--- Log opened Wed Mar 21 00:00:08 2018 20180321 00:00:13<+discordbot1> So, we know we want this class to draw the gamemap (the terrains and such), and the things on it, such as units 20180321 00:00:24< gfgtdf> this is basicially display 20180321 00:00:28<+discordbot1> yes 20180321 00:00:33< gfgtdf> well display except theme stuff 20180321 00:00:36<+discordbot1> now, what EVENTS should it handle 20180321 00:00:52< celticminstrel> I don't think the teams/sides are logically part of the map, unless it's required in order to show village flags. 20180321 00:01:08< celticminstrel> Events? Mouse up, mouse down, mouse move, key down, key up. :P 20180321 00:01:15<+discordbot1> ....no no no 20180321 00:01:25<+discordbot1> I mean, for example, the map needs to scroll 20180321 00:01:25< celticminstrel> I can go into more detail too. 20180321 00:01:43<+discordbot1> you need to be able to select units 20180321 00:02:03< celticminstrel> Things that happen on mouse up/down include select unit, move unit, deselect unit, open context menu (have I missed one)? 20180321 00:02:22<+discordbot1> I'm attempting to figure out how I can split the map event handling from hotkey handling 20180321 00:02:25< celticminstrel> On mouse move, the hex under the mouse needs to be hilited. 20180321 00:02:51< celticminstrel> I don't think you want to split them exactly? You do want the keys to be rebindable after all. 20180321 00:03:05<+discordbot1> no, I mean 20180321 00:03:07<+discordbot1> ugh 20180321 00:03:17< celticminstrel> Mouse move and key events can both trigger scrolling, which becomes another needed thing. 20180321 00:03:25< octalot> model, view, controller? 20180321 00:03:26< gfgtdf> note that mouseclick in gamemap (&minimap?) is currently also implemented via hotkey sytem. 20180321 00:03:35<+discordbot1> ...what the fuck 20180321 00:03:37<+discordbot1> 😐 20180321 00:03:41< celticminstrel> Another thing that could qualify as a map event is "view moves" 20180321 00:03:54< gfgtdf> leftmouseclick that is, iirc? 20180321 00:03:59< gfgtdf> i'm actualyl not sure 20180321 00:04:03< celticminstrel> Yeah, don't go removing the mouse click hotkey. 20180321 00:04:27<+discordbot1> fucking hotkey system... 20180321 00:05:11< celticminstrel> There are like ten different keys for "view moves" and most of them are hardcoded (not rebindable), but IIRC the rebindable one is Ctrl+V? 20180321 00:05:28< celticminstrel> (The non-rebindable ones are digit keys.) 20180321 00:06:03<+discordbot1> right now, it seems a lot of events are implemente din the controller classes 20180321 00:06:18<+discordbot1> stuff like process_keydown_event 20180321 00:06:32< celticminstrel> Well, that makes sense, because they're the controller classes. 20180321 00:06:38< celticminstrel> That's what controller classes are for. 20180321 00:07:16< celticminstrel> (IOW what octalot said is something that already exists in 1.14.) 20180321 00:07:31<+discordbot1> I'm not touching MVC 20180321 00:07:44< celticminstrel> (I think game_board is the model, display is the view, and the controllers are obvious.) 20180321 00:07:57< celticminstrel> TBH I think this is a place where MVC should be used. 20180321 00:08:02<+discordbot1> and it makes me want to jump off a cliff 20180321 00:08:23< celticminstrel> Because we have more than one possible model, and IIRC more than one possible controller. Not quite sure, maybe even more than one possible view? 20180321 00:08:24<+discordbot1> look, WHAT I'M THINKING 20180321 00:08:34<+discordbot1> can you let me finish 😐 20180321 00:08:35< gfgtdf> i'd say game_state the model actually 20180321 00:08:39< celticminstrel> And the strength of MVC is that you can swap out any of the three components. 20180321 00:08:43< celticminstrel> gfgtdf: Okay, sure. 20180321 00:09:19< gfgtdf> in particualr since it contains tod_manager_ which had a direct influence on the game map display 20180321 00:09:42<+discordbot1> I'm pondering, what if instead of mouse_handler_base::mouse_motion_event, we just add handling for SDL_MOUSEMOTION in display::handle_event? 20180321 00:09:48<+discordbot1> and similar for other events 20180321 00:09:50< celticminstrel> The MVC in the Lua interpreter was certainly overengineering IMO, but the use of MVC for the main game seems pretty reasonably to me. 20180321 00:10:12< celticminstrel> Well, I suppose that's possible, but what are the downsides? 20180321 00:10:30<+discordbot1> I see no downsides 20180321 00:10:37< gfgtdf> and what are it's upsides ? 20180321 00:10:39< celticminstrel> There are definitely downsides. 20180321 00:10:54<+discordbot1> the upside would be the code would be a lot simpler. 20180321 00:11:05< celticminstrel> I don't feel well-versed enough in MVC to articulate the downsides of dropping it. 20180321 00:11:11< celticminstrel> But they definitely exist. 20180321 00:11:14<+discordbot1> What I'm trying to do right now is untangle the whole fucking mess that is the controller and display interface 20180321 00:11:15<+discordbot1> handle_event would be more complex 20180321 00:11:28< celticminstrel> Also, I don't think it's reasonable to say that MVC is more complicated than non-MVC. 20180321 00:11:46<+discordbot1> do you want me to dig up the log? 20180321 00:12:08< celticminstrel> Why are you bringing up logs all of a sudden? 20180321 00:12:46< celticminstrel> MVC isn't even that complicated, at least in principle. The model holds the data, the view puts it onscreen, and the controller handles user interaction. 20180321 00:13:06< celticminstrel> Exactly which component calls the others can vary. 20180321 00:13:10<+discordbot1> celmin: read this please https://pastebin.com/cgYaY7rc 20180321 00:13:46< celticminstrel> Well, the first question that pops into my head is whether that's really a problem that needs solving, but that aside... 20180321 00:13:49 * celticminstrel continues reading. 20180321 00:15:08<+discordbot1> even you must admit that's a fucking mess. 20180321 00:15:17<+discordbot1> of epic proportions 20180321 00:15:26< celticminstrel> Got as far as line 24 and my conclusion so far is that "mouse controller should probably just be merged into controller base somehow". 20180321 00:15:38< celticminstrel> Unless I misread or misinterpreted something. 20180321 00:15:49< gfgtdf> cleaning up includes shodul not be a main aspect when doing thse decisions. 20180321 00:15:57<+discordbot1> oh ffs... 20180321 00:16:06< gfgtdf> mouse_handler is more a part of controller than of display. 20180321 00:16:18< celticminstrel> Yeah. 20180321 00:16:33< gfgtdf> but really i don't like mouse_handler much eigher. 20180321 00:16:37<+discordbot1> YES there are elements that belong in the controller 20180321 00:17:15< celticminstrel> Well, keep in mind that I'm saying this entirely based on the log you posted, so I don't necessarily know everything there is to know about mouse handler, so calm down. >_> 20180321 00:17:18<+discordbot1> I'm pondering whether STUFF THAT DIRECTLY CONTROLS THE VIEWABLE REPRESENTATION OF THE MAP, such as scrolling, should be handled in the same place as the implementation of said view 20180321 00:17:32<+discordbot1> and you guys aren't helping 20180321 00:17:36< gfgtdf> scrolling is done via hotkey system 20180321 00:17:46< celticminstrel> IMO? Scroll events should be handled in the controller, which then calls into the display to perform the actual scroll. 20180321 00:18:04<+discordbot1> fucking hotkey system.. 20180321 00:18:26< celticminstrel> Calm down... 20180321 00:18:44<+discordbot1> it makes everything 1000% more complicated 20180321 00:19:23< celticminstrel> How? Logically it should make everything simpler. Without it you'd have to handle like seven different kinds of event. With it, you just need to handle hotkey events. 20180321 00:19:57<+discordbot1> I'm looking at the code 20180321 00:20:06<+discordbot1> it does not look like what gfgtdf says is the case? 20180321 00:20:18< gfgtdf> hm ? 20180321 00:20:22<+discordbot1> a SDL_MOUSEBUTTONUP calls mh_base.mouse_press(event.button, is_browsing()); 20180321 00:20:23< celticminstrel> Scrolling is indeed handled in the hotkey system if that's what you're talking about. 20180321 00:20:37< celticminstrel> ...what the heck is is_browsing. 20180321 00:20:47<+discordbot1> if it's a left click event, ht calls left_click, which handles scrolling 20180321 00:20:48< gfgtdf> vulprobably mh_base.mouse_press invlΓΆokes the hotkey system 20180321 00:20:51< gfgtdf> probably* 20180321 00:20:54<+discordbot1> @hrubymar10 xcode build is failing: clang: error: no such file or directory: '/Users/travis/build/wesnoth/wesnoth/src/floating_textbox.cpp' Dunno if anything reported on travis is worth fixing atm though. 20180321 00:21:07<+discordbot1> it does not 20180321 00:21:08<+discordbot1> @celticministrel That's for observers, IIRC 20180321 00:21:34<+discordbot1> I'm just seeing multiple layers of abstraction 20180321 00:21:38< gfgtdf> ah riogth the line below calls hotkey::mbutton_event 20180321 00:21:41< gfgtdf> hmm 20180321 00:21:54<+discordbot1> yes 20180321 00:22:00<+discordbot1> but it doesn't do the scrolling 20180321 00:22:38<+discordbot1> @Vultraz Speaking generically... I would start not with the current implementation, but with minimal API (set of public methods) necessary to achieve its goal. The first thing to do is to identify its responsibility. If there are multiple, then it's the way to split it into several objects. Let me take a more specific look at it... 20180321 00:23:00< gfgtdf> hm ye, but scrolling via keyboard is done via hotkey sysrtme is assume? not aure anymore. 20180321 00:23:18< celticminstrel> Oh yeah, mouse scrolling might not be handled via hotkeys, I guess. 20180321 00:23:22<+discordbot1> I'm trying to break the logic down into a coherent design in order to proceed 20180321 00:23:25<+discordbot1> Scrolling is done by directly testing with SDL if a given key is pressed. 20180321 00:23:29< celticminstrel> And scroll wheel scrolling may also be hard-coded. 20180321 00:23:50< celticminstrel> @sinda There are definitely hotkeys for that though, and last I checked they worked. 20180321 00:24:15<+discordbot1> god dammit discord stop scrolling up 20180321 00:24:45<+discordbot1> ftr, I've also considered we should move the hotkey handling to global scope 20180321 00:24:54<+discordbot1> and filter it by context 20180321 00:25:00<+discordbot1> ie, game, editor, UI 20180321 00:25:50<+discordbot1> but that's different 20180321 00:27:10<+discordbot1> celmin: why should the controller handle scrolling? 20180321 00:27:42< celticminstrel> Because scrolling is user interaction, and that's what the controller is for. However, the controller should not attempt to visualize the result of scrolling; that belongs in the display. 20180321 00:29:43<+discordbot1> well then I guess nothing changes 20180321 00:29:56<+discordbot1> god dammit. 20180321 00:30:14<+discordbot1> I can't do this 20180321 00:30:14< celticminstrel> If you say so? 20180321 00:30:20< celticminstrel> :/ 20180321 00:30:38<+discordbot1> I'm not a good enough programmer to understand the design needed here 20180321 00:31:00< celticminstrel> I do think there would be a benefit to figuring out how many different views, controllers, and models we have. 20180321 00:31:17<+discordbot1> too many 20180321 00:31:23<+discordbot1> alright, here's a question for you 20180321 00:31:31<+discordbot1> who should own the UI dialog object 20180321 00:31:43<+discordbot1> right now i have it as a unique ptr in display 20180321 00:31:58<+discordbot1> initialized by derived classes with the appropriate object 20180321 00:32:13< celticminstrel> Off the top of my head, I'd say no more than three models (a game, a map, and a scenario, the latter two being for the editor); two or three controllers (editor, player, possibly separate player controllers for single and multi); and two models (main map, minimap). 20180321 00:32:30<+discordbot1> for some reason, I've become confused with the way derived classes function... 20180321 00:32:50<+discordbot1> since display only owns a prt of the dialog base class, I assume it can only access base class functions 20180321 00:33:08<+discordbot1> @loonycyborg SCons 3.0.1 and GCC 7.3.0.. 20180321 00:33:12< celticminstrel> By UI dialog object you mean the ingame_ui object? 20180321 00:33:20<+discordbot1> ingame_ui_base 20180321 00:33:25< celticminstrel> Hmm. 20180321 00:33:33<+discordbot1> ok, quick c++ detour 20180321 00:33:43<+discordbot1> display owns a unique_ptr 20180321 00:34:19<+discordbot1> game_display, which inherits from display, initializes that ptr with a new game_ui, object. game_ui inherits from ingame_ui_base 20180321 00:34:53<+discordbot1> when you do something like that, if you were to do dynamic_cast(ptr) in display, what would you get? 20180321 00:35:32< celticminstrel> I'm confused because this seems like such an obvious question. 20180321 00:35:48<+discordbot1> would you get a valid pointer because the exact type is game_ui? 20180321 00:35:50< celticminstrel> I mean, you'd get back the game_ui object that game_display constructed, right? 20180321 00:36:02<+discordbot1> ok 20180321 00:36:23<+discordbot1> then under what circumstances can an object lose what I've heard referred to an this "exact type"? 20180321 00:36:45< celticminstrel> How can an object lose its exact type? 20180321 00:36:50< celticminstrel> Uhh. It can't? 20180321 00:37:02< celticminstrel> Unless you've done something really bad. 20180321 00:38:14<+discordbot1> for example, in modal_dialog, there's a vector of field_base*. Let say I add a member of type field_bool, which inherits from field_base. So, vec.emplace_back(new field_bool()). Now, I want to return a field_bool from some function. I found I cannot do return vec.back(). 20180321 00:39:04< celticminstrel> You have to cast it. 20180321 00:39:50<+discordbot1> because the compiler cannot know at build time whether the object is field_base or field_bool. Is that right? 20180321 00:40:03< celticminstrel> Pretty much? 20180321 00:42:04<+discordbot1> alright 20180321 00:42:04<+discordbot1> thanks 20180321 00:42:16<+discordbot1> back to the ui thng 20180321 00:42:19<+discordbot1> who should own it? 20180321 00:42:30<+discordbot1> should the controller own both the in-game ui object, and the display? 20180321 00:42:56<+discordbot1> or should the controller own the display which would own the ui object? 20180321 00:43:08<+discordbot1> i don't think that makes sense if we want the display to be ONLY the MAP 20180321 00:43:25<+discordbot1> so let's run with the idea of the controller owning it 20180321 00:43:33<+discordbot1> we'd end up with three parallel class trees. 20180321 00:43:53< celticminstrel> TBH the dialog object kinda is both a controller and a view, so this is a weird kind of question anyway. 20180321 00:44:32<+discordbot1> display as the base of game_display and editor_display. controller_base as the base of play_controller and editor_controller ingame_ui_base as the base of game_ui and (yet to be created) editor_ui 20180321 00:44:32< celticminstrel> Mind you, it may be a view of more than just the map. 20180321 00:44:34< irker760> wesnoth/wesnoth:master Jyrki Vesterinen e70a9c769d Fix size of fake screen AppVeyor: 2/4 builds failed 20180321 00:44:35< irker760> Details vs2015/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-2057 20180321 00:44:36< irker760> Details vs2017/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-master-1761 20180321 00:44:50< celticminstrel> Still, I think perhaps it makes more sense for the dialog to own the controller than the controller to own the dialog. 20180321 00:45:02<+discordbot1> eh? 20180321 00:45:16<+discordbot1> what do you mean 20180321 00:45:24< celticminstrel> If the idea is that display only handles the map, not the theme... 20180321 00:45:46<+discordbot1> the theme should be irrelevant to the display of the map 20180321 00:45:57< celticminstrel> The game_ui is then the theme plus the map, so it makes sense that it should own the display. 20180321 00:46:10< celticminstrel> It could either own the controller or be the controller. More likely the former, I guess. 20180321 00:46:13<+discordbot1> what? 20180321 00:46:24<+discordbot1> ... why would the game_ui own the map? 20180321 00:46:40<+discordbot1> game_ui should have ABSOLUTELY NOTHING AT ALL EVER to do with drawing the map 20180321 00:47:01< celticminstrel> Well, it could also be that I misunderstood what the game_ui is intended to be. 20180321 00:47:22<+discordbot1> the game_ui class is intended to be a modeless dialog that draws over the map. 20180321 00:47:26< celticminstrel> If it's only the theme (sidebar, menubar, etc) then yeah it doesn't make sense for it to own the map. 20180321 00:47:26<+discordbot1> think of it as a HUD 20180321 00:47:37<+discordbot1> it very well might be a HUD later down the road 20180321 00:47:38< celticminstrel> Ah, okay. 20180321 00:48:35< celticminstrel> It's a bit hard to choose between display and controller, but I think I'm leaning towards the latter. 20180321 00:48:50< celticminstrel> Um, I don't think you can say it's not a HUD now, TBH. 20180321 00:49:02<+discordbot1> HUDs are transparent 20180321 00:49:06<+discordbot1> sometimes 20180321 00:49:07< celticminstrel> That's not a requirement. 20180321 00:49:12< celticminstrel> They can be opaque. 20180321 00:49:17<+discordbot1> right now it's not really an HUD since it never draws over the map 20180321 00:49:24<+discordbot1> technically 20180321 00:49:30<+discordbot1> but yeah, i guess you make a point 20180321 00:49:52< celticminstrel> Well, technically is kinda meaningless since when it's not transparent you don't want to draw the map under it anyway. 