--- Log opened Sat Aug 13 00:00:31 2016
20160813 00:04:10< vultraz> gfgtdf: try adding the line use_markup = "true" at data/gui/window/tooltip_large.cfg:225
20160813 00:04:30< gfgtdf> vultraz: ok thy
20160813 00:05:15< celmin> I don't think it's a good idea to enable it in all tooltips though...
20160813 00:05:25< vultraz> no choice here
20160813 00:05:36< celmin> There's always a choice.
20160813 00:05:48< vultraz> not without adding a bloody flag too all widgets that use tooltips
20160813 00:05:54< vultraz> to tcontrol or something..\
20160813 00:08:12< vultraz> and frankly
20160813 00:08:19< vultraz> having to specify use_tooltip_markup is ridiculous 
20160813 00:08:45< celmin> I disagree.
20160813 00:09:05< celmin> Also, the markup itself is ridiculous, so I'm much prefer it not be used anywhere.
20160813 00:09:39< vultraz> Why do you disagree?
20160813 00:09:47< celmin> Is there any chance we can actually implement our own sane markup? Admittedly it wouldn't be as versatile, but who really needs to change the gravity of their text?
20160813 00:09:49< vultraz> And yes, maybe pango is ridiculous, but what else do we have right now?
20160813 00:10:02< celmin> Let's just implement basic Markdown or something.
20160813 00:10:21-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160813 00:10:33< celmin> As for why I disagree, I would assume that it's slower if you enable markup.
20160813 00:10:43< celmin> I don't know how significant this is, mind you.
20160813 00:10:44< vultraz> ....what....
20160813 00:10:55< celmin> But it's enough to say that it's not ridiculous to have the option.
20160813 00:10:58< vultraz> if there's no markup to process it doesn't process that!
20160813 00:11:05< celmin> Huh?
20160813 00:11:16< vultraz> I know there IS a reason we didn't enable markup everywhere
20160813 00:11:20< vultraz> shadowm told me one, years ago
20160813 00:11:23< vultraz> I can't remember it now
20160813 00:11:48< celmin> If markup is enabled, it needs to parse the string to convert the markup.
20160813 00:12:13< celmin> Even if there is no markup, that's certainly going to be at least marginally slower than with markup disabled.
20160813 00:12:33< celmin> I guess it would amount to copying the string, most likely.
20160813 00:15:00-!- gfgtdf_ [~chatzilla@x4e369c08.dyn.telefonica.de] has joined #wesnoth-dev
20160813 00:15:03-!- gfgtdf [~chatzilla@x4e369c08.dyn.telefonica.de] has quit [Read error: Connection reset by peer]
20160813 00:15:10-!- gfgtdf_ is now known as gfgtdf
20160813 00:15:28-!- travis-ci [~travis-ci@ec2-54-92-211-36.compute-1.amazonaws.com] has joined #wesnoth-dev
20160813 00:15:29< travis-ci> wesnoth/wesnoth#10345 (master - f5210fa : Charles Dang): The build is still failing.
20160813 00:15:29< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/151923599
20160813 00:15:29-!- travis-ci [~travis-ci@ec2-54-92-211-36.compute-1.amazonaws.com] has left #wesnoth-dev []
20160813 00:15:48-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 250 seconds]
20160813 00:20:35< shadowm> There isn't a special reason. It's simply the principle of least surprise.
20160813 00:21:16< shadowm> I could elaborate on that but I'm fairly sure those last four words should suffice for this audience.
20160813 00:33:32-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160813 00:38:30-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160813 00:39:58-!- travis-ci [~travis-ci@ec2-54-157-217-171.compute-1.amazonaws.com] has joined #wesnoth-dev
20160813 00:39:59< travis-ci> wesnoth/wesnoth#10346 (master - 450f16f : Charles Dang): The build has errored.
20160813 00:39:59< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/151928605
20160813 00:39:59-!- travis-ci [~travis-ci@ec2-54-157-217-171.compute-1.amazonaws.com] has left #wesnoth-dev []
20160813 00:45:29-!- louis94 [~~louis94@91.178.241.1] has quit [Ping timeout: 260 seconds]
20160813 00:47:36< vultraz> it seems to me that in 2016, we shouldn't have to worry about the impact of parsing text markup!
20160813 00:47:45< celmin> I dunno.
20160813 00:47:58< celmin> Parsing is a fairly intensive task though.
20160813 00:50:39< gfgtdf> but we are usualyl onyl show one tooltip per time, so i agree with vultrat thats we shoudlnt worry about in this case
20160813 00:52:54< shadowm> Okay, looks like I actually need to elaborate somehow.
20160813 00:53:30< shadowm> When making decisions on this kind of thing you need to start with the base assumption that user-defined input is somehow involved.
20160813 00:54:22< shadowm> For example, GUI1 lacked a trivial mechanism to disable formatting markup on items that shouldn't parse any.
20160813 00:54:51< shadowm> This was abused time and again by people making their add-on or era or campaign names colored or large-sized.
20160813 00:55:25< shadowm> So in a way it's a combination of what I said above, _and_ the principle of least privilege.
20160813 00:56:04< shadowm> If you don't need unexpected behavior that should be reserved for sanitized or (preferably) hardcoded input, then it shouldn't be enabled by default.
20160813 00:57:12< shadowm> Maybe you can make an exception for tooltips and only tooltips, but that still seems dubious to me since those are known to be used in lieu of larger table cells in some places (e.g. listboxes) that may take user-defined input.
20160813 00:57:38< shadowm> If pretty much all of the time people won't need markup then there's nothing with requiring them to use their brains before enabling it.
20160813 00:58:06-!- Bonobo [~Bonobo@2001:44b8:254:3200:c4dd:7e1c:b7c5:ebc2] has joined #wesnoth-dev
20160813 01:03:27-!- Bonobo [~Bonobo@2001:44b8:254:3200:c4dd:7e1c:b7c5:ebc2] has quit [Ping timeout: 264 seconds]
20160813 01:03:47-!- Bonobo [~Bonobo@2001:44b8:254:3200:59c6:97d3:2da7:c730] has joined #wesnoth-dev
20160813 01:03:51< celmin> If I understand correctly, the tooltip definition is global to a given dialog, right? 
20160813 01:03:59< vultraz> yes
20160813 01:04:01< celmin> Meaning you can't use a different definition for each widget?
20160813 01:04:34< vultraz> no
20160813 01:04:39-!- gfgtdf_ [~chatzilla@x4e36920b.dyn.telefonica.de] has joined #wesnoth-dev
20160813 01:04:51< celmin> If enabling markup is done in the tooltip definition, you could have a separate definition that does this, but… what shadowm said makes me think this might be a bad idea… what happens if people try to use markup in their ability descriptions?
20160813 01:07:45-!- gfgtdf [~chatzilla@x4e369c08.dyn.telefonica.de] has quit [Ping timeout: 265 seconds]
20160813 01:07:49-!- gfgtdf_ is now known as gfgtdf
20160813 01:08:05-!- travis-ci [~travis-ci@ec2-54-157-217-171.compute-1.amazonaws.com] has joined #wesnoth-dev
20160813 01:08:06< travis-ci> wesnoth/wesnoth#10347 (master - 312a77c : Charles Dang): The build is still failing.
20160813 01:08:06< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/151933218
20160813 01:08:06-!- travis-ci [~travis-ci@ec2-54-157-217-171.compute-1.amazonaws.com] has left #wesnoth-dev []
20160813 01:11:31< irker712> wesnoth: gfgtdf wesnoth:master d668dc98a5db / src/ (game_state.cpp team.cpp): fix segfault when loading saves with undoable moves. https://github.com/wesnoth/wesnoth/commit/d668dc98a5dbfc86888a2bd36ad45958901aef78
20160813 01:11:33< irker712> wesnoth: gfgtdf wesnoth:master 2ea30d565d65 / src/gui/dialogs/ (loadscreen.cpp loadscreen.hpp): clean an include https://github.com/wesnoth/wesnoth/commit/2ea30d565d652b003435e0feed7ed715edb405f9
20160813 01:11:35< irker712> wesnoth: gfgtdf wesnoth:master 6f179bf4d0ab / / (6 files in 2 dirs): show trait and ability tooltips in recall dialog https://github.com/wesnoth/wesnoth/commit/6f179bf4d0abec40731dd059b9c92d7ecf2d9c6e
20160813 01:12:44< celmin> gfgtdf: It looks to me like d668dc9 potentially introduces a different bug.
20160813 01:12:46< gfgtdf> vultraz: ^ i didnt do the this for the recuit dialog today, spend too much time finding out the bug in my previous code (that used 'child_label.set_label(label)' )
20160813 01:13:08< celmin> gfgtdf: Also, you shouldn't've removed the check for empty teams.
20160813 01:13:18-!- noy [~Noy@wesnoth/developer/noy] has quit [Read error: Connection reset by peer]
20160813 01:13:35< vultraz> and that stray 'else if' is still there
20160813 01:13:52< gfgtdf> celmin: whcih do do mean ?
20160813 01:14:01< celmin> gfgtdf: In the segfault fix.
20160813 01:14:25-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160813 01:14:52< gfgtdf> celmin: well vultraz changes !resources::teams  to resources::gameboard->teams().empty() i just fixed that..
20160813 01:15:04< gfgtdf> changed*
20160813 01:15:07< celmin> gfgtdf: No, you've changed the semantic meaning of that conditional.
20160813 01:15:13< celmin> It should return if there are no teams.
20160813 01:15:23< celmin> But with your commit, it no longer does.
20160813 01:15:25< celmin> ...
20160813 01:15:26< celmin> Wait.
20160813 01:15:30< celmin> That goes for vultraz.
20160813 01:15:32< celmin> Sorry.
20160813 01:15:45< vultraz> what?
20160813 01:16:06< celmin> vultraz: !resources::teams is not equivalent to resources::gameboard->teams().empty()
20160813 01:16:17< vultraz> where is this?
20160813 01:16:36< celmin> vultraz: The commit gfgtdf just pushed?
20160813 01:17:10< vultraz> I see
20160813 01:17:21< vultraz> the what he did was the proper fix
20160813 01:18:08< celmin> gfgtdf: With the recall dialog change, you didn't bother to address any of our concerns as far as I can see.
20160813 01:18:27< gfgtdf> qhihc concerns ?
20160813 01:18:31< vultraz> our?
20160813 01:18:37< celmin> vultraz: Yours
20160813 01:18:39< vultraz> ah
20160813 01:18:51< celmin> gfgtdf: Read up to when we were commenting on it when you linked to it on your fork.
20160813 01:19:07< vultraz> either way, I shall address them
20160813 01:19:11< celmin> gfgtdf: I'm not especially eager to repeat myself.
20160813 01:19:35< vultraz> oh geez, this is bad
20160813 01:19:37< celmin> gfgtdf, vultraz: Do you really need both tree_details and label_details? It seems redundant to me.
20160813 01:19:41< vultraz> the dialog resizes now :|
20160813 01:19:52 * vultraz ;_;
20160813 01:19:59< vultraz> celmin: will look into it
20160813 01:20:17< vultraz> recruit dialog is totally broken..
20160813 01:20:22< vultraz> ok, I think I know the fix
20160813 01:20:24< gfgtdf> vultraz: in idth?
20160813 01:20:27< gfgtdf> width*
20160813 01:20:32< vultraz> gfgtdf: height
20160813 01:21:13< vultraz> just let me try to fix these problems
20160813 01:21:24< gfgtdf> vultraz: hmm that didnt happen to me, i onyl had some issues to width
20160813 01:22:52< gfgtdf> celmin: the onyl concern i see is about the elseif, and since theat code will mostlike be rewritten when the tooltios in thr unit type case is added i dont really sonsider this important
20160813 01:23:25< celmin> gfgtdf: It's important. Don't commit half-working code.
20160813 01:23:47< vultraz> I will fix
20160813 01:23:49< celmin> And don't break existing code with the argument that a fix will come very soon, either.
20160813 01:24:46< vultraz> hm
20160813 01:25:06< vultraz> I find it hard to understand his code sometimes..
20160813 01:25:30< vultraz> looks like you're doing one node per line?
20160813 01:25:40< vultraz> honestly, I think it'd be better to have all traits in a tooltip
20160813 01:25:44< vultraz> and all abilities
20160813 01:25:50< celmin> What? No.
20160813 01:26:01< vultraz> you *just* said earlier that's acceptable
20160813 01:26:04< vultraz> make up your mind
20160813 01:26:06< vultraz> :|
20160813 01:26:23< celmin> I said it's acceptable for descriptions, not for the list of traits/abilities.
20160813 01:26:58< vultraz> yes..
20160813 01:27:01< vultraz> that's what I'm saying
20160813 01:27:05< vultraz> put all the descriptions in a tooltip
20160813 01:27:09< celmin> Okay, so I misunderstood.
20160813 01:27:12< celmin> Carry on.
20160813 01:27:18< vultraz> and make it so when you hover over the title they also come up
20160813 01:27:21< vultraz> so, basically
20160813 01:27:26< vultraz> reduce it to one node per section
20160813 01:27:29< vultraz> as opposed to per line
20160813 01:27:43< celmin> Well, the traits are supposed to only be one line anyway.
20160813 01:27:50< celmin> And the abilities are also only supposed to be one line.
