--- Log opened Fri Sep 25 00:00:04 2015
--- Day changed Fri Sep 25 2015
20150925 00:00:04-!- [Relic] [~Relic]@2602:306:33a3:6d30:e10f:acfe:840c:2c05] has joined #wesnoth-dev
20150925 00:01:01-!- Guest18895 [~quassel@2604:a880:800:10::1c2:b001] has quit [Changing host]
20150925 00:01:01-!- Guest18895 [~quassel@wikipedia/Legoktm] has joined #wesnoth-dev
20150925 00:01:56-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20150925 00:10:05< celticminstrel> It helps that several of these have a class/namespace name that matches the filename.
20150925 00:10:46-!- Guest18895 is now known as legoktm
20150925 00:35:23-!- fabi [~quassel@176.0.79.57] has joined #wesnoth-dev
20150925 00:35:23-!- fabi [~quassel@176.0.79.57] has quit [Changing host]
20150925 00:35:23-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20150925 00:42:03< celticminstrel> Well, number of link errors is now under 200.
20150925 00:55:42-!- prkc [~prkc@catv-89-134-159-103.catv.broadband.hu] has quit [Remote host closed the connection]
20150925 00:59:21-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20150925 01:08:11-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20150925 01:16:02< Aginor> celticminstrel: sounds like progress :)
20150925 01:17:57< celticminstrel> Yes, because 200 is the cut-off point before it gives up and stops reporting them, so that means that (probably) every single link error is now listed. ...except I fixed some and it's still 198 (same as before). :|
20150925 01:18:36< Aginor> ah :/
20150925 01:18:59< Aginor> maybe a good time to check out cmake's xcode genration and see if it's been fixed? :D
20150925 01:19:22< celticminstrel> Meh.
20150925 01:19:59-!- danni [~quassel@adsl-au-4-115.ozonline.com.au] has quit [Ping timeout: 264 seconds]
20150925 01:21:16-!- danni [~quassel@adsl-au-4-115.ozonline.com.au] has joined #wesnoth-dev
20150925 01:25:11-!- irker327 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20150925 01:25:11< irker327> wesnoth: Andreas Löf wesnoth:sdl2 5bd53b9eaeb3 / src/video.cpp: Fix CPP unit tests under SDL2 http://git.io/vn5ch
20150925 01:25:36< celticminstrel> Yay!
20150925 01:25:55< Aginor> still need to figure out what's breaking WML tests though
20150925 01:26:02< Aginor> that's going to be a lot harder
20150925 01:26:10< celticminstrel> Is it?
20150925 01:26:44< Aginor> I have no idea what might be causing it or even where to start debugging the problem :)
20150925 01:26:56< Aginor> ./wesnoth -u alice_kills_bob_levelup --validcache --log-strict=debug --noaddons
20150925 01:27:00< Aginor> Error (strict mode, strict_level = 3): wesnoth reported on channel error engine
20150925 01:27:03< Aginor> 20150925 13:25:48 error engine: unit at 13,4tried to advance more than 20 times.
20150925 01:27:10< Aginor> that seems to be the relevant bits
20150925 01:27:16< Aginor> but it's not telling me much
20150925 01:27:19< Aginor> (and the test fails)
20150925 01:28:53< celticminstrel> Gonna need to ask mattsc or whoever to add Boost.Test to the downloadable set of libraries.
20150925 01:29:20< celticminstrel> I think I'd better not shove my own copy in since it's 1.57 while Wesnoth is using Boost 1.54.
20150925 01:29:41< gfgtdf> Aginor: since that problem only occurs in you SDl branch i think its relates to the advancement choice window
20150925 01:31:22< Aginor> gfgtdf: advancement choice window?
20150925 01:31:36< Aginor> celticminstrel: better ask for SDL2_* too while you're at it :)
20150925 01:31:45< celticminstrel> Good point.
20150925 01:32:14< gfgtdf> Aginor: thhat message you posted comes from here: https://github.com/wesnoth/wesnoth/blob/master/src/actions/attack.cpp#L1469
20150925 01:32:30< gfgtdf> Aginor: you know whether taht errormessage also is on the non-sdl2 branch ?
20150925 01:32:46< Aginor> gfgtdf: yes, I've managed to grep and find out as much myself
20150925 01:32:54< Aginor> gfgtdf: no, it's not present on sdl1.2
20150925 01:33:54< gfgtdf> Aginor: and afaik the only part of taht code that coudl interact with your changes is the advance_unit_dialog function and rthe non_interactive() function
20150925 01:34:42< Aginor> I think I've found it
20150925 01:34:53< Aginor> bool non_interactive()
20150925 01:34:53< Aginor> {
20150925 01:34:53< Aginor> 	if (fake_interactive)
20150925 01:34:53< Aginor> 		return false;
20150925 01:34:53< Aginor> #if SDL_VERSION_ATLEAST(2, 0, 0)
20150925 01:34:55< Aginor> 	return false;
20150925 01:34:58< Aginor> #else
20150925 01:35:00< Aginor> 	return SDL_GetVideoSurface() == NULL;
20150925 01:35:03< Aginor> #endif
20150925 01:35:05< Aginor> that'd break stuff
20150925 01:35:08< Aginor> thanks gfgtdf 
20150925 01:37:15< Aginor> the test is now passing :)
20150925 01:43:39< gfgtdf> Aginor: ok
20150925 01:43:52-!- gfgtdf [~chatzilla@f054129203.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 41.0/20150917150946]]
20150925 01:55:29< vultraz> shikadibot: seen iceiceice
20150925 01:55:30< shikadibot> vultraz: The person with the nick iceiceice last spoke 9d 4h ago.  47m 18s ago they left with the message: Quit: Leaving
20150925 01:55:50< irker327> wesnoth: Andreas Löf wesnoth:sdl2 85dad75cc0d6 / src/video.cpp: Fix the non_interactive() function used in WML tests http://git.io/vn5Bn
20150925 01:56:01< Aginor> and now I wait for travis to trundle through
20150925 02:10:17-!- travis-ci [~travis-ci@ec2-54-81-35-112.compute-1.amazonaws.com] has joined #wesnoth-dev
20150925 02:10:18< travis-ci> wesnoth/wesnoth#7485 (sdl2 - 5bd53b9 : Andreas Löf): The build is still failing.
20150925 02:10:18< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/82077207
20150925 02:10:18-!- travis-ci [~travis-ci@ec2-54-81-35-112.compute-1.amazonaws.com] has left #wesnoth-dev []
20150925 02:10:48< celticminstrel> Trundle through indeed.
20150925 02:12:38< celticminstrel> Ooh! Link errors stopped at 197 this time instead of 198. I wonder if it means anything.
20150925 02:17:15< celticminstrel> Now it's starting to list missing vtables and global symbols (eg main).
20150925 02:17:48< Aginor> I'd help, but I have no idea how the xcode project is structured
20150925 02:17:57< Aginor> but it sounds like you're missing files :)
20150925 02:17:59< celticminstrel> Very poorly.
20150925 02:18:14< celticminstrel> The files aren't missing, just not set to be compiled as part of the unit tests.
20150925 02:21:29< vultraz> Is the sdl2 branch nearing completion, BTW?
20150925 02:22:19< Aginor> vultraz: it's getting there
20150925 02:22:58< Aginor> vultraz: I think I managed to solve the absolutely biggest problem yesterday but there's still a few things to do
20150925 02:23:20< vultraz> Ok
20150925 02:23:28< vultraz> I should start looking for sdl2 libs, then...