20180321 00:50:07<+discordbot1> depends on the game 20180321 00:51:05<+discordbot1> but whatever. 20180321 00:51:07<+discordbot1> ok. 20180321 00:51:18<+discordbot1> should the controller base class have any knowledge of the ui object? 20180321 00:52:01<+discordbot1> either though a protected member or a virtual function? 20180321 00:52:33< celticminstrel> Probably not, I'd say... 20180321 00:52:48< celticminstrel> The UI would be different for each controller, I think. 20180321 00:53:00<+discordbot1> yes 20180321 00:53:02< irker760> wesnoth/wesnoth:master pentarctagon 803db572d1 Add the updated Dunefolk race descriptio AppVeyor: vs2017/Release Failed 20180321 00:53:03< irker760> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-master-1764 20180321 00:53:12< celticminstrel> Oh yeah, I suppose replays could have a separate controller... 20180321 00:54:50< gfgtdf> actualyl they did in 1.12, but i changed that in 1.13 to impleemnt the 'continue play' and 'network mp replay' features 20180321 00:55:26<+discordbot1> celmin: so, basically say 20180321 00:55:48<+discordbot1> play_controller would own game_display and a game_ui object? 20180321 00:56:18< celticminstrel> gfgtdf: I don't see how those two are mutually exclusive though. 20180321 00:56:33< celticminstrel> @Vultraz That sounds reasonable to me. 20180321 00:56:38<+discordbot1> OK 20180321 00:56:40<+discordbot1> i can make some progress 20180321 00:56:46<+discordbot1> thanks 20180321 00:57:01< celticminstrel> It would probably also own the model (game_state I guess?). 20180321 00:57:18<+discordbot1> it does 20180321 00:58:35< gfgtdf> celticminstrel: depodens on what you mean with "separate controller" 20180321 00:59:04< gfgtdf> celticminstrel: i mean the _is_ a replay only controller https://github.com/wesnoth/wesnoth/blob/master/src/replay_controller.cpp 20180321 00:59:34< gfgtdf> bute it's more like a plugin to the play_controller 20180321 01:01:06-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection] 20180321 01:05:14< celticminstrel> Actually, we should probably consider deprecating wesnoth.theme_items at level 3, because the chances of any new theme API being compatible is fairly low. 20180321 01:05:59< celticminstrel> (I mean on 1.14 branch. Unless it's too late to deprecate things? I don't think it'd cause problems with comatibility between stable versions, but I might be missing some detail.) 20180321 01:06:18<+discordbot1> it can be done 20180321 01:06:38< celticminstrel> Assuming it's not deprecated on the 1.14 branch it might well be one of those rare cases of using level 4 deprecation. 20180321 01:06:59< gfgtdf> celticminstrel: hmm why shodulwe drop compability for it? we just need to have a [report] gui2 widget and reuse the code form reports.cpp 20180321 01:07:08<+discordbot1> ......... 20180321 01:07:23< gfgtdf> i mean i have no problem with deprecating it as loong as we replace it with sometihng of eiaquvalent power 20180321 01:07:43<+discordbot1> I have not even considered what the new theme interface will be 20180321 01:07:51< celticminstrel> gfgtdf: My reasoning is that we may want to retain it in spirit but in a form completely incompatible with its current form. 20180321 01:08:12< celticminstrel> Thus, deprecating it in 1.14, then removing it in 1.15 and adding in something new to replace it. 20180321 01:08:55< celticminstrel> I assume the code from reports.cpp will be involved somehow in the new theme interface (though probably not entirely unchanged). 20180321 01:09:01< gfgtdf> well no, you cannto deprecate it since 1.14 will ahve no replacement for it 20180321 01:09:11<+discordbot1> uh what? 20180321 01:09:18<+discordbot1> no? 20180321 01:09:20<+discordbot1> of course you can 20180321 01:09:24< gfgtdf> i mean what shodul umc aouthor do ? 20180321 01:09:34<+discordbot1> oh wait, I see what you mean 20180321 01:09:48< celticminstrel> It's certainly possible to deprecate things that have no replacement. My thought was to include a message saying something to the effect of "This feature will be redesigned in 1.15". 20180321 01:10:15< celticminstrel> However I don't know if it's appropriate to deprecate this specifically without having a replacement. 20180321 01:10:59< celticminstrel> I just thought it might be useful to use deprecation as a sort of heads-up that they would likely need to rewrite their theme_items code when 1.16 / 2.0 comes around. 20180321 01:11:02< gfgtdf> celticminstrel: well yes you can add such a message on the wiki page. but not somewhere ingame where it annoys users. 20180321 01:11:29< celticminstrel> IIUC there aren't a lot of people using theme_items? And 90% of those who use it are probably using it for custom status effects. 20180321 01:11:45< celticminstrel> gfgtdf: Deprecation messages won't annoy users unless they have debug mode enabled. 20180321 01:11:58< gfgtdf> celticminstrel: even level 3 deprecation messages ? 20180321 01:12:02< celticminstrel> Yeah. 20180321 01:12:19< celticminstrel> Level 1 deprecation messages won't even appear with debug mode unless you take special measures. 20180321 01:12:37< celticminstrel> I think level 4 would appear without debug mode, because it raises a Lua error. 20180321 01:13:24<+discordbot1> play_controller is quite bloated 20180321 01:14:07< gfgtdf> ye we shodul probably somehow split the 'input' and the 'main game logic (tunr conunter)' etc. part. 20180321 01:14:37<+discordbot1> a whole bunch of functions are just accessors for stuff in game_state. 20180321 01:15:07< gfgtdf> also the is no real difference betwen play_controller and playsingle_controller, the onyl reason why i didn'tmerge them is exactly that play_controller was already "big enough" 20180321 01:15:39<+discordbot1> indeed... 20180321 01:24:01< Ravana_> Lua errors are only displayed in Lua console, which needs debug 20180321 01:24:40< gfgtdf> is this new behviour ? 20180321 01:24:51< celticminstrel> Oh yeah, unless they include the magic ~wml marker... I don't remember whether the deprecation ones do though. 20180321 01:24:52< gfgtdf> i see "lua error" post in the wiki all the time 20180321 01:25:04< Ravana_> new of 1.13 20180321 01:25:22< gfgtdf> so what heppen if i don't ahev debug mode on ? 20180321 01:25:25< gfgtdf> happens* 20180321 01:25:43< Ravana_> game breaks without any visible reason for it 20180321 01:25:55< gfgtdf> this is bad 20180321 01:26:06< gfgtdf> i mean we do want them to be able to report bugs 20180321 01:26:25< gfgtdf> i am sure even mainline has still bugs. 20180321 01:27:14< gfgtdf> and most persosns do not use debug mode unless they are debuggin oen particular bug. 20180321 01:27:22< celticminstrel> gfgtdf does have a point. 20180321 01:28:10< celticminstrel> The information will also be in the log files, mind you... 20180321 01:28:20< celticminstrel> At least on Windows (do other platforms save the logs to files?). 20180321 01:30:06< gfgtdf> but it should also tell you that a bug appeared in a fiurst place, i mean it doesn't need to display the full error message in the chat aera like it does in 1.12 . But we do not want peopel to contiue playing with a probalby corrupded gamestate unaware of that. 20180321 01:30:33< celticminstrel> Showing a simple message with a pointer to the logs for more details seems reasonable. 20180321 01:40:27<+discordbot1> i wonder if VS 2017 supports nested namespace definitons 20180321 01:45:37<+discordbot1> let us see! 20180321 01:46:04<+discordbot1> looks like it does, if building inc++17 mode 20180321 01:52:55< celticminstrel> But IIRC we were going to be building in C++14 mode? 20180321 01:54:02<+discordbot1> yes. 20180321 01:55:07<+discordbot1> But jyrki did agree we can actually make VS 2017 the min version to deal with lack of full c++14 support in 2015, so we might be able to use some minor c++17 features like nested namespace definitions since we'll likely be using at least gcc 6. 20180321 01:58:32< celticminstrel> I don't see how that follows at all? 20180321 01:59:09< celticminstrel> Making VS2017 the min version does not mean requiring C++17. 20180321 01:59:45<+discordbot1> No, but c++17 support is quite good in 2017 20180321 02:00:00< celticminstrel> That's entirely beside the point. 20180321 02:00:12<+discordbot1> so I'm saying it allows the use of some c++17 features when considered in aggregate with our other compiler versions 20180321 02:01:56< celticminstrel> I don't think we should even be thinking about this right now. Stick with C++14 for the time being. We can upgrade to C++17 in a few months to a year maybe, when everything is more stable. 20180321 02:02:14<+discordbot1> sure 20180321 02:02:48<+discordbot1> I wonder when VS 2019 comes out... 20180321 02:03:12< celticminstrel> I'd guess in 2019. :P 20180321 02:03:33<+discordbot1> but didn't 2017 come out before 2017? 20180321 02:03:35< celticminstrel> Maybe I'll actually use it if it removes the dumb login requirement. 20180321 02:03:43< celticminstrel> It might've, I don't remember. 20180321 02:04:01< celticminstrel> Still, even if it's before 2019 I would thing late 2018. 20180321 02:04:05< celticminstrel> ^think 20180321 02:05:19<+discordbot1> prolly 20180321 02:05:56<+discordbot1> i doubt they'll remove the login thing though 20180321 02:06:09<+discordbot1> it's really not that stupid 20180321 02:06:44< celticminstrel> It's the most stupidest thing imaginable. 20180321 02:07:55<+discordbot1> whether it is or is not, it's an odd ideological hill to make a stand on, especially when in practical terms it would cause you no inconvenience 20180321 02:08:26< celticminstrel> It totally would? 20180321 02:08:32<+discordbot1> in face, it's causing you even MORE inconvenience to not do so 20180321 02:08:49<+discordbot1> fact 20180321 02:08:52< celticminstrel> Not really, because I still have VS2013 which works just fine. 20180321 02:09:05<+discordbot1> for 1.14 20180321 02:09:09< celticminstrel> It's only Wesnoth that has decided to force me to use something else. 20180321 02:09:22< celticminstrel> Anything else I work on can continue to be VS2013. 20180321 02:09:30<+discordbot1> I can't imagine 2013 is very fast either 20180321 02:09:41< celticminstrel> True. 20180321 02:09:46<+discordbot1> IIRC they made a lot of effort to rewrite the compiler after 2013 to make it faster 20180321 02:09:53< celticminstrel> But when building something other than Wesnoth, it's fast enough. 20180321 02:11:55<+discordbot1> really? 20180321 02:12:01<+discordbot1> I don't remember it being all that fast 20180321 02:12:27<+discordbot1> and wesnoth is huge 20180321 02:12:29< celticminstrel> Yup. It's not too slow when I build BoE, and IIRC that's the largest besides Wesnoth that I work on. 20180321 02:14:39< irker760> wesnoth: Charles Dang wesnoth:master a5344ae52a3f / src/ (play_controller.cpp play_controller.hpp): Play Controller: formatting cleanup https://github.com/wesnoth/wesnoth/commit/a5344ae52a3f79fe2bc2b3c0592d914f9bb44b8f 20180321 02:14:42< irker760> wesnoth: Charles Dang wesnoth:master 3254bb2c7d24 / src/ (7 files in 2 dirs): Moved UI object from display to play_controller https://github.com/wesnoth/wesnoth/commit/3254bb2c7d244296680af37a6d1d9f7767e4df7c 20180321 02:14:45< irker760> wesnoth: Charles Dang wesnoth:master 05521756935c / INSTALL.md: Merge branch 'master' of github.com:wesnoth/wesnoth https://github.com/wesnoth/wesnoth/commit/05521756935cba176e9705989d79689000089968 20180321 02:14:54< celticminstrel> Blech merge commits. 20180321 02:15:16< celticminstrel> (To be clear I only dislike merge commits of the form "Merge master into my branch".) 20180321 02:15:56<+discordbot1> I didn't really want to rebase just for the update of INSTALL 20180321 02:16:00<+discordbot1> it'd mean another large rebuild 20180321 02:16:16<+discordbot1> yaknowwhatimsayin 20180321 02:16:59< celticminstrel> It'd be so much better if Visual Studio judged whether to rebuild by considering the actual file contents rather than just the last modified time... 20180321 02:17:29< celticminstrel> I don't remember whether I had to rebuild after a rebase in Xcode... 20180321 02:18:06<+discordbot1> play_controller really has a lot of accessors into the gameboard... 20180321 02:21:39<+discordbot1> The question is which is more important: supporting VS 2015 or supporting C++17. 20180321 02:21:41<+discordbot1> Microsoft has a chronic aversion to file content accesses. 20180321 02:22:02<+discordbot1> That's the whole reason why they stuck with the "file extension = file type" paradigm. 20180321 02:23:04<+discordbot1> @Tad Carlucci the idea is to support 2017 in order to have cull C++14 support. stuff like extended constxpr is not supported in 2015. Any C++17 support thereby gained by use of 2017 in pursuit of c++14 support is incidental. 20180321 02:25:49< irker760> wesnoth/wesnoth:master Charles Dang 0552175693 Merge branch 'master' of github.com:wesn AppVeyor: vs2015/Release Failed 20180321 02:25:50< irker760> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-2061 20180321 02:27:00< celticminstrel> Full not cull 20180321 02:27:45<+discordbot1> yes 20180321 02:27:56<+discordbot1> you don't see message edits 20180321 02:27:59<+discordbot1> πŸ˜› 20180321 02:28:22< celticminstrel> Unfortunate. 20180321 02:30:34<+discordbot1> kinda impossible to do since the bot is just a pipe 20180321 02:30:42<+discordbot1> and on irc,messages never change 20180321 02:31:02<+discordbot1> Butts. 20180321 02:31:21<+discordbot1> Can anyone remind me what it is I was supposed to look at as soon as I was back on my desktop? 20180321 02:31:37<+discordbot1> Besides the 1.14 wesnothd. 20180321 02:32:05<+discordbot1> changing the label of modifications from "MP Modification" to just "Modification" on the addon server web interface 20180321 02:32:17<+discordbot1> Oh that's a good one. 20180321 02:32:45<+discordbot1> I also need to learn git worktree. 20180321 02:33:39< celticminstrel> There you go again claiming things are impossible. 20180321 02:34:00< celticminstrel> While it's true there's no way to actually edit the message on the IRC side, it would still be possible to convey the edits to IRC users. 20180321 02:34:18<+discordbot1> He is right though, it is impossible because the bot is a pipe. It doesn't handle state. 20180321 02:34:24<+discordbot1> I said it's impossible since all the bot does is deliver messages 20180321 02:34:28<+discordbot1> But I said butts because I want you both to change back to a more productive topic. 20180321 02:35:41<+discordbot1> For example 20180321 02:35:55<+discordbot1> Someone sell git worktree to me. 20180321 02:36:02-!- travis-ci [~travis-ci@ec2-54-91-233-154.compute-1.amazonaws.com] has joined #wesnoth-dev 20180321 02:36:03< travis-ci> wesnoth/wesnoth#17069 (master - 0552175 : Charles Dang): The build failed. 20180321 02:36:03< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/356163862 20180321 02:36:03-!- travis-ci [~travis-ci@ec2-54-91-233-154.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180321 02:36:25<+discordbot1> I've lived the git-new-workdir life for years and now I need an effective salesperson to sell me the new thing. 20180321 02:36:34-!- gfgtdf_ [~chatzilla@x4e368626.dyn.telefonica.de] has joined #wesnoth-dev 20180321 02:37:14<+discordbot1> hey, don't look at me 20180321 02:37:23<+discordbot1> Yeah, I know you're the GUI scrub here. 20180321 02:37:44<+discordbot1> zookeeper uses the same GUI πŸ˜› 20180321 02:37:51<+discordbot1> He's also a GUI scrub. 20180321 02:37:59<+discordbot1> I've known this ever since the Subversion days. 20180321 02:38:48-!- gfgtdf [~chatzilla@x4e3687a6.dyn.telefonica.de] has quit [Ping timeout: 256 seconds] 20180321 02:38:59-!- gfgtdf_ is now known as gfgtdf 20180321 02:40:41< irker760> wesnoth: Charles Dang wesnoth:master a971af89d7af / src/game_initialization/playcampaign.cpp: Playcampaign: formatting cleanup https://github.com/wesnoth/wesnoth/commit/a971af89d7af820e9e02d52ea4dc4121af2d4142 20180321 02:40:44< irker760> wesnoth: Charles Dang wesnoth:master 9bc19d7e027b / src/display.cpp: Fixup 3254bb2 (forgot to include this) https://github.com/wesnoth/wesnoth/commit/9bc19d7e027b3d3bdbd50456f4c9be274de4bd3b 20180321 02:41:36<+discordbot1> Casual reminder that I hate subjects that are nothing but gibbering along the lines of "Fixup ". 20180321 02:41:44<+discordbot1> *gibberish 20180321 02:42:44< celticminstrel> Agreed. Though I think I've done them too sometimes. 20180321 02:43:11< celticminstrel> The --fixup option actually uses the old commit's message rather than its hash (which is safer if you're going to rebase it out later). 20180321 02:43:29< celticminstrel> On the other hand it does mean github wouldn't be able to link to the old commit. 20180321 02:43:32<+discordbot1> wait, it's an actual option? 20180321 02:43:41< celticminstrel> "git commit --fixup hash" yes. 20180321 02:43:47<+discordbot1> huh 20180321 02:43:56<+discordbot1> People who learned under CVS/SVN got into the bad habit of using commit ids to reference commits in lieu of meaningful descriptions, yes. 20180321 02:43:58<+discordbot1> I thought it was just a message thing iceiceice did 20180321 02:44:02<+discordbot1> Subjects, rather. 20180321 02:44:51<+discordbot1> It's always fun when they refer to objects whose hashes have changed over time for one reason or another (e.g. rebases or rewrites requested by the legal dept.) 20180321 02:46:07< celticminstrel> What do you mean by "a message thing iceiceice did" 20180321 02:46:23< celticminstrel> @shadowm Yeah that's why I said --fixup is safer if you intend to rebase it out later. 