--- Log opened Sun Apr 09 00:00:19 2017
20170409 00:02:00-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170409 00:02:12-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170409 00:03:47< vultraz_iOS> hmmmm
20170409 00:03:54 * vultraz_iOS ponders the rounded rectangle
20170409 00:06:58-!- atarocch [~atarocch@93.56.160.28] has quit [Ping timeout: 264 seconds]
20170409 00:07:12-!- Shiki [~Shiki@p54855738.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170409 00:08:14< irker263> wesnoth: Charles Dang wesnoth:master ec6191a68e0b / changelog: Updated changelog for editor unit crash fix https://github.com/wesnoth/wesnoth/commit/ec6191a68e0b0d2497399a357aa2b3f82a1dbf94
20170409 00:12:22-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20170409 00:15:09-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170409 00:19:26< vultraz_iOS> Still tempted to mark the changelog/plays changelog/INSTALL as markdown
20170409 00:32:24< irker263> wesnoth: Celtic Minstrel wesnoth:master d536d6ab7620 / src/ (23 files in 8 dirs): Implement a destructor subscription system for formula_callable https://github.com/wesnoth/wesnoth/commit/d536d6ab7620bde84ca84c064f5200ea19d3460c
20170409 00:32:26< irker263> wesnoth: Jyrki Vesterinen wesnoth:master 0b02ba08767c / src/formula/ (variant.cpp variant.hpp): Make variant constructor from std::shared_ptr a template https://github.com/wesnoth/wesnoth/commit/0b02ba08767c5bbd85d605102812f1eeb3e295a8
20170409 00:32:28< irker263> wesnoth: Matthias Krüger wesnoth:master 165204f159e3 / src/units/unit.cpp: unit.cpp: remove unused variable. Found by cppcheck. https://github.com/wesnoth/wesnoth/commit/165204f159e385607fa7ec10805d464d7fed2cc1
20170409 00:32:30< irker263> wesnoth: Celtic Minstrel wesnoth:master 32681af9783c / src/game_initialization/multiplayer.cpp: fix unused code, found by cppcheck. https://github.com/wesnoth/wesnoth/commit/32681af9783c98a5f28141323904299e8f38e951
20170409 00:32:32< irker263> wesnoth: Celtic Minstrel wesnoth:master abb47d13ab16 / src/ai/testing.cpp: Minor cleanup of ai_testing output https://github.com/wesnoth/wesnoth/commit/abb47d13ab163c838001314bc4153a3313ea4fd1
20170409 00:32:34< irker263> wesnoth: Celtic Minstrel wesnoth:master ac6a62046e95 / src/ (4 files in 2 dirs): Add wesnoth.log_replay which adds stuff to the saved game in [upload_log] https://github.com/wesnoth/wesnoth/commit/ac6a62046e95c9a9fc75230077f7c95c2fca600b
20170409 00:32:36< irker263> wesnoth: Celtic Minstrel wesnoth:master 795592fe12f5 / src/ (23 files in 8 dirs): Merge branch 'wfl_memleak_fix' https://github.com/wesnoth/wesnoth/commit/795592fe12f51cba769b8d114f7a37e0b76f002b
20170409 00:32:38< irker263> wesnoth: Celtic Minstrel wesnoth:master dd7296a35957 / changelog: Update changelog https://github.com/wesnoth/wesnoth/commit/dd7296a3595741eee998033e9154dd093abbb710
20170409 00:36:50< irker263> wesnoth: Celtic Minstrel wesnoth:master 30447aa7f7e3 / src/ (playsingle_controller.cpp playsingle_controller.hpp): Make the fallback() FormulaAI action transfer control only until end of turn https://github.com/wesnoth/wesnoth/commit/30447aa7f7e37d8f4b7bf4a7464ce3d6f0304014
20170409 00:36:52< irker263> wesnoth: Celtic Minstrel wesnoth:master 3e5a5626f37a / src/ (ai/lua/core.cpp menu_events.hpp): Add fallback_human() action to the Lua AI https://github.com/wesnoth/wesnoth/commit/3e5a5626f37a1e76c6839440c11844bce981ed7c
20170409 00:36:54< irker263> wesnoth: Celtic Minstrel wesnoth:master abcd235fdb50 / src/ (4 files in 2 dirs): Merge pull request #972 from wesnoth/lua_ai_fallback https://github.com/wesnoth/wesnoth/commit/abcd235fdb504e3f3e6461359b3fa451ebf20aac
20170409 00:52:09-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170409 00:53:46-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170409 01:08:35-!- Shiki [~Shiki@p54855738.dip0.t-ipconnect.de] has quit [Remote host closed the connection]
20170409 01:13:17< mattsc> vultraz_iOS or celticminstrel: I am now getting 208 of these:
20170409 01:13:19< mattsc> src/formula/callable_fwd.hpp:27:8: warning: 'wfl::callable_die_subscriber' has virtual functions but non-virtual destructor [-Wnon-virtual-dtor]
20170409 01:13:33< celticminstrel> Whoops!
20170409 01:15:21< irker263> wesnoth: Celtic Minstrel wesnoth:master 88280285ee59 / src/formula/callable_fwd.hpp: Fixup non-virtual destructor https://github.com/wesnoth/wesnoth/commit/88280285ee59efbe420786fa81039594fc03a810
20170409 01:21:21-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20170409 01:39:10-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170409 01:59:48-!- gfgtdf_ [~chatzilla@x4e3684f9.dyn.telefonica.de] has joined #wesnoth-dev
20170409 02:01:56-!- gfgtdf [~chatzilla@x4e363f37.dyn.telefonica.de] has quit [Ping timeout: 260 seconds]
20170409 02:01:57-!- gfgtdf_ is now known as gfgtdf
20170409 02:12:21< irker263> wesnoth: mattsc wesnoth:master be3cf5fa6241 / projectfiles/Xcode/Wesnoth.xcodeproj/project.pbxproj: Update Xcode project to fix linker warnings https://github.com/wesnoth/wesnoth/commit/be3cf5fa62416e5aa83c50d39bd6f27cb96c6373
20170409 02:12:27< mattsc> celticminstrel: ^
20170409 02:13:19< celticminstrel> That is indeed the general idea of how to fix the errors. IIRC I tried fixing it in a similar way and failed, though.
20170409 02:13:55< mattsc> It took me several attempts of different combinations before it actually worked for me.
20170409 02:33:38-!- gfgtdf [~chatzilla@x4e3684f9.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.93 [Firefox 52.0.2/20170323105023]]
20170409 03:00:26-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20170409 03:07:00< irker263> wesnoth: Celtic Minstrel wesnoth:wml_tag_porting 7e79147ffd1c / data/lua/wml-tags.lua src/scripting/lua_audio.cpp: fixup! Properly port [music] to Lua https://github.com/wesnoth/wesnoth/commit/7e79147ffd1cd2ded2f7f16b07cf39eafcbdbc66
20170409 03:31:29< vultraz_iOS> celticminstrel: so, are the memleaks fixed?
20170409 03:31:39< celticminstrel> Ask matthiaskrgr?
20170409 03:32:37< vultraz_iOS> matthiaskrgr: are the memleaks fixed?
20170409 03:33:48< celticminstrel> I mean, I doubt that fixes every memleak in the game though. :P
20170409 03:34:41< celticminstrel> A good test case is probably -tformula since it would result in a lot of action objects being created. If you want you could test before and after the merge.
20170409 03:35:10< celticminstrel> (BTW, someone other than matthiaskrgr is also welcome to investigate memleaks too.)
20170409 03:35:51< vultraz_iOS> well, it should fix this at least https://gna.org/bugs/index.php?25076
20170409 03:36:52< celticminstrel> Indeed, there's another test case.
20170409 03:37:15< celticminstrel> I wonder if those memleaks were actually from the Multiplayer window though? Or were they from the tooltips?
20170409 03:38:05< vultraz_iOS> perhaps the tooltips
20170409 03:39:00< celticminstrel> Mind you, I guess leaks are possible even in dialogs that don't use any formulas, if callables need to be allocated to pass into the typed_formula.
20170409 03:40:10< vultraz_iOS> I wonder if matthiaskrgr could also test the memleak bugs marked as "ready for test"
20170409 03:42:07< vultraz_iOS> would like to know where this memleak comes from.. :P
20170409 03:42:08< vultraz_iOS> https://gna.org/bugs/index.php?25092
20170409 03:42:58< vultraz_iOS> I wonder if it's us, though.
20170409 03:43:06< celticminstrel> I'd guess something in sound.?pp if it's us.
20170409 03:43:21< celticminstrel> Oh.
20170409 03:43:29< vultraz_iOS> oh wait
20170409 03:43:41-!- Guest53659 [~chrisdotc@n219079125169.netvigator.com] has joined #wesnoth-dev
20170409 03:43:52< vultraz_iOS> hm
20170409 03:43:56< vultraz_iOS> only one case of Mix_LoadMUSType_RW 
20170409 03:44:04< vultraz_iOS> in sound.cpp
20170409 03:44:57< vultraz_iOS> Mix_Music* const music = Mix_LoadMUSType_RW(rwops, MUS_NONE, true); // SDL takes ownership of rwops
20170409 03:45:16< celticminstrel> vultraz_iOS: How many cases of Mix_FreeMusic?
