--- Log opened Wed Oct 07 00:00:38 2015
20151007 00:05:29-!- ancestral [~ancestral@71-220-62-196.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20151007 00:06:08-!- sfan786 [~sfan786@c-98-239-230-28.hsd1.pa.comcast.net] has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
20151007 00:09:45-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Quit: wedge009]
20151007 00:10:02-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20151007 00:19:20-!- jinblack_ [~jinblack@169-231-98-142.wireless.ucsb.edu] has joined #wesnoth-dev
20151007 00:25:47-!- jinblack_ [~jinblack@169-231-98-142.wireless.ucsb.edu] has quit [Ping timeout: 252 seconds]
20151007 00:30:31-!- sfan786 [~sfan786@c-98-239-230-28.hsd1.pa.comcast.net] has joined #wesnoth-dev
20151007 00:32:02-!- sfan786 [~sfan786@c-98-239-230-28.hsd1.pa.comcast.net] has quit [Client Quit]
20151007 00:41:58-!- jinblack_ [~jinblack@128.111.48.6] has joined #wesnoth-dev
20151007 00:44:43-!- kinow [~kinow@apache/committer/kinow] has quit [Quit: Leaving]
20151007 00:47:48-!- ideuler [~textual@a89-153-77-183.cpe.netcabo.pt] has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
20151007 00:49:35-!- sfan786 [~sfan786@c-98-239-230-28.hsd1.pa.comcast.net] has joined #wesnoth-dev
20151007 00:54:49-!- horrowind [~Icedove@2a02:810a:8b00:1c54:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20151007 00:59:04-!- sfan786 [~sfan786@c-98-239-230-28.hsd1.pa.comcast.net] has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
20151007 01:06:09< iceiceice> shikadibot: seen ancestral?
20151007 01:06:09< shikadibot> iceiceice: The person with the nick ancestral 1h ago they left with the message: Quit: i go nstuf kthxbai
20151007 01:12:23-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20151007 01:25:07-!- Appleman1234 [~Appleman1@KD106161091082.au-net.ne.jp] has quit [Ping timeout: 256 seconds]
20151007 01:33:14-!- irker175 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20151007 01:40:51-!- ancestral [~ancestral@71-220-62-196.mpls.qwest.net] has joined #wesnoth-dev
20151007 01:54:23-!- jinblack_ [~jinblack@128.111.48.6] has quit [Ping timeout: 250 seconds]
20151007 01:58:58-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20151007 02:00:23-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20151007 02:12:46< vultraz> celticminstrel/gfgtdf: would you like to work on a read_file function? apparently, it can be made safe
20151007 02:13:12< celticminstrel> Uh. I'm not sure.
20151007 02:14:50-!- irker429 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20151007 02:14:50< irker429> wesnoth: Celtic Minstrel wesnoth:master af7ba3e52751 / src/ (editor/controller/editor_controller.cpp scripting/game_lua_kernel.cpp unit.hpp): Fix spelling http://git.io/vcAdY
20151007 02:16:23< irker429> wesnoth: Charles Dang wesnoth:master c1e8acd119e5 / src/game_events/action_wml.cpp: Add [replace_map] file= which will load the specified map at runtime http://git.io/vcAdi
20151007 02:16:26< irker429> wesnoth: Charles Dang wesnoth:master 3d06f689d763 / src/ (dialogs.cpp game_initialization/multiplayer_lobby.cpp saved_game.cpp): Rename [scenario] map= key to map_file for consistency with [replace_map] http://git.io/vcAdP
20151007 02:16:29< irker429> wesnoth: Charles Dang wesnoth:master 2dfed2c3e477 / src/game_events/action_wml.cpp: [replace_map]: clarified an error message http://git.io/vcAdX
20151007 02:16:32< irker429> wesnoth: Charles Dang wesnoth:master 0173161b2d72 / src/ (4 files in 3 dirs): Merge pull request #516 from Vultraz/master http://git.io/vcAd1
20151007 02:16:54< vultraz> If we get a lua read_file later I'll use that instead
20151007 02:20:26< vultraz> celticminstrel: good work with the [for] fixes, btw
20151007 02:21:09< celticminstrel> I actually ended up using a half-open range internally, if you didn't notice.
20151007 02:21:51-!- Appleman1234 [~Appleman1@KD111239008063.au-net.ne.jp] has joined #wesnoth-dev
20151007 02:22:17< celticminstrel> I think it simplified the logic considerably, though I didn't completely get the logic working for fully-closed ranges. There's a good reason why eg STL uses half-open ranges, after all.
20151007 02:26:50< celticminstrel> Need to expand [test_condition] to give some useful output for [have_unit] and [have_location] failure.
20151007 02:28:27< vultraz> [have_unit][/have_location] failed because this unit/location doesn't exist?
20151007 02:29:19< celticminstrel> Well, currently it does print out the contents of the [have_unit] or [have_location] (excluding further subtags).
20151007 02:29:57< celticminstrel> But if there's several filter conditions, how to determine which one is the most likely cause of the failure?
20151007 02:37:31< vultraz> You can't
20151007 02:38:03< celticminstrel> If there's an ID, you can.
20151007 02:38:11-!- oldlaptop_ is now known as oldlaptop
20151007 02:38:26< celticminstrel> Otherwise, the best I can think of is rerunning the filter several times with certain conditions removed.
20151007 02:38:34< celticminstrel> It's a best that isn't very good.
20151007 02:39:20< celticminstrel> (Also if there's x,y you may be able to determine the likely cause of the failure, or anything else that can only select only one location/unit.)
20151007 02:49:18-!- MotorMe [~MotorMe@27.189.18.27] has joined #wesnoth-dev
20151007 02:49:47< irker429> wesnoth: Celtic Minstrel wesnoth:master 21e06ef48268 / src/ (scripting/game_lua_kernel.cpp unit_attack_type.hpp): Add accuracy/parry to Lua unit attacks table http://git.io/vcAxi
20151007 02:50:50< celticminstrel> I could still add unit:find_reach() and unit_find_path()
20151007 02:50:56-!- ancestral [~ancestral@71-220-62-196.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20151007 02:51:00< celticminstrel> ^unit:find_path()
20151007 02:51:12< celticminstrel> Those are both fairly complicated functions though, maybe it's better not to.
20151007 02:55:23-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 264 seconds]
20151007 02:57:14-!- ancestral [~ancestral@71-220-62-196.mpls.qwest.net] has joined #wesnoth-dev
20151007 03:02:41< vultraz> Are you adding everything to the lua table?
20151007 03:02:51< celticminstrel> No idea.
20151007 03:13:52-!- jinblack_ [~jinblack@128.111.48.6] has joined #wesnoth-dev
20151007 03:22:21-!- MotorMe [~MotorMe@27.189.18.27] has quit [Ping timeout: 250 seconds]
20151007 03:36:32-!- jinblack_ [~jinblack@128.111.48.6] has quit [Ping timeout: 250 seconds]
20151007 04:05:03-!- MotorMe [~MotorMe@27.189.18.27] has joined #wesnoth-dev
20151007 04:19:39-!- Appleman1234 [~Appleman1@KD111239008063.au-net.ne.jp] has quit [Ping timeout: 240 seconds]
20151007 04:38:59-!- jinblack_ [~jinblack@ip184-187-174-106.sb.sd.cox.net] has joined #wesnoth-dev
20151007 04:40:54-!- Shackra [~Jorge@186.177.2.148] has quit [Ping timeout: 255 seconds]
20151007 04:42:08-!- Kwandulin [~Miranda@p5B008841.dip0.t-ipconnect.de] has joined #wesnoth-dev
20151007 04:51:08-!- ancestral [~ancestral@71-220-62-196.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20151007 04:59:38-!- ancestral [~ancestral@71-220-62-196.mpls.qwest.net] has joined #wesnoth-dev
20151007 05:07:08-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Quit: wedge009]
20151007 05:07:25-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20151007 05:11:08-!- timotei_ [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 272 seconds]
20151007 05:14:03-!- timotei_ [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20151007 05:16:53-!- Appleman1234 [~Appleman1@KD111239021008.au-net.ne.jp] has joined #wesnoth-dev
20151007 05:31:24< irker429> wesnoth: Charles Dang wesnoth:master 2a466bd56775 / src/ (4 files in 3 dirs): Remove handling of [map] as a tag http://git.io/vcxcz
20151007 05:32:07< celticminstrel> Apparently there's a half-implemented max/min-range from fabi, maybe something should be done about that.
20151007 05:35:42< vultraz> Where?
20151007 05:35:53< celticminstrel> In attacks.
20151007 05:35:59< celticminstrel> I dunno if it should be completed or removed.
20151007 05:36:21< celticminstrel> unit_attack_type or whatever.
20151007 05:36:32< celticminstrel> What was the reason for removing [map] again?
20151007 05:37:19-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20151007 05:41:05< vultraz> We were going to switch to it in 1.11 but it was reverted for some reason
20151007 05:41:28< vultraz> And in 1.13 we removed usage= and border_size=
20151007 05:41:49< celticminstrel> Ah? So map data no longer includes those embedded in it?
20151007 05:41:59< vultraz> Correct  
20151007 05:42:16< vultraz> So the irregular wml style of having keys/attributes inside a key is gone
20151007 05:42:24< vultraz> it's just map_data=data
20151007 05:42:39< celticminstrel> Well, that makes sense. Usage should be determined by where it's used - if it's in [scenario] or [repace_map], it's a map, and if it's in [terrain_mask] it's a mask.
20151007 05:42:45< vultraz> Rendering [map] both undocumented and irrelevant 
20151007 05:43:03< celticminstrel> I dunno about border_size, but it's probably easiest if it's just always 1.
