--- Log opened Sat Apr 02 00:00:10 2016
20160402 00:02:38< gfgtdf> vultraz: please fix this issue, its mostlikeley casues by using std::tostring in config.cpp
20160402 00:06:24-!- irker659 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160402 00:06:24< irker659> wesnoth: Charles Dang wesnoth:master c5ac29d505a6 / src/config.cpp: Revert use of std::to_string in config https://github.com/wesnoth/wesnoth/commit/c5ac29d505a6d8e5b9845a01e43268067dfbad8a
20160402 00:06:35< vultraz> gfgtdf: yeah that fixes it
20160402 00:06:37-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160402 00:06:42-!- AI0867 [~ai@wesnoth/developer/ai0867] has quit [Ping timeout: 244 seconds]
20160402 00:06:50-!- AI0867 [~ai@wesnoth/developer/ai0867] has joined #wesnoth-dev
20160402 00:09:33-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 246 seconds]
20160402 00:17:45-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20160402 00:17:52-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20160402 00:26:05-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20160402 00:27:00-!- prkc [~prkc@gateway/vpn/privateinternetaccess/prkc] has quit [Remote host closed the connection]
20160402 00:28:42-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160402 00:32:21-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has joined #wesnoth-dev
20160402 00:37:10-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160402 00:37:52< vultraz> ancestral: so what're we doing with the font PR? should we merge it, or wait for gui1, or what.
20160402 00:38:31< ancestral> Has anyone with Linux tested it?
20160402 00:39:25-!- travis-ci [~travis-ci@ec2-54-198-118-249.compute-1.amazonaws.com] has joined #wesnoth-dev
20160402 00:39:26< travis-ci> wesnoth/wesnoth#9171 (master - c5ac29d : Charles Dang): The build has errored.
20160402 00:39:26< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/120207571
20160402 00:39:26-!- travis-ci [~travis-ci@ec2-54-198-118-249.compute-1.amazonaws.com] has left #wesnoth-dev []
20160402 00:40:36< vultraz> I don't believe so
20160402 00:40:40< ancestral> I just want to make sure it looks OK in Ubuntu, Debian or CentOS
20160402 00:40:48< ancestral> aeth: Up to anything?
20160402 00:40:57< vultraz> what even is CentOS
20160402 00:41:20< ancestral> It’s a cereal
20160402 00:41:30< ancestral> https://en.wikipedia.org/wiki/CentOS
20160402 00:41:48< ancestral> I was just throwing it out as an example
20160402 00:41:55< vultraz> ah
20160402 00:42:07< vultraz> (I know it was a distro, but I thought it was special or something)
20160402 00:46:43< Aginor> vultraz: it's the community version of redhat enterprise
20160402 00:47:22< Aginor> it's popular in academic institutions
20160402 00:47:58< celticminstrel> ancestral: Any chance of trying to build the unit tests?
20160402 00:48:13< celticminstrel> There's something wrong with them still that I can't figure out.
20160402 00:48:16< ancestral> Yeah, between now and Sunday
20160402 00:48:34< celticminstrel> (Also, we probably won't be supporting CentOS anymore, at least the current stable version.)
20160402 00:48:40< celticminstrel> (Unstable should be fine.)
20160402 00:51:10< ancestral> celticminstrel: Is it an extension of the boost libs, or is it a separate package, or?
20160402 00:51:28< ancestral> I should be building 1.60.0?
20160402 00:51:49< celticminstrel> I'm using 1.57 at the moment. I think vultraz is using 1.58.
20160402 00:52:01< ancestral> What should I be building?
20160402 00:52:04< gfgtdf> we could use c++11 custom literals to use "string"t to make strings translatable, not sure if this is better though.
20160402 00:52:13< celticminstrel> I don't know any reasons to not use 1.60, though vultraz had problems with it.
20160402 00:52:29< celticminstrel> ancestral: I was talking about the unit tests target in the XCode project, sorry for confusion.
20160402 00:52:31< vultraz> Yes, I cannot link with my 1.60 build
20160402 00:52:33< gfgtdf> i'm using 1.60 without problems
20160402 00:52:36< ancestral> celticminstrel: Ah okay so
20160402 00:52:37< celticminstrel> Though that does require having Boost around.
20160402 00:52:41< ancestral> Link up the targets in Xcode?
20160402 00:52:58< celticminstrel> Yeah, just switch to the target and press Run.
20160402 00:53:17< ancestral> Something to this effect (obviously out of date): http://rdingwall.com/2008/06/01/using-the-boost-unit-test-framework-with-xcode-3/
20160402 00:53:35< celticminstrel> gfgtdf: We can't use custom literals since I and MSVC2013 don't support them.
20160402 00:53:53< celticminstrel> ancestral: It's all set up to build, and it works for me at last check, but I had runtime errors.
20160402 00:54:00< ancestral> Okay
20160402 00:54:07< ancestral> So you just want me to pull and build
20160402 00:54:09< ancestral> And test
20160402 00:54:17< celticminstrel> I was wondering if you might have any insight into it. Or, if you don't get them, maybe it's just a problem with my libs somehow.
20160402 00:54:18< celticminstrel> Yeah.
20160402 00:54:23 * ancestral starts up Xcode
20160402 00:54:26< ancestral> 7.3
20160402 00:55:18< celticminstrel> gfgtdf: Also it's not really any better than the current way, and in fact slightly worse since you'd need a custom version of xgettext.
20160402 00:55:25< celticminstrel> Speaking of which... what about iceiceice's po parser?
20160402 00:56:39-!- aeonchild [enchilado@defocus/yummy/enchilado] has quit [Ping timeout: 276 seconds]
20160402 00:57:00< gfgtdf> don't know too muhc of its internals, currently we are using boost::locale to read out gettext files
20160402 00:58:31< ancestral> celticminstrel: Building now
20160402 00:58:37< celticminstrel> gfgtdf: xgettext is used to generate the pofiles; it's an external parser that searches the C++ files for _("...") strings.
20160402 00:58:44< gfgtdf> ancestral: boost 1.60 ?
20160402 00:58:57< celticminstrel> Wesnoth unit tests.
20160402 00:58:58< ancestral> The boost unit tests
20160402 00:59:09< ancestral> gfgtdf: I did try 1.60 a few weeks ago
20160402 00:59:17< ancestral> Got soe nasty errors
20160402 00:59:18< ancestral> *some
20160402 00:59:20< gfgtdf> celticminstrel: so iceiceice's parser replaces xgettext ?
20160402 00:59:31< aeth> ancestral: https://github.com/Aethaeryn/wesnoth-umc/blob/22478f270f4a509932f0def3cdf6fd750da99273/Aethaeryns_Mod/lua/menu.lua
20160402 00:59:35< aeth> ancestral: high-level menu code in Lua
20160402 00:59:45< ancestral> Slick
20160402 00:59:53< aeth> because the Lua API is frustratingly frustrating to use directly
20160402 01:00:07< aeth> This is just a series of tables and higher order functions (sorta callbacks, but fancier because Lua)
20160402 01:00:39< ancestral> aeth: Can you do me a favor?
20160402 01:00:55< ancestral> Can you build from ancestral-lato and tell me how things look for you?
20160402 01:01:19< aeth> The actual Lua wrapper is https://github.com/Aethaeryn/wesnoth-umc/blob/22478f270f4a509932f0def3cdf6fd750da99273/Aethaeryns_Mod/lua/gui2.lua
20160402 01:01:40< ancestral> Wow, you’re becoming quite the lua coder
20160402 01:02:11< aeth> Which is funny because in another channel a few hours ago a Lua fan was accusing me of not knowing how to program in Lua because I pointed out the language's flaws in a discussion.
20160402 01:02:39< celticminstrel> gfgtdf: No, iceiceice's parser doesn't replace anything.
20160402 01:03:02< celticminstrel> xgettext generates the po files, which are then compiled into a binary mo file.
20160402 01:03:18< celticminstrel> This is fine for Wesnoth's own strings.
20160402 01:03:24< celticminstrel> However, mo files are not portable.
20160402 01:03:37< aeth> ancestral: Afaik, I'm writing some of the most elaborate Lua code for Wesnoth.
20160402 01:03:45< aeth> ancestral: where is ancestral-lato?
20160402 01:03:47-!- aeonchild [enchilado@gateway/shell/blinkenshell.org/x-aupxozcvqdtgxhkc] has joined #wesnoth-dev
20160402 01:03:49< ancestral> It’s a branch
20160402 01:03:53< celticminstrel> Thus, iceiceice's parser would enable custom campaigns to be translated more easily, because they could ship po files instead of mo files for their translations.
20160402 01:03:54< gfgtdf> celticminstrel: you mean you need different mo files for each cpu architeture ?
20160402 01:04:00< celticminstrel> Something like that.
20160402 01:04:01< ancestral> You should be able to switch to it from the official repo
20160402 01:04:09< aeth> well first I have to pull
20160402 01:04:21< ancestral> It’s fine, it’s not urgent
20160402 01:04:30< aeth> new branch! :o
20160402 01:04:55< aeth> ancestral: so just run scons like usual or is there something special I need to do?
20160402 01:04:56< celticminstrel> Maybe we could import aeth's GUI API into the repo... (with his permission obviously)
20160402 01:05:16< celticminstrel> Or redo it in C++ if that seems desirable.
20160402 01:05:31< ancestral> font-variant-ligatures
20160402 01:05:33< ancestral> Oops
20160402 01:06:02< ancestral> aeth: If you wouldn’t mind taking a screenshot of the main menu, the advanced preferences, and maybe a screen in the game
20160402 01:06:02< vultraz> celticminstrel, gfgtdf: yes, we should use iceiceice's thing, but we should talk to him first in order to properly deploy it
20160402 01:06:05< ancestral> That would be amazing
20160402 01:06:41< aeth> celticminstrel: I'm not the best at C++, but I probably will be able to do it if it's just a straight translation of my Lua. My wrapper is nowhere near finished, though. Currently, each window only has one value. I suppose it shouldn't be too hard to extend it now that I use callbacks.
20160402 01:06:57< aeth> So e.g. function foo(choice_list, choice_checkbox) instead of function foo(option)
20160402 01:07:05< ancestral> celticminstrel: I got one error
20160402 01:07:16< celticminstrel> Translating an API written in Lua into one written in C++ is not straightforward at all.
20160402 01:07:23< ancestral> Ahh ld: library not found for -lboost_unit_test_frameworkw
20160402 01:07:46< celticminstrel> Translating Lua code to C++ is probably fairly straightforward, if you're only trying to replicate the functionality.
20160402 01:08:22< celticminstrel> But the goal is to make it an API exposed to Lua, so it wouldn't be that straightforward.