20170409 03:45:42< vultraz_iOS> 1
20170409 03:45:46< celticminstrel> Hmm.
20170409 03:46:16< vultraz_iOS> predicated on unload_music being true
20170409 03:46:39< vultraz_iOS> which is only true if stop_music is called
20170409 03:46:55< celticminstrel> Let's switch it to store shared_ptrs to Mix_Music.
20170409 03:47:00< celticminstrel> And pass Mix_FreeMusic as the deleter.
20170409 03:47:38< celticminstrel> But also, test to make sure that doesn't make the game crash on exit.
20170409 03:48:13< celticminstrel> If it does you'll probably also need to add something to main to ensure the music cache is flushed before main exits.
20170409 03:48:15< vultraz_iOS> I assume we want a shared_ptr<const Mix_Music>?
20170409 03:48:31< celticminstrel> Uh. Not sure about that, but if you're wrong, it won't build, so go ahead and try.
20170409 03:48:47< celticminstrel> (Specifically, I'm not sure about the const.)
20170409 03:49:20< vultraz_iOS> (since this is uh...a const ptr)
20170409 03:49:27< celticminstrel> Oh.
20170409 03:49:46< vultraz_iOS> (I keep mixing up const ptrs and ptrs to const :| )
20170409 03:49:49< celticminstrel> Mix_Music*const would translate to const shared_ptr<Mix_Music>, but IIRC you can't use that as a map element type.
20170409 03:50:13< celticminstrel> You can however use it at the point where the music is loaded.
20170409 03:50:41< celticminstrel> shared_ptr<const Mix_Music> as a raw pointer would be const Mix_Music*
20170409 03:51:01< vultraz_iOS> aka ptr to const
20170409 03:51:04< vultraz_iOS> right?
20170409 03:51:25< celticminstrel> Yes.
20170409 03:51:34< celticminstrel> But that code was using a const pointer, not a pointer to const.
20170409 03:52:57-!- Guest53659 [~chrisdotc@n219079125169.netvigator.com] has quit [Remote host closed the connection]
20170409 03:54:38< vultraz_iOS> C:\TDM-GCC-32\lib\gcc\mingw32\5.1.0\include\c++\bits\shared_ptr_base.h|886|error: invalid application of 'sizeof' to incomplete type '_Mix_Music'|
20170409 03:55:12< celticminstrel> Ooh fun.
20170409 03:55:18< vultraz_iOS> even though #include <SDL_mixer.h>is included
20170409 03:55:37< celticminstrel> Yeah, SDL probably defines it in a c file rather than in the header.
20170409 03:55:59< vultraz_iOS> eeevvviilll C!
20170409 03:55:59< celticminstrel> But do you get that even when passing the deleter?
20170409 03:56:17< vultraz_iOS> oh right
20170409 03:56:22< vultraz_iOS> the deleter..
20170409 03:56:37< celticminstrel> I just wouldn't be surprised if the error is actually from std::default_deleter.
20170409 03:56:52< celticminstrel> And if so, passing Mix_FreeMusic should solve it.
20170409 03:56:56< vultraz_iOS> how do you pass the deleter again
20170409 03:57:01< celticminstrel> Second argument to constructor.
20170409 03:57:19< celticminstrel> I suggest &Mix_FreeMusic actually, for clarity.
20170409 03:57:31< celticminstrel> Should work without the & too, but...
20170409 03:58:01< vultraz_iOS> ah that works
20170409 03:58:04< vultraz_iOS> ok, so I have
20170409 03:58:05< vultraz_iOS> const std::shared_ptr<Mix_Music> music(Mix_LoadMUSType_RW(rwops, MUS_NONE, true), &Mix_FreeMusic);
20170409 03:58:13< celticminstrel> Yup yup
20170409 03:58:58< vultraz_iOS> the game does not crash
20170409 03:59:09< celticminstrel> Great.
20170409 03:59:19< celticminstrel> Hopefully it doesn't crash for others either.
20170409 03:59:35< celticminstrel> I guess SDL_mixer probably isn't based on OpenAL like SFML is, so that might be part of it.
20170409 04:02:08< celticminstrel> I wonder if we should use shared_ptr for Mix_Chunk(?) too.
20170409 04:02:32< irker263> wesnoth: Charles Dang wesnoth:master 92d1a26caf17 / src/sound.cpp: Attempt to fix memleak when quitting the game with music enabled (bug #25092) https://github.com/wesnoth/wesnoth/commit/92d1a26caf172bde8b880f210b22e80d4135e83b
20170409 04:03:32< celticminstrel> You sure like your emplace.
20170409 04:03:47< vultraz_iOS> \ o / PRAISE EMPLACE
20170409 04:03:51< celticminstrel> Line 609 is wrong.
20170409 04:03:56< celticminstrel> It should be track.second.reset()
20170409 04:04:10< celticminstrel> Which will then call Mix_FreeMusic for you since that was passed as the deleter before.
20170409 04:04:19< vultraz_iOS> ahh
20170409 04:04:34< celticminstrel> Though actually.
20170409 04:04:44< celticminstrel> You can just remove the loop altogether, since .clear() will handle all that for you.
20170409 04:07:10< irker263> wesnoth: Charles Dang wesnoth:master 3b880448169c / src/sound.cpp: Minor fixup to 92d1a26caf17 https://github.com/wesnoth/wesnoth/commit/3b880448169cc142870fd95a8588d63963aedcd1
20170409 04:08:36< vultraz_iOS> looks like https://gna.org/bugs/index.php?25203 is also formula-related
20170409 04:08:39< vultraz_iOS> he should eval it too
20170409 04:09:00< celticminstrel> Looks like your commit didn't cause conflicts with wml_tag_porting.
20170409 04:09:09< vultraz_iOS> why would it?
20170409 04:09:15< celticminstrel> Touching the same file?
20170409 04:09:24< celticminstrel> But different parts of the file, so I guess it was fine.
20170409 04:09:33< vultraz_iOS> ah
20170409 04:09:44< vultraz_iOS> matthiaskrgr has many leaks to test now :P
20170409 04:10:37< celticminstrel> Ugh, stupid GNA, why do I have to download your stupid attachment.
20170409 04:10:49< vultraz_iOS> because gna is antiquated and evil
20170409 04:11:07< celticminstrel> Pretty sure it would've been easier not to.
20170409 04:11:12< celticminstrel> Not to force it I mean.
20170409 04:13:25< irker263> wesnoth: Charles Dang wesnoth:master 76f92a205fb9 / src/storyscreen/interface.cpp: Enable new storyscreen by default (but not remove the old one yet) https://github.com/wesnoth/wesnoth/commit/76f92a205fb94cb9744c595831c67d22b436d6c0
20170409 04:13:52< celticminstrel> Does still need the tiling after all.
20170409 04:15:13< vultraz_iOS> yes
20170409 04:15:18< vultraz_iOS> and delayed storybox showing
20170409 04:16:16< celticminstrel> I think we can do the text fade-in rather easily BTW.
20170409 04:16:38< celticminstrel> Assuming there is a timer-like variable in the storyscreen's canvas.
20170409 04:17:07< celticminstrel> If you set h = "(min(timer, height))" for example, I think the text would slowly show.
20170409 04:17:20< vultraz_iOS> there is not
20170409 04:17:37< celticminstrel> But it would be possible to add one.
20170409 04:18:09< celticminstrel> How does the debug clock do its animations BTW?
20170409 04:18:44< vultraz_iOS> very badly
20170409 04:18:48< vultraz_iOS> the layout is totally unstable
20170409 04:18:54< celticminstrel> Uh. They look pretty good to me.
20170409 04:19:04< celticminstrel> What do you mean it's unstable?
20170409 04:19:34< vultraz_iOS> the entire dialog sprouts a scrollbar if you let it sit too long and it's impossible to keep it in a single position
20170409 04:19:44< vultraz_iOS> but as for the anim itself..
20170409 04:20:10< vultraz_iOS> ...ooo...
20170409 04:20:13< vultraz_iOS> this is interesting...
20170409 04:20:26< vultraz_iOS> it hooks into DRAW events P_P
20170409 04:20:35< celticminstrel> At 800x600 it starts out with a scrollbar.
20170409 04:21:04< vultraz_iOS> is calls a func on every draw and sets canvas variables
20170409 04:21:12< celticminstrel> BTW, do we use a progress bar in the addons manager when uploading / downloading? If not, we should.
20170409 04:21:15< vultraz_iOS> this...this is interesting
20170409 04:21:18< vultraz_iOS> we do
20170409 04:21:24< vultraz_iOS> its just broken on windows
20170409 04:21:30< celticminstrel> Maybe I should take a look at the progress bar.
20170409 04:21:31< celticminstrel> What?
20170409 04:21:34< celticminstrel> Why's it broken?
20170409 04:21:43< celticminstrel> I mean take a look at the addons manager.
20170409 04:21:56< vultraz_iOS> https://gna.org/bugs/index.php?23083
20170409 04:22:06< vultraz_iOS> ASIO bug
20170409 04:22:47< celticminstrel> If it's really an ASIO bug, the status should be set to upstream.