20151007 05:43:10< celticminstrel> Even for masks.
20151007 05:43:29< vultraz> I agree
20151007 05:43:37< vultraz> but there's a problem
20151007 05:43:57< vultraz> gamemap::overlay chokes on certain masks if border_size is 1
20151007 05:44:34< celticminstrel> Well, border size of 0 is probably only used for masks anyway, so it could be a key in [terrain_mask[
20151007 05:44:40< celticminstrel> ^[terrain_mask]
20151007 05:44:47< vultraz> I attempted to fix this awhile back with pydsigner but shit only hit the fan since we were doing it wrong
20151007 05:44:58< celticminstrel> And I doubt border_size > 1 ever makes sense.
20151007 05:45:26< vultraz> I can't exactly remember what shadowm's conclusions were
20151007 05:45:41< vultraz> Something about invalid memory access outside of a loop
20151007 05:46:54< vultraz> Optimally we should fix this before 1.13.2 bc LoW is broken ATM
20151007 05:47:30< celticminstrel> Ah...
20151007 05:48:36< vultraz> Want to try to help me fix it?
20151007 05:48:47< celticminstrel> I could look at it.
20151007 05:52:28< vultraz> The basics are this: if you set border_size to 1 in terrain_mask, LoW's maps get realigned correctly
20151007 05:52:38< vultraz> However, single-hex masks will no longer work
20151007 05:53:01< celticminstrel> ...single-hex masks? o.O
20151007 05:53:20< vultraz> Also, then if you set [terrain_mask] border=true, you get an invalid access error (IIRC)
20151007 05:53:29< vultraz> celticminstrel: certain unit tests use that for some reason
20151007 05:53:59< vultraz> Still, it means there would then be a 'minimum' map size
20151007 05:54:08< celticminstrel> Well, I suppose it sorta makes sense if you want to add a terrain overlay to a single hex...
20151007 05:54:09< vultraz> s/map/mask
20151007 05:54:22< vultraz> celticminstrel: you can do that with [terrain] layer=overlay
20151007 05:54:30< celticminstrel> Ah, okay.
20151007 05:54:41< celticminstrel> Then maybe those unit tests should be changed to use that instead.
20151007 05:54:47< vultraz> Yes
20151007 05:58:03< vultraz> Though actually those tests dont need overlay at all
20151007 05:58:06< vultraz> just terrain
20151007 06:00:40-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20151007 06:01:05< irker429> wesnoth: Charles Dang wesnoth:master 798c59c7bdce / data/test/scenarios/ (feeding.cfg test_relative_dir.cfg test_unit_map.cfg): Replace most instances of a single-hex [terrain_mask] action with [terrain] http://git.io/vcx8l
20151007 06:02:18< celticminstrel> vultraz: Missed some in test_victory_attacks.cfg
20151007 06:04:44< shadowm> Sorry I haven't gotten around to it, but it would be nice if someone added dedicated [terrain_mask] tests instead.
20151007 06:05:16< shadowm> It would also be nice if the behavior tested was the one from 1.12 assuming any changes between 1.12 and master aren't intentional.
20151007 06:05:21< vultraz> celticminstrel: deliberate 
20151007 06:05:31< celticminstrel> Having [terrain_mask] tests I guess would make it easier to fix this issue.
20151007 06:05:37< vultraz> Until we have a dedicated test
20151007 06:05:59< celticminstrel> Ah, so you left that in so that there's still at least one test that uses [terrain_mask]?
20151007 06:06:08< vultraz> yes
20151007 06:13:05< vultraz> And yes, we should check against the 1.12 behavior
20151007 06:13:19< vultraz> in particular there's a case in AtS that for the love of god I cannot make act the same in 1.13
20151007 06:13:22< vultraz> no matter what I changed >_>
20151007 06:15:25< celticminstrel> Given that usage and border_size have been removed, everyone will have to update their scenarios anyway...
20151007 06:15:39< vultraz> Not really
20151007 06:15:41< celticminstrel> What's different about the behaviour in 1.13?
20151007 06:15:44< vultraz> those keys didn't really do anything
20151007 06:15:48< vultraz> except in masks, it seems..
20151007 06:15:53< vultraz> celticminstrel: look at LoW s1 in master
20151007 06:15:55< vultraz> everything is offset
20151007 06:16:08< vultraz> passing border_size as 1 in terrain_mask fixes that but screws up single hex masks
20151007 06:16:13< celticminstrel> Why can't you just update the x,y so that it's not offset?
20151007 06:16:18< vultraz> and then for some reason border=yes causes memory problems in gamemap::overlay
20151007 06:16:24< vultraz> (be back in a bit)
20151007 06:16:41< celticminstrel> Or remove the first and last row and column?
20151007 06:16:52< vultraz> uh...
20151007 06:16:53< vultraz> hm..
20151007 06:17:06< vultraz> celticminstrel: well it's not uniform
20151007 06:17:13< vultraz> you need to seee it
20151007 06:17:16< vultraz> see
20151007 06:17:24< celticminstrel> Maybe tomorrow.
20151007 06:17:30< celticminstrel> Sleep time.
20151007 06:19:32-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20151007 06:19:49-!- celticminstrel is now known as celmin|sleep
20151007 06:49:30-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20151007 06:52:13-!- travis-ci [~travis-ci@ec2-54-157-220-110.compute-1.amazonaws.com] has joined #wesnoth-dev
20151007 06:52:14< travis-ci> wesnoth/wesnoth#7585 (master - 798c59c : Charles Dang): The build has errored.
20151007 06:52:14< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/84033428
20151007 06:52:14-!- travis-ci [~travis-ci@ec2-54-157-220-110.compute-1.amazonaws.com] has left #wesnoth-dev []
20151007 07:28:59-!- MotorMe [~MotorMe@27.189.18.27] has quit [Ping timeout: 264 seconds]
20151007 07:41:44< vultraz> translation y u fail
20151007 07:45:51-!- jinblack_ [~jinblack@ip184-187-174-106.sb.sd.cox.net] has quit [Ping timeout: 255 seconds]
20151007 08:03:24-!- Coffee_irc [~david@ppp118-210-17-216.lns20.adl2.internode.on.net] has joined #wesnoth-dev
20151007 08:04:03-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Quit: No Ping reply in 180 seconds.]
20151007 08:05:30-!- fabi [~quassel@176.6.10.9] has joined #wesnoth-dev
20151007 08:05:30-!- fabi [~quassel@176.6.10.9] has quit [Changing host]
20151007 08:05:30-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20151007 08:05:36-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20151007 08:11:03-!- ancestral [~ancestral@71-220-62-196.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20151007 08:11:09-!- roberpot [~roberpot@84.122.85.148.dyn.user.ono.com] has joined #wesnoth-dev
20151007 08:15:44-!- ancestral [~ancestral@71-220-62-196.mpls.qwest.net] has joined #wesnoth-dev
20151007 08:34:50< Aginor> 'lo
20151007 08:35:12< Aginor> so this SDL2 resize bug has had me looking at SDL1 and SDL2 resiz issues
20151007 08:35:47< Aginor> which has led me in a merry chase through the SDL library, and now it's leading me into xlib
20151007 08:36:16< Aginor> wedge009: would you be able to help with some testing for windows if I go and make a diff?
20151007 08:36:34< Aginor> I'd like to see if the problem persists on a non-X11 system
20151007 08:38:58-!- mjs-de [~mjs-de@x4db52b06.dyn.telefonica.de] has joined #wesnoth-dev
20151007 08:40:47-!- ypnos [~ypnos@boron.physiol.uni-freiburg.de] has joined #wesnoth-dev
20151007 08:43:16-!- ancestral [~ancestral@71-220-62-196.mpls.qwest.net] has quit [Quit: Smell ya later]
20151007 09:07:22-!- roberpot [~roberpot@84.122.85.148.dyn.user.ono.com] has quit [Read error: No route to host]
20151007 09:08:22-!- roberpot [~roberpot@84.122.85.148.dyn.user.ono.com] has joined #wesnoth-dev
20151007 09:25:27< irker429> wesnoth: Charles Dang wesnoth:master abd5dfdcc063 / src/ (game_events/action_wml.cpp map.cpp map.hpp): Remove handling of maps' usage= key http://git.io/vcpmE
20151007 09:26:32-!- shadowm_desktopW [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 244 seconds]
20151007 10:03:44-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Ping timeout: 272 seconds]
20151007 10:06:35-!- zombah [~zombah@2a02:28:3:1:214:4fff:fe47:5920] has joined #wesnoth-dev
20151007 10:10:28-!- travis-ci [~travis-ci@ec2-184-72-153-75.compute-1.amazonaws.com] has joined #wesnoth-dev
20151007 10:10:29< travis-ci> wesnoth/wesnoth#7586 (master - abd5dfd : Charles Dang): The build passed.