20160402 01:08:26< aeth> Nothing I am doing relies on Lua-specific functionality, although quite a bit of it relies on first class functions.
20160402 01:08:33< celticminstrel> ancestral: Yeah, it's assuming that's in your libs.
20160402 01:08:47< aeth> ancestral: it builds
20160402 01:08:51< celticminstrel> aeth: Mainly the reason it's not straightforward is because of the stack-based Lua C API.
20160402 01:08:53-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20160402 01:09:02< celticminstrel> Maybe it's more straightforward than I think, though.
20160402 01:09:35< celticminstrel> You're not just writing the same functionality in C++ - you're using the Lua C API to expose that functionality to the Lua kernel.
20160402 01:10:15< aeth> celticminstrel: It would probably be easiest if I just extend the Lua API where necessary and keep most of the wrapper itself in Lua.
20160402 01:10:34< celticminstrel> I'm sure that would be easier, yes.
20160402 01:10:40< aeth> ancestral: the fonts look very different in your version, is that on purpose?
20160402 01:10:44< ancestral> Yes
20160402 01:10:50< ancestral> That is what we’re trying to test
20160402 01:10:52< vultraz> aeth: new font
20160402 01:10:53< vultraz> Lato
20160402 01:11:26< ancestral> I’d like to see some of the buttons a bit larger myself in the main menu
20160402 01:11:32< ancestral> s/some/all
20160402 01:11:41< vultraz> ancestral: working on it
20160402 01:11:49< ancestral> vultraz: Nice!
20160402 01:12:44< aeth> ancestral: well it significantly changes the look of all the menus
20160402 01:12:47< vultraz> ancestral: https://drive.google.com/file/d/0B-mR9s8FduLLYzFuMjBxd0FxQzg/view?usp=sharing
20160402 01:13:13< celticminstrel> In a good way?
20160402 01:13:53< ancestral> ^ That’s the question
20160402 01:14:03< celticminstrel> Yes I know
20160402 01:14:21< celticminstrel> vultraz: Is that the larger buttons I saw before with border re-added?
20160402 01:14:31< vultraz> celticminstrel: not exactly
20160402 01:14:42< vultraz> it's the design from yesterday, 1 px shorter
20160402 01:14:52< vultraz> I could try the one you saw with a border
20160402 01:15:00< celticminstrel> Not important.
20160402 01:15:16< celticminstrel> Unless you think it's a good idea.
20160402 01:15:32< vultraz> Not sure
20160402 01:16:00< vultraz> I used [image] for the background, so you no longer need to hurry on gradients
20160402 01:16:06< ancestral> vultraz: My advice? Don’t make these buttons any smaller
20160402 01:16:07< vultraz> but [point] would be very useful
20160402 01:16:21< ancestral> The actuall spacing with these, I might shift the text down 1 px, which is no problem
20160402 01:16:28< vultraz> ancestral: what you're looking at is 25hm up 3 px from the current standard of 22.
20160402 01:16:32< vultraz> h,*
20160402 01:16:36< ancestral> Cool
20160402 01:16:54< ancestral> Hell, I’d be curious how it looks at like 30
20160402 01:17:21< ancestral> My theory is: people are going to be clicking these buttons every time they load the game. Therefore, they should be very easy to find and click
20160402 01:17:24< celticminstrel> I would never even consider implementing [point].
20160402 01:17:33< ancestral> s/load the game/open Wesnoth
20160402 01:17:37< celticminstrel> [points], maybe, if there was a pressing need.
20160402 01:18:07< vultraz> ancestral: https://drive.google.com/file/d/0B-mR9s8FduLLRWFqR0ZidmV4dXM/view?usp=sharing
20160402 01:18:16< aeth> uploading it to an imgur album... it might compress the images if they're too big
20160402 01:18:19< vultraz> (keep in mind I'm doing this for all buttons)
20160402 01:18:34< ancestral> Ah
20160402 01:18:34< celticminstrel> That's a bit big in my opinion.
20160402 01:18:36< aeth> interesting bug in imgur, since "Title" is the default text for the title field, if I begin an image title with "title" it deletes it.
20160402 01:18:41< aeth> Makes it frustrating to caption the "title screen"
20160402 01:18:47< celticminstrel> At least for the titlescreen buttons; could be appropriate in some places, maybe.
20160402 01:18:51< ancestral> celticminstrel: Or maybe they just look a bit short width-wise
20160402 01:18:58< celticminstrel> Ehh...
20160402 01:19:04< celticminstrel> Maybe0
20160402 01:19:06< celticminstrel> ^.
20160402 01:19:16< ancestral> vultraz: Just curious, can you make them wider?
20160402 01:19:21< ancestral> For the sake of experimentation
20160402 01:19:21< vultraz> yes
20160402 01:19:31< ancestral> Like an extra 25% or something
20160402 01:19:59< aeth> ancestral: https://imgur.com/a/EP34B
20160402 01:20:19< aeth> The aliasing doesn't look that good in the multiplayer scenario description
20160402 01:20:22< aeth> Otherwise it looks fine
20160402 01:20:45< ancestral> Awesome
20160402 01:20:49< celticminstrel> The MP screen may change drastically, though.
20160402 01:21:04< ancestral> Is that screen GUI1 or GUI2?
20160402 01:21:20< celticminstrel> As far as I know, all the MP select screens are GUI1.
20160402 01:21:30< ancestral> aeth: my branch doesn’t change GUI1 stuff yet
20160402 01:21:43< aeth> ancestral: the font is a bit taller, which could cause bugs where people hard coded some expected behavior based on the font size
20160402 01:21:53< ancestral> So you may be seeing Deja Vu Sans there with the uglier renderer
20160402 01:21:58< aeth> ah, probably
20160402 01:22:01< vultraz> ancestral: https://drive.google.com/file/d/0B-mR9s8FduLLWkxnaW50dDI3NkE/view?usp=sharing
20160402 01:22:09< ancestral> aeth: It’s taller because the size is bigger
20160402 01:22:23< ancestral> aeth: Yes
20160402 01:22:23< vultraz> this is with an 8px boost to both the width and height from the current (so108, 22 to 116, 30)
20160402 01:22:36< ancestral> aeth: That could be an issue with theming in the game. Which, does not use Lato yet
20160402 01:22:43< aeth> ancestral: if someone's using custom GUI2 menus in their scenario, they might have scrollbars where they didn't used to.
20160402 01:22:45< celticminstrel> I guess it's better than the previous one, but still feels a bit big to me.
20160402 01:22:49< ancestral> But I’m willing to update the theme anyway
20160402 01:23:06< celticminstrel> Then again, I suppose bigger is better for mobile...
20160402 01:23:07< ancestral> aeth: Are the menus hard-coded to specific sizes?
20160402 01:23:18< celticminstrel> vultraz: Are you testing that in-game? Does it still fit in 800x600?
20160402 01:23:21< ancestral> Or number of entries? until a scroll bar shows?
20160402 01:23:23 * celticminstrel assumes so, but should be sure.
20160402 01:23:28< aeth> ancestral: no, the GUI2 menus expand afaik.
20160402 01:23:45< aeth> They're roughly centered, and then expand in any direction until they run out of room, and then a scrollbar appears.
20160402 01:23:48< celticminstrel> ancestral: Yeah, GUI2 dialogs generally auto-expand.
20160402 01:23:51< vultraz> celticminstrel: in-game, yes
20160402 01:24:10< vultraz> celticminstrel: and it does fit
20160402 01:24:27< ancestral> celticminstrel: Okay, I have the unit_test framework I built from boost 1.60 apparently
20160402 01:24:31< ancestral> Wonder if that will work
20160402 01:24:40< vultraz> ancestral: thoughts?
20160402 01:25:24< celticminstrel> ancestral: It should, as long as your other Boost libs are also 1.60...
20160402 01:26:32< ancestral> vultraz: I like the bigger buttons
20160402 01:26:38< ancestral> However, if they’re too big for other places
20160402 01:26:46< ancestral> If it’s not too much trouble
20160402 01:26:57< ancestral> Could we have bigger buttons only for the main menu?
20160402 01:27:06< ancestral> That being said, I don’t want to make the final call
20160402 01:27:07< vultraz> I can do that in 30 seconds
20160402 01:27:39< vultraz> I still need to tweak the design
20160402 01:27:48< ancestral> vultraz: If you don’t mind, at some point, take a full screenshot of the main menu?
20160402 01:27:49< vultraz> since zookeeper says it's not good enough
20160402 01:28:11< ancestral> I would certainly defer to him
20160402 01:28:13-!- gfgtdf_ [~chatzilla@x4e36a6e9.dyn.telefonica.de] has joined #wesnoth-dev
20160402 01:29:59< celticminstrel> I also need to implement rounded rectangles for these buttons.
20160402 01:30:20< vultraz> ancestral: https://drive.google.com/file/d/0B-mR9s8FduLLclJDTklCY1FlYXM/view?usp=sharing
20160402 01:30:27-!- gfgtdf [~chatzilla@x50ab6f7b.dyn.telefonica.de] has quit [Ping timeout: 276 seconds]
20160402 01:30:38-!- gfgtdf_ is now known as gfgtdf
20160402 01:31:04< ancestral> vultraz: I think it’s good
20160402 01:31:52< celticminstrel> Looks okay, apart from the obvious border issue.
20160402 01:32:02< vultraz> The great thing about having these done procedurally is that we can have buttons of any size
20160402 01:32:08< vultraz> without worrying about bad scaling
20160402 01:32:33< ancestral> Excellent
20160402 01:32:44< ancestral> What’s happened over the span of 10 years is people have gotten larger monitors, and the buttons are taking up a smaller and smaller amount of the screen, meaning it’s becoming harder to click on stuff
20160402 01:33:19-!- aeonchild [enchilado@gateway/shell/blinkenshell.org/x-aupxozcvqdtgxhkc] has quit [Ping timeout: 260 seconds]
20160402 01:34:31< ancestral> aeth: Thanks, that’s all I needed
20160402 01:34:42< ancestral> I appreciate your time
20160402 01:34:52< celticminstrel> I wonder if WFL is fast enough to be used for per-pixel image modifications...
20160402 01:36:50< vultraz> you'd need to test
20160402 01:37:13< celticminstrel> Well, I did add the debug_profile() function.
20160402 01:38:46< celticminstrel> 163 left.
20160402 01:40:12-!- NosajIRL [~nos@208.91.185.104] has quit [Ping timeout: 276 seconds]
20160402 01:45:23< celticminstrel> So ancestral, how about the unit tests?
20160402 01:45:35< ancestral> Failed
20160402 01:45:43< ancestral> Just a second
20160402 01:45:50-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160402 01:46:20< celticminstrel> Failed as in test failure, or failed as in crashing before they start?