20170409 04:23:32< vultraz_iOS> oh my god the debug clock floods the console
20170409 04:23:43< celticminstrel> Yup.
20170409 04:23:53< celticminstrel> I didn't pay attention to the actual messages.
20170409 04:24:13< celticminstrel> (I was using the debug clock to test my filled circles and also the rounded rectangle. In fact, you can now see the filled circle in action there.)
20170409 04:24:42< vultraz_iOS> it's a lot more stable than last i checked
20170409 04:24:49< celticminstrel> The scrollbar doesn't seem to be growing at all.
20170409 04:24:52< vultraz_iOS> but the scroll panel's scrollbars are rather borked
20170409 04:24:56< vultraz_iOS> yeah
20170409 04:25:07< celticminstrel> The close button doesn't work, but you can close it by clicking the clock button again.
20170409 04:25:41< vultraz_iOS> for the record, it's running faster than 1 second per tick :P
20170409 04:25:50< celticminstrel> ???
20170409 04:26:11< vultraz_iOS> compare your computer's clock to the debug clock
20170409 04:26:59< vultraz_iOS> but again
20170409 04:27:03< vultraz_iOS> it's hooking into DRAW
20170409 04:28:03< vultraz_iOS> tl'dr it's a nice demo but doesn't actually keep proper time
20170409 04:28:11< celticminstrel> I can't seem to find a clock app, but I looked at it again and you're right. It's probably not properly using SDL_GetTicks or whatever to sync to real-time.
20170409 04:29:27< vultraz_iOS> i wonder if hooking into DRAW is of any use to me
20170409 04:29:32< vultraz_iOS> don't think so
20170409 04:29:50< vultraz_iOS> unless...
20170409 04:30:04< vultraz_iOS> well, at least not for the floating images
20170409 04:30:07< vultraz_iOS> those are fine
20170409 04:30:14< celticminstrel> Yeah, it's not calling SDL_GetTicks at all. A more proper implementation would pass something to the step() function based on the result of SDL_GetTicks minus the previous result of SDL_GetTicks.
20170409 04:30:16< vultraz_iOS> I can't think of any better way than scheduled draws
20170409 04:31:22< celticminstrel> Basically it's not even trying to keep proper time.
20170409 04:36:36< vultraz_iOS> yup
20170409 04:38:12< Aginor> scheduled draws isn't a reliable way to keep absolute progress based on time
20170409 04:38:49< Aginor> you still want the progress to continue smoothly even if frames are dropped because it takes too long to render one frame
20170409 04:40:04< Aginor> we'd be better off implementing a slerp function and feed it progress based on sdl ticks
20170409 04:40:18< Aginor> https://docs.unity3d.com/ScriptReference/Vector3.Slerp.html for an example of a slerp function
20170409 04:40:19< vultraz_iOS> slerp?
20170409 04:40:37< Aginor> or a lerp
20170409 04:40:38< Aginor> https://docs.unity3d.com/ScriptReference/Vector3.Lerp.html
20170409 04:40:44< Aginor> different maths backing them
20170409 04:40:47< irker263> wesnoth: Charles Dang wesnoth:master e1a579da51a8 / src/ (21 files in 13 dirs): Deployed std::map::emplace en-masse https://github.com/wesnoth/wesnoth/commit/e1a579da51a88ff787081de621ad4477e8ba5662
20170409 04:42:10< celticminstrel> Implementing that sort of function in WFL should be relatively easy.
20170409 04:42:18< celticminstrel> Could probably even be done as a user-defined function.
20170409 04:42:31< Aginor> it should be in C++ as a part of the general graphics library
20170409 04:42:53 * celticminstrel shrugs.
20170409 04:43:00< celticminstrel> Well, if it was, then it could be trivially exposed to WFL.
20170409 04:43:05< Aginor> yeah
20170409 04:43:32< vultraz_iOS> Aginor: wasn't there some sort of SDL2 renderer library you advised us to use?
20170409 04:43:46< vultraz_iOS> instead of shipping our own shape drawing?
20170409 04:44:11< Aginor> vultraz_iOS: I think that was a fully-featured 3rd party library that was already doing all of the canvas stuff
20170409 04:44:26< vultraz_iOS> ah
20170409 04:45:23< Aginor> here's a good generic lerp example: https://docs.unity3d.com/ScriptReference/Mathf.Lerp.html
20170409 04:45:32< Aginor> instead of providing examples from the 3d API
20170409 04:47:06< celticminstrel> Well, 2D versions could be useful too though.
20170409 04:47:06< Aginor> given that we're doing 2d graphics, yes :D
20170409 04:47:06< celticminstrel> Isn't that particular lerp a really trivial calculation?
20170409 04:47:06< Aginor> it is
20170409 04:47:06< celticminstrel> I wouldn't be surprised if the exact calculation is used explicitly in places in Wesnoth.
20170409 04:47:29< Aginor> but it's also very common, so it's better to librarise it instead of spreading all over the place
20170409 04:47:55< celticminstrel> Mmhm.
20170409 04:49:26< vultraz_iOS> celticminstrel: more [animate_unit] bugs https://forums.wesnoth.org/viewtopic.php?p=610519#p610519
20170409 04:49:30< Aginor> anyhow, back to consulting for me
20170409 04:55:08< vultraz_iOS> ya know, im rather proud I was able to fix that editor crash
20170409 04:55:20< vultraz_iOS> wonder if there's something else similar i can work on
20170409 04:55:22 * vultraz_iOS checks tracker
20170409 05:04:41< vultraz_iOS> lot of planning mode stuff..
20170409 05:11:35< vultraz_iOS> celticminstrel: why did you make the lobby inherit from quit_confirmation?
20170409 05:11:54< celticminstrel> So you get the quit confirmation when pressing X while it's open.
20170409 05:12:06< celticminstrel> Maybe the addons manager should also inherit from it BTW?
20170409 05:12:30< vultraz_iOS> there's a bug now
20170409 05:12:34< vultraz_iOS> just noticed it
20170409 05:12:37< vultraz_iOS> create an mp game
20170409 05:12:39< vultraz_iOS> try to quit
20170409 05:12:43< vultraz_iOS> you get two prompts
20170409 05:12:53< vultraz_iOS> just to return to lobby
20170409 05:13:07< vultraz_iOS> "DO you want to quit" and "do you want to log out"
20170409 05:13:18< vultraz_iOS> not good
20170409 05:13:47< celticminstrel> That's weird.
20170409 05:13:56< celticminstrel> BTW, try to quit from the menu or by pressing X?
20170409 05:14:22< vultraz_iOS> menu
20170409 05:14:38< celticminstrel> What if you press X?
20170409 05:15:27< vultraz_iOS> both
20170409 05:16:01< vultraz_iOS> the quit confirmation stack is static, keep in mind
20170409 05:16:16< vultraz_iOS> so we have two qc objects, so it displays twice
20170409 05:17:48< vultraz_iOS> I think the lobby object isn't destroyed when you enter a game
20170409 05:17:53< vultraz_iOS> even though the dialog is closed
20170409 05:18:04< vultraz_iOS> since the rest of the workflow is invoked in the same scope
20170409 05:19:12< vultraz_iOS> i wonder if this is a problem..
20170409 05:19:55< vultraz_iOS> well, actually it's loop-local
20170409 05:20:22< celticminstrel> Theoretically it shouldn't matter whether the lobby is destroyed.
20170409 05:20:40< celticminstrel> The point of quit_confirmation is supposed to be that the most recently opened scope gets the question, I thought.
20170409 05:20:53< celticminstrel> I should probably try not to stay up all night again...
20170409 05:21:05< vultraz_iOS> i never understood why the qc was implemented as a stack
20170409 05:22:05< celticminstrel> Uh. What else would it be implemented as?
20170409 05:22:37-!- Kwandulin [~Kwandulin@p5DDD1725.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170409 05:23:13< vultraz_iOS> just a single caller?
20170409 05:23:27< celticminstrel> How is that supposed to work?
20170409 05:23:49< vultraz_iOS> nevermind
20170409 05:29:33< irker263> wesnoth: Celtic Minstrel wesnoth:master 7835e05d9372 / data/lua/wml/animate_unit.lua src/scripting/game_lua_kernel.cpp: Fix [animate_unit][facing] not working. https://github.com/wesnoth/wesnoth/commit/7835e05d93727f0f7e276480cee54b0222064bc7
20170409 05:30:20< Kwandulin> Is this regarding [animate_unit] already fixed? https://forums.wesnoth.org/viewtopic.php?p=610407#p610407 Is there a workaround?
20170409 05:31:00< vultraz_iOS> is that this? https://gna.org/bugs/?25625
20170409 05:31:01< celticminstrel> I think there's a workaround... let me check the removed API...
20170409 05:31:06< vultraz_iOS> it's marked Ready For Test
20170409 05:31:24< celticminstrel> Or wait.
20170409 05:31:53< celticminstrel> Oh right, there should be a workaround still.
20170409 05:32:03< vultraz_iOS> but was it fixed
20170409 05:32:17< celticminstrel> In data/lua/wml/animate_unit.lua, you can explicitly use tostring when passing the text to add().
20170409 05:32:22< celticminstrel> That should fix it.
20170409 05:32:23-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20170409 05:32:35< celticminstrel> (A proper fix is already in master.)