20151007 10:10:29< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/84055812
20151007 10:10:29-!- travis-ci [~travis-ci@ec2-184-72-153-75.compute-1.amazonaws.com] has left #wesnoth-dev []
20151007 10:50:05< wedge009> Aginor: Sure, perhaps tomorrow. I'm sure it's late for you already as well
20151007 10:50:24-!- ypnos [~ypnos@boron.physiol.uni-freiburg.de] has quit [Ping timeout: 246 seconds]
20151007 10:59:51< Aginor> wedge009: yes :)
20151007 11:13:22-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20151007 11:21:25-!- MotorMe [~MotorMe@27.189.240.209] has joined #wesnoth-dev
20151007 11:21:58-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20151007 11:24:42-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20151007 11:37:35-!- ypnos [~ypnos@boron.physiol.uni-freiburg.de] has joined #wesnoth-dev
20151007 11:38:37-!- Appleman1234 [~Appleman1@KD111239021008.au-net.ne.jp] has quit [Ping timeout: 252 seconds]
20151007 11:42:13-!- Kwandulin [~Miranda@p5B008841.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20151007 11:51:03-!- joet [~joet@host86-163-222-44.range86-163.btcentralplus.com] has joined #wesnoth-dev
20151007 12:13:36-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Quit: wedge009]
20151007 12:13:58-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20151007 12:23:36-!- destr [5862aeb2@gateway/web/freenode/ip.88.98.174.178] has joined #wesnoth-dev
20151007 12:23:56-!- destr [5862aeb2@gateway/web/freenode/ip.88.98.174.178] has quit [Client Quit]
20151007 12:25:38-!- irker429 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20151007 12:30:25-!- roberpot [~roberpot@84.122.85.148.dyn.user.ono.com] has quit [Quit: • IRcap • 8.72 •]
20151007 12:30:38-!- MotorMe [~MotorMe@27.189.240.209] has quit [Remote host closed the connection]
20151007 12:30:59-!- MotorMe [~MotorMe@27.189.240.209] has joined #wesnoth-dev
20151007 12:36:11-!- Appleman1234 [~Appleman1@KD106161089229.au-net.ne.jp] has joined #wesnoth-dev
20151007 12:45:40-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Quit: wedge009]
20151007 12:46:01-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20151007 12:46:05-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Client Quit]
20151007 12:54:31-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20151007 13:03:45-!- Coffee_irc [~david@ppp118-210-17-216.lns20.adl2.internode.on.net] has quit [Ping timeout: 240 seconds]
20151007 13:11:46-!- Kwandulin [~Miranda@p5B008841.dip0.t-ipconnect.de] has joined #wesnoth-dev
20151007 13:24:44-!- louis94 [~~louis94@42.133-243-81.adsl-dyn.isp.belgacom.be] has joined #wesnoth-dev
20151007 13:33:44-!- Appleman1234 [~Appleman1@KD106161089229.au-net.ne.jp] has quit [Ping timeout: 255 seconds]
20151007 13:38:39-!- louis94 [~~louis94@42.133-243-81.adsl-dyn.isp.belgacom.be] has quit [Ping timeout: 255 seconds]
20151007 13:46:04-!- Shackra [~Jorge@186.177.2.148] has joined #wesnoth-dev
20151007 13:52:40 * celmin|sleep thought Aginor was working on OSX?
20151007 13:53:25< vultraz> I think that's the point
20151007 13:53:29< vultraz> he wants wedge009 to test on windows
20151007 13:53:43< celmin|sleep> OSX is not an xlib system.
20151007 13:53:53< celmin|sleep> X11 system, sorry.
20151007 13:54:34< celmin|sleep> Well okay, it has X11 available, but Cocoa is not related to it.
20151007 14:05:58-!- celmin|sleep is now known as celticminstrel
20151007 14:29:57< celticminstrel> Why do I get all these warnings about unused attributes being discarded?
20151007 14:30:04< celticminstrel> ^unknown
20151007 14:30:38< celticminstrel> In LoW I get warnings for no_recruit, flag, flag_icon, team_name, user_team_name, village_gold, controller
20151007 14:31:01< celticminstrel> ...most of which sound like they should be known, unless I'm missing something.
20151007 14:31:54 * vultraz doesn't know
20151007 14:32:19< celticminstrel> no_recruit is the only attribute I've never actually head of on that list.
20151007 14:32:58< celticminstrel> Based on the coordinates, the mask is offset by 1 on each axis.
20151007 14:33:25< celticminstrel> The dirt is at (2,3) in 1.12 but at (3,4) in 1.13.
20151007 14:34:01-!- roberpot [~roberpot@84.122.85.148.dyn.user.ono.com] has joined #wesnoth-dev
20151007 14:34:24< celticminstrel> (The dirt hex on the north edge of the northwest camp.)
20151007 14:35:07< celticminstrel> Also in 1.13 the orcs already own the villages near them, but I guess that was probably an intentional change.
20151007 14:35:23< celticminstrel> Wait, no.
20151007 14:35:26< vultraz> As you can see, the leaders are not in their keeps
20151007 14:35:32< vultraz> it's more than 1,1 offset
20151007 14:35:44< vultraz> depending on where you look on the map
20151007 14:35:50< celticminstrel> Oh, that's just an artifact of the different team colours.
20151007 14:36:00< celticminstrel> Were the team colours intentionally changed?
20151007 14:36:32< vultraz> Eh?
20151007 14:36:56< celticminstrel> Velon is blue in 1.13 and teal in 1.12.
20151007 14:37:10< celticminstrel> The orcs all have different colours to.
20151007 14:37:12< celticminstrel> ^too
20151007 14:37:26< vultraz> .....huh
20151007 14:37:49< celticminstrel> And it's not more than 1,1 offset. The southern two orcs might look like they're further away than the northern two orcs, but if you look at the actual coordinates, they're still off by exactly one.
20151007 14:38:31< vultraz> oh
20151007 14:38:33< vultraz> huh
20151007 14:38:40< celticminstrel> Wait, that's green, not teal. Sorry.
20151007 14:39:30< celticminstrel> Still different though.
20151007 14:39:37< vultraz> celticminstrel: quick OT question.. what type would a function argument be if you wanted to pass an enum value?
20151007 14:39:46< celticminstrel> The enum type.
20151007 14:40:18< celticminstrel> Unless I'm missing something.
20151007 14:41:50< vultraz> the case is here https://github.com/Vultraz/wesnoth/commit/42bc93eec92bfa535b34a06d11c3aebd4e3f9bc1
20151007 14:42:39< celticminstrel> The map and the mask in LoW are exactly the same size, according to the map editor.
20151007 14:42:40< vultraz> I'm trying to pass an argument fromeither team::CONTROLLER or team::SHARE_VISION to register_radio_toggle
20151007 14:45:20-!- gfgtdf [~chatzilla@f054134217.adsl.alicedsl.de] has joined #wesnoth-dev
20151007 14:45:26< celticminstrel> Were those defined with MAKE_ENUM?
20151007 14:46:20-!- Appleman1234 [~Appleman1@KD118156252235.au-net.ne.jp] has joined #wesnoth-dev
20151007 14:48:17< celticminstrel> If I change the mask to overlay at 0,0 instead of 1,1, the northern leaders are now on their keeps, but the southern ones still aren't. :|
20151007 14:49:26-!- jinblack_ [~jinblack@ip184-187-174-107.sb.sd.cox.net] has joined #wesnoth-dev
20151007 14:49:34< celticminstrel> The keeps themselves are quite different in 1.13, but I assume that's intentional.
20151007 14:50:43< vultraz> celticminstrel: if you pass border_size = 1 from within terrain_mask, the keeps become identical to 1.12
20151007 14:50:48< vultraz> celticminstrel: and yes, MAKE_ENUM
20151007 14:51:01< celticminstrel> So the different keeps are not intentional?
20151007 14:51:10< vultraz> AFAIK
20151007 14:51:31< vultraz> Now you see why this puzzles me so
20151007 14:51:38< gfgtdf> you are working on LoW in 1.13 ?
20151007 14:52:07< vultraz> gfgtdf: no he's investigating the terrain_mask issue
20151007 14:52:19< vultraz> gfgtdf: I'm trying to make editor edit side use the tristate
20151007 14:52:27< celticminstrel> So the keep edges are shifted down one hex while the centre is not.
20151007 14:52:55< gfgtdf> vultraz: you are using tristate buttons ot radio buttons ?
20151007 14:53:05< vultraz> gfgtdf: radio like the controller
20151007 14:53:26< vultraz> gfgtdf: do we have tristate?
20151007 14:53:51< celticminstrel> On the east-middle keep it's the centre that's shifted down one hex.
20151007 14:54:16< celticminstrel> And on the south, the centre is shifted up one.
20151007 14:54:23< gfgtdf> vultraz: yes but we dont have a fitting defainition for that usecase yet.
20151007 14:57:52< vultraz> I'm having a hard time refactoring these functions for more general use
20151007 14:58:12< celticminstrel> [Oct 07@10:50:43am] vultraz: celticminstrel: if you pass border_size = 1 from within terrain_mask, the keeps become identical to 1.12
20151007 14:58:13< celticminstrel> How do I do that?
20151007 14:58:19-!- MotorMe [~MotorMe@27.189.240.209] has quit [Ping timeout: 240 seconds]
20151007 14:58:37< celticminstrel> A key in the mask action?
20151007 14:59:24< vultraz> celticminstrel: game_events/action_wml.cpp:1186
20151007 14:59:24< celticminstrel> Oh, you mean I need to edit the source code?
20151007 14:59:24-!- Kwandulin [~Miranda@p5B008841.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20151007 14:59:26< vultraz> yes
20151007 14:59:26< gfgtdf> vultraz: y i saw your commit, your problme is that for the vision thignk your need a register_radio_toggle function that takes a team::SHARE_VISION as parameter and for the controller thing you need a function taht takes a team::CONTROLLER as parameter. The best solution woudl be to make register_radio_toggle a template
20151007 14:59:36< celticminstrel> ...or I could enter the debugger and edit the memory directly so that I don't have to recompile. >_>
20151007 14:59:49< vultraz> gfgtdf: a template?
20151007 14:59:56< vultraz> I've never used those 
20151007 15:00:00 * vultraz googles
20151007 15:00:13< celticminstrel> Templates are complicated.