20160813 01:27:56< vultraz> no no no no no
20160813 01:27:58< celmin> Yes.
20160813 01:28:08< celmin> At least, make it something like three or four per line.
20160813 01:28:08< vultraz> god dammit
20160813 01:28:10-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20160813 01:28:15< vultraz> that will make it too LONG!
20160813 01:28:21< celmin> Don't just list each one on its own line, that's a waste of space.
20160813 01:28:29< celmin> There's plenty of room for two or three or four on one line.
20160813 01:28:47 * vultraz groans
20160813 01:29:22-!- travis-ci [~travis-ci@ec2-54-157-217-171.compute-1.amazonaws.com] has joined #wesnoth-dev
20160813 01:29:22< celmin> And listing them all on their own line is bad for 800x600 as well - it makes it take up too much vertical space and generate a scrollbar.
20160813 01:29:23< travis-ci> wesnoth/wesnoth#10348 (master - 12f7107 : Lari Nieminen): The build is still failing.
20160813 01:29:23< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/151943248
20160813 01:29:23-!- travis-ci [~travis-ci@ec2-54-157-217-171.compute-1.amazonaws.com] has left #wesnoth-dev []
20160813 01:29:26-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160813 01:29:41 * vultraz curses 800x600 to oblivion
20160813 01:29:55< gfgtdf> i like one per line better, havin multiple per line looks kinda sqashed.
20160813 01:30:01< celmin> And yet, Aginor puts forth a very good reason to continue supporting it.
20160813 01:30:09< gfgtdf> also it doesnt show a scrollbar to me on 800y600 evne if i have 4 traits
20160813 01:31:15< celmin> gfgtdf: You probably didn't try an elvish druid - not only does it have two traits and abilities, but also three attacks.
20160813 01:31:21< gfgtdf> well it shows a greyed-out scrollbar for some reason, but this doesnt mena that there is not enough space.
20160813 01:31:22-!- noy [~Noy@wesnoth/developer/noy] has quit [Client Quit]
20160813 01:32:04< vultraz> i cannot stand the mention of these bloody tiny resolutions >_>
20160813 01:32:18-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160813 01:32:26< celmin> That won't prevent me from bringing them up every time you do something that breaks them.
20160813 01:32:41< vultraz> it is so frustrating to have to sacrifice aesthetics for the sake of squashing everything into a tiny BOX
20160813 01:33:06< celmin> 800x600 is small, but not really tiny in my opinion.
20160813 01:33:12< celmin> I don't normally play at that resolution, admittedly.
20160813 01:33:36< vultraz> WHO WOULD EVER PLAY AT THAT RESOLUTION
20160813 01:33:38< vultraz> WHO
20160813 01:33:40< vultraz> WHO
20160813 01:34:12< celmin> Actually, maybe I should try fiddling with the MP lobby to make it work better at that resolution… just to prove that it can be done...
20160813 01:34:35< vultraz> be my guest 
20160813 01:34:40-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160813 01:34:43< vultraz> I'll be very surprised if you can
20160813 01:34:51< celmin> Because if I understand correctly, you don't have to sacrifice aesthetics to squash it into a tiny box.
20160813 01:35:39< shadowm> 800x600 will only go away the moment people become unable to make their windows not take up the whole screen.
20160813 01:35:55< celmin> Huh?
20160813 01:36:06< shadowm> I said, 800x600 will only go away the moment people become unable to make their windows not take up the whole screen.
20160813 01:36:31< celmin> I don't get it.
20160813 01:36:41-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 244 seconds]
20160813 01:36:56< shadowm> Windows can be horizontally 800 and vertically 600 pixels in size.
20160813 01:37:12< shadowm> No matter how large your screen's pixel count is.
20160813 01:38:24< shadowm> This is a crucial fact that our resident cantankerous revolution leader doesn't ever seem to acknowledge (this has actually come up a million times in conversations with him over the years).
20160813 01:39:00-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 250 seconds]
20160813 01:39:36< celmin> For some reason the GUI2 unit tests fail on my computer due to untested dialogs, even though they work on Travis.
20160813 01:39:49< shadowm> I surely hope you get it now.
20160813 01:39:55< celmin> I think so.
20160813 01:40:27< celmin> vultraz: Do you understand the MP lobby layout enough to help me if I'm confused about something?
20160813 01:40:36< vultraz> yes
20160813 01:40:45< vultraz> in fact I was going to make a fix to it
20160813 01:40:54< celmin> A WML fix?
20160813 01:41:01< vultraz> first, you can make the chat box minimum size smaller on smaller resolutions
20160813 01:41:17< vultraz> then, you need to make the game list row grow_factor = 1 and the chat box grow_factor = 0
20160813 01:41:48-!- gfgtdf [~chatzilla@x4e36920b.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 48.0/20160726073904]]
20160813 01:42:14< vultraz> please do that if you're tweaking it
20160813 01:43:03< celmin> I'm not quite sure what you're referring to, but I doubt what I'm doing will conflict with it, so if you've already done it or know exactly what you mean, go ahead and do it yourself.
20160813 01:43:41< celmin> BTW, rooms are currently a listbox?
20160813 01:43:50< celmin> Surely they should be a treeview?
20160813 01:44:03< vultraz>  ..what?
20160813 01:44:20< celmin> Since you click a disclosure triangle to see who's in the room?
20160813 01:44:44< vultraz> i don't know what in hell is going on with the user lists
20160813 01:45:09< celmin> Hmm, what does GUI_WINDOW_FULLSCREEN really do...
20160813 01:45:50< vultraz> makes the window fullscreen
20160813 01:46:03< celmin> I see… so it sets width and height to screen width and screen height, respectively...
20160813 01:46:20< celmin> But it goes in [resolution].
20160813 01:47:15< celmin> …wait, according to the wiki, [resolution] doesn't have width and height keys...?
20160813 01:47:29< vultraz> I don't know
20160813 01:47:31< vultraz> it works
20160813 01:49:27< celmin> Okay, doesn't look like it conflicts with resolution selection, if the wiki can be trusted.
20160813 01:52:12-!- travis-ci [~travis-ci@ec2-54-92-211-36.compute-1.amazonaws.com] has joined #wesnoth-dev
20160813 01:52:13< travis-ci> wesnoth/wesnoth#10349 (master - 0526808 : Charles Dang): The build has errored.
20160813 01:52:13< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/151952524
20160813 01:52:13-!- travis-ci [~travis-ci@ec2-54-92-211-36.compute-1.amazonaws.com] has left #wesnoth-dev []
20160813 01:52:25< celmin> In the recall dialog I get a scrollbar for the whole dialog. I think I showed you that already right?
20160813 01:52:49< vultraz> what??
20160813 01:52:53< vultraz> what resolution are you on
20160813 01:53:00< celmin> 800x600
20160813 01:53:10< celmin> Haven't I mentioned several times that I do all my testing on 800x600?
20160813 01:53:11 * vultraz curses
20160813 01:53:34< celmin> Guessing the preview has slightly too large min_height or something.
20160813 01:53:48< celmin> I wish there were a few games on the server.
20160813 01:53:59< celmin> It'd make it easier to see how stuff looks.
20160813 01:54:09< celmin> I guess I could always connect a second instance and make one.
20160813 01:57:16< vultraz> celmin: you know what you can do
20160813 01:57:41< vultraz> make sure spacers are resized 
20160813 01:57:43-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has quit [Quit: ChipmunkV]
20160813 01:57:48< vultraz> right now
20160813 01:57:54< vultraz> if you have a formula for spacer size
20160813 01:58:03< vultraz> it isn't resized if you do a resize
20160813 01:59:55< celmin> I wonder if that's even possible.
20160813 02:00:20< celmin> It probably implies rebuilding the window.
20160813 02:00:27< celmin> I bet the formulas are evaluated at build time.
20160813 02:00:39< celmin> Which would be like the tdialog constructor or something.
20160813 02:00:45< celmin> Alternatively, before pre_show.
20160813 02:01:03< celmin> Oh right, I think there was a hook for that wasn't there...
20160813 02:08:26< celmin> Which means it's before pre_show, which means currently the only way to do it is to close the dialog and reopen it.
20160813 02:08:44< celmin> Which isn't really useful at all.
20160813 02:08:44< irker712> wesnoth: Charles Dang wesnoth:master bacece459c11 / data/gui/ (widget/scroll_label_description.cfg window/lobby_main.cfg): MP Lobby: fixed a few layout issues https://github.com/wesnoth/wesnoth/commit/bacece459c114b4cf51305d179c9cbecbc3bbdd3
20160813 02:08:58< vultraz> celmin: also improves layout at 800x600
20160813 02:09:02< vultraz> ^
20160813 02:09:18< vultraz> width is still fucked, but if you want, move the filter box above the box
20160813 02:09:23< vultraz> that should fix it
20160813 02:09:37< vultraz> then figure out why typing in the box causes a disabled window vertical scrollbar to apepar
20160813 02:09:48< vultraz> anyway, I appear to have been mistaken with my use of grow_factors
20160813 02:10:08< vultraz> typing in the box now doesn't mean the games list gets smaller
20160813 02:10:21< vultraz> turns out the chat needed to grow, to hit the minimum height of the list
20160813 02:11:51< vultraz> anyway
20160813 02:11:54< vultraz> proceed
20160813 02:12:11< vultraz> if you fix the issue with spaces, you'll have done a huge thing to improve support of smaller resolutions
20160813 02:12:34< vultraz> it absolutely sucks if you resize the window and are stuck with the old sizes for the spacers and nothing else
20160813 02:12:36< celmin> You mean what you said about re-evaluating formulas if the window size changes?
20160813 02:12:40< vultraz> yes
20160813 02:12:44< celmin> Wait, only spacers?
20160813 02:12:51< vultraz> only spacers
20160813 02:12:57< celmin> Hmm, that's actually kinda weird...
20160813 02:13:04< vultraz> spacers are fixed-sized widgets
20160813 02:13:10< vultraz> other widgets evaluate their "best size"
20160813 02:13:37< vultraz> technically, you could make spacers either fixed-sized OR best size
20160813 02:13:47< celmin> Well, I don't see how it could help smaller resolutions though.
20160813 02:14:06< celmin> It's not like I make the resolution smaller while the dialog is open. It's that size all along.
20160813 02:14:06< vultraz> add new keys width/height_formula and if they're used, use those formulas to calculate the best size on resize
20160813 02:14:23< celmin> Eh? Why would I want new keys?
20160813 02:14:31< vultraz> celmin: spacers are used in the minimum size macros
20160813 02:14:45< vultraz> say you want a minimum size of 200 w for a widget
20160813 02:14:55< vultraz> but then, your game window gets smaller
20160813 02:14:59< vultraz> 200 is now too much
20160813 02:15:04< vultraz> you want, say, 100
20160813 02:15:35< vultraz> i guess you wouldn't need new keys
20160813 02:15:55< vultraz> just check if a formulas has been passed
20160813 02:16:03< vultraz> and add get_best_size methods to tspacerf
20160813 02:16:22< vultraz> and calculate best spacer size if formulas were provided
20160813 02:16:28< vultraz> else return the fixed size
20160813 02:16:33< vultraz> celmin: make sense?
20160813 02:16:36< celmin> ie, use tformula so that enclosing the entire value in parentheses makes it a formula.
20160813 02:16:54< vultraz> however it works internally
20160813 02:16:56< vultraz> i dont know
20160813 02:17:15< celmin> Your MP create compile error is bugging me.
20160813 02:17:21< celmin> I might add a new method to tree views.
20160813 02:17:38< celmin> Or declare the map on the line above, then pass it.
20160813 02:17:38< vultraz> btw, travis will fail on the UPP commit
20160813 02:17:41< celmin> Not sure yet.
20160813 02:17:41< vultraz> i am going to lunch
20160813 02:17:44< celmin> UPP?
20160813 02:17:50< vultraz> unit preview pane
20160813 02:17:55< celmin> Oh, gfg's?
20160813 02:17:57< vultraz> i shall commit my fixes/improvements when i return
20160813 02:17:59< vultraz> yes
20160813 02:18:05< celmin> 'kay whatever
20160813 02:19:15-!- travis-ci [~travis-ci@ec2-54-92-211-36.compute-1.amazonaws.com] has joined #wesnoth-dev
20160813 02:19:16< travis-ci> wesnoth/wesnoth#10350 (master - 63c1b7f : Charles Dang): The build is still failing.
20160813 02:19:16< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/151957150
20160813 02:19:16-!- travis-ci [~travis-ci@ec2-54-92-211-36.compute-1.amazonaws.com] has left #wesnoth-dev []
20160813 02:19:37< celmin> It's failing already anyway, not like it'll make any difference if he added an additional failure path.
20160813 02:24:32-!- fabi [~fabi@176.7.89.74] has joined #wesnoth-dev
20160813 02:29:00< celmin> I don't really see a difference after your tweaks in the MP lobby, though it's not like I took screenshots to properly compare. The horizontal scrollbar does seem maybe slightly less...
20160813 02:33:40-!- un214 [~un214@104.220.56.173] has joined #wesnoth-dev
20160813 02:51:32-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160813 02:52:14-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has joined #wesnoth-dev
20160813 03:01:21< celmin> So basically, I was 100% right about the resolutions thing.
