not aware of any outright bugs with it, no
20100615 19:19:45< zookeeper> there used to be a lot of bugs with the events duplicating and so on
20100615 19:20:16< gabba> fendrin: around?
20100615 19:20:39< alink> zookeeper: yeah the event is potentially triggered a lot, but a second mechanism filter the ones already encountered, maybe that part is new
20100615 19:21:00< fendrin> gabba: yes
20100615 19:21:20< gabba> fendrin: you were interesting is using arrows in the editor, right?
20100615 19:21:34< fendrin> gabba: yes
20100615 19:21:46< fendrin> gabba: pathes
20100615 19:21:58< fendrin> gabba: arrows that symbolize unit pathes.
20100615 19:22:32< zookeeper> alink, anyway, they are a wanted feature, or at least if they'd be removed then something very similar would have to be added to replace them
20100615 19:22:38< gabba> fendrin: problem is, I'm highly tempted to use "if resources::screen == NULL" checks in various places of the arrow code, and I think that would bother you, right?
20100615 19:23:05< alink> zookeeper: ok, I will not remove it, but it need some cleaning
20100615 19:23:11< alink> zookeeper: for example, does unit in recall list should fire it? and move_unit_fake?
20100615 19:23:34< zookeeper> alink, undefined :P
20100615 19:23:35< fendrin> gabba: well, I paint on that screen don't I? There must be one around.
20100615 19:23:45< alink> zookeeper: ok I see :)
20100615 19:23:47< gabba> managing the lifecycle of those arrows proves to be very annoying... I'd like them to auto-unregister themselves from the display on destruction, but they outlive the display and try to notify it, resulting in a nice segfault
20100615 19:24:07< zookeeper> alink, move_unit_fake shouldn't, but not sure about the recall list...i'd say probably yes
20100615 19:24:30< fendrin> gabba: That looks like a shutdown problem?
20100615 19:24:33< gabba> fendrin: yes, but resources::screen has the guarantee of being NULL if the display isn't "live"... I don't know any other way of checking for that
20100615 19:25:15-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20100615 19:25:24< alink> zookeeper: maybe it's simpler to list the unit creation cases which should fire it : recruit, recall (or recall list), WML unit spawn. plague, anything else?
20100615 19:25:33< gabba> I saw a lot of such "if resources::screen == NULL" checks in the code, so I'm not the only one with this problem apparently
20100615 19:25:42< boucman> hey all
20100615 19:25:48< alink> boucman: hi
20100615 19:25:49< fendrin> hi boucman
20100615 19:25:51< gabba> hi boucman
20100615 19:25:58 * boucman is not going to stay for long
20100615 19:26:19< alink> zookeeper: my main problem is that any c++ unit manipulation seems to fire it, instead of just real game stuff
20100615 19:26:28< fendrin> gabba: There is nothing wrong with display being a Null pointer during the shutdown process. 
20100615 19:26:33< zookeeper> alink, sounds like it's easy to miss something...no, nothing else that i can think of. except leaders which are already there.
20100615 19:26:56< gabba> boucman: you arrive just in time for a little display pointer discussion :D
20100615 19:27:09< boucman> hehe
20100615 19:27:12< zookeeper> alink, is that a problem? presumably the game still checks if they've already been added so they don't get duplicated.
20100615 19:27:48< DrJones-wesnoth> I have to go, see you
20100615 19:27:57< alink> zookeeper: well, in some cases it could be fired when the unit was not really added to the map
20100615 19:28:16-!- DrJones-wesnoth [5479bc31@gateway/web/freenode/ip.84.121.188.49] has quit [Quit: Page closed]
20100615 19:28:22< gabba> fendrin: it's not wrong in itself, but I only have two ways of dealing with it: 1- ensure that my whiteboard manager and the team objects (that hold planned moves) die before the display, or 2- use the check I just mention, which excludes your editor I think, since you don't use the resources.hpp/cpp
20100615 19:28:25< alink> zookeeper: it's more a question of incorrect event's triggering
20100615 19:29:20< zookeeper> alink, when the events actually trigger doesn't have anything to do with whether it's a unit type event or a regular scenario event
20100615 19:29:21< fendrin> gabba: I do use resources cpp. I have a team with units to be displayed. That can't be done easily without the resources system.
20100615 19:29:28< zookeeper> or at least i very much doubt it does
20100615 19:30:26< zookeeper> or maybe i just don't understand what you're talking about ;)
20100615 19:30:30< gabba> fendrin: if you do use resources::, then my problem is solved I guess... I didn't really want to mess with the teams initialization/destruction order WRT display
20100615 19:31:41< fendrin> gabba: Okay, did you enable the terrain animation for the editor as well?
20100615 19:32:17< gabba> fendrin: terrain animation? I never touched that...