20151007 15:00:56< celticminstrel> And I don't know if using a template would automagically fix the problem. It really depends on how register_radio_toggle actually uses that parameter.
20151007 15:01:17< vultraz> oh geez, templates are complicated D:
20151007 15:01:34< gfgtdf> celticminstrel: actuall  dinth think templates are comlicated
20151007 15:01:50< celticminstrel> They can be.
20151007 15:02:20< vultraz> celticminstrel: basically it sets the radio button to true or false if controller_/share_vision_ == the enum value
20151007 15:02:48< vultraz> and then pushes it to a pair value in a vector
20151007 15:03:02< celticminstrel> Do you happen to know where register_radio_toggle is defined?
20151007 15:03:13< vultraz> uh
20151007 15:03:16< vultraz> right there in the file
20151007 15:03:28< celticminstrel> Same file you linked?
20151007 15:03:47< vultraz> gui/dialogs/editor/editor_edit_side.cpp:100 yes
20151007 15:03:55< vultraz> or 78 in the header
20151007 15:05:15< celticminstrel> Just for future reference, "defined" refers to the place where the function body is; in the header it's only "declared". :)
20151007 15:05:24 * celticminstrel found it.
20151007 15:06:43< celticminstrel> So your problem is that register_radio_toggle is currently set up for controller, and you also want to use it for vision.
20151007 15:06:55< vultraz> yes
20151007 15:06:57< vultraz> for both
20151007 15:07:00< vultraz> without duplicating it
20151007 15:07:35< celticminstrel> Looking at the implementation, that'll require several more parameters.
20151007 15:07:42< celticminstrel> I think.
20151007 15:07:47< celticminstrel> At least two?
20151007 15:08:22< celticminstrel> And gfgtdf is right, you need a template here.
20151007 15:08:40< vultraz> D:
20151007 15:08:44< vultraz> how dost I template
20151007 15:09:49< celticminstrel> controller_, toggle_controller_callback, and controller_tgroup_ probably need to be turned into parameters, too.
20151007 15:10:38< gfgtdf> i wouldnt put toggle_controller_callback into parameters, instead i'd make toggle_controller_callback a template and forward some if the parmeaters to it
20151007 15:10:52< celticminstrel> True, you could do that. I was even about to mention it myself.
20151007 15:11:12< vultraz> I have no idea how to do any of that
20151007 15:12:17< celticminstrel> To make it a template, add "template<typename Type>" right above the function declaration (and implementation).
20151007 15:12:51< celticminstrel> You can replace Type with whatever name you like.
20151007 15:13:18< celticminstrel> Then Type is a new type that can be used anywhere in the function to refer to whatever type actually gets passed as the second parameter (assuming it's used to declare the second parameter).
20151007 15:14:45< vultraz> Above the implementation?
20151007 15:14:49< vultraz> Huh
20151007 15:15:00< celticminstrel> Right above "void your_function" or whatever.
20151007 15:15:28< vultraz> Yes I know
20151007 15:16:10< vultraz> So then I can make value
20151007 15:16:12< vultraz> T value?
20151007 15:16:18< celticminstrel> Yes.
20151007 15:17:56< vultraz> So can T be used... for multiple arguments?
20151007 15:17:59< vultraz> or what
20151007 15:18:11< vultraz> I'm confused as to how this is more than a 'generic type'
20151007 15:18:41< iceiceice> vultraz: you can think of it as a generic type if you like
20151007 15:18:55< iceiceice> if you want to really understand what happens you need to have a good understanding of linkage in C++
20151007 15:19:13< iceiceice> basically the templates are somewhere in between macros and regular C++ code
20151007 15:19:20< celticminstrel> T can be used exactly like any other type.
20151007 15:19:24< iceiceice> the template itself is not a function and cannot be called
20151007 15:19:36< iceiceice> but when you instantiate it, the compiler generates a corresponding function
20151007 15:19:38< celticminstrel> I wouldn't compare them to macros normally, but I suppose you could do so.
20151007 15:19:40< iceiceice> where T has been deduced with some actual type
20151007 15:19:48< iceiceice> celticminstrel: they are a form of metaprogramming
20151007 15:19:51< iceiceice> code generation
20151007 15:20:08< iceiceice> the C++ preferred way of doing it i guess
20151007 15:20:30< celticminstrel> vultraz: When I force border_size to 1 I get an error that it's not 0.
20151007 15:20:34< iceiceice> vultraz: if you want to udnerstand templates in detail the way to learn is to read about it, not over irc
20151007 15:20:42< iceiceice> but in general yeah it can be thought of as a generic type, sure
20151007 15:21:18< vultraz> celticminstrel: oh, yeah, you have to disable the assert
20151007 15:21:46< celticminstrel> Where?
20151007 15:22:11< vultraz> celticminstrel: what does the message say? I removed usage handling earlier so it shouldn't still be checking...
20151007 15:22:21-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20151007 15:22:31< celticminstrel> 'border_size' should be '0' when 'usage = mask'
20151007 15:23:02< celticminstrel> Maybe I rebuilt before that change, I'll try rebuilding first.
20151007 15:23:13< celticminstrel> ^built
20151007 15:23:30-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 255 seconds]
20151007 15:24:15-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20151007 15:25:34< vultraz> celticminstrel: so I'm reading and it says you can have function templates
20151007 15:25:54< celticminstrel> Yes.
20151007 15:26:10< vultraz> are you saying I should make register_radio_toggle a template?
20151007 15:26:17< celticminstrel> Yes.
20151007 15:26:49< celticminstrel> Normally you put template implementations in the header file, but it's a private function so it should be fine to put it in the source file.
20151007 15:27:42< celticminstrel> You should probably also make the other private function a template (the toggle callback function).
20151007 15:27:54< vultraz> sooo.... template <typename T> inline T void register_radio_toggle(twindow& window, const std::string& toggle_id, T value, morestuffheremaybe) ?
20151007 15:28:13< celticminstrel> You don't need the inline unless it's in the header, which I also don't think is necessary.
20151007 15:28:23< celticminstrel> But otherwise, yes.
20151007 15:28:36< celticminstrel> Though I thought you removed the toggle_id parameter.
20151007 15:28:55< vultraz> no, that's the widget to find
20151007 15:29:05< vultraz> (keep in mind I'm getting all this from google, lol)
20151007 15:29:14< celticminstrel> Oh right, you didn't remove it, just changed it. Never mind.
20151007 15:29:42< celticminstrel> But yeah, don't worry about inline unless you're moving the whole thing to the header file, which isn't really needed since it's private.
20151007 15:30:32< vultraz> don't I need the 'inline' for the declaration?
20151007 15:30:38< celticminstrel> No.
20151007 15:30:49< vultraz> oh
20151007 15:31:17< celticminstrel> The only think "inline" actually does is tell the compiler to make sure there isn't multiple copies of the function if the header is included in several places.
20151007 15:31:21< celticminstrel> ^thing
20151007 15:31:32< celticminstrel> It's only needed if the whole definition is in the header.
20151007 15:31:45-!- sfan786 [~sfan786@c-24-131-93-63.hsd1.pa.comcast.net] has joined #wesnoth-dev
20151007 15:31:49< celticminstrel> It has nothing to do with templates, though is often used with them because they're often defined in the header.
20151007 15:32:18< vultraz> while definition as in the implementation too?
20151007 15:32:44< vultraz> whole*
20151007 15:32:55< celticminstrel> Yes.
20151007 15:33:10< vultraz> C:\Users\Charles\Documents\wesnoth-git\src\gui\dialogs\editor\editor_edit_side.cpp|101|error: two or more data types in declaration of 'register_radio_toggle'|
20151007 15:33:15< vultraz> eh/
20151007 15:33:30< celticminstrel> I would need to see the code to understand the error.
20151007 15:33:55< vultraz> template <typename T>
20151007 15:33:56< vultraz> T void teditor_edit_side::register_radio_toggle(twindow& window, const std::string& toggle_id, T value)
20151007 15:34:04< vultraz> do I not need this T?
20151007 15:34:11 * vultraz is so confused
20151007 15:34:12< celticminstrel> There's an extraneous T before void.
20151007 15:34:55< celticminstrel> Okay, so yes, using a border size of 1 fixes the mask issue.
20151007 15:35:21< vultraz> now same error point to the deceleration :( (code the same)
20151007 15:35:28< vultraz> do I not need a T before void there either?
20151007 15:35:43< celticminstrel> No. Why would you need a T there?
20151007 15:36:05< celticminstrel> Are there other masks for which using a border size of 1 would break them?
20151007 15:36:08< vultraz> google is misleading me
20151007 15:36:20< vultraz> celticminstrel: single-hex masks, I believe
20151007 15:36:32< celticminstrel> Oh, the T was there in your first example.
20151007 15:36:52< celticminstrel> Besides single-hex masks.
20151007 15:37:01< vultraz> I don't think so
20151007 15:37:26< vultraz> though there is that weird mask in ats with a deliberate border_size = 1 in its file that acts really screwy in 1.13
20151007 15:38:35< celticminstrel> The masks in my campaign all have border_size=1 as well.
20151007 15:38:52< celticminstrel> As well as usage=map.
20151007 15:38:59< vultraz> map usage for masks? o_O
20151007 15:39:13< celticminstrel> Maybe that just proves that usage didn't do anything. :P
20151007 15:39:42< celticminstrel> All but one of my masks were created by simply editing a base mask to replace most of it with shroud.
20151007 15:39:43< vultraz> well usage did used to be used to validate border_size...
20151007 15:39:45< vultraz> but..