20150925 02:24:03< Aginor> might be a good idea :)
20150925 02:29:14< Aginor> hmm. still not good in travis
20150925 02:29:37< Aginor> it looks like there's a segfault when running cpptests
20150925 02:30:25< Aginor> but wml tests are looking promising now
20150925 02:31:58< vultraz> I wonder if this would be a time to try to get windows 64 bit wesnoth running
20150925 02:32:48< Aginor> that shouldn't be too hard, as long as you get the dependencies in 64-bit too
20150925 02:32:57< celticminstrel> Yay.
20150925 02:33:25< celticminstrel> I wouldn't think 64-bit is urgent, as 32-bit programs run fine on 64-bit processors.
20150925 02:34:00< vultraz> Not urgent, no
20150925 02:34:14< vultraz> But it'd be nice, since OS X only has 64 bit 
20150925 02:34:17< vultraz> Consistency and all
20150925 02:35:32< vultraz> Big hurdle is pango, though
20150925 02:35:42< vultraz> Well
20150925 02:35:48< vultraz> I can just use the 1.30
20150925 02:37:02< Aginor> vultraz: you'd need to maintain both 32 and 64 bit for windows then
20150925 02:37:19< Aginor> so you in effect add another platform with not much gain
20150925 02:37:35< celticminstrel> OSX still supports 32-bit, right?
20150925 02:37:59< celticminstrel> Admittedly I'm three versions behind so I might not be up on the latest...
20150925 02:45:00-!- travis-ci [~travis-ci@ec2-54-81-35-112.compute-1.amazonaws.com] has joined #wesnoth-dev
20150925 02:45:01< travis-ci> wesnoth/wesnoth#7486 (sdl2 - 85dad75 : Andreas Löf): The build was fixed.
20150925 02:45:01< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/82079223
20150925 02:45:01-!- travis-ci [~travis-ci@ec2-54-81-35-112.compute-1.amazonaws.com] has left #wesnoth-dev []
20150925 02:45:17< celticminstrel> Yay!
20150925 02:45:40< celticminstrel> Ooh, my link errors are finally under 100.
20150925 02:50:33< Aginor> and I've had a travis build pass :)
20150925 03:03:11-!- iceiceice [~chris@ext-74.ias.edu] has joined #wesnoth-dev
20150925 03:03:12-!- iceiceice [~chris@ext-74.ias.edu] has quit [Changing host]
20150925 03:03:12-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20150925 03:07:30< celticminstrel> ...leader_scroll_dialog.cpp contains... status_table and scenario_settings_table.
20150925 03:07:32< celticminstrel> Why...
20150925 03:07:47-!- [Relic] [~Relic]@2602:306:33a3:6d30:e10f:acfe:840c:2c05] has quit [Quit: I press the magic X and all the weirdos go away!]
20150925 03:12:32< iceiceice> vultraz: howdy
20150925 03:19:58< celticminstrel> Alright! Down to 18 missing symbols, and 17 of them are in boost::unit_test.
20150925 03:20:10< celticminstrel> The other one is main, which I'm ignoring in case Boost.UnitTest defines it for you.
20150925 03:20:46< celticminstrel> Oh, 17 includes main. Its error count is always off by one, because it includes "Linker command failed with exit code 1" as an error.
20150925 03:26:51< celticminstrel> I like how my scroll wheel can be allowed to spin freely and thus go ridiculously fast.
20150925 03:33:34< vultraz> hey iceiceice 
20150925 03:36:23< vultraz> iceiceice: I can't seem to get strict lua to turn of on master
20150925 03:37:44< iceiceice> what did you try?
20150925 03:38:17< vultraz> ilua.strict = false
20150925 03:41:39< iceiceice> vultraz: i don't really remember how it works, you could try to trouble shoot it by putting debug output in the ilua script
20150925 03:42:14< iceiceice> i guess maybe its better to have an "ilua.deactivate()" function
20150925 03:42:25< iceiceice> so you have to explicitly call something so it will check for typos
20150925 03:43:50-!- [Relic] [~Relic]@2602:306:33a3:6d30:55ea:de86:fb63:d12a] has joined #wesnoth-dev
20150925 03:44:59< vultraz> It's confusing though, since the actual strict checking is in set_strict() yet that's not called from anywhere
20150925 03:45:51< irker327> wesnoth: Andreas Löf wesnoth:sdl2 f89056487e51 / src/sdl/window.cpp: Add fullscreen support to SDL2 http://git.io/vn5Du
20150925 03:47:32< vultraz> fullscreen!
20150925 03:47:44< Aginor> yes!
20150925 03:48:07< Aginor> another few-line fix
20150925 03:51:12< vultraz> Does the window icon look any less garbled for you in sdl2?
20150925 03:51:31< Aginor> which icon?
20150925 03:52:40< vultraz> The one set with SDL_SetWindowIcon()
20150925 03:52:47< celticminstrel> OSX doesn't have one.
20150925 03:52:50< celticminstrel> I think.
20150925 03:53:15< celticminstrel> Windows shows it on the titlebar - it's what you click to get the system menu.
20150925 03:53:28< celticminstrel> Maybe OSX could show it in the titlebar - it does have space for an icon there. Not sure.
20150925 03:53:48< Aginor> vultraz: yes, it looks good for me
20150925 03:55:22< vultraz> could I see a screenshot to compare?
20150925 03:55:50< vultraz> or wait, I will
20150925 03:56:38< vultraz> Aginor: better than this? https://www.dropbox.com/s/4b8m8s2t33k1szk/windowicon.PNG?dl=0
20150925 04:03:27< Aginor> http://www.alternating.net/agi/wesnoth.png
20150925 04:03:36< Aginor> you decide
20150925 04:04:32< Aginor> also, my window manager doesn't use any icons in the windows, there's only the "task bar" one
20150925 04:05:08< vultraz> holy shit do you work on  50" TV or something O_O
20150925 04:05:20< Aginor> no :D
20150925 04:05:31< Aginor> two 22" monitors running 1680x1050
20150925 04:06:36-!- knotwork [~markm@unaffiliated/knotwork] has quit [Read error: No route to host]
20150925 04:09:30< vultraz> dat space tho
20150925 04:10:06< Aginor> it's nice, but I've been thinking about upgrading to newer monitors
20150925 04:13:04< vultraz> I have also been considering that
20150925 04:13:16< vultraz> Since I'm currently a 20" 1600x900 peasant :P
20150925 04:13:27< vultraz> (single monitor setup)
20150925 04:13:33< Aginor> 2 monitors is where it's at
20150925 04:13:44< Aginor> although I might go 3 next time :)
20150925 04:14:09< Aginor> just make sure it's all the same size and bought from the same batch so they have the same colours and brightnesses
20150925 04:14:51< Aginor> hmm, segmentation fault in /usr/src/debug/SDL2-2.0.3/src/video/SDL_blit_A.c:350
20150925 04:15:00< Aginor> not ideal :D
20150925 04:17:27< vultraz> I don't have space for 2 :P
20150925 04:17:36< Aginor> get a bigger desk :D
20150925 04:17:37< vultraz> Or a powerful enough system
20150925 04:17:47< vultraz> I'm running this off my laptop since its screen blacked out
20150925 04:17:56< vultraz> And that only has an iHD4k
20150925 04:18:07< Aginor> mmm
20150925 04:18:14< Aginor> I have a humble little desktop machine
20150925 04:40:25-!- Shackra [~Jorge@186.177.2.148] has quit [Read error: Connection reset by peer]
20150925 04:43:22-!- knotwork [~markm@unaffiliated/knotwork] has joined #wesnoth-dev
20150925 04:45:45-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150925 04:55:12< Aginor> hmm, gui1 text input will be slightly trickier than I thought there's refactoring needed....