20100615 19:32:25< alink> zookeeper: yeah. I explain better. Currently any unit *creation* (without placing on the map or use it for real game purposes) will add a unit_type event to the event queue. That means, that if the AI or whiteboard want to create a temporary unit for internal reason, it triggers the event
20100615 19:32:49< fendrin> gabba: Okay
20100615 19:32:51< zookeeper> alink, s/triggers/adds, i presume
20100615 19:33:03< alink> zookeeper: yes, indeed, sorry
20100615 19:33:08< zookeeper> ah, ok, i get it now
20100615 19:33:29< alink> zookeeper: so I would prefer add the event handling only in the specific where the unit is really added to the game map
20100615 19:33:40< alink> +cases
20100615 19:34:02< zookeeper> well, that does sound like a bug but also one which should basically never have any ill effects, since there's hardly any sane usage of unit type events which doesn't actually do something based on the real units on the map
20100615 19:34:06< gabba> fendrin: however, after today's final refactoring arrows should be pretty much ready to use (i.e. API should stop changing so much). Graphics are still pretty basic though. If you use them, let me know how you fare...
20100615 19:34:26< zookeeper> alink, yeah, that sounds ok
20100615 19:35:30< alink> zookeeper: so the list of cases is simply : insertion into the game map and into the recall list
20100615 19:35:53< zookeeper> alink, yep
20100615 19:35:58< alink> zookeeper: ok thanks
20100615 19:36:16< gabba> boucman: executive summary of my input discussion with alink: the way I'm doing things now is probably the best one. Maybe some obvious refactorings will surface later, though.
20100615 19:36:53< zookeeper> alink, you can spawn single [unit]s with to_variable=variablename in which case they just get put into a variable instead of the game map (i don't know if they go through the game map internally though), but it's not a problem if the events don't get added then
20100615 19:37:13< boucman> the way you are doing being "acting at the calls to the real game structure" IIUC
20100615 19:37:25< zookeeper> (in fact the events definitely should _not_ be added in that case, but i thought i'd mention it anyway)
20100615 19:38:03< gabba> which bot is the acronym bot again :P ?
20100615 19:38:27< alink> zookeeper: ok, should not be a problem, but thanks for the warning
20100615 19:38:53< alink> shikadibot: IIUC
20100615 19:38:53 * shikadibot is offended
20100615 19:39:10< gabba> lol
20100615 19:39:21< alink> shikadibot: SiO
20100615 19:39:28< gabba> I keep forgetting some of those
20100615 19:39:42< alink> didn't work :-(
20100615 19:39:42-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20100615 19:40:06< gabba> shikadibot: elitist bastard
20100615 19:40:15< alink> IIUC : If I Understand Correctly
20100615 19:40:36< gabba> thanks alink, I looked it up in the meanwhile
20100615 19:42:39< alink> gabba: btw I have a question about wb and recruit. How do you will handle traits (esp. quick). Assume no traits ?
20100615 19:43:17< gabba> alink: no random traits
20100615 19:43:38< alink> ok, make sense and solve various coding problems
20100615 19:44:11< boucman> alink: traits don't influence the turn the unit appears since the unit can't attack/move afaict
20100615 19:44:49< alink> yes, but does wb is limited to 1-turn planning? (real question)
20100615 19:45:10< alink> note that they could be attacked
20100615 19:45:26< boucman> alink: for the time being yes, in the long term most likely yes, since making a good UI would be unmanageable
20100615 19:45:29< alink> or have adjacent ability effect 
20100615 19:45:43< boucman> alink: not during the player's turn, onlyt the oponent's turn
20100615 19:45:56< boucman> abilities are not linked to traits currently :P
20100615 19:46:13< alink> anyway, the mains problems that I saw are solved, all is fine :)
20100615 19:46:20< gabba> alink: I want to allow multi-turn planning, and planned actions that persist until the next turn if you don't execute them. boucman you don't think it'll work?
20100615 19:46:40-!- billynux [~billy@wesnoth/developer/billynux] has joined #wesnoth-dev
20100615 19:46:49< alink> gabba: the enemy moves will often change a lot of these plans
20100615 19:47:04< gabba> I'm not sure I understand "they could be attacked"
20100615 19:47:28< gabba> alink: yes, but keeping a few multi-turn planned moves would be a nice alternative to gotos
20100615 19:47:48< boucman> gabba: I tend to prefer to tackle a simple problem and then improve
20100615 19:48:05< boucman> if you manage single turn we'll discuss multi turn, but i'd rather have single turn work
20100615 19:48:09< boucman> and network second
20100615 19:48:16< boucman> multi turn is only third for me :P
20100615 19:48:45< gabba> boucman: in this case I'm not convinced that multi-turn is harder than single-turn; they each involve extra checks
20100615 19:48:52< alink> yes start by the simple (but already very complex) one turn planning
20100615 19:49:16< boucman> gabba: maybe, but you should still focus on single turn and "suprise us" later :P
20100615 19:49:41< alink> gabba: there is UI aspects, and all new turns changes (healing, ToD, for example )
20100615 19:50:40< billynux> most of the time multi-turn strategy doesn't make sense (think chess)... your strategy could become faulty... etc... but you probably already talked about it (oh, and hi)
20100615 19:50:41< boucman> and ennemy/allies turns... multi turn planning never really made sense, except far from the front line...
