--- Log opened Tue Mar 15 00:00:09 2016
--- Day changed Tue Mar 15 2016
20160315 00:00:09-!- SpoOkyMagician [~chatzilla@2607:fcc8:be59:b00:d5fb:e8aa:3e47:6b94] has quit [Quit: brb getting gitbash]
20160315 00:03:52-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20160315 00:05:23-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160315 00:06:37< vultraz> I'm considering removing the helptip code from gui2
20160315 00:06:45< vultraz> I have no idea how it's any different from a tooltip
20160315 00:08:20< shadowm> I explained this before. :\
20160315 00:09:03-!- SpoOkyMagician [~chatzilla@2607:fcc8:be59:b00:d5fb:e8aa:3e47:6b94] has joined #wesnoth-dev
20160315 00:09:20-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160315 00:13:38< vultraz> [06:54:02]	vultraz	I think you guys said it's an archaic help thingy
20160315 00:13:51-!- iceiceice [~chris@50.245.222.235] has joined #wesnoth-dev
20160315 00:14:00< vultraz> something about a helpful tooltip or something 
20160315 00:14:07-!- iceiceice [~chris@50.245.222.235] has quit [Changing host]
20160315 00:14:07-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20160315 00:14:33-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160315 00:17:57-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160315 00:22:54-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160315 00:23:06< gfgtdf> vultraz: did you test how much memeory saved with LOW_MEM
20160315 00:23:08< gfgtdf> ?
20160315 00:26:24< vultraz> gfgtdf: no, but it's mostly things like TCing sprites
20160315 00:27:57< vultraz> gfgtdf: or similar small things
20160315 00:28:01< vultraz> so I'd imagine not much at all
20160315 00:28:11< gfgtdf> vultraz: hmm i still think its better to test that
20160315 00:28:19< vultraz> how
20160315 00:28:27< gfgtdf> vultraz: i men it coudl be that team colors take quite soe memory
20160315 00:28:48< gfgtdf> vultraz: compile wesnoth with LOW_MEM and without and compare the memoryconsuption at runtime
20160315 00:28:54< vultraz> gfgtdf: I'm pretty sure LOW_MEM is a relic from an age where these operations were actually expensive
20160315 00:28:59< vultraz> I think shadowm would agree
20160315 00:29:32< zookeeper> what are you trying to accomplish by dropping it?
20160315 00:29:45< vultraz> cleanup of unnecessary code
20160315 00:29:47< zookeeper> let me guess: the usual, simplifying the codebase
20160315 00:29:49< zookeeper> yep
20160315 00:30:07< gfgtdf> well simplyfying the codebase is a good thing
20160315 00:30:20< vultraz> the most laudable goal :)
20160315 00:31:15< iceiceice> did any official releases use LOW_MEM?
20160315 00:31:19< iceiceice> like hte pandora ?
20160315 00:31:29< vultraz> ask shadowm 
20160315 00:31:36< iceiceice> i guess pandora is gone now but concievably it could come back
20160315 00:32:16< iceiceice> vultraz, its one thing to like, refactor code that is unnecessarily complicated that improves it and doesn't destroy functionality
20160315 00:32:27< zookeeper> does LOW_MEM get in anyone's way when coding? i can't see how it would. so let's drop it because it's cleanup and that's good because... otherwise it gets in someone's way? oh wait, it doesn't.
20160315 00:32:35< iceiceice> its another thing to just go around deleting features
20160315 00:32:51< iceiceice> i mean i guess i dont know how much code is involved in deleting this
20160315 00:33:11< iceiceice> one thing to consider is that, it may be that the risk of causing a bug by deleting a ton of code is greater than the benefit
20160315 00:33:41< vultraz> 39 cases of LOW_MEM
20160315 00:33:54< vultraz> a chunk involve disabling potential debug code
20160315 00:34:04< gfgtdf> i wonder whether our config class will benefit from c++14 heterogeneous lookup
20160315 00:34:27< gfgtdf> so that when doing cfg["aasd"] it woudlnt need to create a temporary string from the char*
20160315 00:44:17-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20160315 00:46:55-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Ping timeout: 252 seconds]
20160315 00:50:14-!- gfgtdf [~chatzilla@f050176167.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 45.0/20160303134406]]
20160315 00:51:47-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 248 seconds]
20160315 00:53:03< shadowm> vultraz: I quite explicitly compared it to the older "What's This" contextual help feature Windows had/has.
20160315 00:53:18< vultraz> Yes, I recall that
20160315 00:53:29< shadowm> As I recall, mordante intended to develop it further but other things took priority.
20160315 00:53:33< vultraz> But I only vaguely recall what that does 
20160315 00:53:46< shadowm> ...
20160315 00:54:00< shadowm> At the time, I said "vultraz surely doesn't know what that is".
20160315 00:54:06< vultraz> I also recall that
20160315 00:54:09< shadowm> Then you said you knew, so I decided I didn't need to explain further.
20160315 00:54:19< vultraz> I said I vaguely know 
20160315 00:55:09< vultraz> I *believe* they're popups with explanations of functionality 
20160315 00:55:16< shadowm> Tooltips were classically used for displaying short descriptions (or names) of controls, not more elaborate help text.
20160315 00:55:36< shadowm> Contextual help was designed to be a middle ground between both.
20160315 00:55:40-!- louis94 [~~louis94@91.178.241.103] has quit [Quit: Konversation terminated!]
20160315 00:56:19< shadowm> I guess it started to disappear when task-oriented UI design came into the picture.
20160315 00:59:01< vultraz> oh wait
20160315 00:59:11< vultraz> was it those clickable question marks?
20160315 00:59:45< vultraz> I recall such things in ancient windows versions such as XP and Vista
20160315 01:00:44< vultraz> I don't think we should have such things in a game
20160315 01:03:17< shadowm> "Ancient" he says.
20160315 01:04:31< vultraz> well, XP is ancient 
20160315 01:04:39< vultraz> Vista is 'old'
20160315 01:04:57< shadowm> I wouldn't call software that was still supported two years ago (and still is, depending on the exact SKU you're using) "ancient".
20160315 01:05:44< shadowm> Especially not when major ISVs like Google too support it.
20160315 01:11:08< shadowm> Regarding LOW_MEM, you are going to make choices on a whim if it makes the pseudo-perfectionist in you happier, so I'm not sure why you want to bring me into the discussion.
20160315 01:44:28-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20160315 01:59:36-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160315 02:35:00-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160315 02:39:30-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160315 02:49:33-!- mjs-de [~mjs-de@x5ce490ca.dyn.telefonica.de] has quit [Ping timeout: 246 seconds]
20160315 03:12:05-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160315 03:30:29-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20160315 03:36:08-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20160315 04:13:57-!- oldlaptop [~quassel@50-37-53-169.mskg.mi.frontiernet.net] has quit [Read error: Connection reset by peer]
20160315 04:24:30-!- oldlaptop [~quassel@50-37-53-169.mskg.mi.frontiernet.net] has joined #wesnoth-dev
20160315 04:41:18< wedge009> celticminstrel: I probably count as half. And I'm usually active in times when the Americas are asleep, so I wouldn't be surprised if people don't recognise me.
20160315 05:09:57-!- ancestral [~ancestral@75-168-168-95.mpls.qwest.net] has joined #wesnoth-dev
20160315 05:27:32-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds]
20160315 05:30:18-!- irker239 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160315 05:35:12-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160315 05:39:40-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 250 seconds]
20160315 05:44:17-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20160315 05:48:42-!- ancestral [~ancestral@75-168-168-95.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160315 06:17:24< shadowm> vultraz: So I guess this is what you were alluding to the other day: https://github.com/cbeck88/spirit-po
20160315 06:35:54-!- SpoOkyMagician [~chatzilla@2607:fcc8:be59:b00:d5fb:e8aa:3e47:6b94] has quit [Quit: ChatZilla 0.9.92 [Firefox 44.0.2/20160210153822]]
20160315 06:57:58-!- Kwandulin [~Miranda@p200300760F7D73291D50DC339569BFBE.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160315 07:00:43-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160315 07:29:22< vultraz> shadowm: yes
20160315 08:02:17-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160315 08:06:16< vultraz> Is it just me or are translations broken :|
20160315 08:06:26< vultraz> No matter what language I select I always see english
20160315 08:06:32-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160315 08:31:26-!- boucman_work [~boucman@247.37.0.109.rev.sfr.net] has joined #wesnoth-dev
20160315 08:38:21< vultraz> Huh
20160315 08:38:30< vultraz> Ok, so helptips work in gui2 as such:
20160315 08:38:52< vultraz> if a helptip is present, the tooltip will read "tooltip text. (Fress f10 for more information)"
20160315 08:39:22< vultraz> then pressing f10 shows the helptip as another tooltip
20160315 08:40:28< shadowm> Yes.
20160315 08:41:09< shadowm> Hmmm.
20160315 08:41:17< shadowm> vultraz, that tooltip design you added...
20160315 08:41:37< shadowm> Is that the toggle_button/panel background?
20160315 08:42:11< vultraz> yes, the selected one, but darkened 
20160315 08:42:24< shadowm> I feel like they ought to stand out on their own.
20160315 08:42:38< vultraz> I'm open to suggestions
20160315 08:42:41< shadowm> Not look like a floating listbox entry.
20160315 08:43:42< shadowm> White text, darker background?
20160315 08:43:58< shadowm> *darker solid background
20160315 08:44:44< shadowm> I don't know. If I knew how to design a UI's visuals I'd have done so ages ago and I wouldn't have had to put up with LordBob's for so long.
20160315 08:44:44< vultraz> shadowm: https://drive.google.com/file/d/0B-mR9s8FduLLY3VNamZHYmNQUlE/view?usp=sharing 
20160315 08:44:50< vultraz> (except now it's wider)
20160315 08:45:06< shadowm> (Yes, I just said I don't like our current UI design. Deal with it.)
20160315 08:45:14< vultraz> ?!
20160315 08:45:27< vultraz> what! 
20160315 08:45:47< fabi> Or ignore him.
20160315 08:45:58< vultraz> I mean, I know I don't like the scrollbar design but come on :|
20160315 08:46:02< shadowm> I don't know, but that still doesn't look like a proper tooltip to me.
20160315 08:46:07-!- Kwandulin [~Miranda@p200300760F7D73291D50DC339569BFBE.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160315 08:47:06< shadowm> vultraz: The blues are rather muddy, the overall contrast is too high and annoying when staring at it for prolonged periods of time (e.g. the MP lobby), and the dialog frames are too noisy.
20160315 08:47:25< shadowm> s/annoying/eye-straining/
20160315 08:47:33< shadowm> (That's even worse than just plain annoying.)