20150925 04:57:46< vultraz> Sad that you have to refactor something slated  for refactoring out anyway :(
20150925 04:58:28< celticminstrel> If no-one shows up to maintain GUI2 it could be a long time before GUI1 is removed.
20150925 04:59:22< vultraz> You mean if no one shows up with the ability to refactor GUI2 
20150925 05:04:30-!- oldlaptop [~quassel@50.36.238.180] has quit [Ping timeout: 240 seconds]
20150925 05:04:43< celticminstrel> Yeah that.
20150925 05:14:07-!- kinow [~kinow@apache/committer/kinow] has joined #wesnoth-dev
20150925 05:15:12-!- oldlaptop [~quassel@50.36.238.180] has joined #wesnoth-dev
20150925 05:20:50< Aginor> from what I've looked at GUI2, it's actually pretty sensible
20150925 05:21:23< Aginor> it's convoluted, but I think it's very hard to make a flexible GUI without making it convoluted
20150925 05:34:51< vultraz> The problem is it can't handle anything but windows 
20150925 05:35:01< vultraz> that for some reason were designed to totally stop any other action
20150925 05:36:06< Aginor> vultraz: I'm not sure I follow what you're trying to say, could you please clarify?
20150925 05:36:29< vultraz> When a GUI2 dialog is open, all other interactions stop
20150925 05:36:41< vultraz> Even passive actions such as animated terrains
20150925 05:36:43< vultraz> They freeze
20150925 05:36:48< Aginor> ah
20150925 05:36:54< vultraz> GUI2 completely takes control of the game display
20150925 05:37:02< Aginor> that should be fixable though, presumably
20150925 05:37:04< vultraz> And relegates it totally to the open window
20150925 05:37:06< vultraz> Yes
20150925 05:37:14< vultraz> It also functions totally in terms of 'windows'
20150925 05:37:24< Aginor> by window, you mean dialog?
20150925 05:37:30< vultraz> The experimental gui2 MP lobby is in fact a borderless window
20150925 05:37:46< Aginor> (I've never seen that)
20150925 05:38:04-!- celticminstrel is now known as celmin|sleep
20150925 05:38:27< celmin|sleep> The title screen is also a GUI2 dialog.
20150925 05:38:48< vultraz> The titlescreen is also a borderless window, I believe
20150925 05:39:03< celmin|sleep> What vultraz is complaining about is basically that it can only handle modal windows.
20150925 05:39:08< celmin|sleep> No modeless windows.
20150925 05:41:09< vultraz> Basically, yes
20150925 05:41:25< Aginor> yup...
20150925 05:41:29< vultraz> In order for GUI2 to become our UI manager, it will need to A: not hog the entire gamestate and display
20150925 05:41:58< vultraz> B: support different types of ... windows or whatever you want to call it
20150925 05:42:02< Aginor> vultraz: A is actually not too hard, don't feed all events to GUI2, spread them around
20150925 05:42:16< Aginor> B is rather unclear though :)
20150925 05:42:32< vultraz> It needs to be able to replace the in-game UI
20150925 05:42:39< vultraz> currently set up as horribly hacky 'themes'
20150925 05:42:51< Aginor> yeah...
20150925 05:43:11< fabi> vultraz: Note that theme wml is by far easier than gui2 wml.
20150925 05:43:27< vultraz> fabi: I disagree
20150925 05:43:30< vultraz> rect is hell
20150925 05:43:58< fabi> vultraz: grid is hell
20150925 05:44:21< vultraz> Grid doesn't involve arbitrary magic numbers 
20150925 05:44:45< Aginor> layout managers are tricky no matter what
20150925 05:45:00< vultraz> True dat
20150925 05:45:01< Aginor> and I'm sorry to have started a fklamewar with my innocent question
20150925 05:45:12< vultraz> nah, it's ok
20150925 05:45:14< fabi> gui2 is not even giving you usable error reports if something went wrong with the gui2 wml.
20150925 05:45:35< vultraz> fabi: that's something that will have to be fixed
20150925 05:45:43< vultraz> we just need someone able to doi t
20150925 05:45:47< vultraz> do it
20150925 05:46:00< fabi> vultraz: No, gui2 should just be removed and replaced with a ready to use gui system.
20150925 05:46:01< Aginor> vultraz: how's you c++ coming along? :D
20150925 05:46:27< vultraz> Aginor: ironically, GUI2 fits well into wesnoth's 'one action at a time' system. IE, there's no way to have simultaneous WML events or actions fire at once.
20150925 05:47:16< vultraz> Aginor: uh... not really good xD
20150925 05:47:27< vultraz> I'm getting better at Lua, though
20150925 05:48:04< Aginor> fabi, vultraz; Replacing the existing way with something different will also be extremely painful, the GUI code has deep roots. Continuing the rewrite to GUI2 will also be painful. The entire state of GUIness is between two uncomfortable places.
20150925 05:48:58< Aginor> vultraz: I've found a perfect place for you to practice more c++ though :D :D :D
20150925 05:50:25< vultraz> I know how stuff generally works, but stuff like objects, inheritance, pointers and references all confuse the hell out of me 
20150925 05:52:40< vultraz> I just don't... *get* it
20150925 05:52:41< vultraz> yet
20150925 05:53:17< fabi> Aginor: Well, writting a gui system is a bigger task and project than a game like Wesnoth. 
20150925 05:53:32< fabi> We are talking that speech for more than 5 years now.
20150925 05:53:39< fabi> I bet the situation will never change.
20150925 05:54:59< Aginor> fabi: have you found a good toolkit that won't force a complete rewrite of all code that touches graphics?
20150925 05:55:54< fabi> Aginor: Why would that be necessary?
20150925 05:56:17< Aginor> fabi: because it's all pretty tightly coupled at the moment
20150925 05:56:57< Aginor> the entire game window is one big bitmap that is drawn upon in various orders, and if you break that paradigm you break the game
20150925 05:57:19< iceiceice> Aginor: i thnk there are SDL-based gui tool-kits available
20150925 05:57:38< iceiceice> Wesnoth's gui is not so integrated with it's graphics as some games, in that the gui never really gets under the units or anything
20150925 05:57:51< iceiceice> unless i'm mistaken the gui can more or less just be drawn on top of the SDL-based game graphics
20150925 05:58:07< iceiceice> they could even be compeltely different graphics pipelines most likely
20150925 05:59:24< Aginor> iceiceice: wesnoth has done some unorthodox decisions in the past that will also make these things hard.
20150925 05:59:33< Aginor> however, if itäs SDL based it's more feasible
20150925 05:59:41< Aginor> http://libagar.org/ may be an option
20150925 06:01:01< iceiceice> Aginor: i think the most significant cost is that
20150925 06:01:12< iceiceice> we already have a huge amount of stuff implemented in the gui2 system
20150925 06:01:20< iceiceice> and reimplementing all of that is just a stupid amount of work
20150925 06:01:26< Aginor> yeah, I agree
20150925 06:01:29< iceiceice> even if we could get a new system in place without breaking the graphics
20150925 06:01:48< Aginor> it's a mess :/
20150925 06:01:52< iceiceice> yes
20150925 06:01:57< Aginor> and not something that will be fixed today
20150925 06:02:03< iceiceice> yes
20150925 06:02:03< Aginor> but...