20180321 02:46:35<+discordbot1> I got the habit of "fixup foo" commits from iceiceice 20180321 02:47:27<+discordbot1> Trust me, I tried to get him to drop that. 20180321 02:48:01<+discordbot1> Some people just want to watch the world burn with opaque hashes. 20180321 02:48:38<+discordbot1> (Either that or they were born gifted with the ability to map said hashes to meaningful objects in their squishy human brains somehow.) 20180321 02:49:52< celticminstrel> Github does nicely link to it for you which is nice, but it's still kinda... 20180321 02:50:38< celticminstrel> I can say that it's nice to have a link to the commit it's fixing, but it would be nicer if it were a bit more descriptive. 20180321 02:51:43<+discordbot1> Hold the phone. 20180321 02:52:18<+discordbot1> Did anyone ever remember that the in-game help has a Using Add-ons section explaining the individual add-on types? 20180321 02:52:33<+discordbot1> oh 20180321 02:52:37<+discordbot1> And that in particular it wasn't updated for commit a09e2bb8583c6125f8c4e3157d6eba8bbd800b62? 20180321 02:52:43<+discordbot1> no 20180321 02:52:50<+discordbot1> https://github.com/wesnoth/wesnoth/commit/a09e2bb8583c6125f8c4e3157d6eba8bbd800b62 20180321 02:53:17<+discordbot1> sigh One more issue for 1.14.1 I guess. 20180321 02:55:03<+discordbot1> 😦 20180321 02:59:28<+discordbot1> Repurposed the same ticket for the whole Add-ons section since Installing Add-ons also seems to have drifted behind the times. 20180321 03:00:06<+discordbot1> Oh, yeah, you also wanted to play with the Sylph sprite when you were back at your desktop. 20180321 03:00:14<+discordbot1> "To install an add-on, select it from the list and click text='OK', or simply double-click on the add-on’s title." 20180321 03:00:29<+discordbot1> Hm, yeah, I remember that now. 20180321 03:01:00<+discordbot1> It might have a few rough edges from the palette update 20180321 03:01:19<+discordbot1> But I must give jet credit for that palette, it looks 10x better with it. 20180321 03:01:47<+discordbot1> I have opinions on the palette. 20180321 03:02:44<+discordbot1> I do like the palette, don't get me wrong (although the Standard Elvish Blondβ„’ has always seemed a tad oversaturated to me on different screens with different colour configurations). 20180321 03:03:32-!- travis-ci [~travis-ci@ec2-54-166-113-143.compute-1.amazonaws.com] has joined #wesnoth-dev 20180321 03:03:33< travis-ci> wesnoth/wesnoth#17070 (master - 9bc19d7 : Charles Dang): The build is still failing. 20180321 03:03:33< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/356168906 20180321 03:03:33-!- travis-ci [~travis-ci@ec2-54-166-113-143.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180321 03:03:39<+discordbot1> I've just never understood why sprites went in that direction while terrain under Eleazar's command went into the tropey Real Is Brown direction. 20180321 03:05:12<+discordbot1> To this day I'm still not a fan of the resulting colour dissonance, especially where water is involved. 20180321 03:06:44<+discordbot1> I just don't usually bring it up since I don't feel anywhere near competent enough to discuss the terrain art style on a level ground with my complete lack of experience in that field. 20180321 03:07:42-!- travis-ci [~travis-ci@ec2-54-166-113-143.compute-1.amazonaws.com] has joined #wesnoth-dev 20180321 03:07:43< travis-ci> shikadiqueen/wesnoth#32 (master - e70a9c7 : Jyrki Vesterinen): The build is still failing. 20180321 03:07:43< travis-ci> Build details : https://travis-ci.org/shikadiqueen/wesnoth/builds/356170579 20180321 03:07:43-!- travis-ci [~travis-ci@ec2-54-166-113-143.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180321 03:07:53<+discordbot1> None of these puns are intentional. 20180321 03:09:45<+discordbot1> Where are SP modifications listed? 20180321 03:10:01<+discordbot1> I need to get the 1.14 worktree set up to actually have a look since master is dead. 20180321 03:12:10<+discordbot1> there are no SP modifications 20180321 03:12:15<+discordbot1> modifications work in both sp and mp 20180321 03:12:23<+discordbot1> I know. 20180321 03:12:25<+discordbot1> if you mean where you can access them in sp, the campaign menu 20180321 03:12:29<+discordbot1> I phrased that backwards. Where are-- okay. 20180321 03:14:55< irker760> wesnoth: Iris Morelle wesnoth:master 39b05dd0dbc1 / data/tools/addon_manager/html.py: wam: Update description and name for Modifications https://github.com/wesnoth/wesnoth/commit/39b05dd0dbc1fc14e50e7db19401651ec3eae154 20180321 03:15:30< irker760> wesnoth/wesnoth:master Charles Dang fef1310890 Attempt to fix build AppVeyor: 1/2 builds failed 20180321 03:15:31< irker760> Details vs2017/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-master-1762 20180321 03:18:16< irker760> hoi im irker 20180321 03:18:45< shadowm> Huh, didn't expect that to work. 20180321 03:19:08< shadowm> I guess ai0867.net still has the irker filter thing running. 20180321 03:20:22<+discordbot1> @Vultraz The changes will be reflected next time the thing runs. It'll also apply to the indexes for other instances though, not just 1.14. 20180321 03:21:27<+discordbot1> I for one am not too bothered about that. I mean, how many people read those descriptions anyway, 0.5% of the 170 people who use the add-ons web index every week? 20180321 03:22:03<+discordbot1> So, 8.5 people 20180321 03:22:30<+discordbot1> One of them is half-alive, half-dead, locked inside a box. 20180321 03:23:14<+discordbot1> Try not to think about it too much lest their life come to a swift end after having their predicament observed by a rude interloper. 20180321 03:26:28<+discordbot1> Interesting. Trees created by git worktree have .git/ replaced by a plain text file called .git. 20180321 03:27:36<+discordbot1> So that means git worktree would also portably work on Windows without messing with system-global security policies to enable symbolic links. 20180321 03:29:35< irker760> wesnoth: Iris Morelle wesnoth:1.14 2879d10dc53a / data/tools/addon_manager/html.py: wam: Update description and name for Modifications https://github.com/wesnoth/wesnoth/commit/2879d10dc53af88d04860e42108d2c9aa020f4ac 20180321 03:31:12<+discordbot1> TLDR Color: I wish the various units had some different hair colorings, maybe even celtic-style humans with blue body tattoos and such similar to historical times πŸ˜ƒ 20180321 03:31:28< celticminstrel> Gack pings/ 20180321 03:31:32<+discordbot1> or since we're talking a pure fantasy world, there's no reason for there not to be variety for the sake of variety πŸ˜‰ 20180321 03:32:10< celticminstrel> Since the Wesnoth personal naming scheme is Welsh-based, some tatoos might almost be appropriate thogh. 20180321 03:32:13< celticminstrel> ^tattoos 20180321 03:32:21<+discordbot1> lemme find a picture 20180321 03:32:21< celticminstrel> Only on outlaw units IMO. 20180321 03:32:37< celticminstrel> IIRC currently the only tattooed unit is actually the sarks. 20180321 03:32:52< celticminstrel> Which is a dwarf. 20180321 03:32:55< celticminstrel> So that doesn't count. 20180321 03:35:40< celticminstrel> ...what does this mean? "Disallowed access to blacklisted file paths such as hidden files and directories." 20180321 03:35:43< celticminstrel> From the changelog. 20180321 03:36:51<+discordbot1> https://cdn.discordapp.com/attachments/259976436490829825/425860306376327169/2018-03-20_23_36_26-nethergate_resurrection_at_DuckDuckGo_-_Chromium.png 20180321 03:37:04< celticminstrel> Ooh Nethergate. 20180321 03:37:06<+discordbot1> can you see that in IRC? 20180321 03:37:10<+discordbot1> πŸ‘πŸΎ 20180321 03:37:20< celticminstrel> Oh it's just the splash screen. 20180321 03:37:24<+discordbot1> yeah on sale on steam for under $2 USD 20180321 03:37:30< celticminstrel> (It shows up as a clickable link on IRC.) 20180321 03:37:38<+discordbot1> yup 20180321 03:37:41< celticminstrel> Well, I'm not going to buy it on Steam though. 20180321 03:37:48< celticminstrel> That said IIRC I already have a copy of it. 20180321 03:37:50<+discordbot1> Romans vs Celts would be a great campaign in Wesnoth too πŸ˜ƒ 20180321 03:38:02< celticminstrel> Sure. 20180321 03:38:12<+discordbot1> btw CM I know what that means 20180321 03:38:19<+discordbot1> "Disallowed access to blacklisted file paths such as hidden files and directories." 20180321 03:38:38< celticminstrel> Just because Wesnoth is a fantasy game at core doesn't mean you can't also have historical eras or campaigns. ;_ 20180321 03:38:40< celticminstrel> ^ ;) 20180321 03:38:46< celticminstrel> What does it mean? 20180321 03:38:47<+discordbot1> basically, the wesnoth file browser for saving and stuff had built-in file browser even on windows that would ignore some OS security settings 20180321 03:39:01<+discordbot1> for example allowing you to save and load save games in your C:\Windows directory πŸ˜› 20180321 03:39:05<+discordbot1> if you so desired 20180321 03:39:15<+discordbot1> obviously lead to a lot of weird permissions issues later on 20180321 03:39:18< irker760> wesnoth: Charles Dang wesnoth:master e1e7f65ec12b / src/ (7 files in 3 dirs): Display: removed interface that set up theme buttons https://github.com/wesnoth/wesnoth/commit/e1e7f65ec12bac4a19d01f5cf1366e24d64e01c0 20180321 03:39:21< irker760> wesnoth: Charles Dang wesnoth:master 72c451d2ae6b / src/ (display.cpp display.hpp): Display: restored background drawing https://github.com/wesnoth/wesnoth/commit/72c451d2ae6bf2cb3e2753f82e1ec8c3685adec9 20180321 03:39:22< celticminstrel> We're talking about the file dialog here, or the load game dialog or both? 20180321 03:39:24< irker760> wesnoth: Charles Dang wesnoth:master 5145b009a5d9 / src/ (display.cpp display.hpp): Display: removed code to draw theme panels/labels https://github.com/wesnoth/wesnoth/commit/5145b009a5d9dcfc9ab7a78bccea05e207f49e6b 20180321 03:39:38<+discordbot1> I believe it is the file dialog 20180321 03:39:40< celticminstrel> (File dialog is what you see in eg the map editor.) 20180321 03:39:53<+discordbot1> basically, any 'sensitive' directory now won't allow you to save or load from those directories; it'll properly error out 20180321 03:40:02<+discordbot1> at least that's how I understand it now 20180321 03:40:18< celticminstrel> So does that just means it's paying close attention to OS permissions? 20180321 03:40:27-!- gfgtdf [~chatzilla@x4e368626.dyn.telefonica.de] has quit [Read error: Connection reset by peer] 20180321 03:40:29< celticminstrel> (If it was about permissions though you wouldn't've been able to do it before anyway...) 20180321 03:40:48< celticminstrel> Or OS hidden-file attributes or what? 20180321 03:41:11< celticminstrel> Using "hidden" in the message seems to imply dot-dirs. 20180321 03:41:19< celticminstrel> So for example, not allowing access to a .git folder. 20180321 03:41:47< celticminstrel> @Vultraz Opinion on including the wml module in the release notes? 20180321 03:41:58<+discordbot1> might as welll 20180321 03:42:05< celticminstrel> And opinion on wesnoth.persistent_tags? 20180321 03:42:29<+discordbot1> eh? 20180321 03:42:44< celticminstrel> It's pretty new though, maybe better not to? I dunno. 20180321 03:43:02< celticminstrel> Basically a simpler way to do on_load/on_save for custom data in the save files. 20180321 03:43:21< celticminstrel> I'll edit your post to add stuff. 20180321 03:43:30< celticminstrel> I feel like the dunefolk should be mentioned too? 20180321 03:43:58<+discordbot1> yes 20180321 03:44:01<+discordbot1> please add 20180321 03:44:12< celticminstrel> Speaking of the dunefolk, the rename is also in this version right? 20180321 03:44:23<+discordbot1> previous 20180321 03:44:31< celticminstrel> Oh okay. 20180321 03:45:15<+discordbot1> celticminstrel: https://github.com/wesnoth/wesnoth/commit/c5e972b8b4b2dc0355dd05f2527c90c985de40a9 20180321 03:45:49<+discordbot1> I think you'll have to ask jyrkive directly if the code isn't obvious enough (the commit description is deliberately vague). 20180321 03:46:04< celticminstrel> If the wml module is going into release notes it'll need to be properly documented on the wiki, of course. 20180321 03:46:12<+discordbot1> Needless to say, it's not about the user dialogs. 20180321 03:47:10< celticminstrel> __MACOSX IIRC is a way for the Mac to represent HFS-specific metadata on a FAT filesystem. 20180321 03:47:28<+discordbot1> (It seems rather pointless to me that this was done in a hush-hush manner without even getting a CVE assigned to the issue but whatever.) 20180321 03:47:39< celticminstrel> What even is a CVE anyway? 20180321 03:48:03< celticminstrel> I keep seeing them and it gives off the impression that there's some central repository somewhere. 20180321 03:48:23<+discordbot1> CVE is one of several providers of identifiers and reports for documented security vulnerabilities. 20180321 03:48:49< celticminstrel> Oh, Jyrki extended the glob syntax o.o 20180321 03:48:57<+discordbot1> The entry isn't listed under "Security reason" for no reason, you know. 20180321 03:49:06<+discordbot1> *fixes 20180321 03:49:13< celticminstrel> Don't think it'll be useful in my schema branch but who knows. 20180321 03:50:36< celticminstrel> ... 20180321 03:50:51< celticminstrel> I just realized, ai.synced_command should've been a level 4 deprecation rather than an outright removal. 20180321 03:51:16< celticminstrel> I mean that's still an outright removal though, just with a better error message. 20180321 03:55:11< celticminstrel> @Vultraz Please take a look at my edits. 20180321 03:55:18<+discordbot1> in abit 20180321 04:03:44<+discordbot1> @loonycyborg Looks like I can't build Wesnoth 1.12 with SCons 3.0.1... scons: Reading SConscript files ... Saved options: default_targets = 'campaignd,test,wesnoth,wesnothd', extra_flags_config = '-pipe -march=native -Wno-implicit-fallthrough', extra_flags_release = '-O3', prefsdir = '.wesnoth-1.12', server_gid = '1000', server_uid = '1000', strict = True, ccache = True, fast = True, lockfile = True Building Wesnoth version 20180321 04:03:45<+discordbot1> 1.12.6+dev Creating lockfile AttributeError: 'module' object has no attribute 'config_checks': File "/home/shadowm/src/wesnoth-1.12/SConstruct", line 287: configure_args = dict(custom_tests = init_metasconf(env, ["cplusplus", "python_devel", "sdl", "boost", "pango", "pkgconfig", "gettext", "lua"]), config_h = "$build_dir/config.h", File "scons/metasconf.py", line 8: config_checks.update(module.config_checks) 20180321 04:09:38< celticminstrel> Wait why are you building 1.12 20180321 04:10:18<+discordbot1> I have to admit that my only surprise is that the author of that question is not Vultraz. 20180321 04:10:37<+discordbot1> Last time I checked the add-ons and MP servers for 1.12 were as active as ever. 20180321 04:11:20< irker760> wesnoth/wesnoth:master gfgtdf 439340ce78 update INSTALL.md for c++14 requirement AppVeyor: 2/4 builds failed 20180321 04:11:21< irker760> Details vs2015/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-2059 20180321 04:11:22< irker760> Details vs2017/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-master-1763 20180321 04:15:39< irker760> wesnoth/wesnoth:1.14 Iris Morelle 2879d10dc5 wam: Update description and name for Mod AppVeyor: vs2017/Release Failed 20180321 04:15:40< irker760> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-1.14-1768 20180321 04:16:56< celticminstrel> We should add an ai.run_formula() command to call FormulaAI. >_> 20180321 04:29:39-!- travis-ci [~travis-ci@ec2-54-161-172-140.compute-1.amazonaws.com] has joined #wesnoth-dev 20180321 04:29:40< travis-ci> wesnoth/wesnoth#17071 (master - 5145b00 : Charles Dang): The build has errored. 20180321 04:29:41< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/356180218 20180321 04:29:41-!- travis-ci [~travis-ci@ec2-54-161-172-140.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180321 04:49:28<+discordbot1> btw, I just enabled force-push and delete protection on the 1.14 branch. 20180321 04:57:33<+discordbot1> πŸ‘ 20180321 05:00:54< celticminstrel> I just discovered I can use tabs in the forum compose box. 20180321 05:00:58< celticminstrel> \o/ 20180321 05:12:05-!- TadCarlucci [~lundberg@74.193.219.119] has quit [Ping timeout: 268 seconds] 20180321 05:20:14-!- TadCarlucci [~lundberg@74.193.219.119] has joined #wesnoth-dev 20180321 05:24:29<+discordbot1> pondering the best way to get the GUI2 minimap drawing units... 20180321 05:26:25< irker760> wesnoth/wesnoth:master Charles Dang 5145b009a5 Display: removed code to draw theme pane AppVeyor: vs2015/Release Failed 20180321 05:26:26< irker760> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-2065 20180321 05:27:19< celticminstrel> I should have been in bed hours ago. >_> Good night. 20180321 05:27:27-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Quit: KABOOM! It seems that I have exploded. Please wait while I reinstall the universe.] 20180321 05:30:45<+discordbot1> Nighty. 20180321 05:30:53<+discordbot1> πŸ‘‹ 20180321 05:31:03<+discordbot1> Ugh nooooo. 20180321 05:31:12<+discordbot1> It's Wednesday. That means Wesnoth is doing the backup thing. 20180321 05:31:16<+discordbot1> Wesnoth.org. 20180321 05:31:31<+discordbot1> My wiki edit is gonna time ouuuuuut 20180321 05:32:41<+discordbot1> Oh, phew, it went through. 20180321 05:33:46<+discordbot1> People. 20180321 05:34:09<+discordbot1> You don't need to use [https://wiki.wesnoth.org/...] links in the wiki. There's a syntax for wiki links for a reason. 