20151007 15:39:46< celticminstrel> ^base map
20151007 15:40:19< vultraz> er
20151007 15:40:25< vultraz> wait, that mask in AtS has a border_size of 0
20151007 15:40:27< vultraz> the fuck...
20151007 15:40:31< vultraz> oh I guess uh
20151007 15:40:42< vultraz> changing it to 1 would screw up masks with explicit 0?
20151007 15:40:44< vultraz> gah
20151007 15:41:12< vultraz> celticminstrel: your masks work bc you had usage=map :P
20151007 15:41:17< vultraz> usage = map allows b_s = 1
20151007 15:41:46 * celticminstrel looks at the LoW masks... turns out they're the same as mine.
20151007 15:42:19< celticminstrel> All of them have border_size=1 and usage=map.
20151007 15:42:27< vultraz> uhh
20151007 15:42:32< vultraz> are you sure you're in master o_O
20151007 15:42:44< celticminstrel> That's not in master.
20151007 15:42:51< celticminstrel> That's in my 1.12 installation.
20151007 15:42:55< vultraz> oh
20151007 15:42:57< vultraz> ok
20151007 15:43:11< vultraz> so the masks work because of the 1 border size
20151007 15:43:20< vultraz> explains why 0 breaks them
20151007 15:43:38< vultraz> but changing it to 1 would also break masks that are 0 :/
20151007 15:43:41< celticminstrel> Looks like all but one of AtS's masks is also the same, border_size=1 and usage=map.
20151007 15:43:53< vultraz> I had a proposal
20151007 15:43:56< celticminstrel> One is border_size=0 and usage=mask.
20151007 15:44:11< vultraz> that being that the existing terrain_mask border=bool key be re purposed 
20151007 15:44:15< vultraz> if true, b_s = 1
20151007 15:44:18< vultraz> (default true)
20151007 15:44:20< vultraz> if false, 0
20151007 15:44:53< vultraz> however... gamemap::overlay has this line
20151007 15:44:54< vultraz> int actual_border = (m.border_size() == border_size()) && border ? border_size() : 0;
20151007 15:45:19-!- Appleman1234 [~Appleman1@KD118156252235.au-net.ne.jp] has quit [Ping timeout: 246 seconds]
20151007 15:45:20< celticminstrel> What about that line?
20151007 15:45:25< vultraz> if it's ever 1 you get errors
20151007 15:45:31< vultraz> ie, if it's ever true
20151007 15:45:46< celticminstrel> If border is true?
20151007 15:45:56< vultraz> yes
20151007 15:46:14< vultraz> except, it can't be, because m.border_size() == border_size() won't pass... for some reason...
20151007 15:46:38< vultraz> if you change it to just int actual_border = border ? border_size() : 0;
20151007 15:46:41< vultraz> you will get errors
20151007 15:46:55< vultraz> at *runtime*, mind you
20151007 15:47:03 * celticminstrel changes it to int actual_border = border
20151007 15:47:04< vultraz> not compile-time
20151007 15:47:18< vultraz> celticminstrel: uh... int = bool?
20151007 15:47:28< celticminstrel> Can border_size() ever be >1?
20151007 15:48:42< vultraz> border_size() returns border_size_ which is initialized in the constructor as gamemap::SINGLE_TILE_BORDER which is equal to ... nothing
20151007 15:48:48< vultraz> 	enum tborder {
20151007 15:48:49< vultraz> 		NO_BORDER = 0,
20151007 15:48:51< vultraz> 		SINGLE_TILE_BORDER
20151007 15:48:52< vultraz> 		};
20151007 15:49:03< celticminstrel> So that means SINGLE_TILE_BORDER is 1.
20151007 15:49:12< vultraz> oh?
20151007 15:49:19< celticminstrel> Because enum constants with no definition are always the previous one plus one.
20151007 15:49:24< vultraz> oh, i see
20151007 15:50:03< celticminstrel> And "int_val = bool_val" is the same as "int_val = bool_val ? 1 : 0"
20151007 15:50:50< vultraz> oh, is that because true/false = 0/1?
20151007 15:51:00< celticminstrel> Yeah.
20151007 15:51:26< vultraz> ahh
20151007 15:51:28< vultraz> ok
20151007 15:51:30< vultraz> makes sense
20151007 15:52:17< vultraz> celticminstrel: to get the error, you just need to have a terrain_mask with border=true (recommend using LoW/utils/map_utils.cfg) as a case and loading LoW
20151007 15:52:37< celticminstrel> Trying that now.
20151007 15:53:26< celticminstrel> std::bad_alloc? o.O
20151007 15:53:35< vultraz> yeah, not invalid_access
20151007 15:53:38< vultraz> sorry
20151007 15:53:48< vultraz> shadowm did extensive debugging for that
20151007 15:53:55< vultraz> took him a few hours, poor him :P
20151007 15:54:05< vultraz> concluded it was to do with index access in gamemap::overlay
20151007 15:54:11 * celticminstrel enables the __cxa_throw breakpoint.
20151007 15:55:28< celticminstrel> ...apparently GUI2 misuses exceptions. :(
20151007 15:55:51< vultraz> *wesnoth* overuses exceptions 
20151007 15:56:10< celticminstrel> I've noticed that a little, yeah.
20151007 15:56:38< vultraz> A lot could probably be handled with bools
20151007 15:56:45< vultraz> like what I did to the savegame loading code awhile back
20151007 15:57:10< vultraz> changed it from void functions throwing empty structs to bools checked in the call locations
20151007 15:57:33< celticminstrel> Okay, found it, but had to disable breakpoints until the story screen came up because all of the enums throw exceptions.
20151007 15:57:53< vultraz> ....why would THEY throw exceptions O_O
20151007 15:58:14< celticminstrel> Attempting to convert a value that's not a valid enum value, presumably.
20151007 15:58:29< vultraz> #justwesnoththings
20151007 15:59:01< celticminstrel> This isn't Twitter. :P
20151007 15:59:29< celticminstrel> I think I see why this error is occurring.
20151007 15:59:33< vultraz> oh?
20151007 15:59:34< celticminstrel> Border size is added twice.
20151007 15:59:47< celticminstrel> Once in overlay, once in the overloaded index operator.
20151007 15:59:53< celticminstrel> Wait.
20151007 16:00:06< celticminstrel> No, that's wrong.
20151007 16:00:15-!- ypnos [~ypnos@boron.physiol.uni-freiburg.de] has quit [Ping timeout: 250 seconds]
20151007 16:00:25< celticminstrel> There's two index operators called in a row, and I think I'm in the first one, not the second.
20151007 16:00:50-!- sfan786 [~sfan786@c-24-131-93-63.hsd1.pa.comcast.net] has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
20151007 16:01:31< celticminstrel> Okay, so the problem is that border_size is 0 and thus it's trying to access row -1.
20151007 16:03:07< vultraz> but border_size is 1
20151007 16:03:54< celticminstrel> That's what I thought too.
20151007 16:04:16< vultraz> but it's... not?
20151007 16:08:18< celticminstrel> No. I think I know why, too.
20151007 16:15:50-!- roberpot [~roberpot@84.122.85.148.dyn.user.ono.com] has quit [Read error: Connection reset by peer]
20151007 16:17:02-!- roberpot [~roberpot@84.122.85.148.dyn.user.ono.com] has joined #wesnoth-dev
20151007 16:20:31< celticminstrel> Okay, so the goal is for [terrain_mask]border=yes to use a border size of 1, while [terrain_mask]border=no uses a border size of 0, correct?
20151007 16:20:46< vultraz> Correct
20151007 16:21:04< vultraz> Since terrain masks are the only thing that would require a variable border size
20151007 16:24:59< celticminstrel> Does anything rely on using the border_size/data keys in [terrain_mask]?
20151007 16:25:04< celticminstrel> (Rather than using mask=)
20151007 16:25:12< celticminstrel> Can I just remove them?
20151007 16:27:00< vultraz> yes, remove them
20151007 16:27:25< vultraz> they aren't even documented 
20151007 16:27:35< celticminstrel> Nor are old_not and new_not... :O
20151007 16:28:10< vultraz> uh, where is THAT from O_O
20151007 16:28:17< celticminstrel> Terrain mask rules.
20151007 16:28:21< vultraz> Oh
20151007 16:28:25< vultraz> Ok, then remove that too
20151007 16:28:26< celticminstrel> map.cpp, line 340
20151007 16:28:33< vultraz> yay for housecleaning
20151007 16:28:41< celticminstrel> Why?
20151007 16:29:05< celticminstrel> ...I suppose the same affect is achievable with something like old=!,...
20151007 16:29:34< gfgtdf> vultraz, zookeeper: ayny opionion on http://gna.org/bugs/?23490 ?
20151007 16:30:12-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has joined #wesnoth-dev
20151007 16:30:33< celticminstrel> gfgtdf: Have you seen PR518 yet?
20151007 16:31:41< celticminstrel> Okay, LoW works, need to test SoF too.
20151007 16:31:56< celticminstrel> Or actually, I'll try DF first.
20151007 16:31:58-!- jinblack_ [~jinblack@ip184-187-174-107.sb.sd.cox.net] has quit [Ping timeout: 272 seconds]
20151007 16:33:06< celticminstrel> usage=mask is also found in HTTT and Liberty.
20151007 16:33:12< celticminstrel> Oh and TSG.
20151007 16:34:00< vultraz> heh
20151007 16:34:07< vultraz> celticminstrel: test a single-hex mask
20151007 16:34:35< celticminstrel> Sure, I'll run the unit test, which one was it... test_victory_something...
20151007 16:35:53< celticminstrel> Oh huh, DF defaults to green and purple, rather than red and... whatever P2 default usually is.
