--- Log opened Mon Mar 14 00:00:31 2016
20160314 00:11:35-!- travis-ci [~travis-ci@ec2-54-167-252-66.compute-1.amazonaws.com] has joined #wesnoth-dev
20160314 00:11:36< travis-ci> gfgtdf/wesnoth-old#615 (side_name - cd40c9f : gfgtdf): The build has errored.
20160314 00:11:36< travis-ci> Build details : https://travis-ci.org/gfgtdf/wesnoth-old/builds/115751042
20160314 00:11:36-!- travis-ci [~travis-ci@ec2-54-167-252-66.compute-1.amazonaws.com] has left #wesnoth-dev []
20160314 00:25:45-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20160314 00:48:44-!- travis-ci [~travis-ci@ec2-54-226-42-114.compute-1.amazonaws.com] has joined #wesnoth-dev
20160314 00:48:45< travis-ci> gfgtdf/wesnoth-old#616 (side_name - c73b822 : gfgtdf): The build failed.
20160314 00:48:45< travis-ci> Build details : https://travis-ci.org/gfgtdf/wesnoth-old/builds/115755302
20160314 00:48:45-!- travis-ci [~travis-ci@ec2-54-226-42-114.compute-1.amazonaws.com] has left #wesnoth-dev []
20160314 00:55:12-!- louis94 [~~louis94@91.178.240.1] has quit [Quit: Konversation terminated!]
20160314 00:59:38< vultraz> gfgtdf: why does a blindfold get applied for the turn dialog?
20160314 01:07:39< gfgtdf> vultraz: hmm this doesnt happen to me.
20160314 01:08:03< gfgtdf> vultraz: maybe you accidently enabled this while testing the new preferences dialog ?
20160314 01:08:17< vultraz> playsingle_controller.cpp:428
20160314 01:09:08< gfgtdf> vultraz: hmm still i think you must have aciently enabled if this appears to you
20160314 01:09:24< vultraz> I dunno
20160314 01:09:40< gfgtdf> vultraz: this feature was implemented by iceiceice for hotseat multiplayer games
20160314 01:09:40< vultraz> let me test without that line
20160314 01:09:52< gfgtdf> vultraz: don't remove that code in cingleplayer.cpp
20160314 01:09:59< vultraz> yes but why do you need it for the turn dialog
20160314 01:10:14< gfgtdf> vultraz: well so taht you dotn see teh map behind it
20160314 01:10:52< vultraz> that doesn't make sense in this context
20160314 01:10:58< gfgtdf> vultraz: so taht you dont get the vision for the next team which might be an enemy
20160314 01:11:19< vultraz> doesn't that only matter if you're an observer
20160314 01:11:27< celticminstrel> vultraz: Hotseat
20160314 01:11:40< celticminstrel> Two people playing together, using the same computer.
20160314 01:11:49< vultraz> oh
20160314 01:11:51< vultraz> huh
20160314 01:11:55< vultraz> yeah I guess... that is a thing
20160314 01:12:01< celticminstrel> Player 1 moves, ends turn. leaves it at the turn dialog. Player 2 comes along, moves, etc etc.
20160314 01:12:07< vultraz> seems a weird concept
20160314 01:12:12< vultraz> to me, for some reason
20160314 01:14:24< vultraz> well at least I know it's deliberate 
20160314 01:14:32< vultraz> not a bug
20160314 01:42:41-!- irker864 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160314 01:54:12-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160314 02:05:33< celticminstrel> The Lua console unfortunately doesn't capture local variables.
20160314 02:06:45< celticminstrel> At least, not when invoked through the gamestate inspector.
20160314 02:10:52< gfgtdf> celticminstrel: which local variables you mean ?
20160314 02:11:09< celticminstrel> Local variables from the scope that invoked it.
20160314 02:11:31< celticminstrel> eg, when writing "wesnoth.wml_actions.inspect{}" in my code.
20160314 02:11:51< celticminstrel> (Even better would be something like wesnoth.show_console().)
20160314 02:12:02< celticminstrel> (Maybe easier to implement, too.)
20160314 02:13:02-!- gfgtdf_ [~chatzilla@f050176167.adsl.alicedsl.de] has joined #wesnoth-dev
20160314 02:14:05< gfgtdf_> celticminstrel: hmm yes i use [inspect] very rareley so i never noticed that.
20160314 02:14:29< gfgtdf_> celticminstrel: well but it doesnt surpise mean i mean liek in other cases when you call a function taht function cannot see your local variables eigher
20160314 02:14:36< celticminstrel> Right.
20160314 02:15:16< gfgtdf_> celticminstrel: can't you temorarily add lua code in your functions that copied those values to global variables ?
20160314 02:15:24< celticminstrel> But this is a situation specifically intended for debugging, so it would be useful to break that rule.
20160314 02:15:32< celticminstrel> Ah, that's true, I could remove "local".
20160314 02:15:34-!- gfgtdf [~chatzilla@f054160176.adsl.alicedsl.de] has quit [Ping timeout: 252 seconds]
20160314 02:15:42-!- gfgtdf_ is now known as gfgtdf
20160314 02:16:14< gfgtdf> celticminstrel: hmm mabye teh lua debugging tools allow you to show variables up the stack ?
20160314 02:16:26< celticminstrel> Not sure.
20160314 02:18:37< gfgtdf> celticminstrel: i'D think taht debug.getlocal does exactly that
20160314 02:18:46< celticminstrel> Ah.
20160314 02:18:52< celticminstrel> Okay, I'll try that next time.
20160314 02:19:11< celticminstrel> For now, I've resolved the issue that was leading me to want access to local variavles.
20160314 02:19:14< celticminstrel> ^variables
20160314 02:22:24-!- gfgtdf [~chatzilla@f050176167.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 45.0/20160303134406]]
20160314 02:49:15-!- ancestral [~ancestral@71-220-49-102.mpls.qwest.net] has joined #wesnoth-dev
20160314 02:50:14-!- travis-ci [~travis-ci@ec2-54-167-252-66.compute-1.amazonaws.com] has joined #wesnoth-dev
20160314 02:50:15< travis-ci> gfgtdf/wesnoth-old#617 (side_name - b3b4eb5 : gfgtdf): The build is still failing.
20160314 02:50:15< travis-ci> Build details : https://travis-ci.org/gfgtdf/wesnoth-old/builds/115766366
20160314 02:50:15-!- travis-ci [~travis-ci@ec2-54-167-252-66.compute-1.amazonaws.com] has left #wesnoth-dev []
20160314 02:51:13< celticminstrel> What's with all these side_name failures...
20160314 03:43:36-!- gfgtdf [~chatzilla@f050176167.adsl.alicedsl.de] has joined #wesnoth-dev
20160314 03:44:06< gfgtdf> celticminstrel: hmm i somehow get warnings 'put_unit(x,y,unit) is deprecated' when i use u:to_map(x, y)
20160314 03:44:26< celticminstrel> Weird...
20160314 03:44:30-!- gfgtdf [~chatzilla@f050176167.adsl.alicedsl.de] has quit [Client Quit]
20160314 03:44:32< celticminstrel> That's not supposed to happen.
20160314 04:12:13-!- ancestral [~ancestral@71-220-49-102.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160314 05:14:16-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20160314 05:22:48-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds]
20160314 06:14:29-!- Kwandulin [~Miranda@p200300760F7D736640A5B2E0E19852FB.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160314 06:48:02-!- celticminstrel is now known as celmin|Zzzzzz
20160314 07:05:18-!- nurupo_ [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20160314 07:07:29-!- Aginor [~andreas@unaffiliated/aginor] has quit [Remote host closed the connection]
20160314 07:09:22-!- Aginor [~andreas@unaffiliated/aginor] has joined #wesnoth-dev
20160314 07:13:33-!- mjs-de [~mjs-de@x5ce490ca.dyn.telefonica.de] has joined #wesnoth-dev
20160314 07:15:30-!- Netsplit *.net <-> *.split quits: DDR, nurupo, Greywhind, crimson_penguin
20160314 07:15:30-!- nurupo_ is now known as nurupo
20160314 07:16:14-!- Netsplit over, joins: Greywhind, DDR, crimson_penguin
20160314 07:16:56< Aginor> vultraz: thanks for fixing those bugs
20160314 07:39:19< vultraz> :)
20160314 07:39:21< vultraz> no prob
20160314 07:39:38< vultraz> Aginor: anything else you'd like me to work on?
20160314 07:42:32-!- ancestral [~ancestral@184-100-103-220.mpls.qwest.net] has joined #wesnoth-dev
20160314 07:59:28-!- daMark [5cd3e742@gateway/web/freenode/ip.92.211.231.66] has joined #wesnoth-dev
20160314 08:02:28-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160314 08:06:07-!- ancestral [~ancestral@184-100-103-220.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160314 08:39:45-!- Crendgrim [~crend@wesnoth/forum-moderator/crendgrim] has joined #wesnoth-dev
20160314 08:40:55-!- Kwandulin [~Miranda@p200300760F7D736640A5B2E0E19852FB.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160314 09:14:31-!- boucman_work1 [~boucman@193.56.60.161] has joined #wesnoth-dev
20160314 09:17:56-!- shadowm [~ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 244 seconds]
20160314 09:18:03-!- shikadibot [~shikadi@wesnoth/umc-dev/bot/shikadibot] has quit [Ping timeout: 240 seconds]
20160314 09:18:22-!- AI0867 [~ai@wesnoth/developer/ai0867] has quit [Ping timeout: 260 seconds]
20160314 09:18:27-!- boucman_work1 [~boucman@193.56.60.161] has quit [Remote host closed the connection]
20160314 09:18:53-!- boucman_work1 [~boucman@bob75-2-81-56-46-209.fbx.proxad.net] has joined #wesnoth-dev
20160314 09:19:36-!- boucman_work1 [~boucman@bob75-2-81-56-46-209.fbx.proxad.net] has quit [Client Quit]
20160314 09:19:55-!- boucman_work1 [~boucman@193.56.60.161] has joined #wesnoth-dev
20160314 09:42:41-!- Appleman1234 [~Appleman1@KD106161147064.au-net.ne.jp] has quit [Ping timeout: 244 seconds]
20160314 09:43:45-!- shadowm [~ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20160314 09:48:17-!- AI0867 [~ai@wesnoth/developer/ai0867] has joined #wesnoth-dev
20160314 09:48:23-!- shikadibot [~shikadi@wesnoth/umc-dev/bot/shikadibot] has joined #wesnoth-dev
20160314 09:49:19-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160314 09:50:34-!- zombah [~zombah@2a02:28:3:1:214:4fff:fe47:5920] has joined #wesnoth-dev
20160314 09:53:15-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 264 seconds]
20160314 09:53:16-!- wedge010 is now known as wedge009
20160314 10:02:47-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Quit: Quitte]
20160314 10:03:10-!- boucman_work1 is now known as boucman_work
20160314 10:03:19-!- boucman_work [~boucman@193.56.60.161] has quit [Changing host]
20160314 10:03:19-!- boucman_work [~boucman@wesnoth/developer/boucman] has joined #wesnoth-dev
20160314 10:03:35-!- boucman_work [~boucman@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20160314 10:04:08-!- boucman_work [~boucman@bob75-2-81-56-46-209.fbx.proxad.net] has joined #wesnoth-dev
20160314 10:05:45-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20160314 10:09:39< vultraz> zookeeper: moar terrains from doofus
20160314 10:11:29< zookeeper> i know
20160314 10:17:15-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Ping timeout: 264 seconds]
20160314 10:20:37-!- fabi [~quassel@176.4.138.100] has joined #wesnoth-dev
20160314 10:20:37-!- fabi [~quassel@176.4.138.100] has quit [Changing host]
20160314 10:20:37-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20160314 10:22:39< fabi> vultraz: https://forums.wesnoth.org/viewtopic.php?p=593267#p593267
20160314 10:23:29< fabi> I think she might have a good text there. At least she deserves some answer.