20160315 08:47:35< vultraz> that's why I made it gold-small
20160315 08:47:50< shadowm> ... I was talking about the overall UI design.
20160315 08:48:05< vultraz> blah
20160315 08:48:29< shadowm> And talking about contrast, I think that might be the issue with the new tooltips: no contrast at all.
20160315 08:48:50< shadowm> Both the dialog behind and the tooltip box have the same color scheme.
20160315 08:49:04< shadowm> But really, you don't have to pay attention to me. Do like fabi said and ignore me.
20160315 08:49:22< shadowm> I'm not terribly interested in complaining any further about an issue I don't know how to fix.
20160315 08:50:51< vultraz> DO you at least think they're better than the semi-transparent black blob at the bottom of the screen
20160315 08:51:15< shadowm> They are at least 400% superior to the bottom tooltips, yes.
20160315 08:53:42< shadowm> I don't think bigger text is the solution, by the way. The current size in master is just good.
20160315 08:53:57< vultraz> the tooltip text size?
20160315 08:54:14< shadowm> Yes.
20160315 08:54:27< shadowm> Maybe what they need is a shadow -- wait, WE CAN'T DO SHADOWS.
20160315 08:54:48< shadowm> Not dynamic shadows, at least.
20160315 08:55:47< shadowm> So here's a simple test case: when you go to Preferences → General, and mouse over the Cache button from the bottom, the tooltip will usually overlap the button itself.
20160315 08:56:17< shadowm> The effect seems a bit... inexpensive.
20160315 08:56:25< shadowm> Like, very obviously inexpensive.
20160315 08:56:47< vultraz> should it not overlap the button?
20160315 08:56:57< shadowm> Eh, no, it's not the placement that's an issue.
20160315 08:57:10< Aginor> gfgtdf: The tool I was using must have picked up on #if 0 && ... and decided to take out the segment. Should it be reinstated or can it stay dropped?
20160315 08:57:14< shadowm> Hence I suggested shadows.
20160315 08:57:25< Aginor> does gfgtdf read his blacklog? :)
20160315 08:57:31< shadowm> Aginor: Yes.
20160315 08:58:00< shadowm> vultraz: I guess what I'm trying to say is that I want to see more shadows.
20160315 08:58:26< vultraz> easier said than done
20160315 08:58:44< shadowm> Nested dialogs have the same issue. The dialog frame does have a shadow plastered on the image files, but it doesn't stand out enough.
20160315 08:58:53< shadowm> So when you have a dialog on top of another, it looks a bit...
20160315 08:59:00< shadowm> I mean come on, I am not the 2000s kid here.
20160315 08:59:28< vultraz> https://www.dropbox.com/s/cyazkc7hja82mks/slightly%20different%20tooltips.png?dl=0
20160315 08:59:30< shadowm> You probably should be even more bothered than I am by the lack of shadows on active elements.
20160315 08:59:34< vultraz> will work on shadows later
20160315 08:59:39< shadowm> Back in my days we didn't have shadows at all.
20160315 09:00:08< Aginor> And you had walk uphill both ways, in a perpetual blizzard. But milk was cheaper.
20160315 09:00:38 * shadowm glares at vultraz.
20160315 09:00:38< Aginor> (sorry, but I couldn't resist :) )
20160315 09:00:55< shadowm> Why did you randomly switch back to Dropbox?
20160315 09:01:41< shadowm> Ohhhh, I finally figured it out. Double-clicking shows the image at 100% zoom properly when in full size mode.
20160315 09:01:57< shadowm> You'd think full size mode would start at 100% but no.
20160315 09:02:10< vultraz> shadowm: I only use google drive for celmin
20160315 09:02:13< vultraz> since he can't see dropbox
20160315 09:02:16< vultraz> you can see dropbox
20160315 09:02:41< shadowm> I can but I've complained a trillion times about Dropbox's piece of crap image view JS.
20160315 09:02:43< vultraz> anyway, I must say, I think I prefer the nice 1-px procedurally drawn border
20160315 09:02:48< vultraz> it looks cleaner, somehow
20160315 09:03:11< shadowm> Yeah, that looks like a step in the right direction.
20160315 09:03:11< vultraz> I might adopt the paradigm in more places
20160315 09:03:14 * vultraz out to dinner
20160315 09:03:31< shadowm> That's my other issue with the UI: way too painterly.
20160315 09:03:52< shadowm> Sure, it doesn't inconvenience you people. It doesn't inconvenience our users either.
20160315 09:04:00< shadowm> It makes it a royal PITA to replicate in CSS, however.
20160315 09:05:06< shadowm> (Which is why I decided to use a simplified version of our color scheme in wesmere instead.)
20160315 09:05:20< Aginor> shadowm: are you trying to make the theme in css?
20160315 09:05:42< shadowm> Not the full thing, just a couple of major elements.
20160315 09:05:48-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20160315 09:06:01< Aginor> that strikes me as a rather substantial task :)
20160315 09:06:35< shadowm> It's not that bad, really.
20160315 09:08:55< shadowm> vultraz: Are we going to have three tooltip styles around from now on, or is the theme UI's floating tooltip (see the sidebar) going to be replaced with this?
20160315 09:09:47< Aginor> Make the main gamescreen the next major thing that gets ported to GUI2 :)
20160315 09:09:56< shadowm> Because I strongly suggest cutting down on UI inconsistencies rather the other way around.
20160315 09:10:22< shadowm> If vultraz can port the theme UI to GUI2 then I'll eat my pillow.
20160315 09:10:40< shadowm> *without substantial assistance
20160315 09:10:53< Aginor> here I was going to demand a video :)
20160315 09:12:51-!- zombah [~zombah@2a02:28:3:1:214:4fff:fe47:5920] has joined #wesnoth-dev
20160315 09:14:12< shadowm> I feel like there are a lot of bordered elements in our UI that could have their borders removed or toned down.
20160315 09:14:58< zookeeper> gfgtdf, how exactly does limit=n work? does it make the filter only match if there are at least n units which pass (rest of) the filter? and/or does it make only the first n units pass the filter?
20160315 09:15:10< zookeeper> calling it "limit" implies the latter
20160315 09:15:20< shadowm> The in-game UI notably has: double borders on the top and side bars, bright borders around every single black box element, borders around the minimap area containing the minimap and minimap options.
20160315 09:15:23< Aginor> I suspect there's room for a lot of facelift within the UI to make it feel more modern
20160315 09:15:37< shadowm> The combination of all these borders in such a small space make it look really crowded.
20160315 09:16:09< shadowm> (And the terrain info area has a divider between both lines that doesn't contribute much.)
20160315 09:16:28< shadowm> (Oh, and even the map zoom slider has a border around it for some reason???)
20160315 09:16:56< shadowm> Yeah, modern is not the image that all this evokes for me.
20160315 09:17:26< shadowm> Oh, and the menu buttons have borders too.
20160315 09:17:59< shadowm> And dead space between them (for some reason???).
20160315 09:18:29< shadowm> I guess in that case the issue is the fact that they are buttons in the first place.
20160315 09:18:44< shadowm> You don't really see applications using buttons to implement their menu bars these days.
20160315 09:19:38< shadowm> Also, minor usability niggle: the End Turn button's activation area does not include the bottom right corner of the screen/side bar.
20160315 09:20:12< Aginor> I still wonder if trying to make the game run under QT might not be worthwhile
20160315 09:20:26< shadowm> So if you have the mouse cursor over a unit and quickly move it to the corner, you'll miss the button.
20160315 09:20:40< shadowm> Yeah, I'm going to pass on that subject.
20160315 09:20:50< Aginor> :D
20160315 09:20:55< shadowm> I've used Qt before, so I don't have anything against it.
20160315 09:21:38< Aginor> likewise. It's solved a lot of the problems we're struggling with
20160315 09:21:41< shadowm> But I feel it promotes a coding style that is completely against Wesnoth and modern C++(11/14) in general.
20160315 09:22:25< shadowm> (new and naked pointers everywhere, for the curious.)
20160315 09:23:27< shadowm> Nevertheless, it's not my business anymore so it doesn't really matter to me.
20160315 09:27:48< Aginor> I'm trying to stay out of those bits myself
20160315 09:28:12< Aginor> so I shall go back to trying to figure out which mouse to order
20160315 09:28:28< Aginor> why are most mice wireless nowadays unless they're silly gaming mice?
20160315 09:30:41< shadowm> Hm, I've not used wired mice for almost a decade, and I can positively say I prefer wireless in this case.
20160315 09:31:08< shadowm> It wasn't very nice having to fight entangled cables all the time, especially while drawing.
20160315 09:35:53< Aginor> looks like I'll be going wireless
20160315 09:35:55< Aginor> http://www.logitech.com/en-us/product/marathon-mouse-m705
20160315 09:35:58< Aginor> for the curious
20160315 09:37:48< shadowm> Asymmetric mice. ¬_¬
20160315 09:37:48< Aginor> shadowm: in my previous job, mice in the 2G4 spectrum were pretty useless, by the time you have 40+APs, lots of BT devices and a people doing signal generators, most wireless is pretty crap
20160315 09:38:06< shadowm> Oh yeah, that would be an issue.
20160315 09:41:33< Aginor> yes ;)
20160315 09:42:38-!- Appleman1234 [~Appleman1@KD106161150240.au-net.ne.jp] has quit [Ping timeout: 244 seconds]
20160315 09:43:33< Aginor> and I'm finally done doing everything that needed sorting
20160315 09:57:35-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20160315 09:59:17< vultraz> [20:19:36]	shadowm	Also, minor usability niggle: the End Turn button's activation area does not include the bottom right corner of the screen/side bar.
20160315 09:59:20< vultraz> ancestral has a fix for this
20160315 10:00:39< vultraz> [20:08:54]	shadowm	vultraz: Are we going to have three tooltip styles around from now on, or is the theme UI's floating tooltip (see the sidebar) going to be replaced with this?
20160315 10:00:42< vultraz> latter
20160315 10:00:56< vultraz> somehow
20160315 10:01:05< celticminstrel> But not right away, I'm sure.
20160315 10:01:14 * Aginor goes to bed
20160315 10:01:21< celticminstrel> What was the reason for not yet using iceiceice's po loader?
20160315 10:01:36< celticminstrel> Something something C++11, right?
20160315 10:03:40< vultraz> celticminstrel: c++11 is needed
20160315 10:04:06< celticminstrel> So, it doesn't compile on MSVC2010?