20160813 03:02:45< celmin> For some reason the chat area is now surrounded by a ton of empty space, though...
20160813 03:03:01< celmin> Not sure what's causing that.
20160813 03:04:30-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Remote host closed the connection]
20160813 03:06:37-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160813 03:10:03< celmin> Apart from the chat area problem, it's done.
20160813 03:10:17< celmin> This is more a proof of concept than anything else, but I'll commit it anyway so you can see, I guess.
20160813 03:15:21< celmin> Just can't figure out how to make sure that the chat area takes up all the remaining space...
20160813 03:17:37 * celmin tries adding vertical_grow=true to the [row]… >_>
20160813 03:18:07< celmin> vultraz: Any ideas?
20160813 03:22:59-!- Polsaker [~Polsaker@wikimedia/botters.Polsaker] has joined #wesnoth-dev
20160813 03:26:11< vultraz> celmin: code please
20160813 03:26:24< vultraz> celmin: also, that's a [column] key, not [row]
20160813 03:26:35< celmin> Yeah, figured, but thought I'd try it just in case.
20160813 03:27:01< celmin> I tried putting the whole grid in the minimum size macro, but that didn't help either.
20160813 03:27:14< vultraz> if you have two rows and one needs to take up more space, you give it grow_factor = 1 and the other grow_factor = 0 in their respective [row]s
20160813 03:27:39< celmin> Okay, let me try increasing its grow factor then...
20160813 03:28:10 * celmin tries 10
20160813 03:28:23< vultraz> holy crap
20160813 03:28:32< vultraz> no, you should never need to have more than 1
20160813 03:28:34< celmin> The aforementioned thing (moving the whole grid) did seem to help with the extra horizontal space at least.
20160813 03:28:39< vultraz> some old dialogs used random large numbers
20160813 03:28:41< celmin> Why not?
20160813 03:28:48< vultraz> because it's all relative
20160813 03:29:00< vultraz> i suppose at some point one would need 2
20160813 03:29:01< celmin> Exactly, so if you give it a higher number the row gets more space.
20160813 03:29:10< vultraz> no..
20160813 03:29:14< vultraz> that's... not how it works
20160813 03:29:27< vultraz> commit what you have and I'll fiddle with it
20160813 03:29:38< celmin> It didn't help, anyway.
20160813 03:29:54< celmin> I want to fix the extra space to the left of the title first.
20160813 03:30:26< vultraz> what..?
20160813 03:30:33< vultraz> look, I can't help you without seeing the code
20160813 03:30:37< celmin> Somehow I got extra space there.
20160813 03:30:53< vultraz> in the title at the very top?
20160813 03:30:58< celmin> Yeah.
20160813 03:31:07< vultraz> use horizontal_alignment= "left" in that label's column.
20160813 03:31:10< celmin> I dunno, maybe it's unavoidable with this layout.
20160813 03:31:22< celmin> Or border=0 somewhere...
20160813 03:31:32< vultraz> what...
20160813 03:31:34< celmin> …no that's probably already the default.
20160813 03:31:43< vultraz> what are you doing
20160813 03:32:04< celmin> I'll try one or two more things and then commit it regardless of whether it works.
20160813 03:32:15< vultraz> please do
20160813 03:32:21< vultraz> I need to see what it is you're doing 
20160813 03:32:32< celmin> Isn't there a macro for inserting a horizontal spacer?
20160813 03:32:37< vultraz> no
20160813 03:32:44< celmin> Oh, okay then.
20160813 03:32:45< vultraz> a horizontal line, yes
20160813 03:32:59< vultraz> and that is GUI_HORIZONTAL_SPACER_LINE
20160813 03:35:51< celmin> Still, I think restricting grow_factor to 1 is definitely silly.
20160813 03:36:05< celmin> Actually, I also set it to 10 for the filter box.
20160813 03:37:42-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev
20160813 03:38:48< tad_> A few days ago when testing I noticed an issue with move_unit_fake leaving an artifact at the start when under fog. Today it's not there. Has someone been working in this area?
20160813 03:39:10< celmin> Not to my knowledge, but it's not like I know what everyone is working on.
20160813 03:39:23-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160813 03:39:40< tad_> Well, that got me thinking about that lua hack in HttT S05b for a hidden speaker.
20160813 03:40:04< tad_> So I retested and, today, [hide_unit] is working fine and not flashing the unit on the screen.
20160813 03:40:47< tad_> Which means, today, at least, that hack isn't needed.
20160813 03:41:43< tad_> I'd prefer to use [hide_unit] and no hack but I'm worried that it's changing.
20160813 03:44:52< celmin> So, the button row looks better now, though not quite what I was going for (I wanted the space after refresh, not before).
20160813 03:45:06< celmin> (Plus there's still extra space to the left.)
20160813 03:45:52< celmin> It's close enough to commit though.
20160813 03:46:17< celmin> Anyway, this was more intended as "hey look what you can do" than anything permanent (though that doesn't mean it's okay to revert it wholesale).
20160813 03:47:16 * vultraz awaits commit
20160813 03:47:50< irker712> wesnoth: Celtic Minstrel wesnoth:master e38446aa4cf2 / data/gui/window/lobby_main.cfg: Add lower-resolution-friendly version of MP lobby https://github.com/wesnoth/wesnoth/commit/e38446aa4cf23a05f97d134f46848f968377bafd
20160813 03:48:35< celmin> BTW, you'll have to actually change your resolution to fiddle with it.
20160813 03:48:55< celmin> I can screenshot it if that's too much trouble though. :P
20160813 03:49:21< vultraz> interesting
20160813 03:49:32< vultraz> so [resolution] is useful after all
20160813 03:49:47< celmin> I'm pretty sure I've occasionally been trying to tell you this for awhile.
20160813 03:50:17< vultraz> anyway, i'll fix the layout
20160813 03:50:19< celmin> Since it didn't seem to get through, I figured an example might help. :P
20160813 03:50:23< vultraz> and no more grow_factor = 10
20160813 03:50:27< celmin> No, leave that.
20160813 03:50:36< celmin> If you don't, there's extra space between the checkboxes.
20160813 03:50:42< vultraz> I'll fix that
20160813 03:50:47< vultraz> jeez
20160813 03:51:16< celmin> If you can fix that and make the text box take all the remaining space without using grow_factor=10, then sure.
20160813 03:51:29< vultraz> what extra space are you talking about
20160813 03:51:34< celmin> But it's really silly to insist on only using grow_factor=1
20160813 03:51:38< vultraz> i just tested with that set to 1 and there's no extra space :|
20160813 03:51:53< celmin> What's your resolution?
20160813 03:51:57< vultraz> 800x600
20160813 03:52:08< celmin> There was definitely extra space for me… :|
20160813 03:52:46< celmin> Oh wait, there's two uses of grow_factor=10, is it the one ont line 594 that you changed?
20160813 03:53:00< celmin> (The other I don't care about. I was just trying things.)
20160813 03:53:05< celmin> ^on
20160813 03:53:14< vultraz> no
20160813 03:53:16< vultraz> the other one
20160813 03:53:20< vultraz> I will deal with that
20160813 03:53:31< vultraz> first, let me fix the obvious issues
20160813 03:53:38< celmin> Like the chat area? :P
20160813 03:53:50< vultraz> yes
20160813 03:53:55< vultraz> just fixed that
20160813 03:54:03< celmin> Amazing.
20160813 03:54:31< vultraz> changed the one at 594 and i still see no space
20160813 03:54:47< celmin> I definitely saw extra space between the filter's checkboxes without it..
20160813 03:55:00< celmin> It was uniformly distributed between all checkboxes.
20160813 03:55:02< celmin> More or less.
20160813 03:55:12< vultraz> that's good
20160813 03:55:27< celmin> No, it's not good, because that space is better used to make the actual filter box longer.
20160813 03:55:28< vultraz> hm
20160813 03:55:32-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Quit: Page closed]
20160813 03:55:36< vultraz> i shall take screenshots and compare
20160813 03:55:59< celmin> Maybe it would work if the checkboxes were set to not grow? I dunno.
20160813 03:56:49< vultraz> plus, we want it evenly distributed at large resolutions
20160813 03:57:13< celmin> Even at large resolutions I think it's better to squish the text boxes together so that the filter box is larger.
20160813 03:57:18< celmin> ^checkboxes
20160813 03:57:28< vultraz> but yes, setting grow_factor = 0 on the others is what one wants
20160813 03:57:36< vultraz> *this* is why I said don't use 10
20160813 03:57:38< celmin> Why would you want them evenly distributed?
20160813 03:57:41< vultraz> there's a way to do it with 0 and 1
20160813 03:58:07< celmin> Sure, not using 10 may be the right solution here, but it's still silly to never use grow_factors other than 0 or 1.
20160813 03:58:15< vultraz> There's no need to
20160813 03:58:25< celmin> There's definitely a (theoretical) need to.
20160813 03:58:25< vultraz> So I never do
20160813 03:58:36< vultraz> what.. have you done here
20160813 03:58:37< celmin> But sure, it might just happen that none of our dialogs require it.
20160813 03:58:39< vultraz> so many spacers :|
20160813 03:58:50< celmin> That's just so the two rows have the same number of columns.
20160813 03:58:56 * vultraz groans
20160813 03:58:58 * vultraz fixes
20160813 03:59:04< celmin> I suppose I could have used two separate one-row grids, but that doesn't feel any less ugly.
20160813 03:59:17< celmin> (ie, both ways are ugly)
20160813 03:59:26< vultraz> rule of thumb:
20160813 03:59:31< vultraz> try to minimize use of spacers
20160813 03:59:40< vultraz> unless they're for, ya know, spacing
20160813 03:59:50< vultraz> or you *explicitly* want column alignment 
20160813 04:00:20< celmin> True, precise alignment between the columns isn't really required here...
20160813 04:00:41< celmin> I mean, it might look good for "Log Out" to line up with "Join Game", I suppose...
20160813 04:00:47< vultraz> secondly
20160813 04:01:06< vultraz> always see if you can move a row to a parent grid
20160813 04:01:43< celmin> The parent grid is one column, so that won't work.
20160813 04:02:06< vultraz> then you use a child grid for that row
20160813 04:02:13< vultraz> I'll show you want I mean
20160813 04:03:08< vultraz> (I should really finish my wiki page on gui2 layout tactics)
20160813 04:03:54< celmin> How does the GUI2 engine work again? Does it take the contents of a [resolution] tag or those of a [window] tag?
20160813 04:04:01< vultraz> resolution
20160813 04:04:15< vultraz> at least, I think so
20160813 04:04:23< celmin> Can Lua detect the current screen resolution?
20160813 04:04:34< vultraz> i think widgets are fed the relevant resolution definition
20160813 04:04:38< vultraz> I don't know
20160813 04:05:11< celmin> Oh, by the way, I think the indentation in the filter box macro is a bit off.
20160813 04:09:26< irker712> wesnoth: Charles Dang wesnoth:master 7fb8d7dade97 / data/gui/window/lobby_main.cfg: MP Lobby: improved low-res layout https://github.com/wesnoth/wesnoth/commit/7fb8d7dade97bdadc63b5ae8fb1cbdbae03091ae
20160813 04:09:32< vultraz> celmin: ^
20160813 04:11:06< vultraz> notice the use of horizontal_grow for the chat box's *containing* column
20160813 04:11:30< vultraz> anyway, that should have fixed all your issues
20160813 04:11:56< celmin> Looks a lot better, indeed. The chat box still isn't taking up all the available vertical space though.
20160813 04:24:23< vultraz> celmin: i hope you know more about gui2 layout now
20160813 04:24:34< celmin> Maybe?
20160813 04:35:42-!- hk238 [~kvirc@t224.ip7.netikka.fi] has joined #wesnoth-dev
20160813 04:47:25-!- Kwandulin [~Miranda@p200300760F35BF2A986950BCBE579C6F.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160813 04:56:07-!- celmin [~celticmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The minstrel departs, to spread the music to the masses!]
20160813 04:57:44-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160813 04:59:02-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160813 04:59:05< vultraz> blah
20160813 04:59:13< vultraz> i am failing to fix the UPP
20160813 04:59:43< vultraz> the dialog wont stop resizing
20160813 05:00:09-!- noy [~Noy@wesnoth/developer/noy] has quit [Read error: Connection reset by peer]
20160813 05:02:34< celticminstrel> UPP?
20160813 05:02:41< celticminstrel> Though I seem to recall asking that earlier...
20160813 05:02:51< celticminstrel> Oh, unit preview pane...
20160813 05:02:54< celticminstrel> Right, okay.
20160813 05:03:33-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20160813 05:07:13-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160813 05:08:30< vultraz> I don't even know why
20160813 05:09:43-!- un214 [~un214@104.220.56.173] has quit [Remote host closed the connection]
20160813 05:27:07< shadowm> Yeah, I agree that these ad-hoc acronyms aren't a great idea.
20160813 05:27:32< celticminstrel> Ad-hoc acronyms are rarely a good idea.
20160813 05:27:51< celticminstrel> If you'd recently been speaking of it unacronymized, then maybe.
20160813 05:28:57< vultraz> except we had :P
20160813 05:29:08< celticminstrel> Not that recently.