20160314 10:27:45< vultraz> zookeeper?
20160314 10:27:55< zookeeper> getting timeout
20160314 10:33:02< zookeeper> well it's a bit... meh. i can't tell if it's better or worse than the current one.
20160314 10:34:28< fabi> Well, the world would be boring if every person shared the same taste :-)
20160314 10:39:21-!- Appleman1234 [~Appleman1@KD106161150240.au-net.ne.jp] has joined #wesnoth-dev
20160314 10:54:41-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160314 11:18:55-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160314 11:23:16-!- daMark [5cd3e742@gateway/web/freenode/ip.92.211.231.66] has quit []
20160314 11:27:31-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160314 11:27:55-!- irker239 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160314 11:27:55< irker239> wesnoth: Charles Dang wesnoth:master e13cf1e5912c / / (3 files in 2 dirs): Procedurally draw combobox borders https://github.com/wesnoth/wesnoth/commit/e13cf1e5912c666f79992e21fc9d420bbaac297c
20160314 11:29:32< vultraz> I could probably do the whole thing procedurally... hm
20160314 11:32:32< fabi> vultraz: What is preceduarlly?
20160314 11:33:04< vultraz> Using GUI2 to draw the appropriate images on the canvas instead of with images
20160314 11:33:13< vultraz> s/appropriate images/appropriate design
20160314 11:36:35< fabi> I see.
20160314 11:37:18-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160314 11:43:40-!- louis94 [~~louis94@91.178.241.103] has joined #wesnoth-dev
20160314 11:52:48-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds]
20160314 11:55:48-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160314 11:55:49-!- stikonas_ is now known as stikonas
20160314 12:01:04-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160314 12:02:42-!- louis94 [~~louis94@91.178.241.103] has quit [Quit: Konversation terminated!]
20160314 12:33:00-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160314 13:16:07-!- boucman_work [~boucman@bob75-2-81-56-46-209.fbx.proxad.net] has quit [Ping timeout: 252 seconds]
20160314 13:17:46-!- boucman_work [~boucman@193.56.60.161] has joined #wesnoth-dev
20160314 13:19:48-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 276 seconds]
20160314 13:23:41-!- boucman_work [~boucman@193.56.60.161] has quit [Ping timeout: 268 seconds]
20160314 13:24:53-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160314 13:33:50< vultraz> gfgtdf: adding checkboxes to the dropdown list dialog works. you can select them on/off without closing the dialog
20160314 13:35:09< vultraz> I'm pondering the best way to make use of this
20160314 13:35:49< vultraz> bc I think we could actually make right-click menus GUI2 
20160314 13:37:01< vultraz> at least 4 columns are needed...
20160314 13:37:21< vultraz> icon - checkbox - name - descrip 
20160314 13:37:28-!- travis-ci [~travis-ci@ec2-54-167-252-66.compute-1.amazonaws.com] has joined #wesnoth-dev
20160314 13:37:29< travis-ci> gfgtdf/wesnoth-old#618 (side_name - 49ffe50 : gfgtdf): The build was fixed.
20160314 13:37:29< travis-ci> Build details : https://travis-ci.org/gfgtdf/wesnoth-old/builds/115848371
20160314 13:37:29-!- travis-ci [~travis-ci@ec2-54-167-252-66.compute-1.amazonaws.com] has left #wesnoth-dev []
20160314 13:38:02< vultraz> but we don't always want to use all of those
20160314 13:38:20-!- gfgtdf [~chatzilla@f050176167.adsl.alicedsl.de] has joined #wesnoth-dev
20160314 13:38:41< gfgtdf> vultraz: where exactly doy ou want to use comboboxes in dropdown  lists ?
20160314 13:38:56< gfgtdf> vultraz: afaik also our gui2 contextmenus don't have that
20160314 13:39:27< gfgtdf> oh my build finally passed
20160314 13:39:54< vultraz> gfgtdf: no not comboboxes in the dropdown list. checkboxes
20160314 13:40:10< vultraz> so like, we could have a dropdown in Campaigns where you can select mods to activate
20160314 13:40:15< vultraz> instead of having them in the list
20160314 13:40:43< gfgtdf> vultraz: did you say you want to have comboboxes inside the dropdown menus or did i understood something wrong ?
20160314 13:41:05< gfgtdf> vultraz: didn't
20160314 13:41:15< vultraz> checkboxes inside the dropdown menus
20160314 13:41:21< vultraz> not comboboxes inside the dropdown menus
20160314 13:41:25< gfgtdf> vultraz: ah ok
20160314 13:43:29-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 244 seconds]
20160314 13:44:32< irker239> wesnoth: gfgtdf wesnoth:master 50a9a25ae7a8 / / (3 files in 3 dirs): move [unstore_unit] definition to lua https://github.com/wesnoth/wesnoth/commit/50a9a25ae7a84da43a0bcc80287a3a9c119adc22
20160314 13:44:34< irker239> wesnoth: gfgtdf wesnoth:master 413d8cf3d2a2 / src/scripting/game_lua_kernel.cpp: implement wesnoth.get_unit(string) directly https://github.com/wesnoth/wesnoth/commit/413d8cf3d2a21e9752b3330054b526a9782c30f0
20160314 13:44:36< irker239> wesnoth: gfgtdf wesnoth:master a91cb9ad0b27 / src/ (config.hpp unit_filter.cpp unit_filter.hpp): add limit=n parameter to standard unit filters https://github.com/wesnoth/wesnoth/commit/a91cb9ad0b27e50c7b74f0ed8e3ad6ba927d74c5
20160314 13:44:38< irker239> wesnoth: gfgtdf wesnoth:master 43275cd9bac2 / src/ (actions/create.cpp actions/create.hpp game_events/action_wml.cpp): don't copy units when recruiting/recalling https://github.com/wesnoth/wesnoth/commit/43275cd9bac2a2c812c518cdaad357cdef6e72bf
20160314 13:44:40< irker239> wesnoth: gfgtdf wesnoth:master ff5a37a0648c / src/scripting/ (game_lua_kernel.cpp game_lua_kernel.hpp): lock unit_map while running filter or applying effects https://github.com/wesnoth/wesnoth/commit/ff5a37a0648cdd1cedbdfbe63544b805e8ef8871
20160314 13:44:42< irker239> wesnoth: gfgtdf wesnoth:master 986550c04059 / / (6 files in 3 dirs): move [teleport] actionwml to lua. https://github.com/wesnoth/wesnoth/commit/986550c0405900d80a3e28dbae093419c484db7a
20160314 13:44:44< irker239> wesnoth: gfgtdf wesnoth:master 310ba2670487 / src/ (3 files in 2 dirs): remove unused queued_event parameter from WML_HANDLER_FUNCTION https://github.com/wesnoth/wesnoth/commit/310ba26704873a89268d7dfef10f384a12ab2559
20160314 13:44:46< irker239> wesnoth: gfgtdf wesnoth:master e89abbeb91ec / src/ (actions/unit_creator.cpp team.cpp): cleanup unit_creator https://github.com/wesnoth/wesnoth/commit/e89abbeb91ecb2a11e450ef3ba0cc3d44cca5b2a
20160314 13:44:48< irker239> wesnoth: gfgtdf wesnoth:master bf2ec4df8208 / src/map.cpp: improve map::overlay() https://github.com/wesnoth/wesnoth/commit/bf2ec4df82086a1ba1e3a10bec8bdcce6dc8ab5a
20160314 13:44:50< irker239> wesnoth: gfgtdf wesnoth:master f1052b0c2730 / src/ (map.cpp map.hpp): remove map::operator[](int) https://github.com/wesnoth/wesnoth/commit/f1052b0c2730c97d2a97e888f904617bc174ea13
20160314 13:44:52< irker239> wesnoth: gfgtdf wesnoth:master 7a8957e91834 / src/ (5 files in 4 dirs): use std::vector to store starting locations https://github.com/wesnoth/wesnoth/commit/7a8957e918342f4b7caf7765037a80341fa52647
20160314 13:44:54< irker239> wesnoth: gfgtdf wesnoth:master b2fff5afd938 / src/scripting/game_lua_kernel.cpp: removed wesnoth.put_unit() warning https://github.com/wesnoth/wesnoth/commit/b2fff5afd938974ede0eb83ff3b57714efef6506
20160314 13:47:18< vultraz> I'm just trying to figure out the best way to make use of this
20160314 13:48:12< vultraz> a dropdown list with checkboxes would probably be best as a toggle button with a label, not two columns with a toggle button and a label
20160314 13:49:13< gfgtdf> vultraz: why woudl you still need the toggle button for the campaogns modifications case ?