20100615 19:51:12< boucman> and even there gotos are not used that much because they can easily become wrong during your oponent's turn
20100615 19:51:16< gabba> alink: just another case where I need to validate all remaining planned actions (same as after every wml event)
20100615 19:51:59< gabba> boucman: for gotos it makes a lot of sense, since they auto-execute and you can't correct them outside of your turn... planned actions are different
20100615 19:52:15< boucman> gabba: ok, I think I understand better
20100615 19:52:34< boucman> keeping planned actions from one turn to the next makes sense and could indeed easily be done
20100615 19:53:21< boucman> multi turn planning (i.e planning both your move, your oponent's and your allies) is way too hard
20100615 19:53:34< boucman> so planning as a replacement of goto is fine, 
20100615 19:53:43< boucman> ok, I must leave now, see you all later
20100615 19:53:52< gabba> ok, see you boucman
20100615 19:53:58< billynux> see ya
20100615 19:54:14< billynux> how is it going gabba?
20100615 19:54:27< billynux> I have to start working on the client network code now...
20100615 19:54:37< gabba> hi billynux, I'm doing great thanks :)
20100615 19:55:01-!- Blueblazed [~nick@adsl-99-182-54-45.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100615 19:55:03< alink> billynux: btw did you get my IRC question about the Doxyfile in ana/ ?
20100615 19:55:09< billynux> the other two gsocers were....? upthorn & timotei?
20100615 19:55:30< billynux> alink, no, sorry, I should check IRC backlogs more often
20100615 19:55:36< gabba> billynux: yes, those are the ones
20100615 19:55:53< billynux> gabba, how are they doing? I don't see them much here
20100615 19:56:01< alink> billynux: just grep your name is enough
20100615 19:56:14< gabba> I think timotei is in exams or something like that
20100615 19:56:27-!- Blueblaze [~nick@99.171.160.117] has quit [Ping timeout: 245 seconds]
20100615 19:56:34< gabba> dunno about upthorn, except he commits from time to time
20100615 19:56:37< billynux> alink, do you know the wget command to get all wesnoth-dev logs?
20100615 19:57:22< alink> billynux: no sry
20100615 19:58:11< gabba> billynux: all of the logs? how, by generating the dates in the URLs somehow?
20100615 19:58:40< alink> just click every few days on few links ;-)
20100615 19:58:48< billynux> I know...
20100615 19:59:19< billynux> gabba, yes... wget -R -Alog -regex wesnoth-dev irclogs... something like that, but it eludes me ATM
20100615 19:59:21< alink> using some firefox tools if help is needed to search faster
20100615 19:59:25< gabba> billynux: I have the command for one day if you want :P; I just CTRL+r and type "wg" to recall the last one and edit it again
20100615 19:59:52< billynux> :)
20100615 20:00:09< gabba> readline is your friend when memory fails
20100615 20:00:39< billynux> alink, damn... did I screw doxygen that bad?
20100615 20:01:06< alink> billynux: depends, I don't remember how it was before
20100615 20:01:22< billynux> if there was something there, it probably wasn't what it is now
20100615 20:01:44< alink> yeah I figured
20100615 20:02:17< billynux> did I move ana's Doxyfile to trunk/doc directly? (I thought I put it in doc/ana)
20100615 20:02:27< alink> but maybe it's a problem on mordante's side (assuming that it's still him who generate them)
20100615 20:02:44< billynux> I dunno, he should probably log on shortly
20100615 20:03:21< alink> there is a Doxyfile in src/ana
20100615 20:04:04< alink> but I am not familiar with doxygen details
20100615 20:04:42< billynux> there is... and it has OUTPUT_DIRECTORY       = doc/code
20100615 20:05:31< billynux> if you run doxygen on that file it could overwrite what is on doc/code (relative path), but it does sound kind of a longshot
20100615 20:06:03< alink> gabba: wb crash when I reselect a unit having already a planned move
20100615 20:12:33-!- drusepth [~drusepth@dhcp-077-249-151-209.chello.nl] has joined #wesnoth-dev
20100615 20:19:29-!- YogiHH [YogiHH@d095212.adsl.hansenet.de] has joined #wesnoth-dev
20100615 20:19:50-!- YogiHH [YogiHH@d095212.adsl.hansenet.de] has quit [Changing host]
20100615 20:19:50-!- YogiHH [YogiHH@wesnoth/developer/yogihh] has joined #wesnoth-dev
20100615 20:22:43< billynux> anybody have any clue as to why CMake wouldn't find installed Boost sublibraries? -> Boost_IOSTREAMS_LIBRARY
20100615 20:27:45< gabba> alink: known bug, there are a lot of rough edges
20100615 20:35:05-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20100615 20:35:17< mordante> servus
20100615 20:36:51< billynux> hi mordante
20100615 20:36:56< mordante> hi billynux 
20100615 20:37:40< billynux> mordante,  I removed getopt_pp as a dep. -> I manually parse the options in order, simple but not elegant
20100615 20:37:53< mordante> ok nice
20100615 20:38:00< billynux> I'm trying to test an SDL_net server with an ana client
20100615 20:38:21< billynux> looking at http://gpwiki.org/index.php/SDL:Tutorial:Using_SDL_net (besides the brasil match)
20100615 20:45:43< mordante> billynux, how hard would it be to test with our own server?