20180321 05:50:42< irker760> wesnoth: Charles Dang wesnoth:master 112ada179cb1 / src/ (18 files in 8 dirs): GUI2: added display and execute wrappers for a bunch more dialogs https://github.com/wesnoth/wesnoth/commit/112ada179cb12306c09f2c02da18442fb7c621f5 20180321 05:50:55<+discordbot1> Variadic templates are the best 20180321 05:53:02< irker760> wesnoth/wesnoth:master pentarctagon 803db572d1 Add the updated Dunefolk race descriptio AppVeyor: 2/4 builds failed 20180321 05:53:03< irker760> Details vs2017/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-master-1764 20180321 05:53:04< irker760> Details vs2015/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-2060 20180321 06:06:47< irker760> wesnoth: Charles Dang wesnoth:master 29ee9a71502f / src/ (6 files in 5 dirs): Convert uses of boost::regex and related functions/types to their stdlib counter https://github.com/wesnoth/wesnoth/commit/29ee9a71502fbc1c57b88daa533ffcfdbbbbd0fb 20180321 06:17:55-!- travis-ci [~travis-ci@ec2-54-166-113-143.compute-1.amazonaws.com] has joined #wesnoth-dev 20180321 06:17:56< travis-ci> wesnoth/wesnoth#17072 (master - 112ada1 : Charles Dang): The build failed. 20180321 06:17:56< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/356203473 20180321 06:17:56-!- travis-ci [~travis-ci@ec2-54-166-113-143.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180321 06:52:08< irker760> wesnoth/wesnoth:master Charles Dang 29ee9a7150 Convert uses of boost::regex and related AppVeyor: vs2015/Release Failed 20180321 06:52:09< irker760> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-2067 20180321 07:01:04-!- TadCarlucci [~lundberg@74.193.219.119] has quit [Ping timeout: 260 seconds] 20180321 07:01:07< irker760> wesnoth: Charles Dang wesnoth:master 1e9f2548a6fe / src/ (5 files in 5 dirs): Cleaned up MSVC 2013 compatibility code https://github.com/wesnoth/wesnoth/commit/1e9f2548a6fe83f933a4e45c398093fa6cb2746b 20180321 07:01:10< irker760> wesnoth: Charles Dang wesnoth:master 4cf1f8dc9d8e / src/gui/widgets/grid.hpp: GUI2/Grid: made use of std::exchange https://github.com/wesnoth/wesnoth/commit/4cf1f8dc9d8e90a5b273a01221b84d226361671f 20180321 07:01:23< irker760> wesnoth: Sofartin wesnoth:master 00dac71a9036 / projectfiles/Xcode/Wesnoth.xcodeproj/project.pbxproj: Fixed Xcode Project https://github.com/wesnoth/wesnoth/commit/00dac71a9036c2bc897bb4adf3ea510dfe37ee12 20180321 07:04:34-!- TadCarlucci [~lundberg@74.193.219.119] has joined #wesnoth-dev 20180321 07:06:19-!- travis-ci [~travis-ci@ec2-54-166-113-143.compute-1.amazonaws.com] has joined #wesnoth-dev 20180321 07:06:20< travis-ci> wesnoth/wesnoth#17073 (master - 29ee9a7 : Charles Dang): The build has errored. 20180321 07:06:20< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/356206749 20180321 07:06:20-!- travis-ci [~travis-ci@ec2-54-166-113-143.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180321 07:14:05< irker760> wesnoth: Charles Dang wesnoth:master f081a1c13390 / src/ (6 files in 5 dirs): Revert "Convert uses of boost::regex and related functions/types to their stdlib https://github.com/wesnoth/wesnoth/commit/f081a1c13390282936e926fd52a635fa64387877 20180321 07:14:06<+discordbot1> blah. ok. reverting that... 20180321 07:14:10<+discordbot1> will talk to celmin tomorrow 20180321 07:18:26< vn971> compilation problem solved. IDK what exactly helped, but I've updated to today-s master (+24hours master) and wiped Ccache. It compiled just fine. 20180321 07:25:51< irker760> wesnoth/wesnoth:master Charles Dang 0552175693 Merge branch 'master' of github.com:wesn AppVeyor: 2/4 builds failed 20180321 07:25:52< irker760> Details vs2015/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-2061 20180321 07:25:53< irker760> Details vs2017/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-master-1765 20180321 07:30:02<+discordbot1> So the macOS people haven't figured out yet how to fix the problem with the case-insensitive sorter? 20180321 07:31:06<+discordbot1> (Which, by the way, lacks a ticket number in the release announcement.) 20180321 07:36:00-!- travis-ci [~travis-ci@ec2-54-166-113-143.compute-1.amazonaws.com] has joined #wesnoth-dev 20180321 07:36:01< travis-ci> shikadiqueen/wesnoth#33 (1.14 - 34b6538 : Jyrki Vesterinen): The build passed. 20180321 07:36:01< travis-ci> Build details : https://travis-ci.org/shikadiqueen/wesnoth/builds/356209106 20180321 07:36:01-!- travis-ci [~travis-ci@ec2-54-166-113-143.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180321 07:36:44<+discordbot1> @shadowm we disabled case-insensitive sorting on macOS 20180321 07:36:55<+discordbot1> Then why is it still in the release announcement? 20180321 07:37:03<+discordbot1> good point 20180321 07:42:25< irker760> wesnoth/wesnoth:master pentarctagon 0b28afa05c Adds a flag to globally enable or disabl AppVeyor: vs2015/Release Failed 20180321 07:42:26< irker760> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-2068 20180321 07:49:09<+discordbot1> "JPG screenshots" 20180321 07:49:23<+discordbot1> Seriously. The name of the format is JPEG, not JPG. 20180321 07:49:52<+discordbot1> Well, JPG is still commonly used because it's the most common file extension. 20180321 07:49:53<+discordbot1> Yes I know the most popular extension is .jpg thanks to Windows 9x. 20180321 07:50:06<+discordbot1> (Apparently some people have missed that it's not 1992 anymore.) 20180321 07:50:07<+discordbot1> But then you should be saying ".jpg screenshots". 20180321 07:51:43<+discordbot1> Also while I can understand that the home dir obfuscation is intended to aid with "privacy" concerns, it doesn't make for great parseability of any affected output. 20180321 07:52:23<+discordbot1> At least --config-path seems to be honest. 20180321 07:52:35<+discordbot1> Yeah, that kind of change shouldn't have been rushed in at the last minute without time for proper discussion. 20180321 07:52:57<+discordbot1> Also, is it necessary to keep the old main menu music in the title screen playlist? :p 20180321 07:53:35<+discordbot1> Just had it blaring right after the new main menu music. 20180321 07:55:09<+discordbot1> And the game chooses' Aleksi's battle theme as the first track for AToTB scenario 1... the most dull battle scenario in existence. 20180321 07:55:34<+discordbot1> narrows eyes 20180321 07:55:38<+discordbot1> lol 20180321 07:55:55<+discordbot1> I like the frank and open opinions you put here; incredibly refreshing πŸ˜ƒ 20180321 07:56:33<+discordbot1> yes the format (technically) is .jpeg but we abbreviated it to .jpg because all extensions must be only 3 characters so says our dear overlord Microsoft of the Windows fame πŸ˜› 20180321 07:56:47<+discordbot1> That's been untrue since Windows 95. 20180321 07:57:01<+discordbot1> (Unless you forcefully disable long filename support.) 20180321 07:57:02<+discordbot1> The three-character file extension limit last existed in Windows 3.x. 20180321 07:58:38<+discordbot1> uh 20180321 07:58:39<+discordbot1> http://warp.povusers.org/grrr/short_filename_extensions.html 20180321 07:59:30<+discordbot1> And I beleive the idea didn't come from Microsoft. IIRC DEC did it years before. 20180321 07:59:31-!- discordbot1 [~discordbo@baldras.wesnoth.org] has quit [Remote host closed the connection] 20180321 07:59:40-!- discordbot2 [~discordbo@baldras.wesnoth.org] has joined #wesnoth-dev 20180321 07:59:42-!- mode/#wesnoth-dev [+v discordbot2] by ChanServ 20180321 07:59:42<+discordbot2> https://superuser.com/questions/315393/there-is-a-length-limit-to-windows-file-extensions 20180321 07:59:49<+discordbot2> I assume it came from CP/M first. 20180321 07:59:49<+discordbot2> convention and tradition, basically 20180321 07:59:55<+discordbot2> nah, it was mostly from DOS 20180321 08:00:02<+discordbot2> DOS had the 8 character limits 20180321 08:00:06<+discordbot2> and 3 character file extensions 20180321 08:00:10<+discordbot2> CP/M is DOS' direct predecessor. 20180321 08:00:16<+discordbot2> ouch 20180321 08:00:22<+discordbot2> CP/M didn't impose those limits though 20180321 08:00:33<+discordbot2> oh, here it is 20180321 08:00:48<+discordbot2> Wikipedia seems to believe otherwise. 20180321 08:00:49<+discordbot2> ok so there's a limit of 255 characters total on directory path and file name and extension combined 20180321 08:00:52<+discordbot2> please dont 20180321 08:00:54<+discordbot2> 🀦🏾 20180321 08:01:05<+discordbot2> Wikipedia is not a valid source and if you try to claim it is such I will LAUGH at you 20180321 08:01:12<+discordbot2> if you want another source, feel free 20180321 08:01:16<+discordbot2> but damn no wikipedia plz >_> 20180321 08:01:18<+discordbot2> Hm, okay. I feel laughed at. 20180321 08:01:24<+discordbot2> Thanks for that. 20180321 08:01:31<+discordbot2> Well, no one has been suggesting a .jointphotographicexpertsgroup extension. πŸ˜‰ 20180321 08:01:40<+discordbot2> Also, the 255 path limit is actually artificial and specific to the Win32 API. 20180321 08:01:43<+discordbot2> wikipedia is trash, it is not a valid source nor should be counted as one for any reliable piece of information that cannot be easily found elsewhere 20180321 08:01:51<+discordbot2> >.> 20180321 08:01:57<+discordbot2> anyhoo 20180321 08:02:04<+discordbot2> getting back to the file extension thing 20180321 08:02:19<+discordbot2> Internally everything using the Native API (including the kernel itself and the filesystem drivers) have a 32767 limit. 20180321 08:02:22<+discordbot2> NTFS file system actually imposes the 255 limit thing 20180321 08:02:31<+discordbot2> I don't remember if the size unit is characters or wide characters or bytes. 20180321 08:02:40<+discordbot2> No, it's not a NTFS limit. NTFS allows much longer paths. 20180321 08:02:59<+discordbot2> You can actually break out of the limit using the \\.\ prefix with the right APIs. 20180321 08:03:00<+discordbot2> It's the Win32 API that limits paths to 259 characters by default. 20180321 08:03:03<+discordbot2> Damn. I never realized computer operating systems didn't exist until Microsoft stol .. er ,, invented them. 20180321 08:03:19<+discordbot2> Or \??\. 20180321 08:03:33<+discordbot2> 4. Attribute - $FILE_NAME (0x30) 4.1. Overview This Attribute stores the name of the file attribute anl is always resident. As defined in $AttrDef, this attribute has a minimum size of 68 bytes and a maximum of 578 bytes. This equates to a maximum filename length of 255 Unicode characters. 20180321 08:03:34<+discordbot2> One or both of the two, I don't remember which. There are some tricky semantics behind both. 20180321 08:03:48<+discordbot2> Last year I made Windows BSOD by creating too long of a path. Couldn't even delete what it did create through explorer - I had to use robocopy or something like that. 20180321 08:03:52<+discordbot2> it is NTFS, @jyrkive 20180321 08:03:57<+discordbot2> (Were we talking about paths or filenames?) 20180321 08:04:05<+discordbot2> source: http://dubeyko.com/development/FileSystems/NTFS/ntfsdoc.pdf 20180321 08:04:08<+discordbot2> filenames 20180321 08:04:39<+discordbot2> there's also some older programs that break if the paths are too long...but that's more of a legacy issue 20180321 08:04:39<+discordbot2> "ok so there's a limit of 255 characters total on directory path and file name and extension combined" 20180321 08:04:44<+discordbot2> That's literally what you said. 20180321 08:04:51<+discordbot2> I did say that and I believe I mixed it up πŸ˜„ 20180321 08:04:55<+discordbot2> meant to say just file name... 20180321 08:04:57<+discordbot2> been a while 20180321 08:05:16<+discordbot2> GDI I KEEP PRESSING CTRL+R ON DISCORD INSTEAD OF WESNOTH 20180321 08:05:18<+discordbot2> @Pentarctagon just boot from a thumb drive and fix the hard drive when you do that. 20180321 08:05:43<+discordbot2> oh btw 20180321 08:05:47< DeFender1031> It's not entirely true that 8.3 died with DOS. IIRC, the later "you can make longer filenames" stuff implemented "long_filename_that_is_long.ext" as a lookup table where the actual file name in the filesystem is "long_f~1.ext" and just displayed it using the longer name from the lookup table for most things. 20180321 08:05:50<+discordbot2> any chance of a 'portable' wesnoth? 20180321 08:06:16<+discordbot2> I'm getting dΓ©jΓ  vu all of a sudden. 20180321 08:06:35<+discordbot2> DeFender1031: that's called LFN, and it's used with FAT16 and FAT32 file systems. 20180321 08:06:45<+discordbot2> NTFS has native support for long file names. 20180321 08:06:47<+discordbot2> NTFS is long file names first, short file names as an addition. 20180321 08:06:52< DeFender1031> @jyrkive yeah, exactly. 20180321 08:07:11<+discordbot2> https://support.microsoft.com/en-us/help/894630/you-receive-a-the-specified-file-or-folder-name-is-too-long-error-mess 20180321 08:07:13< DeFender1031> But FAT32 isn't dead. 20180321 08:07:16<+discordbot2> It is possible to ignore NTFS' short filenames or disable generation of such entirely (in particular on Linux). 20180321 08:07:29<+discordbot2> https://answers.microsoft.com/en-us/windows/forum/windows_7-files/how-do-i-change-how-long-file-folder-names-can-be/cfb58a04-1efc-41ac-a4df-178825ca7ae1 20180321 08:08:01<+discordbot2> Did you know that Microsoft's support platforms are user-oriented? 20180321 08:08:13<+discordbot2> Better: https://betanews.com/2016/05/29/long-paths-windows-10/ 20180321 08:08:15<+discordbot2> The 3-char filename extension was introduced around 1962. 20180321 08:08:23<+discordbot2> There's a lot more going on behind the scenes that they won't tell you through those because they only concern software developers. 20180321 08:08:35<+discordbot2> https://msdn.microsoft.com/en-us/library/windows/desktop/aa365247(v=vs.85).aspx#maxpath 20180321 08:09:10<+discordbot2> go down to 'maximum path length limitation' 20180321 08:09:37<+discordbot2> so yes, 260 is (technically) what it should support, but is around 255 due to various api quirks and hidden bits and bytes πŸ˜› 20180321 08:09:48<+discordbot2> and due to unicode extension we can have long path names like normaly 20180321 08:10:05<+discordbot2> rolls eyes 20180321 08:10:11<+discordbot2> The Windows API has many functions that also have Unicode versions to permit an extended-length path for a maximum total path length of 32,767 characters. This type of path is composed of components separated by backslashes, each up to the value returned in the lpMaximumComponentLength parameter of the GetVolumeInformation function (this value is commonly 255 characters). 20180321 08:10:12<+discordbot2> Still, you're likely doing something wrong if you somehow manage to exceed 255 characters. 20180321 08:10:21<+discordbot2> so basically 20180321 08:10:35<+discordbot2> each individual part (folder name) of a path can be only 255 unicode characters 20180321 08:10:47<+discordbot2> but if you use backslashes and additional folders you can get up to that 32k+ number 20180321 08:10:47<+discordbot2> You know, I just realized why I was getting a sense of dΓ©jΓ  vu. 20180321 08:11:43<+discordbot2> unless you do a registry hack to disable 8.3 filenames, Windows still stores them in the file system 20180321 08:12:01<+discordbot2> Yes, but that's irrelevant for the particular point that started this discussion. 20180321 08:12:26<+discordbot2> why are extensions only 3 characters 20180321 08:12:31<+discordbot2> because tradition/convention 20180321 08:12:33<+discordbot2> Β―_(ツ)_/Β― 20180321 08:12:47<+discordbot2> They are not always "only" 3 characters, that was my whole point. 20180321 08:12:50<+discordbot2> Thanks for derailing it. 20180321 08:12:56<+discordbot2> A "convention" that no one is trying to maintain. 20180321 08:13:06<+discordbot2> eh, seems to be pretty maintained to me >_> 20180321 08:13:14<+discordbot2> Microsoft is not advising developers to stick to three-character extensions. 20180321 08:13:21<+discordbot2> http://filext.com/ 20180321 08:13:29<+discordbot2> .torrent is an exception 20180321 08:13:31<+discordbot2> If you pay closer attention you'll notice that even Microsoft themselves use extensions larger than 3 characters in key system components and software. 20180321 08:13:33<+discordbot2> so is .html 20180321 08:13:40<+discordbot2> And .jpeg. 20180321 08:13:55<+discordbot2> yup 20180321 08:14:08<+discordbot2> many places still abbreviate HTML to just HTM and JPEG to JPG 20180321 08:14:12<+discordbot2> Β―_(ツ)_/Β― 20180321 08:14:31<+discordbot2> Like I said at the very beginning, just a bad habit carried over from the DOS-based Windows days. 20180321 08:14:39<+discordbot2> I agree with that 20180321 08:14:50<+discordbot2> there's no real necessity for it and there's only so many file extensions before we run out of space 20180321 08:15:01<+discordbot2> Reminds me of some of the files in the game Starbound, which went rather silly lengths the other direction, resulting in files such as "flashlight.flashlight". 20180321 08:15:03<+discordbot2> can't have conflicting extensions and make weird compatibility errors 20180321 08:15:19-!- travis-ci [~travis-ci@ec2-54-166-113-143.compute-1.amazonaws.com] has joined #wesnoth-dev 20180321 08:15:20< travis-ci> wesnoth/wesnoth#17077 (master - 00dac71 : Sofartin): The build has errored. 20180321 08:15:20< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/356217929 20180321 08:15:20-!- travis-ci [~travis-ci@ec2-54-166-113-143.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180321 08:16:06<+discordbot2> You mean "carried over from the DEC PDP-11 days." If you're going to complain about JPEG vs JPG you should complain about PDP-11 vs DOS 20180321 08:17:06<+discordbot2> with 3 letters, there are 18,278 different combinations before you start running out of room 20180321 08:17:25<+discordbot2> 4 letters ups that to 475,254 combos 20180321 08:17:28<+discordbot2> A bigger problem is the need to abbreviate in the first place. 