--- Log opened Fri Apr 21 00:00:53 2017
20170421 00:02:12-!- RatArmy [~ratarmy@om126229088058.12.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170421 00:06:39-!- irker824 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170421 00:08:31-!- RatArmy [~ratarmy@om126229088058.12.openmobile.ne.jp] has joined #wesnoth-dev
20170421 00:12:19< vultraz_iOS> hmmm
20170421 00:15:26< vultraz_iOS> celticminstrel: how would one specialize an overload if one has a variadic template?
20170421 00:15:53< celticminstrel> There's no such thing as "specializing an overload".
20170421 00:16:26< vultraz_iOS> blah terminology 
20170421 00:16:58< vultraz_iOS> right, it's just a specilization 
20170421 00:17:05< vultraz_iOS> specialization 
20170421 00:17:16< celticminstrel> Or an overload.
20170421 00:17:48< celticminstrel> IIRC non-template functions always take priority over template functions during overload resolution.
20170421 00:18:31< vultraz_iOS> celticminstrel: https://pastebin.com/6FTssjzZ
20170421 00:18:37< vultraz_iOS> how do I handle this
20170421 00:19:01< vultraz_iOS> it's sending the contex_ptr to the first... whatever
20170421 00:19:16< vultraz_iOS> specialization/overload/function/unicorn
20170421 00:20:59< vultraz_iOS> C:\Users\exodi\Documents\wesnoth\src\editor\map\context_manager.cpp|934|error: no matching function for call to 'editor::map_context::map_context(const std::unique_ptr<editor::map_context>&)'|
20170421 00:21:02< vultraz_iOS> of course
20170421 00:21:10< vultraz_iOS> because it's supposed to use the other one :/
20170421 00:24:45< celticminstrel> Don't like the name context_ptr, but...
20170421 00:25:27< celticminstrel> std::forward(args)... is apparently the proper way to do that.
20170421 00:25:38< celticminstrel> (I also messed that up somewhere, I think.)
20170421 00:25:42< vultraz_iOS> where?
20170421 00:25:50< celticminstrel> Line 4.
20170421 00:25:54< celticminstrel> Anyway...
20170421 00:26:18< celticminstrel> Why is it an r-value reference? Maybe it'll work if it's an l-value reference or not a reference?
20170421 00:26:36< vultraz_iOS> what?
20170421 00:27:30< celticminstrel> Why is it context_ptr&& rather than context_ptr& or context_ptr?
20170421 00:27:48< celticminstrel> Or const context_ptr& even.
20170421 00:28:05< vultraz_iOS> because it's a unique_ptr
20170421 00:28:46< vultraz_iOS> though i see a deleted function error further down...
20170421 00:28:55< vultraz_iOS> looks like unique_ptr is not MoveInsertable?
20170421 00:29:05< vultraz_iOS> but it is MoveConstructible and MoveAssignable?
20170421 00:29:52< vultraz_iOS> why..
20170421 00:30:39< vultraz_iOS> ah wait
20170421 00:30:40< vultraz_iOS> std::move
20170421 00:31:49< celticminstrel> If you want move-construction, use emplace, not push.
20170421 00:31:57< celticminstrel> emplace_back in this case.
20170421 00:32:08< celticminstrel> Though I question why you're adding the context to a list in the first place...
20170421 00:32:32< vultraz_iOS> oh right, emplace has push semantics 
20170421 00:32:35< vultraz_iOS> er
20170421 00:32:36< vultraz_iOS> move
20170421 00:32:52< celticminstrel> Well no, not automatically.
20170421 00:32:58< celticminstrel> It can move or copy.
20170421 00:33:07< celticminstrel> But it's in-place construction.
20170421 00:33:38< vultraz_iOS> I'm not constructing hee, just ftr
20170421 00:35:26< vultraz_iOS> still hasn't helped my original problem at all
20170421 00:38:14-!- Greg-Boggs [~greg_bogg@c-73-37-6-51.hsd1.or.comcast.net] has quit [Remote host closed the connection]
20170421 00:43:51< vultraz_iOS> blagh..
20170421 00:45:09-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170421 00:49:27-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20170421 00:53:14< vultraz_iOS> i guess this just isn't possible
20170421 00:55:09< vultraz_iOS> since variadic templates eat All The Things
20170421 00:55:10-!- RatArmy [~ratarmy@om126229088058.12.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170421 01:00:16< vultraz_iOS> heh
20170421 01:00:21< vultraz_iOS> simple solution: don't overload :P
20170421 01:00:25< vultraz_iOS> now I'm here..
20170421 01:00:35< vultraz_iOS> error: cannot bind 'editor::context_manager::context_ptr {aka std::unique_ptr<editor::map_context>}' lvalue to 'editor::context_manager::context_ptr&& {aka std::unique_ptr<editor::map_context>&&}'|
20170421 01:01:34< vultraz_iOS> celticminstrel: what does this mean?
20170421 01:01:44-!- gfgtdf_ [~chatzilla@x4e369253.dyn.telefonica.de] has joined #wesnoth-dev
20170421 01:04:40-!- gfgtdf [~chatzilla@x4e32b5b9.dyn.telefonica.de] has quit [Ping timeout: 240 seconds]
20170421 01:04:53-!- gfgtdf_ is now known as gfgtdf
20170421 01:05:30-!- RatArmy [~ratarmy@om126229088058.12.openmobile.ne.jp] has joined #wesnoth-dev
20170421 01:10:42< vultraz_iOS> ah, i need to use std::forward in the call
20170421 01:15:14< vultraz_iOS> celticminstrel: please review https://github.com/Vultraz/wesnoth/commit/25a4718554ba3481888105316330e03f4a005a60
20170421 01:22:07-!- RatArmy [~ratarmy@om126229088058.12.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170421 01:36:26< celticminstrel> vultraz_iOS: Pretty sure line 851/854 should be move, not forward.
20170421 01:36:46< celticminstrel> And I don't think you need the template arguments? move(mc) / forward(mc) should be sufficient.
20170421 01:37:48< celticminstrel> I think I'd also recommend enclosing lines 846-855 in a scope to ensure that mc isn't used after it's destroyed.
20170421 01:37:53< vultraz_iOS> and the other forward?
20170421 01:38:37< celticminstrel> The other forward is on the wrong line, it should be on the line before (947 rather than 948).
20170421 01:38:48< celticminstrel> std::forward(args)...
20170421 01:39:32< vultraz_iOS> doesn't work there
20170421 01:39:43< celticminstrel> There's no point in using it if it's not there.
20170421 01:40:06< celticminstrel> context_ptr new_mc(new map_context(std::forward(args)...)); should work
20170421 01:40:21< vultraz_iOS> alright, ill use move
20170421 01:40:26< celticminstrel> ???
20170421 01:40:27< vultraz_iOS> it does not
20170421 01:40:38< vultraz_iOS> celticminstrel: i'll replace all forwards with move
20170421 01:40:42< celticminstrel> Oh, try T... instead of T&...?
20170421 01:40:54< celticminstrel> No, that one should be forward though, not move.
20170421 01:41:01< vultraz_iOS> why?
20170421 01:41:12< celticminstrel> Because that's the way to forward arguments to the constructor.
20170421 01:41:27< vultraz_iOS> I'm so confused
20170421 01:41:36< vultraz_iOS> it works fine as-is
20170421 01:41:36< celticminstrel> But I think you should also use "T..." instead of "const T&..." for the arguments type?
20170421 01:42:04< celticminstrel> Well, sure, just "args..." will work in 99% of cases, and is probably "good enough" here since this isn't a truly fully-generic interface.
20170421 01:42:13-!- gfgtdf [~chatzilla@x4e369253.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.93 [Firefox 52.0.2/20170323105023]]
20170421 01:42:35< celticminstrel> I think the "correct" way though is...
20170421 01:42:55< vultraz_iOS> and when i said i'd make the third forward move I meant if I leave it at line 948
20170421 01:43:20< celticminstrel> template<typename... T> void context_manager::replace_map_context(T... args){
20170421 01:43:21< celticminstrel> 	context_ptr new_mc(new map_context(std::forward(args)...));
20170421 01:43:22< celticminstrel> 	replace_map_context_with(std::move(new_mc));
20170421 01:43:23< celticminstrel> }
20170421 01:43:54< vultraz_iOS> would rather not take the args by copy
20170421 01:43:59< celticminstrel> You won't.
20170421 01:44:05< celticminstrel> I think.
20170421 01:44:16< celticminstrel> (What are the actual args?)
20170421 01:44:18< vultraz_iOS> I'm still confused as to why forward is needed
20170421 01:44:22< vultraz_iOS> you have't made it clear
20170421 01:45:12< vultraz_iOS> celticminstrel: https://github.com/wesnoth/wesnoth/blob/master/src/editor/map/map_context.hpp#L67-L77
20170421 01:45:34< vultraz_iOS> (only case we might not want a ref is for the bool.. but oh well)
20170421 01:46:57< vultraz_iOS> (i don't even understand what std::forward does)
20170421 01:47:03< celticminstrel> http://eli.thegreenplace.net/2014/perfect-forwarding-and-universal-references-in-c/
20170421 01:47:19< celticminstrel> I think that means the "correct" way is in fact "T&&... args"
20170421 01:47:46< celticminstrel> And then use std::forward like I said.
20170421 01:48:13< celticminstrel> It's probably "good enough" the way you did it, mind you. You probably don't need absolutely perfect forwarding here because there's only one or two possible target constructors.
20170421 01:48:31< vultraz_iOS> two
20170421 01:48:55< celticminstrel> Three if you count the implicit move constructor.
20170421 01:49:35< celticminstrel> So, maybe if you use "T&&... args" the forward would work like I said?
20170421 01:49:47< celticminstrel> But, if not, then it's not really a big deal in this case.
20170421 01:49:56< celticminstrel> You could just forget about the forward.
20170421 01:51:13< vultraz_iOS> i'll do that
20170421 01:51:19< vultraz_iOS> still need the move, though
20170421 01:51:35< celticminstrel> Well, I did include both forward and move in what I pasted here, so...
20170421 01:52:16< vultraz_iOS> celticminstrel: https://github.com/Vultraz/wesnoth/commit/2f16ef619194ab7dff92e22d5b34e8445807d9a8
20170421 01:52:49< celticminstrel> So you're not bothering with the forward?
20170421 01:53:03< celticminstrel> Anyway, that looks fine, I guess.
20170421 01:53:03< vultraz_iOS> nah
20170421 01:54:24< vultraz_iOS> well, ill try and see if && works
20170421 01:54:33< vultraz_iOS> no
20170421 01:54:45< celticminstrel> Obviously you'd need to drop const for && to work.
