--- Log opened Thu Aug 11 00:00:30 2016
20160811 00:00:36< Aginor> anyway, I need to focus on work, not on this discussion
20160811 00:00:39< Aginor> I'm detaching
20160811 00:00:42< Aginor> laters
20160811 00:00:50< fabi__> Well, I have done a lot work on the Wesnoth franchise and in different branches. You just get the impression that I am inactive because I don't cotribute into master.
20160811 00:00:58< vultraz> If you're pissed that developers have no say in the expenditure of funds, that's one thing. But you as an individual developer don't get special treatment.
20160811 00:01:10< celmin> I'd still count that as inactive, fabi.
20160811 00:01:30< vultraz> And there's nothing that says wesnoth.inc can't commission art for another game.
20160811 00:01:49< celmin> That's what he's complaining about, silly.
20160811 00:02:06< fabi__> ^
20160811 00:02:36< vultraz> In fact, I would very much be happy if wesnoth inc. rebranded as a game studio
20160811 00:02:41< vultraz> and handled multiple games.
20160811 00:02:52< celmin> Honestly, I don't think that's such a good idea.
20160811 00:03:50< vultraz> if you want a portion of the budge dedicated exclusively to *this* project, great.
20160811 00:07:56< celmin> I don't think anything you say makes the situation any better, vultraz. It might even be making it worse.
20160811 00:10:14< vultraz> hey, I agree that stuff needs to change
20160811 00:10:21< vultraz> Im not justifying the way things are
20160811 00:10:50< vultraz> but some things, like using the fund for other game commissions, are alright, at least to t a point
20160811 00:11:18< celmin> At least one person clearly disagrees with you.
20160811 00:11:26< celmin> So you can't really say "it's alright".
20160811 00:11:42< celmin> …admittedly, fabi's barely a contributor at this point.
20160811 00:12:04< fabi__> https://gist.github.com/fendrin/75a1a83c3811319fee110ac713f0aa6d
20160811 00:12:14< celmin> At most the official LoW maintainer? Not sure of that though.
20160811 00:12:19< fabi__> The transcompiler in action.
20160811 00:13:56< celmin> I think I've mentioned this before, but I don't see how that improves upon WML.
20160811 00:14:08< celmin> It's no prettier, for example. (Admittedly it's also no uglier.)
20160811 00:14:24< fabi__> Well, it can be more pretty.
20160811 00:14:33< celmin> Doubt it?
20160811 00:14:55< fabi__> The current translation wants to match the original code very much.
20160811 00:15:02< fabi__> To not scare away WML coders.
20160811 00:15:18< fabi__> But you can code quite elegant in Lua or MoonScript.
20160811 00:15:38< celmin> There's really nothing wrong with WML in the first place. (Barring the preprocessor.)
20160811 00:15:54< fabi__> I strongly disagree here.
20160811 00:16:30< celmin> It's an XML-like syntax.
20160811 00:16:54< celmin> Do you also dislike XML?
20160811 00:17:09< fabi__> Well, not as a format that is meant to be human readable.
20160811 00:17:14< fabi__> XML is not that human readable.
20160811 00:17:24< celmin> True.
20160811 00:17:35< celmin> WML is more readable than XML IMO
20160811 00:17:41< fabi__> Indeed
20160811 00:18:41< fabi__> The main advantages of WSL vs WML is the lua controll structures in event scripting.
20160811 00:19:25< fabi__> There are linters for lua and moonscript out there.
20160811 00:19:30< fabi__> Compilers.
20160811 00:19:39< celmin> Well, events never should've been done in WML anyway.
20160811 00:19:45< fabi__> Agreed.
20160811 00:20:37< fabi__> celmin: Do you know that text? https://www.lua.org/history.html
20160811 00:28:36-!- Greg-Bog_ [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160811 00:31:45< celmin> I don't understand why you're linking that...
20160811 00:32:09-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 260 seconds]
20160811 00:32:23-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has joined #wesnoth-dev
20160811 00:32:24< fabi__> The text is interresting and shows parts of my motiviation.
20160811 00:32:54< celmin> I think I'm gonna switch all Lua registry keys to strings instead of light userdata.
20160811 00:33:00< celmin> Unless someone objects.
20160811 00:33:02< celmin> Like gfgtdf 
20160811 00:33:03< fabi__> Lua envolved from two languages, one pretty much serving the same purpose as WML is doing for us in non event context.
20160811 00:33:18< ancestral> So yeah I still am not sure what to do about SHA changes
20160811 00:34:46< vultraz> ancestral: honestly, at this point just leave it
20160811 00:37:45< ancestral> It’s dev anyway
20160811 00:38:12< celmin> Should probably publish the fixed SHA though so people stop complaining.
20160811 00:38:38< celmin> Anyway, if it really is byte-for-byte identical, there shouldn't be anything to worry about.
20160811 00:39:09< ancestral> Probably that’s the case
20160811 00:45:01-!- Bonobo [~Bonobo@2001:44b8:254:3200:10d2:81a8:da7c:7079] has joined #wesnoth-dev
20160811 00:47:19-!- Greg-Bog_ [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160811 00:50:15-!- Bonobo [~Bonobo@2001:44b8:254:3200:10d2:81a8:da7c:7079] has quit [Ping timeout: 264 seconds]
20160811 00:50:28-!- gfgtdf [~chatzilla@x4e369e4d.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 48.0/20160726073904]]
20160811 00:51:11-!- Bonobo [~Bonobo@2001:44b8:254:3200:688c:4cc:c6ff:ef62] has joined #wesnoth-dev
20160811 00:55:25-!- iceiceice [~chris@unaffiliated/iceiceice] has joined #wesnoth-dev
20160811 00:55:51< iceiceice> celmin, fwiw i think using strings instead of lightuserdata for keys is the way to go
20160811 00:55:55< iceiceice> its a lot simpler i think
20160811 00:56:00< celmin> Oh hi.
20160811 00:56:21< celmin> I was actually going to ask you, but you weren't around.
20160811 00:56:23< iceiceice> also fwiw i don't think there's anything terribly wrong with wml
20160811 00:56:28< celmin> Yeah.
20160811 00:56:31< iceiceice> i dont like the preprocessor though
20160811 00:56:34< celmin> Yeah.
20160811 00:56:37< iceiceice> and i wish that it could be more efficient
20160811 00:56:48< celmin> That what could be more efficient?
20160811 00:56:52< iceiceice> evaluataing WML
20160811 00:57:00< celmin> Does that mean parsing, or the config class?
20160811 00:57:06< iceiceice> it means also the execution
20160811 00:57:21< iceiceice> like, it would be cool if the game logic can run in like a stack machine
20160811 00:57:24< iceiceice> so all the info is local
20160811 00:57:38< iceiceice> and you dont derference a ton of pointers every time you do something
20160811 00:57:41< celmin> Would be cool, yeah.
20160811 00:57:49< iceiceice> lua already runs in a stack machine
20160811 00:58:02< iceiceice> and even beyond that, there are like special distributions of lua
20160811 00:58:06< iceiceice> that can do like JIT compilation
20160811 00:58:08< iceiceice> to native code
20160811 00:58:16< celmin> Well, the ActionWML is really a thin veneer over Lua, so I suppose you could say it has a stack.
20160811 00:58:20< iceiceice> without changing syntax or me having to maintain that nonsense
20160811 00:59:02< vultraz> iceiceice: I thought you weren't an active dev anymore :P
20160811 00:59:34< iceiceice> well i'm still interestd in the project, i mean i contributed for several years
20160811 00:59:39< iceiceice> and i fixed like a hundred bugs in 1.12 :p
20160811 00:59:58< iceiceice> but yeah i'm not that interested in contibuting to the current version
20160811 01:00:08< iceiceice> not for any one reason i guess
20160811 01:00:28< iceiceice> i'm sort of more motivated by like, open source generally than any one particular open source project
20160811 01:01:18< iceiceice> i like to follow the dev work though, and sometimes i can provide useful info (i think / hope)
20160811 01:01:44< vultraz> you could be more useful fixing bugs :P
20160811 01:01:52< vultraz> or cleaning up cruft
20160811 01:02:45< iceiceice> i guess mostly what i worked on in the last month is my variant type,
20160811 01:02:53< iceiceice> i submitted it to boost library incubator :p
20160811 01:03:05< iceiceice> i think it fills a useful niche
20160811 01:04:32< vultraz> i like to think we're making the codebase cleaner
20160811 01:04:47< vultraz> it's honestly scary sometimes how much there is to do :P
20160811 01:07:41< vultraz> and shit, I seem to have broken the editor 
20160811 01:08:46-!- Ravana_ [~Ravana@unaffiliated/ravana/x-2327071] has quit [Read error: Connection reset by peer]
20160811 01:10:56< celmin> Hmmm… mp_options.cpp:213
20160811 01:11:01< celmin> Not sure what to do here.
20160811 01:11:25< celmin> Oh, I have an idea.
20160811 01:12:20< celmin> …oh, these functions are private.
20160811 01:14:23-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160811 01:18:04< vultraz> ok, so I'm just going to define the layout for MP Create
20160811 01:18:09< vultraz> it's not gonna be functional yet
20160811 01:18:24< shadowm> Excuse me, but using funds held by a company that was created for the express purpose of handling Wesnoth's finances, to support other games without telling anyone is not "alright".
20160811 01:18:58-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160811 01:18:58< celmin> ^
20160811 01:19:03< shadowm> But then again, I don't think this is a thing that anyone has said is factually true, surely? Last time I checked it was just a rumor going around of a future possibility.
20160811 01:19:09< fabi__> ^ Damn  Ihate to agree with shadowm
20160811 01:19:15< vultraz> HAH
20160811 01:19:37< vultraz> I spoke to dave about it once.
20160811 01:19:50< Aginor> yes, rumours are certainly not helping enything
20160811 01:19:55< Aginor> transparency would
20160811 01:20:03< vultraz> I *think* he said some money was going to AA>
20160811 01:20:09< vultraz> But I cannot remember exactly.
20160811 01:20:34< vultraz> as far as I know, that's it
20160811 01:21:21< vultraz> I am curious how our budget can handle the sheer amount of art being made for AA
20160811 01:21:32< vultraz> WHich leads me to suspect there might be another money pool somewhere
20160811 01:21:33< Aginor> that sounds like this project is funding his other hobby project
20160811 01:21:34< vultraz> But I have no idea.
20160811 01:21:38< iceiceice> vultraz, iirc he considers work on anura engine to be R & D for wesnoth
20160811 01:21:50< iceiceice> vultraz, also there was a tweet sent out by jetrel on frogatto channel
20160811 01:21:59< iceiceice> where he discussed frogatto revenue
20160811 01:22:17< celmin> Emphasis on "without telling anyone"
20160811 01:22:18< vultraz> As far as I know (from him), none of the frogatto devs are going to get paid.
20160811 01:22:21< iceiceice> i dont know what other revenue wesnoth inc. has besides frogatto and wesnoth
20160811 01:22:31< shadowm> Anyway, basically what needs to happen here is that the development team sends Dave/Wesnoth Inc a message, as a concerted effort, including all your questions and demands, in order to actually move things forward since communication originating from their end is clearly not their forte.
20160811 01:22:45< shadowm> A letter, I mean.
20160811 01:22:48< vultraz> Honestly
20160811 01:22:57< vultraz> If you look at it objectively, it is suspicious :P
20160811 01:23:15 * Aginor strongly agrees with shadowm 
20160811 01:23:21-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160811 01:23:52< vultraz> Work by multiple people on connected projects whose profits go into mysterious pools with only one or two people privy to the info regarding their expenditure or exact numbers.
20160811 01:24:10< vultraz> So yes, I second that motion
20160811 01:24:16< vultraz> We should draft a letter.
20160811 01:24:24< celmin> Then let's do it.
20160811 01:24:24< shadowm> That way we can hopefully put an end to the rumors and make informed decisions about our continuance in the project OR our ties to Wesnoth Inc.
20160811 01:24:27< vultraz> Emphasis on "we"
20160811 01:26:00< shadowm> This is relevant for the team that maintains wesnoth.org as well, since Wesnoth Inc pays for our hosting.
20160811 01:32:50-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160811 01:40:29< vultraz> blah, running out of space
20160811 01:41:20< celmin> Space?
20160811 01:42:03< vultraz> for the configure options
20160811 01:42:14< vultraz> I still haven't added a modifications list somewhere :|
20160811 01:46:38< celmin> So uhh, whose brilliant idea was it to extend std::vector… :|
20160811 01:46:59< celmin> vultraz: I think keeping it as two separate screens like currently is actually better.
20160811 01:47:10< vultraz> why?
20160811 01:47:34-!- RatArmy [~RatArmy@om126212249091.14.openmobile.ne.jp] has joined #wesnoth-dev
20160811 01:47:41< celmin> It makes sense to think of "selecting your scenario" and "choosing settings" as two separate operations, to me.
20160811 01:48:07< iceiceice> celmin, i didnt know someone extended std::vector in wesnoth
20160811 01:48:17< iceiceice> they made a specialization of it?
20160811 01:48:30< celmin> iceiceice: src/pathfind/pathfind.hpp:93
20160811 01:48:54< celmin> I don't really know what the point of this is. They seem to have reimplemented lower_bound?
20160811 01:49:29< celmin> I guess it's private inheritance, at least.
20160811 01:49:46< iceiceice> yeah thats kind of wierd
20160811 01:50:06< iceiceice> i mean they surely could have used composition instead of inheritance
20160811 01:50:11< iceiceice> but probably they didnt want to type out all the fields?
20160811 01:50:19< iceiceice> if its private inheritance then i dont konw why they did that
20160811 01:50:26< celmin> No ida.
20160811 01:50:27< iceiceice> celmin, i think that is public though
20160811 01:50:28< celmin> ida
20160811 01:50:29< iceiceice> because it is struct
20160811 01:50:35< celmin> Oh, it might be...
20160811 01:50:49< celmin> Does struct's "default public" apply also to inheritance?
20160811 01:50:52< iceiceice> i think so
20160811 01:51:15< iceiceice> most of my template code i wrote in the last year would not work otherwise i think :)
20160811 01:52:12< celmin> Okay, so I was hoping I could avoid it, but it looks like that's not an option. I need to implement a non-const all_children_iterator.
20160811 01:53:06< celmin> I'll do that separately after I finish with the bulk of const_cast stuff, because there's a mass rename involved too.
20160811 02:04:28< celmin> Whoops, I accidentally used reset instead of checkout.
20160811 02:04:50-!- celmin [~celticmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The minstrel departs, to spread the music to the masses!]
20160811 02:06:29< vultraz> Ok, I think I could add two tabs
20160811 02:07:20< celticminstrel> Fortunately I was using git add -p, so I think I can recover my work easily.
20160811 02:07:37< celticminstrel> vultraz: I was going to suggest something like that actually.
20160811 02:08:37< vultraz> celticminstrel: current layout https://drive.google.com/file/d/0B-mR9s8FduLLbWtxVGh5MnljMkE/view?usp=sharing
20160811 02:09:00< celticminstrel> I'll look after I push the initial const_cast cleanup.
20160811 02:10:13< vultraz> (anyone else welcome to comment)
20160811 02:10:36-!- louis94 [~~louis94@91.178.241.79] has quit [Ping timeout: 250 seconds]
20160811 02:12:03< shadowm> The text in the description is missing some padding so that it's not directly adjacent to the scrollbar.