20170409 05:32:48< Kwandulin> Great, thanks
20170409 05:32:50< celticminstrel> I thought at first you were linking the same forum post that vultraz_iOS linked.
20170409 05:33:46< celticminstrel> There should be a workaround for that too, which basically means adding a line "unit.facing = dir", but might additionally require passing dir through some function whose name I don't remember (which was removed from the API right after the release).
20170409 05:34:24< celticminstrel> I mean, I can look it up if you care about [animate_unit][facing].
20170409 05:34:34< vultraz_iOS> we should probably release 1.13.8 soon
20170409 05:34:38< vultraz_iOS> and make 1.13.9 RC1
20170409 05:34:42< vultraz_iOS> just to get these bug fixes out
20170409 05:34:51< celticminstrel> We should probably merge wml_tag_porting then.
20170409 05:35:09< vultraz_iOS> before release, yes
20170409 05:35:14< celticminstrel> Though there were still two more tags I wanted to do.
20170409 05:35:22< vultraz_iOS> no hurry
20170409 05:35:29< vultraz_iOS> i want to finish the storyscreen at least
20170409 05:36:44< vultraz_iOS> and the outtro screen
20170409 05:37:03< celticminstrel> ???
20170409 05:37:37< vultraz_iOS> the The End screen
20170409 05:37:49< celticminstrel> That's gonna be a GUI2 dialog too?
20170409 05:38:11< vultraz_iOS> well, it's rather small
20170409 05:38:16< vultraz_iOS> but it should optimally be, yes
20170409 05:38:52< vultraz_iOS> to avoid flip() calls
20170409 05:38:56< vultraz_iOS> and manual layout
20170409 05:38:58< vultraz_iOS> and ttf calls
20170409 05:39:01< celticminstrel> I'd like to get the help dialog done before release too. Maybe I'll try working on that next.
20170409 05:39:18< vultraz_iOS> and it's a good opportunity to implemnt text fade in gui2
20170409 05:39:40< celticminstrel> By "before release" I mean 1.14 though.
20170409 05:41:36< vultraz_iOS> i think we can do text fade
20170409 05:41:40< vultraz_iOS> just not 100% sure hot
20170409 05:41:43< vultraz_iOS> how
20170409 05:42:11< celticminstrel> Just to clarify, what is this text fade you speak of?
20170409 05:44:36< vultraz_iOS> fade in text from 0 alpha
20170409 05:44:42< vultraz_iOS> though techically it's pure black
20170409 05:44:50< vultraz_iOS> but it looks like 0 alpha since the bg is black
20170409 05:44:53< vultraz_iOS> but we should use lpha
20170409 05:44:58< celticminstrel> Ah.
20170409 05:45:24< celticminstrel> It's really kinda unfortunate how GUI2 handles colours.
20170409 05:45:37< celticminstrel> Since it makes it nontrivial to accept formulas in them.
20170409 05:46:56< vultraz_iOS> why?
20170409 05:47:06< vultraz_iOS> oh, you mean the 'eval once'?
20170409 05:47:32< celticminstrel> Huh?
20170409 05:47:46< celticminstrel> I'm talking about how colours are a comma-separated list of their components.
20170409 05:49:07< vultraz_iOS> why not ((r), (g), (b), (a))?
20170409 05:49:25< celticminstrel> Well, that's not a valid formula.
20170409 05:49:43< celticminstrel> ([r, g, b, a]) would be one possibility.
20170409 05:50:55< celticminstrel> (r), (g), (b), (a) would work if red, green, blue, and alpha were each their own key.
20170409 05:52:24< vultraz_iOS> then let's use the former
20170409 05:52:42< celticminstrel> You know how to implement it, right?
20170409 05:55:52< vultraz_iOS> no
20170409 05:56:07< celticminstrel> You need to add a template specialization for typed_formula.
20170409 05:56:44< vultraz_iOS> of what type?
20170409 05:57:45< celticminstrel> Whatever type you're adding a conversion for?
20170409 05:58:35< celticminstrel> Just take a look at the bottom of typed_formula.hpp for examples.
20170409 06:03:14< vultraz_iOS> we already have one for color
20170409 06:04:09< celticminstrel> Eh?
20170409 06:04:17< vultraz_iOS> a specilization 
20170409 06:04:25< vultraz_iOS> specialization 
20170409 06:04:33< celticminstrel> Oh huh.
20170409 06:05:05< celticminstrel> Well in that case, you can just replace any color_t with typed_formula<color_t> and it should work.
20170409 06:05:14< celticminstrel> (I mea, obviously you also need to evaluate the formula somewhere.)
20170409 06:05:20< celticminstrel> ^mean
20170409 06:05:43< vultraz_iOS> but the problem is, even if we do..
20170409 06:05:48< vultraz_iOS> oh right
20170409 06:05:59< vultraz_iOS> the formulas are evaluated every draw
20170409 06:07:31< irker263> wesnoth: Celtic Minstrel wesnoth:master 57efa5cf4735 / src/gui/core/canvas.cpp: Ad f_color to GUIVariable wiki page https://github.com/wesnoth/wesnoth/commit/57efa5cf473512e5354b0545d172b6dfe8c36d39
20170409 06:07:58< celticminstrel> If you change any color_t to typed_formula, be sure to also update the wikidoc comment to say f_color instead of color.
20170409 06:08:13< celticminstrel> ...I should've [ci skipped] that one. Oh well.
20170409 06:08:22< celticminstrel> ...also I misspelled Add. :(
20170409 06:26:30< vultraz_iOS> i assume i only have to call evaluate on the list?
20170409 06:26:34< vultraz_iOS> and variant handles each ?
20170409 06:26:53< celticminstrel> Sorry, what?
20170409 06:28:00< vultraz_iOS> i don't havr t mnaually split the list in typed_formula
20170409 06:28:07< vultraz_iOS> right?
20170409 06:28:20< celticminstrel> typed_formula handles all conversion for you.
20170409 06:30:17< vultraz_iOS> ok
20170409 06:31:53< vultraz_iOS> celticminstrel: how should we indicate we want fading, though?
20170409 06:31:55< vultraz_iOS> not in color=
20170409 06:32:39< vultraz_iOS> we need some other key...
20170409 06:32:43< celticminstrel> What? Wouldn't fading mean animating the alpha channel?
20170409 06:33:00< celticminstrel> Unless I misunderstood something.
20170409 06:33:11< vultraz_iOS> yes
20170409 06:33:35< vultraz_iOS> ah wait i know
20170409 06:33:39< vultraz_iOS> make alpha dependent on a variant
20170409 06:33:40< vultraz_iOS> er
20170409 06:33:42< vultraz_iOS> variable
20170409 06:33:44< vultraz_iOS> set variable in a timer
20170409 06:33:48< vultraz_iOS> ?
20170409 06:33:57< celticminstrel> Something like that yes.
20170409 06:34:16< vultraz_iOS> should the timer be part of the canvas?
20170409 06:34:47< celticminstrel> Why?
20170409 06:35:04< vultraz_iOS> so it should be handled on a per-case basis?
20170409 06:35:10< vultraz_iOS> actually that's probably a good idea
20170409 06:35:22< vultraz_iOS> ok, this should be easy enough to implement
20170409 06:35:29< vultraz_iOS> semi-related
20170409 06:35:34< vultraz_iOS> what about easings?
20170409 06:35:45< vultraz_iOS> we can probably do it much the same way but where should we get the formulas from
20170409 06:36:36< vultraz_iOS> (easings as in these http://easings.net/ )
20170409 06:37:27< celticminstrel> AFAICT that page merely describes ways of interpolating between the initial and final point, so...
20170409 06:37:37< celticminstrel> The choice of formula would handle that.
20170409 06:38:11< celticminstrel> Like, if you want a linear easing, your formula is something like "(timer)", and if you want a cubic one it's maybe "(timer ^ 3)" or something.
20170409 06:39:07< celticminstrel> I'm not sure if it makes sense to include a timer right in the canvas.
20170409 06:39:17< vultraz_iOS> yeah
20170409 06:39:19< celticminstrel> If the canvas is redrawn every tick, I suppose it could.
20170409 06:39:19< vultraz_iOS> that's true
20170409 06:39:26< vultraz_iOS> but it is not
20170409 06:39:29< vultraz_iOS> yet
20170409 06:42:20< Aginor> you can't redraw it every frame without also redrawing everything that may potentially be on top of it
20170409 06:43:22< celticminstrel> Obviously.