20160402 01:47:26< ancestral> Test failure
20160402 01:47:37< celticminstrel> Okay, so there must be something wrong with my libs...
20160402 01:47:40< ancestral> Okay so probably the unit_test_framework from 1.60 is a bad idea
20160402 01:47:46< ancestral> celticminstrel: Maybe not
20160402 01:47:48< celticminstrel> Eh?
20160402 01:47:51-!- aeonchild [enchilado@defocus/yummy/enchilado] has joined #wesnoth-dev
20160402 01:47:54< ancestral> Where did you get your unit_test_framework lib?
20160402 01:48:03< ancestral> That is missing from the compile stuff
20160402 01:48:23< celticminstrel> The compile stuff no longer works, so I copied by 1.57 libs from /usr/local.
20160402 01:48:26< celticminstrel> ^my
20160402 01:48:32< ancestral> 1.57 from usr local okay
20160402 01:48:44< celticminstrel> I think those are the ones I compiled myself.
20160402 01:48:51< ancestral> Boost 155
20160402 01:48:54< ancestral> I could try that
20160402 01:49:45< ancestral> celticminstrel: Otherwise, if you built yours as a universal build, I should be able to use it
20160402 01:50:01< celticminstrel> I have no idea if it's a universal build.
20160402 01:50:10< ancestral> And it might work even if it’s not
20160402 01:50:12< celticminstrel> I think I can find out somehow...
20160402 01:50:22< ancestral> It matters little
20160402 01:50:38< celticminstrel> "Mach-O 64-bit dynamically linked shared library x86_64"
20160402 01:50:39< ancestral> Since technically universal would mean PPC… except some libs throw in 32-bit/64-bit into that flag too
20160402 01:50:43< celticminstrel> (output of file)
20160402 01:50:48< ancestral> Okay
20160402 01:50:49< celticminstrel> (The "file" command I mean)
20160402 01:50:54< ancestral> So there’s a chance it might be 64-bit only
20160402 01:51:07< celticminstrel> Probably is, yeah.
20160402 01:51:25< ancestral> Well, I can build 1.55 pretty quickly with Homebrew, I guess we’ll see if that will work
20160402 01:51:34< ancestral> Otherwise I can try to manually find 1.57
20160402 01:51:49< celticminstrel> Is XCode still set up for building 32-bit?
20160402 01:52:00< celticminstrel> I changed some settings with the C++11 stuff.
20160402 01:52:52< ancestral> celticminstrel: Yes
20160402 01:53:05< ancestral> You can still choose to build as a 32-bit/64-bit standard binary
20160402 01:56:52-!- ruxpendp [8cb4f58b@gateway/web/freenode/ip.140.180.245.139] has joined #wesnoth-dev
20160402 01:57:34-!- ruxpendp [8cb4f58b@gateway/web/freenode/ip.140.180.245.139] has quit [Client Quit]
20160402 01:59:13< vultraz> celticminstrel: https://drive.google.com/file/d/0B-mR9s8FduLLUm1zak0zMi1lcHc/view?usp=sharing
20160402 02:01:37< celticminstrel> Does look better than before. I think that's the thing I was thinking you could try but didn't suggest because I thought it would be pointless given that I plan to implement proper rounded rectangles.
20160402 02:04:39< vultraz> might be a little too rounded..
20160402 02:04:48< celticminstrel> No, definitely not.
20160402 02:05:02< celticminstrel> I actually think it could still be a little more rounded.
20160402 02:05:05< vultraz> O_O
20160402 02:05:56< celticminstrel> Something like this (example for the bottom left):
20160402 02:06:05< celticminstrel> X___
20160402 02:06:11< celticminstrel> XX__
20160402 02:06:17< celticminstrel> _X__
20160402 02:06:22< celticminstrel> _XX_
20160402 02:06:24< celticminstrel> __XX
20160402 02:07:02< celticminstrel> Oh, I guess ignore the middle line. Whoops.
20160402 02:07:50< celticminstrel> But, you really don't want to do that by drawing a bunch of points... or even lines...
20160402 02:08:31< vultraz> only benefit of this method is it makes sure that the background image is drawn over
20160402 02:17:15-!- aeonchild [enchilado@defocus/yummy/enchilado] has quit [Ping timeout: 276 seconds]
20160402 02:22:45-!- aeonchild [enchilado@defocus/yummy/enchilado] has joined #wesnoth-dev
20160402 02:37:19-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20160402 02:48:14< celticminstrel> Hmm... where would I look for the code that uploads addons...
20160402 02:49:55< vultraz> addons/manager
20160402 02:49:58< vultraz> I think
20160402 02:51:29< vultraz> how's the for stuff?
20160402 02:51:44-!- Waste [~Cracker@blk-138-75-15.eastlink.ca] has joined #wesnoth-dev
20160402 02:51:48< celticminstrel> Almost done. I'm at U.
20160402 02:52:17< vultraz> U?
20160402 02:52:23< vultraz> utils?
20160402 02:52:45< celticminstrel> No, XCode's search feature lists files in alphabetical order by their name, not their path.
20160402 02:53:11< vultraz> ah
20160402 02:54:52< celticminstrel> Now V.
20160402 02:55:24-!- gfgtdf [~chatzilla@x4e36a6e9.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 45.0.1/20160315153207]]
20160402 02:56:30< celticminstrel> It's bad enough that people have said "write 'if ()' instead of 'if()'", but it's even more annoying to then find that not only does half the code utterly ignore this, it even mixes the two styles in the same vicinity.
20160402 02:58:58< celticminstrel> Only two files left now!
20160402 03:00:57< celticminstrel> Now I just have to build, fix any compile errors, and commit.
20160402 03:01:36-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Quit: wedge009]
20160402 03:02:54< celticminstrel> That could still take awhile though.
20160402 03:04:21< celticminstrel> I wonder if iterator_pair.hpp will enable for(auto& v : {c.begin(), c.end()})
20160402 03:04:39< celticminstrel> Will it be able to infer that that's an std::pair? Or will it get confused and give up?
20160402 03:05:19< celticminstrel> (Mind you, I'm not sure why you'd ever want to do that when you can just do for(auto& v : c).)
20160402 03:06:31-!- irker659 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160402 03:10:11< celticminstrel> I know it's pointless and a case of "fixing what's not broken" and such, but I still kinda want to try to implement the FormulaAI engine in Lua.
20160402 03:11:02-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160402 03:12:41< celticminstrel> (And by extension the idea of enabling the creation of entire new engines using Lua strikes me as kinda cool, if maybe a little pointless.)
20160402 03:13:30< celticminstrel> (Note that "engine" in this context means primarily "a parser for AI configurations".)
20160402 03:20:14< aeth> celticminstrel: go for it
20160402 03:20:21< aeth> The more that's in Lua the better imo
20160402 03:20:26< celticminstrel> XD
20160402 03:20:41< celticminstrel> It would require more work than just Lua.
20160402 03:21:20< celticminstrel> Currently, you can write [aspect], [facet], [stage], [candidate_action], and [goal] in Lua, but you cannot write [engine] in Lua, which is what that would require.
20160402 03:21:41< celticminstrel> An engine's responsibility is to convert WML snippets into callable AI components.
20160402 03:22:36< celticminstrel> The other side, allowing Lua to compile and call formulas, I've sort of already finished, though I want to redo some parts of it.
20160402 03:37:42-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160402 03:41:01-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has quit [Quit: End Transmission.]
20160402 03:42:13-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 252 seconds]
20160402 03:50:54< celticminstrel> Only 23 errors! And I think most of them are from the other thing.
20160402 03:57:06< celticminstrel> Also, BOOST_REVERSE_FOREACH, which kinda complicates things a bit...
20160402 04:00:56< celticminstrel> Okay, gonna use the Boost reverse adaptor for that.
20160402 04:01:13< celticminstrel> Part of Boost.Range apparently.
20160402 04:02:32< aeth> Since boost was mentioned, I'd make some remark about C++, but every language has one of those "should have been in the standard library" libraries.
20160402 04:02:39< aeth> So this is a case of all languages being bad, not C++ being bad :P
20160402 04:04:23-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20160402 04:09:52< celticminstrel> There were only four cases of BOOST_REVERSE_FOREACH though.
20160402 04:12:40< celticminstrel> There are things in Boost that shouldn't be in the standard library as well, though.
20160402 04:13:06< celticminstrel> Boost.Spirit is pretty specialized for example.
20160402 04:13:25< aeth> Well, yeah. That's one of the problems with a "should have been in the standard library" libraries. They usually have scope creep and are more than just their main use.
20160402 04:14:16< aeth> Or alternatively, you could take a Python development approach and say "Why not include all the things in the standard library?" :P
20160402 04:14:30< celticminstrel> Java went that way too.
20160402 04:14:35< aeth> Some languages have web servers.
20160402 04:14:39< aeth> Java might be one of those
20160402 04:14:46< celticminstrel> I think both Java and Python do.
20160402 04:14:58< celticminstrel> Though, I'm not sure if it counts as part of the standard library in Java.
20160402 04:15:12< celticminstrel> I think it's in net.sun or something, not java.* or javax.*.
20160402 04:17:29< aeth> It's been years since I did Java programming, but iirc, one of the best parts about Java is their collections library.
20160402 04:17:52< aeth> It's frustrating when something basic isn't part of the core language, and Java probably includes too much, but at least they include a lot of basic things that many languages miss.
20160402 04:18:27< celticminstrel> Java collections are fairly good, yeah, but I one thing I feel they did wrong is making all iterators "fail-fast".
20160402 04:18:31< aeth> e.g. The core Scheme language has no hash tables (even in r7rs, the latest), so each implementation has to basically invent their own.
20160402 04:18:41< celticminstrel> Which basically renders one of the points of linked lists inoperable.
20160402 04:18:59< aeth> A lot of languages go the opposite direction of Scheme and make *everything* roughly like a hash table, which is also frustrating.
20160402 04:19:26< aeth> At the very least, I want linked lists, arrays, and hashes.
20160402 04:19:28< celticminstrel> (The fact that you can always save a reference to any element and it will remain valid as long as that element is not removed.)
20160402 04:19:42< celticminstrel> In Java, you can't save iterators pointing to elements of a linked list.
20160402 04:20:19-!- NosajIRL [~nos@208.91.185.104] has joined #wesnoth-dev
20160402 04:20:23< celticminstrel> You could save the index, but that's inefficient, so you might as well just use an ArrayList instead if you need that, even though the LinkedList is more efficient for arbitrary insertions and removals.
20160402 04:20:56< celticminstrel> (Fail-fast is fine for some containers, I just think it shouldn't be applied to all containers.)