20151007 16:36:52< vultraz> p2 should be blue
20151007 16:37:13< vultraz> if it's different it just means a different color= value in [side]
20151007 16:37:47-!- mjs-de [~mjs-de@x4db52b06.dyn.telefonica.de] has quit [Remote host closed the connection]
20151007 16:44:03-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20151007 16:45:22-!- irker856 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20151007 16:45:22< irker856> wesnoth: gfgtdf wesnoth:master dd1b681d59c8 / src/scripting/game_lua_kernel.cpp: add support for overlays to lua units poxies http://git.io/vcjPJ
20151007 16:45:48< celticminstrel> It looks like the masks do work in DF.
20151007 16:46:21< celticminstrel> So only masks that used border_size=1 (and usage=map) will need to be updated to have border=yes. ...probably.
20151007 16:46:32-!- roberpot [~roberpot@84.122.85.148.dyn.user.ono.com] has quit [Ping timeout: 272 seconds]
20151007 16:46:36-!- new_one [~new_one@2604:a880:1:20::22e:d001] has left #wesnoth-dev ["WeeChat 1.1.1"]
20151007 16:46:38< vultraz> celticminstrel: make border default
20151007 16:46:49< vultraz> celticminstrel: a border of 1 is kinda supposed to be the 1.13 standard
20151007 16:47:09< vultraz> celticminstrel: and it seems that 1 (and usage=map) was used more widely
20151007 16:47:15< celticminstrel> Well, then DF, SoF, Liberty, TSG, and HTTT will need to be updated. (And a few other multiplayer scenarios.)
20151007 16:47:25< vultraz> updated how?
20151007 16:47:35< celticminstrel> To have border=no.
20151007 16:47:52< celticminstrel> In mainline, LoW appears to be the only one that already has border_size=1 and usage=map.
20151007 16:47:54< vultraz> (DF?)
20151007 16:48:01< celticminstrel> Dark Forecast.
20151007 16:48:06< vultraz> ah
20151007 16:48:14< celticminstrel> MP scenario
20151007 16:48:47< zookeeper> gfgtdf, eeehh...
20151007 16:49:49< zookeeper> gfgtdf, it doesn't even suggest how those variables would be accessed. 
20151007 16:50:00< celticminstrel> 6p_Team_Survival and the test scenario are the only ones missing from that previous list.
20151007 16:50:16-!- louis94 [~~louis94@42.133-243-81.adsl-dyn.isp.belgacom.be] has joined #wesnoth-dev
20151007 16:50:22< vultraz> celticminstrel: ok, you deal with the API, I'll deal with the campaigns
20151007 16:50:33< gfgtdf> zookeeper: for lua my idea were function wesnoth.get/set_side_variable
20151007 16:51:16< celticminstrel> (For the record, I got that list with the following command: git grep "usage=mask" main/1.12 )
20151007 16:51:28< gfgtdf> zookeeper: for wml there can maybe soem tag store side variable or something.
20151007 16:52:06< celticminstrel> (Where "main" is my shortcut for https://github.com/wesnoth/wesnoth.git )
20151007 16:52:33< celticminstrel> (Might be "origin" for some people.)
20151007 16:52:53< vultraz> Ok
20151007 16:53:01< celticminstrel> So vultraz/zookeeper - make border=yes the default and remove old_not/new_not, right?
20151007 16:53:18< vultraz> and data/border_size
20151007 16:53:21< zookeeper> i haven't been following
20151007 16:56:27< zookeeper> gfgtdf, sure, but i just don't see the rationale. if you can write an inventory system then making the variables for a given side carryover to the next scenario should be... pretty trivial. and the engine solution proposed sounds unusual and somewhat complicated.
20151007 16:56:58-!- roberpot [~roberpot@84.122.85.148.dyn.user.ono.com] has joined #wesnoth-dev
20151007 16:57:07< zookeeper> i kinda see the point, but... i just don't see why someone would see it as big enough point to warrant an engine feature
20151007 16:57:10-!- roberpot [~roberpot@84.122.85.148.dyn.user.ono.com] has left #wesnoth-dev []
20151007 16:57:17< celticminstrel> If they're making an inventory system, perhaps they should identify the sides by save_id instead of by side number.
20151007 16:57:23< celticminstrel> Is there a way to get a side by save_id?
20151007 16:57:57< zookeeper> does carryover even entail anything other than units and gold currently? any other side attributes carried over?
20151007 16:58:15< celticminstrel> Recall list, I think.
20151007 16:58:22< gfgtdf> zookeeper: the side scpecif carryover on<l caontiaoins godl and units
20151007 16:58:24< zookeeper> "units"
20151007 16:58:24< celticminstrel> But that's units, I guess.
20151007 16:58:26< zookeeper> yeah
20151007 16:59:06< zookeeper> gfgtdf, contains? :P
20151007 16:59:11< celticminstrel> They could always store the inventory in the side leader or something.
20151007 16:59:29< gfgtdf> celticminstrel: its rather hard to identify sides by save_id specially becase wml variables current doesn't really  support maps
20151007 16:59:29< zookeeper> precisely
20151007 17:00:03< celticminstrel> vultraz: There are several campaigns/scenarios whose maps still have usage=map.
20151007 17:00:09< gfgtdf> actuall< storing it in teh side leaders sees liek a very hacky solution to me, specialyl if fails when teh leader dies or when there ar emultiple leaders.
20151007 17:00:12< vultraz> celticminstrel: mainline?
20151007 17:00:18< celticminstrel> Master, yes.
20151007 17:00:26< vultraz> I thought I got them all..
20151007 17:00:28< gfgtdf> zookeeper: yes, contains
20151007 17:00:33< vultraz> I'll deal with that too
20151007 17:00:35< celticminstrel> (Found with git grep -E "usage=(mask|map)" )
20151007 17:00:44< vultraz> git grep is a thing?
20151007 17:00:45< zookeeper> if the implementation would be really simple then i guess i don't have anything against the idea. but it seems like such a niche case that's easy to workaround that i wouldn't complicate the carryover code for its sake...
20151007 17:00:50< celticminstrel> Some of them are directly embedded in scenarios (like some of the test scenarios).
20151007 17:01:01< celticminstrel> Yes vultraz, it's a shortcut for searching the whole repo. :)
20151007 17:01:06< vultraz> huh
20151007 17:01:24< celticminstrel> [Oct 07@12:51:16pm] celticminstrel: (For the record, I got that list with the following command: git grep "usage=mask" main/1.12 )
20151007 17:01:31< celticminstrel> [Oct 07@12:52:06pm] celticminstrel: (Where "main" is my shortcut for https://github.com/wesnoth/wesnoth.git )
20151007 17:01:31< vultraz> oh
20151007 17:01:33< vultraz> derp
20151007 17:01:48< celticminstrel> Dead Water appears to not have its maps updated, in particular.
20151007 17:01:52< zookeeper> gfgtdf, everything about an inventory is already incredibly hacky, and variable workarounds like that seem insignificant in comparison
20151007 17:02:06< zookeeper> anyway, i'll go make some dinner for now ->
20151007 17:02:24< gfgtdf> zookeeper: whats hacky about that?
20151007 17:05:13< vultraz> celticminstrel: will deal with now\
20151007 17:07:27< celticminstrel> gfgtdf: PR518?
20151007 17:07:45< gfgtdf> celticminstrel: looks fine to to me but i didnt test it
20151007 17:07:57< celticminstrel> I tested it, shall I merge it then?
20151007 17:08:08< gfgtdf> ok
20151007 17:14:46-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20151007 17:24:40-!- prkc [~prkc@catv-89-134-159-103.catv.broadband.hu] has joined #wesnoth-dev
20151007 17:27:34< vultraz> geez
20151007 17:27:40< vultraz> so many campaigns I hadn't converted to new format yet
20151007 17:27:44< vultraz> campaigns' maps*
20151007 17:27:45 * celticminstrel hopes no-one pushes yet.
20151007 17:27:53< celticminstrel> Yeah. :P
20151007 17:27:54 * vultraz hopes no one pushes yet
20151007 17:28:03< celticminstrel> :|
20151007 17:28:11< celticminstrel> These can't both be true.
20151007 17:29:57-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 250 seconds]
20151007 17:31:05< vultraz> celticminstrel: let me push the format updates first
20151007 17:31:10< vultraz> before you start pushing stuff
20151007 17:31:32 * celticminstrel sigh, okay.
20151007 17:35:38-!- skyfaller [~skyfaller@wikipedia/Skyfaller] has joined #wesnoth-dev
20151007 17:36:02-!- skyfaller [~skyfaller@wikipedia/Skyfaller] has quit [Client Quit]
20151007 17:37:30-!- louis94 [~~louis94@42.133-243-81.adsl-dyn.isp.belgacom.be] has quit [Ping timeout: 240 seconds]
20151007 17:37:51-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20151007 17:38:02-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20151007 17:38:59-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20151007 17:40:40< vultraz> celticminstrel: usage=mask border_size=0 now needs border=no, right?
20151007 17:40:50< celticminstrel> Yes.
20151007 17:44:34< vultraz> celticminstrel: hm... SoF's had usage=mask but no border_size, so I'm wondering what I should do with those
20151007 17:46:02< celticminstrel> vultraz: It looks like border_size would be 0 if not specified.
20151007 17:46:14< vultraz> so SoF's need border=no?
20151007 17:46:18< celticminstrel> Yeah.
20151007 17:46:18< vultraz> Ok
20151007 17:50:31< celticminstrel> Almost forgot to run the unit tests that use single-hex masks.
20151007 17:50:42< celticminstrel> One timed out.