20180321 08:17:42<+discordbot2> You can quite intuitively know that .jpeg stands for JPEG. 20180321 08:17:48<+discordbot2> For .jpg, you have to remember it. 20180321 08:18:24<+discordbot2> ANYWAY THE POINT WAS 20180321 08:18:46<+discordbot2> After the string freeze I'm not going to be able to resist the urge to relabel that string in the version info dialog to "JPEG support". 20180321 08:18:59<+discordbot2> *string freeze lift 20180321 08:19:24<+discordbot2> Go for it. 20180321 08:19:29<+discordbot2> That change should be trivially pofixable so I won't bother many translators. 20180321 08:19:42<+discordbot2> This reminds me I should be playing in Spanish... 20180321 08:20:19<+discordbot2> So uh, the title screen music keeps playing when reloading a scenario after startup? 20180321 08:20:33<+discordbot2> On 1.12 and earlier it'd switch back to the scenario's playlist immediately. 20180321 08:20:40<+discordbot2> so it IS a bug 20180321 08:20:42<+discordbot2> Didn't I see a bug report about that? 20180321 08:21:00<+discordbot2> I hope so, because this is exactly the kind of cosmetic thing that matters a lot for my campaigns. πŸ˜› 20180321 08:21:15<+discordbot2> @Tad Carlucci yes, but jyrki said it had been 1.13 behavior since forever that the current track finish when reloading a save 20180321 08:22:21<+discordbot2> Yep, just took 5 seconds to test 1.12, it switches back to the scenario's playlist instantly. 20180321 08:22:30<+discordbot2> Ah, yes. And I remember thinking that sounded reasonable even if it's a regression 20180321 08:22:33<+discordbot2> I wasn't imagining things. 20180321 08:22:39<+discordbot2> Ugh. 20180321 08:22:43<+discordbot2> so neither was I! 20180321 08:22:50<+discordbot2> Developer mentality at work there. 20180321 08:23:00<+discordbot2> For me, as a content creator, this kind of regression pisses me off to no end. 20180321 08:23:29<+discordbot2> I take the time to design every little detail in my campaigns to suit my needs, so when mainline developers break stuff and then claim it isn't important I feel like they don't want me producing content for Wesnoth in the first place. 20180321 08:23:30<+discordbot2> For me, the very first version of Wesnoth I ever player was 1.13.4+dev (self-compiled). 20180321 08:23:43<+discordbot2> So, of course I assumed it to be expected behavior. 20180321 08:24:04<+discordbot2> Β―\_(ツ)_/Β― 20180321 08:24:13<+discordbot2> ....geez. Broke that long ago? 20180321 08:24:41<+discordbot2> Personally, I don't hold with technology making noise so I generally cut the vocal cords and enjoy blessed silence. 20180321 08:24:56<+discordbot2> That's exactly what I'm talking about. 20180321 08:25:21<+discordbot2> This kind of stuff, BTW, is one reason why bug fixing should always have the highest priority. 20180321 08:25:30<+discordbot2> Bugs should be fixed immediately when they appear. 20180321 08:25:45<+discordbot2> I tried searching for open issues with "music" in them but none of the results seem relevant. 20180321 08:25:47<+discordbot2> Only then you can move back to new features or whatever. 20180321 08:26:45<+discordbot2> https://github.com/wesnoth/wesnoth/issues/2602 20180321 08:27:11<+discordbot2> Somehow that didn't come up in a search with is:issue is:open music 20180321 08:27:19<+discordbot2> Oh. It's closed. 20180321 08:27:20<+discordbot2> Because it isn't open. 20180321 08:27:21<+discordbot2> No wonder. 20180321 08:32:42<+discordbot2> @jyrkive it may be ideal to just say 'fix bugs and nothing else' but the reality of developing a game (or any software) says otherwise 20180321 08:33:01<+discordbot2> there's always that balance between new features and bug fixing that any game dev has to deal with 20180321 08:33:03<+discordbot2> There, commented on it and readded the requisite labels. 20180321 08:33:15<+discordbot2> Not Blocker though, because as much as it irritates me, it's not technically one. 20180321 08:33:17<+discordbot2> At least when I break something, I fix it before moving on to anything else. 20180321 08:33:32<+discordbot2> I don't believe devs break things on purpose πŸ˜› 20180321 08:33:42<+discordbot2> that's a pretty cynical view especially for BfW 20180321 08:33:50<+discordbot2> if you were talking Steam trash then sure πŸ˜„ 20180321 08:34:39<+discordbot2> I was actually banned from the official Discord 'dev' server because I told them it was irresponsible of them to release alpha-quality patches without patch notes or warning to their end-users on the stable non-PTR builds 20180321 08:35:09<+discordbot2> their response? 'making changelogs takes too long and nobody cares anyways' 'who cares if we break it that is the only way they'll report bugs since nobody uses PTR version' 20180321 08:35:19<+discordbot2> >_> 20180321 08:35:39<+discordbot2> I persisted in saying it was irresponsible for devs to do that and they didn't take kindly to my criticism 🀣 20180321 08:35:55<+discordbot2> Of course we aren't adding bugs just for fun. My point is just that developers should never ignore new regressions. 20180321 08:36:19<+discordbot2> In this case, it has been long lost when the scenario music change occurred and why. 20180321 08:36:21<+discordbot2> so if you ever wonder why discord randomly breaks it is because the discord devs purposefully have been pushing out alpha-quality patches to force-break things so users would report the bugs and get them fixed faster 20180321 08:36:25<+discordbot2> I finished the first scenario of Two Brothers on turn 8. 20180321 08:36:37<+discordbot2> Am I a Wesnoth pro yet 20180321 08:36:46<+discordbot2> If fixing it became a priority immediately when it was noticed, it would have been much easier to fix. 20180321 08:37:04<+discordbot2> I'm more along the side of 'fix bugs until things are stable and not game-breaking or blocking' and then keep adding in new content/features as time permits while testing them as much as feasible 20180321 08:37:32<+discordbot2> (maybe cause people just mute the music and play it in the background) 20180321 08:37:34< irker760> wesnoth/wesnoth:master Charles Dang f081a1c133 Revert "Convert uses of boost::regex and AppVeyor: vs2017/Release Failed 20180321 08:37:35< irker760> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-master-1776 20180321 08:37:38<+discordbot2> Right now we don't know what the logic of the code was supposed to be. (Having read it, the code very much looks like continuing to play the previous music is intentional.) 20180321 08:37:55<+discordbot2> Same thing with the unit test failures in master, BTW. 20180321 08:38:23<+discordbot2> find and isolate the regression and track the patch/patches that implemented it and fix it from there? 20180321 08:38:52<+discordbot2> I can tell that they were definitely broken in accelerated_rendering in July 2017 (at that time, Vultraz made a commit that broke them, if they weren't broken already). 20180321 08:39:22<+discordbot2> But because he ignored the unit test failure, we no longer know in which commit they regressed or why. 20180321 08:39:57<+discordbot2> I have to fall back to a slower method - comparing the behavior between 1.13.12+dev and master, and finding out the differences. 20180321 08:40:22<+discordbot2> part of the issue the tests are having is with event contexts. 20180321 08:41:04<+discordbot2> The test I was debugging fails because the game believes that the game map area is 0x0 pixels. 20180321 08:41:40<+discordbot2> again, very possibly to do with the video surface being null. 20180321 08:41:54<+discordbot2> No, it's not that. 20180321 08:42:14<+discordbot2> We query the theme for the size of the game map object. Nothing to do with rendering. 20180321 08:43:21-!- travis-ci [~travis-ci@ec2-54-166-113-143.compute-1.amazonaws.com] has joined #wesnoth-dev 20180321 08:43:22< travis-ci> wesnoth/wesnoth#17078 (master - f081a1c : Charles Dang): The build failed. 20180321 08:43:22< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/356220796 20180321 08:43:22-!- travis-ci [~travis-ci@ec2-54-166-113-143.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180321 08:44:08<+discordbot2> I feel like while I start working on the 1.14 press release announcement template it might be a good idea to start listing items on a wiki page as some sort of general outline for what vultraz or whoever will actually write later on. 20180321 08:44:27<+discordbot2> I like the village ToD effects in particular. 20180321 08:44:57<+discordbot2> It's not a game changer by any means, but it's one of those little things that make the game feel more polished and alive. 20180321 08:45:35<+discordbot2> thank doofus-01 for those 20180321 08:45:52<+discordbot2> (Of course the expectation is that every developer or artist involved in 1.13.x will add to that page.) 20180321 08:46:07<+discordbot2> (Time and presence permitting. I know some of them left.) 20180321 08:46:09<+discordbot2> also thank him for the amazing new gate graphics 20180321 08:46:19<+discordbot2> i know you wanted new gate graphics 20180321 08:46:27<+discordbot2> Yeah, can't wait to use those some day next year maybe. 20180321 08:46:53<+discordbot2> (how do I know you wanted new gate graphics? Because I clearly remember reading a comment you left somewhere saying "Gate graphics suck") 20180321 08:47:18<+discordbot2> I'm pretty sure I said that a million times. 20180321 08:47:56<+discordbot2> It was hard to draw some maps without acknowledging the fact that the minimal graphics I had weren't up to the task. 20180321 08:48:50<+discordbot2> Even harder than ignoring the Escheresquian geometries that resulted from certain common terrain configurations. 20180321 08:49:42<+discordbot2> and glorious carpets 20180321 08:50:24<+discordbot2> I bet Uria would demand anyone who summon her put down a carpet first. 20180321 08:50:26<+discordbot2> I don't think the enemies in my campaigns believe in the existence of carpets. 20180321 08:51:02<+discordbot2> Nah. 20180321 08:51:02<+discordbot2> oh yeah, SCONCES! 20180321 08:51:06<+discordbot2> She'd just float around. 20180321 08:51:12<+discordbot2> Lit Stone Wall is deprecated 20180321 08:51:14<+discordbot2> Being all threatening and crap. 20180321 08:51:21<+discordbot2> because we have proper sconces now 20180321 08:51:31<+discordbot2> including ground sconces 20180321 08:51:52-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev 20180321 08:52:10<+discordbot2> I still haven't decided exactly how I'm going to go about the terrain changes this time, since I still want to keep the 1.12 version around for a while. Probably some #ifver+#ifhave magic in the {MAP} macro. 20180321 08:52:15<+discordbot2> Or whatever the macro was called. 20180321 08:52:27<+discordbot2> It's been what, a year since I last touched my campaigns? 20180321 08:52:29<+discordbot2> 1.12 version of? 20180321 08:52:54<+discordbot2> Of the campaigns. 20180321 08:52:56<+discordbot2> oh 20180321 08:53:10<+discordbot2> Which means loading the 1.12 versions of the maps on 1.12 and the 1.14 versions on 1.14. 20180321 08:53:28<+discordbot2> some stuff looks a bit different 20180321 08:53:34<+discordbot2> I know. 20180321 08:53:35<+discordbot2> especially with the adjustment of the walls 20180321 08:53:42<+discordbot2> which means you need to reposition all the banners 20180321 08:53:45<+discordbot2> The Hewn Wall in particular is... a whole new thing. 20180321 08:53:45<+discordbot2> and moss overlays 20180321 08:54:05<+discordbot2> yeah, the hewn wall was completely replaced... 20180321 08:54:11<+discordbot2> i looked at some of your maps 20180321 08:54:13<+discordbot2> it looks ok 20180321 08:54:20<+discordbot2> but you'll have to make a judtment clallll 20180321 08:54:24<+discordbot2> I only used it in the first place because I think Eleazar had more elaborate plans for it that nver came to fruition. 20180321 08:54:32<+discordbot2> (We're talking back in the 1.9.x days here.) 20180321 08:54:56<+discordbot2> heh, i was just gonna go look in the editor before remembering i can't because the editor crashes 20180321 08:55:31<+discordbot2> The 1.14 editor is a-ok here. 20180321 08:55:34<+discordbot2> master 20180321 08:55:47<+discordbot2> Well, that's your fault. :p 20180321 08:55:50<+discordbot2> it is 20180321 08:55:54<+discordbot2> i'll get to it eventually 20180321 08:56:06<+discordbot2> And also, as release manager, if you don't keep a 1.14 working tree and builds of your own you should be fired. 20180321 08:56:37<+discordbot2> Isn't the editor used for map development? 20180321 08:56:58<+discordbot2> Git master must remain in a state where developing the game is possible. 20180321 08:57:53<+discordbot2> Look, there is going to be a period wherein a whole bunch of stuff is broken and unusable. 20180321 08:58:12<+discordbot2> Sure. 20180321 08:58:24<+discordbot2> Then make sure you can use 1.14 as well. 20180321 08:58:29<+discordbot2> Unless you personally intend to develop maps on 1.15, please stop repeating "we must fix master!" 20180321 08:58:39<+discordbot2> It will be fixed. 20180321 08:59:03<+discordbot2> You realize that master being broken slows down progress? 20180321 09:00:22<+discordbot2> (Whoah, the elves are in cahoots with the undead in Two Brothers now??) 20180321 09:00:26<+discordbot2> I am doing the best I can. I've spent all day on Wesnoth today. 20180321 09:00:42<+discordbot2> OK, that's good. 20180321 09:01:00<+discordbot2> I just have the issue with you saying that the editor will be fixed eventually. 20180321 09:01:11<+discordbot2> Using that word suggests that it's low priority to you. 20180321 09:01:12<+discordbot2> My priority is the main game 20180321 09:01:42<+discordbot2> Oh, so that's what you meant. 20180321 09:01:50<+discordbot2> figure out the proper method for the ingame un in-game, then do the editor 20180321 09:01:58<+discordbot2> Okay, in that case it's fine (and indeed, expected). 20180321 09:03:46<+discordbot2> Anyway I hope the announcement outline idea I posted doesn't go ignored because the effort has to come from everyone, including me. 20180321 09:04:40<+discordbot2> Don 20180321 09:04:48<+discordbot2> Even though my contributions this time around amount to a single dialog. 20180321 09:06:16< octalot> I'm curious, why not make a feature-branch pointing to the current master, and force-rewind master to a non-broken version? 20180321 09:06:40<+discordbot2> That would cause merge conflicts. 20180321 09:06:47< galegosimpatico> Don't push the Mikado. Revert every offending rev, squash it down to a single rev, and let Vultraz do the things with calm on a feature branch. 20180321 09:07:19<+discordbot2> Don't push the what now? 20180321 09:08:21< galegosimpatico> Don't let unimportant decorations lose your focus on the important stuff. 20180321 09:09:44<+discordbot2> Unfortunately, the underlying change isn't an "unimportant decoration", but a rewrite of a large part of the engine. 20180321 09:10:05<+discordbot2> If it was in a branch, it would be difficult to merge that branch later. 20180321 09:11:22<+discordbot2> (I think the GUI2 sidebar would have been small enough that it could have lived in a branch, though. In fact, my original plan was to implement that sidebar in the OpenGL branch, and continue to use the old sidebar in master.) 20180321 09:11:39<+discordbot2> (Until Vultraz started implementing that GUI2 sidebar himself.) 20180321 09:14:06< octalot> I confess I haven't looked at this change, but generally I've found that any large branch will have changes that can be cherry-picked (or edited and cherry picked), which can merge earlier 20180321 09:14:50<+discordbot2> That doesn't really help. The main problem is all the changes which are too large to cherry-pick. 20180321 09:15:19< octalot> So the whole branch ends up getting merged in a different order to the order it was written, but the code is cleaner for doing that, and the unit tests for master can be kept working 20180321 09:15:41< irker760> wesnoth/wesnoth:1.14 Iris Morelle 2879d10dc5 wam: Update description and name for Mod AppVeyor: 2/4 builds failed 20180321 09:15:42< irker760> Details vs2017/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-1.14-1768 20180321 09:15:43< irker760> Details vs2015/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-1.14-2064 20180321 09:16:11<+discordbot2> Unit tests aren't testing graphics or the sidebar. 20180321 09:16:39<+discordbot2> They have broken as an unintentional side effect of some change (right now, I have no idea which). 20180321 09:18:17< octalot> Ok, I'll have to wait until the weekend for time to look at the branch and see if my ideas fit it 20180321 09:19:25<+discordbot2> It's starting to feel like people care more about a green checkmark on travis than actually progressing 20180321 09:19:26<+discordbot2> @bumbadadabum Maybe you want to look into the bandit stuff here: https://forums.wesnoth.org/viewtopic.php?p=624932#p624932 20180321 09:19:27<+discordbot2> 😐 20180321 09:19:53<+discordbot2> @Vultraz Your POV is completely wrong. 20180321 09:20:13<+discordbot2> The reason for automated testing is that it notifies us when commits break things. 20180321 09:20:32<+discordbot2> We do not have that safety net when the tests are failing. 20180321 09:20:54<+discordbot2> Right now, if a commit breaks something, we won't know about it, since the tests are already failing. 20180321 09:21:35<+discordbot2> fair enough 20180321 09:21:45<+discordbot2> Therefore, once I have finally fixed this "0x0 game area" bug, it's very much possible that I find another problem. 