20160811 02:13:28< shadowm> The vertical separator dividing the screen in half-ish is shorter than the box immediately to its left.
20160811 02:14:35< shadowm> The "Debug OOS" option looks just as mysterious as before.
20160811 02:15:08< shadowm> The "Load Game" entry probably could use an ellipsis so it's clear it does something more.
20160811 02:15:30< vultraz> I might make that a button again
20160811 02:15:42< vultraz> And probably hide Debug OOS outside of debug mode.
20160811 02:15:49< shadowm> The Players/Map Size/Era labels should probably be aligned column-wise.
20160811 02:16:17< shadowm> I think gfgtdf wanted "Debug OOS" to be visible at all times. I don't remember what it does, though.
20160811 02:16:44< shadowm> I believe it just causes the game to make stricter synchronization checks.
20160811 02:16:51< vultraz> Agreed RE alignment of labels
20160811 02:17:09< shadowm> Either way, the core of the issue is that "debug" and "OOS" are both insider terms.
20160811 02:17:24< shadowm> Someone who's new to the game or computers in general doesn't know what either thing means.
20160811 02:17:47< vultraz> Right now I'm thinking of moving the Details area into a tab, with a modifications list as the other tab
20160811 02:17:56< celticminstrel> Actually, I think I'll go downstairs again. I only really came up so that I could easily view both monitors simultaneously in order to recover from reset --hard.
20160811 02:18:34< vultraz> Era and Set Password should probably go in the mods tab, too
20160811 02:19:47< vultraz> (can we once again acknowledge how awesome shadowm is for making tabs possible)
20160811 02:21:08< shadowm> What I'm saying doesn't sound like a big thing but it may cause confusion. "Is this a cheat option?" "I'm having issues with the game ((with an aspect unrelated to gamestate sync)), surely checking this option that I don't know what it does will help somehow." "I'm going to start a thread in the forums asking what this is without bothering to use the Search function." "Someone's saying they ...
20160811 02:21:14< shadowm> ... checked the debug OOS option when hosting, they must be cheating!"
20160811 02:21:17< shadowm> Etcetera.
20160811 02:22:15-!- RatArmy [~RatArmy@om126212249091.14.openmobile.ne.jp] has quit [Ping timeout: 244 seconds]
20160811 02:22:24< shadowm> Is it really necessary to have the Filter box require its own row to span the entire left half of the screen.
20160811 02:22:34< shadowm> s/\./?/
20160811 02:22:35< vultraz> how about "Enable Stricter Out-of-sync Checks"
20160811 02:22:38< vultraz> and no, it is not
20160811 02:22:46< shadowm> Scratch that, it spans both halves.
20160811 02:23:00-!- celmin [~celticmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20160811 02:23:14< shadowm> Well, the wording is good but again I don't remember for sure what it does.
20160811 02:23:32< celmin> I find it really weird how search fields in different apps synchronize.
20160811 02:23:41< shadowm> I'd be more inclined towards "Strict Synchronization".
20160811 02:24:02< shadowm> The reason being that "Out-of-sync" translates to a rather long phrase in Spanish.
20160811 02:24:19< vultraz> WIll use that
20160811 02:24:24< celmin> For example, I search for something in XCode, then switch to TextWrangler and discover that its search field has changed to the same value.
20160811 02:24:57< shadowm> "Activar chequeos de fuera-de-sincronización más estrictos"
20160811 02:25:10< shadowm> "Activar chequeos de sincronización más estrictos"
20160811 02:25:19< vultraz> Jeez
20160811 02:25:32< shadowm> Furthermore, "Enable" next to a checkbox is more than a bit redundant.
20160811 02:26:29< shadowm> (Also, perhaps a better translation for "OOS" to Spanish would be "desincronización", but then the sentence above falls into the dreaded can-can territory.)
20160811 02:26:31< celmin> Do you need más?
20160811 02:26:37< shadowm> Yes.
20160811 02:26:54< shadowm> Unless the English version makes the adjective not a comparative.
20160811 02:27:30< celmin> Well, yeah, I'd translate "sincronización más estrictos" as "Stricter synchronization".
20160811 02:28:22< celmin> vultraz: Looks nice, but that's not gonna fit.
20160811 02:28:26< shadowm> I was translating vultraz's original, which did have the comparative.
20160811 02:28:32< celmin> Pretty sure.
20160811 02:28:52< shadowm> My proposal would simply be "Sincronización estricta".
20160811 02:28:57< celmin> shadowm: True, it did; I was looking at your substitute suggestion - ei "Strict Synchronization"
20160811 02:29:07< celmin> ^ie
20160811 02:29:28< celmin> vultraz: Please make sure it fits in 800x600 or whatever our smallest resolution is.
20160811 02:29:49< celmin> It'd be acceptable to use the layout you posted at higher resolutions, though.
20160811 02:30:00< celmin> And split it into tabs on lower resolutions.
20160811 02:30:36< celmin> (In theory that's probably possible by using multiple [resolution] tags, though I dunno how well that actually works since as far as I know no dialog actually does it.)
20160811 02:31:27 * celmin also agrees that "Debug OOS" is a terrible name. People will be asking "what the hell is an OOS".
20160811 02:33:28< shadowm> it means u r cheating noob gtfo
20160811 02:36:03< shadowm> I have actually heard of people being accused of cheating because they got OOS errors due to some random bug (or just mismatched add-on versions).
20160811 02:36:49< iceiceice> i also did play with people who were cheating :p
20160811 02:37:00< iceiceice> and they try to pretend teh OOS error can be ignored
20160811 02:37:20< iceiceice> after recruiting like 10 gryphon riders
20160811 02:37:34< celmin> Does that debug message pop up on both sides if you use debug create?
20160811 02:37:49< celmin> Or other debug commands.
20160811 02:38:00< shadowm> That's new to 1.13.x and yes, I think.
20160811 02:39:25< celmin> It'd be nice if it can be suppressed when there are no remote sides.
20160811 02:39:38< shadowm> But in 1.12.x and previous versions you should be able to spawn a million Yetis without raising suspicions as long as they don't interact with anything and nothing interacts with them.
20160811 02:39:56< celmin> Sounds impossible...
20160811 02:40:37< shadowm> It's not practical for anything other than map cheating, yes.
20160811 02:40:43< shadowm> And there are more efficient ways to do that.
20160811 02:41:01< celmin> So "clearing fog/shroud" doesn't count as interaction or something?
20160811 02:41:39< shadowm> Clearing fog didn't in 1.10.x. I just remembered that the implementation of fog changed in 1.11.x, so maybe it's synced nowadays.
20160811 02:41:59< shadowm> (I may have both version numbers wrong and it's actually 1.8.x and 1.9.x, respectively.)
20160811 02:56:51< celmin> iceiceice: Do you have any idea what the executeKey value in the Lua registry is for? It seems to just be set to the global debug.traceback, unless I'm missing something.
20160811 02:57:17< iceiceice> yeah
20160811 02:57:17< iceiceice> so
20160811 02:57:24< iceiceice> i do know :)
20160811 02:57:43< iceiceice> so if you didn't already, should read documentation for "lua_pcall"
20160811 02:58:04< iceiceice> the lua_pcall function executes a function call in a protected context
20160811 02:58:14< iceiceice> you are supposed to put function on the stack, followed by all its arguments
20160811 02:58:27< iceiceice> but also, if you want to use a custom error handler,
20160811 02:58:38< iceiceice> that needs to go on the stack also, somewhere beneath the function
20160811 02:58:47< iceiceice> and you pass the index to the error handler as the 4th argument
20160811 02:59:07< iceiceice> in wesnoth we just use the lua function debug.backtrace
20160811 02:59:23< iceiceice> if you wanted to use a different function for the error handler, you could change that
20160811 02:59:26< celmin> Okay, so this all makes sense, except for one little thing.
20160811 02:59:39< celmin> If we just use debug.backtrace, why does it need to be stored in the registry?
20160811 02:59:47< iceiceice> i think its like a caching thing
20160811 03:00:17< celmin> Hmm.
20160811 03:00:20< iceiceice> loading the debug lib and getting backtrace from its table is like, idk maybe 20-50 lua operations or something?
20160811 03:00:26< celmin> Well, I suppose it's one lookup instead of two.
20160811 03:00:28< vultraz> maybe I should keep Eras in a list?
20160811 03:00:35< vultraz> so descriptions can be shown
20160811 03:00:39< celmin> vultraz: Huh?
20160811 03:00:50< vultraz> celmin: I was going to use a combobox for eras
20160811 03:00:52< iceiceice> i see, i guess if debug lib is already cached then you could do it in two
20160811 03:00:53< vultraz> but eras have descriptions
20160811 03:01:12< vultraz> if they're in a list those descrips can be displatyed
20160811 03:01:14< vultraz> displayed
20160811 03:02:07< celmin> iceiceice: What I meant is that evaluating "debug.traceback" means "look up the key 'debug' in the _G table" followed by "look up the key 'traceback' in that table", whereas if it's in the registry it's "look up the key executeKey in the registry".
20160811 03:02:33< celmin> vultraz: You can still do descriptions with a combobox.
20160811 03:02:37< iceiceice> celmin, do we actually expose debug to the user scripts?
20160811 03:02:39< vultraz> how so?
20160811 03:02:45< iceiceice> i thought that we dont
20160811 03:03:07< celmin> vultraz: Update the description each time they select a different era.
20160811 03:03:10< iceiceice> i think OS and debug were considered unsafe for sandboxing reasons
20160811 03:03:17< vultraz> hm
20160811 03:03:20< vultraz> yes
20160811 03:03:21< celmin> iceiceice: Uhh, it seems to be available at the lua_kernel_base stage, let me check further.
20160811 03:04:06< celmin> iceiceice: Ah, looks like it allows debug.traceback and debug.getinfo only.
20160811 03:04:11< iceiceice> i see
20160811 03:04:21< iceiceice> it would still be a little more robust probably to use the registry
20160811 03:04:37< iceiceice> because if the user for whatever reason did something else with debug table in _G
20160811 03:04:44< iceiceice> then i guess all wesnoth's execute commands would fail
20160811 03:04:49< celmin> Makes sense.
20160811 03:04:56< celmin> If they assigned something else to debug.traceback.
20160811 03:05:12< celmin> Okay.
20160811 03:06:11< celmin> lua_insert(L, -2 - nArgs);
20160811 03:06:37< celmin> …oh right, that means 2 from the top if nArgs is 0.
20160811 03:06:44< celmin> Okay.
20160811 03:07:12< celmin> I feel like this is the same function twice...
20160811 03:08:06< celmin> I guess they differ in their response to errors...
20160811 03:08:17< celmin> (luaW_pcall and lua_kernel_base::protected_call)
20160811 03:08:50< iceiceice> yeah i think they basically do the same thing
20160811 03:09:04< iceiceice> unless it changed
20160811 03:09:31< celmin> The first few lines are identical, they only differ in how they report errors if lua_pcall reported failure.
20160811 03:09:50< iceiceice> maybe not
20160811 03:09:55< iceiceice> yeah i see
20160811 03:10:02< iceiceice> it is vaguely coming back to me :)
20160811 03:10:11< celmin> One is missing a lua_remove. Not sure if that matters.
20160811 03:12:04< iceiceice> hmm
20160811 03:12:20< iceiceice> i'm looking at this again,
20160811 03:12:24< iceiceice> https://github.com/wesnoth/wesnoth/blob/master/src/scripting/lua_kernel_base.cpp#L509
20160811 03:12:37< iceiceice> i wonder if the "jailbreak::rethrow()" should clean up the stack before it throws
20160811 03:12:50< celmin> So, lua_remove before rethrow?
20160811 03:13:10< celmin> Seems like it wouldn't hurt to do so.
20160811 03:13:21< iceiceice> yeah i agree about that
20160811 03:13:28< iceiceice> you might also leak any return values i guess
20160811 03:14:28< iceiceice> i think the post condition of the function is that any return values are supposed to end up on the stack,
20160811 03:14:37< iceiceice> i dont remember if the error message is supposed to get popped also
20160811 03:14:41< celmin> I could lua_pop(L, nRets), unless nRets is variable...
20160811 03:14:43< iceiceice> i guess it does that
20160811 03:14:52< celmin> I dunno.
20160811 03:15:30< iceiceice> it might not matter
20160811 03:15:44< iceiceice> i dont remember what actual exceptions are still thrown through lua
20160811 03:16:07< iceiceice> celmin, you could do something like "settop(error_handler_index - 1)" or something
20160811 03:16:29< iceiceice> but it would have to be like
20160811 03:16:36< celmin> Only on failure, right?
20160811 03:16:48< iceiceice> try { jailbreak::rethrow() } catch (...) { settop(...); rethrow; }
20160811 03:16:53< iceiceice> which is a bit ugly
20160811 03:16:56< iceiceice> idk if its worth it
20160811 03:17:03< celmin> I think I'll pass for now.
20160811 03:18:43< iceiceice> i guess its not even clear when that would matter
20160811 03:18:54< iceiceice> the jailbreak mechanism is suposed to stash the exception and raise a lua error
20160811 03:19:07< iceiceice> so you are supposed to get a lua error back from the pcall, unless the user lua trapped it
20160811 03:56:00-!- hk238 [~kvirc@t224.ip7.netikka.fi] has joined #wesnoth-dev
20160811 04:19:23-!- hk238 [~kvirc@t224.ip7.netikka.fi] has quit [Quit: http://www.kvirc.net/ 4.9.1 Aria]
20160811 04:40:04< celmin> Okay, so if I were to override operator new to take a lua_State*, what would the corresponding operator delete do? 
20160811 04:49:57< celmin> Uhhh… what does lua_newuserdata(L, 0) do?
20160811 04:50:35-!- Netsplit *.net <-> *.split quits: matthiaskrgr
20160811 04:51:05< celmin> Hmm, I guess it's really valid…
20160811 04:52:00< celmin> (It's used for wesnoth.game_config and wesnoth.current.)
20160811 04:53:14< celmin> Oh, interesting, wesnoth.textdomain stores the textdomain's name as the userdata.
20160811 05:01:49-!- Netsplit over, joins: matthiaskrgr
20160811 05:07:45-!- JyrkiVesterinen [~JyrkiVest@87-100-213-159.bb.dnainternet.fi] has joined #wesnoth-dev
20160811 05:11:34-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20160811 05:20:45< vultraz> huh, I actually have too little MUCH space now
20160811 05:22:52< celmin> Heh.
20160811 05:24:16< vultraz> I added a seperate tab for mods
20160811 05:24:18< vultraz> seperate 
20160811 05:24:23< vultraz> separate 
20160811 05:24:29< vultraz> buttttt
20160811 05:24:42< vultraz> you're not even likely to have mods installed, sometimes
20160811 05:25:12< vultraz> so what do I fill the rest of this area with :|
20160811 05:26:40< vultraz> i could put era
20160811 05:26:41< vultraz> set password..
20160811 05:27:19-!- RatArmy [~RatArmy@om126212249091.14.openmobile.ne.jp] has joined #wesnoth-dev
20160811 05:27:28-!- Ravana_ [~Ravana@unaffiliated/ravana/x-2327071] has joined #wesnoth-dev
20160811 05:28:58< vultraz> but that's not enough
20160811 05:32:13-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160811 05:36:42-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 265 seconds]
20160811 05:36:46-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20160811 05:38:25< vultraz> blah, what to do..