20160813 05:44:42-!- noy [~Noy@wesnoth/developer/noy] has quit [Read error: Connection reset by peer]
20160813 05:49:48-!- travis-ci [~travis-ci@ec2-54-157-217-171.compute-1.amazonaws.com] has joined #wesnoth-dev
20160813 05:49:49< travis-ci> wesnoth/wesnoth#10359 (master - 6f179bf : gfgtdf): The build has errored.
20160813 05:49:49< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/151974280
20160813 05:49:49-!- travis-ci [~travis-ci@ec2-54-157-217-171.compute-1.amazonaws.com] has left #wesnoth-dev []
20160813 05:51:15-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160813 05:57:35-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20160813 06:01:52-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
20160813 06:03:35-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20160813 06:08:33-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Ping timeout: 276 seconds]
20160813 06:12:57-!- travis-ci [~travis-ci@ec2-54-227-209-210.compute-1.amazonaws.com] has joined #wesnoth-dev
20160813 06:12:58< travis-ci> wesnoth/wesnoth#10360 (master - bacece4 : Charles Dang): The build failed.
20160813 06:12:58< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/151981078
20160813 06:12:58-!- travis-ci [~travis-ci@ec2-54-227-209-210.compute-1.amazonaws.com] has left #wesnoth-dev []
20160813 06:32:33< irker712> wesnoth: Charles Dang wesnoth:master 9d3f3442b499 / / (3 files in 2 dirs): Obvious fixed to the Unit Preview Pane. More reworking needed https://github.com/wesnoth/wesnoth/commit/9d3f3442b4996eb2000abefd3ac60db8ee6dc90e
20160813 06:47:20-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160813 06:50:18-!- travis-ci [~travis-ci@ec2-54-157-217-171.compute-1.amazonaws.com] has joined #wesnoth-dev
20160813 06:50:19< travis-ci> wesnoth/wesnoth#10362 (master - e38446a : Celtic Minstrel): The build is still failing.
20160813 06:50:20< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/151990077
20160813 06:50:20-!- travis-ci [~travis-ci@ec2-54-157-217-171.compute-1.amazonaws.com] has left #wesnoth-dev []
20160813 06:51:42-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160813 07:07:12-!- travis-ci [~travis-ci@ec2-54-227-209-210.compute-1.amazonaws.com] has joined #wesnoth-dev
20160813 07:07:13< travis-ci> wesnoth/wesnoth#10363 (master - 7fb8d7d : Charles Dang): The build is still failing.
20160813 07:07:13< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/151992243
20160813 07:07:13-!- travis-ci [~travis-ci@ec2-54-227-209-210.compute-1.amazonaws.com] has left #wesnoth-dev []
20160813 07:13:48-!- JyrkiVesterinen [~JyrkiVest@87-100-192-6.bb.dnainternet.fi] has joined #wesnoth-dev
20160813 07:29:01-!- Kwandulin [~Miranda@p200300760F35BF2A986950BCBE579C6F.dip0.t-ipconnect.de] has quit [Ping timeout: 258 seconds]
20160813 07:30:26-!- travis-ci [~travis-ci@ec2-54-157-217-171.compute-1.amazonaws.com] has joined #wesnoth-dev
20160813 07:30:27< travis-ci> wesnoth/wesnoth#10364 (master - 9d3f344 : Charles Dang): The build is still failing.
20160813 07:30:27< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/152004208
20160813 07:30:27-!- travis-ci [~travis-ci@ec2-54-157-217-171.compute-1.amazonaws.com] has left #wesnoth-dev []
20160813 07:35:55-!- Kwandulin [~Miranda@p200300760F35BFDD986950BCBE579C6F.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160813 08:11:12-!- Kwandulin [~Miranda@p200300760F35BFDD986950BCBE579C6F.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160813 08:14:16-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20160813 08:29:17-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160813 08:31:56-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 258 seconds]
20160813 08:31:56-!- wedge010 is now known as wedge009
20160813 08:35:38-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160813 08:38:15< zookeeper> shadowm, actually, what was the reason for 800x600 as the minimum allowed, instead of 1024x768? i thought it was because of some mobile device.
20160813 08:38:25< zookeeper> or, s/was/is
20160813 08:39:59-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 258 seconds]
20160813 08:43:14< JyrkiVesterinen> Maybe netbooks with 1024x600 screens?
20160813 08:43:48< zookeeper> yeah, maybe. i don't know how common those are.
20160813 08:44:39< vultraz> what kind of screen is 1024 x 600
20160813 08:45:42< zookeeper> on an unrelated note, i'm wondering whether https://forums.wesnoth.org/viewtopic.php?f=12&t=44454 could be done in a non-intrusive manner, for some units at least. it's been suggested before and surprisingly enough i think it's in principle a sound idea. but it'd probably mix poorly with named characters.
20160813 08:46:04-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160813 08:46:44< vultraz> JyrkiVesterinen: be glad we dropped 800x480
20160813 08:47:22< JyrkiVesterinen> vultraz: During the netbook trend several years back, many popular netbook models such as Asus Eee PC 901 were equipped with 1024x600 screens.
20160813 08:47:33< vultraz> hm
20160813 08:47:37< vultraz> yeah, what happened to netbooks
20160813 08:47:45< vultraz> I haven't heard anything about them in awhile
20160813 08:49:00< vultraz> anyway
20160813 08:49:11< vultraz> out minimum res was 800x480 until earlier this year
20160813 08:49:25< vultraz> but that became too restrictive 
20160813 09:06:55< wedge009> vultraz: They got pushed out because of tablets. Which I hate because I prefer 10" netbooks but I'm definitely in the minority. But even then the smallest I have - which are several years old - is 1280x720. The smallest resolution I have (not even functional any more) is 1024x768.
20160813 09:08:17< wedge009> Oh, and I used to have a 10" 1024x600, but that was a crappy Atom so I sold it (apparently people still buy them).
20160813 09:11:34-!- Kwandulin [~Miranda@p200300760F35BFDDCC4834CD5EEA6D59.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160813 09:21:02-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has joined #wesnoth-dev
20160813 09:44:29-!- Kwandulin_2 [~Miranda@p200300760F35BFDD5011BEAFE8B6B7EA.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160813 09:45:09-!- Kwandulin [~Miranda@p200300760F35BFDDCC4834CD5EEA6D59.dip0.t-ipconnect.de] has quit [Ping timeout: 258 seconds]
20160813 09:50:47-!- Kwandulin_2 is now known as Kwandulin
20160813 09:55:56 * vultraz owns a tablet
20160813 10:07:25-!- JyrkiVesterinen [~JyrkiVest@87-100-192-6.bb.dnainternet.fi] has quit [Quit: .]
20160813 10:17:50< wedge009> Most people do. I'm just weird.
20160813 10:23:58-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160813 10:28:24-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 258 seconds]
20160813 10:30:29-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has quit [Quit: ChipmunkV]
20160813 10:30:59< vultraz> wedge009: you can't really hold a netbook
20160813 10:36:17< wedge009> Who says you need to?
20160813 10:36:30< wedge009> Like I said, I know I'm in the minority.
20160813 10:37:03< wedge009> But that's besides the point. The point is that even netbooks have a reasonable resolution.
20160813 10:37:42< wedge009> And besides, you asked the question.
20160813 10:41:53 * zookeeper doesn't own a tablet
20160813 10:42:25< zookeeper> although, miraculously, i am currently using an actual smartphone. but that's just because there was some weird problem with the mobile network which made my old phone not function right in it.
20160813 10:42:48< loonycyborg> my smartphone has resolution of800x480
20160813 10:42:59< loonycyborg> yet android port of wesnoth works on it
20160813 10:43:37< loonycyborg> I'd say such silly resolutions will exist for a long time yet
20160813 10:43:55< loonycyborg> and by not supporting them in mainline we're creating more work for android porters :P
20160813 10:48:54< vultraz> actually think that's also MY android phone's res
20160813 10:55:12< shadowm> zookeeper: 800x480 was. 800x600 was the minimum long before that for historical reasons.
20160813 10:55:32< zookeeper> mmkay.
20160813 11:00:15< wedge009> Huh, I thought 'smart' phones tend to have higher resolutions than desktop/laptop monitors because they're closer to the eye and all that.
20160813 11:01:26< loonycyborg> wedge009: some phones have hugeass resolutions, but they're expensive
20160813 11:02:20-!- RatArmy [~RatArmy@om126229084035.12.openmobile.ne.jp] has joined #wesnoth-dev
20160813 11:02:57< loonycyborg> older phones tend to have low res both because such small displays are harder to manufacture and because their cheapass gpus might have hard time supporting high res
20160813 11:09:35-!- irker712 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160813 11:15:02-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has joined #wesnoth-dev
20160813 11:18:50-!- gfgtdf [~chatzilla@x4e36920b.dyn.telefonica.de] has joined #wesnoth-dev
20160813 11:19:17< gfgtdf> is there somehwere a bugreport about savegame list beein empty on small horizontal resolution ?
20160813 11:20:59< vultraz> what?
20160813 11:21:01< vultraz> no
20160813 11:23:23< gfgtdf> when i play wesnoth on low horizontal resolution like 800x600 or 800x1000 the savegame list appears to be empty. Does this also happen to you ?
20160813 11:28:09< vultraz> no
20160813 11:28:45-!- JyrkiVesterinen [~JyrkiVest@78-27-108-95.bb.dnainternet.fi] has joined #wesnoth-dev
20160813 11:29:40< vultraz> it displays fine
20160813 11:36:40< JyrkiVesterinen> Regarding the smartphone screen resolution discussion: 800x480 is the so-called baseline resolution on Android. Anything lower than that is called small, and anything higher is called large.
20160813 11:37:33< JyrkiVesterinen> Of the commercial mobile games I have worked on, at least two of them supported resolutions down to 480x320 pixels (HVGA). I don't recall the minimum for the third one.
20160813 11:38:02< JyrkiVesterinen> Of course, this is a volunteer project and we can put the limit wherever we want.
20160813 11:40:46< JyrkiVesterinen> On the smartphone market, the screen resolution depends on both the age of the device and the price segment. The trend has been towards higher resolutions: the standard in high-end has gradually come up from 800x480 to 2560x1440 over the years.
20160813 11:41:21< JyrkiVesterinen> And, like loonycyborg said, cheaper devices use lower resolutions than high end.
20160813 11:41:59< vultraz>  2560x1440 on a smartphone? 
20160813 11:42:02< vultraz> I assume Retina?
20160813 11:43:11< JyrkiVesterinen> 1440p is the standard for high-end smartphones now. For example Samsung Galaxy Note 7 uses that resolution.
20160813 11:43:30< vultraz> Huh
20160813 11:43:33< JyrkiVesterinen> I even recall seeing a piece of news about a smartphone with a 4K screen.
20160813 11:43:46< vultraz> isn't the the S7?
20160813 11:44:00< JyrkiVesterinen> Such resolutions don't make any sense on such small displays, but they probably help marketing...
20160813 11:44:38< loonycyborg> they assume that all consumers are equipped with portable microscopes
20160813 11:45:15< vultraz> Not really.
20160813 11:45:24< vultraz> It doesn't mean stuff is tiny.
20160813 11:45:30< vultraz> Just really, really crisp.
20160813 11:45:43< loonycyborg> not necessarily
20160813 11:45:55< loonycyborg> at some point human eye can't resolve it anymore
20160813 11:46:01< vultraz> Last year I went from a gen 1 iPad mini to a gen 4 and holy *crap* the Retina display is so crisp and clean.
20160813 11:46:11< vultraz> and I've seen the iPhone 6s in action too
20160813 11:46:24< vultraz> the image crispness is incredible 
20160813 11:46:48< JyrkiVesterinen> I found the 4K phone I mentioned: https://en.wikipedia.org/wiki/Sony_Xperia_Z5_Premium
20160813 11:47:32< vultraz> huh
20160813 11:47:35< vultraz> impressive
20160813 11:47:37< vultraz> but it's sony
20160813 12:04:02< wedge009> I just see what my friends/colleagues use and they all seem to have high-res phones. So I thought that's what they all were. How does one play Wesnoth on a phone anyway? Hmm, I suppose its turn-based nature makes it easier than other games.
20160813 12:04:32< loonycyborg> it has mouse emulation
20160813 12:04:39< loonycyborg> I think proper touch support would be better
20160813 12:05:37< wedge009> The 'crispness' is only there because your eye can't resolve the individual pixels (hence the 'retina' marketing term). But if font sizes are left at what's suitable for desktop monitors, then yes, you'd have a hard time seeing things on a super-high-res phone.
20160813 12:06:15< wedge009> That's why there's all this stuff about how OSs handle high-DPI displays like phones.
20160813 12:06:57< vultraz> wesnoth doesn't have retina support, officially
20160813 12:07:05< vultraz> that is, it's not optimized for it
20160813 12:07:10-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has quit [Quit: ChipmunkV]
20160813 12:07:11< vultraz> even though we do have some 2x images
20160813 12:07:34< vultraz> anyway, i wouldn't know how it plays on mobile, since i dont have it on mobile :P
20160813 12:07:59< vultraz> my phone is low-end but my tablet is high-end, so i assume they'd play very differently
20160813 12:08:12< wedge009> Recently I was looking for a 27 inch monitor for my father that was still 'only' HD resolution - 1920x1080. He has trouble reading the small fonts on higher DPI displays. Looking around, I saw comments from a lot of people bemoaning why anyone would bother buying a larger monitor with only HD resolution.