20160314 13:49:28< vultraz> bc you can select more than one mod
20160314 13:50:27< gfgtdf> vultraz:  i somehow thought you takled about the lisbox toggle panels
20160314 13:50:27< vultraz> gfgtdf: https://www.dropbox.com/s/efqrcuyaqad45yz/dropdown%20example.PNG?dl=0 quick example of what I'm talking about
20160314 13:51:00< gfgtdf> vultraz: well i dont think it makes sense in _that_ dialog.
20160314 13:51:15< vultraz> right
20160314 13:51:17< vultraz> it doesn't
20160314 13:51:24< vultraz> it's just an example to show you what Im ean
20160314 13:51:26< vultraz> i mean
20160314 13:52:05< gfgtdf> vultraz: hmm but it still has the background toggle panels, so i wonder whether it makes snes to have thos background toggle panels in the modifications dialog
20160314 13:53:19< vultraz> hm
20160314 13:53:20< vultraz> true
20160314 13:53:47< vultraz> basically we need different styles of the dropdown dialog
20160314 13:54:38< vultraz> also for the modifications dialog, we probably don't want to use a combobox since a combobox is about selecting one option from a list 
20160314 13:54:44< vultraz> probably just a regular button
20160314 13:54:46< vultraz> or something
20160314 13:55:32< gfgtdf> vultraz: hmm
20160314 13:56:05< vultraz> gfgtdf: so I'm thinking
20160314 13:56:14< vultraz> gfgtdf: maybe we can make the dropdown behavior a dialog 'mode'
20160314 13:56:16< vultraz> or something
20160314 13:56:28< vultraz> meaning it has a specific border style and it opens at the mouse location
20160314 13:56:34< vultraz> but you can assign that to any dialog
20160314 13:56:35< gfgtdf> vultraz: y i had a similar thought about dialog mode.
20160314 13:56:36< vultraz> so like
20160314 13:56:47< vultraz> we could make different dialogs
20160314 13:56:53< vultraz> and assign them "dropdown mode"
20160314 13:57:21< vultraz> so when they open they open at mouse location or a specified location
20160314 13:57:54-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20160314 13:57:56< gfgtdf> vultraz: hmm note that the currently dropdown menu doesnt open at 'mouse position' but rather vlow or on top of the activaing button.
20160314 13:58:04< vultraz> yeah
20160314 13:58:30< gfgtdf> vultraz: maybe we coudl just make it a window defuinidefnition or something
20160314 13:58:34< gfgtdf> vultraz: not sure though
20160314 13:59:21< vultraz> yeah I'm not sure either
20160314 13:59:38< vultraz> right now there's a 'menu' definition that's used by the dropdown dialog
20160314 13:59:44< vultraz> and that gives it the thin border
20160314 14:00:24-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20160314 14:03:07-!- jimmy_ [642546f3@gateway/web/freenode/ip.100.37.70.243] has joined #wesnoth-dev
20160314 14:07:33< vultraz> hmm
20160314 14:08:39< vultraz> ok so dropdown menus use the "menu" window definition, they open at an arbitrary  position, and dismiss on resize/click
20160314 14:10:15< vultraz> I don't think it can be made a window definition unless we expand the code that defines a window 
20160314 14:11:23< vultraz> orrr....
20160314 14:12:16< vultraz> yeah I'm not sure
20160314 14:13:45< vultraz> I don't really want to add a new key to [window_definition] [resolution] and then add specific code if that key is present 
20160314 14:14:11< vultraz> and sadly dialogs can't have their own variations
20160314 14:14:24< vultraz> s/definitions
20160314 14:17:09-!- boucman_work [~boucman@bob75-2-81-56-46-209.fbx.proxad.net] has joined #wesnoth-dev
20160314 14:17:55< vultraz> gfgtdf: ok I think the best way right now is just to have a few shared attributes in a menu and then have a small helper class that defines functions that a dropdown dialog would have
20160314 14:17:59< vultraz> s/a menu/a macro
20160314 14:18:12< vultraz> like this part:
20160314 14:18:15< vultraz> 	//Dismiss on click outside the window
20160314 14:18:16< vultraz> 	window.connect_signal<event::SDL_LEFT_BUTTON_UP>(boost::bind(&click_callback, boost::ref(window), _3, _4, _5), event::tdispatcher::front_child);
20160314 14:18:18< vultraz> 	//Dismiss on resize
20160314 14:18:19< vultraz> 	window.connect_signal<event::SDL_VIDEO_RESIZE>(boost::bind(&resize_callback, boost::ref(window)), event::tdispatcher::front_child);
20160314 14:18:21< vultraz> etc
20160314 14:27:00-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20160314 14:28:47-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20160314 14:36:18-!- jimmy_ [642546f3@gateway/web/freenode/ip.100.37.70.243] has quit [Ping timeout: 252 seconds]
20160314 14:39:13-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20160314 14:43:19-!- fabi [~quassel@176.4.138.100] has joined #wesnoth-dev
20160314 14:43:19-!- fabi [~quassel@176.4.138.100] has quit [Changing host]
20160314 14:43:19-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20160314 15:19:01-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20160314 15:20:31-!- fabi [~quassel@176.4.138.100] has joined #wesnoth-dev
20160314 15:20:31-!- fabi [~quassel@176.4.138.100] has quit [Changing host]
20160314 15:20:31-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20160314 15:23:30< gfgtdf> vultraz: i wodner whether we shoudl maybe drop the border= parameter in [terrain_mask] completeley and have x,y=0,0 by default instead (which is the somehow the upper-left corner on the map including the border.)
20160314 15:24:08< vultraz> gfgtdf: the border parameter is to help migrate old masks 
20160314 15:26:11< gfgtdf> vultraz: hmm [terrain_mask] is also often used to replace just a littel part of them amp, for example on a 30x30 map you sometimes want to use x,y=10,10 and have mask=<10x10 map data> which woudl then onyl replace the middle third of the map
20160314 15:26:18< irker239> wesnoth: Charles Dang wesnoth:master 11b941ab58f1 / src/gui/dialogs/ (107 files in 3 dirs): GUI2: removed CVideo argument from tdialog::preshow https://github.com/wesnoth/wesnoth/commit/11b941ab58f13f2417189b9b5990aa30ca3da036
20160314 15:26:21< irker239> wesnoth: Charles Dang wesnoth:master 3f582310c9a0 / src/gui/dialogs/ (8 files): GUI2: removed CVideo argument from tdialog::post_build https://github.com/wesnoth/wesnoth/commit/3f582310c9a0238a934cd00b70620b2bf2f48afe
20160314 15:26:22-!- boucman_work [~boucman@bob75-2-81-56-46-209.fbx.proxad.net] has quit [Remote host closed the connection]
20160314 15:26:44-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20160314 15:27:19< vultraz> 107? really? wow O_O
20160314 15:27:29< gfgtdf> vultraz: i think that removing the border= parameter makes this function much easier to understand.
20160314 15:28:22< vultraz> well masks that are the full size of the map might not apply properly
20160314 15:28:57< gfgtdf> vultraz: why that ?
20160314 15:29:14< vultraz> bc some old masks rely on 1,1 as their start
20160314 15:29:33< gfgtdf> vultraz: hmm then you can just pass 1,1 as x,y in [terrain_mask]
20160314 15:29:41< vultraz> hmmm
20160314 15:29:45-!- boucman_work [~boucman@bob75-2-81-56-46-209.fbx.proxad.net] has joined #wesnoth-dev
20160314 15:29:48< gfgtdf> vultraz: also i wouldnt call those 'full size'
20160314 15:29:49< vultraz> good point
20160314 15:30:00< vultraz> you make a good point...
20160314 15:33:23-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Read error: Connection reset by peer]
20160314 15:34:08-!- fabi [~quassel@176.4.138.100] has joined #wesnoth-dev
20160314 15:34:08-!- fabi [~quassel@176.4.138.100] has quit [Changing host]
20160314 15:34:08-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20160314 15:39:18-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20160314 15:41:52-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20160314 15:41:57-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160314 15:50:50-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20160314 15:52:24-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20160314 16:00:51-!- travis-ci [~travis-ci@ec2-54-224-162-220.compute-1.amazonaws.com] has joined #wesnoth-dev
20160314 16:00:52< travis-ci> wesnoth/wesnoth#8873 (master - 3f58231 : Charles Dang): The build has errored.
20160314 16:00:52< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/115882199
20160314 16:00:52-!- travis-ci [~travis-ci@ec2-54-224-162-220.compute-1.amazonaws.com] has left #wesnoth-dev []
20160314 16:03:34< celmin|Zzzzzz> vultraz: Icon and checkbox could maybe share a column?
20160314 16:03:35< celmin|Zzzzzz> ...how is the Lua implementation of [unstore_unit] so much shorter than the C++ one? o.o
20160314 16:04:00 * celmin|Zzzzzz is also dubious about removing the border key, gfgtdf 
20160314 16:04:26< vultraz> celmin|Zzzzzz: possibly
20160314 16:04:29< celmin|Zzzzzz> BTW vultraz, if you use a dropdown box for the modifications, then what do you show when it's not open?
20160314 16:04:32< gfgtdf> celmin|Zzzzzz: becasue most of the work is already done in u:to_map() and u;to_recll_list
20160314 16:04:42< vultraz> celmin|Zzzzzz: a button saying Modifications
20160314 16:05:08< celmin|Zzzzzz> gfgtdf: But it's for compatibility, as vultraz said.
20160314 16:05:24< celmin|Zzzzzz> vultraz: Ahh, so it wouldn't be a combobox.
20160314 16:05:35-!- celmin|Zzzzzz is now known as celticminstrel
20160314 16:05:39-!- Kwandulin [~Miranda@p200300760F7D7329210B18523F5AB63B.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160314 16:07:48< gfgtdf> celticminstrel: well i think that in most cases where this could break compability, the compability is already borken by making it default to true.
20160314 16:08:22< celticminstrel> Isn't it already explicitly set to false in masks that require the behaviour?