20160811 05:38:26-!- mjs-de [~mjs-de@x4db5a58f.dyn.telefonica.de] has quit [Remote host closed the connection]
20160811 05:38:38< celmin> Run around in circles.
20160811 05:38:49< vultraz> I need room for a potentially long mods list
20160811 05:39:09< vultraz> or
20160811 05:39:17< vultraz> i could implement that dropdown with checkboxs thing
20160811 05:39:57< JyrkiVesterinen> 20160810 23:47:39< vultraz> I hope jyrk...whathisname does more stuff.
20160811 05:40:03< JyrkiVesterinen> Yes, that't the plan.
20160811 05:40:15< vultraz> :D
20160811 05:40:25< JyrkiVesterinen> From the start, I decided to help this project by doing something every day.
20160811 05:40:36< JyrkiVesterinen> I chose to focus on endurance, not speed.
20160811 05:40:59< JyrkiVesterinen> I'm afraid that I might run out of motivation if I tried to move much faster.
20160811 05:41:01< celmin> So, with the overloaded new operator I get a crash in lua_cpp_function...
20160811 05:41:16< JyrkiVesterinen> I have worked this way in my earlier one-man open source projects, too.
20160811 05:43:02< celmin> Though I think it's not specifically the fault of the overload.
20160811 05:45:07-!- TheJJ_ [~rofl@ipbcc36896.dynamic.kabel-deutschland.de] has joined #wesnoth-dev
20160811 05:47:30< vultraz> ugh
20160811 05:47:33< vultraz> what to DO
20160811 05:48:11< celmin> I have no idea
20160811 05:48:33< vultraz> let's see..
20160811 05:48:39-!- Elsi [~Elsi@luwin.ulrar.net] has quit [Ping timeout: 250 seconds]
20160811 05:48:39-!- TheJJ [~rofl@ipbcc36896.dynamic.kabel-deutschland.de] has quit [Ping timeout: 250 seconds]
20160811 05:48:41< vultraz> I could merge the Set Password field too
20160811 05:49:11-!- Elsi [~Elsi@luwin.ulrar.net] has joined #wesnoth-dev
20160811 05:50:03< celmin> Okay, so yeah, the crash wasn't specifically caused by the overloaded new operator, just by one specific use of it. Hmm...
20160811 05:51:16< vultraz> maybe i should include a chat area after all
20160811 05:51:24< celmin> new(lua_newuserdata(L, sizeof(lua_function)) lua_function(f); luaL_setmetatable(L, cpp_function);
20160811 05:51:25< celmin> vs
20160811 05:52:08< celmin> void* p = lua_newuserdata(L, sizeof(lua_function)); luaL_setmetatable(L, cpp_function); new(p) lua_function(f);
20160811 05:55:22< vultraz> void pointer?
20160811 05:55:55< celmin> There must be some semantic difference between them that I'm missing?
20160811 05:56:07< celmin> Unless it's something about the change from that to new(L)
20160811 05:56:07< vultraz> ask iceiceice 
20160811 05:56:16< celmin> Yes void pointer, what about it?
20160811 05:57:07< vultraz> nothing, I just dont see those often
20160811 05:57:12< celmin> Oh.
20160811 05:57:41< celmin> Well, it's what lua_newuserdata returns, though rarely in Wesnoth is that actually assigned to a variable.
20160811 05:57:52< iceiceice> i guess i dont see a major difference in those two lines
20160811 05:58:24< celmin> Let's see if it crashes with normal placement new, as shown above, rather than with new(L)...
20160811 05:58:31< celmin> It doesn't. Hmm.
20160811 05:58:37< iceiceice> celmin, i think i read something that, the secon is preferred?
20160811 05:58:43< iceiceice> but only in some like obscure case
20160811 05:58:46< celmin> Huh?
20160811 05:59:04< iceiceice> so the only time it would make a difference is if you are concerned about `new(p) lua_functoin(f)` throwing an exception
20160811 05:59:09< celmin> Ah.
20160811 05:59:21< iceiceice> but in wesnoth i dont think you should worry about htat
20160811 05:59:24< celmin> Yeah.
20160811 05:59:28< iceiceice> i mean thats like, memory allocation failure or something
20160811 05:59:59< celmin> new(L) should be exactly equivalent… was the crash just a fluke?
20160811 06:00:37< celmin> I don't think std::function throws exceptions on construction anyway.
20160811 06:00:39< iceiceice> i dont know, i never tried overloading operator new
20160811 06:06:32< celmin> Hmm, well, I guess I can just leave that bit of code to call lua_newuserdata directly...
20160811 06:09:18< celmin> Maybe it's something about std::function.
20160811 06:09:27< celmin> (Unlikely though that seems.)
20160811 06:33:44< celmin> Weird that lua_function is the only one that causes problems with custom new.
20160811 06:34:12< celmin> I tested almost every other case without problems.
20160811 06:37:45-!- Kwandulin [~Miranda@p200300760F35BF0564437385458EA484.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160811 06:47:27-!- celticminstrel is now known as celmin|sleep
20160811 06:49:45-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has joined #wesnoth-dev
20160811 06:52:27-!- boucman_work [~boucman@193.56.60.161] has joined #wesnoth-dev
20160811 06:57:49-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has joined #wesnoth-dev
20160811 07:01:43< vultraz> what does the Random Faction: Independent/No Mirror/No Ally Mirror option do?
20160811 07:05:50< vultraz> sounds like something needed more descriptive names
20160811 07:09:15< vultraz> celmin, shadowm: https://drive.google.com/file/d/0B-mR9s8FduLLSHd2U3JTRWRGVkU/view?usp=sharing and https://drive.google.com/file/d/0B-mR9s8FduLLQlVzcTM0X0FQRTA/view?usp=sharing
20160811 07:09:59 * vultraz is committing this before he loses it
20160811 07:11:13-!- irker010 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160811 07:11:13< irker010> wesnoth: Charles Dang wesnoth:master 8651dabfe9b1 / / (4 files in 3 dirs): MP Create: major progress on layout of GUI2 MP Create dialog https://github.com/wesnoth/wesnoth/commit/8651dabfe9b1db4a5f498f290e6564f2f45b5980
20160811 07:18:34< Kwandulin> Looks like the changes to the interface are coming along really nicely
20160811 07:18:48< vultraz> indeed
20160811 07:20:22-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160811 07:20:32-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160811 07:24:32-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20160811 07:26:44-!- Kwandulin_2 [~Miranda@p200300760F35BFBD64437385458EA484.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160811 07:29:02-!- Kwandulin [~Miranda@p200300760F35BF0564437385458EA484.dip0.t-ipconnect.de] has quit [Ping timeout: 258 seconds]
20160811 07:45:23-!- travis-ci [~travis-ci@ec2-50-19-32-6.compute-1.amazonaws.com] has joined #wesnoth-dev
20160811 07:45:24< travis-ci> wesnoth/wesnoth#10303 (master - 8651dab : Charles Dang): The build has errored.
20160811 07:45:24< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/151428644
20160811 07:45:24-!- travis-ci [~travis-ci@ec2-50-19-32-6.compute-1.amazonaws.com] has left #wesnoth-dev []
20160811 08:01:14< wedge009> vultraz: Regarding conversion to GUI2 - it sounds like you're a little burned out. Maybe take a break? Even if only for a day. As someone who doesn't know *anything* about the details of GUI1 vs GUI2, the progress I see purely as a player from the builds on master, the changes to the dialogues have been largely positive even if only from an aesthetic sense.
20160811 08:01:49< vultraz> The process involves writing copious amounts of WML
20160811 08:01:55< vultraz> hundreds of lines 
20160811 08:02:08< wedge009> If GUI2 has problems of its own, well, it seems to me the commitment is there to move things to it. And by your own admission it seems to be an improvement over GUI1.
20160811 08:02:13< vultraz> (though there usually is a space between each line)
20160811 08:02:18< wedge009> Is this a process that can be taught easily?
20160811 08:02:33< wedge009> Perhaps if I knew what to do, I could help.
20160811 08:02:43< vultraz> Not particularly
20160811 08:04:08< wedge009> As for the Wesnoth company issue... I honestly don't know anything about it other than what I've heard here. So I don't really have an opinion on it other than the apparent lack of transparency does seem to be a problem that needs to be addressed. Not necessarily right this very minute or by you specifically.
20160811 08:04:23< wedge009> Oh, well, I'll try to keep watch on what you've been doing, then.
20160811 08:04:43< vultraz> any critiques on the mp create layout?
20160811 08:04:57< wedge009> Is that a change from today?
20160811 08:05:19 * wedge009 just got back from work maybe 10 minutes ago
20160811 08:05:30< vultraz> yes, I just committed it
20160811 08:05:34< vultraz> it's not functional
20160811 08:05:40< vultraz> so you have to check the screenshots
20160811 08:05:51< vultraz> https://drive.google.com/file/d/0B-mR9s8FduLLSHd2U3JTRWRGVkU/view?usp=sharing and https://drive.google.com/file/d/0B-mR9s8FduLLQlVzcTM0X0FQRTA/view?usp=sharing
20160811 08:06:00 * wedge009 looks
20160811 08:16:10< wedge009> vultraz: As best as I can tell, we've sacrificed the chat and lobby functions and squeezed the three screens into one. Aesthetically, I think it looks nice - an improvement in my opinion. I don't use the chat stuff often, but I imagine having that missing would be bad - is this just a design concept?
20160811 08:17:00< wedge009> I should restart - been putting off an update restart from yesterday, plus some weird stuff is happening in my Wesnoth. Unless we've removed the buggy 'experimental MP interface' for both 1.12 and 1.13?
20160811 08:17:16< vultraz> well, when I say it's not functional I meant it's not activated in-game yet, but it is a dialog you can access 
20160811 08:17:21< vultraz> if you edit a line in the code 
20160811 08:17:27< vultraz> so no, it's not a mock-up
20160811 08:17:30< vultraz> it's the new layout
20160811 08:17:59< vultraz> also, it only merged 2 screens, not 3
20160811 08:18:04< wedge009> I think it should work well.
20160811 08:18:16< vultraz> the third, the waiting for players one, will have a chat area again
20160811 08:18:50< wedge009> Depends if people want to talk about the game options while they're setting up.
20160811 08:20:25< vultraz> well, as you can see there's some empty space there
20160811 08:20:34< vultraz> I might add a small chat area
20160811 08:25:04-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Read error: Connection reset by peer]
20160811 08:33:11-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160811 08:33:31< wedge009> Well that was timely. I crashed out instead of restarting.
20160811 08:35:08-!- JyrkiVesterinen [~JyrkiVest@87-100-213-159.bb.dnainternet.fi] has quit [Quit: .]
20160811 08:36:28< vultraz> converting stuff to gui2 is mostly tedious 
20160811 08:36:36< vultraz> when gui2 cooperates 
20160811 08:36:43< vultraz> frustrating when it doesnt
20160811 08:38:32< wedge009> What things are still on GUI1?
20160811 08:39:41-!- nore [~ncourant@sas.eleves.ens.fr] has joined #wesnoth-dev
20160811 08:39:58< vultraz> Statistics, File browser, Addons Manager (already completed the layout), Help (impossible to convert currently), MP Create (working on it), Mp lobby (needs gui2 bugfixes), and right click and theme menus.
20160811 08:40:06< vultraz> also, themes themselves
20160811 08:40:13< vultraz> but that's a whole different kettle of fish.
20160811 08:41:24< vultraz> menus already have the required framework, just need some tweaks and wiring together
20160811 08:41:25< wedge009> Help is complex, but handy and generally seems okay for the time being.
20160811 08:41:51< vultraz> the file browser is on hold pending deep, dark gui2 event handling improvements
20160811 08:41:52< wedge009> If/when GUI2 is done, I'll be someone will say let's convert it to something else! XD
20160811 08:42:01< vultraz> wedge009: everyone already is :P
20160811 08:42:05< wedge009> Heh.
20160811 08:42:08< wedge009> Darn.
20160811 08:42:25< wedge009> Oh, the renderpath redo?
20160811 08:42:47-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20160811 08:44:06< vultraz> not that
20160811 08:44:36< vultraz> it's a chicken-egg problem that would require careful study of gui2's internals to fix.
20160811 08:44:50< vultraz> shadowm actually has the prototype of the dialog, but for that issue cannot finish it.
20160811 08:45:12< vultraz> out of all the remaining things, Statistics and menus are fairly simple
20160811 08:45:15< vultraz> (I think)
20160811 08:45:23< vultraz> oh, and Attack Predictions
20160811 08:45:27< vultraz> that's  gui1..
20160811 08:47:20< wedge009> Damage graphs?
20160811 08:47:39< vultraz> yes
20160811 08:47:44< vultraz> that's the problem :|
20160811 08:48:58< wedge009> Wait, damage graphs are a problem in converting to GUI2? Why? Or the problem is in not knowing?
20160811 08:49:27-!- Kwandulin_2 [~Miranda@p200300760F35BFBD64437385458EA484.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160811 08:50:16< vultraz> yes, the damage graphs might be hard to draw in gui2
20160811 08:50:21< vultraz> but I do have some idea of what to do.
20160811 08:50:42< vultraz> dunno if it'll work
20160811 08:51:51< zookeeper> vultraz, you have a funny definition of "develop" :p
20160811 08:52:07< vultraz> what?
20160811 08:53:10< zookeeper> if by that definition i don't develop anything. although if you just meant currently right now, then fine.
20160811 08:54:08< vultraz> you're working on utbs
20160811 08:54:10< vultraz> maybe
20160811 08:54:16< vultraz> but that's not on the engine
20160811 08:56:44< zookeeper> yes, "development" doesn't exclusively mean "development of the engine"
20160811 08:57:59-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has quit [Quit: ChipmunkV]
20160811 09:00:40-!- RatArmy [~RatArmy@om126212249091.14.openmobile.ne.jp] has quit [Ping timeout: 250 seconds]
20160811 09:03:33-!- vincent_c [~bip@vcheng.org] has quit [Quit: Coyote finally caught me]
20160811 09:04:06-!- vincent_c [~bip@vcheng.org] has joined #wesnoth-dev
20160811 09:05:24< nore> Hi, I'm considering trying to add a way for several players to control a side together (so, both players would be able to control units, attack, etc); however, since I don't know the codebase at all yet, I was wondering how difficult this would be, and if done, if it would be considered for merging or if it is an unwanted feature
20160811 09:05:39-!- horrowind [~Icedove@2a02:810a:83c0:404:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20160811 09:06:11< nore> I'm aware that there would be problems such as concurrency that would need to be solved one way or another as well
20160811 09:06:24 * Aginor passes that one on to gfgtdf
20160811 09:06:56< Aginor> nore: hang around and someone will answer you, eventually :)
20160811 09:07:05< nore> but well, do you consider this doable, given some time, or I should just dismiss the idea and give up? :)
20160811 09:07:43< Aginor> I myself is slightly sceptical, but I don't really touch any of the multiplayer stuff, so I can't really judge
20160811 09:07:57< Aginor> what's the overall usecase you're trying to satisfy by doing this?
20160811 09:07:59< nore> Aginor: yeah, I know that, I am myself a core developper or another game and I often see people leave before someone even noticed their message ;)
20160811 09:08:30< nore> usecase for me is being able to play singleplayer campaigns with a friend
20160811 09:08:48-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160811 09:09:10< nore> so discussing stategy, etc while seeing the map for both of us
20160811 09:10:17< wedge009> vultraz: I managed to confirm https://gna.org/bugs/?23734 after all. It shouldn't be closed... yet.