20180321 09:21:58<+discordbot2> And I'll have no idea in which commit it broke. 20180321 09:22:14<+discordbot2> Therefore I need to spend more effort, and time, to fix it. 20180321 09:22:30< irker760> wesnoth: Charles Dang wesnoth:master 94f172dbdd47 / src/display.cpp: Display: removed unnecessary getSurface check https://github.com/wesnoth/wesnoth/commit/94f172dbdd47286abdfbb3a1dae3969135d73c1f 20180321 09:22:31<+discordbot2> In the meantime, other developers may make more commits and break more stuff without knowing it. 20180321 09:22:40<+discordbot2> Just like the music bug. πŸ˜› 20180321 09:23:05<+discordbot2> The workload gets larger and larger. Recursively. 20180321 09:23:34< galegosimpatico> People want to build. I could write a couple of ideas I want to request pull of, I just have to put them on hold until master is more calm, it's just that. I understand the points clarified by Jyrki. The last thing you should do is to feel hurried because people complain. Be cool. 20180321 09:23:42<+discordbot2> I've checked the commit log for sound.cpp... I see nothing from the past 3 years that looks like an obvious culprit 20180321 09:24:02<+discordbot2> I'll probably take a stab at tracking down the source of the music bug. I strongly doubt it's an issue in the sound module, though. 20180321 09:24:19<+discordbot2> It's probably something related to the play*controller maze. 20180321 09:24:47<+discordbot2> @shadowm from my debugging, it seems the only reason it stops at all when starting a name game is that the {SCENARIO MUSIC} macro has immediate=yes 20180321 09:25:13<+discordbot2> Did that stop being the case? 20180321 09:25:36<+discordbot2> Nope, seems to still be the case in master. 20180321 09:25:38<+discordbot2> new* 20180321 09:26:02<+discordbot2> Is the attribute no longer written along with the playlist in saved games? 20180321 09:26:14<+discordbot2> consequently, the lack of immediate=yes is why it continues when simply reloading 20180321 09:26:35<+discordbot2> the attribute it not written that I can tell, but I tried adding it and it made no difference 20180321 09:26:53<+discordbot2> Then that's not the answer. 20180321 09:27:06<+discordbot2> It might also be two bugs. 20180321 09:27:24<+discordbot2> First, the attribute not being written. And second, the attribute being ignored for some reason. 20180321 09:27:58<+discordbot2> This is the music playlist written to the save in AToTB in version 1.12.x: https://pastebin.com/LeZBkL6d 20180321 09:27:58<+discordbot2> Again, it's very difficult to tell when so much time has passed that there may now be multiple bugs causing it. 20180321 09:28:03< irker760> wesnoth: ln-zookeeper wesnoth:master 3be2ff518781 / data/campaigns/Eastern_Invasion/scenarios/03_An_Unexpected_Appearance.cfg: EI S03: Changed Mal-Kallat back to male https://github.com/wesnoth/wesnoth/commit/3be2ff518781c7a7104e810707be64567cc26e27 20180321 09:28:04<+discordbot2> This is it in 1.14.12: https://pastebin.com/fMwkMhdU 20180321 09:28:35<+discordbot2> 1.14.12? 20180321 09:28:58<+discordbot2> Other than the tracks being changed a bit and the addition of the implicit shuffle=yes option at some point in 1.13.x, it's really the same. 20180321 09:29:02<+discordbot2> 1.13.12. You know what I mean. 20180321 09:29:54<+discordbot2> (As far as I remember Pent or someone who at the time looked like Pent to me submitted a PR adding the possibility to opt into not randomizing the playlist using shuffle=no, which is not a default value.) 20180321 09:30:02<+discordbot2> is it under [snapshot]? 20180321 09:30:08<+discordbot2> in the 1.12 save 20180321 09:30:15<+discordbot2> In both. 20180321 09:30:21< irker760> wesnoth: ln-zookeeper wesnoth:1.14 2e43a5658884 / data/campaigns/Eastern_Invasion/scenarios/03_An_Unexpected_Appearance.cfg: EI S03: Changed Mal-Kallat back to male https://github.com/wesnoth/wesnoth/commit/2e43a5658884eaad1bcc144adbddd898d4335a5a 20180321 09:30:59<+discordbot2> @zookeeper don't forget to cherry-pick 20180321 09:31:09<+discordbot2> That should be the only one that counts as far as I understand, since the others are in the replay data (including the replay's prestart event). 20180321 09:31:33<+discordbot2> The game doesn't take the replay data into account unless you actually choose to watch the replay. 20180321 09:31:40<+discordbot2> my testing did confirm the music was found 20180321 09:31:44<+discordbot2> the playlist was cleared 20180321 09:31:49<+discordbot2> and the appropriate music added 20180321 09:31:57<+discordbot2> everything appears to work 20180321 09:32:04<+discordbot2> except the current track does not stop 20180321 09:32:28<+discordbot2> And again, looking at the code, that's what it appears to be designed to do. 20180321 09:33:18<+discordbot2> Anyway, I really need to sleep, later. 20180321 09:34:06<+discordbot2> I thought perhaps this was the issue 20180321 09:34:07<+discordbot2> https://github.com/wesnoth/wesnoth/commit/7dea2b92abad957004e57030963dcb0f458286ac 20180321 09:34:26<+discordbot2> but IIRC, reverting it breaks stuff spectacularly 20180321 09:34:50<+discordbot2> There is likely a lot of stuff that depends on it now. 20180321 09:35:06<+discordbot2> Probably it would have reverted cleanly when it was made. 20180321 09:35:22<+discordbot2> I can't even see what the bug is 20180321 09:36:04<+discordbot2> maybe archive.org has something.. 20180321 09:36:48<+discordbot2> ah, here 20180321 09:36:50<+discordbot2> https://web.archive.org/web/20170204102655/https://gna.org/bugs/index.php?23065 20180321 09:38:12<+discordbot2> well, reverting the commit does make the music stop on load 20180321 09:38:27<+discordbot2> but themain menu music is wrong 20180321 09:38:32<+discordbot2> it's a random track 20180321 09:39:12<+discordbot2> Even worse, reverting it would clearly break compatibility with UMC at this point. 20180321 09:39:42<+discordbot2> No. 20180321 09:39:46<+discordbot2> It's a 1.13-only change 20180321 09:40:05<+discordbot2> Some people have also developed UMC for 1.13. 20180321 09:40:33<+discordbot2> And they're told to expect things to break. Granted, right now we've supposedly made it so things NOT break 20180321 09:40:45<+discordbot2> πŸ˜’ 20180321 09:41:11<+discordbot2> This goes beyond "expect things to break". 20180321 09:41:39<+discordbot2> There has been a way to change the playlist without stopping the current track for over three years. 20180321 09:42:12<+discordbot2> And it would be changed just now, after the RC1?!?! 20180321 09:42:32<+discordbot2> Then we need to figure out how to make it not apply when loading a save. 20180321 09:42:46<+discordbot2> Yes. That would work. 20180321 09:43:32<+discordbot2> anyway, unit test failures reduced wiht last commit 20180321 09:44:18<+discordbot2> Interesting. 20180321 09:44:40<+discordbot2> I expected that might happen 20180321 09:44:59<+discordbot2> That line of code had been removed on a_r but got back in during conflict resolution 20180321 09:49:54< irker760> wesnoth: Charles Dang wesnoth:master a9afee610447 / src/floating_label.cpp: Floating Label: don't attempt to render empty text https://github.com/wesnoth/wesnoth/commit/a9afee61044751d8edc92afa9d0a14ca7be97d4f 20180321 10:04:04<+discordbot2> @jyrkive I seem able to fix the sound bug by adding a force parameter to commit_music_changes that bypasses the if(current_track->play_once()) check 20180321 10:04:45<+discordbot2> not sure if the solution should somehow utilize immediate= in some form instead, though 20180321 10:05:05<+discordbot2> Shouldn't it be in the if(current_track) line instead? 20180321 10:05:21<+discordbot2> hm. perhaps. 20180321 10:06:20<+discordbot2> I think set_previous_track() should also be skipped when force-committing. 20180321 10:12:08<+discordbot2> looks like the number of "could not create floating label's text" errors has gone down 20180321 10:12:22<+discordbot2> still a bunch, but that's the only thing causing the remaining 8 wml tests to fail 20180321 10:13:08<+discordbot2> Eight? I had a much larger number of tests failing (due to printing an error message to the log because of the 0x0 gamemap). 20180321 10:13:38<+discordbot2> "160 out of 168 tests were correct." 20180321 10:15:48<+discordbot2> must've screwed something up with the text texture cache 20180321 10:16:43-!- travis-ci [~travis-ci@ec2-54-166-113-143.compute-1.amazonaws.com] has joined #wesnoth-dev 20180321 10:16:44< travis-ci> wesnoth/wesnoth#17082 (master - a9afee6 : Charles Dang): The build is still failing. 20180321 10:16:44< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/356265347 20180321 10:16:44-!- travis-ci [~travis-ci@ec2-54-166-113-143.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180321 10:18:09<+discordbot2> Interesting. Judging from Travis logs, looks like https://github.com/wesnoth/wesnoth/commit/94f172dbdd47286abdfbb3a1dae3969135d73c1f fixed the 0x0 gamemap issue. 20180321 10:18:26<+discordbot2> That commit made affected tests such as test_end_turn pass. 20180321 10:18:49<+discordbot2> That's why I was saying it likely had something to do with the video surface. I had been considering that bit of code 20180321 10:18:55<+discordbot2> (It may be worthwhile to wait for the AppVeyor build, though.) 20180321 10:19:28<+discordbot2> Well, I still have no idea how. 20180321 10:20:44< zookeeper> okay, so why have i been stuck at "Building terrain rules..." for over 2 minutes now... 20180321 10:21:38<+discordbot2> MacBook-Pro:MacOS user$ ./Wesnoth --fps ... Battle for Wesnoth v1.15.0-dev (a9afee6104-Modified) Started on Wed Mar 21 11:19:40 2018 ... Setting mode to 1440x878 Checking lua scripts... ok Floating point exception: 8 20180321 10:21:50< zookeeper> cpu keeps churning, memory use doesn't change one byte. must be stuck in a loop somewhere. 20180321 10:22:24<+discordbot2> what branch 20180321 10:23:12<+discordbot2> @hrubymar10 that doesn't look good 20180321 10:24:43< zookeeper> a week old 1.13 build. 20180321 10:24:44< vn971> I've built wesnoth master 3 hours ago, everything loads fine. 20180321 10:25:08< vn971> I don't use --fps though. 20180321 10:26:27< irker760> wesnoth/wesnoth:master Charles Dang 5145b009a5 Display: removed code to draw theme pane AppVeyor: 1/2 builds failed 20180321 10:26:27< irker760> Details vs2015/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-2065 20180321 10:26:38< irker760> wesnoth/wesnoth:1.14 ln-zookeeper 2e43a56588 EI S03: Changed Mal-Kallat back to male AppVeyor: vs2017/Release Failed 20180321 10:26:39< irker760> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-1.14-1779 20180321 10:30:30<+discordbot2> wesnoth:master 20180321 10:31:10< zookeeper> so does the gender of the speaker affect the translation of what they say in any/many languages? like if there are gender-specific "I" pronouns or something like that. 20180321 10:31:34<+discordbot2> I’ll give backtrace in 45 min 20180321 10:31:53<+discordbot2> Yes. 20180321 10:33:58<+discordbot2> Well, I'm not sure about the gender of the speaker 20180321 10:34:25<+discordbot2> I know gender definitely plays a role when translating who you're talking about. 20180321 10:34:59< zookeeper> that's why i asked what i asked, and not that. 20180321 10:36:29< zookeeper> anyway, indeed EI S05 is joyfully broken because of some problem with the bandits lua stuff. it's interesting how these conversions of WML to lua have a tendency to break. 20180321 10:37:07< galegosimpatico> I think there are some languages where the translation of some words can depend on the gender of the speaker. 20180321 10:39:59-!- octalot [~steve@213-225-3-81.nat.highway.a1.net] has quit [] 20180321 10:40:25-!- octalot [~steve@213-225-3-81.nat.highway.a1.net] has joined #wesnoth-dev 20180321 10:40:30< zookeeper> @vultraz, so uh, did you test whatever you did to Eastern_Invasion/lua/bandits.lua? 20180321 10:40:39<+discordbot2> no 20180321 10:40:53< zookeeper> because now, you know, "[spread_bandit_villages] is not supported" 20180321 10:41:44<+discordbot2> zookeeper, in many romance languages sentences like 'I am surprised' would translate differently. Past participles and adjectives referred to the speaker are gender-matched. I think verbs change in russian too. Not sure about languages where the pronoun changes 20180321 10:42:34<+discordbot2> @zookeeper and it says nothing else? 20180321 10:44:03< zookeeper> oh right, there was more in the log. yep, syntax error: Lua runtime error: campaigns/Eastern_Invasion/lua/bandits.lua:78: ')' expected near ']' 20180321 10:45:01<+discordbot2> oh yeah 20180321 10:45:08<+discordbot2> ok, that's just missing a ) 20180321 10:45:12< zookeeper> indeed 20180321 10:45:14<+discordbot2> to close string.format 20180321 10:45:27<+discordbot2> if you add that does it work 20180321 10:46:37< zookeeper> yes 20180321 10:48:09<+discordbot2> ok 20180321 10:48:11<+discordbot2> my fault, then 20180321 10:57:40<+discordbot2> @bumbadadabum Nevermind the previous. 20180321 11:04:05< zookeeper> vn971, it's kind of relevant what terrain(s) those blackbox village hexes have? 20180321 11:06:14< vn971> zookeeper: https://github.com/wesnoth/wesnoth/issues/2706 just some of the default ones. Those squares are for any scenario maps I previewed. Those on screenshot are Gs^Vh 20180321 11:06:29< zookeeper> ah, okay. 20180321 11:06:38< vn971> the link I pasted - I presume you're talking about this issue. 20180321 11:07:10< zookeeper> of course 20180321 11:07:58<+discordbot2> ragarding to --fps crash https://pastebin.com/JaJJ9NdZ 20180321 11:09:05< zookeeper> vn971, yeah, you're right... i wonder what's up with that. 20180321 11:11:05< zookeeper> it only affects villages. 20180321 11:13:22<+discordbot2> Don’t villages always show up as black dots? 20180321 11:16:20< zookeeper> doesn't strike me as making much sense, seeing how walls show up as black as well. in any case, they show up as inconsistent hex/square shapes anyway. 20180321 11:18:57< vn971> Vultraz, that's why I didn't mark my issue as "Bug". I'm not totally sure. But it's weird yes. 20180321 11:20:13< zookeeper> right, i guess the idea is for neutral villages to show up as black. and i agree there should be _some_ marker like that for villages, but perhaps black for neutral isn't the best choice. in any case, the shapes are still buggy, even if it's only particularly visible on a tiny map. 20180321 11:21:19< zookeeper> in-game, units show up as squares, but actually even they aren't centered on the hexes. 20180321 11:22:10< zookeeper> also, none of those rendering style toggles below the minimap in-game seem to do anything, except for zoom. 20180321 11:22:38-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20180321 11:23:00< vn971> zookeeper: indeed! And that's a serious enough issue it seems. 20180321 11:23:03< vn971> will you report? 20180321 11:23:08< zookeeper> sure 20180321 11:23:52< zookeeper> or rather i'll just comment more 20180321 11:29:02< vn971> zookeeper: if you mean buttons in Map Editor, then I think it's a separate issue. 20180321 11:31:00-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection] 20180321 11:44:08< irker760> wesnoth/wesnoth:master Charles Dang a9afee6104 Floating Label: don't attempt to render AppVeyor: vs2017/Release Failed 20180321 11:44:09< irker760> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-master-1780 20180321 11:52:10< irker760> wesnoth/wesnoth:master Charles Dang 29ee9a7150 Convert uses of boost::regex and related AppVeyor: 1/2 builds failed 20180321 11:52:11< irker760> Details vs2015/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-2067 20180321 11:52:12<+discordbot2> zookeeper: That's https://github.com/wesnoth/wesnoth/issues/2681 20180321 12:16:24-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20180321 12:42:24< irker760> wesnoth/wesnoth:master pentarctagon 0b28afa05c Adds a flag to globally enable or disabl AppVeyor: 1/2 builds failed 20180321 12:42:25< irker760> Details vs2015/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-2068 20180321 13:11:29<+discordbot2> how can I make Wesnoth pass --fps by default 20180321 13:11:44<+discordbot2> Why are you asking? 20180321 13:12:06<+discordbot2> I want debug Floating point exception: 8 crash 20180321 13:15:10<+discordbot2> Don't bother. The problem is obvious from your call stack. 20180321 13:15:20<+discordbot2> We aren't recording frametimes in master. 20180321 13:16:00<+discordbot2> To calculate the average FPS, the code divides the sum of all frametimes by the number of samples. Without any samples, it's division by zero. 20180321 13:16:28<+discordbot2> Here is the code that records frametimes in the 1.14 branch: https://github.com/wesnoth/wesnoth/blob/1.14/src/display.cpp#L1687 20180321 13:37:34< irker760> wesnoth/wesnoth:master Charles Dang f081a1c133 Revert "Convert uses of boost::regex and AppVeyor: 2/4 builds failed 20180321 13:37:35< irker760> Details vs2017/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-master-1776 20180321 13:37:36< irker760> Details vs2015/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-2072 20180321 13:48:50-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer] 20180321 13:48:58-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20180321 13:49:05-!- stikonas_ is now known as stikonas 20180321 14:16:40-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180321 14:16:46-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180321 14:27:44-!