20150925 06:02:03< vultraz> No
20150925 06:02:06< vultraz> Eventually
20150925 06:02:09< Aginor> let's start to make a plan
20150925 06:02:19< vultraz> There are plans to use the Anura engine for wesnoth 2
20150925 06:02:27< Aginor> figure out what the proper options are, the upsides, downsides and so on
20150925 06:02:33< vultraz> I believe iceiceice had some qualms about this, though
20150925 06:02:42< vultraz> Again, still up in the air until Frogatto is finished
20150925 06:03:24-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Ping timeout: 250 seconds]
20150925 06:03:26< iceiceice> vultraz: did dave make any kind of announcement about wesnoth 2?
20150925 06:03:45< vultraz> No
20150925 06:03:55< iceiceice> you know he made some kind of repo, right?
20150925 06:04:11< vultraz> He spoke about it a few days ago, though
20150925 06:04:16< iceiceice> but there hasn't been a whole lot of development
20150925 06:04:18< iceiceice> what did he say?
20150925 06:04:42< vultraz> TO look into after frogatto, basically
20150925 06:05:46< iceiceice> Aginor: fabi and i had a thread a long time ago
20150925 06:06:05< iceiceice> where we contemplated trying to use Qt and make qt render widgets in our sdl window
20150925 06:06:20< iceiceice> we didn't really know mucha bout qt, we pretty much backed out of that :)
20150925 06:06:21< vultraz> That was discarded, though, i think
20150925 06:06:30< iceiceice> it could work but its a huge, huge change
20150925 06:06:42< iceiceice> and basically Qt is a lot heavier than it hogut it was
20150925 06:06:55< Aginor> the benefit is that Qt comes with nice tools to develop the GUI in
20150925 06:07:00< iceiceice> yes
20150925 06:07:01< Aginor> but it is a whole ecosystem
20150925 06:08:45< iceiceice> Aginor: if you would redesign the whole wesnoth engine from the start what technologies would you use?
20150925 06:08:53< iceiceice> starting today
20150925 06:09:41< vultraz> Source 2 :P
20150925 06:10:00< vultraz> And DX12
20150925 06:10:10-!- Kwandulin [~Miranda@p5B009693.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150925 06:10:52< iceiceice> yeah i mean thats a reasonable thing to point out, maybe the game shouldn't have its own engine, maybe its better to use unity or something
20150925 06:11:05< iceiceice> fabi pointed out a project called "Fife" to me once
20150925 06:11:07< Aginor> vultraz: I'd pick something cross platform
20150925 06:11:33< iceiceice> Fife is foss and pretty close to the mark
20150925 06:11:45< Aginor> iceiceice: I don't know to be honest, I haven't done an inventory
20150925 06:11:46< vultraz> Source 2 and OGL?
20150925 06:12:28< iceiceice> Aginor: i'll say one thing though, i would definitely use an external widget tool kit and not a hand-rolled one
20150925 06:12:41< Aginor> iceiceice: definitively
20150925 06:12:45< iceiceice> this gui development stuff is so very painful
20150925 06:12:53< iceiceice> you do all this work and its only for this game
20150925 06:12:55< vultraz> iceiceice: apparently, anura is moving away from widgets to just objects
20150925 06:13:04< vultraz> I don't know what that practically means, though
20150925 06:13:20< iceiceice> if i'm going to fix some bug so that mouse clicks on scroll bars make the button light up or something mind numbing like that,
20150925 06:13:32< iceiceice> i want ti to be like, something i can contribute upstream to a project that other people care about also
20150925 06:14:26< iceiceice> wesnoth had a pretty unusual project development arc
20150925 06:14:36< iceiceice> basically they had tons and tons of people and interest early on 
20150925 06:14:41< iceiceice> way more than most projects get early on 
20150925 06:14:47< iceiceice> and they decided to roll everything themselves by hand
20150925 06:14:52< iceiceice> like, everything
20150925 06:14:56< iceiceice> make a custom programming language
20150925 06:15:00< iceiceice> a custom gui
20150925 06:15:09< iceiceice> now, the project is 300,000 lines of code last time i checked
20150925 06:15:15< iceiceice> which si like, comparable to firefox or something
20150925 06:15:16< iceiceice> which is crazy
20150925 06:15:25< vultraz> I don't think our forebearers gave thought to futureproofing
20150925 06:15:29< iceiceice> and theres only a tiny number of programmers now
20150925 06:16:06< iceiceice> vultraz: if you look at Fife, and you remove all the opengl boiler plate (one of their libs has 40k lines of opengl headers for compatibility or something)
20150925 06:16:16< iceiceice> their whole engine is like 40,000 lines of C++
20150925 06:16:54< vultraz> You have to think of what they thought wesnoth was at the time. Just a small hobby project. If they had aimed to make an easily accessible, modular project for a large audience, we'd be in a different place
20150925 06:17:18< iceiceice> vultraz: i mean the sad thing is like,
20150925 06:17:27< iceiceice> there are a ton of people who love this game and want to see it continue
20150925 06:17:32< vultraz> The problem now is we're trying to aim big while still working from the code from when we aimed small, and with a comparative number of devs
20150925 06:17:34< iceiceice> and are willing to put the time in 
20150925 06:17:51< iceiceice> but the code base is so huge and so unmaintainable, that its not really clear its possible at all
20150925 06:17:57< vultraz> Yeah
20150925 06:17:58< vultraz> Honestly
20150925 06:18:14< vultraz> I think it's time to just start over
20150925 06:18:22< vultraz> Ground up, day 1
20150925 06:18:38< vultraz> Kinda like Valve and Dota 2 in source 1 to 2
20150925 06:19:00< vultraz> By their own admission, it was difficult to implement their custom event gamemodes
20150925 06:19:14< vultraz> So they said, well, we're not doing that until Source 2 lands
20150925 06:19:15< iceiceice> I don't think working on the current wesnoth code base is a waste of time, i mean you can learn alot about what can go wrong
20150925 06:19:47< vultraz> Well, no
20150925 06:19:59< iceiceice> but i do think theres like at some point people should be making plans, what would be best for the future
20150925 06:20:09< vultraz> There are
20150925 06:20:29< iceiceice> vultraz: i think people should think critically about those plans also though
20150925 06:20:39< iceiceice> did you ever do a source code count on anura?
20150925 06:20:45< vultraz> No
20150925 06:20:48< iceiceice> did you ever count how many unit tests anura has?
20150925 06:20:54< iceiceice> orhow many C++ programmers they have?
20150925 06:21:05< iceiceice> or how many different parallel gui systems they have?