20160315 10:04:11< vultraz> as soon as gfgtdf updates his compiler we're going to c++11
20160315 10:25:12< vultraz> blah, I have to fixup rectangle drawing *again*
20160315 10:28:26< vultraz> ok, got it
20160315 10:29:00-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160315 10:29:00-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Read error: Connection reset by peer]
20160315 10:29:27-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160315 10:33:48-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160315 10:38:33-!- Elvish_Hunter [~elvish_hu@wesnoth/developer/elvish-hunter] has joined #wesnoth-dev
20160315 10:39:00< Elvish_Hunter> Hi all
20160315 10:39:06< celticminstrel> Hi
20160315 10:39:11< vultraz> allos
20160315 10:40:04-!- Appleman1234 [~Appleman1@106.161.140.138] has joined #wesnoth-dev
20160315 10:40:15< Elvish_Hunter> I read about the number separator in the attack selection dialog, and I have a proposal...
20160315 10:40:48< Elvish_Hunter> What do you think of this? " ⋅ " It's the dot used as multiplication sign.
20160315 10:41:48< vultraz> hmmm
20160315 10:41:52< vultraz> that's an idea
20160315 10:42:17< zookeeper> i hope that with the in-game font it'd be bigger than the single pixel i see right now :p
20160315 10:42:57< Elvish_Hunter> Well, DejaVu Sans has plenty of bigger dots available. It's not a problem choosing a different one ;)
20160315 10:43:07< zookeeper> but i don't think it'd really be more intuitive or otherwise clearer than the x symbol
20160315 10:44:39-!- louis94 [~~louis94@91.178.242.179] has joined #wesnoth-dev
20160315 10:47:38-!- Nobun [~nobun@host62-49-dynamic.12-79-r.retail.telecomitalia.it] has joined #wesnoth-dev
20160315 10:48:22< Nobun> a little question: the 'master' branch on github repository is the development branch of wesnoth 1.13, right?
20160315 10:48:50< Nobun> (and I assume that 1.12 branch is for the 1.12.x version, right?)
20160315 10:49:23< Elvish_Hunter> Nobun: you're right.
20160315 10:49:39< celticminstrel> Bullet sign is also a dot.
20160315 10:52:29< celticminstrel> Rhonda: SSL certificate still not valid for irclogs.wesnoth.org
20160315 10:58:41-!- louis94 [~~louis94@91.178.242.179] has quit [Quit: Konversation terminated!]
20160315 11:05:04< wedge009> Aginor: I've been using that mouse for years. In fact, I have two of them.
20160315 11:06:48-!- irker575 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160315 11:06:48< irker575> wesnoth: Charles Dang wesnoth:master fe4623e15215 / src/gui/auxiliary/canvas.cpp: Fixup canvas background drawing again https://github.com/wesnoth/wesnoth/commit/fe4623e15215195f278777c4cab923329adb1764
20160315 11:06:48< irker575> wesnoth: Charles Dang wesnoth:master 53af40b2f324 / data/gui/default/ (widget/window_tooltip_large.cfg window/tooltip_large.cfg): ttip: improved look of floating tooltips further https://github.com/wesnoth/wesnoth/commit/53af40b2f324379985114b7852fd23b2f565958d
20160315 11:06:49< irker575> wesnoth: Charles Dang wesnoth:master 06788955c480 / data/gui/default/ (macros/_initial.cfg widget/combobox_default.cfg widget/window_tooltip_large.cfg): GUI2: added macro for border color https://github.com/wesnoth/wesnoth/commit/06788955c480333b1617d6fe40b7f2f14c6c9769
20160315 11:06:51< irker575> wesnoth: Charles Dang wesnoth:master 4f106b4d1f6f / data/gui/default/widget/ (panel_title_screen.cfg window_default.cfg): GUI2: gave windows procedurally-drawn minimal borders https://github.com/wesnoth/wesnoth/commit/4f106b4d1f6f07a27b190b10018078073ae9571a
20160315 11:07:08< celticminstrel> BTW vultraz, TC is definitely quite an extensive operation.
20160315 11:07:18< celticminstrel> ^expensive
20160315 11:07:22< vultraz> oh?
20160315 11:07:35< celticminstrel> It's basically std::transform.
20160315 11:07:47< celticminstrel> Applied to a gigantic 2D array of pixels.
20160315 11:08:32< vultraz> ...really?
20160315 11:08:50< celticminstrel> Well, I don't know exactly how it works, there may be some optimizations I'm unaware of.
20160315 11:09:08< celticminstrel> But that's the effect, at least.
20160315 11:09:37< Elvish_Hunter> Given brunomacabeusbr's answer, I was thinking of merging PR #461. Does anyone know how of get rid of the merge commits, if it's possible at all?
20160315 11:09:41< celticminstrel> It probably needs to test every pixel, see if it matches the magenta, and then replace it if it does.
20160315 11:09:58< Elvish_Hunter> * how to get rid of
20160315 11:09:59< vultraz> Elvish_Hunter: rebase
20160315 11:10:05< celticminstrel> Elvish_Hunter: The way I did it with whatsisname is git pull --squash
20160315 11:10:11< celticminstrel> However that removes author information.
20160315 11:10:32< celticminstrel> So I then had to git commit --author "his name <email>"
20160315 11:10:57< celticminstrel> Also, that does shove the entire thing into one commit. If that's undesirable, rebase is better.
20160315 11:11:34< celticminstrel> Rebase (by default) essentially ignores merge commits and flattens the history tree.
20160315 11:12:17< celticminstrel> Oh yeah vultraz, helptip is indeed the clickable question mark.
20160315 11:12:33< celticminstrel> Click the ? button, then click a control to get more info.
20160315 11:12:34< vultraz> celticminstrel: do you think we need to keep helptip functionality
20160315 11:12:50< celticminstrel> Is it used anywhere?
20160315 11:13:02< vultraz> no
20160315 11:13:07< celticminstrel> Huh, so wedge009 counts as half a dev? :P
20160315 11:13:19< vultraz> also, did you see my comment on your formula pr
20160315 11:13:19< celticminstrel> vultraz: Can it be removed cleanly, without lots of side-effects?
20160315 11:13:25< celticminstrel> I didn't see anything.
20160315 11:13:51< vultraz> i think so
20160315 11:13:52< celticminstrel> I spent the last... whoa, almost 12 hours struggling to shut down my computer cleanly (and giving up at the very last minute).
20160315 11:14:12< celticminstrel> vultraz: Then I don't see any reason t keep it.
20160315 11:14:25< celticminstrel> However, I suggest getting at least two more opinions.
20160315 11:14:59< celticminstrel> My take is, "why do we need both tooltip and helptip, they're more or less the same in concept", but I might be missing something.
20160315 11:15:54 * celticminstrel just got to the part in the logs where you describe them. Okay, so maybe they could have a little use here and there ...
20160315 11:17:05< celticminstrel> 20160315 08:54:27< shadowm> Maybe what they need is a shadow -- wait, WE CAN'T DO SHADOWS.
20160315 11:17:22< celticminstrel> This isn't quite the same, but isn't there a blur functionality in GUI2?
20160315 11:18:10< celticminstrel> In fact, since there's a blur, maybe adding shadows wouldn't be all that hard.
20160315 11:39:57-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160315 11:40:03< wedge009> celticminstrel: Yeah, not exactly worth a full one. x.x
20160315 11:40:12< celticminstrel> Why not?
20160315 11:43:42< Rhonda> celticminstrel: That domain doesn't do anything but redirect.  And it's even just a .htaccess file instead of its own vhost …
20160315 11:44:29< Rhonda> I would know how to do the exceptions in the vhost, didn't dare to touch the .htaccess to make the acme directory still accessible under the different name, too.
20160315 11:44:33< celticminstrel> Oh, you're right, it redirects to wesnoth.org/irclogs.
20160315 11:44:34< Rhonda> Too much on my head, sorry.
20160315 11:58:29-!- Elvish_Hunter [~elvish_hu@wesnoth/developer/elvish-hunter] has left #wesnoth-dev ["Ciao!"]
20160315 12:00:53-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160315 12:22:41< vultraz> Aginor: next time you're around, could you help me with something
20160315 12:40:33-!- boucman_work [~boucman@247.37.0.109.rev.sfr.net] has quit [Ping timeout: 276 seconds]
20160315 12:46:43-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160315 12:47:04-!- Appleman1234 [~Appleman1@106.161.140.138] has quit [Quit: Leaving]
20160315 12:47:25-!- Appleman1234 [~Appleman1@KD106161140138.au-net.ne.jp] has joined #wesnoth-dev
20160315 13:08:59-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20160315 13:09:00-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Read error: Connection reset by peer]
20160315 13:09:37< vultraz> shadowm: btw, take a look at my new window borders
20160315 13:10:24-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20160315 13:11:40 * zookeeper wonders if someone else could run data/tools/hexometer
20160315 13:12:33-!- Nobun [~nobun@host62-49-dynamic.12-79-r.retail.telecomitalia.it] has quit [Ping timeout: 240 seconds]
20160315 13:13:55-!- Nobun [~nobun@host62-49-dynamic.12-79-r.retail.telecomitalia.it] has joined #wesnoth-dev
20160315 13:14:01-!- Nobun [~nobun@host62-49-dynamic.12-79-r.retail.telecomitalia.it] has quit [Read error: Connection reset by peer]
20160315 13:14:07< irker575> wesnoth: Charles Dang wesnoth:master 33a68d33fd99 / data/themes/_initial.cfg: Remove background from theme status boxes and tweak position slightly https://github.com/wesnoth/wesnoth/commit/33a68d33fd99862968ff131c9c4d0601803f581a
20160315 13:14:11-!- Nobun [~nobun@host62-49-dynamic.12-79-r.retail.telecomitalia.it] has joined #wesnoth-dev
20160315 13:19:10< zookeeper> well, looks like i can't get hexometer to produce meaningful output
20160315 13:22:02< zookeeper> anyone want to pastebin what hexometer spits out if run on the terrain dir? that is, hexometer -f ../../core/images/terrain
20160315 13:25:07< fabi> hexometer?
20160315 13:25:37< zookeeper> hexometer
20160315 13:25:41< fabi> zookeeper: hexometor: command not found
20160315 13:25:47< vultraz> "command not found" for basically everything
20160315 13:25:54< vultraz> fabi: use ./hexometer
20160315 13:26:10< fabi> at toplevel
20160315 13:26:10< zookeeper> and don't typo the command
20160315 13:26:15< vultraz> zookeeper: nothing meaningful
20160315 13:26:17< celticminstrel> What is this thing.
20160315 13:26:47< vultraz> no idea
20160315 13:26:58< zookeeper> checks images for out-of-hex pixels
20160315 13:26:59< vultraz> sorcery
20160315 13:27:09 * celticminstrel blink
20160315 13:27:15< zookeeper> which would be convenient because i don't want to write a gimp script for that
20160315 13:27:18< celticminstrel> It's a directory.