20170421 01:54:57< celticminstrel> Or, at least, I don't think there's any point in ever having a const &&?
20170421 01:56:56< vultraz_iOS> yeah, doesn't work like that either
20170421 01:57:03< vultraz_iOS> tries to call a ctor with only the first argument
20170421 01:57:35-!- irker370 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20170421 01:57:35< irker370> wesnoth: Charles Dang wesnoth:master bf79dbd8ee8f / src/editor/map/ (context_manager.cpp context_manager.hpp): Editor: refactor handling of map_contexts https://github.com/wesnoth/wesnoth/commit/bf79dbd8ee8fb5ee592a3d3c268c3280f62b7f81
20170421 02:00:34< celticminstrel> BTW, is Travis passing recently?
20170421 02:01:34< vultraz_iOS> yes
20170421 02:02:32< vultraz_iOS> celticminstrel: last build all passed, except the mac one which timed out
20170421 02:02:42< celticminstrel> Yay,
20170421 02:02:49< vultraz_iOS> so you can rebase the crypt
20170421 02:03:22< celticminstrel> There were no further comments on it, I see.
20170421 02:03:37< celticminstrel> Pondering if/how I can split it into two "phases" like tad suggested.
20170421 02:06:36-!- un214 [~un214@76-14-112-211.rk.wavecable.com] has joined #wesnoth-dev
20170421 02:26:13-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20170421 02:30:55< mattsc> vultraz_iOS: thanks for fixing the Xcode bracket problem in display.cpp.  That fixes it for that file.  There’s one more instance of the same thing at src/units/unit.cpp:83:55
20170421 02:32:41< irker370> wesnoth: Charles Dang wesnoth:master 7f37a0856f53 / src/units/unit.cpp: Fixup XCode build again (cont 3252348) https://github.com/wesnoth/wesnoth/commit/7f37a0856f53a5a271cb2aaf87761eeb2e16d5ec
20170421 02:32:43< vultraz_iOS> mattsc: ^
20170421 02:32:45< mattsc> I can commit that if you want, but I’ll leave it for you if you prefer.
20170421 02:32:52< mattsc> Ooo, too late.  :D
20170421 02:37:53< irker370> wesnoth: mattsc wesnoth:master ea9c4b849fd7 / projectfiles/Xcode/Wesnoth.xcodeproj/project.pbxproj: Update Xcode project https://github.com/wesnoth/wesnoth/commit/ea9c4b849fd70a9302f5726082e77151536f0666
20170421 02:38:12< mattsc> vultraz_iOS: ^ all good now as far as picky Xcode is concerned
20170421 02:38:41< vultraz_iOS> :D
20170421 02:47:00-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20170421 02:50:52< mattsc> Btw, for those of you who have not seen this, here is why we cannot get rid of irc:  https://xkcd.com/1782/
20170421 02:54:06< vultraz_iOS> hipsters :P
20170421 02:56:02< irker370> wesnoth: Charles Dang wesnoth:master 90de2269f06f / src/editor/ (3 files in 2 dirs): Editor: context_manager code cleanup https://github.com/wesnoth/wesnoth/commit/90de2269f06fb4f38d8284119ce87b2e88ae228c
20170421 03:01:05-!- RatArmy [~ratarmy@om126212252255.14.openmobile.ne.jp] has joined #wesnoth-dev
20170421 03:03:58-!- un214 [~un214@76-14-112-211.rk.wavecable.com] has quit [Read error: Connection reset by peer]
20170421 03:15:27< celticminstrel> I guess you deleted and re-added the renamed files, rather than renaming them in-place.
20170421 03:15:36< celticminstrel> I guess that's fine though, just not the way I would've done it.
20170421 03:16:24< vultraz_iOS> what?
20170421 03:16:36< celticminstrel> The XCode project update.
20170421 03:19:32< mattsc> Right.
20170421 03:29:04-!- RatArmy_ [~ratarmy@om126212252255.14.openmobile.ne.jp] has joined #wesnoth-dev
20170421 03:29:05-!- RatArmy [~ratarmy@om126212252255.14.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170421 03:36:53< vultraz_iOS> what the hell is wrong with my compiler
20170421 03:37:07< vultraz_iOS> i'm modifying a cpp and it's building All The Things
20170421 03:46:35-!- RatArmy_ [~ratarmy@om126212252255.14.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170421 04:15:30< vultraz_iOS> this is so weird
20170421 04:15:39< vultraz_iOS> if i modify map/label.cpp it rebuilds everything
20170421 04:16:14< celticminstrel> I hope nothing is including it by mistake...
20170421 04:16:47< vultraz_iOS> not that i can see
20170421 04:19:02-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Quit: KABOOM! It seems that I have exploded. Please wait while I reinstall the universe.]
20170421 04:19:26-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20170421 04:22:27-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170421 04:26:59-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 255 seconds]
20170421 04:30:35< irker370> wesnoth: Charles Dang wesnoth:master 86d00a479dc0 / src/ (6 files in 3 dirs): Refactored display member out of map_labels in favor of the singleton accessor https://github.com/wesnoth/wesnoth/commit/86d00a479dc03047e0ca94a46be6588e598f1a78
20170421 05:21:55-!- mjs-de [~mjs-de@x4db6e2df.dyn.telefonica.de] has joined #wesnoth-dev
20170421 05:22:48-!- JyrkiVesterinen [~JyrkiVest@87-100-254-108.bb.dnainternet.fi] has joined #wesnoth-dev
20170421 05:33:07-!- mjs-de [~mjs-de@x4db6e2df.dyn.telefonica.de] has quit [Remote host closed the connection]
20170421 05:35:57-!- JyrkiVesterinen [~JyrkiVest@87-100-254-108.bb.dnainternet.fi] has quit [Quit: .]
20170421 05:36:02-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20170421 05:38:12-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 260 seconds]
20170421 05:38:12-!- wedge010 is now known as wedge009
20170421 05:42:39-!- RatArmy_ [~ratarmy@om126229092116.12.openmobile.ne.jp] has joined #wesnoth-dev
20170421 05:51:01-!- Kwandulin [~Kwandulin@p200300760F6D80B36877D6CE172B1F72.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170421 05:58:46-!- RatArmy_ [~ratarmy@om126229092116.12.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170421 06:10:44-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170421 06:15:22-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 268 seconds]
20170421 06:45:27-!- JyrkiVesterinen [~JyrkiVest@85-76-70-126-nat.elisa-mobile.fi] has joined #wesnoth-dev
20170421 07:03:21-!- zookeeper [zookeeper@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20170421 07:05:12-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170421 07:10:51-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has joined #wesnoth-dev
20170421 07:29:28< celticminstrel> I feel there's something rather fundamentally wrong with the [objectives] implementation.
20170421 07:29:47< zookeeper> i was just about to ask you about bob's variable update issue
20170421 07:30:07< zookeeper> (i confirmed he's right)
20170421 07:30:13< celticminstrel> Looking at the implementation, it's obvious that there's no way variables would automagically update like he says they should.
20170421 07:30:42< celticminstrel> Nor would [show_if] sections be re-evaluated, I think.
20170421 07:30:48< zookeeper> O.o
20170421 07:30:57< celticminstrel> Though they would be if you used [show_objectives] if I'm reading this right.
20170421 07:31:06< vultraz_iOS> other way around
20170421 07:31:15< vultraz_iOS> not updated in that case
20170421 07:31:28< celticminstrel> [show_if], but not variables, because the very first line of the [objectives] implementation parses the entire config.
20170421 07:31:33< celticminstrel> What, vultraz_iOS?
20170421 07:32:06< vultraz_iOS> or perhaps it is so..
20170421 07:32:09< vultraz_iOS> I can't remember
20170421 07:32:17< celticminstrel> See, [objectives] stores the parsed config away somewhere, and [show_objectives] regenerates from that.
20170421 07:32:35< celticminstrel> But pressimg Cmd+J or whatever to view objectives won't invoke [show_objectives], right?
20170421 07:32:42< celticminstrel> So how are they regenerated in that case?
20170421 07:32:58< celticminstrel> Plus, the config has already been parsed, so the variables are already substituted.
20170421 07:33:14< celticminstrel> So regenerating would re-evaluated [show_if] but not variables.
20170421 07:33:29< celticminstrel> More disturbingly, why are team objectives just a raw string?
20170421 07:33:51< celticminstrel> I suppose it could work that way though if it was a raw string with the variables unsubstituted...
20170421 07:34:12< celticminstrel> Maybe simply removing the line that parses the config would be enough? Oh and making sure the C++ code substitutes variables when showing objectives.
20170421 07:34:19< vultraz_iOS> looks like [show_objectives] indeed refreshes the conditionals
20170421 07:34:21< vultraz_iOS> "refreshes the objectives defined by [objectives] and its [show_if] tags, and displays them. (It is also called whenever the user explicitly asks for the objectives; this matters only if the tag was overridden by a Lua script."
20170421 07:34:45< celticminstrel> Uhhh. How the heck is it called when the user explicitly asks for the objectives?
20170421 07:35:02< celticminstrel> Literally the only WML tag ever called from C++ AFAIK is [command].
20170421 07:35:12< celticminstrel> Unless I've missed something.
20170421 07:35:41< vultraz_iOS> don't ask me
20170421 07:36:07< celticminstrel> So if you add a line "wesnoth.message('Hello world!')" to the [show_objectives] implementation, do you see it in chat when you press Cmd+J or whatever?
20170421 07:36:15< vultraz_iOS> let me chec some code..
20170421 07:36:59< vultraz_iOS> ok yes, variables are only set when [objectives] is used
20170421 07:37:08< vultraz_iOS> https://github.com/Vultraz/Shadows_of_Deception/blob/master/episode1/scenarios/02_Dwarvish_Trail.cfg#L369-L398
20170421 07:37:12< vultraz_iOS> i remember because I did this
20170421 07:38:27< celticminstrel> So is it really true that [show_objectives] is called when the user manually requests the objectives?
20170421 07:38:52 * celticminstrel blink gasp.
20170421 07:38:54< celticminstrel> Whoa!
20170421 07:39:11< celticminstrel> Line 128 of menu_events.cpp! o.O
20170421 07:39:22< celticminstrel> How did I miss that the other time when I was searching run_wml_action...
20170421 07:40:25< celticminstrel> Mind you, [show_objectives] is also misleading. It doesn't actually show the objectives. There's no way from Lua to show the objectives.
20170421 07:40:26-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170421 07:40:45< celticminstrel> Al it can do is mark the objectives as "changed" so that they come up naturally at the beginning of your turn.