20150925 06:21:35< Aginor> there's a difference between doing game engine stuff and doing everything from the ground up
20150925 06:21:52< vultraz> I don't mean do *everything* from the ground up again
20150925 06:22:00< vultraz> Just that we should do a major refactor of the codebase
20150925 06:22:07< iceiceice> vultraz: also, when i spoke to dave and krista, they emphatically ruled out supporting WML inside anura
20150925 06:22:13< iceiceice> and specifically, supporting the preprocessor
20150925 06:22:17< vultraz> Look how complicated even this SDL2 transition is
20150925 06:22:29< vultraz> That may or may nor include using a 3rd party engine
20150925 06:22:30< iceiceice> they just said that people woudl hav eto make new campaigns
20150925 06:22:32< Aginor> I made a javascript-backed game engine as a technology test in https://libgdx.badlogicgames.com/ a year or so ago, and that was pretty neat
20150925 06:23:19< Aginor> vultraz: lipkab actually did 99% of it, but he mixed in HW acceleration with simply moving to SDL2
20150925 06:23:40< Aginor> He'd easily have finished it if he'd focused on that first
20150925 06:25:03< Aginor> libgdx is not a game engine, but it facilitates one, and it's certainly a layer or two above SDL. Something on that level is probably what I'd look at for wesnoth moving forward
20150925 06:27:43< iceiceice> Aginor: i think the hardware acceleration was his goal
20150925 06:28:10< iceiceice> it took him a while before he realized that it wasn't really going to accelerate things because of all the passes / pixel grabbing itrequires
20150925 06:28:37< iceiceice> also there were some technical issues he has with SDL2 regarding alpha blending and some other things
20150925 06:28:53< iceiceice> so at some point he decided SDL2 was not mature enough and was not the goal
20150925 06:28:59< iceiceice> but i think he did indeed get pretty close
20150925 06:29:10< Aginor> as I said, he got 99% there
20150925 06:29:20< iceiceice> i mean fwiw just getting SDL2 is still a big improvement right?
20150925 06:29:22< Aginor> I've just been fixing bugs, he's done everything
20150925 06:29:27< iceiceice> even if the graphics aren't faster
20150925 06:29:37< iceiceice> if full screen works on OS X that is like... the most requested feature ever
20150925 06:29:44< iceiceice> hehe
20150925 06:29:59< Aginor> it improves input handling and comes with support and have people actively developing against it
20150925 06:30:02< Aginor> I think that's a win
20150925 06:30:37< iceiceice> yeah
20150925 06:31:26< Aginor> it make sit feasible to replace some parts with 3rd party libraries and all of that
20150925 06:36:48-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20150925 06:42:35< Aginor> having just changed the textbox.?pp files, I can say that there's a lot of pure game lodgic that has dependencies on the header
20150925 06:43:24-!- [Relic] [~Relic]@2602:306:33a3:6d30:55ea:de86:fb63:d12a] has quit [Quit: I press the magic X and all the weirdos go away!]
20150925 06:46:27-!- irker327 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20150925 06:46:38 * Aginor swears at event ordering and the lack of proper event delegation
20150925 06:48:16< Aginor> I've added proper UTF-8 input to gui1 textbox now, but I get a stray character in there when I enter command mode because the text event happens after the keydown event that triggers the command mode
20150925 06:49:34-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 265 seconds]
20150925 06:51:44-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 06:54:46-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 06:55:51-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 06:58:10-!- Appleman1234 [~Appleman1@KD111239023081.au-net.ne.jp] has quit [Ping timeout: 240 seconds]
20150925 07:00:06-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 240 seconds]
20150925 07:00:39-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 07:02:31-!- kinow [~kinow@apache/committer/kinow] has quit [Quit: Leaving]
20150925 07:08:05-!- boucman_work [~jrosen@80-15.80-90.static-ip.oleane.fr] has joined #wesnoth-dev
20150925 07:08:05-!- boucman_work [~jrosen@80-15.80-90.static-ip.oleane.fr] has quit [Changing host]
20150925 07:08:05-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20150925 07:08:19-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20150925 07:08:48-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 264 seconds]
20150925 07:11:17-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 07:15:01-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 07:15:13-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 07:16:04-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 07:19:25-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 07:24:33-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 250 seconds]
20150925 07:27:12-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 07:37:02-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 246 seconds]
20150925 07:40:17-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 07:44:46-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 240 seconds]
20150925 07:50:41-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 07:51:35-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 07:53:15-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 08:05:34-!- genbattle_ [~genbattle@122-57-91-5.jetstream.xtra.co.nz] has joined #wesnoth-dev
20150925 08:11:18-!- Appleman1234 [~Appleman1@KD111239023081.au-net.ne.jp] has joined #wesnoth-dev
20150925 08:13:30-!- irker525 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20150925 08:13:30< irker525> wesnoth: Andreas Löf wesnoth:sdl2 96dd2980bcfc / src/widgets/ (textbox.cpp textbox.hpp): Add proper text input to GUI1 textbox under SDL2 http://git.io/vndaJ
20150925 08:13:38< Aginor> wedge009: ^ :)
20150925 08:13:48< Aginor> wedge009: I've also given you fullscreen support
20150925 08:14:09< Aginor> wedge009: and recreated your crash, but that felt difficult so I decided to leave it alone
20150925 08:14:49-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Ping timeout: 250 seconds]
20150925 08:15:26-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20150925 08:24:06-!- genbattle_ [~genbattle@122-57-91-5.jetstream.xtra.co.nz] has quit [Ping timeout: 244 seconds]
20150925 08:29:59-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 240 seconds]
20150925 08:30:50-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 08:39:46-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 265 seconds]
20150925 08:41:33-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20150925 08:42:22-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 08:44:10-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 08:49:48-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 08:52:38-!- aeonchild [enchilado@defocus/yummy/enchilado] has quit [Quit: leaving]
20150925 08:52:50-!- enchilado [enchilado@defocus/yummy/enchilado] has joined #wesnoth-dev
20150925 08:53:40-!- enchilado is now known as aeonchild
20150925 08:55:23-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 264 seconds]
20150925 08:56:23-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 08:59:46-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 09:00:27-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 09:01:23-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 09:02:59-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 09:06:20-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 09:07:01-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 09:08:15-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 09:08:25-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 09:13:27-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150925 09:18:47-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 264 seconds]
20150925 09:20:58-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 09:24:34-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 09:25:51-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 09:26:13-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Quit: No Ping reply in 180 seconds.]
20150925 09:26:48-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 09:26:59-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 09:27:38-!- fabi [~quassel@176.4.42.171] has joined #wesnoth-dev
20150925 09:27:38-!- fabi [~quassel@176.4.42.171] has quit [Changing host]
20150925 09:27:38-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20150925 09:31:29-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 252 seconds]
20150925 09:32:00-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 09:47:12-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 264 seconds]
20150925 09:51:11-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Ping timeout: 264 seconds]
20150925 09:53:39-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 09:54:22-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 09:55:04-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 09:55:48-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 09:56:31-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 09:57:08-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 09:57:42-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 10:02:33-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 256 seconds]
20150925 10:07:03-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 10:11:46-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 240 seconds]
20150925 10:15:31-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 10:17:50-!- zombah [~zombah@2a02:28:3:1:214:4fff:fe47:5920] has joined #wesnoth-dev
20150925 10:23:31-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 256 seconds]
20150925 10:26:24-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 10:26:44< wedge009> Aginor: Thanks, verified it works. But I've found a couple of other strange things.
20150925 10:27:45-!- horrowind [~Icedove@2a02:810a:8b00:1c54:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20150925 10:28:10< wedge009> Aginor: Often I don't get the Wesnoth icon in the title bar of the window (on Windows). Also, it may be a driver issue, but there's a weird inverse blending issue on radeon open-source drivers. The colour cursor has an opaque white box around it.
20150925 10:28:18-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 10:28:21< wedge009> Aginor: I can't reproduce it on anything else, though.