20160315 13:27:27< vultraz> celticminstrel: yes, with a file inside
20160315 13:27:34< celticminstrel> Okay found it.
20160315 13:28:24< celticminstrel> Do you guys have ImageMagick installed?
20160315 13:28:41< celticminstrel> Also, do I just run it from the repo root with no parameters?
20160315 13:28:54< vultraz> I ran it from that dir
20160315 13:28:59< vultraz> and no I don't have imagemagick
20160315 13:29:02< celticminstrel> With no parameters?
20160315 13:29:10< vultraz> with the parameters zookeeper provided
20160315 13:29:22< celticminstrel> Oh, I see it.
20160315 13:29:23< zookeeper> the parameters i provided naturally assume you run it from its dir
20160315 13:29:32< zookeeper> adjust accordingly if not
20160315 13:29:54< fabi> it works
20160315 13:30:12< celticminstrel> My "find" doesn't support the -regextype flag.
20160315 13:30:44< celticminstrel> ...wait what.
20160315 13:30:51< fabi> zookeeper: Seems to take a while... I paste it when the task is finished.
20160315 13:31:23< zookeeper> fabi, great, thanks
20160315 13:32:17< celticminstrel> Well whatever. Doesn't work for me, due to "find" command incompatibilities.
20160315 13:32:20< fabi> zookeeper: http://pastebin.com/9K57CHMD
20160315 13:32:21< fabi> you are welocme
20160315 13:33:08< zookeeper> fabi, did you use -f?
20160315 13:33:36< fabi> let me check
20160315 13:33:44< zookeeper> right, yeah looks like you did
20160315 13:33:44< zookeeper> excellent
20160315 13:33:46< fabi> yes
20160315 13:34:06< zookeeper> so many files to fix D:
20160315 13:34:19< fabi> Indeed. Looks like every file has out of hex pixels
20160315 13:34:25< fabi> Why do they need fixing?
20160315 13:35:52< zookeeper> well, not all of them do. but ones which are placed with multihex terrain graphics rules will cause extra images to be placed for each adjacent hex onto which the extra pixels bleed onto.
20160315 13:36:22< zookeeper> which is of course hardly ever a big problem as such, but...
20160315 13:36:40< fabi> zookeeper: Shouldn't fixing them just applying the hex field mask upon them?
20160315 13:36:47< fabi> I hope you don't do that by hand.
20160315 13:37:07< zookeeper> yes, that's how they can be fixed
20160315 13:37:30< zookeeper> i've done a few by hand already because i don't have an automatic way to do it
20160315 13:38:15< fabi> I have never scripted stuff like gimp or imagemagik...
20160315 13:39:14< fabi> But applying a mask should be a regular procedure. Most likely scripting it isn't that much work.
20160315 13:39:42< zookeeper> yeah, i'll do it as a gimp script if i decide it's worth it to fix them
20160315 13:42:19< fabi> Wouldn't it be better to extend the "hexometer" script=
20160315 13:42:23< mattsc> celticminstrel: when running your AI refactor branch, I’m getting this “error general: Unknown key: ^”.  I assume that that has nothing to do with your changes and that I can ignore it?
20160315 13:42:47< celticminstrel> I have never seen that before.
20160315 13:42:52-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160315 13:43:20< celticminstrel> I hope it's nothing to do with my changes.
20160315 13:44:01< mattsc> I’ve checked my preferences and didn’t see me using the key anywhere.
20160315 13:44:43< mattsc> Anyways, it doesn’t cause any trouble (other than cluttering up stderr), so I’ll just leave it for now.
20160315 13:44:58-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 244 seconds]
20160315 13:44:58-!- wedge010 is now known as wedge009
20160315 13:48:10< mattsc> Oh, wow, I am also getting thousands of unhandled gui event warnings.
20160315 13:48:25< mattsc> (in this case when the preferences menu was open)
20160315 13:48:37< celticminstrel> Yeah, I've been getting those too.
20160315 13:48:41< celticminstrel> On any branch.
20160315 13:50:30< mattsc> I assume it’s something that might be fixed by now?
20160315 13:50:42 * vultraz shrugs
20160315 13:50:49< vultraz> I ignore them as noise
20160315 13:50:50< celticminstrel> Nope.
20160315 13:50:54< celticminstrel> Not fixed.
20160315 13:50:58< celticminstrel> It affects master.
20160315 13:51:33< celticminstrel> It actually means that if I run WML unit tests with the GUI, they fail because of strict mode and those warnings.
20160315 13:51:51< celticminstrel> Though it's not hard to disable strict mode... but it's a little annoying.
20160315 13:51:52< mattsc> I see.  That makes looking for errors in stdout/err a bit harder …
20160315 13:52:04< mattsc> celticminstrel: btw, I did find the source of the ^ key error.  It was indeed a custom hotkey in my preferences file.  And I guess hotkeys work differently now, so it did not show up ingame.
20160315 13:52:14< celticminstrel> Ah, okay.
20160315 13:52:40< mattsc> I haven’t worked with 1.13 in a while  :P
20160315 13:52:51< celticminstrel> You shouldn't get too many GUI errors while in-game, unless you have a lot of [message] or are using the inspector or something.
20160315 13:53:00< mattsc> ok
20160315 13:54:44< mattsc> celticminstrel: on commit a77e90cc2654ee, why do you do this only for leader_goal?  Do the other aspects do it like that already?
20160315 13:55:37< celticminstrel> What's the commit message?
20160315 13:55:40< celticminstrel> Oh wait.
20160315 13:55:43< celticminstrel> I think I found it.
20160315 13:55:58< mattsc> “In [leader_goal], never set facet ID to * or a number “
20160315 13:56:31< celticminstrel> That whole block is because of leader_goal's auto_remove feature.
20160315 13:56:40< celticminstrel> So it's irrelevant for any other aspect.
20160315 13:57:04< mattsc> Oh, I see.  Makes sense.
20160315 13:57:32< celticminstrel> From AiWML: "However, setting this id does not automatically set the id of the [leader_goal] facet."
20160315 13:57:32< mattsc> Was there anything specifically you wanted me to test while going through these?  Or just do what we said yesterday?
20160315 13:57:41< celticminstrel> That changes that.
20160315 13:57:50< mattsc> ok
20160315 13:57:57< celticminstrel> There was nothing specific, no.
20160315 13:58:17< mattsc> ok
20160315 13:58:19< celticminstrel> Except Micro AIs, but I think you were going to do that anyway.
20160315 13:58:53< mattsc> right, and I’ve had a first look at that, but only a few spot checks so far.
20160315 13:59:16-!- gfgtdf [~chatzilla@f054170174.adsl.alicedsl.de] has joined #wesnoth-dev
20160315 13:59:25< mattsc> I’ve seen a couple problems, but I think they are with the test scenarios, rather than with the MAIs themselves.
20160315 13:59:39< gfgtdf> Aginor: well mostlikeley it will be reinserted anyway when we update to newwer lua versions
20160315 14:00:00< mattsc> As in, a couple of the scenarios use engine definitions.  Should be easy to fix though, once I get to this for real.
20160315 14:00:20< gfgtdf> zookeeper: only teh first n units match the filter
20160315 14:00:53< celticminstrel> I think I mentioned that fixing that would probably be as simple as adding one line to the engine definition - something similar to what I did already in the Experimental AI, though it would be a bit simpler if they don't require the data field.
20160315 14:01:21< mattsc> Yes, you did.  I just haven’t gotten around to trying that yet.
20160315 14:01:51< mattsc> I first want to go through all the commits so that I know what I want to be looking for.
20160315 14:01:56< gfgtdf> zookeeper: often you only need one unit that macthes a certain filter, for example in some mainline tags like [recall] or [object] in such cases you coudl for example use limit=1 so that the game engine doesnt bother searching for mor units as soon as one is found
20160315 14:03:01< celticminstrel> In other words, it means exactly what he thought it means. :P
20160315 14:04:59-!- mjs-de [~mjs-de@x4db5c5ef.dyn.telefonica.de] has joined #wesnoth-dev
20160315 14:08:33< zookeeper> gfgtdf, okay, sounds good. it's a feature i've wanted for a long time
20160315 14:09:03< zookeeper> (i think)
20160315 14:13:32< gfgtdf> zookeeper: [recall] and [object] are currently only applied to one unit, unrelated to limit=<n>, but in other tags like [store_unit], [put_to_recall_list] or [move_unit] limit= can be used
20160315 14:13:59< gfgtdf> zookeeper: i wonder whetehr we shoudl change [object] and [recall] to be applied to all units matching the filter so that you'd hvae to use limit=1 to get the old baheiour
20160315 14:14:48< celticminstrel> Uh...
20160315 14:15:00< zookeeper> gfgtdf, only if enabled by some new added key
20160315 14:15:49< celticminstrel> ...oh, DW just hard-codes all the IDs, rather than something like [recall]upkeep=loyal
20160315 14:16:18-!- zombah [~zombah@2a02:28:3:1:214:4fff:fe47:5920] has quit [Quit: Leaving]
20160315 14:16:28< zookeeper> yeah i think objects and recalls usually pretty explicitly refer onto to one unit, but still that'd be an uncomfortable amount of potential breakage to introduce
20160315 14:16:34< zookeeper> s/onto/only
20160315 14:16:39-!- zombah [~zombah@2a02:28:3:1:214:4fff:fe47:5920] has joined #wesnoth-dev
20160315 14:18:24< celticminstrel> gfgtdf: Is it possible to explicitly specify something like "limit=unlimited"? If so, then you could make [recall] and [object] take all units matching the filter, but also add an implicit "limit=1" if no limit is otherwise defined.
20160315 14:19:42< gfgtdf> celticminstrel: well you cna use limit =999 which does basially the same
20160315 14:19:45< zookeeper> well, limit=9999 would be pretty e...
20160315 14:20:16< celticminstrel> I guess that's true.
20160315 14:20:31< celticminstrel> Looks nicer to say "all" or "unlimited" or something, though.
20160315 14:20:39< mattsc> celticminstrel: the point of allowing arbitrary aspects is so that you could write a custom AI with a simple way of transferring configuration parameters via WML?
20160315 14:21:44< celticminstrel> mattsc: It's not currently allowing arbitrary aspects, but yes, that is a possible future goal.
20160315 14:22:01-!- Appleman1234 [~Appleman1@KD106161140138.au-net.ne.jp] has quit [Ping timeout: 240 seconds]
20160315 14:22:04< celticminstrel> mattsc: At present, only the parsing stage that converts simple to full form allows multiple aspects.