20160314 16:08:39-!- esr [~esr@wesnoth/developer/esr] has quit [Read error: No route to host]
20160314 16:09:42-!- esr [~esr@wesnoth/developer/esr] has joined #wesnoth-dev
20160314 16:09:50< vultraz> hmmmm
20160314 16:09:53< gfgtdf> celticminstrel: well afaik we only fixed it in the 2p dark forecast scenario, and didnt test all the oter scaenrios that use it
20160314 16:09:58< gfgtdf> celticminstrel: so  we don'T know
20160314 16:10:38< vultraz> gfgtdf: I'm going to try to make a helper struct for dropdown functions
20160314 16:10:55< vultraz> then it can be used in different dialog definitions as needed
20160314 16:14:41< celticminstrel> Ah.
20160314 16:15:34< vultraz> blah
20160314 16:22:11< vultraz> gahh why does oop confuse me such, sometimes
20160314 16:25:08< vultraz> ok maybe I don't need a struct
20160314 16:25:13< vultraz> just a file with utility functions
20160314 16:27:08< vultraz> celticminstrel: a classless file with utility functions would all need to be static, right?
20160314 16:35:14< vultraz> blagh
20160314 16:35:45< mattsc> celticminstrel: do you know why the OS X package of 1.13.4 expects the cairo libs in /usr/local/opt/ (which I assume is the homebrew directory) ?
20160314 16:35:54< mattsc> I know this is a question for ancestral, but he’s not here atm.
20160314 16:37:23< mattsc> And I also know that I can fix this by throwing in my own dylibs, just wondering why the released package gives me the cairo-libs-not-found error.
20160314 16:41:29< celticminstrel> vultraz: Uh, no.
20160314 16:42:56< celticminstrel> mattsc: No clue. I know /opt/local is MacPorts, but I think Homebrew is something like /usr/local/Cellar/
20160314 16:44:03< mattsc> celticminstrel: okay, thanks.  I’ll wait for ancestral then.
20160314 16:44:14< celticminstrel> vultraz: They would need to be static only if they were enclosed by a class, but in C++, if you have a class with only static functions, then you really need to ask yourself why it's a class and not a namespace.
20160314 16:44:30< vultraz> ah...
20160314 16:44:31< vultraz> yes
20160314 16:44:33< vultraz> I see
20160314 16:44:34< vultraz> ok
20160314 16:44:48< celticminstrel> ("only static functions" also means no data.)
20160314 16:48:57< gfgtdf> celticminstrel: well namespaces have some disadvantages over classes. For example you can pass a class as a template arguments even if its fully static.
20160314 16:49:29< celticminstrel> gfgtdf: True. But, I said "you need to ask yourself why it's a class", not "you should really use a namespace".
20160314 16:49:53< celticminstrel> If you need to pass the class as a template argument, then you've successfully answered that question of why it's a class.
20160314 16:52:47-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160314 16:55:50-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160314 17:10:47-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160314 17:15:15-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160314 17:15:47< irker239> wesnoth: Celtic Minstrel wesnoth:master c1b6060e4bb6 / data/lua/wml-tags.lua: Fix cleanup of [foreach] scope https://github.com/wesnoth/wesnoth/commit/c1b6060e4bb628c8002adba68773f8a16187378b
20160314 17:15:49< irker239> wesnoth: Celtic Minstrel wesnoth:master ecb3dc6c462f / data/lua/wml-tags.lua: Error if looping tags lack [do] https://github.com/wesnoth/wesnoth/commit/ecb3dc6c462f2f02880319e2a5b7b6714a846dd6
20160314 17:15:51< irker239> wesnoth: Celtic Minstrel wesnoth:master 7d9a0d90b968 / src/scripting/ (lua_common.cpp lua_common.hpp): Allow array attribute values in Lua config syntax https://github.com/wesnoth/wesnoth/commit/7d9a0d90b96859b87939eac795e96831624a29a9
20160314 17:15:53< irker239> wesnoth: Celtic Minstrel wesnoth:master 0f39cc3fe015 / src/scripting/ (game_lua_kernel.cpp lua_common.cpp lua_common.hpp): Fix function case https://github.com/wesnoth/wesnoth/commit/0f39cc3fe01573cd6a1fdc0e0c2413d3ef049be0
20160314 17:18:07-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160314 17:20:26-!- boucman_work [~boucman@bob75-2-81-56-46-209.fbx.proxad.net] has quit [Remote host closed the connection]
20160314 17:42:00 * celticminstrel creates PR628, will probably merge by the end of the day (like, in 10 hours or so) unless either Travis fails or someone says something.
20160314 17:47:13-!- travis-ci [~travis-ci@ec2-54-159-66-6.compute-1.amazonaws.com] has joined #wesnoth-dev
20160314 17:47:14< travis-ci> wesnoth/wesnoth#8874 (master - 0f39cc3 : Celtic Minstrel): The build failed.
20160314 17:47:14< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/115915393
20160314 17:47:14-!- travis-ci [~travis-ci@ec2-54-159-66-6.compute-1.amazonaws.com] has left #wesnoth-dev []
20160314 17:47:40< gfgtdf> celticminstrel: i think that luaW_pushlocation shodul oynl returns a 2 enry array not also x and y attributes
20160314 17:48:06< celticminstrel> There are some places that already accepted both versions.
20160314 17:48:20< celticminstrel> Oh wait, pushlocation?
20160314 17:48:23< celticminstrel> Or tolocation?
20160314 17:49:00< gfgtdf> pushlocation
20160314 17:49:11< celticminstrel> I think for pushlocation I wanted it to be compatible with interfaces that relied on array format and also with interfaces that relied on x and y.
20160314 17:49:16< vultraz> gfgtdf: where does the click_callback for the dropdown dialog get its coordinate parameter value?
20160314 17:49:49< gfgtdf> celticminstrel: hmm which interfaces exactly relied on the x,y forat ?
20160314 17:50:02< celticminstrel> src/gui/dialogs/synced_choice_wait.hpp:38:7: error: 'gui2::tsynced_choice_wait::pre_show' hides overloaded virtual function
20160314 17:50:09< celticminstrel> That's bad, should be fixed asap.
20160314 17:50:10< gfgtdf> vultraz: which ccordinate value you mean ?
20160314 17:50:42< vultraz> void click_callback(twindow& window, bool& handled, bool& halt, tpoint coordinate)
20160314 17:50:45< vultraz> last argument
20160314 17:50:48< celticminstrel> gfgtdf: Not quite sure off the top of my head, perhaps some of the unit manipulation ones.
20160314 17:51:47< celticminstrel> Does tsynced_choice_wait::pre_show require the CVideo parameter?
20160314 17:52:21< vultraz> celticminstrel: I removed cvideo from preshow
20160314 17:52:48-!- zombah [~zombah@2a02:28:3:1:214:4fff:fe47:5920] has quit [Quit: Leaving]
20160314 17:53:12< celticminstrel> vultraz: Ah, that's the problem then. Put it back, as an unnamed parameter.
20160314 17:53:18< vultraz> why?
20160314 17:53:29< celticminstrel> Because if you don't, the function will never be called.
20160314 17:53:43< gfgtdf> celticminstrel: what exactly does wesnoth.match_unit(unkit, location) do ?
20160314 17:53:52< vultraz> celticminstrel: I changed the signature 
20160314 17:53:59< vultraz> celticminstrel: seems I forgot to remove that case
20160314 17:54:01< celticminstrel> If it doesn't exactly match the signature of the function in dialog.hpp, then dialog.hpp will not call it.
20160314 17:54:16< vultraz> celticminstrel: I changed the signature in dialog.hpp
20160314 17:54:24< celticminstrel> Oh, I see.
20160314 17:54:44< celticminstrel> There were no pre_show functions that used it? That kind of surprises me a little, honestly.
20160314 17:54:51< vultraz> only three did
20160314 17:55:01< vultraz> but twindow has a video field now
20160314 17:55:06< vultraz> so you can call window.video
20160314 17:55:08< vultraz> ()
20160314 17:55:36< celticminstrel> gfgtdf: As I understand it, it runs the filter as if the unit were on that target location, without actually moving the unit.
20160314 17:55:56< celticminstrel> gfgtdf: It was an already existing feature in the unit_filter class, so I thought I'd expose it to the Lua API.
20160314 17:56:06< vultraz> celticminstrel: were you having trouble with that dialog?
20160314 17:56:09< celticminstrel> vultraz: Ahh. Okay then.
20160314 17:56:09< irker239> wesnoth: Charles Dang wesnoth:master b311135482bf / src/gui/dialogs/ (synced_choice_wait.cpp synced_choice_wait.hpp): tsynced_choice_wait: removed a CVideo argument from pre_show missed in 11b941ab5 https://github.com/wesnoth/wesnoth/commit/b311135482bf814b80f4f2d517f3c045908ef37a
20160314 17:56:16< celticminstrel> vultraz: Hmm?
20160314 17:56:33< vultraz> why did you ask about this dialog specifically which turns out to be the only case I missed
20160314 17:56:50< celticminstrel> Because it's what caused the Travis to fail.
20160314 17:56:54< vultraz> ahh
20160314 17:57:01< vultraz> well it should pass now
20160314 17:57:28< gfgtdf> celticminstrel: well 'becasue it was already a feature of the c++ code' doesnt sound like a good reason to expose something to lua.
20160314 17:57:46< celticminstrel> By itself, no.
20160314 17:57:54< celticminstrel> It seemed like a useful feature though.
20160314 17:58:16< celticminstrel> Like an AI wanting to see what might happen if the unit moves.
20160314 17:58:17-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160314 18:00:14< celticminstrel> ...actually, I wonder if it would be good to use that map_locker in CA evaluation routines.
20160314 18:01:04< celticminstrel> I guess it wouldn't really prevent most gamestate changes though.
20160314 18:02:13< vultraz> hmmmmmm
20160314 18:02:14< vultraz> https://wiki.libsdl.org/SDL_GetMouseState
20160314 18:03:11-!- louis94 [~~louis94@91.178.241.103] has joined #wesnoth-dev
20160314 18:04:34< gfgtdf> celticminstrel: i actualyl think we should change all those functions to return  {x, y} instead of {x = y, y = y}, let me chack te diff of your pr for whether there are any backwards-compability-relevant functions that us eteh x= y= syntax
20160314 18:05:01< celticminstrel> I personally prefer {x=x,y=y}.