20160811 09:10:30< zookeeper> i don't know how hard it'd be technically, but considering how our animations block input and unit actions things happen strictly sequentially, it's hard to see how in practise responsiveness wouldn't become a problem.
20160811 09:10:40< zookeeper> -things
20160811 09:11:08< vultraz> wedge009: well that will go away if i get this converted to gui2
20160811 09:11:10< vultraz> so
20160811 09:11:15< wedge009> That was my thought.
20160811 09:11:41< nore> yeah, the concurrency problem is an important one, I have yet to think more about that
20160811 09:11:52< zookeeper> i mean, if you perform a move, but just after the other player performed another move, does your move still go through after the other player's move has been done and displayed? it probably shouldn't, because their move very often would affect your plans
20160811 09:12:01< Aginor> wedge009: I did some initial work with fixing it in gui1 but I think it rotted again
20160811 09:12:04< zookeeper> in which case, you'd constantly end up having your moves canceled
20160811 09:12:09< zookeeper> (or so i'd expect)
20160811 09:12:17< vultraz> Aginor: no point fixing if it's going to gui2
20160811 09:12:27< Aginor> vultraz: this was much earlier this year
20160811 09:12:32< vultraz> yes
20160811 09:12:33< wedge009> Multiple players on a single-player game... interesting concept. But from what little I know, it does sound difficult to implement.
20160811 09:12:41< wedge009> Aginor: It's okay.
20160811 09:12:42< Aginor> when I was fixing all resize bugs
20160811 09:12:52< vultraz> i remember
20160811 09:12:52< zookeeper> nore, random thought as an alternative: let the players take action one after another: player 1 can do one move, player 2 can do one move, player 1 can do one move, ...
20160811 09:13:19< nore> or maybe have some way to give control to a specific player
20160811 09:13:25-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 252 seconds]
20160811 09:13:43< nore> so you could share the control as you want
20160811 09:13:57< nore> like 1 turn each, or 1 scenario each, etc
20160811 09:14:15< nore> this should solve the concurrency problems
20160811 09:15:20< zookeeper> i don't think it's fundamentally (again, from a non-technical perspective) an unwanted feature, i'm just not sure what the best form for it would be.
20160811 09:16:39< nore> ok, I will try to look in more detail then
20160811 09:17:46< nore> it looks like it will be a bit easier than I thought if we get rid of the concurrency problems with solutions such as yours
20160811 09:18:59< zookeeper> nore, if the planning mode was better then one way might be to limit the concurrency to that (as in, players can only do plans simultaneously). maybe not as convenient in casual co-op though.
20160811 09:19:04 * wedge009 forgot that 'experimental MP interface' only applies to non-local MP lobbies.
20160811 09:19:44< nore> though, we want moves that can be undone to be shown to other players controlling the same side, so it's not that simple
20160811 09:19:53< nore> wedge009: what is this?
20160811 09:20:34< wedge009> Something I mentioned earlier, I thought the experimental interface wasn't working any more but I was only looking at the local MP game lobby.
20160811 09:20:45< wedge009> Before you joined.
20160811 09:20:50< nore> oh, ok
20160811 09:24:06< zookeeper> nore, oh right. undo is a big problem.
20160811 09:24:58< nore> well, I guess the moves could be shown to players controlling the same side only, but that means they would have to be sent as well
20160811 09:25:49< nore> so each move that would be sent would need to have flag saying whether it can be undone (though the server could maybe know that)
20160811 09:26:09< zookeeper> nore, and even in my idea of alternative pseudo-turns would basically eliminate undo (you'd never be able to undo multiple moves, no matter how innocuous).
20160811 09:26:21-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has joined #wesnoth-dev
20160811 09:26:26< zookeeper> -in
20160811 09:27:12< nore> yeah, we don't really want that either, I'd say
20160811 09:28:37-!- Kwandulin [~Miranda@p200300760F35BFBDB4268AF7BE0035EA.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160811 09:29:20< nore> well, we *could* keep what would currently be done if the other player was an observer (i.e., don't know undoable moves), and then change that later, so that players controlling the same side can see these moves
20160811 09:30:03< zookeeper> vultraz, WRT the game creation screen: if it lacks the lobby chat box, then the absolute requirement would be that whatever chat you miss while in that screen will be there when you proceed or cancel. it shouldn't be possible for someone to inadvertently miss some lobby chat due to them setting up a game.
20160811 09:30:35< nore> (oh, btw: I noticed some time ago that undo could be exploited to get some information on the map, though this is probably known and marked as "won't fix" given the other problems that would appear if such moves were prevented from being undone)
20160811 09:30:51< zookeeper> you mean polling for ambushers?
20160811 09:31:09< nore> yes, and for enemy units with fog as well
20160811 09:31:29< zookeeper> with fog? how does that work
20160811 09:31:41< zookeeper> you can't undo if you uncover any fog
20160811 09:31:50< nore> oh, has that changed?
20160811 09:31:55< zookeeper> that's always been the case
20160811 09:32:04< zookeeper> don't tell me there's a bug like that :P
20160811 09:32:16< nore> I thought you could undo your moves if it didn't show any enemy unit
20160811 09:32:52< nore> but I remember that from at least 5 years ago, so maybe the bug has been fixed or I just have a bad memory
20160811 09:33:16< zookeeper> yeah if you remember correctly, then it must have been a bug
20160811 09:33:29< nore> in any case, you can poll for ambushers anyway
20160811 09:33:32< zookeeper> the point has always been to invalidate undo if uncovering any fog
20160811 09:34:14< zookeeper> the ambushers issue is real, but IIRC there's not really anything that can be done about that (which wouldn't be a huge inconvenience otherwise)
20160811 09:35:52< nore> well, about the fog thing, I just tested, and if there has ever been a bug, it works correctly now
20160811 09:36:15< nore> about the ambushers thing, yes, that was what I though
20160811 09:37:04< nore> +t
20160811 09:45:55< wedge009> vultraz: Did you say you'd finished the Status dialogue conversion to GUI2? Or are you still working on it? Looks like it's not giving enough space to show all the teams. Looks good, otherwise.
20160811 09:46:32< vultraz> wedge009: it's done, but what you see is https://gna.org/bugs/?24780
20160811 09:46:47< vultraz> someone NEEDS to look into that\
20160811 09:47:13< wedge009> vultraz: Right, I thought it might be.
20160811 09:50:04< wedge009> Actually, it's a problem no matter how many teams there are. It seems to be just too short to show the last team properly. Though the root causes may well be the same, I don't know.
20160811 09:50:13-!- JyrkiVesterinen [~JyrkiVest@87-100-213-159.bb.dnainternet.fi] has joined #wesnoth-dev
20160811 09:51:14< vultraz> it is
20160811 10:11:17-!- irker010 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160811 10:22:53-!- RatArmy [~RatArmy@om126212249091.14.openmobile.ne.jp] has joined #wesnoth-dev
20160811 10:38:30-!- fabi [~fabi@176.4.66.117] has joined #wesnoth-dev
20160811 10:41:44-!- fabi__ [~fabi@176.5.34.179] has quit [Ping timeout: 260 seconds]
20160811 10:57:08-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160811 11:01:50-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 258 seconds]
20160811 11:06:11-!- molt [~molt@dynamic-213-198-235-143.adsl.eunet.rs] has quit [Quit: Leaving]
20160811 11:09:01-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160811 11:14:45-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160811 11:30:34-!- gfgtdf [~chatzilla@x4e368369.dyn.telefonica.de] has joined #wesnoth-dev
20160811 11:30:43< gfgtdf> 20160811 09:05:24< nore> Hi, I'm considering trying to add a way for several players to control a side together (so, both players would be able to control units, attack, etc); however, since I don't know the codebase at all yet, I was wondering how difficult this would be, and if done, if it would be considered for merging or if it is an unwanted feature
20160811 11:32:26< gfgtdf> nore: i think this woudl be very hard, for exmple it can happen that both players try to move the same unit at the same time, so you need to think about ways on how to prevent this.
20160811 11:33:31< gfgtdf> nore: the point is that often moves are sensed to the server after the glient knows that it cannot undone them, so its not really posible to 'revert'an impossible move in sucha  case.
20160811 11:35:09-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160811 11:36:25-!- RatArmy [~RatArmy@om126212249091.14.openmobile.ne.jp] has quit [Ping timeout: 252 seconds]
20160811 11:37:11< zookeeper> just brainstorming for fun: perhaps one way would be to pass the control between the players, but not between each move, but based on a timer. player 1 gets 10 seconds of playtime, then player 2 gets 10 seconds, then player 1... undo wouldn't be affected, and there wouldn't be much concurrency issues.
20160811 11:37:12< DeFender1031> gfgtdf, seems like that would require real-time move processing, which is, as far as I understand, contrary to the style of game wesnoth is.
20160811 11:43:05< zookeeper> or, possibly, players would just have a yield button which would let the other player have a go.
20160811 11:44:26< zookeeper> actually, that would probably be the simplest way to implement it, and it'd allow the players themselves maximal control over it
20160811 11:45:32 * zookeeper suggests that
20160811 11:53:11< zookeeper> it wouldn't allow strictly simultaneous play, but the players would be free to decide how frequently they yield. maybe some players would prefer to do it constantly, and maybe others would prefer to each control about half the units and yield only once or twice per turn.
20160811 12:01:21-!- RatArmy [~RatArmy@om126212249091.14.openmobile.ne.jp] has joined #wesnoth-dev
20160811 12:01:45< gfgtdf> zookeeper: hmm yes i think that would work.
20160811 12:01:48< gfgtdf> nore: ^
20160811 12:10:08-!- horrowind [~Icedove@2a02:810a:83c0:404:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20160811 12:16:49< nore> gfgtdf: yeah, that was I was going to try, I just first need to finish cloning the repo (which is *huge*)
20160811 12:18:07-!- louis94 [~~louis94@91.178.242.169] has joined #wesnoth-dev
20160811 12:18:26-!- RatArmy [~RatArmy@om126212249091.14.openmobile.ne.jp] has quit [Ping timeout: 265 seconds]
20160811 12:18:57< nore> oh, btw, a question I asked a few days ago on #wesnoth, maybe someone here will know the answer better: Am I right saying that in multiplayer campaigns, victory events are not synced for observers? If so, wouldn't that cause OOS for said observers?
20160811 12:24:39< gfgtdf> nore: with the changes in 1.13.5 the are synced. THere is now a new 'local victory' event in case you really need the unsynced victory events.
20160811 12:25:07< nore> that's interesting, I didn't know that
20160811 12:25:27< nore> so, which remaining events are unsynced?
20160811 12:25:36-!- RatArmy [~RatArmy@om126212249091.14.openmobile.ne.jp] has joined #wesnoth-dev
20160811 12:26:59< gfgtdf> nore: well  preload, sleect, aiturn, local victory, localdefeat
20160811 12:27:49< nore> ok, thanks
20160811 12:34:32< iceiceice> celmin, randomly was looking at an unrelated thing in regards to lua,
20160811 12:34:43< iceiceice> http://stackoverflow.com/questions/32046828/lua-userdata-alignment
20160811 12:34:47< iceiceice> http://www.lua.org/source/5.3/llimits.h.html#L_Umaxalign
20160811 12:34:57< iceiceice> i wonder if your thing was crashing because of an alignment issue?
20160811 12:35:08< iceiceice> i never actually saw that before when i used lua
20160811 12:35:15< iceiceice> actually i dont know why they dont just set it to "maxalign_t" or whatever
20160811 12:35:22< iceiceice> but maybe thats not available in C
20160811 12:36:46< iceiceice> i guess maybe it should have been crashing before also if there was an alignment problem
20160811 12:41:00-!- RatArmy [~RatArmy@om126212249091.14.openmobile.ne.jp] has quit [Quit: Konversation terminated!]
20160811 12:45:28-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160811 12:45:44< celmin> vultraz: So about your latest screenshot, it's actually rather problematic. Given what wedge009 said about bug 23734, I'm seriously questioning the choice of merging the two stages of MP create, though the bug admittedly could be solved by duplicating a portion of MP create in a new dialog. Additionally, your MP create takes up too much space - I'm pretty sure it won't fit in an 800x600 window (though I consider that a minor problem 
20160811 12:45:46< celmin> since I think it's possible to use a different layout on smaller resolutions). Finally, what happened to the era and modification selection? I don't see it at all.
20160811 12:46:04< celmin> gfgtdf: Do you have to advertise the hacky "local victory" event? :|
20160811 12:46:51< gfgtdf> celmin: uhm i didn tfeel like i was advertiing it
20160811 12:47:09< celmin> Mentioning it then
20160811 12:48:53< celmin> iceiceice: I have no idea if that could possibly be the problem...
20160811 12:48:54-!- Kwandulin [~Miranda@p200300760F35BFBDB4268AF7BE0035EA.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160811 12:49:28< celmin> Seems kinda unlikely though since the problem disappeared when using regular placement new rather than custom placement new...
20160811 12:50:36-!- celmin|sleep is now known as celticminstrel
20160811 12:52:03-!- Bonobo [~Bonobo@2001:44b8:254:3200:688c:4cc:c6ff:ef62] has quit [Ping timeout: 264 seconds]
20160811 12:52:08< gfgtdf> vultraz: is there aready code for the gui2 mp configure dialog ? spwcialyl about the handlin of [options] ?
20160811 12:52:24-!- Bonobo [~Bonobo@2001:44b8:254:3200:688c:4cc:c6ff:ef62] has joined #wesnoth-dev
20160811 12:52:33-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20160811 12:54:05-!- Kwandulin [~Miranda@p200300760F35BFBD40527093C4168585.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160811 12:55:57-!- ideuler [~textual@0.213.62.94.rev.vodafone.pt] has joined #wesnoth-dev
20160811 13:21:25-!- Kwandulin [~Miranda@p200300760F35BFBD40527093C4168585.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160811 13:22:12< zookeeper> Espreon, want to double-check some text? here would be some: https://github.com/wesnoth/wesnoth/pull/743
20160811 13:24:12-!- horrowind [~Icedove@2a02:810a:83c0:404:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20160811 13:24:59-!- hk238 [~kvirc@t224.ip7.netikka.fi] has joined #wesnoth-dev
20160811 13:25:43< iceiceice> celmin, agree
20160811 13:32:32< celmin> Hmm. When pushing a list of units, does it really save much to not re-push the metatable for each one?
20160811 13:37:11< celmin> Hmm. There seems to be a distinction between "expected unit, got zzz" and "unit not found"...
20160811 13:37:52< zookeeper> urgh. MOVE_UNIT doesn't have check_passability= so it defaults to yes. since it's used for scripted movement, i think check_passability=no needs to be added. there might be more cases like https://forums.wesnoth.org/viewtopic.php?p=600086#p600086
20160811 13:38:17< celmin> The former means you passed something that's not even a unit proxy; the latter could probably happen if you passed an invalid unit proxy (one with u.valid = nil).
20160811 13:38:25< celmin> I wonder if this is an important distinction.
20160811 13:39:43< gfgtdf> celmin: i'd say it is, sicne the both error indicated usualla have quite distice sources
20160811 13:40:07-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has joined #wesnoth-dev
20160811 13:40:07< gfgtdf> zookeeper: since its a trap (accorign t the forum post) i wonder wher a enter_hex event might be more fitting ?