20151007 17:52:00< vultraz> gaah. 6p Team Survival has many maps/masks D:
20151007 17:52:08< vultraz> fml fml fml
20151007 17:56:54-!- Appleman1234 [~Appleman1@KD111239004240.au-net.ne.jp] has joined #wesnoth-dev
20151007 17:57:46< vultraz> my thump hurts from pressing ctrl so much :(
20151007 17:57:50< vultraz> thumb*
20151007 18:02:43-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20151007 18:04:46-!- jinblack_ [~jinblack@128.111.48.6] has joined #wesnoth-dev
20151007 18:09:44< irker856> wesnoth: Charles Dang wesnoth:master b8c5641bafb6 / data/campaigns/Dead_Water/maps/ (10 files): DW: convert maps to new format http://git.io/vCeOO
20151007 18:09:47< irker856> wesnoth: Charles Dang wesnoth:master b253a331b58a / data/campaigns/Dead_Water/scenarios/02_Flight.cfg: DW S2: convert inline mask to new format http://git.io/vCeO3
20151007 18:09:50< irker856> wesnoth: Charles Dang wesnoth:master 997221352076 / data/campaigns/Sceptre_of_Fire/ (24 files in 3 dirs): SoF: convert maps and masks to new format http://git.io/vCeOs
20151007 18:09:53< irker856> wesnoth: Charles Dang wesnoth:master 6009729fd180 / data/campaigns/Delfadors_Memoirs/maps/ (21 files): DM: convert maps to new format http://git.io/vCeOG
20151007 18:09:56< irker856> wesnoth: Charles Dang wesnoth:master a97cd1c74ed2 / data/campaigns/Heir_To_The_Throne/maps/ (27 files): HttT: convert maps to new format http://git.io/vCeOZ
20151007 18:09:59< irker856> wesnoth: Charles Dang wesnoth:master d6a31d2be159 / data/campaigns/Heir_To_The_Throne/ (masks/17_Scepter_of_Fire.mask utils/httt_utils.cfg): HttT: split S17 mask into its own file and converted it to new format http://git.io/vCeOn
20151007 18:10:02< irker856> wesnoth: Charles Dang wesnoth:master 7f8645ebc04f / data/campaigns/Liberty/maps/ (8 files): L: convert maps to new format http://git.io/vCeOc
20151007 18:10:05< irker856> wesnoth: Charles Dang wesnoth:master 6cdc849aa276 / data/campaigns/Liberty/scenarios/08_Glory.cfg: L S8: convert inline mask to new format http://git.io/vCeOC
20151007 18:10:08< irker856> wesnoth: Charles Dang wesnoth:master fe3057173bbf / data/campaigns/Son_Of_The_Black_Eye/maps/ (17 files): SotBE: convert maps to new format http://git.io/vCeOW
20151007 18:10:11< irker856> wesnoth: Charles Dang wesnoth:master 8e58764517e5 / data/campaigns/The_Rise_Of_Wesnoth/maps/ (24 files): TRoW: convert maps to new format http://git.io/vCeOl
20151007 18:10:14< irker856> wesnoth: Charles Dang wesnoth:master 31e822d7c641 / data/campaigns/Eastern_Invasion/maps/ (7 files): EI: convert remaining maps to new format http://git.io/vCeO8
20151007 18:10:17< irker856> wesnoth: Charles Dang wesnoth:master f932da26bebc / data/campaigns/The_South_Guard/scenarios/06b_The_Long_March.cfg: TSG S6b: convert inline masks to new format http://git.io/vCeO4
20151007 18:10:20< irker856> wesnoth: Charles Dang wesnoth:master a96253982b6d / data/multiplayer/scenarios/6p_Team_Survival.cfg: Team Survival: convert masks to new format http://git.io/vCeOB
20151007 18:10:23< irker856> wesnoth: Charles Dang wesnoth:master 68363b3e2107 / data/multiplayer/scenarios/ (Wesbench_AI.cfg Wesbench_Scroll.cfg Wesbench_Shroud_Walk.cfg): Convert benchmark scenario maps to new format http://git.io/vCeOR
20151007 18:10:28< vultraz> celticminstrel: DONE 
20151007 18:10:36< celticminstrel> Whoa.
20151007 18:11:10-!- skyfaller [~skyfaller@wikipedia/Skyfaller] has joined #wesnoth-dev
20151007 18:11:23< celticminstrel> Looks like the tests are all fine, just randomly timing out for some reason.
20151007 18:11:42-!- Coffee_irc [~david@ppp118-210-17-216.lns20.adl2.internode.on.net] has joined #wesnoth-dev
20151007 18:12:42< irker856> wesnoth: Celtic Minstrel wesnoth:master 1b22a974c212 / src/ (game_events/action_wml.cpp map.cpp): Make [terrain_mask]border default to yes and imply the required border size http://git.io/vCe3u
20151007 18:20:02< irker856> wesnoth: Celtic Minstrel wesnoth:master 9859091c7c49 / / (21 files in 6 dirs): Convert test scenario maps to new format http://git.io/vCeZk
20151007 18:20:06< irker856> wesnoth: Celtic Minstrel wesnoth:master a2125ad3fd26 / src/formula_string_utils.cpp: Make closing | mandatory in $var?default| syntax http://git.io/vCeZL
20151007 18:20:08< irker856> wesnoth: CelticMinstrel wesnoth:master 57c79de7b3de / src/formula_string_utils.cpp: Merge pull request #518 from CelticMinstrel/var-default http://git.io/vCeZt
20151007 18:21:01-!- Kwandulin [~Miranda@p5B008841.dip0.t-ipconnect.de] has joined #wesnoth-dev
20151007 18:31:23< vultraz> celticminstrel: I'm off, bbl.
20151007 18:32:00< vultraz> Are you going to be pushing the overlay fix soon?
20151007 18:34:03-!- Appleman1234 [~Appleman1@KD111239004240.au-net.ne.jp] has quit [Ping timeout: 252 seconds]
20151007 18:34:15< celticminstrel> The what?
20151007 18:34:36< celticminstrel> You mean 1b22a974c212?
20151007 18:34:49< vultraz> That fixed everything? o_O
20151007 18:34:52< celticminstrel> Despite the close timing, that wasn't part of the pull request.
20151007 18:34:56< vultraz> I thought there would be more to it
20151007 18:34:59< celticminstrel> I think so.
20151007 18:35:19< celticminstrel> I think it fixed everything.
20151007 18:35:26< vultraz> Wow, ok
20151007 18:35:48< celticminstrel> I used the border key to determine what border size should be passed to the map when reading it.
20151007 18:36:05< celticminstrel> So if border=yes, the map always has a border size of 1.
20151007 18:36:18< vultraz> ah, the bool to int thing
20151007 18:36:19< celticminstrel> And if border=no, then the mask always has a border size of 0.
20151007 18:36:41< celticminstrel> I actually reverted that, because once the border key was used, the existing conditional worked.
20151007 18:37:17< celticminstrel> And I thought it wouldn't hurt to leave it as a sanity test.
20151007 18:37:35< vultraz> the third argument of gamemap::read is an int, and you pass a bool
20151007 18:37:37< vultraz> is what I meant
20151007 18:37:54< celticminstrel> Yeah.
20151007 18:38:27< vultraz> Do you have any objection to a mask_file key like map_file in [replace_map], BTW?
20151007 18:38:33< vultraz> serves same function
20151007 18:38:39< vultraz> ie, load at runtime
20151007 18:39:12 * celticminstrel shrugs.
20151007 18:41:57 * vultraz adds
20151007 18:42:49< irker856> wesnoth: Charles Dang wesnoth:master 35b227bbc25e / src/game_events/action_wml.cpp: Add [terrain_mask] mask_file= key to load mask at runtime http://git.io/vCe4h
20151007 18:45:56 * vultraz does some minor cleanup regarding border size stuff
20151007 18:50:52-!- ancestral [~ancestral@c-66-41-61-49.hsd1.mn.comcast.net] has joined #wesnoth-dev
20151007 18:53:36-!- louis94 [~~louis94@42.133-243-81.adsl-dyn.isp.belgacom.be] has joined #wesnoth-dev
20151007 18:55:12-!- roberpot1 [~roberpot@84.122.85.148.dyn.user.ono.com] has joined #wesnoth-dev
20151007 18:55:15-!- roberpot1 [~roberpot@84.122.85.148.dyn.user.ono.com] has left #wesnoth-dev []
20151007 19:00:29< irker856> wesnoth: Charles Dang wesnoth:master ae1ebd09cd30 / src/ (4 files in 3 dirs): Minor cleanup regarding border size http://git.io/vCezx
20151007 19:00:39< vultraz> ok, now I'm REALLY off
20151007 19:01:25< vultraz> celticminstrel: big thanks for figuring this out
20151007 19:08:12-!- ancestral [~ancestral@c-66-41-61-49.hsd1.mn.comcast.net] has quit [Quit: i go nstuf kthxbai]
20151007 19:15:29-!- Coffee_irc [~david@ppp118-210-17-216.lns20.adl2.internode.on.net] has quit [Ping timeout: 265 seconds]
20151007 19:25:58-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20151007 19:28:02-!- jinblack_ [~jinblack@128.111.48.6] has quit [Ping timeout: 272 seconds]
20151007 19:29:36-!- shadowm_desktopW [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20151007 19:30:56-!- Appleman1234 [~Appleman1@KD111239024043.au-net.ne.jp] has joined #wesnoth-dev
20151007 19:32:04-!- Nobun [~nobun@5.170.129.45] has joined #wesnoth-dev
20151007 19:51:24-!- louis94 [~~louis94@42.133-243-81.adsl-dyn.isp.belgacom.be] has quit [Ping timeout: 264 seconds]
20151007 19:54:24-!- Nobun [~nobun@5.170.129.45] has quit [Quit: Salve a tutti]
20151007 19:54:31-!- jinblack_ [~jinblack@128.111.48.6] has joined #wesnoth-dev
20151007 19:56:15-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20151007 19:59:21-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20151007 20:27:03-!- ideuler [~textual@a89-153-77-183.cpe.netcabo.pt] has joined #wesnoth-dev
20151007 20:35:21-!- roberpot [~roberpot@84.122.85.148.dyn.user.ono.com] has joined #wesnoth-dev
20151007 20:35:58-!- roberpot [~roberpot@84.122.85.148.dyn.user.ono.com] has left #wesnoth-dev []
20151007 20:59:00-!- Kwandulin [~Miranda@p5B008841.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20151007 21:10:14-!- Coffee_irc [~david@ppp118-210-17-216.lns20.adl2.internode.on.net] has joined #wesnoth-dev
20151007 21:14:56-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20151007 21:15:29-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20151007 21:32:02-!- louis94 [~~louis94@42.133-243-81.adsl-dyn.isp.belgacom.be] has joined #wesnoth-dev
20151007 21:44:00-!- fendrin [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20151007 21:47:26-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Ping timeout: 244 seconds]
20151007 22:01:23-!- irker856 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20151007 22:08:12-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20151007 22:23:13-!- c74d3a [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has quit [Ping timeout: 252 seconds]
20151007 22:48:10-!- fendrin [~quassel@wesnoth/developer/fendrin] has quit [Ping timeout: 272 seconds]
20151007 22:55:14-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20151007 22:55:28-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has quit [Quit: tomreyn]
20151007 22:56:18-!- Appleman1234 [~Appleman1@KD111239024043.au-net.ne.jp] has quit [Ping timeout: 246 seconds]
20151007 22:58:00-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 264 seconds]
20151007 23:00:56-!- Coffee_irc [~david@ppp118-210-17-216.lns20.adl2.internode.on.net] has quit [Quit: Konversation terminated!]