20160314 18:05:10< celticminstrel> In fact, I kind of wish locations were userdata.
20160314 18:05:35< celticminstrel> Or at least had a custom metatable.
20160314 18:06:16< gfgtdf> celticminstrel: hmm i just noticed that src/scripting/lua_map_location_ops.cpp and  src/ai/lua/core.cpp  have their own copy of static void push_map_location
20160314 18:06:34< celticminstrel> Oh huh.
20160314 18:06:39< celticminstrel> I didn't notice those.
20160314 18:06:40< gfgtdf> celticminstrel: well but {x, y} does integer lookups while { x= x, y = y} means string lookups
20160314 18:07:13< celticminstrel> Yes, and?
20160314 18:07:54< gfgtdf> celticminstrel: integer lookups are faster
20160314 18:08:19< celticminstrel> In the general case, I'm sure you're right.
20160314 18:08:40< celticminstrel> However, if you specifically compare {x,y} with {x=x,y=y}, I seriously doubt you'd find any difference.
20160314 18:09:05< celticminstrel> If it's {x=x,y=y, lots_of_other_stuff}, then maybe it'd be a bit slower.
20160314 18:10:21< gfgtdf> celticminstrel: but its quite releay x=x, y=y unoless you implement some wml tags maybe.
20160314 18:10:38< celticminstrel> Huh?
20160314 18:13:01 * celticminstrel a) doesn't know what you meant by "releay" and b) has no idea what wml tags have to do with this.
20160314 18:13:39-!- elias [~allefant@allegro/developer/allefant] has quit [Ping timeout: 260 seconds]
20160314 18:14:11< gfgtdf> what i meant is that in those cases where perfromance matters, you usually don't have {x=x, y=y, lots of other stuff anyway.}
20160314 18:14:40< celticminstrel> Right.
20160314 18:14:53< gfgtdf> mattsc: you know what these get_new_enemy_src_dst, get_new_src_dst functions are ?
20160314 18:15:05< celticminstrel> I think those return move maps.
20160314 18:15:28< celticminstrel> A multimap<location,location> or something containing all possible AI moves.
20160314 18:15:48< gfgtdf> celticminstrel: y but when are they used= i cannot find any usecases in mainline
20160314 18:16:03< celticminstrel> The Lua-facing versions are kinda poor, which is probably why they're wrapped by the "dummy Lua engine".
20160314 18:16:15< celticminstrel> get_enemy_src_dst for example.
20160314 18:16:42< gfgtdf> celticminstrel: what do you mean by "lua facing versions" ?
20160314 18:16:51< celticminstrel> See data/ai/lua/stdlib.lua
20160314 18:17:09< celticminstrel> gfgtdf: I mean the ones defined in src/ai/lua/core.?pp
20160314 18:17:22< celticminstrel> There are similar functions in the C++ API.
20160314 18:17:32< celticminstrel> (Which they presumably delegate to.)
20160314 18:18:20< gfgtdf> mattsc: most of those functions currently returns locations in the {x=x, y=y} but from looking at the code it seems like those locations are eigher unused or passed to loaction_set function which turn convert them to the {x,y} format anyway so i wonder whether we can justchange those to return {x,y} in the first place
20160314 18:18:57< celticminstrel> That reminds me, I'd like to promote the location_set thing to something known by the C++ code.
20160314 18:18:59< mattsc> gfgtdf: yes, they are supposed to make movemaps available to the Lua AI coder that are faster than calculating them yourself using Lua
20160314 18:19:39< gfgtdf> mattsc: could you point me to a lua code that uses them, preferably in mainline ?
20160314 18:19:47< mattsc> the reason why they are set up as location sets is so that you can query whether they contain a location without having to loop through on of the coordinates, I think
20160314 18:19:56< mattsc> gfgtdf: there is none
20160314 18:20:08< gfgtdf> mattsc: do you use them in your addon ?
20160314 18:20:13< celticminstrel> Is there any that you know of not in mainline?
20160314 18:20:23< mattsc> the reason is that they were never really working the way how they’d really be useful
20160314 18:20:57< mattsc> by the time these got change to a way that _might_ be working, I was already doing things differently.
20160314 18:21:13< celticminstrel> That way is what stdlib.lua does, I think.
20160314 18:21:15< mattsc> So no, I am not aware of them being used anywhere.  But that doesn’t mean that somebody else is not.
20160314 18:21:35< celticminstrel> So gfgtdf, if you're changing them, you might want to look at that file to.
20160314 18:21:46< celticminstrel> As well as cache.lua.
20160314 18:21:58< celticminstrel> (Which is used by stdlib.lua)
20160314 18:22:20< mattsc> One of the downside the old versions had is that they did not adapt to changing the situation on the map during the AI evaluation.
20160314 18:22:31< mattsc> “What would happen is I did this?” kind of scenarios.
20160314 18:22:46< mattsc> I do not know whether the get_new_* functions now do that or not.
20160314 18:22:55< celticminstrel> Ah, so locking out gamestate changes during CA evaluation is not, in fact, a good idea.
20160314 18:23:35< mattsc> Umm, no, if you do that lots of my AIs will break and I am going to quit.
20160314 18:23:38< mattsc> :P
20160314 18:24:13< celticminstrel> mattsc: People using no explicit [engine] don't have get_new_* and the is_whatever equivalent; instead they have eg get_src_dst, get_enemy_dst_src, etc.
20160314 18:24:26< mattsc> You absolutely _need_ some way of figuring out how, e.g., moving one of your own uints will block the enemy.
20160314 18:24:28< celticminstrel> I don't know if they adapt to "temporary" gamestate changes though.
20160314 18:24:47< gfgtdf> mattsc: so ai_helper.lua created its own function for doing these things ?
20160314 18:24:55< celticminstrel> Not equivalent, sorry.
20160314 18:25:00< celticminstrel> Counterpart, I guess.
20160314 18:25:23< mattsc> well, as I said, I have my own ways of doing so anyway, because they didn’t use to support that.
20160314 18:26:06< mattsc> gfgtdf: something like that; but that function is really old and I don’t remember exactly what it does either; not using that any more either (although it might still be leftover in one of my AIs)
20160314 18:26:17< celticminstrel> Might also be slower.
20160314 18:26:36< gfgtdf> mattsc: is there a documentation of lua ai function? I cannot fund a link in the wiki.wesnoth.org/LuaWML 
20160314 18:26:49< celticminstrel> I think it's at LuaAI.
20160314 18:26:54< mattsc> yep
20160314 18:27:01-!- elias [~allefant@allegro/developer/allefant] has joined #wesnoth-dev
20160314 18:27:06< celticminstrel> They're not available outside of AI code.
20160314 18:27:36< celticminstrel> So it would probably be inappropriate to list them on LuaWML along with all the other API functions.
20160314 18:27:47< celticminstrel> (A link to the page would be fine, I guess.)
20160314 18:28:17< celticminstrel> gfgtdf: But anyway, why are you considering changing this {x=x,y=y} to {x,y} if not for speed concerns?
20160314 18:30:06< gfgtdf> celticminstrel: first, it is also for speed concerns
20160314 18:30:32< celticminstrel> Have you benchmarked this?
20160314 18:30:33< gfgtdf> celticminstrel: but also it woudl be nice if those function returnsing locatiosn had the same interface, most of the functions that i know specially wesnoth.find_path, location_set and wesnoth.get_locations already return the {x, y} format
20160314 18:30:59< celticminstrel> I think my PR628 was changing that though.
20160314 18:31:38< celticminstrel> I see location_set does return {x,y}. I didn't change that.
20160314 18:31:43< gfgtdf> celticminstrel: and if oen format is used by more coomonly used function and it also faster than we shodul definiteled use that one.
20160314 18:32:09< celticminstrel> But is it really faster?
20160314 18:32:41< celticminstrel> Okay, looking up an integer key is a simple comparison and addition or whatever - "is this in range? if so, add offset to beginning of array" or something like that.
20160314 18:32:56< celticminstrel> I don't know whether Lua uses a hash map or a tree map.
20160314 18:33:57< gfgtdf> celticminstrel: i'd guess a hash map
20160314 18:34:13< gfgtdf> celticminstrel: for non low-integer keys i mean.
20160314 18:35:00< celticminstrel> When you consider the {x=x,y=y} case specifically, if it's a hash map, you need to compute the hash, then look it up in the array, then test for possible hash collisions. So, that's 1 point for computing the hash of a 1-character string; 1 point, maybe 2 for looking it up in the array; and 0 or 1 points for testing collisions (probably 0). Only 1 point more than the array case.
20160314 18:35:28< celticminstrel> With a string, testing for collision means lexicographical comparison.
20160314 18:35:49< celticminstrel> So, 1 point per character in the shorter string.
20160314 18:36:34< celticminstrel> If it's a tree map, then I think all you have is a comparison, so that might actually be faster in this degenerate case.
20160314 18:41:09-!- mjs-de [~mjs-de@x5ce490ca.dyn.telefonica.de] has quit [Remote host closed the connection]
20160314 18:44:53< celticminstrel> Basically, I'd prefer to see proof that it really is faster in this specific case.
20160314 18:44:58-!- louis94 [~~louis94@91.178.241.103] has quit [Ping timeout: 268 seconds]
20160314 18:45:31< celticminstrel> (I don't doubt that integer keys are faster than string keys in general, but this is, as I mentioned a degenerate case, so the generality may not apply.)
20160314 18:45:39< celticminstrel> ^mentioned,
20160314 18:45:54-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20160314 18:55:25-!- travis-ci [~travis-ci@ec2-54-159-66-6.compute-1.amazonaws.com] has joined #wesnoth-dev
20160314 18:55:26< travis-ci> wesnoth/wesnoth#8877 (master - b311135 : Charles Dang): The build was fixed.
20160314 18:55:26< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/115924540
20160314 18:55:26-!- travis-ci [~travis-ci@ec2-54-159-66-6.compute-1.amazonaws.com] has left #wesnoth-dev []
20160314 18:57:13< gfgtdf> celticminstrel: hmm im currently traing to do some testing
20160314 19:05:34-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160314 19:06:39< gfgtdf> celticminstrel: hmm accoring to my testing theose integer loopups are not faster than the strign lookups in this case
20160314 19:07:00-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160314 19:07:02< gfgtdf> celticminstrel: hmm maybe thats becaseu all lua number are doubles and it need to do the double -> int cast first
20160314 19:07:10 * celticminstrel assumes you did something like millions of locations.