20170421 07:40:48< celticminstrel> ^All
20170421 07:41:30< vultraz_iOS> anyway, either we change this or we point Bob at the established solution (what I linked you)
20170421 07:41:56< celticminstrel> Ask zookeeper, but I suspect this is a real bug.
20170421 07:42:03< vultraz_iOS> speak, zookeeper 
20170421 07:42:30< celticminstrel> I think the fix would be twofold.
20170421 07:43:00< zookeeper> [show_if]s and variables should be re-evaluated whenever the objectives dialog is shown, i think.
20170421 07:43:04< celticminstrel> 1. Delete line 181 of data/lua/wml/objectives.lua
20170421 07:43:21< celticminstrel> 2. In play_controller::show_objectives, substitute variables in.
20170421 07:43:46< celticminstrel> Hopefully variables are available from the play_controller. >_>
20170421 07:43:55< zookeeper> arguably someone might want to just use variables to set objectives once and not have substitution happen later, but that's far easier to accomplish with a workaround than the inverse.
20170421 07:44:05-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has joined #wesnoth-dev
20170421 07:44:16-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has quit [Remote host closed the connection]
20170421 07:44:17 * vultraz_iOS is attempting to transition menu items to the new syntax...
20170421 07:44:30-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has joined #wesnoth-dev
20170421 07:44:43< zookeeper> in any case, [show_if] is a clear-cut case as it was intended to work so that it always gets re-evaluated whenever the dialog is shown
20170421 07:44:45< celticminstrel> You could even add support for "delayed_variable_substitution=false" in [objectives]
20170421 07:44:56< zookeeper> celticminstrel, yeah, that'd be ideal i suppose
20170421 07:45:15< celticminstrel> vultraz_iOS: Good luck with that. Is my attempt at it still on the repo?
20170421 07:45:17< zookeeper> hmh
20170421 07:45:22< vultraz_iOS> is it?
20170421 07:45:40< celticminstrel> zookeeper: Since [show_objectives] is in fact automatically called before showing them, it seems like [show_if] will work as intended already.
20170421 07:46:17< celticminstrel> vultraz_iOS: Local branch name is menus_refactor...
20170421 07:46:39< celticminstrel> Looks like it's not, huh.
20170421 07:46:50< zookeeper> alternative: instead of adding [objectives]-wide delayed_variable_substitution=, just keep current behavior but allow $|var to be used for delayed substitution.
20170421 07:46:51< celticminstrel> IIRC it broke all the menus.
20170421 07:46:59< vultraz_iOS> no, you threw it out
20170421 07:47:06< celticminstrel> No I didn't?
20170421 07:47:12< celticminstrel> It's still there on my computer.
20170421 07:47:24< vultraz_iOS> it was in a PR and you closed it
20170421 07:47:28< celticminstrel> Was it?
20170421 07:47:40< vultraz_iOS> yes
20170421 07:47:55< celticminstrel> zookeeper: I guess that would work with just the play_controller change.
20170421 07:48:46< celticminstrel> IIRC variables are in the game_state, right?
20170421 07:49:25< celticminstrel> Ah, nope, game_data.
20170421 07:50:43< vultraz_iOS> might as well be in the same place as donald trump's tax returns
20170421 07:51:00< celticminstrel> Looks like they're accessible from the play_controller.
20170421 07:55:04< vultraz_iOS> good god this menu code has so many arms
20170421 07:57:24< celticminstrel> Hm, now that I think of it, the variable substitution could just as easily be done in [show_objectives] as in play_controller::show_objectives... wonder which would be better...
20170421 07:57:24-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170421 07:57:54< vultraz_iOS> is it "label" or "name"...
20170421 07:58:11< vultraz_iOS> label
20170421 07:58:57< irker370> wesnoth: Celtic Minstrel wesnoth:objectives_with_variables e9895f8a102f / src/play_controller.cpp: Objectives: Always interpolate variables before showing https://github.com/wesnoth/wesnoth/commit/e9895f8a102ff1c12cb28e49cb4873f4a3296c5e
20170421 08:00:09< matthiaskrgr> https://pastebin.com/C77wPRz6  
20170421 08:00:30< vultraz_iOS> :|
20170421 08:00:32< vultraz_iOS>  :| :| :| :| :|
20170421 08:01:06< matthiaskrgr> |: 
20170421 08:01:13< celticminstrel> Is that the case where I put an unnecessary fallthrough even though it was at the end of the switch block?
20170421 08:01:21< celticminstrel> If that's the case, just remove the fallthrough.
20170421 08:01:30< celticminstrel> Looking at the filename, that's probably what that is.
20170421 08:03:05< irker370> wesnoth: Celtic Minstrel wesnoth:objectives_with_variables 96f7018030d9 / data/lua/wml/objectives.lua: Objectives: Support delayed_variable_substitution (default true) https://github.com/wesnoth/wesnoth/commit/96f7018030d9471522abdd976fa413b5921709e2
20170421 08:04:00< irker370> wesnoth: Celtic Minstrel wesnoth:objectives_with_variables 699853dfc3a1 / data/lua/wml/objectives.lua: Objectives: Always interpolate variables before showing (alternative to e9895f8a https://github.com/wesnoth/wesnoth/commit/699853dfc3a18a4092e8d1d53178b8c66610aa65
20170421 08:04:44< celticminstrel> Okay, there you go, that might fix the objectives issue. Just need to decide between those two commits (and potentially do fixups).
20170421 08:05:37< zookeeper> celticminstrel, if $| works then there shouldn't be any need for delayed_variable_substitution, right?
20170421 08:05:57< celticminstrel> You're right that the middle commit is technically unnecessary.
20170421 08:06:38< celticminstrel> Well, not quite. Deleting the line would suffice, the delayed_variable_substitution attribtue is unnecessary.
20170421 08:07:02< celticminstrel> But if you have a lot of variables, using delayed_variable_substitition is neater, I think.
20170421 08:07:56< zookeeper> sure, it doesn't hurt
20170421 08:08:15< vultraz_iOS> we need to carefully document this change
20170421 08:10:30< vultraz_iOS> celticminstrel: is command_executor::get_menu_images supposed to extract image attributes from the strings?
20170421 08:12:30< celticminstrel> https://github.com/wesnoth/wesnoth/pull/992
20170421 08:12:35< celticminstrel> vultraz_iOS: I don't remember...
20170421 08:13:47< celticminstrel> (Obviously, don't merge that PR until the questions have been addressed.)
20170421 08:14:03< celticminstrel> (Which at the very least means reverting one of the three commits.)
20170421 08:19:52-!- JyrkiVesterinen [~JyrkiVest@85-76-70-126-nat.elisa-mobile.fi] has quit [Quit: Rebooting]
20170421 08:21:47-!- JyrkiVesterinen [~JyrkiVest@85-76-70-126-nat.elisa-mobile.fi] has joined #wesnoth-dev
20170421 08:24:56-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has joined #wesnoth-dev
20170421 08:26:00< irker370> wesnoth: Celtic Minstrel wesnoth:lua_mapgen_stuff 140499bc5f68 / data/lua/cave_map_generator.lua: Lua Cave Mapgen: Support comma-separated chamber locations https://github.com/wesnoth/wesnoth/commit/140499bc5f682eabc35abee0278279fb90e13efa
20170421 08:26:02< irker370> wesnoth: Celtic Minstrel wesnoth:lua_mapgen_stuff ced1e66b95be / data/ (4 files in 3 dirs): Lua mapgen: Use helper.rand for chamber locations https://github.com/wesnoth/wesnoth/commit/ced1e66b95be9966847fbfbb3850c30e5a19cebd
20170421 08:26:04< irker370> wesnoth: Celtic Minstrel wesnoth:lua_mapgen_stuff aa37e7b3657f / data/lua/cave_map_generator.lua: Lua Cave Mapgen: Allow alternate clear and village terrains https://github.com/wesnoth/wesnoth/commit/aa37e7b3657fc97be695f3036439eaca3c852fc2
20170421 08:26:06< irker370> wesnoth: Celtic Minstrel wesnoth:lua_mapgen_stuff 04926f96655b / data/lua/cave_map_generator.lua: Lua Cave Mapgen: Allow a chamber's existence to depend on number of players https://github.com/wesnoth/wesnoth/commit/04926f96655b7234c480c092fdcbf0da87574115
20170421 08:26:08< irker370> wesnoth: Celtic Minstrel wesnoth:lua_mapgen_stuff 9b623f9040e3 / data/multiplayer/scenarios/Random_Scenario_Cave.cfg: Add a multiplayer cave random map https://github.com/wesnoth/wesnoth/commit/9b623f9040e3bc0ba0714c50ed84d2c61b6f9864
20170421 08:27:57< celticminstrel> I'm gonna open PRs for that and gui2_help so they don't get totally forgotten.