20100615 20:46:04< mordante> billynux, btw did you see alink's remark regarding devdocs.wesnoth.org?
20100615 20:46:51-!- kevg [~kevg@94.232.5.102] has quit [Read error: Connection reset by peer]
20100615 20:47:23< billynux> mordante, yes I saw it... was that my fault? how come? there is only one Doxyfile in src/ana
20100615 20:47:32-!- kevg [~kevg@94.232.5.102] has joined #wesnoth-dev
20100615 20:48:02< billynux> mordante, I don't know how hard it would be to test to wesnoth, but I wanted to test a sample app first
20100615 20:48:34< billynux> (just a successful connection and a TCP_Write32)
20100615 20:49:24< billynux> I'm having trouble linking: undefined reference to `SDLNet_Init'
20100615 20:49:26< mordante> billynux, I guess doxygen uses that one also not sure why, but better move it in a separate directory out of the src tree see whether that fixes the issue
20100615 20:49:55< billynux> find_package( SDL 1.2.7 REQUIRED COMPONENTS net) and target_link_libraries( sdl_test ana ${SDL_LIBRARY} )
20100615 20:50:21< billynux> do you want me to delete it? or where would it be better off?
20100615 20:51:04< mordante> s/SDL_LIBRARY/SDLNET_LIBRARY/ should fix the issue
20100615 20:51:55< billynux> 1'
20100615 20:52:01< mordante> ?
20100615 20:52:20< billynux> mordante, nope (1' means 1 minute)
20100615 20:52:32< mordante> ah ok
20100615 20:52:56< billynux> ccmake doesn't show any SDLNET_LIBRARY defined, SDL_LIBRARY is both the .a, the  .so and -lpthread
20100615 20:53:05< billynux> (libSDLmain.a etc)
20100615 20:53:14< mordante> maybe move it the doc/ana/doxygen/Doxyfile
20100615 20:53:49< billynux> ok, sure that will fix it? :)
20100615 20:54:17< mordante> no but we can try ;-)
20100615 20:54:37< billynux> k
20100615 20:55:07< mordante> keep the SDLNET_LIBRARY and add find_package(SDL_net REQUIRED)
20100615 20:55:25-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has joined #wesnoth-dev
20100615 20:55:26< mordante> guess the net isn't a component of SDL
20100615 20:56:01< billynux> k
20100615 20:56:22< billynux> that fixed it
20100615 20:56:55< mordante> good
20100615 20:57:04< billynux> if I move the Doxyfile there... any idea how to set up access to the ana sources from the config? -> looking tat the options
20100615 20:58:09< mordante> use STRIP_FROM_PATH to strip the src/ana
20100615 20:58:30< billynux> good
20100615 20:59:01< mordante> and INPUT where the files are and I prefer RECURSIVE = YES to traverse the subdirectories
20100615 20:59:45-!- Gambit [~quassel@pa-67-234-73-7.dhcp.embarqhsd.net] has quit [Ping timeout: 240 seconds]
20100615 21:00:04< billynux> mordante, so... STRIP_FROM_PATH = src/ana and INPUT = ../../src/ana ?
20100615 21:00:20< billynux> (sorry, doxygen config. not my forte
20100615 21:00:22< billynux> )
20100615 21:00:41< mordante> probably yes but best test
20100615 21:00:53< billynux> ok
20100615 21:01:16< mordante> I'm also not intimate with it, most of the times you set it up once and then forget about it ;-)
20100615 21:01:23< billynux> true..
20100615 21:01:33-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20100615 21:01:33< billynux> any recommendations to start on the cient code?
20100615 21:02:28< mordante> client code as Wesnoth's client code?
20100615 21:02:51< billynux> yep
20100615 21:03:20< billynux> doxygen went kaput... RECURSIVE = yes searches all of wesnoth?
20100615 21:04:01< mordante> it searches recursively from the directories listed in INPUT
20100615 21:04:44< mordante> before we had src src/widgets src/serialization etc etc and whenever a directory was add it wasn't to the INPUT list...
20100615 21:04:59< billynux> however its trying to read all of wesnoth
20100615 21:05:21< mordante> then turn it off, if you want I can look into it later this week
20100615 21:06:16< billynux> where is the on/off switch? (delete Doxyfile?)
20100615 21:06:45< mordante> I meant turn of the RECURSIVE ;-)
20100615 21:09:32< mordante> gabba did you see the information I left for you in the logs
20100615 21:09:35< billynux> ok... I was running doxygen from wesnoth's trunk... now CREATE_SUBDIRS = yes refuses to work
20100615 21:10:29< mordante> do you want me to look at it?
20100615 21:10:48< Ivanovic> noy: have you seen the logs regarding 
20100615 21:11:37< billynux> mordante, ok, commiting
20100615 21:11:47< gabba> mordante: nope, looking
20100615 21:11:53< CIA-86> billynux * r43454 /trunk/ (6 files in 3 dirs): Moved ana's Doxyfile to doc/ana and added a test subdir.