20160314 19:07:14< celticminstrel> Maybe.
20160314 19:12:21-!- esr [~esr@wesnoth/developer/esr] has quit [Quit: WeeChat 1.3]
20160314 19:15:04< gfgtdf> celticminstrel: hmm we should test again with lua 5.3 which has native integers :p 
20160314 19:15:13< celticminstrel> Sure.
20160314 19:17:39-!- esr [~esr@static-71-162-243-5.phlapa.fios.verizon.net] has joined #wesnoth-dev
20160314 19:17:39-!- esr [~esr@static-71-162-243-5.phlapa.fios.verizon.net] has quit [Changing host]
20160314 19:17:39-!- esr [~esr@wesnoth/developer/esr] has joined #wesnoth-dev
20160314 19:26:13-!- mjs-de [~mjs-de@wh.Uni-Dortmund.DE] has joined #wesnoth-dev
20160314 19:28:55-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Remote host closed the connection]
20160314 19:29:40-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20160314 19:34:02-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160314 19:41:39-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160314 19:46:00-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 250 seconds]
20160314 19:49:38-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160314 19:53:53< vultraz> what is a helptip again?
20160314 19:54:04< vultraz> I think you guys said it's an archaic help thingy
20160314 19:57:03-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160314 19:59:22< vultraz> celticminstrel: you like formulas? look in data/gui/default/window/tooltip_large.cfg p_p
20160314 19:59:36< vultraz> what is this sorcery 
20160314 20:01:36< celticminstrel> Oh wow. I had no idea that GUI2 allowed function definitions.
20160314 20:01:52< vultraz> can you make sense of this?
20160314 20:02:32< celticminstrel> Probably, gimme some time.
20160314 20:03:44< celticminstrel> Okay so.
20160314 20:04:23< celticminstrel> x, y, width, and height each are calculated by a function.
20160314 20:04:34< celticminstrel> And I guess reevaluate_best_size too.
20160314 20:04:43< celticminstrel> Is that supposed to be a boolean?
20160314 20:05:46< vultraz> no idea..
20160314 20:05:54< vultraz> I think 'I' 'II' are variables
20160314 20:06:02< celticminstrel> They aren't.
20160314 20:06:21< celticminstrel> For x, it calls set_x(...) to calculate the value.
20160314 20:06:37< celticminstrel> The parameters passed to this function are: the mouse location, the window size, and the screen size.
20160314 20:07:15< celticminstrel> set_x calls placement_method(...) to decide how to do its job.
20160314 20:08:27< celticminstrel> m.x,m.y is the mouse location; w.x is window width, w.y is window height, s.x is screen width s.y is screen height.
20160314 20:08:42 * celticminstrel missed a comma, whoops.
20160314 20:08:53< celticminstrel> Line 71
20160314 20:09:06< vultraz> ah, I think I'm figuring it out
20160314 20:09:12< celticminstrel> If that condition is true, set_x uses the formula on line 90 - that is, set_x() ultimately returns 5.
20160314 20:11:03< celticminstrel> Lines 73-76 kind of don't make much sense since they're comparing mouse coordinates to window dimensions...
20160314 20:19:08< vultraz> celticminstrel: what do you think https://drive.google.com/file/d/0B-mR9s8FduLLY3VNamZHYmNQUlE/view?usp=sharing
20160314 20:21:53-!- gfgtdf [~chatzilla@f050176167.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 45.0/20160303134406]]
20160314 20:23:38< vultraz> should I deploy?
20160314 20:24:04< celticminstrel> One minute, got distracted by gfgtdf.
20160314 20:31:19< celticminstrel> This may just be personal preference, but I think I'd prefer those formulas to be inline in the config file rather than in macros.
20160314 20:31:45< celticminstrel> (It would also make it easier for me to set up my syntax highlighter to highlight them as formulas. >_> )
20160314 20:32:10< celticminstrel> (Though... I suppose I can add a case for macro definitions that start with ( and end with )...)
20160314 20:32:40< celticminstrel> vultraz: About your tooltips - that looks amazing! What have you done? O_O
20160314 20:32:58< celticminstrel> The tooltip's actually by the mouse rather than at the bottom of the screen!
20160314 20:33:10< vultraz> almost nothing! the functionality's been there the entire time
20160314 20:33:20< celticminstrel> Whyyyy did no-one use it!? :(
20160314 20:33:38< vultraz> we've simply been copying the blocks that say [tooltip] id=tooltip_large [/tooltip] in *every*dialog we use without a second thought
20160314 20:33:59< vultraz> you need id=tooltip for this 
20160314 20:34:12< vultraz> I just changed the look a bit
20160314 20:34:14< celticminstrel> So id=tooltip_large puts it at the bottom?
20160314 20:34:17< vultraz> yup
20160314 20:34:40< vultraz> all the formulas you're looking at are for putting the tooltip around mouse position
20160314 20:35:37< celticminstrel> "# TODO tooltips in this window make little sense.
20160314 20:35:38< celticminstrel> # Have to think of a nice solution."
20160314 20:35:38< vultraz> I actually made it so it centers the tooltip on mouse position 
20160314 20:36:03< celticminstrel> I don't suppose you could post a screenshot that includes the cursor?
20160314 20:36:21< celticminstrel> I think Windows needs external software for that or something.
20160314 20:36:41< celticminstrel> Mac has it built-in, as long as you're using a non-custom cursor.
20160314 20:36:54< vultraz> oh, huh, printscreen discards the cursor
20160314 20:37:03< celticminstrel> Yeah, that's normal.
20160314 20:37:16< celticminstrel> Even Mac does that unless you use the Grab app and tell it not to.
20160314 20:37:26< vultraz> I don't think I can get one with the cursor, no
20160314 20:37:57< celticminstrel> Is the cursor entirely outside the tooltop box?
20160314 20:38:01< vultraz> yes
20160314 20:38:12< celticminstrel> Near the top-left corner, or...?
20160314 20:38:32< vultraz> ok, if the cursor is at 0,0
20160314 20:38:40< vultraz> and the box is 100 px long
20160314 20:38:49< vultraz> with my setup, left edge is at -50, right at 50
20160314 20:38:56< vultraz> and bottom edge at +15 up
20160314 20:39:05< celticminstrel> Fine, fine, but in that specific situation in the screenshot, where's the cursor?
20160314 20:39:25< vultraz> hovered over the option text
20160314 20:39:37< vultraz> remember, the text and button are part of the same widget
20160314 20:39:53< celticminstrel> Sorry, where's the cursor relative to the tooltip box?
20160314 20:40:21< celticminstrel> Ah, wait.
20160314 20:40:26< celticminstrel> I can see where the cursor is.
20160314 20:40:30< celticminstrel> The checkbox is highlighted.
20160314 20:40:39< celticminstrel> So that's where the cursor is. By the bottom left.
20160314 20:40:40< celticminstrel> Okay.
20160314 20:40:54< vultraz> the tooltip stays in the same place as long as the mouse is over the widget
20160314 20:41:14< celticminstrel> Please deploy.
20160314 20:41:36< vultraz> this is on a per-dialog basis, so we should decide what dialogs get this treatment 
20160314 20:41:48< vultraz> (I'd say All Of Them, except maybe the titlescreen)
20160314 20:41:56< celticminstrel> Why not titlescreen?
20160314 20:42:05< celticminstrel> Though hmm, you might be right.
20160314 20:45:00< vultraz> so, all of them
20160314 20:45:02< vultraz> ?
20160314 20:45:07< vultraz> (except titlescreen)
20160314 20:45:15< celticminstrel> It's hard to say "all" without knowing what "all" is.
20160314 20:45:34< celticminstrel> Some others to maybe skip: new addons list, new lobby, gamestate inspector.
20160314 20:45:51< vultraz> yes, good point
20160314 20:45:56< celticminstrel> "maybe"
20160314 20:49:53< vultraz> I'll just mass-scale deploy it except for the aforementioned dialogs
20160314 20:50:01< vultraz> we can always switch back if we discover something off
20160314 20:50:05< vultraz> in a specific dialog
20160314 20:50:06< celticminstrel>  Are there any other fullscreen ones?
20160314 20:50:21< vultraz> not that I know of
20160314 20:50:32< celticminstrel> I'm not saying all fullscreen ones should keep the old method, mind you.
20160314 20:50:37< celticminstrel> Merely that they may not need it.
20160314 20:54:12< irker239> wesnoth: Charles Dang wesnoth:master 2fdc08d92ec9 / data/gui/default/ (widget/window_tooltip_large.cfg window/tooltip_large.cfg): ttip: added a separate window definition for floating tooltips https://github.com/wesnoth/wesnoth/commit/2fdc08d92ec9f02eab9d891335cd3dbac92a5843
20160314 20:54:15< irker239> wesnoth: Charles Dang wesnoth:master 75c3446c7753 / data/gui/default/window/tooltip_large.cfg: ttip: tweaked layout of floating tooltips https://github.com/wesnoth/wesnoth/commit/75c3446c77531270bdbc5ae4ca29e9f7a614df65
20160314 20:54:18< irker239> wesnoth: Charles Dang wesnoth:master 4ebf8a258b5d / data/gui/default/window/ (65 files): GUI2: use floating tooltips (almost) everywhere https://github.com/wesnoth/wesnoth/commit/4ebf8a258b5daf72a3fcc53151ea4dee253f4e7b
20160314 20:55:17< celticminstrel> I see.
20160314 20:55:25-!- Kwandulin [~Miranda@p200300760F7D7329210B18523F5AB63B.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160314 20:55:34< celticminstrel> So now you can say you use WFL. :P
20160314 20:57:01< vultraz> barely :P
20160314 20:58:40< vultraz> I'd like to add one of those little 'triangles' you usually see under tooltips
20160314 20:58:45< vultraz> just need to figure out how
20160314 20:58:50< celticminstrel> Huh?