- octalot [~steve@213-225-3-81.nat.highway.a1.net] has quit [] 20180321 15:26:39< irker760> wesnoth/wesnoth:1.14 ln-zookeeper 2e43a56588 EI S03: Changed Mal-Kallat back to male AppVeyor: 2/4 builds failed 20180321 15:26:40< irker760> Details vs2017/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-1.14-1779 20180321 15:26:41< irker760> Details vs2015/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-1.14-2075 20180321 16:06:01-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev 20180321 16:29:20-!- Oebele [~quassel@143.177.58.202] has joined #wesnoth-dev 20180321 16:43:43< irker760> wesnoth: Charles Dang wesnoth:master d0f7bfbf379e / src/floating_label.cpp: Floating Label: added text to debug output if texture creation fails https://github.com/wesnoth/wesnoth/commit/d0f7bfbf379e37629fc742b988cff6a1e2059475 20180321 16:44:09< irker760> wesnoth/wesnoth:master Charles Dang a9afee6104 Floating Label: don't attempt to render AppVeyor: 2/4 builds failed 20180321 16:44:10< irker760> Details vs2017/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-master-1780 20180321 16:44:11< irker760> Details vs2015/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-2076 20180321 16:52:09-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!] 20180321 16:52:25-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20180321 17:09:26<+discordbot2> well it's definitely not trying to render bad text 20180321 17:35:39< irker760> wesnoth/wesnoth:master Charles Dang d0f7bfbf37 Floating Label: added text to debug outp AppVeyor: vs2017/Release Failed 20180321 17:35:40< irker760> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-master-1781 20180321 17:35:48< irker760> wesnoth: Charles Dang wesnoth:master 37dee4b1ff51 / src/ (22 files in 8 dirs): Display: removed all code related to report drawing https://github.com/wesnoth/wesnoth/commit/37dee4b1ff51d1fffa39983771bf55fd5cc8da05 20180321 17:35:51< irker760> wesnoth: Charles Dang wesnoth:master d6cc93be1bf5 / src/font/text.cpp: Font/Pango Text: added debug assertion https://github.com/wesnoth/wesnoth/commit/d6cc93be1bf53028f54f62e586e9b0fc81e426f8 20180321 17:35:54< irker760> wesnoth: Charles Dang wesnoth:master db64bc71265e / src/ (5 files in 4 dirs): Command Executor: removed set_button_state() https://github.com/wesnoth/wesnoth/commit/db64bc71265e78eeaf738afd5ae51e59492c49d1 20180321 17:38:12-!- Polsaker [hfmEwsuZBt@donger/wielder/Polsaker] has quit [Read error: Connection reset by peer] 20180321 17:40:21-!- Polsaker [QZXspfMPf7@donger/wielder/Polsaker] has joined #wesnoth-dev 20180321 17:40:31-!- travis-ci [~travis-ci@ec2-54-224-223-228.compute-1.amazonaws.com] has joined #wesnoth-dev 20180321 17:40:32< travis-ci> wesnoth/wesnoth#17083 (master - d0f7bfb : Charles Dang): The build is still failing. 20180321 17:40:32< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/356449625 20180321 17:40:32-!- travis-ci [~travis-ci@ec2-54-224-223-228.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180321 17:51:47< irker760> wesnoth: Charles Dang wesnoth:master eeffa82e2bf5 / src/actions/undo.cpp: Fixup 37dee4b (unused variable) https://github.com/wesnoth/wesnoth/commit/eeffa82e2bf56aecbdf83d2de23f5bcdfc054150 20180321 17:52:11-!- Oebele [~quassel@143.177.58.202] has quit [Remote host closed the connection] 20180321 17:55:32-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection] 20180321 17:59:44<+discordbot2> @zookeeper Japanese has many gendered first-person pronouns that depend on the speaker's age/preference/status, if it helps. And in addition to that, in Spanish pretty much every adjective and pronoun has gender inflections. 20180321 18:01:33<+discordbot2> Articles are gendered as well. 20180321 18:01:43<+discordbot2> e.g. "I am a healer" should be translated as either "soy un sanador" for a male speaker or "soy una sanadora" for a female speaker. 20180321 18:02:04<+discordbot2> "I am surprised" is either "estoy sorprendido" or "estoy sorprendida". 20180321 18:04:01<+discordbot2> I believe the Japanese translators deal with the first person pronoun issue by using more stilted and gender-neutral language throughout Wesnoth in certain cases, or so I was told ages ago. However, in Spanish (and other romance languages probably) there is no such a thing at all, so you'd have to use extremely artificial-sounding language to get around that kind of thing. 20180321 18:04:48<+discordbot2> (Source: I'm a native speaker of Spanish, and even was a translator and translation maintainer for a while.) 20180321 18:06:15<+discordbot2> All that said I believe back in the day there were very few cases in mainline where we knew we needed to write around gender inflections. But that was back in 2009, and several new campaigns have been added since then. 20180321 18:06:21< loonycyborg> gettext has ways to handle stuff like that though 20180321 18:06:32< loonycyborg> at least plurals 20180321 18:07:12<+discordbot2> On my own campaign there's a few cases of generic dialogue lines for arbitrary units where I preemptively added code to select either a regular message or an alternative with the female^ disambiguation prefix. 20180321 18:08:01<+discordbot2> Probably didn't cover all of them because at the end of the day I wrote them in English and for English, and I don't intend to try to translate them myself. 20180321 18:09:31<+discordbot2> @loonycyborg In this case we're talking about WML. I believe even the syntax for plurals didn't ever make it past the preliminary patch/discussion phase. 20180321 18:09:57<+discordbot2> (Correct me if I'm wrong, I might've missed it if it was added in 1.13.x.) 20180321 18:11:06<+discordbot2> I also don't think I've ever heard of GNU gettext being aware of the concept of gender. It's trickier than you might think because a single sentence can have multiple objects with different genders. 20180321 18:11:29< loonycyborg> well we're feeding all our wml through gettext extractor. 20180321 18:11:46<+discordbot2> Marking all of them in code seems like it'd require a lot of creative syntax. 20180321 18:12:23<+discordbot2> That doesn't mean anything. 20180321 18:12:46< loonycyborg> at least it would handle cases confined to particular string 20180321 18:12:52<+discordbot2> The point is that WML has its own syntax for translatable strings. It's not the same as if you were calling the gettext API in C directly because WML is designed so you cannot do that. 20180321 18:13:13< loonycyborg> though I expect it would be harder task to carry info about plurals/genders between different strings 20180321 18:13:36<+discordbot2> And to my knowledge the only syntax it handles is the plain dgettext() call. 20180321 18:13:52< loonycyborg> ah oh well 20180321 18:14:04<+discordbot2> (Through a combination of marking strings as translatable and establishing a textdomain by default or using the #textdomain directive.) 20180321 18:14:12< loonycyborg> I definitely remember that gettext handles plurals at least on C api level 20180321 18:14:19<+discordbot2> Yes, it does. 20180321 18:14:27<+discordbot2> But we were talking about gender, not plurals 20180321 18:14:28< loonycyborg> I wonder whether it has something for gender too 20180321 18:14:57<+discordbot2> I just pointed out that plurals were proposed for WML but I don't think the proposal made it anywhere because the syntax was a bit cumbersome. 20180321 18:15:09-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity] 20180321 18:15:55<+discordbot2> The patch was from a translator and on the now defunct Gna.org tracker too. 20180321 18:17:00<+discordbot2> What I do see and did correctly recall made it at some point is plural support in the wesnoth.textdomain Lua API call. Version 1.13.6. 20180321 18:17:46<+discordbot2> The problem is, most WML coders are not big fans of Lua (it is kind of a big context switch for a brain reading the code after all). 20180321 18:38:34< irker760> wesnoth/wesnoth:master Charles Dang eeffa82e2b Fixup 37dee4b (unused variable) AppVeyor: vs2017/Release Failed 20180321 18:38:35< irker760> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-master-1783 20180321 18:46:59<+discordbot2> Also you shouldn't try to pass es_ES as es_MX and vice versa. 20180321 18:47:43<+discordbot2> There's only one Spanish translation here and it's closer to mainland Spanish than anything else. 20180321 18:48:19<+discordbot2> Considering how it's been chronically understaffed I don't think a Latin American Spanish translation would be the best use of anyone's time. 20180321 18:50:21<+discordbot2> At least since this is a text-only game, the syntax differences are not too much of a deal-breaker. Plus in this country there are many books for which we only get the vanilla Spanish translations, so we've just learned to deal with that kind of crap. 20180321 18:51:02<+discordbot2> Give us an European Spanish dub of a TV show though and people will throw a fit. 20180321 18:54:58<+discordbot2> It's a lot more troublesome than for instance en_US vs en_UK. 20180321 18:56:48<+discordbot2> βœ‹πŸΎ 20180321 18:57:12<+discordbot2> for now, consider duplication of es_ES and es_MX until someone can come along to volunteer to maintain the translation separately? 20180321 18:57:24<+discordbot2> It's not necessary. 20180321 18:57:25<+discordbot2> there are minor differences but they are there 20180321 18:57:51<+discordbot2> There is so much text that adding another translation would be a massive task. 20180321 18:57:56< vn971> I have an scons question if you don't mind. How can I build/install wesnoth to a non-standard location? I'm currently using `scons destdir=my_dir install` and it works almost as I would want to, except one thing. I have to specify --data-dir when running the binary. By default it looks up on /share/wesnoth 20180321 18:58:04<+discordbot2> oh wait...the translation isn't already in the game? 20180321 18:58:08<+discordbot2> I already said it's not worth the effort with an understaffed translation team. 20180321 18:58:38<+discordbot2> I've always wondered how Wesnoth got translated into so many languages in the first place πŸ˜„ 20180321 18:58:58<+discordbot2> Yes, the Spanish translation is already in the game and near 100% completion, but it's like one and a half people working on it. 20180321 18:59:32<+discordbot2> And every development cycle one way or another the devs find a way to set it back to below 40% of the campaign catalogues. 20180321 19:00:21<+discordbot2> my Spanish isn't super-perfect or I'd volunteer to translate the Spanish version πŸ˜„ 20180321 19:00:32<+discordbot2> I'd probably cause a heart-attack in native speakers πŸ˜„ 20180321 19:01:34<+discordbot2> In general, it's a bad idea to translate if you're not a native speaker of the target language. 20180321 19:02:11<+discordbot2> Non-native speakers pretty much can't use the full expressive power of the language, resulting in, at best, fairly stiff and monotonic writing. 20180321 19:03:18< irker760> wesnoth: Iris Morelle website:master 90581fe637bb / gettext.wesnoth.org/ (bin/update-gettext-stats includes/config.php): gettext.w.o: 1.14 is now the stable branch https://github.com/wesnoth/website/commit/90581fe637bb602d483b746a02657984e425a86f 20180321 19:03:36<+discordbot2> @Daedalus do you maintain any translation? 20180321 19:03:57< vn971> jyrkive which is exactly a kind of translations I like :D nevermind though, I don't really like any translations at all anyway. Simplest most-standard English is all I want, and it's not typical behavior. 20180321 19:04:44<+discordbot2> When I first started playing Wesnoth I needed the Spanish translation, as incomplete as it was. 20180321 19:05:01< vn971> Since we're on "wesnoth-dev" channel, are there maybe some thoughts about `scons` and installation to non-standard path? 20180321 19:07:47<+discordbot2> I don't maintain any translation and I'm not a native speaker of the languages I do know πŸ˜„ 20180321 19:12:34< loonycyborg> vn971: what thoughts exactly? 20180321 19:13:37< vn971> loonycyborg: how to fix it. How to make wesnoth look up files on a custom path, not /share/wesnoth 20180321 19:13:58< loonycyborg> pass datadir as argument to wesnoth 20180321 19:14:23< vn971> loonycyborg: that's one way, but can I specify this at build/compile time ? 20180321 19:14:35< loonycyborg> also prefix=something would make scons build and install it at another prefix 20180321 19:15:03< vn971> it's quite tedious to repeat this all the time in certain scenarios. An example is gdb where you can't even use stuff like bash aliases. You have to re-type --data-dir etc 20180321 19:15:03< loonycyborg> there are also vars for finer grained control of paths like this 20180321 19:15:30< loonycyborg> I generally run wesnoth right from the checkout dir 20180321 19:15:33< loonycyborg> like ./wesnoth 20180321 19:15:35< vn971> loonycyborg: so what should I use to install wesnoth to a custom path so it'll use the right files by default ? 20180321 19:15:56< loonycyborg> it looks in current directory for datafiles by default 20180321 19:16:29-!- travis-ci [~travis-ci@ec2-54-224-223-228.compute-1.amazonaws.com] has joined #wesnoth-dev 20180321 19:16:30< travis-ci> wesnoth/wesnoth#17085 (master - eeffa82 : Charles Dang): The build failed. 20180321 19:16:30< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/356484321 20180321 19:16:30-!- travis-ci [~travis-ci@ec2-54-224-223-228.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180321 19:16:33< loonycyborg> you sure you need to install it? you can run it right from checkout, ./wesnoth 20180321 19:16:43< loonycyborg> it'll find datadir there without extra args 20180321 19:17:16< loonycyborg> to which custom path do you want it to be installed anyway? 20180321 19:17:29< vn971> loonycyborg: hmm.. So if I `cd` to the directory where "data" files live and start wesnoth somehow like ../../bin/wesnoth , all will be good? 20180321 19:17:45< loonycyborg> no, from checkout root 20180321 19:17:50< loonycyborg> where SConstruct etc is 20180321 19:17:59< loonycyborg> and where wesnoth binary will be built by default 20180321 19:18:13< vn971> loonycyborg: I want to have "branches". Like master, 1.14 etc. Anything that'll give me such result is fine by me. 20180321 19:19:38< loonycyborg> I'm using git worktree for that 20180321 19:20:37< loonycyborg> like 'git worktree add ../wesnoth-1.14 1.14" 20180321 19:21:19< loonycyborg> so I have master checkout, 1.14 checkout, 1.12 checkout etc 20180321 19:30:05<+discordbot2> Same. 20180321 19:30:24<+discordbot2> @loonycyborg By the way, any news on building 1.12 with SCons 3.0.1? 20180321 19:31:09<+discordbot2> If not I guess I can switch to the competition. This must be why we keep two build systems for POSIX platforms, after all. πŸ˜‰ 20180321 19:32:43< loonycyborg> well I can cherry pick those commits to 1.12 branch 20180321 19:32:57< loonycyborg> even though we won't have any more releases from it 20180321 19:33:26<+discordbot2> Like I said, the 1.12 add-ons server and MP server are still active. 20180321 19:33:34<+discordbot2> Expect them to be active for months after 1.14.0 gold is out. 20180321 19:34:36<+discordbot2> (The obvious corollary is that people like want to be able to build 1.12.) 20180321 19:35:24<+discordbot2> (And so does wesnoth.org, especially if we upgrade VMs or need security patches backported.) 20180321 19:35:48< loonycyborg> hmm will look into it rn 20180321 19:36:52<+discordbot2> (In particular, I expect our Glorious Leader to force my hand at some point and have me upgrade campaignd and wesnothd to Debian stretch, which has SCons 2.5.1.) 20180321 19:37:20< irker760> wesnoth: Gregory A Lundberg wesnoth:master cdb11c2109d8 / src/tests/utils/fake_display.cpp: Fix bad initializer https://github.com/wesnoth/wesnoth/commit/cdb11c2109d878cfe0b60de608a0af9d0e99c7e0 20180321 19:38:42<+discordbot2> @Tad Carlucci you forgot to remove the member 20180321 19:38:55<+discordbot2> I'll do that 20180321 19:39:21<+discordbot2> I just removed the parameter from the invokation so I'd get a compile. 20180321 19:42:48-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20180321 19:46:15< vn971> the ability to build 1.12 tree on a modern system would be nice indeed, if it's not too hard. Even considering 1.12 is basically not updated anymore, yes. 20180321 19:48:35< irker760> wesnoth: Charles Dang wesnoth:master 75954109c53d / src/tests/utils/fake_display.cpp: Fake Display: cleanup https://github.com/wesnoth/wesnoth/commit/75954109c53d28f2c45269dfab5b339390c208a1 20180321 19:53:59<+discordbot2> @jyrkive may 1st is the new target date 20180321 19:55:46<+discordbot2> That goes for everyone looking at this on IRC too. 20180321 20:15:28-!- octalot [~steve@178.115.131.231.wireless.dyn.drei.com] has joined #wesnoth-dev 20180321 20:23:58<+discordbot2> we need to update https://wiki.wesnoth.org/ReportingBugs 20180321 20:24:08<+discordbot2> and add the bug report template to the repo 20180321 20:33:43< irker760> wesnoth/wesnoth:master Charles Dang 75954109c5 Fake Display: cleanup AppVeyor: vs2015/Release Failed 20180321 20:33:44< irker760> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-2081 20180321 20:33:48<+discordbot2> "If a bug has been in a named release, add a milestone for the next release when it is fixed." 20180321 20:33:49<+discordbot2> huh 20180321 20:34:08<+discordbot2> yeah, actually, I seem to remember proposing this back when we first switched trackers. 20180321 20:34:16<+discordbot2> But we certainly haven't been adhering by that. 20180321 20:34:50<+discordbot2> We've been using the milestones for "we want this to be fixed for this release" 20180321 20:35:12< Soliton> i suppose the idea is to have an overview of what's all been fixed in one version. 