20150925 10:29:25< wedge009> Aginor: Also, the resizing window crash seems more non-deterministic now. Before it would happen on any resize. Now it seems to only happen some of the time. Will try to investigate further. This one affects Windows and Linux.
20150925 10:33:59-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 10:34:06< wedge009> (When it does show, the title/system icon does look a lot nicer in SDL2. Probably using better scaling.
20150925 10:34:08< wedge009> )
20150925 10:39:11-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 264 seconds]
20150925 10:41:07< Aginor> wedge009: so icon is intermittently showing up?
20150925 10:42:18-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 10:43:34< Aginor> wedge009: is your radeon issue on linux or windows?
20150925 10:47:22-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 265 seconds]
20150925 10:48:34-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 10:50:07-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 10:50:40-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 10:56:11< wedge009> Aginor: Yes, intermittently, though mostly not showing. :S White-boxed colour-cursor is on Linux.
20150925 10:56:33< wedge009> Aginor: Icon issue is on Windows. I'll fix the PR in a bit.
20150925 11:01:50< wedge009> Aginor: Done.
20150925 11:13:31-!- irker525 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20150925 11:21:32< Aginor> wedge009: sweet, it's looking good
20150925 11:21:46< Aginor> I'm just waiting for travis to finish
20150925 11:24:17< wedge009> Aginor: A quick Google search on the colour cursor issue suggests it's a problem with SDL2 on AMD GPUs. I'm using Mobility HD 2600 so it only has radeon as an option (fglrx drops support for this GPU on newer kernels), but the poster was using fglrx.
20150925 11:26:21< wedge009> Aginor: Maybe need some more people to test on ATI/AMD GPUs? If it is a problem external to Wesnoth, I wonder if it's worth providing three options: standard BW cursors, the original surface-based colour cursor hack currently used, and then the SDL2 colour cursors.
20150925 11:26:48< Aginor> wedge009: ironicaly, I'm using AMD CPU and graphics :D
20150925 11:27:19< wedge009> In your Macbook?
20150925 11:27:55< Aginor> no, on my main workstation
20150925 11:28:16< wedge009> Me too... but that's Windows.
20150925 11:30:12< Aginor> I'm not using the fglrx driver though, I got annoyed with having to make a new patch for the driver ever so often
20150925 11:30:37< Aginor> especially since AMD saw no interest in picking up the patches I tried to contribute back :D
20150925 11:30:44< wedge009> Boo.
20150925 11:34:04< wedge009> I still need fglrx for OpenCL.
20150925 11:34:45< Aginor> fair enough
20150925 11:35:12< Aginor> it feels like we're making progress towards getting this entire sdl2 thing finished though
20150925 11:36:05< Aginor> wedge009: is this the post you came across?
20150925 11:36:07< Aginor> http://trac.wildfiregames.com/ticket/2823
20150925 11:38:01< wedge009> It does seem like a lot of progress has been made recently. I want to try the multi-player game joining scenario that caused the same sdl_blit unhandled exception as the windows resizing.
20150925 11:38:21< wedge009> And also see what's going on with the system icon, though it's by no means a breaker.
20150925 11:38:28< Aginor> I haven't fixed the window resize yet
20150925 11:38:35< Aginor> I can replicate that one at the ttle screen
20150925 11:38:37< wedge009> I know. It's a tricky one.
20150925 11:38:45< wedge009> And the post was https://forums.libsdl.org/viewtopic.php?t=10422
20150925 11:39:36< Aginor> all of that isn't looking entirely applicable because wesnoth uses a software renderer
20150925 11:40:48< wedge009> Hmm.
20150925 11:41:27< wedge009> Still in the middle of copying to the Linux machine with AMD GPUs and fglrx drivers.
20150925 11:41:30< Aginor> wedge009: which cursor, the normal one?
20150925 11:43:31< wedge009> Normal one.
20150925 11:44:17< wedge009> Just checked - it seems to happen for all of them.
20150925 11:44:54< Aginor> thanks
20150925 11:45:01< Aginor> the pngs are looking good
20150925 11:45:25-!- aquileia [95acd0d3@gateway/web/freenode/ip.149.172.208.211] has joined #wesnoth-dev
20150925 11:45:49-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20150925 11:46:17< wedge010> Yeah, I checked those as well. They definitely have transparency.
20150925 11:47:58< aquileia> Aginor: Is it worth updating the projectfiles (with sdl/keyboard.cpp) in your sdl2 branch yet or do you plan to add more files?
20150925 11:48:51-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 255 seconds]
20150925 11:48:51-!- wedge010 is now known as wedge009
20150925 11:54:48< wedge009> aquileia: I've just been adding files manually, though I'm using VC14 now.
20150925 11:55:12< wedge009> Aginor: Interesting - happens on GCN GPU with fglrx as well.
20150925 11:56:22< wedge009> Aginor: It does appear to be specific to AMD GPUs. ):
20150925 11:57:35-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 256 seconds]
20150925 11:59:18< Aginor> aquileia: I have no plans to add more files at this stage
20150925 11:59:59-!- irker029 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20150925 11:59:59< irker029> wesnoth: Wedge009 wesnoth:sdl2 354673875050 / src/joystick.cpp: Clarify implicit cast to boolean http://git.io/vnF4k
20150925 11:59:59< irker029> wesnoth: Wedge009 wesnoth:sdl2 0ba49fae04af / src/ (15 files in 6 dirs): Replace 255 with SDL_ALPHA_OPAQUE where appropriate. http://git.io/vnF4I
20150925 12:00:00< irker029> wesnoth: Andreas wesnoth:sdl2 1181a401edc3 / src/ (16 files in 6 dirs): Merge pull request #503 from Wedge009/sdl2_minor_changes http://git.io/vnF4L
20150925 12:00:39< Aginor> wedge009: thanks for those changes :)
20150925 12:01:03< wedge009> Aginor: Minor, but I think important in long-term. (:
20150925 12:05:05 * vultraz is working on a redesign of the Play Campaign window
20150925 12:05:26-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 12:09:41-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 246 seconds]
20150925 12:16:50-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 12:20:04-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 12:25:46-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 12:26:24-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 12:26:50< irker029> wesnoth: aquileia wesnoth:master 74bf61c424ef / projectfiles/VC9/wesnoth.vcproj: vcproj: Update for 3c11aa6 & 3ed4eed http://git.io/vnFgk
20150925 12:27:35-!- Kwandulin [~Miranda@p5B009693.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150925 12:29:32-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 12:34:38-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 260 seconds]
20150925 12:36:59< vultraz> ok, damn, this is harder than I thought :|
20150925 12:37:13< vultraz> (this bit specifically)
20150925 12:39:12-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 12:43:37-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 252 seconds]
20150925 12:48:23-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 12:51:55-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 12:53:10-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 12:57:30-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 240 seconds]
20150925 13:01:59-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20150925 13:05:01< vultraz> ok, i need to figure out how to reuse the code for parsing the difficulty descriptions
20150925 13:05:17-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 13:10:06-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 260 seconds]
20150925 13:10:51-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 13:15:36-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 13:16:44-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20150925 13:17:22-!- pax2you [~pax@host213-156-dynamic.20-79-r.retail.telecomitalia.it] has joined #wesnoth-dev
20150925 13:17:25-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 13:18:03-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 13:19:00-!- pax2you [~pax@host213-156-dynamic.20-79-r.retail.telecomitalia.it] has quit [Client Quit]
20150925 13:19:22-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 13:19:59-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 13:20:30-!- pax2you [~pax@host213-156-dynamic.20-79-r.retail.telecomitalia.it] has joined #wesnoth-dev
20150925 13:21:22-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 13:21:54-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 13:22:38-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 13:23:15-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Excess Flood]
20150925 13:36:44-!- pax2you [~pax@host213-156-dynamic.20-79-r.retail.telecomitalia.it] has left #wesnoth-dev ["Leaving"]
20150925 13:37:18-!- pax2you [~pax@host213-156-dynamic.20-79-r.retail.telecomitalia.it] has joined #wesnoth-dev
20150925 14:00:50-!- Kwandulin [~Miranda@p5B009693.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150925 14:02:21-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20150925 14:06:47-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150925 14:08:59-!- noy [~Noy@wesnoth/developer/noy] has quit [Client Quit]
20150925 14:22:30-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Ping timeout: 240 seconds]
20150925 14:22:53-!- pax2you [~pax@host213-156-dynamic.20-79-r.retail.telecomitalia.it] has quit [Remote host closed the connection]
20150925 14:23:01-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20150925 14:24:09< aquileia> wedge009, Aginor, aquileia: VC9 reports a heap of errors src/ai when trying to build the sdl2 branch... I'm retrying with VC12, but I'll have to look into VC9 later (pinging myself deliberately)
20150925 14:42:50-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 14:46:56-!- Shackra [~Jorge@186.177.2.148] has joined #wesnoth-dev
20150925 14:47:41-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150925 14:59:06-!- Shackra [~Jorge@186.177.2.148] has quit [Read error: Connection reset by peer]
20150925 15:01:34-!- Guest24707 [~Jorge@186.177.2.148] has joined #wesnoth-dev
20150925 15:17:58-!- Guest24707 [~Jorge@186.177.2.148] has quit [Quit: No Ping reply in 180 seconds.]