20160314 20:59:24< vultraz> you know, so it's kinda like a speech bubble
20160314 21:00:38< vultraz> I think I know how
20160314 21:09:21< vultraz> hmmmm
20160314 21:09:24< vultraz> guess not
20160314 21:10:40< vultraz> drawing it past the bottom goes outside canvas 
20160314 21:13:53< celticminstrel> Oh. I don't really thing that's necessary.
20160314 21:14:02< celticminstrel> Tooltips often don't have that.
20160314 21:20:28-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160314 21:20:54-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160314 21:25:03-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20160314 21:30:20< celticminstrel> ^think
20160314 21:34:49< celticminstrel> Grr, Travis getting tons of template errors in lua_push...
20160314 21:35:45-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20160314 21:39:10< celticminstrel> Okay, so the Travis GCC build has template errors, but the Travis clang build does not.
20160314 21:39:25-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160314 21:39:52< celticminstrel> Ah, some unit tests failed. I'd better see if that happens for me, then.
20160314 21:41:05< celticminstrel> I'll let someone else deal with the template errors. >_>
20160314 21:41:29< celticminstrel> (I wonder which compiler is more "correct" with those, actually.)
20160314 21:41:49< celticminstrel> (Is GCC erroneously giving errors, or is clang erroneously accepting code that should be rejected?)
20160314 21:54:26< celticminstrel> Does anyone understand the arcane magic of the dispatch template?
20160314 21:55:20< celticminstrel> Or rather, of lua_kernel_base::get_lua_kernel_base_ptr()
20160314 21:57:31< vultraz> not likely
20160314 21:59:41-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20160314 22:04:54-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160314 22:05:36< irker239> wesnoth: Charles Dang wesnoth:master f92936125b9c / data/gui/default/ (widget/window_tooltip_large.cfg window/tooltip_large.cfg): ttip: few more tweaks to floating tooltip look https://github.com/wesnoth/wesnoth/commit/f92936125b9ca76a448c3a929dbaa47266149642
20160314 22:05:44< vultraz> celticminstrel: see if you like that better or the other text ^
20160314 22:07:02< vultraz> it's now gold_small as opposed to default, and the background is a little different
20160314 22:07:59< celticminstrel> Blargh.
20160314 22:08:14< celticminstrel> I can let you know eventually.
20160314 22:08:34< vultraz> no rush
20160314 22:11:15< celticminstrel> Need to figure out why tests are failing in 628.
20160314 22:11:46< celticminstrel> Okay, it's related to wesnoth.put_unit, seemingly.
20160314 22:12:22< celticminstrel> Incidentally, vultraz, it seems that Lua 5.3 changed LUAI_EXTRASPACE to LUA_EXTRASPACE.
20160314 22:12:32< vultraz> lua 5.3 changes a lot of names
20160314 22:12:55< celticminstrel> Well, that's one that hits the configuration header.
20160314 22:13:02< vultraz> I had to figure them out from a PR from another project when I did the 5.3.1 PR :P
20160314 22:17:36-!- louis94 [~~louis94@91.178.241.103] has joined #wesnoth-dev
20160314 22:22:09< zookeeper> err. i'm ignorant. what's the usual function/whatever we use to make a string from an int?
20160314 22:22:33< celticminstrel> In C++? I beleive that's str_cast().
20160314 22:22:35< celticminstrel> ^believe
20160314 22:23:40< zookeeper> yes, so it seems, thanks
20160314 22:24:16< vultraz> or lexical_cast<std::string>()
20160314 22:24:28< vultraz> or in c++11, to_string()
20160314 22:24:39< celticminstrel> I'd use str_cast and reserve lexical_cast for the reverse operation, probably.
20160314 22:24:43< shadowm> We don't  use C++11 and zookeeper clearly wants to write code for Wesnoth.
20160314 22:25:12< vultraz> we will soon use c++11
20160314 22:25:26< celticminstrel> I dunno about soon, but I hope you're right...
20160314 22:25:27< shadowm> According to who?
20160314 22:25:34< celticminstrel> Exactly which people are active here anyway?
20160314 22:25:37< vultraz> Me
20160314 22:25:39< vultraz> According to me
20160314 22:25:42< vultraz> I shall make it happen
20160314 22:25:48< shadowm> I was hoping for someone with a technical background.
20160314 22:26:12< celticminstrel> As devs, not other roles.
20160314 22:26:17< celticminstrel> There's vultraz, gfgtdf, zookeeper, iceiceice, maybe shadowm.
20160314 22:26:28< vultraz> and Aginor :P
20160314 22:26:30< vultraz> and you
20160314 22:26:31< shadowm> As opposed to some random overzealous neophyte, I mean.
20160314 22:26:33< celticminstrel> I guess mattsc counts too.
20160314 22:26:45< celticminstrel> shadowm: Well, I want it to happen too.
20160314 22:27:03< shadowm> If you knew Ruby you'd probably go and try to force people to use Ruby too without really caring about making an educated decision. <_<
20160314 22:27:09< celticminstrel> Oh right, loonycyborg too.
20160314 22:27:27< celticminstrel> And wedge009 and SigurdFD.
20160314 22:27:37< celticminstrel> Though I think the latter is more a WML person?
20160314 22:27:41< vultraz> yeah
20160314 22:27:41< shadowm> iceiceice isn't active, celticminstrel.
20160314 22:27:43< celticminstrel> So maybe doesn't count for this.
20160314 22:27:53< shadowm> Neither am I.
20160314 22:27:58< celticminstrel> shadowm: Well, he's active enough to show up and chat frequently... >_>
20160314 22:28:19< shadowm> Active and present are completely different things.
20160314 22:28:45< celticminstrel> So that totals seven or eight, depending whether you count wedge009.
20160314 22:29:03< shadowm> Though I will say that if vultraz makes a decision that leads to wesnothd being unbuildable on baldras I will not be impressed, because I'm still in charge of that.
20160314 22:29:19< celticminstrel> And I think only 2-4 of those use MSVC.
20160314 22:29:48< celticminstrel> What's the setup on baldras?
20160314 22:30:27< shadowm> On the wesnothd VM we're (since February) running Debian jessie, whose default compiler is gcc 4.9. That should cover most of the C++11 requirements, I believe.
20160314 22:30:51< shadowm> On basilic (server3) I don't know what we are running.
20160314 22:31:17< shadowm> Debian wheezy.
20160314 22:31:37< shadowm> gcc 4.7.
20160314 22:31:39< celticminstrel> GCC4.9 appears to fully support C++11.
20160314 22:31:54< shadowm> But I'm not too concerned about losing server3 since we already lost server2.
20160314 22:32:49< celticminstrel> 4.7 misses explicit virtual overrides, extended friend declarations, user-defined literals, inheriting constructors, alignment support, generalized attributes, template aliases, something about decltype that I don't quite understand, non-static data member initializers, and rvalue *this references.
20160314 22:33:11< celticminstrel> And one or two others that aren't really important, like thread-local storage.
20160314 22:33:18< shadowm> Inheriting constructors are a big thing.
20160314 22:33:34< shadowm> As well as the override keyword.
20160314 22:33:42< celticminstrel> I think my own compiler misses some of those, particularly generalize attributes, but inheriting constructors and non-static data member initializers are certainly big misses.
20160314 22:34:09< celticminstrel> Oh, "explicit virtual overrides" refers to the override keyword? Then that's a big thing too.
20160314 22:34:20< shadowm> All that said, I have said this a million times to vultraz, my bigger concern has always been the MSVC++ crowd (zookeeper and gfgtdf).
20160314 22:34:29< vultraz> I know
20160314 22:34:33< celticminstrel> Ah, so zookeeper does use MSVC.
20160314 22:34:33< vultraz> I'm just waiting for them
20160314 22:34:33< zookeeper> <3
20160314 22:34:37< celticminstrel> loonycyborg does too, right?
20160314 22:34:44< shadowm> No, loonycyborg uses GCC.
20160314 22:34:48< celticminstrel> Ah, okay.
20160314 22:34:50< vultraz> zookeeper: do you run a c++11 compatible version?
20160314 22:34:59< celticminstrel> Which version do you run?
20160314 22:34:59< zookeeper> as far as i know i have no problem with c++11, since i've compiled c++11 stuff before
20160314 22:35:06< shadowm> IIRC he moved to Mingw-w64 recently, so here's presumably running GCC 4.9 or later.
20160314 22:35:15< shadowm> loonycyborg, that is.
20160314 22:35:27 * zookeeper runs msvc 2013 community edition
20160314 22:35:39< celticminstrel> Then you have a lot of C++11 features already.
20160314 22:36:18< shadowm> aqulieia used to be a point of concern too, until he kind of vanished.
20160314 22:36:21< celticminstrel> I have MSVC 2013 ultimate, and it can compile my C++11 code with only minor tweaks (for example it doesn't work on deleted move constructors).
20160314 22:36:53< zookeeper> right, so it presumably doesn't have full support
20160314 22:36:59< celticminstrel> Right, it doesn't.
20160314 22:37:00< zookeeper> which could or could not become a problem
20160314 22:37:09< vultraz> you can get vc15 community for free
20160314 22:37:17< vultraz> if it's really necessary 
20160314 22:37:18< celticminstrel> So we'd have to limit our C++11 usage by what MSVC 2013 supports.
20160314 22:37:33< vultraz> so it's just gfgtdf
20160314 22:37:38< celticminstrel> What's his version?
20160314 22:37:46< zookeeper> yeah if i could just install a different version to get things working again, then sure i would
20160314 22:37:53< vultraz> I seem to recall something about 2010...
20160314 22:37:57< celticminstrel> Ah.
20160314 22:39:07< shadowm> I was a proponent of "deploying" C++11 piecewise to avoid forcing people to upgrade compilers, by the way, rather than allowing the full featureset to be used right away.
20160314 22:39:42< shadowm> ("deploying" is the closest term I could think of, it's really just a matter of flipping a switch and telling people what features they can use, and have them start cleaning up code.)
20160314 22:39:43< celticminstrel> Does MSVC2010 support any C++11?
20160314 22:39:48< vultraz> some
20160314 22:39:54< celticminstrel> How much?
20160314 22:39:58< vultraz> not sure
20160314 22:40:04< shadowm> iceiceice was a proponent of the "all or nothing" approach instead.