20180321 20:35:26<+discordbot2> Yeah, that was my idea originally, I guess 20180321 20:35:28< Soliton> which would certainly be nice. 20180321 20:35:32<+discordbot2> there is no such thing 20180321 20:36:04<+discordbot2> Going to remove that entry 20180321 20:36:51< Soliton> perhaps there is a way to filter by date on github so you can get an approximation of that for free if you know the previous release date. 20180321 20:37:48<+discordbot2> There is such a thing, yes 20180321 20:37:57<+discordbot2> https://help.github.com/articles/searching-issues-and-pull-requests/#search-based-on-when-an-issue-or-pull-request-was-created-or-last-updated 20180321 20:40:01-!- gfgtdf [~chatzilla@x4e368626.dyn.telefonica.de] has joined #wesnoth-dev 20180321 20:40:09< gfgtdf> Ravana_: 20180321 01:24:01< Ravana_> Lua errors are only displayed in Lua console, which needs debug 20180321 20:40:33< gfgtdf> Ravana_: i justz tested this with a simple [set_menu_item]+[lua] can could not reproduce it. 20180321 20:41:07< Ravana_> few days ago on 1.13.12 I caused Lua error and it was only visible in lua console 20180321 20:41:20< gfgtdf> i tested on 1.13.12 20180321 20:42:21< gfgtdf> hmm wait, maybe it only happens during the 'initalisation' lua code (where there is no display yet), not during loa codes which are invoked later 20180321 20:42:27< gfgtdf> lua* 20180321 20:42:57< gfgtdf> which would in particular mean that thi was no intentional change. 20180321 20:44:50<+discordbot2> come to think of it, actually, we could probably fulfill the requirements for "issues fixed in release" without altering our existing practices at all 20180321 20:45:36<+discordbot2> if you mark an open bug with a milestone, it means you want it fixed for that release.... 20180321 20:45:41<+discordbot2> if it does get fixed, then it's tagged 20180321 20:45:46<+discordbot2> for that milestone 20180321 20:46:04< irker760> wesnoth: Alexander van Gessel wesnoth:1.12 6af2eb00dfd6 / / (9 files in 4 dirs): Basic 2to3 fixes https://github.com/wesnoth/wesnoth/commit/6af2eb00dfd6736aaa1f4e185333d0e959c2360f 20180321 20:46:06< irker760> wesnoth: Alexander van Gessel wesnoth:1.12 3776c578771c / SConstruct src/SConscript: Get rid of commands module https://github.com/wesnoth/wesnoth/commit/3776c578771ce3046a3b31c816cadce77562a4a4 20180321 20:46:08< irker760> wesnoth: Alexander van Gessel wesnoth:1.12 39d393307322 / SConstruct scons/boost.py: Further python3 compatibility updates. https://github.com/wesnoth/wesnoth/commit/39d3933073220f0d0a2767c3505980f3f4d751f5 20180321 20:46:08<+discordbot2> all we'd have to do is mark bugs we fixed that have been in releases but we didn't specifically asy "we must fix" 20180321 20:46:10< irker760> wesnoth: Alexander van Gessel wesnoth:1.12 cf017dc1684d / SConstruct scons/install.py: Don't use comma-separated arguments to print function https://github.com/wesnoth/wesnoth/commit/cf017dc1684d3bdaffa50cfb798b3da6cfd3b523 20180321 20:46:12< irker760> wesnoth: Alexander van Gessel wesnoth:1.12 2cba6c2cdb65 / SConstruct scons/gettext.py scons/gettext_tool.py: Rename gettext scons module to avoid naming conflict https://github.com/wesnoth/wesnoth/commit/2cba6c2cdb652aa563f0d036bc48ba8403ff306c 20180321 20:46:14< irker760> wesnoth: loonycyborg wesnoth:1.12 5a6df3d253db / SConstruct po/SConscript: Made build with scons 3.0.0 work fixing issue #2093 https://github.com/wesnoth/wesnoth/commit/5a6df3d253db4939d218f0b24a4b126835daeebd 20180321 20:46:14-!- travis-ci [~travis-ci@ec2-54-161-182-138.compute-1.amazonaws.com] has joined #wesnoth-dev 20180321 20:46:15< travis-ci> wesnoth/wesnoth#17087 (master - 7595410 : Charles Dang): The build is still failing. 20180321 20:46:16< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/356535173 20180321 20:46:16-!- travis-ci [~travis-ci@ec2-54-161-182-138.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180321 20:46:16< irker760> wesnoth: loonycyborg wesnoth:1.12 cefa806bf579 / SConstruct po/SConscript scons/scanreplace.py: scons: fixup python3 backport https://github.com/wesnoth/wesnoth/commit/cefa806bf579f48fa53e7b33bf157b88303c976c 20180321 20:46:20<+discordbot2> hmmmmm 20180321 20:48:41< loonycyborg> shadowm: building 1.12 with scons 3.0.1 should work now 20180321 20:48:58< loonycyborg> but iirc another packbort is needed for pot-update 20180321 20:49:04< loonycyborg> *backport 20180321 20:49:20< loonycyborg> will look for it later 20180321 20:50:04< shadowm> This means it'll still build with older versions of SCons, right? 20180321 20:52:48< gfgtdf> Ravana_: added a bugrpoert for lua errros notvisible issue 20180321 20:53:01< irker760> wesnoth: Iris Morelle wesnoth:1.14 ef8fc9df939f / README.md: Check the Travis status for 1.14 in the README for this branch https://github.com/wesnoth/wesnoth/commit/ef8fc9df939f8875d8c69ef41619c40a1287a352 20180321 20:55:03< Soliton> @vultraz: with closed:>[insert last release date] you're guaranteed to get everything. 20180321 21:01:17< Soliton> btw, if i may be annoying as usual... kinda weird to announce the release date on twitter and afterwards tell your fellow devs. 20180321 21:01:50<+discordbot2> I told you immediately after. 20180321 21:02:08< Soliton> exactly. 20180321 21:02:21<+discordbot2> Your point being? 20180321 21:03:27< Soliton> why do you not mention it here first so someone can give you input on it? 20180321 21:04:30<+discordbot2> Because I had already discussed it with someone else and made a decision. 20180321 21:04:52< Soliton> especially the way it was worded on twitter if now some dev told you some info you missed or didn't know that makes the date impossible or whatever... 20180321 21:05:25< Soliton> sure you're the release manager, doesn't mean you have to make decisions in isolation or private anyway. 20180321 21:05:41< Soliton> anyway not a big complaint just seems weird to me. 20180321 21:06:08< loonycyborg> shadowm: yes 20180321 21:06:21< loonycyborg> I tested it both bith 3.0.1 and 2.5.1 20180321 21:06:46< shadowm> It's the announcement date, so there's less than a handful of people required to be available on that day. 20180321 21:07:26< shadowm> Mainly vultraz, one of the website staff (which means I'm implicitly volunteering to be there), and whoever he needs to help him with the Steam stuff if anyone at all. 20180321 21:08:05< shadowm> I also made sure to make it clear that he needs at least a week between the gold tag and the announcement date for everyone to sort things out. 20180321 21:08:56< shadowm> This should give everyone plenty of leeway to get things prepared for 1.14.0 with at least one or two more RCs in between now and 7 days before May 1st. 20180321 21:08:59< Soliton> i'm sure it was all well though out. i just don't understand why it was thought out in private. 20180321 21:09:11<+discordbot2> I agree with Soliton there 20180321 21:09:39<+discordbot2> I mean it is great to announce a release date but considering there have been issues even compiling a working master build.... 20180321 21:09:59< shadowm> We're ignoring master for now since 1.14 is what we need to ship Daedalus. 20180321 21:10:06<+discordbot2> why not announce the release date on Steam after the 'gold' version has been sent to Steam for verification and uploading? 20180321 21:10:33<+discordbot2> unless ya'll got an updated trailer planned πŸ˜„ 20180321 21:10:50< shadowm> A lot of opinions keep being thrown around with regards to the schedule and ultimately it's impossible to reach a consensus for longer-term stuff like this. 20180321 21:11:49< shadowm> Planning for individual RCs should continue to be done with the devs' consensus, and if we reach a point where 1.14.0 7 days before May 1st doesn't seem viable then vultraz will push that announcement date back again. But ideally that won't happen. 20180321 21:12:30<+discordbot2> better to wait a week before announcing and make the date, than to have something come up and push it back 20180321 21:12:35<+discordbot2> public perception and all that 20180321 21:13:02< shadowm> Like I told him, I trust that no-one will get hit by a bus in the meantime because otherwise that will set everything back *again*. But the dev team also needs to show the community that they're working on things as fast as they can, and that includes pushing out as many RCs as are needed as bugs are discovered and promptly fixed. 20180321 21:13:30<+discordbot2> that's ideally all well and good but development is pretty rough, just ask Vultraz how minor stuff can randomly derail plans 20180321 21:13:56<+discordbot2> game development/coding can be unpredictable at times, regressions can happen and tracking down bugfixes is relative to how much time people have to do so 20180321 21:13:57< shadowm> That last part I just said applies regardless of whether gold is to be announced on May 1st this year or December 1st next year. 20180321 21:14:02<+discordbot2> all-volunteer team and all that 20180321 21:14:05< shadowm> Daedalus, I know all this. 20180321 21:14:13< Soliton> Daedalus could you please stop talking? 20180321 21:14:14< shadowm> I was release manager for 1.12.1 through 1.13.2. 20180321 21:14:17<+discordbot2> I feel the announcement was a bit ambitious and premature 20180321 21:14:28-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 260 seconds] 20180321 21:14:42<+discordbot2> Soliton seems to want me to shut up 🀣 20180321 21:14:57-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20180321 21:19:59<+discordbot2> For some reason people don't like Captain Obvious 20180321 21:20:17<+discordbot2> in opensource projects people also tend to have neophilia 20180321 21:20:27<+discordbot2> so repetition of obvious things triggers them 20180321 21:21:41< Soliton> i was just annoyed by someone having little insight into how wesnoth works preaching about how it should be done. 20180321 21:22:16< Soliton> no matter whether they agree with me or not. 20180321 21:23:35< Soliton> anyhow i agree with all things shadowm said. still weird to learn the release date from twitter. 20180321 21:24:02< Soliton> but i'd like to repeat that it's not a big deal, was just wondering. 20180321 21:24:26<+discordbot2> I don't even know much about how it works 20180321 21:24:40<+discordbot2> will valve people even do any checking on the build? 20180321 21:25:01< shadowm> I think vultraz said the other day they do a "quick review". Who knows exactly what that entails. 20180321 21:25:08<+discordbot2> I'm pretty sure they won't check it every update 20180321 21:25:16< shadowm> Nothing too exhaustive for sure, because I vaguely recall malware getting on Steam before. 20180321 21:25:31<+discordbot2> will they try to run it? 20180321 21:25:53< shadowm> I think they do have to check every update. 20180321 21:26:20< shadowm> I also vaguely recall a publisher announcing at some point that a critical update was pending review from Valve before it would be rolled out. 20180321 21:26:33<+discordbot2> Probably about like Apple or Google. Make sure it looks like what we say it is and doesn't do anything suprising like sending password files and usernames to some random site. 20180321 21:26:43<+discordbot2> anyway 20180321 21:26:48<+discordbot2> it doesn't look like to me 20180321 21:27:16< shadowm> In other words, just run an AV on it that'll check it/parts of it against a list of well-known checksums for malware or something. 20180321 21:27:25< shadowm> Nothing terribly sophisticated. 20180321 21:27:26<+discordbot2> seems builds just immediately get available to player as soon as I set newly uploaded build as default 20180321 21:27:43< shadowm> That could be because it's still private though. 20180321 21:27:49<+discordbot2> maybe 20180321 21:28:11<+discordbot2> but I expect checking even little bugfixes would be too much drag 20180321 21:29:56<+discordbot2> Not if it's a dumb automated system 20180321 21:37:51<+discordbot2> I suppose AV run could be fast 20180321 21:37:56<+discordbot2> and unnoticeable to me 20180321 21:53:03< irker760> wesnoth/wesnoth:1.14 Iris Morelle ef8fc9df93 Check the Travis status for 1.14 in the AppVeyor: vs2017/Release Failed 20180321 21:53:04< irker760> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-1.14-1787 20180321 22:07:03<+discordbot2> "Note: Your store page will need to go through a quick review prior to posting. Please allow 2-3 business days." "Note: Your build will need to go through a quick review prior to posting. Please allow 3-5 business days." 20180321 22:07:08<+discordbot2> I think this is just for the initial release 20180321 22:11:26<+discordbot2> btw vult you already tried running builds I uploaded? 20180321 22:11:30<+discordbot2> "Once your store page and game build have been completed, reviewed, and approved, then you're ready for release." 20180321 22:11:34<+discordbot2> Not yet 20180321 22:11:53<+discordbot2> I've asked shadowm to try out the Linux one as we speak. 20180321 22:12:12<+discordbot2> shadowm has perner access too? 20180321 22:12:19<+discordbot2> *partner 20180321 22:12:21<+discordbot2> I added her today 20180321 22:13:36-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!] 20180321 22:13:49-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20180321 22:14:01<+discordbot2> cool 20180321 22:14:23-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection] 20180321 22:19:20<+discordbot2> BTW, I still need someone to help make screenshots for the steam page 20180321 22:19:25<+discordbot2> they'll also be used on the frontpage 20180321 22:19:57<+discordbot2> need at least 5, preferably at 1920 x 1080 20180321 22:22:40<+discordbot2> 1280 x 720 is also acceptable 20180321 22:23:00<+discordbot2> I have a set for Appstore, in multiple languages. They are HiDPI, twice as big, though, and not the most attractive scenes. 20180321 22:23:22<+discordbot2> hm 20180321 22:23:27<+discordbot2> well, we kinda need attractive scenes 20180321 22:24:22<+discordbot2> at least one needs be of the editor 20180321 22:26:38-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20180321 22:26:48<+discordbot2> hmm btw 20180321 22:27:01<+discordbot2> I have some games installed on other steam acc 20180321 22:27:17<+discordbot2> yet my upload acc still sees them in library 20180321 22:27:25<+discordbot2> I'd appreciate players' help too πŸ˜ƒ 20180321 22:27:35<+discordbot2> are you supposed to be able to play even game installed from other steam acc? 20180321 22:27:53<+discordbot2> like could I play wesnoth on my main acc like this 20180321 22:30:24<+discordbot2> Yes 20180321 22:30:38<+discordbot2> You can see locally installed games when logged in on another account 20180321 22:30:51<+discordbot2> But you do not own them 20180321 22:31:15<+discordbot2> I didn't try to run unowned by that acc games 20180321 22:31:17<+discordbot2> will it work? 20180321 22:31:29-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 248 seconds] 20180321 22:32:02<+discordbot2> what? 20180321 22:32:29<+discordbot2> Try it, it's not illegal πŸ˜ƒ 20180321 22:34:03-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20180321 22:35:39< irker760> wesnoth/wesnoth:master Charles Dang d0f7bfbf37 Floating Label: added text to debug outp AppVeyor: 1/2 builds failed 20180321 22:35:40< irker760> Details vs2017/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-master-1781 20180321 22:36:14<+discordbot2> seems it works 20180321 22:36:25<+discordbot2> it's like family sharing but with yourself πŸ˜› 20180321 22:39:38<+discordbot2> only I was running a free game 20180321 22:40:02<+discordbot2> I cba to run nonfree ones atm 20180321 22:40:15<+discordbot2> they all are huge and will likely cause ram overflow 20180321 22:45:41-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection] 20180321 22:46:38-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20180321 22:50:20< zookeeper> shadowm, right, so a spanish translator basically always has to check each speaker's gender before they can translate a message properly? 20180321 22:50:32-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Client Quit] 20180321 22:51:42<+discordbot2> in russian sometimes verbs are affected by speaker's gender 20180321 22:51:49<+discordbot2> depending on tense 20180321 22:51:57-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20180321 22:52:35<+discordbot2> well not speaker's 20180321 22:52:38<+discordbot2> noun's 20180321 22:53:25<+discordbot2> subject's that is 20180321 22:53:36<+discordbot2> verbs, nouns, adjectives... 20180321 22:55:25< zookeeper> must be a lot more laborious to translate into those types of languages :x 20180321 22:56:26<+discordbot2> zookeeper: Yes. 20180321 22:56:54<+discordbot2> Says the Finnish speaker πŸ˜› 20180321 22:58:43< zookeeper> oh i don't think translating into or from finnish is difficult, even if the language itself can be. 20180321 22:59:07< zookeeper> our pronouns are gender-neutral, for one 20180321 22:59:40< zookeeper> well, okay, it is tricky when you try to translate something that communicates someone's gender by pronouns only 20180321 23:00:28< loonycyborg> afaik chinese and japanese have a lot less of such stuff 20180321 23:03:50<+discordbot2> is thankful english doesn't have gendered nouns. 20180321 23:38:34< irker760> wesnoth/wesnoth:master Charles Dang eeffa82e2b Fixup 37dee4b (unused variable) AppVeyor: 2/4 builds failed 20180321 23:38:35< irker760> Details vs2017/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-master-1783 20180321 23:38:36< irker760> Details vs2015/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-2079 20180321 23:40:27-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 240 seconds] 20180321 23:56:39-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180321 23:56:45-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev --- Log closed Thu Mar 22 00:00:10 2018