20160811 13:40:32< zookeeper> wouldn't make any difference AFAICT
20160811 13:40:47< zookeeper> since none of the units that can trigger the trap could walk through that hex
20160811 13:41:09< zookeeper> also, traps like that are very dubious in the first place so i'm inclined to just remove it
20160811 13:41:10< celmin> Then the distinction should probably be coded into luaW_checkunit.
20160811 13:41:55< celmin> So… luaW_tounit could return nullptr for three reasons...
20160811 13:52:30-!- coproduit [~coproduit@hott.coq.sexy] has joined #wesnoth-dev
20160811 14:01:50< celmin> Hmm, using luaW_checkunit means I lose access to lua_unit::on_recall_list...
20160811 14:51:58-!- louis94 [~~louis94@91.178.242.169] has quit [Quit: Konversation terminated!]
20160811 14:52:23-!- louis94 [~~louis94@91.178.242.169] has joined #wesnoth-dev
20160811 14:57:01-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160811 15:05:22< gfgtdf> do we have a sp mod type in the addon manager already?
20160811 15:06:00< celmin> No? Mods don't seem to have MP/SP distinction.
20160811 15:07:40< gfgtdf> the wiki lists only mod_mp, which somehow seems unfitting for mods that are designed mainyl for sp
20160811 15:09:27-!- Bonobo [~Bonobo@2001:44b8:254:3200:688c:4cc:c6ff:ef62] has quit [Ping timeout: 264 seconds]
20160811 15:11:55< zookeeper> shikadibot: seen bumbadadabum
20160811 15:12:08< zookeeper> (...i figured that might not work anymore)
20160811 15:12:20< celmin> BTW, has anyone tested the asio campaignd?
20160811 15:12:54< celmin> Other than (I assume) loonycyborg 
20160811 15:13:16< gfgtdf> vultraz, loonycyborg: can i add a sp mod addon type now or would that cinflics with your current work ?
20160811 15:13:54< celmin> gfgtdf: Does it really make sense to distinguish SP/MP mods?
20160811 15:14:32< gfgtdf> celmin: so your suggestion is just mking  the mp_mod catagotes a more genreal mod catagors?
20160811 15:14:45< celmin> Possibly.
20160811 15:15:45< celmin> But note that that was a question to which I don't know the answer.
20160811 15:15:55< zookeeper> better to add new ones
20160811 15:16:58-!- Kwandulin [~Miranda@p200300760F35BFBDD454BB521CFE1723.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160811 15:17:50< zookeeper> good grief... why is it always bumdadadabum? https://forums.wesnoth.org/viewtopic.php?f=4&t=44445
20160811 15:18:14< celmin> Huh?
20160811 15:18:40-!- ideuler [~textual@0.213.62.94.rev.vodafone.pt] has quit [Quit: Chakalaka.]
20160811 15:19:27< gfgtdf> celmin: well form looking at mp modiication on the 1.12.6 addon server thate are quite some that wouldnt make sense in mp, for example i wrote a 'check same version' mod that allows you to see the wesnoth version used by other players, i also see an 'expaendable leaders' mod that makes another unit your leader when your leader dies which woudl mostlikely break most sp campaigns too.
20160811 15:19:53< celmin> I guess you mean ones that wouldn't make sense in single-player.
20160811 15:19:57< gfgtdf> wouldnt make sense in sp i mean
20160811 15:20:39< gfgtdf> or 'no mirror mod'
20160811 15:21:04< celmin> No mirror? That sounds obsolete.
20160811 15:21:54-!- irker443 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160811 15:21:54< irker443> wesnoth: ln-zookeeper wesnoth:master 087a2822d16b / changelog data/campaigns/Eastern_Invasion/lua/bandits.lua: Fixed broken bandits code https://github.com/wesnoth/wesnoth/commit/087a2822d16b8eb129a271e4a6e66edd51a5be6f
20160811 15:22:50-!- JyrkiVesterinen [~JyrkiVest@87-100-213-159.bb.dnainternet.fi] has quit [Quit: .]
20160811 15:23:01< gfgtdf> zookeeper: you could change that code to just use wesnoth.random(3,4) and wesnoth.random(2, rand1)
20160811 15:23:03< Espreon> zookeeper: Sure, why not? (The Sergeant's will need some work. In particular, I don't like the use of pronouns therein)
20160811 15:23:19< zookeeper> gfgtdf, i don't care about these pointless lua conversions
20160811 15:23:52< Espreon> (Well, the way they're being used, I meant)
20160811 15:24:00< zookeeper> Espreon, great
20160811 15:25:01< celmin> They all seemed fine to me, so I didn't say anything.
20160811 15:25:16< Espreon> Eh... no...
20160811 15:25:36< celmin> I have no problem with the use of singular they, BTW.
20160811 15:25:52< Espreon> It sounds weird to me.
20160811 15:25:57< celmin> However, sergeant units are always male, so it would make sense to change it to use he.
20160811 15:26:03< Espreon> Exactly.
20160811 15:26:39< celmin> Singular they has been used for centuries in English, so it's pretty much part of the language by now, as much as some people try to deny it.
20160811 15:27:00< celmin> Mind you… the unit description kinda is a plura context...
20160811 15:27:11< celmin> Except for the first sentence.
20160811 15:27:31< Espreon> Eh, there are some places where it sounds wrong to me.
20160811 15:27:34< celmin> So if you changed the first "they" to "Sergeants" it would work better, without altering the other pronouns.
20160811 15:27:39< celmin> ^plural
20160811 15:27:43< Espreon> ... like if the speaker actually knows the gender of the person in question.
20160811 15:28:32< Espreon> I tend to think it's only 100% OK if it refers to things like "someone" or "anyone".
20160811 15:28:39< celmin> Espreon: I've seen people claim that the usage of singular they is limited in ways similar to that. I'm not sure how true that is in practice. It's admittedly a bit odd to use it if you know the gender, but I don't mind it.
20160811 15:29:00< Espreon> It feels weird to me, so... meh.
20160811 15:29:15< Espreon> I'll have to take a look and think about it.
20160811 15:29:43-!- ancestral [~ancestral@21.sub-174-219-130.myvzw.com] has joined #wesnoth-dev
20160811 15:30:08< celmin> So basically there's three options: pluralize first sentence; replace first "they" with noun; use he/his instead of they/their.
20160811 15:30:20< Espreon> I would just pluralize it.
20160811 15:30:52< Espreon> Using a definite singular like that already feels funny in such a context.
20160811 15:32:40< Espreon> "The Sergeant is a low-ranking officer in the ranks of an army" ... please, get real
20160811 15:33:25< celmin> Using "an army" is also a bit odd.
20160811 15:33:31< Espreon> Yup.
20160811 15:34:48< Espreon> I'm not sure if I like how the Royal Warrior's description narrows it down to kings and princes, but eh... what to do?
20160811 15:34:59< zookeeper> gfgtdf, actually this helper.rand(3,4) doesn't seem to cause any problems in 1.13.4, but does in 1.13.5. do you happen to know what's changed?
20160811 15:35:03< Espreon> The way it stands in that request is better than "The royal warriors ..."
20160811 15:35:18< Espreon> Oh, it was already like that.
20160811 15:35:20< Espreon> Never mind, then.
20160811 15:35:22< celmin> Isn't the Royal Warrior the one that has a comment saying it's meant just as a base unit for specific princes and such?
20160811 15:35:33< Espreon> Yeah.
20160811 15:35:44< Espreon> I didn't notice that it's already narrowed down to kings and princes.
20160811 15:35:57< gfgtdf> zookeeper: bumbadabum move dthe helper.rang implementation completeley to lua, so that it dpoesnt accept ints or tstrings anymore
20160811 15:36:16< celmin> tstrings? What?
20160811 15:36:23< celmin> That one's easily fixed, at least...
20160811 15:36:24< gfgtdf> zookeeper: but even in 1.12 i think helper.rand(3,4)  would just be teh same as helper.rand("3") 
20160811 15:36:28< zookeeper> hilarious. so it's his fault after all :>
20160811 15:37:17< gfgtdf> zookeeper: the second argument beeing just ignored i mean.
20160811 15:37:25< celmin> Looking at the old code, it seems gfgtdf is right.
20160811 15:37:31< Espreon> Too bad his laptop is damaged.
20160811 15:37:44-!- horrowind [~Icedove@2a02:810a:83c0:404:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20160811 15:38:22< zookeeper> gfgtdf, right
20160811 15:38:38< celmin> So it's his fault, but also helped to uncover a previously silent bug.
20160811 15:39:31< Espreon> Whoever wrote these seems to have a tendency of having coordinated subjects agree with singular verbs. "Their resilience and skill ... makes" rather than "Their resilience and skill ... make" ... oh well.
20160811 15:40:16< zookeeper> whatever you say :p
20160811 15:40:30< celmin> Hmm.
20160811 15:40:38< celmin> I can't decide which one sounds more correct.
20160811 15:41:01< Espreon> I would lean towards the latter.
20160811 15:41:11< zookeeper> you can add whatever revisions you feel are warranted as comments to the PR or just clearly point them out to me here, whichever you prefer
20160811 15:41:19< Espreon> I know it seemed funny when I found something semblant in my younger brother's speech.
20160811 15:41:31< Espreon> Right.
20160811 15:41:45< Espreon> I'm probably just going to modify them myself anyway, so...
20160811 15:42:17< celmin> BTW, what was the reason for asking Espreon specifically on this?
20160811 15:42:22< celmin> (Just curious)
20160811 15:42:42< celmin> Also, Google says Espreon is right on the verb agreement, though the other still sounds okay to me.
20160811 15:45:35< Espreon> OK, it starts going downhill with the Triton's.
20160811 15:45:37< zookeeper> because AFAIK he likes to obsess over this kind of stuff
20160811 15:45:37< Espreon> Gah.
20160811 15:45:43< Espreon> Not really.
20160811 15:45:52 * Espreon hates writing and proofreading
20160811 15:45:53< Espreon> :D
20160811 15:46:11< celmin> I feel like your words ring hollow.
20160811 15:46:31< Espreon> Oh well.
20160811 15:46:32< zookeeper> well, if you feel an irresistible pull to correct mistakes in writing even though you hate it then that's just as good if you ask me :>
20160811 15:46:49< Espreon> Well, I mean, if it's for the good of Wesnoth, then why the hell not?
20160811 15:47:46< celmin> Hmm, what does getmetatable return for a tstring...
20160811 15:47:46< zookeeper> it's for the glory of all the people of the west-north
20160811 15:47:47-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160811 15:47:49< zookeeper> charge!
20160811 15:47:58< Espreon> Haha.
20160811 15:48:10< Espreon> Good God... the Grunt's is a monstrosity.
20160811 15:48:12< Espreon> Yup...
20160811 15:48:25< Espreon> I'm definitely going to have to just modify these...
20160811 15:48:39< celmin> Ugh, seems to be "translatable string"?
20160811 15:49:45-!- ancestral [~ancestral@21.sub-174-219-130.myvzw.com] has quit [Quit: i go nstuf kthxbai]
20160811 15:50:01< celmin> Well, helper.rand could easily be changed to allow translatable strings, but maybe it's best not to encourage that. (If someone really needs to pass one, they can call tostring first.0
20160811 15:50:03< celmin> ^)
20160811 15:51:18< Espreon> zookeeper: Please poke me about this tomorrow. I'm going to try to work on these tonight, but... yeah...
20160811 15:53:07< celmin> What's so monstrous about the grunt?
20160811 15:54:01< Espreon> So many things strung together with commas.
20160811 15:54:14< Espreon> I think it could be a bit much...
20160811 15:54:24-!- louis94 [~~louis94@91.178.242.169] has quit [Ping timeout: 260 seconds]
20160811 15:54:37< celmin> Hmm. I suppose the first sentence, at least, could be improved.
20160811 15:54:50< celmin> Though it doesn't seem all that bad to me, really.
20160811 15:55:42< Espreon> Perhaps "monstrosity" was a bit harsh...
20160811 15:56:20< Espreon> I don't really like the flow of it, for some reason.
20160811 15:57:48-!- travis-ci [~travis-ci@ec2-54-82-73-21.compute-1.amazonaws.com] has joined #wesnoth-dev
20160811 15:57:49< travis-ci> wesnoth/wesnoth#10305 (master - 087a282 : ln-zookeeper): The build failed.
20160811 15:57:49< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/151540981
20160811 15:57:49-!- travis-ci [~travis-ci@ec2-54-82-73-21.compute-1.amazonaws.com] has left #wesnoth-dev []
20160811 15:59:19-!- boucman_work [~boucman@193.56.60.161] has quit [Ping timeout: 265 seconds]
20160811 15:59:56-!- hk238 [~kvirc@t224.ip7.netikka.fi] has quit [Quit: http://www.kvirc.net/ 4.9.1 Aria]
20160811 16:04:34< zookeeper> Espreon, sure
20160811 16:06:52-!- hk238 [~kvirc@t224.ip7.netikka.fi] has joined #wesnoth-dev
20160811 16:08:43-!- JyrkiVesterinen [~JyrkiVest@87-100-213-159.bb.dnainternet.fi] has joined #wesnoth-dev
20160811 16:10:42< Espreon> Thanks.
20160811 16:32:52-!- midzer [~quassel@p57B4551A.dip0.t-ipconnect.de] has quit [Ping timeout: 252 seconds]
20160811 16:32:58-!- fabi_ [~fabi@176.7.89.74] has joined #wesnoth-dev
20160811 16:35:20-!- fabi [~fabi@176.4.66.117] has quit [Ping timeout: 258 seconds]
20160811 16:44:21-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit []
20160811 16:47:13-!- louis94 [~~louis94@91.178.242.169] has joined #wesnoth-dev
20160811 16:50:20-!- fabi_ [~fabi@176.7.89.74] has quit [Quit: Konversation terminated!]
20160811 16:50:37-!- mjs-de [~mjs-de@x4e31469b.dyn.telefonica.de] has joined #wesnoth-dev
20160811 16:50:50-!- fabi [~fabi@176.7.89.74] has joined #wesnoth-dev
20160811 17:00:58-!- louis94 [~~louis94@91.178.242.169] has quit [Ping timeout: 258 seconds]
20160811 17:04:01< loonycyborg> gfgtdf: celmin: shadowm has also tested asio campaignd
20160811 17:04:29< celmin> Is there any particular reason why it hasn't been merged yet?
20160811 17:04:36< loonycyborg> It's unlikely that adding new addon types gonna result in conflicts
20160811 17:05:16< loonycyborg> yes, shadowm needs to vet it thoroughly
20160811 17:05:33< loonycyborg> because he's original author of the codebase :P
20160811 17:05:44< loonycyborg> or at least a major contributor
20160811 17:06:58< celmin> I was asking because I want it merged, not because gfgtdf was suggesting a new category.
20160811 17:08:55-!- Kwandulin [~Miranda@p200300760F35BFBDD454BB521CFE1723.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160811 17:23:07< JyrkiVesterinen> Heads up: https://github.com/wesnoth/wesnoth/pull/723#issuecomment-239229650
20160811 17:23:31< JyrkiVesterinen> In other words, you can't rely on overridden virtual methods being marked with the override specifier.
20160811 17:23:55< JyrkiVesterinen> Since not all of them are marked (unlike we thought after merging PR723).