20170421 08:28:47< vultraz_iOS> Still think we should merge gui2 help so the work is in master
20170421 08:29:04< vultraz_iOS> Not like I haven't committed half-complete dialogs before
20170421 08:31:08< irker370> wesnoth: Celtic Minstrel wesnoth:master 0a6d6f3f6097 / src/gui/widgets/tree_view_node.cpp: Tree View: Allow find_widget to recurse into the nodes https://github.com/wesnoth/wesnoth/commit/0a6d6f3f6097e240c5c46b692181aaa06a394d71
20170421 08:31:10< irker370> wesnoth: Celtic Minstrel wesnoth:master df619399c53a / src/gui/widgets/ (tree_view_node.cpp tree_view_node.hpp): Tree View Node: Add functionality to auto-expand parents when selecting a node https://github.com/wesnoth/wesnoth/commit/df619399c53a2ad3e5845bc7e435aed2ed23af63
20170421 08:33:22< celticminstrel> Why should I follow your bad example? ;)
20170421 08:33:41< celticminstrel> I cherry-picked those two from gui2_help because they might be useful or something. ^
20170421 08:35:02< irker370> wesnoth: Celtic Minstrel wesnoth:gui2_help da22fff14ad4 / / (4 files in 2 dirs): Help Browser: Actually show the correct topics with the correct icons https://github.com/wesnoth/wesnoth/commit/da22fff14ad46bc42ffa3c7db059864ae5114a2d
20170421 08:35:04< irker370> wesnoth: Celtic Minstrel wesnoth:gui2_help 300af0b2e9c4 / src/ (4 files in 2 dirs): Help Viewer: Show topics with generated text https://github.com/wesnoth/wesnoth/commit/300af0b2e9c495e86be39e66c81cf2512c86867b
20170421 08:35:06< irker370> wesnoth: Celtic Minstrel wesnoth:gui2_help 15e49c2adcac / src/help/ (help_impl.cpp help_impl.hpp): Help: That parameter isn't really needed https://github.com/wesnoth/wesnoth/commit/15e49c2adcac0c65197d75a787d45621176f6b39
20170421 08:35:08< irker370> wesnoth: Charles Dang wesnoth:gui2_help 71faf691720c / data/gui/window/help_browser.cfg: Help Browser: improved layout stability and added next/back buttons (non-functio https://github.com/wesnoth/wesnoth/commit/71faf691720c86ed98ef3c6dcd4a17b3ee6c064c
20170421 08:35:11< irker370> wesnoth: Celtic Minstrel wesnoth:gui2_help 6cc95dde98a3 / src/ (5 files in 2 dirs): Help Viewer: Parse help markup to Pango markup... mostly. https://github.com/wesnoth/wesnoth/commit/6cc95dde98a3acf877997af323e6030e2caac348
20170421 08:35:13< irker370> wesnoth: Charles Dang wesnoth:gui2_help f3da8fcfe29d / / (3 files in 2 dirs): Help Browser: implemented viewing of sub-sections/topics https://github.com/wesnoth/wesnoth/commit/f3da8fcfe29d79b3d29a8313272b61605baff1b1
20170421 08:35:16< irker370> wesnoth: Charles Dang wesnoth:gui2_help e9786f1dcc09 / src/ (gui/dialogs/title_screen.cpp help/help.cpp): Moved new help entry point to a common location https://github.com/wesnoth/wesnoth/commit/e9786f1dcc09574fe00795d72d68d1e86f1e5604
20170421 08:35:19< irker370> wesnoth: Celtic Minstrel wesnoth:gui2_help ddd0cb2ce57e / src/gui/dialogs/ (help_browser.cpp help_browser.hpp): Help Browser: Implement history https://github.com/wesnoth/wesnoth/commit/ddd0cb2ce57e2923f112a6cb79858fad030dbbd2
20170421 08:35:21< irker370> wesnoth: Celtic Minstrel wesnoth:gui2_help 7e8287837640 / src/help/help_text_area.cpp: Fix GUI1 help topics not showing https://github.com/wesnoth/wesnoth/commit/7e8287837640b2836224d5d98bc4f98f768deef7
20170421 08:35:23< irker370> wesnoth: Charles Dang wesnoth:gui2_help 23da45feb18e / / (3 files in 3 dirs): Help Browser: properly implement toggling by clicking the book icons https://github.com/wesnoth/wesnoth/commit/23da45feb18e2869ae85a9908faf189b145fa324
20170421 08:35:26< irker370> wesnoth: Celtic Minstrel wesnoth:gui2_help 38b175868085 / src/gui/dialogs/help_browser.cpp: Help Browser: Fix back button being visible when dialog opens https://github.com/wesnoth/wesnoth/commit/38b175868085ee4bcb3c0b747891343405fb4f79
20170421 08:35:28< irker370> wesnoth: Celtic Minstrel wesnoth:gui2_help edaba7a5796f / data/gui/window/help_browser.cfg src/gui/dialogs/help_browser.cpp: Help Browser: Show topic title https://github.com/wesnoth/wesnoth/commit/edaba7a5796f01a4ba3d237c0f01b4278cfcc8ab
20170421 08:35:30< irker370> wesnoth: Celtic Minstrel wesnoth:gui2_help 5d1e0e20a83c / src/gui/dialogs/help_browser.cpp: Help Browser: Don't double-list a section's root topic https://github.com/wesnoth/wesnoth/commit/5d1e0e20a83cee6017d8fc450309aba5cd34f7a1
20170421 08:35:32< irker370> wesnoth: Celtic Minstrel wesnoth:gui2_help 2adf9d3f5bd3 / src/ (gui/dialogs/help_browser.cpp gui/dialogs/help_browser.hpp help/help.cpp): Help Browser: Support arbitrary initial topic https://github.com/wesnoth/wesnoth/commit/2adf9d3f5bd321c618effa1f096fa9837468f5ec
20170421 08:37:28< celticminstrel> Should I open a PR for lua_formula_bridge, I wonder...
20170421 08:37:41< celticminstrel> Maybe I should test it a little first.
20170421 08:38:34< celticminstrel> PR 993 is probably possible to merge for 1.13.8 (assuming it all works correctly; I don't remember what the state of it was when I last worked on it). PR 994 probably won't make it. lua_formula_bridge is even less likely to make it.
20170421 08:39:32-!- celticminstrel is now known as celmin|sleep
20170421 08:41:32-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170421 08:52:48-!- JyrkiVesterinen [~JyrkiVest@85-76-70-126-nat.elisa-mobile.fi] has quit [Quit: .]
20170421 09:02:41-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170421 09:02:47-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170421 09:05:09-!- atarocch [~atarocch@37.176.107.35] has joined #wesnoth-dev
20170421 09:15:20-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has joined #wesnoth-dev
20170421 09:25:17-!- Kwandulin [~Kwandulin@p200300760F6D80B36877D6CE172B1F72.dip0.t-ipconnect.de] has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
20170421 09:29:58-!- atarocch [~atarocch@37.176.107.35] has quit [Remote host closed the connection]
20170421 09:30:22-!- sevu [~Unknown@141.57.59.163] has joined #wesnoth-dev
20170421 09:30:23-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170421 09:34:22-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has joined #wesnoth-dev
20170421 09:36:13-!- sevu [~Unknown@141.57.59.163] has quit [Quit: Verlassend]
20170421 09:36:32-!- sevu [~Unknown@141.57.59.163] has joined #wesnoth-dev
20170421 09:43:01-!- sevu [~Unknown@141.57.59.163] has quit [Quit: Verlassend]
20170421 09:48:14-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20170421 10:07:22-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170421 10:07:30-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has joined #wesnoth-dev
20170421 10:22:18-!- sevu [~Unknown@141.57.59.163] has joined #wesnoth-dev
20170421 10:26:04-!- sevu is now known as sevu_
20170421 10:26:05-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170421 10:26:31-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has joined #wesnoth-dev
20170421 10:26:41-!- sevu_ is now known as sevu__
20170421 10:27:10-!- sevu__ is now known as sevu_
20170421 10:28:13-!- sevu_ [~Unknown@141.57.59.163] has left #wesnoth-dev ["Verlassend"]
20170421 10:55:36-!- sevu [~Unknown@141.57.59.163] has joined #wesnoth-dev
20170421 11:01:31-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170421 11:07:40-!- gfgtdf [~chatzilla@x4e32b333.dyn.telefonica.de] has joined #wesnoth-dev
20170421 11:13:59-!- atarocch [~atarocch@37.176.107.35] has joined #wesnoth-dev
20170421 11:36:30-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has joined #wesnoth-dev
20170421 11:39:55-!- irker370 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170421 11:42:12-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170421 11:51:23-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20170421 11:51:29-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170421 12:34:56-!- atarocch [~atarocch@37.176.107.35] has quit [Ping timeout: 260 seconds]
20170421 12:44:33-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20170421 13:00:58-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has joined #wesnoth-dev
20170421 13:30:41-!- RatArmy_ [~ratarmy@om126211121095.13.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170421 13:32:04-!- Appleman1234 [~Appleman1@pl33323.ag1212.nttpc.ne.jp] has joined #wesnoth-dev
20170421 13:37:50-!- atarocch [~atarocch@93.56.160.28] has joined #wesnoth-dev
20170421 13:48:59-!- DeFender1031 [~DeFender1@46-116-90-83.bb.netvision.net.il] has quit [Quit: I'm not back now.]
20170421 14:03:33-!- sevu [~Unknown@141.57.59.163] has quit [Remote host closed the connection]
20170421 14:37:48-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has quit [Quit: Connection closed for inactivity]
20170421 15:13:53-!- gfgtdf [~chatzilla@x4e32b333.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.93 [Firefox 52.0.2/20170323105023]]
20170421 15:23:53-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170421 15:24:00-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170421 16:01:35-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20170421 16:45:38-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170421 16:45:48-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170421 16:56:03-!- vultraz_iOS [uid24821@gateway/web/irccloud.com/x-uisyagfcxayjinqt] has joined #wesnoth-dev
20170421 16:56:09-!- vultraz_iOS [uid24821@gateway/web/irccloud.com/x-uisyagfcxayjinqt] has quit [Changing host]
20170421 16:56:09-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has joined #wesnoth-dev
20170421 17:35:25-!- celmin|sleep is now known as celticminstrel
20170421 17:37:36-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170421 17:39:55< vultraz_iOS> celticminstrel: https://gna.org/bugs/index.php?25670
20170421 17:41:24< celticminstrel> Not sure why you're pinging me about it...
20170421 17:41:47< vultraz_iOS> your commit introduced it
20170421 17:41:55< celticminstrel> Oh...?
20170421 17:42:07< vultraz_iOS> https://github.com/wesnoth/wesnoth/commit/1954bb67b4421fb763043d10d227842b6bb0b2f2
20170421 17:42:35< celticminstrel> Does his dialog use menu buttons?
20170421 17:42:54< vultraz_iOS> i don't know
20170421 17:43:01< celticminstrel> Or wait, that's the Lua GUI2 commit...
20170421 17:44:32< vultraz_iOS> I have most of the menus converted to the vector-of-config syntax
20170421 17:44:32< celticminstrel> Well, I can't tell what the problem is without a stack trace though...
20170421 17:47:03< vultraz_iOS> generated stuff like the tod menu are broken tho
20170421 17:47:18< celticminstrel> Well, that's no good.
20170421 17:47:31< celticminstrel> BTW, I assume you know that the tod menu needs to use image=
20170421 17:47:39< vultraz_iOS> better than "everything" broken, though :D
20170421 17:47:41< vultraz_iOS> indeed
20170421 17:47:48< celticminstrel> Well, slightly better, sure.
20170421 17:49:20< vultraz_iOS> the entire system needs a refactor, though
20170421 17:49:23< vultraz_iOS> very inefficient
20170421 17:49:25< celticminstrel> But among that generated stuff is the "back to turn" menu.
20170421 17:49:33< celticminstrel> So, having it broken is kinda unacceptable.
20170421 17:50:09< vultraz_iOS> well, I can repro his crash
20170421 17:51:15< celticminstrel> Do you have a stack trace? (It's no good if it's a stack trace in the toplevel catch handler though.)
20170421 17:51:18< vultraz_iOS> celticminstrel: segfault in #0 0xcd31c5	luaH_setint(lua_State*, Table*, long long, lua_TValue*) () (??:??)
20170421 17:51:21< vultraz_iOS> that's it
20170421 17:51:28-!- JyrkiVesterinen [~JyrkiVest@87-100-254-108.bb.dnainternet.fi] has joined #wesnoth-dev
20170421 17:51:38< celticminstrel> Uh, that's not good enough. I need more stack frames than that.