20160402 04:21:09< celticminstrel> (I don't know of any besides linked lists where this is relevant.)
20160402 04:32:11< aeth> Oh, I forgot. Bit arrays are pretty important too. A lot of languages overlook this. e.g. Try writing a prime sieve without one.
20160402 04:32:47< celticminstrel> You can just use an array of boolean. Less space-efficient, admittedly, but still works.
20160402 04:33:24< aeth> Well, yeah, that's why I said prime sieve. Using a bit array you're going to be a lot more memory efficient unless the language does something like convert the boolean array into a bit array or something.
20160402 04:33:38< aeth> I'm sure there's a few other things where you can easily hit a memory wall without a bit array.
20160402 04:34:06< aeth> https://en.wikipedia.org/wiki/Succinct_data_structure
20160402 04:37:21-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160402 04:55:58-!- ScegfOd [637f4b7c@gateway/web/freenode/ip.99.127.75.124] has quit [Ping timeout: 250 seconds]
20160402 05:34:08< vultraz> celticminstrel: I used to think if () was the correct style, then I realized it was actually if()
20160402 05:34:26< celticminstrel> "correct style"
20160402 05:35:07< vultraz> hm?
20160402 05:35:10< celticminstrel> There is no correct style. There are good styles, bad styles, and okay styles, but no correct style.
20160402 05:35:32< vultraz> "wesnoth's style"
20160402 05:35:34< vultraz> I should say
20160402 05:36:10< celticminstrel> I'm pretty sure someone said it was "if ()". Though I'd be happy if that were to change.
20160402 05:36:25< vultraz> in all recent code I've been using if()
20160402 05:36:35< celticminstrel> (Since the person who said it considers himself retired anyway.)
20160402 05:36:45< celticminstrel> (If I recall correctly.)
20160402 05:36:45< vultraz> my older code uses if ()
20160402 05:36:48-!- vincent_c [~bip@vcheng.org] has quit [Ping timeout: 276 seconds]
20160402 05:36:55< celticminstrel> Okay, I'll keep that in mind.
20160402 05:37:06< celticminstrel> I generally use if() anyway unless I really pay attention to style.
20160402 05:40:54< celticminstrel> I wonder if the Boost reverse range adaptor will use the rbegin()/rend() functions I added to iterable_pair.hpp.
20160402 05:41:04< celticminstrel> It looks like these functions are a C++14 thing.
20160402 05:47:59< celticminstrel> vultraz: Any objections to pulling in part of my formula stuff right after this BOOST_FOREACH removal?
20160402 05:48:10< vultraz> no
20160402 05:48:21< vultraz> I haven't started the function replacement yet
20160402 05:48:56< celticminstrel> I also have one or two little things sitting around, like a Lua function to set the current time.
20160402 05:49:06< vultraz> why in the world would you want that
20160402 05:49:10< vultraz> unless you mean ToD
20160402 05:49:13< vultraz> not clock time
20160402 05:49:14< celticminstrel> Oh yeah, I can delete a few of these branches, can't I...
20160402 05:49:17< celticminstrel> Yeah, ToD.
20160402 05:49:20< vultraz> ah
20160402 05:49:21< vultraz> good, good
20160402 05:49:25< aeth> If we're going to bikeshed, "if ()" definitely looks more like the correct style to me, at least. It's a keyword, not a function call.
20160402 05:49:38 * vultraz groans
20160402 05:49:42< aeth> So that means Wesnoth should standardize on if() because it tends to use the style I don't :P
20160402 05:49:43-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160402 05:50:20< celticminstrel> BTW aeth, do you think rewriting FormulaAI in Lua is better than the other discussed option? (There's been talk of removing it altogether.)
20160402 05:51:13< celticminstrel> (The argument is that anything you can do in FormulaAI can also be done in Lua, so there's not much point having two ways of doing it.)
20160402 05:51:45< celticminstrel> (Though currently some things are more convenient in FormulaAI, like giving specific units unique behaviour rules.)
20160402 05:51:58< aeth> celticminstrel: If everything you can do in FormulaAI you can do in Lua, then why not write it in Lua?
20160402 05:52:07< aeth> celticminstrel: It'll wind up making the Lua more convenient to use.
20160402 05:52:23< celticminstrel> I'm not sure how it would make the Lua more convenient to use though...
20160402 05:52:35< celticminstrel> Or did I misunderstand what you said...
20160402 05:53:06< aeth> celticminstrel: If it's anything like the GUI stuff I've been doing, you'll find a bunch of edge cases of things that should be in the Lua AI, and would be trivial to put in the Lua API, but aren't exposed to the Lua API.
20160402 05:53:15< celticminstrel> My XCode has this weird syntax highlighting bug whereby it can't detect the end of a comment of the form /* TODO: */
20160402 05:54:01< celticminstrel> aeth: If we're talking about API, I think the Lua AI API is more featured at the moment, though it's possible that FormulaAI does have one or two functions that Lua lacks,
20160402 05:54:03-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20160402 05:54:11< celticminstrel> I haven't looked through the FormulaAI functions in detail.
20160402 05:54:15< aeth> Yes.
20160402 05:54:47< aeth> One way to be sure (though perhaps not the easiest way) would be to implement Formula AI in Lua.
20160402 05:54:57< celticminstrel> I don't suppose you have any opinion specifically on whether or not FormulaAI should be removed.
20160402 05:55:09 * celticminstrel wouldn't expect you too if you're not doing AI work, though.
20160402 05:55:20< aeth> I haven't done anything related to the Wesnoth AI.
20160402 05:55:28-!- NosajIRL [~nos@208.91.185.104] has quit [Ping timeout: 244 seconds]
20160402 05:55:38< celticminstrel> I did actually discover some things missing from Lua by looking at WFL.
20160402 05:56:02-!- vincent_c [~bip@vcheng.org] has joined #wesnoth-dev
20160402 05:56:12< celticminstrel> Like details of units not exposed directly (though at least Lua does expose the full config, so you could still get at that info even if it was a bit less convenient).
20160402 05:56:30< celticminstrel> I wonder if there'd be any reason to expose the unit animations to Lua.
20160402 05:57:30< aeth> The Lua API really seems like people just add to it as they need it. So there probably hasn't been any need until now, sort of like there wasn't any need until recently for the terrain images to be exposed in Lua.
20160402 06:00:36< aeth> For me, my add-on is basically racing against the feature freeze because I'm sure there's at least a dozen things that I can't do that I need to be able to do, but I'm not sure what until I get there. Known unknowns.
20160402 06:04:24-!- mjs-de [~mjs-de@p508C8FF9.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160402 06:05:59< vultraz> https://drive.google.com/file/d/0B-mR9s8FduLLcXZBVzVUcGlqS1k/view?usp=sharing decided to use the old button bg texture for the new buttons
20160402 06:06:21< celticminstrel> I feel like ca.?pp might be better split up into one file per candidate action (plus one for the base class possibly).
20160402 06:07:40< celticminstrel> I also feel like I'm trying to do fifty things at once and am forgetting about forty of them.
20160402 06:08:15< celticminstrel> The buttons don't look bad.
20160402 06:08:28< celticminstrel> Is that still just one-pixel rounding?
20160402 06:09:06< celticminstrel> BTW, am I wrong in thinking that Lua doesn't give direct access to the game map?
20160402 06:09:49< celticminstrel> Oh, it does have wesnoth.get_terrain(), I guess.
20160402 06:10:04< celticminstrel> So you can do it, it's just not super-convenient.
20160402 06:10:22-!- Kwandulin [~Miranda@p200300760F191C871CA8C268173E01A8.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160402 06:10:34< celticminstrel> Does anyone else take issue with the three "helper.set_xxx_metatable" functions?
20160402 06:13:50< aeth> celticminstrel: I think you'd have to currently combine get_map_size with get_terrain, unless the wiki is misleading me
20160402 06:15:00< celticminstrel> Yeah, that makes sense, make sure you don't access out of bounds.
20160402 06:15:02< aeth> I personally don't see why there couldn't be a get_map and even a set_map if you want to get the map, modify a lot of it, and then set it all at once.
20160402 06:15:29< aeth> You could probably create some interesting generated things that change every turn that way.
20160402 06:15:45< celticminstrel> It's not like that's currently impossible, but...
20160402 06:15:56< celticminstrel> Speaking of generated, what about Lua map generators?
20160402 06:16:07< celticminstrel> They do exist, apparently.
20160402 06:16:22< celticminstrel> But I wonder if there's any way to call one from a Lua script...
20160402 06:16:26< vultraz> celticminstrel: yes, 1 px
20160402 06:16:28< vultraz> it's good enough
20160402 06:16:35< celticminstrel> vultraz: I disagree.
20160402 06:16:40 * vultraz groans
20160402 06:17:09-!- Waste [~Cracker@blk-138-75-15.eastlink.ca] has quit [Quit: Leaving]
20160402 06:17:23< celticminstrel> I'm not expecting you to make it any better without an actual round rectangle shape, though.
20160402 06:17:37< celticminstrel> I mean, you can if you want, but I'm not expecting it.
20160402 06:20:28< vultraz> I can't use a rounded rectangle because a draw an image over the background!
20160402 06:20:30< vultraz> not a rectangle
20160402 06:20:32< vultraz> !
20160402 06:21:33< aeth> celticminstrel: What would a hypothetical get_map() give you? A string or a table?
20160402 06:22:00< celticminstrel> aeth: I was thinking of a table whose metatable called get_terrain and get_terrain_info.
20160402 06:22:07< vultraz> unless you can give me a way to take a rounded rectangle and cut off anything outside it
20160402 06:22:21< celticminstrel> vultraz: Clipping...
20160402 06:22:56< aeth> celticminstrel: How are maps represented internally in the C++? I have never looked at that part of the code.
20160402 06:23:28< celticminstrel> aeth: I haven't looked too closely either, but I was always assuming they were some kind of 2D array...
20160402 06:25:01< vultraz> celticminstrel: I don't see how you want it more rounded when this is what they originals did
20160402 06:25:29< celticminstrel> vultraz: I'll compare them side-by-side again...
20160402 06:28:13< aeth> celticminstrel: Perhaps consider the hypothetical problem of rewriting parts of the map editor in Lua (I am assuming the map editor doesn't use Lua anywhere) and you'll get some ideas as to where the terrain Lua is lacking, or at least lacking efficient or convenient ways of doing it even if there are ways.
20160402 06:28:56< aeth> I'm not saying rewrite the map editor in Lua, I'm just saying that if the map editor *could* be written in Lua, then there would be a pretty much complete map and terrain API.