20160811 17:27:41< gfgtdf> JyrkiVesterinen: since that pr oynl changes 9 files id actuall be rather supoirsed if that foud all cases of missing override
20160811 17:27:59< gfgtdf> JyrkiVesterinen: unless someone ddi wokr on addon override before.
20160811 17:28:02< gfgtdf> else*
20160811 17:28:59< JyrkiVesterinen> Yes, I also thought that the PR is way too small. I merged it anyway because it was still an improvement.
20160811 17:29:45< JyrkiVesterinen> And now we know for sure that PoignardAzur's g++ missed some missing override specifiers.
20160811 17:52:36-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has quit [Quit: ChipmunkV]
20160811 17:57:11< vultraz> gfgtdf: I don't handle [option]s
20160811 17:57:20< celmin> vultraz: You need to.
20160811 17:57:28< vultraz> yes
20160811 17:57:42< vultraz> someone point me to some example of what the hell they're supposed to look like
20160811 17:57:51-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has joined #wesnoth-dev
20160811 17:58:02< celmin> I think it's kinda like advanced preferences.
20160811 17:58:20< celmin> Based on the contents of the [option] you add various controls to the window.
20160811 17:58:41< vultraz> screenshot would be nice
20160811 17:59:12< vultraz> ive never seen an example, ever
20160811 17:59:15< celmin> Oh sorry, it really is [options] rather than [option].
20160811 17:59:22< celmin> You can easily make an example.
20160811 17:59:27< gfgtdf> vultraz: i think orcia ageless has some [options]
20160811 17:59:30< celmin> http://wiki.wesnoth.org/OptionWML
20160811 17:59:43< gfgtdf> vultraz: you coudl start a orcia survival map in 1.12.6
20160811 18:00:04< celmin> So it's really kinda similar to how advanced preferences work.
20160811 18:00:13< vultraz> ugh
20160811 18:00:40< gfgtdf> vultraz: dark foecast mianline map has also [options] but it contains just one slider
20160811 18:00:51< celmin> BTW vultraz, what happened to era/mod selection?
20160811 18:00:53< gfgtdf> vultraz: in .13 thta is
20160811 18:00:58< gfgtdf> 1.13*
20160811 18:01:08< vultraz> celmin: what about it
20160811 18:01:17< celmin> I didn't see it in your screenshot.
20160811 18:01:22< vultraz> it's there
20160811 18:01:30< vultraz> there are two screenshots
20160811 18:01:56< celmin> What?
20160811 18:03:58< vultraz> ugh
20160811 18:04:02< vultraz> ugh ugh
20160811 18:04:06< vultraz> this will be ugly :|
20160811 18:04:15< celmin> About as ugly as advanced prefs?
20160811 18:04:37< vultraz> while I'm proud of what I did with advanced prefs, it's in no way elegant.
20160811 18:06:05< vultraz> well, guess that's what the empty space will be for
20160811 18:06:48< celmin> I suppose the way GUI2 works you'll need something like a listbox… since there's no such thing as a "sequence layout manager"...
20160811 18:06:58< celmin> Only grids.
20160811 18:07:16< vultraz> yes
20160811 18:07:21< vultraz> I know how to do it
20160811 18:07:41< celmin> Well, I suppose the listbox is sort of a sequence layout manager.
20160811 18:07:52< gfgtdf> celmin: we  you still have treeviews which are somhow like more gnertic listboxes that allow you to put diffenet types of rows in it.
20160811 18:08:04< gfgtdf> generic*
20160811 18:08:11< celmin> Or treeviews...
20160811 18:08:27< celmin> Wait, are treeviews really treeviews?
20160811 18:08:48< vultraz> I hate treeviews
20160811 18:08:49< gfgtdf> celmin: how do you mean ?
20160811 18:08:56< vultraz> they're so bloody ugly.
20160811 18:09:10< gfgtdf> vultraz: how are they ugly ?
20160811 18:09:16< celmin> An example of what I think of when I see/hear "treeview" would be the Windows Explorer sidebar.
20160811 18:09:32< gfgtdf> celmin: yes they are designed for such cases
20160811 18:09:41< celmin> You have a list of folders or whatever, and each has a disclosure triangle and so on.
20160811 18:09:46< vultraz> i guess objectively they're not THAT ugly, but I don't really like using them
20160811 18:09:58< celmin> Can you do that with a single treeview, or does it require nesting treeviews arbitrarily deep?
20160811 18:10:22< gfgtdf> celmin: a single treeview can do that.
20160811 18:10:34< vultraz> I feel the code interface is rather messy, though
20160811 18:10:36-!- louis94 [~~louis94@91.178.242.169] has joined #wesnoth-dev
20160811 18:10:52< celmin> I think it would be cool to convert the inspector dialog to use a treeview instead of a listbox pair, but I couldn't figure out how to do it.
20160811 18:12:12< gfgtdf> vultraz: hmm their lua iterface is simple enough, and i dont really think thwir  c++ interface is harder.
20160811 18:12:53< gfgtdf> vultraz: also not that you dont want selcetable rows here, so if you want to use listboxes you will need to do quiet soem changes to work around that.
20160811 18:13:17< vultraz> gfgtdf: I assume all I need to do is not use a toggle panel
20160811 18:13:25< vultraz> toggle panels aren't built-in to listboxes
20160811 18:14:30< vultraz> what I was thinking is an empty toggle panel, then construct the options, then swap the grid
20160811 18:14:42< vultraz> but I suppose I *could* use a treeview
20160811 18:14:51< vultraz> define 4 types of node
20160811 18:14:55< vultraz> then add each as necessary..
20160811 18:15:52< celmin> …hold on, could that be done for advanced prefs too?
20160811 18:16:11-!- hk238 [~kvirc@t224.ip7.netikka.fi] has quit [Quit: http://www.kvirc.net/ 4.9.1 Aria]
20160811 18:16:21< vultraz> the advanced prefs are inside a listbox
20160811 18:16:33< vultraz> dont really want a treeview inside a listbox :P
20160811 18:16:33-!- louis94 [~~louis94@91.178.242.169] has quit [Ping timeout: 240 seconds]
20160811 18:17:52< vultraz> unless you propose reworking that whole list with treeview nodes
20160811 18:18:04< gfgtdf> vultraz: i'd asume that the listbox will soiply refuse to load if you omit the togglepanel but maybe tis is lua specific behaviour
20160811 18:18:12< vultraz> wat :|
20160811 18:18:21< gfgtdf> vultraz: why how is it inisde a listobx ?
20160811 18:18:27< celmin> Huh?
20160811 18:18:35< celmin> Wouldn't the treeview replace the listbox in that case?
20160811 18:18:48< vultraz> celmin: that's what I was thinking
20160811 18:19:02< vultraz> i suppose i should experiment
20160811 18:19:09< celmin> That's what I meant, using a treeview instead of a listbox for prefs.
20160811 18:19:25< celmin> I forget, does the campaign list use a treeview at the moment?
20160811 18:19:46< gfgtdf> celmin: it does
20160811 18:19:58< gfgtdf> celmin: you can intall addons form tha addon server to see it
20160811 18:20:12< gfgtdf> celmin: the current solution is not that pretty though
20160811 18:21:39 * vultraz installs.. AGELESS ERA
20160811 18:21:57< celmin> Gasp.
20160811 18:21:58< gfgtdf> celmin: i think we shodul replace teh checkboxes the the campaign list that fite better to the style of the rest to the campaigsn list
20160811 18:22:07< celmin> Huh?
20160811 18:22:29-!- irker443 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160811 18:22:42< vultraz> gfgtdf: I think the campaign list mods section could be changed to a combobox if we added a version of dropdown with checkboxes
20160811 18:22:46< vultraz> which we need to do anyway
20160811 18:30:49< gfgtdf> celmin: this is how the campaign modification list currently looks like: https://s9.postimg.org/3pkp5crz3/modlist.png , i at lest for me it looks rather bad specailly becasue the checkboxes cannot be aligned matching to the campaign items above.
20160811 18:31:23< celmin> Oh hey, I see my campaign.
20160811 18:31:43< celmin> Ooh, sudden downpour.
20160811 18:32:03< gfgtdf> celmin: anther issue is thqt you currently cannot see the desciptions in sp, but that can mostlieey simply fixed with tooltips
20160811 18:39:02< gfgtdf> zookeeper: didnt you just fix https://gna.org/bugs/index.php?24951 ?
20160811 18:39:33< gfgtdf> zookeeper: from bugnumber sry, https://gna.org/bugs/index.php?24954
20160811 18:39:44< zookeeper> yeah you got me confused for a moment there
20160811 18:40:04< zookeeper> yes, yes i did. i didn't realize there was a report for it already.
20160811 18:49:22< gfgtdf> zookeeper: do you know which textdaomain the gettext parser chooses for heeader files ?
20160811 18:49:54< zookeeper> what, the c++ header files? no idea
20160811 18:50:45< gfgtdf> hmm ok
20160811 18:50:47< gfgtdf> Ivanovic: do you know which textdomain the gettext parser uses for header files?
20160811 18:51:38< celmin> gfgtdf: It's set by a #define, I believe.
20160811 18:51:47< celmin> Probably defaults to "wesnoth", but not sure.
20160811 18:52:19< celmin> You probably shouldn't use translatable strings in header files, really.
20160811 18:52:22< gfgtdf> celmin: hmm yes but headers can e included in multiple c files and its possibel that one has a #defeine A and another has a #define B so i wanst sure
20160811 18:52:31< celmin> If you do you should #undef the textdomain.
20160811 18:53:14< celmin> Why're you asking?
20160811 18:55:02< gfgtdf> celmin: i was invecstgating https://gna.org/bugs/?24887
20160811 18:55:34< Ravana_> notice that while gameplay is boring, you do not need ageless to test options with orocia
20160811 18:55:41< celmin> N_() is not a translatable string.
20160811 18:56:27< celmin> So you need to manually look it up, either by constructing a t_string with the correct textdomain, or by using the gettext wrapper function.
20160811 18:56:33< gfgtdf> celmin: i know
20160811 18:56:50< celmin> Well, reading the bug, it seemed like not doing that was a likely cause.
20160811 18:57:13< gfgtdf> celmin: i think it doesn that i notied in the bugreport where it shodul do that
20160811 18:57:50< gfgtdf> celmin: i think the problem my be more about gettext parder ignorng hreaders or soemthing liek that
20160811 19:03:28< celmin> I did say you probably shouldn't include translatable strings in headers...
20160811 19:04:29< gfgtdf> celmin: well i ddint write that code and im still not sure what teh source of that bu is
20160811 19:05:13< vultraz> what's the point of the [entry] option?
20160811 19:05:54< celmin> Text field, right?
20160811 19:06:05< gfgtdf> its combobox iirc
20160811 19:06:27< gfgtdf> vultraz: in LoW you can choose differnet scneario to start with, like starting at scenario 4
20160811 19:06:46< vultraz> there's a [combo] option
20160811 19:06:49< gfgtdf> vultraz: i personalyl am not convinced that its s useful since you can just load a stat-of-scenario save from scenario 4 if youwant
20160811 19:06:51< vultraz> [entry] seems like just text
20160811 19:07:04< vultraz> why would you want to just display text :|
20160811 19:07:26< gfgtdf> vultraz: ah rght i though you were talking about the campaign entry pont combobox in the mp configure screen
20160811 19:07:52< gfgtdf> vultraz: well i think you can enter text there
20160811 19:08:02< vultraz> ahh
20160811 19:08:10< vultraz> ok
20160811 19:09:15< vultraz> Ok I think I have an idea how to handle this
20160811 19:10:47< celmin> Well yeah, it's a text field.
20160811 19:10:49< celmin> Like I said.
20160811 19:13:30< JyrkiVesterinen> The initializer list of GUI1 textbox class is intended using one tab *and* five spaces. Weirdest indenting practice I have seen. :/
20160811 19:13:37< JyrkiVesterinen> Will change to just the tab.
20160811 19:14:47< vultraz> ahah
20160811 19:15:08< vultraz> ok, Ive decided to put options in their own tab
20160811 19:15:59< gfgtdf> vultraz: is it still an improvement over using 2 pages then like the gui1 dialog did ?
20160811 19:17:45< vultraz> anything is an improvement over gui1's handling of options: https://drive.google.com/file/d/0B-mR9s8FduLLbEpxN1A1THNrNTg/view?usp=sharing
20160811 19:17:47< vultraz> :|
20160811 19:18:07< vultraz> unless it's better in 1.13
20160811 19:18:23< vultraz> but still
20160811 19:19:00-!- irker830 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160811 19:19:00< irker830> wesnoth: Jyrki Vesterinen wesnoth:master 3d8c29bc299b / changelog players_changelog src/widgets/textbox.cpp src/widgets/textbox.hpp: Fix a stray ; character appearing in the command console https://github.com/wesnoth/wesnoth/commit/3d8c29bc299bacf58894638734a728e5550ee65b
20160811 19:20:17< vultraz> \ o /
20160811 19:20:29-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160811 19:23:35< Espreon> Thank God.
20160811 19:24:31< gfgtdf> loonycyborg: maybe you know wbaou the gettext c++ parser ?
20160811 19:25:04< loonycyborg> know what exactly?
20160811 19:25:12< vultraz> ok, yes, [entry] is a textbox
20160811 19:25:34< gfgtdf> loonycyborg: it seems to me like it doesn find translatable string in header files.
20160811 19:26:23< gfgtdf> loonycyborg: so i wonder whether this is something can can/should be fixed or not.
20160811 19:26:36< loonycyborg> hmm
20160811 19:26:50< loonycyborg> I'm not sure that translatable strings should even be in headers
20160811 19:27:14< loonycyborg> it might be hard to find appropriate textdomain for them
20160811 19:27:42< gfgtdf> loonycyborg: hmm just using the default textdomain should be fine i think.
20160811 19:31:17< loonycyborg> files that are scanned by xgettext are explicitly listed in po/domain/POTFILES.in
20160811 19:31:26< loonycyborg> and they don't seem to list any headers
20160811 19:32:30< gfgtdf> loonycyborg: hmm i cannot find a domain filter in po
20160811 19:32:39< gfgtdf> loonycyborg: https://github.com/wesnoth/wesnoth/tree/master/po
20160811 19:33:14< loonycyborg> actually I meant po/$DOMAIN/POTFILES.in :P
20160811 19:34:44-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160811 19:34:50< loonycyborg> ah yes it's generated file
20160811 19:36:12< loonycyborg> gfgtdf: https://github.com/wesnoth/wesnoth/blob/master/po/SConscript#L35
20160811 19:36:49< celmin> I said twice that translatable strings probably wouldn't be in header files, but it seems like gfgtdf isn't listening.
20160811 19:37:16< celmin> Instead of making gettext parse header files, why don't you move those strings out of the header?
20160811 19:37:41< gfgtdf> celmin: becasue they are used in multiple files, (but onyl one files trnallates them)
20160811 19:38:11< celmin> That's not an excuse.
20160811 19:38:22< gfgtdf> celmin: you can also try to fix it if you wan to
20160811 19:38:31< celmin> Do you know what header files?
20160811 19:38:36< celmin> ^which
20160811 19:38:56< gfgtdf> units/types.hpp and mp_game_settings.hpp
20160811 19:39:26< celmin> Hmm...
20160811 19:40:12< celmin> …what, this is utterly crazy...
20160811 19:41:01< celmin> Well anyway, I think it can probably be fixed by moving the strings to the source file. I can look into it after I've finished what I'm currently doing.