20160315 14:22:17< celticminstrel> mattsc: Which means that in the event of new aspects being added, there is less code to update.
20160315 14:22:38< gfgtdf> celticminstrel: does c++ creeate a defautl copy ctor even if anoter constructor is already given ?
20160315 14:22:45< mattsc> celticminstrel: okay.  That’s just how I interpreted this  “This is no longer needed, since the new generalized simplified aspects syntax assumes any unknown tag or key is intended to be an aspect.”
20160315 14:22:59< zookeeper> celticminstrel, i dunno, there's a certain charm to a magic value like 999, one could argue that it's _more_ obvious what it does at a glance than a special value
20160315 14:23:24< Nobun> is there any documentation about what any module actually does (for example: src/game_launcher.cpp)
20160315 14:23:55< celticminstrel> mattsc: Right, it assumes it's intended to be an aspect and converts it to full [aspect][facet] form under that assumption; however, the AI engine will later complain about an unrecognized ID in that aspect. I had thought about actually allowing arbitrary aspects, but hadn't gotten to it.
20160315 14:24:06< Nobun> I'm trying to figure myself how to solve a bug I found, but I don't know where I have to touch the code
20160315 14:24:41< mattsc> celticminstrel: okay, no worries; I personally have no use for this feature, so that’s fine by me.
20160315 14:24:59< mattsc> Maybe I should say: no _current_ use
20160315 14:25:58< celticminstrel> gfgtdf: I think it creates a default copy ctor if and only if there is no constructor that is callable with a single argument of the class's type (either by value or by reference). So, cls(cls&), cls(cls), cls(cls&, int=3), all prevent generation of a copy constructor... I think. Not 100% sure. Other constructors only suppress the default no-args constructor.
20160315 14:26:35< gfgtdf> zookeeper: did you see this post https://forums.wesnoth.org/viewtopic.php?f=5&p=595007#p594954 ?
20160315 14:26:41< celticminstrel> zookeeper: Uhh... I don't agree... though at least in this case the magic value isn't really all that magical.
20160315 14:27:03< zookeeper> gfgtdf, yeah, will probably change the code
20160315 14:30:21-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: mattsc]
20160315 14:34:14-!- atarocch [~atarocch@37.176.181.181] has joined #wesnoth-dev
20160315 14:34:48-!- boucman_work [~boucman@247.37.0.109.rev.sfr.net] has joined #wesnoth-dev
20160315 14:51:21< vultraz> huh. the zoom slider appears to be broken
20160315 14:52:14-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20160315 14:52:59< vultraz> no matter
20160315 15:06:38-!- prkc [~prkc@gateway/vpn/privateinternetaccess/prkc] has quit [Ping timeout: 244 seconds]
20160315 15:14:30< Soliton> Nobun: generally no.
20160315 15:15:41< vultraz> "The formula language doesn't support variables so macros are used to emulate them."
20160315 15:15:44< vultraz> celticminstrel: is this true?
20160315 15:20:59-!- prkc [~prkc@108.61.123.79] has joined #wesnoth-dev
20160315 15:25:32-!- noy_ [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160315 15:28:10-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 268 seconds]
20160315 15:28:10-!- noy_ is now known as noy
20160315 15:28:44-!- noy [~Noy@wesnoth/developer/noy] has quit [Client Quit]
20160315 15:38:00-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160315 15:46:42< Nobun> thank, Soliton: I will try to use cat + grep to find something useful
20160315 15:48:05< Soliton> grep is quite capable of reading files, no cat needed.
20160315 15:49:36-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160315 15:49:38-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160315 15:49:44-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160315 15:49:57-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20160315 15:51:37< Nobun> eh eh... this is the magic thing that my own "search context string in file" script does... and it works ;)
20160315 15:59:49< shadowm> Aginor, vultraz: https://forums.wesnoth.org/viewtopic.php?p=595051#p595051  second paragraph might be of interest, especially to Aginor.
20160315 16:00:06< shadowm> Third paragraph, even.
20160315 16:06:22< Nobun> Soliton, shadowm: I cannot figure myself what file I should look. I'm trying to figure how to solve a bug related to the scrollbar that appears on startup when loading files and cache
20160315 16:06:47< shadowm> Nobun: That's not a scroll bar, that's a progress bar.
20160315 16:07:00< Nobun> yep, shadowm... sorry :)
20160315 16:07:20< shadowm> src/loadscreen.cpp
20160315 16:07:28< Nobun> thank shadowm 
20160315 16:07:38< shadowm> Yes, let's thank shadowm.
20160315 16:08:00< Nobun> :)
20160315 16:08:04< shadowm> (Should I point out the difference between 'thanks' as an expression and 'thank' as a verb?)
20160315 16:09:22< Nobun> yep... sounds good
20160315 16:09:47< Nobun> no problem on improving my (poor :P) english
20160315 16:12:00-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160315 16:14:57-!- irker575 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160315 16:18:44< vultraz> hmmmm
20160315 16:18:52< vultraz> yes, that is interesting
20160315 16:19:04< vultraz> actually, he brings up a very good point about dialog size handling
20160315 16:19:11< vultraz> we use absolute restrictions
20160315 16:19:17< vultraz> when we should really use relative ones
20160315 16:20:05< vultraz> I hadn't thought of that
20160315 16:21:34-!- Kwandulin [~Miranda@p200300760F7D73B10904786F05F474FD.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160315 16:24:37< vultraz> shadowm: oh, and look at in-game and see if it looks better
20160315 16:26:34-!- atarocch [~atarocch@37.176.181.181] has quit [Ping timeout: 260 seconds]
20160315 16:32:31-!- atarocch [~atarocch@37.176.181.181] has joined #wesnoth-dev
20160315 16:35:03-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160315 16:35:53-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160315 16:37:27-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160315 16:37:41< Nobun> sorry guys, but I'm unable to what I could touch to solve the bug xD
20160315 16:38:02< Nobun> so I will simply report it :P
20160315 16:38:17< Nobun> to find what *
20160315 16:39:34-!- Nobun [~nobun@host62-49-dynamic.12-79-r.retail.telecomitalia.it] has quit [Quit: Salve a tutti]
20160315 16:50:13-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20160315 17:02:36-!- boucman_work [~boucman@247.37.0.109.rev.sfr.net] has quit [Read error: Connection reset by peer]
20160315 17:05:49-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160315 17:42:40-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160315 18:01:54-!- irker474 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160315 18:01:54< irker474> wesnoth: Charles Dang wesnoth:master 8e8b4a38356e / images/buttons/button_menu/ (4 files): Added empty images for an imageless menu button https://github.com/wesnoth/wesnoth/commit/8e8b4a38356ecedc6c126d351a251d51c9d95a76
20160315 18:01:55< irker474> wesnoth: Charles Dang wesnoth:master a4c4f18a7c45 / data/themes/ (default.cfg editor.cfg): Don't show any images for buttons in menu bar https://github.com/wesnoth/wesnoth/commit/a4c4f18a7c45ef8c652fe9be72d166ab3e798309
20160315 18:02:45< vultraz> gui1 hackz
20160315 18:03:54< vultraz> Aginor: small thing, though, about that ^ in-game, but not in-editor, mousing over those gives a black background. probably a misplaces draw call somewhere, but I can't find it
20160315 18:04:00< vultraz> minor thing, tbh
20160315 18:04:43-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160315 18:06:53-!- littlehephy [827eff34@gateway/web/freenode/ip.130.126.255.52] has joined #wesnoth-dev
20160315 18:11:12< vultraz> gfgtdf: do you think making the side 1 leader show in the sidebar by default when you enter the game is a good idea?
20160315 18:11:26< vultraz> bc right now when you first enter there's nothing there
20160315 18:11:35< vultraz> until you mouse over/select a unit
20160315 18:11:38< gfgtdf> vultraz: you mean in mp ?
20160315 18:12:07< vultraz> no any game
20160315 18:13:02< gfgtdf> well i think selecting side 1 s leader is generally a bad idea, what we coudl discussis is showing the leader of the currently visible side
20160315 18:13:10< vultraz> hm 
20160315 18:13:14< vultraz> yes that sounds like a better idea
20160315 18:13:34< vultraz> just so there's something in the sidebar when you enter a game
20160315 18:13:46< gfgtdf> hmm whats bad about having nothign in thtqa sidebar ?
20160315 18:14:20< vultraz> it looks weird
20160315 18:14:27< irker474> wesnoth: Charles Dang wesnoth:master a8e5036c24f0 / data/themes/editor.cfg: editor: No panel background for top bar status areas https://github.com/wesnoth/wesnoth/commit/a8e5036c24f07506b2de24153af7c160ae6948b7
20160315 18:14:31< vultraz> also since I removed the panel backgrounds it's very empty
20160315 18:15:11< gfgtdf> why did you remove the backrounds ?
20160315 18:15:37< vultraz> experiment to make the UI cleaner 
20160315 18:15:42< vultraz> I think it works
20160315 18:15:49< vultraz> waiting to hear from shadowm/other people
20160315 18:29:38-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160315 18:29:38-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20160315 18:37:29< irker474> wesnoth: ln-zookeeper wesnoth:master d6be4711083e / data/core/images/terrain/ (57 files in 8 dirs): Cleaned up off-hex stray pixels https://github.com/wesnoth/wesnoth/commit/d6be4711083e79264f48f38dffdb981ff4f72bdd
20160315 18:44:00< vultraz> zookeeper: could you maybe fix the overlay of lightbeams
20160315 18:57:28< zookeeper> as soon as i take a look at it and try to find whatever is supposed to be wrong with it
20160315 18:58:12< gfgtdf> vultraz: also note that there are other was to get an 'empty' right side panel, specially when the currently sslected unit dies
20160315 19:06:50< vultraz> zookeeper: the edges overlap
20160315 19:11:51-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Ping timeout: 264 seconds]
20160315 19:12:17< zookeeper> what does that mean?