20170409 06:43:23< vultraz_iOS> exactly
20170409 06:43:27< vultraz_iOS> which means HW accl
20170409 06:44:55< Aginor> more importantly, it means changing the graphics stack, just doing hardware acceleration on it as it stands now won't work any better
20170409 06:45:04< vultraz_iOS> indeed
20170409 06:45:05< Aginor> it might even be slower when I think about it
20170409 06:48:14< vultraz_iOS> possibly
20170409 06:48:26< vultraz_iOS> but at least we'll just be working with a common rendering base
20170409 06:48:34< vultraz_iOS> instead of all the evil gui1 stuff
20170409 06:48:38< vultraz_iOS> will be easier to refactor
20170409 07:38:11-!- Ravana_ [~Ravana@unaffiliated/ravana/x-2327071] has quit [Ping timeout: 260 seconds]
20170409 07:50:08-!- atarocch [~atarocch@93.56.160.28] has joined #wesnoth-dev
20170409 07:57:00-!- celticminstrel is now known as celmin|sleep
20170409 08:20:00-!- zookeeper [zookeeper@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20170409 08:20:08-!- zookeeper [zookeeper@wesnoth/developer/zookeeper] has quit [Client Quit]
20170409 08:20:37-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20170409 08:21:31-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20170409 08:38:09-!- Kwandulin [~Kwandulin@p5DDD1725.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20170409 08:44:40-!- JyrkiVesterinen [~JyrkiVest@87-100-237-65.bb.dnainternet.fi] has joined #wesnoth-dev
20170409 08:54:26< zookeeper> are the rounded rectangles around the scenario name intentionally kind of... lumpy?
20170409 08:55:01< vultraz_iOS> yes
20170409 08:55:03< vultraz_iOS> it's a WIP
20170409 08:55:12< vultraz_iOS> we might even go with rounded rectangles
20170409 08:55:14< zookeeper> okay
20170409 08:55:26< vultraz_iOS> anything else i should address with the new screen?
20170409 08:55:31< vultraz_iOS> (and do you like the shading at the top)
20170409 08:55:34< zookeeper> well, look at TSG for starters
20170409 08:56:00< zookeeper> umm, what's the point of the top shading?
20170409 08:56:55< vultraz_iOS> I found it made things look a little better
20170409 08:57:01< vultraz_iOS> instead of having the images just starkly end
20170409 08:58:22< zookeeper> well if you want something like that then you at least have to make it consistent, as in always drawn in story screens
20170409 08:58:40< vultraz_iOS> it is
20170409 08:58:47< vultraz_iOS> except if there's a title
20170409 08:58:57< zookeeper> yes. except.
20170409 08:59:12< vultraz_iOS> so I should restore it to always?
20170409 08:59:20< zookeeper> either always or not at all
20170409 08:59:45< zookeeper> also regarding "images just starkly end", pillarboxing seems quite a bit more common than letterboxing anyway, so...
20170409 09:00:43< zookeeper> the story text box seems a lot taller now
20170409 09:02:00< vultraz_iOS> it's shorter o_O
20170409 09:04:09< zookeeper> ...no it's not. 200px vs 280px
20170409 09:04:22< vultraz_iOS> ohh
20170409 09:04:23< vultraz_iOS> right
20170409 09:04:39< vultraz_iOS> the difference is it no longer shrinks to min size if a part caused it to grow
20170409 09:04:52< vultraz_iOS> I see TSG filling the screen
20170409 09:05:17< vultraz_iOS> hm
20170409 09:05:23< vultraz_iOS> I seem to recall it did not do that before..
20170409 09:05:36< vultraz_iOS> uh..
20170409 09:05:37< vultraz_iOS> oh jesus
20170409 09:05:44< vultraz_iOS> the journey images aren't sticking
20170409 09:05:46< vultraz_iOS> the fuck.
20170409 09:07:03< zookeeper> that too
20170409 09:07:36-!- Kwandulin [~Kwandulin@p200300760F3E7D0360733A801C004BFE.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170409 09:09:31-!- irker263 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170409 09:10:02< vultraz_iOS> honestly ive only really tested with httt and atotb
20170409 09:11:06< zookeeper> the background is not horizontally centered correctly
20170409 09:11:11< vultraz_iOS> what?
20170409 09:11:13< vultraz_iOS> p_p
20170409 09:11:27< zookeeper> the background is not horizontally centered correctly
20170409 09:12:02< vultraz_iOS> how so?
20170409 09:12:05< vultraz_iOS> too left?
20170409 09:12:07< vultraz_iOS> too right?
20170409 09:12:40< zookeeper> i'm not going to analyze it
20170409 09:12:54< vultraz_iOS> i don't see it :|
20170409 09:13:07< zookeeper> use your brain
20170409 09:14:51< vultraz_iOS> to be clear, are you referring to the main TSG story images
20170409 09:15:17< vultraz_iOS> (which apparently don't specify the necessary keys to display correctly)
20170409 09:15:26< vultraz_iOS> or there's a bug where defaults are not set correctly..
20170409 09:15:32< zookeeper> the wooden background
20170409 09:15:40< vultraz_iOS> the wooden background..
20170409 09:17:58< vultraz_iOS> i don't see it
20170409 09:18:08< vultraz_iOS> what resolution are you?
20170409 09:19:07< zookeeper> 1920x1200
20170409 09:19:08-!- mjs-de [~mjs-de@x4e304e6e.dyn.telefonica.de] has joined #wesnoth-dev
20170409 09:19:17< vultraz_iOS> that's more than me...
20170409 09:19:21< zookeeper> or whatever that translates as considering the taskbar etc
20170409 09:19:56< vultraz_iOS> there should never be anything more than a 0.5 pixel error
20170409 09:20:04< vultraz_iOS> can I see a screenshot?
20170409 09:20:06< zookeeper> and you're saying you're actually making a comparison to 1.13.7 or whatever instead of just staring at the screen and trying to see a problem?
20170409 09:20:20< vultraz_iOS> the latter, of course
20170409 09:20:43< zookeeper> that's why i said use your brain
20170409 09:20:43< vultraz_iOS> let me pull up the old screen and see...
20170409 09:23:50-!- Kwandulin [~Kwandulin@p200300760F3E7D0360733A801C004BFE.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds]
20170409 09:24:29< vultraz_iOS> ok, there seems to be a tiny, tiny difference between the two screens
20170409 09:24:31< vultraz_iOS> gui1 and 2
20170409 09:24:39< vultraz_iOS> just like with the pixel positions of floating images
20170409 09:24:45< vultraz_iOS> a few tiny pixels difference
20170409 09:25:14< zookeeper> sigh
20170409 09:27:13< vultraz_iOS> y u do dis :|
20170409 09:27:49< vultraz_iOS> I'm looking at a side by side comparison of the same screen with both versions
20170409 09:27:56< vultraz_iOS> and I don't see anything!
20170409 09:28:49< vultraz_iOS> what am I supposed to see!
20170409 09:29:01< zookeeper> https://www.dropbox.com/s/nq5hwta0cyc68za/vultraz_wtf.gif?dl=0
20170409 09:30:08< vultraz_iOS> ahhhhhh
20170409 09:30:19< vultraz_iOS> ok, see, now I see it, but the effect is *much* subtler for me here
20170409 09:35:38< vultraz_iOS> ok I *think* i know what's going on
20170409 09:36:04< vultraz_iOS> you don't have a widescreen monitor - I can tell since there're no black borders
20170409 09:36:20< vultraz_iOS> therefor, the old dialog must have trimmed equal areas from both sides of the image
20170409 09:36:30< vultraz_iOS> while my version draws at 0,0 and cuts off to the right
20170409 09:36:47< vultraz_iOS> since I have a widescreen monitor, it's impossible for me to observe this
20170409 09:37:15< vultraz_iOS> (unless i decrease resolution, of course)
20170409 09:37:33< vultraz_iOS> so no need to get all passive-aggressive :|
20170409 09:38:27< zookeeper> no, you can see the exact same thing with for example 1680x1050
20170409 09:40:47< vultraz_iOS> which i do not have.
20170409 09:40:50< vultraz_iOS> but doesn't matter
20170409 09:41:01< zookeeper> how do you "not have" it?
20170409 09:41:03< vultraz_iOS> I have an idea for a fix
20170409 09:41:40< vultraz_iOS> I don't have it since wesnoth is running at an up-scaled 1536x801 since we don't have high DPI support for my 4k monitor
20170409 09:42:13-!- Kwandulin [~Kwandulin@p200300760F3E7D03D87D4BCDA6716761.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170409 09:42:25< zookeeper> oh, so all resolutions above 1536x801 are unavailable to you?
20170409 09:43:30< vultraz_iOS> unless i disable DPI compatibility mode, yes
20170409 09:43:35< zookeeper> well, you didn't say you have some quirky setup where you can't use common resolutions
20170409 09:45:05< vultraz_iOS> if we *did* have high-dpi support I'd be running at 3840 × 2160
20170409 09:47:12< zookeeper> so where does 1536x801 come from?
20170409 09:48:46< vultraz_iOS> automatic Windows non-dpi compatibility scaling
20170409 09:49:12< zookeeper> so confusing
20170409 09:49:12< vultraz_iOS> "virtualizes" 90 DPI
20170409 09:49:30< vultraz_iOS> and some some reason, that translates to 1536x801
20170409 09:49:41< vultraz_iOS> so wesnoth fills my screen, but stuff is up-scaled
20170409 09:51:05< vultraz_iOS> windows does it rather well, actually
20170409 09:51:09< vultraz_iOS> you stop noticing after awhile
20170409 09:51:20< vultraz_iOS> obviously it's not as crisp as native res, sadly
20170409 09:53:46< zookeeper> i've never understood this thing about DPI. i don't know or care what my DPI is, why would i care even if i had a 4k monitor? presumably i'd just want some things to run upscaled 2x
20170409 09:54:29< vultraz_iOS> high DPI means you can no longer measure by pixels size
20170409 09:54:32< vultraz_iOS> pixel*
20170409 09:54:41< vultraz_iOS> because everything would become tiny
20170409 09:54:50< vultraz_iOS> it's incredibly crisp, though 
20170409 09:54:54< vultraz_iOS> you see no pixelation
20170409 10:15:27-!- fkhodkov [~user@ppp-188-65-8-129-bras1.istra.ru] has joined #wesnoth-dev
20170409 10:15:55-!- fkhodkov [~user@ppp-188-65-8-129-bras1.istra.ru] has quit [Client Quit]
20170409 10:29:28-!- JyrkiVesterinen [~JyrkiVest@87-100-237-65.bb.dnainternet.fi] has quit [Quit: .]