20170421 17:51:43< celticminstrel> Maybe I'll try it myself.
20170421 17:51:53< vultraz_iOS> sorry, can't do better r/n
20170421 17:55:22< vultraz_iOS> blah
20170421 17:55:45< vultraz_iOS> been using config["id"] for hkey ids, but now I've somehow made it that WML menu item descriptions won't show if id= is not used
20170421 17:55:46< vultraz_iOS> fuck
20170421 17:55:59< vultraz_iOS> that's not what I wanted
20170421 17:56:10< celticminstrel> Wait what?
20170421 17:56:14-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170421 17:56:26-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170421 17:56:44< celticminstrel> BTW, if necessary you can add extra keys to the config that aren't used by the dropdown menu.
20170421 17:56:54< vultraz_iOS> yes
20170421 17:57:00< vultraz_iOS> I was using an id= key
20170421 17:57:11< celticminstrel> Oh, I guess the dropdown menu doesn't use that then.
20170421 17:57:16< celticminstrel> So what's not showing?
20170421 17:57:34< vultraz_iOS> the text
20170421 17:58:06< celticminstrel> Did you edit the dropdown menu implementation?
20170421 17:58:26< vultraz_iOS> no
20170421 18:00:39< vultraz_iOS> ah
20170421 18:00:43< vultraz_iOS> the problem is in get_menu_images
20170421 18:00:57< vultraz_iOS> well
20170421 18:00:58< vultraz_iOS> heh
20170421 18:01:03< vultraz_iOS> now I have the opposite problem
20170421 18:01:07< celticminstrel> IIRC get_menu_images looks for the correct images based on the hotkey ID.
20170421 18:01:20< vultraz_iOS> all generated menu items appear but none of the static ones xD
20170421 18:01:25< celticminstrel> XD
20170421 18:01:53-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170421 18:02:59-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170421 18:04:37< vultraz_iOS> well now nothing is showing ;_;
20170421 18:05:43< vultraz_iOS> ok, let's take a reference
20170421 18:08:15-!- irker216 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20170421 18:08:15< irker216> wesnoth: Charles Dang wesnoth:master 8b8065cb2785 / src/editor/map/context_manager.cpp: Fixup 90de226 https://github.com/wesnoth/wesnoth/commit/8b8065cb27853bb10a421b4920b9866ee4fb67c3
20170421 18:16:51< vultraz_iOS> \o/
20170421 18:16:53< vultraz_iOS> it works!
20170421 18:17:11< vultraz_iOS> ah, except WML items :|
20170421 18:17:15< vultraz_iOS> those show ID not name
20170421 18:17:29< vultraz_iOS> ponder ponder ponder
20170421 18:18:57< vultraz_iOS> hmm
20170421 18:19:00< vultraz_iOS> now they don't show at all ;_;
20170421 18:20:38< matthiaskrgr> everything will be ok
20170421 18:20:41 * matthiaskrgr hugs vultraz_iOS 
20170421 18:20:52< vultraz_iOS> heh :P
20170421 18:21:05< matthiaskrgr> :p 
20170421 18:21:12< vultraz_iOS> ahh
20170421 18:21:16< vultraz_iOS> indeed it works if I use ID
20170421 18:21:44 * zookeeper hands matthiaskrgr the squeaky mallet
20170421 18:21:49< zookeeper> we don't use soft methods here
20170421 18:22:02 * vultraz_iOS flees into the night
20170421 18:22:28 * matthiaskrgr squeaks into the night 
20170421 18:24:28< vultraz_iOS> it wooorks \ o /
20170421 18:25:53< vultraz_iOS> ah, just remembered I forgot something
20170421 18:26:47< vultraz_iOS> celticminstrel: you may peruse https://github.com/Vultraz/wesnoth/commit/d4aba16f16492abdc1eef6b21152f364d564bcaap
20170421 18:27:24< celticminstrel> vultraz_iOS: So just to clarify, have you tested every possible menu in both the editor and the game?
20170421 18:28:11< vultraz_iOS> every possible menu, you say
20170421 18:28:39< celticminstrel> I think one of them may be a right-click menu for units...?
20170421 18:28:49< celticminstrel> So place a unit and right-click it? Not quite sure though.
20170421 18:29:03< celticminstrel> Mind you, I didn't test that one when I switched them to GUI2... >_>
20170421 18:29:10< celticminstrel> Because placing a unit was causing crashes at the time.
20170421 18:29:31< celticminstrel> Your commit seems to not exist.
20170421 18:29:54< vultraz_iOS> https://github.com/Vultraz/wesnoth/commit/d4aba16f16492abdc1eef6b21152f364d564bcaa
20170421 18:30:33< celticminstrel> There was a random p on the end, huh...
20170421 18:33:44< matthiaskrgr> clang build is still failing |: 
20170421 18:34:18< matthiaskrgr> https://pastebin.com/nXwD8R6m  
20170421 18:34:37< vultraz_iOS> matthiaskrgr: just try removing that line?
20170421 18:35:55< matthiaskrgr> then the one above fails
20170421 18:36:13< matthiaskrgr> and then the one above ( FALLTHROUGH; ) fails 
20170421 18:36:31< matthiaskrgr> but removing all 3 lines only leaves a -Wswitch behind
20170421 18:36:39< vultraz_iOS> ;_;
20170421 18:37:04 * matthiaskrgr hugs vultraz_iOS while holding a squeaky mallet 
20170421 18:37:25< matthiaskrgr> ../src/gui/auxiliary/iterator/walker_widget.cpp:41:9: warning: enumeration values 'internal' and 'child' not handled in switch [-Wswitch]
20170421 18:37:55< vultraz_iOS> celticminstrel: I wonder if I need to use std::move with emplace_back(config_of to relaly get moves?
20170421 18:42:39-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170421 18:43:53< celticminstrel> vultraz_iOS: I suspect so.
20170421 18:43:58< vultraz_iOS> ;_;
20170421 18:44:20< celticminstrel> And don't remove the whole line, just remove the fallthrough on that line.
20170421 18:44:31-!- Kwandulin [~Kwandulin@p200300760F6D8096B181ECDE39D95415.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170421 18:45:25< celticminstrel> case internal:
20170421 18:45:28< celticminstrel> case child:
20170421 18:45:30< celticminstrel>         ;
20170421 18:45:34< celticminstrel> Like that ^
20170421 18:45:49< celticminstrel> So there's only one fallthrough left, I think.
20170421 18:46:15< celticminstrel> Though even that one may not really be needed, since it's only falling through to a blank statement.
20170421 18:50:36< vultraz_iOS> celticminstrel: more stuff https://github.com/Vultraz/wesnoth/commit/5f9b4667c01adcd6e68c9600c0f490f61d830aef
20170421 18:50:42< vultraz_iOS> before you give feedback
20170421 18:50:59< celticminstrel> Does that replace the previous commit or augment it?
20170421 18:51:24< vultraz_iOS> augment
20170421 18:51:44< celticminstrel> So I should finish looking over the previous commit first, then.
20170421 18:58:34-!- Kwandulin [~Kwandulin@p200300760F6D8096B181ECDE39D95415.dip0.t-ipconnect.de] has quit [Ping timeout: 245 seconds]
20170421 19:00:49< celticminstrel> All this items.front() nonsense is annoying. Why do the generated placeholders have to appear first?
20170421 19:01:43< celticminstrel> Wait! You commented out the palette manager expansion! That's unacceptable!
20170421 19:03:03-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20170421 19:03:09< vultraz_iOS> celticminstrel: restored in second commit
20170421 19:03:14< celticminstrel> Ah okay.
20170421 19:03:48-!- Kwandulin [~Kwandulin@p200300760F6D8096B181ECDE39D95415.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170421 19:04:00< celticminstrel> ...wonder what ->is_embedded() means.
20170421 19:07:27< celticminstrel> Are you sure this is no longer needed? https://github.com/Vultraz/wesnoth/commit/d4aba16f16492abdc1eef6b21152f364d564bcaa#diff-6042ddb155eaeb75d4794d483a34429eL469
20170421 19:07:54< celticminstrel> I guess it might not be...
20170421 19:08:52< celticminstrel> You know, maybe it should check if(menu) before if(hk == hotkey::HOTKEY_NULL), since the outcome doesn't depend on hk when there's a menu.
20170421 19:10:24< vultraz_iOS> it should nto be since that's the old markup parsing
20170421 19:13:29< vultraz_iOS> celticminstrel: space indent all fixed in second commit
20170421 19:13:39< vultraz_iOS> ignore it
20170421 19:13:51< celticminstrel> I guess that means I don't need to tell you that I made some commit comments. XD
20170421 19:14:14< vultraz_iOS> i simplified the conditional https://github.com/wesnoth/wesnoth/commit/1b4c0fd9b58699d0f825decd1d8dbfb3a4a1dc13
20170421 19:15:00< celticminstrel> Passing a config_of to emplace_back should automatically use move construction, FTR.
20170421 19:15:14< celticminstrel> Because the config_of is an r-value.
20170421 19:15:26< vultraz_iOS> ok
20170421 19:16:19< vultraz_iOS> I should probably convert newsaves from a string vec to a config vec..
20170421 19:16:21< celticminstrel> Not sure if that's why you were asking about move with emplace earlier, but...
20170421 19:16:22< vultraz_iOS> not the other way around 
20170421 19:16:45< vultraz_iOS> celticminstrel: yes
20170421 19:17:09< celticminstrel> Of course, that's assuming config actually supports move construction...
20170421 19:17:16< vultraz_iOS> IIRC it does
20170421 19:17:25< vultraz_iOS> at least, it damn well should
20170421 19:17:44< vultraz_iOS> config::config(config &&cfg):
20170421 19:18:05< celticminstrel> The absence of that wouldn't necessarily mean it's not supported though.
20170421 19:20:11< vultraz_iOS> eh, ill leave the newsaves handling as-is
20170421 19:20:12< vultraz_iOS> easier
20170421 19:20:17< vultraz_iOS> celticminstrel: ok, so I think im done
20170421 19:20:21< vultraz_iOS> barring any feedback
20170421 19:20:36< celticminstrel> No idea what you're saying about newsaves.
20170421 19:20:37< vultraz_iOS> (is it just me or is my method a lot less intrusive than your method)
20170421 19:20:46< celticminstrel> I don't remember.
20170421 19:21:32< celticminstrel> If you want to address the comment about vgettext, that should be a separate commit (ie don't squash it in with the first two).
20170421 19:22:00< vultraz_iOS> alright
20170421 19:26:22< celticminstrel> Well, I don't see any partcular problems.