20160314 22:40:53 * zookeeper wrote a wonderful piece of terrain graphics debugging code which renders a number of each hex telling you how many images there are on it
20160314 22:41:08< vultraz> niice
20160314 22:41:21< celticminstrel> Okay, so MSVC2010 sounds like it supports most of the biggest ones - auto, decltype, lambdas, static_assert, and nullptr.
20160314 22:41:27< shadowm> vultraz: http://lmgtfy.com/?q=msvc2010+C%2B%2B11+status
20160314 22:41:29< celticminstrel> And rvalue references.
20160314 22:41:45< celticminstrel> I don't know if that's the full list.
20160314 22:42:55< shadowm> MSVC++ 2010 doesn't support initializer_list, it sucks.
20160314 22:42:55< celticminstrel> Its standard library has new algorithms, unique_ptr, shared_ptr, forward_list (not important), cbegin et al, and a few other useful C++11 things.
20160314 22:43:04< celticminstrel> Yeah, that does.
20160314 22:43:14< celticminstrel> Doesn't support variadic templates either, I guess.
20160314 22:43:37< celticminstrel> Which means no emplace_back et al.
20160314 22:43:48< shadowm> No range-for either!
20160314 22:43:52< celticminstrel> :(
20160314 22:43:59< shadowm> That's even worse than just sucks.
20160314 22:44:08< celticminstrel> I think I found a different list than you did.
20160314 22:44:21< shadowm> I'm looking at this one: https://msdn.microsoft.com/en-us/library/hh567368.aspx
20160314 22:44:37< shadowm> It's on MSDN, so it must be pretty authoritative.
20160314 22:44:51< celticminstrel> Oh, four versions all in one nice convenient table.
20160314 22:45:06< shadowm> Yup.
20160314 22:45:11< celticminstrel> 2015 is still missing one C++11 feature.
20160314 22:46:51-!- SpoOkyMagician [~chatzilla@2607:fcc8:be59:b00:d5fb:e8aa:3e47:6b94] has joined #wesnoth-dev
20160314 22:46:52< shadowm> I think expression SFINAE is only of concern for people who indulge more in metaprogramming though?
20160314 22:46:58< celticminstrel> Also doesn't ahve full C99 preprocessor.
20160314 22:47:00< celticminstrel> Yeah probably.
20160314 22:47:04< celticminstrel> ^have
20160314 22:47:07< shadowm> Without mordante around, I rather doubt anyone will miss it.
20160314 22:47:14< celticminstrel> Might be right.
20160314 22:47:43-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160314 22:49:44< shadowm> Our C code (Lua) targets C89 and our existing C++ code targets C++98, so the lack of complete C99 preprocessor support isn't a concern either.
20160314 22:50:11< shadowm> (Side note: Microsoft has never really cared much about C99 for some reason.)
20160314 22:52:42< zookeeper> hurr, durr... if my aforementioned code is to be trusted, there's a lot awfulness in the terrain rules
20160314 22:54:14< shadowm> Hm, the C++11 attribute syntax is supported first by MSVC++ 2015? :\
20160314 22:54:47< shadowm> I guess they didn't care too much about that because people can use macros to select the MSVC++ attribute syntax.
20160314 22:56:08< celticminstrel> Only really nice thing I know of C99 preprocessor is variadic macros.
20160314 22:56:41< celticminstrel> zookeeper: It wouldn't surprise me. :)
20160314 22:56:55< shadowm> "That covers the Core Language. As for the C++11 Standard Library, we don't have a pretty comparison table of features, but Visual C++ in Visual Studio 2012 implemented it, with two exceptions."
20160314 22:57:26< celticminstrel> Heh, there are two yes's on the C++17 table.
20160314 22:57:35< celticminstrel> "Removing trigraphs" is one of them.
20160314 22:57:56< celticminstrel> I'm sure every compiler supports that at this point. :P
20160314 22:59:31-!- gfgtdf [~chatzilla@f050176167.adsl.alicedsl.de] has joined #wesnoth-dev
20160314 22:59:47< shadowm> I feel the C++11 table ought to mention dropping `export` since that wouldn't require any effort of Microsoft's part.
20160314 22:59:57-!- mjs-de [~mjs-de@wh.Uni-Dortmund.DE] has quit [Ping timeout: 244 seconds]
20160314 23:00:05< celticminstrel> They probably never supported it.
20160314 23:00:12< celticminstrel> I think only one compiler ever did.
20160314 23:00:13< shadowm> No-one did.
20160314 23:00:15< zookeeper> 1 hex of water in the middle of grass consists of 23 images -.- it does add up so there's nothing weird about it, but seeing the number like that... urgh.
20160314 23:00:26< shadowm> Yes, one compiler vendor went to the trouble of implementing export.
20160314 23:00:43< shadowm> They were amongst the people who proposed dropping it for C++11 IIRC.
20160314 23:00:44< celticminstrel> zookeeper: Wow. o.o
20160314 23:01:16< celticminstrel> What the heck is a SCARY iterator... it sounds scary...
20160314 23:01:49< shadowm> So the thing that vultraz is most obsessed with isn't documented, alas.
20160314 23:01:58< celticminstrel> Which thing is that?
20160314 23:02:08< shadowm> C++11 standard library  support.
20160314 23:02:16< celticminstrel> Oh.
20160314 23:02:33< shadowm> "Visual C++ in Visual Studio 2012 implemented it, with two exceptions. First, when a library feature depended on functionality that was missing in the compiler, it was either simulated—for example, simulated variadic templates for make_shared<T>()—or it wasn't implemented. (There were only a few cases—most notably, <initializer_list>—which were fully implemented in Visual C++ in Visual Studio 2013.)"
20160314 23:02:42< gfgtdf> vultraz: about the lua jailbreak exception thing: this shodul actually be reworked. Iirc the problem is that the currently the lua_cjailbreak exception is a staic variable but since we have athe application lua kernel we might support more lua_states at once so it woudl be better o have on per lua_state
20160314 23:03:03< celticminstrel> https://msdn.microsoft.com/en-us/library/dd465215.aspx - scroll to "Visual C++ Libraries"
20160314 23:03:03-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160314 23:03:09< celticminstrel> Not complete, but does give some idea.
20160314 23:03:27< celticminstrel> That's for MSVC 2010.
20160314 23:03:51-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160314 23:04:41< shadowm> Huh, MFC is still alive?
20160314 23:04:51< shadowm> Or was back then, anyway.
20160314 23:04:53< celticminstrel> No idea.
20160314 23:05:04< celticminstrel> I think it still exists by 2013.
20160314 23:05:22< shadowm> I was under the impression that people had already stopped caring about it circa 2002.
20160314 23:05:27< gfgtdf> Aginor: i woner why this commit https://github.com/wesnoth/wesnoth/commit/358bd1235f07d8d577bdb2bb0b4aafcff2e93093 changes src/lua/luaconf.h ?
20160314 23:05:27< celticminstrel> Though I'm not entirely sure of that either, because I was avoiding using it.
20160314 23:05:44< celticminstrel> gfgtdf: What compiler version do you use?
20160314 23:05:53< gfgtdf> celticminstrel: still 2010
20160314 23:05:59< celticminstrel> 'kay
20160314 23:06:38< gfgtdf> celticminstrel: i want to upgrade to 2015 but im unsure abut whether i shoudl download commity or ask university for a ultimate version.
20160314 23:06:43< celticminstrel> gfgtdf: I suspect that's because that was a dead preprocessor section - note how it's "0 && something".
20160314 23:06:59< shadowm> celticminstrel: Yeah, but it still doesn't belong in that commit.
20160314 23:07:23< celticminstrel> shadowm: Agreed, but Aginor used a tool to do that, so it's not too surprising that it slipped through the cracks.
20160314 23:07:53< shadowm> I'd be a bit concerned if the tool did it.
20160314 23:07:53< celticminstrel> I must've missed it too when going through the diff.
20160314 23:08:35< celticminstrel> shadowm: I dunno, if the goal of the tool is "remove all bits of code that are never reached if these defines are set to this), then I'd say it did exactly what it was told to.
20160314 23:09:16< shadowm> But it didn't remove the other #if 0 blocks that we have.
20160314 23:09:24< celticminstrel> Huh, really? Okay then.
20160314 23:09:30< shadowm> Or maybe he had to filter those out by hand, I don't know.
20160314 23:09:44< celticminstrel> Or maybe it has a special exception for plain "#if 0"
20160314 23:10:04< celticminstrel> Since that's a common idiom for "commenting out" a bit of code.
20160314 23:10:53< shadowm> (Cue vultraz proposing dropping every #if 0 block in existence.)
20160314 23:11:26< shadowm> (In all seriousness, no, don't do that. Some of them are debugging hints or WIP code.)
20160314 23:22:06< vultraz> I propose dropping #if LOW_MEM
20160314 23:22:11< vultraz> #ifdef*
20160314 23:22:18< vultraz> proposed*
20160314 23:22:32< celticminstrel> If I'm lucky, that might fix my TC issue (but probably not).
20160314 23:22:47< celticminstrel> ...wait, there isn't an actual option in prefs to completely disable TC, is there?
20160314 23:23:03< celticminstrel> Or IPF or whatever.
20160314 23:23:08< vultraz> There is
20160314 23:23:10< vultraz> Show Team Colors
20160314 23:23:28< celticminstrel> No, I'm pretty sure that only affects whether an ellipse is shown on non-selected units.
20160314 23:23:42< vultraz> ah, yes
20160314 23:24:01< vultraz> god, the new tooltips look SO DAMN GOOD :D
20160314 23:36:22-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160314 23:39:37-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 240 seconds]
20160314 23:39:39-!- wedge010 is now known as wedge009
20160314 23:42:01-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160314 23:44:43-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160314 23:53:29< irker239> wesnoth: ln-zookeeper wesnoth:master 30bd1c132445 / data/core/images/terrain/water/ (reef.png reef2.png reef3.png reef4.png): Cleaned up stray pixels and fixed alignment of reef4 https://github.com/wesnoth/wesnoth/commit/30bd1c1324450a6931f45e5fa6c97c485ab7de17
20160314 23:59:46-!- mjs-de [~mjs-de@x5ce490ca.dyn.telefonica.de] has joined #wesnoth-dev
--- Log closed Tue Mar 15 00:00:09 2016