20170409 10:37:56-!- irker162 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20170409 10:37:56< irker162> wesnoth: Charles Dang wesnoth:master 10f867bd7e1d / data/gui/schema.cfg src/gui/core/canvas.cpp: GUI2/Canvas: evaluate text color as formula https://github.com/wesnoth/wesnoth/commit/10f867bd7e1d7520ba9de46b7c1636309e0bfba0
20170409 10:37:57< irker162> wesnoth: Charles Dang wesnoth:master dec0fd6dfdc5 / / (11 files in 7 dirs): Convert outro screen to GUI2 https://github.com/wesnoth/wesnoth/commit/dec0fd6dfdc51f125ef44f380061d539fe3e5e96
20170409 10:38:23< vultraz_iOS> celmin|sleep: F A D I N G T E X T ^
20170409 10:39:07-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170409 10:40:32-!- Ravana_ [~Ravana@unaffiliated/ravana/x-2327071] has joined #wesnoth-dev
20170409 10:45:10< vultraz_iOS> as a bonus, now the text fades OUT too
20170409 10:50:46< vultraz_iOS> zookeeper: oh, btw, after looking at your gif i know why the box looks bigger. before it had a 200 px minimum size and tried to shrink to that every part. The new screen has a screen height / 4 minimum
20170409 10:51:01< vultraz_iOS> s/new screen/new box
20170409 10:53:37< vultraz_iOS> perhaps i should enforce a min(200/height/4) to guard aginst really tiny resolutions
20170409 10:53:47< vultraz_iOS> 200, height/4*
20170409 10:58:59< irker162> wesnoth: Charles Dang wesnoth:master 53adf0ba1a5e / data/gui/window/story_viewer.cfg: Story Viewer: enforce at least 200 height for the story box https://github.com/wesnoth/wesnoth/commit/53adf0ba1a5ed46eaf9f01b0b79a41378cb3afec
20170409 11:05:04< irker162> wesnoth: Charles Dang wesnoth:master 8ed11e750632 / data/gui/schema.cfg data/gui/widget/text_box_default.cfg src/gui/core/canvas.cpp: GUI2/Canvas: removed [line] alpha= key in favor of evaluating color as a formula https://github.com/wesnoth/wesnoth/commit/8ed11e75063225b4e7f284e3d4566c0446a81fe9
20170409 11:08:44< zookeeper> vultraz_iOS, is it intentional that the main text box background is now much darker, too?
20170409 11:09:01< vultraz_iOS> it is? it shouldn't be...
20170409 11:09:12< vultraz_iOS> but it doesn't really matter
20170409 11:10:06< zookeeper> you see it even in the gif
20170409 11:11:07< zookeeper> it doesn't matter otherwise, except by making the parts of the map/whatever behind it harder to read if one wants to. if it's needed for legibility of the text then it's fine, but it should be only as dark as it needs to be for the text legibility to not suffer
20170409 11:26:19-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170409 11:26:31-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170409 11:26:37-!- stano [~stano@84.245.120.53] has joined #wesnoth-dev
20170409 11:26:41< stano> hi all
20170409 11:29:05-!- prkc [~prkc@d0.85.7a9f.ip4.static.sl-reverse.com] has joined #wesnoth-dev
20170409 11:33:51-!- Shiki [~Shiki@p5485540F.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170409 11:34:08< vultraz_iOS> hello
20170409 11:39:50-!- JyrkiVesterinen [~JyrkiVest@87-100-237-65.bb.dnainternet.fi] has joined #wesnoth-dev
20170409 11:43:51< Shiki> hi
20170409 11:52:34< stano> i like to try the new campaign, so i am installing 1.3.7 from source. very slow and i have no idea if it will work :)
20170409 12:21:32-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170409 12:23:16-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170409 12:25:33-!- Kwandulin [~Kwandulin@p200300760F3E7D03D87D4BCDA6716761.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20170409 13:03:02< irker162> wesnoth: Charles Dang wesnoth:master 08d25b05427e / src/gui/auxiliary/typed_formula.hpp: GUI2: refactor typed_formula to make code cleaner https://github.com/wesnoth/wesnoth/commit/08d25b05427e113cfdcf2b5d71137dadac89ffa4
20170409 13:03:05< irker162> wesnoth: Charles Dang wesnoth:master c379b7ccc9c4 / src/ (color.cpp color.hpp): Added getter for null color https://github.com/wesnoth/wesnoth/commit/c379b7ccc9c440ca935228fe78241d63a9b841f6
20170409 13:03:08< irker162> wesnoth: Charles Dang wesnoth:master 61aa601a6ff6 / data/gui/schema.cfg src/gui/core/canvas.cpp: GUI2/Canvas: converted other shapes to use formulas for their color handling https://github.com/wesnoth/wesnoth/commit/61aa601a6ff63fdcaca3545c1ade4d8851fa6b3f
20170409 13:13:17-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170409 13:15:49-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170409 13:19:33-!- Kwandulin [~Kwandulin@p200300760F3E7D03D87D4BCDA6716761.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170409 13:56:02-!- Shiki [~Shiki@p5485540F.dip0.t-ipconnect.de] has quit [Remote host closed the connection]
20170409 13:57:54-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20170409 14:06:50-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170409 14:08:36-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170409 14:10:57< irker162> wesnoth: mattsc wesnoth:master 4d76be9d6104 / projectfiles/Xcode/Wesnoth.xcodeproj/project.pbxproj: Update Xcode project https://github.com/wesnoth/wesnoth/commit/4d76be9d610447c42e35625d7e77d428367f55d9
20170409 14:11:25< mattsc> 162 warnings  :P
20170409 14:11:28< mattsc> src/gui/auxiliary/typed_formula.hpp:260:50: warning: unused parameter 'v' [-Wunused-parameter]
20170409 14:11:28< mattsc> inline T typed_formula<T>::execute(wfl::variant& v) const
20170409 14:13:26-!- Shiki [~Shiki@p5485540F.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170409 14:16:09< irker162> wesnoth: Charles Dang wesnoth:master 528e6b89b83e / src/gui/auxiliary/typed_formula.hpp: Fixup 08d25b0 https://github.com/wesnoth/wesnoth/commit/528e6b89b83ec3265d661f2b8a2c92f03632defc
20170409 14:16:12< vultraz_iOS> mattsc: ^
20170409 14:24:58< mattsc> vultraz_iOS: thanks; it’s probably not necessary to say that this does the trick
20170409 14:53:01 * celmin|sleep wonders what zookeeper means by lumpy
20170409 14:56:57< celmin|sleep> vultraz_iOS: You forgot to update the doc comment for [text].
20170409 14:59:36-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170409 15:01:20< celmin|sleep> What's with https://github.com/wesnoth/wesnoth/commit/08d25b05427e113cfdcf2b5d71137dadac89ffa4#diff-202bd17361bd425e29ae5f38327e3e74R117 ?
20170409 15:01:28-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170409 15:01:53< celmin|sleep> Why'd you remove the execute function?
20170409 15:02:34< celmin|sleep> Oh wait, I see, it's private.
20170409 15:06:59< celmin|sleep> But anyway vultraz_iOS, update the appropriate wikidoc comments in canvas.cpp to indicate that certain keys are now f_color instead of color.
20170409 15:13:30-!- Shiki [~Shiki@p5485540F.dip0.t-ipconnect.de] has quit [Remote host closed the connection]
20170409 15:14:48-!- timotei_ [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 240 seconds]
20170409 15:17:18-!- timotei_ [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20170409 15:22:15< vultraz_iOS> I'll get to it tomorrow if you don't first
20170409 15:33:30-!- celmin|sleep is now known as celticminstrel
20170409 15:52:29-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170409 15:52:40< zookeeper> celticminstrel, there's nothing that you think i could reasonably be referring to? okay, i'll take a screenshot
20170409 15:54:15-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170409 15:55:13< zookeeper> https://www.dropbox.com/s/ta922d09hbbagcc/lumpy_title_box.jpg?dl=0
20170409 15:57:00< irker162> wesnoth: Jyrki Vesterinen wesnoth:master cec90f537997 / projectfiles/VC12/ (wesnoth.vcxproj wesnoth.vcxproj.filters): Update Visual Studio project https://github.com/wesnoth/wesnoth/commit/cec90f537997b685ca3d80ca8a52f94cc5fb585d
20170409 15:57:10< celticminstrel> Okay, I think I get what you mean, but why is that shot zoomed in so much?
20170409 15:57:42< zookeeper> for your convenience
20170409 15:57:48< celticminstrel> I'm assuming you're referring to the lumps on the corners, not the layering.