20160315 19:12:46< zookeeper> oh
20160315 19:13:00< zookeeper> if you put multiple next to each other, the part in between gets overly bright
20160315 19:13:14< zookeeper> yeah i guess that's a bit annoying
20160315 19:13:42< irker474> wesnoth: Charles Dang wesnoth:master 0aa8900dce34 / src/construct_dialog.cpp: Removed an extra CVideo::flip() call from GUI1 https://github.com/wesnoth/wesnoth/commit/0aa8900dce348d5ca590250351b2ffafbc0befab
20160315 19:14:17-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20160315 19:15:11< irker474> wesnoth: ln-zookeeper wesnoth:master 3223ea29b1e4 / data/core/images/terrain/ (50 files in 3 dirs): Cleaned up off-hex stray pixels https://github.com/wesnoth/wesnoth/commit/3223ea29b1e458bbbc2356aee9e193e2f8b35778
20160315 19:15:28< vultraz> didn't...you just do that
20160315 19:18:20< zookeeper> different files
20160315 19:18:25< vultraz> ah
20160315 19:18:35< zookeeper> i've done maybe... less than half now? :p
20160315 19:19:10< vultraz> that's a lot of pixels
20160315 19:19:41< zookeeper> not all in fabi's output need fixing, and arguably most of the ones i've done so far didn't (because they've applied with single-hex rules), but i dislike that sort of crap anyway
20160315 19:19:49-!- atarocch [~atarocch@37.176.181.181] has quit [Ping timeout: 260 seconds]
20160315 19:21:42< vultraz> oh *@#&$()*@#()
20160315 19:22:09< vultraz> I hate when you test a commit, and then commit it, and then as soon as you commit it, find a case that you missed :|
20160315 19:22:36< irker474> wesnoth: Charles Dang wesnoth:master 819c352e7c32 / src/construct_dialog.cpp: Revert "Removed an extra CVideo::flip() call from GUI1" https://github.com/wesnoth/wesnoth/commit/819c352e7c320c882bbf42b127dc17e6774df28b
20160315 19:22:39< vultraz> I'm not dealing with this now
20160315 19:24:54-!- louis94 [~~louis94@91.178.242.179] has joined #wesnoth-dev
20160315 19:30:49-!- Kwandulin [~Miranda@p200300760F7D73B10904786F05F474FD.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160315 19:36:01-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160315 19:52:16-!- littlehephy [827eff34@gateway/web/freenode/ip.130.126.255.52] has quit [Ping timeout: 252 seconds]
20160315 19:55:02-!- louis94 [~~louis94@91.178.242.179] has quit [Ping timeout: 260 seconds]
20160315 19:55:48< celticminstrel> [Mar 15@11:15:41am] vultraz: "The formula language doesn't support variables so macros are used to emulate them."
20160315 19:55:49< celticminstrel> No.
20160315 19:55:51-!- SpoOkyMagician [~chatzilla@2607:fcc8:be59:b00:d5fb:e8aa:3e47:6b94] has joined #wesnoth-dev
20160315 19:57:30-!- atarocch [~atarocch@37.177.98.117] has joined #wesnoth-dev
20160315 19:59:50-!- Samual [~Samual@xonotic/core-team/Samual] has quit [Ping timeout: 250 seconds]
20160315 20:01:38-!- atarocch [~atarocch@37.177.98.117] has quit [Ping timeout: 244 seconds]
20160315 20:06:30-!- Samual [~Samual@xonotic/core-team/Samual] has joined #wesnoth-dev
20160315 20:10:10< vultraz> huh
20160315 20:10:23< vultraz> well it seems [message]'s dialog thinks that's the case
20160315 20:10:30< celticminstrel> Point me to it?
20160315 20:12:19< vultraz> gui/default/window/wml_message.cfg:18
20160315 20:16:44< vultraz> this seems inelegant 
20160315 20:22:35< celticminstrel> vultraz: I'm looking at it now.
20160315 20:38:38< celticminstrel> vultraz: I think it would be simple to add [vars] to [resolution] to set additional formula variables.
20160315 20:50:03< vultraz> wel
20160315 20:50:11< vultraz> I'm thinking of dropping [resolution]
20160315 20:50:28-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20160315 20:50:53< celticminstrel> vultraz: Why? Is it that simple to excise?
20160315 20:51:05< celticminstrel> vultraz: If it's a headache to drop it, don't bother.
20160315 20:51:26< celticminstrel> vultraz: Also I think it's not bad to keep it - it does serve a useful purpose as far as I know, even if it's not currently being used.
20160315 20:53:38< gfgtdf> do both window and widget definition support [resolution] ?
20160315 20:54:01< celticminstrel> Think so?
20160315 20:55:06< gfgtdf> well i think its defeinteley useful to be able to amke different layouts depnding on window soze, liek for example ueing a tabbar for low windows sizes and putting the panels next to each other on large screens
20160315 20:55:33< celticminstrel> This is weird. The __GUI_IMAGE_DISPLAYED_X macro takes an argument, but only X is passed.
20160315 20:56:11< celticminstrel> Which implies to me that, if image_original_width > 100, x ends up being null.
20160315 20:56:29< celticminstrel> I believe converting null to a number produces 0.
20160315 20:56:41< celticminstrel> So I guess that's not terrible, but...
20160315 21:00:01< celticminstrel> I have [vars] added here (untested as of yet), but am trying to figure out just what to do with the actual dialog definition.
20160315 21:01:41< celticminstrel> The problem is that I can't think of a way for [vars] to have access to any of the variables that the other formulas can see.
20160315 21:02:08< celticminstrel> functions= also doesn't have access to them, but I think function formulas are re-evaluated on each call, so that shouldn't be a problem.
20160315 21:02:20< celticminstrel> Perhaps __GUI_IMAGE_WIDTH could be implemented as a function.
20160315 21:13:13< celticminstrel> vultraz: So uh, I've managed to combine all those macros into one mega-macro.
20160315 21:13:24< vultraz> good, good
20160315 21:14:01< celticminstrel> Or rather, ___GUI_IMAGE_WIDTH no longer includes any macros.
20160315 21:14:23< celticminstrel> I'm going to post it, let me know if you think it's more understandable like this.
20160315 21:14:51< gfgtdf> did we recently disczuss about changing the config implementatio to use a hash map ?
20160315 21:16:20< mattsc> celticminstrel: ooo, you even fixed the compile time warning!  (I’m “back in business” concerning PR 615)
20160315 21:16:32< celticminstrel> vultraz: https://wiki.wesnoth.org/User:Celtic_Minstrel/FormulaTests
20160315 21:16:49< vultraz> 'where'?
20160315 21:17:02< celticminstrel> That defines variables within a formula.
20160315 21:17:31< celticminstrel> So essentially, the part after where is executed first.
20160315 21:17:44< celticminstrel> (Subject to grouping parentheses.)
20160315 21:17:53< vultraz> It's a little confusing
20160315 21:18:07< vultraz> but that might just be my lack of WFL knowledge
20160315 21:18:22< celticminstrel> Is it less confusing than before?
20160315 21:18:44< vultraz> it's always confusing to me :P
20160315 21:18:53< vultraz> but I'd say yes
20160315 21:19:23< celticminstrel> The comments interpolated into the source are literally copied from the comments on individual submacros, so it's possible some may be no longer relevant.
20160315 21:20:00< celticminstrel> That replaces the ___GUI_IMAGE_WIDTH_MACRO (three initial underscores) and everything prior to it. Though I haven't tested it runs.
20160315 21:20:21< celticminstrel> (Well, it goes inside said macro, with an extra set of enclosing parentheses.)
20160315 21:20:45< celticminstrel> (Oh wait, those aren't needed, right. Because that macro is then included somewhere else.)
20160315 21:21:20< celticminstrel> Three lines after "best_width = " there's an extraneous set of parentheses around the else clause.
20160315 21:22:31< celticminstrel> I have a feeling that the algorithm was updated and the comments describing it not.
20160315 21:22:38< vultraz> random thought: I wonder if WFL could be extended to call WML
20160315 21:22:45< celticminstrel> Uhh, what?
20160315 21:23:34< vultraz> I've been thinking about FFL since yesterday\
20160315 21:23:39< celticminstrel> FFL?
20160315 21:23:43< celticminstrel> WFL?
20160315 21:23:53< vultraz> whatever we're calling this
20160315 21:23:55< vultraz> WSL?
20160315 21:23:58< celticminstrel> WFL is a language where everything is immutable.
20160315 21:24:16< vultraz> Frogatto Formula Language or whatever they called it
20160315 21:24:21< celticminstrel> Oh.
20160315 21:24:31< celticminstrel> I haven't looked at it.
20160315 21:24:32< vultraz> it derived from the same source as wesnoth's
20160315 21:24:35< vultraz> (Silvertree)
20160315 21:24:43< celticminstrel> Silvertree?
20160315 21:24:58< vultraz> ask shadowm 
20160315 21:25:07< vultraz> some derivitive project
20160315 21:25:11< celticminstrel> Okay, so, how does this lead to "WFL calls WML"?
20160315 21:25:22< vultraz> well, I was thinking two things:
20160315 21:25:53< vultraz> * about how FFL is used entirely for Frogatto/Anura's scripting and data needs
20160315 21:26:20< vultraz> * how our implementation is basically restricted to data manipulation to get values to pass to other functions
20160315 21:26:52< celticminstrel> What do you mean with the second point>
20160315 21:26:54< celticminstrel> ^?
20160315 21:27:06-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20160315 21:27:18< vultraz> well, look at the formulas we're working with
20160315 21:27:30< vultraz> it's all to get the value of the dimension keys in [image]
20160315 21:27:44< celticminstrel> BTW, are you going to commit that new formula or should I do it a bit later?
20160315 21:27:57< vultraz> you should
20160315 21:28:04< celticminstrel> 'kay, will do.
20160315 21:28:08< celticminstrel> So back to topic.
20160315 21:28:30-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160315 21:28:34< vultraz> I'm only now realized the formula engine is a lot more feature-full than I had thought
20160315 21:28:50< vultraz> so in essence, we have *four* code layers
20160315 21:28:58< celticminstrel> Eh?
20160315 21:29:11< vultraz> C++, Lua, WML, and WFL
20160315 21:29:15< celticminstrel> Oh, right.
20160315 21:29:32< celticminstrel> Though WML is more a data layer than code (ActionWML notwithstanding).
20160315 21:30:04< vultraz> For all our efforts to add features to our Lua layer of WFL layer, everything is very much seperated
20160315 21:30:09< vultraz> s/of/or
20160315 21:30:50< vultraz> so for devs, we work with all four layers to implement features
20160315 21:30:57< vultraz> for umc authors, they have to learn all three
20160315 21:31:15< celticminstrel> Not quite true, they can get away with just WML for most situations.
20160315 21:31:35< vultraz> so, actually, ignore my suggestion about calling wml from wfl
20160315 21:31:45< vultraz> since it actually goes against my point
20160315 21:31:52< vultraz> that being that we have a very fragmented system here
20160315 21:31:54< celticminstrel> I don't really know what it meant for WFL to call WML anyway.
20160315 21:32:40< vultraz> yeah, bad phrasing
20160315 21:32:59< celticminstrel> I'm still curious what it meant even if you've retracted the idea.