20160402 06:30:12-!- mjs-de [~mjs-de@p508C8FF9.dip0.t-ipconnect.de] has quit [Ping timeout: 250 seconds]
20160402 06:30:43-!- NosajIRL [~nos@208.91.185.104] has joined #wesnoth-dev
20160402 06:30:59-!- mjs-de [~mjs-de@p508C8FF9.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160402 06:40:10-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 250 seconds]
20160402 06:41:14-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160402 07:43:22-!- aeonchild [enchilado@defocus/yummy/enchilado] has quit [Quit: Reconnecting]
20160402 07:43:31-!- aeonchild [enchilado@defocus/yummy/enchilado] has joined #wesnoth-dev
20160402 07:46:58-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20160402 07:53:51< zookeeper> vultraz, you should still trim the bright blue lines by 1px both
20160402 08:05:39< Ivanovic> celticminstrel: was the cmake build also adjusted for cxx11 support?
20160402 08:05:52< Ivanovic> i just tried to build master after ages and it fails badly
20160402 08:06:05< celticminstrel> No, CMake has not been adjusted.
20160402 08:06:30< Ivanovic> do you have the commit at hand that fixed scons?
20160402 08:06:38< Ivanovic> then i can have a look at fixing the cmake recipe
20160402 08:07:35< celticminstrel> I think you want commit 0795ad879bf120684e2afc763fd9976e9ce34c17
20160402 08:07:55< celticminstrel> The following one (af7286c4214fc6770de2b1dcdb8bf99a1016ac94) may also be relevant.
20160402 08:11:03< celticminstrel> Okay vultraz, I guess you're right. The old buttons are no rounder than the current ones. However, I do think the lighter right/top border looks better.
20160402 08:11:15< celticminstrel> Or is it lighter all around... not sure.
20160402 08:17:10-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160402 08:22:18-!- Kwandulin [~Miranda@p200300760F191C871CA8C268173E01A8.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160402 08:24:43-!- irker882 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160402 08:24:43< irker882> wesnoth: Nils Kneuper wesnoth:master f758706ea2a4 / CMakeLists.txt: fixed compiling with CMake in a hacky way https://github.com/wesnoth/wesnoth/commit/f758706ea2a4306aa67e00662ab6bf5f5173e5c6
20160402 08:24:45< irker882> wesnoth: Nils Kneuper wesnoth:master adb0abfb6dec / src/CMakeLists.txt: add boost-thread to cmake recipe, part2 https://github.com/wesnoth/wesnoth/commit/adb0abfb6dec9449f309149e9e2d662e2ac9873d
20160402 08:27:38-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160402 08:27:53< Ivanovic> celticminstrel: now the cmake build for master should work again
20160402 08:27:58< Ivanovic> (at least on my box it does)
20160402 08:28:04< celticminstrel> Yay, thanks.
20160402 08:32:05-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160402 08:34:04< celticminstrel> Apparently resources::tod_manager is null even though a game has started...
20160402 08:35:25< vultraz> zookeeper: both what?
20160402 08:35:32< celticminstrel> I'm still getting the corrupt cache error, but no crash anymore; maybe it wasn't related in the first place.
20160402 08:35:53< celticminstrel> vultraz: I think he's referring to highlights/shadows just within the button border.
20160402 08:36:06< zookeeper> vultraz, the leftmost pixel of the horizontal bright blue line, and the bottom pixel of the vertical blue line.
20160402 08:36:39-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160402 08:40:04< celticminstrel> ...oh. :/
20160402 08:41:05< vultraz> zookeeper, celticminstrel: https://drive.google.com/file/d/0B-mR9s8FduLLVWw4WVp3YjI2ZnM/view?usp=sharing
20160402 08:41:27< vultraz> good to go?
20160402 08:42:09< vultraz> if it's really necessary I guess I can make the top/right borders lighter..
20160402 08:42:30< celticminstrel> It means not using [rectangle].
20160402 08:42:42< vultraz> Or I can just draw over the top lines
20160402 08:42:48< zookeeper> and/or bottom/left borders darker
20160402 08:42:51< celticminstrel> I suppose.
20160402 08:43:10< vultraz> I dunno...I think there's something to say for the uniform look
20160402 08:43:15< vultraz> all other widgets have uniform borders
20160402 08:43:39< celticminstrel> There is certainly something to say for the uniform look.
20160402 08:43:47< celticminstrel> I don't think it's right for Wesnoth though.
20160402 08:45:50< zookeeper> all other widgets, what? certainly not true for checkboxes, sliders and scrollbars which are the most common ones.
20160402 08:46:47< vultraz> out of those three, only checkboxes apply
20160402 08:48:44< zookeeper> no
20160402 08:49:10< vultraz> well, the slider grips..
20160402 08:49:20< vultraz> but not scrollbars anymore
20160402 08:49:59< zookeeper> that's irrelevant, stop being a dummy
20160402 08:50:25< vultraz> alright, I'll produce a version with darker bottom/left borders
20160402 08:50:37< zookeeper> perfect
20160402 08:51:07< celticminstrel> Might be nice if all widgets had that, but I suppose that's not high priority...
20160402 08:51:55< zookeeper> aside from that, i can't see why you wanted to make them so darn tall, they're massive now. surely all buttons everywhere are not going to be that tall=
20160402 08:51:56< zookeeper> ?
20160402 08:52:12< celticminstrel> I think the taller ones were to be just on the title screen?
20160402 08:52:39< vultraz> yes, these are 116x30 and will be titlescreen only
20160402 08:52:45< celticminstrel> Though others will still be a few pixels taller than currently (to account for slightly larger font size)
20160402 08:52:49< vultraz> but I am going to bump the other buttons slightly
20160402 08:53:00< vultraz> probably to 112x26 or something
20160402 08:54:08< celticminstrel> BTW vultraz, check it on 800x600 before committing anything.
20160402 08:54:26< vultraz> zookeeper, celticminstrel: https://drive.google.com/file/d/0B-mR9s8FduLLRUZZWU9XTEd2QUU/view?usp=sharing
20160402 08:54:52< vultraz> celticminstrel: I still need to do the other three state variations after you guys are satisfied with the normal state :|
20160402 08:55:10< vultraz> So Much WML
20160402 08:55:20< celticminstrel> Well, it does look good... I can't seem to tell the difference in brightness of the borders though...
20160402 08:55:21< vultraz> wish I could do this with WFL
20160402 08:55:49< celticminstrel> How would you do it with WFL?
20160402 08:55:52< zookeeper> the difference is super small
20160402 08:56:05< celticminstrel> Hypothetically speaking.
20160402 08:57:12< vultraz> draw("rectangle" { x1 = 0, y1 = 0, x2 = 0, y2 = 0, border_thickness = 1, color = [GUI_BORDER_COLOR]})
20160402 08:57:16-!- travis-ci [~travis-ci@ec2-23-20-20-92.compute-1.amazonaws.com] has joined #wesnoth-dev
20160402 08:57:17< travis-ci> wesnoth/wesnoth#9172 (master - adb0abf : Nils Kneuper): The build has errored.
20160402 08:57:17< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/120250461
20160402 08:57:18-!- travis-ci [~travis-ci@ec2-23-20-20-92.compute-1.amazonaws.com] has left #wesnoth-dev []
20160402 08:57:32< celticminstrel> Hmm.
20160402 08:57:41< celticminstrel> So maybe something like this?
20160402 08:57:42< vultraz> (or however you can insert macros into the stream)
20160402 08:58:14< celticminstrel> rectangle(0, 0, 0, 0, 1, {GUI_BORDER_COLOR})
20160402 08:58:34< celticminstrel> Hmm, or you could pass the extra args as a map...
20160402 08:59:04< vultraz> zookeeper, celticminstrel https://drive.google.com/file/d/0B-mR9s8FduLLODhSNjhkYUxMUzA/view?usp=sharing
20160402 08:59:09< celticminstrel> rectangle(0, 0, 0, 0, ['border_thickness' -> 1, color -> {GUI_BORDER_COLOR}])
20160402 08:59:12< vultraz> celticminstrel: I was just thinking of how FFL does it
20160402 08:59:18< zookeeper> anyway, i don't even know what kind of "good to go" you're asking for. good to go in a development release, yeah sure, but not good enough that you won't need to make further improvements (whatever those might be).
20160402 08:59:19< zookeeper> and really, with the amount of complaining about it being so hard to do with WML, what exactly makes you think it's easier with WML than by constructing it from images as previously discussed?
20160402 08:59:19< celticminstrel> BTW, what does GUI_BORDER_COLOR expand to?
20160402 08:59:44< celticminstrel> I think the latest looks good.
20160402 08:59:46< vultraz> celticminstrel: "162, 127, 68, 255"
20160402 09:00:01-!- mjs-de [~mjs-de@p508C8FF9.dip0.t-ipconnect.de] has quit [Remote host closed the connection]
20160402 09:00:08< celticminstrel> So a comma-separated list of numbers.
20160402 09:00:15< vultraz> might tone down top/right a bit
20160402 09:00:18< celticminstrel> You could wrap it either in [] or a function call, then.
20160402 09:00:33< celticminstrel> Yeah, the contrast between the borders could actually be a tiny bit less.
20160402 09:01:29< celticminstrel> BTW, WFL identifiers cannot contain digits.
20160402 09:01:37< vultraz> celticminstrel, zookeeper https://drive.google.com/file/d/0B-mR9s8FduLLamdqcmdReGhveHc/view?usp=sharing
20160402 09:01:56< vultraz> zookeeper: good enough that you don't get after me for not being round enough, or something :|
20160402 09:02:05< celticminstrel> Looks good to me.
20160402 09:02:10< zookeeper> the latest one is pretty good
20160402 09:02:17< celticminstrel> Is that gradientish effect still there? It's hard to tell.
20160402 09:02:21< zookeeper> so what does the WML for the whole thing look like?
20160402 09:04:05< vultraz> zookeeper: http://pastebin.com/zEGii3Ra
20160402 09:04:55< zookeeper> huh, okay, not completely ghastly then.
20160402 09:05:37< zookeeper> except that alpha point business O.o
20160402 09:06:36< zookeeper> so can you use those alpha points only in widgets but not in dialogs/windows, or something?
20160402 09:07:01< zookeeper> because it looks like you could do pretty much anything with them, in this context
20160402 09:07:05< vultraz> I hate the way I had to round the widget
20160402 09:07:14< vultraz> But I can do that wherever [draw] is supported
20160402 09:07:20< vultraz> so I can do it in dialogs too
20160402 09:07:24< vultraz> with the [drawing] widget
20160402 09:08:05-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160402 09:08:29< zookeeper> do you have to explicitly draw 0,0,0,0 to a pixel to have it be transparent, or is transparency the default if you don't draw anything there?