20160813 12:08:41< wedge009> I would have thought they'd play the same or very similar way.
20160813 12:09:11< JyrkiVesterinen> Mobile platforms handle resolution differences much better than PCs.
20160813 12:09:50< wedge009> By design, I suppose, since they come in a wider variety of them.
20160813 12:09:54< JyrkiVesterinen> There have been attempts to improve the situation (for example, Microsoft has made some improvements in Windows 8 and more in Windows 10), but the situation still has a lot to improve.
20160813 12:10:20< wedge009> On Windows, certainly. Though I heard OSX does pretty well because of all those high-resolution screens.
20160813 12:10:22< JyrkiVesterinen> And it doesn't help the PC situation that most users are still using an operating system from 2009. Namely Windows 7.
20160813 12:10:30< wedge009> Hey! :p
20160813 12:10:48< wedge009> Actually, I know about that, because I read a lot of remarks about Win7 not handling DPI differences well.
20160813 12:11:16< wedge009> I haven't had too much trouble with it, but I only bumped mine to 114%
20160813 12:11:33-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160813 12:12:23< wedge009> Firefox handles custom DPI settings, but IE and Chromium don't - or I don't know how to configure them to.
20160813 12:16:27-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 276 seconds]
20160813 12:26:55-!- Kwandulin [~Miranda@p200300760F35BFDD5011BEAFE8B6B7EA.dip0.t-ipconnect.de] has quit [Ping timeout: 258 seconds]
20160813 12:28:05-!- Kwandulin [~Miranda@p200300760F35BFDD5011BEAFE8B6B7EA.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160813 12:33:37-!- Kwandulin [~Miranda@p200300760F35BFDD5011BEAFE8B6B7EA.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160813 12:40:50-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160813 12:47:25-!- louis94 [~~louis94@91.178.241.1] has joined #wesnoth-dev
20160813 12:48:28-!- boucman_work [~boucman@221.86.207.77.rev.sfr.net] has quit [Remote host closed the connection]
20160813 12:59:06-!- irker327 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160813 12:59:06< irker327> wesnoth: gfgtdf wesnoth:master 714661490601 / src/gui/widgets/unit_preview_pane.cpp: fix wrong size calculations in unit preview pane https://github.com/wesnoth/wesnoth/commit/714661490601af5268a5982e2f3a5668f94282ef
20160813 12:59:07< gfgtdf> vultraz: ^ that might fix the resize issues in recall dialog 
20160813 13:00:07< gfgtdf> vultraz: shoudl i change set_displayed_type to also sow tooltios or do you have another plan ?
20160813 13:01:02-!- boucman_work [~boucman@2a02-8428-034f-f800-9e32-0c7c-b391-6223.rev.sfr.net] has joined #wesnoth-dev
20160813 13:05:32< vultraz> gfgtdf: sure, go ahead
20160813 13:05:48< vultraz> i can always consolidate some of the nodes later.
20160813 13:13:19-!- louis94 [~~louis94@91.178.241.1] has quit [Ping timeout: 260 seconds]
20160813 13:15:28-!- fabi_ [~fabi@176.6.126.227] has joined #wesnoth-dev
20160813 13:15:31-!- fabi [~fabi@176.7.89.74] has quit [Read error: Connection reset by peer]
20160813 13:15:58< gfgtdf> vultraz: i made a bugreport about my loadscreen issue: https://gna.org/bugs/index.php?24967
20160813 13:17:08-!- RatArmy [~RatArmy@om126229084035.12.openmobile.ne.jp] has quit [Ping timeout: 258 seconds]
20160813 13:19:29-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160813 13:23:09-!- Kwandulin [~Miranda@p200300760F35BFDD8C5BC3219A3AD0B8.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160813 13:23:44-!- louis94 [~~louis94@91.178.241.1] has joined #wesnoth-dev
20160813 13:33:35-!- travis-ci [~travis-ci@ec2-54-92-211-36.compute-1.amazonaws.com] has joined #wesnoth-dev
20160813 13:33:36< travis-ci> wesnoth/wesnoth#10365 (master - 7146614 : gfgtdf): The build is still failing.
20160813 13:33:36< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/152040900
20160813 13:33:36-!- travis-ci [~travis-ci@ec2-54-92-211-36.compute-1.amazonaws.com] has left #wesnoth-dev []
20160813 13:37:24-!- louis94 [~~louis94@91.178.241.1] has quit [Ping timeout: 258 seconds]
20160813 13:43:32-!- louis94 [~~louis94@91.178.241.1] has joined #wesnoth-dev
20160813 13:57:50-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20160813 13:58:15-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 250 seconds]
20160813 14:02:52-!- louis94 [~~louis94@91.178.241.1] has quit [Ping timeout: 240 seconds]
20160813 14:05:04-!- celmin [~celticmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20160813 14:11:59-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160813 14:16:54-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20160813 14:17:17-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Client Quit]
20160813 14:20:05-!- louis94 [~~louis94@91.178.241.1] has joined #wesnoth-dev
20160813 14:22:31< celmin> Oh, another advantage of the abilities refactor is that traits such as "aged" and weapons or ability specials only ever added by [object] can get help descriptions.
20160813 14:22:44< celmin> ^help topics
20160813 14:29:27< zookeeper> humh. currently they only get topics if they're part of race or unit type definitions?
20160813 14:29:52-!- Kwandulin [~Miranda@p200300760F35BFDD8C5BC3219A3AD0B8.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160813 14:30:15< celmin> Yeah.
20160813 14:30:52< celmin> Because the game only sees the final result of preprocessing, so it has no way of knowing about them if they don't appear somewhere under [units].
20160813 14:31:23< zookeeper> right... i wonder if there's any other cases than aged in mainline that should receive a workaround for that.
20160813 14:36:12-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 276 seconds]
20160813 14:37:53< celmin> Loyal.
20160813 14:37:59< celmin> Just loyal and aged.
20160813 14:38:14< celmin> I don't know of any weapon specials or abilities in mainline that don't get help topics.
20160813 14:38:22< celmin> (That doesn't mean there aren't any, of course.)
20160813 14:40:46-!- mjs-de [~mjs-de@x4db5911b.dyn.telefonica.de] has joined #wesnoth-dev
20160813 14:42:42< zookeeper> loyal and aged both show up in the traits section.
20160813 14:43:41< celmin> I don't see them.
20160813 14:44:01< celmin> I think a quick and easy workaround would be to add them to fog clearer unit.
20160813 14:44:58< celmin> Though if that's a hidden unit, it might hide the traits too...
20160813 14:45:24 * zookeeper sees them both in 1.12.6 and 1.13.5
20160813 14:45:55< celmin> Ah, so as I thought it's the result of the new trait topic generator.
20160813 14:46:18< celmin> Prior to that, trait descriptions were all hard-coded rather than being generated.
20160813 14:46:23< celmin> ^trait topics
20160813 14:46:33< celmin> (Hard-coded in help.cfg or whatever.)
20160813 14:48:07< zookeeper> well, fixing that by adding the traits/abilities/etc to the fog clearer or other dummy unit would have been the fix i'd suggest
20160813 14:53:45-!- louis94 [~~louis94@91.178.241.1] has quit [Ping timeout: 276 seconds]
20160813 14:55:58-!- horrowind [~Icedove@2a02:810a:83c0:404:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20160813 15:02:15-!- louis94 [~~louis94@91.178.241.1] has joined #wesnoth-dev
20160813 15:08:20-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160813 15:15:42-!- Bonobo [~Bonobo@2001:44b8:254:3200:59c6:97d3:2da7:c730] has quit [Quit: Leaving]
20160813 15:16:03-!- louis94 [~~louis94@91.178.241.1] has quit [Ping timeout: 264 seconds]
20160813 15:16:56-!- noy [~Noy@wesnoth/developer/noy] has quit [Read error: Connection reset by peer]
20160813 15:40:40< irker327> wesnoth: Jyrki Vesterinen wesnoth:staging-event-handlers c10a1dbc0802 / changelog players_changelog src/widgets/textbox.cpp src/widgets/textbox.hpp: Revert "Fix a stray ; character appearing in the command console" https://github.com/wesnoth/wesnoth/commit/c10a1dbc0802b907b115852af1d725f4db6bfac1
20160813 15:40:42< irker327> wesnoth: Jyrki Vesterinen wesnoth:staging-event-handlers f621d7fa4602 / changelog players_changelog src/events.cpp src/events.hpp: Add new event handlers to a staging area https://github.com/wesnoth/wesnoth/commit/f621d7fa46025300a314da3004db8ddc53982c18
20160813 15:47:36< gfgtdf> i wonder why that ^ was reverted? I mena i agree that a lower level fix would be better but we shoudl still wait with reverting until we actuall have abetter fix i'd say.
20160813 15:47:58< JyrkiVesterinen> Note that I only reverted it in a branch. My earlier fix is still in master.
20160813 15:48:17< celmin> Oh. Huh.
20160813 15:48:19< JyrkiVesterinen> In addition, the following commit fixes the problem in a different way.
20160813 15:48:24< celmin> I didn't notice that. XD
20160813 15:48:24< gfgtdf> JyrkiVesterinen: ah ok its fine then
20160813 15:48:32< celmin> (That it was a branch.)
20160813 15:50:35< irker327> wesnoth: gfgtdf wesnoth:master 4d84d42c47d5 / src/ (20 files in 9 dirs): iterator_range instead of pair for config::child_itors https://github.com/wesnoth/wesnoth/commit/4d84d42c47d5ae66d8cbe2134b01a74ca2836a22
20160813 15:50:37< irker327> wesnoth: gfgtdf wesnoth:master 59284452c6dc / / (3 files in 2 dirs): show ability and wepon specials tooltips in unit recruti dialog. https://github.com/wesnoth/wesnoth/commit/59284452c6dcd009b685eba865a4317d2b44b679
20160813 15:52:21< celmin> gfgtdf: If you're doing config stuff anyway, what about 1) the same for all_children_iterator() and attribute_range() and 2) a non-const version of all_children_iterator()?
20160813 15:53:05< gfgtdf> celmin: maybe i'll do that later.
20160813 15:53:29< celmin> I was planning to do #2 at some point, not sure when though, so let me know if you start it.
20160813 15:54:05-!- crimson_penguin [~crimson_p@wesnoth/developer/crimsonpenguin] has quit [Quit: ZNC - http://znc.sourceforge.net]
20160813 15:54:05-!- Jetrel_bot [~Jetrel@ec2.happyspork.com] has quit [Quit: ZNC - http://znc.in]
20160813 15:54:05-!- DDR [~david@ec2.happyspork.com] has quit [Quit: DDR is not Dance Dance Revolution]
20160813 15:55:12< celmin> Operator bool is the opposite of .empty()?
20160813 15:55:21< gfgtdf> celmin: i dont have plans for #2 maybe i'll do #1
20160813 15:56:32-!- louis94 [~~louis94@91.178.241.1] has joined #wesnoth-dev
20160813 15:56:55-!- crimson_penguin [~crimson_p@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20160813 15:57:55-!- Jetrel_bot [~Jetrel@ec2.happyspork.com] has joined #wesnoth-dev
20160813 15:58:27-!- DDR [~david@ec2.happyspork.com] has joined #wesnoth-dev
20160813 16:07:19< vultraz> what... is a zip iterator
20160813 16:07:49< celmin> Good question!
20160813 16:08:24< vultraz> I sense sarcasm 
20160813 16:08:32< gfgtdf> vultraz: it allows you to iterate over 2 collections at the same time.
20160813 16:08:43< vultraz> huh
20160813 16:08:51< celmin> Not sure if it's related, but zip() is normally an operation mapping eg [[1,2], [3,4]] to [[1,3], [2,4]].
20160813 16:09:21< celmin> So if you zip two vectors, you should get a vector of pairs.
20160813 16:09:36< celmin> Where the first pair is the first element of each vector and so forth.
20160813 16:12:57< gfgtdf> celmin: it doesnt the same, except that it doesnt realy create a vector of pairs but a iterator_range that behaves like iterating over a vector of pairs
20160813 16:13:31< celmin> Close enough.
20160813 16:13:42< vultraz> hm
20160813 16:13:47< celmin> It only does two containers though? Not an arbitrary number of containers?
20160813 16:13:51< vultraz> so, i really...really don't understand anything going on in this line
20160813 16:13:54< vultraz> auto zip(const T&... containers) -> boost::iterator_range<boost::zip_iterator<decltype(boost::make_tuple(std::begin(containers)...))>>
20160813 16:13:55< vultraz> what is ->
20160813 16:13:59< vultraz> what is decltype
20160813 16:14:01< vultraz> what is >>
20160813 16:14:07< gfgtdf> celmin: arbitay number o containers that have to have teh same length
20160813 16:14:33< celmin> vultraz: That's not even the full line. :P
20160813 16:14:48< celmin> It's clearly a template declaration.
20160813 16:15:00-!- louis94 [~~louis94@91.178.241.1] has quit [Ping timeout: 258 seconds]
20160813 16:15:06< celmin> So, "auto" as a return type is a signal that the return type will be given lambda-style, using ->.