20170409 15:57:55< zookeeper> yes, the corners
20170409 15:58:10< celticminstrel> Maybe my circle-filling calculations are off-by-one somewhere...
20170409 15:58:29< celticminstrel> Or the calculations for the round rectangle...
20170409 16:01:29< zookeeper> if you're drawing the corners as 1/4ths of a circle then seems like there's an off-by-1 (or several) somewhere.
20170409 16:13:34-!- Appleman1234 [~Appleman1@pl26331.ag1212.nttpc.ne.jp] has quit [Ping timeout: 264 seconds]
20170409 16:14:45-!- SigurdFD [~SigurdFD@dynamic-acs-72-23-110-196.zoominternet.net] has joined #wesnoth-dev
20170409 16:39:54-!- Kwandulin [~Kwandulin@p200300760F3E7D03D87D4BCDA6716761.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20170409 16:45:16-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170409 16:47:13-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170409 16:57:47< SigurdFD> 20170408 19:13:48< gfgtdf> SigurdFD: i think [store_side] income shoudl returns the same as the the [side]income= key, everythign else sounds just confusing
20170409 16:59:14< SigurdFD> gfgtdf: I agree with this. currently & for several years [store_side] income is different than [side] or [modify_side] versions
20170409 17:00:45< SigurdFD> zookeeper, celticminstrel: thoughts? ^
20170409 17:03:54-!- stano [~stano@84.245.120.53] has quit [Quit: Leaving]
20170409 17:08:23-!- Kwandulin [~Kwandulin@p200300760F3E7D03D87D4BCDA6716761.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170409 17:12:20-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20170409 17:12:59-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170409 17:17:56< zookeeper> SigurdFD, i don't have anything new to add
20170409 17:23:35-!- Appleman1234 [~Appleman1@pl26740.ag1212.nttpc.ne.jp] has joined #wesnoth-dev
20170409 17:41:03< vultraz_iOS> celticminstrel: I was wondering if we could get rid of that wfl/wflend stuff and replace it with {} or even a [wfl] key.
20170409 17:41:17< celticminstrel> Why bother? It works as it is.
20170409 17:41:32< celticminstrel> SigurdFD: What exactly was the current situation, again?
20170409 17:43:43< vultraz_iOS> syntax improvement
20170409 17:43:53< celticminstrel> I don't see it as an improvement.
20170409 17:44:03< celticminstrel> I see it as no worse and no better than the current situation.
20170409 17:44:16< celticminstrel> Well okay, I suppose {} is slightly worse.
20170409 17:44:19< vultraz_iOS> granted
20170409 17:44:23< SigurdFD> celticminstrel: I'll have to explain later, out of time now.
20170409 17:44:27< vultraz_iOS> it's really low on my list anyway
20170409 17:44:29-!- SigurdFD [~SigurdFD@dynamic-acs-72-23-110-196.zoominternet.net] has quit []
20170409 17:54:27< vultraz_iOS> I can't figure out why the joyney markers vanish from the storyscreen :/
20170409 17:55:13< celticminstrel> BTW, I think maybe the storyscreen and outtro should both use the animation method of the debug clock.
20170409 17:55:31< celticminstrel> Plus usd of SDL_GetTicks to keep it properly timed.
20170409 17:55:33< celticminstrel> ^use
20170409 17:55:42< vultraz_iOS> hook into DRAW, you mean?
20170409 17:55:53< celticminstrel> DRAW is fired incessantly, correct?
20170409 17:56:12< vultraz_iOS> isn't draw essentially "frame drawing"?
20170409 17:56:19< celticminstrel> The canvas may not be every tick, but the event is fired every tick, isn't it?
20170409 17:56:20< vultraz_iOS> meaning it's very frequent
20170409 17:56:30< celticminstrel> Right, that's why you use SDL_GetTicks.
20170409 17:56:41< vultraz_iOS> but why are you saying I should use this method
20170409 17:57:29< celticminstrel> Well, your current method is scheduled events, right? Something like JS's setTimeour? ie, using the timer interface in gui/auxiliary for example.
20170409 17:58:36< celticminstrel> I haven't looked at your code, just basing on things you've said in IRC.
20170409 18:00:00< vultraz_iOS> yes, scheduled events. In the outtro, I fire an event every 50 ms to set the canvas fade_step variable, and once it's faded in, I spawn another 1-time timer to set a flag to allow the process to function in reverse after the specified duration
20170409 18:00:20< celticminstrel> What was the guaranteed granularity of the timer, again?
20170409 18:00:41< vultraz_iOS> in the storyscreen, I use consecutive timers to spawn more timers
20170409 18:00:49< vultraz_iOS> ie, i call a function to draw the first image
20170409 18:00:59< vultraz_iOS> which spawns a timer to draw the second after a duration
20170409 18:01:06< vultraz_iOS> and that the third
20170409 18:01:26< vultraz_iOS> I don't remember the guaranteed granularity 
20170409 18:01:33< celticminstrel> So basically like JavaScript's setTimeout, indeed.
20170409 18:01:46< celticminstrel> Well, minus closures I guess.
20170409 18:03:00-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170409 18:03:26-!- gfgtdf [~chatzilla@x4e3684f9.dyn.telefonica.de] has joined #wesnoth-dev
20170409 18:03:46< gfgtdf> should config::merge_attributes ignore (=treat as nonexistent) blank attributes? I think yes but i'm not 100% sure
20170409 18:04:25< vultraz_iOS> id say yes
20170409 18:04:34< celticminstrel> By "blank attributes" do you mean attributes specified in the config as just the name with no value?
20170409 18:04:39-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170409 18:04:45< celticminstrel> Or do you mean attributes set in the C++ code explicitly to boost::none?
20170409 18:04:55< gfgtdf> celticminstrel: set explicitly to boost::none
20170409 18:05:40< celticminstrel> Setting an attribute to boost::none (though probably bad practice) should be entirely equivalent to deleting it. In fact, such attributes should probably be excluded when serializing.
20170409 18:05:57< gfgtdf> afaik they are
20170409 18:07:08< irker162> wesnoth: gfgtdf wesnoth:master 7e5e1e762f7e / src/game_initialization/connect_engine.cpp: attempt to fix faction info sometimes missing in [side] https://github.com/wesnoth/wesnoth/commit/7e5e1e762f7e0477a88b16ff6626660850fb35cf
20170409 18:07:44< vultraz_iOS> celticminstrel: I think I know why the images disappear
20170409 18:08:15< vultraz_iOS> celticminstrel: but it's 5 am and i haven't been able to sleep so tell me if this sounds reasonable...
20170409 18:08:35< vultraz_iOS> the position of each is controlled by variables
20170409 18:08:36< vultraz_iOS> in wfl
20170409 18:08:45< vultraz_iOS> but those are reset for every shape
20170409 18:09:04< vultraz_iOS> so even though the shapes are on the canvas
20170409 18:09:15< vultraz_iOS> they all have the same *variables*
20170409 18:09:18< vultraz_iOS> but the values change
20170409 18:09:29< vultraz_iOS> which means they're all getting drawn at the same spot 
20170409 18:09:52< vultraz_iOS> celticminstrel: is there some way to serialize the wfl result to the canvas config
20170409 18:10:12< vultraz_iOS> though again, that's a shape modifying the canvas...
20170409 18:10:26< vultraz_iOS> i mean, we *get* the value..
20170409 18:10:29< celticminstrel> I don't quite get what you're saying...
20170409 18:10:31< vultraz_iOS> I'd just need to be able to write it backwards..
20170409 18:11:53< vultraz_iOS> celticminstrel: imagine you have two images. you say, draw image 1 at foo, bar, and foo = 10 and bar = 10, and then you change foo and bar to equal 20 and draw image 2 *without* serializing the '10' in the first case
20170409 18:12:00< vultraz_iOS> think... uh.. references instead of copies
20170409 18:12:31< celticminstrel> I don't understand why you say there are references instead of copies.
20170409 18:12:44< celticminstrel> I mean, I'm still not even sure what you're talking about.
20170409 18:12:49< vultraz_iOS> storyscreen
20170409 18:12:52< vultraz_iOS> floating images
20170409 18:13:01< vultraz_iOS> they're not drawing as expected
20170409 18:13:09< vultraz_iOS> you only see them in one place at once
20170409 18:13:34< celticminstrel> So "the position of each is controlled by variables" mean...
20170409 18:13:37< vultraz_iOS> and I think that's because in the canvas config, *every floating image has it's x= key, say, this:
20170409 18:13:39< vultraz_iOS> "(trunc(fi_ref_x * base_scale_x) + base_origin.x - (image_original_width  / 2))"
20170409 18:13:46< vultraz_iOS> so, when draw happens...
20170409 18:13:58< vultraz_iOS> each image is drawn and that's evaluated
20170409 18:14:04< vultraz_iOS> and it all evaluates to the same image
20170409 18:14:11< celticminstrel> Wait, how?
20170409 18:15:18< vultraz_iOS> because ... ahh
20170409 18:15:46< vultraz_iOS> because shapes (or in this case image) do not set formula evaluations
20170409 18:15:51< vultraz_iOS> in their member values
20170409 18:16:11< vultraz_iOS> so for all intents and purposes, the shape is storing "(trunc(fi_ref_x * base_scale_x) + base_origin.x - (image_original_width  / 2))" in x_
20170409 18:17:01< celticminstrel> ...huh?