20170421 19:28:02-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170421 19:29:25-!- JyrkiVesterinen [~JyrkiVest@87-100-254-108.bb.dnainternet.fi] has quit [Quit: .]
20170421 19:29:49< vultraz_iOS> tbh this was a lot simpler than I expected
20170421 19:29:53< irker216> wesnoth: Charles Dang wesnoth:master 7d2d100cf88f / src/ (21 files in 7 dirs): Convert menu item handling to the new vector-of-config method https://github.com/wesnoth/wesnoth/commit/7d2d100cf88f7bd675ad02ca5180ec1ccce5e0c3
20170421 19:29:56< irker216> wesnoth: Charles Dang wesnoth:master 1310e03161d6 / src/hotkey/command_executor.cpp: Simplified a conditional https://github.com/wesnoth/wesnoth/commit/1310e03161d683bd1e2c779f944d7143396e9982
20170421 19:30:45< vultraz_iOS> you made it sound impossible :P
20170421 19:30:57< celticminstrel> Looks like what you did is a subset of what I was attempting.
20170421 19:31:20< vultraz_iOS> the system still needs more refactoring, yes
20170421 19:31:28< vultraz_iOS> but this is a good intermediate step
20170421 19:31:33< vultraz_iOS> strips out the GUI1 markup
20170421 19:31:45< celticminstrel> Wonder if I could rebase my branch onto master now... obviously there'd be a ton  of conflicts though...
20170421 19:32:36-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 260 seconds]
20170421 19:33:27< vultraz_iOS> it's months old, so verily 
20170421 19:35:34< vultraz_iOS> celticminstrel: did we address this? https://gna.org/bugs/?25663
20170421 19:40:30< celticminstrel> Not as far as I know.
20170421 19:41:57< celticminstrel> vultraz_iOS: Shouldn't grow_factor,horizontal_grow=1,true in column 1 and grow_factor,horizontal_alignment=0,right in column 2 force the columns to opposite sides?
20170421 19:42:48< vultraz_iOS> horizontal_alignment doesn't take a number.
20170421 19:43:22< vultraz_iOS> oh
20170421 19:43:24< vultraz_iOS> misread
20170421 19:43:25< vultraz_iOS> uhhh
20170421 19:43:41< vultraz_iOS> dunno
20170421 19:44:22< celticminstrel> Problem is in MP staging low-res version.
20170421 19:44:26< celticminstrel> 1024x769 and smaller.
20170421 19:44:50< celticminstrel> The buttons at the top should be right-aligned.
20170421 19:44:57< vultraz_iOS> ill take a look..
20170421 19:45:25< irker216> wesnoth: Celtic Minstrel wesnoth:master 22980629f7d2 / src/scripting/lua_gui2.cpp: Fixup 1954bb67b https://github.com/wesnoth/wesnoth/commit/22980629f7d241bc825df4c99628b5ad7bb790f2
20170421 19:46:17< vultraz_iOS> do post if that fixes it
20170421 19:46:54< vultraz_iOS> actually i can confirm it fixes it
20170421 19:48:16< celticminstrel> If you're talking about SFD's bug report, I already confirmed it fixes it.
20170421 19:48:24< celticminstrel> And even posted and marked it fixed.
20170421 19:49:58< irker216> wesnoth: Charles Dang wesnoth:master e36023aaabfa / data/gui/window/mp_staging.cfg: MP Staging: fixed button placement on low resolutions https://github.com/wesnoth/wesnoth/commit/e36023aaabfa402ca960f9ecfcd40035cb7de09f
20170421 19:49:59< vultraz_iOS> celticminstrel: ^
20170421 19:50:40< celticminstrel> ...
20170421 19:50:49< celticminstrel> The right-aligned one needs the grow factor?
20170421 19:51:09< vultraz_iOS> if the parent grid grows, yes
20170421 19:51:37< vultraz_iOS> I'm surprised this dialog's low res version actually looks good
20170421 19:51:57< vultraz_iOS> you've been making good low-res versions recently
20170421 19:52:11< vultraz_iOS> sadly, Create is not up the same standard :(
20170421 19:52:28-!- RatArmy_ [~ratarmy@om126211112080.13.openmobile.ne.jp] has joined #wesnoth-dev
20170421 19:52:37< vultraz_iOS> celticminstrel: remember, grow just controls how the cells grow in a gri
20170421 19:52:38< vultraz_iOS> d
20170421 19:52:58< vultraz_iOS> h/v alignment/grow controls how the cell contents are laid out
20170421 19:53:24< celticminstrel> IIRC I also wanted the proportions at the bottom to be different...
20170421 19:53:33< celticminstrel> So the chat takes up more space and the list takes up less...
20170421 19:53:33< vultraz_iOS> celticminstrel: how so?
20170421 19:53:37< vultraz_iOS> ok
20170421 19:54:00< celticminstrel> It crashed when I tried to make that work though.
20170421 19:59:36< irker216> wesnoth: Charles Dang wesnoth:master 02cf739db70e / data/gui/window/mp_staging.cfg: MP Staging: better size ratio for chat and players list at low resolution https://github.com/wesnoth/wesnoth/commit/02cf739db70ef0e5f9ff1d5d0bf2b712a2b96559
20170421 19:59:37< vultraz_iOS> celticminstrel: ^
20170421 19:59:47< vultraz_iOS> a rare use of non-1 grow_factor appears!
20170421 19:59:57< celticminstrel> :P
20170421 20:00:38< vultraz_iOS> perhaps non 0/1 grow_factor is useful after all...
20170421 20:01:51< vultraz_iOS> I wonder if perhaps I could use it for certain layouts...
20170421 20:01:59< vultraz_iOS> instead of forcing min size...
20170421 20:02:03< vultraz_iOS> hmmmmmmmmmmmmmmmmmmmmmmmmmmm
20170421 20:02:15< vultraz_iOS> have I too easily dismissed a useful feature?
20170421 20:03:16< celticminstrel> Quite likely. :P
20170421 20:03:43< vultraz_iOS> I still recommend grow_factor in its simplest form when possible, though
20170421 20:03:52< vultraz_iOS> ie, don't use 10 when you can use 1
20170421 20:04:15< celticminstrel> Yeah, it's probably a good idea to keep the numbers as small as possible, I guess.
20170421 20:06:48< vultraz_iOS> i guess the reason i haven't liked it is i've seen some blatantly wrong uses
20170421 20:06:54< vultraz_iOS> and i guess that made me assume 0/1 was best
20170421 20:08:09-!- gfgtdf [~chatzilla@x4e32b333.dyn.telefonica.de] has joined #wesnoth-dev
20170421 20:08:22< gfgtdf> vultraz_iOS: uhm what eaxtly is tha advantage of 7d2d100cf88f7 ?
20170421 20:08:45< vultraz_iOS> gfgtdf: getting rid of GUI1 handling
20170421 20:09:01< vultraz_iOS> stuff like IMAGE_SEPARATOR
20170421 20:09:02-!- RatArmy_ [~ratarmy@om126211112080.13.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170421 20:09:15< vultraz_iOS> is there something wrong with it?
20170421 20:09:49< gfgtdf> vultraz_iOS: bot most of tehse std::vector<std::string> contains basically poen id, and not gui1 encodes gui strings
20170421 20:09:50< gfgtdf> #
20170421 20:09:58< vultraz_iOS> yes
20170421 20:10:08< gfgtdf> vultraz_iOS: so why did you do it ?
20170421 20:10:25< vultraz_iOS> because the new menus use a vector of config
20170421 20:10:32< vultraz_iOS> so it's better to have a unified interface
20170421 20:10:57< gfgtdf> vultraz_iOS: but they dont are the same  ones containonyl an id while other contsins gui things
20170421 20:11:28< celticminstrel> ?
20170421 20:11:55< celticminstrel> I don't even understand what gfgtdf is asking.
20170421 20:12:39< gfgtdf> it seems to me kikt cultras replaces a vector of id string with vector of configs that only have one id value
20170421 20:13:06< vultraz_iOS> yes
20170421 20:13:13< vultraz_iOS> initially
20170421 20:13:18< vultraz_iOS> and then they also get a label value
20170421 20:15:39< gfgtdf> hmm
20170421 20:15:57< vultraz_iOS> celticminstrel: looking into the sota bug
20170421 20:16:08< vultraz_iOS> celticminstrel: i wonder if it's a bug with [unit]...
20170421 20:16:22< vultraz_iOS> celticminstrel: if you use placement=map instead of map_overwrite, the unit appears and animates
20170421 20:16:58< gfgtdf> vultraz_iOS: still when you know what a strict contains in the end it better to use a cusotm struct like "struct menu_item_info { std::string& label; std::string& id; }"
20170421 20:17:29< gfgtdf> vultraz_iOS: the problem with configs in c++ code are that:
20170421 20:17:46< vultraz_iOS> gfgtdf: well then I'd need to do that for all menu handling
20170421 20:18:03< celticminstrel> I think it'd be less convenient to use a struct here.
20170421 20:18:03< vultraz_iOS> not just the game content generation
20170421 20:18:15< celticminstrel> A struct must have every member, but in fact some of those members are optional.
20170421 20:18:20< gfgtdf> 1) from looking at the code its qute hard to see what the data contains, this is ok if is really some "arbitary data that is pased toa  unkown-to-this code function"
20170421 20:19:02< gfgtdf> 2) it's rather slow, things like cfg["id"] requite: 1) a strign lookup, 2) copyign the result string
20170421 20:19:15< gfgtdf> celticminstrel: then use optional (or leave the string as empty is that good enough)
20170421 20:19:38< celticminstrel> "rather slow" sounds like an exaggeration.
20170421 20:20:33< celticminstrel> It's true that in the most general case it's not as efficient as perhaps it could be.
20170421 20:20:43-!- RatArmy_ [~ratarmy@om126211112080.13.openmobile.ne.jp] has joined #wesnoth-dev
20170421 20:20:52< celticminstrel> But, that string lookup will be pretty fast, because there are only around 10 or so elements in the tree.
20170421 20:21:04< celticminstrel> And copying the result string won't be all that slow, because it's a fairly short string.
20170421 20:21:23< celticminstrel> In this specific case, that is.
20170421 20:22:09< gfgtdf> celticminstrel: well yes if you repce on thign with config you probalby won't notice, but it accumulates. if you replace all you data with config your programm will be slower.
20170421 20:22:30< vultraz_iOS> gfgtdf: https://github.com/wesnoth/wesnoth/commit/c4625197d7c6d0f37f broke [unit] placement=map_overwrite
20170421 20:22:36< vultraz_iOS> cause of this https://gna.org/bugs/?25663
20170421 20:22:50< celticminstrel> Well, sure, that might be true.