20100615 21:11:55< Ivanovic> noy: regarding DrJones-wesnoth and the wesnoth summer art scholarship
20100615 21:12:41< mordante> gabba, the 13th starting at 20:19
20100615 21:13:36< gabba> ok, reading
20100615 21:15:38< gabba> mordante: noted for doubles; do we use "double" or a typedef?
20100615 21:16:06< mordante> no normally simply a double, the normal floating point type in C++
20100615 21:16:20< gabba> ok
20100615 21:17:31< gabba> about initialization, as you saw I'm careful about initializing basic types especially pointers; but given the possibility of errors you pointed out I'm ok with just initializing all
20100615 21:18:12< mordante> it's really a shame g++ can't only warn about the types where it really matters :-(
20100615 21:18:50< mordante> but I prefer spending time on `fixing' all initialization over having to valgrind again
20100615 21:19:05< CIA-86> crab * r43455 /trunk/src/ai/ (actions.cpp actions.hpp): refactor src/ai/actions.?pp to use resources:: instead of ai::game_info
20100615 21:19:07-!- Gambit [~quassel@pa-67-234-73-7.dhcp.embarqhsd.net] has joined #wesnoth-dev
20100615 21:19:13< Crab_> alink: ^ done
20100615 21:19:14< gabba> right, that would be a nice FR for them actually
20100615 21:20:16< mordante> I'm not really sure whether they would do it, the switch is already activating a lot more warnings...
20100615 21:21:56< mordante> btw gabba it's always a good idea to grep the logs for your name
20100615 21:22:22< mordante> we tend to use it quite often to leave messages for other developers
20100615 21:22:56< gabba> mordante: I do it. Actually I had seen the two lines starting with "gabba", but not the whole explanation following them.
20100615 21:23:04< mordante> ah ok
20100615 21:26:13< mordante> alink, devdocs is generated on the server for a long time now
20100615 21:27:05< gabba> mordante: probably much faster alternative to Valgrind: http://sourceforge.net/apps/mediawiki/cppcheck/index.php?title=Main_Page -- never used it, but the list of bugs they found with it http://sourceforge.net/apps/mediawiki/cppcheck/index.php?title=Found_bugs seems to indicate a good tool.
20100615 21:27:49< gabba> more an alternative to gcc's static analysis, actually, since valgrind is runtime
20100615 21:28:02< mordante> gabba, do you use git-svn?
20100615 21:28:08< gabba> mordante: yes
20100615 21:28:42< mordante> run  git log |grep -C5 cppcheck  :-P
20100615 21:29:34< gabba> lol, point taken
20100615 21:30:15< mordante> but I used valgrind to find runtime bugs and static code analysis doesn't always work 
20100615 21:30:16< billynux> mordante, good... connection works (no surprise there)... working on message decryption
20100615 21:30:37< gabba> well, maybe cppcheck could be customized to check only base types for initializations or something like that... just an idea
20100615 21:31:24< mordante> but it's a rather nice tool and upstream is rather active and fast with fixing bugs
20100615 21:31:59< mordante> they already report uninitialized types, only not sure which types
20100615 21:32:06< mordante> billynux, nice
20100615 21:32:22< mordante> billynux, did you get boost iostreams working with cmake?
20100615 21:33:10< billynux> yep, I had a version I compiled... I just installed the ubuntu provided -dev packages
20100615 21:35:20< mordante> ok
20100615 21:35:57< alink> Crab_: thanks, that looked indeed a bit more tricky than a simple rename
20100615 21:37:06< mordante> billynux, regarding the client code how far are you to start on it?
20100615 21:37:30< alink> Crab_: btw i see get_info in ai/actions.hpp and ai/contexts.hpp. and only one is const. Any reason? 
20100615 21:38:39< billynux> I can start ASAP, if you give me the go-ahead, but I just wanted to test how would SLD_net / asio would work
20100615 21:39:19< alink> also using "const virtual ..." is weird and my IDE highlight it as a syntax error
20100615 21:40:00< Crab_> alink: ai/actions.cpp are representations of ai actions, they are not in context of an ai. changes to game state do not change the action itself
20100615 21:40:59< Crab_> alink: ai/contexts.hpp is the ai context, changes in game info there are really changes in the state of a particular thing which is in ai context.
20100615 21:41:48< Crab_> alink: other files shouldn't be tricky
20100615 21:42:11< Crab_> alink: it's just there was an opportunity to get rid of some unused stuff, simplifying code.
20100615 21:42:35< mordante> billynux, I've no objection if you start to replace the Wesnoth SDLNet code with ana code
20100615 21:42:39< alink> Crab_: mmmh ok and so does this constness is important enough to need introduction of const ressources:: before this change ?
20100615 21:42:53< Crab_> alink: no, it's not important in there
20100615 21:43:13< billynux> mordante, why do you read a remote_handle in network.cpp:720? (what does that remote_handle mean? )
20100615 21:43:15< mordante> do you already know whether you want to keep the API the same or change everything completely?