20160402 09:09:05< vultraz> hm
20160402 09:09:08< vultraz> passing nothing works
20160402 09:09:42< zookeeper> so you could just avoid drawing anything to the corners, instead of first drawing stuff and then marking the corners as transparent? maybe it'd make it easier, maybe not, just a thought.
20160402 09:09:56< vultraz> I can't
20160402 09:10:11< vultraz> I have to erase the image background
20160402 09:10:15< vultraz> on those pixels
20160402 09:10:35< vultraz> though
20160402 09:10:42< vultraz> I suppose I can just not draw on the borders..
20160402 09:11:07< vultraz> then I'd need 8 [line]s for the border..
20160402 09:12:14< vultraz> which I guess is two less [line]s and a [rectangle]...
20160402 09:12:22< vultraz> fewer
20160402 09:12:53< zookeeper> i'd imagine you could avoid the GUI__POINT's by just drawing the orange lines as 2px-tall/wide rectangles instead
20160402 09:13:55< vultraz> No, it needs to be one pixel
20160402 09:14:11< zookeeper> why?
20160402 09:14:30< vultraz> 2px borders?
20160402 09:14:32< vultraz> ugly
20160402 09:14:38< zookeeper> duh
20160402 09:14:39< zookeeper> no
20160402 09:15:14< zookeeper> i'm talking about drawing exactly the same thing, but in simpler way
20160402 09:16:15< vultraz> well I can imagine how to do it with 4 [lines] and 4 uses of GUI__POINT
20160402 09:16:58< vultraz> which is probably good... the less WML, the better..
20160402 09:17:17< vultraz> if you can propose something simpler, I'd be glad to hear it
20160402 09:17:32< zookeeper> nah
20160402 09:17:35 * zookeeper is afk
20160402 09:52:41-!- Kwandulin [~Miranda@p200300760F191C87E5AF605E28676CEC.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160402 10:29:19-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 244 seconds]
20160402 10:46:19< vultraz> before I do any more refactoring I'm going to ask jet his opinion on rounded vs rectangular.
20160402 10:46:45< celticminstrel> I think I've now enabled WML variable access in $(...)
20160402 10:47:05< celticminstrel> That is, the WML variables are now also accessible as formula variables.
20160402 10:47:05< vultraz> celticminstrel: uh... that's already a thing. $($foo)
20160402 10:47:22< celticminstrel> vultraz: That's semantically different.
20160402 10:47:38< celticminstrel> The idea is that you can now do $(foo).
20160402 10:47:53< vultraz> but using $ makes it clear it's a wml variable
20160402 10:48:00< celticminstrel> With $($foo + 1), if there is no variable foo defined, you'll get a formula syntax error.
20160402 10:48:17< vultraz> what about
20160402 10:48:22< celticminstrel> Wit $(foo + 1), if there is no variable foo defined, you'll get the result of adding null and 1, which is 1.
20160402 10:48:27< celticminstrel> ^With
20160402 10:48:38< vultraz> $(foo + 1 where foo is $foo or 0)
20160402 10:48:43< celticminstrel> I think in most cases, the latter is preferable.
20160402 10:48:46< vultraz> (do we have 'is'?)
20160402 10:49:00< celticminstrel> No.
20160402 10:49:05< vultraz> (that's more like FFL, I think)
20160402 10:49:07< zookeeper> and you can't just "fix" that problem with undefined variables in some other way?
20160402 10:49:43< celticminstrel> I suspect not. The problem arises from variable substitution proceding right-to-left, I think.
20160402 10:50:08< celticminstrel> In other words, by the time it's processing the $(...), it has forgotten that variables were substituted into it.
20160402 10:50:30< celticminstrel> Okay, so you could do some kind of preanalysis of the entire string, I suppose.
20160402 10:50:31< vultraz> celticminstrel: is $(foo + 1 where foo = ($foo != nil, $foo, 0)) valid?
20160402 10:51:00< celticminstrel> vultraz: Close, but I don't believe it'll solve the undefined problem.
20160402 10:51:18< celticminstrel> Variable defaulting does help with the problem, mind you.
20160402 10:51:28< celticminstrel> The $foo?8| syntax.
20160402 10:51:36< vultraz> why the '8
20160402 10:51:55< celticminstrel> If foo is not defined, substitute 8 instead. It could be any string.
20160402 10:52:34< vultraz> that's... kinda ugly syntax, I just realized
20160402 10:52:55< zookeeper> no, it's great
20160402 10:53:00< vultraz> 'where foo is a or b' would be nicer, but that'd only work in formulas, I guess?
20160402 10:53:07< zookeeper> exactly
20160402 10:53:24< celticminstrel> It is, but I don't think there's really any way you could make it not-ugly without completely changing the variable substitution syntax.
20160402 10:53:27< vultraz> (can we even add such a thing)
20160402 10:53:37< vultraz> 'is' and 'or', that is
20160402 10:53:40< celticminstrel> I'm not sure what you're asking.
20160402 10:54:01< vultraz> can we add the keywords 'is' and 'or' to WFL
20160402 10:54:07< celticminstrel> "or" already exists.
20160402 10:54:12< celticminstrel> Why would we add "is"?
20160402 10:54:23< celticminstrel> It can be done, but I have no idea what it would be for.
20160402 10:55:12< vultraz> I'll have to see what it does in FFL..
20160402 10:55:25< vultraz> I guess = is better anyway
20160402 10:55:29< vultraz> where foo = a or b
20160402 10:55:54< celticminstrel> Well, that does mean overloading = with two meanings, but it's not ambiguous, so probably not really an issue.
20160402 10:56:12< celticminstrel> "where foo = a or b" will result in foo being set to a if it's truthy, b otherwise.
20160402 10:56:43< celticminstrel> In WFL, falsy values are null, 0, 0.0, '', [], and the empty map.
20160402 10:56:48< celticminstrel> I think that's all of them.
20160402 10:58:50-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160402 10:59:51< celticminstrel> I'm not actually sure how useful it is to access the WML variables as formula variables, but some other things I've done made it easy, so I thought I might as well... >_> If you prefer, I can skip it.
20160402 10:59:55< celticminstrel> zookeeper: ^
20160402 10:59:57< Ravana_> Instead of $($foo + 1) you currently write $($foo -1 +2). But I like what you do, make sure to include that in the formula page
20160402 11:00:43< celticminstrel> The only downsides are a) You can't easily access variables containing digits and b) Container variables are always arrays, so you need unit[0].name instead of unit.name.
20160402 11:01:15< celticminstrel> b) is probably solvable with some more complex logic.
20160402 11:01:53< celticminstrel> (Something like "if there is only one element, then expose it as an object instead a list of objects".)
20160402 11:02:06< celticminstrel> a) would require changing the formula tokenizer.
20160402 11:02:26< celticminstrel> And possibly break formulas using the dice operator (whether that's important though, I don't know).
20160402 11:03:24-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160402 11:04:22< zookeeper> well it sounds pretty iffy to me to introduce an alternative way of accessing variables if it's severely limited and has all sorts of exceptions
20160402 11:04:42< Ravana_> making dice function instead of operator would get around that
20160402 11:04:57< celticminstrel> That would still break formulas using it. :P
20160402 11:05:22< celticminstrel> Also, even if digits were allowed in formula identifiers, you could still write "3 d 6".
20160402 11:07:20< zookeeper> since we have the default value syntax, couldn't one basically just automatically append |0 if no default is provided?
20160402 11:07:43< zookeeper> i'm not surprised at all if not, but just a thought
20160402 11:08:19< zookeeper> err, i mean ?0
20160402 11:08:47< celticminstrel> zookeeper: It's not severely limited; you can access variables containing digits through an alternate (though ugly) syntax. It has some (possibly minor) advantages over the other way when used within a more complex formula, since it doesn't have the potential for breaking formulas if the variable is undefined. (Though, unlike the current method, it wouldn't support substituting actual formula snippets from variables.)
20160402 11:09:18< celticminstrel> It would certainly be possible to make the default variable substitution 0 instead of an empty string, but then you'd probably get unexpected 0's showing up in your strings.
20160402 11:09:42< celticminstrel> The variable substitution algorithm (currently) has no way of knowing whether the variable currently being substituted is within a formula substitution.
20160402 11:11:16< celticminstrel> Hmm, apparently scoped variables are not stored in the same place as the others...
20160402 11:11:36< zookeeper> right
20160402 11:12:04< celticminstrel> Unless of course the are scoped variables defined from Lua.
20160402 11:12:07< celticminstrel> ^they
20160402 11:12:13 * zookeeper is afk again
20160402 11:24:47-!- irker882 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160402 11:27:09-!- aeonchild [enchilado@defocus/yummy/enchilado] has quit [Ping timeout: 276 seconds]
20160402 11:34:44-!- aeonchild [enchilado@defocus/yummy/enchilado] has joined #wesnoth-dev
20160402 11:58:34-!- gfgtdf [~chatzilla@x4e36a6e9.dyn.telefonica.de] has joined #wesnoth-dev
20160402 11:58:39-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160402 11:59:27< gfgtdf> celticminstrel, aeth: there are [replace_map] ( afaik and also a store map) wml tags that can be used form lua like any otzher wml tag
20160402 12:00:22< gfgtdf> celticminstrel, aeth: to use a lua map generator you don't need any special function, just use dofile+call+replace_map.
20160402 12:00:26-!- nos_ [~nos@208.91.185.104] has joined #wesnoth-dev
20160402 12:01:05< celticminstrel> Are you talking about the mapgen kernel or whatever?
20160402 12:02:50< gfgtdf> celticminstrel: i answered 20160402 06:16:22< celticminstrel> But I wonder if there's any way to call one from a Lua script... and 20160402 06:15:02< aeth> I personally don't see why there couldn't be a get_map and even a set_map if you want to get the map, modify a lot of it, and then set it all at once.
20160402 12:03:48< celticminstrel> If I recall correctly, [store_map] isn't terribly useful for large-scale map modification - doesn't it store it as a string? I could be wrong though.
20160402 12:05:02< gfgtdf> celticminstrel: yes it does, and replace_map expectes a string.
20160402 12:05:29< celticminstrel> So yeah, not very useful.
20160402 12:05:47< celticminstrel> Useful for loading maps from files, though.
20160402 12:05:54< gfgtdf> celticminstrel: well it not hard to parse such a string
20160402 12:06:02< celticminstrel> Or storing the map, changing it to something else, and then restoring it.
20160402 12:06:16< celticminstrel> True, but why should you have to parse a string?
20160402 12:06:33< gfgtdf> celticminstrel: becasue store_map retuns a string
20160402 12:06:43< celticminstrel> That's not a reason.
20160402 12:07:28< celticminstrel> I do want to do this thing about making WML variables accessible in $(...), but it turns out that the way I attempted it didn't work, so I guess I'll try again later.