20160813 16:15:30< celmin> decltype() means "the type that would be returned by this expression".
20160813 16:16:02< celmin> The >> is just two closing angle brackets for template argument lists.
20160813 16:16:08< vultraz> oh
20160813 16:16:18< vultraz> i thought that had to be > >
20160813 16:16:20< celmin> Where is that line anyway?
20160813 16:16:28< celmin> vultraz: Not since C++11.
20160813 16:16:32< vultraz> ahhh
20160813 16:16:41< vultraz> he just added it in https://github.com/wesnoth/wesnoth/commit/59284452c6dcd009b685eba865a4317d2b44b679
20160813 16:16:56< celmin> Aha.
20160813 16:17:08< celmin> I only looked at the other commit.
20160813 16:17:49-!- louis94 [~~louis94@91.178.241.1] has joined #wesnoth-dev
20160813 16:18:06< celmin> BTW gfgtdf, if that works without the trailing return type, feel free to wrap that in an ifdef.
20160813 16:18:22< celmin> (I think C++14 can automatically deduce the return type.)
20160813 16:19:21< celmin> ("that" means the -> clause.)
20160813 16:19:26< gfgtdf> vultraz: i just copied that function form the net without thinking about it too much about its implementation (which doesn't mean that i wouldn't understand it if i did.)
20160813 16:19:46< vultraz> gfgtdf: you mean celmin?
20160813 16:19:56 * celmin shrug
20160813 16:20:11< vultraz> is that why it uses boost::make_tuple
20160813 16:20:15< vultraz> and boost::get
20160813 16:20:44< gfgtdf> vultraz: mostlileley becasue make_zip_iterator epects a bost::tuple dont know whether it also wokrs with std::tuple
20160813 16:21:02< celmin> Worth trying, I suppose.
20160813 16:21:17 * vultraz is doing some minor cleanup
20160813 16:21:21< celmin> But I wouldn't be too surprised if gfgtdf is right.
20160813 16:21:26< celmin> What kind of minor cleanup?
20160813 16:21:31< vultraz> mostly formatting
20160813 16:21:52< vultraz> might be able to remove the label_details_ member.
20160813 16:21:59< vultraz> don't think it's used anymore..
20160813 16:22:32< vultraz> gfgtdf: btw, what about the attack dialog? that doesn't have tooltips for its traits.
20160813 16:23:50< gfgtdf> vultraz: hmm might be possible to do thae same way in the attack dialog, don't know for sure.
20160813 16:24:38< gfgtdf> vultraz: there mogth be some problems though, like with right-alined elements in treeview, not sure whether that's possible.
20160813 16:24:50< vultraz> yeah...
20160813 16:24:53< vultraz> might be problems :|
20160813 16:25:15< vultraz> i kinda feel like we're abusing treeviews :P
20160813 16:25:24< celmin> Why wouldn't right-aligned elements be possible in a tree view?
20160813 16:25:36< celmin> (Also, you're definitely abusing treeviews. :P )
20160813 16:25:55< vultraz> I wonder if this functionality could not be generalized
20160813 16:26:09< vultraz> for the mp create stuff, their behavior as advertised is perefect
20160813 16:26:12< gfgtdf> vultraz: the main reason why i implemented this was for the recall dialog (its annoying when you play worldconquest2 and cannot see what training aour units have unless you recall them.)
20160813 16:26:21< celmin> You're basically using a tree view as a sequential layout manager.
20160813 16:26:35< vultraz> for the rest, that ^
20160813 16:26:51< celmin> You used tree views as intended in MP Create?
20160813 16:26:57< celmin> How can I see MP Create anyway?
20160813 16:27:28< vultraz> invert the new_widgets conditional in game_initialization/multiplayer.cpp
20160813 16:27:33< vultraz> and yes, I did
20160813 16:27:57< vultraz> https://drive.google.com/file/d/0B-mR9s8FduLLMGh0QmRNZUVCbFk/view?usp=sharing
20160813 16:28:09< vultraz> i realized being able to compress the options is a great thing
20160813 16:28:10< celmin> I bet it'll crash if I try to open it in 800x600.
20160813 16:28:20< vultraz> no, you just get scrollbars :)
20160813 16:28:21< celmin> …oh wait, you removed that assert didn't you.
20160813 16:28:34< vultraz> did I?
20160813 16:28:42< vultraz> I recall doing something of the sort
20160813 16:28:45< celmin> So if it would have crashed before, now you just get missing elements.
20160813 16:28:55< vultraz> i get no missing elements
20160813 16:28:58< celmin> Sorry, why'd you link the screenshot suddenly?
20160813 16:29:11< vultraz> to refresh your memory
20160813 16:29:14< celmin> BTW, any ideas for how to fix the low-res lobby?
20160813 16:29:17< vultraz> and prove I was using them s intended
20160813 16:29:29< celmin> Does the tree view have a node for each source of options then?
20160813 16:29:39< vultraz> yes
20160813 16:29:43< celmin> Nice.
20160813 16:29:57< celmin> We should use a tree view for the rooms/player list in the MP lobby, too.
20160813 16:30:08< vultraz> players list DOES use a treeview
20160813 16:30:10< vultraz> somehow
20160813 16:30:15< vultraz> it's ugly
20160813 16:30:19< celmin> Huh? Isn't it a listbox?
20160813 16:30:26< celmin> The WML looks like it's a listbox.
20160813 16:30:38< vultraz> it confuses me :|
20160813 16:30:45< vultraz> i still haven't been able to decipher it
20160813 16:30:50< vultraz> and I know gui2 wml better than anyone
20160813 16:30:51< celmin> I see...
20160813 16:31:40< vultraz> I'll have to look closer at it
20160813 16:32:09< vultraz> honestly
20160813 16:32:17< vultraz> that entire area needs to be redesigned
20160813 16:32:24< gfgtdf> vultraz: i think we should show the [options] in he order they appear in the config instead of first checkboxes then comboboxes....   . It might be that the author grouped  them do categories or somethign and that'd be messed up otherwise.
20160813 16:32:45< celmin> Yeah, you should probably use all_children_range()
20160813 16:32:57-!- louis94 [~~louis94@91.178.241.1] has quit [Ping timeout: 250 seconds]
20160813 16:33:01< vultraz> perhaps..
20160813 16:33:06< fabi_> vultraz: hi
20160813 16:33:06< vultraz> but I still want like options grouped together
20160813 16:33:08< celmin> As for the rooms/player list, I suppose you could just reimplement it from scratch.
20160813 16:33:08-!- louis94 [~~louis94@91.178.241.1] has joined #wesnoth-dev
20160813 16:33:10< vultraz> that's what gui2 does too
20160813 16:33:14< vultraz> er
20160813 16:33:16< vultraz> gui1
20160813 16:33:28< celmin> GUI1 did not respect the order of the options?
20160813 16:34:02< celmin> I think it's better to respect the order though, not group them by type. If the author wants them grouped by type, they can do it themselves.
20160813 16:34:09< vultraz> no, no :|
20160813 16:34:11< fabi_> vultraz: Let me ask, when do you want to freeze for 1.14?
20160813 16:34:18< vultraz> don't encourage bad ui practices
20160813 16:34:27< vultraz> fabi_: late this year
20160813 16:34:47< celmin> It's not bad practices, it's giving the author the freedom to group related options together even if they are of different types.
20160813 16:34:49-!- travis-ci [~travis-ci@ec2-54-157-217-171.compute-1.amazonaws.com] has joined #wesnoth-dev
20160813 16:34:50< travis-ci> wesnoth/wesnoth#10366 (staging-event-handlers - f621d7f : Jyrki Vesterinen): The build has errored.
20160813 16:34:50< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/152062484
20160813 16:34:50-!- travis-ci [~travis-ci@ec2-54-157-217-171.compute-1.amazonaws.com] has left #wesnoth-dev []
20160813 16:35:15< fabi_> vultraz: So 1.14 is to be expected in mid 2017?
20160813 16:35:24< vultraz> early, hopefully
20160813 16:35:39< fabi_> hmmm
20160813 16:35:40< celmin> I hope I can get my stuff done by then… >_>
20160813 16:35:42< vultraz> celmin: hm, it looks like gui1 did respect the order
20160813 16:35:47< vultraz> I don't like this
20160813 16:35:49< vultraz> it looks very ugly
20160813 16:35:54< celmin> I think it's better to respect the order.
20160813 16:36:19< vultraz> also, you'll need to tell me how to check what the tag name is
20160813 16:36:49< celmin> for(config::any_child blah : cfg.all_children_range() {if(blah.key == "whatever") {…}}
20160813 16:36:58< vultraz> I see
20160813 16:37:02< vultraz> I'll consider this
20160813 16:37:05< celmin> blah.cfg is the contents
20160813 16:37:06< gfgtdf> vultraz: when your addon has for example optiosn to change the map generation and other option to change the ai strength then i'd epct that the map settigns are grouped together and the ai setting not first checboxed then sliders.
20160813 16:37:08< vultraz> for now, layout stays fixed
20160813 16:37:24< vultraz> dialog is still very wip, after all.
20160813 16:37:31< vultraz> and the order can be easily changed
20160813 16:39:44< vultraz> you make a good point, but i also feel it looks better grouped by type
20160813 16:39:52< vultraz> easier for players to look at
20160813 16:40:09< vultraz> and it's a step away from gui1's hodgepodge layout practices
20160813 16:40:40< celmin> It's not better if the options are all mixed up with regards to their meaning.
20160813 16:41:02< vultraz> as I said, you make a good point
20160813 16:41:07< vultraz> I shall have to consider this
20160813 16:41:30< celmin> Please do more than consider it.
20160813 16:42:35< vultraz> by that you mean "do it", I suppose :p
20160813 16:45:01< gfgtdf> celmin: im just playing scenairo 3 of your campaign and somwhoe the merman in the village that you shodul save doesn't gets healed by that village.
20160813 16:45:16-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has joined #wesnoth-dev
20160813 16:45:38< celmin> gfgtdf: 1.12 or 1.13?
20160813 16:45:57< gfgtdf> celmin: 1.13, but i think this effect 1.12 aswell
20160813 16:46:02< celmin> Weird.
20160813 16:46:19< gfgtdf> celmin: the problem is mostlikeley that he has controller=no so he doesnt get a turn and thus no healing.
20160813 16:46:26< gfgtdf> contorller=null
20160813 16:46:34< celmin> Ah, I suppose that makes sense as an explanation.
20160813 16:47:10< vultraz> eh
20160813 16:47:12< vultraz> hm
20160813 16:47:21< vultraz> the attack names weren't supposed to be indented
20160813 16:47:24< vultraz> :
20160813 16:47:27< celmin> ?
20160813 16:48:17< vultraz> tree view made it so it shows as
20160813 16:48:20< vultraz> Attacks
20160813 16:48:22< vultraz>   Name
20160813 16:48:25< vultraz>     Details
20160813 16:48:42< vultraz> before, Name was flush with the header..
20160813 16:48:48< celmin> Honestly, that seems okay to me.
20160813 16:49:27< vultraz> I feel it throws the aesthetic 
20160813 16:49:39< celmin> I think it makes a lot of sense.
20160813 16:49:56< vultraz> obviously, details should be indented under name
20160813 16:50:07< vultraz> the question is whether both should be indented under the header
20160813 16:50:11< celmin> I think it makes a lot of sense.
20160813 16:50:46< vultraz> traits and abilities are, but I didn't do that that with attacks specifically to avoid multiple levels of indent
20160813 16:51:30-!- travis-ci [~travis-ci@ec2-54-157-217-171.compute-1.amazonaws.com] has joined #wesnoth-dev
20160813 16:51:31< travis-ci> wesnoth/wesnoth#10368 (master - 5928445 : gfgtdf): The build is still failing.
20160813 16:51:31< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/152063627
20160813 16:51:31-!- travis-ci [~travis-ci@ec2-54-157-217-171.compute-1.amazonaws.com] has left #wesnoth-dev []
20160813 16:51:48< celmin> There's nothing wrong with multiple levels of indent.
20160813 16:51:53< vultraz> there very much is
20160813 16:52:00< vultraz> anyway, im off to get sleep
20160813 16:52:00< celmin> There very much isn't.
20160813 16:52:30< vultraz> i would rather remove the headers than have two levels of indent
20160813 16:52:40< celmin> That's ridiculous.
20160813 16:52:52< vultraz> perhaps
20160813 16:52:59< vultraz> but it just... *bothers* me
20160813 16:53:13< vultraz> dunno why
20160813 16:53:16< gfgtdf> i dont realyl see an iassue with multiple levels of indention,
20160813 16:53:20< gfgtdf> however
20160813 16:53:22< gfgtdf> to change the level of indention change the first parmerter of add_name_tree_node to uese the next hight node
20160813 16:53:28< gfgtdf> higher*
20160813 16:53:45< celmin> Hmm, the sidebar does do it that way, though.
20160813 16:54:13< celmin> There's no attacks header, but the attacks are as a block indented relative to the race name.
20160813 16:54:22< vultraz> that is acceptable
20160813 16:54:46< celmin> Well, I'm not sure the sidebar is a good model here though.
20160813 16:55:00< gfgtdf> hmm but this then means te sidebar has also multiple levls of indention.
20160813 16:55:03< celmin> It's race name -> [traits, attacks]
20160813 16:55:28< celmin> Yes.