20160811 19:41:36< celmin> Though using them as enums is a bit...
20160811 19:41:36< vultraz> hmm
20160811 19:41:53< vultraz> gfgtdf: do you know how to get a list of custom options 
20160811 19:42:51< irker830> wesnoth: Jyrki Vesterinen wesnoth:master 8d1cb23f8288 / src/game_initialization/create_engine.cpp: MP create: fix UB if filter excludes all campaigns/scenarios/maps https://github.com/wesnoth/wesnoth/commit/8d1cb23f828815ac002748e794c08f1a6d2ae1bd
20160811 19:43:16< gfgtdf> vultraz: hmm i'd guess just cipoying the way the gui1 mo create did that
20160811 19:43:23< celmin> vultraz: Well, you could try using .get_child("options")?
20160811 19:43:24< vultraz> mp::options::manager has too much gui1 crap stuffed in
20160811 19:43:43< JyrkiVesterinen> So, when I fixed the ; bug just now, I wanted to verify that GUI1 text fields still work if in normal dialogs.
20160811 19:44:10< JyrkiVesterinen> I opened the first dialog that used one I could find: Multiplayer -> Local Game, and tried to type "asdf" into the Filter field.
20160811 19:44:16< gfgtdf> vultraz: hmm i see
20160811 19:44:19< JyrkiVesterinen> The game crashed when I typed the D.
20160811 19:44:26< vultraz> JyrkiVesterinen: "current_level_index_ = 0u;" ?
20160811 19:44:30< vultraz> confused
20160811 19:44:54< JyrkiVesterinen> The above commit fixes the crash, but the game will still happily proceed to the next screen even with all the scenarios filtered out.
20160811 19:45:05< JyrkiVesterinen> Sometimes the instability of Wesnoth amazes me...
20160811 19:45:09< celmin> vultraz: That's a numeric suffix.
20160811 19:45:12< vultraz> ah
20160811 19:45:25< vultraz> JyrkiVesterinen: well, I'm trying to port those screens to gui2 as we speak
20160811 19:45:33< celmin> It's really not necessary in most cases, but it guarantees the integer literal has a specific type.
20160811 19:45:39< celmin> In this case, unsigned int.
20160811 19:45:52< vultraz> iinteresting
20160811 19:46:09< celmin> It's definitely not necessary in that specific case.
20160811 19:46:12-!- louis94 [~~louis94@91.178.242.169] has joined #wesnoth-dev
20160811 19:46:33< gfgtdf> vultraz: hmm it seems like it just ready every [modification]/[era]/[campaign]/[multiplayer] in the game config and check whether it macthes the currently actived, and adds its options if thats the case
20160811 19:47:15< vultraz> gfgtdf: so should I be able to fetch the [options] child from the config?
20160811 19:47:29< gfgtdf> vultraz: maybe we shodul move that code somehow to configure_engine so that it can be used by both
20160811 19:47:36< gfgtdf> vultraz: hmm usually yes
20160811 19:48:12< vultraz> i already have access to the scenario config in the dialog
20160811 19:49:22< gfgtdf> vultraz: you need all tags that contain [options], in the worst case just use game_config_manager::get()->game_config()
20160811 19:49:47< vultraz> all tags that contains [options]?
20160811 19:50:03< gfgtdf> vultraz: and the something like foreach chils... fi (tag == "multiplayer" || tag = "campaign") && i_id_active (child["id"])
20160811 19:50:13< gfgtdf> vultraz: thats more or less what the gui1 optiosn do i think
20160811 19:50:18< vultraz> wow that's ugly
20160811 19:50:23< vultraz> I'll figure something out
20160811 19:51:13< vultraz> the dialog keeps a pointer to the config 
20160811 19:51:19< vultraz> of the current scenario
20160811 19:51:21< vultraz> selected
20160811 19:51:22< celmin> You need to combine [options] from all possible sources.
20160811 19:51:38< celmin> So also campaign, era, and any active modifications.
20160811 19:51:51< vultraz> from the *selected* ones, right?
20160811 19:52:27< vultraz> i assume only options for selected things are shown
20160811 19:52:34< gfgtdf> vultraz: yes, for some reason the savefiles currently also coatins options for the mods but thins is a differetn bu.
20160811 19:52:43< gfgtdf> vultraz: yes only options for selected thing are shown
20160811 19:52:46< vultraz> ok
20160811 19:52:50< celmin> Of course, that's why campaign/era were singular.
20160811 19:52:51< vultraz> first let me get the interface working
20160811 19:53:03< vultraz> then I'll worry about getting them from multiple sources
20160811 19:53:42< gfgtdf> celmin: does your campaign work in 1.13.5 ?
20160811 19:53:52< celmin> Probably, but I haven't tested it.
20160811 19:55:12-!- louis94 [~~louis94@91.178.242.169] has quit [Ping timeout: 276 seconds]
20160811 19:55:25-!- travis-ci [~travis-ci@ec2-54-91-149-250.compute-1.amazonaws.com] has joined #wesnoth-dev
20160811 19:55:26< travis-ci> wesnoth/wesnoth#10306 (master - 3d8c29b : Jyrki Vesterinen): The build is still failing.
20160811 19:55:26< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/151609086
20160811 19:55:26-!- travis-ci [~travis-ci@ec2-54-91-149-250.compute-1.amazonaws.com] has left #wesnoth-dev []
20160811 20:13:39-!- JyrkiVesterinen [~JyrkiVest@87-100-213-159.bb.dnainternet.fi] has quit [Quit: Going to bed]
20160811 20:20:40-!- travis-ci [~travis-ci@ec2-54-91-149-250.compute-1.amazonaws.com] has joined #wesnoth-dev
20160811 20:20:41< travis-ci> wesnoth/wesnoth#10307 (master - 8d1cb23 : Jyrki Vesterinen): The build has errored.
20160811 20:20:41< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/151615283
20160811 20:20:41-!- travis-ci [~travis-ci@ec2-54-91-149-250.compute-1.amazonaws.com] has left #wesnoth-dev []
20160811 20:26:42-!- mjs-de [~mjs-de@x4e31469b.dyn.telefonica.de] has quit [Remote host closed the connection]
20160811 20:29:19-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160811 20:32:37-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 258 seconds]
20160811 20:32:37-!- wedge010 is now known as wedge009
20160811 20:43:26-!- tad__ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev
20160811 20:43:39-!- fabi [~fabi@176.7.89.74] has quit [Ping timeout: 264 seconds]
20160811 20:46:37< tad__> This new status panel looks like a work in process. What is the intention of game vs scenario on it?
20160811 20:47:39 * celmin isn't sure what the status panel refers to.
20160811 20:48:29< tad__> Menu/"Status Table"
20160811 20:48:29-!- fabi [~fabi@176.7.89.74] has joined #wesnoth-dev
20160811 20:51:02< tad__> The old panel simply said "More -->" and "<--Back"
20160811 20:52:30< tad__> It makes me wonder if the idea is that each team should have a consistent ID throughout a campaign and the panel will show status info for a team defeated in a previous scenario.
20160811 20:57:56< tad__> Perhaps it simply needs better words chosen for the buttons?
20160811 20:58:02< zookeeper> (there's a new status panel?)
20160811 20:58:50< tad__> Under the Menu button, the "Status Table" menu item. alt+s
20160811 20:59:03-!- fabi [~fabi@176.7.89.74] has quit [Quit: Konversation terminated!]
20160811 21:00:26< zookeeper> looks the same as always in 1.13.5
20160811 21:00:32< zookeeper> to my eyes anyway
20160811 21:01:30< tad__> I think voltraz updated it in the past couple days. It says "Game" and "Scenario" now, no scroll bars, and the third team is chopped off mid-way.
20160811 21:01:44< zookeeper> right
20160811 21:01:49 * zookeeper knows nothing
20160811 21:02:49< tad__> When I saw it I thought "Game? I am supposed to re-align sides in all scenario so they're on a consistent team for an entire campaign?"
20160811 21:03:57< tad__> After playing with it I see he simply changed "More ->" to "Scenario" and "Back <-" to "Game" and left both buttons up all the time.
20160811 21:05:03< tad__> I'll ask voltraz about it next time I see  him, I guess.
20160811 21:06:13-!- louis94 [~~louis94@91.178.242.169] has joined #wesnoth-dev
20160811 21:07:59< gfgtdf> vultraz: i think the new statsutable needs a scrollbar too son, for examepl even with just 3 sides they dont fit all on the window
20160811 21:10:33< zookeeper> well that sure sounded a bit patronizing :>
20160811 21:13:17< tad__> Sorry. I'm assuming he knows about it since it looks like his change. You said you don't.
20160811 21:14:22< tad__> The "I guess" means I may forget all about it, or more changes may appear and all will be revealed, in time.
20160811 21:14:46< gfgtdf> too soon*
20160811 21:15:04< zookeeper> tad__, no i meant gfgtdf's typo :P
20160811 21:15:16< tad__> Ah.
20160811 21:16:03< zookeeper> tad__, anyway... i'm going through #729's commits atm
20160811 21:16:19< zookeeper> i'll probably get that done tomorrow :P
20160811 21:16:44< tad__> Like I know my commit numbers. Hang on while I look.
20160811 21:17:45< tad__> Ah, the errors part for HttT .. OK. Want me to hang if you have questions? I'm fiddling with AToTB in another window but can check in occassionally.
20160811 21:18:23< zookeeper> ok, well...
20160811 21:18:30< zookeeper> what was inconsistent about simyr not being intelligent?
20160811 21:19:20< tad__> It is not intelligent until he's created for side=1 .. just in case someone compares the end-game from one run to the next.
20160811 21:20:02< tad__> So I made him intelligent in the victory dance for side=2 so he's the same when you see him but don't get him.
20160811 21:20:34< zookeeper> oh right, i see
20160811 21:22:41< celmin> Victory dance?
20160811 21:23:35< tad__> When I get to making choose_level work, I move all player-visible stuff from "victory" to "victory dance" and fire the custom event just before [endlevel]
20160811 21:23:52< celmin> Huh? There's a victory dance event?
20160811 21:24:01< celmin> Or wait, I guess it's custom.
20160811 21:24:09< celmin> Right, okay.
20160811 21:24:14< tad__> I am only letting "victory" have the clean-up work like clearing variables, and such.
20160811 21:24:28< celmin> Feels slightly hacky somehow.
20160811 21:24:36< celmin> But I guess it's not that bad.
20160811 21:24:58< tad__> You remember thinking it might be nice to fast-forward over all the dialogs on next_level and choose_level .. well this does that
20160811 21:25:02< zookeeper> tad__, umm, i'm not sure if you're joking or not
20160811 21:25:11< tad__> Where?
20160811 21:25:20< zookeeper> about a victory dance event
20160811 21:26:14< tad__> Well, it's sort of a joke on the name. "Yeah! I won" now lets talk about it, move some ships around and do some things which can be skipped if we're fast-forwarding.
20160811 21:26:42< zookeeper> i don't mean the name
20160811 21:27:00< celmin> Sounds like zookeeper agrees with me about it being hacky. >_>
20160811 21:28:03< zookeeper> anyway, WRT https://github.com/wesnoth/wesnoth/pull/729/commits/5f52825e2d didn't anyone look at the bug yet?
20160811 21:28:03< tad__> Well it's a name. "before victory event" ..
20160811 21:28:49< zookeeper> oh you were talking only about blackwater port? sure, nevermind then, split it if you want
20160811 21:29:20< celmin> I've noticed the bug in my test scenario (pretty sure you shouldn't have a unit preselected when a scenario starts) but not investigated.
20160811 21:29:38< celmin> I was meaning to investigated, but someone else might get to it first.
20160811 21:29:56< tad__> Well, I split it into "victory dance" and "victory" a few places when you get to the choose_level commit at the end of "general improvements"
20160811 21:30:03< celmin> …though recently I'm not sure I'm if I'm still getting it...
20160811 21:30:58< tad__> As to the bug I did some digging and am sure it's inside there but, for the life of me, could not see any reason for a unit to become selected by what's being done in those macro.
20160811 21:31:42< zookeeper> it's a bug in [object]
20160811 21:31:51< tad__> Somewhere, yes.
20160811 21:35:21< tad__> Don't know if it happens outside the prestart. Never noticed it anywhere else. But tested and in those three HttT scenario it bites you during prestart. In the last one, you may not notice because it will be a player-owned unit selected.
20160811 21:35:38< zookeeper> hilarious, https://github.com/wesnoth/wesnoth/pull/729/commits/92a9b13c <- those filters have been wrong since they were added in 2006
20160811 21:35:45< celmin> So it's not in src/lua/object.lua then.
20160811 21:36:07< gfgtdf> tad__: endign the level with :n shoudl usually also fire the victory evne so i wonder why you need a seperate event?
20160811 21:36:20< celmin> That suggests it's a bug in unit::add_modification.
20160811 21:36:33< celmin> Or something closely related.
20160811 21:37:00< gfgtdf> unit::add_modification shouldn't select anything
20160811 21:37:00< zookeeper> Ivanovic, ^ sorry that it took 10 years for someone to notice, but your code never worked :>
20160811 21:37:04< celmin> gfgtdf: I know
20160811 21:37:20< gfgtdf> that is unless it uses some lua defeined effect (apply_to=) that selects the unti
20160811 21:37:29< celmin> gfgtdf: It doesn't
20160811 21:37:41< tad__> gfgtdf: Consider Blackwater Port. In the "victory" event you have a lot of talking and a long move_unit_fake .. you can skip the talking with Esc but still have to wait out the move. By splitting the player-visible stuff you don't need Esc.
20160811 21:38:08-!- prkc [~prkc@192.40.89.71] has joined #wesnoth-dev
20160811 21:38:35< zookeeper> well, esc should really be made to skip move_unit_fakes too
20160811 21:38:53< gfgtdf> tad__: sounds like a rather small advantage to me but well..
20160811 21:38:58< tad__> That may be. But I'm after getting things to work best I can as it is now.
20160811 21:39:46< tad__> To the player it's no difference at all. But we debuggers can fly with :n and choose_level, at least.
20160811 21:42:49< tad__> @celmin: yeah, I got down to it being in [object] and code-read the Lua for [object] and did not see it there. If I had, I'd have PR'd a separate change for the Lua.
20160811 21:42:51-!- Appleman1234_ [~Appleman1@KD036012020177.au-net.ne.jp] has joined #wesnoth-dev
20160811 21:43:18< celmin> I'm thinking it could be something in wesnoth.add_modification, though maybe there are other possible points for the problem.
20160811 21:43:41< celmin> Actually, come to think of it… does it use add_modication during rebuilding...
20160811 21:43:50-!- Appleman1234 [~Appleman1@KD036012028079.au-net.ne.jp] has quit [Disconnected by services]
20160811 21:43:58-!- Appleman1234_ is now known as Appleman1234
20160811 21:44:44< tad__> What concerns me is I have only ever seen it during prestart and that might mean it's something uninitialized. And those can be a real bitch to find.
20160811 21:45:12< celmin> I wonder if it ever happens when a unit levels up.
20160811 21:45:31< celmin> (A unit that has the object, naturally.)
20160811 21:45:52< celmin> If it's related to [object] and it happens on prestart, it could be part of the unit::advance_to process...
20160811 21:46:39< tad__> As I read the code: In HttT that [object] is applied during prestart AND whenever Li'sar recruits. And it does not affect recruits.
20160811 21:47:15< celmin> It's applied through ActionWML on prestart?