20170409 18:17:03< vultraz_iOS> celticminstrel: we need to add a serialized_value field to typed_formula
20170409 18:17:12< vultraz_iOS> gahhh
20170409 18:17:18< vultraz_iOS> how are you not understanding this :(
20170409 18:17:23< celticminstrel> I don't know.
20170409 18:17:32< vultraz_iOS> the image shapes
20170409 18:17:36< vultraz_iOS> take the data from the config
20170409 18:17:42< vultraz_iOS> store the formula string
20170409 18:17:46< vultraz_iOS> evaluate it every draw
20170409 18:17:57< vultraz_iOS> but I don't want evaluate every draw here
20170409 18:18:08< vultraz_iOS> I want to save the result of the first evaluation of the formula
20170409 18:18:21< vultraz_iOS> so I need to add such ability to typed_formula
20170409 18:18:23< celticminstrel> Why?
20170409 18:18:27< celticminstrel> Why would you want that?
20170409 18:18:40< vultraz_iOS> .........
20170409 18:19:09< vultraz_iOS> because if you have 5 images of the same size and their x formula is the same they will end up in the same location!
20170409 18:19:19< vultraz_iOS> it's even talked about in the todo comment!
20170409 18:19:21< vultraz_iOS> 	/**
20170409 18:19:21< vultraz_iOS> 	 * @todo formulas are now recalculated every draw cycle which is a  bit
20170409 18:19:22< vultraz_iOS> 	 * silly unless there has been a resize. So to optimize we should use an
20170409 18:19:22< vultraz_iOS> 	 * extra flag or do the calculation in a separate routine.
20170409 18:19:22< vultraz_iOS> 	 */
20170409 18:19:25< celticminstrel> Why would they end up in the same location?
20170409 18:19:37< celticminstrel> They should have different values of fi_ref_x
20170409 18:19:47< celticminstrel> ...
20170409 18:20:07< vultraz_iOS> no!
20170409 18:20:18< vultraz_iOS> the formula says "use the value of fi_ref_x"
20170409 18:20:24< vultraz_iOS> there's no variable scope
20170409 18:20:32< vultraz_iOS> it's a *canvas variable*
20170409 18:20:48< vultraz_iOS> so if i have images all using fi_ref_x they will all produce the same calculation!
20170409 18:21:22-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20170409 18:21:44-!- Kwandulin [~Kwandulin@p200300760F3E7D03D87D4BCDA6716761.dip0.t-ipconnect.de] has quit [Ping timeout: 255 seconds]
20170409 18:21:49< vultraz_iOS> or i need to implement some way to set the variables in the local context....
20170409 18:22:00< celticminstrel> I get the issue now.
20170409 18:22:05< celticminstrel> Looking at the code.
20170409 18:22:10< vultraz_iOS> *thank you*
20170409 18:22:28< vultraz_iOS> *what do you propose*
20170409 18:23:20-!- Kwandulin [~Kwandulin@p200300760F3E7D03D87D4BCDA6716761.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170409 18:24:12-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20170409 18:25:07< vultraz_iOS> it's essentially me saying, let x = a. draw something at x. now let x = b. draw something at x. now draw both of these.
20170409 18:25:10< vultraz_iOS> both will be at b
20170409 18:27:20< vultraz_iOS> I see 2 possibilities
20170409 18:27:45< vultraz_iOS> 1. Implement the ability to store serialized data in typed_formula
20170409 18:28:32-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170409 18:28:39< vultraz_iOS> 2. Make the local variables class member variables of the relevant shape class and set the variable in that context so it persists between draws
20170409 18:28:44< vultraz_iOS> Not sure fully if that would work
20170409 18:28:46< celticminstrel> Hmm, I think what I would propose is eliminating fi_ref_x and fi_ref_y altogether.
20170409 18:28:57< celticminstrel> Instead, hard-code them into each image's formula.
20170409 18:29:11< vultraz_iOS> I can't!
20170409 18:29:22< celticminstrel> You can.
20170409 18:29:31< vultraz_iOS> How?
20170409 18:29:42< celticminstrel> What you do now is set them on the canvas, then set a constant string for the formula. So you do have the data you'd need when you need it.
20170409 18:30:01< celticminstrel> Instead of setting a constant string as the formula, build the string so that fi_ref_x is coded into the formula as an integer constant.
20170409 18:30:32< celticminstrel> So if the image's reference x is 12, the formula is now set to "(trunc(12 * base_scale_x) + base_origin.x - (image_original_width  / 2))"
20170409 18:31:05< vultraz_iOS> I see...
20170409 18:31:09< vultraz_iOS> Yes, I could do this...
20170409 18:31:12-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170409 18:31:25< vultraz_iOS> But are any of the two suggestions worth implementing g
20170409 18:32:04< celticminstrel> I actually thought typed_formula already had some caching mechanism, though it probably wouldn't apply in this particular case.
20170409 18:32:43< celticminstrel> I don't think it's especially useful to serialize the result to produce an "evaluate once" formula, but maybe there are uses I hadn't thought of.
20170409 18:33:12< celticminstrel> As for option 2...
20170409 18:34:04< celticminstrel> I suppose you could add a [varaibles] subtag to shapes which has the same function as the [variables] subtag in FormulaAI, if you wanted.
20170409 18:34:18< celticminstrel> ^variables
20170409 18:34:40< celticminstrel> It could be useful if there are variables that are used by the formulas of multiple keys
20170409 18:34:59< celticminstrel> But which need to be set to some complex value.
20170409 18:35:24< celticminstrel> If you're adding it to shapes and/or canvas I'd say you might as well also add it to window.
20170409 18:35:31< celticminstrel> Oh, but there might be one problem with it.
20170409 18:36:00< celticminstrel> I'm not sure if the schema syntax has a way to specify "any attribute is allowed", so it might be necessary to extend the schema syntax too.
20170409 18:36:27< vultraz_iOS> .
20170409 18:36:34< vultraz_iOS> .....
20170409 18:36:37< vultraz_iOS> Well then
20170409 18:36:46< celticminstrel> You could look into it, I suppose.
20170409 18:37:19< celticminstrel> If the schema doesn't support that, it definitely should, because general WML has a number of cases where any attribute is alllowed.
20170409 18:40:40< celticminstrel> ...I have an extra L there. :(
20170409 18:41:07< celticminstrel> Anyway, if you want to implement [variables] and the schema format does need extending, I wouldn't mind handling the schema side of it.
20170409 18:42:00< celticminstrel> The idea would be that every key in [variables] is a formula which sets a variable named after the key (so actually, keys with numbers aren't allowed).
20170409 18:43:04< gfgtdf> i just started 1.12.7 relese and there are a whole 7 games player on the mp server
20170409 18:43:08< gfgtdf> 1.13.7*
20170409 18:43:35< gfgtdf> i always thought deve mp server are generally empty
20170409 18:47:54-!- markus_ [~mjs-de@x4e30a981.dyn.telefonica.de] has joined #wesnoth-dev
20170409 18:47:54-!- mjs-de [~mjs-de@x4e304e6e.dyn.telefonica.de] has quit [Read error: Connection reset by peer]
20170409 18:49:05-!- Shiki [~Shiki@p5485540F.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170409 19:06:42< zookeeper> gfgtdf, well... 7 games isn't much. but probably dev versions get more players the older the stable branch gets.
20170409 19:22:14-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170409 19:23:51-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170409 19:38:25< gfgtdf> also there seems to be a bug that results in OOS every other game you observe
20170409 19:39:12< gfgtdf> might be a reace condition not sure
20170409 19:49:58-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20170409 19:52:38-!- timotei_ [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 252 seconds]
20170409 20:16:52-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170409 20:18:47-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170409 20:56:57-!- JyrkiVesterinen [~JyrkiVest@87-100-237-65.bb.dnainternet.fi] has quit [Quit: .]
20170409 21:08:49-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170409 21:09:01-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170409 21:09:20-!- irker162 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170409 21:17:37-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20170409 21:27:47-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has quit [Quit: Connection closed for inactivity]
20170409 21:33:01-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170409 21:33:31-!- Kwandulin [~Kwandulin@p200300760F3E7D03D87D4BCDA6716761.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds]
20170409 21:34:44-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170409 21:50:45-!- Shiki [~Shiki@p5485540F.dip0.t-ipconnect.de] has quit [Remote host closed the connection]
20170409 21:52:14-!- markus_ [~mjs-de@x4e30a981.dyn.telefonica.de] has quit [Remote host closed the connection]
20170409 21:53:57-!- atarocch [~atarocch@93.56.160.28] has quit [Remote host closed the connection]
20170409 22:04:35-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170409 22:04:41-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170409 22:14:04-!- gfgtdf [~chatzilla@x4e3684f9.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.93 [Firefox 52.0.2/20170323105023]]
20170409 22:37:12-!- un214 [~un214@76-14-112-211.rk.wavecable.com] has joined #wesnoth-dev
20170409 23:34:58-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 252 seconds]
--- Log closed Mon Apr 10 00:00:54 2017