20160315 21:33:15< vultraz> but I think it would be cool, if, say, we could do something like this:
20160315 21:35:01< vultraz> draw(if(x > 0, 1, 2), 1, image["foo/foo.png]) 
20160315 21:35:08< vultraz> just random psuedosyntaxc
20160315 21:35:12< vultraz> pseudosyntax
20160315 21:35:31< celticminstrel> Note that formula strings are always 'single quoted'.
20160315 21:35:47< celticminstrel> So uh, what exactly does that do?
20160315 21:35:56< celticminstrel> Draw an image on a specific hex?
20160315 21:36:01< vultraz> draw an image at a, b
20160315 21:36:17< vultraz> tbh, I need to look at FFL more
20160315 21:36:32< vultraz> it revolves around objects, I think..
20160315 21:38:13< vultraz> much as I like the power of WFL, I think it's being wasted as a WML attribute formula calculator 
20160315 21:38:23< vultraz> even lua can be called from wml with [lua] 
20160315 21:38:43< celticminstrel> WFL can be called anywhere where variable substitution is supported.
20160315 21:39:28< vultraz> I think we need to eventually come up with some unifying global syntax here
20160315 21:39:35< celticminstrel> Some what now?
20160315 21:39:46< vultraz> not syntax, sorry
20160315 21:40:07< celticminstrel> Support for WFL could be extended to more places in WML.
20160315 21:40:09< vultraz> whatever you call what FFL does
20160315 21:40:13< celticminstrel> For example, within a [unit_type].
20160315 21:40:16< vultraz> the problem is exactly that
20160315 21:40:21< vultraz> we're just tacking more things on
20160315 21:40:40< vultraz> WFL was tacked on to WML
20160315 21:40:45< vultraz> Lua was tacked  on to WML
20160315 21:41:17< vultraz> all designed around the config/wml paradigm
20160315 21:41:36< vultraz> I'm not sure how viable this system is in the long run
20160315 21:41:48< celticminstrel> It seems to be working well enough so far.
20160315 21:41:52-!- louis94 [~~louis94@91.178.242.179] has joined #wesnoth-dev
20160315 21:41:58< celticminstrel> I won't say it's optimal or anything.
20160315 21:45:00< vultraz> this is a ffl file: https://github.com/frogatto/frogatto/blob/master/data/objects/playable/frogatto_playable.cfg
20160315 21:45:42< vultraz> it's not the most nicely formatted, but for us, it would take at lest half a doze files and multiple abstraction layers just to define all of that
20160315 21:45:49< celticminstrel> Looks like JSON at a glance.
20160315 21:46:00< celticminstrel> Although with unquoted keys.
20160315 21:46:31< celticminstrel> I see some actual formula-language-looking stuff too.
20160315 21:46:33-!- louis94 [~~louis94@91.178.242.179] has quit [Ping timeout: 240 seconds]
20160315 21:47:04< celticminstrel> Seems more strongly-type than WFL?
20160315 21:47:08< celticminstrel> ^typed
20160315 21:47:08< vultraz> it's frogatto's 'object'
20160315 21:47:13< vultraz> not a scripting language
20160315 21:47:21< vultraz> which I find quite interesting
20160315 21:47:28< celticminstrel> I'm looking under "properties" where there's a "def".
20160315 21:48:19< celticminstrel> It does look very similar to WFL though.
20160315 21:48:45< vultraz> random fact: apparently, anura's (the engine)'s UI system consist entirely of objects. they don't use widgets.
20160315 21:49:09< celticminstrel> That sounds like redefining terms.
20160315 21:49:21< vultraz> ask then the details
20160315 21:49:25< vultraz> them
20160315 21:49:30< vultraz> I do know it's somehow better
20160315 21:50:30< vultraz> but imagine if we somehow had the ability to say, in a gui2 widget, put something like
20160315 21:50:34< vultraz> on_draw(do_thing)
20160315 21:50:44< vultraz> and it would be piped to the draw cycle 
20160315 21:50:52< celticminstrel> Regarding the FFL language itself, it basically seems like WFL with more type-checks.
20160315 21:51:01< vultraz> instead of filling out the wml config for [draw], passing that through a huge number of c++ layers..
20160315 21:51:08< celticminstrel> And possibly first-class functions.
20160315 21:51:10< vultraz> which is then called by the rendering engine..
20160315 21:51:17< celticminstrel> First-class functions would be good, I think.
20160315 21:51:29< celticminstrel> But I don't know how much use that would really see.
20160315 21:51:44< vultraz> we're continuing to think in the WML paradigm
20160315 21:51:55< vultraz> I think wesnoth needs a new paradigm
20160315 21:52:12< vultraz> I mean, this whole formula language was conceived as a replacement to WML, I think
20160315 21:52:24< celticminstrel> I don't see how it could replace WML.
20160315 21:52:33< vultraz> or at least to allow what WML's design did not
20160315 21:52:36< vultraz> speak to sirp
20160315 21:52:40< vultraz> or Jetrel
20160315 21:52:50< vultraz> they would know
20160315 21:53:31< celticminstrel> You seem to be talking about GUI2 more than general WML.
20160315 21:53:45< celticminstrel> Oh, FFL has + for string concatenation.
20160315 21:54:14< vultraz> gui2 was fully designed around wml as its data type
20160315 21:54:51< vultraz> almost every single function of the game is based on the config class's design
20160315 21:55:19< celticminstrel> I dunno about that...
20160315 21:56:57-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20160315 21:58:14< vultraz> weall, speak to Jetrel
20160315 21:58:17< vultraz> he always has insight
20160315 21:59:09< celticminstrel> Is there documentation for this FFL thing?
20160315 22:03:42< celticminstrel> Because I'm looking at strange syntaxes without really knowing what they mean.
20160315 22:18:52-!- gfgtdf [~chatzilla@f054170174.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 45.0/20160303134406]]
20160315 22:19:21-!- gfgtdf [~chatzilla@f054170174.adsl.alicedsl.de] has joined #wesnoth-dev
20160315 22:19:41< gfgtdf> vultraz: you know whether there is a wml to get a rect of the the current map into a variable?
20160315 22:20:59-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20160315 22:21:22< celticminstrel> I think I've seen such a thing somewhere...
20160315 22:21:23-!- louis94 [~~louis94@91.178.242.179] has joined #wesnoth-dev
20160315 22:21:39< fabi> hi
20160315 22:21:46< celticminstrel> Hi.
20160315 22:23:32< iceiceice> vultraz, fwiw, i read what you wrote before about WFL
20160315 22:23:50< iceiceice> (1) I think WFL is actually deeply integrated into Gui2
20160315 22:24:11< iceiceice> for reasons that I don't fully understand, the gui2 canvas object uses WFL formulas to keep track of parameters of widgets
20160315 22:24:26< iceiceice> (2) I don't think that WFL should really be compared to FFL, they are really totally different
20160315 22:24:28< vultraz> gfgtdf: https://wiki.wesnoth.org/InternalActionsWML#.5Bstore_map_dimensions.5D
20160315 22:24:40< iceiceice> WFL is like some extremely primitive version of FFL with very poor error reporting,
20160315 22:24:49< celticminstrel> iceiceice: From Looking at the sample he posted, I don't think they are totally different.
20160315 22:24:57< celticminstrel> ^looking
20160315 22:25:03< iceiceice> they are like 6 years of development apart at least
20160315 22:25:13< shadowm> Divergent evolution.
20160315 22:25:22< vultraz> That's what I said
20160315 22:25:26< celticminstrel> They certainly are quite a bit different, but they seem to share several things, such as "where" clauses.
20160315 22:25:34 * celticminstrel nods to shadowm.
20160315 22:25:38< iceiceice> celticminstrel, its like saying "windows 95 is not totally different from windows xp"
20160315 22:26:07< celticminstrel> iceiceice: Nevertheless, different or not, it wouldn't be amiss to take an idea or two from FFL for WFL, right?
20160315 22:26:20< iceiceice> sure i mean if you want
20160315 22:26:20< gfgtdf> vultraz: no i meant somethign thats stores that map data of a certain rectable-shaped part of the map.
20160315 22:26:28 * celticminstrel only got one idea while reading through the sample; might get a few more from reading documentation.
20160315 22:26:36< iceiceice> if it were all up to me WFL would just be remoed
20160315 22:26:39< iceiceice> *removed
20160315 22:26:53< celticminstrel> That would break everything.
20160315 22:26:59< iceiceice> yeah but it would be worth it
20160315 22:27:06< iceiceice> you could use lua or wml in those caes
20160315 22:27:06< iceiceice> *cases
20160315 22:27:20< iceiceice> it's not actually used in any essential way
20160315 22:27:25< iceiceice> formula AI is dead
20160315 22:27:27< celticminstrel> $(...)
20160315 22:27:32< iceiceice> the gui2 stuff could be changed without much difficulty
20160315 22:27:32< shadowm> ^
20160315 22:27:39< iceiceice> the $(...) syntax is pretty horrible anyways
20160315 22:27:45< shadowm> I greatly appreciate the ability to write inline formulas instead of using verbose WML or having to switch into a Lua context.
20160315 22:27:52< gfgtdf> celticminstrel: we coudl also change the $() syntax to do lua
20160315 22:27:53< iceiceice> you could write inline lua
20160315 22:28:04< celticminstrel> Horrible or not, I think $(a + b) or similar is probably quite common
20160315 22:28:05< iceiceice> then you will at least get reaosnable erorr messages
20160315 22:28:32< gfgtdf> celticminstrel: in most cases people use simepl things liek 6 +7  anyways which have exactly the same syntax in lua
20160315 22:28:33< shadowm> People who use WML are used to unreasonable error messages anyway. :P
20160315 22:28:35< celticminstrel> iceiceice: Part of the reason you don't get reasonable error messages is that people write $($a + $b).
20160315 22:28:47< iceiceice> it's really a pretty huge amount of work to just develop a modern garbage collected programming language from the ground up
20160315 22:28:51< iceiceice> which is what FFL is
20160315 22:28:58< iceiceice> and why they are still working on it after X many years
20160315 22:29:02< celticminstrel> So if a and b undefined, the formula tries to compiled " + ".
20160315 22:29:10< celticminstrel> Which is, obviously, not a valid formula.
20160315 22:29:41< iceiceice> i dont quite get your point
20160315 22:29:57< celticminstrel> There are two distinct things interacting here.
20160315 22:29:59< iceiceice> my criticism was more general than any specific syntax mistake
20160315 22:30:25< celticminstrel> When experimenting with WFL from the AI formula prompt, I don't recall any unreasonable error messages.
20160315 22:30:43< iceiceice> usually the issue is, lack of error messages
20160315 22:30:44< celticminstrel> Though they could be made more clear, as I recall they did at least contain sufficient information to realize what's wrong.