20160811 21:47:32< celmin> ie, it's not an object the unit already has before the scenario starts?
20160811 21:47:40-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160811 21:47:40-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Read error: Connection reset by peer]
20160811 21:47:44< zookeeper> tad__, ok i made it through. i think the only one i might find actually objectionable is https://github.com/wesnoth/wesnoth/pull/729/commits/864cb0537c7
20160811 21:48:16< gfgtdf> celmin: i wonder why you think its not in lua/object.lua ?
20160811 21:48:18< celmin> I was assuming that's because it's a cutscene scenario?
20160811 21:48:26< celmin> gfgtdf: I might be misunderstanding the situation.
20160811 21:48:44< zookeeper> tad__, i understand the rationale, but if the player is not given a save, then they have to replay the whole previous scenario to see it again, if they need to
20160811 21:48:46< celmin> gfgtdf: I thought it was units that had an object already in their [modifications] from the carryover data.
20160811 21:49:00< tad__> @zookeeper: That says we don't need a scenario-start save file for the cut-sceen (no player interaction) and it's done that way elsewhere.
20160811 21:49:09< celmin> But if it's a prestart event with [object], then it could be in the Lua code after all.
20160811 21:49:15< gfgtdf> i means, object.lua is clearly calling wesnoth.select_unit and wesnoth.highlight_hex
20160811 21:49:25< celmin> Ooh.
20160811 21:49:39< celmin> So maybe it is in the Lua.
20160811 21:50:14< tad__> The save=no applies to the scenarion-start save for the next scenario. I know, poor choice of words for the attribute.
20160811 21:50:34< celmin> Yeah, in Lua it's callde prescenario_save instead.
20160811 21:50:37< celmin> ^called
20160811 21:50:41< celmin> (And in C++ too.)
20160811 21:50:46< tad__> You're thinking "savereplay" or something like that.
20160811 21:50:51< celmin> replay_save
20160811 21:50:56< tad__> yeah
20160811 21:51:44< zookeeper> tad__, it's used mainly for epilogues, where it does make sense. but we can't have a long scenario followed by a long discussion the player might not have any interest in reading right now, and force them to replay the whole scenario if they need to see the discussion again, if they happen to not have an autosave handy.
20160811 21:53:18< tad__> Ah. OK. That's a good argument. I will back out that change. And when I get to DM, I'll yank it from the existing places it appears instead of adding it where it was missed, for the same reason.
20160811 21:54:13< zookeeper> i wonder why in EI someone has put save=no in a result=defeat [endlevel]...
20160811 21:54:27< celmin> An excellent question.
20160811 21:54:35< zookeeper> likely answer: sloppy copy-paste :p
20160811 21:54:48< tad__> Or they wanted to be DAMNED sure? 
20160811 21:56:40< tad__> What I miss is getting a scenario-start save for S01. I've taken to doing a choose_level right off so I have one. Makes it a lot easier to Load all the time instead of having to remember S01 needs New Campaign.
20160811 21:57:00< celmin> Eh, did I break put_unit...
20160811 21:58:20< zookeeper> tad__, i think save=no probably won't make sense anywhere where you do have linger mode. where linger mode isn't used (final scenarios, or for some lesser cutscenes), it might be appropriate.
20160811 21:59:47< tad__> Well, I heard your argument and thought .. ya'know .. there have been times I Esc'd past a cut scene and regretted it and some places you have to go back and hope you have a saved turn to see it again
20160811 22:01:08< zookeeper> especially with a particularly important little cutscene as this one it's better to not make that mistake easier :>
20160811 22:02:14< tad__> Well, as a new player (to a campaign), you don't know what it important. And as an old player (other campaigns) you suddenly wonder if you just missed something important and it is a bummer to have to replay just to read it.
20160811 22:03:58< tad__> So, I'm going over all my old work and using what I've learned as a guide. I will pull that change from HttT and keep an eye out everywhere else for save=no where it should not be.
20160811 22:05:16< zookeeper> in some situations it might make sense to let the cutscene scenario have the save, but not the following scenario. arguably not in this case since it's such a major cutscene, but in some smaller cutscenes which directly ties into the next scenario, it might.
20160811 22:06:17< tad__> Well, consistency is good. I'd suggest every scenario get a scenario-start with the sole exception of the Epilog ..and suggest we might even re-consider that.
20160811 22:07:40< tad__> sigh .. I have to go. Wife has scheduled us for a party in 2 hours and is just now telling me about it !!! 
20160811 22:07:45< zookeeper> sure
20160811 22:08:04< zookeeper> but as said, i think we can merge the PR once that one is taken care of
20160811 22:08:54< tad__> I will push the change this evening .. I have 2 other changes for other PRs in the series I want to make but will hold them because once "errors" is pushed I may need to push all the others to remvoe it from the mess.
20160811 22:09:09< zookeeper> obviously i can't haven't reviewed all the actual code besides cursory glances, but there wasn't anything else objectionable on paper
20160811 22:10:11< tad__> Oh .. and I ahve a 1-liner PR which I added this AM. Moves the random seed message from cage generator using in HttT to info log level. Once that is in, I can run HttT with NO messages on screen or log .. at all.  YEAH!
20160811 22:10:35< celmin> Oh yeah, I remember seeing that and thinking "I should merge this".
20160811 22:10:55< celmin> I don't remember why I didn't.
20160811 22:11:32< zookeeper> urgh. lots to correct in this thread about donations, it seems: https://forums.wesnoth.org/viewtopic.php?f=6&t=44448
20160811 22:11:45< irker830> wesnoth: Gregory A Lundberg wesnoth:master db7d3b764611 / src/generators/cave_map_generator.cpp: Use info level https://github.com/wesnoth/wesnoth/commit/db7d3b76461141b25e39b1575162cdd3b6bdd0c0
20160811 22:11:47< irker830> wesnoth: Celtic Minstrel wesnoth:master d83acc004a2d / src/generators/cave_map_generator.cpp: Merge pull request #742 from GregoryLundberg/GL_cave_seed_to_info https://github.com/wesnoth/wesnoth/commit/d83acc004a2d4aee62a4da721136a005785138df
20160811 22:11:53< zookeeper> "you can donate to devs by buying iOS port" sounds incorrect based on recent discussions... :p
20160811 22:12:07< tad__> I was going to say .. I started all this because I was thinking of a UMC and had questions of the main campaigns. And saw an error message when I looked. "I can fix that." and then I saw the commetn about polishing for Steam and it all snowballed. Hope y'all don't mind.
20160811 22:12:48-!- louis94 [~~louis94@91.178.242.169] has quit [Ping timeout: 250 seconds]
20160811 22:14:25< tad__> need to go. ttyl
20160811 22:14:29-!- tad__ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Quit: Page closed]
20160811 22:15:21< zookeeper> well i don't mind reviewing improvements... but it would be easier if i could just check and merge like one or two individual simple PR's per day :p
20160811 22:15:28< celmin> Heh.
20160811 22:22:47< gfgtdf> is there a way to filter in filter_wml whether an attribute is non-empty (existent)?
20160811 22:24:15< celmin> Uh. Not quite sure.
20160811 22:24:18< gfgtdf> i have [object]s that have an item_number=<number> and i want to use wesnoth.remove_modifications t remvoe all objects that have an item_number= attribute
20160811 22:24:27< celmin> The code for that would be in config.cpp, I think.
20160811 22:24:58< gfgtdf> of course i coudl add a is_numberd_item=yes on those objects but that woudl somehow be dubblcate information.
20160811 22:25:00< celmin> Writing [not]item_number= might work.
20160811 22:25:17< celmin> I guess [not][filter_wml]item_number=
20160811 22:25:29-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has quit [Quit: ChipmunkV]
20160811 22:25:33< celmin> Ah, right, you're working in Lua though.
20160811 22:25:43< celmin> Hmm.
20160811 22:25:51< vultraz> gfgtdf: no, the status table doesn't have a problem, listboxes do https://gna.org/bugs/?24780
20160811 22:26:22< vultraz> gfgtdf: if you (or anyone) could please take a look at that bug because it stumps me >_>
20160811 22:27:21< vultraz> gfgtdf: I know the cause: the header grid is no longer inside the _content_grid grid, meaning it's excluded from certain calculations  
20160811 22:27:38< celmin> I'm not sure if blank is considered equal to ""
20160811 22:27:50< vultraz> gfgtdf: but the header grid needs to be outside the _content_grid if you don't want the header scrolled
20160811 22:28:10< gfgtdf> celmin: its not i just looked at the code
20160811 22:29:01< celmin> Oh, [not] is allowed in filter_wml, huh.
20160811 22:29:50< zookeeper> gfgtdf, to test variables elsewhere, one way has been to do [variable] name=doesthisexist value=$emptyfoobarasdf [/variable], dunno if the same would work in [filter_wml].
20160811 22:31:14< zookeeper> gfgtdf, in your case i'd just give the object another attribute as well and filter by that
20160811 22:31:20< celmin> I suppose we could extend filter_wml with some magic attributes.
20160811 22:31:34< celmin> Yeah, what zookeeper said is probably simplest though.
20160811 22:32:11< gfgtdf> zookeeper: ok i'll just add the other attribute.
20160811 22:33:03< zookeeper> i usually abuse the name= key for that
20160811 22:33:28< zookeeper> but i think you can now just put arbitrary keys there
20160811 22:34:14< celmin> add_modification copies the [object] tag verbatim, so yeah, arbitrary keys work.
20160811 22:34:38< zookeeper> arguably it might be unwise to rely on that kind of undocumented behavior, but...
20160811 22:34:54 * vultraz has abused that
20160811 22:35:01< gfgtdf> zookeeper: i'm quite sure that there are anough other addons that also bause that :)
20160811 22:35:17< gfgtdf> relied in that i mean
20160811 22:35:20< gfgtdf> on*
20160811 22:35:31< zookeeper> yeah. many unwise authors... like myself!
20160811 22:35:46< celmin> Someone should make a WML unit test to preserve it. (Maybe I'll do that. >_> )
20160811 22:36:32< gfgtdf> well the disadvantage of the currentl beviosu is tha the unleesary code in[object] (liek [else] and [filter]) will also be copyied into the unit
20160811 22:36:45< celmin> That'd be easily fixed actually, in the Lua.
20160811 22:37:58< vultraz> celmin: this feels very hacky http://pastebin.com/3B1BpGLu
20160811 22:38:07< vultraz> or at least, inelegant.. 
20160811 22:38:46< celmin> I feel like there's a method in config that would satisfy that requirement...
20160811 22:39:34< celmin> Try cfg.child("multiplayer", index)
20160811 22:39:48< celmin> Maybe with & in front.
20160811 22:40:05< celmin> The child iterator should probably be random-access though, actually...
20160811 22:40:11< celmin> Maybe I can do that.
20160811 22:42:18< celmin> Still, if you need the equivalent of $child[$n], you should probably be using config::child like I just said.
20160811 22:45:50-!- travis-ci [~travis-ci@ec2-54-80-121-62.compute-1.amazonaws.com] has joined #wesnoth-dev
20160811 22:45:51< travis-ci> wesnoth/wesnoth#10308 (master - d83acc0 : Celtic Minstrel): The build failed.
20160811 22:45:51< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/151655907
20160811 22:45:51-!- travis-ci [~travis-ci@ec2-54-80-121-62.compute-1.amazonaws.com] has left #wesnoth-dev []
20160811 22:45:59< celmin> Huh?
20160811 22:46:12< celmin> Oh right.
20160811 22:46:45-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 244 seconds]
20160811 22:47:02< vultraz> celmin: that works
20160811 22:47:04< vultraz> thanks
20160811 22:47:49< celmin> Oh, looks like the broken build is vultraz's fault?
20160811 22:47:53< celmin> Somehow?
20160811 22:48:08< vultraz> well I have broken the editor
20160811 22:48:08< celmin> It's the GUI2 tests and the experimental MP Create dialog.
20160811 22:48:35< celmin> You need to pass the engine to the constructor.
20160811 22:49:53< vultraz> in the tests?
20160811 22:50:28< celmin> Not in the tests.
20160811 22:50:40< celmin> Wait, no.
20160811 22:50:43< celmin> Maybe it is in the tests?
20160811 22:51:03< vultraz> I added an argument the gui2 mp create ctor 
20160811 22:51:14< vultraz> so if it's used in the tests, it needs to be added
20160811 22:51:49-!- louis94 [~~louis94@91.178.242.169] has joined #wesnoth-dev
20160811 22:55:23-!- enchi [enchilado@defocus/yummy/enchilado] has quit [Ping timeout: 244 seconds]
20160811 23:00:06-!- enchi [enchilado@defocus/yummy/enchilado] has joined #wesnoth-dev
20160811 23:16:22-!- fabi [~fabi@176.7.89.74] has joined #wesnoth-dev
20160811 23:16:36< fabi> hi
20160811 23:18:42< vultraz> hi
20160811 23:19:55< vultraz> celmin: is it good to loop through all [options] tags in a scenario, should there be more than 1?
20160811 23:20:36-!- prkc [~prkc@192.40.89.71] has quit [Ping timeout: 276 seconds]
20160811 23:20:52< celmin> I don't think there'd be more than one, but I don't think it would hurt to iterate over a child_range either. You could check out ScenarioWML, ModificationWML, and related wiki pages if you want though.
20160811 23:21:21< celmin> EraWML and CampaignWML I guess are the others. Might be ModificationsWML rather than without the S.
20160811 23:26:49< vultraz> how... is there no clear() member in ttree_view
20160811 23:27:13< celmin> Good question. Add one?
20160811 23:27:19< celmin> Seems like a necessary thing to have.
20160811 23:27:31< celmin> Although, what does ttree_view inherit from?
20160811 23:27:36< vultraz> hm.
20160811 23:27:39< vultraz> ttree_view_node does, though
20160811 23:28:24< vultraz> let's see if this even works
20160811 23:29:53< vultraz> nope
20160811 23:30:03< vultraz> what am I doing wrong..
20160811 23:33:02< celmin> Is wesnoth.put_unit broken?
20160811 23:33:16< celmin> If used on an existing unit with new coordinates, shouldn't it move the unit?
20160811 23:33:21< Aginor> hi
20160811 23:33:23< celmin> Hi
20160811 23:34:18< celmin> Example: wesnoth.put_unit( wesnoth.get_unit(15, 15), 16, 16) should move the unit on (15,15) to (16,16), right?
20160811 23:34:53-!- prkc [~prkc@192.40.89.71] has joined #wesnoth-dev
20160811 23:35:42< celmin> (Without animating it of course)
20160811 23:35:55< gfgtdf> yes i think it shoudl do that
20160811 23:36:12< celmin> Okay, so does it work on master?
20160811 23:36:28< gfgtdf> celmin: dont know i think i never used it liek that
20160811 23:36:47< celmin> It doesn't work for me, but I've got quite a few changes here, some of which might have affected it.
20160811 23:49:04< vultraz> why doesn't this work
20160811 23:49:06< vultraz> ughh
20160811 23:50:14< vultraz> oh, I had the wrong node ID
20160811 23:51:43< vultraz> still doesn't work :|
20160811 23:53:23< vultraz> oh
20160811 23:53:25< vultraz> duhh 
20160811 23:53:35< vultraz> im checking something with no custom options :|
20160811 23:54:01< vultraz> ok, so an entry comes up..
--- Log closed Fri Aug 12 00:00:57 2016