20100615 21:43:15< alink> Crab_: ok, thanks. And I can still add it later if needed
20100615 21:43:22< Crab_> alink: the simple rule is 'in ai, do not ever change the game state outside of src/ai/actions.cpp '
20100615 21:43:57< alink> Crab_: seems to be a nice simple rule :-)
20100615 21:44:14< Crab_> alink: you lose compiler protection by removing the const-ness in some places, yes. but it's not that important for now.
20100615 21:44:30< billynux> mordante, did we agree on a compilation option to choose SDL / ana ?
20100615 21:45:16< Crab_> alink: the rule is nonenforceable anyway thanks to lua ai support :)
20100615 21:45:27< billynux> I could 1- Move current SDL_net code to a particular file. 2- Add autotools/CMake/SCons options to select SDL or ana 3- Start to work on my own ana version of the client
20100615 21:45:46< mordante> billynux, yes/no we can add it you think it makes it easier for the transition, but after gsoc SDLNet should be gone
20100615 21:47:14< billynux> mordante, agreed, but I don't want to break the current wesnoth client network capabilities
20100615 21:47:50< mordante> billynux, http://www.libsdl.org/cgi/docwiki.cgi/SDLNet_Read32
20100615 21:49:21< mordante> billynux, no, so I think making it a choice with #ifdefs is the easiest for the transition phase but in the end all SDL code should be removed
20100615 21:49:33< billynux> yes
20100615 21:50:07< billynux> RE read32... that wasn't my question... but, let me look at the code and figure it out myself
20100615 21:50:33-!- Crab_ [~Crab_@wesnoth/developer/crab] has quit [Quit: Leaving.]
20100615 21:51:24< mordante> ok
20100615 21:51:59< mordante> I would leave the SDL_net code where it is and add new files, maybe in a network subdirectory although it might be overkill
20100615 21:52:20< alink> Crab_: one infraction to the last rule : default_ai_context_impl::multistep_move_possible uses temporary_unit_placer which break constness, but it's not a real definitive change
20100615 21:53:50< alink> Crab_: we must work on this timing problem :-p
20100615 21:57:36< billynux> mordante, ok
20100615 22:03:27< mordante> billynux, do have more questions at the moment?
20100615 22:03:45-!- Ken_Oh [~briang@static-71-178-174-220.washdc.fios.verizon.net] has quit [Read error: Connection reset by peer]
20100615 22:03:49< billynux> mordante,  do you have one more minute to look at the test code?
20100615 22:04:08< mordante> billynux, is it already committed?
20100615 22:04:14< billynux> wait for CIA :)
20100615 22:04:31< CIA-86> billynux * r43456 /trunk/src/ana/test/ (4 files): Finished the simple SDL_net server / ana client test.
20100615 22:04:32< billynux> cmake -Dtests=on ....
20100615 22:04:35< mordante> if CIA was reliable ;-P
20100615 22:05:02< billynux> I always trust them to be rummaging through my facebook pics :)
20100615 22:06:02< billynux> mordante, there are a couple of ugly hacks to get things going, but the test works fine
20100615 22:06:14< billynux> damn, sdl_server.cpp doesn't have any header
20100615 22:06:21< mordante> billynux, FYI I get this warning `Warning: Source file "/src/wesnoth/src/ana/src/asio_listener.cpp" is listed multiple times for target "ana"'
20100615 22:07:09< billynux> yes... it is need for both client and server implementations (since they both listen to incoming messages the same way)
20100615 22:07:24< billynux> I guess I should fix the CMake vars adding a common_srcs
20100615 22:08:12< billynux> also... I didn't unify an asio_sender.cpp for both... a bit of an implementation decision, but there is some code repetition due to that
20100615 22:08:58< gabba> fendrin: still around?
20100615 22:10:08< mordante> yes making a common source group (and maybe a static lib for that part) would solve that issue
20100615 22:11:06< billynux> mordante, yes. Any idea what license I should use for the code I stole from http://gpwiki.org/index.php/SDL:Tutorial:Using_SDL_net ?
20100615 22:12:17< mordante> stealing code is a bad idea
20100615 22:12:48< mordante> but in general all code in Wesnoth is GPL v2 (or later)
20100615 22:13:10< mordante> so if the code from that site is compatible with that you can include it else not
20100615 22:13:27< billynux> variable renaming FTW? :) ... ok, I don't want to incur in anything illegal: Content is available under GNU Free Documentation License 1.2.
20100615 22:13:50< billynux> so... put a reference to the original URL and use GPLv2?
20100615 22:14:35< Espreon> Refactoring... is fun.
20100615 22:15:10< Espreon> billynux: As for the variable names, use Translation Party.
20100615 22:15:28< mordante> I'm not sure whether that's allowed, but renaming variables is not a way to avoid copyrights 
20100615 22:15:39< billynux> Espreon, never heard of it... it was a joke about stealing code
20100615 22:15:41< mordante> but why do you want to add that code?