20151007 23:01:23-!- Coffee_irc [~david@ppp118-210-17-216.lns20.adl2.internode.on.net] has joined #wesnoth-dev
20151007 23:05:40-!- mjs-de [~mjs-de@f048001133.adsl.alicedsl.de] has joined #wesnoth-dev
20151007 23:07:49-!- ideuler [~textual@a89-153-77-183.cpe.netcabo.pt] has quit [Quit: Chakalaka.]
20151007 23:08:31-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20151007 23:15:33< vultraz> celticminstrel: does a template return value such as T foo means the return type is dynamic?
20151007 23:15:57< vultraz> which is why I don't need T void foo because it returns nothing and can just use void foo?
20151007 23:20:07< celticminstrel> Yes.
20151007 23:24:24-!- skyfaller [~skyfaller@wikipedia/Skyfaller] has quit []
20151007 23:25:27< vultraz> random though: I haven't had need of pointers in anything I've been doing
20151007 23:25:44< celticminstrel> Probably not a bad thing.
20151007 23:28:57< vultraz> note:   deduced conflicting types for parameter 'T' ('team::CONTROLLER::type' and 'team::CONTROLLER*')|
20151007 23:28:59< vultraz> blagh
20151007 23:30:01< celticminstrel> The former is the raw enum, the latter is the wrapper type created by MAKE_ENUM (or, a pointer to that type).
20151007 23:30:18< vultraz> POINTERS D:
20151007 23:30:24< celticminstrel> I don't know how that error could've happened from just that, though.
20151007 23:33:08< vultraz> ...huh
20151007 23:33:47-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20151007 23:33:55< vultraz> why do sometimes arguments get detected as char and not string
20151007 23:34:29< celticminstrel> The type of a string literal is const char*
20151007 23:34:41< celticminstrel> That's probably why.
20151007 23:35:17< vultraz> Because I'm getting error: no matching function for call to 'gui2::teditor_edit_side::register_radio_toggle(gui2::twindow&, const char [17], team::CONTROLLER::type, team::CONTROLLER*)'|
20151007 23:35:25< vultraz> note: candidate: template<class T> void gui2::teditor_edit_side::register_radio_toggle(gui2::twindow&, const string&, T&, T&)|
20151007 23:35:28< celticminstrel> With C++14 I believe you can append _s to the literal to make it a std::string (ie "your string"_s) but that's unlikely to happen anytyime soon.
20151007 23:35:48< celticminstrel> Oh, so it's detecting it as const char[n], close enough.
20151007 23:35:59< celticminstrel> ^const char[len+1]
20151007 23:36:25-!- louis94 [~~louis94@42.133-243-81.adsl-dyn.isp.belgacom.be] has quit [Quit: Konversation terminated!]
20151007 23:36:43< celticminstrel> If I recall correctly, all you're doing with tha tparameter is passing it on to another function, right?
20151007 23:37:00< celticminstrel> So it's probably fine to declare it as const char* rather than const string&
20151007 23:37:21< celticminstrel> Though I dunno if that's really your problem.
20151007 23:37:45< celticminstrel> Why are you passing the team::CONTROLLER value twice in two different ways?
20151007 23:37:45< vultraz> yeah, me neither..
20151007 23:38:03< vultraz> 	register_radio_toggle(window, "controller_human", team::CONTROLLER::HUMAN, &controller_);
20151007 23:38:10< celticminstrel> I'd deal with that first, the string discrepancy may not be an actual issue.
20151007 23:38:11< vultraz> it needs to see if uh...
20151007 23:38:14< celticminstrel> Ahh.
20151007 23:38:16< vultraz> arg 3 == 4
20151007 23:44:54< celticminstrel> Okay, so declare the fourth argument as const T& and don't use & when passing it (the latter is your problem here, the former is more just good practice).
20151007 23:45:16< vultraz> oh, I see
20151007 23:45:30< celticminstrel> (Might not be your whole problem, since templates are picky, but...)
20151007 23:45:50-!- mjs-de [~mjs-de@f048001133.adsl.alicedsl.de] has quit [Remote host closed the connection]
20151007 23:46:01< celticminstrel> team::CONTROLLER::type should be implicitly convertible to team::CONTROLLER.
20151007 23:46:40< celticminstrel> If it still doesn't work, call it as register_radio_toggle<team::CONTROLLER>(other args);
20151007 23:47:11< celticminstrel> That bypasses template argument type detection and forces T to be team::CONTROLLER.
20151007 23:51:51< gfgtdf> vultraz: you can pass argument 4 eigher by reference (  ..., controller_);  ) then you need T& as parameter or by pointer (  ..., &controller_);  ) then you need T*. the thir parameterr shoudl be passed by value so just T
20151007 23:52:06< gfgtdf> vultraz: hmm sry didnd t see the rest of teh conversation
20151007 23:52:15-!- travis-ci [~travis-ci@ec2-54-162-147-254.compute-1.amazonaws.com] has joined #wesnoth-dev
20151007 23:52:16< travis-ci> wesnoth/wesnoth#7591 (master - 1b22a97 : Celtic Minstrel): The build was broken.
20151007 23:52:16< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/84151817
20151007 23:52:16-!- travis-ci [~travis-ci@ec2-54-162-147-254.compute-1.amazonaws.com] has left #wesnoth-dev []
20151007 23:52:19< celticminstrel> Actually, both T's should probably be const.
20151007 23:52:21< celticminstrel> o.o
20151007 23:52:26< vultraz> C:\Users\Charles\Documents\wesnoth-git\src\gui\dialogs\editor\editor_edit_side.cpp|91|note:   cannot convert 'HUMAN' (type 'team::CONTROLLER::type') to type 'team::CONTROLLER&'|
20151007 23:52:27< vultraz> o_O
20151007 23:52:29< gfgtdf> celticminstrel: no none of then shodulnbe const
20151007 23:52:50< celticminstrel> Why not?
20151007 23:52:58< gfgtdf> celticminstrel: the first shoudl be by value, and the second will  be changes in teh callback so it cannot be cpnst
20151007 23:54:20-!- Appleman1234 [~Appleman1@KD106161095058.au-net.ne.jp] has joined #wesnoth-dev
20151007 23:54:28< celticminstrel> Okay, so it's not that broken, just still mysteriously timing out on some of the tests that use terrain masks.
20151007 23:55:03< celticminstrel> Ah, you're right, second one shouldn't be const, sorry.
20151007 23:55:23< celticminstrel> First one still should be, though you could pass by value instead (no &) in which case don't const it.
20151007 23:55:25< gfgtdf> celticminstrel: hmm all travis tests fail  :(
20151007 23:55:39< celticminstrel> gfgtdf: Which build are you looking at?
20151007 23:55:47< celticminstrel> The one I looked at only had two failures.
20151007 23:56:15< gfgtdf> celticminstrel: https://travis-ci.org/wesnoth/wesnoth/builds/84151817 (the one just posted by travis-ci)
20151007 23:56:24< celticminstrel> No, I mean whih sub-build.
20151007 23:56:27< celticminstrel> ^which
20151007 23:56:37< celticminstrel> Oh wait, you're just looking at that, aren't you.
20151007 23:56:42< gfgtdf> celticminstrel: all sub build failed thats what i meant
20151007 23:57:47< celticminstrel> Two WML tests failed due to timeout. I think rerunning the build could have different effects, as I saw the same issue myself with the tests, but inconsistently (I ran those particular tests three times and each of the tests passed at least once, and the errors.log didn't show anything at all).
--- Log closed Thu Oct 08 00:00:25 2015