20150925 15:18:02-!- louis94 [~~louis94@109.133.201.27] has joined #wesnoth-dev
20150925 15:19:17-!- Shackra_ [~Jorge@186.177.2.148] has joined #wesnoth-dev
20150925 15:27:55-!- irker029 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20150925 15:32:47-!- prkc [~prkc@catv-89-134-159-103.catv.broadband.hu] has joined #wesnoth-dev
20150925 15:39:47-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Ping timeout: 265 seconds]
20150925 15:49:54-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20150925 16:10:09-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20150925 16:15:25< aquileia> mp_replay_controller.cpp @ l.380: MSVC seems to be ignorant to the fact that specifying a return value after an assert is futile...
20150925 16:17:01< aquileia> Should I locally disable the warning or may I just add a "return false" to silence MS' stupidity?
20150925 16:17:28-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has joined #wesnoth-dev
20150925 16:30:50-!- celmin|sleep is now known as celticminstrel
20150925 16:34:55< Soliton> you can disable asserts with some define, so the return would be a good idea.
20150925 16:36:42< celticminstrel> Specifying a return value after an assert is not futile. 
20150925 16:36:58< celticminstrel> Not even if it's assert(false).
20150925 16:46:26-!- Kwandulin [~Miranda@p5B009693.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150925 16:48:06< celticminstrel> So basically, yeah, return false.
20150925 16:51:11< aquileia> ok
20150925 16:52:04< aquileia> thanks for the quick response, Soliton & celticminstrel
20150925 16:56:22-!- zombah [~zombah@2a02:28:3:1:214:4fff:fe47:5920] has quit [Quit: Leaving]
20150925 17:05:37-!- iceiceice [~chris@50.245.222.235] has joined #wesnoth-dev
20150925 17:05:37-!- iceiceice [~chris@50.245.222.235] has quit [Changing host]
20150925 17:05:37-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20150925 17:06:38< iceiceice> aquileia: it's not actually futile, its because assert is more or less implemented differently on msvc
20150925 17:06:55< iceiceice> in ms you can recover from an assertion failure if the user clicks some dialog
20150925 17:06:59< iceiceice> so "assert" is not actually no return
20150925 17:07:05< iceiceice> and code after an assert is not dead code
20150925 17:07:53< celticminstrel> Code after an assert is not dead code on any platform.
20150925 17:08:00< iceiceice> celticminstrel: yes, it is.
20150925 17:08:06< celticminstrel> Because as Soliton said, asserts can be disabled, and typically are for release builds.
20150925 17:08:07< iceiceice> it is dead on linux and is removed by the compiler.
20150925 17:08:18< celticminstrel> Not even on Linux can you count it as dead code.
20150925 17:08:32< celticminstrel> If asserts are enabled, then yes, the compiler will notice and should remove it.
20150925 17:08:40< iceiceice> if you want a cross-platform equivalent assert, you should make one using a macro definition, and 'if' and 'abort' function
20150925 17:08:45< iceiceice> because abort is no return on all platforms
20150925 17:09:09< iceiceice> celticminstrel, we dont need to argue, you can compile a simple test program with gcc and see when it generates warnings for falling off the end of nonvoid functions.
20150925 17:09:35< celticminstrel> Sure.
20150925 17:12:36< iceiceice> celticminstrel: the thing Soliton is mentioning is the NDEBUG macro
20150925 17:12:46< iceiceice> and it is indeed a macro, it is fixed at compile time
20150925 17:12:48< celticminstrel> Yes.
20150925 17:13:01< iceiceice> code doesn't become dead or not dead because "in another build you might define a different macro"
20150925 17:13:05< iceiceice> so that isn't actually the reason.
20150925 17:13:25< iceiceice> it's because the platforms just implement assert slightly differnetly.
20150925 17:13:59< celticminstrel> That's fair, but I still think it's not good to consider code after an assert(false) to be dead code, partly because of different implementations and partly because of NDEBUG.
20150925 17:14:11< iceiceice> yeah i mean, imo we shouldn't use assert
20150925 17:14:22< iceiceice> it would be simpler to just use something based on abort
20150925 17:14:25< iceiceice> so it works the same everywhere
20150925 17:14:32< celticminstrel> clang does give the warning if NDEBUG is defined.
20150925 17:15:02< celticminstrel> assert is okay when it's something that you're pretty sure will never happen.