20160315 22:30:50< celticminstrel> iceiceice: Example?
20160315 22:30:51< iceiceice> when it just decides to do something stupid instead of telling you what is wrong
20160315 22:30:58< iceiceice> look just play around with it
20160315 22:31:05< iceiceice> i'm just giving you my 2 cents
20160315 22:31:26< celticminstrel> I think I can imagine a few things.
20160315 22:31:46< celticminstrel> Like how invalid operations often return null rather than signalling an error.
20160315 22:33:15< fabi> Why don't use MoonScript for everything ;-P
20160315 22:33:29< celticminstrel> Because that would be pointless.
20160315 22:33:43< fabi> aha
20160315 22:36:19< gfgtdf> are there advantages to use forlua over lua ?
20160315 22:36:38< celticminstrel> It's a different paradigm. I don't know if either has advantages.
20160315 22:37:14< celticminstrel> Formula is functional programming, so it has things like filter and reduce that Lua lacks; of course, Lua can do the same thing with loops though.
20160315 22:37:32< celticminstrel> WFL also has (mostly) immutable values.
20160315 22:38:00< celticminstrel> Which just means you return a new filtered list rather than removing elements from the list (for example).
20160315 22:38:18< gfgtdf> celticminstrel: well, in lua you coudl most likeley easily write a reduce function.
20160315 22:38:27< gfgtdf> celticminstrel: and you can also write it in a way that returns a new list
20160315 22:38:30< celticminstrel> Probably, yeah.
20160315 22:38:54< celticminstrel> Reduce doesn't return a list, gfgtdf.
20160315 22:39:00< celticminstrel> It returns a value.
20160315 22:39:04< gfgtdf> celticminstrel: i meant filter
20160315 22:39:08< celticminstrel> 'kay
20160315 22:39:42< celticminstrel> WFL doesn't have things like while-loops or for-loops.
20160315 22:40:01< celticminstrel> It doesn't need them, really.
20160315 22:40:15< gfgtdf> celticminstrel: and reduce is most likeley as easy as function reduce (list, f) local res = nil; for i,v in ipairs(list) res = f(res, v) end return res rend
20160315 22:40:54< gfgtdf> celticminstrel: for me taht main disadvantage of forula is simpyl taht it requires wml authors to learn a new (third) lanugage besides wml and lua
20160315 22:41:01< celticminstrel> WFL does have a minor advantage of not using {}  or " so as not to confuse the WML preprocessor.
20160315 22:41:27< celticminstrel> Which means you can easily macro-substitute into a formula.
20160315 22:41:38< celticminstrel> You can't really macro-substitute into Lua.
20160315 22:41:59< celticminstrel> (I tried it once by mixing <<>> and "", with confusing results. Might try again later to figure out what was really happening.)
20160315 22:42:33< shadowm> The social movement for the global abolishing of WML preprocessor capabilities would consider that a massive disadvantage rather than an advantage.
20160315 22:42:59 * celticminstrel shrugs.
20160315 22:43:21< celticminstrel> Abolishing the WML preprocessor is a noble goal, but one that's unlikely to happen before, oh I dunno, 1.18.
20160315 22:43:37< zookeeper> i wish we had as much enthusiasm for content as for the engine.
20160315 22:43:46< gfgtdf> celticminstrel: i actualyl never needed to do macro substutution in lua and negher i had problems with << >>
20160315 22:44:06< celticminstrel> gfgtdf: The problem was with stuff like <<>> + "" + <<>>
20160315 22:44:18< shadowm> zookeeper: Yeah, me too. If only we had someone in the dev team able to work on mainlining new and refreshing content, for example.
20160315 22:44:35< celticminstrel> So much of Wesnoth's main config leans heavily on the preprocessor that to rip it out is such a huge task as to be untenable.
20160315 22:45:14< gfgtdf> celticminstrel: hmm whem exactly did you use << >> + << >> ?
20160315 22:45:21< gfgtdf> celticminstrel: im just curious
20160315 22:45:29< celticminstrel> gfgtdf: Because I wanted to macro-substitute into the Lua code.
20160315 22:45:45< gfgtdf> :O why that ?
20160315 22:45:49< celticminstrel> So I had something like <<lua code {>> + "{macro}" + <<} more lua code>>
20160315 22:45:58< zookeeper> shadowm, if we did, then they'd be so swamped with old content that they'd never have time to work on _new_ content
20160315 22:46:07< celticminstrel> I ended up setting a WML variable from the macro instead.
20160315 22:46:34< gfgtdf> celticminstrel: sounds liek if good idea to do it
20160315 22:46:43< gfgtdf> celticminstrel: you coudl also use [args] in lua
20160315 22:46:46< celticminstrel> Which in retrospect is probably a better way of doing it in the first place.
20160315 22:47:12< shadowm> zookeeper: So you are actually expecting developers to play the game and make content for it? Gadzooks!
20160315 22:47:38< shadowm> I wasn't suggesting for them to do any work beyond moving files around.
20160315 22:48:31< celticminstrel> gfgtdf: I couldn't use [args] since it was AI engine code, although... there was a similar thing to [args] that I added to the AI engine, not sure if that would've worked.
20160315 22:49:50< zookeeper> shadowm, frankly, i'm not sure on which level of sarcasm and insinuation we're operating at currently, so i can't really respond :p
20160315 22:51:21< gfgtdf> i think i can close pr278 since i mostly fixed the underlying when implemeting side_name in [side].
20160315 22:52:02-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160315 22:57:53< irker474> wesnoth: ln-zookeeper wesnoth:master 091370494a25 / data/core/images/terrain/flat/ (24 files): Removed feathering and fixed visual mismatch between convex and concave https://github.com/wesnoth/wesnoth/commit/091370494a2567094e6aa5fab017ba8c0cffdff8
20160315 23:00:08< zookeeper> (i think that introduces a little corner glitch in a few places, but i'll fix the rules accordingly tomorrow)
20160315 23:08:27-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 276 seconds]
20160315 23:12:52-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20160315 23:18:29-!- mjs-de [~mjs-de@x4db5c5ef.dyn.telefonica.de] has quit [Remote host closed the connection]
20160315 23:27:50-!- Appleman1234 [~Appleman1@KD106161140181.au-net.ne.jp] has joined #wesnoth-dev
20160315 23:30:23< fabi> are Wesnoth's hex grids regular?
20160315 23:30:40< fabi> Meaning all sides are equal in length?
20160315 23:30:53< celticminstrel> That's not what regular means.
20160315 23:31:01< celticminstrel> I don't think they are regular.
20160315 23:31:10< celticminstrel> But they might be equilateral.
20160315 23:31:19-!- SpoOkyMagician [~chatzilla@2607:fcc8:be59:b00:d5fb:e8aa:3e47:6b94] has quit [Read error: Connection reset by peer]
20160315 23:33:12< fabi> "A regular hexagon is defined as a hexagon that is both equilateral and equiangular"
20160315 23:34:50< fabi> zookeeper might know that.
20160315 23:36:22< celticminstrel> Yup.
20160315 23:36:29< celticminstrel> That is indeed what regular means.
20160315 23:36:44< celticminstrel> Where's the hex outline image again?
20160315 23:38:44 * celticminstrel uses a protractor on the screen.
20160315 23:38:51< celticminstrel> As I thought, they are not regular.
20160315 23:39:00< celticminstrel> The angles at the left and right vertices are a bit larger.
20160315 23:40:04< fabi> My poor mind asumed that equilateral would determine the shape of a hex field exactly.
20160315 23:41:00< fabi> But yeah, one can strech them without destroying the equilaterality.
20160315 23:41:39< celticminstrel> Indeed. This is true for any polygon of more than 3 sides.
20160315 23:42:11< fabi> Yes, some geometry knowledge wouldn't hurd.
20160315 23:42:16< celticminstrel> I don't feel like taking a ruler to the screen as well, but it does look like they're equilateral.
20160315 23:42:33< celticminstrel> It's at images/misc/hover-hex.png if you want to check for yourself.
20160315 23:42:47< fabi> Thank you :-)
20160315 23:43:36 * celticminstrel does calculations and finds that the bottom and top angles are probablt 115° (since the side angles seem to be 130°).
20160315 23:43:39< celticminstrel> ^probably
20160315 23:43:51< celticminstrel> Not that anyone cares about that probably. >_>
20160315 23:44:53< fabi> Hmmm, I need the exact dimensions orelse the images won't be compatible....
20160315 23:45:05< fabi> So I care about it.
20160315 23:47:19< fabi> I wonder why Dave did not try to compensate for the maps not using a metric that corresponds to human perception.
20160315 23:47:44< fabi> A 20x20 map is not as high as it is width.
20160315 23:49:19< celticminstrel> I have no idea what you're talking about.
20160315 23:49:27< celticminstrel> Also you want to say "wide" not "width" there.
20160315 23:49:37< fabi> Well 
20160315 23:49:51< fabi> that is easy to get. Open a square map.
20160315 23:50:02< fabi> Look at the minimap.
20160315 23:50:10< fabi> It is filled completely.
20160315 23:50:28< fabi> I designed the minimap so that square maps fit exactly.
20160315 23:50:41< fabi> But the minimap is not a square.
20160315 23:51:26< fabi> s/square/quadrat
20160315 23:51:50< celticminstrel> Eh?
20160315 23:52:08< fabi> quadratic maps
20160315 23:52:11< fabi> 20x20
20160315 23:52:18< celticminstrel> I don't think you mean quadratic here.
20160315 23:52:24< fabi> x*x
20160315 23:52:30< fabi> Yes. but you get what I mean.
20160315 23:52:34< celticminstrel> So square, right?
20160315 23:52:36< fabi> x == y
20160315 23:52:38< fabi> yes
20160315 23:52:50< fabi> or better w == h
20160315 23:54:41< fabi> The hex field geometry of Wesmere does not produce a metric.
20160315 23:55:12< celticminstrel> I'm pretty sure it does.
20160315 23:55:14< fabi> Well, not a metric we usually assume.
20160315 23:55:54< celticminstrel> I'm still not convinced the Euclidean metric is useful, specifically.
20160315 23:56:07< celticminstrel> But there's an obvious metric nevertheless.
20160315 23:58:11< fabi> Anyway, the thing is done. Changing it is not possible anymore. So no reason to discuss it.
20160315 23:58:44< celticminstrel> That applies to a lot of things that get discussed anyway, at great length.
20160315 23:59:26< fabi> Indeed. 
20160315 23:59:39< celticminstrel> Often by yourself, even. :P
--- Log closed Wed Mar 16 00:00:26 2016