20100615 22:15:57< billynux> yes mordante. I know
20100615 22:16:06< billynux> the wiki page has  Content is available under GNU Free Documentation License 1.2
20100615 22:16:21< Espreon> billynux: http://www.translationparty.com/
20100615 22:16:28< billynux> mordante, I think it is good for the users of ana to see it interacting with a different network library
20100615 22:17:00< billynux> specially for guys porting their current network implementations, which should be the typical case
20100615 22:18:29< billynux> Espreon, :D :D -> Top lazy dog The quick brown fox
20100615 22:18:46< billynux> Top lazy dog... that sounds just like me :)
20100615 22:20:30< Espreon> Heh...
20100615 22:20:43< mordante> I'm too tired to look into that now so need to look at it next time
20100615 22:21:06-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: GO, GET TO THE CHOPPAH!!!]
20100615 22:21:46< billynux> mordante, ok... thanks and sleep tight
20100615 22:22:15< mordante> thanks
20100615 22:22:16< mordante> night
20100615 22:22:32-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20100615 22:30:32-!- scibotic [~scibotic@124-171-15-201.dyn.iinet.net.au] has joined #wesnoth-dev
20100615 22:33:29< CIA-86> billynux * r43457 /trunk/src/ana/ (src/CMakeLists.txt test/sdl_server.cpp): Improved header in sdl_server.cpp and added a common_srcs list in CMakeFile.txt
20100615 22:37:12< billynux> g'night y'all
20100615 22:37:20< CIA-86> gabba * r43458 /trunk/src/ (arrow.cpp arrow.hpp): Arrows: use double instead of float, as per mordante's instructions
20100615 22:37:41< CIA-86> gabba * r43459 /trunk/src/ (3 files in 2 dirs): Whiteboard: start of mechanism to track selected units.
20100615 22:37:41< CIA-86> gabba * r43460 /trunk/src/ (3 files in 2 dirs): Whiteboard: selection mechanism
20100615 22:37:42< CIA-86> gabba * r43461 /trunk/src/whiteboard/ (5 files): Erase unit's previous move when defining a new move
20100615 22:37:48< CIA-86> gabba * r43462 /trunk/src/ (11 files in 2 dirs): Arrows/Whiteboard: big and hopefully final refactoring; got rid of observer pattern and settled on a lifecycle management strategy.
20100615 22:38:52-!- billynux [~billy@wesnoth/developer/billynux] has quit [Quit: Leaving]
20100615 22:46:58-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100615 22:50:14-!- Upth [ogmar@adsl-75-26-202-12.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100615 22:53:36-!- Upthorn [ogmar@adsl-75-26-202-12.dsl.scrm01.sbcglobal.net] has quit [Ping timeout: 248 seconds]
20100615 22:53:41-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100615 22:54:37-!- Greywhind_ [~Greywhind@138.16.59.131] has quit [Quit: leaving]
20100615 23:03:16-!- norbert_ [~norbert@82-171-70-54.ip.telfort.nl] has joined #wesnoth-dev
20100615 23:03:57-!- Blueblazed [~nick@adsl-99-182-54-45.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 245 seconds]
20100615 23:04:47-!- dtiger [~dtiger@dynamic-vpdn-213-184-233-16.telecom.by] has quit [Ping timeout: 260 seconds]
20100615 23:08:09< CIA-86> espreon * r43463 /trunk/ (54 files in 3 dirs): Added missing conversions; ran pofix on everything.
20100615 23:10:50< CIA-86> espreon * r43464 /trunk/data/hardwired/tips.cfg: Added some cola to some source strings.
20100615 23:14:05-!- scibotic [~scibotic@124-171-15-201.dyn.iinet.net.au] has quit [Quit: Leaving]
20100615 23:22:15< CIA-86> alink * r43465 /trunk/src/ai/contexts.hpp: use 'virtual const' instead of 'const virtual'
20100615 23:22:33< CIA-86> alink * r43466 /trunk/src/ (16 files in 6 dirs): Remove unit_map from ai::game_info, and uses resources::units instead
20100615 23:22:35< CIA-86> alink * r43467 /trunk/src/ai/formula/ (callable_objects.cpp callable_objects.hpp function_table.cpp): Remove now useless parameter
20100615 23:22:39< CIA-86> alink * r43468 /trunk/src/ (6 files in 2 dirs): Remove now useless parameter
20100615 23:22:42< CIA-86> alink * r43469 /trunk/src/ (generate_report.cpp menu_events.cpp unit.cpp unit.hpp): Remove now useless parameter
20100615 23:22:47< CIA-86> alink * r43470 /trunk/src/ (14 files in 6 dirs): Remove teams from ai::game_info, and uses resources::teams instead
20100615 23:22:53< CIA-86> alink * r43471 /trunk/src/ (11 files in 5 dirs): Remove map from ai::game_info, and uses resources::game_map instead
20100615 23:23:50< alink> pfff, that was boring, I will finish the remaining fields tomorrow 
20100615 23:24:05< alink> at least it will allow to simplify a lot of things
20100615 23:24:12< alink> anyway, gn
20100615 23:24:16-!- alink [~alink@wesnoth/developer/alink] has quit [Remote host closed the connection]
20100615 23:27:02< Espreon> gabba: http://wesnoth.pastebin.com/pCWbb59c ... Oh noez!