20150925 17:15:17< iceiceice> yeah i mean, it's ok,
20150925 17:15:22< iceiceice> but it just creates these compiler issues
20150925 17:15:31< iceiceice> because you'll get warnings on msvc that you don't get on clang / gcc
20150925 17:15:39< iceiceice> because the code analysis works out differently
20150925 17:15:48-!- louis94 [~~louis94@109.133.201.27] has quit [Ping timeout: 272 seconds]
20150925 17:16:21< iceiceice> i mean folks can use whatever they want, but imo its simpler to have something thats more consistent
20150925 17:17:48< iceiceice> you'll get the same warning on mingw also actually
20150925 17:29:05-!- [Relic] [~Relic]@2602:306:33a3:6d30:a4b2:f20a:9fb0:e27a] has joined #wesnoth-dev
20150925 17:37:35-!- louis94 [~~louis94@109.133.201.27] has joined #wesnoth-dev
20150925 17:49:25-!- louis94 [~~louis94@109.133.201.27] has quit [Ping timeout: 240 seconds]
20150925 17:54:03-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150925 18:10:46-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20150925 18:43:57-!- esr [~esr@wesnoth/developer/esr] has quit [Quit: WeeChat 1.0.1]
20150925 18:49:12-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20150925 18:49:13-!- lok [c81162fe@gateway/web/freenode/ip.200.17.98.254] has joined #wesnoth-dev
20150925 18:49:13-!- lok [c81162fe@gateway/web/freenode/ip.200.17.98.254] has quit [Client Quit]
20150925 18:50:10-!- horrowind [~Icedove@2a02:810a:8b00:1c54:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20150925 18:53:34-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 260 seconds]
20150925 18:53:46-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20150925 19:11:55-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150925 19:20:38-!- louis94 [~~louis94@109.133.201.27] has joined #wesnoth-dev
20150925 19:25:30-!- louis94 [~~louis94@109.133.201.27] has quit [Ping timeout: 250 seconds]
20150925 19:27:17-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20150925 19:38:30-!- Appleman1234 [~Appleman1@KD111239023081.au-net.ne.jp] has quit [Ping timeout: 268 seconds]
20150925 19:48:25-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has quit [Quit: tomreyn]
20150925 20:01:04-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150925 20:02:27-!- genbattle_ [~genbattle@122-57-91-5.jetstream.xtra.co.nz] has joined #wesnoth-dev
20150925 20:22:50-!- genbattle_ [~genbattle@122-57-91-5.jetstream.xtra.co.nz] has quit [Ping timeout: 240 seconds]
20150925 20:34:33-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Read error: Connection reset by peer]
20150925 20:50:54-!- Appleman1234 [~Appleman1@KD111239000197.au-net.ne.jp] has joined #wesnoth-dev
20150925 20:51:24-!- iceiceice [~chris@ext-74.ias.edu] has joined #wesnoth-dev
20150925 20:51:25-!- iceiceice [~chris@ext-74.ias.edu] has quit [Changing host]
20150925 20:51:25-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20150925 20:54:21< aquileia> Aginor, wedge009: "error general: Unknown key: print screen" - German keyboard layout
20150925 20:58:46< aquileia> Renaming my preference file fixed it - we might want to clean any pre-existing hotkey preferences when we release 1.13.5
20150925 20:59:32< celticminstrel> Yeah, the SDL2 branch doesn't understand any of the old hotkeys, I think.
20150925 20:59:47< celticminstrel> Do you mean 1.13.2?
20150925 21:00:30< aquileia> right, mixed it up with the next minor stable version number
20150925 21:04:58< aquileia> I've a fix for the vcproj files that converts them to use SDL2; it doesn't work with VC9 yet but VC12 runs and passes all WML tests.
20150925 21:07:19< aquileia> VC9 chokes on the syntax of some of the AI code, I'll test whether that applies to master as well
20150925 21:12:48-!- irker110 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20150925 21:12:48< irker110> wesnoth: aquileia wesnoth:master 112fd79f88c2 / src/mp_replay_controller.cpp: Fix MSVC warning due to missing return value http://git.io/vnAow
20150925 21:26:40-!- louis94 [~~louis94@109.133.201.27] has joined #wesnoth-dev
20150925 21:31:36-!- louis94 [~~louis94@109.133.201.27] has quit [Ping timeout: 250 seconds]
20150925 21:32:02-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 250 seconds]
20150925 21:35:19< wedge009> aquileia: Thanks, I wanted that one fixed. (: All I've done on VC9 so far is change the SDL lib references to their SDL2 equivalents. It's still compiling. I tried it once before, but then it wanted the boost libraries so I just restarted using your external package.
20150925 21:37:05< wedge009> aquileia, celticminstrel: I think the hot keys should be okay for the most part - letter-based ones, at least. The more esoteric ones like Print Screen, Pause/Break, etc, may be trouble. I did find last night, though, that the default Ctrl+Alt+m combination for mute works properly for SDL2 (is broken with SDL1.2).
20150925 21:40:00< celticminstrel> I think colon wasn't working when I tried it.
20150925 21:43:37< wedge009> SDL2 branch?
20150925 21:45:53< wedge009> Seems okay for me, but I'm using regular US layout. I don't know how it handles non-US keyboards (not even UK layout) but I think that's always going to be difficult to deal with. I think the way it's done in the SDL2 branch is much better than where it currently stands in master, though.
20150925 21:47:14< celticminstrel> I'm using US layout. Well, US extended, but should be the same.
20150925 21:49:14< wedge009> Aginor may be able to remark on it. I believe he did switch the command-line default from : to ;, possibly because of something going on there.
20150925 21:49:41< celticminstrel> I thought the idea there was to shift from : to shift-;
20150925 21:55:08< wedge009> Something like that.
20150925 21:55:34< wedge009> : is printed as 'shift+;' in the preferences now.
20150925 21:58:22-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 260 seconds]
20150925 21:58:26-!- louis94 [~~louis94@109.133.201.27] has joined #wesnoth-dev
20150925 22:00:50-!- Shackra_ [~Jorge@186.177.2.148] has quit [Ping timeout: 240 seconds]
20150925 22:03:06< wedge009> aquileia: Is it contexts.hpp that's throwing issues for you in VC9?
20150925 22:04:38< aquileia> Don't remember whether it was the cpp or hpp, but yeah, I think most issues were there
20150925 22:08:05< aquileia> wedge009: IIRC it spread over ai/akihara/recruitment, ai/composite/contexts & ai/contexts
20150925 22:08:14< wedge009> Yeah, that's the one.
20150925 22:11:22-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20150925 22:16:54< aquileia> wedge009: I assume you have a suspicion?
20150925 22:17:28< wedge009> I'm trying to look into it. Seems to be complaining about the TYPE enumeration (error message talks about an int), but not sure why it's complaining.
20150925 22:17:53< wedge009> I gave up on VC9 after finding the map editor kept hitting the assertion failure.
20150925 22:18:06< wedge009> Which doesn't happen in more recent VCs.
20150925 22:20:43< aquileia> Well, I wouldn't bother with VC9 either if we weren't using the old vcproj format
20150925 22:21:31< aquileia> But it doesn't hurt to check from time to time whether it actually builds in that version
20150925 22:22:45< aquileia> I guess we'll drop support when C++11 starts to spread throughout the code
20150925 22:25:15-!- [Relic] [~Relic]@2602:306:33a3:6d30:a4b2:f20a:9fb0:e27a] has quit [Quit: I press the magic X and all the weirdos go away!]
20150925 22:26:40< celticminstrel> VC12 is the first with decent C++11 support, right?
20150925 22:26:49< celticminstrel> I think VC12 is what I have.
20150925 22:27:45-!- [Relic] [~Relic]@2602:306:33a3:6d30:dd3b:55c8:506e:5e08] has joined #wesnoth-dev
20150925 22:29:35< aquileia> They added C++11 bit by bit, so I'd have to look up what they included in each version...
20150925 22:30:36< aquileia> They actually provide a table: https://msdn.microsoft.com/en-us/library/hh567368.aspx
20150925 22:30:51< celticminstrel> I've seen the table.
20150925 22:34:45-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Quit: wedge009]
20150925 22:38:27-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20150925 22:45:19-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150925 23:14:37-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20150925 23:14:42-!- aquileia [95acd0d3@gateway/web/freenode/ip.149.172.208.211] has quit [Ping timeout: 246 seconds]
20150925 23:56:34-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20150925 23:57:00-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150925 23:58:00-!- genbattle_ [~genbattle@122-57-91-5.jetstream.xtra.co.nz] has joined #wesnoth-dev
--- Log closed Sat Sep 26 00:00:39 2015