20160813 17:14:24-!- Kwandulin [~Miranda@p200300760F35BFDD08D6214154F79916.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160813 17:24:27-!- JyrkiVesterinen [~JyrkiVest@78-27-108-95.bb.dnainternet.fi] has quit [Quit: .]
20160813 17:48:36-!- hk238 [~kvirc@t224.ip7.netikka.fi] has quit [Quit: http://www.kvirc.net/ 4.9.1 Aria]
20160813 17:49:33-!- louis94 [~~louis94@91.178.241.1] has quit [Ping timeout: 240 seconds]
20160813 17:52:09< irker327> wesnoth: gfgtdf wesnoth:master 3cc2d0962ac5 / / (3 files in 2 dirs): more fixes to unit_preview_pane and treeviews. https://github.com/wesnoth/wesnoth/commit/3cc2d0962ac58668b3bf9ae18328f945b7a562fa
20160813 17:58:29< gfgtdf> vultraz: pleasy try if you still ahve issues with resizing ^
20160813 18:01:17-!- Kwandulin [~Miranda@p200300760F35BFDD08D6214154F79916.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160813 18:04:03-!- ancestral [~ancestral@63.84.240.98] has joined #wesnoth-dev
20160813 18:11:57-!- travis-ci [~travis-ci@ec2-54-157-217-171.compute-1.amazonaws.com] has joined #wesnoth-dev
20160813 18:11:58< travis-ci> wesnoth/wesnoth#10369 (master - 3cc2d09 : gfgtdf): The build is still failing.
20160813 18:11:58< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/152081215
20160813 18:11:58-!- travis-ci [~travis-ci@ec2-54-157-217-171.compute-1.amazonaws.com] has left #wesnoth-dev []
20160813 18:14:56-!- ancestral [~ancestral@63.84.240.98] has quit [Quit: i go nstuf kthxbai]
20160813 18:38:17-!- gfgtdf [~chatzilla@x4e36920b.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 48.0/20160726073904]]
20160813 19:17:35-!- mjs-de [~mjs-de@x4db5911b.dyn.telefonica.de] has quit [Remote host closed the connection]
20160813 19:34:28-!- horrowind [~Icedove@2a02:810a:83c0:404:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20160813 19:53:45-!- gfgtdf [~chatzilla@x4e36920b.dyn.telefonica.de] has joined #wesnoth-dev
20160813 19:54:08< gfgtdf> celmin: any idea here https://forums.wesnoth.org/viewtopic.php?p=600305 ?
20160813 19:55:38< celmin> Looks like the same problems ancestral was having.
20160813 19:55:53< gfgtdf> celmin: below, he got linking errors
20160813 19:56:34< celmin> Ah.
20160813 19:57:15< celmin> I think I know the problem, but not how to fix it.
20160813 19:57:33< celmin> Actually, you said it yourself in the thread.
20160813 19:57:54< celmin> BTW, .mm is Objective-C++
20160813 19:58:07< celmin> .m is Objective-C and I think stands for "module", which is weird but whatever.
20160813 20:07:11< celmin> Huh, it looks like I "accidentally" fixed an issue with fake units.
20160813 20:19:15< gfgtdf> did something happen with attack ccalulations? i feel the liek default selected option in the attack dialog is quiet stupid now
20160813 20:20:02< fabi_> Is there a syntax like x,y = (1,2,3),(1,2,3)?
20160813 20:20:23< gfgtdf> i mean i have a lvl3 elvish sorcess, the enemy has 9hp and it stil has teh 5x4 slow instead of 7x4 macical attack.
20160813 20:21:16< zookeeper> fabi_, not that i know of
20160813 20:22:12< celmin> I doubt that would work, since as I understand multi-assignment is resolved when the config is parsed, while comma-separated values are resolved when the value is parsed for use.
20160813 20:22:32< celmin> I don't think it would be a good idea for it to work, either.
20160813 20:23:12< gfgtdf> 9x4 macical attack i  meant.
20160813 20:23:50< celmin> gfgtdf: Well, there was Jyrki's Monte Carlo thing, not sure if that would affect it though.
20160813 20:24:26< celmin> gfgtdf: It also depends on the attacker's HP, right?
20160813 20:25:35< gfgtdf> celmin: it shoudl but if attack2 has 1) better accuracy (magical), 2) can one-hit the defender, 3) has at lest as many strikes then its oibvious for me that attack2 is teh better option.
20160813 20:25:55< celmin> So what if it doesn't one-hit the defender?
20160813 20:26:12< zookeeper> celmin, it's obviously not a big problem, but i've occasionally thought that it'd be nice if you could somehow mix multi-assignment and comma-separated lists.
20160813 20:26:25< gfgtdf> then the enemy surives? i dotn realyl understand that question
20160813 20:26:58< celmin> I mean, if the sorceress is also low on HP, and not one-hitting the defender means a risk to her life, then I'd say slow might still be better.
20160813 20:27:29< celmin> I don't really know if there's a problem here, anyway.
20160813 20:27:52-!- louis94 [~~louis94@91.178.241.1] has joined #wesnoth-dev
20160813 20:27:56< gfgtdf> celmin: well yes but the chanches of attack2 killing the defender are still ight than attack1 slowing the defender, becasue of what i said aboive
20160813 20:27:58< zookeeper> i'd like to see the exact situation to be able to say anything
20160813 20:28:22< celmin> ^
20160813 20:28:40< celmin> It's impossible to say whether there's a problem without knowing the exact situation.
20160813 20:29:06< celmin> Exactly which units are present, their hit points, whether they're already slowed, etc.
20160813 20:29:14-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160813 20:32:51-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 264 seconds]
20160813 20:32:52-!- wedge010 is now known as wedge009
20160813 20:34:19< gfgtdf> how so? can you give any situiation like what i said above where atack1 would be better?
20160813 20:35:06< zookeeper> yes, if the slowing attack gives a lower chance to die
20160813 20:35:27< zookeeper> or even a lower expectation of retaliation damage
20160813 20:36:23< zookeeper> (which, if there is no chance to die with either attack, it always does... unless the target is already slowed)
20160813 20:37:59< gfgtdf> hmm ok this is an attack without retalation damage. (against a melee only unit)
20160813 20:39:45< zookeeper> well you should have said so :p in that case, eh... yeah i'd always pick the magical attack because it has a higher chance to kill, but that isn't a generalizable formula
20160813 20:40:10< zookeeper> if it required 3 hits to kill with the magical attack, i might rather slow it instead
20160813 20:40:15< gfgtdf> ok from looking at the code im quite sure it was caused  from  265e41dbdad3ece6ae45374dcf2174bdc0f385cc
20160813 20:40:27< zookeeper> and that'd require on what other units are available
20160813 20:40:35< gfgtdf> zookeeper: i said above that it takes onyl 1 hit to kill the enemy with the magical attack in this case.
20160813 20:40:41< zookeeper> eh, s/require/depend (weird typo)
20160813 20:41:03< zookeeper> gfgtdf, oh right
20160813 20:42:16 * zookeeper wonders why tortoisegit is so dreadfully slow in locating commits based on hash
20160813 20:44:41< gfgtdf> JyrkiVesterinen: since 265e41dbdad3ece6ae45 mouse_handler::fill_weapon_choices return soff-by-one indicies please fix that.
20160813 20:45:08< celmin> If you know the problem why can't you fix it yourself?
20160813 20:45:32< gfgtdf> off*
20160813 20:48:26< gfgtdf> celmin: well i dont feel save enought in this code that i never wokr on before to commit this untested on github webpage.
20160813 20:48:47-!- fabi [~fabi@176.0.89.115] has joined #wesnoth-dev
20160813 20:48:54< celmin> Oh, okay.
20160813 20:51:43-!- fabi_ [~fabi@176.6.126.227] has quit [Ping timeout: 252 seconds]
20160813 20:52:31-!- irker327 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160813 20:53:38-!- gfgtdf [~chatzilla@x4e36920b.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 48.0/20160726073904]]
20160813 20:56:34-!- mjs-de [~mjs-de@x4db5911b.dyn.telefonica.de] has joined #wesnoth-dev
20160813 20:57:08-!- irker978 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160813 20:57:08< irker978> wesnoth: Jyrki Vesterinen wesnoth:master c10a1dbc0802 / changelog players_changelog src/widgets/textbox.cpp src/widgets/textbox.hpp: Revert "Fix a stray ; character appearing in the command console" https://github.com/wesnoth/wesnoth/commit/c10a1dbc0802b907b115852af1d725f4db6bfac1
20160813 20:57:08< irker978> wesnoth: Jyrki Vesterinen wesnoth:master f621d7fa4602 / changelog players_changelog src/events.cpp src/events.hpp: Add new event handlers to a staging area https://github.com/wesnoth/wesnoth/commit/f621d7fa46025300a314da3004db8ddc53982c18
20160813 20:57:09< irker978> wesnoth: Andreas wesnoth:master 9469b9b6c101 / src/ (events.cpp events.hpp widgets/textbox.cpp widgets/textbox.hpp): Merge pull request #748 from wesnoth/staging-event-handlers https://github.com/wesnoth/wesnoth/commit/9469b9b6c101b125df34bedf80555223a9474d99
20160813 20:59:48< Aginor> I looked at the failing travis build and it appears that the problems there are unrelated
20160813 21:00:13-!- louis94 [~~louis94@91.178.241.1] has quit [Ping timeout: 244 seconds]
20160813 21:00:25< Aginor> it'd be nice if we could keep master in a pretty reasonable state to avoid doubts like those :)
20160813 21:00:38< celmin> Yeah, I think I know how to fix Travis but haven't gotten around to it.
20160813 21:02:03-!- mjs-de [~mjs-de@x4db5911b.dyn.telefonica.de] has quit [Remote host closed the connection]
20160813 21:16:20-!- travis-ci [~travis-ci@ec2-54-92-211-36.compute-1.amazonaws.com] has joined #wesnoth-dev
20160813 21:16:21< travis-ci> wesnoth/wesnoth#10370 (master - 9469b9b : Andreas): The build is still failing.
20160813 21:16:21< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/152104507
20160813 21:16:21-!- travis-ci [~travis-ci@ec2-54-92-211-36.compute-1.amazonaws.com] has left #wesnoth-dev []
20160813 21:21:51-!- louis94 [~~louis94@91.178.241.1] has joined #wesnoth-dev
20160813 21:28:09-!- gfgtdf [~chatzilla@x4e36920b.dyn.telefonica.de] has joined #wesnoth-dev
20160813 21:29:21< gfgtdf> hmm this seems to be partly caused by my iterator_range commit, it seems like iterator_range::drop_front is not avialabe in eerlier boost versions.
20160813 21:30:13< gfgtdf> the best fix is mostlikley to replace dro_front with advance_begin(1), will do this later
20160813 21:31:05< gfgtdf> added in boost 1.56 that is
20160813 21:34:53< celmin> Or we could update Travis to use a newer Boost.
20160813 21:44:51-!- Appleman1234 [~Appleman1@KD036012030225.au-net.ne.jp] has quit [Ping timeout: 264 seconds]
20160813 21:49:35< celmin> Actually, I think that would be the better option anyway, gfgtdf.
20160813 21:50:34< gfgtdf> celmin: i agree that updating  travis toa  newer version of boost woudl be nice, this might then also give us a c++14 test on boost.
20160813 21:50:57< celmin> Why do we list separate version requirements for each Boost library?
20160813 21:51:01< celmin> shadowm: Any idea?
20160813 21:51:05 * celmin is looking at INSTALL
20160813 21:51:19< shadowm> Nah, it's pointless.
20160813 21:51:29< celmin> I'm pretty sure that also doesn't cover everything Boost that's used.
20160813 21:51:53< celmin> Okay, if it's pointless, I'm just going to shorten it to specify 1.56, for now.
20160813 21:52:01< celmin> Might bump it up later, if necessary.
20160813 21:52:05< gfgtdf> i woundt even assume that differnt version of boost worth together, 
20160813 21:52:12< celmin> 1.56 for all Boost libs
20160813 21:52:19< shadowm> Well, I guess there is the fact that for a while Asio was an optional dependency and it bumped the requirements a bit IIRC.
20160813 21:52:31< shadowm> gfgtdf: No, they obviously don't, at least not reliably.
20160813 21:53:13< gfgtdf> yes, that 'woundt' meant 'woundn't'
20160813 21:53:40< shadowm> Anyway right now I think there's no optional compile options/targets that change the requirements Boostways so you should be able to just group it all under a single version that's the greatest of the bunch.
20160813 21:53:41< gfgtdf> i'd just take the hghtext of what we current need
20160813 21:54:05< shadowm> But then again I've been out of touch with master for 5 months.
20160813 21:54:08< celmin> Any point still listing the ones that aren't header-only? I suppose maybe not.
20160813 21:54:22< shadowm> That's a good question.
20160813 21:55:28< shadowm> In Debian all header only libraries are provided together as one big Boost package. There are only separate packages for libraries that provide shared objects (think DLLs).
20160813 21:55:50< shadowm> I have no idea what's the approach taken by other distributions.
20160813 21:56:01< gfgtdf> celmin: i think wesnoth currentl build with boost 1.48 but im not sure.
20160813 21:56:12< celmin> I see, so it could be worth listing them then.