20170421 20:23:05< vultraz_iOS> since it can't override the unit at a location
20170421 20:23:16< vultraz_iOS> this is a serious regression
20170421 20:24:10< celticminstrel> So basically an erase call needs to be added.
20170421 20:24:36< celticminstrel> Possibly only if the placement attribute requested an overwrite.
20170421 20:25:07< celticminstrel> If it's a problem with [unit], then it would be in actions/unit_creator.cpp most likely.
20170421 20:25:15< vultraz_iOS> it i
20170421 20:25:16< vultraz_iOS> s
20170421 20:25:19< vultraz_iOS> i traced it there
20170421 20:25:25< irker216> wesnoth: gfgtdf wesnoth:master 3b63dc99ecba / src/actions/unit_creator.cpp: fix [unit] placement=map_overwrite https://github.com/wesnoth/wesnoth/commit/3b63dc99ecbaa47a2a3be99659fd3ddaee46b004
20170421 20:25:49< celticminstrel> Eh, close enough I guess.
20170421 20:26:02< celticminstrel> I suppose find_location wouldn't return an occupied location if placement forbade it.
20170421 20:26:13-!- atarocch [~atarocch@93.56.160.28] has quit [Remote host closed the connection]
20170421 20:27:39< gfgtdf> shoudl prbablychange unitmap::replace to take unit_ptr, was just easier to find all occurenaces this way.
20170421 20:28:26< celticminstrel> Pretty sure I said that back when you were doing that "less unit copying".
20170421 20:31:53-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170421 20:45:58-!- Herk_ [9b04855f@gateway/web/freenode/ip.155.4.133.95] has joined #wesnoth-dev
20170421 20:47:21< gfgtdf> i'm wondering should we try to fix the wrong map appering in the map lobby if the map was change from wml?
20170421 20:47:41< gfgtdf> i mean it coudl inveove some more network overhead updataing the server with the latest map infow whn it was changed
20170421 20:47:48< gfgtdf> involve*
20170421 20:50:05< gfgtdf> i man we cold like keep the host track of whehther teh map hash changes since the last server update and do a server opdate once per turn
20170421 20:50:15< gfgtdf> could*
20170421 20:50:26< celticminstrel> Well, a naive implementation would resend the whole map, but you could easily just send the changes to the map, too.
20170421 20:52:27< vultraz_iOS> I wonder of maybe I should use a struct for the menu content generation after all..
20170421 20:53:01< vultraz_iOS> Ponder ponder
20170421 20:53:21< gfgtdf> vultraz_iOS: in always in vaour of the struct, i remember debuggin the unit clas which previously had a config memeber and it was quite hard to keep track of all possible attributes
20170421 20:53:51< gfgtdf> vultraz_iOS: that's also why i then replaces it in favour of more unit attributes
20170421 20:54:11< vultraz_iOS> Maybe I'll do that then at the end of the refactoring
20170421 20:54:20< gfgtdf> more unit memeber fields
20170421 20:54:23< gfgtdf> i meant
20170421 20:54:45< celticminstrel> Keep in mind that if you use a struct, you need to convert the config to the struct at some point.
20170421 20:55:07< celticminstrel> I mean, you can't fully eliminate the use of config for menu content, because it's directly read from the GUI2 WML.
20170421 20:56:32< vultraz_iOS> What?
20170421 20:56:39< vultraz_iOS> Nothing is read from the gui2 wml
20170421 20:56:47< celticminstrel> Yes it is?.
20170421 20:56:56< celticminstrel> [menu_button] supports [options], right?
20170421 20:57:02< vultraz_iOS> It does?
20170421 20:57:16< celticminstrel> Well, it sure looks like it from the code.
20170421 20:57:35< vultraz_iOS> Easy enough to convert from a config tho
20170421 20:57:45< celticminstrel> Well yes, it's easy to do.
20170421 20:58:47< vultraz_iOS> Anyway, I need to do some more refactoring
20170421 20:58:51< vultraz_iOS> The current interface is flawed
20170421 20:58:57< celticminstrel> ???
20170421 20:59:36< vultraz_iOS> Too much spurious iteration
20170421 21:00:33< celticminstrel> Are you talking about the commit you just made, or the menu_button interface?
20170421 21:01:33< vultraz_iOS> The theme menu generation expansion interface
20170421 21:02:55< vultraz_iOS> The menu button interface is perfectly fine
20170421 21:06:25-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20170421 21:28:31-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170421 21:37:23< vultraz_iOS> hmm
20170421 21:37:29< vultraz_iOS> ok, let's see, what is this code doing
20170421 21:37:49< celticminstrel> Which code?
20170421 21:38:06< vultraz_iOS> editor_controller::show_menu
20170421 21:38:15< vultraz_iOS> where you complained about the front()
20170421 21:38:20< vultraz_iOS> excessive front()*
20170421 21:40:24< celticminstrel> Ah, right.
20170421 21:41:13< vultraz_iOS> so, it looks like it's's basically checking for which menu is being invoked
20170421 21:42:35< vultraz_iOS> it copies the IDs from the argument, then checks which one is first...
20170421 21:42:36-!- RatArmy_ [~ratarmy@om126211112080.13.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170421 21:45:39-!- mkdroid [~null@unaffiliated/matthiaskrgr] has joined #wesnoth-dev
20170421 21:46:53< celticminstrel> Since it's always checking front(), it looks like it won't work if a theme decides to place a generated menuitem anywhere other than at the top of the menu... though I could be missing something.
20170421 21:47:20< vultraz_iOS> first let me clean up the code a little
20170421 21:51:48-!- Kwandulin [~Kwandulin@p200300760F6D8096B181ECDE39D95415.dip0.t-ipconnect.de] has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
20170421 21:56:08-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20170421 22:03:59-!- mkdroid [~null@unaffiliated/matthiaskrgr] has quit [Remote host closed the connection]
20170421 22:04:21-!- RatArmy_ [~ratarmy@om126211112080.13.openmobile.ne.jp] has joined #wesnoth-dev
20170421 22:07:15< vultraz_iOS> ok, I can remove some of these unnecessary loops
20170421 22:07:32< vultraz_iOS> in the editor expand_ functions
20170421 22:07:43< vultraz_iOS> for some reason they loop the list looking for the first valid entry
20170421 22:07:56< vultraz_iOS> but in show_menu, the entry *needs* to be the first one
20170421 22:08:28< vultraz_iOS> so the loops are useless
20170421 22:08:58< vultraz_iOS> then again, what if you wanted to have multiple items in a menu
20170421 22:08:59< vultraz_iOS> hmm
20170421 22:09:04< vultraz_iOS> maybe I should leave the loops
20170421 22:09:27< irker216> wesnoth: Charles Dang wesnoth:master d18b94d65d06 / src/editor/controller/editor_controller.cpp: Editor: cleaned up show_menu code https://github.com/wesnoth/wesnoth/commit/d18b94d65d06404c3e57cf50b3ea7092fb63727a
20170421 22:09:41< vultraz_iOS> celticminstrel: the bothersome front() is gone
20170421 22:12:32-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170421 22:13:05-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170421 22:15:03< celticminstrel> Not really what I meant...
20170421 22:15:09< celticminstrel> [Apr 21@6:08:58pm] vultraz_iOS: then again, what if you wanted to have multiple items in a menu
20170421 22:15:10< celticminstrel> That ^
20170421 22:17:03< vultraz_iOS> well, that should be handled by the main function not by the expansion functions, I guess
20170421 22:17:03< vultraz_iOS> hmm
20170421 22:21:02 * celticminstrel is getting rid of the loops in the expand_* functions.
20170421 22:21:16< vultraz_iOS> oh
20170421 22:21:17< vultraz_iOS> yay!
20170421 22:24:44< gfgtdf> speaking about menus: the 'back to turn' menu is still bugged: the parent meanu disappears when the child menu pops up.
20170421 22:25:09< celticminstrel> That's hard to fix.
20170421 22:26:08< vultraz_iOS> yes, it's hard to fix
20170421 22:29:36< celticminstrel> If you wanted to make a loop in editor_controller::show_menu, what I'm doing now is one step towards it... the other thing you'd need to do is do something about the active_menu_ value.
20170421 22:29:56< celticminstrel> That member means that no menu can contain more than one generated section, I think.
20170421 22:30:03< celticminstrel> If I understand it correctly.
20170421 22:30:19< vultraz_iOS> well, do we need it?
20170421 22:30:36< vultraz_iOS> multiple generated sections
20170421 22:30:41< celticminstrel> I don't know.
20170421 22:31:02< celticminstrel> I think we do need generated sections that aren't at the top of the menu, but...
20170421 22:31:08< celticminstrel> Changing stuff in editor_controller won't really help there.
20170421 22:31:14< vultraz_iOS> yeah
20170421 22:31:15< celticminstrel> Because that's the WML Menuitems section.
20170421 22:31:22< celticminstrel> Which is in the game, not the editor.
20170421 22:31:52< vultraz_iOS> we need to implement actual widget uses in the multimenu
20170421 22:32:08< vultraz_iOS> er
20170421 22:32:22< vultraz_iOS> I mean, we need actual widgets for the playlist menu, say
20170421 22:32:57< vultraz_iOS> not sure how to do that...
20170421 22:33:05< vultraz_iOS> oh, wait, the dropdown handles everything
20170421 22:33:17< celticminstrel> This might actually be a little less efficient since it's using std::inserter... oh well, probably won't make that big a difference.
20170421 22:33:26< celticminstrel> What are you talking about?
20170421 22:33:51< celticminstrel> Oh right, you want to use the checkbox option for the music track menu?
20170421 22:34:00< vultraz_iOS> yes
20170421 22:34:13< vultraz_iOS> and implement another option for the radio buttons
20170421 22:34:15< celticminstrel> Well, show_menu doesn't currently support that, and there's not really any reason for it to, but maybe if you bypassed show_menu in that case.
20170421 22:34:46< celticminstrel> As for the radio buttons, I actually think there's no advantage to implementing an option for it; you don't expect the menu to stay open when you change the selection, anyway.
20170421 22:34:54< vultraz_iOS> why would I need to bypass it?
20170421 22:34:56< celticminstrel> And that also applies to any other checkboxes other than the one in the music track menu.
20170421 22:35:00< vultraz_iOS> the checkbox key is in the data config
20170421 22:35:08< celticminstrel> But keep open is not.
20170421 22:35:14< celticminstrel> And shouldn't be, really.
20170421 22:35:26< celticminstrel> And there's no point in using the checkbox key without keep open.