20160402 12:08:50< gfgtdf> celticminstrel: i thought you are currently wokring on the boost_foreach port ?
20160402 12:09:13< celticminstrel> I already finished that.
20160402 12:09:45< gfgtdf> celticminstrel: hmm but its not pushed
20160402 12:09:50< celticminstrel> Right.
20160402 12:12:05-!- NosajIRL [~nos@208.91.185.104] has quit [Quit: Leaving]
20160402 12:19:53-!- Kwandulin [~Miranda@p200300760F191C87E5AF605E28676CEC.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160402 12:41:15-!- nos_ [~nos@208.91.185.104] has quit [Ping timeout: 276 seconds]
20160402 12:42:53-!- gfgtdf [~chatzilla@x4e36a6e9.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 45.0.1/20160315153207]]
20160402 12:54:55-!- aeonchild is now known as dlihcnoea
20160402 13:24:11< celticminstrel> I wonder if the loading screen would look better with the bar above the text...
20160402 13:24:22< celticminstrel> Ever since adding the animation, I've felt the bar was too low.
20160402 13:24:32< celticminstrel> ^Ever since the animation was added
20160402 13:36:20< celticminstrel> o.o Trying to assign a string to a unit variable (in Lua) seems to produce '': Event not found
20160402 13:36:27< celticminstrel> (Also, this is from the console.)
20160402 13:39:58< vultraz> celticminstrel: you can easily test that
20160402 13:40:15< vultraz> doesn;t look bad
20160402 13:40:28< celticminstrel> Yeah, I can easily test it, but was in the middle of something.
20160402 13:41:02-!- irker092 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160402 13:41:02< irker092> wesnoth: Celtic Minstrel wesnoth:master 104ccf6e36bc / src/ai/formula/ (ai.cpp ai.hpp candidates.cpp candidates.hpp): Remove obsolete class from FormulaAI https://github.com/wesnoth/wesnoth/commit/104ccf6e36bc13db26aaecf884c378e3bd28d3e9
20160402 13:41:02< irker092> wesnoth: Celtic Minstrel wesnoth:master ca382018b7a9 / src/ (206 files in 41 dirs): BOOST_FOREACH -> range for https://github.com/wesnoth/wesnoth/commit/ca382018b7a9b5b744f0669bf1442a30ead63c15
20160402 13:41:02< irker092> wesnoth: Celtic Minstrel wesnoth:master fe818f80051e / / (5 files in 3 dirs): Add Lua API function to set the current time of day https://github.com/wesnoth/wesnoth/commit/fe818f80051ece8f1fb084fd9e5bc072a7cd1188
20160402 13:41:03< irker092> wesnoth: Celtic Minstrel wesnoth:master bfc6744b03c0 / src/gui/dialogs/ (loadscreen.cpp loadscreen.hpp): Workaround for MSVC2013's buggy atomic https://github.com/wesnoth/wesnoth/commit/bfc6744b03c01c3bf7440a073f9c6f6e399dae02
20160402 13:41:04< irker092> wesnoth: Celtic Minstrel wesnoth:master 4d1c9bdd437c / / (4 files in 2 dirs): Add wesnoth.eval_formula which compiles and evaluates a formula in the formula e https://github.com/wesnoth/wesnoth/commit/4d1c9bdd437cc4a9e3cb2ea2c2cc01abf1b27316
20160402 13:41:06< irker092> wesnoth: Celtic Minstrel wesnoth:master 38198e915172 / src/ (5 files in 2 dirs): Move config_callable to callable_objects.hpp and add a comparator function https://github.com/wesnoth/wesnoth/commit/38198e915172edb64c9d43b76168b72cfc860953
20160402 13:41:08< irker092> wesnoth: Celtic Minstrel wesnoth:master ac4479090fef / src/scripting/lua_formula_bridge.cpp: Add a comparator function for lua_callable https://github.com/wesnoth/wesnoth/commit/ac4479090fef362bd491d72ee011619700dc21eb
20160402 13:41:10< irker092> wesnoth: Celtic Minstrel wesnoth:master 5bb6c30b057d / src/scripting/ (5 files): Lua API: Add wesnoth.compile_formula which returns a callable userdata https://github.com/wesnoth/wesnoth/commit/5bb6c30b057d8b78363867e178d525123d4d698a
20160402 13:41:12< irker092> wesnoth: Celtic Minstrel wesnoth:master 5fdad345f243 / src/ (6 files in 2 dirs): Formula engine: Generalize system of defining and instantiating functions https://github.com/wesnoth/wesnoth/commit/5fdad345f243d7d9a7a72fe24c7e31dc24ff3cab
20160402 13:41:14< irker092> wesnoth: Celtic Minstrel wesnoth:master b503b6cfe107 / src/scripting/lua_formula_bridge.cpp: Lua formula bridge now recognizes location objects https://github.com/wesnoth/wesnoth/commit/b503b6cfe107720f9f05e497cb9158e526e3cbf8
20160402 13:41:16< irker092> wesnoth: Celtic Minstrel wesnoth:master cc7c1247445e / / (15 files in 4 dirs): Merge branch 'lua_formula_bridge' https://github.com/wesnoth/wesnoth/commit/cc7c1247445ea484a2e43f931f6a95ed8c1ed4fb
20160402 13:41:18< irker092> wesnoth: Celtic Minstrel wesnoth:master 5c9923dabaf2 / changelog src/formula/function.cpp src/formula/function.hpp: Formula engine: Add pair() function https://github.com/wesnoth/wesnoth/commit/5c9923dabaf2806e0ba45708a958f64f94f9a316
20160402 13:41:20< irker092> wesnoth: Celtic Minstrel wesnoth:master 0688e6df8f9e / changelog src/formula/callable_objects.cpp src/formula/callable_objects.hpp: Formula engine: Expose WML unit/side variables https://github.com/wesnoth/wesnoth/commit/0688e6df8f9ecf3c589da916fbd2e41b952f146d
20160402 13:41:42< vultraz> zomg
20160402 13:41:53< celticminstrel> ?
20160402 13:41:59< vultraz> so many commits
20160402 13:42:05-!- Kwandulin [~Miranda@p200300760F191C1E75FB33483B30152D.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160402 13:42:07< celticminstrel> Anything specific, or... that answers my question.
20160402 13:42:37-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160402 13:44:45< celticminstrel> Still need to do the exposing WML variables to $(...); the gradients and possibly rounded rectangle or pointset; uhh, I think there was something else too...
20160402 13:45:17< vultraz> Gradients aren't really a priority anymore
20160402 13:45:26< vultraz> I used a custom image texture background
20160402 13:45:50< celticminstrel> That's nice. I still want to try it, though.
20160402 13:46:01< celticminstrel> But yeah, it might not be soon.
20160402 13:46:28< celticminstrel> Oh right, the formula refactor.
20160402 13:46:36< celticminstrel> Bleh.
20160402 13:46:55< celticminstrel> Well, I should try that branch again though, maybe it'll actually build without changes now that we're C++11.
20160402 13:46:59< celticminstrel> If I'm lucky.
20160402 13:48:13< celticminstrel> I said "formula refactor", but it's mainly focusing on the variant.
20160402 13:48:56< celticminstrel> There's also that AMLA thing that I forgot about... and possibly the enhanced Lua in filters thing...
20160402 13:49:31< vultraz> So Many Changes
20160402 13:49:33< celticminstrel> I still feel like I've forgotten something though. Oh well. I'm sure it'll come back to me at the least convenient time.
20160402 13:49:40< celticminstrel> Hmm?
20160402 13:49:54< vultraz> you're working on so many things :P
20160402 13:50:05< celticminstrel> Yeah...
20160402 13:51:14< celticminstrel> It means that some of them might never get done. :P
20160402 13:52:13< celticminstrel> I still haven't finished the Lua-formula bridge, either - it's missing function support.
20160402 13:55:16< celticminstrel> Maybe what I was forgetting is related to all the tests.
20160402 13:55:24< celticminstrel> ie, fixing Travis.
20160402 14:00:03-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 248 seconds]
20160402 14:01:24-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160402 14:03:15< vultraz> you missed a case
20160402 14:03:21< celticminstrel> ?
20160402 14:03:58< vultraz> in src/tracer.cpp
20160402 14:04:04< vultraz> 2 cases of BOOST_FOREACH 
20160402 14:04:07< vultraz> I'll fix it
20160402 14:04:08< celticminstrel> ...
20160402 14:04:29< vultraz> (what does that file even do)
20160402 14:04:37< celticminstrel> That must be an unused source file then. (Or Windows-only.)
20160402 14:05:50< vultraz> Its header is included once, but that's behind an include guard
20160402 14:05:54< vultraz> #ifdef FLOATING_POINT_EMULATION_TRACER_ENABLE
20160402 14:06:03< vultraz> but that's never defined
20160402 14:07:22< vultraz> so in essence it's never used
20160402 14:07:46< vultraz> I guess it's not in your projectfile
20160402 14:08:02< irker092> wesnoth: Charles Dang wesnoth:master 9fc19a9fbde3 / src/tracer.cpp: Convert two remaining cases of BOOST_FOREACH to range-for https://github.com/wesnoth/wesnoth/commit/9fc19a9fbde34fcf90cd2efe44e7d0de68781cd2
20160402 14:08:56< celticminstrel> Yeah, I didn't use git grep this time.
20160402 14:09:20< vultraz> I'm going to deal with the std::bind stuff now
20160402 14:09:27< celticminstrel> Which means I might've missed more in tools or similar.
20160402 14:09:34< vultraz> what should I call the util file...
20160402 14:09:41< vultraz> utils/functional.hpp I guess
20160402 14:09:53< celticminstrel> That seems reasonable.
20160402 14:09:59< celticminstrel> Or utils/function.hpp
20160402 14:18:06< vultraz> what is boost::function2...
20160402 14:19:26< vultraz> or function0
20160402 14:29:30< vultraz> oh come on
20160402 14:29:35< vultraz> now there's function3!
20160402 14:31:50< vultraz> and of course, function1
20160402 14:39:53< vultraz> std::function doesn't have empty()?
20160402 14:55:02< irker092> wesnoth: mattsc wesnoth:master 20b07c63791c / data/ai/micro_ais/ (micro_ai_helper.lua scenarios/goto.cfg): Micro AIs: internal changes to AI and CA ID handling https://github.com/wesnoth/wesnoth/commit/20b07c63791c1c1add1ed2b9655808739a4a12f6
20160402 14:55:54-!- prkc [~prkc@catv-80-98-243-98.catv.broadband.hu] has joined #wesnoth-dev
20160402 15:03:09< vultraz> celticminstrel: could you help me with this: src\gui\widgets\repeating_button.cpp|143|error: no matching function for call to 'bind(<unresolved overloaded function type>, gui2::twindow*&, gui2::event::tevent, std::reference_wrapper<gui2::trepeating_button>)'|
20160402 15:03:24< vultraz> I checked tdispatcher, and all 4 fire() prototypes don't seem to match that argument list..