20160813 21:56:18< celmin> gfgtdf: Didn't you just say 1.56?
20160813 21:56:28< gfgtdf> celmin: there are also some BOOST_VERSION < vesion checks that coudl be rmeoved
20160813 21:56:35< celmin> Oh fun...
20160813 21:57:06< gfgtdf> celmin: 1.56 for current master, 1.48 was for 1.13.5
20160813 21:57:38< gfgtdf> celmin: i dont think we s hdoul incrrease depndency just for that, but sicne it dont realyl know how liby wokr on lnux/ma i cannot tell for sure
20160813 21:57:49< shadowm> When was this discussed in the mailing list? Bumping the mnimum requirement to 1.56, I mean.
20160813 21:57:57< gfgtdf> celmin: for windows support i see no problems with raising boost version requirements
20160813 21:58:37< celmin> Ugh.
20160813 21:59:31-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev
20160813 22:00:28< tad_> Before I waste time on writing a solution, is there a way to get gender-possessive forms like 'his' and 'hers'?
20160813 22:00:43< celmin> I feel like I need a lot more context.
20160813 22:02:59< Espreon> Uh...
20160813 22:03:00< tad_> As "$unit.name| grabs the burning stick and it burns $unit.possessive| hand." .. Konrad/his .. Li'sar/her
20160813 22:03:17< celmin> Aha.
20160813 22:04:01< celmin> Well, you could duplicate the line and use female_text (assuming this is a [message]).
20160813 22:04:17< celmin> The only difference of course being that one version uses his and the other uses hers.
20160813 22:04:19< tad_> That would give me 
20160813 22:04:43< Espreon> Wait, there's "female_text"? Since when?
20160813 22:04:54< celmin> Since I added it.
20160813 22:05:00< Espreon> When?
20160813 22:05:11< celmin> I don't remember exactly, a few months ago?
20160813 22:05:11< tad_> Ah. OK. Yes, it's for [message]. Forgot about gender-text there. Thanks.
20160813 22:05:16< Espreon> Hmm, cool.
20160813 22:05:18-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Quit: Page closed]
20160813 22:07:00< Espreon> This will only do for languages that do things on a basis of sex-animacy, but whatever. Actually implementing something more proper would be a monumental task.
20160813 22:07:30< celmin> True..
20160813 22:09:25< Espreon> Why was it that they eliminated the female Lich strings?
20160813 22:09:41< celmin> Huh?
20160813 22:09:58< Espreon> There used to be "female^Lich" and such
20160813 22:10:09< Espreon> (Or maybe just "female^Lich")
20160813 22:10:12< celmin> Why would the eliminate that?
20160813 22:10:16< celmin> ^they
20160813 22:10:17< Espreon> I have no clue.
20160813 22:10:46< Espreon> Needless to say, whoever did so did not understand the implications.
20160813 22:11:32< celmin> A dark sorcerer levelling up to a lich should not suddenly start using male strings.
20160813 22:11:45< celmin> ^sorceress
20160813 22:12:14< Espreon> Exactly.
20160813 22:13:42< fabi> hi Espreon
20160813 22:13:51< Espreon> Hello, fabi
20160813 22:14:18< Espreon> They just better not hope they don't find Tocharian A-speaking ghosts wanting to play Wesnoth. After all, that language has gender distinction in the first person.
20160813 22:14:24< Espreon> Dunno if any other language does.
20160813 22:14:45< celmin> Japanese?
20160813 22:14:53< Espreon> Ah yes.
20160813 22:15:13< Espreon> I was thinking "real pronouns" rather than noun-like pronominals.
20160813 22:15:20< celmin> Fair enough.
20160813 22:15:28< Espreon> But regardless, silly me.
20160813 22:20:33< Espreon> (It goes far beyond pronominals and inflexions, really, but... well... whatever)
20160813 22:20:53< zookeeper> to answer why "they" eliminated that, one would need to first know who "they" are
20160813 22:21:07< celmin> Indeed.
20160813 22:21:13< Espreon> For some reason, I want to say "mordante".
20160813 22:21:15< Espreon> Dunno why.
20160813 22:21:20< Espreon> I'll dig it up...
20160813 22:21:37< zookeeper> sounds like a very a weird guess
20160813 22:21:38< zookeeper> -a
20160813 22:21:50< Espreon> I know...
20160813 22:21:57< Espreon> It really is weird...
20160813 22:28:45< Espreon> OK, shadowm added it only to remove it right after: https://github.com/wesnoth/wesnoth/commit/3623ba4ceded360f9458676a99ae32942d29a1a8#diff-4bf438459c52513d50c02ed1f67c4cfb
20160813 22:29:16< Espreon> Dunno if it existed before.
20160813 22:29:25< Espreon> My guess is "no"
20160813 22:29:34< Espreon> Yeah, doesn't look like it
20160813 22:29:38< celmin> I'd say it should be put back.
20160813 22:30:02< celmin> Though, I'm a bit curious about the referenced IRC discussion...
20160813 22:30:32< Espreon> And I can't find logs that go back to 2007
20160813 22:30:43< Espreon> ... nor do I remember where any sort of archives would go
20160813 22:30:55< zookeeper> i have the logs if you want to read them
20160813 22:31:01< Espreon> Yes, please.
20160813 22:31:17< zookeeper> ok, give me a while
20160813 22:34:39< Espreon> OK
20160813 22:36:04< zookeeper> http://pastebin.com/VeUZZjEX
20160813 22:36:15< zookeeper> i could dig up the discussion of the original addition too, if one exists
20160813 22:36:50< zookeeper> oh wait, it's from the same day
20160813 22:36:51< celmin> Who the heck is Noyga?
20160813 22:37:09< Espreon> The French translation maintainer of the time
20160813 22:37:15< Espreon> zookeeper: Please do
20160813 22:37:17< zookeeper> celmin, someone who everyone who isn't a newbie remembers :>
20160813 22:38:55< zookeeper> nope, i don't have logs from that
20160813 22:40:06< celmin> I think I basically understand the argument.
20160813 22:40:30< celmin> Trait names should agree not with logical gender but with grammatical gender, which causes problems in languages where they don't align.
20160813 22:40:44-!- Appleman1234 [~Appleman1@KD036012012154.au-net.ne.jp] has joined #wesnoth-dev
20160813 22:41:07< Espreon> What is the current result in the French translation for Liches anyway, though?
20160813 22:41:40< Espreon> Do traits assigned to liches use the male strings or the female strings?
20160813 22:42:00< celmin> The only traits normally assigned to liches would be "undead", I think.
20160813 22:42:30< celmin> Oh, wait, I guess they would also keep traits that they had before levelling up.
20160813 22:44:23< Espreon> I don't think any language that does this by grammatical gender is going to get the right result, except by chance.
20160813 22:44:50< Espreon> I suspect the male strings get applied to all liches, and if that's the case, we are only creating a gap for languages that do this by sex-animacy.
20160813 22:45:53< Espreon> Besides, a lich is not necessarily to be treated inanimate, for it still has a persona. Although a lich may be male by grammar, it can still be a she by persona.
20160813 22:46:11< Espreon> Like how girls are neuter by grammar but "sie" by persona in Standard German.
20160813 22:47:48< Espreon> I guess you could refer to a GIRL (by virtue of the word) "es" 'it' rather than "sie" 'she', but... I don't think this would viewed very positively.
20160813 22:48:38< fabi> Espreon: In fact the local language in the Saarland allows to exchange "sie" with "es".
20160813 22:48:58< Espreon> In all cases, or just with this one term?
20160813 22:48:58< fabi> Espreon: And yes, people from elsewhere consider this rude.
20160813 22:49:16< fabi> all cases
20160813 22:51:16< Espreon> Well, I meant: just for words that refer to girls, or for all words of neuter gender?
20160813 22:51:22< Espreon> I suspect the former.
20160813 22:51:37< fabi> the former
20160813 22:51:42< Espreon> Gotcha.
20160813 22:51:46< fabi> :-)
20160813 22:52:01< shadowm> Espreon: IIRC people chastised me for introducing the change in the first place.
20160813 22:52:15< shadowm> Back then I didn't really know what I was doing or why I could potentially be right in doing it.
20160813 22:53:19< Espreon> Hmmm, I see.
20160813 22:53:46< shadowm> I don't know why your first thought is to attribute things to everyone's favorite punching bag when English prose and WML have never been his domain.
20160813 22:54:29< Espreon> Probably because I remembered him committing changes to that file and forgot what he actually did.
20160813 22:54:48< Espreon> It was not at all done out of mere spite or whatever.
20160813 22:55:07< Espreon> But I guess the fact that he is everyone's punching bag could have had some influence on my decision to name him.
20160813 22:55:12< Espreon> Oh well. Poor mordante.
20160813 22:57:09< shadowm> I think it's more like "stupid post-2006 Wesnoth devs".
20160813 22:57:57< shadowm> Or post-2007, actually. I'm trying to exclude mordante from that group.
20160813 22:58:49< shadowm> The point being is that time and energy tend to be more visibly spent blaming people for things instead of fixing them.
20160813 22:59:16< shadowm> s/is//
20160813 23:02:05< Espreon> I really don't care who's responsible. I just care about why it was done and if the decision really has merit.
20160813 23:02:18< Espreon> I myself, at least...
20160813 23:06:06 * zookeeper doesn't understand the argument for removing the female lich
20160813 23:06:34< celmin> I understand but don't really agree.
20160813 23:07:24< Espreon> The argument is that in X language, a lich will be Y gender by virtue of grammar regardless of persona and that if we were to keep female liches, it would result in gender agreement errors.
20160813 23:07:34< Espreon> I don't see how this can be true for French, though.
20160813 23:08:02< Espreon> He claimed that LICH is feminine by virtue of grammar in French.
20160813 23:08:16< Espreon> ... and that the male string for undead is good for both masculine and feminine nouns.
20160813 23:08:31< Espreon> ... so even if we were to introduce female liches, it would still be good for French.
20160813 23:09:04< Espreon> On the other hand, if traits other than "undead" get applied to a lich and if they have differing masculine and feminine forms, it would introduce a gender agreement error in the French translation.
20160813 23:09:27< Espreon> ... since the male strings would get applied to this unit that is by virtue of grammar feminine.
20160813 23:09:51< celmin> Strong, intelligent, resilient, and quick would probably be commonly found on liches.
20160813 23:09:58< shadowm> AFAICT this is the responsibility of the translators.
20160813 23:10:05< celmin> Actually, strong would be quite a bit rarer though.
20160813 23:10:08< Espreon> Right.
20160813 23:10:16< celmin> Because most liches start as a dark adept, which is never strong.
20160813 23:10:26< shadowm> We just need to provide the venue for such a responsibility to exist in the first place.
20160813 23:10:30< Espreon> If any successful gender agreement is done in the game outside of the sex-animacy realm, it is done by chance and only chance.
20160813 23:11:37< Espreon>  http://pastebin.com/VeUZZjEX
20160813 23:11:51< Espreon> Whoops.
20160813 23:12:59< Espreon> Since Wesnoth's treatment of gender is only suitable for sex-animacy, removing it restored the gap for languages that do it by sex-animacy.
20160813 23:13:30< Espreon> Perhaps he was thinking that if we introduced what are technically female liches, more and more female units would start appearing.
20160813 23:13:44< Espreon> ... making the situation for languages that have grammatical gender worse.
20160813 23:15:10< Espreon> 01:09 < shadowm> AFAICT this is the responsibility of the translators.
20160813 23:15:15< Espreon> What should the translators do, though?
20160813 23:15:35< shadowm> Decide if and how they want to translate female^Lich.
20160813 23:15:46< shadowm> They are not babies. They can figure this stuff out.
20160813 23:15:57< celmin> They're free to translate female^Lich the same as Lich.
20160813 23:16:10< shadowm> For some languages (e.g. English) that's a no-brainer, yes.
20160813 23:16:29< Espreon> 01:15 < celmin> They're free to translate female^Lich the same as Lich.
20160813 23:16:42< shadowm> But right now they have no choice because there's no such a string.
20160813 23:16:42< Espreon> This won't guarentee the avoidance of an agreement error.
20160813 23:18:43< shadowm> I doubt the Lich will be the first or last unit in that regard.
20160813 23:19:04< Espreon> Well, anyway. Since successful agreement for grammatical gender primarily happens by chance and this creates a gap for languages that do it by sex-animacy, I say it should go back in.
20160813 23:19:08< Espreon> Oh no, definitely not.
20160813 23:21:50< Espreon> If it disturbs anyone enough, they can simply translate it in such a way that fits with the current system. Yes, you're right/
20160813 23:46:27< irker978> wesnoth: Celtic Minstrel wesnoth:master 16786f990587 / src/help/help_impl.cpp: Generate help topics for weapon specials obtained through AMLAs https://github.com/wesnoth/wesnoth/commit/16786f99058716c6a34c851e47f2854382ec1810
20160813 23:46:29< irker978> wesnoth: Celtic Minstrel wesnoth:master bbb547dd3925 / src/tests/gui/test_gui2.cpp: Fix unit tests compilation https://github.com/wesnoth/wesnoth/commit/bbb547dd39253728dbfec584184195b9fff16979
--- Log closed Sun Aug 14 00:00:56 2016