20170421 22:35:32< vultraz_iOS> this would be easier if the UI were in GUI2...
20170421 22:35:38< celticminstrel> I dunno about that...
20170421 22:35:55< vultraz_iOS> (though the multimenu would need to be themed)
20170421 22:36:00< celticminstrel> Maybe it would, maybe not.
20170421 22:36:13< celticminstrel> (Be easier, I mean.)
20170421 22:36:23-!- Herk_ [9b04855f@gateway/web/freenode/ip.155.4.133.95] has quit [Quit: Page closed]
20170421 22:51:13< celticminstrel> ...oh wow, the editor toolbar is just a little bit too long for 800x600. The music track list button overlaps another button.
20170421 22:51:14-!- RatArmy_ [~ratarmy@om126211112080.13.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170421 22:51:33-!- zookeeper [zookeeper@wesnoth/developer/zookeeper] has quit [Ping timeout: 258 seconds]
20170421 22:52:20< vultraz_iOS> huh
20170421 22:52:22< vultraz_iOS> it is
20170421 22:52:35< irker216> wesnoth: Celtic Minstrel wesnoth:master 2e988809e889 / src/editor/ (8 files in 3 dirs): Eliminate pointless loops in editor menu generation https://github.com/wesnoth/wesnoth/commit/2e988809e8891c3cc89ff798586eed52df07aaf7
20170421 22:53:02< celticminstrel> With minimal work now, if you wanted, you could implement the ability for the generated portion to not come at the top of the menu.
20170421 22:53:22< celticminstrel> You'd just need to add an offset stored along with the active_menu_, or something like that.
20170421 22:54:49< vultraz_iOS> \o/
20170421 22:54:58< celticminstrel> Of course, that's only for the editor...
20170421 22:55:12< vultraz_iOS> i don't think the game has generated sections
20170421 22:56:07< celticminstrel> It has at least two.
20170421 22:56:17< celticminstrel> The Back to Turn menu and WML-set menuitems.
20170421 22:56:37< celticminstrel> So, I was thinking we should reorder the context menu to put recruit/recall at the top. What do you think?
20170421 22:57:02< vultraz_iOS> hmmm
20170421 22:57:05< vultraz_iOS> sure
20170421 22:58:05< celticminstrel> Then what should the order of the other things be?
20170421 22:58:12< celticminstrel> Besides recruit/recall, we have...
20170421 22:58:36< celticminstrel> Custom WML menuitems
20170421 22:58:41< celticminstrel> Whiteboard actions
20170421 22:58:48< vultraz_iOS> custom WML first
20170421 22:58:51< celticminstrel> Debug actions (eg create, kill, change side)
20170421 22:59:04< celticminstrel> Informational items (unit, terrain)
20170421 22:59:05< vultraz_iOS> those last, maybe
20170421 22:59:15< celticminstrel> Labels
20170421 22:59:17< celticminstrel> Chat
20170421 22:59:21< celticminstrel> Continue
20170421 22:59:37< celticminstrel> Ah wait, I'm looking at classic theme not default here.
20170421 22:59:56< celticminstrel> Chat was removed.
20170421 22:59:57< vultraz_iOS> oh geez
20170421 22:59:59< vultraz_iOS> no wonder
20170421 23:00:00< vultraz_iOS> xD
20170421 23:00:06< vultraz_iOS> i cleaned it up back in 2015
20170421 23:00:13< vultraz_iOS> also, unused variable warnings
20170421 23:00:17< celticminstrel> No, that's the only one that was wrong though.
20170421 23:00:21< celticminstrel> What unused variable?
20170421 23:00:36< celticminstrel> You really think informational items should be last?
20170421 23:00:39< vultraz_iOS> i
20170421 23:00:51< celticminstrel> Where?
20170421 23:00:57< vultraz_iOS> editor_palettes.hpp:66
20170421 23:01:01< vultraz_iOS> palettes
20170421 23:01:07< celticminstrel> The only other items I didn't mention are delay shroud updates and update shroud.
20170421 23:01:09< celticminstrel> Oh.
20170421 23:01:17< celticminstrel> If it's really unused, just delete the name.
20170421 23:01:20< vultraz_iOS> ya know what
20170421 23:01:24< vultraz_iOS> just leave the order
20170421 23:01:25< celticminstrel> "int i" -> "int" or "int /*i*/"
20170421 23:01:28< vultraz_iOS> and move the other items to the top
20170421 23:01:41< celticminstrel> So just move recruit/recall and leave everything else untouched?
20170421 23:01:45< vultraz_iOS> yes
20170421 23:01:48< celticminstrel> What about continue?
20170421 23:02:47< celticminstrel> I'm getting so many WML errors/warnings, it's annoying...
20170421 23:02:51< vultraz_iOS> no opinion
20170421 23:02:57< celticminstrel> I guess that's what I get for copying over my old addons from 1.10. >)>
20170421 23:03:00< celticminstrel> ^ >_>
20170421 23:03:10-!- RatArmy_ [~ratarmy@om126211112080.13.openmobile.ne.jp] has joined #wesnoth-dev
20170421 23:03:31< vultraz_iOS> 1.0 O_O
20170421 23:03:33< vultraz_iOS> 1.10*
20170421 23:03:57< irker216> wesnoth: Charles Dang wesnoth:master d33198c3d527 / src/editor/palette/editor_palettes.cpp: Fixup 2e98880 (unused variable warnings) https://github.com/wesnoth/wesnoth/commit/d33198c3d5274f364296d7cebe27ff823918e19d
20170421 23:04:23< celticminstrel> Looks like no extra work is required - I just rearranged the theme, and the WML menuitems still work perfectly.
20170421 23:04:32< celticminstrel> So in-game it already works to have the generated portions not at the top.
20170421 23:05:20< vultraz_iOS> yes, because WML and saves have loops
20170421 23:05:29< vultraz_iOS> like the ones you just removed
20170421 23:06:00< vultraz_iOS> should apply a similar change
20170421 23:06:07< vultraz_iOS> expand_wml_commands
20170421 23:06:43< celticminstrel> Presumably it also means they aren't just checked using front() like the editor was...
20170421 23:07:06< vultraz_iOS> there's no check
20170421 23:08:06< celticminstrel> Also, should we adopt ancestral's new keybindings as default?
20170421 23:08:08-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170421 23:08:15< celticminstrel> I don't quite remember what they were though...
20170421 23:08:41-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170421 23:09:26< vultraz_iOS> dunno..
20170421 23:09:36< vultraz_iOS> they're in an addon i think
20170421 23:10:57< celticminstrel> Hmm... is it my imagination or are a bunch of menu icons missing now...
20170421 23:11:30< vultraz_iOS> weird..
20170421 23:11:39< vultraz_iOS> just had wesnoth freeze up and now i can't repro
20170421 23:12:22< vultraz_iOS> ah wait, got it
20170421 23:12:31< vultraz_iOS> open help in the editor and then attempt to quit
20170421 23:12:40< vultraz_iOS> wesnoth freezes
20170421 23:12:47< celticminstrel> Fun...
20170421 23:12:58< vultraz_iOS> I'll let someone else fix this
20170421 23:16:48< celticminstrel> Looks like I probably am just imagining it...
20170421 23:17:00< celticminstrel> But, you know, it might be nice to have more menu icons.
20170421 23:17:49< irker216> wesnoth: Celtic Minstrel wesnoth:master a1c22e509887 / data/themes/default.cfg: Rearrange default theme context menu to put most important commands at the top https://github.com/wesnoth/wesnoth/commit/a1c22e5098875a9139e5f3b2a129e4ae3212005a
20170421 23:18:46< celticminstrel> I wonder if it's possible for custom themes to place specific WML items in specific order in the menus...
20170421 23:19:29< vultraz_iOS> ah, good formatting
20170421 23:19:51-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20170421 23:20:01< celticminstrel> I used the quotes with + for its line continuation property.
20170421 23:20:10< vultraz_iOS> hm
20170421 23:20:22< vultraz_iOS> actually, custom WML menu items should be first
20170421 23:20:31< celticminstrel> I think I disagree, TBH.
20170421 23:20:49< celticminstrel> They are important, but core mechanics such as recruit are even more important.
20170421 23:21:03< celticminstrel> Maybe continue shouldn't be first though. I'm unsure about that one.
20170421 23:21:05< vultraz_iOS> well, we'll see what the public things
20170421 23:21:13< celticminstrel> Might be better to put it third, or fourth.
20170421 23:21:24< celticminstrel> So that it doesn't push down recruit/recall when it appears.
20170421 23:30:42-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170421 23:31:14-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170421 23:32:02< gfgtdf> i perosnllay use the ricght click menu e4xclusiveles for cusotm things items since for all imporant core features (recuit/recall) i know the hotkey
20170421 23:33:08< celticminstrel> Well, that's why I'd never consider, for example, putting the custom items at the bottom.
20170421 23:33:12< gfgtdf> actuall i don't event know what the 'continue' item does
20170421 23:33:22< celticminstrel> I think that's the "press t to continue" thing.
20170421 23:33:28< gfgtdf> ok
20170421 23:45:07< celticminstrel> Oh hey, I'd forgotten about this... https://github.com/wesnoth/wesnoth/commit/3b1237af9dcc54723345de6b6fd46850032900e7
20170421 23:45:08-!- RatArmy_ [~ratarmy@om126211112080.13.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170421 23:47:39< celticminstrel> There's also this: https://github.com/wesnoth/wesnoth/commit/c18ca562a81d8e9aa7aef45f248b29fdac303fd5
20170421 23:49:04< celticminstrel> Mind you, if we're not supporting 10.7 anymore, maybe it's easier to drop Growl support and just always use Cocoa notifications which are already available...
20170421 23:49:17< celticminstrel> Of course, if we're not supporting 10.7 anymore, I can't do Mac stuff even if I wanted to.
20170421 23:50:32< celticminstrel> Oh yeah, people should comment on PR 992.
20170421 23:52:44< celticminstrel> This seems like something that should perhaps be merged: https://github.com/wesnoth/wesnoth/commit/1cf4f067ea8ad03fe485c8288d6044769aa00f4d
20170421 23:55:13< celticminstrel> Any objections?
20170421 23:56:06< celticminstrel> Hmm, not sure about the change to the text about movement points...
20170421 23:58:04< celticminstrel> Come to think of it, that's not my only reservation either...
20170421 23:59:09< vultraz_iOS> what is Growl?
20170421 23:59:21< celticminstrel> A desktop notifications service.
20170421 23:59:34< celticminstrel> For Mac.
20170421 23:59:48< celticminstrel> I have no idea if it still exists now that notifications are built into the OS.
--- Log closed Sat Apr 22 00:00:06 2017