20160402 15:03:29< vultraz> so I'm not sure how it worked before
20160402 15:06:13-!- lipkab [~the_new_l@host-91-147-210-58.biatv.hu] has joined #wesnoth-dev
20160402 15:12:48-!- travis-ci [~travis-ci@ec2-54-159-167-2.compute-1.amazonaws.com] has joined #wesnoth-dev
20160402 15:12:49< travis-ci> wesnoth/wesnoth#9175 (master - 20b07c6 : mattsc): The build is still failing.
20160402 15:12:49< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/120287860
20160402 15:12:49-!- travis-ci [~travis-ci@ec2-54-159-167-2.compute-1.amazonaws.com] has left #wesnoth-dev []
20160402 15:14:04-!- aeth [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Ping timeout: 260 seconds]
20160402 16:07:52-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20160402 16:10:11-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160402 16:26:46-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20160402 16:42:47-!- aeth [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20160402 16:50:25-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20160402 17:08:15-!- ideuler [~textual@a89-153-75-168.cpe.netcabo.pt] has joined #wesnoth-dev
20160402 17:17:36-!- Kwandulin [~Miranda@p200300760F191C1E75FB33483B30152D.dip0.t-ipconnect.de] has quit [Ping timeout: 250 seconds]
20160402 17:53:19< celticminstrel> vultraz: boost::functionN<a,b,c> is boost::function<a(b,c)>
20160402 17:54:16< celticminstrel> vultraz: if(f.empty()) is just if(f)
20160402 17:55:29-!- irker092 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160402 18:02:02-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160402 18:06:24< mattsc> celticminstrel: I think I am now done with what I said I would do; or do you remember anything else?
20160402 18:07:55-!- prkc [~prkc@catv-80-98-243-98.catv.broadband.hu] has quit [Ping timeout: 252 seconds]
20160402 18:10:42< celticminstrel> I can't remember anything else.
20160402 18:16:22< celticminstrel> vultraz: Did you solve the fire() issue yet?
20160402 18:16:32< celticminstrel> Or repeating_button or whatever/
20160402 18:19:46-!- prkc [~prkc@46.166.190.190] has joined #wesnoth-dev
20160402 18:20:58< lipkab> celticminstrel, vultraz: For the record, boost::function is part of the C++11 standard (under the name std::function).
20160402 18:21:36< celticminstrel> Yes, that's the point, but there are some small differences.
20160402 18:22:37< lipkab> Oh, sorry, I didn't want to hijack the discussion.
20160402 18:23:21< lipkab> Just saw the boost::function thing there and wanted to point out that it's part of STL... since we're moving to C++11.
20160402 18:23:50< celticminstrel> I think vultraz was in the middle of replacing the one with the other.
20160402 18:23:54< lipkab> If you've known that already, then disregard me :)
20160402 18:29:19-!- Kwandulin [~Miranda@p200300760F0D01D015EC433FECB97F33.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160402 18:30:25-!- ideuler [~textual@a89-153-75-168.cpe.netcabo.pt] has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
20160402 18:42:45-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160402 18:44:13-!- ideuler [~textual@a89-153-75-168.cpe.netcabo.pt] has joined #wesnoth-dev
20160402 19:00:03-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160402 19:09:11< celticminstrel> So, aeth's GUI2 interface seems nice, if limited in use.
20160402 19:09:31< celticminstrel> As far as I can tell, it's currently specifically geared towards making menus.
20160402 19:09:44< celticminstrel> Where you select one option and an action is performed.
20160402 19:12:45-!- ideuler [~textual@a89-153-75-168.cpe.netcabo.pt] has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
20160402 19:16:43-!- irker414 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160402 19:16:43< irker414> wesnoth: Celtic Minstrel wesnoth:master 7c37ad57bd58 / data/ (core/units/humans/Mage_Silver.cfg lua/wml/message.lua): Two new [message] features https://github.com/wesnoth/wesnoth/commit/7c37ad57bd5858bc15a10d1e9f0deaa57a22467a
20160402 19:16:47< celticminstrel> ^ The non-controversial portion of the portrait_variant branch
20160402 19:18:38-!- ideuler [~textual@a89-153-75-168.cpe.netcabo.pt] has joined #wesnoth-dev
20160402 19:23:00-!- Kwandulin [~Miranda@p200300760F0D01D015EC433FECB97F33.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160402 19:23:09-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20160402 19:35:24-!- prkc [~prkc@46.166.190.190] has quit [Ping timeout: 260 seconds]
20160402 19:35:35-!- travis-ci [~travis-ci@ec2-54-159-167-2.compute-1.amazonaws.com] has joined #wesnoth-dev
20160402 19:35:36< travis-ci> wesnoth/wesnoth#9176 (master - 7c37ad5 : Celtic Minstrel): The build is still failing.
20160402 19:35:36< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/120322397
20160402 19:35:36-!- travis-ci [~travis-ci@ec2-54-159-167-2.compute-1.amazonaws.com] has left #wesnoth-dev []
20160402 19:50:15-!- prkc [~prkc@catv-80-98-243-98.catv.broadband.hu] has joined #wesnoth-dev
20160402 19:52:34< celticminstrel> Eh? Since when was it failing?
20160402 19:52:42-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20160402 20:01:50< celticminstrel> There's an unused MP create GUI2 dialog?
20160402 20:02:36< celticminstrel> Oh right, I forgot to add the new files to scons and CMake.
20160402 20:05:45< celticminstrel> ^file
20160402 20:05:55< irker414> wesnoth: Celtic Minstrel wesnoth:master ca512b20c245 / src/ (11 files in 7 dirs): Fix several more cases of missed BOOST_FOREACH and FOREACH https://github.com/wesnoth/wesnoth/commit/ca512b20c245da41d3d6e61344ebd45028b30988
20160402 20:05:57< irker414> wesnoth: Celtic Minstrel wesnoth:master b8441bba2437 / src/ (CMakeLists.txt SConscript): Update scons and CMake https://github.com/wesnoth/wesnoth/commit/b8441bba2437a12d9bd11d01a00488364c62009d
20160402 20:18:40-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160402 20:22:23< mattsc> celticminstrel: I have a vague recollection of you mentioning an AI behavior that cannot be done with the MAIs.  Am I misremembering this?
20160402 20:22:58< celticminstrel> Hmmm... it's possible I said something like that...
20160402 20:27:19-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20160402 20:37:56< celticminstrel> If so though, I don't remember what it could have been.
20160402 20:39:29-!- travis-ci [~travis-ci@ec2-54-159-167-2.compute-1.amazonaws.com] has joined #wesnoth-dev
20160402 20:39:30< travis-ci> wesnoth/wesnoth#9177 (master - b8441bb : Celtic Minstrel): The build has errored.
20160402 20:39:30< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/120329234
20160402 20:39:30-!- travis-ci [~travis-ci@ec2-54-159-167-2.compute-1.amazonaws.com] has left #wesnoth-dev []
20160402 20:50:02-!- trewe [~trewe@bl20-26-110.dsl.telepac.pt] has joined #wesnoth-dev
20160402 20:52:21< mattsc> okay
20160402 21:00:53< celticminstrel> I think I was thinking about variations on the hunter AI at one point, but I never mentioned that here.
20160402 21:05:01< celticminstrel> Like additional options governing when it decides to return home.
20160402 21:15:53-!- mjs-de [~mjs-de@p508C8F63.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160402 21:32:13< celticminstrel> ...the FormulaAI serializer doesn't save functions... I wonder if that means [function] is actually broken...
20160402 21:35:57-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160402 21:51:22-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160402 21:51:58-!- mjs-de [~mjs-de@p508C8F63.dip0.t-ipconnect.de] has quit [Remote host closed the connection]
20160402 22:03:48-!- Appleman1234 [~Appleman1@KD106161142191.au-net.ne.jp] has quit [Ping timeout: 246 seconds]
20160402 22:25:17-!- lipkab [~the_new_l@host-91-147-210-58.biatv.hu] has quit [Remote host closed the connection]
20160402 22:40:30-!- Greywhind [~Greywhind@c-71-232-29-126.hsd1.ma.comcast.net] has quit [Ping timeout: 248 seconds]
20160402 22:54:54-!- dlihcnoea [enchilado@defocus/yummy/enchilado] has quit [Ping timeout: 260 seconds]
20160402 22:56:30-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has quit [Quit: $Famous_last_words]
20160402 22:58:30-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has joined #wesnoth-dev
20160402 23:00:50-!- Greywhind [~Greywhind@c-71-232-29-126.hsd1.ma.comcast.net] has joined #wesnoth-dev
20160402 23:01:39-!- Appleman1234 [~Appleman1@KD106161150075.au-net.ne.jp] has joined #wesnoth-dev
20160402 23:06:14-!- irker414 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160402 23:10:39-!- dlihcnoea [enchilado@defocus/yummy/enchilado] has joined #wesnoth-dev
20160402 23:20:16-!- trewe [~trewe@bl20-26-110.dsl.telepac.pt] has quit [Ping timeout: 268 seconds]
20160402 23:30:02-!- irker332 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160402 23:30:02< irker332> wesnoth: gfgtdf wesnoth:master 1cc86a14af0b / src/game_initialization/ (connect_engine.cpp multiplayer.cpp): simplify start_local_game_commandline https://github.com/wesnoth/wesnoth/commit/1cc86a14af0b46cac24ca662b2ed6ffbcf15f9f4
20160402 23:38:44-!- travis-ci [~travis-ci@ec2-54-205-222-0.compute-1.amazonaws.com] has joined #wesnoth-dev
20160402 23:38:45< travis-ci> wesnoth/wesnoth#9178 (master - 1cc86a1 : gfgtdf): The build failed.
20160402 23:38:45< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/120354260
20160402 23:38:45-!- travis-ci [~travis-ci@ec2-54-205-222-0.compute-1.amazonaws.com] has left #wesnoth-dev []
20160402 23:47:02< vultraz> celticminstrel: didn't solve it yet
20160402 23:48:05-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160402 23:50:24< vultraz> celticminstrel: I was waiting to see if you had some insight 
20160402 23:50:34< vultraz> the call doesn't seem to match any of the functions
20160402 23:54:37-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 260 seconds]
20160402 23:56:46< vultraz> oh, jeez
20160402 23:57:06< vultraz> if I comment that out, similar errors crop up elsewhere
--- Log closed Sun Apr 03 00:00:52 2016