20100615 23:27:52< gabba> Espreon: damn, how comes I didn't get that warning
20100615 23:28:10 * Espreon shrugs
20100615 23:28:24< gabba> type-punned? lol, I think the compiler is making a pun
20100615 23:28:51< gabba> lemme see what I can do
20100615 23:30:01< ABCD> gabba: you probably have an older version of gcc :D
20100615 23:30:16< gabba> Ubuntu karmic, is that old :P ?
20100615 23:30:39< gabba> gcc --version
20100615 23:30:39< gabba> gcc (Ubuntu 4.4.1-4ubuntu9) 4.4.1
20100615 23:31:05< ABCD> ...or maybe not...
20100615 23:31:05< ABCD> :D
20100615 23:31:25< Espreon> Which reminds me, better upgrade to 4.4.4.
20100615 23:31:28< ABCD> but breaking strict-aliasing rules can cause all kinds of issues
20100615 23:31:33< gabba> I have a member variable of the type display*&, it doesn't see to like it
20100615 23:31:37< ABCD> I just upgraded to 4.4.4 here
20100615 23:31:59< shadowmaster> 4.4.4 here too
20100615 23:32:23< gabba> and this is the constructor initialization of said variable: "screen_((display*&) resources::screen)"
20100615 23:32:24< shadowmaster> Espreon: what is "cola"?
20100615 23:32:24< Espreon> ... and eventually, I'll have a complete KDE 4.4.85 installation.
20100615 23:32:33< Espreon> "Colon" is the singular.
20100615 23:32:33< shadowmaster> in the context of r43464, that is.
20100615 23:32:41< ABCD> gabba: what is the type of resources::screen?
20100615 23:32:41< shadowmaster> hm.
20100615 23:32:59-!- thespaceinvader [~chatzilla@wesnoth/artist/thespaceinvader] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539]]
20100615 23:33:04< shadowmaster> I'm sorry, but I don't think I'll be able to say "cola" anyway
20100615 23:33:05< gabba> resources::screen is of type game_display*
20100615 23:33:11< shadowmaster> I'll stick to colons
20100615 23:33:33< shadowmaster> blame Coca-Cola for that
20100615 23:33:45< Espreon> Meh...
20100615 23:33:49< ABCD> I think that's the problem, perhaps...
20100615 23:34:12< gabba> ABDC: what I'd like to do is to keep a reference to resources::screen without copying the pointer (i.e. when it becomes NULL, I need to know it)
20100615 23:34:35 * ABCD has to go...
20100615 23:35:01< gabba> I guess I'll have to typecast it everytime I use it... if *that* works... I want to avoid #include "game_display.hpp" at all costs.
20100615 23:35:15-!- wesbot changed the topic of #wesnoth-dev to: 127 bugs, 283 feature requests, 13 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100615 23:44:35< CIA-86> gabba * r43472 /trunk/src/ (arrow.cpp arrow.hpp): Another way of handling the display pointer, with hopefully better luck
20100615 23:44:45< gabba> Espreon: try compiling with this update
20100615 23:45:03< Espreon> OK.
20100615 23:45:37< shadowmaster> I see that Wesnoth still takes ridiculously long to compile frm scratch
20100615 23:45:57< shadowmaster> and it still wants to be compiled from scratch for some reason
20100615 23:49:03< Espreon> gabba: So far, no explosions.
20100615 23:49:09< gabba> nice
20100615 23:49:11< ABCD> gabba: the problem is that resources::screen is a game_display*, so you can't store a display* in it - which is what could happen if you store it in a display*&
20100615 23:50:35< gabba> ABCD: ok, it makes sense... I think
20100615 23:52:00< gabba> now that I think of it, I could use: display** screen_ = &resources::screen
20100615 23:53:50< ABCD> I think that would be the exact same issue - there isn't much difference between a display*& and a display**
20100615 23:53:56< ABCD> (except for syntax)
20100615 23:56:40< gabba> there has to be a way I can typecast resources::screen before storing the pointer?
20100615 23:57:08< gabba> display** screen_ = &((display*)resources::screen) maybe?
20100615 23:57:23< ABCD> that's still a type-punned pointer
20100615 23:58:07< gabba> TBH I never ran into the "type-punned pointer" expression
20100615 23:58:41< ABCD> maybe "display* const* const screen_" would work
20100615 23:59:00-!- YogiHH [YogiHH@wesnoth/developer/yogihh] has left #wesnoth-dev []
20100615 23:59:19 * ABCD doesn't remember all the issues with type-punned pointers off hand
20100615 23:59:27< gabba> grrrmbl, last resort would be to first cast it as a void*, but I bet gcc will also complain about that one
20100615 23:59:29< ABCD> and I really have to leave this time (/me has a class)
20100615 23:59:40< gabba> thanks for your insights
20100615 23:59:42< ABCD> that's really a problem
--- Log closed Wed Jun 16 00:00:46 2010